AE&M 62-3 by Sociedade Brasileira de Endocrinologia e Metabologia

ISSN 2359-3997

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM Vol. 62 – No. 03 – June 2018

Archives of Endocrinology

and Metabolism

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM

Archives of

Endocrinology

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

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM Vol. 62 – No. 03 – June 2018

editorials 271 The elusive clinical significance of osteocalcin actions in energy metabolism in human

Archives of Endocrinology Bruno Ferraz-de-Souza

OFFICIAL JOURNAL OF THE BRAZILIAN original articles SOCIETY 275 Relationship between adiponectin and leptin on osteocalcin in obese adolescents during weight OF loss therapy Raquel Munhoz da Silveira Campos, Deborah Cristina Landi Masquio, Flávia Campos Corgosinho, Joana Pereira de Carvalho-Ferreira, ENDOCRINOLOGY Bárbara Dal Molin Netto, Ana Paula Grotti Clemente, Lian Tock, Sergio Tufik, Marco Túlio de Mello, Ana Raimunda Dâmaso 285 Waist circumference is an effect modifier of the association between bone mineral density and glucose metabolism AND METABOLISM Lygia N. Barroso, Dayana R. Farias, Marcia Soares-Mota,Heloisa Bettiol, Marco Antônio Barbieri, Milton Cesar Foss, 273 How deep is our anxiety during treatment of thyroid cancer? Flavio Carneiro Hojaij

and Metabolism Antônio Augusto M. da Silva, Gilberto Kac

296 Accuracy of sentinel lymph node mapping in detecting occult neck metastasis in papillary thyroid carcinoma Jose Higino Steck, Elaine Stabenow, Gustavo Baldove Bettoni, Samuel Steck, Claudio Roberto Cernea

303 Estimated costs of hospitalization due to coronary artery disease attributable to familial hypercholesterolemia in the Brazilian public health system Luciana R. Bahia, Roger S. Rosa, Raul D. Santos, Denizar V. Araujo

309 Metabolic syndrome components are associated with oxidative stress in overweight and obese patients

Nayara Rampazzo Morelli, Bruna Miglioranza Scavuzzi, Lucia Helena da Silva Miglioranza, Marcell Alysson Batisti Lozovoy, Andréa Name Colado Simão, Isaias Dichi

Archives of

319 Self-report of psychological symptoms in hypoparathyroidism patients on conventional therapy

Amanda J. Arneiro, Bianca C. C. Duarte, Rodrigo M. Kulchetscki, Victor B. S. Cury, Maicon P. Lopes, Breno S. Kliemann, Ileana B. Bini, Marcos A. Assad, Gleyne L. K. Biagini, Victoria Z. C. Borba,Carolina A. Moreira

325 Bouts of exercise elicit discordant testosterone: cortisol ratios in runners and non-runners

Thiago Paes de Barros De Luccia, José Eduardo Soubhia Natali, Alexandre Moreira, José Guilherme Chaui-Berlinck, José Eduardo Pereira Wilken Bicudo

Endocrinology

332 Maternal hypothyroxinemia in the first trimester of gestation and association with obstetric and neonatal outcomes and iron deficiency: a prospective Brazilian study Pedro Weslley Rosario, Luis Fernando Faria Oliveira, Maria Regina Calsolari

337 Effectiveness and safety of carbohydrate counting in the management of adult patients with type 1 diabetes mellitus: a systematic review and meta-analysis

and Metabolism

Eliege Carolina Vaz, Gustavo José Martiniano Porfírio,Hélio Rubens de Carvalho Nunes, Vania dos Santos Nunes-Nogueira

346 Long term maintenance of glucose and lipid concentrations after Roux-en-Y gastric bypass

Fernanda Cristina Carvalho Mattos Magno, Priscila Alves Medeiros de Sousa, Marcelo Paiva Rodrigues, Lícia Lopes Pio Pereira, José Egídio Paulo de Oliveira, Eliane Lopes Rosado, João Régis Ivar Carneiro

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM Raiane P. Crespo, Tania A. S. S. Bachega, Berenice B. Mendonça, Larissa G. Gomes

review

352 An update of genetic basis of PCOS pathogenesis

brief reports 362 Serum TSH level stability after 5 years in euthyroid adults at low risk for thyroid dysfunction Pedro Weslley Rosario, Maria Regina Calsolari

366 Graves’ ophthalmopathy:low-dose dexamethasone reduces retinoic acid receptor-alpha gene expression in orbital fibroblasts Sarah Santiloni Cury, Miriane Oliveira, Maria Teresa Síbio, Sueli Clara, Renata De Azevedo Melo Luvizotto, Sandro Conde, Edson Nacib Jorge, Vania dos Santos Nunes, Célia Regina Nogueira,Gláucia Maria Ferreira da Silva Mazeto

case reports 370 Potential role of sorafenib as neoadjuvant therapy in unresectable papillary thyroid cancer

Debora L. S. Danilovic, Gilberto Castro Jr., Felipe S. R. Roitberg, Felipe A. B. Vanderlei, Fernanda A. Bonani, Ricardo M. C. Freitas, George B. Coura-Filho, Rosalinda Y. Camargo, Marco A. Kulcsar, Suemi Marui, Ana O. Hoff

376 Inflammatory myopathy in the context of an unusual overlapping laminopathy

Cristina Guillín-Amarelle, Sofía Sánchez-Iglesias, Antonio Mera, Elena Pintos, Ana Castro-Pais, Leticia Rodríguez-Cañete, Julio Pardo, Felipe F. Casanueva, David Araújo-Vilar

FAMÍLIA

NESINA® TAKEDA Completa no cuidado do DM21*

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: 22881.6.18

Endocrinology

Informations: Title: D’après Mondrian Year: 1984 Author: Claudio Tozzi Technique: Acrylic paint on canvas OFFICIAL JOURNAL OF THE BRAZILIAN Measures: 150 x 206 cm

and Metabolism SOCIETY OF ENDOCRINOLOGY AND METABOLISM

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

Circulation of this issue: 3,500 copies Subscription: R$450.00/year – Single issue: R$55.00 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 2359-3997 (printed issues) 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

Supported by:

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

2015-2018 EDITOR-IN-CHIEF Marcello D. Bronstein (SP)

CO-EDITORS

REPRESENTATIVES OF COLLABORATING SOCIETIES SBD

Larissa Gomes (SP)

ABESO

Léa Maria Zanini Maciel (SP)

OFFICIAL JOURNALLeandro Kasuki (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

Ana Luiza Silva Maia (RS)

Maria Izabel Chiamolera (SP)

André Fernandes Reis (SP)

Maria Marta Sarquis (MG)

Andrea Glezer (SP)

Mario Saad (SP)

Tânia S. Bachega (SP)

Antônio Marcondes Lerário (SP)

Mário Vaisman (RJ)

INTERNATIONAL ASSOCIATE EDITOR

Antônio Roberto Chacra (SP)

Marise Lazaretti Castro (SP)

Ayrton Custódio Moreira (SP)

Milena Caldato (PA)

Shlomo Melmed (Los Angeles, EUA)

Berenice B. Mendonça (SP)

Raquel Soares Jallad (SP)

Bruno Halpern (SP)

Rodrigo Moreira (RJ)

Carlos Alberto Longui (SP)

Ruth Clapauch (RJ)

Archives of

FOUNDER

ASSOCIATE EDITORS

Waldemar Berardinelli (RJ)

PRESIDENTS OF THE SBEM DEPARTMENTS

EDITORS-IN-CHIEF, EDITORIAL OFFICE*

ADRENAL AND HYPERTENSION

Endocrinology César Luiz Boguszewski (PR)

Sandra R. G. Ferreira (SP)

Clarisse Ponte (CE)

Simão A. Lottemberg (SP)

Delmar Muniz Lourenço Jr. (SP)

Sonir Roberto Antonini (SP)

Luiz Alberto Andreotti Turatti (SP)

Denise Momesso (RJ)

Suemi Marui (SP)

DYSLIPIDEMIA AND ATHEROSCLEROSIS

Eder Carlos R. Quintão (SP)

Madson Queiroz de Almeida (SP)

1951-1955 Waldemar Berardinelli (RJ) Thales Martins (RJ)

DIABETES MELLITUS

1957-1972 Clementino Fraga Filho (RJ)

Cynthia Melissa Valério (RJ)

Edna Nakandakare (SP)

1964-1966* Luiz Carlos Lobo (RJ)

BASIC ENDOCRINOLOGY

Edna T. Kimura (SP)

Maria Izabel Chiamolera (SP)

1966-1968* Pedro Collett-Solberg (RJ) 1969-1972* João Gabriel H. Cordeiro (RJ)

and Metabolism

Weiss (RJ)

PEDIATRIC ENDOCRINOLOGY

Julienne Ângela Ramires de Carvalho (PR)

1983-1990 Antônio Roberto Chacra (SP)

BONE AND MINERAL METABOLISM

1995-2006 Claudio Elias Kater (SP) 2007-2010 Edna T. Kimura (SP) 2011-2014 Sergio Atala Dib (SP)

Elaine Maria Frade Costa (SP)

OFFICIAL JOURNAL OF THE FelipeBRAZILIAN Gaia (SP) SOCIETY OF ENDOCRINOLOGY AND METABOLISM Rita de Cássia Viana Vasconcellos Flavio Hojaij (SP) FEMININE ENDOCRINOLOGY AND ANDROLOGY

1978-1982 Armando de Aguiar Pupo (SP)

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

Laércio Joel Franco (SP)

Luiz Alberto Andreotti Turatti (SP)

Archives of Endocrinology 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) Rui M. B. Maciel (SP)

Julio Z. Abucham (SP)

Carolina Aguiar Moreira (PR) NEUROENDOCRINOLOGY

Marcello Delano Bronstein (SP) OBESITY

Maria Edna de Melo (SP) THYROID

Célia Regina Nogueira (SP)

Victória Borba (PR)

International Editorial Commission Andrea Giustina (Itália)

Gil Guerra-Júnior (SP)

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)

SBEM – Brazilian Society of Endocrinology and Metabolism SBEM BRAZILIAN BOARD OF DIRECTORS 2017-2018 President Vice-President Executive Secretary Adjunct Executive Secretary Treasurer-General Adjunct Treasurer

Fábio Rogério Trujilho Alexandre Hohl Paulo Augusto Carvalho de Miranda Neuton Dornelas Gomes Rodrigo de Oliveira Moreira Marcio Corrêa Mancini

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 - 2017/2018 Brazilian Society of Endocrinology and Metabolism ADRENAL AND HYPERTENSION

DIABETES MELLITUS

President Madson Queiroz de Almeida madsonalmeida@usp.br

President Luiz Alberto Andreotti Turatti turatti@uol.com.br Directors Amely Pereira Silva Balthazar Gustavo José Caldas Pinto Costa Sergio Alberto Cunha Vêncio Walter José Minicucci Thaísa Dourado Guedes Treasurer João Eduardo Nunes Salles Alternates Marcos Cauduro Troian Victor Gervásio e Silva

Vice-President Directors

Marivânia da Costa Santos Alexis Dourado Guedes Flávia Amanda Costa Barbosa Milena Coelho Fernandes Caldato Sonir Roberto Rauber Antonini Tânia Aparecida Sanchez Bachega

DYSLIPIDEMIA AND ATHEROSCLEROSIS

BASIC ENDOCRINOLOGY

President Cynthia Melissa Valério cy_valerio@yahoo.com.br

President Maria Izabel Chiamolera mchiamolera@unifesp.br Vice-President Bruno Ferraz de Souza Directors Catarina Segreti Porto Dóris Rosenthal Maria Tereza Nunes Marisa Maria Dreyer Breitenbach Tania Maria Ruffoni Ortiga Alternates Vânia Maria Corrêa da Costa Ubiratan Fabres Machado

Vice-President Directors

Renan Magalhães Montenegro Júnior Fernando de Mello Almada Giuffrida Marcello Casaccia Bertoluci

Scientific Departments - 2017/2018 WOMEN ENDOCRINOLOGY AND ANDROLOGY President Rita de Cássia Viana Vasconcellos Weiss rcvweiss@gmail.com Vice-President Directors Alternates

Dolores Perovano Pardini Amanda Valéria Luna de Athayde Mônica de Oliveira Poli Mara Spritzer Ricardo Martins da Rocha Meirelles Ruth Clapauch Izydorczyk Antônio Mendes Fontanelli Larissa Garcia Gomes

PEDIATRIC ENDOCRINOLOGY President Julienne Angela Ramires de Carvalho julienne@endocrinoped.com.br Vice-President Directors Alternate

Carlos Alberto Longui Aline da Mota Rocha Angela Maria Spinola e Castro Cláudia Braga Monteiro Paulo César Alves da Silva Suzana Nesi França Marilia Martins Guimarães

BONE AND MINERAL METABOLISM

NEUROENDOCRINOLOGY

President Vice-President Directors Alternate

President Marcello D. Bronstein mdbronstein@uol.com.br Vice-President César Luiz Boguszewski Directors Heraldo Mendes Garmes Luciana Ansanelli Naves Lucio Vilar Rabelo Filho Luiz Antonio de Araujo Mônica Roberto Gadelha Alternates Andrea Glezer Manoel Ricardo Alves Martins

Carolina Aguiar Moreira carolina.aguiar.moreira@gmail.com Miguel Madeira Barbara Campolina Carvalho Silva Francisco Alfredo Bandeira e Farias Marise Lazaretti Castro Sergio Setsuo Maeda Victória Zeghbi Cochenski Borba Tatiana Munhoz da Rocha Lemos Costa

OBESITY

THYROID

President Maria Edna de Melo medna@usp.br Vice-President Rosana Bento Radominski Director/Secretary Walmir Ferreira Coutinho Director/Treasurer Erika Paniago Guedes Directors Cintia Cercato Leila Maria Batista Araujo Director/Treasurer Fabio Ferreira de Moura Alternate Jacqueline Rizzolli

President Célia Regina Nogueira nogueira@fmb.unesp.br Vice-President José Augusto Sgarbi Secretary Janete Maria Cerutti Directors Ana Luiza Silva Maia Laura Sterian Ward Patricia de Fátima dos Santos Teixeira Gisah Amaral de Carvalho Mario Vaisman Alternate Danilo Glauco Pereira Villagelin Neto

Permanent Commissions - 2017/2018 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

César Luiz Boguszewski

President Erika Bezerra Parente ebparente@gmail.com Members Érika Paniago Guedes, Teresa Arruti Rey

cesarluiz@hc.ufpr.br

Members

Ruy Lyra da Silva Filho, Valéria Cunha C. Guimarães, Ana Cláudia Latrônico

SCIENTIFIC COMISSION President Alexandre Hohl alexandrehohl@uol.com.br Indicated by the directories Joao Eduardo Nunes Salles, Alexis Dourado Guedes, Erika Bezerra Parente, Ana Mayra Andrade de Oliveira, Margaret Cristina da Silva Boguszewski, Guilherme Alcides Flores Rollin, Milena Coelho Fernandes Caldato, Mônica de Oliveira, Nina Rosa de Castro Musolino

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

CONTINUOUS MEDICAL EDUCATION President Alexandre Hohl alexandrehohl@uol.com.br Members Lireda Meneses Silva, Walter José Minicucci, Cleber Favaro

STATUTES, RULES AND REGULATIONS President Nina Rosa de Castro Musolino ninamusolino@gmail.com Members Airton Golbert, Henrique de Lacerda Suplicy, Luiz Henrique Maciel Griz Representative of the Evandro de Souza Portes brazilian directory

PROFESSIONAL ETHICS AND DEFENCE President Itairan da Silva Terres itairan.terres@gmail.com Vice-Inspector Maite Trojaner Salona Chimeno 1ST member Diana Viegas Martins 2ND member João Modesto Filho 3RD member Evandro de Souza Portes 4TH member Marcelo Henrique da Silva Canto Costa 5TH member Luiz Henrique Santos Canani

ENDOCRINE DYSREGULATORS President Elaine Frade Costa elainefradecosta@gmail.com Vice-President Margaret Cristina da Silva Boguszewski Members Tania Aparecida Sanchez Bachega, Ricardo Martins da Rocha Meirelles, Marilia Martins Guimarães, Eveline Gadelha Pereira Fontenele, Maria Izabel Chiamolera

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 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: Yuri Galeno Pinheiro Chaves de Freitas yurigaleno@gmail.com Members: Fábio Ferreira de Moura, Clayton Luiz Dornelles Macedo, Roberto Luís Zagury, Ricardo de Andrade Oliveira, Fulvio Clemo Thomazelli, Felipe Henning Duarte Gaia

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 Joaquim Custodio da Silva Junior jocsjunior@uol.com.br Members Joaquim Custodio da Silva Junior, Eduardo Quadros Araújo, Marcelo Fernando Ronsoni, Manoel Ricardo Alves Martins, Marcio Weissheimer Lauria

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 secretaria@diabetes.org.br www.diabetes.org.br Administrative Manager: Anna Maria Ferreira

ABESO – BRAZILIAN ASSOCIATION FOR THE STUDY OF OBESITY AND METABOLIC SYNDROME ABESO BRAZILIAN BOARD OF DIRECTORS (2017-2018)

President

Maria Edna de Melo

Vice-President

Alexander Koglin Benchimol

1ST Secretary General

Bruno Halpern

2ND Secretary General

Fábio Ferreira de Moura

Treasurer

Erika Paniago Guedes

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

editorial

The elusive clinical significance of osteocalcin actions in energy metabolism in human Bruno Ferraz-de-Souza1

Arch Endocrinol Metab. 2018;62/3

Laboratório de Endocrinologia Celular e Molecular LIM-25 e Unidade de Doenças Osteometabólicas, Divisão de Endocrinologia, Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HCFMUSP), São Paulo, SP, Brasil 1

