AE&M 63-3 by Sociedade Brasileira de Endocrinologia e Metabologia

ISSN 2359-4292

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM Vol. 63 – No. 03 – June 2019

editorial 187 Is testosterone involved in low female sexual desire? Carmita H. N. Abdo

consensus

Archives of Endocrinology

190 Testosterone therapy for women with low sexual desire: a position statement from the Brazilian Society of Endocrinology and Metabolism

OFFICIAL JOURNAL OF THE BRAZILIAN original articles SOCIETY OF without 199 Hypothyroid offspring replacement with euthyroid wet nurses during lactation improves thyroid programming modifying metabolic programming ENDOCRINOLOGY Jorge Tapia-Martínez, Margarita Franco-Colín, Rocio Ortiz-Butron, Marisol Pineda-Reynoso, Edgar Cano-Europa 208 Does having Turner syndrome affect quality of life in Brazilian women compared to common population? AND METABOLISM Maria Bernarda Estevez, Patricia Teofilo Monteagudo, Kelly Christina Oliveira, Ieda Therezinha do Nascimento Verreschi

Rita V. Weiss, Alexandre Hohl, Amanda Athayde, Dolores Pardini, Larissa Gomes, Monica de Oliveira, Ricardo Meirelles, Ruth Clapauch, Poli Mara Spritzer

and Metabolism

215 Familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families from Muzambinho (Minas Gerais, Brazil)

João Paulo dos Anjos Souza Barbosa, Luciano Basso,Teresa Bartholomeu, Januária Andrea Souza Rezende, Jorge Alberto de Oliveira, António Prista, Go Tani, José António Ribeiro Maia, Cláudia Lúcia de Moraes Forjaz

222 The differences in homeostasis model assessment values in type 2 diabetic patients with different lengths of history of diabetes Chen Wang, Zaibo Liu, Peng Zhang, Xiaolong Ma, Kui Che4, Yangang Wang

228 An early stage in T4-induced hyperthyroidism is related to systemic oxidative stress but does not influence the pentose cycle in erythrocytes and systemic inflammatory status

Archives of

Rayane Brinck Teixeira, Tânia Regina Gattelli Fernandes-Piedras, Adriane Belló-Klein, Cristina Campos Carraro, Alex Sander da Rosa Araujo

235 Clinical and functional variables can predict general fatigue in patients with acromegaly: an explanatory model approach

André da Cunha Michalski, Arthur de Sá Ferreira, Leandro Kasuki, Monica R. Gadelha, Agnaldo José Lopes, Fernando Silva Guimarães

241 The relationship of flavonoid intake during pregnancy with excess body weight and gestational diabetes mellitus

Mariana de Andrade Balbi, Lívia Castro Crivellenti, Daniela Cristina Candelas Zuccolotto, Laércio Joel Franco, Daniela Saes Sartorelli

Endocrinology

250 Searching for mutations in the HNF1B gene in a Brazilian cohort with renal cysts and hyperglycemia

Renata P. Dotto, Lucas Santos de Santana, Susan C. Lindsey, Lilian Araújo Caetano, Luciana F. Franco, Regina Célia M. S. Moisés, João R. Sa, José Luiz Nishiura, Milena Gurgel Teles, Ita P. Heilberg, Magnus R. Dias-da-Silva, Fernando M. A. Giuffrida, André F. Reis

258 Arterial stiffness in hyperthyroid patients is deteriorated due to thyroid hormones

Canan Yildiz, Mustafa Altay, Sedat Yildiz, Yavuz Çag6ir, Tolga Akkan, Yasemin Aydog6an Ünsal, Esin Beyan

and Metabolism

265 Body composition and nutritional and metabolic parameters in postmenopausal women sufficient, insufficient and deficient in vitamin D Luisa Amábile Wolpe Simas, Leila Caroline Bianchet Zanatta, Carolina Aguiar Moreira, Victoria Zeghbi Cochenski Borba, Cesar Luiz Boguszewski

272 Paraoxonase 1 serum activity in women: the effects of menopause, the C(-107)T polymorphism and food intake

Mauren Castro Ritta1, Aline Marques Baldez, Isabel Oliveira de Oliveira, Driele Neske Garcia, Paola Spiering Souza, Kelvin Ruan da Silva Andrade, Sandra Costa Valle, Simone Pieniz, Carlos Castilho Barros, Michal M. Masternak, Augusto Schneider

OFFICIAL JOURNAL OF THE BRAZILIAN

280 MTHFR gene polymorphisms in hypothyroidismSOCIETY and hyperthyroidism among Jordanian females OF ENDOCRINOLOGY AND METABOLISM

Diala W. Abu-Hassan, Abdullah N. Alhouri, Nadera A. Altork, Zakaria W. Shkoukani, Tamer Salhab Altamimi, Omar M. Alqaisi, Baha Mustafa

288 Renal resistive index in patients with polycystic ovary syndrome Gülsen Tüfekçiog6lu, SÇakir Özgür KesÇkek, Okan Dilek, Cengiz Yilmaz

293 In properly selected patients with differentiated thyroid cancer, antithyroglobulin antibodies decline after thyroidectomy and their sole presence should not be an indication for radioiodine ablation Luis Felipe Zavala, María Inés Barra, Roberto Olmos1, Michael Tuttle, Hernán González, Nicolás Droppelmann, Lorena Mosso, José M. Dominguez

300 Diagnostic performance of thyroid ultrasound in Hürthle cell carcinomas

Nathalie Oliveira Santana, Ricardo Miguel Costa Freitas, Vinicius Neves Marcos, Maria Cristina Chammas, Rosalinda Yossie Asato Camargo, Cláudia Kliemann Schmerling, Felipe Augusto Brasileiro Vanderlei5, Ana Oliveira Hoff, Suemi Marui, Debora Lucia Seguro Danilovic

review 306 Insufficient iodine intake in pregnant women in different regions of the world: a systematic review

Aline C. Candido, Núbia de S. de Morais, Luiza V. Dutra, Carina A. Pinto, Sylvia do C. C. Franceschini, Rita de Cássia G. Alfenas

AND METABOLISM OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM

Financial and editorial assistant: Roselaine Monteiro roselaine@endocrino.org.br Rua Botucatu, 572 – conjunto 83 – 04023-062 – São Paulo, SP Telefax: (11) 5575-0311 / 5082-4788

Archives of Endocrinology Online submission / Electronic publishing www.aem-sbem.com • www.scielo.br/abem

and Metabolism

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM

Archives of

Rua Anseriz, 27, Campo Belo 04618-050 – São Paulo, SP. Fone: 11 3093-3300 www.segmentofarma.com.br • segmentofarma@segmentofarma.com.br

Publication code: 23728.7.19

Endocrinology

and Metabolism

All adds have to comply with RDC no 96/08 Commercial advising: Estela Kater estela.kater@gmail·com

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM

Indexed in Biological Abstracts, Index Medicus, Latindex, Lilacs, MedLine, SciELO, Scopus, ISI-Web of Science BRAZILIAN ARCHIVES OF ENDOCRINOLOGY AND METABOLISM Brazilian Society of Endocrinology and Metabolism – São Paulo, SP: Brazilian Society of Endocrinology and Metabolism, volume 5, 1955Six issues/year Continued from: Brazilian Archives of Endocrinology (v. 1-4), 1951-1955 ISSN 2359-4292 (online issues) 1. Endocrinology – journals 2. Metabolism – journals I. Brazilian Society of Endocrinology and Metabolism II. Brazilian Medical Association CDU 612.43 Endocrinology CDU 612.015.3 Metabolism

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM

Archives of endocrinology and metabolism Official journal of SBEM – Brazilian Society of Endocrinology and Metabolism (Department of the Brazilian Medical Association), SBD – Brazilian Diabetes Society, ABESO – Brazilian Association for the Study of Obesity and Metabolic Syndrome

2019-2022 EDITOR-IN-CHIEF Marcello D. Bronstein (SP)

CO-EDITORS

REPRESENTATIVES OF COLLABORATING SOCIETIES SBD

Léa Maria Zanini Maciel (SP)

ABESO

Leandro Kasuki (SP)

OFFICIAL JOURNALLuiz Alberto Andreotti Turatti (SP) OF THE BRAZILIANMadson Queiroz Almeida (SP) Brazilian Editorial Commission Manoel Ricardo Alves Martins (CE) Alexander A. L. Jorge (SP) OF SOCIETY Marcio Mancini (SP) Alexandre Hohl (SC) ENDOCRINOLOGY Margaret Cristina S. Boguszewski (PR) Ana Amélia Hoff (SP) AND METABOLISM Maria Candida B. V. Fragoso (SP) Ana Claudia Latronico (SP) Maria Edna de Melo (SP)

and Metabolism FOUNDER Waldemar Berardinelli (RJ)

1951-1955 Waldemar Berardinelli (RJ) Thales Martins (RJ) 1957-1972 Clementino Fraga Filho (RJ) 1964-1966* Luiz Carlos Lobo (RJ)

Maria Marta Sarquis (MG)

Andrea Glezer (SP)

Mario Saad (SP)

Ayrton Custódio Moreira (SP)

Marise Lazaretti Castro (SP)

INTERNATIONAL ASSOCIATE EDITOR

Berenice B. Mendonça (SP)

Milena Caldato (PA)

Bruno Halpern (SP)

Raquel Soares Jallad (SP)

Carlos Alberto Longui (SP)

Rodrigo Moreira (RJ)

César Luiz Boguszewski (PR)

Ruth Clapauch (RJ)

Archives of

ASSOCIATE EDITORS

Endocrinology

PRESIDENTS OF THE SBEM DEPARTMENTS

ADRENAL AND HYPERTENSION

Madson Queiroz de Almeida (SP) DIABETES MELLITUS

Clarisse Ponte (CE)

Sandra R. G. Ferreira (SP)

Delmar Muniz Lourenço Jr. (SP)

Simão A. Lottemberg (SP)

Denise Momesso (RJ)

Sonir Roberto Antonini (SP)

Eder Carlos R. Quintão (SP)

Suemi Marui (SP)

and Metabolism

Hermelinda Cordeiro Pedrosa (DF)

Edna Nakandakare (SP)

DYSLIPIDEMIA AND ATHEROSCLEROSIS

Edna T. Kimura (SP)

1969-1972* João Gabriel H. Cordeiro (RJ)

Maria Izabel Chiamolera (SP)

BASIC ENDOCRINOLOGY

Mônica de Oliveira (PE)

PEDIATRIC ENDOCRINOLOGY

1991-1994 Rui M. de Barros Maciel (SP)

BONE AND MINERAL METABOLISM

Carlos Alberto Longui (SP) Miguel Madeira (RJ)

1995-2006 Claudio Elias Kater (SP)

NEUROENDOCRINOLOGY

2007-2010 Edna T. Kimura (SP)

OBESITY

Heraldo Mendes Garmes (SP) Mario Kehdi Carra (SP) THYROID

Jose Augusto Sgarbi (SP)

Victória Borba (PR)

Elaine Maria Frade Costa (SP) Felipe Gaia (SP)

OFFICIAL JOURNAL OF THE FlavioBRAZILIAN Hojaij (SP) FEMININE ENDOCRINOLOGY AND SOCIETY OF ENDOCRINOLOGY AND METABOLISM Gil Guerra-Júnior (SP) ANDROLOGY

1983-1990 Antônio Roberto Chacra (SP)

2015-2018 Marcello D. Bronstein (SP)

André Fernandes Reis (SP)

Mário Vaisman (RJ)

Cynthia Melissa Valério (RJ)

2011-2014 Sergio Atala Dib (SP)

Maria Izabel Chiamolera (SP)

Antônio Roberto Chacra (SP)

1966-1968* Pedro Collett-Solberg (RJ)

1978-1982 Armando de Aguiar Pupo (SP)

Ana Luiza Silva Maia (RS)

Tânia S. Bachega (SP)

Shlomo Melmed (Los Angeles, EUA)

EDITORS-IN-CHIEF, EDITORIAL OFFICE*

Larissa Gomes (SP)

Luiz Alberto Andreotti Turatti (SP)

Archives of Endocrinology Ana Luiza Maia (RS) Beatriz D’ Agord Schaan (RS) Bruno Ferraz de Souza (SP) Erika Parente (SP) Francisco Bandeira (PE) Fernanda Vaisman (RJ) Fernando M. A. Giuffrida (BA) João Roberto Maciel Martins (SP) Melanie Rodacki (RJ) Monica R. Gadelha (RJ) Nina Rosa C. Musolino (SP) Poli Mara Spritzer (RS) Ricardo Meirelles (RJ) Rogerio Friedman (RS)

Laércio Joel Franco (SP)

International Editorial Commission Antonio C. Bianco (EUA)

Giovanna Balarini Lima (RJ)

Décio Eizirik (Bélgica)

Gisah M. do Amaral (SP)

Franco Mantero (Itália)

Hans Graf (SP)

Fernando Cassorla (Chile)

José Augusto Sgarbi (SP)

Gilberto Paz-Filho (Austrália)

José Gilberto H. Vieira (SP)

John P. Bilezikian (EUA)

Julio Z. Abucham (SP)

Ken Ho (Austrália)

SBEM – Brazilian Society of Endocrinology and Metabolism SBEM BRAZILIAN BOARD OF DIRECTORS 2019-2020 President Vice-President Executive Secretary Adjunct Executive Secretary Treasurer-General Adjunct Treasurer

Rodrigo de Oliveira Moreira César Luiz Boguszewski Neuton Dornelas Gomes Marcio Corrêa Mancini Paulo Augusto Carvalho Miranda Maria Amazonas

Rua Humaitá, 85, cj. 501 22261-000 – Rio de Janeiro, RJ Fone/Fax: (21) 2579-0312/2266-0170 www.endocrino.org.br sbem@endocrino.org.br

Scientific Departments - 2019/2020 Brazilian Society of Endocrinology and Metabolism ADRENAL AND HYPERTENSION

DIABETES MELLITUS

President Madson Queiroz de Almeida madsonalmeida@usp.br

President Hermelinda Cordeiro Pedrosa pedrosa.hc@globo.com Vice-President Rosane Kupfer Directors Erika Bezerra Parente Thaisa Dourado Guedes Trujilho Fábio Ferreira de Moura Janice Sepulveda Reis Ana Cristina Ravazzani de Almeida

Vice-President Directors

Milena Coelho Fernandes Caldato Marivânia da Costa Santos Flávia Amanda Costa Barbosa Leonardo Vieira Neto

DYSLIPIDEMIA AND ATHEROSCLEROSIS

BASIC ENDOCRINOLOGY

President Cynthia Melissa Valério cy_valerio@yahoo.com.br

President Maria Izabel Chiamolera mchiamolera@unifesp.br Vice-President Maria Tereza Nunes Directors Tania Maria Ruffoni Ortiga Bruno Ferraz de Souza Marisa Maria Dreyer Breitenbach Luciani Renata Silveira Carvalho Vania Maria Correa da Costa

Vice-President Directors

Renan Magalhães Montenegro Júnior Joaquim Custódio da Silva Júnior Marcello Casaccia Bertoluci Joana Rodrigues Dantas Pereira

Scientific Departments - 2019/2020 WOMEN ENDOCRINOLOGY AND ANDROLOGY President Mônica de Oliveira morecife@hotmail.com Vice-President Directors

Rita de Cassia Viana Vasconcellos Weiss Dolores Perovano Pardini Ruth Clapauch Larissa Garcia Gomes Marcelo Fernando Ronsoni Alexis Dourado Guedes

PEDIATRIC ENDOCRINOLOGY President Carlos Alberto Longui carloslongui@msn.com Vice-President Directors

Julienne Ângela Ramires de Carvalho Sonir Roberto Rauber Antonini Claudia Braga Monteiro Paulo César Alves da Silva

BONE AND MINERAL METABOLISM

NEUROENDOCRINOLOGY

President Vice-President Directors

President Heraldo Mendes Garmes heraldmg@uol.com.br Vice-President Manoel Ricardo Alves Martins Directors Andrea Glezer Leandro Kasuki Jomori de Pinho Mauro Ântonio Czepielewski Nina Rosa de Castro Musolino Silvia Regina Correa da Silva

Miguel Madeira migmadeira@gmail.com Francisco Alfredo Bandeira e Farias Carolina Aguiar Moreira Sérgio Setsuo Maeda Marise Lazaretti Castro Bárbara Campolina Carvalho Silva Catarina Brasil D´alva

OBESITY

THYROID

President Mario Kehdi Carra mariocarra@gmail.com Vice-President Fábio Rogério Trujilho Directors Maria Edna de Melo Rosana Bento Radominski Rodrigo Lamounier Bibiana Prada de Camargo Colenci Jacqueline Rizzolli

President José Augusto Sgarbi jose.sgarbi@gmail.com Vice-President Patricia de Fátima dos Santos Teixeira Directors Rafael Selbach Scheffel Gisah Amaral de Carvalho Danilo Glauco Pereira Villagelin Neto João Roberto Maciel Martins (SP)

Permanent Commissions - 2019/2020 Brazilian Society of Endocrinology and Metabolism STRATEGIC PLANNING FOLLOW-UP

HISTORY OF ENDOCRINOLOGY

President Alexandre Hohl alexandrehohl@uol.com.br Members Nina Rosa de Castro Musolino, Airton Golbert, Ricardo Martins da Rocha Meirelles, Ruy Lyra da Silva Filho

President Henrique de Lacerda Suplicy hsuplicy@gmail.com Members Adriana Costa e Forti, Thomaz Rodrigues Porto da Cruz

ENDOCRINOLOGY CAMPAIGNS

President

Mônica Roberto Gadelha

mgadelha@hucff.ufrj.br

Members

César Luiz Boguszewski

President Erika Bezerra Parente ebparente@gmail.com Members Alina Coutinho Rodrigues Feitosa, Carolina Ferraz da Silva, Mariana Guerra Paulino Guerra

SCIENTIFIC COMISSION President César Luiz Boguszewski clbogus@uol.com.br Indicated by the directories Sonir Roberto Rauber Antonini, Carolina Aguiar Moreira, Fernanda Vaisman, Bruno Halpern, Manoel Ricardo Alves Martins,Fábio Rogério Trujilho, Luciana Antunes de Almeida Secchi, Karen Faggioni de Marca Seidel, Fernando Gerchman, Milena Coelho Fernandes Caldato

SOCIAL COMMUNICATION President Ricardo Martins da Rocha Meirelles r.meirelles@terra.com.br Nominated by the president Alexandre Hohl ABEM Editor Marcello D. Bronstein Members Nina Rosa de Castro Musolino

INTERNATIONAL

NORMS, QUALIFICATION AND CERTIFICATION President Vivian Carole Moema Ellinger vivianellinger@gmail.com Members Ronaldo Rocha Sinay Neves, Marisa Helena César Coral, Maria Emilia Pereira de Almeida, Milena Coelho Fernandes Caldato

JOINT COMMISSION – CAAEP President Julienne Angela Ramires de Carvalho julienne@endocrinoped.com.br Members Marilia Martins Guimarães, Suzana Nesi França

RESEARCH President Members

Freddy Eliaschewitz freddy.g@uol.com.br Antônio Roberto Chacra, Luiz Augusto Tavares Russo

GUIDELINES PROJECT CONTINUOUS MEDICAL EDUCATION President João Eduardo Nunes Salles jensalles@yahoo.com.br Members César Luiz Boguszewski, Alexandre Hohl

STATUTES, RULES AND REGULATIONS President Fábio Rogério Trujilho trtruji@terra.com.br Members Alessandra Ferri Casini, Cynthia Melissa Valério, Ruy Lyra da Silva Filho, Wellington Santana da Silva Júnior

PROFESSIONAL ETHICS AND DEFENCE President Maite Trojaner Salona Chimeno etica@endocrino.org.br Vice-Inspector Itairan da Silva Terres 1ST member Diana Viegas Martins 2ND member João Modesto Filho 3RD member Luis Henrique Santos Canani 4TH member Márcio Weissheimer Lauria 5TH member Cristina Bardou Pizarro

ENDOCRINE DYSREGULATORS President Elaine Maria Frade Costa elainefradecosta@gmail.com Vice-President Maria Izabel Chiamolera Members Eveline Gadelha Pereira Fontenele, Tânia Aparecida Sanchez Bachega, Ricardo Martins da Rocha Meirelles

Coordinator Alexis Dourado Guedes dr.alexis@uol.com.br Adrenal and hypertension Madson Queiroz de Almeida Dyslipidemia and atherosclerosis Cynthia Melissa Valério Diabetes Mellitus Luiz Alberto Andreotti Turatti Basic endocrinology Maria Izabel Chiamolera Feminine and Andrology Rita de Cássia Viana Vasconcellos Weiss Pediatric Endocrinology Julienne Angela Ramires de Carvalho Bone and mineral metabolism Carolina Aguiar Moreira Neuroendocrinology Marcello D. Bronstein Obesity Maria Edna de Melo Thyroid Célia Regina Nogueira

TEMPORARY – SPORT AND EXERCISE ENDOCRINOLOGY - CTEEE President: Clayton Luiz Dornelles Macedo clayton.macedo@uol.com.br Members: Yuri Galeno Pinheiro Chaves de Freitas, Roberto Luís Zagury, Fulvio Clemo Santos Thomazelli, Felipe Henning Gaia Duarte, Ricardo de Andrade Oliveira, Andrea Messias Britto Fiorettiia

TITLE OF SPECIALIST IN ENDOCRINOLOGY AND METABOLISM President: Josivan Gomes de Lima josivanlima@gmail.com Vice-President: Márcio Corrêa Mancini Members: Marise Lazaretti Castro, Mauro Antônio Czepielewski, Milena Coelho Fernandes Caldato, Renan Magalhães Montenegro Júnior, Rogério Friedman

VALORIZATION OF NEW LEADERSHIPS President Wellington Santana da Silva Junior wellingtonsantanajr@gmail.com Members Mateus Dornelles Severo, Ricardo Mendes Martins

Brazilian Societies and Associations for Endocrinology and Metabolism

SBD – BRAZILIAN DIABETES SOCIETY SBD BRAZILIAN BOARD OF DIRECTORS (2018/2019)

President

Hermelinda Cordeiro Pedrosa

Vice-Presidents

Gustavo Caldas

Janice Sepulveda Reis

João Eduardo Nunes Salles

Rosane Kupfer

Rosângela Réa

1ST Secretary

Karla Melo

2ND Secretary

Fernanda Thomé

1ST Treasurer

Antonio Carlos Lerário

2ND Treasurer

Luiz Antônio Araújo

Supervisory Board

Silmara Leite

Regina Calsolari

Nely Calegaro

Estela Muskat Jatene

Rua Afonso Brás, 579, cj. 72/74 04511-011– São Paulo, SP Fone/Fax: (11) 3842-4931 www.diabetes.org.br

ABESO – BRAZILIAN ASSOCIATION FOR THE STUDY OF OBESITY AND METABOLIC SYNDROME ABESO BRAZILIAN BOARD OF DIRECTORS (2019-2020)

President

Mario Kehdi Carra

Vice-President

Rodrigo Lamounier

1ST Secretary General

Jacqueline Rizzoli

2ND Secretary General

Bibiana Colenci

Treasurer

Cintia Cercato

Rua Mato Grosso, 306, cj. 1711 01239-040 – São Paulo, SP Fone: (11) 3079-2298/Fax: (11) 3079-1732 info@abeso.org.br www.abeso.org.br

editorial

Is testosterone involved in low female sexual desire? Carmita H. N. Abdo1

Arch Endocrinol Metab. 2019;63/3

Psiquiatra, Professora Livre-Docente do Departamento de Psiquiatria da Faculdade de Medicina da Universidade de São Paulo (FMUSP). Fundadora e Coordenadora do Programa de Estudos em Sexualidade (ProSex) do Instituto de Psiquiatria do Hospital das Clínicas da FMUSP. Presidente da Associação Brasileira de Psiquiatria (ABP), São Paulo, SP, Brasil 1

Correspondence to: Carmita H. N. Abdo carmita.abdo@uol.com.br Received on June/19/2019 Accepted on June/19/2019 DOI: 10.20945/2359-3997000000153

e are living an era of growing longevity due to the new technological and medical advances. However, we are also going through a time of unsuitable habits, in which women are increasingly exposed to stress, little physical activity, few hours of sleep, unbalanced diet, result of these same resources and/or the imperative need to access them. Since the beginning of the 1900s, the higher life expectancy gave rise to a new population contingent: postmenopausal women (1). It should be remembered that menopause, besides being universal (occurs with all women in their 50s) is physiological and despite the increasing amount of women who do not produce estrogen anymore, the ‘’replacement” of this hormone is not a consensus among experts. In parallel with this lack of consensus, the use of testosterone to improve low female sexual desire has been another arena, including women of all ages. A recently published systematic review on benefits and harms of testosterone administrated to male sexual dysfunction concluded that testosterone therapy can be considered for men with low or low-normal testosterone levels and issues with sexual desire, erectile function and satisfaction derived from overall sexual life. The exact formulation, dosage and duration of treatment still needs to be clarified, the safety profile also remains unclear (2). A higher incidence of low sexual desire and low sexual satisfaction, as well as dyspareunia, characterizes the sex life of a significantly larger proportion of women, compared with those in the premenopausal period. Hypoactive sexual desire dysfunction (HSSD) includes some symptoms with a duration of at least 6 months and is associated with clinically significant personal distress (frustration, grief, guilt, incompetence, loss, sadness, sorrow or worry). These manifestations (symptoms) are: lack of motivation for sexual activity, either by decreased or absent spontaneous desire (no sexual thoughts or fantasies) or by decreased or absent responsive desire to erotic cues and stimulation, inability to maintain desire or interest through sexual activity, loss of desire to initiate or participate in a sexual activity, including behavioral responses, such as avoidance of situations that could lead to sexual activity, which is not secondary to sexual pain disorders (3). Specialized literature warns that testosterone has been frequently prescribed to women with low sexual desire, despite the lack of evidence of association between low testosterone levels and low libido and improvement in sexual dysfunctions when testosterone therapy is administered. While serum androgen levels decline steeply in the early reproductive years, they do not fluctuate as a direct consequence of natural menopause (4).

187

Is testosterone involved in low female sexual desire?

Testosterone therapy is not routinely recommended for women with low androgen levels caused by adrenal insufficiency, hypopituitarism or surgical menopause, nor to enhance cardiometabolic parameters, cognitive health or to improve general wellbeing in healthy women. Therefore, the use of androgens in postmenopausal women remains controversial (5-9). In a systematic review, nine of ten studies failed to find a correlation between total testosterone levels and sexual desire, and four small studies showed little or none improvement in libido when compared to placebo (10). Therefore, the current position statement of the Brazilian Society of Endocrinology and Metabolism (SBEM) was developed to review the existing literature on the off-label use of testosterone to treat low sexual desire in women. Nine members of the Female Endocrinology and Andrology Department from SBEM were asked about it. The main results of this review can be summarized as follows (11): • The relative hyperandrogenic status of menopause is associated with lower sex hormone binding globulin (SHBG) levels, resulting in increased free testosterone levels; • Surgical menopause induced by bilateral oophorectomy, however, is associated with a different pattern: significantly lower total and free testosterone levels and dehydroepiandrosterone sulfate (DHEAS) decreased as compared with age-matched controls with both ovaries; • In the presence of a proper HSDD diagnosis, once informed consent is obtained from the patient, an individualized trial of testosterone therapy for 3 to 6 months could be suggested, aiming midnormal premenopausal testosterone levels during treatment; • In the absence of clear improvement and if adverse events are observed at 6 months, the use of testosterone should be ceased. There are no safety and efficacy evidences for testosterone therapy in women with HSDD beyond 24 months; • Ideally, testosterone measurements should be obtained in the morning hours and in the follicular phase of the menstrual cycle. In normally cycling premenopausal women, testosterone levels should be measured at baseline and 3-6 weeks after starting treatment, especially to avoid supraphysiological concentrations, since response to therapy does not correlate with testosterone levels; • Considering side effects and long-term safety, the role of testosterone in breast cancer and 188

cardiovascular disease pathophysiology requires further elucidation; • The abuse of androgens in sports and in the community, for aesthetic purposes, remains a major concern. Female athletes reported increased aggressiveness. Dyslipidemia, hypertension, arrhythmia, coagulation disorders, fibrosis, and cardiac hypertrophy have also been observed; • There are currently no testosterone formulations approved for women by regulatory agencies in the United States, Brazil and most countries. Testosterone formulations approved for men are not recommended for women; • When considering testosterone therapy, all risks and benefits should be thoroughly discussed with the patient before prescription. This position statement was very well developed and leaves no doubt that female sexual dysfunction is a common complaint, specially in postmenopausal women, and may have a negative impact on quality of life. Testosterone seems to exert a positive effect on sexual desire in women with HSDD, however, requires caution and criteria in its management. In addition to the above explained, there is a warning about cases of sexual problems in which testosterone therapy definitely does not work; for instance, when libido is impaired by the partner’s lack of attraction, resentment, anger, fear, embarrassment, or even by myths, taboos, misconceptions about sexual activity. In these cases, non-pharmacological methods are the therapeutic alternatives. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Soules MR, Bremner WJ. The menopause and climacteric: endocrinologic basis and associated symptomatology. J Am Geriatr Soc. 1982 Sep;30(9):547-61. 2. Dimitropoulos K, Verze P, Van den Broeck T, Salonia A, Yuan CY, Hatzimouratidis K, et al. What are the benefits and harms of testosterone therapy for male sexual dysfunction? - a systematic review. Int J Impot Res. 2019. [Epub ahead of print] 3. Clayton AH, Goldstein I, Kim NN, Althof SE, Faubion SS, Faught BM, et al. The International Society for the Study of Women’s Sexual Health Process of Care for Management of Hypoactive Sexual Desire Disorder in Women. Mayo Clin Proc. 2018;93(4): 467-87. 4. Davison SL, Bell R, Donath S, Montalto JG, Davis SR. Androgen levels in adult females: changes with age, menopause, and oophorectomy. J Clin Endocrinol Metab. 2005;90(7):3847-53. Arch Endocrinol Metab. 2019;63/3

Is testosterone involved in low female sexual desire?

5. Abdo CH, Valadares AL, Oliveira WM Jr, Scanavino MT, Afif-Abdo J. Hypoactive sexual desire disorder in a population-based study of Brazilian women: associated factors classified according to their importance. Menopause. 2010;17(6):1114-21.

9. Traish A, Guay AT, Spark RF. Testosterone therapy in Women Study Group. Are the Endocrine Society’s Clinical Practice Guidelines on Androgen Therapy in Women misguided? A commentary. J Sex Med. 2007;4(5):1223-34.

6. Basson R, Brotto LA, Petkau AJ, Labrie F. Role of androgens in women’s sexual dysfunction. Menopause. 2010;17(5):962-71.

10. Reed BG, Nemer LB, Carr BR. Has testosterone passed the test in premenopausal women with low libido? A systematic review. Int J Womens Health. 2016;8:599-607.

7. Davis SR, Moreau M, Kroll R, Bouchard C, Panay N, Gass M, et al. APHRODITE Study Team. Testosterone for low libido in postmenopausal women not taking estrogen. N Engl J Med. 2008;359(19):2005-17.

8. Santen RJ, Allred DC, Ardoin SP, Archer DF, Boyd N, Braunstein GD et al.; Endocrine Society. Postmenopausal Hormone Therapy: An Endocrine Society Scientific Statement. J Clin Endocrinol Metab. 2010;95(7 Suppl 1):s1-66.

11. Weiss RV, Hohl A, Athayde A, Pardini D, Gomes L, de Oliveira M, et al. Testosterone therapy for women with low sexual desire: a position statement from the Brazilian Society of Endocrinology and Metabolism. Arch Endocrinol Metab. 2019;63(3):190-8.

Arch Endocrinol Metab. 2019;63/3

189

consensus

Testosterone therapy for women with low sexual desire: a position statement from the Brazilian Society of Endocrinology and Metabolism Instituto Estadual de Diabetes e Endocrinologia Luiz Capriglione, Pontifícia Universidade Católica do Rio de Janeiro, Escola Médica de Pós-Graduação em Endocrinologia (IEDE-PUC/RJ), Rio de Janeiro, RJ, Brasil 2 Serviço de Endocrinologia e Metabolismo, Departamento de Medicina Interna, Hospital Universitário (HU), Universidade Federal de Santa Catarina (UFSC), Florianópolis, SC, Brasil 3 Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brasil 4 Disciplina de Endocrinologia, Universidade Federal de São Paulo (UNIFESP), São Paulo, SP, Brasil 5 Disciplina de Endocrinologia, Faculdade de Medicina, Hospital de Clínicas, Universidade de São Paulo (USP), São Paulo, SP, Brasil 6 Instituto de Medicina Integral Professor Fernando Figueira, Recife, PE, Brasil 7 Departamento de Fisiologia Endócrina e Fisiologia e Laboratório de Pesquisas Clínicas e Experimentais em Biologia Vascular (BIOVASC), Rio de Janeiro, RJ, Brasil 8 Unidade de Endocrinologia Ginecológica, Serviço de Endocrinologia, Hospital de Clínicas de Porto Alegre e Departamento de Fisiologia, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brasil 1

Rita V. Weiss¹

http://orcid.org/0000-0002-6294-6825

Alexandre Hohl²

http://orcid.org/0000-0002-8073-5837

Amanda Athayde1,3

http://orcid.org/0000-0003-1865-4785

Dolores Pardini4

http://orcid.org/0000-0002-4091-1621

Larissa Gomes5

http://orcid.org/0000-0002-0902-0142

Monica de Oliveira6

http://orcid.org/0000-0001-9140-2555

Ricardo Meirelles1

http://orcid.org/0000-0002-1981-9855

Ruth Clapauch7

http://orcid.org/0000-0002-3153-7199

Poli Mara Spritzer8

http://orcid.org/0000-0002-67347688

ABSTRACT Objective: To summarize current evidence regarding testosterone treatment for women with low sexual desire. Materials and methods: The Female Endocrinology and Andrology Department of the Brazilian Society of Endocrinology and Metabolism invited nine experts to review the physiology of testosterone secretion and the use, misuse, and side effects of exogenous testosterone therapy in women, based on the available literature and guidelines and statements from international societies. Results: Low sexual desire is a common complaint in clinical practice, especially in postmenopausal women, and may negatively interfere with quality of life. Testosterone seems to exert a positive effect on sexual desire in women with sexual dysfunction, despite a small magnitude of effect, a lack of long-term safety data, and insufficient evidence to make a broad recommendation for testosterone therapy. Furthermore, there are currently no testosterone formulations approved for women by the relevant regulatory agencies in the United States, Brazil, and most other countries, and testosterone formulations approved for men are not recommended for use by women. Conclusion: Therefore, testosterone therapy might be considered if other strategies fail, but the risks and benefits must be discussed with the patient before prescription. Arch Endocrinol Metab. 2019;63(3):190-8 Keywords Testosterone therapy; female sexual disorder; position

Correspondence to: Poli Mara Spritzer Serviço de Endocrinologia, Hospital de Clínicas de Porto Alegre Rua Ramiro Barcelos, 2350 Porto Alegre, RS, Brasil 90035-003 spritzer@ufrgs.br Received on Jan/2/2019 Accepted on Apr/22/2019

DOI: 10.20945/2359-3997000000152

190

Arch Endocrinol Metab. 2019;63/3

Testosterone therapy in women

exual function in women is a complex issue, impacted upon by several factors. Female testosterone levels decrease slightly during the reproductive years, but do not change substantially after menopause, because the ovaries continue to produce androgens after depletion of follicular capital (1,2). Reduced androgen levels have been frequently associated with reduced libido in women, mainly in the peri- and postmenopausal periods (1-4). While testosterone has been frequently prescribed to women with low sexual desire, there is no evidence of a consistent association between low testosterone levels and low libido, nor of consistent improvement in sexual problems when testosterone therapy is administered. Therefore, testosterone therapy is not routinely recommended for women with low androgen levels caused by adrenal insufficiency, hypopituitarism, or surgical menopause, nor to enhance cardio-metabolic parameters, cognitive health, or general well-being in healthy women. In fact, the use of androgens in postmenopausal women remains controversial, with strong positions being taken by a variety of sexual medicine experts and professional societies (5-9). In 1997 and 1998, two meetings were held in Cape Cod, Massachusetts, to discuss the clinical detection and management of female sexual dysfunction. These two meetings, and their subsequent iterations, would lead to the formation of the International Society for the Study of Womenâ&#x20AC;&#x2122;s Sexual Health (ISSWSH) in 2001 (10). Later in 2002, the Princeton Consensus proposed the existence of a female androgen insufficiency syndrome, defined as a pattern of clinical symptoms in the presence of decreased bioavailable testosterone and normal estrogen status (11). The diagnostic criteria were criticized because the syndrome definition implied that sexual dysfunction, the predominant clinical finding, was a consequence of androgen insufficiency (12). In 2005, a study revealed that, while serum androgen levels decline steeply in the early reproductive years, they do not vary as a direct consequence of natural menopause. The investigators pointed out that the significant age-related variations in androgens must be taken into account when normal ranges are reported and in studies on the role of androgens in women (1). Recently, a systematic review assessed the available evidence regarding the evaluation of serum testosterone levels and testosterone treatment for premenopausal women with low libido from 1995 to 2015. Nine out of ten studies failed to find a correlation between total Arch Endocrinol Metab. 2019;63/3

testosterone levels and sexual desire, and four small studies showed little, if any, improvement in libido when compared to placebo (13). Therefore, the present position statement was developed to review the existing literature on the off-label use of testosterone to treat low sexual desire in women.

MATERIALS AND METHODS Nine members of the Female Endocrinology and Andrology Department of the Brazilian Society of Endocrinology and Metabolism, experts in clinical practice, clinical research and/or translational research in the field, were asked to critically review the topic of testosterone use and develop a position statement on testosterone therapy for women with low sexual desire. MEDLINE and SciELO/LILACS databases were searched for relevant studies and clinical practice guidelines regarding testosterone therapy in women. The level of evidence of each publication was graded using the Oxford Centre for Evidence-Based Medicine recommendations (14). In the present position statement, these levels of evidence are reported as: A: experimental or observational studies with consistent results; B: experimental or observational studies with less consistent results; C: case reports (uncontrolled studies); D: opinion lacking critical evaluation or based on guidelines, physiological studies, or animal models (15).

Testosterone levels in women across the lifespan During childhood, testosterone levels are very low. At puberty, kisspeptin triggers pulsatile gonadotropin releasing hormone (GnRH) secretion in the medial basal hypothalamus and preoptic area (16), enhancing luteinizing hormone (LH) secretion and consequently stimulating estradiol and testosterone secretion. Upon puberty and during the reproductive years, mainly in young adult women, testosterone is produced equally by the ovarian theca cells and the adrenal cortex, accounting for approximately 300 mcg daily (17,18). Only one-third of circulating testosterone results from direct ovarian and adrenal secretion; the remaining two-thirds arise from peripheral conversion of precursors, including delta 4-androstenedione (A4) and DHEA, in non-steroid-producing tissues. Testosterone is converted to estradiol by aromatase and to dihydrotestosterone (DHT) by 5-alpha reductase, in target tissues as well as in the periphery (mainly in adipose tissue) (19). 191

INTRODUCTION

Testosterone therapy in women

As in men, testosterone production in women follows a circadian rhythm, with higher levels in the morning and a nadir around midnight (20). Moreover, a modest mid-cycle testosterone peak, coincident with the LH surge in ovulatory cycles, has been confirmed by recent high-sensitivity assays (21). Apart from this mid-cycle peak, testosterone levels in the follicular and luteal phases seem to be stable (1,21,22). During the late reproductive years, the increment of anovulatory cycles leads to a failure of this mid-cycle rise, despite preservation of baseline physiological ovarian testosterone secretion at other phases of the cycle (3). Aging promotes a progressive reduction in testosterone production by the ovaries and adrenal glands (Table 1). At that stage, the peripheral conversion of A4 and DHEA into testosterone becomes more important (23-25). DHEA serves as an intermediate steroid within the testosterone and estradiol biosynthetic pathways. Peripheral tissues involved with DHEA/ DHEA-S uptake from the circulation typically generate more potent androgens and estrogens that act locally (26,27). During the menopause transition, the decline in circulating estrogen is more pronounced than the decline in androgen levels, and the ensuing gradual increase in the testosterone to estradiol ratio (T/E) leads to a relative hyperandrogenic state (28). This relative androgen excess status has been associated with central abdominal adiposity and the metabolic syndrome (28-30). Also, serum levels of sex-hormone binding protein (SHBG) contribute to testosterone transport and availability, limiting the amount of free and bioactive testosterone (1). Thus, the relative hyperandrogenic status of menopause is associated with lower SHBG levels, resulting in increased free testosterone levels (24,25). Surgical menopause induced by bilateral oophorectomy, however, is associated with a different pattern. In an epidemiological cross-sectional survey,

Australian women aged 55 years or older who had undergone bilateral oophorectomy showed significantly lower total and free testosterone levels than agematched controls with both ovaries. The same study observed a very gradual decrease in the ovarian and adrenal function of 1,423 Australian women starting in the early reproductive years: from 18 to 75 years of age, total testosterone decreased 55%, free testosterone decreased 49%, and DHEA-S decreased 77% (1).

Testosterone therapy for sexual desire dysfunction The diagnosis of sexual desire dysfunction has historically been guided by the criteria set out in the Diagnostic and Statistical Manual of Mental Disorders (DSM). From 1980 to 2013, the diagnosis of hypoactive sexual desire disorder (HSDD) depended on the symptom of absent or reduced desire for sexual activity and lack of sexual fantasies, with associated clinically significant distress. Recently, the DSM-5 introduced broader criteria for the diagnosis of sexual desire dysfunction, and merged desire and arousal into a single category, assigning this condition the diagnostic entity “female sexual interest/arousal disorder” (SIAD) (31). A recent study has reported that around 73% of women with a diagnosis of HSDD also met criteria for SIAD (32). However, it is important to point out that the DSM-5 is a psychiatric compendium, and is not intended for clinical use by the disciplines that care for sexual medicine patients in clinical settings. In this sense, both the International Consultation on Sexual Medicine publications and the International Society for the Study of Women’s Sexual Health (ISSWSH) Nomenclature indicate that current evidence does not support the merger of desire and arousal into a single category, supporting the maintenance of an HSDD diagnostic category (33-35). A recent publication (36) indicates that ICD-11 intends to preserve HSDD as a separate category for coding (Table 2).

Table 1. Testosterone levels across the lifespan in a reference group of Australian women Age (years)

18-24

25-34

35-44

45-54

55-64

65-75

Total testosterone (25-83 ng/dL) (direct manual RIA method)

45.53

31.98

26.9

23.34

19.0

20.46

Free testosterone, calculated (370-1,110 pg/dL) Sodergard equation (based on total testosterone, SHBG, and albumin levels)

680.4

497.1

393.9

340.6

311.5

281.2

Modified from Davison and cols. (1).

192

Arch Endocrinol Metab. 2019;63/3

Testosterone therapy in women

Table 2. Proposed classification of female sexual dysfunction in ICD-11 and current ICD-10, and DSM-5 definitions Proposed ICD-11

ICD-10

DSM-5

Hypoactive sexual desire and dysfunction

Absence or loss of sexual desire

Female sexual interest/arousal disorder

Category: Female sexual arousal dysfunction obs: the psychological component of arousal that corresponds to the ICD-10 Lack of sexual enjoyment is also included in this proposed classification

Failure of genital response; lack of sexual enjoyment

Female sexual interest/arousal disorder

A diagnosis of HSDD must include any of the following symptoms, with a duration of at least 6 months: “1) lack of motivation for sexual activity, either by decreased or absent spontaneous desire (sexual thoughts or fantasies) or by decreased or absent responsive desire to erotic cues and stimulation or inability to maintain desire or interest through sexual activity; 2) loss of desire to initiate or participate in sexual activity, including behavioral responses such as avoidance of situations that could lead to sexual activity, that is not secondary to sexual pain disorders” (37, p. 468). These manifestations are “combined with clinically significant personal distress that includes frustration, grief, guilt, incompetence, loss, sadness, sorrow, or worry” (37, p. 468). A sexual history and/or the use of a validated instrument, such as the Decreased Sexual Desire Screener (DSDS) (38), are recommended for confirming the diagnosis and determining the type of HSDD (33,37). In women going through the natural menopausal transition, a comprehensive, longitudinal study has found that prior function and relationship factors have a stronger impact on female sexual function than hormone concentrations (A) (39). Clinical management of HSDD encompasses sex therapy, central nervous system drugs, and hormonal agents, and should always consider menopausal status. Concerning hormone therapy, testosterone is currently only approved for that purpose in Australia; it remains an off-label treatment in other countries. Transdermal testosterone has been widely studied to treat HSDD. A systematic review of seven randomized trials, enrolling more than 3,000 women, evaluated the efficacy of a 300 mcg testosterone patch and found that the testosterone group had more satisfying sexual episodes, sexual activity, orgasm, desire, and improvement in Personal Distress Scale scores, but also increased acne and hair growth, suggesting a supraphysiologic dose (40). According to the authors, short-term transdermal testosterone was efficient in improving sexual function in naturally and surgically menopausal women affected by HSDD, both Arch Endocrinol Metab. 2019;63/3

on or off estrogen-progestin hormone therapy (38). This evidence has been incorporated into the clinical practice guidelines issued by at least four Societies (A) (3,37,41,42). Therefore, in the presence of a proper HSDD diagnosis, once informed consent is obtained from the patient, an individualized trial of testosterone therapy for 3 to 6 months could be suggested, aiming at mid-normal premenopausal testosterone levels during treatment. The existing guidelines state that in the absence of clear improvement and if adverse events are observed at 6 months, the use of testosterone should be ceased (D) (3,37,41). There are no safety and efficacy data for testosterone therapy in women with HSDD beyond 24 months (3,37,41,42), although one single 4-year openlabel extension trial in surgically menopausal women with HSDD showed no relevant safety issues during this period (43). While non-oral preparations (such as a transdermal patch, gel, or cream) are recommended for therapy in women, such preparations are not available in most countries, including Brazil. It is important to note that testosterone preparations formulated for men are not recommended for use by women, due to lack of data concerning efficacy and safety of these preparations in women (3,37,42). In women who are receiving a trial of testosterone therapy, testosterone levels should be measured at baseline and 3-6 weeks after starting treatment, especially to avoid supraphysiologic concentrations, since response to therapy does not correlate with testosterone levels (B) (3,34,37). During ongoing testosterone therapy, signs of clinical androgen excess and laboratory testosterone levels should be assessed every 6 months to monitor potential side effects. Current evidence from women with adrenal insufficiency (44) or normal adrenal function (45) does not support DHEA use for HSDD treatment (A). Besides, two drug combinations, based on known neurobiological mechanisms that are critically important in sexual excitation and inhibition (46,47), 193

Adapted from Reed et al. (36).

Testosterone therapy in women

are in development for low sexual desire in women. One consists of a tablet that combines a testosterone coating for sublingual administration and an innercore component containing the phosphodiesterase type 5 inhibitor sildenafil. Sublingual testosterone could potentially increase the brain’s responses to sexual cues, leading to increases in sexual motivation, while sildenafil could increase genital vasocongestion (46,47). The other drug is a tablet which combines an innercore component containing the 5-hydroxytryptamine (5-HT)1A receptor agonist buspirone coated with a delayed-release matrix impregnated with testosterone, which could ensure that the pharmacologic effects of both components coincide (46,47). However, these investigational drugs have not advanced yet to Phase III clinical trials and further progress are needed before considering as a treatment option.

Laboratory evaluation of testosterone levels Reliable testosterone assays are essential in the diagnosis and management of a number of clinical conditions in females, including precocious pubarche, androgensecreting tumors, and the polycystic ovary syndrome (PCOS) (48). However, most testosterone assays have poor sensitivity and accuracy in the female ranges (normal range, 20-60 ng/dL). Circulating testosterone levels in women vary according to reproductive status, phase of the menstrual cycle, and time of day (49). There are no universally normal ranges applicable to all ages and sexes, and, in fact, there is no testosterone standard on which to base an assay. Only 1-3% of testosterone is not bound to plasma proteins, raising questions about which measure of testosterone is the most clinically useful – total (TT) or free testosterone (FT) (50). The commonly used methods for measuring TT or FT are immunoassays and mass spectrometry (MS). Automated chemiluminescence is the most common method overall, although it has lower accuracy. In immunoassays, extraction of testosterone and purification by chromatography before the assay decreases crossreactivity with other steroids and improves the accuracy of the method (49); however, these procedures are unwieldy for commercial use (48). Despite significant advances to improve the accuracy and precision of currently available assays, limited comparability exists between assays at the lower and upper extremes of the testosterone range. Because of this lack of comparability, there is no current standard immunoassay for the assessment of total testosterone levels (49). 194

In turn, extraction and high-performance chromatography followed by MS is currently considered the gold standard for measuring total testosterone (3,49,51,52). However, obtaining and maintaining a mass spectrometer is expensive, and sample throughput is generally slower and more costly than with automated immunoassays. Moreover, standardization is still needed to ensure good comparability of results across laboratories. While TT measurement may not be ideal in the context of evaluating androgen production and availability, FT determination also poses a concern due to its low accuracy for values in the female range. Quantitation of FT levels by dialysis is perhaps the most accurate test for measuring androgens in women, but also the most costly, and is not performed by most laboratories. Ideally, testosterone measurements should be obtained in the morning hours and in the follicular phase of the menstrual cycle in normally cycling premenopausal women.

Testosterone therapy: side effects and long-term safety The main side effects of testosterone therapy in women are related to masculinizing symptoms and abnormal endometrial bleeding, but breast cancer and cardiovascular risk are also a reason for concern (37,53). The effects are dose-dependent; supraphysiologic doses must always be avoided. The most common clinical side effects are acne, hirsutism, androgenic alopecia, and deepening of the voice. Clitoromegaly has not emerged as a frequent side effect at physiological doses (41,42,53). In the APHRODITE trial, a randomized study that used testosterone therapy at doses ranging from 150-300 µg/d, dose was not associated with increased rates of acne, voice change or alopecia over 12 months. However, a higher rate of terminal hair growth was observed at the 300 µg/d dose (54). Androgen excess is usually associated with endometrial atrophy when given without concomitant estrogen. In the APHRODITE trial, women with endometrial bleeding who were exposed to the high study testosterone dose (300 µg/d) were found to have endometrial atrophy on endometrium biopsy (55). However, as most androgens can be aromatized to estrogens by aromatase, endometrial abnormalities associated with high estrogen concentrations, such as endometrial hyperplasia, dysplasia, and cancer, could also be expected during testosterone therapy (49,55). Arch Endocrinol Metab. 2019;63/3

The effects of testosterone therapy on the breast are conflicting (3,56,57). Testosterone seems to decrease breast tissue proliferation and prevent the pro-apoptotic effect mediated by estrogen (56). Patients on estrogen therapy did not show mammographic density changes, and no breast cell proliferation was observed when testosterone was added (58). Data from the Womenâ&#x20AC;&#x2122;s Health Initiative (WHI), an observational study in which a combination of estrogen and testosterone therapy was used, show a non-significant impact on invasive breast cancer risk (adjusted HR 1.42; 95%CI, 0.95-2.11) (59). Regarding cardiovascular and metabolic outcomes, the relationship between endogenous postmenopausal hormone levels and atherosclerosis is unclear (60-62). Oral testosterone therapy seems to negatively impact lipid profile, usually decreasing HDL and increasing LDL. However, parenteral and transdermal androgens do not seem to have this effect (60). A systematic review did not find adverse effects of testosterone therapy on blood pressure, vascular reactivity, blood viscosity, insulin sensitivity, hemoglobin concentration, or coagulation factors in postmenopausal women (63). One study showed an inverse correlation between circulating androgen levels and arterial intimamedia thickness assessed by carotid ultrasonography in postmenopausal women. This result provides preliminary evidence suggesting a protective effect of androgens against the development of atherosclerosis (64). However, these data were obtained from women with a lower risk of carotid artery atherosclerosis and considering physiologic testosterone levels, and cannot be extrapolated to exogenous testosterone therapy (65). Therefore, the role of testosterone in breast cancer and cardiovascular disease pathophysiology requires further elucidation, and evidence of safety from longterm studies is lacking.

Testosterone misuse and abuse The first studies to investigate the ability of supraphysiologic doses of testosterone to increase libido in postmenopausal women were conducted by Salmon and Geist, in 1943 (66). These studies showed increased libido with testosterone therapy in this population, leading to an increased interest in hormonal treatment with testosterone for lower libido in women. Arch Endocrinol Metab. 2019;63/3

Testosterone therapy could potentially be prescribed to women with HSDD, but no testosterone formulation has been approved for females by the regulatory agencies of the United States, Brazil, or most other countries. Consequently, testosterone formulations approved for men have been (mis)used off-label. Testosterone derivatives have distinct levels of relative anabolic and androgenic activities, exerting their ergogenic and cosmetic effects (67). Exogenous androgens can increase muscle mass and strength in athletes, regardless of gender or age. Therefore, since the early 1970s, the International Olympic Committee has banned the use of anabolic-androgenic steroids (AAS) by male and female athletes (68). However, androgens continue to be used abusively as ergogenic drugs in sports. In 2006, 34 international laboratories accredited by the International Anti-Doping Agency recorded 4,332 positive findings in tests, of which 45% were due to use of androgens (67). Furthermore, the aesthetic misuse of testosterone in non-athletes has been growing dramatically and uncontrolledly, and a clear increase in the use of these steroids by women has been observed. However, in the current medical literature, the vast majority of research data on the use of anabolic steroids in women comes from athletes (69). Randomized studies exploring the effects of AAS in women are scarce; case reports remain the leading source of information in this regard. The type and intensity of adverse effects depend on the AAS used, its dose, and the duration of use. Also, oral formulations can be hepatotoxic. AAS administration in women can induce masculinization, with development of acne vulgaris, change in libido, and deepening voice as the adverse effects most often seen in the first weeks of anabolic steroid use. Prolonged AAS administration can induce hair loss (alopecia), changes in pubic hair growth, smaller breasts, and clitoromegaly (A). AAS use can suppress pituitary LH and FSH secretion, leading to menstrual irregularities and even infertility. Also, in some studies, female athletes reported increased aggressiveness with the use of steroids (70). Dyslipidemia, hypertension, arrhythmia, coagulation disorders, fibrosis, and cardiac hypertrophy have been observed (71,72). The abuse of androgens in sports and in the community, for aesthetic purposes, remains a major concern. AAS abuse in the general population requires well-designed psychological prevention programs 195

Testosterone therapy in women

aiming to deter teenagers from future abuse of androgens. However, to date, such programs have proven effective only to increase knowledge, but not to reduce the intention to abuse or abusive behavior. In general, prevention of AAS abuse requires both demand-side and supply-side reductions, involving not only specific education for women of all ages, with attention to psychosocial predisposing factors, but also enforcement of laws to curb illegal networks that enable and maintain AAS abuse. In conclusion, female sexual dysfunction is a common complaint in clinical practice, especially in postmenopausal women, and may have a negative impact on quality of life. Testosterone seems to exert a positive effect on sexual desire in women with HSDD, despite a small magnitude of effect, a lack of long-term safety data, and insufficient evidence for a broad recommendation of testosterone therapy, as also acknowledged in previous clinical guidelines (3,37,42). In addition, there are currently no testosterone formulations approved for women by regulatory agencies in the United States, Brazil, and most countries. Finally, testosterone formulations approved for men are not recommended for use by women. Therefore, when considering testosterone therapy, all risks and benefits should be thoroughly discussed with the patient before prescription. Disclosure: the authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

REFERENCES 1. Davison SL, Bell R, Donath S, Montalto JG, Davis SR. Androgen levels in adult females: changes with age, menopause, and oophorectomy. J Clin Endocrinol Metab. 2005;90(7):3847-53. 2. Davis SR, Wahlin-Jacobsen S. Testosterone in women – the clinical significance. Lancet Diabetes Endocrinol. 2015;3(12):980-92. 3. Wierman ME, Arlt W, Basson R, Davis SR, Miller KK, Murad MH, et al. Androgen therapy in women: a reappraisal: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2014;99(10):3489-510.

4. Wierman ME, Nappi RE, Avis N, Davis SR, Labrie F, Rosner W, et al. Endocrine aspects of women’s sexual function. J Sex Med. 2010;7(1 Pt 2):561-85. 5. Abdo CH, Valadares AL, Oliveira WM Jr, Scanavino MT, Afif-Abdo J. Hypoactive sexual desire disorder in a population-based study of Brazilian women: associated factors classified according to their importance. Menopause. 2010;17(6):1114-21. 6. Basson R, Brotto LA, Petkau AJ, Labrie F. Role of androgens in women’s sexual dysfunction. Menopause. 2010;17(5):962-71. 7. Davis SR, Moreau M, Kroll R, Bouchard C, Panay N, Gass M, et al.; APHRODITE Study Team. Testosterone for low libido in

196

postmenopausal women not taking estrogen. N Engl J Med. 2008;359(19):2005-17. 8. Santen RJ, Allred DC, Ardoin SP, Archer DF, Boyd N, Braunstein GD, et al.; Endocrine Society. Postmenopausal hormone therapy: an Endocrine Society scientific statement. J Clin Endocrinol Metab. 2010;95(7 Suppl 1):s1-66. 9. Traish A, Guay AT, Spark RF. Testosterone therapy in Women Study Group. Are the Endocrine Society’s Clinical Practice Guidelines on Androgen Therapy in Women misguided? A commentary. J Sex Med. 2007;4(5):1223-34. 10. Perelman MA. The history of sexual medicine. In: Lipshultz LI, Pastuszak AW, Goldstein AT, Giraldi A, Perelman MA, editors. Management of sexual dysfunction in men and women. New York: Springer-Verlag; 2016. p. 137-79. 11. Bachmann G, Bancroft J, Braunstein G, Burger H, Davis S, Dennerstein L, et al.; Princeton. Female androgen insufficiency: the Princeton consensus statement on definition, classification, and assessment. Fertil Steril. 2002;77(4):660-5. 12. Arlt W. Androgen therapy in women. Eur J Endocrinol. 2006;154(1):1-11. 13. Reed BG, Nemer LB, Carr BR. Has testosterone passed the test in premenopausal women with low libido? A systematic review. Int J Womens Health. 2016;8:599-607. 14. Levels of evidence and grades of recommendation. Oxford Centre for Evidence-Based Medicine. Available from: http://www.cebm. net/oxford-centre-evidence-based-medicine-levelsevidencemarch-2009/. 15. Maeda SS, Moreira CA, Borba VZC, Bandeira F, Farias MLF, Borges JLC, et al. Diagnosis and treatment of hypoparathyroidism: a position statement from the Brazilian Society of Endocrinology and Metabolism. Arch Endocrinol Metab. 2018;62(1):106-24. 16. Murphy KG. Kisspeptins: regulators of metastasis and the hypothalamic-pituitary-gonadal axis. J Neuroendocrinol. 2005;17(8):519-25. 17. Burger HG. Androgen production in women. Fertil Steril. 2002;77 Suppl 4:S3-5. 18. Abraham GE. Ovarian and adrenal contribution to peripheral androgens during the menstrual cycle. J Clin Endocrinol Metab. 1974;39(2):340-6. 19. Longcope C. Adrenal and gonadal androgen secretion in normal females. Clin Endocrinol Metab. 1986;15(2):213-28. 20. Ankarberg C, Norjavaara E. Diurnal rhythm of testosterone secretion before and throughout puberty in healthy girls: correlation with 17beta-estradiol and dehydroepiandrosterone sulfate. J Clin Endocrinol Metab. 1999;84(3):975-84. 21. Bui HN, Sluss PM, Blincko S, Knol DL, Blankenstein MA, Heijboer AC. Dynamics of serum testosterone during the menstrual cycle evaluated by daily measurements with an ID-LC-MS/MS method and a 2nd generation automated immunoassay. Steroids. 2013;78(1):96-101. 22. Nóbrega LH, Azevedo GD, Lima JG, Ferriani RA, Spritzer PM, Sá MF, et al. Analysis of testosterone pulsatility in women with ovulatory menstrual cycles. Arq Bras Endocrinol Metabol. 2009;53(8):1040-6. 23. Montalcini T, Migliaccio V, Ferro Y, Gazzaruso C, Pujia A. Androgens for postmenopausal women’s health? Endocrine. 2012;42(3):514-20. 24. Kim C, Harlow SD, Zheng H, McConnell DS, Randolph JF Jr. Changes in androstenedione, dehydroepiandrosterone, testosterone, estradiol, and estrone over the menopausal transition. Womens Midlife Health. 2017;3. pii: 9. 25. Burger HG, Dudley EC, Cui J, Dennerstein L, Hopper JL. A prospective longitudinal study of serum testosterone, dehydroepiandrosterone sulfate, and sex hormone-binding globulin levels through the menopause transition. J Clin Endocrinol Metab. 2000;85(8):2832-8. Arch Endocrinol Metab. 2019;63/3

Testosterone therapy in women

27. Labrie F, Bélanger A, Pelletier G, Martel C, Archer DF, Utian WH. Science of intracrinology in postmenopausal women. Menopause. 2017;24(6):702-12. 28. Torréns JI, Sutton-Tyrrell K, Zhao X, Matthews K, Brockwell S, Sowers M, et al. Relative androgen excess during the menopausal transition predicts incident metabolic syndrome in midlife women: study of Women’s Health Across the Nation. Menopause. 2009;16(2):257-64. 29. McCarthy AM, Menke A, Visvanathan K. Association of bilateral oophorectomy and body fatness in a representative sample of US women. Gynecol Oncol. 2013;129(3):559-64. 30. Maturana MA, Moreira RM, Spritzer PM. Lipid accumulation product (LAP) is related to androgenicity and cardiovascular risk factors in postmenopausal women. Maturitas. 2011;70(4):395-9. 31. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders. Fifth edition. Arlington: American Psychiatric Association; 2013.

43. Nachtigall L, Casson P, Lucas J, Schofield V, Melson C, Simon JA. Safety and tolerability of testosterone patch therapy for up to 4 years in surgically menopausal women receiving oral or transdermal oestrogen. Gynecol Endocrinol. 2011;27(1):39-48. 44. Alkatib AA, Cosma M, Elamin MB, Erickson D, Swiglo BA, Erwin PJ, et al. A systematic review and meta-analysis of randomized placebo-controlled trials of DHEA treatment effects on quality of life in women with adrenal insufficiency. J Clin Endocrinol Metab. 2009;94(10):3676-81. 45. Elraiyah T, Sonbol MB, Wang Z, Khairalseed T, Asi N, Undavalli C, et al. Clinical review: The benefits and harms of systemic dehydroepiandrosterone (DHEA) in postmenopausal women with normal adrenal function: a systematic review and metaanalysis. J Clin Endocrinol Metab. 2014;99(10):3536-42. 46. Both S. Recent developments in psychopharmaceutical approaches to treating female sexual interest and arousal disorder. Curr Sex Health Rep. 2017;9(4):192-9.

32. O’Loughlin JI, Basson R, Brotto LA. Women with Hypoactive Sexual Desire disorder versus Sexual Interest/Arousal Disorder: an empirical test of raising the bar. J Sex Res. 2018;55(6):734-46.

47. Tuiten A, van Rooij K, Bloemers J, Eisenegger C, van Honk J, Kessels R, et al. Efficacy and safety of on-demand use of 2 treatments designed for different etiologies of female sexual interest/arousal disorder: 3 randomized clinical trials. J Sex Med. 2018;15(2):201-16.

33. Parish SJ, Goldstein AT, Goldstein SW, Goldstein I, Pfaus J, Clayton AH, et al. Toward a more evidence-based nosology and nomenclature for female sexual dysfunctions-Part II. J Sex Med. 2016;13(12):1888-906.

48. Legro RS, Schlaff WD, Diamond MP, Coutifaris C, Casson PR, Brzyski RG, et al. Total testosterone assays in women with polycystic ovary syndrome: precision and correlation with hirsutism. J Clin Endocrinol Metab. 2010;95(12):5305-13.

34. Davis SR, Worsley R, Miller KK, Parish SJ, Santoro N. Androgens and female sexual function and dysfunction - findings from the Fourth International Consultation of Sexual Medicine. J Sex Med. 2016;13(2):168-78.

49. Rosner W, Auchus RJ, Azziz R, Sluss PM, Raff H. Position statement: Utility, limitations, and pitfalls in measuring testosterone: an Endocrine Society position statement. J Clin Endocrinol Metab. 2007;92(2):405-13.

35. McCabe MP, Sharlip ID, Atalla E, Balon R, Fisher AD, Laumann E, et al. Definitions of sexual dysfunctions in women and men: a Consensus Statement From the Fourth International Consultation on Sexual Medicine 2015. J Sex Med. 2016:13(2);135-43.

50. Stanczyk FZ, Lee JS, Santen RJ. Standardization of steroid hormone assays: why, how, and when? Cancer Epidemiol Biomarkers Prev. 2007;16(9):1713-9.

36. Reed GM, Drescher J, Krueger RB, Atalla E, Cochran SD, First MB, et al. Disorders related to sexuality and gender identity in the ICD-11: Revising the ICD-10 classification based on current scientific evidence, best clinical practices, and human rights considerations. World Psychiatry. 2016;15(3):205-21. 37. Clayton AH, Goldstein I, Kim NN, Althof SE, Faubion SS, Faught BM, et al.The International Society for the Study of Women’s Sexual Health Process of Care for Management of Hypoactive Sexual Desire Disorder in Women. Mayo Clin Proc. 2018;93(4):467-87. 38. Clayton AH, Goldfischer ER, Goldstein I, Derogatis L, LewisD’Agostino DJ, Pyke R. Validation of the Decreased Sexual Desire Screener (DSDS): a brief diagnostic instrument for generalized acquired female hypoactive sexual desire disorder (HSDD). J Sex Med. 2009;6(3):730-8. 39. Dennerstein L, Lehert P, Burger HG. The relative effects of hormonal and relationship factors on sexual function of women through the natural menopausal transition. Fertil Steril. 2005;84(1):174-80. 40. Achilli C, Pundir J, Ramanathan P, Sabatini L, Hamoda H, Panay N. Efficacy and safety of transdermal testosterone in postmenopausal women with hypoactive sexual desire disorder: a systematic review and meta-analysis. Fertil Steril. 2017;107(2):475-82.e15. 41. Simon JA, Davis SR, Althof SE, Chedraui P, Clayton AH, Kingsberg SA, et al. Sexual well-being after menopause: An International Menopause Society White Paper. Climacteric. 2018;21(5):415-27. 42. Shifren JL, Gass ML; NAMS Recommendations for Clinical Care of Midlife Women Working Group. The North American Menopause Society recommendations for clinical care of midlife women. Menopause. 2014;21(10):1038-62.

Arch Endocrinol Metab. 2019;63/3

51. Zendjabil M, Chellouai Z, Abbou O. Role of mass spectrometry in steroid assays. Ann Endocrinol (Paris). 2016;77(1):43-8. 52. Thienpont LM, Van Uytfanghe K, Blincko S, Ramsay CS, Xie H, Doss RC, et al. State-of-the-art of serum testosterone measurement by isotope dilution-liquid chromatography-tandem mass spectrometry. Clin Chem. 2008;54(8):1290-7. 53. Shifren JL, Davis SR, Moreau M, Waldbaum A, Bouchard C, DeRogatis L, et al. Testosterone patch for the treatment of hypoactive sexual desire disorder in naturally menopausal women: results from the INTIMATE NM1 Study. Menopause. 2006;13(5):770-9. 54. Davis SR, Moreau M, Kroll R, Bouchard C, Panay N, Gass M, et al. Testosterone for low libido in postmenopausal women not taking estrogen. N Engl J Med. 2008;359(19):2005-17. 55. Brotto LA. Evidence-based treatments for low sexual desire in women. Front Neuroendocrinol. 2017;45:11-7. 56. Hofling M, Hirschberg AL, Skoog L, Tani E, Hägerström T, von Schoultz B. Testosterone inhibits estrogen/progestogen-induced breast cell proliferation in postmenopausal women. Menopause. 2007;14(2):183-90. 57. Tamimi RM, Hankinson SE, Chen WY, Rosner B, Colditz GA. Combined estrogen and testosterone use and risk of breast cancer in postmenopausal women. Arch Intern Med. 2006;166(14):1483-9. 58. Davis SR, Hirschberg AL, Wagner LK, Lodhi I, von Schoultz B. The effect of transdermal testosterone on mammographic density in postmenopausal women not receiving systemic estrogen therapy. J Clin Endocrinol Metab. 2009;94:4907-13. 59. Ness RB, Albano JD, McTiernan A, Cauley JA. Influence of estrogen plus testosterone supplementation on breast cancer. Arch Intern Med. 2009;169:41-6.

197

26. Klinge C1, Clark BJ, Prough RA. Dehydroepiandrosterone research: past, current, and future. Vitam Horm. 2018;108:1-28.

Testosterone therapy in women

60. Mudali S, Dobs AS, Ding J, Cauley JA, Szklo M, Golden SH, et al. Endogenous postmenopausal hormones and serum lipids: the atherosclerosis risk in communities study. J Clin Endocrinol Metab. 2005;90(2):1202-9. 61. Karim R, Hodis HN, Stanczyk FZ, Lobo RA, Mack WJ. Relationship between serum levels of sex hormones and progression of subclinical atherosclerosis in postmenopausal women. J Clin Endocrinol Metab. 2008;93(1):131-8. 62. Bernini GP, Sgroâ&#x20AC;&#x2122; M, Moretti A, Argenio GF, Barlascini CO, Cristofani R, et al. Endogenous androgens and carotid intimal-medial thickness in women. J Clin Endocrinol Metab. 1999;84(6):2008-12. 63. Spoletini I, Vitale C, Pelliccia F, Fossati C, Rosano GM. Androgens and cardiovascular disease in postmenopausal women: a systematic review. Climacteric. 2014;17(6):625-34. 64. Burger HG, Davis SR. The role of androgen therapy. Best Pract Res Clin Obstet Gynaecol. 2002;16(3):383-93. 65. YasuiT, Matsui S,Tani A, Kunimi K,Yamamoto S, Irahara M. Androgen in postmenopausal women. J Med Invest. 2012;59(1-2):12-27.