Correspondence to: Bruno Ferraz-de-Souza Av. Dr Arnaldo, 455, sala 4344 (LIM-25) 01246-903 – São Paulo,SP, Brasil bruno.ferraz@hc.fm.usp.br Received on June/4/2018 Accepted on June/6/2018 DOI: 10.20945/2359-3997000000050

n spite of its stagnant gross appearance, the skeleton is a highly metabolically active organ. The continuous process of bone remodeling, allowing homeostatic availability of calcium and renovation of the important structural material of bone, requires ample energy supply (1). Extreme examples of such energy demand had been noticed a long time ago in high bone turnover states, for instance in the cachexia accompanying severe hyperparathyroidism or in slender children with osteogenesis imperfecta, but lacked an explanation. Therefore, when the first mechanistic pieces of evidence linking bone and energy metabolism in animal models started arising (2), it not only made sense in physiological terms, but it was also felt as if by identifying these missing links we would potentially be able to tackle simultaneously two major health care concerns, adiposity and osteoporosis. It has now been more than 10 years since the paramount work of Lee and cols. placed osteocalcin (Oc) as the foremost bone-derived hormone to influence energy metabolism (3), rocketing Oc from its role as a (perhaps second rate) bone formation marker to a means by which bone, as an endocrine gland, would direct insulin sensitivity and production, and fat accumulation. From then onwards, subsequent evidence have solidified that, in mice, Oc favors glucose tolerance and insulin sensitivity (4,5). Its role as an energy metabolism effector in humans, however, remains elusive. Osteocalcin is produced by osteoblasts and secreted in a fully carboxylated form into the bone matrix, where it acts to regulate bone mineralization. Under acidic pH in the osteclastic resorption lacuna, it suffers decarboxylation generating undercarboxylated Oc – importantly, this is the form of Oc that has been chiefly linked to a metabolic role, but its determination in serum is not straightforward in clinical or even in research settings (5). Several studies in humans have investigated the relationship of total serum Oc and a variety of metabolic parameters, and results have been conflicting (6). Indeed, an effect, when seen, was considerably smaller in magnitude to what was observed in mice. Two articles in this issue of the Archives of Endocrinology and Metabolism (AE&M) have aimed to shed light on this topic through different approaches. Campos and cols. have analyzed clinical, laboratory and anthropometric and body composition parameters in a cohort of 34 obese adolescents submitted to a one-year interdisciplinary intervention including exercise, diet and psychological support (7). As expected, the intervention resulted in a reduction in body mass and fat was seen, accompanied by a significant increase in adiponectin and in lean mass. Even though an increase in bone mineral content was observed after the 1-year intervention, the expected progression of bone accretion during this period of life precludes ascertaining a positive effect of exercise on bone formation. The authors focused their analysis on the relation

271

Osteocalcin and human energy metabolism

points towards confusion in this matter, and it is unlikely that further association studies will help elucidate it. In fact, osteoporosis treatment with antiresorptive agents for the past decades has effectively manipulated levels of Oc in hundreds of thousands of individuals, and a striking deleterious effect in energy metabolism has not been seen. As human Oc is yet unavailable as a drug to be tested in randomized controlled intervention trials, the question remains whether, in humans, circulating osteocalcin is physiologically relevant as a marker or an effector of metabolic health. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. de Paula FJ, Rosen CJ. Bone Remodeling and Energy Metabolism: New Perspectives. Bone Res. 2013;1(1):72-84. 2. Karsenty G. The complexities of skeletal biology. Nature. 2003;423(6937):316-8. 3. Lee NK, Sowa H, Hinoi E, Ferron M, Ahn JD, Confavreux C, et al. Endocrine regulation of energy metabolism by the skeleton. Cell. 2007;130(3):456-69. 4. Zoch ML, Clemens TL, Riddle RC. New insights into the biology of osteocalcin. Bone. 2016;82:42-9. 5. Oldknow KJ, MacRae VE, Farquharson C. Endocrine role of bone: recent and emerging perspectives beyond osteocalcin. J Endocrinol. 2015;225(1):R1-19. 6. Levinger I, Brennan-Speranza TC, Zulli A, Parker L, Lin X, Lewis JR, et al. Multifaceted interaction of bone, muscle, lifestyle interventions and metabolic and cardiovascular disease: role of osteocalcin. Osteoporos Int. 2017;28(8):2265-73. 7. Campos RMS, Masquio DCL, Corgosinho FC, Carvalho-Ferreira JP, Molin Netto BD, Clemente APG, et al. Relationship between adiponectin and leptin on osteocalcin in obese adolescents during weight loss therapy. Arch Endocrinol Metab. 2018;62(3):275-84. 8. Barroso LN, Farias DR, Soares-Mota M, Bettiol H, Barbieri MA, Foss MC, et al. Waist circumference is an effect modifier of the association between bone mineral density and glucose metabolism. Arch Endocrinol Metab. 2018;62(3):285-95.

between adipokines and bone turnover markers, finding a positive correlation between the adiponectin/ leptin ratio (a marker of adipose tissue dysfunction and insulin resistance) and osteocalcin at baseline. While it is tempting to suggest that the metabolic improvement in insulin resistance and adiposity that resulted from their concerted intervention efforts led to bone formation, it should be noted that at the end of the intervention period, the observed (and expected) increase in the adiponectin/leptin ratio was not accompanied by significant changes in Oc. Barroso and cols. report a cross-sectional analysis of a large longitudinal birth cohort in Ribeirao Preto consisting of 468 young adults, evaluating bone and metabolic parameters (8). The mean age of subjects was 23 years, meaning that peak bone mass might had not been attained by most participants and that the physiologic scale could still be tipped in favor of bone formation. The majority of the cohort had normal body mass index and waist circumference, and moderate to high physical activity, showing that these were mainly metabolically healthy subjects. While positive findings included a correlation between bone mineral density and HOMA-estimated glucose metabolism, that was modified by visceral adiposity as assessed by waist circumference, no significant association between osteocalcin and glucose metabolism was seen. Collectively, these two papers exemplify the frustrating quest for the role of osteocalcin in human energy metabolism. Notably, both papers analyzed total osteocalcin, demonstrating the persisting difficulties of regularly studying undercarboxylated Oc; it could be speculated that results might have been different if the more “metabolically active” form of Oc had been analyzed. However, the bulk of the literature by now

272

Arch Endocrinol Metab. 2018;62/3

editorial

How deep is our anxiety during treatment of thyroid cancer? Flavio Carneiro Hojaij1

Arch Endocrinol Metab. 2018;62/3

Departamento de Cirurgia da Faculdade de Medicina da Universidade de São Paulo (FMUSP). Disciplina de Topografia Estrutural Humana da FMUSP. Laboratórios de Investigação Médica (LIM/02) do Hospital das Clínicas da FMUSP

Correspondence to: Flavio Carneiro Hojaij fchojaij@uol.com.br Received on June/4/2018 Accepted on June/7/2018 DOI: 10.20945/2359-3997000000051

he sentinel lymph node procedure is a technique that aims, with minimal surgical invasion, to diagnose lymph node metastasis and, later, to define the prescribed treatment: full lymph node dissections, radiotherapy, chemotherapy and others. It requires tools such as lymphographies, scintigraphies and surgical approaches followed by anatomopathological and immunohistochemical analyses (1). Its use began in the 1960s in facial epidermoid carcinomas. In the 1970s, it was incorporated in the treatment of penile carcinomas and, in the last 20 years, it has been used in the treatment of cutaneous melanomas and malignant neoplasms in the mammary exocrine glandular tissue (2). In Head and Neck Surgery, the sentinel lymph node procedure has gained strength in the last decade, mainly for the treatment of oral cavity tumors; but, due to the learning curve and the complicated logistics, it has not become popular in our midst (1). The study conducted by Steck and cols., present in this issue of Archives of Endocrinology and Metabolism (AE&M) (3), poses legitimate questions about the comprehensiveness of the treatment of well differentiated thyroid carcinoma. The authors state that the treatment should vary according to its need. Should there be a metastasis in a lymph node, the best conduct, according to the authors, is a therapeutic neck dissection. However, this work arrived in the middle of a hurricane, which is how I’ve been calling the present scenario. We are experiencing a “hurricane” of indagations, directives and controversial information about the aggressivity of well differentiated thyroid carcinomas. The authors of the article even express their anguish about this in their Introduction and Discussion (3,4). Certainly, the tumors we treat nowadays are smaller than the ones we have treated 20 years ago. These new features are directly influenced by an increase in the number of diagnostics, which is the result of better exams being performed and in a greater amount (4-6). Faced with so much data and those clinical characteristics, we, the ones who treat the patients, live in an atmosphere of bipolarity. In one hand, we have the active observation suggested for thyroid neoplasms (4,5) but, on the other hand, another authors (3,6-8) stress the danger of lymph node metastases and give us tools with which to handle with them. Due to this duality, some questions must be raised: 1. Do we really need to be concerned about lymph node metastases that aren’t detected neither in the preoperative ultrasound nor by the surgeon during the procedure?

273

Thyroid cancer treatment

2. Does the detection of micrometastases (submillimetric) in millimetric lymph nodes play a clinical role and an impact on the survival of patients? 3. Should the sentinel lymph node procedure, which has already been validated, follow the same precepts and logistics established for other tumors in the case of malignant thyroid neoplasms? 4. If so, can it be put into practice? The answers to these questions will surely be controversial and, considering the polarization we face in oncologic thyroidology, I think the moment demands calm. We should adapt extreme conducts to intermediary positions, opting for partial surgeries instead of limiting ourselves to two radically opposite postures: mere observation or comprhensive surgeries. We should also individualize treatment, so that its comprehensiveness will be adequate to the patients’ background, culture and expectations. And, finally, we should wait for the development of molecular tools capable of guiding and further helping the daily clinical practice (8,9).

1. Paleri V, Rees G, Arullendran P, Shoaib T, Krishman S. Sentinel node biopsy in squamous cell cancer of the oral cavity and oral pharynx: a diagnostic meta-analysis. Head Neck. 2005;27(9):739-47. 2. Morton DL, Wen DR, Wong JH, Economou JS, Cagle LA, Storm FK, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg. 1992;127(4):392-9. 3. Steck JH, Stabenow E, Bettoni GB, Steck S, Cernea CR. Accuracy of sentinel lymph node mapping in detecting occult neck metastasis in papillary thyroid carcinoma. Arch Endocrinol Metab. 2018;62(3):296-302. 4. Tuttle RM, Fagin JA, Minkowitz G, Wong RJ, Roman B, Patel S, et al. Natural History and Tumor Volume Kinetics of Papillary Thyroid Cancers During Active Surveillance. JAMA Otolaryngol Head Neck Surg. 2017;143(10):1015-20. 5. Ito Y, Miyauchia A, Oda H. Low-risk papillary microcarcinoma of the thyroid: A review of active surveillance trials. Eur J Surg Oncol. 2018;44(3):307-15. 6. Haymart MR, Esfandiari NH, Stang MT, Sosa JA. Controversies in the management of low-risk differentiated thyroid cancer. Endocr Rev. 2017;38(4):351-78. 7. Zaydfudim V, Feurer ID, Griffin MR, Phay JE. The impact of lymph node involvement on survival in patients with papillary and follicular thyroid carcinoma. Surgery. 2008;144(6):1070-7. 8. de la Fouchardière C, Decaussin-Petrucci M, Berthiller J, Descotes F, Lopez J, Lifante JC, et al. Predictive factors of outcome in poorly differentiated thyroid carcinomas. Eur J Cancer. 2018;92:40-7. 9. 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.

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

REFERENCES

274

Arch Endocrinol Metab. 2018;62/3

original article

Relationship between adiponectin and leptin on osteocalcin in obese adolescents during weight loss therapy Raquel Munhoz da Silveira Campos1, Deborah Cristina Landi Masquio2, Flávia Campos Corgosinho3, Joana Pereira de Carvalho-Ferreira4, Bárbara Dal Molin Netto5, Ana Paula Grotti Clemente6, Lian Tock7, Sergio Tufik8, Marco Túlio de Mello5,8,9, Ana Raimunda Dâmaso5

ABSTRACT Objectives: Obesity is a multifactorial disease characterized by the presence of the pro-inflammatory state associated with the development of many comorbidities, including bone turnover marker alterations. This study aimed to investigate the role of the inflammatory state on bone turnover markers in obese adolescents undergoing interdisciplinary weight loss treatment for one year. Subjects and methods: Thirty four post-pubescent obese adolescents with primary obesity, a body mass index (BMI) greater than > 95th percentile of the CDC reference growth charts, participated in the present investigation. Measurements of body composition, bone turnover markers, inflammatory biomarkers and visceral and subcutaneous fat were taken. Adolescents were submitted to one year of interdisciplinary treatment (clinical approach, physical exercise, physiotherapy intervention, nutritional and psychological counseling). Results: Reduction in body mass, body fat mass, visceral and subcutaneous fat, as well as, an increase in the body lean mass and bone mineral content was observed. An improvement in inflammatory markers was seen with an increase in adiponectin, adiponectin/leptin ratio and inteleukin-15. Moreover, a positive correlation between the adiponectin/ leptin ratio and osteocalcin was demonstrated. Further, both lean and body fat mass were predictors of osteocalcin. Negative associations between leptin with osteocalcin, adiponectin with Beta CTXcollagen, and visceral fat with adiponectin were observed. Conclusions: It is possible to conclude that the inflammatory state can negatively influence the bone turnover markers in obese adolescents. In addition, the interdisciplinary weight loss treatment improved the inflammatory state and body composition in obese adolescents. Therefore, the present findings should be considered in clinical practice. Arch Endocrinol Metab. 2018;62(3):275-84 Keywords Inflammation; bone turnover markers; obesity; adolescents; weight loss

1 Departamento de Fisioterapia, Laboratório de Recursos Terapêuticos, Universidade Federal de São Carlos (UFSCar), São Carlos, SP, Brasil 2 Centro Universitário São Camilo, São Paulo, SP, Brasil 3 Universidade Federal de Goiás (UFG), Goiânia, GO, Brasil 4 Programa de Pós-Graduação Interdisciplinar em Ciências da Saúde, Universidade Federal de São Paulo (Unifesp), Santos, SP, Brasil 5 Programa de Pós-Graduação em Nutrição, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil 6 Universidade Federal de Alagoas (Ufal), Maceió, AL, Brasil 7 Weight Science, São Paulo, SP, Brasil 8 Departamento de Psicobiologia, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil 9 Escola de Educação Física, Fisioterapia e Terapia Ocupacional, Universidade Federal de Minas Gerais (UFMG), Belo Horizonte, MG, Brasil

Correspondence to: Raquel Munhoz da Silveira Campos Laboratório de Recursos Terapêuticos, Departamento de Fisioterapia, Universidade Federal de São Carlos Rodovia Washington Luís, km 235 13565-905 – São Carlos, SP, Brasil raquelmunhoz@hotmail.com Received on Mar/30/2017 Accepted on Dec/20/2017

INTRODUCTION

besity is a multifactorial disease associated with a pro-inflammatory state including a lower adiponectinemia and hyperleptinemia framework. Many comorbidities are associated with obesity, including bone mineral alterations, metabolic syndrome, cardiovascular complications, non-alchoolic fatty liver disease, sleep disorders and asthma (1,2). Concomitant with this, body fat distributions, especially visceral adipocytes, are linked to the secretion of pro-inflammatory adipokines that can act negatively on bone metabolism (3). In a previous study, it was shown that visceral fat, as well as the visceral/ Arch Endocrinol Metab. 2018;62/3

subcutaneous ratio, were independent, negative predictors of bone mineral density (BMD) (4). Beta-CTx collagen and osteocalcin are bone metabolism biomarkers that may represent bone turnover. Beta-CTx collagen is the C-terminal telopeptide of type I collagen, the main component (approximately 90%) of the protein matrix of bone. Beta-CTx collagen is released into the bloodstream during bone resorption and is almost entirely excreted by the kidneys. Its quantification serves as a specific marker for the degradation of mature type I collagen from bone (5). Osteocalcin is the major noncollagenous protein that acts locally in the bone mineralization. It is 275

DOI: 10.20945/2359-3997000000039

Inflammation, obesity and osteocalcin

synthesized by the osteoblasts and has been utilized as a marker of bone formation or bone turnover (6). Leptin, an adipokine that is primarily expressed by adipose tissue, is considered to be involved in neuroendocrine control of energy balance. However, in human obesity, hyperleptinemia was associated with a reduction of bone formation biomarkers, especially osteocalcin (7). On the other hand, adiponectin seems to improve bone formation (8), promoting the proliferation, differentiation, and mineralization of osteoblastic cells (9). In addition, serum osteocalcin levels were significantly associated with plasma adiponectin levels and inversely related to leptin levels, in the presence of metabolic syndrome (10). The adiponectin/leptin ratio is associated as a better inflammatory biomarker of inflammation in metabolic syndrome patients than these adipokines analyzed in isolation (11). However, the relationship between the adiponectin/leptin ratio and bone biomarkers, mostly considering its role in osteocalcin in obese adolescents, has yet to be explored. Moreover, considering obesity to be a multifactorial disease, clinical strategies to promote weight loss, associated with physical exercise, nutrition and psychological interventions can be an interesting approach to promote weight loss and health benefits (12). In this way, the aim of the present investigation is to analyze the effects of interdisciplinary weight loss therapy on pro/anti-inflammatory adipokines and their role in bone turnover markers in obese adolescents.

SUBJECTS AND METHODS

Population For this study, it was involved 34 post-puberty obese adolescents of both genders, with age of 15-19 years. Inclusion criteria were Tanner stage five (13), primary obesity, body mass index BMI > 95th percentile of the CDC reference growth charts (14). Non-inclusion criteria were the use of birth control pills, cortisone, anti-epileptic drugs, history of renal disease, alcohol intake, smoking and secondary obesity due endocrine disorders. There were no obese adolescents with diagnoses of ferritin alteration, autoimmune diseases and virus of Hepatitis A, B and C. The main reasons for dropping out (n = 4) in our study were financial and family problems, followed by school and job opportunities. No sex differences were observed in adherence rates. The study was conducted 276

with the principles of the Declaration of Helsinki and was approved by the ethics committee on research at the Universidade Federal de São Paulo (Unifesp) (0135/04; 152.281), Clinical Trial: NCT01358773. All procedures were clear to those responsible for the volunteers and it was obtained consent for research. All evaluations were performed at two different times (baseline and after interdisciplinary intervention).

Anthropometric measurements Weight was measured by plethysmography scale (BODPOD equipment), where patients wore minimum clothing possible and height was measured using a stadiometer (Sanny – model ES 2030). After obtaining the data was calculated using the body mass index (BMI) by dividing the weight by height squared (kg/m²). Body composition, including body fat mass (percentage and kilograms) and body lean mass (percentage and kilograms), was obtained through air displacement pletismography (BODPOD).

Bone mineral density (BMD) and bone mineral content (BMC) A whole-body DXA absorptiometry scan was performed per unit of bone densitometry to determine the whole-body of bone mineral density-BMD (g/cm²) and bone mineral content-BMC (grams) using a Lunar Prodigy Advance System (GE Healthcare). The whole-body scan required the subjects to be placed in a supine position with their arms and legs positioned according to the manufacturers’ specifications (15). Quality control was performed daily using a phantom, and measurements were maintained within the manufactures standards of ≤ 1%. In order to obtain statistically precise measurements, 68% of the exams were repeated within a coefficient of variation of 1DP (± 0.010 g/cm³ for total body size).

Serum analysis Blood samples were collected at the outpatient clinic at approximately 8:00 A.M. after an overnight fast (12 hours). The adipokines concentrations (adiponectin, leptin and interleukin-15) were measured using a commercially available multiplex assay (EMD Millipore: HMHMAG-34K; HCYTOMAG-60K). Manufacturesupplied controls were included to measure assay variation and all samples were analyzed on the same day to minimize day-to-day variation. A minimum of 100 Arch Endocrinol Metab. 2018;62/3

Inflammation, obesity and osteocalcin

Visceral and subcutaneous adiposity measurements The abdominal ultrasonography procedures and the measurements of visceral and subcutaneous fat tissue and fatty liver were performed by the same physician, who was blinded to the subject assignment groups at the baseline time-point and following 1 year therapy. This physician was a specialist in imaging diagnostics. A 3.5-MHz multifrequency transducer (broad band) was used to reduce the risk of misclassification. The intraexamination coefficient of the variation for ultrasound (US) was 0.8%. US measurements of intra-abdominal (visceral) and subcutaneous fat were obtained. US determined subcutaneous fat was defined as the distance between the skin and external face of the rectus abdominal muscle, and visceral fat was defined as the distance between the internal face of the same muscle and the anterior wall of the aorta. The cut-off points for the definition of visceral obesity by ultrasonography were based on the previous methodological descriptions made by Ribeiro-Filho and cols. (18).

Descriptive methodology of interdisciplinary weight loss therapy All sessions were conducted by an interdisciplinary group of health professionals. Three times of week during two hours per day not consecutive, the adolescents participated of supervised therapy in physical exercise, nutrition and psychological attendances during one year. Once each month the adolescents were followed by an endocrinologist (Figure 1).

Clinical intervention All obese adolescents visited the endocrinologist with their parents once each month. In all of these visits, the entire GEO team (Study Group of Obesity) was also present. The doctor monitored and evaluated all clinical exams of adolescents and treated health problems Arch Endocrinol Metab. 2018;62/3

during therapy. The medical follow-up included the initial medical history and a physical examination of blood pressure, cardiac frequency and body mass, and the adolescents were checked for their adherence to all interdisciplinary therapies. The team discussed with the adolescents and their parents some possible changes in lifestyle to promote their health (Figure 1).

Physical exercise intervention: aerobic plus resistance training (AT + RT) During the 1-year therapy period, the adolescents followed a combined exercise training therapy. The protocol was performed three times per week for 1 year and included 30 min of aerobic training plus 30 min of resistance training per session. The subjects were instructed to reverse the order of the exercises (aerobic and resistance) at each training session. The aerobic training consisted of running on a motordriven treadmill (Life Fitness – model TR 9700HR) or bicycle at a cardiac frequency intensity representing the ventilatory threshold I (±4 bpm), which was determined by the results of an initial oxygen uptake test for aerobic exercises (ergoespirometry). The exercise therapy was based on the guidelines from the American College of Sports Medicine (ACSM). Resistance training was also designed based on ACSM recommendations (19,20) (Figure 1).

Nutrition intervention Energy intake was set at the levels recommended by the dietary reference intake for subjects with low levels of physical activity and of the same age and gender following a balanced diet (21). No pharmacotherapies or antioxidants were recommended. Once a week, adolescents had dietetics lessons educating participants on the food pyramid, diet record assessment, weight loss diets and fad diets, food labels, dietetics, fatfree and low-calorie foods and other related topics. Monthly, they had individual consultations (Figure 1).

Physiotherapy intervention The adolescents were accompanied by a physiotherapist during the therapy, in order to prevent musculoskeletal injuries. Once a week, the volunteers had lessons regarding such topics as the postural orientation, prevention of musculoskeletal injuries, diaphragmatic breathing, hydrotherapy, isostretching, and balance. 277

beads were collected for each analyzed using a Luminex MagPix System (Austin, Texas), which was calibrated and verified prior to sample analysis. Unknown sample values were calculated offline using Milliplex Analyst Software (EMD Millipore) (16). Serum osteocalcin and Beta CTx-collagen levels were obtained using the ECLIA (Electrochemiluminescent immune Assay). For the leptin hormone, the following values were adopted: males, between 1 and 20 ng/mL and females, between 4.9 and 24 ng/mL previously described by Gutin and cols. (17).

Inflammation, obesity and osteocalcin

Ethical Committee 0135/04; 152.281

Media

Pubertal evaluation ≥ Tanner stage 5

BMI ≥ 95th percentile

15 to 19 years

Exercise electrocardiography (ECG)

34 volunteers selected Blood collect

Body composition plethysmography

Abdominal ultrasonography

Densitometry

Ergospirometry

30 volunteers concluded the treatment 4 dropout (12%)

Interdisciplinary weight loss therapy

Psychological Intervention Once a week 1 hour

Nutritional Intervention Once a week 1 hour

Physical Exercise 3x week 1 hour

Clinical Intervention Once a month

1 year

Physiotherapy Intervention Once a week 1 hour

Figure 1. Descriptive methodology of study.

Psychological intervention

Psychological therapy treatment plans were established on the basis of validated questionnaires that considered some of the psychological problems caused by obesity, as described in the literature. These include depression, eating disorders, anxiety, decreased self-esteem and body image disorders. Interdisciplinary therapy consisted of a weekly 1h group session. Individualized psychological therapy was recommended when it was necessary according psychological assessment (Figure 1).

were expressed as mean ± SD, and non-parametric data were expressed as median, minimum and maximum values. The effects of interdisciplinary therapy during 1 year were analyzed by t test dependent by samples. For the non-parametric data the Wilcoxon test was applied. Correlations were established through the Pearson test for parametric data and Spearman for nonparametric data. Finally, it was verified the dependencies of variables by simple linear regression. The correlations and regression test were analyzed with baseline and final values.

Statistical analysis Statistical analysis was performed using the program STATISTICA version 7.0 for Windows Vista. The adopted significant value was α ≤ 5%. Data normality was verified with the Shapiro Wilk test. Parametric data 278

RESULTS The present study was composed by 34 obese adolescents: 18 girls (11 ± 1age of menarche) and 16 boys with 16 ± 1 years old, primary obesity (diagnosis Arch Endocrinol Metab. 2018;62/3

Inflammation, obesity and osteocalcin

of obesity around 10 ± 3 years old), body mass 94.5 ± 15.45 (kg) and body mass index 33.21 ± 3 (kg/m²).

Effects of interdisciplinary weight loss therapy One year of interdisciplinary weight loss therapy demonstrated significantly reduction in body mass (kg), BMI (kg/m²), body fat mass (kg and %), visceral fat (cm), subcutaneous fat (cm) and visceral/subcutaneous ratio. Increases in the values of body lean mass (%), bone mineral content (g), adiponectin (µg/L), adiponectin/ leptin ratio, and interleukin-15 (IL-15) (pg/mL) were observed (Tables 1 and 2).

Correlations analysis

demonstrated between bone mineral content (g) with body lean mass (% and kg), visceral fat (cm) with body mass index (kg/m²) and body fat mass (kg), leptin (ng/ mL) with body mass index (kg/m²) and body fat mass (kg) (Table 3).

Final values Negative correlations were showed between bone mineral content (g) with body fat mass (%); osteocalcin (ng/mL) with leptin (ng/mL); and adiponectin/leptin ratio with subcutaneous fat (cm). Positive correlations were demonstrated between bone mineral content (g) with body lean mass (% and kg) (Table 3).

Baseline values

Regression analysis

Negative correlations were showed between bone mineral content (g) with body fat mass (%), visceral fat (cm) with adiponectin (ng/mL) and Beta CTX-collagen (ng/ml) with adiponectin (ng/mL). Positive correlations were

As shown in the Figure 2 leptin (β -0.41; p = 0.04), adiponectin/leptin ratio (β 0.79; p < 0.001), body fat mass (β -0.66; p < 0.001) and body lean mass (β 0.66; p < 0.001) were predictors for changes in osteocalcin concentration.

Table 1. Effects of interdisciplinary weight loss therapy in body composition of obese adolescents Variables Body mass (kg)

Baseline

1 year

p value

Δ value

94.50 ± 15.45

88.70 ± 14.62*

< 0.001

-5.27 ± 4.5

Height (m)

1.68 ± 0.10

1.69 ± 0.12

0.60

0 ± 0.002

Body mass index (kg/m²)

33.21 ± 3.09

31.52 ± 3.19*

< 0.001

-1.90 ± 1.61

Body fat mass (%)

40.40 ± 6.50

37.35 ± 7.03*

< 0.001

-3.31 ± 2.98

Body fat mass (kg)

38.19 ± 8.75

33.17 ± 8.32*

< 0.001

-4.98 ± 4.22

Body lean mass (%)

59.60 ± 6.50

62.65 ± 7.03*

< 0.001

3.32 ± 2.98

Body lean mass (kg)

56.38 ± 10.89

57.55 ± 10.84

0.36

1.17 ± 2.00

1.25 ± 0.08

1.26 ± 0.08

0.82

0 ± 0.02

3192.21 ± 465.89

3338.12 ± 547.06*

< 0.001

140.34 ± 121.74

4.46 ± 1.34

3.69 ± 1.20*

< 0.001

-0.66 ± 0.87

Bone mineral density (g/cm²) Bone mineral content (g) Visceral fat (cm) Subcutaneous fat (cm)

3.66 ± 0.79

3.40 ± 0.84*

0.01

-0.44 ± 0.54

Visceral/subcutaneous ratio

1.27 ± 0.45

0.29 ± 1.46*

0.001

-0.91 ± 1.20

* Statistical difference p ≤ 0.05. Effects of therapy: comparison between baseline and 1 year of therapy.

Table 2. Effects of interdisciplinary weight loss therapy in bone turnover markers and inflammation biomarkers of obese adolescents Beta-CTx collagen (ng/mL)

Baseline

1 year

p value

Δ value

0.77 ± 0.46

0.64 ± 0.18

0.26

-0.16 ± 0.54

Osteocalcin (ng/ml)

30.3 (19/57.20)

34.60 (23.10/56.20)

0.28

3.60 (-7.2/43.3)

Adiponectin (µg/L)

1.9 ± 1.06

3.2 ± 1.2*

< 0.001

1.3 ± 1.4

24.65 ± 13.12

24.66 ± 14.53

0.50

-1.16 ± 8.53

Leptin (ng/ml) Adiponectin/leptin ratio

0.09 ± 0.06

0.17 ± 0.16*

< 0.001

0.08 ± 0.13

IL-15 (pg/mL)

0.05 (0/1.23)

0.10 (0/1.28)*

0.04

0.06 (-0.98/0.81)

Variables

Beta-CTx collagen: C-terminal telopeptides of type I collagen; IL-15: interleukin-15. * Statistical difference p ≤ 0.05. Effects of therapy: comparison between baseline and 1 year of therapy. Arch Endocrinol Metab. 2018;62/3

279

Inflammation, obesity and osteocalcin

Table 3. Correlations analysis Variables

p value

Baselines values Bone mineral content (g) Body fat mass (%)

-0.48

0.012

Body lean mass (%)

0.48

0.012

Body lean mass (kg)

0.56

0.003

Visceral fat (cm) Body mass index (kg/mÂ˛)

0.64

0.0001

Body fat mass (kg)

0.61

0.001

Adiponectin (ng/mL)

-0.47

0.013

-0.45

0.03

Body mass index (kg/mÂ˛)

0.50

0.009

Body fat mass (kg)

0.46

0.016

Body fat mass (%)

-0.47

0.037

Body lean mass (%)

0.47

0.037

Body lean mass (kg)

0.51

0.020

-0.53

0.01

-0.52

0.020

Adiponectin (ng/mL) Beta CTX-collagen (ng/mL) Leptin (ng/mL)

Final values Bone mineral content (g)

Adiponectin/leptin (ng/mL) Subcutaneous fat (cm) Osteocalcin (ng/mL) Leptin (ng/mL)

DISCUSSION The first aim of the present investigation was to analyze the role of inflammatory biomarkers in bone turnover. The most important findings were the negative association with leptin concentration; and the positive correlation between the adiponectin/leptin ratio and osteocalcin. Therefore, we were able to confirm the hypothesis that pro/anti-inflammatory adipokine is a key mediator of bone turnover markers in obese adolescents. According to our understanding, our group was the first to demonstrate the positive correlation between the adiponectin/leptin ratio and osteocalcin in obese adolescents, showing the beneficial effect of an improved adiponectin/leptin ratio on bone metabolism. Osteocalcin is a biomarker only secreted by osteoblasts, which enables the recognition of cell turnover in the skeletal system (7). Leptin concentration has an important influence on bone metabolism. In an experimental study leptin-deficient ob/ob and leptin280

resistant db/db showed increased osteocalcin levels (22), indicating that leptin is an inhibitor of osteoblastic bone formation. Corroborating with this data, we were able to show a negative correlation between leptin and osteocalcin. In accordance, recently, investigations have shown that a reduction in leptin concentration correlated with an improvement of osteocalcin levels in obese adults (7). Additionally, weight loss promotes an increase in the adiponectin concentration, which plays an important role in bone mass. Another investigation showed that an increase in adiponectin concentration was associated with an improvement in the osteoblastic activity, suggesting an increase in osteocalcin concentration only in the experimental study. However, these authors suggest that adiponectin exerts an activity that increases bone mass by suppressing osteoclastogenesis and by activating osteoblastogenesis, indicating that adiponectin manipulation could be therapeutically beneficial for patients with osteopenia (8). It is relevant to observed that, in our results a positive correlation was demonstrated between the adiponectin/ leptin ratio and osteocalcin in obese adolescents, considering the baseline values. Additionally, at the end of treatment an increase in the adiponectin/leptin ratio without significant statistical changes in osteocalcin was observed. However, it is important to note a discreet change in the osteocalcin concentration was observed in the median baseline values compared with the final value (increase of 12%). In this way, it is possible to suggest that a positive change in the adiponectin/leptin ratio could promote benefits in osteocalcin concentration in obese adolescents. Moreover, a previous study considered the adiponectin/leptin ratio to be a better inflammatory marker for the metabolic syndrome population. This ratio has high sensitivity and specificity for the diagnosis of metabolic syndrome (11). In fact, in the present study adiponectin was negatively correlated with visceral fat; and adiponectin/ leptin correlated negatively with subcutaneous fat. It is likely that the impaired actions of adiponectin are clinically important in obese patients because adiponectin is the most abundant adipocyte-derived hormone with established anti-inflammatory effect (1). Additionally, in the present investigation, it was found that both lean and body fat mass are predictors of osteocalcin. Concomitant to this, negative and positive correlations were observed between bone mineral content and body fat and lean body mass, respectively. Arch Endocrinol Metab. 2018;62/3

Inflammation, obesity and osteocalcin

A 60

B 60

β -0.41; p = 0.04

40 30 Adiponectin/leptin ratio

40 Leptin (ng/mL)

β 0.79; p < 0.001

30 20 10

20 10 0 -10 -20 -30

0 15

35 40 45 Osteocalcin (ng/mL)

-40 15

C 60

β 0.66; p < 0.001

65 Body lean mass (%)

Body fat mass (%)

D 75 β -0.66; p < 0.001

45 40

60 55

25 10

35 40 45 Osteocalcin (ng/mL)

40 50 Osteocalcin (ng/mL)

40 10

40 50 Osteocalcin (ng/mL)

In a previous investigation, it was demonstrated that the serum level of total osteocalcin was positively associated with fat-free mass independent of age, fat mass, leptin, and other confounders in premenopausal women. The hypothesis suggested to explain this association is based on the fact that fat-free mass could indirectly reflect the mechanical load on the bone, which can further stimulate bone formation, whereas low skeletal muscle mass is considered a risk factor for low bone mineral density (23). Moreover, corroborating with our findings, an inverse association was observed between osteocalcin and body fat mass in men with type 2 diabetes and in the elderly population. It is suggested that body fat mass accumulation is closely linked to bone turnover. Adipocytes are responsible for secreting many pro/anti-inflammatory cytokines such as TNF-alpha, interleukin-6, interleukin-10, leptin and Arch Endocrinol Metab. 2018;62/3

adiponectin, which are capable of modulating bone metabolism (24,25). Notably, Beta-CTx collagen correlates negatively with adiponectin and positively with visceral fat. Adiponectin, an anti-inflammatory adipokine, could promote the acceleration of osteogenesis (26). BetaCTx collagen is the C-terminal telopeptide of type I collagen, the main component (approximately 90%) of the protein matrix of bone. Beta-CTx collagen is released into the bloodstream during bone resorption and is almost entirely excreted by the kidneys. Its quantification serves as a specific marker for the degradation of mature type I collagen from the bone (5). In addition, visceral fat promotes a secretion of many pro-inflammatory adipokines that are associated with an increase in the bone reabsorption and a decrease 281

Figure 2. Simple linear regression established between osteocalcin (ng/mL) with: A) leptin (ng/mL); B) adiponectin/leptin ratio; C) body fat mass (%) and D) body lean mass (%).