67. Handelsman DJ. Indirect androgen doping by oestrogen blockade in sports. Br J Pharmacol. 2008;154(3):598-605. 68. Pope HG, Wood RI, Rogol A, Nyberg F, Bowers L, Bhasin S. Adverse health consequences of performance-enhancing drugs: an Endocrine Society scientific statement. Endocr Rev. 2014;35(3):341-75. 69. Hartgens F, Kuipers H. Effects of androgenic-anabolic steroids in athletes. Sports Med. 2004;34(8):513-54. 70. Onakomaiya MM, Henderson LP. Mad men, women and steroid cocktails: a review of the impact of sex and other factors on anabolic androgenic steroids effects on affective behaviors. Psychopharmacology (Berl). 2016;233(4):549-69. 71. Blouin K, Boivin A, Tchernof A. Androgens and body fat distribution. J Steroid Biochem Mol Biol. 2008;108(3-5):272-80. 72. Achar S, Rostamian A, Narayan SM. Cardiac and metabolic effects of anabolic-androgenic steroid abuse on lipids, blood pressure, left ventricular dimensions, and rhythm. Am J Cardiol. 2010;106(6):893-901.

66. Salmon UJ, Geist SH. Effect of Androgens upon Libido in Women. J Clin Endocrinol Metab. 194;3(4):235-8.

198

Arch Endocrinol Metab. 2019;63/3

original article

Hypothyroid offspring replacement with euthyroid wet nurses during lactation improves thyroid programming without modifying metabolic programming Jorge Tapia-Martínez1, 2, Margarita Franco-Colín1, Rocio Ortiz-Butron2, Marisol Pineda-Reynoso3, Edgar Cano-Europa1

ABSTRACT Objective: Determine the milk quality effect during lactation on the metabolic and thyroid programming of hypothyroid offspring. Materials and methods: Ten-week-old female Wistar rats were divided into two groups: euthyroid and thyroidectomy-caused hypothyroidism. The rats were matted and, one day after birth, the pups were divided into three groups: euthyroid offspring (EO), hypothyroid offspring (HO) and hypothyroid with a euthyroid replacement wet nurse (HRO). During lactation, the milk quality and offspring body length were evaluated. The body weight and energy intake were determined on a weekly basis, as well as the metabolic profile at the prepubertal (P3536) and postpubertal (P55-56) ages. At P56, the animals were sacrificed, the adipose tissues were weighed and the thyroid glands were dissected for histological processing. Results: The milk of the hypothyroid wet nurse decreases proteins (16-26%), lipids (22-29%) and lactate (22-37%) with respect to euthyroid. The HO has a lower body weight gain (23-33%), length (11-13%) and energy intake (1521%). In addition, HO presents impaired fasting glucose and dyslipidemia, as well as a reduction in seric thyroid hormone (18-34%), adipose reserves (26-68%) and thyroid gland weight (25-34%).The HO present thyroid gland cytoarchitecture alteration. The HRO develop the same metabolic alterations as the HO. However, the thyroid gland dysfunction was partially prevented because the HRO improved under about 10% of the serum thyroid hormone concentration, the thyroid gland weight although histological glandular changes presented. Conclusions: The replacement of hypothyroid offspring with a euthyroid wet nurse during lactation can improve the thyroid programming without modifying metabolic programming. Arch Endocrinol Metab. 2019;63(3):199-207

Departamento de Fisiología, Escuela Nacional de Ciencias Biológicas. Instituto Politécnico Nacional, Ciudad de México, México 2 Departamento de Fisiología, Escuela Nacional de Ciencias Biológicas. Instituto Politécnico Nacional, Ciudad de México, México 3 Academía de Histología, Departamento de Formación Básica Disciplinaria, Escuela Superior de Medicina. Instituto Politécnico Nacional, Ciudad de México, México 1

Correspondence to: Edgar Cano-Europa edgarcanoeuropa@yahoo.com.mx Received on Feb/10/2017 Accepted on Mar/10/2017 DOI: 10.20945/2359-3997000000132

INTRODUCTION

n recent years, several observations have assumed that metabolic diseases in adulthood emerge in utero or during lactation as a change result in the organs and systems under suboptimum conditions (1). In these periods, the neuronal and endocrine systems undergo critical periods of growth and maturation. They also develop their interrelationship signaling pathways (2). Thus, there are irreversible metabolic consequences when the gestation and/or lactation is/are perturbed. Several experiments in animal models have shown that nutrition (3), hormone-regulated metabolism (4,5), and other toxic influences such as thyroid disruptors Arch Endocrinol Metab. 2019;63/3

affect metabolic and thyroid programming (6). In these conditions, adaptive metabolic strategies in the offspring change the metabolic set points in order to permit survival under critical conditions (7); for example, malnutrition in lactating rats was associated with changes in body weight (3,8) and thyroid dysfunction in their adult offspring (9). Congenital hypothyroidism is a clinical and biochemical disease caused by low mother-transferred fetus thyroid hormones during gestation (10); it compromises hypothalamus-pituitary-thyroid gland (HPT) axis development and functionality (11,12). There is evidence that congenital hypothyroidism 199

Keywords Congenital hypothyroidism; lactation; metabolic programming; thyroid programming; thyroid gland

Hypothyroid offspring replacement during lactation improves thyroid programming

affects the offspring metabolic programming because it reduces growth hormone (13) and insulin-like growth factor binding protein (IGBFP)-2 responses (14). In addition, the wet nurse thyroid state is very important for the offspring metabolic and thyroid programming. It has been shown that thyroid hormones are necessary for preparing the breast for lactation (15) because they participate in the mammary gland lipolytic enzymes and proteases synthesis and expression. Thus, the milk of a hypothyroid wet nurse has low energetic metabolites, such as lipids and proteins (16,17); moreover, it contains a low hormone concentration (i.e., thyroid hormone IGF-1, growth hormone and leptin), which altered the metabolic actions in the offspring (17,18). In particular, thyroid hormones are important for the HPT development during the first ten days of lactation (19), and the adequate protein supply during lactation ensures the correct thyroid programming (20). Altogether, congenital hypothyroidism causes metabolic programming changes. Further, the low quality of the milk produced by a hypothyroid wet nurse could modify thyroid gland programming. Therefore, the study aimed to prove that lactation in hypothyroid offspring with a euthyroid wet nurse, could improve hypothyroidism by causing metabolic and thyroid programming alteration.

observation, the sternothyroid muscle was cut and the trachea was exposed. The parathyroid gland was located, dissected from the thyroid gland, and reimplanted into the surrounding neck muscle. The thyroid gland was carefully dissected to avoid injury to the laryngeal nerve and was completely excised. After surgery, enrofloxacin (10 mg/kg, im) and mefenamic acid (1 mg/kg ig) were administered over three days to alleviate pain and prevent infection. During surgery and in the recovery period, 0.05% of death was presented. Seven days post-surgery, three females were placed into a plastic cage with one male for mating. Only the thyroidectomized rats received pulses of 20 µg/kg sc T4 twice; first at the beginning of mating and, second, one day before labor; in previous experiments we found a low fertilization rate and a high delivery mortality rate of the HWN if these pulses of thyroid hormones were not administered. One day after birth, the pups were randomly assigned to groups (each group consisted of four males and four females raised by a wet nurse wet nurse) and were divided into three groups: euthyroid group (EO, n=12), hypothyroid group (HO, n=12) and hypothyroid replacement with euthyroid wet nurse (HRO, n=12). The HRO group consists of placing hypothyroid offspring with a euthyroid wet nurse.

Metabolic evaluations

MATERIALS AND METHODS

Energetic metabolites quantification in the nurses’ milk

Animals and experimental design

The milk quality from each wet nurse in the different groups was tested. On day 7, 14 and 21 a dose of oxytocin (0.2 UI/kg i.p.) was administered to make the milk ejection process faster, then 10 min after oxytocin administration the rats were lightly anesthetized with ether and the mammary gland was gently pressed to simulate stimulating suction; the milk was then collected into individual vials (16). During the process the offspring were separated from the wet nurse for a period of 10 minutes to avoid stress by maternal separation. We obtained 500 µL stored at -20 °C. Until use, the amount of proteins, triglycerides, lactate and non-esterified fatty acid (NEFA) were quantified by using a kit from Randox.

All procedures were performed in accordance with the provisions of the laws and codes of Mexico in the seventh article of the General Health Law regarding health research (NOM-062-ZOO-2007). Additionally, the protocol was approved by the Internal Bioethical Committee (CEI-ENCB-021/2014). Twenty female ten-week-old (250-280g) Wistar rats were used. The animals were housed in acrylic cages (20 × 30 × 18 cm) in a room with a light-dark cycle (12:12 h) and controlled temperature (21 ± 1 °C); food and water were offered ad libitum. The rats were conditioned for a week before starting the experiment; they were randomly divided into two groups: euthyroid wet nurse (EWN, n = 8) and hypothyroid wet nurse (HWN, n = 12). Thyroidectomy with parathyroid reimplantation was performed to the hypothyroid group. Thyroidectomy was carried out on rats anesthetized with ketamine (10 mg/kg, im)–xylazine (5 mg/kg, im). The previously described method was used (21). Briefly, by using a stereomicroscope (Zeiss, Germany) for better 200

Metabolic evaluations in offspring During the whole treatment (P56), we measured the body weight and length two times a week during lactation. From P21 to P56, body weight and energy intake were determined weekly. We evaluated the rats at P35 and P55 because those dates are associated with prepubertal and postpubertal ages. After six hours of Arch Endocrinol Metab. 2019;63/3

Hypothyroid offspring replacement during lactation improves thyroid programming

Thyroid programming in offspring

Determination of serum concentrations of T3 and T4 We collected blood samples from the tail vein at 36 and 56 days post birth. For the determination of serum concentrations of T3 and T4 animals were not starved because the prolonged fast modifies the serum concentration of thyroid hormones. The samples were centrifuged at 3000 rpm for 5 min to obtain serum for thyroid hormone determination. We used the IMMULITE 2000 System (Siemens) to determine free T3 (Calibration Range: 40 to 600 ng/dL, Analytical Sensitivity: 19 ng/dL) and T4 (Calibration Range: 1.0 – 24 μg/dL, Analytical Sensitivity: 0.4 μg/dL).

Morphological analysis of the thyroid gland All animals were anesthetized at 56 days post birth with monosodium pentobarbital (45 mg/kg) after blood sampling. Six animals from each group underwent a total thyroidectomy using a stereoscopic microscope. We carefully preserved the thyroid gland lobes and the isthmus. The thyroid gland was immediately weighed on an analytical balance. Meanwhile, six animals from each group were used for histology. Briefly, 15 mm were dissected from the trachea containing the thyroid gland. It was fixed in 4% paraformaldehyde in PBS during 48 h. After that, the organs were processed by the conventional paraffin embedding technique. Serial sections of 7 µm were obtained and stained with hematoxylin-eosin.

Statistical analysis All of the variables evaluated were given as the mean ± standard error. The body weight, length, milk quantifications, clinical biochemistry of energetic metabolites and thyroid hormones were evaluated by repeated-measure two-way ANOVA and Student-Newman-Keuls post hoc. The two factors were age and thyroid state. The thyroid gland and adipose tissue weight were analyzed by one-way ANOVA and Student-Newman-Keuls post hoc. P < 0.05 was considered statistically significant. Arch Endocrinol Metab. 2019;63/3

RESULTS In Table 1, we observed the wet nurses milk quality during lactation. It was shown that hypothyroid wet nurses have a lower concentration of proteins (16-26%), triglycerides (22-29%), lactate (22-37%) and NEFA (10-11%) in all days that were evaluated in comparison to the euthyroid wet nurses. Figure 1 shows the hypothyroid metabolic effects on body weight (panel A and B), length (panel C and D) and energy intake (panel E and F). Perinatal and postnatal hypothyroidism reduces body weight gain, length and energy intake in male (body weight gain 23%, length 13%, energy intake 15% respect to AUC of EO) and female rats (body weight gain 33%, length 11%, energy intake 21% respect to AUC of EO). Meanwhile, the hypothyroid group with a euthyroid wet nurse has the same metabolic pattern as the hypothyroid group (male body weight gain 28%, length 11%, energy intake 11%, female body weight gain 28%, length 18%, energy intake 15%, respect to AUC of EO). The clinical biochemistry of the energetic metabolites that were analyzed is shown in Table 2. The hypothyroid and hypothyroid with euthyroid wet nurse groups suggested increased glucose (12-25%), triglycerides (47-67%) and cholesterol (55-62%) and reduced NEFA (17-28%) in both ages and genders. Figure 2 shows the reserves of adipose tissue on P56. We found that females in the HO and HRO groups reduced all reserves of adipose tissue (mesenteric 31%, retroperitoneal 68%, paraendometrial 67% for HO and mesenteric 21%, retroperitoneal 59%, paraendometrial 58% for HRO with respect to EO). Meanwhile, males of Table 1. Milk quality evaluation during lactation Milk energetic metabolites Proteins (mg/dL) Triglycerides (mg/dL) Lactate (mg/dL) NEFA (mmol/mL)

Day lactation

Females Euthyroid (EWN, n = 8)

Hypothyroid (HWN, n = 12)

53.23 ± 0.81a

45.04 ± 1.14b

59.08 ± 1.48

43.91 ± 0.984b

57.73 ± 1.16

a a

47.26 ± 1.67a

163.06 ± 7.95

128.33 ± 6.53d

155.24 ± 9.59b

110.14 ± 7.63a

153.13 ± 7.12c

119.75 ± 4.20c

7.90 ± 0.38a

6.13 ± 0.55b

6.93 ± 0.57

4.38 ± 0.475d

9.77 ± 0.64

14.12 ± 0.79

12.73 ± 0.65b

13.48 ± 0.42

12.03 ± 1.12b

14.92 ± 0.39

12.66 ± 1.54b

b c a a a

fasting, at P35 and P56 blood samples from the tail vein were obtained and the serum was separated and stored at −70 °C until assayed. Glucose, triglycerides, cholesterol and NEFA were measured. At P56, the animals were sacrificed by decapitation; the thyroid gland and mesenteric, retroperitoneal, epididymal or paraendometrial adipose tissue were dissected and were weighed immediately after removal.

7.03 ± 0.72e

(a≠b≠c≠d≠e≠f) p < 0.05. RM two-way ANOVA and Student-Newman-Keuls post hoc.

201

Hypothyroid offspring replacement during lactation improves thyroid programming

Euthyroid (EO) Hypothyroid (HO) Hypothyroid replacement with euthyroid wet nurse (HRO)

A 360

B 360

300

300 240

*’**

9000

60 40

*’** *’**

20 *

0 1

14 21 28 Days post-birth

Body weight (g)

120 100 Body weight (Area under curve)

7750 b

20 1

*’**

7750 6500 4000

5250

Euthyroid Hypothyroid Hypothyroid with euthyroid 42 49 56 wet nurse

28 35 Days post-birth

*’** 10 *’**

300 200 100

5 1

7 14 Days post-birth

a b

*’**

300

*’**

* Euthyroid Hypothyroid Hypothyroid with euthyroid wet nurse 21

E 120

200 100

5 1

Body length (Area under curve)

Length (cm)

*’**

15 Body length (Area under curve)

Length (cm)

* **

9000

D 25

7 14 Days post-birth

Euthyroid Hypothyroid Hypothyroid with euthyroid wet nurse 21

F 120

*’**

0 28

3000

*’**

Energy intake (Kcal/d)

Energy intake (Area under curve)

Energy intake (Kcal/d)

*’**

Euthyroid Hypothyroid Hypothyroid with euthyroid 42 49 56 wet nurse

C 25

80 40

5250 4000

*’**

120 100 60

6500

* **

180

2600

80 *

2200 1800

Euthyroid Hypothyroid Hypothyroid with euthyroid wet nurse 56

42 Days post-birth

*’**

0 28

*’**

Energy intake (Area under curve)

Body weight (g)

180

Body weight (Area under curve)

240

3000 2600 2200 1800

42 Days post-birth

a b

Euthyroid Hypothyroid Hypothyroid with euthyroid 56 wet nurse

Figure 1. Effect of pre-natal and post-natal hypothyroid on body weight (A and B) and length (C and D) and energy intake (E and F) of male (A, C and E) and female (B, D and F) rats (n = 12). The data represent the mean ± the standard error, the bar graphs of the sides represent the area under the curve of the graph. (*) p < 0.05 vs the euthyroid group; (**) vs hypothyroid group at same day; a≠b p < 0.05. RM two-way ANOVA and Student Newman Keuls post hoc. 202

Arch Endocrinol Metab. 2019;63/3

Hypothyroid offspring replacement during lactation improves thyroid programming

Table 2. Effect of pre-natal and post-natal hypothyroidism in clinical biochemistry of energetic metabolites in males and females at the pre-pubertal and post-pubertal age Prepubertal age (P35) Euthyroid (EO; n = 12)

Hypothyroid (HO; n = 12)

Hypothyroid with Euthyroid wet nurse (HRO; n = 12)

Euthyroid (EO; n = 12)

Hypothyroid (HO; n = 12)

Hypothyroid with Euthyroid wet nurse (HRO; n = 12)

Glucose (mg/dL)

95.60 ± 1.25a

117.00 ± 2.72b

110.60 ± 4.65b

95.40 ± 0.872a

119.40 ± 4.37b

107.80 ± 4.02b

Triglycerides (mg/dL)

78.80 ± 3.61a

120.00 ± 6.74b

116. 60 ± 7.45b

79.00 ± 3.78a

132.20 ± 9.49b

130.60 ± 7.22b

Cholesterol (mg/dL)

50.00 ± 1.92a

80.00 ± 2.45b

77.60 ± 2.79b

56.80 ± 3.18a

91.60 ± 3.59b

89.80 ± 1.50b

NEFA (mmol/ mL)

1.36 ± 0.07

1.08 ± 0.06

0.98 ± 0.05

1.74 ± 0.11

0.96 ± 0.05

0.96 ± 0.19b

Glucose (mg/dL)

95.20 ± 1.16a

114.80 ± 4.26b

104.80 ± 4.64b

94.40 ± 1.81a

112.80 ± 6.65b

108.00 ± 4.50b

Triglycerides (mg/dL)

85.60 ± 4.23

117.20 ± 7.46

128.40 ± 9.96

87.80 ± 3.15

130.40 ± 9.84

133.00 ± 10.38b

Cholesterol (mg/dL)

57.20 ± 3.41a

88.00 ± 3.44b

86.20 ± 3.31b

57.40 ± 2.23a

90.80 ± 2.52b

89.40 ± 1.47b

NEFA (mmol/ mL)

1.34 ± 0.06

1.15 ± 0.06

1.01 ± 0.05

1.65 ± 0.09

0.88 ± 0.06

0.83 ± 0.14b

Serum concentration

Male

Female

Postpubertal age (P55)

(a≠b) p < 0.05. RM two-way ANOVA and Student-Newman-Keuls post hoc.

D 1.6 a

1.2

0.8 0.4

Mesenteric tissue (g/100 g tissue)

0.0

0.4

0.4 0.0 Hypothyroid

Euthyroid

Retroperitoneal tissue (g/100 g tissue)

0.8

E 1.6

0.0 Euthyroid

1.6

a a

0.4 0.0 Euthyroid

Hypothyroid

Hypothyroid with euthyroid wet nurse

b b

0.4

Hypothyroid with euthyroid wet nurse

0.8

Hypothyroid with euthyroid wet nurse

1.2

Hypothyroid

1.2

0.8

Hypothyroid with euthyroid wet nurse

Paraendometrial tissue (g/100 g tissue)

Retroperitoneal tissue (g/100 g tissue)

Hypothyroid

B 1.6

Euthyroid

Epididymal tissue (g/100 g tissue)

1.2

0.0 Euthyroid

1.2

Hypothyroid

Hypothyroid with euthyroid wet nurse

1.2 0.8 b

0.4

0.0 Euthyroid

Hypothyroid

Hypothyroid with euthyroid wet nurse

Figure 2. Effect of pre-natal and post-natal hypothyroidism on reserves adipose tissue quantification of males [(A) mesenteric, (B) retroperitoneal, (C) epididymal] and females [(D) mesenteric, (E) retroperitoneal, (F) paraendometrial] (n = 12). (*) p < 0.05 vs the euthyroid group; a≠b p < 0.05. RM oneway ANOVA and Student Newmann Keuls post hoc. Arch Endocrinol Metab. 2019;63/3

203

Mesenteric tissue (g/100 g tissue)

A 1.6

Hypothyroid offspring replacement during lactation improves thyroid programming

DISCUSSION

the same groups reduced only mesenteric (26% for HO and 17% for HRO respect to EO) and retroperitoneal (57% for HO and 53% for HRO respect to EO) adipose tissue. The thyroid hormone concentrations and thyroid weight glands are shown in Table 3. It was observed that pre-natal and post-natal hypothyroidism caused a reduction on T3 and T4 at P35 (male reductions: T3 19% and T4 20%; female reductions: T3 22% and T4 18%) and P56 (male reductions: T3 22% and T4 34%; female reductions: T3 22% and T4 25%) in both genders. However, the hypothyroid group that had a euthyroid wet nurse presented a lower thyroid hormone reduction than the hypothyroid group at P35 (male reduction: T3 3% and T4 8%; female reduction: T3 7% and T4 6%) and P56 (male reduction: T3 14% and T4 19%; female reduction: T3 11% and T4 7%). Also, males and females of the hypothyroid group had a lower weight of the thyroid gland (34% for males and 25% for females). The arrows in Figure 3 show in the male (A) and female group (D) of EO, the thyroid gland has hemispherical follicles with follicular proper organization, abundant colloid and reabsorption lacunae are observed. In the male (B) and female (E) of HO, the gland has the presence of larger follicles without the hemispherical characteristic shape with little colloid; in addition, central areas with follicular disorganization were observed. With respect to the HRO male (C) and female (F) groups, the follicles have hemispherical morphology, presenting abundant larger follicles without colloid and flat cells, as well as areas with follicular disorganization.

Lactation is a critical period because important cognitive and metabolic events occur (22). Adverse changes in the environment, such as malnutrition and abnormal weight-regulatory hormonal concentration in lactating rats, are associated with metabolic disease in adulthood (23). Thyroid hormones (TH) are important in the mammary gland function because they participate in the adequate initiation and progress of lactation (15). In addition, the TH regulates energetic metabolic pathways as carbohydrates, lipids and proteins. Thus, the thyroid state in the wet nurses modifies the milk quality and the metabolic programming in the offspring (17). However, research has not yet determined whether the metabolic and thyroid programming alterations in hypothyroid offspring (HO) are the result of low nutritional intake of proteins and lipids and/or the low thyroid hormone contribution from the wet nurse during lactation. Thus, the aim of this study was to determine if hypothyroid offspring replacement with euthyroid wet nurses (HRO) during lactation develops thyroid and metabolic programming alterations during pre-pubertal and post-pubertal ages. We observed that hypothyroid wet nurses (HWN) produce milk with lower energetic metabolites (proteins and lipids) than that of euthyroid wet nurses (EWN). Hypothyroidism affects mammary function and lactation because it reduces the activity and expression of enzymes involved in energetic metabolic pathways as lipogenesis, lipolytic and protein synthesis. Thus, the HWN compromise the offspring nutritional state by causing nutrient restriction during lactation and

Table 3. Effect of pre-natal and post-natal hypothyroidism on serum thyroid hormones quantification and the thyroid gland weight of males and females at the pre-pubertal and post-pubertal age Prepubertal age (P36) Euthyroid (EO)

Hypothyroid (HO)

Hypothyroid with Euthyroid wet nurse (HRO)

Euthyroid (EO)

Hypothyroid (HO)

Hypothyroid with Euthyroid wet nurse (HRO)

T3 (ng/dL) (n = 12)

104.34 ± 5.37a

83.78 ± 3.89b

96.46 ± 1.68c

126.80 ± 4.62a

84.58 ± 3.83b

103. 02 ± 2.26c

T4 (µg/dL) (n = 12)

9.20 ± 0.14a

7.50 ± 0.08b

8.96 ± 0.20c

10.92 ± 0.22a

8.58 ± 0.27b

9.40 ± 0.12c

-----

18.20 ± 1.21a

12.10 ± 0.48b

14.94 ± 1.146c

Serum concentration

Male

Thyroid gland weight (mg) (n = 6) Female

T3 (ng/dL) (n = 12)

88.66 ± 2.69a

73.36 ± 2.18b

94.34 ± 2.61c

124.26 ± 8.96a

94.16 ± 5.22b

90.78 ± 2.52c

T4 (µg/dL) (n = 12)

10.22 ± 0.21

7.98 ± 0.26

9.54 ± 0.16

10.60 ± 0.22

8.32 ± 0.19

9. 00 ± 0.17c

-----

17.2 ± 1.02a

12.94 ± 0.37b

15.92 ± 0.36c

Thyroid gland weight (mg) (n = 6)

(a≠b≠c) p < 0.05. RM two-way ANOVA and Student-Newman-Keuls

204

Postpubertal age (P56)

. (---- not evaluated). Arch Endocrinol Metab. 2019;63/3

Hypothyroid offspring replacement during lactation improves thyroid programming

In mammals, many neuroendocrine networks that are responsible for metabolic and thyroid programming become organized during the intrauterine and perinatal stages of development. During puberty a complex developmental event of continuum changes occurs, leading a sexual and somatic maturation of the individual (27). However, nutrimental restriction during pregnancy and lactation may induce changes within the programming that affects the developmental

Figure 3. Effect of pre-natal and post-natal hypothyroidism on thyroid gland cytoarchitecture. The photomicrographs show histological cuts with HE stain. Euthyroid males (A) and females (D); hypothyroid males (B) and females (E); hypothyroid replacement with euthyroid wet nurse males (C) and females (F). Arrows show the follicular organization, colloid and reabsorption lacunae. Arch Endocrinol Metab. 2019;63/3

205

metabolic readjustment (24). We observed that HO has metabolic alterations and thyroid dysfunction in females and males. There are hormonal inputs that are related to the metabolic programming such as glucocorticoid, insulin and leptin (25). However, in recent years the relationship between TH and metabolic programming has been studied because TH can probably participate in metabolic programming during critical stages of development (26).

Hypothyroid offspring replacement during lactation improves thyroid programming

process of puberty and the presence of metabolic and thyroid disorders that are exacerbated in the long term (28). Even though the mechanism of this effect is still unknown, metabolic programming may be affected permanently by an imbalance in the TH supply during fetal life as it occurs in congenital hypothyroidism and causes long-term changes in the offspring. It has been observed that congenital hypothyroidism decreased insulin secretion in adulthood (29,30), which is related to high glucose concentration (31). In addition, lactation is an important period in which the correct metabolic and thyroid programming is established (23). It is possible that, by restoring lactation on HO, metabolic and thyroid programming could be improved. To analyze it, we employed the HRO to restore the hormonal and nutrimental supply during lactation. It has been shown that cross fostering during the first postnatal days can be an effective method to improve the newborn nutritional status in some species such as pigs (32). However, in rats, cross fostering is usually used to study behavior and thyroid disruptors in relationship with some metabolic parameters (33). So, this is the first study to use cross fostering in rats during lactation through the HRO with a EWN. However, this methodological strategy does not prevent metabolic programming; nevertheless, the thyroid programming was mildly improved. It has been observed that HWN reduces IGF-I and TH concentration in the milk. Thus, the metabolic restoration during lactation with EWN through quality milk provides the correct milk metabolic hormones concentration, but it does not prevent hypothyroidism. On other hand, the euthyroid milk provides more energetic metabolites, such as lipid and proteins, than hypothyroid milk (17). Perhaps the metabolic programming was not improved in HRO because the congenital hypothyroidism causes metabolic programming alteration. Furthermore, regarding thyroid programming, in the HRO, thyroid gland dysfunction was partially prevented. It has been observed that maternal TH during the end of gestation and throughout the ten lactation days programmed the HPT in the offspring (19,34). In addition, the nutritional supply during lactation it is important for the normal thyroid function (3). Thus, in our model, when the HO was fed with the euthyroid milk, the energetic metabolites and the thyroid hormone improve the HPT axis. In conclusion, we suggest that the thyroid status of the mother during gestation stage is very important for the correct metabolic programming in the offspring. 206

Moreover, lactation plays an important role in thyroid programming. Acknowledgements: this study was partially supported by CONACyT (221057) and SIP-IPN (20171416; 20171284). We thank Instituto Politécnico Nacional and CONACyT for financial support. The researchers are fellows of EDI, COFAA and SNI. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Fowden AL, Forhead AJ. Endocrine mechanisms of intrauterine programming. Reproduction. 2004;127(5):515-26. 2. Fernandez-Twinn DS, Ozanne SE. Mechanisms by which poor early growth programs type-2 diabetes, obesity and the metabolic syndrome. Physiol Behav. 2006;88(3):234-43. 3. Ayala-Moreno R, Racotta R, Anguiano B, Aceves C, Quevedo L. Perinatal undernutrition programmes thyroid function in the adult rat offspring. Br J Nutr. 2013;110(12):2207-15. 4. Ozdemir H, Akman I, Coskun S, Demirel U, Turan S, Bereket A, et al. Maternal thyroid dysfunction and neonatal thyroid problems. Int J Endocrinol. 2013;2013:987843. 5. Zahediasl S. Importance of thyroid hormones in intrauterine programming. 2010;8(4):186-7. 6. Wilcoxon JS, Redei EE. Prenatal programming of adult thyroid function by alcohol and thyroid hormones. Am J Physiol Endocrinol Metab. 2004;287(2):E318-26. 7. Patel MS, Srinivasan M, Laychock SG. Metabolic programming: Role of nutrition in the immediate postnatal life. J Inherit Metab Dis. 2009;32(2):218-28. 8. Passos MCF, Ramos CF, Moura EG. Short and long term effects of malnutrition in rats during lactation on the body weight of offspring. Nutr Res. 2000;20(11):1603-12. 9. Passos MC, da Fonte Ramos C, Dutra SC, Mouço T de ME. Longterm effects of malnutrition during lactation on the thyroid function of offspring. Horm Metab Res. 2002;34(1):40-3. 10. Jain V, Agarwal R, Deorari AK, Paul VK. Congenital hypothyroidism. Indian J Pediatr. 2008 Apr;75(4):363-7. 11. Obregon M, Calvo R, Escobar Del Rey F, Morreale De Escobar G. Ontogenesis of thyroid function and interactions with maternal function. Endocr Dev. 2007;10:86-98. 12. Patel J, Landers K, Li H, Mortimer RH, Richard K. Delivery of maternal thyroid hormones to the fetus. Trends Endocrinol Metab. 2011;22(5):164-70. 13. Seo H, Wunderlich C, Vassart G, Refetoff S. Growth hormone responses to thyroid hormone in the neonatal rat. Resistance and anamnestic response. J Clin Invest. 1981;67(2):569-74. 14. Ramos S, Goya L, Martin MA, Escrivá F, Pascual-Leone AM. Influence of hypothyroidism on circulating concentrations and liver expression of IGF-binding proteins mRNA from neonatal and adult rats. J Endocrinol. 2002;172(2):363-73. 15. Campo Verde Arboccó F, Sasso CV, Nasif DL, Hapon MB, Jahn G. Effect of hypothyroidism on the expression of nuclear receptors and their co-regulators in mammary gland during lactation in the rat. Mol Cell Endocrinol. 2015;412:26-35. 16. Hapon MB, Varas SM, Giménez MS, Jahn GA. Reduction of Mammary and Liver Lipogenesis and Alteration of Milk Composition during Lactation in Rats by Hypothyroidism. Thyroid. 2007;17(1):11-8. 17. Hapon MB, Simoncini M, Via G, Jahn GA. Effect of hypothyroidism on hormone profiles in virgin, pregnant and lactating rats, and on lactation. Reproduction. 2003;126(3):371-82. Arch Endocrinol Metab. 2019;63/3

Hypothyroid offspring replacement during lactation improves thyroid programming

18. Passos MCF, Lisboa PC, Pereira-Toste F, Oliveira E, Moura EG. Developmental plasticity in thyroid function primed by maternal hyperleptinemia in early lactation: a time-course study in rats. Horm Metab Res. 2012;44(7):520-6. 19. Brown RS, Shalhoub V, Coulter S, Alex S, Joris I, De Vito W, et al. Developmental regulation of thyrotropin receptor gene expression in the fetal and neonatal rat thyroid: relation to thyroid morphology and to thyroid-specific gene expression. Endocrinology. 2000;141(1):340-5. 20. Ramos CF, Lima APS, Teixeira CV, Brito PD, Moura EG. Thyroid function in post-weaning rats whose dams were fed a low-protein diet during suckling. Brazilian J Med Biol Res. 1997;30(1):133-7. 21. Pineda-Reynoso M, Cano-Europa E, Blas-Valdivia V, HernandezGarcia A, Franco-Colin M, Ortiz-Butron R. Hypothyroidism during neonatal and perinatal period induced by thyroidectomy of the mother causes depressive-like behavior in prepubertal rats. Neuropsychiatr Dis Treat. 2010;6:137-43. 22. Rodrigues AL, de Moura EG, Fonseca Passos MC, Potente Dutra SC, Lisboa PC. Postnatal early overnutrition changes the leptin signalling pathway in the hypothalamic-pituitary-thyroid axis of young and adult rats. J Physiol. 2009;587(11):2647-61. 23. Patel MS, Srinivasan M. Metabolic programming in the immediate postnatal life. Ann Nutr Metab. 2011 Jan [cited 2014 May 4];58 Suppl 2(suppl 2):18-28.

27. Castellano JM, Bentsen AH, Sánchez-Garrido MA, Ruiz-Pino F, Romero M, Garcia-Galiano D, et al. Early metabolic programming of puberty onset: Impact of changes in postnatal feeding and rearing conditions on the timing of puberty and development of the hypothalamic kisspeptin system. Endocrinology. 2011;152(9):3396-408. 28. Engelbregt MJ, Houdijk ME, Popp-Snijders C, Delemarre-van de Waal HA. The effects of intra-uterine growth retardation and postnatal undernutrition on onset of puberty in male and female rats. Pediatr Res. 2000;48(6):803-7. 29. Farahani H, Ghasemi A, Roghani MZS. The effect of maternal hypothyroidism on the carbohydrate metabolism and insulin secretion of isolated islets in adult male offspring of rats. Horm Metab Res. 2010;42:792-7. 30. Karbalaei N, Ghasemi A, Hedayati M, Godini AZS. The possible mechanisms by which maternal hypothyroidism impairs insulin secretion in adult male offspring in rats. Exp Physiol. 2014;99(4):701-14. 31. Karbalaei N, Ghasemi A, Faraji FZS. Comparison of the effect of maternal hypothyroidism on carbohydrate metabolism in young and aged male offspring in rats. Scand J Clin Lab Invest. 2013;73:87-94. 32. Deen MGH, Bilkei G. Cross fostering of low-birthweight piglets. Livest Prod Sci. 2004;90(2-3):279-84.

24. Hapon MB, Varas SM, Jahn GA, Gimenez MS. Effects of hypothyroidism on mammary and liver lipid metabolism in virgin and late-pregnant rats. J Lipid Res. 2005;46(6):1320-30.

33. Crofton KM. PCBs, Thyroid Hormones, and Ototoxicity in Rats: Cross-Fostering Experiments Demonstrate the Impact of Postnatal Lactation Exposure. Toxicol Sci. 2000;57(1):131-40.

25. Fowden AL, Forhead AJ. Hormones as epigenetic signals in developmental programming. Exp Physiol. 2009;94(6):607-25.

34. Calvo R, Obregón MJ, Ruiz de Oña C, Escobar del Rey F, Morreale de Escobar G. Congenital hypothyroidism, as studied in rats. Crucial role of maternal thyroxine but not of 3,5,3’-triiodothyronine in the protection of the fetal brain. J Clin Invest. 1990;86(3): 889-99.

26. Bagheripuor F, Ghanbari M. Comparison of the effects of fetal hypothyroidism on glucose tolerance in male and female rat offspring. 2015;(24):179-85.

Arch Endocrinol Metab. 2019;63/3

207

original article

Does having Turner syndrome affect quality of life in Brazilian women compared to common population? Maria Bernarda Estevez1, Patricia Teofilo Monteagudo2, Kelly Christina Oliveira3, Ieda Therezinha do Nascimento Verreschi4

Ambulatório de Endocrinologia, Escola Paulista de Medicina, Universidade Federal de São Paulo (EPM/Unifesp), São Paulo, SP, Brasil 2 Departamento de Endocrinologia, Escola Paulista de Medicina, Universidade Federal de São Paulo (EPM/Unifesp), São Paulo, SP, Brasil 3 Laboratório de Esteroides, Escola Paulista de Medicina, Universidade Federal de São Paulo (EPM/ Unifesp), São Paulo, SP, Brasil 4 Departamento de Medicina, Escola Paulista de Medicina, Universidade Federal de São Paulo (EPM/Unifesp), São Paulo, SP, Brasil 1

Correspondence to: Patricia Teófilo Monteagudo Departamento de Endocrinologia, Escola Paulista de Medicina, Universidade Federal de São Paulo Rua Estado de Israel, 639, 04022-001 – São Paulo, SP, Brasil ptmonteag@gmail.com

ABSTRACT Objectives: We aimed to measure the quality of life (QoL) of patients with Turner syndrome (PTS) and determine the extent to which their clinical or laboratory alterations influence QoL compared to reference women (RW) of the same age range. Subjects and methods: From Dec-2013 to Dec2014, 90 participants were recruited. They were 18 years and older: 48 with Turner syndrome (TS) (PTS) and 42 without (RW). Recruited subjects completed the Portuguese version of Short Form 36 (SF-36) questionnaire, and blood was drawn to measure LH, FSH, oestradiol (E2), progesterone (P4), SHBG, and SDHEA (by ECLIA) and testosterone (by LC MS/MS). Results: Age and schooling were similar between groups. The most common occupations for PTS were health worker, administration and education, and health worker or cashier for RW. Most participants were Catholic or Evangelical. Eighty-one percent (39/48) of cases used Hormonal Replacement Therapy (HRT), mostly transdermal (23/39). RW and PTS scored similarly on the SF-36 questionnaire. RW had higher oestradiol (p = 0,01), lower FSH (p = 0,01) and higher testosterone (p = 0,01) than PTS. Concentrations of P4, LH, SHBG or SDHEA were similar. Significant associations were found among QoL and hormones (E2 with Vitality and LH with Physical Role) only in the PTS group. Conclusions: PTS do not consider that TS affects their QoL as measured by domains on the SF-36. Oestradiol was related with QoL, emphasising the importance of HRT. Arch Endocrinol Metab. 2019;63(3):208-14 Keywords Quality of life; Turner syndrome; sexual steroids; hypogonadism

Received on Aug/24/2017 Accepted on Mar/14/2018 DOI: 10.20945/2359-3997000000136

INTRODUCTION

n 1938, Dr. Henry Turner described a group of signs and symptoms that were named after him as Turner syndrome (TS) (1). In 1954, clinicians at the the Escola Paulista de Medicina (EPM for short) as reported in Décourt and cols., associated an absence of Barr’s corpuscles in skin biopsy with the clinical features described by Turner (2). In 1959, Ford and cols. described the karyotype, demonstrating altered sexual chromosomes (3). The rate of this genetic condition is ½,000 female newborns, and the most common karyotype in TS is the 45,X (monosomy) condition (60%) (4), though in 99% of cases, the 45,X embryo does not reach full term (5). TS-related genetic anomalies may manifest as clinically observable characteristics such as short stature, low-set ears and hair, webbed neck, genu varum, or syndactilia. Clinical complications 208

may include: gonadal failure, osteoporosis, hearing loss, and sometimes severe cardiovascular diseases (6). Additionally, some psychological problems that can appear include low self-esteem, difficulty with body image and self-consciousness about clinical characteristics (7). Quality of Life (QoL) is an idea linked to one’s welfare and the capability to live to the fullest. Moreover, concepts like QoL, social functioning, and health conditions are considered synonymous (8). As with any chronic disease, patients need informed, attentive and professional follow-up to maintain a satisfactory QoL (9). Therefore, treatment aims to help the patient reach the best physiological and equilibrated state (10). The “health level” of a person is subjective and can only be verified using standardized clinical procedures, such as measuring vital signs, or by consensus between experts. Arch Endocrinol Metab. 2019;63/3

Thus, guidelines or validated resourceful questionnaires were designed to add objectivity to the measurement of QoL, and the most widely used of these have been the World Health Organization Quality of Life (WHOQoL) instrument (8), and the Short Form 36 Health Survey (SF-36, J.E. Ware, QualityMetrics and Medical Outcomes Trust) (11). The SF-36 has been used extensively in Brazil to assess QoL in rheumatoid arthritis (12) and renal insufficiency (13), having been adapted and translated to Brazilian Portuguese in UNIFESP by Ciconelli and cols. in 1999 (14). Additionally, Naess and cols. (15) previously used the SF-36 in a TS population in Norway. We hypothesised that patients diagnosed with TS may have QoL impairments and when these patients are clinically stable with normal laboratory results, they may have different specific domain scores on the SF-36 questionnaire compared to reference women. We also examined the influence of sexual steroids on patient QoL.

SUBJECTS AND METHODS From Dec-2013 to Dec-2014, 90 participants were recruited, being 48 patients with Turner syndrome (PTS), from an endocrinology outpatient clinic at a Federal University in Sao Paulo, Brazil, and from the Support Group for Women with Turner Syndrome (Grupo de Apoio às Mulheres com Síndrome de Turner, GAMT). To serve as a reference group, 42 healthy women of similar age and normal weight and height, who could be companions, relatives, or students of Unifesp (Medicine or Nursing schools) or clinic workers were included. The project was approved by Universitary Hospital Sao Paulo, Unifesp Research Ethics Committee (CAAE number 24501813.6.0000.5505), and it complies with the Helsinki Declaration. The enrolled participants signed the informed consent term attached to the study questionnaire. This study also followed all biohazard measures and complied with the local regulations of the Ministry of Health of Brazil MS 466/12 for Research in Human Subjects. According to data collected in 2010 by the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística, IBGE) (16), Brazil’s population at the time was 190,732,694, while São Paulo city had 5.924.872 women. Considering the TS rate of 1/2,000 female newborns, this means that as many as 2.962 women in the city could have TS, and Arch Endocrinol Metab. 2019;63/3

our studied population is representative of the studied population as in Unifesp Outpatient centerthere were 102 clinical files of girls or women with TS and almost half of them participated. Participants were 18 years or older and were eligible for the study as PTS if they had any karyotype classified as TS. Participants were eligible for participation as Reference Women (RW) if they were healthy and had no exclusion criteria, as are the following: 1) severe illness, 2) currently using birth control medication, 3) had any history of ovarian disease (such as policystic ovary syndrome), or 4) currently pregnant or breastfeeding. Collected information included the presence of TS, age, schooling, occupation classified according to the guidelines of the Brazilian Ministry of Work and Employment (2002) (17), religion, use of hormone replacement therapy (HRT) and the method of administration, and the use of any other medications. For PTS, clinical and laboratorial data for thyroid function, comorbidities and medicines were collected from the medical records. All patients were asked to bring their audiometry evaluation. Two PTS had been undergone a webbed neck correction plastic surgery. We chose the SF-36 as our QoL questionnaire because it is an easy, general, and widely used patient self-reported outcome survey (18). The SF-36 contains 36 self-report items assessing 8 health domains (Physical Functioning, Physical Role, Bodily Pain, General Health Perceptions, Vitality, Social Functioning, Emotional Functioning and Mental Health). Each domain receives its own score ranging from 0 = worst QoL to 100 = best QoL. We utilised the Portuguese version of the SF-36 developed by Ciconelli and cols. (14). Serum concentrations measured included: LH (mlU/mL), FSH (mUl/mL), oestradiol (E2) (pg/mL), progesterone (P4 ng/mL), sex hormone-binding globulin (SHBG nmol/l) and dihydrotestosterone sulphate (DHEA-S µg/dL) by electrochemiluminescence immunoassay (ECLIA) and testosterone by liquid chromatography mass spectrometry (LC MS/MS). Blood samples were collected at follicular stage of menstrual cycle for RW, and follicular-like period for PTS (7-12 days after start of menses). Sera were obtained by centrifugating 7 to 10 mL of blood per patient at 2.000 rpm per 13 to 15 minutes in a SORVALL brand centrifuge. Samples were then frozen and stored at -20 degrees Celsius until processing. Thawing occurred twelve months after the 209

Does having TS affect QoL in Brazil?

blood draw of the first participant; measurement of all the ECLIA-measured hormones was executed at the Laboratory of Steroids of Unifesp in December 2014. Roche Cobas laboratory reactives were used in an ELECSYS 2010 spectrophotometer. Testosterone was measured in an AB sciex QTrap 5500. In addition to testosterone, all the samples and reactions by ECLIA were processed together and on the same occasion dosed in the Special Chemistry Laboratory of Albert Einstein Hospital; later, testosterone was processed in March 2015. In the case of an undetectable result, the lower limit of detection was used as follows: E2 – 5 pg/mL; P4 – 0,03 ng/mL; LH – 0,1 mUI/mL; FSH – 0,1 mUI/mL; SDHEA – 0,100 mcg/dL; SHBG – 0,350 nmol/L; Testosterone – 10 ng/dL. We excluded too high values. Results from statistical analyses are presented as medians and absolute or relative frequencies for qualitative or categorical variables, and as means ± standard deviations for quantitative continuous variables. The Kolmogorov Test of Normality was conducted to identify normally distributed variables. The Pearson correlation coefficient was calculated to examine the relationship between two quantitative normally distributed variables and the Spearman quotient for non-parametric quantitative variables. The association between qualitative variables was verified with contingency tables and analysed with the chi-square test. Student’s t-tests were used to identify significant differences between two quantitative normally distributed variables, and the Mann Whitney test was used for non-parametric quantitative variables. ANOVA was used to compare qualitative variables having three or more categories; differences were considered significant if p < 0.05. Analyses were conducted using IBM SPSS Statistics 20 software (New York, United States).

RESULTS We analysed 48 PTS and 42 RW. Thirty-seven PTS had the 45,X karyotype (monosomy), two had 45,X/Xip, 4 had 45,X/Xiq, and 5 had other karyotypes. Their final height was 145.6 ± 6.1 cm (from 134 to 158 cm). They were 12.7 ± 9.8 years old when TS diagnose was confirmed (from birth to 53 yo). Twenty (42%) received GH treatment. When we compare the highest (150.5 ± 4.1 cm of height) with the lowest TPS patients (141.0 ± 3.6 cm), divided by the Median, there was no 210

diference in their QoL. However, there was a tendence to Functional Capacity (p 0.065) and General Health Perceptions (p = 0.72) aspects of QoL to be higher in higher TPS patients them in lower ones. Age in the PTS group varied from 18 to 65 years (Median 29.5 years) and from 18-57 years (Median 31.5 years) in the RW group. No significant difference in age between the two groups was observed. Twenty PTS (42%) and fifteen RW (36%) reported having an undergraduate schooling level, and therefore, could have a job and career. Employment domains reported among PTS included education (3/48), business administration (6/48), health (4/48), and food preparation (4/48). Employment areas among RW included health worker (13/42), cashier (4/42) and food preparation (2/42). Whereas two PTS reported that they worked with machines (one in a factory, another in an amusement park), no RW reported that occupation type. In addition, 9 PTS and 11 RW declined to state their religion; however, the majority of the participants was Christian (Catholic, 52% of PTS, 31% of RW, or Evangelical, 18,8% PTS and 28,6% RW). Thirty-nine PTS (81%) were taking HRT, three were at menopausal age with no HRT, one used to have migraine with any kind of estrogen replacent, and the other 5 were non-adherents. Of the 39, 23 patients (59%) used transdermal administration (patch or gel), and 16 used oral pills (41%), with the cicle starting at day 1 each month, estrogen replacement everyday and progestogen replacement from days 10 to 13 till 20 to 23 every month. Regarding other reported medications in PTS group, 26 (54%) patients took none, considering separately the use of HRT and eleven took more than one. The most common medicines included calcium or vitamin D, used by 15 patients, and levothyroxine, taken by 12 patients. Four patients were taking serotonine reuptake inhibitors, ten hypertensive TPS were taking losartan and/or amlodipine, and 2 diabetics were taking metformin. All had clinical and laboratorial exams well compensated. Thirty-one participants (73%) from the RW group reported taking no medication, five used thyroid hormone replacement, 2 used serotonine reuptake inhibitors, 1 used a bloodthinner (blood clot prevention), 1 used metformin, one was taking losartan plus clortalidone, and the last one used an antidopaminergic medication. Two RW were at menopause age. Regarding QoL, analyses revealed no difference in scores from SF-36 between both groups (Table 1). Arch Endocrinol Metab. 2019;63/3

Does having TS affect QoL in Brazil?

levels in one PTS and one RW, and for testosterone levels in 17 PTS and two RW with undetectable results. Table 2 presents hormone results for the study. Levels of oestradiol and testosterone were significantly higher in RW than PTS, and FSH was higher in PTS (all p < 0.01). The observed correlations for the PTS group between hormone concentrations and QoL were only with E2 and Vitality (rS = 0.32, p = 0.04) and LH with Physical Role (rS = 0.33, p = 0.02). None such correlations were found in the RW group.

Table 1. Scores from the SF-36 for patients with Tuner syndrome and for the reference women

DOMAINS

Patients with Turner Syndrome

Reference women

n = 48

n = 42

Mean ± SD

General Health Perceptions

59 ± 17

61 ± 16

0.93 ns

Mental Health

57 ± 20

58 ± 20

0.62 ns

Vitality

60 ± 22

62 ± 21

0.23 ns

DOMAINS

Median (min-max)

Physical function

85 (10-100)

87,5 (35-100)

0.73 ns

Physical role

100 (0-100)

0.63 ns

Social role

75 (0-100)

0.72 ns

Emotional role

100 (0-100)

0.60 ns

Bodily pain

84 (0-100)

73 (10-100)

0.34 ns

DISCUSSION In our study, QoL was similar between both groups when measured using the SF-36 questionnaire. Additionally, we observed no differences in education, occupation or religion. However, we noticed that even when PTS under HRT had lower levels of oestradiol than controls, they were similar to those of a normal woman in her follicular stage according to the Unifesp laboratory reference ranges. The levels of oestradiol correlated with the Vitality domain of the SF-36 only in PTS, showing that levels of oestradiol may affect QoL if lower than reference range. Schooling is not impaired in the TS group according to our study, showing that a diagnosis of TS was not influential on patients’ education level or on the possibility of patients completing their studies or becoming professionals. This finding was supported further by the observation that people in the PTS and RW groups had similar careers and occupations. This is consistent with findings from Naess and cols. on education, but they found a difference on work areas (more PTS in education, more RW in Economics or Administration) (15) that we did not, probably due to characteristics

Scores from the SF-36 do not have a unit of measurement.

Twenty-seven percent (13 of 48) of TS had hypoacusia, though only one had severe hypoacusia, and that was approprietadly corrected with an earpiece. When we compared QoL data, the TS patients with hypoacusia did not differ with normal hearing TS nor with RW. Blood samples were not collected in two PTS and three RW (because of certain difficulties returning to the facility); therefore, those individuals were excluded from laboratory analyses, leaving a total of 45 PTS and 40 RW. One RW presented a level of oestradiol of 1,157 pg/mL, and thus was excluded because it exceeded limits (12.5 to 166 pg/mL in the follicular phase). Her other hormonal levels were normal. In seven PTS and one RW, the result was undetectable and rounded up to the minimum value in the detection range for oestradiol levels. The same was done for LH

Turner Women

Reference Women

Laboratory Reference

0.01

12.5-166

Mean ± SD

Range (min/max)

Mean ± SD

Range (min/max)

50.1 ± 67.2

5.0-357.7

105.6 ± 91.5

5.0-398.8

P4 (ng/mL)

1.1 ± 0.4

0.1-21.7

2.5 ± 4.9

0.2-19.6

0.3

0.2-1.5

LH (mIU/mL)

26.6 ± 20.5

0.1-99.1

41.3 ± 167.4

0.1-1066

0.56

2.4-12.6

FSH (mIU/mL)

65.6 ± 46.5

15.2-216.7

11 ± 21.7

0.4-135.1

0.01

3.5-12.5

SDHEA (µg/dL)

189.5 ± 111.3

21.7-498.7

195.5 ± 108.5

42.1-449

0.10

33.9-337 (age 10-44)

SHBG (nmol/L)

65.6 ± 46.5

15.2-216.7

84 ± 56.6

18.2-282.6

0.8

32.4-128 (age 20-44)

10-35

10-67

0.01

10-38*

E2 (pg/mL)

T (ng/dL)

E2: oestradiol; P4: progesterone; LH: luteinizing hormone; FSH: follicle stimulant hormone; SDHEA: sulphate-dehydriepiandrosterone; SHBG: sex hormone-binding globulin; T: total testosterone; pg/mL: picograms per mililitre; ng/mL: nanograms per millilitre; mUI/ml: mili international units per millilitre; μg/dL: micrograms per decilitre; nmol/L: nanomol per litre; ng/dL: nanograms per decilitre. * Lab Reference for Tanner stage 5. Arch Endocrinol Metab. 2019;63/3

211

Table 2. Laboratory results for Patients with Tuner syndrome and the reference women

Does having TS affect QoL in Brazil?

of the educational system. Approximately 50% of our participants had an undergraduate or graduate level of education, and 65% of the participants in the Norwegian study had achieved 15 years of schooling, fostering the observation of the difference in the placement of PTS. By the other side, Stocholm et al found in Denmark that even if the patients with Turner syndrome and their controls had no difference in the possibility to get a bachelor´s degree, they did have difference in their income, until the third decade (19). Unfortunately, we did not ask for the income in the present study. The religious affiliations reported by the participants are common in Brazil (20), and there was no significant difference in declared religions between the groups. The feelings of faith and trust in “something or someone” that can provide material and spiritual protection and love doubtlessly influence individual feelings of welfare (21). Religious practice is related to quicker and better recovery in people with physical or mental ailments, and with the general sensation of health and longevity (22). Some mental issues related to TS can be “made up for” or “caught up” with age and appropriate environment stimuli (23). Differently from the Norwegian study of Naess and cols. (15), we noticed that only on some of SF36 domains, in particular scores for General Health Perceptions and Mental Health, our participants’ scores were lower in the PTS than in the RW, though no significance was found. The differences in cultural and social characteristics of two different countries should be considered, as they can affect people’s opinions about QoL. TS-specific QoL has also been measured in Israel by Zucherman-Levin and cols (with the WHOQoL), who showed that treatment with androgens improved general health, mood, feelings of well-being, sexual desire and relationships (24). Ciconelli and cols.’s investigating QoL in patients with rheumatoid arthritis (RA), found lower scores in QoL regarding characteristics specific to RA, such as Physical Function, Physical Role, Emotional Role and Bodily Pain or higher scores for General Health Perceptions, Social Role, Mental Health and Vitality (16). However, the observable characteristics of TS, such as webbed neck or facial stigmata, did not influence the QoL in our patients, though, the shortest PTS tended to have lower scores for domains Functional Capacity and General Healthy Perceptions. This pinpoints the importance of earlier diagnose of TS and GH taking for a greater period, improving final height to these patients and possibly their QoL. 212

In addition to HRT, patients could be using other kinds of medication for eventual comorbidities, such as hypertension, diabetes mellitus, dyslipidaemia or depression. In our study, the PTS received more medications than the RW, independently of HRT. Therapy findings showed 19% of the PTS were not taking HRT, either because of noncompliance or appropriate age for menopause, meaning the PTS’s reasons for noncompliance need to be addressed in the medical consult. Deverney et al described on their study that the compliance to medical care is influenced by type of physician, paternal socioeconomic class, education level, number of Turner syndrome disease components, size of the medical center following the patient in childhood, and physical health dimensions of Short Form 36 questionnaire (25). On a Dutch study by Freriks and cols., 30% of their sample had not accurate follow-up, and 15% were not receiving HRT at the time (26). In the present work, the laboratory findings showed that PTS can obtain laboratory results quite similar to those of women in their follicular phase, though mean levels of oestradiol were lower. Additionally, the correlation between E2 and the Vitality domain of SF36 questionnaire in PTS combined with the lack of such a relationship in RW is attributed to a manifestation of subclinical hypoestrogenism, present in PTS but absent in the eugonadal RW. Suggesting that vitality decreases without HRT would be an argument in favour of compliance with it. A correlation was not found in any group between a domain of QoL and P4, but RW had a wider range of results because they have natural cycles. The reason for this finding is probably related to the duration of the progesterone pills in the scheme the PTS use (recommended for 10 to 14 days during the month) (26), while in women with normal cycles, the levels will tend to vary according to the phase they are in, and the changes in concentration influence one’s feelings of welfare (27). The fact of PTS taking a progestogen for short periods during the month (from days 13 to 23 every month) can be related to the nonassociation of this hormone with QoL FSH levels but not those of LH were significantly different between groups (FSH higher in PTS, achieving levels of menopause). The effects of gonadotrophins in the absence of gonads were described in the classic study of Thales Martins in 1931 (28), and FSH does not diminish with HRT because of the lack of FSH receptors in the dysgenetic gonad (29). In addition, Healy and Arch Endocrinol Metab. 2019;63/3

Does having TS affect QoL in Brazil?

Arch Endocrinol Metab. 2019;63/3

In conclusion, the women with TS did not report negative impacts of the diagnosis on their QoL as measured by the SF-36 questionnaire. Patients with TS adapt themselves well to the challenges of their condition and to factors that affect QoL of patients undergoing treatment with HRT, which are the same in people without TS. We suggest that psychological support be provided in a model based on that genetic characteristic. Furthermore, treatment with HRT is able to equate the QoL of people with or without TS, and the hormonal profile was similar to the profile of a normal woman in her follicular phase. In the age range of the participants in our study, the desirable amount of naïve or untreated people, or patients without follow-up should be low because TS is present at birth. Therefore, because oestradiol was associated with vitality, it shows the importance of HRT, and the association of LH with physical aspects measured by the SF-36 corroborates this statement. Moreover, the absence of effect of HRT on FSH is due to the absence of gonadotrophin feedback from inhibin. Finally, though very low testosterone concentrations are related to underdeveloped or streak gonads, there is no association between low testosterone concentrations and QoL in our patients. Acknowledgements: the authors MBE, PTM and ITNV acknowledge the kind cooperation of the participants of this study, and the academic sponsorship of Unifesp/EPM (Universidade Federal de São Paulo/Escola Paulista de Medicina) for this work. Fundings: MBE declares that this work was financed with a scholarship from SENESCYT (Secretaria Nacional de Educación Superior Ciencia y Tecnología) of Ecuador and PROEX/CAPES Brazil funds for laboratory reactives and materials. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1.

Turner H. A syndrome of infantilism, congenital webbed neck and cubitus valgus. Endocrinology. 1938;23:566-74.

2. Décourt L, Da Silva S, Chiorboli E, Lima MC, Fernandes JM. Sobre o sexo genético nas pacientes com síndrome de Turner. Rev Assoc Med Bras. 1954;1(2):203-6. 3. Ford CE, Jones KW, Polani PE, DE Almeida JC, Briggs JH. A sexchromosome anomaly in a case of gonadal dysgenesis (Turner’s syndrome). Lancet. 1959;1(7075):711-3. 4. Vilar L. Endocrinología clínica. 4ª ed. Rio de Janeiro: Guanabara Koogan; 2009. 5. Bianco B, Lipay M, Guedes A, Oliveira K, Verreschi IT. SRY gene increases the risk of developing gonadoblastoma and/or nontumoral gonadal lesions in Turner syndrome. Int J Gynecol Pathol. 2009;28(2):197-202.

213

cols. suggested the need for a counterregulation factor secreted by the gonads such as inhibin, to keep FSH at normal levels (18). Inhibin in women is produced in the granulosa cells of follicles. In the study by Gravholt and cols., levels of inhibin-A and inhibin-B in Turner women were generally undetectable. However, some of them showed slightly lower levels of gonadotrophins and detectable inhibin-B (at prepubertal levels), showing that the dysgenetic gonad of PTS is not sufficiently able to participate in the counterregulation of FSH due to the early and accelerated destruction of follicles before birth (30,31). SHBG levels were not significantly different between the PTS and RW groups in our study. However, Hampl and cols., in their study in the Czech Republic, reported that untreated PTS had low SHBG, while PTS receiving GH had even lower concentrations, yet PTS under oestrogens and HRT had increased levels (32). On one hand, SHBG is considered as a marker of insulin resistance (33), on the other, Turner patients have greater insulin resistance, Teoretically the counterbalance between the expected increase of SHBG with HRT and its decrease amidst insulin resistance can justify the absence of difference in SHBG levels between PTS and RW in the present study. We observed in our patients that testosterone levels were lower in PTS than in RW, and many PTS had concentrations below the detection limit of method, which was 10 ng/dl. However, the maximum levels found in PTS were in the range of women in Tanner stage 5 of puberty. Tanner stage was not assessed because participants were 18 or older, presumably done with their pubertal development, so it can be assumed that they were M4P4 or more developed. Testosterone levels were not associated with any domains measured by SF-36, probably because the questions did not include inquiries about sexual desire/performance, feelings of aggressiveness or weight changes. Even though testosterone has been found by immunohistochemical methods in hilium cells of PTS gonads, obtained after prophylactic gonadectomy, it is possible that these levels are too low to be found in blood (34). Though our sample was representative of ST population, the fact that only women who willingly presented to the clinic participated may cause some bias of selection in the sample, and is a limitation of our study. Also, this study lacks sufficient data to describe the situation of untreated PTS, as the population without treatment was quite small (9 patients).

Does having TS affect QoL in Brazil?

6. Guimarães MM, Guerra CTG, Alves STF, Cunha CSAM, Martins LA, Barreto LFM, et al. Intercorrências clínicas na síndrome de Turner. Arq Bras Endocrinol Metabol. 2001;45(4):331-4. 7.

Acevedo Lopez YS. La inteligencia emocional en las adolescentes con síndrome de Turner. Círculo de Humanidades. 2010;31:103-10.

8. Fleck A, Pio M, for the Group WHOQoL. Versão em português dos instrumentos de avaliação de qualidade de vida. Organização Panamericana da Saúde/Universidade Federal do Rio Grande do Sul. Grupo de Estudos de Qualidade de Vida; 1998. Available from: http://www.ufrgs.br/psiquiatria/psiq/whoqol1.html. Accessed on: Oct. 15, 2013. 9. Gawlilk A, Kackzor B, Kaminska H. Quality of clinical follow up of Young women with Turner syndrome treated in a clinical center. Horm Res Pediatr. 2012;77(4):222-8. 10. Wierman ME, Bason R, Davis SR. Androgen therapy in women, an Endocrinology Society Clinical Practice. J Clin Endocrinol Metab. 2006;91(10):3697-710. 11. Medical Outcome Trust. Instruments, SF-36. Available from: http:// www.outcomes-trust.org/instruments.htm#SF-36 Accessed on: Oct. 15, 2013. 12. Mota LMH, Laurindo IMM, Santos Neto LL. Avaliação prospectiva da qualidade de vida em uma coorte de pacientes com artrite reumatoidea inicial. Rev Bras Reumatol. 2010;50(3):249-61. 13. Neto JF, Ferraz MB, Cendoroglo M, Draibe S, Yu L, Sesso R. Quality of life at the initiation of maintenance dialysis treatment – a comparison between the SF-36 and the KDQ questionnaires. Qual Life Res. 2000; 9(1):101-7. 14. Ciconelli R, Ferraz MB, Santos W, Meinão I,Quaresma MR. Tradução para língua portuguesa e validação do questionário genérico SF-36 para avaliação de qualidade de vida. Rev Bras Reumatol. 1999;39(3):143-50. 15. Naess E, Bahr D, Gravholt C. Health Status of Women with Turner syndrome, a questionnaire on health status, education, work participation and aspects of sexual functioning. Clin Endocrinol. 2009;72(5):678-84. 16. Instituto Brasileiro de Geografia e Estatística (IBGE). Censo 2010. Available from: http://cidades.ibge.gov.br/xtras/temas.ph p?lang=&codmun=355030&idtema=90&search=sao-paulo|saopaulo|censo-demografico-2010:-resultados-da-amostracaracteristicas-da-populacao. Accessed on: Oct. 15, 2013. 17. Brazilian Ministry of Work and Employment. Classificação Brasileira de Ocupações. 2002. Available from: http://www. mtecbo.gov.br/cbosite/pages/pesquisas/BuscaPorTitulo.jsf. Accessed on: Nov. 2015. 18 Healy DJ, et al. Contributions to clinical and reproductive endocrinology from in vitro fertilization, em Bailliere’s Endocrinology, McKays of Chatham; 1987. Chapter 5. 19. Stochholm K, Hjerrild B, Mortensen KH, Juul S, Frydenberg M, Gravholt CH. Socioeconomic parameters and mortality in Turner syndrome. Eur J Endocrinol. 2012;166(6):1013-9.

21. González-Valdéz TL. Las creencias religiosas y su relación con el proceso salud-enfermedad. Rev Psicol Iztlacala. 2004;7(2):19-29. 22. Alves RR, Alves HDAN, Barboza RR, Souto WDEM. The influence of religiosity on health. Cienc Saude Colet. 2010;15(4):2105-11. 23. Ricardi FF, Zaia LL, Pellegrino-Rosa I, Rosa JT, de Assis MOZ, Saldanha PH. Psychogenetics of Turner syndrome, investigation of 28 subjects and respective controls using the Bender test and piagetian scales. Genet Mol Res. 2010;9(3):1701-5. 24. Zucherman-Levin N, Frolova-Bishara T, Militianu D, Levin M, Aharon-Peretz J, Hochberg Z. Androgen replacement therapy in Turner Syndrome, a pilot study. J Clin Endocrinol Metab. 2009;94(12):4820-7. 25. Devernay M, Ecosse G, Coste J, Carel JC. Determinants of medical care in Young women with Turner syndrome. J Clin Endocrinol Metab. 2009;94(9):3408-13. 26. Freriks K, Timmermans J, Beerendonk CC, Verhaak CM, Netea-Meier R, Otten BJ, et al. Standardized multidisciplinary evaluation yields significant previously undiagnosed morbidity in adult women with Turner syndrome. J Clin Endocrinol Metab. 2009;96(9):E1517-26. 27. Werther G, Zacharin M.Turner Syndrome Management Guidelines – Australasian Paediatric Endocrine Group. 2003. Available from: http://www.thesis.xlibx.info/th-other/257230-1-turner-syndromemanagement-guidelines-australasian-paediatric-endoc.php. Accessed on: Nov. 2013. 28. Martins T. Physiologia do lobo anterior da hyphopyse e glándulas sexuaes. Arq Fund Oswaldo Cruz. 1931;22:233-41. 29. Clarke IJ, Cummins JT. Pulsatility of reproductive hormones: physiological basis and clinical implications. Baillieres Clin Endocrinol Metab. 1987;1(1):1-21. 30. de Serrano MN, Fernández EG,Mariño A, Peramato PG. Espectro morfológico de las gónadas en los estados intersexuales en la edad pediátrica. An Pediatr. 2005;64(Supl 2):38-44. 31. Gravholt CH, Naeraa RW, Andersson AM, Christiansen JS, Skakkebaek NE. Inhibin A and B in adolescent and young adults with Turner’s syndrome and no sign of spontaneous puberty. Hum Reprod. 2012;17(8):2049-53. 32. Hampl R, Snajderova M, Lebl J, Lisa L, Dvorakova M, Hill M, et al. Sex hormone-binding globulin as a marker of effect of hormonal treatment in Turner’s syndrome. Endocr Regul. 2001;35(1):17-24. 33. Wojcik M, Janus D, Zygmunt-Gorska A, Starzyk JB. Insulin resistance in adolescents with Turner syndrome is comparable to obese peers, but the overall metabolic risk is lower due to unknown mechanism. J Endocrinol Invest. 2015;38(3):345-9. 34. Mendes JR, Strufaldi MW, Delcelo R, Moisés RC, Vieira JG, Kasamatsu TS, et al. Y-chromosome identification by PCR and gonadal histopathology in Turner’s syndrome without overt Y-mosaicism. Clin Endocrinol. 1999;50(1):19-26.

20. Instituto Brasileiro de Geografia e Estatística (IBGE). Censo 2010. Características gerais de população, religião e pessoas

com deficiência. Available from: http://www.ibge.gov.br/home/ estatistica/populacao/censo2010/caracteristicas_religiao_ deficiencia/default_caracteristicas_religiao_deficiencia.shtm Accessed on: Nov. 2013.