Inflammation, obesity and osteocalcin

in bone formation (26). In this regard, our results showed that the pro-inflammatory profile present in obese adolescents is associated with an increase in Beta CTX-collagen and weight loss. A reduction in visceral fat, specifically, is associated with an increase in the adiponectin concentration and a reduction in the bone reabsorption biomarker as shown in the correlation analysis. The second objective was to analyze the effects of interdisciplinary weight loss treatments on pro/antiinflammatory adipokines and their role in bone turnover markers in obese adolescents. Therefore, another important finding from the present investigation is that weight loss therapy promotes a significant reduction in visceral fat (cm), subcutaneous fat (cm), visceral/subcutaneous ratio, total body fat mass and an increase in the lean body mass enhanced by nutritional counseling and physical exercise training. These results are important since it is well-established in the literature that fat deposition in the visceral compartment is related to the development of some diseases (1,27). In fact, visceral fat plays a pathological role, due to the secretion of certain pro-inflammatory adipokines also related to the deterioration of bone mass and is associated with development of many comorbidities, such as metabolic syndrome, dyslipidemia, and cardiovascular complications. A prior study showed that an increase of 1 cm in visceral fat was associated with a 1.97 fold (95% CI 1.06â&#x20AC;&#x201C;3.66) in boys and 2.08 fold (95% CI 1.38â&#x20AC;&#x201C;3.13) in girls increased risk of developing nonalcoholic fatty liver disease (27). In addition to these results, an increase in the IL-15 concentration after weight loss therapy was observed. IL-15 is considered an important fat mass regulator. In a prior investigation, a negative association between plasma IL-15 and fat mass was found, independent of the diagnosis of type 2 diabetes, which suggests that IL15 may be involved in the regulation of body fat mass (28). In experimental research, it was found that the administration of IL-15 seems to decrease circulating triglycerides by decreasing both the liver lipogenic rate and very-low-density lipoprotein (VLDL). Moreover, IL-15 decreases lipoprotein lipase activity and the lipogenic rate in adipose tissue. Experimental studies showed a significant decrease in the intestinal lipid absorption, which may in part explain the antiobesity effects of IL-15. Finally, it has been shown that lipoaspirate-derived human adipocytes treated with IL-15 inhibited pre-adipocyte differentiation. 282

The mechanism of IL-15 signaling in adipocytes however, is currently unknown (29). Together, these findings suggest that IL-15 treatment could result in weight loss and decreased visceral fat, corroborating with the control of the inflammatory state related to obesity (30). Corroborating with our findings, Brunelli and cols. (31), showed that 24 weeks of moderate-highintensity combined training (including aerobic and resistance exercises), in obese middle-aged men promotes an increase in the IL-15, concomitant with an improvement in the adiponectin concentration; and a decrease in the body fat. Also, it is suggested that the increase in the fat free mass and decreased body fat mass, may have contributed to the in IL-15 concentration in humans has a relevant anti-inflammatory function in the metabolic profile and may enhance energy expenditure to protect the body from obesity and type 2 diabetes (32). Additionally, an increase in adiponectin concentration and adiponectin/leptin ratio was observed after interdisciplinary weight loss therapy. Adiponectin is a potent anti-inflammatory adipokine that possesses multiple beneficial effects on obesity-related medical complications (33). It may also have anti-atherogenic and anti-inflammatory properties, and high levels of circulating adiponectin have been related to a lower risk of coronary heart disease (34). However, no statistical difference was observed in the leptin concentration, probably explained because at baseline the adolescents showed a normal leptin concentration as recommended for this population (22). Although, it is relevant to note that hyperleptinemia is consider an important condition that contribute to the pro-inflammatory state observed in obesity population, including adolescents, and associated with comorbidities development (35). Considering bone health, studies demonstrated a negatively association between leptin and bone turnover biomarkers, especially with osteocalcin (7,10,36,37). Another important result was the increase in BMC. This finding is possibly associated with improvement in the inflammatory profile, a reduction in visceral fat and the benefits of physical exercise intervention combined with other therapies realized by volunteers during the development of the study. It has been previously demonstrated that physical exercise can improve osteogenesis, which consequently increases bone mineral density and content (38). Moreover, resistance physical exercise at moderate intensity is Arch Endocrinol Metab. 2018;62/3

Inflammation, obesity and osteocalcin

CONCLUSIONS AND FUTURES DIRECTIONS In the present study, we were able to show that both leptin and adiponectin/leptin ratio were negatively and positively associated with osteocalcin, respectively, modulating bone turnover markers. Finally, the interdisciplinary weight loss treatments were seen to be effective at reducing body fat mass, visceral fat and at increasing lean body mass, bone mineral content, adiponectin and IL-15. Together, these results suggest that this kind of intervention is considered an interesting alternative to prevent and treat obesity and promote bone health. Acknowledgements: Support Foundation of São Paulo Research – Fapesp (2013/041364; 2013/19046-0; 2013/08522-6; 2015/14309-9), National Council for Scientific and Technological Development – CNPq (573587/2008-6; 300654/2013-8; 150177/2014-3) and Coordination of Higher Education Personnel Training – Capes. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Dâmaso AR, de Piano A, Campos RMS, Corgosinho FC, Siegfried W, Caranti DA, et al. Multidisciplinary approach to the treatment of obese adolescents: effects on cardiovascular risk factors, inflammatory profile, and neuroendocrine regulation of energy balance. Int J Endocrinol. 2013;2013:541032. 2. Campos RM, de Mello MT, Tock L, da Silva PL, Corgosinho FC, Carnier J, et al. Interaction of bone mineral density, adipokines and hormones in obese adolescents girls submitted in an interdisciplinary therapy. J Pediatr Endocrinol Metab. 2013;26(78):663-8. 3. Campos RM, Lazaretti-Castro M, Mello MT, Tock L, Silva PL, Corgosinho FC, et al. Influence of visceral and subcutaneous fat in Arch Endocrinol Metab. 2018;62/3

bone mineral density of obese adolescents. Arq Bras Endocrinol Metabol. 2012;56:12-8. 4. Lac G, Cavalie H, Ebal E, Michaux O. Effects of a high fat diet on bone of growing rats. Correlations between visceral fat, adiponectin and bone mass density. Lipids Health Dis. 2008;7:16. 5. Peichl P, Griesmacherb A, Marteau R, Hejc S, Kumpan W, Müller MM, et al. Serum crosslaps in comparison to serum osteocalcin and urinary bone resorption markers. Clin Biochem. 2001;34(2):131-9. 6. Alfadda AA, Masood A, Shaik SA, Dekhil H, Goran M. Association between osteocalcin, metabolic syndrome, and cardiovascular risk factors: role of total and undercarboxylated osteocalcin in patients with type 2 diabetes. Int J Endocrinol. 2013;2013:197519. 7. Suh HS, Hwang IC, Lee KS, Kim KK. Relationships between serum osteocalcin, leptin and the effect of weight loss by pharmacological treatment in healthy, nonsmoking Korean obese adults. Clin Chim Acta. 2013;418:17-21. 8. Oshima K, Nampei A, Matsuda M, Iwaki M, Fukuhara A, Hashimoto J, et al. Adiponectin increases bone mass by suppressing osteoclast and activating osteoblast. Biochem Biophys Res Commun. 2005;331(2):520-6. 9. Kanazawa I, Yamaguchi T, Yano S, Yamauchi M, Yamamoto M, Sugimoto T. Adiponectin and AMP kinase activator stimulate proliferation, differentiation, and mineralization of osteoblastic MC3T3-E1 cells. BMC Cell Biol. 2007;8:51. 10. Saleem U, Mosley TH Jr, Kullo IJ. Serum osteocalcin is associated with measures of insulin resistance, adipokine levels, and the presence of metabolic syndrome. Arterioscler Thromb Vasc Biol. 2010;30(7):1474-8. 11. Mirza S, Qu HQ, Li Q, Martinez PJ, Rentfro AR, McCormick JB, et al. Adiponectin/leptin ratio and metabolic syndrome in a Mexican American population. Clin Invest Med. 2011;34(5):E290. 12. Choi BC, Pak AW. Multidisciplinarity, interdisciplinarity and transdisciplinarity in health research, services, education and policy: 1. Definitions, objectives, and evidence of effectiveness. Clin Invest Med. 2006;29(6):351-64. 13. Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. Arch Dis Child. 1976;51(3):170-9. 14. Centers for Disease Control and Prevention. Hyattsville: National Center for Health Statistics. (Updates on 11 January 2007). Prevalence of overweight among children and adolescents: United States 1999-2002. Disponível em: http://www.cdc.gov/ nchs/products/pubs/pubd/hestats/overwght99.htm. Acesso em: 11 ago. 2013. 15. Black E, Petersen L, Kreutzer M, Toubro S, Sørensen TI, Pedersen O, et al. Fat mass measured by DXA varies with scan velocity. Obes Res. 2002;10(2):69-77. 16. Dossus L, Becker S, Achaintre D, Kaaks R, Rinaldi S. Validity of multiplex-based assays for cytokine measurements in serum and plasma from “non-diseased” subjects: comparison with ELISA. J Immunol Methods. 2009;350(1-2):125-32. 17. Gutin B, Ramsey L, Barbeau P, Cannady W, Ferguson M, Litaker M, et al. Plasma leptin concentrations in obese children: changes during 4-mo periods with and without physical training. Am J Clin Nutr. 1999;69(3):388-94. 18. Ribeiro-Filho FF, Faria AN, Azjen S, Zanella MT, Ferreira SR. Methods of estimation of visceral fat: advantages of ultrasonography. Obes Res. 2003;11(12):1488-94. 19. Donnelly JE, Blair SN, Jakicic JM, Manore MM, Rankin JW, Smith BK; American College of Sports Medicine. American College of Sports Medicine Position Stand. Appropriate physical activity intervention strategies for weight loss and prevention of weight regain for adults. Med Sci Sports Exerc. 2009;41(2):459-71.

283

related to decreased bone resorption markers (39). We know that aerobic physical exercises are associated with weight loss, but the combination of two kinds of physical exercise in the same session (aerobic plus resistance training) could optimize weight loss with greater benefits such as improvement in bone turnover markers and lean body mass as previously shown (2,40) as well as in the present study. Finally, we showed that interdisciplinary weight loss therapy improves pro/anti-inflammatory profile and was related to bone turnover. Together, our results suggest the importance of controlling the inflammatory state and the effects on bone turnover markers related to obesity in adolescence.

Inflammation, obesity and osteocalcin

20. Kraemer WJ, Ratamess NA, French DN. Resistance training for health and performance. Curr Sports Med Rep. 2002;1(3):165-71.

of adipose and muscle mass: a potential role in body weight control. Biochim Biophys Acta. 2001;1526(1):17-24.

21. National Academic Press. Dietary Reference Intake. Applications in Dietary Assessment. Washington, DC; 2001.

31. Brunelli DT, Chacon-Mikahil MP, Gáspari AF, Lopes WA, Bonganha V, Bonfante IL, et al. Combined training reduces subclinical inflammation in obese middle-age men. Med Sci Sports Exerc. 2015;47(10):2207-15.

22. Ducy P, Amling M, Takeda S, Priemel M, Schilling AF, Beil FT, et al. Leptin inhibits bone formation through a hypothalamic relay: a central control of bone mass. Cell. 2000;100(2):197-207. 23. Wu CH, Yang KC, Chang HH, Yen JF, Tsai KS, Huang KC. Sarcopenia is related to increased risk for low bone mineral density. J Clin Densitom. 2013;16(1):98-103. 24. Kanazawa I, Yamaguchi T, Yamauchi M, Yamamoto M, Kurioka S, Yano S, et al. Serum undercarboxylated osteocalcin was inversely associated with plasma glucose level and fat mass in type 2 diabetes mellitus. Osteoporos Int. 2011;22(1):187-94. 25. Kindblom JM, Ohlsson C, Ljunggren O, Karlsson MK, Tivesten A, Smith U, et al. Plasma osteocalcin is inversely related to fat mass and plasma glucose in elderly Swedish men. J Bone Miner Res. 2009;24(5):785-91. 26. Lee HW, Kim SY, Kim AY, Lee EJ, Choi JY, Kim JB. Adiponectin stimulates osteoblast differentiation through induction of COX2 in mesenchymal progenitor cells. Stem Cells. 2009;27(9):2254-62. 27. Dâmaso AR, do Prado WL, de Piano A, Tock L, Caranti DA, Lofrano MC, et al. Relationship between nonalcoholic fatty liver disease prevalence and visceral fat in obese adolescents. Dig Liver Dis. 2008;40(2):132-9. 28. Nielsen AR, Hojman P, Erikstrup C, Fischer CP, Plomgaard P, Mounier R, et al. Association between interleukin-15 and obesity: interleukin-15 as a potential regulator of fat mass. J Clin Endocrinol Metab. 2008;93(11):4486-93. 29. Almendro V, Carbó N, Busquets S, López-Soriano J, Figueras M, Ametller E, et al. Interleukin-15 decreases lipid intestinal absorption. Int J Mol Med. 2005;15(6):963-7.

33. Manigrasso MR, Ferroni P, Santilli F, Taraborelli T, Guagnano MT, Michetti N, et al. Association between circulating adiponectin and interleukin-10 levels in android obesity: effects of weight loss. J Clin Endocrinol Metab. 2005;90(10):5876-9. 34. Wang Y, Zhou M, Lam KS, Xu A. Protective roles of adiponectin in obesity-related fatty liver diseases: mechanisms and therapeutic implications. Arq Bras Endocrinol Metabol. 2009;53(2):201-12. 35. Dâmaso AR, de Piano A, Sanches PL, Corgosinho F, Tock L, Oyama LM, et al. Hyperleptinemia in obese adolescents deregulates neuropeptides during weight loss. Peptides. 2011;32(7):1384-91. 36. Giudici KV, Kindler JM, Martin BR, Laing EM, McCabe GP, McCabe LD, et al. Associations among osteocalcin, leptin and metabolic health in children ages 9-13 years in the United States. Nutr Metab (Lond). 2017;14:25. 37. Jürimäe J, Lätt E, Mäestu J, Saar M, Purge P, Maasalu K, et al. Osteocalcin is inversely associated with adiposity and leptin in adolescent boys. J Pediatr Endocrinol Metab. 2015;28(5-6):571-7. 38. Lanyon LE, Rubin CT. Static vs dynamic loads as an influence on bone remodelling. J Biomech. 1984;17(12):897-905. 39. Whipple TJ, Le BH, Demers LM, Chinchilli VM, Petit MA, Sharkey N, et al. Acute effects of moderate intensity resistance exercise on bone cell activity. Int J Sports Med. 2004;25(7):496-501. 40. Foschini D, Araújo RC, Bacurau RF, De Piano A, De Almeida SS, Carnier J, et al. Treatment of obese adolescents: the influence of periodization models and ACE genotype. Obesity (Silver Spring). 2010;18(4):766-72.

30. Carbó N, López-Soriano J, Costelli P, Alvarez B, Busquets S, Baccino FM, et al. Interleukin-15 mediates reciprocal regulation

32. Ye J. Beneficial metabolic activities of inflammatory cytokine interleukin 15 in obesity and type 2 diabetes. Front Med. 2015;9(2):139-45.