214

Arch Endocrinol Metab. 2019;63/3

original article

Familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families from Muzambinho (Minas Gerais, Brazil) Laboratório de Hemodinâmica da Atividade Motora, Escola de Educação Física e Esporte, Universidade de São Paulo, SP, Brasil 2 Laboratório de Comportamento Motor, Escola de Educação Física e Esporte, Universidade de São Paulo, SP, Brasil 3 Instituto Federal do Sul de Minas – Campus Muzambinho, Muzambinho, MG, Brasil 4 Núcleo de Investigação em Actividade Física e Saúde, Centro de Investigação e Desenvolvimento do Desporto e Actividade Física (CIDAF), Faculdade de Educação Física e Desporto (FEFD), Universidade Pedagógica, Maputo, Moçambique 5 Laboratório de Cineantropometria e Gabinete de Estatística Aplicada, Centro de Investigação, Formação, Intervenção e Inovação em Desporto (CIFI2D), Faculdade de Desporto, Universidade do Porto, Porto, Portugal 1

João Paulo dos Anjos Souza Barbosa1, Luciano Basso2, Teresa Bartholomeu1, Januária Andrea Souza Rezende3, Jorge Alberto de Oliveira2, António Prista4, Go Tani2, José António Ribeiro Maia5, Cláudia Lúcia de Moraes Forjaz1

ABSTRACT Objective: This study investigated the familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families from Muzambinho (Minas Gerais, Brazil). Subjects and methods: The study included members of 139 families, comprising 97 fathers (aged 40 ± 7 years), 129 mothers (35 ± 6 years), 136 sons (12 ± 4 years), and 121 daughters (12 ± 5 years). Evaluated markers included (A) body mass index, waist circumference, glycemia, and cholesterolemia, as metabolic risk markers; (B) total weekly volume of physical activity, as a physical activity marker; and (C) relative muscle strength, as a physical fitness marker. Correlations between family members and heritability (h²) were estimated using the software S.A.G.E. Results: Significant familial correlations were obtained between parents-offspring for glycemia and cholesterolemia (both r = 0.21, p < 0.05) and relative muscle strength (r = 0.23, p < 0.05), and between siblings for waist circumference, glycemia, total weekly volume of physical activity, and relative muscle strength (r variation 0.25 to 0.36, p < 0.05). Heritability values were significant for almost all variables (h² variations: 20% to 57% for metabolic risk markers, 22% for the total weekly volume of physical activity, and 50% for relative muscle strength), except for waist circumference (h² = 15%, p = 0.059). Conclusion: The presence of significant correlations between family members and/or significant heritability strengthens the possible genetic and/or common familial environment influence on metabolic risk markers, total weekly volume of physical activity, and relative muscle strength. Arch Endocrinol Metab. 2019;63(3):215-21 Keywords Cardiovascular risk factors; motor activity; muscle strength; families; genetic epidemiology

Correspondence to: João Paulo dos Anjos Souza Barbosa Rua Coronel Ferreira Leal, 246, ap. 1 05588-090 – São Paulo, SP, Brasil jpdosanjos@usp.br Received on Dec/12/2017 Accepted on Mar/11/2019

INTRODUCTION

ardiovascular diseases are the main cause of death worldwide (1) and in Brazil (2). Several factors, like the presence of metabolic disease (e.g., obesity, diabetes, and dyslipidemia), are known to increase cardiovascular risk (1). On the other hand, regular practice of physical activity is considered a cardiovascular protective factor. Physical fitness also has cardiovascular protective effects (3), and body strength has been recently shown to be inversely related to cardiovascular morbidity and mortality (4). Since all these factors may be affected by Arch Endocrinol Metab. 2019;63/3

genetic and environmental influences, it is important to identify the magnitude of the contribution of each influence (genetic and environmental) to understand how different interventions may affect an individual’s cardiovascular risk. Genetic epidemiology is a research area focused on the investigation of possible genetic influences on a trait (5,6). The first two steps of a genetic epidemiology study are the identification of significant familial aggregation (greater correlations of a trait within family members than expected by chance) and heritability 215

DOI: 10.20945/2359-3997000000137

Familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families

(proportion of a trait variability in a population that may be explained by genetic factors) of the trait, which are assumptions necessary to ensure the presence of genetic influence (6,7). However, familial aggregation and heritability vary among different populations, since environmental influences differ with location and time, reducing or increasing the genetic influence (6-8). Familial aggregation and heritability of a trait have a specific value for each population; therefore, these characteristics must be studied in different contexts because the replication of significant values reinforces the genetic involvement in determining a specific trait. Familial aggregations of markers of metabolic risk, physical activity, and physical fitness have been investigated outside Brazil. Studies from France, United States, and Portugal have found greater correlations between biological relatives (parents-offspring and siblings) than between father-mother for body mass index (BMI) (9), glycemia (10), and metabolic syndrome index (11), supporting a genetic influence on these traits. Regarding total physical activity, familial aggregation between genetically related relatives varies from an absence of correlation (12) to significant correlations (r = 0.34), as found in a study by Maia and cols. (13). For physical fitness, studies of familial aggregation are limited to aerobic fitness (14,15). Similarly, the heritability of markers of metabolic risk, physical activity, and physical fitness has been studied in different countries and in a single study conducted in Brazil. As expected, heritability varies substantially in different populations. Systematic reviews have shown that heritability of BMI ranges from 30% to 76% (8). In the Brazilian study mentioned above, the heritability of metabolic risk markers varied from 29% to 51% (16). Regarding total physical activity, heritability varies from 10% to 30% (17) and was 35% in the Brazilian study (18). To the best of our knowledge, the heritability of muscle strength has not been investigated in Brazilian nuclear families. Based on the exposed above, the study of familial aggregation and heritability is important to ensure the genetic influence in different contexts. It is also important to study these aspects in Brazil, where only a single study in this area has been performed to date. Thus, the objectives of this study were to investigate the familial aggregation and heritability of markers of metabolic risk (BMI, waist circumference, glycemia, and cholesterolemia), physical activity (weekly volume of total physical activity), and physical fitness 216

(relative muscle strength) in nuclear families from the municipality of Muzambinho (Minas Gerais, Brazil). Our hypothesis was that all markers of metabolic risk, physical activity, and physical fitness would present significant familial aggregation and heritability.

SUBJECTS AND METHODS A sample of families was recruited among 10-year-old students participating in a large study about growth, maturation, and motor development in Muzambinho, (19). Data regarding blood pressure heritability from this study have already been published (20). The study was approved by the Ethics Committee of the School of Physical Education and Sport at the University of São Paulo (Brazil), and written informed consent was obtained from all participants. Data were collected between March 2008 and November 2009 in Muzambinho, a small city in the state of Minas Gerais, Brazil. At the time of data collection, the population of Muzambinho was 19,925 and included 5,650 residents aged 5 to 19 years. The municipality has a human developmental index of 0.801, and its main economy is based on agriculture, livestock, and handicrafts (21). The school system, at the time of data collection, consisted of 13 schools (2 private and 11 public schools, including 3 state and 8 municipal ones) with 2,996 school-age children enrolled (321 in the private schools, 1,845 in the state schools, and 830 in the municipal schools) (21). The city had 11 healthcare facilities at data collection. The number of deaths related to cardiovascular diseases was 35 in 2008 and 37 in 2009, representing 28% and 31%, respectively, of the deaths in the municipality each year (22). The study investigators visited the selected families at their homes to explain the study’s procedures, obtain written consent for participation in the study, and schedule a new visit to the home the next morning. In this second visit, fasting blood samples were collected from all family members, and a third appointment was set for an evening visit at one of the city’s schools. In this third appointment, anthropometric measurements, physical activity habits, and physical fitness were assessed.

Measurements Weight (kg) and height (m) were measured with an electronic scale (Filizola, São Paulo, Brazil) following standard procedures, and the BMI was calculated as weight (kg)/height (m)2. Waist circumference was Arch Endocrinol Metab. 2019;63/3

Familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families

Statistical analysis Initial exploratory and descriptive analyses were performed using SPSS 20.0 (SPSS Inc., Chicago, IL, USA). Two-way analysis of variance (ANOVA) was used to compare traits between generations and sexes. The chi-square test was used to compare categorical Arch Endocrinol Metab. 2019;63/3

variables between family groups. The description of the nuclear families was obtained with PEDSTATS (29). Analysis of missing values was carried out using SYSTAT 13 (Systat Software, Inc., San Jose, CA, USA) within each family structure, and missing data distribution was carried out at random. Correlations of each marker between family members were calculated using regression residuals after adjustment for the covariates age, sex, age X sex, age2, and age2 X sex, as suggested by Bouchard and cols. (30). Correlations between parent-offspring (rPO), siblings (rSI), and father-mother (rFM) were calculated using the software S.A.G.E. (31). Heritability (h²) was also estimated with S.A.G.E. (31) using Student’s t distribution for robust estimation of the parameters. This analysis was preceded by adjustments for the covariates age, sex, age X sex, age2, age2 X sex. The level of significance was set at 5% for all analyses.

RESULTS In all, 210 families were invited to participate in the study. Data were effectively obtained from 139 families ranging in size from 3 to 9 members, with most families having 3 or 4 members (81%). In some families, the mother and (particularly) the father were absent in the evaluation. Thus, the final sample consisted of 97 fathers, 129 mothers, and 257 children (136 sons and 121 daughters). The age of the offspring ranged from 6 to 24 years (sons 12 ± 4 years; daughters 12 ± 5 years), and the parents’ age ranged from 24 to 65 years (fathers 40 ± 7 years; mothers 35 ± 6 years). Concerning the socioeconomic status of the participants, only 10.8% of the families had at least one of the parents with an occupational activity requiring a university degree. Regarding family health history, 51.1% of the families reported a history of cardiovascular disease. As expected, BMI and waist circumference values were significantly higher in parents compared with offspring, but were comparable between sexes within the same generation (Table 1). When the BMI was analyzed as a categorical parameter, the frequency of excess weight (overweight and obesity) was similar between men and women of the same generation, and higher in the first generation (fathers [59.5%] = mothers [50.0%] > sons [22.0%] = daughters [20.0%], p < 0.05). The frequency of altered waist circumference was lower 217

measured at the level of the umbilicus (23,24). Based on BMI values, the participants were classified as normal, overweight, or obese according to international guidelines (24,25). International tables were used to classify the participants’ waist circumference (23,24). Glycemia and cholesterolemia were measured using automatic devices (Accu-Chek, Roche, São Paulo, Brazil; and Accutrend GC, Roche Diagnostics, Mannheim, Germany, respectively) after at least 6 hours of fasting (26,27). Physical activity was assessed by a specific and culturally appropriate structured interview conducted with all family members together. The main questions included the family’s usual commuting physical activity (how they commute from one place to another in the city, if walking, cycling, etc.), occupational physical activity (what they do for work, if domestic chores, rural tasks, deliveries, etc.), and leisure time physical activity (what they do during leisure time, if cycling, soccer, swimming, gym classes, ballet, or physical play/games such as tag, hide-and-seek, hopscotch, skip rope, and others). In addition, children were asked about their physical activity at school (including physical education classes and physical activity during breaks). When the participant declared to perform physical activity in any of the mentioned domains (commuting, occupational, school, and/or leisure time), the weekly frequency and duration of the activity were questioned, and the weekly volume of the activity was calculated as the product of frequency and duration. The weekly volume of total physical activity (min/week) was obtained by the unweighted sum of the volumes of all physical activities performed. Muscle strength was assessed by the isometric handgrip test using a hydraulic hand dynamometer (SH 5001, Saehan Corporation, Masan, South Korea). After adjustment of the handle of the equipment to the size of the volunteer’s hand, each volunteer performed a handgrip movement using the maximum strength possible with the dominant hand, and the value obtained was recorded (28). This value was relativized by body weight, yielding the relative muscle strength.

Familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families

in fathers (9.5%) compared with all other family groups and was greater in mothers (44.2%) compared with sons (24.6%) and daughters (31.9%), who presented similar frequencies of altered waist circumference. Glycemia was significantly higher in fathers compared with sons and mothers, while cholesterolemia was higher in both parents compared with offspring and in fathers compared with mothers (Table 1). The weekly volume of total physical activity was higher in parents compared with offspring and was similar between sexes at both generations. The relative muscle strength was higher in the first versus second generation and greater in males than females (father > mother > sons > daughters).

Table 2 presents the results of familial correlations. In regard to metabolic risk markers, there was no significant correlation between family members in terms of BMI. However, significant correlations were observed in terms of waist circumference between siblings (ρ = 0.25), glycemia between all family pairs (parents-offspring, ρ = 0.21; siblings, ρ = 0.28; and father-mother, ρ = 0.36), and cholesterolemia between parents-offspring (ρ = 0.21). Regarding the physical activity marker, the weekly volume of total physical activity correlated significantly between siblings (ρ = 0.36) and father-mother (ρ = 0.42). The physical fitness markers, relative muscle strength, correlated significantly between parents-offspring (ρ = 0.23) and siblings (ρ = 0.35).

Table 1. Descriptive information on markers of metabolic risk, physical activity, and physical fitness among family members from Muzambinho (MG, Brazil) Fathers Mean ± SE (n)

Mothers Mean ± SE (n)

Sons Mean ± SE (n)

Daughters Mean ± SE (n)

25.8 ± 0.36 (n = 84)

25.6 ± 0.41 (n = 120)

18.4 ± 0.31* (n = 127)

18.8 ± 0.39* (n = 115)

WC (cm)

91 ± 0.98 (n = 84)

87 ± 1.00 (n = 120)

66 ± 0.97* (n = 127)

67 ± 1.11* (n = 115)

Glycemia (mg/dL)

96 ± 1.25 (n = 92)

89 ± 0.89# (n = 125)

86 ± 0.77* (n = 134)

84 ± 0.73 (n = 119)

Cholesterolemia (mg/dL)

192 ± 3.67 (n = 71)

177 ± 2.30# (n = 91)

163 ± 1.88* (n = 72)

170 ± 2.75* (n = 64)

2018 ± 99.44 (n = 83)

2044 ± 94.05 (n = 119)

1320 ± 85.70* (n = 126)

1030 ± 70.30* (n = 109)

0.59 ± 0.01 (n = 83)

0.46 ± 0.01# (n = 120)

0.55 ± 0.01* (n = 123)

0.47 ± 0.01*# (n = 109)

Metabolic risk markers BMI (kg/m2)

Physical activity marker WVTPA (min/week) Physical fitness marker Relative strength (kg)

SE: standard error; BMI: body mass index; WC: waist circumference; WVTPA: weekly volume of total physical activity. * p < 0.05 compared with first-generation family member of same sex. # p < 0.05 compared with males of the same generation.

Table 2. Correlations between different pairs of family members regarding markers of metabolic risk, physical activity, and physical fitness in nuclear families from Muzambinho (MG, Brazil) ρPO ± SE

n pairs

ρSI ± SE

n pairs

ρFM ± SE

n pairs

BMI (kg/m2)

0.03 ± 0.06

WC (cm)

0.01 ± 0.06

343

0.17 ± 0.11

109

0.03 ± 0.12

343

0.25 ± 0.09†

109

0.15 ± 0.12

Glycemia (mg/dL)

†

0.21 ± 0.07

379

0.28 ± 0.09

135

†

0.36 ± 0.09

Cholesterolemia (mg/dL)

0.21 ± 0.09†

170

0.14 ± 0.16

-0.15 ± 0.14

0.10 ± 0.07

331

0.36 ± 0.09†

104

0.42 ± 0.10†

0.23 ± 0.06†

337

0.35 ± 0.11†

107

0.16 ± 0.12

Variables

Metabolic risk markers

†

Physical activity marker WVTPA (min/week) Physical fitness marker Relative strength (kg)

BMI: body mass index; WC: waist circumference; WVTPA: weekly volume of total physical activity; ρPO: parents-offspring correlation; ρSI: siblings correlation; ρFM: father-mother correlation. † p < 0.05.

218

Arch Endocrinol Metab. 2019;63/3

Heritability values for markers of metabolic risk, physical activity, and physical fitness were all significant, except for waist circumference, for which a trend toward significance was detected (Table 3). Heritability for metabolic markers ranged from 15% for waist circumference (not significant, p = 0.059) to 57% for cholesterolemia. The heritability value was 22% for the weekly volume of total physical activity and 50% for the relative muscle strength. Table 3. Maximal heritability (h2 ± SE) of markers of metabolic risk, physical activity, and physical fitness in nuclear families from Muzambinho (MG, Brazil) Variables

h2 ± SE

p value

BMI (kg/m2)

0.20 ± 0.09

0.019

WC (cm)

0.15 ± 0.09

0.059

Glycemia (mg/dL)

0.38 ± 0.08

< 0.001

Cholesterolemia (mg/dL)

0.57 ± 0.11

< 0.001

0.22 ± 0.08

< 0.001

0.50 ± 0.09

< 0.001

Metabolic risk markers

Physical activity marker WVTPA (min/week) Physical fitness marker Relative strength (kg)

SE: standard error; BMI: body mass index; WC: waist circumference; SBP: systolic blood pressure; DBP: diastolic blood pressure; WVTPA: weekly volume of total physical activity.

DISCUSSION The main findings of the present study are: (A) metabolic risk markers, except for BMI, present significant correlations between pairs of genetically linked family members. All metabolic risk markers but waist circumference presented low to moderately significant heritability; (B) weekly volume of total physical activity showed significant familial aggregation between siblings and father-mother, as well as significant low heritability; and (C) relative muscle strength presented significant familial aggregation between parents-offspring and siblings, as well as moderate significant heritability. The understanding of the concepts of familial aggregation and heritability is important for the interpretation of studies involving these two concepts. The presence of significant correlations between family members suggests a genetic and/or environmental influence shared by family members (common environment) on the trait (32). When correlations are significant between genetically linked pairs (parentsArch Endocrinol Metab. 2019;63/3

offspring and siblings), a genetic influence is possible, whereas in significant correlations observed only between nongenetically linked relatives (father-mother), a common environmental influence is more likely. When heritability is estimated in nuclear families, as done in the present study, separation of genetic and common environmental influences is unfeasible; thus, what is estimated is the maximal genetic influence supposing there is no effect from the common environment (32). In general, significant familial correlations and/ or heritability are observed for all metabolic risk markers, suggesting an influence of genes and/or common environmental on these traits (9,11) and corroborating the generally accepted idea that genes and familial environment partially determine an individual’s metabolic risk (33). However, considering the obesity markers assessed in our study, BMI showed no familial correlation and low heritability, while waist circumference only presented a significant correlation between siblings and a trend toward significant heritability. Although genetic and common environment may have had a role in determining the obesity risk in this population, their influence was weak. Heritability values of obesity markers vary considerably in the literature, ranging from 30% to 76% in different reviews (8,34), but the results found in the present study (20% for BMI and 15% for waist circumference) are below the limits reported previously. An important determinant of obesity is an individual’s eating habit, which is greatly influenced by cultural traditions (8). However, the eating habits of different families in small cities like Muzambinho are probably very similar, which explains the lower heritability observed in such populations. Unfortunately, the eating habits of the participants in this study were not evaluated and should be included in future studies. Regarding other metabolic risk markers, correlations for glycemia were significant for all family pairs, with the highest value observed between fathermother, who share no genes, suggesting an important influence of common environment on this trait (35). The heritability value of glycemia (38%) was significant and similar to the one found in another Brazilian study conducted in Baependi (33%) and to the highest value reported in the literature (39%) (16). Cholesterolemia showed a significant correlation among parentssiblings and a moderate heritability (57%), which was higher than both the value reported in the Baependi study (29%) and the highest value reported in the 219

Familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families

literature (51%) (36). Discrepancies in cholesterolemia heritability between Baependi and Muzambinho may be explained, at least in part, by different sampling strategies. The Baependi study involved many family generations, while the present study involved nuclear families with school-age offspring, who likely share the family environment more frequently among themselves than with distant family members like cousins, uncles, and grandparents. Together, these results suggest that a common environment may have a great influence on cholesterolemia. Based on previous discussion, it is possible to propose that interventions on unique environments and family common environments are more prone to have positive results in reducing obesity and other metabolic markers, respectively. Weekly volume of total physical activity presented significant correlations between siblings (r = 0.36) and father-mother (r = 0.42), and a significant but low heritability of 22%. As significant correlations were observed between both genetically and nongenetically related relatives, and were higher between the nongenetically related pairs, these results suggest that a common environment may influence this trait (32). The fact that the correlations were identified for pairs within the same generation (siblings and father-mother) suggests a special influence of the common environment shared by members of each generation in determining the weekly volume of physical activity (37). The low heritability value found for this trait is within the range reported in a review (6-62%) (38) and similar to the values found by Chaves and cols. (24%) (39), but lower than those reported in the Baependi study (35%) (18). This low heritability suggests that unique environments may have an important influence in determining this trait. Considering together the results from the familial aggregation and heritability, interventions aiming to increase the weekly volume of physical activity in this population should focus in the common environment shared by relatives from the same generation and the extra-familial environment. To the best of our knowledge, this is the first study to investigate the familial aggregation and heritability of relative muscle strength in Brazilian nuclear families. Relative muscle strength had a significant correlation between parents-offspring and siblings and a moderate heritability (h2 = 0.50). These results suggest that half of the variation of relative muscle strength in Muzambinho’s population might be explained by genetic factors and/or a common familial environment. 220

This result is aligned and within the range of isometric strength shown in other international studies involving nuclear families, showing estimates of heritability involving this type of strength of 27% to 58% (40). Despite the importance of these findings, this study has some limitations. The sample derived from a population with high levels of weekly total physical activity. Also, due to the selection criteria (families of 10-year-old children), the sample consisted of families with young adults as parents. These features may reduce the chance of metabolic risk abnormalities and increase the influence of the common familial environment. Since we had no previous knowledge of the physical activity patterns of this population at the time of data collection, we applied a structured open interview (20,41). However, interviews are vulnerable to some degree of inaccuracy; thus future studies should implement measures that are more objective. The size of the cohort of the present study may be considered insufficiently large, but is similar to that of previous studies (19,20,41) and represented 2.2% of the population of Muzambinho. In conclusion, in families from Muzambinho, markers of metabolic risk, weekly volume of total physical activity, and manual strength showed significant familial aggregation and/or heritability, suggesting a genetic and common environmental influence on these traits. Acknowledgments: we would like to acknowledge the volunteers, heads of the schools, and the Education and Health Secretariats of Muzambinho. We would also like to thank the students, teachers, and informatics technicians who helped with the data collection and analysis. The research was supported by multiple agency grants: CNPq (#478249/2007-1), USP, and CAPES-PROEX (001). Each author contributed individually and significantly to the development of this study. JPASB, AP, GT, JARM, and CLMF designed the study. LB, TB, AP, JASR, JAO, and CLMF collected the data. JPASB, TB, JASR, AP, LB, JAO, JARM, and CLMF tabulated and analyzed the data. All authors participated actively in the discussion of the results. JPASB, JARM, and CLMF were the main contributors in the manuscript preparation. All authors approved the final version of the manuscript. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Cannon CP. Cardiovascular disease and modifiable cardiometabolic risk factors. Clin Cornerstone. 2007;8(3):11-28. 2. Ministério da Saúde. Cadernos de informações de saúde. 2009. Available from: http://www.datasus.gov.br. Accessed in: 24 Mar. 2019. Arch Endocrinol Metab. 2019;63/3

Familial aggregation and heritability of markers of metabolic risk, physical activity, and physical fitness in nuclear families

4. Warburton DER, Nicol CW, Bredin SSD. Health benefits of physical activity: the evidence. CMAJ. 2006;174(6):801-9. 5. Forjaz CLM, Bartholomeu T, Rezende JAS, Oliveira JA, Basso L, Tan G, et al. Desafios no estudo de famílias nucleares: etapas iniciais de análise. Rev Bras Educ Fís Esporte. 2011;25(4):717-30. 6. Burton PR, Tobin MD, Hopper JL. Key concepts in genetic epidemiology. Lancet. 2005;366(9489):941-51. 7.

Maia JAR, Garganta R, Seabra A, Lopes VP. Heterogeneidade nos níveis de actividade física de crianças dos 6 aos 12 anos de idade: Um estudo em gémeos. Rev Port Ciênc Desporto. 2004;4(1): 39-50.

8. Fermino RC, Garganta R, Seabra A, Maia JAR. Efeitos genéticos e ambientais nos indicadores da composição corporal. uma revisão centrada em estudos de agregação familiar. Rev Bras Cineantropom Desempenho hum. 2007;9(4). 9. Trégouët DA, Herbeth B, Juhan-Vague I, Siest G, Ducimetière P, Tiret L. Bivariate familial correlation analysis of quantitative traits by use of estimating equations: Application to a familial analysis of the insulin resistance syndrome. Genet Epidemiol. 1999;16(1):69-83. 10. Tang W, Hong Y, Province MA, Rich SS, Hopkins PN, Arnett DK, et al. Familial Clustering for Features of the Metabolic Syndrome: The National Heart, Lung, and Blood Institute (NHLBI) Family Heart Study. Diabetes Care. 2006;29(3):631-6. 11. Santos DM, Katzmarzyk PT, Trégouet DA, Gomes TN, Santos FK, Maia JA. Familial Aggregation of Metabolic Syndrome Indicators in Portuguese Families. Biomed Res Int. 2013;2013:314823. 12. Mitchell BD, Rainwater DL, Hsueh WC, Kennedy AJ, Stern MP, Maccluer JW. Familial aggregation of nutrient intake and physical activity: results from the San Antonio Family Heart Study. Ann Epidemiol. 2003;13(2):128-35. 13. Maia J, Gomes TN, Trégouët DA, Katzmarzyk PT. Familial resemblance of physical activity levels in the Portuguese population. J Sci Med Sport. 2014;17(4):381-6. 14. Sallis JF, Patterson TL, Buono MJ, Nader PR. Relation of Cardiovascular Fitness and Physical Activity to cardiovascular disease risk factors in children and adults. Am J Epidemiol. 1988;127(5):933-41. 15. Foraita R, Brandes M, Günther F, Bammann K, Pigeot I, Ahrens W. The influence of aerobic fitness on obesity and its parentoffspring correlations in a cross-sectional study among German families. BMC Public Health. 2015;15:638. 16. de Oliveira CM, Pereira AC, de Andrade M, Soler JM, Krieger JE. Heritability of cardiovascular risk factors in a Brazilian population: Baependi Heart Study. BMC Med Genet. 2008;9:32. 17. Wolfarth B, Bray MS, Hagberg JM, Pérusse L, Rauramaa R, Rivera MA, et al. The human gene map for performance and health-related fitness phenotypes: the 2004 update. Med Sci Sports Exerc. 2005;37(6):881-903.

22. Ministério da Saúde. Sistema de Informações sobre Mortalidade – SIM. 2016. Available from: http://www2.datasus.gov.br/DATASUS/index.php?area=060701. Accessed in: 24 Mar. 2019. 23. McCarthy HD, Jarrett KV, Crawley HF. The development of waist circumference percentiles in British children aged 5.0-16.9 y. Eur J Clin Nutr. 2001;55(10):902-7. 24. World Health Organization. Diet, Nutrition and the Prevention of Chronic Disease. Geneva: WHO; 2003. 25. Centers for Disease Control and Prevention (CDC). Growth Charts. 2000. Available from: http://www.cdc.gov/growthcharts. Accessed in: 24 Mar. 2019. 26. del Cañizo FJ, Froilán C, Moreira-Andrés MN. Precisión y exactitud de la medida del colesterol total mediante el reflectómetro Accutrend GC®. Aplicabilidad en atención primaria para la detección de hipercolesterolemias. Atención Primaria. 1996;17(7):463-6. 27. Thomas LE, Kane MP, Bakst G, Busch RS, Hamilton RA, Abelseth JM. A Glucose Meter Accuracy and Precision Comparison: The FreeStyle Flash Versus the Accu-Chek Advantage, Accu-Chek Compact Plus, Ascensia Contour, and the BD Logic. Diabetes Technol Ther. 2008;10(2):102-10. 28. Council of Europe Committee for the Development of Sport. EUROFIT: Hanbook for the EUROFIT tests of physical fitness. Roma; 1988. 29. Wigginton JE, Abecasis GR. Abecasis, PEDSTATS: descriptive statistics, graphics and quality assessment for gene mapping data. Bioinformatics. 2005;21(16):3445-7. 30. Bouchard C, Malina RM, Pérusse L. Genetics of fitness and physical performance. Champaign: Human Kinetics; 1997. 31. Biostatistics. Statistical Analysis for Genetic Epidemiology (SAGE). Available from: http://darwin.cwru.edu/sage/. Accessed in: 24 Mar. 2019. 32. Falconer DS, Mackay TF. Introduction to quantitative genetics. Harlow: Longman; 1996. 33. Mansour-Chemaly M, Haddy N, Siest G, Visvikis S. Family studies: their role in the evaluation of genetic cardiovascular risk factors. Clin Chem Lab Med. 2002;40(11):1085-96. 34. Elks CE, den Hoed M, Zhao JH, Sharp SJ, Wareham NJ, Loos RJ, et al. Variability in the heritability of body mass index: a systematic review and meta-regression. Front Endocrinol (Lausanne). 2012;3:29. 35. Reynolds CA, Baker LA, Pedersen NL. Models of spouse similarity: Applications to fluid ability measured in twins and their spouses. Behav Genet. 1996;26(2):73-88. 36. Santos DM, Katzmarzyk PT, Diego VP, Souza MC, Chaves RN, Blangero J, et al. Genotype by energy expenditure interaction with metabolic syndrome traits: the Portuguese healthy family study. PLoS One. 2013;8(11):e80417. 37. Maia JA, Thomis M, Beunen G. Genetic factors in physical activity levels. Am J Prevent Med. 2002;23(2):87-91.

18. Horimoto AR, Giolo SR, Oliveira CM, Alvim RO, Soler JP, de Andrade M, et al. Heritability of physical activity traits in Brazilian families: the Baependi Heart Study. BMC Med Genet. 2011;12:155.

38. Chaves RN, Souza MC, Santos D, Garganta R, Seabra A, Maia JAR. Agregação familiar nos níveis de atividade física: Um resumo do estado da arte. Rev Bras Ativ Fís Saúde. 2010;15(1):65-9.

19. Basso L, Meira Júnior C, Oliveira JA, Forjaz CLM, Souza JA, Prista A, et al. Crescimento e desenvolvimento motor de escolares de Muzambinho: um estudo com implicações acadêmicas, sociais e de política interinstitucional. Rev Port Ciênc Desporto. 2009;9:247-57.

39. de Chaves RN, Baxter-Jones A, Santos D, Gomes TN, dos Santos FK, de Souza MC, et al. Clustering of body composition, blood pressure and physical activity in Portuguese families. Ann Hum Biol. 2014;41(2):159-67.

20. Forjaz CLM, Bartholomeu T, Rezende JAS, Oliveira JA, Basso L, Tani G, et al. Genetic and environmental influences on blood pressure and physical activity: a study of nuclear families from Muzambinho, Brazil. Braz J Med Biol Res.2012;45:1269-75.

40. Beunen G, Thomis M. Gene powered? Where to go from heritability (h2) in muscle strength and power? Exerc Sport Sci Rev. 2004;32(4):148-54.

21. IBGE - Instituto Brasileiro de Geografia e Estatística. Ministério da Educação, Instituto Nacional de Estudos e Pesquisas Educacionais (INEP) – Censo da Educação Superior. 2010. Arch Endocrinol Metab. 2019;63/3

41. Barbosa JP, Basso L, Seabra A, Prista A, Tani G, Maia JA, et al. Relationship between physical activity, physical fitness and multiple metabolic risk in youths from Muzambinho’s study Eur J Sport Sci. 2016;16(5):618-23.

221

3. Kokkinos P. Physical Activity, Health Benefits, and Mortality Risk. ISRN Cardiol. 2012;2012:718789.

original article

The differences in homeostasis model assessment values in type 2 diabetic patients with different lengths of history of diabetes Endocrinology Department, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China 2 Department of General Surgery, People’s Hospital of Haiyang, Yantai, Shandong, China 3 Department of Gastroenterology, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China 4 Laboratory of Thyroid Disease, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China 1

* Chen Wang and Zaibo Liu contributed equally to this work. Correspondence to: Yangang Wang Endocrinology Department, The Affiliated Hospital of Qingdao University 16 Jiangsu Road 266003 – Qingdao, Shandong, China qdyxylxy@126.com Received on Jan/6/2018 Accepted on Mar/17/2019 DOI: 10.20945/2359-3997000000134

Chen Wang1*, Zaibo Liu2*, Peng Zhang3, Xiaolong Ma1, Kui Che4, Yangang Wang1

ABSTRACT Objective: Type 2 diabetes (T2DM) is characterized by the progressive deterioration of pancreatic islet β-cell function over time and insulin resistance. Knowing more about the differences in pancreatic islet function in T2DM patients who have had diabetes for different lengths of time can help improve therapy for T2DM. Subjects and methods: We conducted a cross-sectional study to compare islet β-cell function and insulin resistance in T2DM patients (n = 3,254) who had had diabetes for different lengths of time and those in normal controls (n = 794) using ANOVA and LSD analysis. Results: We found that compared with that in normal controls, HOMA-β in T2DM patients with a history of diabetes of less than 1 year was lower (approximately 52% of that of normal controls, p = 0.003), while HOMA-IR in these patients was higher (approximately 50% of that of normal controls, p = 0.007). Compared with that in other diabetic patients, HOMA-β in patients with a history of diabetes of more than 30 years was the lowest. HOMA-IR in patients with a history of diabetes of between 20 and 30 years was lower than that in other diabetic patients (p < 0.05). Conclusions: There were obvious decreases in HOMA-β and increases in HOMA-IR in T2DM patients with a history of diabetes of less than 1 year compared with those in normal controls. Therefore, early screening and intervention for T2DM might help improve islet function and delay the progression of diabetes. Arch Endocrinol Metab. 2019;63(3):222-7 Keywords Type 2 diabetes; islet function; insulin resistance; history of disease

INTRODUCTION

ype 2 diabetes (T2DM) is a major global health issue with an incidence of more than 9.7% in China (1). According to a report from the American Diabetes Association (ADA, Philadelphia, PA, USA), the total number of Americans living with diabetes will increase 64% by 2025 (2). Moreover, diabetic complications, such as kidney failure and cardiovascular diseases, can decrease the quality of life and limit the regular activity and productivity of individuals with T2DM (3). T2DM is characterized by the progressive deterioration of islet β-cell function over time and insulin resistance. Initially, islet β cells increase insulin secretion to meet insulin demand, but gradually they cannot compensate for insulin resistance, and 222

hypoinsulinism and hyperglycemia appear (4). Autopsy studies demonstrated a 63% reduction in the islet mass of T2DM patients compared to the islet mass of matched normal controls (5). Subjects with impaired fasting glucose (IFG) were shown to have a reduction of 40% of relative β-cell volume, suggesting that loss of β-cell mass was present at early stages of T2DM (5). It was also reported that β-cell mass was already reduced by 50% at the time of diagnosis of T2DM and that it continued to decline throughout the course of T2DM (6). The reduced β-cell mass was not due to reduced formation of islets or regeneration but was caused by increased rates of apoptosis in islets. The decline in β-cell function in T2DM drives the progressive deterioration of glycemic control and increases the requirement for insulin (7). Arch Endocrinol Metab. 2019;63/3

Islet function and T2DM

However, there have been few studies on the decline of islet β-cell function in T2DM patients with a long history of the disease. In the present study, we conducted a cross-sectional study to compare islet ‘β-cell function and insulin resistance in T2DM patients who had had the disease for different lengths of time and to observe the patterns of insulin secretion and insulin resistance as the length of history of having diabetes changed.

SUBJECTS AND METHODS Study design The study was an open, single-center, cross-sectional study. Participants were involved according to the time of diagnosis of T2DM.

homeostasis model assessment for insulin resistance (HOMA-IR). The formulas were as follows: HOMA-β = 20 * fasting C peptide / (FPG - 3.5) and HOMA-IR = fasting C peptide * FPG / 22.5 (8,9).

Statistical analysis All data were analyzed by Empowerstats statistical software, and the results are presented as x ± S. The differences between groups were analyzed with ANOVA. Multiple comparisons were performed by LSD analysis. A two-sided p < 0.05 was considered statistically significant.

RESULTS The characteristics of the patients

The study protocol was approved by the Ethical Committee of the Affiliated Hospital of Qingdao University. Informed consent in accordance with the Declaration of Helsinki was provided by every patient. T2DM patients who were in the Department of Endocrinology at the Affiliated Hospital of Qingdao University from 2013-2016 were enrolled in the present study. Patients with T1DM, ketoacidosis, hyperosmolar state, and severe acute disease were excluded. According to their medical history, patients were divided into 5 groups: group I (history of T2DM of less than 1 year), group II (history of T2DM between 1 and 10 years), group III (history of T2DM between 10 and 20 years), group IV (history of T2DM between 20 and 30 years), and group V (history of T2DM of more than 30 years). Healthy people with normal glucose who were ageand sex-matched were used as normal controls.

Examinations All medical histories were recorded by two doctors. Height, weight, waist and hip circumference, body mass index (BMI), blood glucose, fasting C peptide, liver function, and hemoglobin were measured. Blood glucose, fasting C peptide, liver function and hemoglobin were measured by hexokinase and chemiluminescence methods. Islet function was evaluated with the modified homeostasis model assessment for β-cell function (HOMA-β), which was calculated using fasting C peptide level (pmol/L) and fasting blood glucose (FPG, mmol/L), and insulin resistance (IR) was evaluated using the modified Arch Endocrinol Metab. 2019;63/3

A total of 3254 patients with T2DM and 794 normal controls were enrolled in this study. The clinical characteristics of all participants are shown in Table 1.

The differences in levels of fasting C peptide, HOMA-β and HOMA-IR in patients with different histories of T2DM As shown in Table 2, HOMA-β in T2DM patients with a history of diabetes of less than 1 year was approximately 52% of that of normal controls (4973.8 vs 9553.7, p = 0.003), although there was no difference in the levels of fasting C peptide between T2DM patients with a history of diabetes of less than 1 year and that of normal controls. As shown in Figure 1, HOMA-β decreased with the development of T2DM. For T2DM patients with a history of diabetes of more than 30 years, there was only approximately 32% of pancreatic islet function left compared to the function in normal controls (3067.0 vs 9533.7, as shown in Table 2). Compared with that in other diabetic patients, HOMA-β in patients with a history of more than 30 years of diabetes was the lowest. HOMA-IR in T2DM patients with a history of diabetes of less than 1 year was higher (approximately 50%) compared to that of normal controls (283.0 vs 185.8, p = 0.007). HOMA-IR in patients with a history of between 20-30 years of diabetes was lower than that of other diabetic patients (p < 0.05). These data were analyzed using Empowerstats statistical software for overall HOMA-β and HOMA-IR, correlation of HOMA-β and HOMAIR with the history of diabetes, and HOMA-β and HOMA-IR in each group as shown in Figure 1. 223

Participants

Islet function and T2DM

Table 1. The clinical characteristics of all participants Normal controls

group I

group II

group III

group IV

group V

794

568

924

978

532

252

58.4 ± 11.6

55.8 ± 13.2

56.4 ± 12.4

58.7 ± 15.3

59.8 ± 15.2

61.3 ± 13.4

–

0.5 ± 0.2

5.0 ± 2.4

13.2 ± 2.9

22.1 ± 2.9

31.1 ± 2.5

Weight (kg)

73.1 ± 12.9

73.8 ± 11.6

73.2 ± 12.4

72.3 ± 12.3

72.1 ± 11.6

69.3 ± 13.0

BMI (kg/m2)

27.0 ± 3.3

26.1 ± 4.1

26.0 ± 3.7

26.5 ± 4.7

26.6 ± 3.7

25.7 ± 3.4

SBP (mmHg)

138.9 ± 80.6

132.0 ± 17.4

137.1 ± 19.4

140.0 ± 19.4

142.3 ± 20.0

142.3 ± 19.8

DBP (mmHg)

87.6 ± 12.7

81.6 ± 10.4

82.5 ± 11.3

80.4 ± 11.0

77.1 ± 11.2

75.7 ± 9.6

HbA1c (%)

5.4 ± 0.4

9.7 ± 2.4*

8.6 ± 2.1*

8.7 ± 1.9*

8.9 ± 1.8*

8.9 ± 1.6*

FBG (mmol/l)

5.3 ± 0.4

7.8 ± 2.5*

8.0 ± 2.6*

7.9 ± 2.9*

7.8 ± 3.0*

7.9 ± 2.6*

Number Age History of T2DM

* p < 0.05, compared with normal controls. Levels of HbA1c and FBG in the diabetic group were higher than those in the normal controls. BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; HbA1c: glycated hemoglobin A1c; FBG: fasting blood glucose.

Table 2. The differences in levels of fasting C peptide, HOMA-β and HOMA-IR in patients with different lengths of history of T2DM Normal controls

group I

group II

group III

2.4 ± 0.8

2.3 ± 1.3

2.2 ± 1.2*

HOMA-β

9533.7 ± 4276.8

4973.8 ± 4144.9*

HOMA-IR

185.8 ± 64.8

283.0 ± 191.5*

Fasting C peptide (ng/mL)

group IV

group V

2.0 ± 1.1*

1.9 ± 1.5*

1.9 ± 1.2*#&

4268.7 ± 2851.8*

4096.4 ± 3155.7*

3826.2 ± 3289.6*

3067.0 ± 1434.6*#&

277.5 ± 251.5*

246.4 ± 193.4*

226.2 ± 212.6*&

278.0 ± 321.9*

* Compared with normal controls, P < 0.05; compared with T2DM patients with a history of disease of less than 1 year, p < 0.05; compared with T2DM patients with a history of disease between 10 and 20 years, p < 0.05. &

* *

*#&

* *

150

2000

Mean of HOMA-IR 200 250 300

Mean of HOMA-β 4000 6000 8000

10000

350

2 3 4 Groups according to the course of T2DM

Figure 1. Differences in the levels of HOMA-β and HOMA-IR as a function of length of history of T2DM. Compared with levels in normal controls, HOMA-β in T2DM patients with a history of diabetes of less than 1 year decreased approximately 48% (p = 0.003), while HOMA-IR increased approximately 50% (p = 0.007). Compared with levels in other diabetic patients, HOMA-β in patients with a history of diabetes of more than 30 years was the lowest. HOMA-IR in patients with a history of diabetes of between 20-30 years was lower than that in other diabetic patients. * Compared with normal controls, p < 0.01; # compared with T2DM patients with a history of diabetes of less than 1 year, P < 0.05; & compared with T2DM patients with a history of diabetes of between 10 and 20 years, p < 0.05.

Differences in the levels of fasting C peptide, HOMA-β and HOMA-IR in T2DM patients with different BMIs We further analyzed differences in the levels of fasting C peptide, HOMA-β and HOMA-IR in T2DM patients with different BMIs. A BMI = 25 kg/m2 was set as the cut-off point. As shown in Figure 2, in group II (history of diabetes between 1-10 years) and group III 224

(history of diabetes between 10-20 years), levels of fasting C peptide and HOMA-β in patients with BMI ≥ 25 kg/m2 were significantly higher than those in patients with BMI < 25 kg/m2 (p < 0.05). There were no significant differences in the levels of fasting C peptide and HOMA-β in patients with different BMIs in group I (history of diabetes < 1 year), group IV Arch Endocrinol Metab. 2019;63/3

Islet function and T2DM

(history of diabetes between 20-30 years) and group V (history of diabetes of more than 30 years). For all T2DM patients, regardless of the history of diabetes, the levels of HOMA-IR in patients with BMI ≥ 25 kg/m2 were significantly higher than those in patients with BMI < 25 kg/m2 (p < 0.05).

with a history of more than 30 years of diabetes was only approximately 32% of the pancreatic islet function left in normal controls. HOMA-IR in T2DM patients with a history of diabetes of less than 1 year was 1.5-fold that of normal controls, while HOMA-IR in patients with a history of diabetes of between 20-30 years was lower than that in other diabetic patients. It was accepted that hyperglycemia and lipotoxicity played roles in the possible mechanisms underlying impaired β-cell function (13,14). Exposure of β-cells to hyperglycemia might promote efflux of insulin secretory granules from the β-cell, leaving less insulin available for release in response to further hyperglycemia (15,16). In addition, it was suggested that increased glucose levels could activate the hexosamine pathway, resulting in excess generation of reactive oxygen species and the inhibition of insulin gene transcription and insulin secretion (17,18). Accumulated fatty acids and their metabolic products might have a negative effect on the conversion of proinsulin to insulin and decrease nitric oxide production, inhibiting glucose oxidation and inducing β-cell apoptosis (19-21).

DISCUSSION Pancreatic β-cells play a critical role in the development of T2DM. With the development of T2DM and an increase in age, both β-cell integrity and function declined and were accompanied by a decline in the ability of β-cells to be restored (10-12). In the present study, we investigated the difference in pancreatic islet secretion function in T2DM patients who have had the disease for different lengths of time by a cross-sectional study and the patterns of insulin secretion and insulin resistance as the length of time of having diabetes changed. We found that HOMA-β in T2DM patients with a history of diabetes of less than 1 year was approximately 52% that of normal controls, and HOMA-β in T2DM patients

Fasting C-pepide

HOMA-β 6000

ng/mL

2.5

5000

4000

1.5

3000

2000

0.5

1000

0 < 25 ≥ 25 < 25 ≥ 25 < 25 ≥ 25 < 25 ≥ 25 < 25 ≥ 25 < 1 year

1-10 years

10-20 years 20-30 years

< 25 ≥ 25 < 25 ≥ 25 < 25 ≥ 25 < 25 ≥ 25 < 25 ≥ 25

> 30 years

< 1 year

1-10 years

10-20 years 20-30 years

> 30 years

HOMA-IR *

< 25

≥ 25

< 1 year

< 25

≥ 25

1-10 years

< 25

≥ 25

10-20 years

< 25

≥ 25

20-30 years

< 25

≥ 25

> 30 years

Figure 2. Differences in the levels of fasting C peptide, HOMA-β and HOMA-IR in patients with different BMIs. In group II (history of diabetes between 1 and 10 years) and group III (history of diabetes of between 10 and 20 years), the levels of fasting C peptide and HOMA-β in patients with BMI ≥ 25 kg/m2 were significantly higher than those in patients with BMI < 25 kg/m2 (*p < 0.05). For all T2DM patients, the levels of HOMA-IR in patients with BMI ≥ 25 kg/m2 were significantly higher than those in patients with BMI < 25 kg/m2 (*p < 0.05). Arch Endocrinol Metab. 2019;63/3

225

400 350 300 250 200 150 100 50 0

Islet function and T2DM

Additionally, we found that in patients with a history of diabetes of between 1-20 years, the levels of fasting C peptide and HOMA-β in patients with a BMI ≥ 25 kg/m2 were significantly higher than those in patients with a BMI < 25 kg/m2. Levels of HOMA-IR in patients with BMI ≥ 25 kg/m2 were significantly higher than those in patients with BMI < 25 kg/m2. It was reported that β-cell mass decreased by approximately 40 and 65% in lean and obese individuals with T2DM, respectively, compared to the cell mass of age- and BMI-matched nondiabetic individuals. In approximately two-thirds of obese individuals who did not develop T2DM, pancreatic islet mass increased to compensate for increased insulin demand (22,23). Changes in β-cell mass might occur several years before hyperglycemia appears, as suggested by the United Kingdom Prospective Diabetes Study (UKPDS) (24). Reduction in β-cell mass might occur following β-cell apoptosis and reduced generation of new cells (25,26). Since a great proportion of patients with diabetes did not know about their disease for many years before the diagnosis, a possible limitation of the present study was the variable about disease history. A larger sample might reduce this bias. As T2DM patients with normal weight are quite different from those with BMI > 25 kg/m2, a further detailed investigation of islet function in T2DM patients with normal weight will be performed. In the present study, we found that compared with that in normal controls, HOMA-β in T2DM patients with a history of diabetes of less than 1 year was lower (approximately 52% that of normal controls), while HOMA-IR was higher (approximately 50% that of normal controls). With the development of diabetes, HOMA-β gradually decreased, while there was no significant difference in HOMA-IR over time. These results could allow us to better understand T2DM and its development, thereby individualizing treatment for T2DM patients with different lengths of history of diabetes according to their condition and islet function.

Acknowledgments: we appreciate all subjects in this study. Ethical approval: all procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Disclosure: no potential conflict of interest relevant to this article was reported. 226

REFERENCES 1. Zhang T, Ji L, Gao Y, Zhang P, Zhu D, Li X, et al. Observational Registry of Basal Insulin Treatment in Patients with Type 2 Diabetes in China: Safety and Hypoglycemia Predictors. Diabetes Technol Ther. 2017;19(11):675-84. 2. Denis GV, Sebastiani P, Andrieu G, Tran AH, Strissel KJ, Lombardi FL, et al. Relationships Among Obesity, Type 2 Diabetes, and Plasma Cytokines in African American Women. Obesity (Silver Spring). 2017;25(11):1916-20. 3. Grice BA, Nelson RG, Williams DE, Knowler WC, Mason C, Hanson RL, et al. Associations between persistent organic pollutants, type 2 diabetes, diabetic nephropathy and mortality. Occup Environ Med. 2017;74(7):521-7. 4. Chen C, Cohrs CM, Stertmann J, Bozsak R, Speier S. Human beta cell mass and function in diabetes: Recent advances in knowledge and technologies to understand disease pathogenesis. Mol Metab. 2017;6(9):943-57. 5. Ribbing J, Hamrén B, Svensson MK, Karlsson MO. A model for glucose, insulin, and beta-cell dynamics in subjects with insulin resistance and patients with type 2 diabetes. J Clin Pharmacol. 2010;50(8):861-72. 6. Murakami R, Saisho Y, Watanabe Y, Inaishi J, Tsuchiya T, Kou K, et al. Pancreas Fat and β Cell Mass in Humans With and Without Diabetes: An Analysis in the Japanese Population. J Clin Endocrinol Metab. 2017;102(9):3251-60. 7. Costes S, Boss M, Thomas AP, Matveyenko AV. Activation of Melatonin Signaling Promotes β-Cell Survival and Function. Mol Endocrinol. 2015;29(5):682-92. 8. Li M, Gu L, Yang J, Lou Q. Serum uric acid to creatinine ratio correlates with β-cell function in type 2 diabetes. Diabetes Metab Res Rev. 2018;34(5):e3001. 9. Ferjan S, Janez A, Jensterle M. DPP4 inhibitor sitagliptin as a potential treatment option in metformin- intolerant obese women with polycystic ovary syndrome: a pilot randomized study. Endocr Pract. 2018;24(1):69-77. 10. Liu S, Du F, Li X, Wang M, Duan R, Zhang J, et al. Effects and underlying mechanisms of irisin on the proliferation and apoptosis of pancreatic β cells. PLoS One. 2017;12(4):e0175498. 11. Tangvarasittichai S. Oxidative stress, insulin resistance, dyslipidemia and type 2 diabetes mellitus. World J Diabetes. 2015;6(3):456-80. 12. Rouse M, Younès A, Egan JM. Resveratrol and curcumin enhance pancreatic β-cell function by inhibiting phosphodiesterase activity. J Endocrinol. 2014;223(2):107-17. 13. Dai C, Kayton NS, Shostak A, Poffenberger G, Cyphert HA, Aramandla R, et al. Stress-impaired transcription factor expression and insulin secretion in transplanted human islets. J Clin Invest. 2016;126(5):1857-70. 14. Shimo N, Matsuoka TA, Miyatsuka T, Takebe S, Tochino Y, Takahara M, et al. Short-term selective alleviation of glucotoxicity and lipotoxicity ameliorates the suppressed expression of key β-cell factors under diabetic conditions. Biochem Biophys Res Commun. 2015;467(4):948-54. 15. Zephy D, Ahmad J. Type 2 diabetes mellitus: Role of melatonin and oxidative stress. Diabetes Metab Syndr. 2015;9(2):127-31. 16. Kramer CK, Zinman B, Choi H, Connelly PW, Retnakaran R. Chronic liraglutide therapy induces an enhanced endogenous glucagon-like peptide-1 secretory response in early type 2 diabetes. Diabetes Obes Metab. 2017;19(5):744-8. 17. Xiong X, Sun X, Wang Q, Qian X, Zhang Y, Pan X, et al. SIRT6 protects against palmitate-induced pancreatic β-cell dysfunction and apoptosis. J Endocrinol. 2016;231(2):159-65. Arch Endocrinol Metab. 2019;63/3

Islet function and T2DM

18. Proietto J. Mechanisms of insulin resistance caused by nutrient toxicity. Hepatol Res. 2005;33(2):87-91. 19. Vernier S, Chiu A, Schober J, Weber T, Nguyen P, Luer M, et al. β-cell metabolic alterations under chronic nutrient overload in rat and human islets. Islets. 2012;4(6):379-92. 20. Huang CN, Wang CJ, Lee YJ, Peng CH. Active subfractions of Abelmoschus esculentus substantially prevent free fatty acid-induced β cell apoptosis via inhibiting dipeptidyl peptidase-4. PLoS One. 2017;12(7):e0180285. 21. Mugabo Y, Zhao S, Seifried A, Gezzar S, Al-Mass A, Zhang D, et al. Identification of a mammalian glycerol-3-phosphate phosphatase: Role in metabolism and signaling in pancreatic β-cells and hepatocytes. Proc Natl Acad Sci U S A. 2016;113(4):E430-9.

24. Lidegaard LP, Hansen AL, Johansen NB, Witte DR, Brage S, Lauritzen T, et al. Physical activity energy expenditure vs cardiorespiratory fitness level in impaired glucose metabolism. Diabetologia. 2015;58(12):2709-17. 25. den Biggelaar LJ, Sep SJ, Eussen SJ, Mari A, Ferrannini E, van Greevenbroek MM, et al. Discriminatory ability of simple OGTTbased beta cell function indices for prediction of prediabetes and type 2 diabetes: the CODAM study. Diabetologia. 2017;60(3):432-41. 26. Courtade JA, Wang EY, Yen P, Dai DL, Soukhatcheva G, Orban PC, et al. Loss of prohormone convertase 2 promotes beta cell dysfunction in a rodent transplant model expressing human pro-islet amyloid polypeptide. Diabetologia. 2017;60(3):453-63.

22. Huang XT, Yue SJ, Li C, Huang YH, Cheng QM, Li XH, et al. A Sustained Activation of Pancreatic NMDARs Is a Novel Factor of β-Cell Apoptosis and Dysfunction. Endocrinology. 2017;158(11):3900-13.

23. Templin AT, Samarasekera T, Meier DT, Hogan MF, Mellati M, Crow MT, et al. Apoptosis Repressor With Caspase Recruitment Domain Ameliorates Amyloid-Induced β-Cell Apoptosisand JNK Pathway Activation. Diabetes. 2017;66(10):2636-45.

Arch Endocrinol Metab. 2019;63/3

227

original article

An early stage in T4-induced hyperthyroidism is related to systemic oxidative stress but does not influence the pentose cycle in erythrocytes and systemic inflammatory status Rayane Brinck Teixeira1, Tânia Regina Gattelli Fernandes-Piedras1, Adriane Belló-Klein1, Cristina Campos Carraro1, Alex Sander da Rosa Araujo1

ABSTRACT Laboratório de Fisiologia Cardiovascular, Departamento de Fisiologia, Instituto de Ciências Básicas da Saúde, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brasil 1

Correspondence to: Alex Sander da Rosa Araujo Laboratório de Fisiologia Cardiovascular, Departamento de Fisiologia, Instituto de Ciências Básicas da Saúde, Universidade Federal do Rio Grande do Sul Rua Sarmento Leite, 500, sala 344 90050170 – Porto Alegre, RS, Brasil alex.rosa@ufrgs.br Received on May/8/2018 Accepted on July/8/2018 DOI: 10.20945/2359-3997000000128

Objective: Hyperthyroidism causes many injuries in its target organs and the consequences are reflected systemically. As systemic alterations in hyperthyroidism at earlier stages have received partial attention, this study aimed to investigate systemic redox and inflammatory status at an early stage of T4-induced hyperthyroidism. Materials and methods: Male Wistar rats were assigned to control and hyperthyroid groups (n = 7/group). The hyperthyroid group received L-thyroxine (12 mg/L) in their drinking water for 14 days whereas control group received only the vehicle. Body weight was measured on the 1st and 14th day of the protocol. On the 14th day, animals were anaesthetized. Blood was then collected from the retro-orbital venous plexus and then the animals were euthanised. The blood was separated into plasma and erythrocytes. Plasma was used to measure ROS levels, sulfhydryl compounds, IL-10, TNF-α and LDH levels; erythrocytes were used for the analysis of thioredoxin reductase activity, glutaredoxin content, and pentose cycle enzymes (total G6PD, G6PD and 6PGD). Results: Hyperthyroid animals presented body weight gain and final body weight reduction, which was associated with increased ROS levels and decreased sulfhydryl content in plasma. Thioredoxin reductase activity, glutaredoxin content, and pentose cycle enzymes levels in erythrocytes, as well as IL-10, TNF-α and LDH plasma levels were unaltered. Conclusion: Taken together, our results suggest an impairment in corporal mass associated with systemic oxidative stress at this stage of hyperthyroidism. Meanwhile, the pentose cycle was not influenced and systemic inflammation and tissue damage seem to be absent at this stage of hyperthyroidism. Arch Endocrinol Metab. 2019;63(3):228-34 Keywords Hyperthyroidism; reactive oxygen species; sulfhydryl compounds; glucose-6-phosphate dehydrogenase; inflammatory cytokines

INTRODUCTION

yperthyroidism is characterised by reduced thyroidstimulating hormone (TSH) concentration associated with increased thyroid hormones (thyroxine – T4 and/or triiodothyronine – T3) levels, elevated heart rate, and cardiac hypertrophy (1,2). Although many studies have focused on advanced phases, such alterations can be found in even earlier moments in hyperthyroidism. Basset and cols. characterised the thyroid hormone-dependent cardiac hypertrophy at 15 days of hyperthyroidism induction (1). In a similar period at approximately 14 days of T4 administration, Fernandes and cols. showed an association among cardiac hypertrophy and pro-apoptotic signalling in the heart (3). 228

On the other hand, increased oxidative stress has been demonstrated in different organs, such as the heart, pancreas and liver, at later stages of thyrotoxicosis (2,4,5). Besides the classical markers of redox imbalance, such as lipid peroxidation and protein oxidation, lactate dehydrogenase (LDH) is noted as an important indicator of tissue injury. In addition, LDH seems to increase in oxidised environments (4). Nevertheless, the oxidative stress may provoke an antioxidant response in order to react to eustress state disruption (5). In this sense, thioredoxin and glutaredoxin, important antioxidant proteins, appear increased in stressful situations, such as Grave’s disease and experimental hyperthyroidism (6,7). The oxidised thiol residues of these antioxidant proteins Arch Endocrinol Metab. 2019;63/3

Systemic effects in early hyperthyroidism

MATERIALS AND METHODS Ethical principles This study was approved by the Ethics Committee for use of animals at the Universidade Federal do Rio Grande do Sul (protocol number 24504). All procedures in this study were conducted according to the ethical principles established by the Universidade Federal do Rio Grande do Sul and the International Principles Guiding for Biomedical research involving animals from the Council for International Organizations of Medical Science. Arch Endocrinol Metab. 2019;63/3

Experimental protocol Male Wistar rats (166 ± 22 g, 34 days old) were obtained from the central animal house at the Universidade Federal do Rio Grande do Sul. Animals were allocated into plastic cages (four animals each) and received water and pelleted food ad libitum. They were maintained under standard laboratory conditions (controlled temperature of 21°C, 12 hours light/dark cycle) and were randomly divided into control (n = 7) and hyperthyroid groups (n = 7). Hyperthyroid group animals were submitted to hyperthyroidism induction by administering 0.15 µM L-thyroxine (T4) (12 mg dissolved in 300 µL of 1N NaOH and diluted in the drinking water, final pH = 7, 1 L of final volume) over 14 days, whereas the control group received only the vehicle (300 µL of 1N NaOH diluted in the drinking water, final pH = 7, 1 L of final volume), as previously described in the literature (14). The drinking water of both hyperthyroid and control groups was changed two times per week. The dose of L-thyroxine and the period of treatment were selected based on previous studies (1,3). Based on a previous study, the estimated daily water intake is 35 ml for each animal, which provides a daily dose of approximately 420 µg of L-thyroxine and increases T4 plasma levels by 4-fold after 14 days of treatment (3). On the 14th day, rats of both the control and the hyperthyroid groups were weighed and anesthetised (ketamine 90 mg/kg and xylazine 10 mg/kg, intraperitoneally). Then 1 ml of blood sample was collected from the retro-orbital venous plexus in heparinised tubes. Each animal was immediately euthanised after blood collection. All the animals were euthanised at the morning. Blood was further centrifuged for 15 minutes at 2415 × g in a refrigerated centrifuge at 4°C (Sorvall RC 5B – Rotor SM 24). Plasma (supernatant) was aliquoted and stored at -80°C for posterior analysis of systemic oxidative stress parameters (total reactive oxygen species [ROS] and sulfhydryl compounds), inflammatory cytokines (plasma levels of IL-10 and TNF-α) and tissue damage (LDH). The erythrocytes present in the pellet were washed through the addition of saline and centrifugation for 3 minutes at 24640 × g. After centrifugation, the supernatant was discarded and the procedure was repeated two more times. Then, the erythrocytes were prepared with an enzyme solution (acetic acid 10 µmol/L and MgCl2 4 mmol/L, 1:50 v./v. of 229

may be recycled through thioredoxin reductase, which reduces the sulfhydryl groups and restores antioxidant protection. In this context, dihydronicotinamide-adenine dinucleotide phosphate (NADPH) is pivotal so that thioredoxin reductase can play on enzymatic function. The NADPH is generated by glucose-6-phosphate dehydrogenase (G6PD) and 6-phosphogluconate dehydrogenase (6PGD) in the pentose phosphate pathway (8). However, the pentose phosphate pathway may be inhibited by negative modulators, such as tumour necrosis factor-alpha (TNF-α) (9). Data in the literature have demonstrated that cellular exposition to TNF-α was previously related to apoptosis and inflammation through the nuclear factor kappa B (NFkB) pathway and caspase-3 activation (10). However, other anti-inflammatory cytokines, such as interleukin 10 (IL-10), may mitigate the inflammation and apoptosis induced by TNF-α (10,11). In this point of view, the IL-10/TNF-α ratio has been used to verify the inflammatory status in different models of disease (11-13). In this context, there is a paucity of information about the role of thyroid hormones on the inflammatory process in earlier stages of hyperthyroidism. Thus far, the majority of experimental studies investigated advanced stages of hyperthyroidism, and there are not many studies investigating the impact of elevated thyroid hormone levels at earlier stages of the disease. Furthermore, systemic alterations have received reduced attention, and there are few studies describing earlier alterations in plasma and erythrocytes in hyperthyroidism. In this sense, the aim in this study was to investigate the presence of redox and inflammatory alterations at 14 days of the experimental model of T4induced hyperthyroidism.

Systemic effects in early hyperthyroidism

erythrocytes and enzyme solution, respectively), and aliquots were stored at -80°C for posterior analysis of systemic oxidative stress measurements (glutaredoxin content, thioredoxin reductase, G6PD and 6PGD activities). In order to provide reliable data, the protocol described above was repeated twice, obtaining three independent experiments in total, and all results of the present study were obtained from a representative sample of these three independent experiments.

Oxidative stress measurements

Evaluation of the total concentration of ROS by oxidation of 2, 7-dichlorofluorescein (DCFH) This method of oxidative stress measurement is based on the measurement of the fluorescence produced by the oxidation of DCFH by the ROS present in the plasma sample. Results were expressed as pmol DCF formed/ mg protein (15).

Determination of the content of sulfhydryl compounds Free sulfhydryl compounds in the plasma were assayed by spectrophotometry using the method described by Ellman and modified by Hu and cols. (16,17). The method is based on the reaction of 5,5-dithio-bis(2nitrobenzoic acid) (DTNB) with thiol groups, resulting in the release of acidic 5-thio-2-nitrobenzoic acid (TNB). Results were expressed as µmol/mg protein.

Measurement of thioredoxin reductase activity The activity of thioredoxin reductase was measured in erythrocyte samples by the reduction of DTNB to TNB, which was detected spectrophotometrically at 412 nm. Data were expressed in nmol/min/mg protein (18).

Evaluation of glutaredoxin content To assess glutaredoxin content in erythrocyte samples, the protocol established by Holmgren & A%slund (19) was utilised. The results were expressed in mmol/mg protein.

Activity evaluation of enzymes G6PD and 6PGD G6PD and 6PGD enzymes determine the production of NADPH, an indicator of redox potential, in order to control the glucose metabolism via the pentose phosphate pathway. Measurement of G6PD was performed in erythrocytes using the method described 230

by Leong & Clark (20). The total activity was expressed in mmol/mg protein. The specific activity of each enzyme was represented as units per mg of protein (U/mg).

Determination of protein content Oxidative stress results were normalised by the amount of protein in plasma and erythrocytes, which was quantified by the method of Lowry and cols. (21). Bovine albumin was used as a stock solution in the concentration of 1 mg/ml and diluted in order to obtain a standard curve of albumin.

Quantification of inflammatory proteins by ELISA and tissue damage by LDH activity in plasma The quantification of inflammatory proteins in plasma was performed by enzyme-linked immunosorbent assay (ELISA) using IgG ELISA kits for IL-10 (rat IL-10) and TNF-α (rat TNF-α) from Invitrogen Corporation (Camarillo, CA, USA). The results were expressed as pg/mL. The activity of LDH was measured using a kinetic test with ultraviolet reading based on the method by Wacker & Snodgrass and Amador and cols., which used a Bioliquid assay kit from Laborclin company (Pinhais, PR, Brazil) for plasma samples (22,23). The results were expressed as units of LDH per litre (U/L).

Statistical analysis Results are expressed as the mean ± standard deviation. The distribution of data was analysed using the Shapiro-Wilk normality test. Parametric data were tested by Student’s t-test while nonparametric results were tested by independent samples Mann-Whitney U test for comparison between groups. The significance level assumed was 5% (p < 0.05). All calculations were performed using SPSS (Statistical Package for Social Sciences, IBM, New York, USA) version 20.0. Graphs were created using GraphPad Prism version 5.00 for Windows (GraphPad Software, San Diego, California, USA).

RESULTS Body weight analysis Values of final body weight were 10% lower in the hyperthyroid group compared to the control group as shown in Figure 1A. In addition, the body weight gain Arch Endocrinol Metab. 2019;63/3

Systemic effects in early hyperthyroidism

from the 1st to the 14th day of the experimental protocol was reduced by 22% in the hyperthyroid group as compared to control (Figure 1B).

The activity of thioredoxin reductase, glutaredoxin content, G6PD total activity, isolated activity of G6PD, and 6PGD activity in erythrocytes was similar between groups (Table 1).

Analysis of systemic oxidative stress Results of total ROS in plasma are presented in Figure 2A. The analysis demonstrated increased ROS levels in the hyperthyroid group (36%) compared to the control group. On the other hand, hyperthyroid rats presented reduced levels of sulfhydryl compounds in plasma (39%) as compared to control group (Figure 2B).

Quantification of inflammatory proteins by ELISA and tissue damage by LDH activity in plasma There was no significant difference between groups in terms of IL-10 (Figure 3A), TNF-α levels (Figure 3B), the IL-10/TNF-α ratio (Figure 3C) and LDH levels (Figure 3D).

300

150 Body weight gain (g)

Body weight (g)

* 200

100

0 Control

Control

Figure 1. Final body weight (A) and body weight gain (B) values of the control and T4 groups. Values represent mean ± standard deviation (n = 7 animals per group). Data were analysed using the Student’s t-test. *p < 0.05 compared with control group.

B 6

2.5 Sulfhydryl compounds (µmol/mg protein)

Total ROS (pmol/mg protein)

* 4

Control

2.0 1.5

* 1.0 0.5 0.0

Control

Figure 2. Analysis of total ROS (A) and sulfhydryl compounds levels (B) in the plasma of control and T4 groups. Values represent mean ± standard deviation (n = 7 animals per group). Data were analysed using the Student’s t-test. *p < 0.05 compared with control group.

Measurement

Control

P value

74.19 (49.26; 119.10)

69.27 (60.59; 133.72)

0.902

0.55 (0.43; 0.94)

0.61 (0.49; 1.13)

0.710

G6PD total activity (mmol/mg protein)

4.08 ± 1.78

4.49 ± 2.99

0.762

G6PD activity (U/mg protein)

2.80 ± 1.53

3.23 ± 2.15

0.671

6PGD activity (U/mg protein)

1.28 ± 0.52

1.26 ± 0.85

0.943

Thioredoxin reductase activity (nmol/min/mg protein) Glutaredoxin content (mmol/mg protein)

Parametric data were analysed using the Student’s t-test and expressed as mean ± standard deviation. Nonparametric results were tested using the Mann-Whitney U test and expressed as median and interquartiles (n = 7 animals per group). Arch Endocrinol Metab. 2019;63/3

231

Table 1. Antioxidant and pentose cycle components measurement

Systemic effects in early hyperthyroidism

B 80

300

60 TNF-α (pg/mL)

IL-10 (pg/mL)

A 400

200 100 0

Control

D 250

200

150

LDH (U/L)

IL-10/ TNF-α ratio

100

Control

Figure 3. Evaluation of IL-10 levels (A), TNF-α levels (B), IL-10/TNF-α ratio (C), and LDH activity (D) in the plasma of control and T4 group. Values represent mean ± standard deviation (n = 5 -7 animals per group). Data were analysed using the Student’s t-test.