284

Arch Endocrinol Metab. 2018;62/3

original article

Waist circumference is an effect modifier of the association between bone mineral density and glucose metabolism Observatório de Epidemiologia Nutricional, Departamento de Nutrição Social e Aplicada, Instituto de Nutrição Josué de Castro, Universidade Federal do Rio de Janeiro (UFRJ), Cidade Universitária, Ilha do Fundão, Rio de Janeiro, RJ, Brasil 2 Instituto de Nutrição Josué de Castro, Universidade Federal do Rio de Janeiro (UFRJ), Cidade Universitária, Ilha do Fundão, Rio de Janeiro, RJ, Brasil 3 Departamento de Puericultura e Pediatria, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, SP, Brasil 4 Departamento de Saúde Pública, Centro de Ciências da Saúde, Universidade Federal do Maranhão (UFMA), São Luís, MA, Brasil 1

ABSTRACT Objective: The role of bone markers on insulin resistance (IR) remains controversial. The objective of this study is to evaluate the association between bone mineral density (BMD) and glucose metabolism and investigate if visceral hyperadiposity, evaluated by waist circumference (WC), is an effect modifier of this association. Subjects and methods: Cross-sectional analysis with 468 young adults from the fourth follow-up of the 1978/79 Ribeirão Preto prospective birth cohort, Brazil. BMD, total osteocalcin (OC), fasting plasma glucose and insulin concentrations were assessed. IR, sensitivity (S) and secretion (β) were estimated by homeostasis model assessment (HOMA) indexes. Multiple linear regression models were constructed to estimate the association between BMD and glucose metabolism. Beta coefficient, R2 and p-values were provided. WC was tested as an effect modifier and OC as a confounder. The covariates were selected based on Direct Acyclic Graph. Results: Significant interaction between BMD (femoral neck and proximal femur areas) and WC on glucose metabolism was observed in the adjusted models. Subjects with increased WC presented a positive association between BMD and log HOMA1-IR while an inverse association was found in those with normal WC (femoral neck R2 = 0.17, p = 0.036; proximal femur R2 = 0.16, p = 0.086). BMD was negatively associated with log HOMA2-S in individuals with increased WC and positively in those with normal WC (femoral neck R2 = 0.16, p = 0.042; proximal femur R2 = 0.15, p = 0.097). No significant associations between BMD, log HOMA2-β and OC and glucose metabolism markers were observed. Conclusions: BMD was associated with glucose metabolism, independently of OC, and WC modifies this association. Arch Endocrinol Metab. 2018;62(3):285-95 Keywords Insulin resistance; glucose metabolism; bone mineral density; bone turnover markers; osteocalcin; bone formation markers

INTRODUCTION Insulin resistance (IR) is characterized by a reduction of the action of the hormone in targets sites, such as muscle, adipose tissue and the liver, resulting in hyperglycemia. To maintain glucose homeostasis, the pancreas adapts through changes in the pancreatic β-cells, resulting in increased insulin secretion. However, exceeding the functional and adaptive capacity can result in the development of type 2 diabetes mellitus (DM2) (1). Several factors, such as being overweight, age, sex, skin color and lifestyle (physical activity, smoking and alcohol intake) (2-8), may be involved in the etiology of IR. Visceral obesity, is associated with a chronic inflammatory response associated with the development Arch Endocrinol Metab. 2018;62/3

Correspondence to: Gilberto Kac Departamento de Nutrição Social e Aplicada, Instituto de Nutrição Josué de Castro, Universidade Federal do Rio de Janeiro Av. Carlos Chagas Filho, 367, CCS, Bloco J2, sala 29, Cidade Universitária, Ilha do Fundão, 21941-590 – Rio de Janeiro, RJ, Brasil gilberto.kac@gmail.com Received on Jun/20/2017 Accepted on Dec/20/2017 DOI: 10.20945/2359-3997000000040

of IR (1). In addition to these classic factors, a possible role of bone markers in IR was found in experimental models (9). Bone mineral density (BMD) results of the remodeling process, i.e., complex process of bone reabsorption and formation, which include the participation of calcitropic hormones that act directly on osteoblasts, osteoclasts and osteocytes. Osteocalcin (OC) is a protein synthesized by osteoblasts during bone formation and therefore affected by the concentration of calcitropic hormones, such as calcitonin and the parathyroid hormone (10). The visceral adipose tissue is associated with the genesis of osteoclasts and therefore with increased bone reabsorption (11). Thus, this 285

Lygia N. Barroso1, Dayana R. Farias1, Marcia Soares-Mota2, Heloisa Bettiol3, Marco Antônio Barbieri3, Milton Cesar Foss3, Antônio Augusto M. da Silva4, Gilberto Kac1

Bone density and glucose metabolism

tissue can affect bone turnover and the concentration of bone turnover markers, such as OC, which seems to be inversely associated with body fat in Chinese men (12). Recent investigations have studied the effects of bone turnover markers in glucose metabolism, adding evidence of the existence of a possible bone-pancreas endocrine axis (9,13). OC seems to be positively associated with proliferation of pancreatic β-cells, insulin secretion and sensitivity and inversely associated with IR in experimental models. In humans, these associations remains controversial in the literature (9,14,15). The association between bone and glucose metabolism is not well defined and few studies have sought to study BMD in this context. We expect an inverse association between BMD and IR in young adults without visceral hyperadiposity, which also appear to have a higher concentration of OC (12), and tested if this is dependent of OC. In individuals with increased waist circumference (WC), we suspect that this association can be modified due to changes in bone metabolism. So we considered WC as a modifying effect on the association between bone and glucose metabolism. The objective of this study was to evaluate if BMD predicts alterations in glucose metabolism, and assess the potential role of OC in this association. In addition, the relationship between OC and WC was tested in our sample.

SUBJECTS AND METHODS

Study design and participants This cross-sectional study was developed with data collected in the fourth phase of the prospective cohort study of individuals born in Ribeirão Preto from the 1st of June 1978 to the 31st of May, 1979. At baseline, information was obtained from 9,067 live newborns delivered in the maternity hospitals of Ribeirão Preto. Infants born to mothers who did not reside in the municipality (n = 2,094) and twins (n = 146) were excluded from the original study. The initial sample comprised 6,827 infants born to mothers residing in Ribeirão Preto. In 2002, when the fourth cohort follow-up was conducted, 5,665 young adults between 23 and 25 years of age were identified as living in the city. Ribeirão Preto consists of 4 geo-economic regions. A sub sample was created from the original study, of which one of 286

three individuals who lived in the same geo-economic area were invited to participate in this phase of the study, resulting in a total of 2,063 young adults. Of the 2,063 individuals included in the fourth phase of the cohort, 513 agreed to undergo the BMD evaluation. Seventeen subjects were excluded due to the presence of a condition that would interfere with the clinical assessment or measures of bone metabolism (i.e., type 1 diabetes, asthmatics using corticosteroids, amaurosis, anorexia nervosa, scoliosis, urolithiasis and stroke). Additionally, 28 subjects were lost due to missing data or because their total OC and markers of glucose metabolism (fasting glucose and insulin) were not measured. The final sample of the present study comprised 468 individuals (females = 235) who underwent BMD, OC and HOMA [(homeostasis model assessment (HOMA), IR, insulin sensitivity (S), and β-cell function (β)] evaluations. More detailed information about this cohort can be obtained from previous publication (16).

Measurements A 40-mL blood sample was collected after 12-hour fasting period. All laboratory tests (fasting insulin and glucose and OC) were analyzed at the time of data collection. Fasting glucose and insulin were determined using commercial kits by GOD/PAP human diagnostic colorimetric enzymatic method (Chronolab AG, Zug, Switzerland) and radioimmunoassay (Insulin kit, DPC, Los Angeles, CA, USA), respectively. OC was determined using an immunoradiometric method (DSL-7600, IRMA, Webster, TX, USA). IR was estimated using the original HOMA1-IR, calculated according to the formula: fasting plasma glucose (mmol/L) x fasting plasma insulin (µU/ mL)/22.5. To estimate insulin sensitivity (HOMA2-S) and secretion (HOMA2-β), we used the HOMA computer model (HOMA2 model), available from https://www.dtu.ox.ac.uk/homacalculator/. BMD (g/cm2) were obtained by dual-energy X-ray absorptiometry (DXA) using Hologic QDR-4500 (Waltham, MA, USA) equipment. Measurements of absolute precision error (and the percentage coefficient of variation) for BMD were 0.007 g/cm2 (0.66%), 0.015 g/cm2 (1.77%) and 0.007 g/cm2 (0.70%) for the three evaluated anatomic areas: the lumbar spine, femoral neck and total proximal femur, respectively. The phantom coefficient of variation throughout the Arch Endocrinol Metab. 2018;62/3

study was 0.38%. A standardized technician performed the quality control and all measurements. The analysis was performed in the nuclear medicine laboratory of the Clinics Hospital, Faculty of Medicine of Ribeirão Preto, University of São Paulo, Brazil. The following socio-demographic and lifestyle variables were obtained through structured questionnaires: sex (male; female), age (23-25), selfclassified skin color (white; mulatto/black/yellow), schooling (≤ 8; 9-11 and ≥ 12 years of study) and smoking (smokers; non-smokers). Physical activity was measured using the short version of the International Physical Activity Questionnaire (IPAQ), validated for the Brazilian population, and categorized as low, moderate, or high activity (17). Caloric intake (kcal/day) was estimated based on an adaptation of a validated food frequency questionnaire (FFQ) (18). The software Dietsys version 4.0 was used (National Cancer Institute, Bethesda, MD, USA). Alcohol consumption was estimated based on the FFQ, expressed as a percentage of total dietary energy per day. Adult weight and height were obtained using standardized techniques. A mechanical scale (Filizola, São Paulo, Brazil) with an accuracy of 100 g and a freestanding wood stadiometer (University of São Paulo, Ribeirão Preto, Brazil) with an accuracy of 0.1 cm were used. Body mass index (BMI) was categorized as < 25 (underweight or normal weight), 25 to 29.9 (overweight) and ≥ 30 kg/m2 (obesity) (19). Waist circumference (WC) was measured by a D-loop non-stretch fiberglass tape as the smallest circumference between the ribs and the iliac crest while the subject stood with the abdomen relaxed at the end of a normal expiration. The individuals were classified as normal/increased WC (women: < 80 cm; ≥ 80 cm; men: < 94 cm; ≥ 94 cm) according to cutoff points of WC proposed by the World Health Organization (WHO) (20).

Ethics This study was approved by the Research Ethics Committee of the Clinics Hospital, Faculty of Medicine of Ribeirão Preto, University of São Paulo, Brazil, in February 2000 (protocol no. 7606/99).

Statistical analyses The subject characteristics were described using means (standard deviation) and p-value refers to the Student’s Arch Endocrinol Metab. 2018;62/3

t-test or median (interquartile range) and MannWhitney U test. Categorical variables are expressed as absolute and relative frequency and compared by quisquared test. The study outcomes were described using medians and interquartile ranges stratified by categories of potentially associated factors. Continuous variables were categorized into tertiles, and comparisons between categories were performed using Mann-Whitney U and Kruskal-Wallis tests. To evaluate the association between BMD and IR (HOMA1-IR), sensitivity (HOMA2-S) and secretion (HOMA2-β), multiple linear regression models were fitted for each outcome. For this analysis, HOMA1-IR, HOMA2-S and HOMA2-β were log-transformed. WC was tested as an effect modifier in multiple linear regression models considering that inflammation associated with visceral adipose tissue can affect bone metabolism markers, which may explain the association between BMD and IR. We constructed linear prediction plots of the associations between each anatomic bone area (spinal, femoral neck and proximal femur) and the outcomes, stratified by WC cutoff points (normal/ increased), in order to interpret the interactions. The covariates were selected for inclusion in the final model based on a directed acyclic graph [(DAG, www. dagitty.net), Figure S1]. A DAG is a graphic model in which potential confounding factors that can distort the causal inference process can be identified and included as covariates in adjusted models (21). DAGs can make more explicit the relationship between exposure and outcome and help avoid inappropriate adjustments. OC concentration was included in the models as a confounder because of its association with both the exposure and the outcome. We tested whether the association between BMD and glucose metabolism is dependent of OC or if there is an independent pathway linking BMD and IR. The relationship between OC and WC was tested by Pearson correlation test, stratified by sex. In the analysis, associations with p-value < 0.05 were considered significant, except in the evaluation of interactions, in which a p-value < 0.1 was considered significant (22). The regression analysis provided a beta coefficient, the co-variable for the p-value, the R2 (variation explained by the models) for each model, and the p-value of all the multiple models. All analyses were performed using Stata Data Analysis and Statistical Software (STATA) version 12.0, 2011, College Station, TX (StataCorp LP). 287

Bone density and glucose metabolism

Figure S1. Causal diagram of the association between BMD and IR. Minimal sufficient adjustment sets for estimating the total effect of BMD and IR, suggested by DAG – age, alcohol consumption, osteocalcin, physical activity, sex, skin color and smoking. Colors of variables: green – exposure; blue – outcome; red – co variables.

RESULTS We evaluated 468 (233 men and 235 women) adults. The study participants had a mean age of 23.5 (0.5) years and a mean BMI of 23.7 (4.2) kg/m². The majority of the sample was white (65%), presented normal WC (78%), had more than 8 years of schooling (88%), reported moderate or high physical activity (74%) and were non-smokers (85%). The mean calorie intake and alcohol consumption were 2,188.9 (713.4) kcal/day and 1.8% (2.3) of EI/day, respectively. The median (interquartile range) of IR, insulin sensitivity and β-cell function were 1.1 (0.7:1.7), 136.6 (92.7:217.9) and 98.1 (73.7:132.4), respectively. Men presented higher IR and lower insulin secretion compared to women (p < 0.05). The mean OC concentration was 12.6 (5.1) ng/mL, and the BMD was 1.0 (0.1) g/cm2 for the spinal anatomic area, 0.9 (0.2) g/cm2 for the femoral neck and 1.0 (0.2) g/cm2 for the proximal femur. 288

Higher mean of OC and BMD (spinal, femoral neck and proximal femur) were detected among men (p < 0.001) (Table 1). A positive association between nutritional status markers (BMI and WC) and HOMA1-IR and HOMA2-β and an inverse association with HOMA2-S was found (p < 0.001 for all). Individuals who reported low physical activity had higher median HOMA1-IR (p = 0.007) and HOMA2-β values (p < 0.001) and lower HOMA2-S values (p = 0.002) than those reporting moderate or high levels of physical activity. The median HOMA1-IR and HOMA2-β levels differed between the sexes, i.e., men presented higher mean IR values (p = 0.016) and lower hormone secretion (p = 0.024) than women. Subjects classified in the 1st tertile of OC presented significantly higher median levels of HOMA2-β than those in the 2nd and 3rd tertiles (p = 0.018). The femoral neck and proximal femur BMD were inversely associated with insulin sensitivity Arch Endocrinol Metab. 2018;62/3

Bone density and glucose metabolism

Table 1. Descriptive characteristics of a young adults sample, 2002-2004 Ribeirão Preto cohort, Brazil, fourth follow-up Characteristics

Total (n = 468)

Men (n = 233)

Women (n = 235)

p-value1

Age (years)

23.5 (0.5)

23.5 (0.03)

0.312

BMI (kg/m2)

23.7 (4.2)

24.7 (0.27)

22.7 (0.26)

< 0.001 0.025

WC (cm)2 Normal

363 (78.0)

180 (49.6)

183 (50.4)

Increased

105 (22.0)

53 (50.5)

52 (49.5)

White

304 (65.0)

142 (46.7)

162 (53.3)

Black/mullato/yellow

164 (35.0)

91 (55.5)

73 (44.5)

Skin color

0.070

Schooling (years)

0.277

≤8

58 (12.0)

27 (46.5)

31 (53.5)

9-11

262 (56.0)

139 (53.0)

123 (47.0)

≥ 12

148 (32.0)

67 (45.3)

81 (54.7)

2188.9 (713.4)

2415.9 (45.8)

1963.8 (42.6)

< 0.001

1.8 (2.3)

2.44 (0.16)

1.13 (0.12)

< 0.001

Low

121 (26)

46 (38.0)

75 (62.0)

Moderate

200 (43)

99 (49.5)

101 (50.5)

High

147 (31)

88 (59.9)

59 (40.1)

Yes

68 (85.0)

43 (63.2)

25 (36.8)

400 (15.0)

190 (47.5)

210 (52.5)

Energy intake (kcal/day) Alcohol consumption (% of EI/day)

0.002

Physical activity

0.016

Smoking

HOMA1-IR

1.1 (0.7;1.7)

1.26 (0.78;1.92)

1.1 (0.7;1.5)

0.016

HOMA2-S

136.6 (92.7;217.9)

131.5 (83.1;204.8)

140.2 (100;223.6)

0.104

HOMA2-β

98.1 (73.7;132.4)

92.4 (69.9;131)

104.3 (76.5;134.2)

0.024

12.6 (5.1)

14.01 (0.33)

11.22 (0.31)

< 0.001

Osteocalcin (ng/mL) BMD Spinal (g/cm2)

1.0 (0.1)

1.06 (0.00)

0.99 (0.00)

< 0.001

BMD Femoral neck (g/cm2)

0.9 (0.2)

1.00 (0.01)

0.84 (0.00)

< 0.001

BMD Proximal femur (g/cm2)

1.0 (0.2)

1.09 (0.01)

0.90 (0.00)

< 0.001

Continuous variables are expressed as mean (standard deviation) and p-value refers to the Student’s t-test or median (interquartile range) and Mann-Whitney U test. Categorical variables are expressed as absolute and relative frequency and compared by qui-squared test. 2 Categorized using World Health Organization cutoff points, normal WC: < 80 cm for women and < 94 cm for men; increased WC: ≥ 80 cm for women and ≥ 94 cm for men. For EI variable, we had 1 exclusion due to high calorie value (> 6000 kcal/day). BMI: body mass index; WC: waist circumference; EI: energy intake; HOMA: homeostatic model assessment; HOMA1-IR: insulin resistance; HOMA2-S: insulin sensitivity; HOMA2-β: β-cell function (insulin secretion); BMD: bone mineral density.

(p = 0.036 and p = 0.002) and were positively associated with IR (p = 0.013 and p < 0.001) (Table 2). A significant inverse correlation between OC and WC was observed in men (r = -0.23, p = 0.002) and women (r = -0.15, p = 0.020) (data not shown). We found a significant interaction between BMD (femoral neck and proximal femur) and WC in the fully adjusted regression (p < 0.1). We observed a positive association between BMD and the log HOMA1-IR level in individuals with increased WC and an inverse association in those with normal Arch Endocrinol Metab. 2018;62/3

WC (femoral neck R2 = 0.17, p=0.036; proximal femur R2 = 0.16, p = 0.086). BMD was negatively associated with the log HOMA2-S level in subjects with increased WC and positively associated in those with normal WC (femoral neck R2 = 0.16, p = 0.042; proximal femur R2 = 0.15, p = 0.097). We did not observe significant associations between BMD (spinal, femoral neck and proximal femur) and the log HOMA2-β level and OC and the log HOMA1IR, HOMA2-S and HOMA2-β levels (Table 3 and Figures 1 and 2). 289

Bone density and glucose metabolism

Table 2. Distribution of insulin resistance (HOMA1-IR1), insulin sensitivity (HOMA2-S2) and β cell function (HOMA2-β2) in 468 young adults according to categories of selected variables, 2002-2004 Ribeirão Preto cohort, Brazil, fourth follow-up n

HOMA1–IR

HOMA2-S

HOMA2-β

< 25

312

1.0 (0.6;1.4)a

157.9 (110.6;247.2)a

92.0 (70.2;122.2)a

≥ 25-29.9

116

1.4 (0.9;1.9)b

116.3 (84.9;170.1)b

108.9 (77.9;139.5)b

≥ 30

2.5 (1.5;3.3)

Normal

363

1.0 (0.6;1.5)

Increased

105

1.8 (1.1;3.2)

BMI (kg/m2)

< 0.001

68.9 (50.3;100.0)

< 0.001

144.4 (109.8;182.0)c

< 0.001

154.5 (108.0;242.7) < 0.001

92.3 (70.5;122.0) < 0.001

88.9 (55.4;135.4)

129.4 (95.0;165.5)

< 0.001

Sex Women

235

1.1 (0.7;1.5)

Men

233

1.3 (0.8-1.9)

140.2 (100.0;224.8)

Low

121

1.3 (0.9;2.0)a

118.5 (80.9;174.7)a

111.8 (88.5;147.5)a

Moderate

200

1.1 (0.7;1.6)b

136.7 (101.4;228.0)b

94.6 (72.8;126.3)b

High

147

1.0 (0.6;1.7)

400

1.1 (0.7;1.7)

Yes

1.1 (0.7;1.9)

White

304

1.1 (0.7;1.6)

Black/mullato/yellow

164

1.2 (0.8;2.0)

0.016

131.5 (83.1;204.8)

104.3 (76.5;134.2) 0.105

92.4 (69.9;131.0)

0.024

Physical activity

0.007

152.1 (93.9;262.4)

0.002

87.5 (67.0;125.8)b

< 0.001

Smoking 136.0 (95.0;217.9) 0.907

146.5 (82.5;214.1)

98.7 (73.7;131.3) 0.890

94.5 (69.9;147.9)

0.055

100.4 (74.9;141.5)

0.955

Skin color 141.1 (97.8;223.9) 0.055

128.8 (81.2;194.3)

95.1 (72.4;129.3) 0.154

Schooling (years) ≤8

1.2 (0.7;2.0)

117.6 (77.2;208.6)

104.9 (76.4;135.8)

9-11

262

1.2 (0.7;1.7)

133.6 (90.2;211.9)

97.5 (72.9;131.7)

≥ 12

148

1.1 (0.7;1.6)

1st and 2nd tertiles (0.0-2.1)

315

1.1 (0.7;1.7)

3rd tertile (2.2-12.1)

153

1.1 (0.7;1.7)

1st and 2nd tertiles (851.1-2377.6)

313

1.1 (0.7;1.6)

3rd tertile (2394.1-4940.7)

154

1.3 (0.7;1.8)

1st tertile (2.9-9.6)

151

1.2 (0.8;1.9)

131.8 (83.5;197.6)

106.6 (78.5;144.6)a

2nd tertile (9.8-14.3)

161

1.1 (0.7;1.6)

139.8 (99.7;219.1)

93.8 (72.9;130.2)b

3 tertile (14.4-32.9)

156

1.1 (0.6;1.7)

1st and 2nd tertiles (0.7-1.1)

309

1.1 (0.7;1.6)

3rd tertile (1.1-1.4)

159

1.2 (0.7;1.8)

1st and 2nd tertiles (0.5-1.0)

310

1.1 (0.7;1.6)

3rd tertile (1.0-1.5)

158

1.3 (0.8;1.9)

1st and 2nd tertiles (0.6-1.0)

309

1.1 (0.7;1.5)

3 tertile (1.0-1.5)

159

1.4 (0.9;2.3)

0.177

151.2 (100.7;221.6)

0.117

95.6 (73.8;132.5)

0.349

Alcohol consumption (% of EI/day) 136.1 (93.9;217.5) 0.727

139.8 (89.6;218.4)

98.3 (74.8;134.0) 0.857

98.0 (69.1;130.2)

0.479

Energy intake (kcal/day) 141.3 (96.6;211.9) 0.221

125.7 (88.5;219.1)

98.9 (73.8;134.2) 0.363

96.5 (72.1;131.0)

0.602

Osteocalcin (ng/mL)

0.174

151.1 (95.0;238.6)

0.110

93.4 (68.6;120.7)b

0.018

BMD Spinal (g/cm2) 141.0 (97.2;224.3) 0.062

123.0 (85.4;200.0)

98.5 (73.1;130.2) 0.089

97.0 (74.5;134.6)

0.415

BMD Femoral neck (g/cm ) 2

140.6 (99.4;229.5) 0.013

125.7 (80.9;197.2)

99.9 (73.7;130.9) 0.036

95.3 (73.7;135.8)

0.864

147.2 (100.0;229.5) < 0.001

115.7 (75.0;180.8)

99.0 (73.7;129.3) 0.002

96.3 (73.7;141.4)

0.447

p-value refers to Kruskall Wallis test and Mann-Whitney U test. Values with differing superscript letters (a, b, c) denote statistically significant differences across the categories. 2 Categorized using World Health Organization cutoff points, normal WC: < 80 cm for women and < 94 cm for men; increased WC: ≥ 80 cm for women and ≥ 94 cm for men. Data are expressed as median (interquartile range). For EI variable, we had 1 exclusion due to high calorie value (> 6000 kcal/day). HOMA: homeostatic model assessment; HOMA1-IR: insulin resistance; HOMA2-S: insulin sensitivity; HOMA2-β: β-cell function (insulin secretion); BMI: body mass index; WC: waist circumference; EI: energy intake; BMD: bone mineral density. 1

290

Arch Endocrinol Metab. 2018;62/3

0.13

0.090

0.248 0.16

-0.0 (-0.0;0.0)

0.8 (-0.1;1.7)

-0.2 (-1.2;0.7)

-0.5 (-1.0;0.0)

0.17

-0.0 (-0.0;0.0)

1.0 (0.1;1.9)

-0.3 (-1.3;0.5)

-0.5 (-1.0;0.0)

0.16

-0.0 (-0.0;0.0)

0.8 (-0.5;2.0)

-0.2 (-1.6;1.2)

-0.5 (-1.1;0.1)

Model 2 β1 (95% CI)

0.113

0.086

0.638

0.080

0.119

0.036

0.440

0.052

0.114

0.246

0.763

0.133

0.12

0.0 (-0.0;0.0)

-0.8 (-1.6; 0.1)

0.2 (-0.7;1.2)

0.1 (-0.3;0.5)

0.13

0.0 (-0.0;0.0)

-1.0 (-1.9;-0.1)

0.4 (-0.5;1.3)

0.2 (-0.2;0.6)

0.12

0.0 (-0.0;0.0)

-0.6 (-1.8;0.6)

0.1 (-1.2;1.4)

0.2 (-0.3;0.8)

Model 1 β1 (95% CI)

0.212

0.091

0.605

0.571

0.228

0.033

0.383

0.385

0.189

0.324

0.885

0.432

0.15

0.0 (-0.0;0.0)

-0.7 (-1.6;0.1)

0.2 (-0.7;1.1)

0.4 (-0.1;0.9)

0.16

0.0 (-0.0;0.0)

-0.9 (-1.8;-0.0)

0.3 (-0.5;1.2)

0.5 (-0.0;1.0)

0.15

0.0 (-0.0;0.0)

-0.7 (-1.9;0.5)

0.2 (-1.1;1.5)

0.4 (-0.2;1.0)

Model 2 β1 (95% CI)

HOMA2-S

0.138

0.097

0.647

0.098

0.144

0.042

0.457

0.066

0.139

0.256

0.760

0.160

0.08

-0.0 (-0.0;0.0)

0.4 (-0.1;1.0)

-0.2 (-0.8;0.4)

-0.3 (-0.6;-0.0)

0.09

-0.0 (-0.0;0.0)

0.6 (-0.0;1.2)

-0.2 (-0.8;0.3)

-0.3 (-0.6;-0.0)

0.08

-0.0 (-0.0;0.0)

0.4 (-0.4;1.2)

-0.1 (-1.0;0.7)

-0.2 (-0.6;0.1)

Model 1 β1 (95% CI)

0.106

0.137

0.591

0.047

0.110

0.061

0.392

0.038

0.079

0.322

0.735

0.196

0.11

-0.0 (-0.0;0.0)

0.4 (-0.2;1.0)

-0.1 (-0.7;0.5)

-0.2 (-0.5;0.2)

0.11

-0.0 (-0.0;0.0)

0.4 (-0.1;1.0)

-0.1 (-0.7;0.4)

-0.2 (-0.5;0.1)

0.11

-0.0 (-0.0;0.0)

0.3 (-0.5;1.2)

-0.1 (-1.0;0.8)

-0.2 (-0.5;0.2)

Model 2 β1 (95% CI)

HOMA2-β

0.373

0.205

0.726

0.307

0.378

0.135

0.606

0.245

0.364

0.401

0.836

0.424

Linear regression coefficient; p-value refers to linear regression. Categorized using World Health Organization cutoff points, normal WC: < 80 cm for women and < 94 cm for men; increased WC: ≥ 80 cm for women and ≥ 94 cm for men. R refers to the outcome variation explained by the models. Model 1 was adjusted only for osteocalcin and the interaction between BMD and WC, Model 2 was further adjusted for physical activity, smoking, alcohol intake, sex, age and skin color. All the multiple models were statically significant (p-value < 0.001). CI: confidence interval; HOMA: homeostatic model assessment; HOMA1-IR: insulin resistance; HOMA2-S: insulin sensitivity; HOMA2-β: β-cell function (insulin secretion); WC: waist circumference; BMD: bone mineral density.

R2 4

0.8 (-0.1;1.7)

0.617

-0.2 (-1.2;0.7)

-0.0 (-0.0; 0.0)

0.799

0.269

-0.1 (-0.5;0.4)

Osteocalcin (ng/mL)

PF#WC

Interaction term

WC (normal/increased)

PF (g/cm2)

BMD - Proximal femur (PF)

0.13

-0.0 (-0.0;0.0)

R2 4

Osteocalcin (ng/mL)

0.395

-0.4 (-1.3;0.5) 0.032

0.553

0.230

-0.1 (-0.6;0.3)

1.0 (0.1;1.9)

FN#WC

Interaction term

WC (normal/increased)

FN (g/cm2)

BMD - Femoral neck (FN)

0.13

-0.0 (-0.0;0.0)

R2 4

Osteocalcin (ng/mL)

0.936

-0.0 (-1.4;1.3) 0.344

0.510

HOMA1-IR

-0.2 (-0.8;0.4)

0.6 (-0.7;1.9)

Spinal#WC

Interaction term

WC (normal/increased)

Spinal (g/cm2)

BMD - Spinal

Model 1 β1 (95% CI)

Table 3. Linear regression between bone mineral density (BMD; g/cm2) and insulin resistance (HOMA1-IR1), insulin sensitivity (HOMA2-S2) and β cell function (HOMA2-β2) in 468 young adults, 2002-2004 Ribeirão Preto cohort, Brazil, fourth follow-up

Bone density and glucose metabolism

291

Bone density and glucose metabolism

Figure 1. Scatter and linear prediction between BMD and Log HOMA1-IR according to WC in 468 young adults, 2002-2004 Ribeirão Preto, Brazil, fourth cross-sectional evaluation. A) Spinal BMD. B) Femoral neck BMD. C) Proximal femur BMD. HOMA1-IR: homeostatic model assessment – insulin resistance; WC: waist circumference; BMD: bone mineral density. Fitted values were predicted using linear regression models; WC was categorized using World Health Organization cutoff points, normal WC: < 80 cm for women and < 94 cm for men; increased WC: ≥ 80 cm for women and ≥ 94 cm for men.

Figure 2. Scatter and linear prediction between BMD and Log HOMA2-S according to WC in 468 young adults, 2002-2004 Ribeirão Preto, Brazil, fourth cross-sectional evaluation. A) Spinal BMD. B) Femoral neck BMD. C) Proximal femur BMD. HOMA2-S: homeostatic model assessment – insulin sensitivity; WC: waist circumference; BMD: bone mineral density. Fitted values were predicted using linear regression models; WC was categorized using World Health Organization cutoff points, normal WC: < 80 cm for women and < 94 cm for men; increased WC: ≥ 80 cm for women and ≥ 94 cm for men.

DISCUSSION The present study has three main results. First, we found that BMD predict alterations in glucose metabolism in young adults. Second, we observed that the direction of the association differed according to WC classification, 292

i.e., adults with increased WC had a positive association between BMD and IR, while those with normal WC had an inverse association between these two markers. The association between BMD and insulin sensitivity occurred in the opposite direction, i.e., we observed Arch Endocrinol Metab. 2018;62/3

an inverse association in individuals with increased WC and a positive association in those with normal WC. Finally, we did not observe any significant association between OC and glucose metabolism in the adjusted models. This study has some potential limitations. Although we used a large sample size from a birth cohort, only 24.9% (n = 513/2,063) of the individuals evaluated in the fourth phase of the birth cohort follow-up consented to undergo DXA assessments, and after exclusions, the final sample comprised 468 subjects who had valid BMD measurements. In addition, although WC is a very practical and internationally used tool to evaluate the deposition of intra-abdominal fat, recommended by WHO (20), its use has as a limitation the fact that it does not separate visceral adipose tissue of the subcutaneous tissue. Moreover, it was not possible to use the WHO protocol to measure waist circumference (WC) in our study, because data collection occurred from 2002 to 2004, while the WHO STEPS protocol was published in 2008 (23). Additionally, this study was based on a cross-sectional analysis, a study design that cannot determine whether the results are merely associations or if BMD exerts a causal effect on glucose metabolism in these young adults. Finally, although in experimental studies OC uncarboxilated has been reported to be the metabolically active form (9,13), we did not differentiate plasma OC by gamma-carboxylation status, and our assessment included all forms of OC. The strength of this study is the number of young adults evaluated by DXA, a very accurate procedure for measuring bone density. Moreover, in the multivariate analysis, we evaluated the inclusion of co-variables based on a DAG that allows for the minimization of bias in epidemiological studies. DAGs allows the identification of the minimum sufficient adjustment to estimate the total and direct effect of a certain exposure on the studied outcome (21). To the best of our knowledge, this is only the second study that has evaluated if BMD, assessed by a gold standard measure (DXA), predict alterations in glucose metabolism (IR, sensitivity and secretion) in young adults. This study provides new information about the association between bone and glucose metabolism. We found a significant association between BMD and IR and insulin sensitivity and a significant interaction between BMD and WC. A non-significant association between BMD and glucose metabolism (plasma glucose and serum insulin) has been found in the unadjusted Arch Endocrinol Metab. 2018;62/3

model, which persisted after adjusting the analysis in 155 healthy young adults (24). In that study, although fat mass was considered a confounder, the adipose tissue was not tested as an effect modifier in the association between bone and glucose metabolism (24), as done in our study. It is known that body fat, particularly visceral fat, may affect bone metabolism markers and BMD. Chronic low-grade inflammation associated with visceral fat is related to the genesis of osteoclasts, increased bone resorption and decreased OC concentration (11,12). Individuals with obesity present increased risk of fractures possibly associated with metabolic dysfunction that result in reduction of bone turnover and bone quality (25). Therefore, considering the effects of inflammation on bone turnover and mass, our conceptual framework considers that WC plays an important role in the association between BMD and IR. We have hypothesized that WC acts as an effect modifier and not as a confounder, and for this reason, this marker of visceral fat deposition was not included in the DAG that depicted the theoretical relationship between all involved variables. OC is one of the most studied bone biomarkers in the association with glucose metabolism. In the current study, it was observed an inverse correlation between OC and WC in a sample of predominantly white young adults of both sexes (data not shown). These results corroborates with the inverse relationship between OC and visceral fat area found in Chinese men (12) and an inverse relationship between OC and trunk fat in men with obesity (26). These findings suggest a negative effect of adipose tissue, especially visceral fat, on OC. Despite this, we did not find a significant association between OC concentrations and HOMA1-IR or HOMA2-S in either the crude or adjusted analysis. Animal studies, however, have demonstrated the positive effect of OC on insulin secretion and the sensitivity and proliferation of pancreatic β-cells (9,27). To exert these effects, OC binds to its receptor GPCR6a in pancreatic β-cells and can also increase the expression of anti-inflammatory adipokines and reduce the secretion of pro-inflammatory cytokines (13). In humans, the findings remain controversial. In line with our results, 137 young adults (18.6 years) were evaluated and no association was found between OC and HOMA1-IR (28). In addition, other studies found no association between OC and HOMA1IR, HOMA2-β, QUICKI insulin sensitivity marker, 293

Bone density and glucose metabolism

blood glucose and insulin in pre- and post-menopausal women (14,29). On the other hand, some studies have found an inverse association between OC and IR and a positive association between OC and insulin sensitivity and secretion (12,26,30). The differences between these studies and ours may be explained by the fact that we studied healthy young adults while the others studies investigated older people (approximately 50 years of age) and/or individuals with obesity, which tend to have higher IR. Moreover, we also found methodological differences as most studies used only correlation statistical procedures (12,26), and only one study performed adjusted regression models like ours (30). Unlike our study, none of the published articles evaluated the selection of covariates with a DAG model. Finally, we expected that the addition of OC in the regression model could explain the association between BMD and IR, however, we observed that associations between bone and glucose metabolism is independent of this bone metabolism marker (because the inclusion of OC in the regression model did not affect the association between BMD and glucose metabolism). Some studies have demonstrated that osteoprotegerin (OPG), that promotes bone formation, appears to be increased in metabolic disorders, such as obesity (31), and in individuals with obesity was found a positive association between OPG and HOMA1-IR (32). In experimental study, OPG increased inflammation in adipose tissue (33). This association of OPG with inflammation may explain its association with IR. In addition to OPG, the amino terminal propeptide of procollagen type 1 (P1NP), a marker of bone formation as OC, was also positively associated with HOMA1IR in young women with overweight or obesity (15). In view of this, we suggest that further studies be performed to investigate the action of biomarkers other than OC, that may explain the positive association between BMD and HOMA1-IR observed in our study. Individuals in the accrual phase present higher speed of bone mass gain, especially until reaching peak bone mass. Considering that in our study we found a positive effect of BMD on IR in young individuals with increased WC, it can be concluded that this is a critical phase of life, associated with increased metabolic risk. It is recognized that IR is involved in the pathophysiology of DM2, a global public health problem. The association between bone and IR suggests the existence of bone-pancreas axis. However, the exact mechanism that links bone mass and glucose metabolism is not 294

fully understood, and this study sought to contribute evidence to clarify this relation. We believe that a better understanding of this association can contribute to improve IR. Corroborating this statement, other studies have been developed with the aim of modulating pharmacologically bone metabolism markers to improve glycemic control (34). Additionally, disorders associated with IR, such as obesity, seem related to reduced bone quality and formation and increased bone fracture risk (25). Thus, the investigation of the relationship between bone and glucose metabolism may not only contribute to the glycemic control but also to bone fragility prevention. Moreover, as expected, we found that subjects with obesity and those with increased WC present higher IR and secretion and lower insulin sensitivity. Individuals with obesity were evaluated and it was identified that those with a higher percentage of lean mass also had higher insulin sensitivity and lower inflammatory status (35). Greater insulin secretion was found in individuals who have greater IR, which characterizes the pancreatic response in compensation of IR (1). It is known that the increased secretion of adiponectin and the positive effect of estrogen on glucose homeostasis contributes to the lower IR observed in women compared to men (6), as found in our results. In addition, we found that men had a higher mean BMI compared to women, which may also explain the higher rate of IR in this group. We conclude that BMD was associated with glucose metabolism and this association is independent of OC. We also found that the WC modifies the association between BMD and IR and sensitivity. These results indicate that bone may play a role in the metabolic profile of IR and obesity. However, further studies are needed to assess the direction of the association between BMD and IR and to test the possible mechanisms involved in this relationship. Author contributions: all authors have made substantial contributions on analysis and interpretation of data; have drafted the article and revised it critically; have seen and approved the contents of the submitted manuscript. Funding: this study was funded by the SĂŁo Paulo Research Foundation â&#x20AC;&#x201C; Fapesp, National Council for Scientific and Technological Development (CNPq), and Brazilian Coordination Body for the Training of University Level Personnel (Capes). Disclosure: no potential conflict of interest relevant to this article was reported. Arch Endocrinol Metab. 2018;62/3

Bone density and glucose metabolism

1. Kahn SE, Hull RL, Utzschneider KM. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature. 2006;444(7121):840-6. 2. Willi C, Bodenmann P, Ghali WA, Faris PD, Cornuz J. Active smoking and the risk of type 2 diabetes: a systematic review and meta-analysis. JAMA. 2007;298(22):2654-64. 3. Kim JY, Lee DY, Lee YJ, Park KJ, Kim KH, Kim JW, et al. Chronic alcohol consumption potentiates the development of diabetes through pancreatic β-cell dysfunction. World J Biol Chem. 2015;6(1):1-15. 4. Li L, Yin X, Yu D, Li H. Impact of physical activity on glycemic control and insulin resistance: a study of community-dwelling diabetic patients in Eastern China. Intern Med. 2016;55(9):1055-60. 5. Jung UJ, Choi MS. Obesity and its metabolic complications: the role of adipokines and the relationship between obesity, inflammation, insulin resistance, dyslipidemia and nonalcoholic fatty liver disease. Int J Mol Sci. 2014;15(4):6184-223. 6. Geer EB, Shen W. Gender differences in insulin resistance, body composition, and energy balance. Gend Med. 2009;6 Suppl 1:60-75. 7.