DISCUSSION The major attribute of this study was to demonstrate that systemic oxidative stress plays a more relevant role on hyperthyroidism impairments at this stage (14 days of T4-induction) as compared to the global inflammatory process. The changes in ROS and sulfhydryl compounds in plasma as well as body weight gain alterations suggest not only a systemic distress state, but also reduced body development in hyperthyroid rats. Alterations in inflammatory and tissue damagerelated parameters were not found in our study. The results of body mass reinforce previous studies, which demonstrated a decrease in body weight and body weight gain after 14 days of hyperthyroidism (3,24). This characteristic is a typical consequence in hyperthyroidism, highly related to increases of T4 hormone levels in plasma. Although we didn’t measure T4 levels, a previous study showed 4-fold increase in T4 levels at 14 days of T4 induced hyperthyroidism (3). In this sense, high T4 levels in hyperthyroidism induce a catabolic condition in which there is increased energy expenditure allied to lipolysis and protein degradation (25,26). As a result of weight loss, the 232

hyperthyroidism-imposed inflammatory response may be refractory (27). Indeed, our results demonstrated that plasma levels of IL-10 and TNF-α, the IL-10/TNF-α ratio and LDH activity remained unaltered at 14 days. This diminished inflammatory reaction may be associated with the reduced mass gain in the hyperthyroid group (28). The adipose tissue can produce proinflammatory cytokines, such as TNF-α and IL-6, and activate macrophages therefore establishing an inflammatory environment. Nevertheless, hyperthyroidism provokes lipolysis, reducing adipose tissue and, consequently, downregulates the production of proinflammatory cytokines. Therefore, even in stressful conditions, which are imposed in a state of hyperthyroidism, the weight loss and reduced lipid tissue may attenuate the increase of plasmatic inflammatory protein. On the other hand, the T4-induced higher demand for energy may augment the mitochondrial ROS formation, which contributes to oxidative stress (29). Indeed, in this experimental model, elevated systemic ROS concentrations were associated with administration of T4 over 14 days. The increased ROS Arch Endocrinol Metab. 2019;63/3

Systemic effects in early hyperthyroidism

Arch Endocrinol Metab. 2019;63/3

by thyrotoxicosis. This study offered some additional data about the systemic status in the hyperthyroidism, indicating that systemic evaluations could be useful as a tool to monitor the impact of hyperthyroidism on the cardiac remodeling. The lack of data at this time makes it difficult to discuss the findings in this study and reinforced that further studies should be conducted regarding this stage of hyperthyroidism. Acknowledgments: veterinary medical support from Dr. André Ricardo Ribeiro Belló. This paper was written during a scholarship supported by the International Cooperation Program CAPES/DFATD at the University of Manitoba, Winnipeg, Canada. Funding: this study was funded by the National Council of Technological and Scientific Development (CNPq), Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (FAPERGS), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Basset A, Blanc J, Messas E, Hagège A, Elghozi JL. Reninangiotensin system contribution to cardiac hypertrophy in experimental hyperthyroidism: an echocardiographic study. J Cardiovasc Pharmacol. 2001;37:163-72. 2. Messarah M, Saoudi M, Boumendjel A, Boulakoud MS, Feki AE. Oxidative stress induced by thyroid dysfunction in rat erythrocytes and heart. Environ Toxicol Pharmacol. 2011;31:33-41. 3. Fernandes RO, Dreher GJ, Schenkel PC, Fernandes TR, Ribeiro MF, Araujo AS, et al. Redox status and pro-survival/pro-apoptotic protein expression in the early cardiac hypertrophy induced by experimental hyperthyroidism. Cell Biochem Funct. 2011;29: 617-23. 4. Messarah M, Boumendjel A, Chouabia A, Klibet F, Abdennour C, Boulakoud MS, et al. Influence of thyroid dysfunction on liver lipid peroxidation and antioxidant status in experimental rats. Exp Toxicol Pathol. 2010;62:301-10. 5. Sajadian M, Hashemi M, Salimi S, Nakhaee A. The Effect of Experimental Thyroid Dysfunction on Markers of Oxidative Stress in Rat Pancreas. Drug Dev Res. 2016;77:199-205. 6. Kihara M, Kontani K, Yamauchi A, Miyauchi A, Nakamura H, Yodoi J, et al. Expression of thioredoxin in patients with Graves’ disease. Int J Mol Med. 2005;15:795-9. 7. Araujo AS, Fernandes T, Ribeiro MF, Khaper N, Belló-Klein A. Redox regulation of myocardial ERK 1/2 phosphorylation in experimental hyperthyroidism: role of thioredoxin-peroxiredoxin system. J Cardiovasc Pharmacol. 2010;56:513-7. 8. Stanton RC. Glucose-6-phosphate dehydrogenase, NADPH, and cell survival. IUBMB Life. 2012;64:362-9. 9. Zhang HS, Wang SQ. Nrf2 is involved in the effect of tanshinone IIA on intracellular redox status in human aortic smooth muscle cells. Biochem Pharmacol. 2007;73:1358-66. 10. Dhingra S, Sharma AK, Arora RC, Slezak J, Singal PK. IL-10 attenuates TNF-alpha induced NF kappaB pathway activation and cardiomyocyte apoptosis. Cardiovasc Res. 2009;82:59-66.

233

levels in the plasma of the hyperthyroid group could accrue from other organs, representing a systemic response to the disruption of tissue redox homeostasis. This hypothesis is reinforced by the cardiac oxidative status. At one time in this same early period, increased levels of oxygen peroxide were detected and associated with cardiac impaired redox status (3). On the other hand, in hyperthyroidism the elevation of oxidants is associated with an impairment in the capacity of antioxidant defences (2). In this context, the assessment of total sulfhydryl groups in the plasma provides a wide vision of the thioldependent defences, such as glutathione, and cysteine, against free radicals. The low levels of these compounds, as determined in our study, suggest impairment in the antioxidant response (30). No previous study was located that utilised a similar experimental protocol where the sulfhydryl content was investigated in plasma at 14 days. However, previous studies reported decreased sulfhydryl content in patients with advanced stages of hyperthyroidism (31,32). Taken together, increased ROS and decreased sulfhydryl levels in plasma indicate distress status induced by hyperthyroidism at this period. The oxidative stress associated with depletion in sulfhydryl content could also reflect a decrement of reducing agents, such as NADPH. This coenzyme is produced by the G6PD enzyme, a component of the pentose cycle whose activity is decreased in hyperthyroid patients (33,34). However, this enzymatic parameter in the erythrocytes sample was similar between groups in our study, indicating no alteration in NADPH formation by the pentose cycle at 14 days. In this scenario, it is possible to suggest that the absence of red cell pentose cycle activity elevation in an oxidised environment could compromise the recycling of sulfhydryl groups, since NADPH levels did not increase in parallel to the elevation of ROS concentrations. In this context, the oxidative stress could predominate systemically. The oxidation of thiol groups may be detrimental to the antioxidant function of thioredoxin reductase and glutaredoxin. In fact, our results showed that thioredoxin and glutaredoxin seem to be unaltered at this stage. However, as previously reported, these proteins could be altered at later stages of hyperthyroidism (2,7,35). In the earlier stages of hyperthyroidism, the oxidative stress seems to be the preponderant factor, inclining tissues to the adverse adaptation imposed

Systemic effects in early hyperthyroidism

11. Kaur K, Sharma AK, Dhingra S, Singal PK. Interplay of TNF-alpha and IL-10 in regulating oxidative stress in isolated adult cardiac myocytes. J Mol Cell Cardiol. 2006;41:1023-30. 12. Batista ML Jr, Rosa JC, Lopes RD, Lira FS, Martins EJr., Yamashita AS, et al. Exercise training changes IL-10/TNF-alpha ratio in the skeletal muscle of post-MI rats. Cytokine. 2010;49:102-8. 13. Dopheide JF, Knopf P, Zeller GC, Vosseler M, Abegunewardene N, Münzel T, et al. Low IL-10/TNFα ratio in patients with coronary artery disease and reduced left ventricular ejection fraction with a poor prognosis after 10 years. Inflammation. 2015;38:911-22. 14. Ladenson PW, Kieffer JD, Farwell AP, Ridgway EC. Modulation of myocardial L-triiodothyronine receptors in normal, hypothyroid, and hyperthyroid rats. Metabolism. 1986;35:5-12. 15. LeBel CP, Ali SF, Bondy SC. Deferoxamine inhibits methyl mercury-induced increases in reactive oxygen species formation in rat brain. Toxicol Appl Pharmacol. 1992;112:161-5. 16. Ellman GL. Tissue sulfhydryl groups. Arch Bioch Biophys. 1959;82:70-7. 17. Hu ML, Louie S, Cross CE, Motchnik P, Halliwell B. Antioxidant protection against hypochlorous acid in human plasma. J Lab Clin Med. 1993;121:257-62. 18. Holmgren A, Björnstedt M. Thioredoxin and thioredoxin reductase. Methods Enzymol. 1995;252:199-208. 19. Holmgren A, Åslund F. “Glutaredoxin”. Methods Enzymol. 1995;252:283-92. 20. Leong SF, Clark JB. Regional development of glutamate dehydrogenase in the rat brain. J Neurochem. 1984;43:106-11. 21. Lowry OH, Rosebrough AL, Farr AL. Randall R. Protein measurement with the folin phenol reagent. J Biol Chem. 1951;193:265-75. 22. Wacker WEC, Snodgrass PJ. Serum LDH Activity in Pulmonary Embolism Diagnosis. JAMA. 1960;174:2142-5. 23. Amador E, Dorfman LE, Wacker WEC. Serum Lactic Dehydrogenase Activity: An Analytical Assessment of Current Assays. Clin Chem. 1963;9:391-9.

25. López M, Alvarez CV, Nogueiras R, Diéguez C. Energy balance regulation by thyroid hormones at central level. Trends Mol Med. 2013;19:418-27.

24. Brzezin%ska-Slebodzin%ska E. Effects of triiodothyronine-induced hyperthyroidism on lipid peroxidation, erythrocyte resistance and iron-binding and iron-oxidizing antioxidant properties of plasma in the rabbit. Vet Res Commun. 2005;29:661-70.

35. Araujo AS, Seibel FE, Oliveira UO, Fernandes T, Llesuy S, Kucharski L, et al. Thyroid hormone-induced haemoglobin changes and antioxidant enzymes response in erythrocytes. Cell Biochem Funct. 2011;29:408-13.

26. Martínez-Sánchez N, Moreno-Navarrete JM, Contreras C, RialPensado E, Fernø J, Nogueiras R, et al. Thyroid hormones induce browning of white fat. J Endocrinol. 2017;232:351-62. 27. Aron-Wisnewsky J, Tordjman J, Poitou C, Darakhshan F, Hugol D, Basdevant A, et al. Human adipose tissue macrophages: m1 and m2 cell surface markers in subcutaneous and omental depots and after weight loss. Clin Endocrinol Metab. 2009;94:4619-22. 28. Kosteli A, Sugaru E, Haemmerle G, Martin JF, Lei J, Zechner R, et al. Weight loss and lipolysis promote a dynamic immune response in murine adipose tissue. J Clin Invest. 2010;120: 3466-79. 29. Venditti P, Di Meo S. Thyroid hormone-induced oxidative stress. Cell Mol Life Sci. 2006;63:414-34. 30. Lemarechal H, Allanore Y, Chenevier-Gobeaux C, Kahan A, Ekindjian OG, Borderie D. Serum protein oxidation in patients with rheumatoid arthritis and effects of infliximab therapy. Clin Chim Acta. 2006;372:147-53. 31. Ademog#lu E, Ozbey N, Erbil Y, Tanrikulu S, Barbaros U, Yanik BT, et al. Determination of oxidative stress in thyroid tissue and plasma of patients with Graves’ disease. Eur J Intern Med. 2006;17:545-50. 32. Andryskowski G, Owczarek T. The evaluation of selected oxidative stress parameters in patients with hyperthyroidism. Pol Arch Med Wewn. 2007;117:285-9. 33. Hecker PA, Leopold JA, Gupte SA, Recchia FA, Stanley WC. Impact of glucose-6-phosphate dehydrogenase deficiency on the pathophysiology of cardiovascular disease. Am J Physiol Heart Circ Physiol. 2013;304:H491-H500.

34. Odçikin E, Ozdemir H, Ciftçi M, Capog#lu I. Investigation of red blood cell carbonic anhydrase, glucose 6-phosphate dehydrogenase, hexokinase enzyme activities, and zinc concentration in patients with hyperthyroid diseases. Endocr Res. 2002;28:61-8.

234

Arch Endocrinol Metab. 2019;63/3

original article

Clinical and functional variables can predict general fatigue in patients with acromegaly: an explanatory model approach Programa de Pós-Graduação em Ciências da Reabilitação, Centro Universitário Augusto Motta, Rio de Janeiro, RJ, Brasil 2 Centro de Pesquisa em Neuroendocrinologia, Serviço de Endocrinologia, Faculdade de Medicina, Hospital Universitário Clementino Fraga Filho, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brasil 3 Divisão de Neuroendocrinologia, Instituto Estadual do Cérebro Paulo Niemeyer, Secretaria Estadual de Saúde do Rio de Janeiro, Rio de Janeiro, RJ, Brasil 4 Departamento de Fisioterapia, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brasil 1

André da Cunha Michalski1, Arthur de Sá Ferreira1, Leandro Kasuki2,3, Monica R. Gadelha2,3, Agnaldo José Lopes1, Fernando Silva Guimarães1,4

ABSTRACT Objective: To evaluate whether hormonal profile, arterial function, and physical capacity are predictors of fatigue in patients with acromegaly. Subjects and methods: This is a cross-sectional study including 23 patients. The subjects underwent a Modified Fatigue Impact Scale (MFIS) assessment; serum growth hormones (GH) and IGF-1 measurements; pulse wave analysis comprising pulse wave velocity (PWV), arterial compliance (AC), and the reflection index (IR1,2); dominant upper limb dynamometry (DYN); and the six-minute walking distance test (6MWT). Multiple linear regression models were used to identify predictors for MFIS. The coefficient of determination R2 was used to assess the quality of the models’ fit. The best model was further analyzed using a calibration plot and a limits of agreement (LOA) plot. Results: The mean ± SD values for the participants’ age, MFIS, PWV, AC, IR1,2, DYN, and the distance in the 6MWT were 49.4 ± 11.2 years, 31.2 ± 18.9 score, 10.19 ± 2.34 m/s, 1.08 ± 0.46 x106 cm5/din, 85.3 ± 29.7%, 33.9 ± 9.3 kgf, and 603.0 ± 106.1 m, respectively. The best predictive model (R2 = 0.378, R2 adjusted = 0.280, standard error = 16.1, and P = 0.026) comprised the following regression equation: MFIS = 48.85 - (7.913 × IGF-I) + (1.483 × AC) - (23.281 × DYN). Conclusion: Hormonal, vascular, and functional variables can predict general fatigue in patients with acromegaly. Arch Endocrinol Metab. 2019;63(3):235-40

INTRODUCTION

cromegaly is a chronic endocrine disease characterized by excessive production of growth hormones (GH) and insulin-like growth factor type I (IGF-I) (1). The insidious process and the difficulty for health professionals to recognize its clinical characteristics often delay diagnosis. Usually, patients are diagnosed 8 to 10 years after their first clinical manifestations, and the disease most commonly occurs in patients between 30 and 50 years of age. There is no significant propensity to sex (2). Clinical manifestations may progress gradually to facial and extremity deformities as well as musculoskeletal, cardiac, respiratory, and metabolic dysfunction and early death. In acromegaly, the normalization of GH and IGF-I levels is associated with clinical improvement (3-7). Although assessing fatigue in individuals with acromegaly is recommended (8), not using adequate Arch Endocrinol Metab. 2019;63/3

Fernando Silva Guimarães fguimaraes_pg@yahoo.com.br Received on May/26/2018 Accepted on Jan/14/2019 DOI: 10.20945/2359-3997000000127

instruments makes it difficult. Fatigue is a subjective symptom, typically defined as extreme and persistent tiredness or physical and mental exhaustion, with characteristics different from those observed in depression or muscle weakness (9,10). It is adequately measured by using a self-report questionnaire in which the individual describes and punctuates the symptoms. One of these instruments is the Modified Fatigue Impact Scale (MFIS), which facilitates the understanding of fatigue and its impact (9). Despite the physiopathology of general fatigue in acromegaly being unclear (11), its identification assists in diagnosis (8), as it is a common manifestation of the disease (12-15). In the study of Woodhouse and cols. (11), the authors speculated that muscle fiber type, muscle strength, physical capacity, and hormone levels might play a role in the fatigue complaints in patients with GH deficiency or excess GH. However, there are no conclusive studies that have identified the determining 235

Keywords Acromegaly; pulse wave analysis; fatigue; exercise tests; muscle strength

Correspondence to:

Fatigue in acromegaly

factors of fatigue in this population. We hypothesize that GH and IGF-I levels, peripheral arterial function, muscle strength, and functional capacity are associated with fatigue scores among this population. Therefore, this study aims to evaluate whether hormonal profile, arterial function, and physical capacity are predictors of fatigue in patients with acromegaly.

SUBJECTS AND METHODS Study design This cross-sectional study aimed at investigating the association between some specific characteristics and fatigue in patients with acromegaly. For this purpose, a multivariate analysis was performed using linear regression. A set of arterial function, functional capacity, and hormonal variables were tested as explanatory in linear regression models to predict fatigue (the dependent variable).

Participants This study was performed between September 2016 and May 2017 with a cohort of patients with acromegaly. Consecutive subjects aged over 18 years were recruited at the Acromegaly Clinic of the Hospital Universitário Clementino Fraga Filho Federal University of Rio de Janeiro, Brazil. They were included if a diagnosis had been confirmed by an IGF-I level above the age-adjusted upper limit of the normal range and by assessment of the absence of GH suppression < 1.0 ng/mL during the oral glucose tolerance test. After treatment, the disease was considered controlled in the presence of a random GH level below 1.0 ng/mL and of an IGF-I level within the age-adjusted reference range. The disease was considered active when IGF-I levels were above the upper limit of the age-specific range of normality or when GH levels remained above 1.0 ng/mL (16). Patients with any of the following conditions were excluded from the study: without regular clinical follow-up, classification of controlled acromegaly for less than six months, severe joint pain, history of medication change in the last six months, respiratory infection in the last month, unable to perform the proposed tests, and women who were fertile and not using contraceptives. All participants signed an informed consent form, and the protocol was approved by the Research Ethics 236

Committee of the Augusto Motta University Center under the number CAAE 53678316.9.0000.5235.

Explanatory (independent) assessments Noninvasive pulse wave analysis (PWA) was used to assess the pulse wave velocity (PWV), arterial compliance (AC), and reflection index (IR1,2). The brachial and radial arterial pulse wave signals were captured through the AFA system (17) for PWV calculation, whereas the same pressure pulse signals were used to determine AC using a four-element Windkessel model. The system consisted of piezoelectric transducers fastened with Velcro straps and connected to a preamplifier. This circuit was coupled to a USB-6009 14-bit model acquisition board (National Instruments, Dallas, TX, USA), which was connected to a computer. The sampling rate was 1.0 kHz per channel. The signals were recorded and processed using a program developed in LabVIEW (National Instruments, Dallas, TX, USA) version 8.0 for Windows (Microsoft Corporation, Seattle, WA, USA). The isometric dominant upper limb dynamometry (DYN) was measured using a handheld hydraulic dynamometer (model SH5001, SAEHAN Corporation, Yangdeok-Dong, Masan, South Korea) according to the American Society of Hand Therapists’ (ASHT) recommendations. Subjects sat comfortably with shoulders slightly adducted, elbows flexed at 90°, and forearms and wrists in neutral positions (18). Three maximal attempts were alternately made in each arm with a contraction duration of three seconds and a rest period of 60 seconds between each trial. The best results were recorded for analysis. The predicted value for manual grip strength was set according to Neves and cols. (19) equation for the Brazilian population. The six-minute walking test (6MWT) followed the American Thoracic Society’s recommendations (20). It was carried out in a level, 50-meter corridor. The total distance (6MWD), dyspnea (as measured by the BORG scale), heart rate (HR), and oxygen saturation were recorded (20). The predicted distance values were calculated according to Britto and cols.’s (21) equation for the Brazilian population.

Outcome (dependent) assessments General fatigue was recorded using the MFIS in which the total score ranges from 0 to 84 (9). Twenty-one questions were divided into three domains: cognitive, Arch Endocrinol Metab. 2019;63/3

Fatigue in acromegaly

Statistics Descriptive analyses were performed using mean ± SD or frequency (absolute, %) depending on the variable type. Several models were tested with laboratory (IGF-I), vascular (PWV, AC, and IR1,2), and functional (DYN and 6MWT) variables used as predictors of MFIS. A forward stepwise method was applied using the determination coefficients (R2) value as an estimate of linear model fit. All analyses were performed using SigmaStat 3.1 (SYSTAT Software Inc., Point Richmond, CA, USA). Calibration was verified by an assessment of the calibration plot (measured vs. predicted along with regression lines showing the slope and intercept) and the limits of agreement (LOA) plot (23).

Model calibration There was no obvious relationship between the bias and the mean (Figure 1, bottom) in the MFIS model. The SD for the bias was a score of 0.0 ± 14.9 (Figure 1, bottom), indicating that the estimate was accurate. The 95% CI for the bias was a score of - 6.5 to 6.5, and the LOA and respective 95% CI for the lower and upper LOA were scores of -29.3 [-40.5; -18.1] and 29.3 [18.1; 40.5], respectively. These intervals are somewhat wide and reflect both the sample size and the moderate variation in the differences. Table 1. Patients’ characteristics Variables

N = 23

Anthropometric Data Body mass (kg)

89.0 ± 12.5

Body height (m)

1.67 ± 0.09

Laboratory Data GH (ng/mL)

8.44 ± 19.39

IGF-I (ng/mL)

RESULTS Sample characteristics Of the 27 recruited patients, four were excluded because of outdated IGF-I recordings (n = 3) and low quality of the PWA signal (n = 1). Seventeen presented with active disease, and six presented with controlled disease. The sample included 14 women and nine men, with a mean age of 49.4 ± 11.2 years. The explanatory (independent) variables, such as IGF-I, DYN, and 6MWD, which have predicted values, were normalized by obtained-predicted ratio. PWV, AC, and IR1,2 do not have predicted values. The general characteristics of the patients are shown in Table 1, and the univariate correlations between MFIS and the predictors are shown in Table 2.

The predictive model The model for MFIS was evaluated using hormonal variables (IGF%), peripheral vascular integrity (PWV, AC, or IR1,2), and physical capacity (DYNO/DYNP or 6MWDO/6MWDP) (Table 3). Six multivariate linear regression models for general fatigue (MFIS) were tested (Table 3). The predictive model with the best fit for MFIS showed R2 = 0.378; R2 adjusted = 0.280; standard error estimate = 16.081; and P = 0.026. The regression equation was MFIS = 48.855 - (7.913 × IGF%) + (1.483 × AC) - (23.281 × DYNO/DYNP). Arch Endocrinol Metab. 2019;63/3

412.39 ± 295.46

IGF%

1.60 ± 0.98

Pulse Wave Analysis PWV (m/s)

10.19 ± 2.34

AC (x 106 cm5/dina)

1.08 ± 0.46

IR1,2 (%)

85.3 ± 29.7

Physical Capacity DYN (kgf)

33.9 ± 9.3

DYNO/DYNP

0.91 ± 11.24

DTC6M (m)

603.0 ± 106.1

6MWDO/6MWDP

1.11 ± 0.50

Fatigue MFIS

31.2 ± 18.9

Results are presented as mean ± standard deviation. GH: growth hormone; IGF-I: insulin-like growth factor type I at the onset of the study; IGF%: percentage of IGF above the normality threshold; PWV: pulse wave velocity; AC: arterial compliance; IR1,2: reflection index; DYN: dominant upper limb dynamometry; DYNO/DYNP: predicted-obtained dominant upper limb dynamometry ratio; 6MWD: six-minute walk test distance; 6MWDO/6MWDP: obtained-predicted six-minute walk test distance ratio; MFIS: modified fatigue impact scale.

Table 2. Univariate correlations between MFIS and the predictors Independent variable

P-value

IGF%

-0.492

0.0171

0.458

0.0280

6MWDO/6MWDP

0.143

0.515

PWV

-0.115

0.602

DYNO/DYNP

-0.0698

0.752

IR1,2

-0.00809

0.971

GF%: percentage of IGF above the normality threshold; PWV: pulse wave velocity; AC: arterial compliance; IR1,2: reflection index; DYNO/DYNP: obtained-predicted dominant upper limb dynamometry ratio; 6MWDO/6MWDP: obtained-predicted six-minute walk test distance ratio.

237

physical, and psychosocial function. The lower the score, the lower the fatigue, and a cutoff score of 38 discriminated fatigued from nonfatigued individuals (22).

Fatigue in acromegaly

Table 3. Regression coefficients for MFIS prediction in subjects with acromegaly p-value

R = 0.378

0.026

IGF%

-7.913 (3.686)

0.045

1.483 (0.789)

0.076

-23.281 (24.309)

0.350

R2 = 0.364

0.032

IGF%

-7.489 (3.706)

0.058

1.400 (0.795)

0.094

16.121 (23.487)

0.501

R2 = 0.276

0.098

IGF%

-10.371 (3.887)

0.015

IR1,2

-7.730 (12.757)

0.552

DYNO/DYNP

-20.924 (26.299)

0.436

R2 = 0.267

0.109

IGF%

-9.744 (3.857)

0.021

PWV

-0.579 (1.621)

0.725

-20.308 (26.527)

0.453

R = 0.265

0.111

IGF%

-9.731 (3,870)

0.021

IR1,2

-4.980 (13.018)

0.706

6MWDO/6MWDP

15.038 (25.746)

0.566

R = 0.263

0.115

-9.304 (3.840)

0.026

DYNO/DYNP Model 2

6MWDO/6MWDP Model 3

Model 4

DYNO/DYNP Model 5

Model 6 IGF% PWV 6MWDO/6MWDP

-0.449 (1.611)

0.784

17.282 (25.295)

0.503

IGF%: obtained-predicted insulin-like growth factor-like type I ratio; PWV: pulse wave velocity; AC: arterial compliance; IR1,2: reflection index; DYN: dominant upper limb dynamometry; DYNO/ DYNP: predicted-obtained dominant upper limb dynamometry ratio; 6MWD: six-minute walking distance; 6MWDO/6MWDP: obtained-predicted six-minute walking distance ratio.

y = 0,378x + 19,392 R2 = 0,378 60 MFIS score, predicted

Model 1

Acromegaly (n = 23)

40 60 MFIS score, measured

Limits of agreement plot

80 Bias in MFIS score (meassured - predicted)

Independent variable

Regression plot

60 40 20 0 0

-20 -40 -60 -80

Average MFIS score (measured and predicted)

DISCUSSION

Figure 1. Analysis of the multivariable linear regression model for Modified Fatigue Impact Scale. Top: Calibration plot of the measured vs. predicted Modified Fatigue Impact Scale. Bottom: Limits of agreement plot of the averaged values and the bias (measured-predicted values).

The present study used the MFIS to develop predictive models of general fatigue in subjects with acromegaly. Along with IGF-I levels, arterial compliance and handgrip dynamometry were important predictors for the model with the best fit. Furthermore, IGF% was an independent predictor in five of the six presented models. This result point to the need for new, longitudinal studies to assess if normalization of IGF-1 during treatment could have a positive impact on physical capacity, AC, and fatigue scores. Conversely, as GH secretion is pulsatile, individuals without the disease may have peaks, and, in turn, patients with acromegaly may present values considered low. Thus, attention is required, as currently there are no reference

ranges for baseline GH. IGF-I has no pulsatile secretion or circadian variation that helps identify and characterize the disease (1). The IMMULITE 2000 assay determined reference values for IGF-I according to different ages of the Brazilian population (24). Several issues remain unanswered regarding the cardiovascular complications and their prognostic implications in acromegaly (25). The increase in the concentration of GH and IGF-I is associated with arterial stiffness (3). In this way, variables of arterial function are important for the early evaluation of these patients. In our study, peripheral arterial function was

238

Arch Endocrinol Metab. 2019;63/3

Fatigue in acromegaly

Arch Endocrinol Metab. 2019;63/3

their submaximal and maximal aerobic performance. They also found an association between ventilatory threshold, maximum oxygen uptake, and a reduction in fatigue perception. Although cross-sectional studies are not designed to establish causality relationships, our results suggest that physical function should be assessed in the follow-up of patients with acromegaly. Moreover, it is likely that such subjects might benefit from rehabilitation strategies aiming at improving peripheral muscle function and aerobic capacity, resulting in the improvement of fatigue perception and quality of life. As the number of predictive (independent) variables in a model depends on the sample size, we could not test models with more than three. Although it can be considered a limitation of this study, we examined many different combinations of hormonal, vascular, and functional variables. Moreover, the Bland-Altman diagram (Figure 1) shows that the estimation was accurate for the best model. Thus, we believe that, in the face of the scarcity of evidence on the causes of fatigue in patients with acromegaly, this study represents an essential contribution to the field. We conclude that, in patients with acromegaly, general fatigue can be predicted by hormonal profile, peripheral vascular integrity, and functional variables. Future longitudinal studies with a larger sample size are needed to elucidate the pathophysiology of fatigue in such patients and assess if supervised exercise programs positively impact their physical performance and thus result in lower fatigue scores. Funding: this study was supported by the Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Vieira Neto L, Abucham J, Araujo LA, Boguszewski CL, Bronstein MD, Czepielewski M, et al. Recomendações do Departamento de Neuroendrocrinologia da Sociedade Brasileira de Endocrinologia e Metabologia para o diagnóstico e tratamento de acromegalia no Brasil. Arq Bras Endocrinol Metab. 2011;55(9):91-105. 2. Donangelo I, Une K, Gadelha M. Diagnóstico e tratamento da acromegalia no Brasil. Arq Bras Endocrinol Metabol. 2003;47(4):331-46. 3. Pivonello R, Auriemma RS, Grasso LFS, Pivonello C, Simeoli C, Patalano R, et al. Complications of acromegaly: cardiovascular, respiratory and metabolic comorbidities. Pituitary. 2017;20(1):46-62.

239

tested by the PWA (PWV, AC, and IR1,2). PWV varies in direct proportion to the tension on the arterial wall (26), whilst IR1,2 is the measurement of pulse wave reflection used to assess total vascular resistance (27). AC is an indicator of arterial stiffness along the path traveled by the pulse wave during its propagation (28). Because PWV is associated with cardiovascular diseases in different populations, it is considered a gold standard for assessing arterial stiffness along with known risk factors (29). However, different from AC, PWV and IR1,2 were not predictors of MFIS when combined with hormonal profile and physical capacity in our study. It is likely that the high variability of PWV and IR1,2 determined a greater dispersion of the data excluding these variables as predictors. The association between vascular impairment and muscle function is well known. Increased risk of peripheral vascular disease progressively restricts perfusion adaptability in skeletal muscle microcirculation (30). It is possible that these vascular impairments play a role in the fiber type changes and muscle dysfunction in these patients, contributing to higher fatigue scores. In this way, many authors have observed that patients with acromegaly present with impaired muscle function and exercise intolerance (12,31,32). As these variables may be associated with general fatigue, Walchan and cols. (31) evaluated a correlation between muscle function variables and Fatigue Impact Scale (FIS). Although they did not find an association with handgrip strength (HGS – isometric dynamometry), a significant correlation was observed between FIS and isokinetic dynamometry of the lower limbs. In our study, physical capacity was assessed by DYNO/DYNP and 6MWDO/6MWDP. Both were statistically significant when associated with AC, the first being the best predictor of fatigue as verified by the determination coefficient. The upper limb isometric dynamometry is easy to perform and commonly used to represent the general muscle strength. Thus, HGS emerges as a practical alternative for the follow-up of subjects with acromegaly by helping to understand the causes of fatigue as reported by the patients in the clinic. The 6MWT it is a low cost, well-standardized method for assessing functional capacity (20). Its contribution to predicting the fatigue scores found in our study (Table 2: models 2, 5, and 6) is in accord with the results of Thomas and cols. (33) who demonstrated that successful medical treatment of symptomatic patients with acromegaly using octreotide-LAR improved

Fatigue in acromegaly

4. Yaron M, Izkhakov E, Sack J, Azzam I, Osher E, Tordjman K, et al. Arterial properties in acromegaly: relation to disease activity and associated cardiovascular risk factors. Pituitary. 2016;19(3):322-31. 5. Fedrizzi D, Czepielewski MA. Cardiovascular disturbances in acromegaly. Arq Bras Endocrinol Metabol. 2008;52(9):1416-29. 6. Melmed S. Acromegaly. In: Jameson JL, De Groot LJ. Endocrinology: Adult and Pediatric. 7th ed. Philadelphia: Elsevier; 2016. p. 209-226.e7. 7. Smith JC, Lane H, Davies N, Evans LM, Cockcroft J, Scanlon MF, et al. The effects of depot long-acting somatostatin analog on central aortic pressure and arterial stiffness in acromegaly. J Clin Endocrinol Metab. 2003;88(6):2556-61. 8. Katznelson L, Atkinson JLD. AACE Guidelines for the diagnosis and treatment of Acromegaly. Endocr Pract. 2011;17(4):1-44. 9. Pavan K, Schmidt K, Marangoni B, Mendes MF, Tilbery CP, Lianza S. Esclerose múltipla: Adaptação transcultural e validação da escala modificada de impacto de fadiga. Arq Neuropsiquiatr. 2007;65(3A):669-73. 10. Dittner AJ, Wessely SC, Brown RG. The assessment of fatigue: A practical guide for clinicians and researchers. J Psychosom Res. 2004;56(2):157-70. 11. Woodhouse LJ, Mukherjee A, Shalet SM, Ezzat S. The influence of growth hormone status on physical impairments, functional limitations, and health-related quality of life in adults. Endocr Rev. 2006;27(3):287-317. 12. Lopes AJ, Guedes da Silva DP, Ferreira ADS, Kasuki L, Gadelha MR, Guimarães FS. What is the effect of peripheral muscle fatigue, pulmonary function, and body composition on functional exercise capacity in acromegalic patients? J Phys Ther Sci. 2015;27(3):719-24. 13. Guedes da Silva DP, Guimarães FS, Dias CM, Guimarães Sde A, Kasuki L, Gadelha MR, et al. On the functional capacity and quality of life of patients with acromegaly: are they candidates for rehabilitation programs? J Phys Ther Sci. 2013;25(11):1497-501. 14. Miller A, Doll H, David J, Wass J. Impact of musculoskeletal disease on quality of life in long-standing acromegaly. Eur J Endocrinol. 2008;158(5):587-93. 15. Drange MR, Fram NR, Herman-Bonert V, Melmed S. Pituitary tumor registry: A novel clinical resource. J Clin Endocrinol Metab. 2000;85(1):168-74.

19. Neves RS, Lopes AJ, Lucia S, Menezes S De, Rafaela T, Lima DL, et al. Hand Grip Strength in Healthy Young and Older Brazilian Adults: Development of a Linear Prediction Model Using Simple Anthropometric Variables. Kinesiology. 2017;49:1-9. 20. Crapo RO, Casaburi R, Coates AL, Enright PL, MacIntyre NR, McKay RT, et al. ATS statement: Guidelines for the six-minute walk test. Am J Respir Crit Care Med.;166(1):111-7. 21. Britto RR, Probst VS, Dornelas De Andrade AF, Samora GAR, Hernandes NA, Marinho PEM, et al. Reference equations for the six-minute walk distance based on a Brazilian multicenter study. Braz J Phys Ther. 2013;17(6):556-63. 22. Parks NE, Eskes GA, Gubitz GJ, Reidy Y, Christian C, Phillips SJ. Fatigue impact scale demonstrates greater fatigue in younger stroke survivors. Can J Neurol Sci. 2012;39(5):619-25. 23. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307-10. 24. Rosario PW. Normal values of serum IGF-1 in adults: results from a Brazilian population. Arq Bras Endocrinol Metab. 2010;54(5):477-81. 25. Mosca S, Paolillo S, Colao A, Bossone E, Cittadini A, Iudice FL, et al. Cardiovascular involvement in patients affected by acromegaly: An appraisal. Int J Cardiol. 2013;167(5):1712-8. 26. Bramwell JC, Hill AV. The Velocity of the Pulse Wave in Man. Proc R Soc B Biol Sci. 1922;93(652):298-306. 27. Ferreira ADS, Filho JB, Cordovil I, Souza MN De. Noninvasive pressure pulse waveform analysis of flow-mediated vasodilation evoked by post-occlusive reactive hyperemia maneuver. Biomed Signal Process Control. 2012;7(6):616-21. 28. Ferreira AS, Santos MAR, Barbosa Filho J, Cordovil I, Souza MN. Determination of radial artery compliance can increase the diagnostic power of pulse wave velocity measurement. Physiol Meas. 2004;25(1):37-50. 29. Pizzi OL, Brandão AA, Pozzan R, Magalhães MEC, Freitas EV, Brandão AP. A velocidade da onda de pulso em jovens: estudo do Rio de Janeiro. Arq Bras Cardiol. 2011;97(1):53-8. 30. Frisbee JC, Butcher JT, Frisbee SJ, Olfert IM, Chantler PD,Tabone LE, et al. Increased peripheral vascular disease risk progressively constrains perfusion adaptability in the skeletal muscle microcirculation. Am J Physiol Heart Circ Physiol. 2016;310(4):H488-504. 31. Walchan EM, Guimarães FS, Soares MS, Kasuki L, Gadelha MR, Lopes AJ. Parameters of knee isokinetic dynamometry in individuals with acromegaly: Association with growth hormone levels and general fatigue. Isokinet Exerc Sci. 2016;24(4):331-40.

17. de Sá Ferreira A, Filho JB, Cordovil I, de Souza MN. Three-section transmission-line arterial model for noninvasive assessment of vascular remodeling in primary hypertension. Biomed Signal Process Control. 2009;4(1):2-6.

32. Lopes AJ, Ferreira AS, Walchan EM, Soares MS, Bunn PS, Guimarães FS. Explanatory models of muscle performance in acromegaly patients evaluated by knee isokinetic dynamometry: Implications for rehabilitation. Hum Mov Sci. 2016;49:160-9.

18. Bohannon RW, Peolsson A, Massy-Westropp N, Desrosiers J, Bear-Lehman J. Reference values for adult grip strength measured with a Jamar dynamometer: a descriptive metaanalysis. Physiotherapy. 2006;92:11-5.

33. Thomas SG, Woodhouse LJ, Pagura SM, Ezzat S. Ventilation threshold as a measure of impaired physical performance in adults with growth hormone excess. Clin Endocrinol (Oxf). 2002;56(3):351-8.

16. Katznelson L, Laws ER, Melmed S, Molitch ME, Murad MH, Utz A, et al. Acromegaly: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2014;99(11):3933-51.

240

Arch Endocrinol Metab. 2019;63/3

original article

The relationship of flavonoid intake during pregnancy with excess body weight and gestational diabetes mellitus Mariana de Andrade Balbi1, Lívia Castro Crivellenti1, Daniela Cristina Candelas Zuccolotto1, Laércio Joel Franco2, Daniela Saes Sartorelli2

ABSTRACT Objective: To investigate the relationship of flavonoid intake during pregnancy with maternal excessive body weight and gestational diabetes mellitus (GDM). Subjects and methods: A crosssectional study was conducted among 785 adult women in singleton pregnancies, and data were collected at the time of the oral glucose tolerance test. For the body mass index (BMI) classification according to the gestational age, the criteria of Atalah was used, and the diagnosis of GDM was based on the World Health Organization of 2014. Two 24-hour dietary recalls were obtained, and the usual intake was determined by the Multiple Source Method. Adjusted multinomial logistic regression was used to investigate the relationship of the flavonoids with overweight and obesity, and adjusted non-conditional logistic regression for the relationship of the flavonoids with GDM. Results: The mean (SD) age of the women was 28 (5) years, 32.1% were overweight, 24.6% were obese and 17.7% were diagnosed with GDM. The median (P25, P75) of total flavonoid intake was 50 (31,75) mg/day. Considering the eutrophic women as the reference, the pregnant women with a higher total flavonoid intake [OR 0.62 (95% CI 0.38; 0.96)] and anthocyanidin intake [OR 0.62 (95% CI 0.40; 0.99)] were less likely to be obese when compared to the women with lower intakes. No association of the flavonoids intake with overweight or GDM was found. Conclusion: A very low intake of flavonoids was observed. The data suggest that the intake of foods naturally rich in total flavonoids and anthocyanidin has a beneficial role regarding obesity among pregnant women. Arch Endocrinol Metab. 2019;63(3):241-9

INTRODUCTION

aternal excess weight is a relevant risk factor for deleterious maternal and fetal health outcomes in the short- and long-term (1,2) and is the main determinant of gestational diabetes mellitus (GDM) (3). Globally, it is estimated that approximately 16% of pregnant women are affected by hyperglycemia (3). Therefore, investigations into factors with potential protective effects for the disease are of paramount importance. Evidence suggests that diets composed of plant-rich foods high in bioactive compounds, such as flavonoids, have a protective role for type 2 diabetes mellitus (DM) (4) and obesity (5). Flavonoids belong to the group of polyphenols and are classified into six subclasses: flavan3-ols, anthocyanidins, flavonols, flavanones, flavones and isoflavones (6). In vitro and in vivo studies have demonstrated that flavonoids have anti-inflammatory Arch Endocrinol Metab. 2019;63/3

Correspondence to: Daniela Saes Sartorelli Departamento de Medicina Social, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo Av. Bandeirantes, 3900 14049-900 – Ribeirão Preto, SP, Brasil daniss@fmrp.usp.br Received on June/4/2018 Accepted on Apr/8/2019 DOI: 10.20945/2359-3997000000143

and antioxidant properties (7), with a high potential for modulation of the intestinal microbiota (8) and the inflammatory response induced by excess weight (9). It has been suggested that flavonoids can act in glycemic homeostasis through several mechanisms: inhibiting glucose absorption; stimulating insulin secretion by pancreatic β cells; favoring the modulation of glucose release by the liver; and acting on the improvement of the action of insulin receptors and absorption of glucose by the tissues (6). Epidemiological studies indicate an inverse relationship between flavonoid intake and adiposity indicators (10-13). Hughes and cols. (14) observed an increase in the body mass index (BMI) over the years among women with lower intakes of flavonols, flavones and catechins. In a study conducted with 2,734 adult twins, it was found that a higher habitual intake of flavonoids, including anthocyanins, flavan-3-ols, 241

Keywords Pregnancy; flavonoids; gestational diabetes mellitus; overweight; obesity

Programa de Pós-Graduação em Saúde Pública, Faculdade de Medicina de Ribeirão Preto (FMRP), Universidade de São Paulo (USP), Ribeirão Preto, SP, Brasil 2 Departamento de Medicina Social, Faculdade de Medicina de Ribeirão Preto (FMRP), Universidade de São Paulo (USP), Ribeirão Preto, SP, Brasil 1

Flavonoid intake and gestational diabetes

flavonols and proanthocyanidins, was associated with a lower percentage of body fat mass, regardless of confounding factors (12). Epidemiological studies conducted in the United States and Europe suggest a protective effect of dietary flavonoids for the development of DM, regardless of confounding factors (4,15-17). European cohort data suggest a lower incidence of DM among individuals with higher habitual flavonoid intake (16). In the United States, cohort findings corroborate this hypothesis (17). In a dose-response analysis of a meta-analysis of prospective studies, Liu and cols. (4) concluded that the intake of 500 mg/day of flavonoids was associated with a 5% reduction in the risk of DM. During pregnancy there is a profound transformation in the intestinal microbiota of women that may interfere with the absorption and metabolism of flavonoids (18). Although investigations in vitro and in animal models suggest a potential protective effect of flavonoids for GDM (6), no epidemiological studies were found that investigated the relationship of the intake of these compounds in pregnancy with excess weight and GDM. The hypothesis of the present study was an inverse relationship between the intake of flavonoids with excess body weight and GDM. The aim of the present study was to investigate the relationship of intake of total flavonoids and subclasses during gestation with maternal excess weight and GDM.

pregnancy, report of use of glycemia altering drugs (such as glucocorticoids); and reports of diseases that alter habitual food consumption. Women with fasting blood glucose ≥126 mg/dL or ≥200 mg/dL two hours after glucose overload were considered to have prior DM and were subsequently excluded from the study. A total of 1,446 women were contacted, however, 19 women refused to participate in the study, 639 were excluded according to the established criteria and three pregnant women presented incomplete data, giving a sample total of 785 women (19). Assuming a prevalence of 20% of GDM among women attended in the Brazilian Nation Health System, a sample size of 512 individuals was necessary, with a margin of error of 5% (20). For the calculation of the gestational week (GW), the date of the last menstruation recorded on the medical card of the pregnant woman was used. The gestational age at the time of the interview ranged from 24 to 39 weeks (70% of the women were between the 24th and 28th GW, 21.5% between the 29th and 32nd GW and 8.5% ≥33 GWs). The study was approved by the Research Ethics Committee of the School Health Center of the Ribeirão Preto Medical School (075/2015-CEP/CSE-FMRPUSP) of the University of São Paulo. The pregnant women agreed to participate in the study by signing the informed consent form.

Classification of BMI according to gestational age

SUBJECTS AND METHODS

Study design and population A cross-sectional study conducted with 785 women, in singleton pregnancies, receiving prenatal care in the Public Health System of Ribeirão Preto, São Paulo state, Brazil (19). The recruitment of women was carried out by trained nutritionists, when the oral glucose tolerance test (OGTT) was performed, and all the pregnant women who attended these laboratories to carry out the OGTT, between 2011 and 2012, were invited to participate in the study. At the time of the assessment, neither the women nor the interviewer knew about the results of the OGTT. The study inclusion criteria were: age 20 years or over and pre-gestational body mass index 20 kg/m2 or over. The exclusion criteria were: report of prepregnancy diabetes; pregnant women with GDM screening prior to the 24th gestational week; twin 242

Data regarding weight (kg) and height (m) were obtained by trained nutritionists at the time of the OGTT using a digital scale (Tanita, model HS 302) and a portable stadiometer (Sanny, model ES 2040), respectively. The current BMI was calculated, with the criteria proposed by Atalah and cols. used for the classification of BMI according to gestational age (21). The criterion proposed by Atalah and cols. (21) was developed using data of Chilean women, and presents limitations to be used in Brazil. However, this criterion was adopted by the Brazilian Ministry of Health to evaluate pregnant women in the country (22).

Diagnosis of gestational diabetes mellitus Blood samples were obtained after 12 hours of fasting, followed by the ingestion of a 75g glucose overload by the participants. Determination of fasting glycemia, one and two hours after overload, was performed using the glucose oxidase test. The diagnosis of GDM Arch Endocrinol Metab. 2019;63/3

Flavonoid intake and gestational diabetes

Estimation of flavonoids from the usual diet The food intake assessment was performed through two 24-hour dietary recalls on non-consecutive days, adopting the “multiple-pass” technique in three stages (24). The first 24-hour dietary recall was obtained at the time of the interview and the second through telephone contact, with at least seven days between replications, regardless the day of the week. The energy intake was estimated using the NutWin® software (Nutrition Support Program, Version 1.5, São Paulo, Brazil, 2002), adopting the database of the Brazilian Food Composition Table. For the estimation of total flavonoids and subclasses of most foods, the Brazilian Food Composition Table (TBCA-USP) (25,26) was used. For the estimative of flavonoids of foods without data on the TBCA, the USDA Database for the Flavonoid Content of Selected Foods was used (27). In the present study, the subclasses of flavonoids explored were: flavonols, flavones, flavanones, flavan-3ols and anthocyanidins. Isoflavones were not quantified due to low intake of food sources among the study participants. To estimate the usual intake of flavonoids and subclasses in the diet the Multiple Source Method (MSM) was employed. The MSM is a statistical modeling program developed by the European Prospective Investigation into Cancer and Nutrition to estimate the usual diet (web-based https://msm.dife.de/) (28). The method estimates the usual intake of nutrients through three steps: 1. Estimation of probability of intake on a random day; 2. Estimation of usual intake on days of consumption, corrected for variability; 3. The product of the probability of intake (step 1) divided by the usual intake on the day of consumption (step 2) results in the usual intake. The MSM dispenses a high number of 24hour dietary recalls (or food records) replications (29). In the present study, the response rate for the second 24-hour dietary recall was 62%, considered adequate for analysis using the MSM (30). In a previous study it was observed that the MSM was an adequate method to estimate the usual food intake during pregnancy (31). Arch Endocrinol Metab. 2019;63/3

Covariables Information on age (years), education (years of study), smoking (never smoked, ex-smoker and current smoker), practice of physical activity (minutes/week of walking and physical exercise), parity, family history of diabetes and history GDM in previous pregnancies were obtained through a structured questionnaire at the time of the interview during the OGTT. Pre-gestational weight was calculated based on data recorded on the obstetric monitoring card of the pregnant woman.

Statistical analysis The study population was characterized using the mean (SD) and median (P25, P75) values for the continuous descriptive variables, while the categorical variables were expressed as frequencies. The dietary intake data were log transformed, and T Student test and ANOVA with post-hoc of Bonferroni were used to investigate differences in the estimated flavonoids of pregnant women according to glycemic homeostasis and the BMI categories, respectively. The MSM was used to estimate the usual intake of total flavonoids and subclasses. For inclusion in the models, the estimate of flavonoids was transformed into its natural logarithm and adjusted for total energy through the residual method. The data were later categorized into intake tertiles. Multinomial logistic regression models were used to evaluate the relationship between total flavonoids and subclasses (in tertiles) and the BMI categories according to gestational week, considering the eutrophic women as the reference. In the analysis, 31 women with low weight according to the gestational week were excluded. The following variables were considered in the adjusted models: age (years), education (years of study), gestational week at the time of interview, smoking (never smoked, ex-smoker, current smoker), practice of physical activity (minutes/week), parity (number of children) and total energy of the diet (logtransformation, kcal/day). Adjusted logistic regression models were used to estimate the relationship of dietary content of flavonoids and subclasses with GDM, adjusted for age (years), education (years of study), gestational week at the time of the interview, smoking (never smoked, ex-smoker, current smoker), practice of physical activity (minutes/week of walking and exercise), parity (number of children) pre-gestational BMI (kg/m2), current BMI (kg/m2), family history of diabetes (yes/ 243

was performed according to the criteria of the World Health Organization of 2014 (23), considering GDM if at least one of the glycemia values were altered in the women, at any stage of the gestation: fasting glycemia between 92 and 125 mg/dl; one hour after glucose overload ≥180 mg/dl; or glucose two hours after glucose overload between 153 and 199 mg/dl.

Flavonoid intake and gestational diabetes

no), history of GDM (yes/no) and total dietary energy (log-transformation, kcal/day). The significance level was set at p<0.05 and all statistical analyses were performed using the SPSS software (Version 24.0, SPSS Inc. Woking, Surrey, UK).

RESULTS

Food and beverages

Contribution (%)

Total flavonoids Mate tea

45.0

Beans (brown)

11.0

Oranges

8.0

Among the 785 study participants, 17.7% were diagnosed with GDM, and 56.7% had excessive body weight. The age of the women ranged from 20 to 45 years and the education level between zero and 15 years of study (Table 1). It was observed that the drinks and foods that contributed the most to the intake of total flavonoids of the pregnant women’s diet were: mate tea, beans, oranges, chocolate powder and orange juice, as presented in Table 2. The median (P25, P75) of total flavonoids of the pregnant women was 50 (31, 75) mg. Women with adequate BMI reported a higher intake of anthocyanidin and flavan-3-ol, when compared with those classified with obesity. Moreover, pregnant women with

Chocolate powder

7.0

Orange juice

6.0

Others

23.0

Table 1. Sociodemographic, anthropometric and lifestyle characteristics of the pregnant women. Ribeirão Preto, SP, 2011-2012. n = 785 Maternal characteristics

Frequency (n%)

Anthocyanidins Açaí

34.0

Grapes (purple)

33.0

Strawberries

8.0

Plums

5.0

Blackberries

5.0

Others

15.0

Flavan-3-ols Chocolate powder

49.0

Apples

19.0

Bananas

16.0

Chocolate

6.0

Mangos

2.0

Others

8.0

Current GDM History of GDM in previous pregnancies

139 (17.7) 34 (4.3)

Flavanones Oranges

49.0

Family history of diabetes

205 (26.1)

Orange juice

39.0

Tangerines

10.0

Lemons

1.0

Wine (red)

0.1

Others

0.9

BMI classification (kg/m2)b Low weight

31 (3.9)

Eutrophic Overweight

309 (39.4) 252 (32.1)

Obese

193 (24.6)

Smoking

Flavones Spring onions

56.0

Chicory

15.0

Watermelon

7.0

Mean (SD)

Pumpkin

6.0

Age (years)

27.6 (5.5)

Grapes (purple)

4.0

Gestational age at time of interview

27.7 (3.2)

Others

12.0

Education (years of study)

9.2 (2.7)

Flavonols

Pre-gestational BMI (kg/m )

25.9 (5.0)

Mate tea

69.0

Parity

1.2 (1.2)

Beans (brown)

16.0

Median (P25; P75)

Arugula

3.0

40 (0; 140)

Apples

3.0

Lettuce

1.0

Others

8.0

Never smoked

624 (79.5)

Ex smoker Current smoker

90 (11.5) 71 (9.0)

Table 2. Contribution of foods and beverages to the intake of total flavonoids and subclasses of the pregnant women. Ribeirão Preto, SP, 2011-2012. n = 785

Physical activityc According to World Health Organization criteria. (22) b Classification of current BMI according to gestational week. (21) c Minutes per week of walking and physical exercise. a

244

Arch Endocrinol Metab. 2019;63/3

Flavonoid intake and gestational diabetes

adequate BMI reported a higher intake of flavanones, when compared with those classified with overweight. However, no significant difference was observed according to the glycemic homeostasis (Table 3). In adjusted multinomial logistic regression models, it was observed that the women classified in the third tertile of the total flavonoid and anthocyanidin estimate had a lower chance of obesity when compared to the women in the first tertile. A direct association between the second tertile of flavanones and obesity was observed, however this association was not verified for the third tertile (Table 4). Table 5 presents the results of the relationship of the estimate of total flavonoids from the usual diet and their subclasses with GDM. In adjusted logistic regression models, no association was observed.

DISCUSSION In the present study, an inverse association of the estimate of the usual intake of total flavonoids and anthocyanidins with obesity during gestation was observed, regardless of confounding factors. The women with higher intakes of total flavonoids and anthocyanidins had a 39% and 38% lower chance, respectively, of being classified as having obesity when compared to the women with lower intakes. However, there was no association between dietary flavonoids and overweight or GDM, probably due to the very low intake of flavonoids observed in the study population. No previous epidemiological studies were found that investigated the relationship between dietary flavonoids during pregnancy and maternal excess weight or GDM.

A direct association between the second tertile of flavanones and obesity was observed. To our knowledge, no published data show a detrimental effect of flavanones on weight control, and this relationship probably results from chance. The median estimate of total flavonoids in the usual diet of the present study (50 mg) was similar to that previously reported among Brazilian adults (55 mg) (32), however, it was much lower than that reported among European adults (246 mg) (16). In the Brazilian study conducted in the city of São Paulo, SP, the main sources of dietary flavonoids were citrus fruits, juices and beans (32); in the present study, the main sources were mate tea, beans and oranges. Worldwide, the main sources of flavonoids are the fruits, vegetables, and wine. In the present study, the median (P25, P75) intake of fruits and vegetables were 67 (38, 136) g/day and 78 (52, 112) g/day, respectively, and the red wine was rarely consumed (data not shown), which could partly explain the very low intake of flavonoids observed. Given the effect of flavonoids on the maintenance of pancreatic β-cell function and insulin sensitivity [6], one of the hypotheses of the present study was an inverse relationship between their intake and GDM, which was not confirmed. It should be noted that many of the epidemiological studies that have verified a protective effect of flavonoids for DM were conducted in Europe (15,16), where the habitual intake of these compounds is much higher than that found among the pregnant women included in the present study, which could partially explain the controversial findings. In the study by Tresserra-Rimbau and cols. (15), it was verified that

Table 3. Median (P25; P75) intake of total flavonoids and subclasses according to the glycemic homeostasis and BMI categories of the pregnant women. Ribeirão Preto, SP, 2011-2012 BMI classificationb (n = 754)

All women (n = 785)

Normal (n = 646)

GDM (n = 139)

Adequate (n = 309)

Overweight (n = 252)

Obesity (n = 193)

Total Flavonoids

50.0 (31.0; 75.0)

51.0 (31.9; 75.1)

47.3 (29.7; 77.3)

57.2 (33.4; 80.9)

49.0 (32.2; 73.3)

45.4 (28.7; 65.0)

Anthocyanidins

1.9 (0.8; 4.7)

1.9 (0.8; 4.6)

1.9 (0.7; 5.7)

2.2 (0.9; 6.2)

1.8 (0.8; 4.7)

1.7 (0.6; 3.9)c,d

Flavan-3-ols

7.8 (2.8; 15.2)

7.9 (2.8; 15.2)

7.3 (2.8; 14.4)

8.5 (3.0; 16.0)

8.0 (2.3; 15.9)

6.7 (2.4; 13.0)c,d

Flavanones

2.3 (1.6; 4.2)

2.3 (1.6; 4.1)

2.2 (1.5; 15.7)

2.4 (1.6; 21.6)

2.1 (1.5; 3.2)

2.4 (1.8; 4.1)c,e

Flavones

0.3 (0.2; 0.6)

0.4 (0.2; 0.7)

0.3 (0.2; 0.6)

0.3 (0.2; 0.5)

Flavonols

24.8 (18.4; 31.9)

24.9 (18.7; 31.7)

24.4 (16.5; 32.8)

25.6 (19.2; 33.1)

24.6 (19.0; 31.1)

24.1 (16.6; 31.5)

According to the WHO criteria. (22) According to the criteria of Atalah. (21) A total of 31 women with low weight according to gestational age were excluded in this analysis. c p < 0.05; according to ANOVA. Variables were log transformed prior to the analysis. d Mean difference between women with adequate BMI, when compared with women classified with obesity. e Mean difference between women with adequate BMI, when compared with women classified with overweight. a

Arch Endocrinol Metab. 2019;63/3

245

Glycemic homeostasisa (n = 785)

Flavonoid intake and gestational diabetes

Table 4. Association of dietary flavonoid intake during pregnancy with excess weight and obesity. Ribeirão Preto, SP, 2011-2012. n = 754a 1st tertile

2nd tertile

3rd tertile

Odds Ratio

95%CI

Odds Ratio

95%CI

1.00 (Ref.)

1.21

0.80; 1.85

0.75

0.49; 1.13

Ref.

1.27

0.81; 1.99

0.61

0.38; 0.96

Ref.

0.85

0.56; 1.30

0.75

0.50; 1.31

Ref.

1.01

0.65; 1.58

0.62

0.40; 0.99

Total Flavonoids Median (mg/day) Overweight Obesity

26.29

50.44

89.29

Anthocyanidins Median (mg/day) Overweight Obesity Flavan-3-ols Median (mg/day)

0.58

1.93

7.06

1.00 2.41

7.82

17.99

Overweight

Ref.

1.23

0.81; 1.88

1.05

0.69; 1.60

Obesity

Ref.

0.89

0.57; 1.40

0.72

0.46; 1.14

Flavanones

Ref. Ref.

1.19

0.79; 1.79

0.67

0.44; 1.02

Ref.

2.17

1.36; 3.47

1.24

0.77; 1.98

Median (mg/day) Overweight Obesity Flavones Median (mg/day)

1.39

2.25

24.36

1.00 0.15

0.32

0.75

Overweight

1.00

1.19

0.79; 1.79

0.88

0.58; 1.32

Obesity

1.00

1.10

0.71; 1.73

0.84

0.54; 1.33

Flavonols Median (mg/day)

1.00 15.82

24.82

35.25

Overweight

1.00

1.21

0.80; 1.83

0.85

0.56; 1.31

Obesity

1.00

0.65

0.41; 1.03

0.75

0.49; 1.17

31 women with low weight according to the gestational week were excluded in this analysis. b Multinomial logistic regression models, considering the eutrophic women as the reference, adjusted for: age (years), gestational week at the time of interview, education (years of study), smoking (never smoked, ex-smoker and current smoker), physical activity (minutes/week of walking, running or physical exercise), number of children and total energy of the diet (log-transformation, kcal/day). a

Table 5. The relationship between dietary flavonoid intake during pregnancy and gestational diabetes mellitus. Ribeirão Preto, SP, 2011-2012. n = 785 1st tertile

2nd tertile

3rd tertile

Total flavonoids Median (mg/day) Odds Ratio (95%CI)a

26.33

50.33

88.32

1.00 (Ref.)

0.75 (0.46; 1.21)

1.11 (0.69; 1.78)

0.68

Anthocyanidins Median (mg/day)

0.58

1.91

7.06

Odds Ratio (95%CI)a

Ref.

1.13 (0.70; 1.83)

1.17 (0.72; 1.89)

0.52

Flavan-3-ols Median (mg/day)

2.41

7.83

17.99

Odds Ratio (95%CI)a

Ref.

1.07 (0.67; 1.71)

0.84 (0.51; 1.37)

0.49

Flavanones Median (mg/day)

1.38

2.24

24.31

Odds Ratio (95%CI)a

Ref.

0.78 (0.48; 1.28)

1.11 (0.69; 1.78)

0.66

Flavones Median (mg/day)

0.15

0.32

0.75

Odds Ratio (95%CI)a

Ref.

1.02 (0.62; 1.69)

1.56 (0.97-2.52)

0.06

Flavonols Median (mg/day) Odds Ratio (95%CI)a

15.94

24.85

35.30

Ref.

0.68 (0.68; 1.10)

0.97 (0.61; 1.54)

0.89

Logistic regression models adjusted for: age (years), education (years of study), gestational week at the time of interview, history of GDM (yes/no), family history of DM (yes/no), smoking (never smoked, ex-smoker and current smoker), practice of physical activity (minutes/week of walking, running or exercise), number of children, pre-gestational BMI (kg/m2), current BMI (kg/m2) and total dietary energy (log-transformation, kcal/day). a

246

Arch Endocrinol Metab. 2019;63/3

Flavonoid intake and gestational diabetes

Arch Endocrinol Metab. 2019;63/3

of the study is that the data were collected by trained nutritionists. The estimation of the flavonoid intake was based on two 24-hour dietary recalls and the estimate of usual intake was performed using the MSM. For the estimation of the flavonoid content of most foods the data determined in Brazilian studies was used. The cross-sectional design constituted the main limitation of the study, however, at the time of the interview neither the women nor the interviewers knew the result of the OGTT, reducing the chance of information bias. Since there is no data on food content of other phenolic compounds available in Brazil, other classes of polyphenols were not determined. The main sources of flavonoids are also rich in other components with antioxidant capacity, which were not estimated in the present study. No data on biomarkers of flavonoid intake was available. In addition, under-reporting of food consumption may have occurred. In addition, due to the lack of information about weight gain in the first trimester of gestation, it was not possible to investigate the relationship between flavonoid intake and gestational weight gain. The adoption of the criterion proposed by Atalah and cols. (21) for the classification of BMI according to gestational age in Brazilian pregnant women is controversial (36), but we do not know another classification available for this population. In addition, this criterion is recommended by the Ministry of Health for the evaluation of pregnant women in the country (22). Due to some evidence for a negative effect of polyphenols in fetal constriction of the ductus arteriosus, intervention studies promoting the intake of flavonoids conducted with pregnant women are not recommended (37). In conclusion, a low intake of flavonoids was found among the pregnant women, however, in sufficient quantity to detect an inverse association of total flavonoids and anthocyanidin with obesity. There was no association between dietary flavonoids and GDM. However, observational studies conducted among populations with a higher usual intake of flavonoids are needed to elucidate their relationship with GDM. Acknowledgments and funding: National Council for Scientific and Technological Development (CNPq: 302498/2015-0 and 472221/2010-8), Coordination for the Improvement of Higher Education Personnel (CAPES), Foundation for Support to Teaching, Research and Assistance of the Hospital das Clínicas of the Ribeirão Preto Medical School of the University of São Paulo (FAEPA) and University of São Paulo (USP), Brazil. Disclosure: no potential conflict of interest relevant to this article was reported. 247

the protective effect of flavonoids in relation to DM was observed among individuals with a flavonoid intake higher than 425 mg/day when compared to those who reported a median intake of 291 mg/day. In a secondary analysis of data from the Nurses Health Study, women with a median intake of 718 mg/day of total flavonoids had a lower risk of DM when compared to women with a median intake of 105 mg/day (17). In a meta-analysis of prospective studies it was concluded that intake of 500 mg/day of flavonoids would be necessary for a 5% reduction in DM risk, ten times higher than the median intake found in the study population (4). However, the flavonoids intake by the pregnant women was sufficient to detect an inverse association with obesity. The inverse association between anthocyanidins in the diet of pregnant women and obesity verified in the present study corroborates the findings of recent epidemiological studies (11,13,17). In secondary analyses of data from the National Health and Nutrition Examination Survey (NHANES 2007-2010 and 20072012) in the United States, it was observed that adult individuals with higher intake of anthocyanidins had a 14% lower chance of having high BMI and waist circumference when compared to individuals with lower intakes (10), with an inverse association with BMI also found (33). In the analysis of data from the Harvard cohort studies, there was a greater chance of weight maintenance in individuals with high anthocyanidin intake when compared to participants with low intake over a four-year monitoring period (13). Experimental studies suggest that the intake of anthocyanidins may contribute to weight control through its modulatory effect on neuropeptide Y and the Y-butyric amino acid receptor in the hypothalamus, thus acting on appetite control (34). However, in the present study, the inverse association between flavonoids and obesity was independent of the total energy intake, as found in previous epidemiological studies (10-12). It has also been suggested that flavonoids can act to reduce visceral fat accumulation and hyperglycemia by inhibiting pancreatic lipase activity and by reducing intestinal lipid absorption (35). Furthermore, recent studies suggest that anthocyanidin exerts a modulating effect on the intestinal microbiota, contributing to the control of obesity [8]. The present study pioneered the investigation of the relationship between flavonoid intake during pregnancy and maternal excess weight and GDM using an epidemiological design. Another strength

Flavonoid intake and gestational diabetes

REFERENCES

1. Godfrey KM, Reynolds RM, Prescott SL, Nyirenda M, Jaddoe VW, Eriksson JG, et al. Influence of maternal obesity on the long-term health of offspring. Lancet Diabetes Endocrinol. 2016;5(1):53-64. 2. Hanson M, Gluckman P, Bustreo F. Obesity and the health of future generations. Lancet Diabetes Endocrinol. 2016;4(12):966-67. 3. International Diabetes Federation. IDF Diabetes Atlas. 8th ed. Brussels: IDF; 2017. 4. Liu YJ, Zhan J, Liu XL, Wang Y, Ji J, He QQ. Dietary flavonoids intake and risk of type 2 diabetes: a meta-analysis of prospective cohort studies. Clin Nutr. 2014;33(1):59-63. 5. Azzini E, Venneria E, Ciarapica D, Foddai MS, Intorre F, Zaccaria M, et al. Effect of red orange juice consumption on body composition and nutritional status in overweight/obese female: a pilot study. Oxid Med Cell Longev. 2017;2017:1-9. 6. Santangelo C, Zicari A, Mandosi E, Scazzocchio B, Mari E, Morano S, et al. Could gestational diabetes mellitus be managed through dietary bioactive compounds? Current knowledge and future perspectives. Br J Nutr. 2016;115(7):1129-44. 7. Anhê FF, Desjardins Y, Pilon G, Dudonné S, Genovese MI, Lajolo FM, et al. Polyphenols and type 2 diabetes: A prospective review. Pharma Nutr. 2013;1:105-14. 8. Jamar G, Estadella D, Pisani LP. Contribution of anthocyanin-rich foods in obesity control through gut microbiota interactions. Bio Factors. 2017;43(4):507-16. 9. Bastos DHM, Rogero MM, Arêas JAG. Mecanismos de ação de compostos bioativos dos alimentos no contexto de processos inflamatórios relacionados à obesidade. Arq Bras Endocrinol Metab. 2009;53(5):646-56. 10. Sebastian RS, Wilkinson Enns C, Goldman JD, Moshfegh AJ. Dietary flavonoid intake is inversely associated with cardiovascular disease risk as assessed by body mass index and waist circumference among adults in the United States. Nutrients. 2017;9(8). 11. Vernarelli JA, Lambert JD. Flavonoid intake is inversely associated with obesity and C-reactive protein, a marker for inflammation, in US adults. Nutr Diabetes. 2017;7(5):e276. 12. Jennings A, MacGregor A, Spector T, Cassidy A. Higher dietary flavonoid intakes are associated with lower objectively measured body composition in women: evidence from discordant monozygotic twins. Am J Clin Nutr. 2017;105(3):626-34. 13. Bertoia ML, Rimm EB, Mukamal KJ, Hu FB, Willett WC, Cassidy A. Dietary flavonoid intake and weight maintenance: three prospective cohorts of 124086 US men and women followed for up to 24 years. BMJ. 2016;352:i17. 14. Hughes LAE, Arts ICW, Ambergen, T, Brants HAM, Dagnelie PC, Goldbohm RA, et al. Higher dietary flavone, flavonol, and catechin intakes are associated with less of an increase in BMI over time in women: a longitudinal analysis from the Netherlands Cohort Study. Am J Clin Nutr. 2008;88:1341-52. 15. Tresserra-Rimbau A, Guasch-Ferré M, Salas-Salvadó J, Toledo E, Corella D, Castañer O, et al. Intake of total polyphenols and some classes of polyphenols is inversely associated with diabetes in elderly people at high cardiovascular disease risk. J Nutr. 2016;146(4):767-77. 16. Zamora-Ros R, Forouhi NG, Sharp SJ, González CA, Buijsse B, Guevara M, et al. Dietary intakes of individual flavanols and flavonols are inversely associated with incident type 2 diabetes in European populations. J Nutr. 2014;144(3):335-43.

19. Barbieiri P, Nunes JC, Torres AG, Nishimura RY, Zuccolotto DC, Crivellenti LC, et al. Indices of dietary fat quality during midpregnancy is associated with gestational diabetes. Nutrition. 2016;32(6):656-61. 20. Trujillo J, Vigo A, Reichelt A, Duncan BB, Schmidt MI. Fasting plasma glucose to avoid a full OG TT in the diagnosis of gestational diabetes. Diabetes Res Clin Pract. 2014;105(3):322-6. 21. Atalah SE, Castillo CL, Castro RS, Aldea AP. Propuesta de um nuevo estándar de evaluación nutricional em embarazadas. Rev Med Chile. 1997;125(12):1429-36. 22. Brasil. Ministério da Saúde, Secretaria de Atenção à Saúde, Departamento de Atenção Básica. Atenção ao pré-natal de baixo risco. Brasília (DF); 2012 [citado 8 out 2018]. (Série A. Normas e Manuais Técnicos) (Cadernos de Atenção Básica, 32). Available in: http://bvsms.saude.gov.br/bvs/publicacoes/ cadernos_atencao_basica_32_ prenatal.pdf. Accessed in: 2 Apr. 2019. 23. World Health Organization. Diagnostic criteria and classification of hyperglycemia first detected in pregnancy: a World Health Organization Guideline. Diabetes Res Clin Pract. 2014;103:341-63. 24. Johnson RK, Soultanakis, RP, Matthews DE. Literacy and body fatness are associated with underreporting of energy intake in US low-income using the multiple-pass 24-hour recall: A doubly labeled water study. J Am Diet Assoc. 1998;98(10):1136-40. 25. Universidade de São Paulo (USP). Tabela Brasileira de Composição de Alimentos (TBCA). Food Research Center (FoRC). Tabelas Complementares favonoides. Versão 6.0. São Paulo, 2017. Available in: http://nware.com.br/tbca/wp-content/uploads/2017/10/Tabelas_complementares-Flavonoides.pdf. Accessed in: 2 Apr. 2019. 26. Menezes EW, Santos NC, Giuntini EB, Dan MCT, Genovese MI, Lajolo FM. Brazilian flavonoid database: Application of quality evaluation system. J Food Compost Anal. 2011;24(4):629-36. 27. Bhagwat S, Haytowitz DB. USDA Database for the Flavonoid Content of Selected Foods – Release 3.2. Maryland: US Department of Agriculture; 2015. 28. Harttig U, Haubrock J, Knuppel S, Boeing H; EFCOVAL Consortium. The MSM program: web-based statistics package for estimating usual dietary intake using the Multiple Source Method. Euro J Clin Nutr. 2011;65(Suppl 11):S87-91. 29. Haubrock J, Nöthlings U, Volatier JL, Dekkers A, Ocké M, Harttig U, et al. Estimating usual food intake distributions by using the multiple source method in the EPIC-Potsdam Calibration Study. J Nutr. 2011;141(5):914-20. 30. Verly E Jr, Castro MA, Fisberg RM, Marchioni DM. Precision of usual food intake estimates according to the percentage of individuals with a second dietary measurement. J Acad Nutr Diet. 2012;112(7):1015-20. 31. Sartorelli DS, Barbieri P, Perdoná GC. Fried food intake estimated by the multiple source method is associated with gestational weight gain. Nutr Res. 2014;34(8):667-73. 32. Miranda AM, Steluti J, Fisberg RM, Marchioni DM. Dietary intake and food contributors of polyphenols in adults and elderly adults of Sao Paulo: a population-based study. Br J Nutr. 2016;115(6):1061-70. 33. Kim K, Vance TM, Chun OK. Greater flavonoid intake is associated with improved CVD risk factors in US adults. Br J Nutr. 2016;115(8):1481-8.

17. Wedick NM, Pan A, Cassidy A, Rimm EB, Sampson L, Rosner B, et al. Dietary flavonoid intakes and risk of type 2 diabetes in US men and women. Am J Clin Nutr. 2012;95(4):925-33.

34. Badshah H, Ullah I, Kim SE, Kim TH, Lee HY, Kim MO. Anthocyanins attenuate body weight gain via modulating neuropeptide Y and GABAB1 receptor in rats hypothalamus. Neuropeptides. 2013;47(5):347-53.

18. Santangelo C, Vari R, Scazzocchio B, Filesi C, Masella R. Management of reproduction and pregnancy complications in maternal obesity: Which role for dietary polyphenols? Biofactors. 2013;40(1):79-102.

35. Takahashi A, Shimizu H, Okazaki Y, Sakaguchi H, Taira T, Suzuki T, et al. Anthocyanin-rich phytochemicals from Aronia Fruits inhibit visceral fat accumulation and hyperglycemia in high-fat diet-induced dietary obese rats. J Oleo Sci. 2015;64(12):1243-50.

248

Arch Endocrinol Metab. 2019;63/3

Flavonoid intake and gestational diabetes

37. Hahn M, Baierle M, CharĂŁo MF, Bubols GB, Gravina FS, Zielinsky P, et al. Polyphenol-rich food general and on pregnancy effects: a review. Drug Chem Toxicol. 2017;40(3):368-74.

36. Kac G, Nucci LB, Spyrides MH, Duncan BB, Schmidt MI. Evaluation of the ability of a Latin American gestational weight curve to predict adverse pregnancy outcomes. Int J Gynaecol Obstet. 2009;106(3):223-6.

Arch Endocrinol Metab. 2019;63/3

249

original article

Searching for mutations in the HNF1B gene in a Brazilian cohort with renal cysts and hyperglycemia Disciplina de Endocrinologia, Departamento de Medicina, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil 2 Grupo de Diabetes Monogênico, Unidade de Endocrinologia Genética/LIM25, Faculdade de Medicina, Universidade de São Paulo (FMUSP), São Paulo, SP, Brasil 3 Disciplina de Nefrologia, Departamento de Medicina, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil 4 Departamento de Ciências da Vida, Universidade do Estado da Bahia (UNEB), Salvador, BA, Brasil 1

Correspondence to: Fernando M. A. Giuffrida Universidade do Estado da Bahia, Departamento de Ciências da Vida Rua Silveira Martins, 2555, Cabula 41150-000 – Salvador, BA, Brasil fernando.giuffrida@me.com Received on July/5/2018 Accepted on Feb/13/2019 DOI: 10.20945/2359-3997000000138

Renata P. Dotto1, Lucas Santos de Santana2, Susan C. Lindsey1, Lilian Araújo Caetano2, Luciana F. Franco1, Regina Célia M. S. Moisés1, João R. Sa1, José Luiz Nishiura3, Milena Gurgel Teles2, Ita P. Heilberg3, Magnus R. Dias-da-Silva1, Fernando M. A. Giuffrida1,4, André F. Reis1

ABSTRACT Objective: To verify the presence of variants in HNF1B in a sample of the Brazilian population selected according to the presence of renal cysts associated with hyperglycemia. Subjects and methods: We evaluated 28 unrelated patients with clinical suspicion of HNF1B mutation because of the concomitant presence of diabetes mellitus (DM) or prediabetes and renal cysts. Genotyping was accomplished using Sanger sequencing or multiplex ligation-dependent probe amplification (MLPA). In positive cases, available relatives were recruited. Results: We found two patients with HNF1B mutations. The first presented the variant p.Pro328Leufs*48(c.983delC) and had DM, renal cysts, and hypomagnesemia. The second presented a heterozygous whole gene deletion in HNF1B, DM, renal cysts, body and tail pancreatic agenesis, and hypomagnesemia; this alteration was also found in his two siblings and his father. Conclusion: The recruitment of suspected cases of HNF1B gene mutations in Brazilians due to hyperglycemia and renal cysts presents two positive cases. Our cases contribute to the annotation of clinical and biochemical phenotypes of this rare form of maturityonset diabetes of the young (MODY). Arch Endocrinol Metab. 2019;63(3):250-7 Keywords HNF1B; MODY; monogenic diabetes; diabetes

INTRODUCTION

aturity-onset diabetes of the young (MODY) is characterized by the occurrence of earlyonset diabetes mellitus (DM), most often before age 25, with dominant autosomal inheritance caused by primary defects in insulin secretion. MODY is currently known to comprise a group of diseases caused by at least 14 different genetic etiologies. In rarer forms, identification of additional families demonstrating the co-segregation of the genetic variation with DM is necessary to confirm that they may be effectively regarded as MODY genes (1,2). The HNF1B gene was associated with cystic kidney disease shortly after being first described as a cause of MODY by Horikawa and cols. (3). This disease was initially called renal cysts and diabetes (RCAD) 250

syndrome (4). However, with the growth in the number of published cases, its broad clinical heterogeneity has been increasingly recognized. It is noteworthy that not all patients carrying an HNF1B mutation have renal cysts or DM (5). Given that HNF1B is expressed in many organs, predominantly in the liver, intestines, kidneys, pancreas, and urogenital tract, phenotypes in these organs are the most common and best described (6). Due to the great clinical heterogeneity among affected patients, various recruitment strategies for patients with suspected HNF1B mutations have been proposed with varying accuracy levels (7-9). The objective of our study was to verify the presence of variants in the HNF1B gene in a sample of the Brazilian population selected according to the presence of renal cysts and hyperglycemia. Arch Endocrinol Metab. 2019;63/3

HNF1B mutations in a Brazilian cohort

Subjects We recruited twenty-eight patients with clinical suspicion of HNF1B gene mutations from the Nephrology and Diabetes outpatient clinics at Universidade Federal de São Paulo (Unifesp), and the Diabetes Unit at Universidade de São Paulo (USP). These patients presented either DM or prediabetes as per American Diabetes Association criteria (10) and renal cysts identified by ultrasound or computed tomography (CT) scans during routine medical care. Nineteen patients had autosomal dominant polycystic kidney disease (ADPKD) based on clinical phenotype, defined by number of renal cysts according to age and one affected parent (11), whereas the other nine had simple renal cysts. The clinical and biochemical features of these patients were collected from their medical records. In the case of patients with identified HNF1B variants, relatives were invited to participate in the study. Patients with type 1 DM, i.e., those with one or more positive autoantibodies against GAD, IA2, or insulin (measured only in individuals without insulin use) and undetectable fasting C-peptide levels, were excluded. Patients were classified as having arterial hypertension and/or dyslipidemia (12). After defining the results with the completed genotypes, we were interested in verifying the accuracy of the Faguer score in our sample. In brief, in this paper’s authors devised a score using a weighted combination of the most discriminative clinical, laboratory, and radiological characteristics of patients from published literature (13). The study was approved by Hospital São Paulo’s ethics committee, and all patients provided informed consent.

Analysis of the 9 exons of HNF1B and the promoter region was performed by DNA amplification through polymerase chain reaction followed by Sanger sequencing using a Big Dye TerminatorTM Cycle Sequencing Ready Reaction Kit and ABI PRISM 3130 xl Genetic Analyzer (Applied Biosystems, Foster City, CA, USA), as previously described (15). Primers used are detailed in Supplementary Table 1. Cloning experiments to confirm the mutations were performed using a TOPO TA CloningTM Kit for Sequencing (Invitrogen/Thermo ScientificTM, Carlsbad, CA, USA) in accordance with the manufacturer’s instructions. Plasmid DNA was then extracted using PureLink Quick Plasmid MiniPrep kits (Invitrogen/Thermo Scientific, Grand Island, NY, USA) or QIAfilter Plasmid Maxi (QIAgenTM, Hilden, Germany) and submitted to direct sequencing as described above.

Statistical analysis Continuous variables were expressed in mean ± SD. Dichotomous variables were described as percentages.

RESULTS Clinical and biochemical characteristics of recruited patients are described in Table 1. Twenty-eight patients with DM or prediabetes and renal cysts were studied. Six (21%) were on insulin therapy and 7 (25%) were on oral antidiabetic drugs. Twenty-six patients had hypertension (92%), 2 (7%) had dyslipidemia, and 2 (7%) had hypomagnesemia. Table 1. Main clinical and laboratory characteristics of studied patients (n = 28) Age (yrs) DM/prediabetes (%)

Genetic analyses Genomic DNA was extracted from peripheral blood using an in-house method (14). Next, genetic analyses were performed in two steps: 1) multiplex ligation dependent probe amplification (MLPA) and 2) Sanger sequencing. MLPA was performed using SALSA® MLPA® Probemix P241-E1 MODY Mix 1 with 4 healthy controls in each run. We then analyzed the data using the CoffalyserTM software, following the manufacturer’s protocol and performing intra- and intersample normalization. Arch Endocrinol Metab. 2019;63/3

52.6 ± 12.1 38/62

IMC (kg/m2)

27.7 ± 6.5

Fasting glucose (mg/dL)

129.5 ± 38

HbA1c (%)

7±2

HbA1c (mmol/mol)b Creatinine (mg/dL)

1.3 ± 0.5

eGFR (mL/min)

67.6 ± 27

Magnesium (mg/dL)

1.8 ± 0.4

Uric Acid (mg/dL)

6.3 ± 1.6

Hypertension (%)

Dyslipidemia (%)

DM and prediabetes according to ADA criteria; variables expressed in mean+SD. a

53.9 ± 22.1

HbA1c measured by HPLC. Continuous

251

SUBJECTS AND METHODS

HNF1B mutations in a Brazilian cohort

We found two patients carrying HNF1B variants (Figure 1 and Table 2). In addition, a few intronic SNPs and a few variants in the promoter region were found, but they present a high minor allele frequency, indicating that they are variants commonly found in the general population and, thus, do not suggest a relevant impact. The first patient with an HNF1B variant was a 38year old male with a BMI of 19.4 (kg/m2) who was hospitalized due to headache. He was diagnosed with DM at age 12, had been using NPH insulin since one year after diagnosis, and had arterial hypertension. He presented with a fasting glucose of 165 mg/dL/9.2 mmol/L, HbA1c 13.2%/121 mmol/mol, serum creatinine 1.2 mg/dL, estimated Glomerular Filtration Rate (eGFR) 76.2 mL/min, and magnesium 1.0 mg/dL (normal range: 1.7-2.6 mg/dL). He had negative pancreatic autoantibodies (GAD, IA2, and anti-insulin) and fasting C-peptide of 0.9 ng/mL. He carried HNF1B variant p.Pro328Leufs*48 (c.983delC), which was identified by Sanger sequence and cloning. This variant was described by Clissold and cols. (16). There were no available relatives for investigation (both parents had died from unknown causes), but the patient denied any familial history of DM or renal cysts.

The other case was a 36-year old male with a BMI of 22.9 (kg/m2) in whom MLPA demonstrated heterozygous whole gene deletion of HNF1B (p.Met1_Trp557del). He had had DM since age 13 and presented with a fasting glucose of 231 mg/dL /12.8 mmol/L, HbA1c 7.1%/54 mmol/mol, serum creatinine 1.17 mg/dL, eGFR 92.2 mL/min, and magnesium 0.9 mg/dL. He had been using NPH insulin and regular insulin. He had negative antibodies against GAD and IA2 and fasting C-peptide of 1.4 ng/mL. Imaging studies demonstrated body and tail pancreatic agenesis. The same mutation was identified in his father, who had DM, renal cysts, and agenesis of the pancreatic tail, and in his two normoglycemic siblings, one brother with body and tail pancreatic agenesis and renal cysts and one sister with pancreatic body hypoplasia and tail agenesis, renal cysts, and bicornuate uterus. These cases illustrate the clinical heterogeneity of HNF1B phenotypes that can be seen even within the same family. After analyzing the whole sample, we applied the score devised by Faguer and cols. All 5 positive cases (including all studied family members) had a minimum score of 8 points, which is suggested by the above cited reference as the optimal cutoff threshold for the negative predictive value to rule out HNF1B mutations in a suspected individual (Table 2).

Family A

Family B

I 1

II 3 Subject 3 • 12 yrs / 38 yrs • Renal cysts • Hypomagnesemia • Hypertension • Hypertrophic cardiomyopathy • p.Pro328Leufs*48(c.983delC) Legend: Index case DM+ / HNF1B +

DM- / HNF1B + DM+ / HNF1B- (Phenocopy)

1 Subject 3 • 55 yrs / 62 yrs • Renal cysts • Tail pancreatic agenesis • Hypertension • Dyslipidemia • p.Met1_Trp557del

II 3

III 5

Subject 5 Subject 6 • NA / 38 yrs • 13 yrs / 36 yrs • Renal cysts • Renal cysts • Body and tail • Hypomagnesemia pancreatic agenesis • Body and tail • Hypertension pancreatic agenesis • Dyslipidemia • Hypertension • p.Met1_Trp557del • Dyslipidemia • p.Met1_Trp557del

7 Subject 7 • 28 yrs • Renal cysts • Body hypoplasia and tail agenesis of the pancreas • Bicornuate uterus • Dyslipidemia • Hypomagnesemia • p.Met1_Trp557del

Figure 1. Pedigrees of families with alterations in the HNF1B gene. The two index cases described in the main text are indicated by arrows in the figure. Genotype and diabetes phenotype are indicated as described in the legend box in the figure. Text bullets alongside each of the described individuals show subject number in the pedigree with initials, age at diagnosis of diabetes/current age, associated phenotypes, and mutation found in each individual. 252

Arch Endocrinol Metab. 2019;63/3

62/55

38/NA

28/NA

B-3 (father)

B-5 (brother)

B-7

19.5

25.2

22.7

22.9

19.4

BMI (kg/m2)

148

231

165

Fasting glucose (mg/dL)

4.3 /23

5.1/32

9.5/ 80

7.1 / 54

13.2 / 121

HbA1c (%)/ mmol/mol

21 / 0.66

56 / 1.89

39 / 1.22

60 / 1.17

43 / 1.2

Urea / creatinine (mg/dL)

139

50.9

92.2

76.2

eGFR (mL/ min)a

eGFR (mL/min): methods CKD-EPI; b Imaging exams: ultrasound or tomography; NA: not applicable.

36/13

B-6 (index)

38/12

Age/Age at diagnosis (y)

A-3 (index)

Patient

0.9

1.42

1.7

0.9

Magnesium (mg/dL)

2.9

11.1

5.6

5.9

Uric Acid (mg/dL)

Table 2. Individual clinical and laboratory characteristics of patients who presented with HNF1B alterations

1.9

2,6

1.4

0.9

Fasting C-Peptide (ng/mL)

Negative

GAD / IA2 antibodies

No DM

Family history

Dyslipidemia Hypomagnesemia

Hypertension, Dyslipidemia

Exocrine pancreatic insufficiency; hypertension; Dyslipidemia; Subclinical hypothyroidism (positive thyroperoxidase antibodies) Hypomagnesemia

Hypertension, Hypertrophic cardiomyopathy, hypomagnesemia

Comorbidities

Faguer Score

Body hypoplasia and agenesis of the tail of the pancreas; Bilateral renal cortical cysts; Bicornuate uterus

Agenesis of body and tail of pancreas; Bilateral renal cysts

Agenesis of the pancreatic tail; Peripyelic cyst in the middle third of the right kidney

Agenesis of body and tail of pancreas; Bilateral renal cortical cysts

Bilateral renal cysts

Imaging examsb

NPH and Regular Insulin + Gliclazide

NPH + Regular Insulin

NPH insulin

Current treatment DM

HNF1B mutations in a Brazilian cohort

253

HNF1B mutations in a Brazilian cohort

DISCUSSION To the best of our knowledge, this is the first systematic investigation of HNF1B variants in Brazilians with hyperglycemia and renal cysts, phenotypes usually seen in patients carrying variants in this gene (17). We detected 2 out of 28 cases (7%). Previously, we described an isolated case of HNF1B mutation in a Brazilian patient with familial DM and hypomagnesemia recruited due to suspicion of MODY (15). The prevalence of mutations in HNF1B is variable, mainly due to the high degree of clinical heterogeneity and various recruitment strategies regarding age range and distinct phenotypes. Kanthimathi and cols. studied 50 Indian patients selected according to the presence of renal disease (such as renal cysts, focal segmental glomerulosclerosis, or tubulointerstitial nephritis) and DM and found HNF1B variants in 6 individuals (12%) (9). Horikawa and cols. studied a total of 230 nonobese Japanese patients aged ≤ 35 years with negative autoantibodies against beta cells and found 6 with mutations in HNF1B (2.6%) (18). Edghill and cols. studied 133 patients from the UK with renal disease of unknown etiology and found variants in 15 patients (11%) (7). Heidet and cols., selecting 377 patients with hyperechogenic multicystic kidney disease or other morphological and functional renal alterations, identified mutations in 75 cases (~20%), of whom 5 (7%) presented with DM (8). Ulinski and cols. investigated 80 children using only sonographic criteria at the first postnatal examination, including those with a renal phenotype, such as unilateral multicystic dysplasia, uni- or bilateral cystic kidneys, uni- or bilateral renal hypo/dysplasia, and single kidney, and found HNF1B variations in 25 patients (31.2%) (19). Conversely, in the present series, the majority of cases consisted of ADPKD, a monogenic renal disease with an autosomal dominant inheritance characterized by the progressive development of cysts and considered an important cause of end-stage renal disease. It is worth mentioning that a clinical suspicion of ADPKD, coupled with ultrasound imaging according to criteria of Pei and cols. (11) and confirmed by family history, is enough to establish the diagnosis of ADPKD with no further need for molecular analysis suggesting either PKD1, PKD2, or the more rare GANAB gene mutations (20,21). However, experimental data have suggested that kidney-specific inactivation of HNF1B leads to a polycystic phenotype and that the expression 254

of several genes involved in cystic diseases is severely affected in conditionally inactivated mice, as evidenced by chromatin immunoprecipitation showing that 3 of those genes (Umod, Pkhd1 and Pkd2) are directly controlled by HNF1B (22). That said, we selected our patients based on the presence of renal cysts because this phenotype is one of the most studied in individuals with HNF1B gene mutations. Cysts have been described in patients with HNF1B gene mutations since the earliest descriptions by Bingham and cols. (4), as well as in several other reports (7,19,23,24). In 2014, Faguer and cols. proposed a score to improve accuracy in patient selection using several clinical, radiologic, and laboratory phenotypes from patients with suspected variations in HNF1B. They analyzed data from 433 patients, including 56 with mutations (13%), and the presence of renal cysts scored high (4 points for each affected kidney) in the identification of such mutations (13). A cutoff point of 8 was the optimal value, with a sensitivity of 98.2%, a specificity of 41.1%, a positive predictive value (PPV) of 19.8%, and a negative predictive value (NPV) of 99.4% (13). Furthermore, Clissold and cols. analyzed the performance of the Faguer score in another cohort of 686 patients from the UK (177 positive cases – 25.8%). The authors found that this score discriminated adequately between patients with and without mutations with area under ROC curve of 0.72 but with a lower NPV (85%), as compared to Faguer’s results (99.4%) (25). Unfortunately, in the Brazilian public health system, where our patients were assisted, data for several phenotypes suggested by Faguer’s score are not routinely available, although in a research context we were able to perform the necessary tests on the sample described here. Even with this important limitation, the score seems to provide good discrimination in our population, taking into account that all positive cases had scores of 8 or higher. On the other hand, according to Verhave and cols. (5), due to the functional promiscuity of the HNF1B transcription factor involved in the development of many organs, HNF1B-associated kidney disease may encompass more than DM and renal cysts. It is noteworthy that data from Faguer and cols. (13) was from a younger population than ours (mean age of 17 years), including a large number of children with several other anatomical abnormalities of the urinary tract. Although 11%-30% of the sample had uniArch Endocrinol Metab. 2019;63/3

or bilateral cysts, they did not characterize polycystic kidney disease. Additionally, patients with suspected ADPKD without a family history are more likely to have mutations in HNF1B (5). Thus, if we did a sub-analysis of our sample, excluding the majority that already had a defined phenotypic diagnosis of ADPKD (19 patients), our positivity rate could reach 22%. Furthermore, Faguer and cols. suggested that this score should be assessed after ruling out recognizable inherited renal disease such as ADPKD, among others (13). Concerning mutation type, the occurrence of whole-gene heterozygous deletions is very frequent, reaching around 50% or more of previously described cases. This whole-gene deletion occurs in the context of a chromosomal microdeletion at 17q12 encompassing 14 genes in addition to HNF1B (9,17,18,23,26,27). Use of MLPA is necessary because Sanger sequencing is not able to identify this type of mutation. Other types of mutations, such as point mutations or small deletions/insertions, are located primarily in the DNA binding domain (13). Furthermore, de novo mutations are common, found in at least ~50% of cases (5,7,19). Our data were similar to those observations, although with a limited number of cases. Our previously published case was a de novo heterozygous whole-gene deletion (15). Of the other 2 cases reported here, the first has a frame shift mutation that is possibly de novo because the patient’s parents had no history of DM and renal disease, but this is not possible to define because the parents have passed away and there is no available DNA for testing. The other case had a heterozygous whole gene deletion inherited from his father. In this case, we tested the paternal grandmother, who had no history of DM, and the result was negative. The paternal grandfather had died, so it is not possible to determine if the patient’s father had a de novo mutation. Few studies have examined the promoter region of the HNF1B gene (9). In the present study, we analyzed this region, but we did not identify any potentially functional variant. We chose to recruit patients with hyperglycemia not only within the diabetic range, but also with prediabetes. DM is present in about 45% to 80% of cases and has high specificity in the Faguer score (13). It is important to take into account that HNF1B mutations are not usually associated with DM in childhood, this being a later phenotype (6,8). A French group studied the diabetes phenotype of patients with HNF1B mutations in more detail. They analyzed 201 adult patients, 159 Arch Endocrinol Metab. 2019;63/3

of whom had DM. Among them, 40% had a familial history of DM suggesting HNF1B-MODY. The age at diagnosis was > 25 years in 57% of the cases, and 80% had BMI < 25 kg/m2. A large portion (80% of cases) had residual pancreatic endocrine function measured by C-peptide even almost 10 years after diagnosis. About 60% of the patients were responsive to the use of sulfonylureas or repaglinide. However, a few years after diagnosis, almost 80% of patients were on insulin therapy for metabolic control. Importantly, the presence of retinopathy and neuropathy was observed in a significant number of cases, and the association of cardiovascular risk factors, such as renal failure, hypertension, and dyslipidemia, was seen in 40% of the cases. In addition, the French researchers noted a genotype-phenotype correlation where those with an entire gene deletion had lower BMI at diagnosis and were more frequently treated with insulin. Interestingly, in this study, the genotype was also associated with different renal function. Compared with point mutations, HNF1B deletions were independently associated with normal renal function at DM diagnosis and at follow-up. (17). Similar results were found in other two previous studies (8,27). Some hypotheses could be raised based on these observations, as other genes affected by whole HNF1B gene deletions could also have a role in the clinical differences (17,28,29). Regarding other pancreatic phenotypes, Faguer and cols. demonstrated a high specificity for pancreas hypoplasia and/or exocrine pancreas insufficiency, each phenotype scoring 4 points on Faguer’s scale. Around 11% of their patients presented some of these pancreatic abnormalities (13). As for our imaging exams, we had 4 patients with agenesis of the pancreatic body and/ or tail, including the positive relatives (80%). In the Dubois-Laforgue cohort, 59 individuals (62%) had morphological abnormalities in the pancreas (17). Clissold and cols. found that, in their cohort, about 4% had pancreatic hypoplasia or exocrine failure. These pancreatic phenotypes were significantly associated with HNF1B gene mutations in adults but not in children (25). Hypomagnesemia seems to be a good discriminator for HNF1B mutations. Van der Made and cols. found HNF1B gene mutations in 3 cases with hypomagnesemia as the initial and predominant symptom (29). In general, the prevalence of hypomagnesemia ranges between 25% and 75% and is slightly lower in pediatric cohorts (5,13,16,17,29,30). Adalat and cols. suggested that the 255

HNF1B mutations in a Brazilian cohort

HNF1B gene is fundamental for the transcription of the FXYD2 gene, encoding the γ-subunit of the sodiumpotassium ATPase expressed in the distal convoluted tubule, which is thought to play a role in transcellular magnesium reabsorption (30). Furthermore, in the Faguer score, hypomagnesemia scores 2 points (13). In our study, including the positive relatives, 3 patients had hypomagnesemia (60%) Our patients, including the positive relatives, also had a high frequency of hypertension and dyslipidemia (80%), whereas in the study by Dubois-Laforgue and cols., 75 (37%) had hypertension and 78 (39%) had dyslipidemia (17). The reason for these associations is not clear and needs to be investigated further. It could be a bias due to the limited number of studied patients, or it could reflect the usual association between these co-morbidities and DM present in the majority of our cases. Finally, we would like to point out some limitations of the present study, such as the small sample size, which prevented a more definitive analysis of the real accuracy of our recruitment strategy for HNF1B gene positive cases. In addition, we used prediabetes, a mild phenotype of hyperglycemia, as an inclusion criterion. From the total number of cases recruited, 62% (n = 18) would fall into this category. If we had selected only those with DM (n = 10), our positivity rate would have been 20% and may have increased the rate of positive cases. In conclusion, the frequency of mutations in HNF1B gene in Brazilian patients selected for renal cysts and hyperglycemia was 7%. Further studies are needed to optimize patient selection criteria for genetic testing, especially in our population. Acknowledgements: we are indebted to Professors Luiz Fernando Onuchic and Manoel Rocha from the Departments of Molecular Medicine and Radiology (Faculdade de Medicina, Universidade de São Paulo) for their assistance in the assessment of renal cysts and pancreatic images. Our funding sources include Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) grants 2013/19920-2 (MGT) and 2015/05123-9 (AFR). Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) grants 160044/2013-8 (RPD) and 454014/2014-7 (FMAG). Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Capes (LAC and LSS).

Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Giuffrida FMA, Reis AF. Genetic and clinical characteristics of maturity-onset diabetes of the young. Diabetes Obes Metab. 2005;7(4):318-26.

256

2. Bonnefond A, Philippe J, Durand E, Dechaume A, Huyvaert M, Montagne L, et al. Whole-Exome Sequencing and High Throughput Genotyping Identified KCNJ11 as the Thirteenth MODY Gene. PLoS One. 2012;7(6):e37423. 3. Horikawa Y, Iwasaki N, Hara M, Furuta H, Hinokio Y, Cockburn BN, et al. Mutation in hepatocyte nuclear factor-1 beta gene (TCF2) associated with MODY. Nat Genet. 1997;17(4):384-5. 4. Bingham C, Bulman MP, Ellard S, Allen LI, Lipkin GW, Hoff WG, et al. Mutations in the hepatocyte nuclear factor-1beta gene are associated with familial hypoplastic glomerulocystic kidney disease. Am J Hum Genet. 2001;68(1):219-24. 5. Verhave JC, Bech AP, Wetzels JFM, Nijenhuis T. Hepatocyte Nuclear Factor 1β-Associated Kidney Disease: More than Renal Cysts and Diabetes. J Am Soc Nephrol. 2016;27(2):345-53. 6.

Bockenhauer D, Jaureguiberry G. HNF1B-associated clinical phenotypes: the kidney and beyond. Pediatr Nephrol. 2016;31(5):707-14.

7. Edghill EL, Oram RA, Owens M, Stals KL, Harries LW, Hattersley AT, et al. Hepatocyte nuclear factor-1beta gene deletions – a common cause of renal disease. Nephrol Dial Transplant. 2008;23(2):627-35. 8. Heidet L, Decramer S, Pawtowski A, Morinière V, Bandin F, Knebelmann B, et al. Spectrum of HNF1B mutations in a large cohort of patients who harbor renal diseases. Clin J Am Soc Nephrol. 2010;5(6):1079-90. 9. Kanthimathi S, Balamurugan K, Mohan V, Shanthirani CS, Gayathri V, Radha V. Identification and molecular characterization of HNF1B gene mutations in Indian diabetic patients with renal abnormalities. Ann Hum Genet 2015;79(1):10-9. 10. American Diabetes Association. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2018. Diabetes Care. 2018;41(Suppl 1):S13-S27. 11. Pei Y, Obaji J, Dupuis A, Paterson AD, Magistroni R, Dicks E, et al. Unified Criteria for Ultrasonographic Diagnosis of ADPKD. J Am Soc Nephrol. 2009;20(1):205-12. 12. American Diabetes Association. 9. Cardiovascular Disease and Risk Management: Standards of Medical Care in Diabetes-2018. Diabetes Care. 2018;41(Suppl 1):S86-S104. 13. Faguer S, Chassaing N, Bandin F, Prouheze C, Garnier A, Casemayou A, et al. The HNF1B score is a simple tool to select patients for HNF1B gene analysis. Kidney Int. 2014;86(5):1007-15. 14. Kizys MML, Cardoso MG, Lindsey SC, Harada MY, Soares FA, Melo MCC, et al. Optimizing nucleic acid extraction from thyroid fine-needle aspiration cells in stained slides, formalin-fixed/paraffin-embedded tissues, and long-term stored blood samples. Arq Bras Endocrinol Metabol. 2012;56(9):618-26. 15. Dotto RP, Giuffrida FMA, Franco L, Mathez ALG, Weinert LS, Silveiro SP, et al. Unexpected finding of a whole HNF1B gene deletion during the screening of rare MODY types in a series of Brazilian patients negative for GCK and HNF1A mutations. Diabetes Res Clin Pract. 2016;116:100-4. 16. Clissold RL, Hamilton AJ, Hattersley AT, Ellard S, Bingham C. HNF1B-associated renal and extra-renal disease-an expanding clinical spectrum. Nat Rev Nephrol. 2015;11(2):102-12. 17. Dubois-Laforgue D, Cornu E, Saint-Martin C, Coste J, BellannéChantelot C, Timsit J, et al. Diabetes, Associated Clinical Spectrum, Long-term Prognosis and Genotype/Phenotype Correlations in 201 Adult Patients With Hepatocyte Nuclear Factor 1 B (HNF1B) Molecular Defects. Diabetes Care. 2017:dc162462. 18. Horikawa Y, Enya M, Fushimi N, Fushimi Y, Takeda J. Screening of diabetes of youth for hepatocyte nuclear factor 1 mutations: clinical phenotype of HNF1β-related maturity-onset diabetes of the young and HNF1α-related maturity-onset diabetes of the young in Japanese. Diabet Med. 2014;31(6):721-7. 19. Ulinski T, Lescure S, Beaufils S, Guigonis V, Decramer S, Morin D, et al. Renal phenotypes related to hepatocyte nuclear factorArch Endocrinol Metab. 2019;63/3

HNF1B mutations in a Brazilian cohort

1beta (TCF2) mutations in a pediatric cohort. J Am Soc Nephrol. 2006;17(2):497-503. 20. Cornec-Le Gall E, Torres VE, Harris PC. Genetic Complexity of Autosomal Dominant Polycystic Kidney and Liver Diseases. J Am Soc Nephrol. 2018;29(1):13-23. 21. Harris PC, Rossetti S. Molecular diagnostics for autosomal dominant polycystic kidney disease. Nat Rev Nephrol 2010;6(4):197-206. 22. Gresh L, Fischer E, Reimann A, Tanguy M, Garbay S, Shao X, et al. A transcriptional network in polycystic kidney disease. EMBO J. 2004;23(7):1657-68. 23. Raaijmakers A, Corveleyn A, Devriendt K, van Tienoven TP, Allegaert K, Van Dyck M, et al. Criteria for HNF1B analysis in patients with congenital abnormalities of kidney and urinary tract. Nephrol Dial Transplant. 2015;30(5):835-42.

26. Edghill EL, Stals K, Oram RA, Shepherd MH, Hattersley AT, Ellard S. HNF1B deletions in patients with young-onset diabetes but no known renal disease. Diabet Med. 2013;30(1):114-7. 27. Clissold RL, Shaw-Smith C, Turnpenny P, Bunce B, Bockenhauer D, Kerecuk L, et al. Chromosome 17q12 microdeletions but not intragenic HNF1B mutations link developmental kidney disease and psychiatric disorder. Kidney Int. 2016;90(1):203-11. 28. Clissold RL, Harries LW, Ellard S, Bingham C, Hattersley AT. Comment on Dubois-Laforgue et al. Diabetes, Associated Clinical Spectrum, Long-term Prognosis, and Genotype/Phenotype Correlations in 201 Adult Patients With Hepatocyte Nuclear Factor 1B (HNF1B) Molecular Defects. Diabetes Care 2017;40:1436-1443. Diabetes Care. 2018;41(1):e7. 29. van der Made CI, Hoorn EJ, la Faille de R, Karaaslan H, Knoers NVAM, Hoenderop JGJ, et al. Hypomagnesemia as First Clinical Manifestation of ADTKD-HNF1B: A Case Series and Literature Review. Am J Nephrol. 2015;42(1):85-90.

25. Clissold R, Shields B, Ellard S, Hattersley A, Bingham C. Assessment of the HNF1B Score as a Tool to Select Patients for HNF1B Genetic Testing. Nephron. 2015;130(2):134-40.

30. Adalat S, Woolf AS, Johnstone KA, Wirsing A, Harries LW, Long DA, et al. HNF1B mutations associate with hypomagnesemia and renal magnesium wasting. J Am Soc Nephrol. 2009;20(5):1123-31.

24. Faguer S, Decramer S, Devuyst O, Lengelé J-P, Fournié GJ, Chauveau D. Expression of renal cystic genes in patients with HNF1B mutations. Nephron Clin Pract. 2012;120(2):c71-8.

Arch Endocrinol Metab. 2019;63/3

257

original article

Arterial stiffness in hyperthyroid patients is deteriorated due to thyroid hormones Canan Yildiz1, Mustafa Altay2, Sedat Yildiz3, Yavuz Çag6ir4, Tolga Akkan5, Yasemin Aydog6an Ünsal6, Esin Beyan6

ABSTRACT Didim State Hospital, Aydin, Turkey 2 Department of Endocrinology and Metabolism, Sag6lik Bilimleri Üniversitesi (University of Health Sciences) Keçiören SUAM, Ankara, Turkey 3 Söke Fehime Faik Kocagöz State Hospital, Aydin, Turkey 4 Halil SÇivgın Çubuk State Hospital, Ankara, Turkey 5 Çaldiran District State Hospital, Van, Turkey 6 Department of Internal Medicine, Sag6lik Bilimleri Üniversitesi (University of Health Sciences) Keçiören SUAM, Ankara, Turkey 1

Correspondence to: Yasemin Aydogan Ünsal yaseminunsalay@gmail.com Received on Jul/26/2018 Accepted on Nov/24/2018 DOI: 10.20945/2359-3997000000135

Objective: The aim of this study is to evaluate and compare arterial stiffness, which is an independent risk indicator for cardiovascular diseases (CVDs), between patients with overt hyperthyroidism, subclinical hyperthyroidism, euthyroidism by antithyroid therapy and healthy volunteers with pulse wave analysis (PWA). Subjects and methods: A total of 102 volunteers were included in the study (30 in the overt hyperthyroid group, 28 in the subclinical hyperthyroid group and 14 with euthyroidism by antithyroid therapy and 30 healthy). The arterial stiffness measurements of the participants in the study were performed with the Mobil-O-Graph PWA device (I.E.M. GmBH, Stolberg, Germany), which makes cuff based oscillometric measurement from the brachial artery. Results: Systolic blood pressure, pulse rate, central systolic blood pressure, cardiac output, heart rate-corrected augmentation index (Aix@75) and pulse wave velocity (PWV) measurements were significantly higher in the hyperthyroid group than in the control group. The heart rate and PWV in the subclinical hyperthyroid group were significantly higher than the control group. In the euthyroid group, systolic blood pressure, central systolic blood pressure, cardiac output, cardiac index and PWV were found significantly higher than the control group. There was also a negative correlation between Aix@75 and thyroid-stimulating hormone (TSH), and a positive correlation between Aix@75 and free thyroid hormones. Conclusion: In our study, we observed that the arterial stiffness was adversely affected by an overt or subclinical increase in thyroid hormones and this correlated with thyroid hormone levels. We recommend that PWV measurement, which is a simple method for detecting CVD risk, can be used in these patients. Arch Endocrinol Metab. 2019;63(3):258-64 Keywords Arterial stiffness; augmentation index; hyperthyroidism; pulse wave analysis

INTRODUCTION

he endocrine system has a wide effect on the cardiovascular system. In particular, the thyroid gland has a specific relationship with cardiovascular system physiology and pathology. Biologically active T3 affects cardiac contractility, heart rate, diastolic function and systemic vascular resistance through genomic and non-genomic effects. Thyroid hormones have the effect of reducing vascular tone on the vascular system and causing normal arteriolar remodeling to be carried out (1). In addition to the direct effects, the thyroid hormones also have indirect effects on cardiovascular function mediated by the autonomic nervous system, renin-angiotensin-aldosterone system, vascular compliance, vasoreactivity and renal function. As with many other systems, hyperthyroidism also reveals some clinical signs in the cardiovascular system. Some of these are systolic hypertension, increased left 258

ventricular mass, exercise intolerance, angina pectoris, and systolic murmur (2). Many invasive, noninvasive methods and scoring systems are currently used for predicting CVDs. Arterial stiffness is also considered an independent risk indicator for cardiovascular diseases (3-5). Increased stiffness in central arteries leads to undesired hemodynamic consequences such as the expansion of pulse pressure (PP) and increased flow to microcirculation. When the deleterious effects of these changes are evaluated, arterial stiffness can be considered as a predictor of cardiovascular risks (6). Numerous studies involving disease-specific and population-based cohorts have demonstrated that higher carotid-femoral pulse wave velocity (cfPWV) is associated with increased risk for primary and recurrent CVDs (7,8). This data supports the assessment of arterial stiffness as an independent CVD risk factor. Arch Endocrinol Metab. 2019;63/3

Arterial stiffness in hyperthyroidism

SUBJECTS AND METHODS This study was approved by the Research Ethics Committee of the University of Health Sciences, Keçiören SUAM (date: 02/02/2016 and reference number 126). All the methods used in the study were carried out in accordance with the approved guidelines and Declaration of Helsinki, Ethical Principles for Medical Research Involving Human Subjects. Patients who have been diagnosed with overt or subclinical hyperthyroidism, patients with euthyroidism who were previously diagnosed with hyperthyroidism and euthyroid healthy volunteers without systemic disease referred to our outpatient clinics for Endocrinology and Metabolic Diseases between 01/06/2016 and 01/12/2016 were included. Written informed consent was obtained from all individual participants included in the study. The inclusion criteria for overt and subclinical hyperthyroidism groups were 18 years old or over, not receiving anti-thyroid or levothyroxine therapy in the last 1 year. The inclusion criteria for the euthyroid group were 18 years old or over, being taken antithyroid therapy due to hyperthyroidism and being in the euthyroid state at least 3 months. Exclusion criteria for all groups were defined as the presence of chronic diseases (diabetes mellitus, hypertension, coronary artery disease, chronic kidney disease, subclinical or obvious hypothyroidism), pregnancy and lactation for all groups. Detailed medical histories of all participants were obtained, and physical examination of all participants were performed. The age and gender of these patients were learned and height and weight measurements were made. The body mass index (BMI) was calculated by the weight (kg)/height (m)2 formula. Arch Endocrinol Metab. 2019;63/3

Blood samples were taken in the morning hours after at least 10 hours of fasting. Free T3, free T4 and TSH levels were studied by chemiluminescence method with Abbott Architect i2000 device (Illinois, USA). It was defined as euthyroidism if the levels of TSH (0.35-4 mIU/mL), fT3 (1.71-4.71 pg/mL) and fT4 (0.8-1.9 ng/mL) were within the normal reference limits. Overt hyperthyroidism was defined as low TSH level with elevated free T3 and free T4 concentrations. Subclinical hyperthyroidism was defined as low TSH level with normal free T3 and free T4 concentrations. TSH Receptor Antibody (TRAB) measurements were performed by Beckman Coulter Immunotech device (Florida, USA) with immunoassay method. anti-thyroglobulin (Anti-TG) and anti-thyroid peroxidase (Anti-TPO) measurements were studied by chemiluminescence method with Abbott Architect i2000 device (Illinois, USA). These biochemical tests were performed for etiology in cases of overt and subclinical hyperthyroidism.

Evaluation of pulse wave analysis and arterial stiffness In this study, a PWA device, Mobil-O-Graph PWA/ ABPM (I.E.M. GmBH, Stolberg, Germany), which performs cuff based oscillometric measurement from the brachial artery, was used to collect data. The MobileO-Graph PWA device is a blood pressure measurement device approved by the British Hypertension Society (BHS) and the European Society of Hypertension (ESH). The measurement reliability of this device has been demonstrated by comparing data obtained by invasive catheterization, magnetic resonance imaging and tonometry devices (11). Participants were kept sitting in a quiet room for 15 minutes before measuring arterial stiffness. Suitable cuffs were selected for the arms of the participants. The cuff was placed just above the right arm elbow. Pulse wave velocity analysis was performed from the brachial artery 4 times at 5-minute intervals at the sitting position. A bluetooth connection was established between the device and the software which is developed to monitor the device’s data. Age, gender, height and weight information of all participants were entered into the software program. Firstly, blood pressure measurements were made with the device and the pressure, where the brachial artery was fully compressed, was determined. After 30 seconds of recovery, the brachial artery was completely re-compressed. When 259

Especially non-invasive approaches in the measurement of arterial stiffness have recently become the foreground. This non-invasive method is called PWA because the calculations are based on the movements of the peripheral pulse waves. PWA is a proven, practical method of detecting arterial stiffness indirectly (9,10). Relatively small number of studies have examined the relationship between thyroid hormones and arterial stiffness, and controversial results have been obtained in these studies. In the light of all this information, our aim is to evaluate and compare arterial stiffness, which is an independent risk indicator for CVDs, between patients with overt hyperthyroidism, subclinical hyperthyroidism, euthyroidism by antithyroid therapy and healthy volunteers, with pulse wave analysis (PWA).

Arterial stiffness in hyperthyroidism

pulse waves reached the point where there is no flow, it is detected and magnified by special sensors in the device. So that the peripheral pulse wave was obtained. The data obtained here is classified as very high quality and low quality by the software. Only very high-quality data were evaluated for the study. The central aortic pulse wave was calculated from measured brachial artery pulse wave by special software. The calculated PWA was completed with aortic pulse wave. As a result of these measurements; systolic blood pressure (SBP), diastolic blood pressure (DBP), mean blood pressure (MBP), heart rate (HR), PP, central SBP, central DBP, central PP, cardiac output (CO), peripheral vascular resistance, cardiac index, body surface, augmentation pressure (AP), augmentation index corrected based on pulse rate 75 beats/min (Aix@75), reflectance magnitude and PWV data were obtained. The obtained data was transferred to the software program and saved.

Statistical analysis SPSS 22.0 program was used for statistical analysis of the data. Statistically, p < 0.05 was considered significant. Descriptive statistics of patients and control groups were performed. Categorical values were reported by number and percentage. The Kolmogorov-Smirnov/ShapiroWilk’s test and histogram graphs of data were used to assess whether the data showed a normal distribution. Data that do not correspond to normal distribution are expressed as median and minimum-maximum values and the nonparametric analysis was performed. Mann-Whitney U test was used to evaluate countable data with non-normal distribution. The Kruskal-Wallis test was performed to compere biochemical values and PWA parameters among groups. Post hoc analysis was

performed and Bonferroni correction was made. The Chi-square test was used to compare the categorical variables. While investigating the associations between non-normally distributed and/or ordinal variables, the correlation coefficients and their significance were calculated using the Spearman correlation test.

RESULTS Thirty patients were included for each overt hyperthyroidism, subclinical hyperthyroidism and euthyroidism groups. Also, 30 volunteers were included in the control group. Patients who were diagnosed with high blood pressure and high blood glucose levels were excluded from the study even though they were not diagnosed during the initial evaluation. In conclusion, we continued to work with 30 patients in the overt hyperthyroid group, 28 in the subclinical hyperthyroid group and 14 in the euthyroid group and 30 healthy volunteers. Thyroid hormone levels and diagnoses of the patients participating in the study are given in Table 1. Control and patient groups were similar in terms of age, sex, and BMI (Table 2). SBP (p = 0.02), HR (p < 0.01), central SBP (p = 0.04), CO (p = 0.02), Aix@75 (p < 0.01), PWV (p < 0.01) showed the significant differences between the groups (Table 2). Other data were not statistically different between the groups. SBP, HR, central SBP, CO, Aix@75 and PWV measurements were significantly higher in the hyperthyroid group than in the control group. HR and PWV measurements were significantly higher in the subclinical hyperthyroid group than in the control group. In the euthyroid group, SBP, central SBP, CO and PWV measurements were significantly higher than the control group (Table 2).

Table 1. The median values of thyroid hormone levels and causes of hyperthyroidism among the groups Overt Hyperthyroidism (n = 30)

TSH (mIU/mL)

Subclinical Hyperthyroidism (n = 28)

Euthyroidism (n = 14)

Control (n = 30)

< 0.01(< 0.01-0.02)

0.17 (0.00-0.49)

1.35 (0.53-3.10)

1.51 (055-3.60)

fT4 (ng/mL)

2.17 (1.46-4.27)

1.13 (0.84-1.70)

1.13 (0.87-1.25)

1.01 (0.83-1.35)

fT3 (ng/mL)

7.45 (4.41-30)

3.18 (2.36-4.30)

2.88 (2.42-3.39)

2.92 (0.83-3.32)

Graves’ disease (n)

Toxic adenoma (n)

TMNG (n)

Others (n)

Others: subclinical granulomatous thyroiditis, Hashitoxicosis, postpartum thyroiditis, subacute lymphocytic thyroiditis. TSH: thyroid stimulating hormone; fT4: thyroxine; fT3: triiodothyronine; TRAB: TSH receptor antibody; Anti-TG: anti-thyroglobulin; Anti-TPO: anti-thyroid peroxidase; TMNG: toxic multinodular goiter. The data in parentheses show the minimum and maximum value range.

260

Arch Endocrinol Metab. 2019;63/3

Arterial stiffness in hyperthyroidism

Table 2. Demographic characteristics and PWA results (median values) in the study groups Overt Hyperthyroidism (n = 30)

Subclinical Hyperthyroidism (n = 28)

Female gender n (%)

22 (73.3%)

Age

32.5 (19-65)

Euthyroidism (n = 14)

Control (n = 30)

25 (89.3%)

12 (85.7%)

22 (73.3%)

35 (20-66)

41.87 (23-69)

31 (19-64)

BMI – kg/m

23.7 (15.7-40)

26.9 (17.3-35)

26.45 (18.7-35.6)

23.8 (18.3-43.8)

SBP – mmHg

120 (102-140)

115.5 (93-140)

125 (110-139)

114.5 (101-138)

0.02a

DBP – mmHg

73 (54-90)

71 (52-87)

75.5 (58-90)

69 (49-90)

MBP – mmHg

93.5 (78-119)

92.5 (70-116)

99 (82-111)

90 (73-112)

HR – beats/min

98.5 (70-117)

87 (56-110)

77 (50-91)

75 (53-96)

< 0.01b,c,d

PP – mmHg

46.5 (33-73)

45 (30-67)

49.5 (43-59)

44.5 (30-63)

cSBP – mmHg

108 (88-134)

107 (85-138)

116.5 (104-130)

104.5 (92-132)

0.04a

cDBP – mmHg

75 (57-90)

72.5 (53-88)

77.5 (59-90)

71.5 (50-90)

sPP – mmHg

34 (23-51)

35 (22-50)

41 (32-57)

36.5 (21-48)

CO – lt/min

4.85 (4.20-45)

4.70 (3.40-6.40)

5.40 (4.50-48)

4.65 (3.30-6.10)

0.02a

PR –smmHg/mL

1.18 (0.95-1.43)

1.17 (0.92-1.70)

1.15 (0.80-1.31)

1.20 (0.80-1.58)

CI – lt/min/m

2.80 (2.40-3.80)

2.85 (2.30-4.00)

2.95 (2.60-3.90)

2.70 (1.70-3.60)

AP – mmHg

9 (2-19)

8 (3-26)

8 (4-21)

7 (2-19)

Aix@75 (%)

35.5 (10-58)

29 (5-45)

20 (8-45)

25 (9-41)

< 0.01c,d

PWV – m/s

5.6 (4.8-9.7)

5.7 (4.5-10.3)

6.35 (4.8-10.1)

5.2 (4.5-8.5)

< 0.01a,b

Correlation of thyroid function tests and thyroid autoantibodies with PWA parameters was assessed. As a result of the analysis, negative correlation was found between Aix@75 and TSH (p < 0.001, r = -0.363); on the other hand, there was positive correlation between Aix@75 and fT3 and fT4 (relatively p = 0.004, r = 0.282 and p = 0.001, r = 0.335) (Figure 1).

DISCUSSION In our study, we found that there was an increase in the arterial stiffness and a significant correlation between Aix@75, the arterial stiffness parameter, and thyroid hormone levels. The relationship between thyroid hormone levels and arterial stiffness has become more current in recent years and has been investigated in some communitybased studies. For example; In the 10,027-person study by Wang and cols., the PWV measurements of 812 individuals were analyzed by an oscillometric method. Unlike similar studies, a negative correlation between PWV and free T4 was observed; but there Arch Endocrinol Metab. 2019;63/3

was no correlation with TSH and free T3. It was noted that the augmentation index was not examined in the study and the cross-sectional nature of the study was the weaknesses of the study. In the populationbased SardiNIA study, PWV and augmentation index measurements were performed by oscillometric method in participants, and as a result, the linear positive correlation between free T4 and PWV was observed. There was no significant correlation with TSH and free T3. It was reported that the increase in fT4 contributes to the increase in PWV, as well as the aging of the vascular system. Another point that was mentioned in the study was the relationship between hypothyroidism and increased arterial stiffness, which was reported in earlier studies, connected to dyslipidemia (12). Obuobie and cols., a first study of the relationship between thyrotoxicosis and arterial stiffness, found lower augmentation index in a small number (n = 20) of thyrotoxicosis cases than the control group during the diagnosis (13). At 6 months after treatment, when the cases were euthyroid, similar data were obtained compared to the control group. In this study, it was 261

NS: not significant; BMI: body mass index. SBP: systolic blood pressure; DBP: diastolic blood pressure; MBP: mean blood pressure; PP: pulse pressure; cSBP: central systolic blood pressure; cDBP: central diastolic blood pressure; CO: cardiac output; PR: peripheral resistance; CI: cardiac index; AP: augmentation pressure; Aix@75: augmentation index (corrected based on pulse rate 75 beats/ min); PWV: pulse wave velocity. a: between euthyroidism and control groups adjusted p < 0.05. b: between subclinical hyperthyroidism and control groups adjusted p < 0.05. c: between overt hyperthyroidism and control groups adjusted p < 0.05. d: between euthyroidism and overt hyperthyroidism groups adjusted p < 0.05.

Arterial stiffness in hyperthyroidism

A 30,00 25,00

fT3

20,00 15,00 10,00 5,00 ,00 ,00

10,00

20,00

30,00 Aix@75

40,00

50,00

60,00

,00

10,00

20,00

30,00 Aix@75

40,00

50,00

60,00

,00

10,00

20,00

30,00 Aix@75

40,00

50,00

60,00

fT4

4,00

2,00

1,00

TSH

4,00

3,00

2,00

1,00

,00

Figure 1. Correlation analysis between TSH (0.35-4 mIU/mL), fT4 (0.81.9 ng/mL) and fT3 (1.71-4.71 pg/mL) with Augmentation Index. “ ” shows the reference values a. Correlation between sT3 and Aix@75 p < 0.001 r = 0.335. b. Correlation between sT4 and Aix@75 p = 0.004 r = 0.282. c. Correlation between TSH and Aix@75 p < 0.001 r = -0.363. 262

reported that thyrotoxicosis reduces this risk by decreasing arterial stiffness as a counter-mechanism to PP and that there was no cardiovascular risk in thyrotoxicosis (14). However, the low number of patients participating in the study, the high average age (mean age 48) and the absence of thyrotoxicosis causes were important limitations of the study. Moulakakis and cols. reported that rats with thyrotoxicosis had a significant arterial stiffness in the aortic wall. An increase in internal and external diameter, smooth muscle cell and collagen density was observed in the aortic wall. On the other hand, a reduction in elastin lamina thickness and elastin density was found. It was reported that arterial stiffness resulting from thyrotoxicosis might have been associated with increased arterial pressure, which increased the risk of developing vascular complications. Despite low lipid levels, an increase in arterial stiffness was found in the study (15). In another study, ambulatory arterial stiffness indices (AASI) and blood pressure changes were assessed using data obtained from 23 participants with overt hyperthyroidism, 36 participants with subclinical hyperthyroidism and 25 healthy volunteers by following 24-hour ambulatory blood pressure monitoring. In this study, systolic and DBP were significantly higher in the hyperthyroid group compared to the other groups. However, AASI results did not show any significant difference between the groups. When all groups were evaluated, the positive correlation between AASI and free thyroid hormone levels was observed but no correlation with TSH was detected. Similarly, in our study, systolic and DBP were higher in the overt and subclinical hyperthyroid groups than the control group. However, unlike this study, Aix@75 and PWV measurements were significantly higher in both the overt and subclinical hyperthyroid groups. In addition, the positive correlation between Aix@75 and free thyroid hormone levels, and the negative correlation between Aix@75 and TSH were observed. In our study, the increase in SBP, HR and CO, especially in the overt hyperthyroid group, can be explained by the known and common effects of thyroid hormones on the cardiovascular system. In the subclinical hyperthyroidism group, the increase in these parameters was not as noticeable as in the overt hyperthyroidism group. This situation may be explained by the lower peripheral thyroid hormone levels. In the Arch Endocrinol Metab. 2019;63/3

Arterial stiffness in hyperthyroidism

Arch Endocrinol Metab. 2019;63/3

system and overt hyperthyroidism is a risk factor for CVDs. In addition to overt hyperthyroidism, subclinical hyperthyroidism also similarly increases the risk of atrial fibrillation and CVDs in a similar way (16). In our study, the relationship between Aix@75 and thyroid hormone levels were emphasized, indicating that these two conditions are related to each other. In current approaches, the decision to treat subclinical hyperthyroidism is based on the additional cardiovascular risk profile of the patients. Assessment of cardiovascular risk in these patient groups with the evaluation of arterial stiffness may be a new and reliable approach in establishing treatment indications. However, there is a need for more extensive and greater studies for this approach. There were some limitations in the study. The first of these; when the euthyroid group was being formed; patients who had previously been treated with hyperthyroidism and became euthyroid were included. The number of these patients was relatively low due to the detection of exclusionary conditions for the study. Second, the diagnosis of dyslipidemia in participants was excluded by the anamnesis. Serum lipid levels of patients were not measured. One of the strengths of the study was the assessment of euthyroid cases with previous hyperthyroidism history, subclinical hyperthyroidism and the overt hyperthyroidism in a single study. Another was that there was no difference in age, gender, and BMI among all groups and those other parameters that might affect arterial stiffness were excluded. As a result; in our study, we found that the overt or subclinical increase in the thyroid hormone levels adversely affected arterial stiffness which is a reliable risk indicator for CVDs. In the case of hyperthyroidism, especially PWV and Aix@75 are significantly altered. From this point of view, patients with thyrotoxicosis, even if taken high doses of thyroid hormone therapy, should be evaluated and informed previously in terms of cardiovascular risks. We hope that our research will lead to more comprehensive and further studies. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Epstein FH, Klein I, Ojamaa K. Thyroid hormone and the cardiovascular system. N Engl J Med. 2001;344(7):501-9.

263

euthyroid group SBP, central SBP, CO and PWV were also higher than the control group. There were some major differences when focusing on PWV and Aix@75 (Table 2). PWV was higher especially in the euthyroidism and the subclinical hyperthyroidism groups compared to the control group. On the other hand, Aix@75 level was higher in the overt hyperthyroidism group. This is an issue that needs to be considered in detail. In the euthyroidism and the subclinical hyperthyroidism groups, until the treatment, the cardiovascular system may be exposed to thyroid hormones, even if small amounts, and prolonged exposure to the thyroid hormones may affect PWV by creating more lasting effects on vascular structures. On the other hand, in the overt hyperthyroidism group, because of its high hormone effect, clinical symptoms occur in a much shorter time. There is not enough time for the changes in vascular structures which cause the increase in PWV. However, short-term high-dose hormone exposure may affect Aix@75 but this exposure time may not be enough to make significant changes in PWV, because of early treatment. These are very recent data and there is no previous study on this topic. With more extensive study, new information can be obtained. Focusing on PWV and Aix@75, a comparative study of efficacy between PWV and Aix@75 for the efficacy of arterial stiffness has not been conducted but it is inferred from the data that Aix@75 is more susceptible to arterial stiffness. This is because of the augmentation index, corrected based on pulse rate 75 beats/min, increases the strength and significance of the obtained data, especially considering the pulse rate increase in thyrotoxicosis studies may lead to wrong evaluations. In addition, a more detailed evaluation of Aix@75 has shown a close relationship between Aix@75 and thyroid hormones. This may explain that Aix@75 is only significant the overt hyperthyroidism group when compared to the control group. A negative correlation between Aix@75 and TSH in the correlation analysis also reinforces our findings. In previous studies, correlation evaluations were made between TSH and PWV and augmentation index, but no significant correlation could be shown. It is important that this is shown in this study. It is known that arterial stiffness is evaluated as a predictive parameter for cardiovascular risks (6). On the other hand, in overt hyperthyroidism, high thyroid hormone levels have many effects on the cardiovascular

Arterial stiffness in hyperthyroidism

2. Marcisz C, Jonderko G, Wróblewski T, Kurzawska G, Mazur F. Left ventricular mass in patients with hyperthyroidism. Med Sci Monit. 2006;12(11):CR481-6.

son with intra-aortic catheter measurements. Blood Press Monit. 2013;18(3):173-6.

3. Akkan T, Altay M, Ünsal Y, Dag6deviren M, Beyan E. Nonfunctioning adrenal incidentaloma affecting central blood pressure and arterial stiffness parameters. Endocrine. 2017;58(3):513-20.

10. Wassertheurer S, Kropf J, Weber T, van der Giet M, Baulmann J, Ammer M, et al. A new oscillometric method for pulse wave analysis: comparison with a common tonometric method. J Hum Hypertens. 2010;24(8):498-504.

4. Sehestedt T, Jeppesen J, Hansen TW, Wachtell K, Ibsen H, Torp-Pedersen C, et al. Risk prediction is improved by adding markers of subclinical organ damage to SCORE. Eur Heart J. 2010;31(7):883-91.

11. Omboni S, Posokhov IN, Kotovskaya Y V, Protogerou AD, Blacher J. Twenty-four-hour ambulatory pulse wave analysis in hypertension management: current evidence and perspectives. Curr Hypertens Rep. 2016;18(10):72.

5. Weber T, Wassertheurer S, Rammer M, Haiden A, Hametner B, Eber B. Wave reflections, assessed with a novel method for pulse wave separation, are associated with end-organ damage and clinical outcomes. Hypertension. 2012;60(2):534-41.

12. Wang J, Zheng X, Sun M, Wang Z, Fu Q, et al. Low serum free thyroxine concentrations associate with increased arterial stiffness in euthyroid subjects: a population-based cross-sectional study. Endocrine. 2015;50(2):465-73.

6. Townsend RR, Wilkinson IB, Schiffrin EL, Avolio AP, Chirinos JA, Cockcroft JR, et al.; American Heart Association Council on Hypertension. Recommendations for Improving and Standardizing Vascular Research on Arterial Stiffness: A Scientific Statement from the American Heart Association. Hypertension. 2015;66(3): 698-722.

13. Delitala AP, Orrù M, Filigheddu F, Pilia MG, Delitala G, Ganau A, et al. Serum free thyroxine levels are positively associated with arterial stiffness in the SardiNIA study. Clin Endocrinol (Oxf). 2015;82(4):592-7.

7. Vlachopoulos C, Aznaouridis K, Stefanadis C. Prediction of Cardiovascular Events and All-Cause Mortality With Arterial Stiffness. A Systematic Review and Meta-Analysis. J Am Coll Cardiol. 2010;55(13):1318-27. 8. Ben-Shlomo Y, Spears M, Boustred C, May M, Anderson SG, Benjamin EJ, et al. Aortic pulse wave velocity improves cardiovascular event prediction: An individual participant meta-analysis of prospective observational data from 17,635 subjects. J Am Coll Cardiol. 2014;63(7):636-46.

15. Alfellani MA, Stensvold CR, Vidal-Lapiedra A, Onuoha ESU, Fagbenro-Beyioku AF, Clark CG. Variable geographic distribution of Blastocystis subtypes and its potential implications. Acta Trop. 2013;126(1):11-8. 16. Moulakakis KG, Sokolis DP, Perrea DN, Dosios T, Dontas I, Poulakou MV, et al. The mechanical performance and histomorphological structure of the descending aorta in hyperthyroidism. Angiology. 2007;58(3):343-52. 17. Surks MI, Ortiz E, Daniels GH, Sawin CT, Col NF, Cobin RH, et al. Subclinical thyroid disease. JAMA. 2004;291(2):228-38.

9. Hametner B, Wassertheurer S, Kropf J, Mayer C, Eber B, Weber T. Oscillometric estimation of aortic pulse wave velocity: compari-

14. Obuobie K, Smith J, John R, Davies JS, Lazarus JH. The effects of thyrotoxicosis and its treatment on central arterial stiffness. Eur J Endocrinol. 2002;147(1):35-40.

264

Arch Endocrinol Metab. 2019;63/3

original article

Body composition and nutritional and metabolic parameters in postmenopausal women sufficient, insufficient and deficient in vitamin D Luisa Amábile Wolpe Simas1, Leila Caroline Bianchet Zanatta1, Carolina Aguiar Moreira1, Victoria Zeghbi Cochenski Borba1, Cesar Luiz Boguszewski1

ABSTRACT Objective: We investigated changes in body composition and nutritional and metabolic parameters in a group of postmenopausal women who were classified as sufficient, insufficient and deficient in vitamin D. Subjects and methods: A total of 106 postmenopausal women were included in this cross-sectional study and classified according to their serum levels of 25-OH-vitamin D as sufficient (≥ 30 ng/mL; group S), insufficient (20.1 and 29.9 ng/mL; group I) or deficient (≤ 20 ng/mL; group D) in vitamin D. Body composition was measured by dual-energy X-ray absorptiometry (DXA); dietary recall questionnaires were completed; and blood samples were analysed to compare the metabolic and nutritional status of the study groups. Results: Eleven (10.4%) of the women were classified in group S, 50 (47.2%) in group I and 45 (42.4%) in group D, with a mean serum level for 25-OH-D of 21.1 ± 7.0 ng/mL in all participants. Body composition did not differ among the groups. Serum levels of 25-OH-D were negatively correlated with serum levels of triglycerides, total cholesterol and LDL cholesterol. Conclusions: Vitamin D insufficiency and deficiency were highly prevalent in our group of postmenopausal women, showing an association with an unfavourable lipid profile. Arch Endocrinol Metab. 2019;63(3):265-71 Keywords Vitamin D; body composition; nutrition; metabolic syndrome; menopause

itamin D deficiency is a worldwide public health problem (1), even in regions and areas with sun exposure for most of the year. In Brazil, several studies have confirmed a high prevalence of hypovitaminosis D (2,3). Serum vitamin D levels may be affected by a number of factors (4), and there is a higher prevalence of vitamin D deficiency in women, with the highest number of daily cases in countries with higher latitude and in people with increased pigmentation of the skin (5). Vitamin D deficiency has been associated with several metabolic and cardiovascular abnormalities, including hypertension, diabetes mellitus, metabolic syndrome, obesity, vascular inflammation, left ventricular hypertrophy and congestive heart failure (6-8). During menopause, women may be particularly susceptible to the consequences of vitamin D deficiency since in this period of life oestrogen decreases, resulting in decreased bone mineral density (BMD) and lean mass, increased fat mass, and an increased risk of metabolic syndrome (MS) and cardiovascular morbidities (9-12). According to the International Diabetes Federation (IDF), the Arch Endocrinol Metab. 2019;63/3

Correspondence to: Cesar Luiz Boguszewski Serviço de Endocrinologia e Metabologia, Hospital de Clínicas, Universidade Federal do Paraná Avenida Agostinho Leão Junior, 285 80030-110 – Curitiba, PR, Brasil clbogus@uol.com.br Received on Aug/15/2018 Accepted on Feb/9/2019 DOI: 10.20945/2359-3997000000121

diagnosis of MS requires the presence of abdominal obesity associated with two of the following criteria: arterial hypertension, hypertriglyceridemia, reduced HDL-cholesterol, and elevated fasting glycaemia (13). The abdominal obesity associated with MS is a wellestablished cardiovascular risk factor and is frequently associated with hypovitaminosis D (12). Nevertheless, there have been divergences among the studies on the relationship between serum vitamin D levels and the various components of MS and body composition (14-16). Thus, the objective of this cross-sectional study was to compare the body composition, nutritional factors and metabolic parameters related to MS in menopausal women sufficient, insufficient and deficient in vitamin D.

SUBJECTS AND METHODS A cross-sectional descriptive study was carried out with postmenopausal women who were recruited from outpatient clinics of the Endocrine Division (SEMPR) of the Federal University Hospital in Curitiba, South 265

INTRODUCTION

Serviço de Endocrinologia e Metabologia (SEMPR), Departamento de Medicina Interna, Universidade Federal do Paraná, Curitiba, PR, Brasil 1

Nutrition: vitamin D in menopause

of Brazil, between February 2012 and March 2013. Menopause was defined through the 12-month criterion of amenorrhea (absence of menstruation) (17). The exclusion criteria were women in the menacme; those older than 70 years; patients with known diabetes mellitus, renal, hepatic or cardiorespiratory dysfunctions, bone diseases, secondary osteoporosis or neoplasias; and women in use of any medication in the last 6 months that according to medical judgement could interfere with glycaemia and/or bone metabolism. The study was approved by the Human Research Ethics Committee of our academic institution. All participants were informed about the research procedures and signed the informed consent term.