Piccolo RS, Subramanian SV, Pearce N, Florez JC, McKinlay JB. Relative contributions of socioeconomic, local environmental, psychosocial, lifestyle/behavioral, biophysiological, and ancestral factors to racial/ethnic disparities in type 2 diabetes. Diabetes Care. 2016;39(7):1208-17.

8. Barzilai N, Ferrucci L. Insulin resistance and aging: a cause or a protective response? J Gerontol A Biol Sci Med Sci. 2012;67(12):1329-31. 9. Lee NK, Sowa H, Hinoi E, Ferron M, Ahn JD, Confavreux C, et al. Endocrine regulation of energy metabolism by the skeleton. Cell. 2007;130(3):456-69.

prevenção de doenças crônicas não transmissíveis. Rev Nutr. 2002;15(2):239-45. 19. World Health Organization (WHO). Physical status: the use and interpretation of anthropometry. Geneva: WHO; 1995. [Technical Report Series 854] 20. World Health Organization (WHO). Obesity: prevention and managing the global epidemic. Geneva: WHO; 1998. [Report of the WHO Consultation on Obesity] 21. Textor J, Hardt J, Knüppel S. DAGitty: a graphical tool for analyzing causal diagrams. Epidemiology. 2011;22(5):745. 22. Twisk J. Applied multilevel analysis: a practical guide. United Kingdom: Cambridge University Press; 2006. 23. World Health Organization (WHO). Waist circumference and waist-hip ratio. Geneva: WHO; 2008. [Report of a WHO expert consultation, 8-11] 24. Pirilä S, Taskinen M, Turanlahti M, Kajosaari M, Mäkitie O, Saarinen-Pihkala UM, et al. Bone health and risk factors of cardiovascular disease – a cross-sectional study in healthy young adults. PLoS One. 2014;9(10):e108040. 25. Nielson CM, Marshall LM, Adams AL, LeBlanc ES, Cawthon PM, Ensrud K, et al.; Osteoporotic Fractures in Men Study Research Group. BMI and fracture risk in older men: the osteoporotic fractures in men study (MrOS). J Bone Miner Res. 2011;26(3):496-502. 26. Migliaccio S, Francomano D, Bruzziches R, Greco EA, Fornari R, Donini LM, et al. Trunk fat negatively influences skeletal and testicular functions in obese men: clinical implications for the aging male. Int J Endocrinol. 2013;2013:182753. 27. Ferron M, Hinoi E, Karsenty G, Ducy P. Osteocalcin differentially regulates beta cell and adipocyte gene expression and affects the development of metabolic diseases in wild-type mice. Proc Natl Acad Sci U S A. 2008;105(13):5266-70.

10. De Paula FJA, Rosen CJ. Back to the future: revisiting parathyroid hormone and calcitonin control of bone remodeling. Horm Metab Res. 2010;42(5):299-306.

28. Polgreen LE, Jacobs DR Jr, Nathan BM, Steinberger J, Moran A, Sinaiko AR. Association of osteocalcin with obesity, insulin resistance, and cardiovascular risk factors in young adults. Obesity (Silver Spring). 2012;20(11):2194-201.

11. Aguirre L, Napoli N, Waters D, Qualls C, Villareal DT, ArmamentoVillareal R. Increasing adiposity is associated with higher adipokine levels and lower bone mineral density in obese older adults. J Clin Endocrinol Metab. 2014;99(9):3290-7.

29. Caglar GS, Ozdemir ED, Kiseli M, Demirtas S, Cengiz SD. The association of osteocalcin and adiponectin with glucose metabolism in nondiabetic postmenopausal women. Gynecol Obstet Invest. 2014;77(4):255-60.

12. BaoY, Ma X,Yang R, Wang F, HaoY, Dou J, et al. Inverse relationship between serum osteocalcin levels and visceral fat area in Chinese men. J Clin Endocrinol Metab. 2013;98(1):345-51.

30. Gravenstein KS, Napora JK, Short RG, Ramachandran R, Carlson OD, Metter EJ, et al. Cross-sectional evidence of a signaling pathway from bone homeostasis to glucose metabolism. J Clin Endocrinol Metab. 2011;96(6):E884-90.

13. Ferron M, Lacombe J. Regulation of energy metabolism by the skeleton: osteocalcin and beyond. Arch Biochem Biophys. 2014;561:137-46. 14. Lu C, Ivaska KK, Alen M, Wang Q, Törmäkangas T, Xu L, et al. Serum osteocalcin is not associated with glucose but is inversely associated with leptin across generations of nondiabetic women. J Clin Endocrinol Metab. 2012;97(11):4106-14. 15. Lucey AJ, Paschos GK, Thorsdottir I, Martínéz JA, Cashman KD, Kiely M. Young overweight and obese women with lower circulating osteocalcin concentrations exhibit higher insulin resistance and concentrations of C-reactive protein. Nutr Res. 2013;33(1):67-75. 16. Barbieri MA, Bettiol H, Silva AA, Cardoso VC, Simões VM, Gutierrez MR, et al. Health in early adulthood: the contribution of the 1978/79 Ribeirão Preto birth cohort. Braz J Med Biol Res. 2006;39(8):1041-55. 17. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381-95. 18. Ribeiro AB, Cardoso MA. Construção de um questionário de frequência alimentar como subsídio para programas de Arch Endocrinol Metab. 2018;62/3

31. Suliburska J, Bogdanski P, Gajewska E, Kalmus G, Sobieska M, Samborski W. The association of insulin resistance with serum osteoprotegerin in obese adolescents. J Physiol Biochem. 2013;69(4):847-53. 32. Gannagé-Yared MH, Yaghi C, Habre B, Khalife S, Noun R, Germanos-Haddad M, et al. Osteoprotegerin in relation to body weight, lipid parameters insulin sensitivity, adipocytokines, and C-reactive protein in obese and non-obese young individuals: results from both cross-sectional and interventional study. Eur J Endocrinol. 2008;158(3):353-9. 33. Bernardi S, Fabris B, Thomas M, Toffoli B, Tikellis C, Candido R, et al. Osteoprotegerin increases in metabolic syndrome and promotes adipose tissue proinflammatory changes. Mol Cell Endocrinol. 2014;394(1-2):13-20. 34. D’Amelio P, Sassi F, Buondonno I, Spertino E, Tamone C, Piano S, et al. Effect of intermittent PTH treatment on plasma glucose in osteoporosis: A randomized trial. Bone. 2015;76:177-84. 35. Fornari R, Francomano D, Greco EA, Marocco C, Lubrano C, Wannenes F, et al. Lean mass in obese adult subjects correlates with higher levels of vitamin D, insulin sensitivity and lower inflammation. J Endocrinol Invest. 2015;38(3):367-72.

295

REFERENCES

original article

Accuracy of sentinel lymph node mapping in detecting occult neck metastasis in papillary thyroid carcinoma Jose Higino Steck1, Elaine Stabenow1, Gustavo Baldove Bettoni1, Samuel Steck1, Claudio Roberto Cernea2

ABSTRACT

Clínica Onccape, Serviço de Cirurgia de Cabeça e Pescoço, Campinas, SP, Brasil 2 Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo (HCFMUSP), Serviço de Cirurgia de Cabeça e Pescoço, São Paulo, SP, Brasil 1

Objectives: The objectives of this study were to evaluate the following: 1) the accuracy of sentinel lymph node mapping (SLNM) in detecting metastasis in papillary thyroid carcinoma (PTC), and 2) if SLNM could modify the American Joint Committee on Cancer (AJCC) staging of previous cN0 PTC patients. Subjects and methods: Forty SLNM were performed prospectively in 38 consecutive cN0 PTC patients, with total thyroidectomy and elective compartment neck dissection (CND). The results of SLNM were compared with CND pathological findings to verify the accuracy of sentinel SLNM. Results: The mean patients’ follow-up was 36 months. A total of 133 SLN were found at levels VI, II, III and IV. The SLN was identified in 95% of the patients with one false negative, 95% sensitivity, a 94% negative predictive value and 97% accuracy. The SLNM upstaging from cN0 to pN+ was 49%, and to stages III and IVa, it was 21%. Conclusions: For this series of cN0 PTC patients: 1) SLNM accuracy was 97%, and 2) SLNM upstaging from cN0 to pN+ was 49%, whereas to stages III and IVa, it was 21%. Arch Endocrinol Metab. 2018;62(3):296-302

Correspondence to: Jose Higino Steck Clínica Onccape, Serviço de Cirurgia de Cabeça e Pescoço Av. Andrade Neves, 295 13013-160 – Campinas, SP, Brasil higinosteck@gmail.com

Keywords Thyroid gland; papillary cancer; sentinel lymph node biopsy; lymphatic metastases; neck dissection

Received on Sep/27/2017 Accepted on Dec/13/2017 DOI: 10.20945/2359-3997000000038

INTRODUCTION

apillary thyroid carcinoma (PTC) is the most common endocrine neoplasm and has shown a global increase in its prevalence in the past three decades (1). The incidence of lymphatic metastasis in PTC at initial diagnosis varies from 30% to 90% in different reports (2,3), and they are more commonly found in the central compartment. However, most lymph nodes harbor micrometastasis, which is not detectable clinically or radiologically. A controversy exists in the literature regarding the necessity of elective or prophylactic central compartment neck dissection (CND) in patients classified as cN0 (neck clinically without metastasis) PTC, according to the American Joint Committee on Cancer (AJCC) (4,5). Uncertainty about the real impact of this metastasis in the prognosis is one of the reasons for this controversy. In addition, an increased risk of postoperative complications is related to elective CND, such as hypoparathyroidism (3). Therefore, a less invasive method, such as sentinel lymph node mapping (SLNM), may be used to evaluate 296

the true pathologic status of clinically normal lymph nodes in these patients. SLNM has been widely used in patients with cutaneous melanoma, breast cancer and other malignant solid neoplasms, to stage them and to indicate lymph node dissection only in cases with histologically proven metastasis. The SLNM has the potential of avoiding CND and its risks to the neck with no metastasis (pN0). Even with no previous indication of elective CND, SLNM can be used to stage the neck and to suggest further treatment with radioiodine therapy if the clinical team considers this to be necessary. SLNM has potential advantages when compared with elective CND: It helps one to avoid the morbidity of CND in negative necks and to better stage the lateral compartment when lateral drainage is present. The objectives of this study were to verify the accuracy of SLNM in detecting occult neck lymph node metastasis in PTC, and to evaluate if the SLNM could modify the stage grouping of patients with PTC cN0. Arch Endocrinol Metab. 2018;62/3

Sentinel node mapping and thyroid carcinoma

After institutional review board approval and written informed consent were obtained, 38 consecutive patients with thyroid nodes were evaluated prospectively. Two of them had bilateral thyroid nodes, so were performed a total of 40 SLNM. The inclusion criteria were as follows: fine needle aspiration (FNA) cytology classified as Bethesda V or VI, clinically N0, evaluated with ultrasonography, at Mario Gatti Hospital in Campinas, Brazil, between 2010 and 2015. The exclusion criteria were as follows: 1) previous neck surgery or radiotherapy, 2) final histopathological exams without thyroid neoplasm confirmation and 3) patients without radioisotope lymphatic drainage. All 38 patients underwent the preoperative ultrasound-guided intratumoral injection of a radioisotope (99mTc-phytate). In 23% of procedures (9/40 procedures), the injection was given the evening before surgery, and the other 77% of procedures were injected in the operating room. The volume of the isotope injected ranged from 0.02 to 0.10 ml, with higher activity for those injected the evening before surgery (400 to 1200 microCi). Total thyroidectomy was performed before the SLNM in all cases to eliminate shine through effect and to facilitate SLN identification. Following total thyroidectomy, the radio guided SLNM was performed following the standard technique adopted for melanoma. The central and lateral compartments were scanned bilaterally with gamma probe Neoprobe 2000® (Neoprobe Corporation, Cincinnati, OH, USA); a collimator was used to obtain a little thickness (7.5 mm) of the detection center. The contralateral side was used to quantify the background (BG) radioactivity. Next, nodes with three times or more radioactivity than the BG were located and removed. Other nodes were removed if they reached 10% of the hottest SLN count (10% rule) (6). The average time course of mapping was 24.5 minutes, with the time range stretching from six to 50 minutes. After the completion of SLNM, elective conventional CND was performed. The final histopathological report of the SLN was compared with the final pathology report of the specimen of the CND, to verify if SLNM was able to detect occult neck lymph node metastasis in cases with PTC. The SLN analysis was conducted through serial sections in paraffin blocks and hematoxylin and eosin Arch Endocrinol Metab. 2018;62/3

staining. After this analysis, patients were classified as the TNM system from the AJCC (5). To verify the effectiveness of SLNM in detecting lymphatic metastasis in PTC, the indexes used to evaluate the diagnostic tests were calculated: quantity of false negatives and positives, sensitivity, specificity, accuracy, and positive and negative predictive values. The statistical power was calculated using program G*Power 3.0.10, 2008, Universitat Kiel, Germany. For a comparison between the averages of two parametric sample populations, Student’s t-test was used (7,8), and for a comparison of the frequency, the Fisher’s exact test (9,10) with IBM SPSS® Statistics 19.0.0, 2010; Illinois, USA, was employed. Statistical significance was defined as p ≤ 0.05.

RESULTS A flowchart revealing the objectives and results is displayed in Figure 1.

38 patients 40 injections 38 with PTC

36 with PTC

Objective 1

Objective 2

4 procedures without drainage = 34

3 patients without drainage = 33

True positives + True negatives / 34

16 pacients cN0 with pN+ SLN

Accuracy 97%

Upstage 49%

(33/34)

(16/33)

PTC: papillary thyroid carcinoma.

Figure 1. Flowchart summarizing the casuistic, objectives and results.