Clinical evaluation

The clinical evaluation included the collection of personal data, general clinical status with emphasis on concomitant medications, presence of chronic diseases, blood pressure (BP), waist circumference (WC), weight, height and body mass index (BMI). The number of calories ingested daily through food was calculated using data from the 24-hour food recall and Clinical Diet Win® Software 2002 (15). The measurements were performed according to WHO recommendations (18) using a 150 kg scale with an accuracy of 1 kg and a stadiometer with an accuracy of 1 mm. The BP was measured twice by the auscultatory method, with the participants seated, and it was defined as normal when less than 130/85 mmHg and abnormal when equal or greater than this cut-off (18). The study group was classified according to their BMI in: normal weight 18.0-24.9 kg/m2; overweight 25.0-29.9 kg/m2; obese grade 1 30.0-34.9 kg/m2, obese grade 2 35.0-39.9 kg/m2 and obese grade 3 ≥ 40 kg/m2 (19). WC was considered normal if less than 80 cm and altered if equal or greater than 80 cm. A WC between 80 and 87 cm was considered of an increased risk for cardiovascular events in overweight/obese women and equal or above 88 cm was considered of very high increased risk (20,21).

Laboratory tests Peripheral venous blood samples were collected after 8 hours of fasting and stored at -80 °C for analysis. Blood glucose was evaluated by hexoquinase/G-6-PD and values ≥ 100 mg/dL were considered abnormal; serum 266

insulin levels were determined by chemiluminescence (CL), with values of reference (VR) of 25 to 30 uUI/ mL and intra-assay coefficient of variation (CV) < 10%. HOMA-IR was calculated with the following formula: glycemia (mg/dL) X insulin (uUI/mL)/405. C-reactive protein (CRP) was measured by the turbidometry method, with a limit of detection below 0.5 mg/dL and CV < 10%. The colorimetric enzymatic method with a VR < 200 mg/dL was used for the determination of total cholesterol. HDL-cholesterol was measured by direct/homogeneous method and values ≤ 50 mg/dL were considered abnormal. LDL-cholesterol was determined by the Friedwald formula and scored at optimal values when ≤ 100 mg/dL, desirable at values between 101-129 mg/dL, borderline 130-159 mg/dL and high ≥ 160 mg/dL. Serum triglycerides (TG) was measured by the glycerolphosphate oxidase method and hypertriglyceridemia was defined by levels ≥ 150 mg/dL. Creatinine was evaluated by the picratecalcalino method with a VR of 0.57-1.11 mg/dL. TGO and TGP were evaluated by NADPH, with a VR varying between 0-55 IU/L and 5-34 IU/L, respectively. Serum calcium was determined by the Arsenazzo III method with a VR of 8.6-10.3 mg/dL and serum phosphorus by the UV phosphomolybdate method with a VR of 2.3-4.7 mg/dL. All measurements of PTH and vitamin D [25(OH)D] were determined simultaneously in one run by CL, with a VR of 11.761.1 pg/mL and CV < 10% for PTH and VR of 30100 ng/mL and CV < 10% for 25(OH)D. According to the 25(OH)D measurements, the participants were classified as having vitamin D deficiency when serum levels were ≤ 20 ng/ml, insufficiency when levels were between 20.1 and 29.9 ng/mL and sufficiency when ≥ 30 ng/mL (22).

Body composition Measurements of lumbar spine (L1 to L4) and total femoral BMD, as well as those of body composition (percentage total fat), were performed with a properly calibrated Lunar Prodigy Advance (GE Healthcare, Madison, WI, USA), with a coefficient of variation of 2% and by the same examiner.

Statistical analysis Statistical analyses were performed with the statistical package GraphPad Prism. All data are presented as Arch Endocrinol Metab. 2019;63/3

Nutrition: vitamin D in menopause

the mean ± SD, except as otherwise noted. To analyse the normality of the data, the Shapiro-Wilk test was used, which resulted in non-normal data. Thus, nonparametric tests were used to analyse the variables. For comparative analysis between groups, the Friedman test (with Dunn’s a posteriori test) was used for the quantitative variables. In the case of qualitative variables, the chi-square test was used. In the insufficient and deficient groups, the Mann-Whitney test was used for comparisons of the quantitative variables, and the chi-square test and Fisher’s exact test were used for the qualitative variables. For the tests between variables, Pearson’s correlations were used, following the theorem of large numbers. From the correlation index, shape and direction can be interpreted according to the concept that directly proportional correlations have a positive sign since inversely proportional correlations have a negative sign (23). In all cases, a level of significance of 5% (p < 0.05) was considered.

RESULTS The study group consisted of 106 postmenopausal women with a mean (±SD) age of 56.5 ± 6.1 years (range 43-70) and BMI of 28.2 ± 4.9 kg/m2 (range 19.7-41.1). In the whole group, serum levels of 25(OH)D were 21.1 ± 7.0 ng/mL (range 7.5-55.0),

with 11 women (10.4%) classified as sufficient, 50 (47.2%) as insufficient and 45 (42.4%) as deficient in vitamin D. The main clinical features of the three study groups are summarized in Table 1. There were no significant differences among the groups. Table 2 shows the comparisons between anthropometric measurements and body composition parameters evaluated by DXA in the three study groups. Again, no statistically significant differences were identified among vitamin D sufficient, insufficient and deficient postmenopausal women. The results of the laboratory tests (Table 3) showed a difference only in the insufficient and deficient groups in relation to serum calcium levels and TGP. Serum calcium was significantly higher in the deficient group (9.4 ± 0.5 vs. 9.2 ± 0.3 mg/dL in the insufficient group, p = 0.04), whereas serum TGP was significantly lower in the deficient group (18.4 ± 9.6 vs. 24.0 ± 16.2 mg/dL in the insufficient group, p = 0.047). There were statistically significant inverse correlations between serum 25(OH)D levels with serum levels of TG (r = 0,27; p = 0,006), total cholesterol (r = -0,25; p = 0,011) and LDL (r = -0,22; p = 0,028). The calcium dosage presented a significant difference when the insufficient and deficient vitamin D groups were compared. Thus, a correlation was performed between the three groups with the glycaemia values and the HOMA calculation to verify if the serum calcium

Table 1. Clinical characteristics of postmenopausal women classified as sufficient, insufficient and deficient in vitamin D Characteristic

Sufficient (n = 11)

Insufficient (n = 50)

Deficient (n = 45)*

54.9 ± 5.9

56.2 ± 5.4

57.3 ± 6.9

28.7 ± 5.4

28.9 ± 4.9

27.4 ± 4.7

Normal

27.3

33.3

Overweight

36.4

37.8

Obesity grade I

18.2

Obesity grade II

18.2

8.9

Obesity grade III

Waist circumference (cm)

91.2 ± 13.0

93.8 ± 12.4

89.5 ± 9.2

27.3

High (WC 81-87 cm)

9.1

18.0

24.4

Very high (WC ≥ 88 cm)

63.6

72.0

55.6

Abnormal

54.5

35.6

Normal

45.5

64.4

Age (yrs) Body mass index (BMI, kg/m ) 2

BMI classification (%)

Normal (WC ≤ 80 cm)

Cardiovascular risk (%)

Blood pressure (%)

* All p values were > 0.05. Arch Endocrinol Metab. 2019;63/3

267

Nutrition: vitamin D in menopause

value would be related to the other groups variables involved with insulin resistance and type 2 diabetes. There were no significant differences between the three groups in relation to the SM criteria for the three groups nor when there was a comparison between the insufficient and vitamin D-deficient groups. Through analysis of the 24-hour recall, it was possible to verify that the average dietary intake of vitamin D, calcium and cholesterol was lower in the insufficient group (Table 4).

DISCUSSION There was a high prevalence of hypovitaminosis D in the studied group, where only 10% of the 106 women had adequate vitamin D levels (considering a cut-off point above 30 ng/mL) and 90% were deficient or insufficient. These values are higher than those found in cohort studies of menopausal women, which showed rates of 67% and 68% for hypovitaminosis D (24,25). In another study with 93 postmenopausal women living

Table 2. Anthropometric characteristics and body composition parameters by DXA assessment of postmenopausal women classified as sufficient, insufficient and deficient in vitamin D Variable

Sufficient (n = 11)

Insufficient (n = 50)

Deficient (n = 45)

Gynoid adipose mass (%)

48.3 ± 5.9

48.5 ± 5.5

47.2 ± 6.5

Android adipose mass (%)

47.6 ± 10.9

48.2 ± 7.5

46.9 ± 8.0

Total adipose, %

41.9 ± 7.4

41.8 ± 6.0

40.5 ± 7.0

Truncal fat (%)

43.3 ± 8.6

43.8 ± 6.8

42.3 ± 7.3

BMD L1-L4 (g/cm )

1.01 ± 0.10

1.07 ± 0.15

1.06 ± 0.13

BMD total femur (g/cm2)

0.90 ± 0.10

0.96 ± 0.15

0.92 ± 0.12

BMC L1-L4 (g)

50.8 ± 7.1

54.2 ± 10.7

53.6 ± 9.1

Normal

9.1

30.0

24.4

Osteopenia

81.8

50.0

51.1

Osteoporosis

9.1

20.0

24.4

Bone mineral density (%)

BMD L1-L4: bone mineral density lumbar spine; BMC: bone mineral content. * All p values were > 0.05.

Table 3. Characteristics of the biochemical exams of postmenopausal women classified as sufficient, insufficient and deficient in vitamin D Variable

Sufficient (n = 11)

Insufficient (n = 50)

Deficient (n = 45)

Total cholesterol (mg/dL)

223.5 ± 37.2 (233.0)

209.3 ± 41.6 (208.0)

226.4 ± 46.5 (228.0)

Triglycerides (mg/dL)

113.7 ± 42.7 (92.0)

122.6 ± 56.9 (107.0)

135.6 ± 61.0 (124.0)

HDL (mg/dL)

52.1 ± 10.0 (52.0)

45.7 ± 10.2 (44.0)

47.9 ± 12.5 (47.0)

LDL (mg/dL)

148.8 ± 34.2 (154.0)

139.8 ± 38.0 (139.5)

151.3 ± 39.4 (152.0)

Glycemia (mg/dL)

95.0 ± 9.6 (92.0)

93.9 ± 12.2 (92.0)

90.7 ± 8.7 (89.0)

Insulin (uUI/mL)

10.4 ± 5.3 (11.1)

10.9 ± 6.1 (9.4)

10.1 ± 5.9 (7.8)

HOMA-IR

2.47 ± 1.30 (2.36)

2.64 ± 1.78 (2.15)

2.33 ± 1.58 (1.85)

Creatinine (mg/dL)

0.73 ± 0.08 (0.7)

0.73 ± 0.12 (0.7)

0.71 ± 0.10 (0.7)

20.2 ± 7.3 (18)

20.8 ± 9.5 (18)

19.0 ± 5.5 (18)

TGO (UI/L) TGP (UI/L)

20.1 ± 6.8 (20)**

24.0 ± 16.2 (19)

18.4 ± 9.6 (16)

Albumin (mg/dL)

4.07 ± 0.20 (4.0)

3.96 ± 0.22 (4.0)

3.96 ± 0.21 (4.0)

Calcium (mg/dL)

9.35 ± 0.27 (9.4)

9.20 ± 0.34 (9.2)

9.38 ± 0.50 (9.3)*

Phosphorus (mg/dL)

3.51 ± 0.45 (3.6)

3.43 ± 0.46 (3.3)

3.31 ± 0.62 (3.4)

PTH (pg/mL)

50.2 ± 11.5 (50.1)

56.3 ± 22.4 (50.8)

59.3 ± 17.7 (3.6)

ng/mL (mg/dL)

0.38 ± 0.10 (0.33)

0.45 ± 0.25 (0.33)

0.62 ± 1.05 (0.33)

HDL: high density lipoprotein; LDL: low density lipoprotein; HOMA-IR: homeostatic model assessment for insulin resistance; TGO: glutamic-oxalacetic transaminase; TPG: pyruvic glutamic transaminase; PCR-C: reactive protein; mg/dL: milligrams per decilitre; uUI/mL: international micro unit per millilitre; UI/L: intentional unit per litre; pg/mL: picogram per millilitre. Values are means ± standard deviations, with medians in parentheses or percentage frequencies. * p = 0.04 vs. insufficient. ** p = 0.04 vs. insufficient.

268

Arch Endocrinol Metab. 2019;63/3

Nutrition: vitamin D in menopause

in the city of Recife, the authors found a prevalence of 24% of vitamin D deficiency, which was lower than this study; this study also found a mean serum vitamin D concentration of 28.8 ng/mL, higher than the 21.1 ng/mL observed in our patients (with cut-off point for deficiency less than 20 ng/mL), because according to with the authors themselves the incidence of solar radiation is higher because Reef is closer to the equator (26). The current study found an inverse correlation between serum vitamin D levels and serum levels of TG, total cholesterol and LDL-cholesterol, indicating a trend towards an unfavourable lipid profile in menopausal women with lower levels of vitamin D. However, we did not find significant differences when we analysed and compared the groups deficient and insufficient in relation to body composition and metabolic variables. Similarly, these groups did not differ when analysed according to the SM criteria presented (27). A study conducted by Iranian researchers has shown that the serum vitamin D concentration was lower in subjects with MS, similar to this study, compared to individuals without MS (27). To identify the relationship between daily intake of vitamin D, serum 25(OH)D levels and prevalence of MS, these researchers compared the plasma and nutritional variables of 128 healthy individuals and 122 individuals with a diagnosis of MS (27). The results showed that 98.4% of the patients in the MS group and 88.3% of the healthy individuals presented vitamin D deficiency [25(OH)D < 20 ng/mL]. This difference was statistically significant (p = 0.005). Furthermore, concentrations of 25(OH)D were significantly associated with waist circumference (p = 0.001), HDL (p = 0.001) and triglyceride levels (p = 0.012). The authors concluded that there is an association between vitamin D and MS. This study, however, failed to confirm the relationship between vitamin D deficiency and a higher prevalence of MS.

A study evaluating micronutrient concentrations in women (n = 8,137) who participated in the National Health and Nutrition Examination Survey III (NHANES III) showed that in postmenopausal women there was a negative association of vitamin D with BMI (25,28). Other researchers have observed that the percentage of body fat was also strongly associated with a low vitamin D concentration (23). The present study did not demonstrate a negative association with BMI and fat percentage. Vitamin D deficiency has been associated with the development of cardiovascular diseases, hypertension, neoplasias, and diabetes, suggesting a crucial role for this vitamin in several homeostasis systems (29). Screening for vitamin D deficiency is cost-effective in some specific populations – postmenopausal, elderly, institutionalized, and pregnant women. In this study, vitamin D levels were not related to blood pressure and glycaemic indexes. We observed differences between serum levels of Ca and TGP among the women who were insufficient and deficient in vitamin D; there were higher values of Ca and lower values of TGP. Ca intake was also significantly higher in the deficient group compared to the insufficient group, as was as the ingestion of cholesterol-containing foods and total daily caloric intake. It is known that adequate plasma concentrations of 25(OH)D are required to allow adequate absorption of calcium administered orally. A recent study showed that calcium absorption was 65% higher when plasma concentrations of 25(OH)D were in the mean of 34,8 ng/mL (30). Although diet represents only 10% to 20% of the vitamin D necessary for the proper functioning of the organism, the dietary sources of vitamin D are scarce, and individuals depend mainly on the skin for production vitamin D, catalysed by solar UVB rays (31). Vitamin D is found in food sources such as cod liver oil and fatty fish (salmon, tuna, and mackerel) or

24-hour recall

Sufficient (n = 11)

Insufficient (n = 50)

Deficient (n = 45)

P value

Vitamin D, µg

3.93 ± 0.80 (3.79)

0.84 ± 1.12 (0.43)

1.32 ± 0.90 (1.27)**

< 0.001*

Calcium, mg

395.0 ± 199.2 (379.3)

238.7 ± 174.1 (199.6)

248.6 ± 130.5 (234.3)

0.021*

Kcal/total, kcal

1649 ± 601.4 (1647)

1406 ± 608.1 (1257)

1985 ± 637.8 (2123)***

< 0.001*

Cholesterol, mg

314.2 ± 601.4 (207.5)

126.3 ± 94.3 (109.5)

199.9 ± 135.1 (168.1) ****

< 0.001*

8.6 ± 3.2 (8.0)

9.1 ± 5.5 (8.5)

7.7 ± 2.6 (7.7)

0.633

Fibre (g)

Table 4. Dietary ingestion in postmenopausal women classified as sufficient, insufficient and deficient in vitamin D

* p < 0.05. ** p = 0.026 vs. insufficient. *** p = < 0.001 vs. insufficient. **** p = 0.003 vs. insufficient. Arch Endocrinol Metab. 2019;63/3

269

Nutrition: vitamin D in menopause

through endogenous cutaneous synthesis, which is the main source of this vitamin for most humans (31). Vitamin D intake was higher in women whose levels were insufficient. Recent studies in the literature have correlated vitamin D supplementation with lower mortality rates; according to these data, we can suggest supplementation for correction of hypovitaminosis D and dietary inadequacy aimed at postmenopausal women (32,33). In conclusion, in this group of menopausal women, there were no significant differences in body composition and metabolic parameters between women with insufficient and deficient levels of vitamin D. Further, lower vitamin D levels were associated with an unfavourable lipid profile; there was a high prevalence of hypovitaminosis D, corresponding to 90% of the sample; and serum vitamin D levels were influenced by dietary intake of vitamin D. Funding statement: no funding statement to make. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Palacios C, Gonzalez L. Is vitamin D deficiency a major global public health problem? J Steroid Biochem Mol Biol. 2014;144: 138-45. 2. Santos AEPD, Queiroz DJM, Neves JPR, Lacerda LM, Gonçalves M, Carvalho AT. Prevalence of hypovitaminosis D and associated factors in adolescent students of a capital of northeastern Brazil. Nutr Hosp. 2017;34(5):1416-23. 3. Silva BCC, Camargos BM, Fujii JB, Dias EP, Soares MMS. Prevalência de deficiência e insuficiência de vitamina D e sua correlação com PTH, marcadores de remodelação óssea e densidade mineral óssea, em pacientes ambulatoriais. Arq Bras Endocrinol Metabol. 2008;52(3):482-8. 4. Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol. 2014;2(1):76-89. 5. Holick M. Vitamina D como um tratamento tão simples pode reverter doenças tão importantes. São Paulo: Editora Fundamento; 2012. 6. Nakashima A, Yokoyama K, Yokoo T, Urashima M. Role of vitamin D in diabetes mellitus and chronic kidney disease. World J Diabetes. 2016;7(5):89-100.

Anderson JL, May HT, Horne BD, Bair TL, Hall NL, Carlquist JF, et al. Relation of vitamin D deficiency to cardiovascular risk factors, disease status, and incident events in a general healthcare population. Am J Cardiol. 2010;106(7):963-8.

8. Mihos CG, Cruz JADL, Hernandez A, Santana O. Vitamin D deficiency and supplementation in cardiovascular disorders. Cardiol Rev. 2017;25(4):189-96. 9. Lerchbaum E. Vitamin D and menopause – a narrative review. Maturitas. 2014;79(1):3-7. 10. Kaur J. A comprehensive review on metabolic syndrome. Cardiol Res Pract. 2014;2014:943162.

270

11. Vitezova A, Zillikens MC, van Herpt TT, Sijbrands EJ, Hofman A, Uitterlinden AG, et al. Vitamin D status and metabolic syndrome in the elderly: the Rotterdam study. Eur J Endocrinol. 2015;172(3):327-35. 12. O’Neill S, O’Driscoll L. Metabolic syndrome: a closer look at the growing epidemic and its associated pathologies. Obes Rev. 2015;16(1):1-12. 13. Simmons RK, Alberti KG, Gale EA, Colagiuri S, Tuomilehto J, Qiao Q, et al. The metabolic syndrome: useful concept or clinical tool? Report of a WHO expert consultation. Diabetologia. 2010;53(4):600-5. 14. Chacko SA, Song Y, Manson JE, van Horn L, Eaton C, Martin LW, et al. Serum 25-hydroxyvitamin D concentrations in relation to cardiometabolic risk factors and metabolic syndrome in postmenopausal women. Am J Clin Nutr. 2011;94(1):209-17. 15. Henn RL, Fuchs SC, Moreira LB, Fuchs FD. Development and validation of a food frequency questionnaire (FFQ-Porto Alegre) for adolescent, adult and elderly populations from Southern Brazil. Cad Saude Publica. 2010;26(11):2068-79. 16. Strange RC, Shipman KE, Ramachandran S. Metabolic syndrome: a review of the role of vitamin D in mediating susceptibility and outcome. World J Diabetes. 2015;6(7):896-911. 17. Butler L, Santoro N. The reproductive endocrinology of the menopausal transition. Steroids. 2011;76(7):627-35. 18. de Onis M, Onyango A, Borghi E, Siyam A, Blossner M, Lutter C. Worldwide implementation of the WHO child growth standards. Public Health Nutr. 2012;15(9):1603-10. 19. World Health Organization. Waist circumference and waist-hip ratio report of a WHO expert consultation. Geneva: World Health Organization; 2008. 20. Lee DC, Sui X, Church TS, Lavie CJ, Jackson AS, Blair SN. Changes in fitness and fatness on the development of cardiovascular disease risk factors hypertension, metabolic syndrome, and hypercholesterolemia. J Am Coll Cardiol. 2012;59(7):665-72. 21. Hamer M, Stamatakis E. Metabolically healthy obesity and risk of all-cause and cardiovascular disease mortality. J Clin Endocrinol Metab. 2012;97(7):2482-8. 22. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, et al. Evaluation, treatment, and prevention of vitamin D deficiency: an endocrine society clinical practice guideline. J Am Coll Cardiol. 2011;96(7):1911-30. 23. Zar J. Biostatistical analysis. Harlow: Pearson; 2009. 24. Arantes HP, Kulak CA, Fernandes CE, Zerbini C, Bandeira F, Barbosa IC, et al. Correlation between 25-hydroxyvitamin D levels and latitude in Brazilian postmenopausal women: from the Arzoxifene generations trial. Osteoporos Int. 2013;24(10):2707-12. 25. Russo LA, Gregorio LH, Lacativa PG, Marinheiro LP. Concentration of 25-hydroxyvitamin D in postmenopausal women with low bone mineral density. Arq Bras Endocrinol Metab. 2009;53(9):1079-87. 26. Bandeira F, Griz L, Freese E, Lima DC, Thé AC, Diniz ET, et al. Vitamin D deficiency and its relationship with bone mineral density among postmenopausal women living in the tropics. Arq Bras Endocrinol Metab. 2010;54(2):227-32. 27. Ghanei L, Ziaee A, Rostami P, Oveisi S, Esmailzadehha N, Kazemifar AM, et al. Association of serum 25-hydroxyvitamin d levels and vitamin D dietary intake with metabolic syndrome: a case control study. J Res Health Sci. 2015;15(1):32-6. 28. Snijder MB, van Dam RM, Visser M, Deeg DJ, Dekker JM, Bouter LM, et al. Adiposity in relation to vitamin D status and parathyroid hormone levels: a population-based study in older men and women. J Clin Endocrinol Metab. 2005;90(7):4119-23. 29. Alves M, Bastos M, Leitão F, Marques G, Ribeiro G, Carrilho F. Vitamin D – importance of laboratory evaluation. Rev Port Endocrinol Diabetes Metab. 2013;8(1):32-9. 30. Bischoff-Ferrari HA, Willett WC, Orav EJ, Lips P, Meunier PJ, Lyons RA, et al. A pooled analysis of vitamin D dose requirements for fracture prevention. N Engl J Med. 2012;367(1):40-9. Arch Endocrinol Metab. 2019;63/3

Nutrition: vitamin D in menopause

31. Maeda SS, Borba VZ, Camargo MB, Silva DM, Borges JL, Bandeira F, et al. Recommendations of the Brazilian society of endocrinology and metabology (SBEM) for the diagnosis and treatment of hypovitaminosis D. Arq Bras Endocrinol Metab. 2014;58(5):411-33.

32. Rejnmark L, Avenell A, Masud T, Anderson F, Meyer HE, Sanders KM, et al. Vitamin D with calcium reduces mortality: patient level pooled analysis of 70,528 patients from eight major vitamin D trials. J Clin Endocrinol Metab. 2012;97(8):2670-81.

33. Rizzoli R, Boonen S, Brandi ML, Bruyere O, Cooper C, Kanis JA, et al. Vitamin D supplementation in elderly or postmenopausal women: a 2013 update of the 2008 recommendations from the European society for clinical and economic aspects of Osteoporosis and Osteoarthritis (ESCEO). Curr Med Res Opin. 2013;29(4):305-13.

Arch Endocrinol Metab. 2019;63/3

271

original article

Paraoxonase 1 serum activity in women: the effects of menopause, the C(-107)T polymorphism and food intake Mauren Castro Ritta1*, Aline Marques Baldez1*, Isabel Oliveira de Oliveira2, Driele Neske Garcia1, Paola Spiering Souza1, Kelvin Ruan da Silva Andrade1, Sandra Costa Valle1, Simone Pieniz1, Carlos Castilho Barros1, Michal M. Masternak3,4, Augusto Schneider1

Faculdade de Nutrição, Universidade Federal de Pelotas (UFPel), Pelotas, RS, Brasil 2 Instituto de Biologia, Universidade Federal de Pelotas (UFPel), Pelotas, RS, Brasil 3 College of Medicine, Burnett School of Biomedical Sciences, University of Central Florida, Orlando, FL, USA 4 Department of Head and Neck Surgery, Greater Poland Cancer Centre, Poznan, Poland * Both authors contributed equally to this work 1

Correspondence to: Augusto Schneider Rua Gomes Carneiro, 1, sala 242 96010-610 – Pelotas, RS, Brasil augusto.schneider@ufpel.edu.br Received on Sept/29/2018 Accepted on Feb/24/2019

ABSTRACT Objective: The aims of this study were to investigate changes in serum paraoxonase 1 (PON1) activity in women at the pre and postmenopausal stages and its association with the PON1 C(-107)T polymorphism and food intake profile. Subjects and methods: A cross-sectional study with female patients aged between 35 and 59 years old was conducted. Women were divided into two groups: premenopausal (n = 40) and postmenopausal (n = 36). Women enrolled in the study had serum PON1, total cholesterol, HDL, LDL, glucose and HbA1c, as well as the BMI measured. Additionally, women were genotyped for the PON1 T(-107)C polymorphism and the food intake profile was obtained through interview. Results: Glucose (p = 0.03), HbA1c (p = 0.002) and total cholesterol (p = 0.002) concentrations were higher in post than premenopausal women, however PON1 activity was not different (p > 0.05). Carriers of the C allele had higher PON1 activity (CC: 88.9 ± 6.5 U/mL and CT: 79.9 ± 4.7 U/mL) than women of the TT genotype (66.6 ± 5.9 U/mL) (p < 0.05). However, the model predicting PON1 activity was slightly better when genotype, total fat and cholesterol content in the diet were all included. Conclusion: In sum, we observed that the PON1 C(-107)T genotype was the major regulator of PON1 activity, and menopause had no effect on PON1 activity. The lipid and glycemic profile were altered in postmenopausal women. Arch Endocrinol Metab. 2019;63(3):272-9 Keywords PON1; aging; HDL; SNPs; nutrigenetics

DOI: 10.20945/2359-3997000000130

INTRODUCTION

enopause is defined as the cessation of ovulation due to the depletion of the ovarian follicular reserve, which occurs around 50 years of age, and is characterized by the absence of menstrual cycles for at least twelve months in older women (1). As a result of the ovarian inactivity, estrogen and inhibin levels are severely reduced in postmenopausal women (1). These hormonal deficiencies lead to changes in the serum lipid profile, as well as body composition and fat distribution (2-4). Therefore, menopause onset is associated with several physiological and biochemical changes, resulting in an increased susceptibility to chronic diseases such as type 2 diabetes (4) and cardiovascular diseases (CVD) (5). The occurrence of CVD in premenopausal women is lower than in men of 272

the same age, however, it increases in postmenopausal women to levels comparable to men (6). Therefore, it is important to understand the physiological and biochemical changes associated with menopause to prevent disease incidence in this high-risk group. Oxidative stress plays an important role in the etiology of atherosclerosis (7). Because of its antioxidant activity, high-density lipoprotein (HDL) prevents low-density lipoprotein (LDL) particles from undergoing oxidative modification, conferring an atheroprotective effect (8). This antioxidant activity of HDL is mostly conferred by paraoxonase 1 (PON1) (9), an enzyme synthesized in the liver found circulating in plasma associated with the HDL particle (10). Serum PON1 activity is influenced by genetic and environmental factors such as atherogenic diets and smoking (11). Interesting, Arch Endocrinol Metab. 2019;63/3

Paraoxonase: menopause, genetics and diet

Arch Endocrinol Metab. 2019;63/3

comorbidities associated with this stage of life, given the increase in life expectancy increases the proportion of the female population at the postmenopausal stage (22). Based on this, the aims of this study were to investigate changes in serum PON1 activity in women at the pre and postmenopausal stages and its association with the PON1 C(-107)T polymorphism and food intake profile. Our hypothesis is that menopause will affect serum PON1 activity response to different PON1 C(-107)T genotypes and dietary intake.

SUBJECTS AND METHODS Design and population A cross-sectional study with female patients aged between 35 and 59 years old, who attended the Units of Family Health in southern Brazil (Rio Grande, RS) in a six-month interval was conducted. The study was submitted to the Ethics Committee of the Universidade Federal de Pelotas and approved under the number 1.708.582. The experiment was performed in accordance to the STROBE guidelines. All participants were informed of the research objective, as well as of the methodological procedures, and signed the Consent Form. The sample consisted of 103 women that met the inclusion criteria in the period of data collection. The inclusion criteria were to have between 35 and 59 year of age, not be receiving hormonal treatment, and for the menopausal group to be at least one year in amenorrhea. Women who used medication that compromised vascular function and the hypothalamic-pituitarygonadal axis, pregnant women, nursing mothers and women with early menopause (less than 40 years), late (over 55 years) or from surgical procedure were excluded from the study. The sample size calculation was based on the literature review and was carried out in the OpenEpi software (online version 3.01). To detect a significant (p < 0.05) effect with power of 95% for a 13% difference in PON1 activity between pre and post menopausal women (13) and a 30% difference in PON1 activity between AA and CC genotypes (16) we would need 9 subjects per genetic group. Therefore, considering the distribution of the three genotypes (28% for the minor frequency TT genotype (target n = 10), 41% for CT (target n = 14) and 31% for the CC genotype (target n = 11) (16) we would need a total of 35 women per menopausal stage considering the 273

serum PON1 activity in females is higher than in males (11). This difference is attributed to sex steroids, since estradiol enhances PON1 activity independent of liver PON1 protein synthesis (12). Therefore, it is believed that higher PON1 activity in reproductive age women contributes to the atheroprotective effects, lowering the risk of CVD. Differences in the PON1 activity were described between pre and postmenopausal women (13), although there is no consensus regarding this issue (14). In this sense, a better understanding of how serum PON1 activity is regulated during menopause and which are the risk factors associated with serum PON1 activity can help improve prevention of chronic diseases. There are more than 160 single nucleotide polymorphisms (SNPs) described in the PON1 gene (11). However, one important SNP is located in the promoter region C(-107)T (rs705379), and exerts a significant effect on PON1 serum activity (15,16) contributing with 25% of serum PON1 activity variation (17). The presence of the C allele results in serum PON1 activity up to two times higher than that observed in the presence of the T allele only (15). We have shown before that women carriers of the C allele also have higher PON1 activity than women carrying the TT genotype (16). Despite this, some suggest that PON1 activity is more important than its genetic variations in preventing CVD (18) and no association between the T(-107)C SNP and CVD has been found (19). However, information on how serum PON1 activity is regulated around menopause of women from different PON1 genotypes is still lacking. The diet plays an important role in the prevention of CVD, as dietary intake of excess saturated fatty acids and cholesterol is associated with increased risk of CVD (20). The activity of PON1 is also modified by diet, as seen in studies that associate the consumption of cholesterol and fatty acids with PON1 activity (21). Women with a diet rich in saturated fatty acids, omega 3 and poor in omega 6 had lower PON1 activity, which was more evident in carriers of the TT genotype (16). This indicates that the effects of the diet on PON1 activity are dependent on the genotype, and given the context and the assumptions presented, could be reinforced that environmental and genetic factors contribute to the risk of CVD. However, there are few scientific reports about the association between menopause, diet, genotype and PON1 activity. Thus, this knowledge will allow more adequate intervention strategies to be implemented, with a view to preventing

Paraoxonase: menopause, genetics and diet

genotype distribution. After applying the inclusion and exclusion criteria described, 76 women were enrolled in the study, and divided into two groups: premenopausal (n = 40) and postmenopausal (n = 36). Women with no menstrual activity in the last twelve months and older than 40 years were considered as postmenopausal (23).

Data and blood collection

A general questionnaire was applied for the evaluation of health conditions, level of education, use of medications and life habits. Additionally, food intake was assessed by application of a food frequency questionnaire, containing 83 foods components, following the standards previously described (24). Food were classified as processed (40 types of food), in natura (26 types of food) and rich in cholesterol/saturated fatty acids (32 types of food), as proposed by the New Food Guide for the Brazilian Population (25). For each food, a weekly consumption frequency was established and from this score the consumption medians were established for each of the 3 categories (processed, in natura, and rich in cholesterol/fatty acids) and patients classified as ingesting above or below the average. In addition, based on the composition of food and ingestion amount reported daily average intake of calorie, protein, carbohydrate, lipid and fiber were calculated. Body mass (kg) and height (centimeters) were measured on a mechanical scale (Welmy, Santa Bárbara d’Oeste, SP, Brazil). The body mass index (BMI) of each participant was obtained through the equation body mass (kg) divided by stature squared (m) and interpreted according to the classification of the World Health Organization (26). Blood collection was scheduled at the time of the interview within a two-week interval. About 5 mL of blood from each patient was collected by venipuncture, after a 12-hour fasting period, in dry vials for biochemical measurements. About 1 mL of blood was collected in an EDTA coated vial for DNA extraction.

1:3 in buffer before added to the working reagent and the change in absorbance was recorded for 60 sec. One unit of arylesterase activity was considered equal to 1 mM of phenol formed per minute. The activity is expressed in U/mL. Blank samples containing water were used to correct non-enzymatic hydrolysis. Blood samples were also sent to a clinical laboratory for analysis in an automated biochemical analyzer. Total cholesterol (TC), HDL cholesterol (HDL), LDL cholesterol (LDL), triglycerides, and glucose were measured by automated enzymatic methods. Glycated hemoglobin (HbA1c) was determined by the immunoturbidimetry method.

Genotyping To extract the genomic DNA, blood samples were processed following an adapted protocol (28). For the determination of the PON1 T(-107)C genotype the procedures were followed as previously described (15,16). Briefly, the amplification of the region where the SNP is located was performed by PCR, using 10 μL of GOTaq® mix for PCR (Promega, Madison, WI, USA), 1 μL (10 μM stock concentration) of the forward primer AGCTAGCTGCGGACCCGGCGGGGAGGaG and 1 μL of the reverse primer GGCTGCAGCCCTCACCACAACCC. Standard conditions for the PCR reaction were used, with an annealing temperature of 67 °C. The lower-case letter in the forward primer indicates a mismatch that introduces a restriction site for the BsrBI enzyme (New England BioLabs, Cambridge, UK), as there is no specific restriction site cleaving the original DNA sequence. After digestion of the PCR product for 2 h at 37 °C, the DNA fragments were separated by 2% agarose gel electrophoresis, stained with SYBR Safe (Applied Biosystems, Foster City, CA, USA). The C allele was identified by fragments of 28 and 212bp, while the T allele resulted in the undigested 240bp fragment. We were unable to process seven samples for DNA genotyping.

Biochemical analysis

Data processing and analysis

PON1 activity was evaluated by the measurement of the arylesterase activity according to previously described (16,27). The activity was measured from the rate of phenol formation by monitoring the absorbance increase at 270 nm. The working reagent consisted of 20 mM Tris/HCl, pH 8.0, containing 1 mM CaCl2 and 4 mM phenylacetate as substrate. Samples were diluted

Statistical analyzes were performed on SAS University Edition (SAS, Cary, NC, USA) and figures were generated in Graphpad Prism 5 (GraphPad, La Jolla, CA, USA). All parameters had passed the preliminary Shapiro-Wilk test for normality. Comparisons between the pre and postmenopausal women were made using the MIXED MODEL procedure using ethnicity and

274

Arch Endocrinol Metab. 2019;63/3

Paraoxonase: menopause, genetics and diet

BMI as co-variates. For testing individual effects of menopause, genotype and food consumption on PON1 activity as well as its interactions, the same MIXED MODEL procedure was used. Post-hoc analysis was performed using the Tukey adjustment. Additionally, to verify the linear or quadratic effect of the presence of the C allele on PON1 activity, the GLM model was used, considering the genotypes TT as 0, CT as 1 and CC as 2. The GLM linear effect was also used to test the effect of the level of consumption of processed, in natura and cholesterol/fatty acids rich diets on serum PON1 activity. In the unadjusted analysis, simple linear regression was used to compare PON1 activity according to genotype. Multivariable analyses included linear regression using three different models. Model one included BMI and total dietary energy content; Model 2 included only the total lipid and cholesterol content of the diet and Model 3 included BMI, total dietary energy content, total dietary lipid and cholesterol content. In order to analyze the effect of several variables on the association between serum PON1 activity and genotype, we ran models including one extra variable at a time, and examined the change in beta (β) coefficient and coefficient of determination (R²). The order in which variables were included in the model was defined by the significance of its association with serum PON1 activity. HardyWeinberg equilibrium and comparison of genotype distribution between pre and postmenopausal women were compared by the Chi-square test. The level of significance was set at p ≤ 0.05.

The genotype frequency for the PON1 C(-107)T SNP is presented in Table 3. The allelic frequency for the C allele was 46.4% and for the T allele it was 53.6%. The population was in Hardy-Weinbeg equilibrium (p = 0.29), and genotype distribution is presented in Table 3. There was no overall difference in genotype distribution between pre and post-menopausal women in the chisquare test (p = 0.22). Regarding the individual effects of genotype and menopause on PON1 activity, there was an effect of the genotype (Figure 1; p = 0.03), but no effect of menopause (p = 0.20) and also no significant menopause by genotype interaction (p = 0.83). There Table 1. General characteristics of women in the pre and postmenopausal stages recruited for the study. Parameter

Premenopausal

Postmenopausal

p value

Sample size

Age (years)

42.0 ± 0.8

53.5 ± 0.6

< 0.0001

46.5 ± 0.7

34.2% (13/38)

White

70.0% (28/40)

86.1% (31/36)

0.11

Mixed/Black

25.0% (10/40)

11.1% (4/36)

0.15

Undeclared

5.0% (2/40)

2.8% (1/36)

Currently

5.4% (2/37)

22.8% (8/35)

0.04

Past

18.9% (7/37)

8.6% (3/35)

0.30

17.5% (7/40)

27.8% (10/36)

0.51

Age at menopause Use of contraceptives Ethnicity2

Smoking

BMI3 Normal range (18-25) Overweight (25-30)

15.0% (6/40)

16.7% (6/36)

Obese (> 30)

67.5% (27/40)

55.5% (20/36)

Means are presented as least square means ± standard error of the mean and where compared using the MIXED model, using BMI and ethnicity as co-variates and the Tukey posthoc test, 2 Proportions are presented as percentages and where compared using Chi-square test, 3 Body mass index. 1

The general characteristics of the study population are presented in Table 1. Postmenopausal women were older (p < 0.0001), as expected, but there was no difference in BMI distribution among groups. Additionally, a higher proportion of postmenopausal women were currently smokers (p = 0.04). Regarding the metabolic profile, glucose (p = 0.03), HbA1c (p = 0.002) and total cholesterol (p = 0.002) concentrations were higher in post than premenopausal women (Table 2). There were no differences for BMI, HDL and LDL cholesterol, triglycerides and PON1 activity between pre and postmenopausal women (p > 0.05; Table 2). Additionally, the use of contraceptives did not affect serum PON1 activity (Yes: 73.8 ± 4.7 U/mL, No: 82.5 ± 7.3 U/mL, p = 0.32). Arch Endocrinol Metab. 2019;63/3

Table 2. Blood biochemistry parameters for women in the pre and postmenopausal stages Parameter

Premenopausal1

Postmenopausal1

p value

Glucose (mg/dL)

84.6 ± 4.3

99.2 ± 4.5

0.03

HbA1c (%)

5.4 ± 0.2

6.2 ± 0.2

0.002

182.6 ± 6.2

211.6 ± 6.7

0.002

Total Cholesterol (mg/dL) HDL (mg/dL)3

48.8 ± 2.2

53.4 ± 2.3

0.16

LDL (mg/dL)4

107.6 ± 6.0

128.7 ± 7.6

0.07

Triglycerides (mg/dL)

138.3 ± 13.1

152.6 ± 14.2

0.46

74.4 ± 4.4

81.3 ± 4.7

0.28

PON1 (U/mL)5

Means are presented as least square means ± standard error of the mean and where compared using the MIXED model, using BMI and ethnicity as co-variates and the Tukey posthoc test, 2 Glycated hemoglobin, 3 High-density lipoprotein cholesterol, 4 Low-density lipoprotein cholesterol, 5 Paraoxonase 1. 1

275

RESULTS

Paraoxonase: menopause, genetics and diet

Table 3. Genotype frequency distribution for the PON1 T(-107)C in women in the pre and postmenopausal stages Genotype

Overall

Premenopausal

Postmenopausal

p value

24.6% (17/69)

21.6% (8/37)

28.1% (9/32)

1.00

43.5% (30/69)

37.8% (14/37)

50.0% (16/32)

0.33

31.9% (22/69)

40.4% (15/37)

21.9% (7/32)

0.12

Proportions are presented as percentages and where compared using Chi-square test.

was a linear effect of the presence of the C allele (p = 0.003), where carriers of the C allele had higher PON1 activity (CC: 88.9 ± 6.5 U/mL and CT:79.9 ± 4.7 U/mL) than women of the TT genotype (66.6 ± 5.9 U/mL) (p < 0.05). There was no effect of the genotype on concentrations of glucose, HbA1c, total cholesterol, HDL, LDL and triglycerides (p > 0.05). Consumption of calories, macronutrients, cholesterol and fiber for each group of pre and postmenopausal women is shown in Table 4. Premenopausal women had a higher intake of calories, 100

a a

Paraoxonase (U/mL)

protein and carbohydrate than postmenopausal women (p < 0.05), but no differences were observed for fat and cholesterol intake (p > 0.05). Additionally, when categorizing foods in processed, in natura and rich in cholesterol/SFA, we observed no difference in the frequency of consumption of these types of food between pre and postmenopausal women (Table 4; p > 0.05). There was a weak association of the levels of in natura food consumption with serum PON1 activity (p = 0.04, R2 = 0.06); however, there was no effect of consumption of processed (p = 0.25) or cholesterol/ SFA foods (p = 0.11) on PON1 activity. The crude and adjusted analyzes of the association between PON1 activity and genotype interaction with intake profile are presented in Table 5. The enzyme variation was estimated at 0.12 (12.0%) by adjusted determination coefficients (R2) with inclusion of total diet Table 5. Unadjusted and adjusted analyzes of the association between PON1 activity and genotype interaction with body mass index and food intake profile

Genotype

66.95

5.31

REF

18.61

8.04

0.32

12.42

7.04

P-value

PON1-EV

REF

0.020

85.56

0.083

79.37

Unadjusted 0

Figure 1. Paraoxonase 1 (PON1) activity (U/mL) among women from the CC, CT and TT genotypes for the PON1 T(-107)C polymorphism. The data are presented as the means ± SEM. Different letters indicate significant differences at p < 0.05. Table 4. Daily estimated profile of food intake in patients in the pre and postmenopausal stages Parameter

Premenopausal

Postmenopausal

p value

Calories (kcal)

1849.5 ± 104.5

1541.2 ± 115.8

0.02

Carbohydrates (g)

223.1 ± 17.4

165.7 ± 20.1

0.01

Protein (g)

109.3 ± 8.5

84.9 ± 9.8

0.03

Lipids (g)

58.4 ± 2.7

52.8 ± 3.0

0.12

108.8 ± 11.3

95.3 ± 13.1

0.35

Processed1

1.8 ± 0.1

1.6 ± 0.1

0.16

In natura1

2.2 ± 0.2

2.4 ± 0.2

0.45

Fat/Cholesterol1

1.2 ± 0.1

1.0 ± 0.1

0.28

Cholesterol (mg)

Means are presented as least square means ± standard error of the mean and where compared using the MIXED model, using BMI and ethnicity as co-variates and the Tukey post-hoc test. 1 Weekly ingestion frequency of each class of food.

276

0.08

0.24 Model 1a

60.96

15.23

REF

17.62

8.32

0.30

11.59

7.04

0.10

0.22

REF

0.038

78.58

0.123

72.55

Model 2b TT

58.78

9.45

REF

17.36

8.08

0.30

13.23

7.11

0.26

0.12

REF

0.036

76.14

0.068

72.01

REF

0.105

77.48

1.115

75.31

Model 3

63.70

14.99

REF

13.78

8.30

0.24

11.61

7.26

0.23

0.17

B: regression coefficient; SE: standard error; R2: adjusted determination coefficient; β: beta coefficient; PON1 EV: estimated value of Paraoxonase 1. a Adjusted for BMI and total dietary energy content, R2 p = 0.147, ANOVA; b Adjusted for total dietary lipid and cholesterol content, R2 p = 0.097, ANOVA; c Adjusted for BMI, total dietary energy, lipid and cholesterol content, R2 p = 0.079, ANOVA. Arch Endocrinol Metab. 2019;63/3

lipid content and dietary cholesterol in the multiple linear regression model (model 2) (p < 0.05). In Table 6 the analysis of the mediators of PON1 change according to menopausal stage is demonstrated. BMI and total dietary lipid content when associated to PON1 genotype were the main factors for enzyme variation in postmenopausal women, with 0.12 (12.0%) and 0.31 (31.0%) of R2, respectively (p < 0.05). An equivalent analysis was performed for premenopausal women and a similar pattern regarding lipid content was observed (p > 0.05). Table 6. Adjusted analyzes of the association between PON1 activity and genotype interaction with food intake profile for pre and postmenopausal women B

Î˛

p-Value

Genotype (1)

67.50

3.90

0.31

0.09

0.010

1+BMI (2)

0.37

0.35

0.13

0.11

0.028

2+Energy (3)

-0.03

0.00

-0.12

0.12

0.046

3+Total fat (4)

0.77

0.45

0.16

0.029

4+Cholesterol (5)

0.07

0.06

0.15

0.18

0.034

Genotype (1)

67.86

5.15

0.25

0.06

0.136

1+BMI (2)

-0.18

0.50

-0.06

0.06

0.362

2+Energy (3)

0.00

-0.12

0.08

0.486

3+Total fat (4)

0.49

0.50

0.35

0.11

0.499

4+Cholesterol (5)

0.18

0.08

0.47

0.24

0.148

Genotype (1)

66.86

13.47

0.34

0.12

0.050

1+BMI (2)

1.10

0.52

0.34

0.24

0.019

2+Energy (3)

0.00

0.01

-0.07

0.24

0.047

3+Total fat (4)

2.00

1.22

0.88

0.31

0.033

4+Cholesterol (5)

0.05

0.10

0.08

0.32

0.063

Parameter (Model) All participants (n = 66)

Premenopausal (n = 34)

Postmenopausal (n = 32)

B: regression coefficient; SE: standard error; R : adjusted determination coefficient; Î˛: beta coefficient. 2

DISCUSSION In this study we investigated changes in serum PON1 activity in pre and postmenopausal women and its association with the PON1 C(-107)T SNP and food consumption during those stages. It was observed that the TT genotype carriers had lower serum PON1 activity than the CT and CC genotypes carriers. However, menopause did not affect serum PON1 activity and there was no combined effect of the genotype and the menopause status on serum PON1 activity. Despite that, postmenopausal women had Arch Endocrinol Metab. 2019;63/3

higher levels of glucose and total cholesterol. Total fat and cholesterol contributed with the genotype to affect serum PON1 activity. We did not found difference in PON1 activity between pre and postmenopausal women which is in agreement with a previous report (29). Despite that, some have reported decreased serum PON1 activity after surgically induced menopause in women, which was restored by the estrogen replacement therapy (13), suggesting estradiol as an important PON1 regulator. In fact, it is suggested that women using oral contraceptives have increased PON1 activity (30), but we did not observe any effect of the contraceptive use on serum PON1 activity in the current study. This lack of difference can be related to the age of the participants, which were all older than 35 years, and already have declining estradiol concentrations (31). Even though PON1 activity did not change between pre and postmenopausal women, it was observed an increase in total cholesterol and also a tendency for higher LDL concentrations. These changes in the lipid profile were previously demonstrated by others (29). The high prevalence of obesity associated with higher cholesterol and glucose concentrations in these postmenopausal women are important risk-factors for chronic diseases such as hypertension and type II diabetes, which also contribute to the occurrence of CVD. The majority of women in the present study was from the PON1(-107) CT genotype, and a smaller proportion was from the CC genotype, which was associated with the highest levels of PON1 activity. This distribution is similar to previous findings (15), including in a study with a group of women from southern Brazil (16). Overall, women carrying the C allele had higher PON1 activity than women of the TT genotype, as observed before (15,16). Despite this, PON1 activity was not different between menopause and pre menopause stage and this lack of difference was independent of the genotype. Since menopause is associated with several biochemical and endocrine changes (3,4), we could expect that it would challenge PON1 activity differently for women carrying different genotypes. However, in the current study this was not observed and the results suggest that PON1 activity remains unchanged after menopause, being mainly regulated by the PON1 T(-107)C genotype. Others had shown interactions of the diet and lifestyle with the PON1 genotype (16,32). In this study, no major influence of lipid and processed foods 277

Paraoxonase: menopause, genetics and diet

on PON1 activity were observed when categorizing women by median food intake. However, the model predicting PON1 activity was slightly better when genotype, total fat and cholesterol content in the diet were all included. This was more evident when considering only post-menopausal women. These findings are in agreement with previous observations in larger populations, indicating that the main modulator of PON1 activity is the genotype and not the diet or lifestyle (21). Others (21) working with very large group of subjects found a contribution lower than 10% of the diet over serum PON1 activity, with the majority of the effects remaining unexplained, and also found genetics as the main regulator. Among the diet effects, some found that the type of lipids and fatty acids have a significant effect on PON1 activity (33), as we currently observed. Previous studies had also found an interaction between food consumption and the PON1 genotype in women (16,32). This is further indication of the complexity of the modulation of serum PON1 activity, with many factors regulating its activity and interacting with the genotype. Although the menopausal stage had no effect in PON1 activity, it seems that this group was slightly more susceptible to the effects of the diet. It is well known that high consumption of fruits and vegetables is associated with lower CVD risk in women (34), and as PON1 is critical for prevention of CVD it is expected that it will also be increased by this type of food. Indeed, in type 2 diabetes patients, higher fruits and vegetables intake increased the activity of enzymes associated with the antioxidant properties of HDL, including PON1 (35), which is in alignment with our findings over consumption of in natura foods. In sum, we observed the PON1 C(-107)T genotype as the major regulator of serum PON1 activity. We did not observe any effect of the menopausal stage on serum PON1 activity. However, the model predicting PON1 activity was slightly better when genotype, total fat and cholesterol content in the diet were all considered together, especially for postmenopausal women. The lipid and glycemic profile were altered in postmenopausal women, which associated with increased BMI, can predispose this population to the development of CVD and diabetes. Our findings should be considered carefully as women in this study had different intervals between menopause and the moment of sample collection, which can shift the trends observed. 278

Acknowledgements: this work was supported by CNPq, CAPES and FAPERGS. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. te Velde ER, Scheffer GJ, Dorland M, Broekmans FJ, Fauser BC. Developmental and endocrine aspects of normal ovarian aging. Mol Cell Endocrinol. 1998;145(1-2):67-73. 2. Stevenson JC, Crook D, Godsland IF. Influence of age and menopause on serum lipids and lipoproteins in healthy women. Atherosclerosis. 1993;98(1):83-90. 3. Kim CJ, Kim TH, Ryu WS, Ryoo UH. Influence of menopause on high density lipoprotein-cholesterol and lipids. J Korean Med Sci. 2000;15(4):380-6. 4. Carr MC. The emergence of the metabolic syndrome with menopause. J Clin Endocrinol Metab. 2003;88(6):2404-11. 5. Wellons M, Ouyang P, Schreiner PJ, Herrington DM, Vaidya D. Early menopause predicts future coronary heart disease and stroke: the Multi-Ethnic Study of Atherosclerosis. Menopause. 2012;19(10):1081-7. 6. Sutton-Tyrrell K, Lassila HC, Meilahn E, Bunker C, Matthews KA, Kuller LH. Carotid atherosclerosis in premenopausal and postmenopausal women and its association with risk factors measured after menopause. Stroke. 1998;29(6):1116-21. 7.

Kattoor AJ, Pothineni NVK, Palagiri D, Mehta JL. Oxidative Stress in Atherosclerosis. Curr Atheroscler Rep. 2017;19(11):42.

8. Hasselwander O, McEneny J, McMaster D, Fogarty DG, Nicholls DP, Maxwell AP, et al. HDL composition and HDL antioxidant capacity in patients on regular haemodialysis. Atherosclerosis. 1999;143(1):125-33. 9. Mackness MI, Arrol S, Durrington PN. Paraoxonase prevents accumulation of lipoperoxides in low-density lipoprotein. FEBS letters. 1991;286(1-2):152-4. 10. James RW, Brulhart-Meynet MC, Singh AK, Riederer B, Seidler U, Out R, et al. The scavenger receptor class B, type I is a primary determinant of paraoxonase-1 association with high-density lipoproteins. Arterioscler Thromb Vasc Biol. 2010;30(11):2121-7. 11. Costa LG, Vitalone A, Cole TB, Furlong CE. Modulation of paraoxonase (PON1) activity. Biochem Pharmacol. 2005;69(4): 541-50. 12. Ahmad S, Scott JE. Estradiol enhances cell-associated paraoxonase 1 (PON1) activity in vitro without altering PON1 expression. Biochem Biophys Res Commun. 2010;397(3):441-6. 13. Kumru S, Aydin S, Aras A, Gursu MF, Gulcu F. Effects of surgical menopause and estrogen replacement therapy on serum paraoxonase activity and plasma malondialdehyde concentration. Gynecol Obstet Invest. 2005;59(2):108-12. 14. Butorac D, Celap I, Kackov S, Robic V, Miletic T, Mestric ZF, et al. Paraoxonase 1 activity and phenotype distribution in premenopausal and postmenopausal women. Biochem Med (Zagreb). 2014;24(2):273-80. 15. Campo S, Sardo MA,Trimarchi G, Bonaiuto M, Fontana L, Castaldo M, et al. Association between serum paraoxonase (PON1) gene promoter T(-107)C polymorphism, PON1 activity and HDL levels in healthy Sicilian octogenarians. Exp Gerontol. 2004;39(7): 1089-94. 16. Santos FG, Becker MK, Correa VS, Garcia DN, Vale SC, CrespoRibeiro JA, et al. The effect of the paraoxonase 1 (PON1) T(-107)C Arch Endocrinol Metab. 2019;63/3

Paraoxonase: menopause, genetics and diet

polymorphism on serum PON1 activity in women is dependent on fatty acid intake. Nutr Res. 2016;36(1):9-15. 17. Brophy VH, Jampsa RL, Clendenning JB, McKinstry LA, Jarvik GP, Furlong CE. Effects of 5’ regulatory-region polymorphisms on paraoxonase-gene (PON1) expression. American journal of human genetics. 2001;68(6):1428-36. 18. Mackness B, Davies GK, Turkie W, Lee E, Roberts DH, Hill E, et al. Paraoxonase status in coronary heart disease: are activity and concentration more important than genotype? Arterioscler Thromb Vasc Biol. 2001;21(9):1451-7. 19. Wheeler JG, Keavney BD, Watkins H, Collins R, Danesh J. Four paraoxonase gene polymorphisms in 11212 cases of coronary heart disease and 12786 controls: meta-analysis of 43 studies. Lancet. 2004;363(9410):689-95. 20. Hu FB, Stampfer MJ, Manson JE, Rimm E, Colditz GA, Rosner BA, et al. Dietary fat intake and the risk of coronary heart disease in women. N Engl J Med. 1997;337(21):1491-9. 21. Kim DS, Maden SK, Burt AA, Ranchalis JE, Furlong CE, Jarvik GP. Dietary fatty acid intake is associated with paraoxonase 1 activity in a cohort-based analysis of 1,548 subjects. Lipids Health Dis. 2013;12:183. 22. Kontis V, Bennett JE, Mathers CD, Li G, Foreman K, Ezzati M. Future life expectancy in 35 industrialised countries: projections with a Bayesian model ensemble. Lancet. 2017;389(10076): 1323-35.

27. Browne RW, Koury ST, Marion S, Wilding G, Muti P, Trevisan M. Accuracy and biological variation of human serum paraoxonase 1 activity and polymorphism (Q192R) by kinetic enzyme assay. Clin Chem. 2007;53(2):310-7. 28. Griffiths L, Chacon-Cortes D. Methods for extracting genomic DNA from whole blood samples: current perspectives. J Biorepos Sci Appl Med. 2014:1. 29. Mascarenhas-Melo F, Sereno J,Teixeira-Lemos E, Ribeiro S, RochaPereira P, Cotterill E, et al. Markers of increased cardiovascular risk in postmenopausal women: focus on oxidized-LDL and HDL subpopulations. Dis Markers. 2013;35(2):85-96. 30. Kowalska K, Sciskalska M, Bizon A, Sliwinska-Mosson M, Milnerowicz H. Influence of oral contraceptives on lipid profile and paraoxonase and commonly hepatic enzymes activities. J Clin Lab Anal. 2018;32(1). 31. Randolph JF, Jr., Sowers M, Bondarenko IV, Harlow SD, Luborsky JL, Little RJ. Change in estradiol and follicle-stimulating hormone across the early menopausal transition: effects of ethnicity and age. J Clin Endocrinol Metab. 2004;89(4):1555-61. 32. Rantala M, Silaste ML, Tuominen A, Kaikkonen J, Salonen JT, Alfthan G, et al. Dietary modifications and gene polymorphisms alter serum paraoxonase activity in healthy women. J Nutr. 2002;132(10):3012-7.

23. NICE. Diagnosis of perimenopause and menopause. Menopause: Full Guideline. National Institute for Health and Care Excellence: Clinical Guidelines. London; 2015.

33. Calabresi L, Villa B, Canavesi M, Sirtori CR, James RW, Bernini F, et al. An omega-3 polyunsaturated fatty acid concentrate increases plasma high-density lipoprotein 2 cholesterol and paraoxonase levels in patients with familial combined hyperlipidemia. Metabolism. 2004;53(2):153-8.

24. Sichieri R, Everhart JE. Validity of a Brazilian food frequency questionnaire against dietary recalls and estimated energy intake. Nutr Res. 1998;18:10.

34. Liu S, Manson JE, Lee IM, Cole SR, Hennekens CH, Willett WC, et al. Fruit and vegetable intake and risk of cardiovascular disease: the Women’s Health Study. Am J Clin Nutr. 2000;72(4):922-8.

25. Saúde Md. Guia alimentar para a população brasileira. In: Básica DdA, editor. 2014. p. 152.

35. Daniels JA, Mulligan C, McCance D, Woodside JV, Patterson C, Young IS, et al. A randomised controlled trial of increasing fruit and vegetable intake and how this influences the carotenoid concentration and activities of PON-1 and LCAT in HDL from subjects with type 2 diabetes. Cardiovasc Diabetol. 2014;13:16.

26. World Health Organization W. Physical status: the use and interpretation of anthropometry. Report of a WHO expert committee; 1995.

Arch Endocrinol Metab. 2019;63/3

279

original article

MTHFR gene polymorphisms in hypothyroidism and hyperthyroidism among Jordanian females Diala W. Abu-Hassan1, Abdullah N. Alhouri2, Nadera A. Altork2, Zakaria W. Shkoukani2, Tamer Salhab Altamimi2, Omar M. Alqaisi2, Baha Mustafa2

ABSTRACT Department of Physiology and Biochemistry, School of Medicine, University of Jordan, Amman, Jordan 2 School of Medicine, University of Jordan, Amman, Jordan 1

Correspondence to: Diala W Abu-Hassan Department of Physiology and Biochemistry, School of Medicine, the University of Jordan, Amman, Jordan 11942 d.abuhassan@ju.edu.jo or d.abuhassan@gmail.com Received on Oct/8/2018 Accepted on Feb/18/2019 DOI: 10.20945/2359-3997000000133

Objective: Methylenetetrahydrofolate reductase (MTHFR) is involved in DNA methylation that is associated with autoimmune pathology. We investigated the association between MTHFR genetic polymorphisms at g.677C>T and g.1298A>C and their haplotypes, and the risk of thyroid dysfunction among Jordanian females. Subjects and methods: A case-control study involving 98 hypothyroidism cases, 66 hyperthyroidism cases and 100 controls was conducted. Polymerase chain reaction/restriction fragment length polymorphism technique was performed to determine genotypes. Statistical analysis using SPSS software was performed. Results: Genetic analysis showed a significant difference in genotype frequency of g.1298A>C between cases, and controls [hypothyroidism: AA (45.9%), AC (37.8%), CC (16.3%); hyperthyroidism: AA (9.1%), AC (69.7%), CC (21.2%); controls: AA (37.8%), AC (29.6%), CC (32.7%); CChypo vs. AAhypo: 2.55, 95% CI: (1.18-5.52); OR at least on Chypo: 1.79, 95% CI: (1.07-2.99)]; CChyper vs. AAhyper: 4.01, 95% CI: (1.79-9.01); OR at least on Chyper: 0.18, 95% CI: (0.07-0.48)]. There was no significant difference in genotype frequency of g.677C>T between cases and controls [hypothyroidism: CC (50.0%), CT (32.7%), TT (17.3%); hyperthyroidism: CC (77.3%), CT (15.2%), TT (7.6%); controls: CC (55.6%), CT (32.3%), TT (12.1%)]. There was a significant difference of MTHFR haplotypes among hypothyroidism cases and controls. TA and CC had a lower hypothyroidism risk whereas; TC showed a higher risk. Conclusions: g.1298A>C genetic polymorphism of MTHFR may modulate the risk of thyroid disease. CC, TA, and TC haplotypes affect the risk of hypothyroidism. Larger samples should be included in the future to verify the role of MTHFR polymorphisms in thyroid diseases. Arch Endocrinol Metab. 2019;63(3):280-7 Keywords deficiency, folic acid, metabolism, thyroid, thyroxine

INTRODUCTION

ypothyroidism is a common health problem with a worldwide annual incidence of 1.5 cases per 1,000 individuals (1,2). It occurs more often in females, with an incidence around 10-15 times higher than in males (1,2). In the United States, it has been estimated to occur in 3.5 per 1,000 women and 0.8 per 1,000 men yearly (3,4). In Jordan, the overall prevalence of thyroid disease is 12.5% (5). The worldwide prevalence of congenital hypothyroidism is 1:4,000, while in Jordan it is 1:1,719 (6). Hypothyroidism during pregnancy has been associated with gestational diabetes, premature deliveries, offspring with low intelligence, risk of peripartum death, and a higher risk of spinal cord malformations (e.g., spina bifida) and Down’s syndrome (7-9). Since the majority of those affected by Hashimoto’s auto-immune thyroiditis are females of child-bearing age, they are more inclined to experience these complications. 280

DNA methylation has been shown to influence gene expression in several studies (10). As an epigenetic change, DNA methylation regulates several biological events, including embryonic development, transcriptional regulation, chromatin modification, X-chromosome inactivation, and genomic imprinting (11). Changes in DNA methylation patterns have been correlated with tumorigenesis and autoimmune disease development (12). The methylenetetrahydrofolate reductase (MTHFR) enzyme catalyzes the conversion of 5, 10-methylenetetrahydrofolate to 5-methyltetrahydrofolate. As the primary circulatory form of folate, 5-methyltetrahydrofolate is utilized to supply methyl groups in several methylation reactions. MTHFR contributes to the hypermethylation of genomic DNA (13,14). The resulting DNA hypermethylation may affect genes that influence the risk of autoimmune thyroid diseases (AITD), leading to Arch Endocrinol Metab. 2019;63/3

MTHFR polymorphisms and thyroid disease

SUBJECTS AND METHODS Study population A total of 264 female patients attending the National Center for Diabetes, Endocrinology and Genetics (NCDEG) in Amman, Jordan were recruited in this case-control study. Ninety-eight hypothyroidism patients, 66 hyperthyroidism patients, and 100 agematched healthy women were involved (Figure 1). Subject recruitment Females who age 20-65 Hypothyroidism, hyperthyroidism and controls

History taking, consent signing and blood sampling DNA extraction PCR, RFLP and gel electrophoresis Genotyping Statistical analysis

Figure 1. Experimental scheme. Arch Endocrinol Metab. 2019;63/3

Hyperthyroidism patients were involved in the study for comparison. The study was approved by the institutional review board (IRB) of the NCDEG in compliance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Declaration of Helsinki. Blood withdrawal from each subject was performed with their permission following an explanation of the purpose of the study. Each participant signed a consent form before giving a sample. The authors declared no potential conflicts of interest to study participants.

MTHFR genotyping Four milliliters of venous blood were collected from patients and healthy subjects in K3EDTA coated tubes. DNA extraction was performed the same day using QIAGEN Puregene Blood Core Kit B (QIAGEN Sciences, Maryland, USA) according to manufacturer’s instructions. The g.677C>T region (rs1801133) was amplified using the forward primer TGAAGGAGAAGGTGTCTGCGGGA (NT_021937.19: 7,861,134 to 7,861,112) and the reverse primer AGGACGGTGCGGTGAGAGTG (NT_021937.19: 7,860,937 to 7,860,956) yielding a 198-bp band. For the g.1298A>C polymorphism (rs1801131), the forward primer CAAGGAGGAGCTGCTGAAGA (NT_021937.19: 7,859,255 to 7,859,236) and the reverse primer CCACTCCAGCATCACTCACT (NT_021937.19: 7,859,128 to 7,859,147) were used, yielding a 128-bp band. The polymerase chain reaction (PCR) conditions for both polymorphisms were: 8 minutes of initial denaturation at 95ºC followed by 40 cycles of 95ºC for 60 seconds, 63ºC for 60 seconds, and 72ºC for 60 seconds, with a final extension at 72ºC for 7 minutes (Bio-Rad, C1000 Thermal cycler™, USA). After PCR was completed, the product of g.677C>T was digested with HinfI restriction enzyme. The resulting fragments were separated by 3% agarose gel electrophoresis and then visualized using RedSafe™ (iNtRON BIOTECHNOLOGY, Korea) staining (Figure 1). The digestion fragment sizes for the g.677C>T genotypes were: 198-bp bands for CC, 175, 23-bp bands for TT, and 198, 175, and 23-bp bands for CT. After PCR was completed, the products of g.1298A>C were digested with MboII restriction enzyme. The resulting fragments were separated by 3% agarose gel electrophoresis and then visualized 281

pathological changes in thyroid gland function. The two most common genetic polymorphisms of the MTHFR gene worldwide are the g.677C>T and g.1298A>C variants, with the g.677C>T variant being present in 24% of the healthy Jordanian population (15). In genetically isolated groups in Jordan, such as Chechens and Circassians, the prevalence of the g.677C>T polymorphism is 27.5% and 50%, respectively (16). The MTHFR +677C/T polymorphism (rs1801133) results in an alanine (C)-to-valine (T) substitution and reduces the enzymatic activity of MTHFR (17,18). The MTHFR +1298A/C polymorphism (rs1801131) results in a glutamic acid (A)-to-alanine (C) substitution leading to a significant decrease of MTHFR enzyme activity in CC genotype individuals (19). Genetic variation in this gene affects the susceptibility to many types of cancer, including thyroid cancer (20-24). In this case-control study, we investigated the relationship between the common genetic polymorphisms of MTHFR, g.677C>T, and g.1298A>C and their effect on susceptibility to thyroid dysfunction among Jordanian females.

MTHFR polymorphisms and thyroid disease

using RedSafe™ (iNtRON BIOTECHNOLOGY, Korea) staining. The digestion fragment sizes for the g.1298A>C genotypes were: 72 and 28-bp bands for AA, 100 and 28-bp bands for CC, and 100, 72 and 28-bp bands for AC. Polymerase chain reactionrestriction fragment length polymorphism (PCRRFLP) findings were validated by running a negative control containing all PCR components except the DNA template in every PCR run, in addition to repeating around 23% of all samples by a different lab personnel.

Statistical analysis Data was coded and entered into SPSS software version 16 (Chicago, IL). Data was expressed as mean ± standard deviation (SD) or as counts (%). The presence of a statistical correlation between categorical variables was evaluated by a Chi-square test. The age difference between cases and controls was evaluated by an independent student t-test. Odds ratios (OR) and 95% confidence intervals (95% CI) were assessed for measuring the association between MTHFR genotypes/alleles/diplotypes/ haplotypes and thyroid disorders. A p value < 0.05 was considered statistically significant in all analyses performed in this study. The Hardy-Weinberg equilibrium was applied to assess genotypes and allele frequencies.

Haplotype analysis The interaction between genetic polymorphisms at the two loci was assessed by evaluating the combined genotypes’ effects and haplotype analysis. We analyzed the haplotype frequencies of the two single nucleotide polymorphisms, SNPs, g.677C>T, and g.1298A>C for hypothyroidism and hyperthyroidism cases and compared them with those of the controls. Haplotype frequencies were calculated using Multiallelic Interallelic Disequilibrium Analysis Software (MIDAS; University of Southampton, Highfield, Southampton, UK) and linkage disequilibrium was represented by D prime (D’).

47.6 ± 10.0 years. No significant age differences were observed between the three arms (p = 0.35). Table 1 presents the distribution of the study subjects according to thyroid status in several age groups. The distribution of MTHFR g.677C>T and g.1298A>C genotypes is shown in Table 2. There were statistically significant differences in the genotype frequency of MTHFR g.1298A>C polymorphism between hypothyroidism patients and the controls. The frequency of the MTHFR C allele was significantly lower in hypothyroidism and hyperthyroidism patients in comparison with healthy controls. Subjects with AC and CC (g.1298A>C) genotypes had significantly lower risks of hypothyroidism than those with AA (g.1298A>C) (p = 0.019 and 0.017, respectively; Table 2). The C (g.1298A>C) allele is less frequent in hyperthyroidism cases when compared to controls (Table 2). The CC (g.1298A>C) genotype is less frequent in hyperthyroidism patients than in controls (Table 2). The T allele was significantly less frequent in hyperthyroidism patients compared to controls (Table 2). No other significant differences were found in the g.677C>T SNP between hypothyroidism cases, hyperthyroidism cases, and healthy controls (Table 2). The haplotypes appearing in our results are CA, CC, TA, and TC. The most frequent haplotypes were CA (677C-1298A) (hypothyroidism cases: 51.5%; hyperthyroidism cases: 45.9%; controls: 35.4%) followed by CC (677C-1298C) (hypothyroidism cases: 14.4%; hyperthyroidism cases: 39.0%; controls: 36.3%). The rarest haplotype in hypothyroidism and hyperthyroidism cases was TA: 13.3% and 5.0%, Table 1. Age characteristics of cases and controls Condition Hypothyroidism

Cases

Controls No. (%)

Age (years)*

50.7 ± 7.786

45.3 ± 10.045

0.37

Age groups

20-29

8 (8.2)

4 (4.0)

30-39

20 (20.4)

6 (6.0)

40-49

26 (26.5)

21 (21.0)

50-59

43 (43.9)

67 (67.0)

Hyperthyroidism

≥ 60

1 (1.0)

2 (2.0)

Age (years)

46.33 ± 11.855

45.3 ± 10.045

0.11

RESULTS

Age groups

20-29

7 (10.6)

4 (4.0)

30-39

12 (18.2)

6 (6.0)

A total of 264 female subjects (98 hypothyroidism patients, 66 hyperthyroidism patients, and 100 age-matched controls) were included in this study. The mean age for patients and control groups was

40-49

15 (22.7)

21 (21.0)

50-59

24 (36.4)

67 (67.0)

≥ 60

8 (12.1)

2 (2.0)

282

*Age was presented in years ± standard deviation. Arch Endocrinol Metab. 2019;63/3

MTHFR polymorphisms and thyroid disease

Table 2. MTHFR g.677C > T and MTHFR A1298 genotypes and allele types and risk of thyroid disease Condition

Genotype or allele

Hypothyroidism

No. (%)* Cases

Controls

Odds Ratio (95% CI)

MTHFR g.677C > T

N = 100

49 (50.0)

55 (55.6)

1 (reference)

32 (32.7)

32 (32.3)

0.63 (0.27-1.45)

0.28

17 (17.3)

12 (12.1)

0.71 (0.29-1.71)

0.44

CC+CT

81 (82.7)

87 (87.9)

1 (reference)

17 (17.3)

12 (12.1)

0.48 (0.20-1.14)

0.096

130 (66.3)

142 (71.7)

1 (reference)

66 (33.7)

56 (28.3)

0.73 (0.44-1.21)

0.22

MTHFR g.1298A > C

45 (45.9)

37 (37.8)

1 (reference)

37 (37.8)

29 (29.6)

2.43 (1.16-5.10)

0.019

16 (16.3)

32 (32.7)

2.55 (1.18-5.52)

0.017

AA+AC

82 (83.7)

66 (67.3)

1 (reference)

16 (16.3)

32 (32.7)

2.96 (1.44-6.11)

0.003

127 (64.8)

103 (52.6)

1 (reference)

69 (35.2)

93 (47.4)

1.79 (1.07-2.99)

0.027

Hyperthyroidism

MTHFR g.677C > T

N = 66

51 (77.3)

55 (55.6)

1 (reference)

10 (15.2)

32 (32.3)

2.94 (0.89-9.74)

0.078

5 (7.6)

12 (12.1)

1.05 (0.27-4.07)

0.95

CC+CT

61 (92.4)

87 (87.9)

1 (reference)

5 (7.6)

12 (12.1)

1.72 (0.57-5.52)

0.333

112 (84.8)

142 (71.7)

1 (reference)

20 (15.2)

56 (28.3)

2.23 (1.12-4.45)

0.023

MTHFR g.1298A > C

6 (9.1)

37 (37.8)

1 (reference)

46 (69.7)

29 (29.6)

0.42 (0.14-1.26)

0.123

14 (21.2)

32 (32.7)

4.01 (1.79-9.01)

0.001

AA+AC

52 (78.8)

66 (67.3)

1 (reference)

14 (21.2)

32 (32.7)

1.83 (0.88-3.79)

0.104

58 (43.9)

103 (52.6)

1 (reference)

74 (56.1)

93 (47.4)

0.18 (0.07-0.48)

0.001

* Data represent actual numbers with percent in parentheses.

Arch Endocrinol Metab. 2019;63/3

DISCUSSION Since thyroid dysfunction and MTHFR polymorphisms are common among Jordanians, investigating the genes and polymorphisms involved in the folate metabolic pathway may assist in determining patient susceptibility to thyroid diseases and may help in early detection and management of the disease, particularly in childbearingage females. 283

respectively. Our results indicated that the two loci 677 and 1298 show linkage disequilibrium (LD) between cases (D’hypo = 0.39, D’hyper = 0.25) and fair LD in controls (D’ = 0.14). Carriers of the CC (677T-1298A) and TA (677T-1298A) haplotypes had significantly lower risks for hypothyroidism, whereas those with TC (677T-1298A) haplotypes had a higher likelihood of having hypothyroidism (Table 3).

MTHFR polymorphisms and thyroid disease

Table 3. Haplotype frequencies of MTHFR among hypothyroidism patients, hyperthyroidism patients and controls Condition

Haplotype

Cases

Controls No. (%)*

Odds Ratio

(95% CI)

p 0.57

Hypothyroidism

677C-1298A

101 (51.5)

69 (35.4)

0.84

0.46-1.53

677C-1298C

29 (14.4)

70 (36.3)

0.24

0.12-0.47

677T-1298A

26 (13.3)

32 (16.7)

0.47

0.23-0.97

0.04

677T-1298C

40 (20.4)

23 (11.7)

2.14

1.03-4.44

0.04

Hyperthyroidism

677C-1298A

61 (45.9)

69 (35.4)

0.88

0.45-1.73

0.72

677C-1298C

51 (39.0)

70 (36.3)

0.51-1.95

677T-1298A

7 (5.0)

32 (16.7)

0.47-2.13

677T-1298C

13 (10.2)

23 (11.7)

0.51-1.95

* Counts reflect the number of chromosomes.

Although many molecular and epidemiological studies have been conducted worldwide in the past few decades concerning the relationship between MTHFR gene polymorphisms and cancer (e.g., lung, breast, and other types of cancers), only a few studies have been conducted on the correlation between these polymorphisms and the dysfunction of thyroid glands (17,18). Among the Japanese population, no association of g.677C>T or g.1298A>C polymorphisms and AITD leading to hypothyroidism or hyperthyroidism was detected; additionally, the genotype and allele frequencies of g.677C>T and g.1298A>C did not inﬂuence the prognosis of AITD (25). A meta-analysis involving 9 studies showed no association of g.677C>T polymorphism with thyroid disease, but an association was detected with thyroid cancer (OR T vs. C = 1.09, 95% CI 0.94–1.26, p = 0.25; OR TT vs. CC=1.04, 95% CI 0.75–1.42, p = 0.83; OR TT vs. CC/CT=1.13, 95% CI 0.86–1.50, p = 0.37; OR TT/CT vs. CC = 1.22, 95% CI 0.88–1.68, p = 0.24) (20). Our study involved two common polymorphisms and only female subjects were recruited; the g.1298A>C polymorphism showed more significant results than the g.677C>T polymorphism in our sample. A familial history of AITD is commonly present. Genome-wide linkage analysis was the first approach employed to screen the genome for the genetic contribution to AITD. However, linkage studies lacked success for almost all complex diseases, suggesting that it can succeed for monogenic diseases but not for complex diseases. Linkage analysis has not revealed novel susceptibility loci for AITD, but the genetic contribution to AITD may have several effects that can be summarized as follows: (i) There was not a large genetic effect discovered for AITD, indicating that disease susceptibility might be due to small effects of 284

multiple genes; (ii) Differences between the genetic contribution in Asian and Caucasian subjects suggest that different sets of genes may contribute to disease susceptibility in different environments and races; (iii) Different combinations of genes may lead to similar clinical phenotypes; (iv) Epigenetic phenomena may have a dominant influence; and (v) More comprehensive screening utilizing the new techniques with greater understanding of the genome is needed. PTPN22, IL-2RA/CD25, CD40, and SCGB3A2 genes were shown to be associated with Graves’ disease (GD) using candidate gene studies (23,24,26-31). HLA-DR3, cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), and the TSHR genes were shown to be major susceptibility genes for GD and Hashimoto’s thyroiditis (HT) using the candidate gene approach (32). As discussed earlier, polymorphisms in various genes have been associated with AITD. The influence of each gene on AITD development when assessed in a population appears weaker than expected from the data that show strong genetic susceptibility to AITD. Explanations of this discrepancy include gene-gene interaction and subset effects. Further studies should be performed to elucidate the mechanism by which MTHFR deficiency would affect thyroid cellular functions and consequently the development of AITD. As discussed previously, genetic factors are important in the pathogenesis of autoimmune thyroiditis. The cooccurrence of HT and GD within one family suggests a common genetic basis for these diseases (33). In the whole-genome screening of families, siblings, and populations with AITD, a number of sites have been located for GD and HT susceptibility, but none of them have shown significant statistical values (3436). This result has been true for other autoimmune diseases, not just for AITD. This can be explained by Arch Endocrinol Metab. 2019;63/3

HLA subtypes; for example, not every patient with GD has the associated HLA-DR3 subtype or even the associated Arg74 in its binding pocket, irrespective of the HLA-DR subtype (37). As a result, the disease can occur even in the absence of the expected HLA association. The etiology of HT and GD involves common pathways that activate the escape of tolerance of thyroid-reactive T-cells and their infiltration of thyroid tissue as well as unique pathways that use the thyroidreactive T-cells to induce thyroid cell death (in HT) or stimulation (in GD). As a result, genetic susceptibility to HT and GD involves shared genes and unique ones. Our results have shown similar effects of g.1298A>C polymorphism on hypothyroidism and hyperthyroidism, implying an influence on the common pathways. Hence, mechanistic studies may assist in understanding the molecular basis of MTHFR’s role in the pathogenesis of AITD. Future studies should also include larger sample sizes to confirm results and establish an association that allows us to include MTHFR polymorphisms in the screening and examination of susceptible individuals. Once a mechanistic and statistical relationship is established, clinicians may introduce preventive guidelines for dealing with susceptible patients, including supplementation with commercially available 5-methyltetrahydrofolate (MTHFR product) to replace the reduced MTHFR function and reduce the risk of AITD. Both genetic and environmental risk factors contribute to the development of thyroid diseases. One mechanism by which environmental factors may promote AITD together with genetic factors is by changing the epigenetic control of gene expression. So far, little is known about these interactions in AITD. However, there has been wide confirmation of the role of X chromosome inactivation (XCI) (38,39). Patients with AITD showed more biased expression of a maternal or paternal X chromosome than normal individuals, supporting the hypothesis that the poorly expressed chromosome may become active in certain tissues (such as the thyroid) and begin to express new antigenic sequences not previously recognized by the immune system. However, this potential mechanism for enhanced susceptibility to AITD needs further investigation. Non-genetic risk factors also contribute to the development of autoimmune thyroid disease. These factors include radiation exposure, both from nuclear Arch Endocrinol Metab. 2019;63/3

fallout and medical radiation, and high iodine intake, in addition to several other environmental contaminants that influence the thyroid (40-42). Exposure to these factors increases the probability of thyroid dysfunction; this may explain the inability of the current study to find a highly significant association between MTHFR polymorphisms and thyroid disease in this sample of patients. The distribution of g.677C>T MTHFR genetic polymorphism was not significantly different between Jordanian females with hypothyroidism or hyperthyroidism and the control subjects. Individual MTHFR genetic polymorphisms might not independently affect the susceptibility to thyroid disease. Adjacent SNPs often show a high correlation between genotypes, meaning that they are in LD and that the interaction of the SNPs within haplotypes might act as a major determinant of disease susceptibility in comparison with the single polymorphisms (43). The analysis of haplotypes was reported by many studies to be more powerful than single polymorphism analysis (43). In this study, LD between g.677C>T and g.1298A>C in the MTHFR gene among hypothyroidism cases was detected. Our results show a significant increase in TC haplotypes among hypothyroidism patients compared to controls, implying a combined effect of g.677C>T and g.1298A>C polymorphisms. The limitations of this study include the relatively small sample size and the difficulty of matching cases and controls on several variables other than age and gender. Additionally, healthy recruits (controls) were not tested to confirm that they were free of thyroid disease at the time of recruitment, nor were follow-up tests performed later. Moreover, data regarding folate intake, folate serum level, and pregnancy and abortion history were not adequately collected or not measured at all. No reliable valid method was developed to assess patients’ dietary intake of folate; this may be significant, as flour in Jordan is fortified with iron and folic acid under the auspices of the Ministry of Health and the vast majority of Jordanians eat bread daily (Food Fortification Initiative, 2014) (44). Finally, the search for genes that increase susceptibility to thyroid autoimmune diseases has identified several candidates, but they only account for a small percentage of the current prevalence of these disorders. Hence, other genes that may be involved in the modulation of thyroid disease risk must be investigated. Recruiting only females in this study was consistent with their higher susceptibility to autoimmune diseases. 285

MTHFR polymorphisms and thyroid disease

In conclusion, the findings of our study suggest that genetic variants of MTHFR at g.1298A>C and its haplotype analysis at 677 and 1298 may modulate the risk of thyroid disorders in Jordanian females. This study is the sole one that has examined the role of MTHFR (g.677C>T and g.1298A>C) genetic polymorphisms and their haplotypes in the modulation of thyroid disease in the Jordanian population. Acknowledgments: we thank the King Abdullah II Fund for Development (KAFD) and King Abdullah Design and Development Bureau (KADDB) for funding this project, grant # 20. We also thank the National Center for Diabetes, Endocrinology and Genetics, Amman, Jordan for facilitating data collection. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1.

Vanderpump MP, French JM, Appleton D, Tunbridge WM, KendallTaylor P. The prevalence of hyperprolactinaemia and association with markers of autoimmune thyroid disease in survivors of the Whickham Survey cohort. Clin Endocrinol (Oxf). 1998;48(1):39-44.

2. Vanderpump MP, Tunbridge WM, French JM, Appleton D, Bates D, Clark F, et al. The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham Survey. Clin Endocrinol (Oxf). 1995;43(1):55-68. 3. Lee SL. Hashimoto Thyroiditis. Medscape. 2018. Available from: https://emedicine.medscape.com/article/120937-overview. Accessed on: Feb. 2019. 4. Vanderpump MP, Tunbridge WM, French JM, Appleton D, Bates D, Clark F, et al. The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham Survey. Clin Endocrinol (Oxf). 1995;43(1):55-68. 5. Radaideh AR, Nusier MK, Amari FL, Bateiha AE, El-Khateeb MS, Naser AS, et al. Thyroid dysfunction in patients with type 2 diabetes mellitus in Jordan. Saudi Med J. 2004;25(8):1046-50. 6. Alawneh H. Incidence of congenital hypothyroidism in Jordan. Menoufia Med J. 2014;27(2):503-6. 7. Negro R, Stagnaro-Green A. Diagnosis and management of subclinical hypothyroidism in pregnancy. BMJ. 2014;349:g4929. 8. Lucock M. Folic acid: nutritional biochemistry, molecular biology, and role in disease processes. Mol Genet Metab. 2000;71(12):121-38. 9. Sadiq MF, Al-Refai EA, Al-Nasser A, Khassawneh M, Al-Batayneh Q. Methylenetetrahydrofolate reductase polymorphisms C677T and A1298C as maternal risk factors for Down syndrome in Jordan. Genet Test Mol Biomarkers. 2011;15(1-2):51-7.

10. Razin A, Riggs AD. DNA methylation and gene function. Science. 1980;210(4470):604-10. 11. Mostoslavsky R, Bergman Y. DNA methylation: regulation of gene expression and role in the immune system. Biochim Biophys Acta. 1997;1333:F29-50.

C677T mutation in the methylenetetrahydrofolate reductase gene. J Nutr. 2000;130(9):2238-42. 14. Chen Z, Karaplis AC, Ackerman SL, Pogribny IP, Melnyk S, LussierCacan S, et al. Mice deficient in methylenetetrahydrofolate reductase exhibit hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic lipid deposition. Hum Mol Genet. 2001;10(5):433-43. 15. Eid SS, Rihani G. Prevalence of factor V Leiden, prothrombin G20210A, and MTHFR C677T mutations in 200 healthy Jordanians. Clin Lab Sci. 2004;17(4):200-2. 16. Dajani R, Fathallah R, Arafat A, AbdulQader ME, Hakooz N, AlMotassem Y, et al. Prevalence of MTHFR C677T Single Nucleotide Polymorphism in Genetically Isolated Populations in Jordan. Biochem Genet. 2013;51(9-10):780-8. 17. Diekman MJ, van der Put NM, Blom HJ, Tijssen JG, Wiersinga WM. Determinants of changes in plasma homocysteine in hyperthyroidism and hypothyroidism. Clin Endocrinol (Oxf). 2001;54(2):197-204. 18. Kang SS, Zhou J, Wong PW, Kowalisyn J, Strokosch G. Intermediate homocysteinemia: a thermolabile variant of methylenetetrahydrofolate reductase. Am J Hum Genet. 1988;43(4):414-21. 19. Lievers KJ, Boers GH, Verhoef P, den Heijer M, Kluijtmans LA, van der Put NM, et al. A second common variant in the methylenetetrahydrofolate reductase (MTHFR) gene and its relationship to MTHFR enzyme activity, homocysteine, and cardiovascular disease risk. J Mol Med. 2001;79(9):522-8. 20. Chen Y, Wang B, Yan S, Wang YG. Significant association between MTHFR C677T polymorphism and thyroid cancer risk: evidence from a meta-analysis. GenetTest Mol Biomarkers. 2014;18(10):695702. 21. Jiang Y, Hou J, Zhang Q, Jia ST, Wang BY, Zhang JH, et al. The MTHFR C677T polymorphism and risk of acute lymphoblastic leukemia: an updated meta-analysis based on 37 case-control studies. Asian Pac J Cancer Prev. 2013;14(11):6357-62. 22. Rai V. Methylenetetrahydrofolate Reductase A1298C Polymorphism and Breast Cancer Risk: A Meta-analysis of 33 Studies. Ann Med Health Sci Res. 2014;4(6):841-51. 23. Chen QY, Nadell D, Zhang XY, Kukreja A, HuangYJ, Wise Jet al.The human leukocyte antigen HLA DRB3*020/DQA1*0501 haplotype is associated with Graves’ disease in African Americans. J Clin Endocrinol Metab. 2000;85(4):1545-9. 24. Ueda H, Howson JM, Esposito L, Heward J, Snook H, Chamberlain G, et al. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature. 2003;423(6939):506-11. 25. Arakawa Y, Watanabe M, Inoue N, Sarumaru M, Hidaka Y, Iwatani Y. Association of polymorphisms in DNMT1, DNMT3A, DNMT3B, MTHFR and MTRR genes with global DNA methylation levels and prognosis of autoimmune thyroid disease. Clin Exp Immunol. 2012;170(2):194-201. 26. Vaidya B, Imrie H, Perros P, Young ET, Kelly WF, Carr D, et al. The cytotoxic T lymphocyte antigen-4 is a major Graves’ disease locus. Hum Mol Genet. 1999;8(7):1195-9. 27. Heward JM, Brand OJ, Barrett JC, Carr-Smith JD, Franklyn JA, Gough SC. Association of PTPN22 haplotypes with Graves’ disease. J Clin Endocrinol Metab. 2007;92:685-90.

12. Rountree MR, Bachman KE, Herman JG, Baylin SB. DNA methylation, chromatin inheritance, and cancer. Oncogene. 2001;20(24):3156-65.

28. Velaga MR, Wilson V, Jennings CE, Owen CJ, Herington S, Donaldson PT, et al. The codon 620 tryptophan allele of the lymphoid tyrosine phosphatase (LYP) gene is a major determinant of Graves’ disease. J Clin Endocrinol Metab. 2004;89(11):5862-5.

13. Stern LL, Bagley PJ, Rosenberg IH, Selhub J. Conversion of 5-formyltetrahydrofolic acid to 5-methyltetrahydrofolic acid is unimpaired in folate-adequate persons homozygous for the

29. Dechairo BM, Zabaneh D, Collins J, Brand O, Dawson GJ, Green AP, et al. Association of the TSHR gene with Graves’ disease: the first disease specific locus. Eur J Hum Genet. 2005;13(11):1223-30.

286

Arch Endocrinol Metab. 2019;63/3

MTHFR polymorphisms and thyroid disease

30. Hiratani H, Bowden DW, Ikegami S, Shirasawa S, Shimizu A, Iwatani Y, et al. Multiple SNPs in intron 7 of thyrotropin receptor are associated with Graves’ disease. J Clin Endocrinol Metab. 2005;90:2898-903. 31. Houston FA, Wilson V, Jennings CE, Owen CJ, Donaldson P, Perros P, et al. Role of the CD40 locus in Graves’ disease. Thyroid. 2004;14(7):506-9. 32. Tome Y. Genetic susceptibility to autoimmune thyroid disease: past, present, and future. Thyroid. 2010;20(7):715-25. 33. Strieder TG, Prummel MF, Tijssen JG, Endert E, Wiersinga WM. Risk factors for and prevalence of thyroid disorders in a crosssectional study among healthy female relatives of patients with autoimmune thyroid disease. Clin Endocrinol (Oxf). 2003;59(3):396-401. 34. Tomer Y, Barbesino G, Greenberg DA, Concepcion E, Davies TF. Mapping the major susceptibility loci for familial Graves’ and Hashimoto’s diseases: evidence for genetic heterogeneity and gene interactions. J Clin Endocrinol Metab. 1999;84(12):4656-64. 35. Tomer Y, Ban Y, Conception E, Barbesino G, Villanueva R, Greenberg DA, et al. Common and unique susceptibility loci in Graves and Hashimoto diseases: results of whole-genome screening in a data set of 102 multiplex families. Am J Hum Genet. 2003;73(4):736-47. 36. Taylor JC, Gough SC, Hunt PJ, Brix TH, Chatterjee K, Connell JM, et al. A genome-wide screen in 1119 relative pairs with autoimmune thyroid disease. J Clin Endocrinol Metab. 2006;91(2):646-53.

39. Yin X, Latif R, Tomer Y, Davies TF (2007) Thyroid epigenetics: X chromosome inactivation in patients with autoimmune thyroid disease. Ann N Y Acad Sci. 2007;1110:193-200. 40. Nagataki S, Shibata Y, Inoue N, Yokoyama N, Izumi M, Shimaoka K. Thyroid diseases among atomic bomb survivors in Nagasaki. JAMA. 1994;272(5):364-70. 41. Imaizumi M, Usa T, Tominaga T, Neriishi K, Akahoshi M, Nakashima E, et al. Radiation dose-response relationships for thyroid nodules and autoimmune thyroid diseases in Hiroshima and Nagasaki atomic bomb survivors 55-58 years after radiation exposure. JAMA. 2006;295(9):1011-22. 42. Vermiglio F, Castagna MG, Volnova E, Lo Presti VP, Moleti M, Violi MA, et al. Post-Chernobyl increased prevalence of humoral thyroid autoimmunity in children and adolescents from a moderately iodine-deficient area in Russia. Thyroid. 1999;9(8):781-6. 43. Zintzaras E, Lau J. Synthesis of genetic association studies for pertinent gene-disease associations requires appropriate methodological and statistical approaches. J Clin Epidemiol. 2008;61(7):634-45. 44. Food Fortification Initiative. Middle East – Fortification Status [Online]. 2014. Available from: http://www.ffinetwork.org/ regional_activity/middle_east.php#middle_east.php?s=2&_ suid=142601829808809411473741271781. Accessed on: Sep 10, 2017.

37. Ban Y, Davies TF, Greenberg DA, Concepcion ES, Osman R, Oashi T, et al. Arginine at position 74 of the HLA-DR beta1 chain is associated with Graves’ disease. Genes Immun. 2004 May;5(3):203-8.

38. Brix TH, Knudsen GP, Kristiansen M, Kyvik KO, Orstavik KH, Hegedüs L. High frequency of skewed x-chromosome inactivation in females with autoimmune thyroid disease: a possible explanation for the female predisposition to thyroid autoimmunity. J Clin Endocrinol Metab. 2005;90(11):5949-53.

Arch Endocrinol Metab. 2019;63/3

287

original article

Renal resistive index in patients with polycystic ovary syndrome Gülsen Tüfekçiog6lu1, SÇakir Özgür KesÇkek1, Okan Dilek2, Cengiz Yilmaz2

ABSTRACT Department of Internal Medicine, Numune Training and Research Hospital, Adana, Turkey 2 Department of Radiology, Numune Training and Research Hospital, Adana, Turkey 1

Correspondence to: SÇakir Özgür KesÇkek City Training and Research Hospital, Department of Internal Medicine Adana, Turkey drkeskek@yahoo.com Received on Oct/11/2018 Accepted on Mar/29/2019 DOI: 10.20945/2359-3997000000141

Objective: Polycystic ovary syndrome (PCOS) is a common endocrine disorder in women of reproductive age. The renal resistive index (RRI) is a measure of renal arterial resistance to blood flow. The aim of this study was to investigate the renal resistive index levels of patients with PCOS. Subjects and methods: A total of 216 women were included in this cross-sectional study. The study group consisted of 109 patients with PCOS, and the control group consisted of 107 healthy subjects. The RRI of all subjects was measured using renal Doppler ultrasonography. Results: The patients with PCOS had higher RRI levels in comparison to the healthy subjects (0.64 ± 0.06 vs. 0.57 ± 0.06, p < 0.001). The RRI levels of the patients with PCOS were correlated with systolic blood pressure (p = 0.004, r = 0.268) and with homeostasis model assessment of insulin resistance (HOMA-IR) (p = 0.02, r = 0.216). Conclusion: In this study, we observed higher RRI levels in patients with PCOS. High RRI levels may be an indicator of cardiovascular and/or cardiovascular-associated diseases in patients with PCOS. Arch Endocrinol Metab. 2019;63(3):288-92 Keywords Polycystic ovary syndrome; renal resistive index; cardiovascular disease; systolic blood pressure; insulin resistance

INTRODUCTION

olycystic ovary syndrome (PCOS) is a common endocrine disorder in women of reproductive age, and it is characterized by menstrual irregularities, signs of androgen excess, obesity, and hirsutism, as well as metabolic abnormalities, including insulin resistance, hypertension, and dyslipidemia (1,2). The diagnosis of this syndrome can be defined according to clinical, biochemical, and ultrasonography criteria (2). Renal resistive index (RRI) is an ultrasound-based measurement that is potentially useful for the evaluation of acute and chronic kidney diseases (3,4). It is calculated as the following: peak systolic velocity/peak end diastolic velocity/peak systolic velocity (5). Additionally, RRI can provide prognostic data relating to systemic vasculature, and increased RRI is associated with hypertension, diabetes, and cardiovascular diseases (6,7). The aim of this study was to investigate the RRI levels of patients with PCOS.

SUBJECTS AND METHODS This cross-sectional cohort study was carried out in the internal medicine outpatient clinics of a tertiary hospital in Turkey. The study was performed according to the terms of the ethical standards of the 288

committee responsible for human experimentation and in accordance with the Declaration of Helsinki. Furthermore, the Institutional Review Board approved the study (04.11.2016/27). Informed consent was obtained from all subjects prior to their enrollment in this study. A total of 257 female subjects, with a minimum age of 18 years, were enrolled in this study and divided into two groups. Thirty-four subjects were excluded due to a lack of radiological or biochemical tests, and seven subjects dropped out without stating a reason. As a result, the study group was composed of 109 patients with PCOS, which was diagnosed in accordance with the Rotterdam diagnostic criteria (8), and the control group consisted of 107 healthy subjects. The subjects’ medical histories and conditions were carefully ascertained. Subjects with a history or prior diagnosis of chronic disease, metabolic disease, malignancy, clinical or biochemical features of Cushing syndrome, hyperprolactinemia, ovarian or adrenal virilizing tumors, or nonclassical congenital adrenal hyperplasia were excluded from this study. Breastfeeding, pregnant, and peri- or post-menopausal women were also excluded from this study. Data collection for the purposes of this study included the following tests: demographic data, blood chemistry, blood pressure, and ultrasonography. Arch Endocrinol Metab. 2019;63/3

Renal resistive index and polycystic ovary syndrome

Arch Endocrinol Metab. 2019;63/3

An experienced radiologist measured the patients’ RRI levels using GE’s Logic 9 high-resolution color doppler ultrasonography (GE medical systems, Milwaukee, USA) with a 5-MHz transducer. Doppler signals were obtained from the interlobar arteries. The patients’ RRI levels were calculated as peak systolic velocity/peak end diastolic velocity/peak systolic velocity. Statistical analyses were performed using MedCalc Statistical Software version 16.8 (MedCalc, Belgium). The variables were investigated using both visual (histograms and probability plots) and analytical methods (the Kolmogorov-Smirnov test) to determine whether or not they were distributed normally. Categorical measurements were reported as numbers and percentages. Quantitative measurements were reported as the mean ± the standard deviation. The chi-square test was used to compare categorical measurements. The t-test or Mann-Whitney U tests were used in the comparison of quantitative measurements between the two groups. The correlation coefficient was used to analyze the degree of association between two variables [Pearson correlation coefficient (r) with p-value and 95% CI for r.]. A log transformation was used for variables that were not normally distributed. Multiple linear regression test (backward method) was used to analyse the relationship between a dependent variable (RRI) and one or more independent variables (age, systolic blood pressure, diastolic blood pressure, fasting glucose, insulin and HOMA). The probability of making a type I error (alpha, significance) is 0.05 in all tests.

RESULTS The groups were comparable according to age. The mean age of the patients with PCOS was 29.7 ± 3.4 while it was 29.6 ± 3.5 years in the healthy group (p = 0.790). Compared to the controls, the women with PCOS had a higher mean BMI (32.5 ± 7.0 vs. 29.3 ± 4.3 p < 0.001). The groups’ demographic and laboratory data are shown in Table 1. There were statistically significant differences in fasting glucose, insulin, HOMA-IR, triglyceride and LDL levels between the groups, with higher levels in patients with PCOS (p < 0.001, for each, respectively, Table 1). Moreover, systolic (114.2 ± 10.8 vs. 110.8 ± 11.0), and diastolic (71.1 ± 9.3 vs. 68.3 ± 10.7) blood pressures were high in patients with PCOS. The differences were statistically significant (p = 0.026, 289

All women were dressed in lightweight clothes during the measurement of their weight. The women’s height was measured in centimeters, and their weight was measured in kilograms. Body mass index (BMI) was calculated as the ratio of weight/height2 (kg/m2). The patients’ blood pressure was measured by trained nurses using periodically calibrated sphygmomanometers (Erka, Germany). Two separate measurements were performed for each subject. A venous blood sample was collected in the morning following an overnight fast on the third day of the follicular phase of each subject’s menstrual cycle. The women’s fasting glucose, triglyceride, high-density lipoprotein (HDL), and low-density lipoprotein (LDL) levels were analyzed using a Roche C-501 (Tokyo, Japan), the hexokinase method (glucose), and a homogeneous colorimetric enzyme test (triglyceride, HDL, and LDL). Serum creatinine levels were analyzed on the Beckman Coulter Synchron LX-20 (Massachusetts, USA), using commercially available kits. Insulin levels were measured using the Abbott ARCHITECT i2000 SR analyzer system (Illinois, USA). Insulin resistance (IR) was measured using homeostasis model assessment [HOMA-IR = fasting glucose (mg/dL) x fasting insulin (μIU/mL)/405]. FSH and LH were measured using the ADVIA Centaur immunoassay system. Free testosterone levels were measured by radioimmunoassay (SRL Inc., Tokyo), and the reference range was between 1.1-4.7 pg/mL. PCOS was diagnosed in cases in which two or more of the following three criteria were present: oligo- or anovulation, hyperandrogenism, and polycystic ovaries. Oligomenorrhea was defined as fewer than 10 menstrual cycles per year. Biochemical hyperandrogenism was determined on the basis of the patient’s serum concentration of free testosterone. The criteria for the diagnosis of polycystic ovaries required the visualization of ≥ 12 follicles per ovary, which were 2 to 9 mm in diameter, or an ovarian volume > 10 cm3, according to transvaginal ultrasonography (8). A transvaginal ultrasound examination was performed between days six and eight of the patient’s menstrual cycle in order to assess ovarian morphology for the diagnosis of PCOS. Ultrasound examinations were performed using a 7-MHz transducer (General Electric Logic 400, Milwaukee, WI, USA). RRIs of both kidneys were measured while the subject was in a supine position. All patients were examined in the morning following an overnight fast.

Renal resistive index and polycystic ovary syndrome

r = 0.216, Figure 2). Multiple regression analysis (backward method) was performed with RRI as a dependent variable and with systolic blood pressure, diastolic blood pressure, glucose, free testosteron and HOMA-IR as independent variables in both groups. A significant correlation persisted between RRI and systolic blood pressure (p < 0.004), and between RRI and HOMA-IR (p = 0.001) in patients with PCOS. On the other hand, no significant correlation persisted between RRI and systolic blood pressure (p = 0.489), and between RRI and HOMA-IR (p = 0.113) in healthy subjects. 0.9

0.8 Renal resistive index

p = 0.039, respectively). Creatinine levels were comparable in both groups (p = 0.303). Women with PCOS had signiﬁcantly lower mean levels of FSH (5.0 ± 1.9 vs. 6.7 ± 2.6, p < 0.001). Conversely, they had signiﬁcantly higher mean levels of LH (12.3 ± 8.6 vs. 6.8 ± 2.1, p < 0.001). Therefore, the LH/FSH levels of patients with PCOS were higher than those of the healthy subjects (2.4 ± 1.1 vs. 1.1 ± 0.5, p < 0.001). Compared to women in the control group, the women with PCOS had higher mean serum concentrations of free testosterone (5.4 ± 2.1 vs. 4.0 ± 1.7, p < 0.001). The frequency of polycystic over was found to be 88.1% (n = 96) in the study group, while it was 9.3% (n = 10) in the control group (Table 2). The mean RRI was 0.64 ± 0.06 in patients with PCOS, and 0.57 ± 0.06 in healthy subjects. The patients with PCOS had significantly higher RRI levels than those in the healthy group (p < 0.001, Table 2). The RRI levels of patients with PCOS were correlated with systolic blood pressure (p = 0.0037, r = 0.276, Figure 1) and with HOMA-IR (p = 0.02, Table 1. Clinical and demographical data of the groups

Age

29.7 ± 3.4

29.6 ± 3.5

0.790

BMI (kg/m2)

32.5 ± 7.0

29.3 ± 4.3

< 0.001

Glucose (mg/dL)

91.7 ± 12.8

80.5 ± 10.5

< 0.001

4.8 ±4.4

1.6 ± 1.2

< 0.001

66 (60.5%)

9 (8.4%)

< 0.001

Systolic blood pressure (mmHg)

114.2 ± 10.8

110.8 ± 11.0

0.026

Diastolic blood pressure (mmHg)

71.1 ± 9.3

68.3 ± 10.7

0.039

Triglyceride (mg/dL)

117.1 ± 58.6

74.4 ± 23.9

< 0.001

LDL (mg/dL)

105.5 ± 27.2

68.2 ± 26.3

< 0.001

Creatinine

0.60 ± 0.13

0.58 ± 0.16

0.303

PCOS: polycystic ovary syndrome; BMI: body mass index; HOMA-IR: homeostasis model assessment of insulin resistance; LDL: low-density lipoprotein.

Table 2. Comparison of the groups according to the hormonal and radiological data PCOS (N = 109)

Healthy (107)

FSH (mU/mL)

5.0 ± 1.9

6.7 ± 2.6

< 0.001

LH (mU/mL)

12.3 ± 8.6

6.8 ± 2.1

< 0.001

LH/FSH

2.4 ± 1.1

1.1 ± 0.5

< 0.001

Free testosterone (pg/mL)

5.4 ± 2.1

4.0 ± 1.7

< 0.001

Frequency of polycystic over N (%)

96 (88.1%)

10 (9.3%)

< 0.001

RRI

0.64 ± 0.06

0.57 ± 0.06

< 0.001

FSH: follicle stimulating hormone; LH: luteinizing hormone; RRI: renal resistive index.

290

100

110

120

130

140

Systolic blood pressure

Figure 1. In a scatter diagram, the relation between renal resistive index and systolic blood pressure is presented graphically. One variable (the variable systolic blood pressure) defines the horizontal axis and the other (variable renal resistive index) defines the vertical axis. The values of the two variables on the same row in the data spreadsheet, give the points in the diagram. 0.9

0.8 Renal resistive index

Healthy (107)

Insulin resistance N (%)

0.6

0.5

PCOS (N = 109)

HOMA-IR

0.7

0.6

0.5 0

HOMA-IR

Figure 2. In a scatter diagram, the relation between renal resistive index and HOMA-IR is presented graphically. One variable (the variable HOMA-IR) defines the horizontal axis and the other (variable renal resistive index) defines the vertical axis. The values of the two variables on the same row in the data spreadsheet, give the points in the diagram. HOMA-IR: homeostasis model assessment of insulin resistance. Arch Endocrinol Metab. 2019;63/3

DISCUSSION In this study, we investigated the RRI in patients with PCOS. We found higher levels of RRI in these patients when compared to healthy subjects. Moreover, we found a correlation between RRI and systolic blood pressure, and HOMA-IR in patients with PCOS. The resistive index measured in the intrarenal segmental arteries is a well-known marker of renal vascular and interstitial damage, corresponding to an increased total cardiovascular risk (9). Tedesco and cols. studied 566 patients with hypertension and reported that patients with higher RRI showed an increased left ventricular mass index and carotid intima-media thickness with a higher prevalence of left ventricular hypertrophy, carotid plaques and microalbuminuria. They concluded that RRI could predict the presence of early CV damage (10). Additionally, KesĂ&#x2021;kek and cols. reported that, as an independent risk factor for cardiovascular diseases, high level of homocysteine is also associated with high level of RRI in patients with hypertension (11). Furthermore, in the study by Liu and cols., 387 patients with DM were analyzed. They reported that higher level of RRI was found to be associated with microvascular complications in patients with diabetes (12). On the other hand, PCOS is one of the most common endocrinopathies in women with considerable negative effects on cardiovascular functions (13). These patients are generally insulin resistant, overweight and have metabolic syndrome, with arterial hypertension, dyslipidemia, impaired glucose tolerance and diabetes. Due to all these cardiovascular risk factors, patients with PCOS are in the high risk group for development of cardiovascular diseases. Moreover, women with PCOS and insulin resistance, high systolic blood pressure, overt vascular disease or kidney disease are in the very highrisk group (14,15). Therefore, high levels of RRI may be an indicator of cardiovascular and/or cardiovascular associated diseases in patients with PCOS. In the current study, we found higher levels of fasting glucose, insulin and HOMA-IR in patients with PCOS. HOMA-IR is the most widely used as a surrogate measure of IR in large population studies and in the PCOS. Insulin resistance in PCOS contributes to both reproductive and metabolic disturbances (16). Therefore, it is clinically important to identify the prevalence and degree of IR in PCOS population. A defect in insulin receptor binding, phosphorylation or post-receptor insulin signaling transduction between the receptor kinase and glucose transport may lead to a decrease Arch Endocrinol Metab. 2019;63/3

in insulin sensitivity in patients with PCOS and cause high blood glucose, insulin and HOMA-IR (17,18). Alebie& and cols. compared 250 PCOS patients with 500 controls. They reported higher levels of HOMA-IR in patients with PCOS (19). In another study, Mu and cols.. found high levels of insulin resistance with adipose insulin resistance index in patients with PCOS (20). High blood glucose, insulin, and HOMA-IR are also associated with high levels of RRI. Bruno and cols. compared 32 newly diagnosed type 2 diabetes patients with 27 healthy subjects. They reported higher RRI levels in patients with type 2 diabetes (21). Additionally, Trovato and cols. investigated 221 patients and concluded that, independent of nutrition, increased RRI is associated with higher degrees of insulin resistance (22). In the present study, we found higher levels of systolic and diastolic blood pressures, triglyceride, LDL and BMI in patients with PCOS. Risk factors for cardiovascular system such as hypertension, diabetes, dyslipidemia and obesity are frequently reported in patients with PCOS (23). The main causes of high blood pressure in patients with PCOS may be insulin resistance, hyperandrogenism, greater sympathetic nerve activity and increased body weight (24). Shi and cols. compared 3396 PCOS patients with 1891 controls. They found higher prevalence of hypertension in patients with PCOS (19.2% vs. 11.9%) (25). Excess androgen levels, insulin resistance, variable amounts of estrogen exposure, and many environmental factors can cause to high triglyceride and LDL levels in patients with PCOS (26). Hong and cols. studied 507 Chinese patients with PCOS. They demonstrated significantly higher concentrations of triglyceride and LDL in patients with PCOS and insulin resistance (27). Serum LH concentrations and the LH to FSH ratio are frequently elevated in women with PCOS (28). FSH levels are normal to slightly suppressed and do not increase to threshold levels required during the early follicular phase of the menstrual cycle to stimulate normal follicular maturation (29). In accordance with this link, when compared with the controls, serum FSH levels were low and LH levels were high in patients with PCOS in the present study. This study has some limitations. First, it would have been beneficial if the clinical hyperandrogenism findings had been shown in addition to the biochemical hyperandrogenism. Second, for better understand the role of the RRI in PCOS patients, it would have been enlightening evaluate two more subgroups: women with 291

Renal resistive index and polycystic ovary syndrome

hypertension without PCOS and women with insulin resistance without PCOS. This could not be done due to the wide exclusion criteria of the study. Further, the crosssectional design of this study may be another limitation. In conclusion, in the present study we found higher levels of RRI in patients with PCOS. High levels of RRI may be caused by cardiovascular risk factors such as high systolic blood pressure and insulin resistance. Furthermore, RRI can be a non invasive test for increased cardiovascular disease risks in patients with PCOS. Ethical statements: all procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent: informed consent was obtained from all individual participants included in the study. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Ehrmann DA, Liljenquist DR, Kasza K, Azziz R, Legro RS, Ghazzi MN, et al. Prevalence and predictors of the metabolic syndrome in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2006;91(1):48-53. 2. Lujan ME, Chizen DR, Pierson RA. Diagnostic criteria for polycystic ovary syndrome: pitfalls and controversies. J Obstet Gynaecol Can. 2008;30(8):671-9. 3. Sugiura T, Wada A. Resistive index predicts renal prognosis in chronic kidney disease: results of a 4-year follow-up. Clin Exp Nephrol. 2011;15(1):114-20. 4. Izumi M, Sugiura T, Nakamura H, Nagatoya K, Imai E, Hori M. Differential diagnosis of prerenal azotemia from acute tubular necrosis and prediction of recovery by Doppler ultrasound. Am J Kidney Dis. 2000;35(4):713-9. 5. Granata A, Zanoli L, Clementi S, Fatuzzo P, Di Nicolò P, Fiorini F. Resistive intrarenal index: myth or reality? Br J Radiol. 2014;87(1038):20140004. 6. Tedesco MA, Natale F, Mocerino R, Tassinario G, Calabro R. Renal resistive index and cardiovascular organ damage in a large population of hypertensive patients. J Hum Hypertens. 2007;21(4):291-6. 7. Di Nicolò P, Granata A. Renal Resistive Index: not only kidney. Clin Exp Nephrol. 2017;21(3):359-66.

8. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and longterm health risks related to polycystic ovary syndrome. Fertil Steril. 2004;81(1):19-25.

12. Liu KH, Chu WC, Kong AP, Ko GT, Ma RC, Chan JW, et al. Intrarenal arterial resistance is associated with microvascular complications in Chinese type 2 diabetic patients. Nephrol Dial Transplant. 2013;28:651-8. 13. Conway G, Dewailly D, Diamanti-Kandarakis E, Escobar-Morreale HF, Franks S, Gambineri A, et al. The polycystic ovary syndrome: A position statement from the European society of endocrinology. Eur J Endocrinol. 2014;171(4):P1-29. 14. Bajuk Studen K, Pfeifer M. Cardiometabolic risk in polycystic ovary syndrome. Endocr Connect. 2018;7(7):R238-51. 15. Wild RA, Carmina E, Diamanti-Kandarakis E, Dokras A, EscobarMorreale HF, Futterweit W, et al. Assessment of cardiovascular risk and prevention of cardiovascular disease in women with the polycystic ovary syndrome: a consensus statement by the Androgen Excess and Polycystic Ovary Syndrome (AE-PCOS) Society. J Clin Endocrinol Metab. 2010;95(5):2038-49. 16. Diamanti-Kandarakis E, Dunaif A. Insulin resistance and the polycystic ovary syndrome revisited: an update on mechanisms and implications. Endocr Rev. 2012;33(6):981-1030. 17. Seow KM, Juan CC, Hsu YP, Hwang JL, Huang LW, Ho LT. Amelioration of insulin resistance in women with PCOS via reduced insulin receptor substrate-1 Ser312 phosphorylation following laparoscopic ovarian electrocautery Hum Reprod. 2007;22(4):1003-10. 18. Ciaraldi TP, el-Roeiy A, Madar Z, Reichart D, Olefsky JM, Yen SS. Cellular mechanisms of insulin resistance in polycystic ovarian syndrome. J Clin Endocrinol Metab. 1992;75(2):577-83. 19. Alebie6 MŠ, Bulum T, Stojanovie& N, Duvnjak L. Definition of insulin resistance using the homeostasis model assessment (HOMA-IR) in IVF patients diagnosed with polycystic ovary syndrome (PCOS) according to the Rotterdam criteria. Endocrine. 2014;47(2):625-30. 20. Mu L, Li R, Lai Y, Zhao Y, Qiao J. Adipose insulin resistance is associated with cardiovascular risk factors in polycystic ovary syndrome. J Endocrinol Invest. 2018 Sep 12. doi: 10.1007/s40618-018-0949-2. 21. Bruno RM, Daghini E, Landini L, Versari D, Salvati A, Santini E, et al. Dynamic evaluation of renal resistive index in normoalbuminuric patients with newly diagnosed hypertension or type 2 diabetes. Diabetologia. 2011;54(9):2430-9. 22. Trovato GM, Pirri C, Martines GF, Trovato F, Catalano D. Coffee, nutritional status, and renal artery resistive index. Ren Fail. 2010;32(10):1137-47. 23. Marciniak A, Nawrocka Rutkowska J, Brodowska A, Wis&niewska B, Starczewski A. Cardiovascular system diseases in patients with polycystic ovary syndrome – the role of inflammation process in this pathology and possibility of early diagnosis and prevention. Ann Agric Environ Med. 2016;23(4):537-41. 24. Kiałka M, Milewicz T, Klocek M. [Blood pressure and polycystic ovary syndrome (PCOS)]. Przegl Lek. 2015;72(6):309-12. 25. Shi Y, Cui Y, Sun X, Ma G, Ma Z, Gao Q, et al. Hypertension in women with polycystic ovary syndrome: prevalence and associated cardiovascular risk factors. Eur J Obstet Gynecol Reprod Biol. 2014;173:66-70. 26. Wild RA. Dyslipidemia in PCOS. Steroids. 2012;77(4):295-9.

9. Lubas A, Kade G, Niemczyk S. Renal resistive index as a marker of vascular damage in cardiovascular diseases. Int Urol Nephrol. 2014;46(2):395-402.

27. Hong Y, Yang D, Liu W, Zhao X, Chen X, Li L. Dyslipidemia in relation to body mass index and insulin resistance in Chinese women with polycystic ovary syndrome. J Biol Regul Homeost Agents. 2011;25(3):365-74.

10. Tedesco MA, Natale F, Mocerino R, Tassinario G, Calabrò R. Renal resistive index and cardiovascular organ damage in a large population of hypertensive patients. J Hum Hypertens. 2007;21(4):291-6.

28. Rebar R, Judd HL, Yen SS, Rakoff J, Vandenberg G, Naftolin F. Characterization of the inappropriate gonadotropin secretion in polycystic ovary syndrome. J Clin Invest. 1976;57(5):1320-9.

11. KesÇkek SÇÖ, Çinar Y, Kirim S, Saler T. High renal resistive index in hypertensive patients is also associated with serum homocysteine level. Clin Exp Nephrol. 2015;19(4):639-45.

29. Hillier SG. Current concepts of the roles of follicle stimulating hormone and luteinizing hormone in folliculogenesis. Hum Reprod. 1994;9(2):188-91.

292

Arch Endocrinol Metab. 2019;63/3

original article

In properly selected patients with differentiated thyroid cancer, antithyroglobulin antibodies decline after thyroidectomy and their sole presence should not be an indication for radioiodine ablation Luis Felipe Zavala1, María Inés Barra1*, Roberto Olmos1*, Michael Tuttle2, Hernán González3, Nicolás Droppelmann3, Lorena Mosso1, José M. Domínguez1

ABSTRACT Objective: Our objective was to evaluate the trend of antithyroglobulin antibodies (TgAb) during follow-up of patients with differentiated thyroid cancer (DTC) treated without RAI, as well as their role in the risk of recurrence. Subjects and methods: This was a prospective, descriptive study. A total of 152 consecutive patients with DTC treated in a single institution undergoing total thyroidectomy without RAI and followed for a median of 2.3 years (0.5-10.3) were divided in two groups: TgAb(-) (n = 111) and TgAb(+) (n = 41). Patients were classified according to AJCC 7th and 8th editions, as well as to their risk of recurrence and response to treatment categories. Results: Both groups, TgAb(-) and TgAb(+), were similar regarding patient and tumor characteristics. At the end of follow-up, 90 (59.2%), 57 (37.5%), 3 (2%) and 2 (1.3%) patients achieved excellent, indeterminate, biochemically incomplete and structurally incomplete response, respectively. The risk of structural recurrence was similar in both groups (TgAb[-] 0.9% vs. TgAb[+] 2.4%, p = 0.46). In the TgAb(+) group, TgAb became negative in 10 (24.4%), decreased ≥ 50% without negativization in 25 (60.9%), decreased < 50% in 4 (9.8%) and remained stable or increased in 2 (4.9%) cases. The only incomplete structural response had increasing TgAb during follow-up. Conclusions: In properly selected patients with DTC, TgAb concentration immediately after total thyroidectomy should not mandate RAI ablation, and their trend during follow-up may impact the risk of recurrence. Arch Endocrinol Metab. 2019;63(3):293-9

INTRODUCTION

ifferentiated thyroid cancer (DTC) is the most common endocrine malignancy, and its incidence has increased in the last decades due, in part, to expanded availability of neck ultrasonography (US) (1-3). Because most of these newly diagnosed tumors have low risk of recurrence, there is a need to personalize their management to minimize the adverse effects of therapy (4). The treatment of DTC includes surgery, selective use of radioactive iodine (RAI) and levothyroxine supplementation (4). RAI has proven to be effective in patients with intermediate and high risk of recurrence, while its utility has been questioned in Arch Endocrinol Metab. 2019;63/3

* These two co-authors contributed similar between them in the development of this article Correspondence to: José Miguel Domínguez Diagonal Paraguay, 362, 4th floor, Santiago, Chile jdomingu@uc.cl Received on Nov/25/2018 Accepted on Feb/17/2019 DOI: 10.20945/2359-3997000000123

low-risk patients (5). Because one of the benefits of RAI is to facilitate patient follow-up, its use has been suggested in the presence of elevated antithyroglobulin antibodies (TgAb) to enable correct interpretation of serum thyroglobulin (Tg) (6). However, RAI use is associated with short- and long-term adverse effects, which can significantly impair patients’ quality of life. For this reason, the indication of RAI must be adjusted to the patient’s individual risk of recurrence and mortality (7-9). After initial treatment, follow-up of patients with DTC includes measurement of serum Tg and imaging studies according to the initial risk of recurrence (10-12). About 20% of patients have TgAb(+), 293

Keywords Thyroid cancer; antithyroglobulin antibodies; radioiodine ablation

Departments of Endocrinology, Faculty of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile 2 Department of Endocrinology Service, Memorial Sloan-Kettering Cancer Center, New York, New York 3 Head and Neck Surgery, Faculty of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile 1

Thyroid cancer and anti thyroglobulin antibodies

which can interfere with Tg measurement, usually underestimating their concentration and thereby potentially leading to false-negative results (13-17). For this reason, most studies evaluating treatment and follow-up of patients with DTC (with or without RAI) exclude TgAb(+) cases (18). The TgAb are not direct tumor markers, and their concentration reflects the immune response to changes in the Tg antigen, secondary to benign or malignant disease (19). To estimate real changes in their concentration, TgAb usually require a minimum follow-up of 6 months, during which the trend in their concentration appears to be more accurate as a marker of disease rather than the absolute level itself (19). For instance, in patients with DTC and TgAb(+) treated with RAI, a reduction greater than 50% in TgAb concentration during the first 12 months of follow-up has been associated with a better prognosis (20,21). Matrone and cols. recently published a prospective follow-up of TgAb(+) patients with microcarcinomas treated without RAI (22). To the best of our knowledge, no other prospective study has compared the clinical outcome of patients with DTC treated without RAI based on the presence or absence of TgAb. The aim of this study was to prospectively evaluate the trend of TgAb concentration during follow-up of patients with DTC treated without RAI, as well as their impact on the risk of recurrence and response to treatment.

SUBJECTS AND METHODS

Subjects This study was approved by the Ethics Committee of Pontificia Universidad Católica de Chile. Patients 18 years old or older, consecutively treated and followed prospectively at our institution between December 2012 and December 2017, who met the following criteria were initially selected: (i) diagnosis of DTC; (ii) subjected to total thyroidectomy, with neck dissection performed only with evidence of clinically apparent nodal disease detected at clinical examination, preoperative neck US or during intraoperative inspection by the surgeon; and (iii) followed for at least 6 months with at least 2 measurements of Tg and TgAb. Nodal disease was classified in low and high volume according to 2015 ATA guidelines (4). 294

The need for RAI was defined in each patient following the recommendations of the 2015 ATA guidelines (4). Patients were classified according to ATA recurrence risk category (low, intermediate and high) as well as the 7th and 8th editions of the AJCC/ UICC staging system (I, II, III and IV) based on the preoperative neck US, intraoperative findings and final surgical pathology report (23). Patients that met any of the following criteria were excluded: (i) partial thyroidectomy; (ii) absence of or incomplete surgical pathology report; (iii) follow-up of less than 6 moths; and (iv) Tg, TgAb or US not performed during follow-up. From the 180 initially included patients, 28 were excluded due to of lack of follow-up (n = 13) or insufficient clinical and pathological information (n = 15).

Follow-up After initial surgery, all patients were treated with a levothyroxine (LT4) dose sufficient to maintain a serum thyrotropin (TSH) concentration < 1.0 UI/mL. The patients underwent clinical examination, Tg, TgAb and neck US every 6 months during the first year and thereafter at 6- to 12-month intervals at their attending physician’s discretion.

Outcomes Cervical neck US was performed by experienced radiologists. Negative neck US was defined as either normal exploration or the presence of any of the following findings: (i) thyroid bed avascular nodules ≤ 5 mm in diameter, stable during follow-up or (ii) reactive cervical lymph nodes. Disease recurrence/persistence was defined as suspicious findings in images, confirmed with cytopathologic study and Tg measurement on aspirate. Response to treatment was defined according to US, Tg and TgAb findings during follow-up. Excellent response was defined when all of the following were present: negative US, Tg < 0.2 ng/mL and negative TgAb. Incomplete biochemical response was defined in the presence of negative US and either non-stimulated Tg > 5 ng/mL plus negative TgAb, or positive TgAb with an increase in concentration. Incomplete structural response was defined as highly suspicious neck US (or any other image modality), confirmed through cytological study, regardless of Tg or TgAb serum concentration. Indeterminate response was defined either as (i) negative neck US plus Tg ≥ 0.2 Arch Endocrinol Metab. 2019;63/3

Thyroid cancer and anti thyroglobulin antibodies

≤ 5 ng/mL plus TgAb(-) or (ii) negative US plus TgAb(+) with decreasing or stable concentration during follow-up or (iii) nonspecific findings through imaging studies (24). To refine the evaluation of the impact of TgAb on prognosis, we classified the TgAb concentration decrease during follow-up as < 50% or ≥ 50%.

Assays Tg was measured using a chemiluminescent immunoassay with functional sensitivity of 0.1 ng/mL (Elecsys II, Roche Diagnostics, Rotkreutz, Switzerland). TgAb were also measured using a chemiluminescent immunoassay (Architect i1000, Abbot Laboratories, Abbott Park, IL), with a reference value of up to 4.11 IU/mL and analytical sensitivity of 1.0 IU/mL. For the purpose of this study, every value below 4.1 UI/mL was considered negative. Both Tg and TgAb were normalized with CRM 457.

Statistical analysis Continuous variables are presented either as mean and standard deviation or median with range, as appropriate. Categorical comparisons were performed using Fisher’s exact test, and continuous variables were compared using Student’s t-test. A p-value of < 0.05 was considered significant. Statistical analysis was performed using SPSS (v.15.0.0: SPSS, Inc., Chicago, IL).

RESULTS A total of 152 patients with a mean age of 40.5 (18-80) years, followed for a median of 2.3 (0.5-10.3) years were

included. Total thyroidectomy alone was performed in 124 (81.6%) patients, and total thyroidectomy with lymph node dissection in 28 (18.4%). Papillary thyroid cancer was the histological diagnosis in 150 (98.7%) and follicular thyroid cancer in 2 (1.3%), both cases minimally invasive and less than 4 cm. Of the total cohort, 97 (64%) had microcarcinoma, 54 (35.5%) were multicentric, 40 (26.3%) were bilateral and 13 (8.5%) had lymph node metastases: 8 with low-volume and 5 with high-volume disease, with highest nodal diameter of 0.5 cm and no cases of extra-nodal extension (Table 1). According to the ATA 2015 guidelines, 123 (80.9%) and 29 (19.1%) patients were classified as low and intermediate risk of recurrence, respectively. No patients were classified as high risk of recurrence. According to the AJCC 7th edition, 138 (90.8%) patients were classified as stage I, and according to the AJCC 8th edition, 100% of the cohort was classified as stage I (Table 1). A median of 4.4 (2.0-7.0) Tg and TgAb measurements were performed. Forty-one (27%) patients had TgAb(+) at diagnosis. Their clinical characteristics were similar to patients with TgAb(-) including age, proportion of microcarcinomas and cases classified either as stage I according to the AJCC 7th-8th editions or low risk of recurrence according to the ATA 2015 guidelines (Table 1). During the study, 2 (1.3%) patients of the whole cohort developed structural incomplete response, while 102 (67.1%) patients achieved excellent response at some point during follow-up. Of this last subgroup, no patients developed structural disease, while at the end of follow-up, 90 (88%) patients

Table 1. Characteristics of the 152 patients studied Antithyroglobulin Antibodies Negative (n = 111)

Positive (n = 41)

p-value

Age (years)

40.5 (18-80)

42 (18-80)

35 (18-63)

0.25

Follow-up (years)

2.3 (0.5-10.3)

2.3 (0.5-9.8)

0.87

Microcarcinoma

97 (64%)

72 (75%)

25 (61%)

0.70

AJCC VII edition I II III

138 (90.8%) 3 (2%) 11 (7.2%)

99 (89.2%) 2 (1.8%) 10 (9%)

39 (95.2%) 1 (2.4%) 1 (2.4%)

AJCC VIII edition I

152 (100%)

111 (100%)

41 (100%)

ATA (2015) Low risk Intermediate risk

123 (80.9%) 29 (19.1%)

90 (81.1%) 21 (18.9%)

33 (80.5%) 8 (19.5%)

Arch Endocrinol Metab. 2019;63/3

Total cohort (N = 152)

0.93

295

Thyroid cancer and anti thyroglobulin antibodies

sustained excellent response and 12 patients were classified as having indeterminate response (Table 2). Of the 95 patients with initial indeterminate response, 53 (55.8%) developed excellent response during follow-up at some point, while at the end of the study, 46 (48.4%) achieved excellent response, 46 (48.4%) indeterminate response, 2 (2.1%) biochemical incomplete response and 1 (1.1%) structural incomplete response (Table 2). The rate of incomplete structural response was similar between TgAb(-) and TgAb(+) patients: 0.9% vs 2.4% (p = 0.46) (Table 2). Regarding the 41 patients with TgAb(+), TgAb concentration became negative in 10 (24.4%), decreased ≥ 50% without negativization in 25 (60.9%), decreased < 50% in 4 (9.8%) and remained stable or increased in 2 (4.9%) cases (Table 3). Among the patients in whom TgAb concentration decreased, the median time to reach its nadir was 1.9 years (range 0.23.1; interquartile range 1.5-3.0). At the end of followup, 10 (24.4%) patients achieved excellent, 28 (68.3%) indeterminate, 2 (4.9%) biochemical incomplete and 1 (2.4%) structural incomplete response. In this group, the only incomplete structural response had increasing TgAb during follow-up. The three patients with incomplete biochemical response underwent active surveillance, and none developed structural disease at the end of follow-up. In

the two patients with a structural incomplete response, both had US/cytologically proven cervical nodal disease at 6 and 18 months after initial therapy. These patients underwent neck dissection and RAI and were both free of disease at the end of follow-up.

DISCUSSION This study prospectively evaluated the clinical and biochemical evolution of patients with DTC treated with total thyroidectomy without RAI, showing that most of them (59.2%) achieved excellent response to treatment at the end of follow-up, and only 1.3% developed structural incomplete response that required further therapy. Our cohort included patients with DTC similar to most previous studies that included patients treated without RAI: predominance of papillary thyroid cancer of initial low risk of recurrence, as well as a similar rate of biochemical and structural recurrence during follow-up (25,26). As in other studies, 27% of our patients had TgAb(+) at diagnosis, with a similar clinical presentation and evolution to those with TgAb(-), including a comparable risk of incomplete structural response during follow-up. Our findings are consistent with other reports that evaluated the impact of TgAb on DTC, with their presence not being associated with a

Table 2. Clinical outcomes in the whole cohort and according to the presence of anti-thyroglobulin antibodies Total cohort (n = 152)

Anti-thyroglobulin antibodies Negative (n = 111)

Positive (n = 41)

Structural incomplete response Yes No

2 (1.3%) 150 (98.7%)

1 (0.9%) 110 (99.1%)

1 (2.4%) 40 (97.6%)

Response to treatment at the end of follow-up Excellent Indeterminate Biochemical incomplete Structural Incomplete

90 (59.2%) 57 (37.5%) 3 (2%) 2 (1.3%)

80 (72.1%) 29 (26.1%) 1 (0.9%) 1 (0.9%)

10 (24.4%) 28 (68.3%) 2 (4.9%) 1 (2.4%)

p-value

0.46

< 0.01

Trend of TgAb concentration during follow-up

N (%)

Structural disease

Negativization

10 (24.4%)

0 (0%)

≥ 50% decrease without negativization

25 (60.9%)

0 (0%)

< 50% decrease

4 (9.8%)

0 (0%)

Stable or increase

2 (4.9%)

1 (50%)

TgAb: antithyroglobulin antibodies.

296

Arch Endocrinol Metab. 2019;63/3

higher risk of recurrence or mortality (27), as well as some studies suggesting that their concentration is not related to tumor mass (19,27). It has been previously reported that in patients with DTC and TgAb(+), the trend of the TgAb concentration during follow-up can be used as a surrogate marker of Tg when evaluating the response to treatment (28). Unfortunately, this information is derived from retrospective studies including patients initially treated with RAI (20). Our study exclusively included patients treated without RAI followed prospectively. In the AcTg(+) group, the trend in antibodies concentration during follow-up was favorable in 85.3% of cases, with negativization in 24.4% and a decrease of ≥ 50% without negativization in another 60.9%. In this subgroup, there were no cases of structural recurrence, and 24.4% of patients achieved excellent response at the end of follow-up (negative neck US plus TgAb[-] and Tg < 0.2 ng/mL). These findings are consistent with studies that included patients treated with RAI, in which a decrease ≥ 50% in TgAb concentration was associated with a risk of recurrence of 0 and 3.3% in lowand intermediate-risk patients, respectively (20,29). Also similar to our cohort, previous studies focusing on patients with TgAb(+) reported a ≥ 50% decrease of their concentration in 75%, with negativization in 30% of cases (28,30). The similar behavior of TgAb in patients treated with or without RAI casts doubt on the role of RAI to facilitate follow-up after initial treatment. Interestingly, at the end of follow-up, 75.6% of the whole cohort achieved either one of the following: i) excellent response to therapy (including initially TgAb[-] and TgAb[+] with negativization during follow-up) or ii) initially TgAb(+), with ≥ 50% decrease of TgAb without negativization, in addition to Tg < 0.2 ng/mL and negative US. It is well known that both aforementioned conditions are associated with a very low risk of recurrence (20). As mentioned before, 27% of patients with initial TgAb(+) achieved excellent response at the end of follow-up. However, a significant decrease in TgAb concentration, even without negativization, is also associated with an excellent prognosis. Our data, hopefully confirmed by further studies with longer follow-up, may challenge the current definition of excellent response, allowing for the inclusion of patients with TgAb(+), in which antibodies concentration decrease ≥ 50% (in conjunction with a negative US) during follow-up. These findings confirm that, in properly selected patients with low Arch Endocrinol Metab. 2019;63/3

or intermediate risk of recurrence, in whom the use of RAI is selective, the behavior of Tg and TgAb during follow-up can modulate the indication for RAI. Furthermore, previous studies have shown that in low- and intermediate-risk patients, the efficacy of RAI is preserved even if administered up to one year after surgery, giving the clinician enough time to thoroughly decide its indication (31-33). It is also notable that of the 95 patients with an indeterminate response at their first follow-up evaluation (representing 63% of the entire cohort), only 1 of them developed structural disease, which was successfully treated with surgery and RAI. This finding, consistent with previous reports, suggests that in patients with DTC of low and selected intermediate risk, a conservative approach with active surveillance is safe and avoids exposure of an important number of patients to therapies with potential risks, such as RAI and surgery (25). As previously described, most patients in our study achieved either excellent or indeterminate response. However, and despite the limited number of cases, half of the patients with stable or increasing TgAb(+) concentration during follow-up developed structural recurrence. Our finding, similar to previous reports, highlights the importance of closer surveillance in this specific group of patients (25). One limitation of this study was the limited length of follow-up (median 2.3 years), which could result in underestimation of the risk of disease recurrence. However, 75.6% of our cohort achieved either excellent response or negative US plus a significant decrease in TgAb levels, findings that in a previous study from Momesso were associated with a risk of recurrence of only 1.3% after a median follow-up of 8.3 years (25). A second limitation was the high proportion (64%) of microcarcinomas in our cohort, considering that according to the 2015 ATA guidelines, these tumors could be submitted to active surveillance with neck US, in the absence of sonographic or cytologic signs of aggressiveness. However, in our study, we began recruitment in 2013, years before the publication of the aforementioned guidelines, when the indication of fine-needle aspiration was mainly driven by clinical and sonographic risk factors, regardless of nodule size (34). Finally, there is still doubt concerning the definition of TgAb(+), with values above both the analytical sensitivity as well as the upper reference limit of the assay being considered TgAb(+). Regarding this 297

Thyroid cancer and anti thyroglobulin antibodies

point, a recent study did not find differences in clinical outcomes when comparing patients with TgAb < 1.0 IU/mL and TgAb < 4.11 IU/mL using the same assay as in our present study (35). We conclude that in properly selected patients with DTC, the concentration of TgAb usually declines during follow-up after total thyroidectomy, and their sole presence should not be an indication of RAI ablation. Funding: the authors received no specific funding for this work. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1.

La Vecchia C, Malvezzi M, Bosetti C, Garavello W, Bertuccio P, Levi F, et al. Thyroid cancer mortality and incidence: a global overview. Int J Cancer. 2015;136(9):2187-95.

2. Davies L, Morris LG, Haymart M, Chen AY, Goldenberg D, Morris J, et al.; AACE Endocrine Surgery Scientific Committee. American Association of Clinical Endocrinologists and American College of Endocrinology Disease State Clinical Review: the increasing incidence of thyroid cancer. Endocr Pract. 2015;21(6):686-96. 3. Lim H, Devesa SS, Sosa JA, Check D, Kitahara CM. Trends in Thyroid Cancer Incidence and Mortality in the United States, 1974-2013. JAMA. 2017;317(13):1338-48. 4. Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016;26(1):1-133. 5. Carhill AA, Litofsky DR, Ross DS, Jonklaas J, Cooper DS, Brierley JD, et al. Long-Term Outcomes Following Therapy in Differentiated Thyroid Carcinoma: NTCTCS Registry Analysis 1987-2012. J Clin Endocrinol Metab. 2015;100(9):3270-9. 6. Chiovato L, Latrofa F, Braverman LE, Pacini F, Capezzone M, Masserini L, et al. Disappearance of humoral thyroid autoimmunity after complete removal of thyroid antigens. Ann Intern Med. 2003;139(5 Pt 1):346-51.

Iyer NG, Morris LG, Tuttle RM, Shaha AR, Ganly I. Rising incidence of second cancers in patients with low-risk (T1N0) thyroid cancer who receive radioactive iodine therapy. Cancer. 2011;117(19): 4439-46.

11. Castagna MG, Maino F, Cipri C, Belardini V, Theodoropoulou A, Cevenini G, et al. Delayed risk stratification, to include the response to initial treatment (surgery and radioiodine ablation), has better outcome predictivity in differentiated thyroid cancer patients. Eur J Endocrinol. 2011;165(3):441-6. 12. Pitoia F, Bueno F, Urciuoli C, Abelleira E, Cross G, Tuttle RM. Outcomes of patients with differentiated thyroid cancer riskstratified according to the American thyroid association and Latin American thyroid society risk of recurrence classification systems. Thyroid. 2013;23(11):1401-7. 13. Kumar A, Shah DH, Shrihari U, Dandekar SR, Vijayan U, Sharma SM. Significance of antithyroglobulin autoantibodies in differentiated thyroid carcinoma. Thyroid. 1994;4(2):199-202. 14. Souza SL, Montalli Da Assumpção LV, Ward LS. Ward, Impact of previous thyroid autoimmune diseases on prognosis of patients with well-differentiated thyroid cancer. Thyroid. 2003;13(5):491-5. 15. Spencer C, Petrovic I, Fatemi S, LoPresti J. Serum thyroglobulin (Tg) monitoring of patients with differentiated thyroid cancer using sensitive (second-generation) immunometric assays can be disrupted by false-negative and false-positive serum thyroglobulin autoantibody misclassifications. J Clin Endocrinol Metab. 2014;99(12):4589-99. 16. Spencer C, Petrovic I, Fatemi S. Current thyroglobulin autoantibody (TgAb) assays often fail to detect interfering TgAb that can result in the reporting of falsely low/undetectable serum Tg IMA values for patients with differentiated thyroid cancer. J Clin Endocrinol Metab. 2011;96(5):1283-91. 17. Spencer CA. Clinical review: Clinical utility of thyroglobulin antibody (TgAb) measurements for patients with differentiated thyroid cancers (DTC). J Clin Endocrinol Metab. 2011;96(12): 3615-27. 18. Lamartina L, Durante C, Filetti S, Cooper DS. Low-risk differentiated thyroid cancer and radioiodine remnant ablation: a systematic review of the literature. J Clin Endocrinol Metab. 2015;100(5):1748-61. 19. Verburg FA, Luster M, Cupini C, Chiovato L, Duntas L, Elisei R, et al. Implications of thyroglobulin antibody positivity in patients with differentiated thyroid cancer: a clinical position statement. Thyroid. 2013;23(10):1211-25. 20. Rosario PW, Carvalho M, Mourão GF, Calsolari MR. Comparison of Antithyroglobulin Antibody Concentrations Before and After Ablation with 131I as a Predictor of Structural Disease in Differentiated Thyroid Carcinoma Patients with Undetectable Basal Thyroglobulin and Negative Neck Ultrasonography. Thyroid. 2016;26(4):525-31. 21. Kim WG, Yoon JH, Kim WB, Kim TY, Kim EY, Kim JM, et al. Change of serum antithyroglobulin antibody levels is useful for prediction of clinical recurrence in thyroglobulin-negative patients with differentiated thyroid carcinoma. J Clin Endocrinol Metab. 2008;93(12):4683-9.

8. Mendoza A, Shaffer B, Karakla D, Mason ME, Elkins D, Goffman TE. Quality of life with well-differentiated thyroid cancer: treatment toxicities and their reduction. Thyroid. 2004;14(2): 133-40.

22. Matrone A, Latrofa F, Torregrossa L, Piaggi P, Gambale C, Faranda A, et al. Changing Trend of Thyroglobulin Antibodies in Patients With Differentiated Thyroid Cancer Treated With Total Thyroidectomy Without 131I Ablation. Thyroid. 2018;28(7):871-9.

Florenzano P, Guarda FJ, Jaimovich R, Droppelmann N, González H, Domínguez JM. Radioactive Iodine Administration Is Associated with Persistent Related Symptoms in Patients with Differentiated Thyroid Cancer Int J Endocrinol. 2016;2016:2586512.

23. Tuttle RM, Haugen B, Perrier ND. Updated American Joint Committee on Cancer/Tumor-Node-Metastasis Staging System for Differentiated and Anaplastic Thyroid Cancer (Eighth Edition): What Changed and Why? Thyroid. 2017;27(6):751-6.

10. Tuttle RM, Tala H, Shah J, Leboeuf R, Ghossein R, Gonen M, et al. Estimating risk of recurrence in differentiated thyroid cancer after total thyroidectomy and radioactive iodine remnant ablation: using response to therapy variables to modify the initial risk estimates predicted by the new American Thyroid Association staging system. Thyroid. 2010;20(12):1341-9.

24. Momesso DP, Tuttle RM, Update on differentiated thyroid cancer staging. Endocrinol Metab Clin North Am. 2014;43(2):401-21.

298

25. Momesso DP, Vaisman F, Yang SP, Bulzico DA, Corbo R, Vaisman M, et al. Dynamic Risk Stratification in Patients with Differentiated Thyroid Cancer Treated Without Radioactive Iodine. J Clin Endocrinol Metab. 2016;101(7):2692-700. Arch Endocrinol Metab. 2019;63/3

Thyroid cancer and anti thyroglobulin antibodies

26. Spencer CA, Takeuchi M, Kazarosyan M, Wang CC, Guttler RB, Singer PA, et al. Serum thyroglobulin autoantibodies: prevalence, influence on serum thyroglobulin measurement, and prognostic significance in patients with differentiated thyroid carcinoma. J Clin Endocrinol Metab. 1998;83(4):1121-7. 27. McLeod DS, Cooper DS, Ladenson PW, Ain KB, Brierley JD, Fein HG, et al.; The National Thyroid Cancer Treatment Cooperative Study Group. Prognosis of differentiated thyroid cancer in relation to serum thyrotropin and thyroglobulin antibody status at time of diagnosis. Thyroid. 2014;24(1):35-42. 28. Spencer C, Fatemi S. Thyroglobulin antibody (TgAb) methods – Strengths, pitfalls and clinical utility for monitoring TgAb-positive patients with differentiated thyroid cancer. Best Pract Res Clin Endocrinol Metab. 2013;27(5):701-12. 29. Durante C, Tognini S, Montesano T, Orlandi F, Torlontano M, Puxeddu E, et al.; PTC Study Group. Clinical aggressiveness and long-term outcome in patients with papillary thyroid cancer and circulating anti-thyroglobulin autoantibodies. Thyroid. 2014;24(7):1139-45.

32. Suman P, Wang CH, Abadin SS, Block R, Raghavan V, Moo-Young TA, et al. Timing of radioactive iodine therapy does not impact overall survival in high-risk papillary thyroid carcinoma. Endocr Pract. 2016;22(7):822-31. 33. Scheffel RS, Zanella AB, Dora JM, Maia AL. Timing of Radioactive Iodine Administration Does Not Influence Outcomes in Patients with Differentiated Thyroid Carcinoma. Thyroid. 2016;26(11):1623-9. 34. American Thyroid Association (ATA) Guidelines Taskforce on Thyroid Nodules and Differentiated Thyroid Cancer, Cooper DS, Doherty GM, Haugen BR, Kloos RT, Lee SL, Mandel SJ, et al. Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid. 2009;19(11):1167-214. 35. Côrtes MCS, Rosario PW, Oliveira LFF, Calsolari MR. Clinical Impact of Detectable Antithyroglobulin Antibodies Below the Reference Limit (Borderline) in Patients with Papillary Thyroid Carcinoma with Undetectable Serum Thyroglobulin and Normal Neck Ultrasonography After Ablation: A Prospective Study. Thyroid. 2018;28(2):229-35.

30. Görges R, Maniecki M, Jentzen W, Sheu SN, Mann K, Bockisch A, et al. Development and clinical impact of thyroglobulin antibodies in patients with differentiated thyroid carcinoma during the first 3 years after thyroidectomy. Eur J Endocrinol. 2005;153(1):49-55.

31. Suman P, Wang CH, Moo-Young TA, Prinz RA, Winchester DJ. Timing of Adjuvant Radioactive Iodine Therapy Does Not Affect Overall Survival in Low- and Intermediate-Risk Papillary Thyroid Carcinoma. Am Surg. 2016;82(9):807-14.

Arch Endocrinol Metab. 2019;63/3

299

original article

Diagnostic performance of thyroid ultrasound in Hürthle cell carcinomas Laboratório de Endocrinologia Celular e Molecular (LIM25), Hospital das Clínicas da Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brasil 2 Radiologia, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brasil 3 Radiologia, Instituto de Radiologia (InRad), Hospital das Clínicas da Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brasil 4 Departamento de Patologia, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brasil 5 Departamento de Cirurgia de Cabeça e Pescoço, Hospital das Clínicas da Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brasil 6 Departamento de Endocrinologia, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina, Universidade de São Paulo, São Paulo, SP, Brasil 1

Nathalie Oliveira Santana1, Ricardo Miguel Costa Freitas2, Vinicius Neves Marcos2, Maria Cristina Chammas3, Rosalinda Yossie Asato Camargo1, Cláudia Kliemann Schmerling4, Felipe Augusto Brasileiro Vanderlei5, Ana Oliveira Hoff6, Suemi Marui1, Debora Lucia Seguro Danilovic1,6

ABSTRACT Objective: Hürthle cell carcinomas (HCCs) of the thyroid have been recently reclassified as a separate entity due to their distinct clinical and molecular profiles. Few studies have assessed the ability of preoperative characteristics in differentiating HCCs from Hürthle cell adenomas (HCAs) due to the low prevalence of both lesions. This study aimed to compare the preoperative features of HCCs and HCAs and evaluate the diagnostic performance of ultrasound in distinguishing between both. Subjetcs and methods: Retrospective study including 101 patients (52 HCCs and 49 HCAs) who underwent thyroid surgery from 2000 to 2016. Clinical, ultrasonographic, and histological data were reviewed. Diagnostic performance of suspicious sonographic features was analyzed in 51 cases (24 HCCs and 27 HCAs). Results: Hürthle cell neoplasms were predominant in females. Subjects ≥ 55 years represented 58% of the cases of HCCs and 53% of those of HCAs. Carcinomas were significantly larger (p < 0.001), and a tumor size ≥ 4 cm significantly increased the risk of malignancy (odds ratio 3.67). Other clinical, cytologic, and sonographic data were similar between HCCs and HCAs. Among the HCCs, the lesions were purely solid in 54.2%, hypoechoic in 37.5%, and had coarse calcifications in 12.5%, microcalcifications in 8.3%, irregular contours in 4.2%, and a taller-than-wide shape in 16.7%. Predominantly/exclusive intranodular vascularization was observed in 52.6%. Overall, 58% of the HCCs were classified asTI-RADS 4 or 5 compared with 48% of the HCAs.TI-RADS 4 or 5 had a specificity of only 51.8% and a positive likelihood ratio of 1.21. Conclusions: Apart from the lesion size, no other preoperative feature adequately distinguished HCCs from HCAs. Sonographic characteristics raising suspicion for malignancy, which are mostly present in papillary carcinomas, were infrequent in HCCs. New tools must be developed to improve preoperative diagnosis and deferral of surgery in cases of adenomas. Arch Endocrinol Metab. 2019;63(3):300-5 Keywords Ultrasound; thyroid cancer; Hürthle cell; Doppler

INTRODUCTION

hyroid carcinomas have increased in prevalence over the last decades (1). Papillary and follicular thyroid cancers are the most common histological types of thyroid cancer, while Hürthle cell carcinomas (HCCs) correspond to only 3 to 4% of all thyroid malignancies. Due to their distinct clinical behavior, HCCs have been recently reclassified as a separate entity (2,3). A thyroid tumor is classified as a Hürthle cell neoplasm (HCN) when over 75% of its tumor cells display oncocytic histologic features without nuclear characteristics of papillary carcinoma. These so-called Hürthle cells are found not only in HCN, but also 300

Correspondence to: Debora Lucia Seguro Danilovic Laboratório de Endocrinologia Celular e Molecular (LIM25), Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo, Av. Dr. Arnaldo, 455, sala 4305 01246-903 – São Paulo, SP, Brasil deboradanilovic@usp.br Received on Jan/16/2019 Accepted on Feb/27/2019 DOI: 10.20945/2359-3997000000131

in benign lesions like nodular goiter, Graves’ disease, and Hashimoto’s thyroiditis, as well as after cervical radiotherapy. Oncocytic changes were previously considered to be a consequence of cellular senescence, but are now seen as a metaplastic process in response to a variety of stimuli inducing cellular stress (4-6). Similar to follicular neoplasms, HCCs are distinguished from Hürthle cell adenomas (HCAs) by the presence on histological assessment of vascular and capsular invasion or identification of nodal and/or distant metastases (4). HCCs are classified as minimally or widely invasive, the last one presenting a poorer response to radioactive iodine therapy and higher recurrence Arch Endocrinol Metab. 2019;63/3

Ultrasound in Hürthle cell carcinomas

SUBJECTS AND METHODS Subjects This retrospective study included 101 patients with pathologically proven HCNs who underwent total thyroidectomy or lobectomy in a tertiary center from 2000 to 2016. Clinical, cytologic, and histopathologic data were collected by retrospective chart review. The histopathology of the lesions was reassessed by a pathologist. The study was approved by the Research Ethics Committee of the University of São Paulo. Data were analyzed anonymously, and any identifying marks in the ultrasound images were removed.

Methods Preoperative ultrasound images were available in 51 cases (24 HCCs and 27 HCAs). Gray-scale and color Doppler images were obtained using high-frequency transducers (7.5-12 MHz) with the following US equipment models: iU22 (Philips, Eindhoven, The Netherlands), MyLab 70XV (Esaote, Florence, Italy), Sonix (Ultrasonix, Burnaby, Canada), Aplio 500 (Toshiba Medical Systems, Tokyo, Japan), or Logic E9 (GE Healthcare, Milwaukee, Wisconsin, USA). Two blinded radiologists retrospectively analyzed the sonographic imaging characteristics and, when their diagnoses were discordant, a third radiologist reviewed the images. The following nodular features were evaluated: size, echogenicity, solid and/or cystic aspect, contours, presence of a hypoechoic peripheral halo (rim) or calcifications, vascularization pattern, and presence Arch Endocrinol Metab. 2019;63/3

of atypical lymph nodes. We compared the longest diameter of the lesions, and evaluated if the nodular height was greater than the width. The echogenicity of the nodule was assessed in relation to the normal thyroid parenchyma. The margins of the nodules were classified as regular or irregular. Microcalcifications were defined as hyperechoic foci up to 2 mm in size, while coarse calcifications were defined as hyperechoic foci > 2 mm in size with posterior acoustic shadowing. The type of vascularization was classified as peripheral, if the vascularization was predominantly perinodular, or central, if it was predominantly intranodular. We analyzed the diagnostic performance of the following features of malignancy suspicion in HCCs: hypoechogenicity, solid aspect, irregular contours, absence of a hypoechoic peripheral halo, presence of microcalcifications or coarse calcifications, taller-thanwide shape, and intranodular vascularization pattern. The nodules were classified according to the Thyroid Imaging, Reporting and Data System (TI-RADS) (22). The diagnostic performance of HCCs in the TI-RADS 4 and 5 categories, which correspond to moderately and highly suspicious nodules, respectively, was also evaluated.

Statistical analysis The data were processed using IBM SPSS Statistics for Windows, Version 24.0 (IBM Corp., Armonk, New York, USA). Two-tailed p values were used, and values < 0.05 were considered statistically significant. Categorical variables are presented as absolute and relative (percentages) frequencies. Differences were evaluated by Pearson’s chi-square test and Fisher’s exact test, when appropriate. Continuous variables are presented as mean ± standard deviation values. Differences among study subgroups were determined using Student’s t test, for normal distributions, or Mann-Whitney U test for non-normal distributions. The diagnostic performance (sensitivity, specificity, positive likelihood ratio [LR], negative LR) of the thyroid ultrasound exams was analyzed. We defined the true positive, true negative, false positive, and false negative results based on the final histological diagnosis (carcinoma or adenoma).

RESULTS AND DISCUSSION Data from 52 HCCs and 49 HCAs were available for review. No significant clinical differences were 301

and mortality rates (7-11). Fine-needle aspiration biopsy (FNA) of HCNs are usually indeterminate for malignancy, challenging the preoperative diagnosis of HCCs. Even current molecular preoperative tests are unable to distinguish HCNs and HCCs correctly (12). Therefore, surgery is still required to distinguish benign and malignant HCNs (13-15). Only a few studies have reported the imaging findings of Hürthle cell tumors. Besides, due to their rarity, HCCs have been poorly represented in these studies, and their sonographic features have been analyzed in combination with those of adenomas (16-21). The aims of this study were to compare the clinical presentation of HCCs and HCAs and evaluate the impact of thyroid ultrasound in distinguishing one lesion from another.

Ultrasound in Hürthle cell carcinomas

observed between benign and malignant cases. HCNs predominated in female subjects (83% HCCs vs. 84% HCAs). The mean age of the patients was 57.2 ± 14.5 years in malignant and 55.2 ± 13.3 years in benign cases. Subjects aged ≥ 55 years comprised 58% of the patients with HCC and 53% of those with HCA. The mean preoperative TSH level was 1.9 ± 1.5 µU/mL in patients with HCC and 2.1 ± 1.2 µU/mL in those with HCA. On pathological analysis, the diameter of the carcinomas was significantly larger (mean 47.4 ± 25.7 mm, median 46.5 mm) than that of the adenomas (mean 28.6 ± 19.5 mm, median 25 mm; p < 0.001). Nodules ≥ 4 cm had an increased risk of malignancy (odds ratio 3.67, 95% confidence interval 1.58 – 8.52) (Figure 1). Multinodular goiter coexisted in 51% of the carcinomas and 40.8% of the adenomas, while lymphocytic thyroiditis was found in 15.7% and 14.3% of the carcinomas and adenomas, respectively. Coexisting papillary thyroid carcinoma was diagnosed in 16% of the subjects with HCC and 20.4% of those with HCA. One patient with HCC had a concomitant poorly differentiated thyroid carcinoma. All HCCs presented as unifocal lesions, and 52.6% of the HCCs were considered widely invasive. Vascular invasion was present in 76% and extrathyroidal extension in 6.5% of the HCCs. Lymph node metastases were diagnosed in 5 cases (10%), while 6 patients (12%) had distant metastases. The mean largest dimensions of the carcinomas and adenomas on the ultrasound images were 50.1 ±

28.3 vs. 37.9 ± 22.5 mm (p = 0.05), respectively. The ultrasound features of HCCs and HCAs are described in Table 1; no significant differences were observed between the groups. A wide spectrum of sonographic patterns was observed in carcinomas (Figure 2). A cytological diagnosis of follicular neoplasm/HCN (Bethesda IV) was observed in 87.2% of HCC and 96.7% of HCA cases (Table 2). The diagnostic performance of ultrasound in HCC is presented in Table 3. Hypoechogenicity and presence of coarse calcifications had the highest positive LR. However, the specificity of hypoechogenicity was only 77.8%. Also, despite having high specificity, the presence of coarse calcifications was observed in only 12.5% of the HCCs. We also evaluated the diagnostic performance of the TI-RADS classification in HCCs. Even though 58% of the HCCs were classified as TI-RADS 4 or 5, no significant difference was observed in this regard when HCCs were compared with HCAs, as 48% of the latter cases were also moderately or highly suspicious. Therefore, TI-RADS 4 or 5 had a specificity of only 51.8% and a positive LR of 1.21. Apart from tumor size, no significant difference occurred between benign and malignant cases that could allow an accurate preoperative diagnosis. In this series, Hürthle cell tumors larger than 4 cm had a significantly increased malignancy risk, which reinforces data from a recent report of 330 pathologically diagnosed oxyphilic cell neoplasms (61 carcinomas), in which a tumor size > 4 cm was an independent predictor of HCC (23).

Table 1. Ultrasound characteristics of Hürthle cell carcinomas and adenomas 120

p < 0.001

Tumor diameter (mm)

100 80 60 40

Carcinomas

Adenomas

Figure 1. Comparison of the largest diameters of Hürthle cell carcinomas and adenomas. The solid lines represent the median diameter of the carcinomas and adenomas. 302

Carcinoma (n = 24)

Adenoma (n = 27)

Hypoechogenicity (%)

9 (37.5)

6 (22.2)

Purely solid (%)

Ultrasound characteristic

13 (54.2)

12 (44.4)

Irregular contours (%)

1 (4.2)

3 (11.1)

Absence of hypoechoic halo (%)

5 (20.8)

11 (40.7)

Microcalcification (%)

2 (8.3)

3 (11.1)

Coarse calcification (%)

3 (12.5)

2 (7.4)

Taller-than-wide shape (%)

4 (16.7)

8 (29.6)

Central or predominantly central blood flow (%)* TI-RADS 2 3 4 5

10 (52.6) 5 (20.8) 5 (20.8) 12 (50) 2 (8.3)

12 (60) 4 (14.8) 10 (37) 10 (37) 3 (11.1)

* Percentage from available exams. Arch Endocrinol Metab. 2019;63/3

Ultrasound in Hürthle cell carcinomas

Figure 2. Variety of sonographic presentations of Hürthle cell carcinomas. (A) Partially cystic isoechoic nodule; (B) solid isoechoic nodule with peripheral halo; (C) markedly solid hypoechoic nodule; (D) solid hypoechoic nodule with coarse calcifications; (E) solid hypoechoic nodule with microcalcifications and coarse calcifications; (F) solid isoechoic nodule with peripheral vascularization; (G) solid hypoechoic nodule with intranodular vascularization. Table 2. Thyroid cytopathology of Hürthle cell carcinomas and adenomas Cytopathology Benign

Carcinomas (n = 38)

Adenomas (n = 30)

p value

4 (10.5%)

1 (3.3%)

0.156

Follicular neoplasm or suspicious for follicular neoplasm

9 (23.7%)

12 (40%)

0.231

Follicular neoplasm or suspicious for follicular neoplasm, Hürthle cell

24 (63.1%)

17 (56.6%)

0.587

1 (2.7%)

Nondiagnostic or unsatisfactory

Sensitivity (95% CI)

Specificity (95% CI)

Positive LR (95% CI)

Negative LR (95% CI)

Accuracy (95% CI)

Hypoechogenicity

37.5% (18.80-59.41)

77.8% (57.74-91.38)

1.69 (0.70-4.05)

0.8 (0.56-1.16)

58.8% (44.17-72.42)

Purely solid

54.2% (32.82-74.45)

55.6% (35.33-74.52)

1.22 (0.70-2.13)

0.83 (0.48-1.43)

54.9% (40.34-68.87)

Irregular contours

4.2% (0.11-21.12)

88.9% (70.84-97.65)

0.37 (0.04-3.37)

1.08 (0.92-1.26)

49% (34.75-63.40)

Absence of peripheral halo

20.8% (7.13-42.15)

59.3% (38.80-77.61)

0.51 (0.21-1.26)

1.34 (0.92-1.94)

41.2% (27.58-55.83)

Microcalcification

8.3% (1.03-27.00)

88.9% (70.84-97.65)

0.75 (0.14-4.12)

1.03 (0.86-1.23)

51% (36.60-65.25)

Coarse calcification

12.5% (2.66-32.36)

92.6% (75.71-99.09)

1.69 (0.31-9.26)

0.94 (0.79-1.14)

54.9% (40.34-68.87)

Taller-than-wide shape

16.7% (4.74-37.38)

70.4% (49.82-86.25)

0.56 (0.19-1.63)

1.18 (0.87-1.60)

45.1% (31.13- 59.66)

Central or predominantly central blood flow

50% (27.20-72.80)

36.8% (16.29- 61.64)

0.79 (0.45-1.38)

1.36 (0.65-2.83)

43.6% (27.81-60.38)

TI-RADS 4 or 5

58.3% (36.64-77.89)

51.9% (31.95-71.33)

1.21 (0.72-2.03)

0.8 (0.44-1.46)

54.9% (40.34-68.87)

Ultrasound characteristic

Table 3. Performance of sonographic features raising suspicion for malignancy in the diagnosis of Hürthle cell carcinomas

LR: likelihood ratio; 95 % CI: 95% confidence interval. Arch Endocrinol Metab. 2019;63/3

303

Ultrasound in Hürthle cell carcinomas

Other studies have also demonstrated a correlation between tumor size and malignancy risk in HCNs. A tumor size ≥ 2.5 cm was an independent predictor of HCC (24), whereas all tumors ≤ 2 cm observed were benign and all of those larger than 6 cm were malignant (25). As previously demonstrated, the cytologic analysis of our tumors failed to identify malignant nodules correctly, as 87% of the HCCs and 97% of the HCAs were diagnosed as follicular neoplasm/HCN (Bethesda IV). The positive predictive value for malignancy of cytologic HCN diagnoses has been previously described to be between 15 and 45% (13,15), decreasing to 9.5% in the presence of Hashimoto’s thyroiditis (14). Therefore, other preoperative aspects should be evaluated to allow a proper diagnosis of malignancy before surgery. Several sonographic features are usually associated with a higher risk of malignancy in a thyroid nodule. According to the American Thyroid Association, a highly suspicious sonographic pattern is represented by a solid hypoechoic nodule or solid hypoechoic component of a partially cystic nodule associated with at least one of the following findings: irregular margins, microcalciﬁcations, taller-than-wide shape, rim calciﬁcations with a small protruding soft tissue component or evidence of extrathyroidal extension. The absence of a peripheral halo and increased intranodular vascularization are also considered suspicious for malignancy (1). These suspicious characteristics are mainly observed in papillary thyroid carcinomas (1,26). In this retrospective analysis, B-mode and color Doppler sonographic features were unable to differentiate malignant from benign HCNs. HCNs may present a wide variety of sonographic findings (1620,24,27), and hypoechogenicity predicted malignancy in this background (28). Ito and cols. (23) observed that round and hypoechoic solid nodules and solid nodules with irregular borders or with psammoma calcifications represented 83% of the carcinomas and 78% of the adenomas. However, the authors concluded that such sonographic presentations were independent predictors of malignancy in patients with Hürthle cells in FNA cytology. In contrast, lesions with a flat, isoechoic, or hyperechoic pattern with a peripheral halo were frequent among the HCCs in the present study at rates of 83.3%, 62.5%, and 79.2%, respectively. The sonographic features of follicular thyroid carcinomas are different from those of papillary 304

carcinomas. The most frequent presentation of follicular carcinomas (with respective presentation rates for each feature) is as a solid (82.6%), flat (72.7%), isoechoic (65.2%) nodule with a hypoechoic rim (86.6%) and without calcifications (82.6%) (26). The peripheral halo has been suggested to represent the tumor’s capsule or pseudocapsule of surrounding fibrous connective tissue, which could be present in both benign and malignant HCNs (19). Unlike previous data on follicular carcinomas (26), cystic changes (45.8%) and calcifications, mainly coarse ones (12.5%), were observed in our HCCs. Lee and cols. (19) reported cystic changes in around 50% and coarse calcifications in 20% of HCNs, without significant differences between benign and malignant nodules, although the authors analyzed a smaller sample of only 3 HCC. Regarding vascularization, our study showed predominantly intranodular vascularization in both HCCs (52.6%) and HCAs (60%). We evaluated the performance of the TIRADS classification in diagnosing HCCs (22). TI-RADS 4 (moderately suspicious) or TI-RADS 5 (highly suspicious) nodules corresponded to 58.3% of our HCCs. However, 48% of the HCAs were also categorized as moderately or highly suspicious according to this classification. Likewise, the Korean TIRADS was also unable to distinguish HCCs from HCAs (29). This study has some limitations. The retrospective analysis was restricted to pathologically proven HCNs, instead of considering all nodules suspected of HCNs on cytology. Although this series included a higher number of HCCs than other reports (due to the rarity of these tumors), the number of cases could be considered small to identify statistically significant differences. In conclusion, apart from tumor size, no other preoperative feature was able to differentiate HCCs from HCAs adequately. Sonographic characteristics considered suspicious for malignancy, which are mostly seen in papillary thyroid carcinomas, were infrequent in HCCs. Besides, the combination of suspicious sonographic characteristics with the TI-RADS classification was also insufficient to diagnose the cases of HCC accurately. Therefore, new tools must be developed, such as molecular tests or other imaging tools like elastography or contrast-enhanced ultrasound, to allow an accurate preoperative diagnosis of malignancy and possibly defer diagnostic surgery in cases of adenomas. Arch Endocrinol Metab. 2019;63/3

Ultrasound in Hürthle cell carcinomas

Acknowledgments: we thank our patients and all the colleagues who contributed to our work. Funding statement: this research was funded by a grant from Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), grant number 2015/14819-7. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016;26(1):1-133. 2. Lloyd R, Osamura R, Kloppel G. WHO Classification of Tumours: Pathology and Genetics of Tumours of Endocrine Organs. 4a. Lyon, France: IARC; 2017. 3. Hundahl SA, Fleming ID, Fremgen AM, Menck HR. A National Cancer Data Base report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985-1995. Cancer. 1998;83(12):2638-48. 4. Auger M. Hürthle cells in fine-needle aspirates of the thyroid: A review of their diagnostic criteria and significance. Cancer Cytopathol. 2014;122(4):241-9. 5. Cannon J. The Significance of Hurthle Cells in Thyroid Disease. Oncologist. 2011;16:1380-7. 6. Mete O, Asa SL. Oncocytes, oxyphils, hürthle, and askanazy cells: Morphological and molecular features of oncocytic thyroid nodules. Endocr Pathol. 2010;21(1):16-24. 7. Carcangiu ML, Bianchi S, Savino D, Voynick IM, Rosai J. Follicular Hurthle cell tumors of the thyroid gland. Cancer. 1991;68(9): 1944-53. 8. Kushchayeva Y, Duh QY, Kebebew E, Clark OH. Prognostic indications for Hürthle cell cancer. World J Surg. 2004;28(12):1266-70. 9. Lopez-Penabad L, Chiu AC, Hoff AO, Schultz P, Gaztambide S, Ordoñez NG, et al. Prognostic factors in patients with Hürthle cell neoplasms of the thyroid. Cancer. 2003;97(5):1186-94. 10. Chindris A-M, Casler JD, Bernet VJ, Rivera M, Thomas C, Kachergus JM, et al. Clinical and Molecular Features of Hürthle Cell Carcinoma of the Thyroid. J Clin Endocrinol Metab. 2015;100(January):55-62. 11. Besic N, Vidergar-Kralj B, Frkovic-Grazio S, Movrin-Stanovnik T, Auersperg M. The Role of Radioactive Iodine in the Treatment of Hürthle Cell Carcinoma of the Thyroid. Thyroid. 2003 Jun [cited 2018 Oct 14];13(6):577-84. 12. Harrell R, Bimston D. Surgical Utility of Afirma: Effects of High Cancer Prevalence and Oncocytic Cell Types in Patients with Indeterminate Thyroid Cytology. Endocr Pract. 2014 Apr;20(4):364-9. Arch Endocrinol Metab. 2019;63/3

13. Pu RT, Yang J, Wasserman PG, Bhuiya T, Griffith KA, Michael CW. Does Hurthle cell lesion/neoplasm predict malignancy more than follicular lesion/neoplasm on thyroid fine-needle aspiration? Diagn Cytopathol. 2006 May;34(5):330-4. 14. Roh MH, Jo VY, Stelow EB, Faquin WC, Zou KH, Alexander EK, et al. The predictive value of the fine-needle aspiration diagnosis “Suspicious for a follicular neoplasm, Hürthle cell type” in patients with Hashimoto thyroiditis. Am J Clin Pathol. 2011;135(1):139-45. 15. Giorgadze T, Rossi ED, Fadda G, Gupta PK, Livolsi VA, Baloch Z. Does the fine-needle aspiration diagnosis of “Hürthle-cell neoplasm/follicular neoplasm with oncocytic features” denote increased risk of malignancy? Diagn Cytopathol. 2004;31(5):307-12. 16. Tuzun D, Ersoy R, Yazgan AK, Kiyak G, Yalcin S, Cakir B. Cytomorphologic features and ultrasonographic characteristics of thyroid nodules with Hurthle cells. Ann Diagn Pathol. 2015;19(3):175-9. 17. Kim TH, Lim JA, Ahn HY, Lee EK, Min HS, Won Kim K, et al. Tumor size and age predict the risk of malignancy in Hürthle cell neoplasm of the thyroid and can therefore guide the extent of initial thyroid surgery. Thyroid. 2010;20(11):1229-34. 18. Parikh PP, Allan BJ, Lew JI. Surgeon-performed ultrasound predictors of malignancy in patients with Hü rthle cell neoplasms of the thyroid. J Surg Res. 2013;184(1):247-52. 19. Lee SK, Rho BH, Woo SK. Hürthle cell neoplasm: Correlation of gray-scale and power doppler sonographic findings with gross pathology. J Clin Ultrasound. 2010;38(4):169-76. 20. Maizlin Z V, Wiseman SM, Vora P, Kirby JM, Mason AC, Filipenko D, et al. Hurthle cell neoplasms of the thyroid: sonographic appearance and histologic characteristics. J Ultrasound Med. 2008;27(5):751-7; quiz 759. 21. Rago T, Di Coscio G, Basolo F, Scutari M, Elisei R, Berti P, et al. Combined clinical, thyroid ultrasound and cytological features help to predict thyroid malignancy in follicular and Hürthle cell thyroid lesions: Results from a series of 505 consecutive patients. Clin Endocrinol (Oxf). 2007;66(1):13-20. 22. Tessler FN, Middleton WD, Grant EG, Hoang JK, Berland LL, Teefey SA, et al. ACR Thyroid Imaging, Reporting and Data System (TI-RADS): White Paper of the ACR TI-RADS Committee. J Am Coll Radiol. 2017;14(5):587-95. 23. Ito Y, Hirokawa M, Miyauchi A, Kihara M, Yabuta T, Masuoka H, et al. Diagnosis and surgical indications of oxyphilic follicular tumors in Japan: Surgical specimens and cytology. Endocr J. 2016;63(11):977-82. 24. Lee KH, Shin JH, Ko ES, Hahn SY, Kim JS, Kim JH, et al. Predictive factors of malignancy in patients with cytologically suspicious for Hurthle cell neoplasm of thyroid nodules. Int J Surg. 2013;11(9):898-902. 25. Sippel RS, Elaraj DM, Khanafshar E, Zarnegar R, Kebebew E, Duh Q-Y, et al. Tumor Size Predicts Malignant Potential in Hürthle Cell Neoplasms of the Thyroid. World J Surg. 2008;32(5):702-7. 26. Jeh SK, So LJ, Bum SK, Yoen SL. Evaluating the degree of conformity of papillary carcinoma and follicular carcinoma to the reported ultrasonographic findings of malignant thyroid tumor. Korean J Radiol. 2007;8(3):192-7. 27. Rago T, Di Coscio G, Basolo F, Scutari M, Elisei R, Berti P, et al. Combined clinical, thyroid ultrasound and cytological features help to predict thyroid malignancy in follicular and Hrthle cell thyroid lesions: results from a series of 505 consecutive patients. Clin Endocrinol (Oxf). 2007;66(1):13-20. 28. Brito JP, Gionfriddo MR, Al Nofal A, Boehmer KR, Leppin AL, Reading C, et al. The Accuracy of Thyroid Nodule Ultrasound to Predict Thyroid Cancer: Systematic Review and Meta-Analysis. J Clin Endocrinol Metab. 2014;99(4):1253-63. 29. Park JW, Kim DW, Kim D, Baek JW, Lee YJ, Baek HJ. Korean Thyroid Imaging Reporting and Data System features of follicular thyroid adenoma and carcinoma: a single-center study. Ultrasonography. 2017;36(4):349-54.

305

Authorship: NOS, RYAC, FABV, and SM collected, analyzed, and interpreted the clinical, cytologic, and sonographic data. RMCF, VNM, and MCC reviewed and analyzed the ultrasound images. CKS reassessed and interpreted the histopathology of the lesions. DLSD conceived the study and its design and performed the statistical analysis. NOS, AOH, SM, and DLSD analyzed and interpreted the compiled data. NOS and DLSD prepared the figures and drafted the main manuscript text. All authors provided critical review and approved the final version of this manuscript.

review

Insufficient iodine intake in pregnant women in different regions of the world: a systematic review ¹ Departamento de Nutrição e Saúde, Universidade Federal de Viçosa (UFV), Viçosa, MG, Brasil

Aline C. Candido¹

http://orcid.org/0000-0002-6259-4786

Núbia de S. de Morais¹ http://orcid.org/0000-0002-6457-3516

Luiza V. Dutra1

http://orcid.org/0000-0002-8351-655

Carina A. Pinto1

http://orcid.org/0000-0002-1186-6702

Sylvia do C. C. Franceschini 1 http://orcid.org/0000-0001-7934-4858

Rita de Cássia G. Alfenas1 http://orcid.org/0000-0003-2290-1611

ABSTRACT

Correspondence to: Aline Carare Candido Departamento de Nutrição e Saúde Ed. Centro de Ciências Biológicas II Av. Peter Henry Rolfs, s/nº, Campus Universitário 36570.900 – Viçosa, MG, Brasil alinecarrare@gmail.com Received on Dec/16/2018 Accepted on Apr/18/2019 DOI: 10.20945/2359-3997000000151

Objective: To determine the prevalence of insufficient iodine intake in pregnant women. Materials and methods: The search was performed in the electronic databases Medline (PubMed), Latin American and Caribbean Literature in Health Sciences (Lilacs) and Scopus. Review studies, experimental studies, those with adolescent pregnant women (< 20 years) and iodine supplementation were excluded. The selection followed the steps of identifying the articles in the databases, deleting the duplicates, and reading the titles, abstracts, and then the entire article.The search for the articles occurred in September 2017, using the descriptors “pregnant” and “iodine deficiency” NOT “supplementation” in English, Portuguese and Spanish. Results: Thirteen articles were included, the deficiency prevalence ranged from 16.1% to 84.0%, and the median of iodine intake was insufficient in 75% of the studies. There is no classification for mild, moderate or severe levels of iodine deficiency in pregnant women, which makes it impossible to know the real dimension of the problem. Conclusion: The high prevalence of insufficient iodine intake in pregnant women, observed worldwide, shows the need for a population classification in order to direct public policies. Arch Endocrinol Metab. 2019;63(3):306-11 Keywords Iodine deficiency; pregnant woman; prevalence

INTRODUCTION

odine is essential for the synthesis of thyroid hormones during pregnancy and for the fetal neurological development (1-3) The main consequences of low intake for pregnant women are goiter, spontaneous abortion, hypothyroidism and thyroid nodules. And for the fetus it can result in neonatal hypothyroidism, cretinism, retardation in growth and neuropsychomotor development (4). In pregnant women, the recommendation for iodine is higher because there is an increase in the production of thyroid hormone, renal losses and transfer of this mineral from the mother to the fetus, all of which increases the need (5). 306

The availability of iodine in nature differs by geographical area and deficiency is more associated with mountainous regions such as the Himalayas and Alps and areas with frequent flooding. In addition, other regions also have a scarcity of this mineral, such as Central Africa, Central Asia, Europe and in places where the soils are poor (6). Universal salt iodination was suggested in 1831 by the French scientist Boussingault to minimize the prevalence of goiter. As a result, this strategy reduced goiter in the population, increased urinary excretion, improved thyroid function and increased iodine intake in pregnant women, so it was implemented in several countries around the world (7,8). Arch Endocrinol Metab. 2019;63/3

Insufficient iodine intake in pregnant women

The search returned 469 articles. After eliminating duplicates by bases and among bases, 243 remained. Arch Endocrinol Metab. 2019;63/3

Articles identified in the searches (n = 469) PubMed: 161 Lilacs: 3 Scopus: 305 Duplicate articles deleted n = 226 Articles selected after deleting duplicates n = 243

Articles selected after reading the titles n = 150

Articles selected after reading the abstracts n = 67

Articles selected and included after complete reading n = 13

Articles excluded after reading titles (n = 93) – Not related to the topic – Thyroid dysfunction – Supplementation – Review articles – Official documents

Articles excluded after reading the abstracts (n = 83) – Supplementation – Thyroid dysfunction – Adolescent – Review articles – Others Articles excluded after complete reading (n = 54) – No prevalence data – Supplementation – Thyroid dysfunction – Adolescent – Pathologies

Figure 1. Flowchart of the process of identification and selection of the articles included. Source: PRISMA (9). 307

RESULTS

Identification

The review followed the recommendations of the Preferred Reporting Items for Systematic Reviews (PRISMA) (9) and was based on the guiding question “Is there a reason for concern about insufficient iodine intake in pregnant women?”. The article search occurred in September 2017 without date delimitation. The authors independently searched the electronic databases Publisher Medline (PubMed), Latin American and Caribbean Literature in Health Sciences (Lilacs), and Scopus. Descriptors indexed in the Health Science Descriptors system (Decs) were combined as follows: “pregnant” AND “iodine deficiency” NOT “supplementation”, in English, Portuguese and Spanish. For the PubMed search, we used the human, pregnant and adult filters, and in the Scopus filters, we used articles and pregnant women. Original articles on the prevalence of insufficient iodine intake in adult pregnant women (≥ 20 years) based on the Urinary Iodine Concentration (UIC), according to data from World Health Organization (WHO), were included (10). Review studies, experimental studies, those with adolescent pregnant women (< 20 years old) and iodine supplementation were excluded. The selection followed the steps of identifying the articles in the databases, deleting the duplicates, and reading the titles, abstracts, and then the entire article. The methodological quality of the studies was evaluated by the questionnaire proposed by Downs and Black (11), which contains 27 questions divided into four categories: study report (main findings described), external validity (evaluates representativeness), internal validity (investigates biases and confounding factors) and study power. We excluded 10 of the 27 questions since they referred to experimental studies. Each answer received a score of “0” (if it did not meet the criterion evaluated) or “1” (if the criterion was met), with a maximum of 17 points.

Selection

MATERIALS AND METHODS

After reading titles, abstracts and articles in full, 13 were included (Figure 1). The years of the studies ranged from 2010 to 2016, with three longitudinal and 10 cross-sectional studies, performed in North and South America (Venezuela and Canada), Africa (Democratic Republic of Congo), Asia (Japan, Iran, Turkey, Bangladesh, India), Europe (England and Spain), and Oceania (Australia). The sample size ranged from 36 to 5,256 pregnant women and the prevalence of insufficient iodine intake ranged from 16.1 to 84.0%. In order to evaluate the urinary iodine concentration (UIC) of pregnant women, the WHO reference was considered, where UIC < 150 μg/L is classified as insufficient iodine intake, 150-249 μg/L adequate, 250-499 μg/L more than necessary, ≥ 500 μg/L excessive intake (10). Table 1 describes the results of the studies included in this review, with a median UIC ranging from 56.8 to 224.5 μg/L. As described in Table 1, we found that 75% (n = 9) of the studies were classified as insufficient iodine intake.

Eligibility and inclusion

Based on this information, the identification of iodine deficient countries allows us to build a global structure for the formulation of targeted and effective public policies. Therefore, our goal is to determine the prevalence of insufficient iodine intake in pregnant women.

Insufficient iodine intake in pregnant women

Table 1. Description of the studies selected for systematic review Authors / Year

Nguyen and cols. (2010) (12)

Caballero (2011) (13)

Fuse and cols. (2011) (14)

Çetinkaya and cols. (2012) (15)

Shamim and cols. (2012) (16)

Bath and cols. (2013) (17)

Aguayo and cols. (2013) (18)

308

Site

Australia

Venezuela

Japan

Turkey

Bangladesh

England

Spain

Study Design

Recruitment

Crosssectional

Pregnant women from any trimester of gestation seen at Canberra Hospital, Australia

Crosssectional

Pregnant women were selected at the prenatal visit. We selected 300 pregnant women, 100 women from each quarter

Crosssectional

Healthy pregnant women were recruited during the 3 quarters and puerperal with 5 to 6 weeks postpartum without history of thyroid disease attended at Yamaguchi Hospital, Funabashi city

Crosssectional

We recruited pregnant women in the three trimesters of gestation at the Medical University of Ataturk, Turkey

Crosssectional

We recruit pregnant women from rural areas in northwestern Bangladesh. A randomized, placebocontrolled study with vitamin A or beta-carotene supplementation

Longitudinal

Pregnant women living in the old Avon area in South West England.

Longitudinal

Women attending obstetric outpatient appointments in the catchment area of Cruces Hospital

Sample size (n)

100

300

934

113

2490

Trimester of pregnancy The authors did not mention the trimester of gestation

First trimester (n = 100), second trimester (n = 100) and third trimester (n = 100) First trimester (n = 243), second trimester (n = 541), third trimester (n = 466) and postpartum (n = 533)

First trimester (n = 30), second trimester (n = 49) and third trimester (n = 34)

First trimester (≤ 16 weeks, n = 1376) and third trimester (≥ 32 weeks, n = 1114)

First trimester (< 13 weeks) 1040

2104

First trimester (n = 2104) and second trimester (n = 1322)

Median Urinary Iodine Concentration (UIC) (μg/L)

Median UIC Classification

Prevalence of Iodine Deficiency

Quality of the Studies Included

62 (12-750)

Insufficient

84%

12 points

224.5

Adequate

25%

14 points

219

Adequate

16.1%

16 points

132.8 (15.1-291.6)

Insufficient

72.6%

9 points

Initial: 66 (34 - 133) Final: 55 (28 -110)

Insufficient

Initial: 78.85¨% Final: 82.94%

16 points

91.1 (53.8 - 143)

Insufficient

67%

13 points

Quarter 1st 88.5 (16 -875) 2nd 140 (21 – 880)

Insufficient

Quarter 1st – 79.8% 2nd – 54.4%

16 points

Arch Endocrinol Metab. 2019;63/3

Insufficient iodine intake in pregnant women

Amouzegar and Azizi (2013) (19)

Site

Turkey

Study Design

Recruitment

Crosssectional

Pregnant women referred to the mother and child health care clinics of two maternity hospitals of Tehran

Habimana and cols. (2013) (20)

Democratic Republic of Congo

Crosssectional

Katz and cols. (2013) (4)

Canada

Crosssectional

Joshi and cols. (2014) (21)

Bath and cols. (2015) (22)

India

England

Delshad and cols. (2016) (23)

Iran

Crosssectional

Longitudinal

Crosssectional

375 pregnant women attending antenatal consultation, 125 women in each of the three maternity units 250 pregnant women from a clinical hospital Pregnant women (n = 5256) attended the Jamnabai General Hospital. 230 British pregnant women recruited for the Selenium in Pregnancy intervention

Pregnant women attended at maternal and child health centers

Median Urinary Iodine Concentration (UIC) (μg/L)

Median UIC Classification

Prevalence of Iodine Deficiency

Quality of the Studies Included

138.4 (24.1 – 404)

Insufficient

34.0%

16 points

225

225 pregnant women in the three quarters, 75 women who gave birth and 75 non-pregnant women as controls

138 (105 – 172)

Insufficient

52.3%

13 points

142

Second or third trimester of a singleton pregnancy

221 (142 – 397)

Adequate

29.6%

12 points

297, 14

More than necessary

16.79%

10 points

230

Pregnant women in the first trimester (12-14 weeks), second trimester (20 weeks) and third trimester (35 weeks)

56.8 (31.1 – 104)

Insufficient

55.7%

15 points

1072

Singleton pregnancy and in the first, second and third trimester of their pregnancy

87.3 (43.5 – 139.1)

Insufficient

78.17%

10 points

Sample size (n)

Trimester of pregnancy First trimester (< 15 weeks)

First trimester (< 15 weeks) 5.256

In the evaluation of the methodological quality of the studies, the lowest score was nine and the highest 16. The best-evaluated criteria were the study report (main findings described), external validity (the same follow-up time for all individuals, appropriate statistical tests, and outcomes with reliable measures) and internal validity (individuals recruited in the same period). Only two studies presented power and adjustment for confounding factors in the analysis.

DISCUSSION Insufficient iodine intake is an obstacle to social and economic development, reaching approximately Arch Endocrinol Metab. 2019;63/3

2 billion people worldwide (10). To eradicate the disorders caused by deficiency, universal iodination of salt is used as a safe, economical and sustainable strategy to ensure adequate intake worldwide (24). Globally, 86.6% of households have access to iodized salt, with the number of people consuming it increasing from 1 billion to approximately 4 billion in the last 10 years (25) In this study, of the evaluated countries, Japan has no legislation for salt iodination, in England and Spain, it is voluntary, and it is mandatory in the others. The highest prevalence of deficiency detected was in Australia (84%), located in Oceania, different from expected, since there is a mandatory salt iodization 309

Authors / Year

Insufficient iodine intake in pregnant women

policy and is surrounded by the Indian and Pacific oceans. However, natural disasters are recurrent in that country, and iodine deficiency can be an ecological phenomenon caused by flooding and soil erosion and, consequently, food crops will be deficient (4). Asian countries: Turkey, Bangladesh and Iran showed insufficient iodine intake in pregnant women. Despite the extensive territory, the exponential growth of the population has led to a shortage of basic survival conditions, making access to adequate food difficult. This deficiency may also be a reflection of geographic characteristics such as mountains, floodplains and distance from the sea, restricting access to iodine sources.10 However, in India, iodine intake was more than necessary, demonstrating that monitoring salt iodization is critical, since overeating can be detrimental to health (21). In contrast, the lower prevalence of deficiency was observed in Japan, which has no legislation for salt iodization. However, the habit of ingesting dietary sources of iodine, without subjecting them to high temperatures, may justify adequate population status (26). Studies in European countries, England and Spain, found insufficient intake. The authors attributed this result to variations in the consumption of dietary sources of iodine and to the fact that pregnant women did not receive supplementation (18,22). In Venezuela and Canada, located in South and North America, iodine intake was adequate, demonstrating the success of awareness campaigns conducted in these countries for the consumption of iodized salt to protect the mother from health problems (4,13). At the 60th WHO World Health Assembly in 2007, about 31% of the world population had insufficient iodine intake and the most affected regions were Asia and Europe, while in the Americas the intake of fortified salt ensured adequate iodine status (25). In Africa, a study in the Democratic Republic of Congo found insufficient iodine intake in the semiurban and rural region, reflecting low socioeconomic status and population location, making access to iodine sources difficult. In addition, in the rural area, the habit of using natural spices instead of fortified salt contributes to this deficiency (20). The difference in the prevalence of insufficient iodine intake can be attributed to geographic characteristics, dietary habits and salt iodination policy. This study was geographically representative allowing an overview of the prevalence of insufficient iodine intake in pregnant women in different regions of the world: the overall state of iodine in pregnant women. 310

The limitation is that the classification for levels of mild, moderate or severe iodine deficiency in pregnant women is not defined, making it impossible to know the magnitude of the problem. In conclusion, pregnant women are a group vulnerable to insufficient iodine intake and the high prevalence observed in this review confirms the severity of this health problem worldwide. Therefore, there is a need for a population classification to guide public policies, as well as strategies such as salt iodization that should receive government support, and an effective monitoring to ensure adequate iodine intake. Acknowledgements: this study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001; the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), process 408295/20171 and the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG) process APQ-03336-18. Funding Source: this study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001; the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), process 408295/2017-1 and the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG) process APQ-03336-18. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Calvo RM, Jauniaux E, Gulbis B, Asunción M, Gervy C, Contempré B, et al. Fetal tissues are exposed to biologically relevant free thyroxine concentrations during early phases of development. J Clin Endocrinol Metab. 2002;87(4):1768-77. 2. Morreale de Escobar G, Obregon MJ, Escobar del Rey F. Role of thyroid hormone during early brain development. Eur J Endocrinol [Internet]. 2004;(Suppl 3):U25-37. 3. Teixeira D, Calhau C, Pestana D, Vicente L, Graça P. Iodo – Importância para a saúde e o papel da alimentação. Programa Nacional para a Promoção da Alimentação Saudável – DireçãoGeral da Saúde; 2014. 4. Katz PM, Leung AM, Braverman LE, Pearce EM, Tomlinson G, He X, et al. Iodine Nutrition During Pregnancy In Toronto, Canada. Endocr Pract. 2013;19(2):206-11. 5. Lazarus JH. Thyroid disease in relation to pregnancy: A decade of change. Clin Endocrinol (Oxf) [Internet]. 2000;53(3):265-78. 6. Linhares DPS, Garcia PV, Almada A, Ferreira T, Queiroz G, Cruz JV, et al. Iodine environmental availability and human intake in oceanic islands: Azores as a case-study. Sci Total Environ. 2015 Dec 15;538:531-8. 7. Knobel M, Medeiros-Neto G. Moléstias associadas à carência crônica de iodo. Arq Bras Endocrinol Metabol. 2004;48(1):53-61. 8. Pontes AAN De, Rocha ADM, Leite DFB, Lessa ADF, Adan LFF. Iodization of salt in Brazil, a controversial subject. Arq Bras Endocrinol Metabol. 2009;53(1):113-4. 9. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, et al. The PRISMA statement for reporting systematic Arch Endocrinol Metab. 2019;63/3

Insufficient iodine intake in pregnant women

reviews and meta-analyses of studies that evaluate healthcare interventions: Explanation and elaboration. BMJ. 2009 Jul 21;339:b2700. 10. WHO. Assessment of the iodine deficiency disorders and monitoring their elimination. World Heal Organ (WHO), Geneva [Internet]. 2007;3:1-107. 11. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377-84. 12. Nguyen B, Baker D, Southcott E, Potter J, Sneddon A, Hickman PE. Iodine deficiency in pregnant women in the ACT. Aust New Zeal J Obstet Gynaecol. 2010;50(6):539-42. 13. Caballero L. Yoduria en escolares y embarazadas del estado Trujillo, Venezuela 2007-2008 Urinary Iodine in school children and pregnant women of Trujillo state, Venezuela 2007-2008. Raem No. 2011;48(4):206-11. 14. Fuse Y, Ohashi T, Yamaguchi S, Yamaguchi M, Shishiba Y, Irie M. Iodine status of pregnant and postpartum Japanese women: Effect of iodine intake on maternal and neonatal thyroid function in an iodine-sufficient area. J Clin Endocrinol Metab. 2011;96(12):3846-54.

18. Aguayo A, Grau G, Vela A, Aniel-Quiroga A, Espada M, Martul P, et al. Urinary iodine and thyroid function in a population of healthy pregnant women in the North of Spain. J Trace Elem Med Biol. 2013;27(4):302-6. 19. Amouzegar A, Azizi F. Variations of urinary iodine during the first trimester of pregnancy in an iodine-replete area. Comparison with non-pregnant women. Hormones. 2013;12(1):111-8. 20. Habimana L, Twite KE, Wallemacq P, De Nayer P, Daumerie C, Donnen P, et al. Iodine and iron status of pregnant women in Lubumbashi, Democratic Republic of Congo. Public Health Nutr. 2013;16(8):1362-70. 21. Joshi K, Nair S, Khade C, Rajan MGR. Early gestation screening of pregnant women for iodine deficiency disorders and iron deficiency in urban centre in Vadodara, Gujarat, India. J Dev Orig Health Dis [Internet]. 2014;5(01):63-8. 22. Bath SC, Furmidge-Owen VL, Redman CW, Rayman MP. Gestational changes in iodine status in a cohort study of pregnant women from the United Kingdom: season as an effect modifier. Am J Clin Nutr. 2015;101(6):1180-7. 23. Delshad H, Touhidi M, Abdollahi Z, Hedayati M, Salehi F, Azizi F. Inadequate iodine nutrition of pregnant women in an area of iodine sufficiency. J Endocrinol Invest. 2016;39(7):755-62.

15. Çetinkaya K, Ingeç M, Çetinkaya S, Kaplan I. Iodine defi ciency in pregnancy and in women of reproductive age in Erzurum, Turkey. 2012;42(4):675-80.

24. World Health Organization (WHO). World Summit for Children – Mid-Decade Goal: Iodine Deficiency Disorders (Idd). UNICEFWHO Jt Comm Heal Policy Spec Sess. 1994;2(27):27-8.

16. Shamim A, Parul C, Schulze K, Ali H, Alamgir K, Rashid M, et al. Iodine status in pregnancy and household salt iodine content in rural Bangladesh. Matern Child Nutr. 2012;8(2):162-73.

25. Pretell E. Boletim de Carências Nutricionais: Distúrbios por Deficiência de Iodo – DDI. Ministério da Saúde. Secretaria de Atenção à Saúde Política de Alimentação e Nutrição (CGPAN). 2008; 1: 1-6. 26. Fisher J, Tran T, Biggs B, Tran T, Dwyer T, Casey G, et al. Iodine status in late pregnancy and psychosocial determinants of iodized salt use in rural northern Viet Nam. Bull World Health Organ. 2011;89(11):813-20.

17. Bath SC, Steer CD, Golding J, Emmett P, Rayman MP. Effect of inadequate iodine status in UK pregnant women on cognitive outcomes in their children: Results from the Avon Longitudinal Study of Parents and Children (ALSPAC). Lancet [Internet]. 2013;382(9889):331-7.

Arch Endocrinol Metab. 2019;63/3

311

Instructions for authors GENERAL INFORMATION We emphasize the importance of following these instructions carefully. Failure to do so will delay the processing of your manuscript. Manuscripts should be submitted solely to the AE&M and should not have been published, or be under consideration for publication in any substantial form, in another periodical-either professional or lay. Manuscripts should be submitted in English. Proofreading by a scientific editing service is strongly recommended; the following companies are suggested: Voxmed Medical Communications, American Journal Experts and PaperCheck. Manuscriptss that successfully complete the peer-review process and are recommended for publication will only be accepted and published upon receipt of a certificate proving professional academic English proofreading. In extraordinary circumstances, the certificate can be waived by editorial decision. Papers that do not meet these requirements will be returned to the author for the necessary revisions before formal review. Authors must include the ORCID (Open Researcher and Contributor ID) in the article submission and in the manuscript file. All submissions are initially evaluated in depth by the scientific editors. Papers that do not conform with the general criteria for publication will be returned to the authors without detailed review, typically within three to five days. Otherwise, manuscripts will be sent to reviewers (most commonly two).

tion of the natural history or management of the condition described. Case Reports should be 2,000 words or less, with no more than four figures and tables, and no more than 30 references. We emphasize that only case reports that offer important basic translational or clinical contributions, preferentially together with a review of the literature, will be considered for publication.

Letters to the Editor Letters to the Editor may be submitted in response to manuscript that has been published in the Journal. Letters should be short commentaries related to specific points of agreement or disagreement with the published manuscript. Letters are not intended for the presentation of original data unrelated to a published article. Letters should be no longer than 500 words, with no more than five complete references, and should not include any figures or tables.

MANUSCRIPT PREPARATION GENERAL FORMAT

MANUSCRIPT CATEGORIES Reports of original research may be submitted to AE&M as Original Articles or Brief Reports. Other special categories of manuscripts are described below. All manuscripts must adhere to the word count limitations, as specified below, for text only; word count does not include the abstract, references, or figures/tables and their legends. Word count must be shown on the title page, along with the number of figures and tables. The format is similar for all manuscript categories, and it is described in detail in the “Manuscript Preparation” section.

The Journal requires that all manuscripts be submitted in a single-column format that follows these guidelines: • The manuscript must be submitted in MS-Word format. • All text should be double-spaced with 2 cm margins on both sides using 11-point type Times Roman or Arial font. • All lines should be numbered throughout the entire manuscript and the entire document should be paginated. • All tables and figures must be placed after the text and must be labeled. Submitted papers must be complete, including the title page, abstract, figures, and tables. Papers submitted without all of these components will be placed on hold until the manuscript is complete.

Original Articles The Original Article is a scientific report of the results of original research that has not been published or submitted for publication elsewhere (either in print or electronically). It represents a substantial body of laboratory or clinical work. In general, Original Articles should not exceed 3,600 words in the main text, include more than six figures and tables, or more than 35 references.

Review Articles The AE&M publishes Review Articles that show a balanced perspective on timely issues within the field of clinical endocrinology. All reviews are submitted upon invitation and are subject to peer review. Articles in this category are requested by the Editors to authors with proven expertise in the field. Authors considering the submission of uninvited reviews should contact the editors in advance to determine whether the topic that they propose is of current potential interest to the Journal. Review articles should be no longer than 4,000 words in the main text, include no more than four figures and tables, and no more than 60 references. The author should mention the source and/or request authorization for use of previously published figures or tables.

Consensus Statements Consensus Statements related to the endocrine and metabolic health standards and healthcare practices may be submitted by professional societies, task forces, and other consortia. All such submissions will be subjected to peer review, must be modifiable in response to criticism, and will be published only if they meet the usual editorial standards of the Journal. Consensus Statements should typically be no longer than 3,600 words in the main text, include no more than six figures and tables, and no more than 60 references.

Brief Report The Brief Report consists of new data of sufficient importance to warrant immediate publication. It is a succinct description of focused study with important, but very straightforward, negative or confirmatory results. Brevity and clarity are always likely to enhance the chance of a manuscript being accepted for publication. A maximum of 1,500 words in the main text plus up to 20 references and normally no more than two illustrations (tables or figures or one of each) are acceptable for Brief Reports.

ALL SUBMISSIONS MUST INCLUDE: • A cover letter requesting the evaluation of the manuscript for publication in AE&M, and any information relevant to the manuscript. The manuscript´s originality and exclusivity should be stated, as well as the contribution of each author, Elsewhere on the submission form, authors may suggest up to three specific reviewers and/or request the exclusion of up to three others.

The manuscript must be presented in the following order: 1. Title page. 2. Structured abstract (or summary for case reports). 3. Main text. 4. Tables and figures. They must be cited in the main text in numerical order. 5. Acknowledgments. 6. Funding statement, competing interests and any grants or fellowships supporting the writing of the paper. 7. List of references.

Title Page The title page must contain the following information: 1. Title of the article (a concise statement of the major contents of the article). 2. Full names, departments, institutions, city, and country of all co-authors. 3. Full name, postal address, e-mail, telephone and fax numbers of the corresponding author. 4. Abbreviated title of no more than 40 characters for page headings. 5. Up to five keywords or phrases suitable for use in an index (the use of MeSH terms is recommended). 6. Word count – excluding title page, abstract, references, figures/tables and their legends. 7. Article type

Case Report

Structured Abstracts

A Case Report is a brief communication presenting collected or single case reports of clinical or scientific significance. These reports should be concise and focused on the issue to be discussed. They should address observations of patients or families that add substantially to the knowledge of the etiology, pathogenesis, and delinea-

All Original Articles, Brief Reports, Reviews, Case Reports should be submitted with structured abstracts of no more than 250 words. The abstract must be self-contained and clear without reference to the text, and should be written for general journal readership. The abstract format should include four sections that reflect the section

headings in the main text. All information reported in the abstract must appear in the manuscript. Please use complete sentences for all sections of the abstract.

of reproducing color figures in print (the publisher will provide price quotes upon acceptance of the manuscript).

Introduction

Photographs

The article should begin with a brief introductory statement that places the study in historical perspective, and explains its objective and significance.

The AE&M strongly prefers to publish unmasked patient photos. We encourage all prospective authors to work with families prior to submission and address the issue of permission for review and possible publication of patient images. If your submission contains ANY identifiable patient images or other protected health information, you MUST provide documented permission from the patient (or the patient’s parent, guardian, or legal representative) before the specific material circulates among editors, reviewers and staff for the purpose of possible publication in AE&M. If it is necessary to identify an individual, use a numerical designation (e.g. Patient 1) rather than using any other identifying notations, such as initials.

Materials and Methods These should be described and referenced in sufficient detail for other investigators to be able to repeat the study. The source of hormones, unusual chemicals and reagents, and special pieces of apparatus should be stated. For modified methods, only the modifications need be described.

Units of Measure The Results section should briefly present the experimental data in text, tables, and/ or figures. For details on preparation of tables and figures, see below. The Discussion should focus on the interpretation and significance of the findings, with concise objective comments that describe their relation to other studies in that area. The Discussion should not reiterate the Results.

Authorship The AE&M ascribes to the authorship and contributorship guidelines defined by the International Committee of Medical Journal Editors (www.ICMJE.org). Unrestricted joint authorship is allowed. A maximum of two corresponding authors is allowed. The uniform requirements for manuscripts submitted to medical journals state that authorship credit should be based only on substantial contribution to: 1. The conception and design, or analysis and interpretation of data. 2. The drafting of the article or its critical review for important intellectual content. 3. The final approval of the version to be published. All these conditions must be met. The corresponding author is responsible for ensuring that all appropriate contributors are listed as authors, and that all authors have agreed with the content of the manuscript and its submission to the AE&M.

Conflict of interest A conflict of interest statement for all authors must be included in the main document, following the text, in the Acknowledgments section. If authors have no relevant conflict of interest to disclose, this should be indicated in the Acknowledgments section.

Acknowledgments The Acknowledgments section should include the names of those people who contributed to a study but did not meet the requirements for authorship. The corresponding author is responsible for informing each person listed in the acknowledgment section that they have been included and providing them with a description of their contribution so they know the activity for which they are considered responsible. Each person listed in the acknowledgments must give permission – in writing, if possible – for the use of his or her name. It is the responsibility of the corresponding author to provide this information.

References References to the literature should be cited in numerical order (in parentheses) in the text and listed in the same numerical order at the end of the manuscript on a separate page or pages. The author is responsible for the accuracy of references. The number of references cited is limited for each category of submission, as indicated above.

Tables Tables should be submitted in the same format as the article (Word), and not in another format. Please note: we cannot accept tables as Excel files within the manuscript. Tables should be self-explanatory and the data they contain must not be duplicated in the text or figures. Tables must be constructed as simply as possible and be intelligible without reference to the text. Each table must have a concise heading. A description of experimental conditions may appear together with footnotes at the foot of the table. Tables must not simply duplicate the text or figures.

Figures and Legends All figures must display the figure number. Sizing the figure: the author is responsible for providing digital art that has been properly sized, cropped, and has adequate space between images. All color figures will be reproduced in full color in the online edition of the journal at no cost to the authors. Authors are requested to pay the cost

Results should be expressed in metric units. Temperature should be expressed in degrees Celsius and time of day using the 24-hour clock (e.g., 0800 h, 1500 h).

Standard Abbreviations All abbreviations must be immediately defined after it is first used in the text.

Experimental Subjects To be considered for publication, all clinical investigations described in submitted manuscripts must have been conducted in accordance with the guidelines of The Declaration of Helsinki, and must have been formally approved by the appropriate institutional review committees or their equivalent. The study populations should be described in detail. Subjects must be identified only by number or letter, not by initials or names. Photographs of patients’ faces should be included only if scientifically relevant. The authors must obtain written consent from the patient for the use of such photographs. For further details, see the Ethical Guidelines. Investigators must disclose potential conflict of interest to study participants and should indicate in the manuscript that they have done so.

Experimental Animals A statement confirming that all animal experimentation described in the manuscript was conducted in accordance with accepted standards of humane animal care, as outlined in the Ethical Guidelines, should be included in the manuscript.

Molecular Genetic Description • Use standard terminology for variants, providing rs numbers for all variants reported. These can be easily derived for novel variants uncovered by the study. Where rs numbers are provided, the details of the assay (primer sequences, PCR conditions, etc.) should be described very concisely. • Pedigrees should be drawn according to published standards (See Bennett et al. J Genet Counsel (2008) 17:424-433 - DOI 10.1007/s10897-008-9169-9).

Nomenclatures • For genes, use genetic notation and symbols approved by the HUGO Gene Nomenclature Committee (HGNC) – (http://www.genenames.org/). • For mutation nomenclature, please use the nomenclature guidelines suggested by the Human Genome Variation Society (http://www.hgvs.org/mutnomen/) • Provide information and a discussion of departures from Hardy-Weinberg equilibrium (HWE). The calculation of HWE may help uncover genotyping errors and impact on downstream analytical methods that assume HWE. • Provide raw genotype frequencies in addition to allele frequencies. It is also desirable to provide haplotype frequencies. • Whenever possible, drugs should be given their approved generic name. Where a proprietary (brand) name is used, it should begin with a capital letter. • Acronyms should be used sparingly and fully explained when first used.

Papers must be written in clear, concise English. Avoid jargon and neologisms. The journal is not prepared to undertake major correction of language, which is the responsibility of the author. Where English is not the first language of the authors, the paper must be checked by a native English speaker. For non-native English speakers and international authors who would like assistance with their writing before submission, we suggest Voxmed Medical Communications, American Journal Experts or PaperCheck. ISSN 2359-3997 © A&EM – Rua Botucatu, 572 – conjunto 83 – 04023-062 – São Paulo, SP, Brazil

Results and Discussion