Tc99 injection was performed in 38 patients; females were 90% (34/38) of them in a proportion of 9:1. The mean age of the casuistic was 44 ± 15 years (range: 20 – 76 years). The FNA biopsy of 40 thyroid nodes revealed that 72% (29/40) had malignant findings (Bethesda VI), and 28% (11/40) were suspicious for malignancy (Bethesda V). The histopathological final diagnosis confirmed papillary thyroid carcinoma in 95% (38/40) of biopsied thyroid nodes. The two cases without thyroid neoplasm were excluded for the metastasis analyses. The primary tumor mean diameter was 14 ± 9 mm (range: 3 - 43 mm). The 38 primary tumors were classified as pT1a: 42% (16/38), pT1b: 24% (9/38), 297

SUBJECTS AND METHODS

Sentinel node mapping and thyroid carcinoma

pT2: 8% (3/38), pT3: 24% (9/38) and pT4a: 2% (1/38). In 4/40 (10%) of the SLN scintigraphy procedures, no Tc99 lymphatic drainage occurred, so it was not possible to localize the SLN. These four procedures were excluded. In addition, one patient presented no drainage at the left-side node but exhibited drainage from the right-side injection. Thus, we had a total of 36 feasible SLNM in 35 patients. The mean size of these four tumors was 33 ± 9 mm compared with 11 ± 6 mm of the tumors with Tc99 drainage (Student’s t-test, p = 0.0001). The three largest tumors of the sample did not have any Tc99 drainage. Two of these four tumors without drainage presented positive CND and two negative without recurrence with 39 and 43 months of follow-up. Considering the 36 procedures with radioisotope drainage, two SLNM did not present any lymphoid tissue at the histologic evaluation. Therefore, the SLN was properly identified in 34 of 36 (95%) procedures in which SLNM was feasible. Thus, 95% of SLNM were effective. In 35/36 (97%) of the feasible SLNM, SLN was found in the central neck, and in 12/36 (33%), it was found in the lateral neck. One case revealed only a lateral SLN. In 4/36 (11%) of feasible mappings, SLN was at level II, 9/36 (25%) at level III and 2/36 (6%) at level IV. The mean number of SLN identified in each procedure was 4 ± 2 nodes and at the central neck the mean number was 3 ± 2 lymph nodes (range: 1 to 8 nodes). Considering cases with feasible SLNM and with confirmed PTC (34 procedures), 16/34 (47%) had a metastatic SLN in the central neck, and 4/12 (33%) of the SLNMs had drainage to the lateral neck. Considering procedures with positive SLN in the central neck, only 9/16 (56%) presented other positive nodes in the CND specimen, and 3/16 (19%) presented node metastases detected in the lateral compartment as well. One patient presented skip metastasis (positive lateral neck lymph node with no positive central compartment lymph node). Surgical complications were as follows: vocal fold paralysis in 4/76 (5%) laryngeal nerves at risk, 3/76 (4%) with temporary paralysis and 1/76 (1%) with definitive paralysis. All paralyses were unilateral. Transient hypoparathyroidism occurred in 3/38 (8%) patients, and definitive hypoparathyroidism was observed in 1/38 patients (3%). Resected parathyroid glands were detected at the final pathology report: only 9% (3/32) were found on 298

SLNM specimens, whereas the other 91% were found on CND specimens. Thus, the risk of inadvertently resected parathyroid glands was 10 times higher in CND specimens compared with SLNM specimens. The mean follow-up period was 36 ± 13 months (range: 3 - 56 months). A minimum follow-up of 24 months was achieved in 83% (30/36) of the cases with PTC. The specific disease survival was 100%, and the disease-free survival was 89 ± 11%. Only one patient developed recurrence in the contralateral neck after 46 months.

SLNM accuracy The comparison between the SLNM and the final pathologic classification of cervical lymph nodes is shown in Table 1. For these analyses, were excluded the two procedures without confirmed papillary carcinoma and the four SLNM without radioisotope lymphatic drainage. Table 1. Comparison between sentinel lymph node mapping results with the final pathologic classification of cervical lymph nodes in patients treated for papillary thyroid carcinoma pN status

SLN mapping pathologic status Positive**

Total

Negative**

pN positive

17 (94%)

1 (6%)

18 (100%)

pN negative

0 (0%)

16 (100%)

17 (50%)

34 (100%)

Total

SLN: sentinel lymph node; pN: pathological status of the cervical lymph nodes. ** Fisher’s exact test: p = 0.0001.

The values used to evaluate the diagnostic tests were calculated to determine the diagnostic ability of the SNLM to detect neck node metastases in PTC (Table 2). The statistical power was 95%. Table 2. Diagnostic test evaluation values of the sentinel lymph node mapping to detect lymph node metastasis of papillary thyroid carcinoma Values False positive

0 (0%)

False negative

1 (3%)

Sensitivity

94%

Specificity

100%

Predictive positive value

100%

Predictive negative value

94%

Accuracy

97%

The false-negative case revealed no lymphoid tissue in the histologic SLN specimen. However, microscopically positive metastatic nodes were detected in the CND pathology report (1/5). This case involved Arch Endocrinol Metab. 2018;62/3

Sentinel node mapping and thyroid carcinoma

55 50

Positivity rate (%)

40 35 30

25 20 15 10 5 0

SLN

NSLN

Figure 2. Comparison between the mean of sentinel lymph node positivity rate of papillary thyroid carcinoma metastasis (and 95% confidence interval) and the non-sentinel lymph node positivity rate (Student’s t-test, p = 0.031).

Upstaging after SLNM For this analysis, we considered the number of patients instead of procedures. Of the 38 initial patients, the following were excluded: two without confirmed carcinoma and three without any radioisotope lymphatic drainage. As all patients included in this study were classified as clinically N0, SLNM upstaging was 49% (16/33). Final pN classification, based only upon the SLNM, was as follows: 17/33 (52%) pN0; 12/33 (36%) pN1a and 4/33 (12%) pN1b. For CND specimens, the final N classifications were 50% (18/36) pN0 and 50% (14/36) pN1. The mean age of the patients with a positive SLN was 38 ± 14 years compared with 49 ± 13 years for those with negative SLN (Student’s t-test, p = 0.015). Arch Endocrinol Metab. 2018;62/3

A negative correlation was found between age and positive SLN number (Pearson’s correlation coefficient = -0.34, p = 0.039). The postoperative TNM staging of the 36 patients with PTC was: stage I, 72% (26/36); stage II, 3% (1/36); stage III, 8% (3/36) and stage IVA, 17% (6/36).

DISCUSSION PTC has a high frequency of occult lymph node metastases in various series (2), confirmed in the present series as 50%. Although still a subject of controversy, the prognostic value of lymph node metastases was revealed in some recent studies, and it has been taken into consideration for determining the AJCC’s TNM staging system, which designates the mortality risk, and the American Thyroid Association risk stratification, which states the recurrence risk (5,11,12). Therefore, there has been an increasing interest in determining the lymph node involvement in PTC to appropriately stage the patient, with the goal being to decrease the locoregional recurrence rate (12). Ultrasonography of the neck is considered the best method for staging the neck before surgery, but the sensitivity for detecting metastasis in level VI when the thyroid gland is still present is low, 35 to 50% (13,14). The most sensitive method for determining the presence of positive nodes in the central compartment in PTC cN0 is CND, which can be performed electively at the same time as thyroidectomy (15,16). Some meta-analysis indicated a higher recurrence-free survival rate in patients undergoing CND (17). However, a controversy exists in the literature regarding whether the elective CND must be indicated in all patients with PTC, due to increased complications, especially transient hypoparathyroidism. Moreover, not all PTC patients receive treatment from experienced surgeons, which is essential for reducing the complication rate. In fact, significant discrepancies can be found among the various evidence-based guidelines of the important societies of experts involved in the treatment of PTC. Some suggest elective CND only for advanced cases of PTC (T3 and T4), whereas others recommend it for all cases (11,18,19). In the present study, CND was performed in all patients after SLNM, to compare the SLN status with the remaining CND and to meticulously evaluate its accuracy. In the present study, SLNM provided the same information for lymph node staging as CND did. Thus 299

a 42-year-old man, with a 10mm PTC macroscopically extended to the trachea, classified as T3N1aM0. He underwent radioiodine therapy and was followed for 47 months. He presented recurrence at the lateral neck nodes 46 months after surgery and was treated via neck dissection. To determine the density of the SLN positivity, the SLN positivity ratio (PR) was idealized and calculated by dividing the number of positive SLN by the total number of SLN. The PR was obtained for non-SLN too. The mean SLN PR (35 ± 7%) was higher compared with the non-SLN PR (16 ± 5%) (p = 0.031) (Figure 2).

Sentinel node mapping and thyroid carcinoma

SLNM could adequately select patients who would benefit from CND. If the SLN was histologically negative, CND did not need to be performed. On the other hand, if the SLN was positive, the indication for the CND was no longer elective (or prophylactic) but rather therapeutic. This was an initial study whose aim was to compare the sentinel lymph node content with the histological examination of the neck dissection, the gold standard method for detecting the presence of neck metastasis, to establish SLNM feasibility in the neck. We will continue the study with an intraoperative frozen section investigation of the SLN. Thus, in the same operative procedure, we will be able to perform neck dissection only for patients with positive sentinel lymph node. Immunohistochemistry could be used for a postoperative micrometastasis search, although the impact of these micrometastases on the prognosis is not well established. Some features of the central compartment of the neck can lead to greater technical difficulty for SLNM, such as the small sizes of the lymph nodes in the region (20), and the proximity of the site of radioisotope injection in the primary tumor. Therefore, the first procedure must be thyroidectomy, to reduce background radioactivity, thus preventing the so-called shine-through effect of the injection site, which could hide nearby lymph nodes. Three meta-analyses (21-23) evaluated the SLNM technique for PTC. The vast majority of studies employed only blue dye to map the SLN. In the two largest and most recent meta-analyses, only six studies using the radioisotope were included (22,23). The injection site of the radioisotope, intra or peritumoral, did not have any influence on the SLN detection rate (23). All meta-analyses concluded that using only the blue dye would be less effective for detecting the SLN than the radioisotope, alone or in combination with the blue dye. The techniques using the radioisotope had higher detection rates (83% with the blue dye compared with 96 to 98% with the radioisotope) (21). In the present study with the use of the radioisotope, the SLN was detected in 94.5% of SLNM. After radioisotope injection, 4/40 SLNM did not demonstrate any lymphatic drainage. These four patients had an average diameter of the primary tumor that was three times larger than that of those with lymphatic drainage. Therefore, although larger primary tumors usually have a higher risk of metastasis, in this 300

series, they had a greater chance of drainage failure. It is possible that in some of these tumors, the lymphatic vessels were already blocked via neoplastic cell invasion, impairing the migration of the radioisotope. SLN drainage difficulty and the increased risk of false negatives have been reported when the SLNM is used for locally advanced tumors from other locations, such as oral cancer (24).

SLN accuracy In 97% of the SLNM, the SLN was located in the central compartment and in lateral compartment in 33%. In one case, drainage to the lateral compartment was found without concomitant drainage to the central compartment. Another patient had central and lateral drainage, but metastasis was confirmed only in the lateral SLN. Thus, another advantage of SLNM is to detect these unusual drainage patterns to the lateral compartment at the time of the initial treatment (skip metastases) (25). In half of the SLNM, proven SLN metastasis was discovered. Additional metastases to non-sentinel lymph nodes were found in 56% of CND specimens. One false negative SLN (negative SLN with other positive lymph nodes in the central compartment) existed. In this patient, the hot SLN contained, in fact, only thymus fragments in the final pathology report. This was the only patient of the series who had lymph node recurrence on the contralateral lateral neck. Sensitivity was 94.4% with a specificity of 100%, the positive predictive value was 100% and the negative predictive value of 94.1%. The accuracy of SLNM in detecting lymph node metastases in PTC was 97.1%. These results are comparable to the best data published in the literature, although few authors who reported the use of the radioisotope have been able to calculate the accuracy of the method, as the gold standard of pN status of the whole CND specimen was not available in all cases. In the meta-analysis of Balasubramanian and Harrison (22), which analyzed 24 studies, the SLNM accuracy was evaluated only in 15. Of these, however, 13 used only the blue dye, and the SLNM sensitivity of this group was 91.6%, with 100% specificity, 95.8% accuracy and a false negative rate of 7.7%. Two metaanalyses evaluated the use of the radioisotope, with or without blue dye, with a sensitivity of 67%, specificity of 100%, accuracy of 88% and false negative of 16%. Another study of 25 patients featured a sensitivity of Arch Endocrinol Metab. 2018;62/3

Sentinel node mapping and thyroid carcinoma

Upstaging The SLNM could improve thyroid cancer staging, as it had better sensitivity and accuracy compared with ultrasound. During the surgery, the accuracy of SLNM in detecting lymph node metastasis was the same as that of the CND, and it was better than surgeon-performed intraoperative palpation (22,26). The presence of positive SLN metastasis in the central compartment increased the cN0 patient’s staging to pN1a (39%). The SLNM was able to make the diagnosis of occult metastases in the lateral compartment, which is an advantage compared with elective CND, which evaluates only the central compartment, with the potential to increase the staging from cN0 to pN1b when the lateral SLN is positive (11%). In the present study, if only CND had been analyzed, the lymph node upstaging would be 50%, similar to SLNM upstaging (49%). The CND staging, however, would be in the central compartment only (pN1a). Without SLNM, information on the lateral compartment would have been lost in this series in 11% of the procedures. Thus, we considered that the SLNM was at least as effective as the CND was for properly staging the central neck, with the advantage of providing information about the lateral compartment. In this series, it was not possible to assess the complications of the SLNM procedure because CND Arch Endocrinol Metab. 2018;62/3

was performed in all patients. However, at least in theory, the main complication of CND, which is hypoparathyroidism, would be reduced in SLNM. In fact, only three out of a total of 32 (9.4%) parathyroid glands were removed accidentally via SLNM, whereas the other (90.6%) were removed in the CND specimen, a 9.6 times greater risk of resection in the CND versus SLNM. This upstaging in the lymph node status could be particularly important in patients older than 45 years old. In this age group, the AJCC staging would be III or IV in the presence of metastases, maybe with a negative prognostic impact (27). In the present series, the upstaging to stage III or IV in this age group was observed in 21% of the cases. Interestingly, occult metastases to SLN were found more frequently among young patients; yet, in this age group, they do not affect the AJCC staging. In conclusion, this study revealed that the SLN concept applies to PTC just as in other solid tumors. Larger tumors were less likely to have lymphatic drainage. The accuracy of SLNM in detecting cervical lymph node metastases in PTC was 97%. The sensitivity was 94.4%, with a specificity of 100%, a positive predictive value of 100% and a negative predictive value of 94.1%. SLNM upstaging from cN0 to pN+ PTC patients was 49%, and to AJCC stages III or IVa, upstaging made up 21% of the cases. The accuracy was comparable to CND but with the advantage of also staging the lateral neck compartment and theoretically with less risk of hypoparathyroidism than the CND. Acknowledgments: we are grateful to all patients participating in this study. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1.

Pellegriti G, Frasca F, Regalbuto C, Squatrito S, Vigneri R. Worldwide increasing incidence of thyroid cancer: update on epidemiology and risk factors. J Cancer Epidemiol. 2013;2013:965212.

2. Caron NR, Clark OH. Papillary thyroid cancer: surgical management of lymph node metastases. Curr Treat Options Oncol. 2005;6(4):311-22. 3. Soto Diaz L, Steck JH. Indicações e técnicas do esvaziamento do compartimento central do pescoço no câncer da tireoide. In: Volpi EM, Steck JH, editores. Cirurgia da tireoide e paratireoide. São Paulo: Editora G Press; 2013. p 187-96. 4. Amdur RJ, Mazzaferri EL. Essentials of thyroid cancer management. New York: Springer; 2005.

301

100%, specificity of 100%, accuracy of 100% and false negative of 0%. This variability of the accuracy can stem from the use of various techniques and the SLNM learning curve in thyroid surgery. In the present study, the team’s experience with the use of SLNM in other cancers, combined with experience in thyroid surgery, may have led to the good accuracy level. In this series, frozen sections were not performed because, by protocol, all patients would be subjected to CND anyway. The authors plan to evaluate the efficacy of SLNM with frozen sections to determine the need for CND in the same surgery. In the literature, frozen sections present a false negative rate of 5-12% in detecting lymph node metastases in PTC (22). SLN’s positive ratio (PR) was higher (35%) than the non-SLN PR (16%) was, suggesting that the concept of SLN would apply to the PTC similarly as in other solid tumors. When metastasis is present, they would be more likely to be found in the SLN than in any other neck node.

Sentinel node mapping and thyroid carcinoma

5. Sobin LH, Gospodarowicz MK, Wittekind C. TNM classification of malignant tumours. New York: Wiley-Blackwell; 2009. 6. Emery RE, Stevens JS, Nance RW, Corless CL, Vetto JT. Sentinel node staging of primary melanoma by the “10% rule”: pathology and clinical outcomes. Am J Surg. 2007;193(5):618-22. 7. Dawson B, Trapp RG. Basic and clinical biostatistics. Illinois: Norwalk Lange; 2004. 8. Rosner B. Fundamentals of biostatistics. Massachusetts: Duxburg Press; 2006. 9. Agresti A. Categorical data analysis. New York: John Wiley & Sons; 1990. 10. Stewart J. Cálculo. São Paulo: Pioneira Thomson Learning; 2002. 11. 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. 12. Podnos YD, Smith D, Wagman LD, Ellenhorn JD. The implication of lymph node metastasis on survival in patients with welldifferentiated thyroid cancer. Am Surg. 2005;71(9):731-4. 13. Choi JS, Kim J, Kwak JY, Kim MJ, Chang HS, Kim EK. Preoperative staging of papillary thyroid carcinoma: comparison of ultrasound imaging and CT. AJR Am J Roentgenol. 2009;193(3):871-8. 14. Morita S, Mizoguchi K, Suzuki M, Iizuka K. The accuracy of (18) [F]-fluoro-2-deoxy-D-glucose-positron emission tomography/ computed tomography, ultrasonography, and enhanced computed tomography alone in the preoperative diagnosis of cervical lymph node metastasis in patients with papillary thyroid carcinoma. World J Surg. 2010;34(11):2564-9. 15. Moreno MA, Edeiken-Monroe BS, Siegel ER, Sherman SI, Clayman GL. In papillary thyroid cancer, preoperative central neck ultrasound detects only macroscopic surgical disease, but negative findings predict excellent long-term regional control and survival. Thyroid. 2012;22(4):347-55.

27. Zaydfudim V, Feurer ID, Griffin MR, Phay JE. The impact of lymph node involvement on survival in patients with papillary and follicular thyroid carcinoma. Surgery. 2008;144(6):1070-7.

16. Shindo M, Wu JC, Park EE, Tanzella F. The importance of central compartment elective lymph node excision in the staging and treatment of papillary thyroid cancer. Arch Otolaryngol Head Neck Surg. 2006;132(6):650-4.

17. Lang BH, Ng SH, Lau LL, Cowling BJ, Wong KP, Wan KY. A systematic review and meta-analysis of prophylactic central neck dissection on short-term locoregional recurrence in papillary thyroid carcinoma after total thyroidectomy. Thyroid. 2013;23(9):1087-98. 18. Takami H, Ito Y, Okamoto T, Yoshida A. Therapeutic strategy for differentiated thyroid carcinoma in Japan based on a newly established guideline managed by Japanese Society of Thyroid Surgeons and Japanese Association of Endocrine Surgeons. World J Surg. 2011;35(1):111-21. 19. Sancho JJ, Lennard TW, Paunovic I, Triponez F, Sitges-Serra A. Prophylactic central neck disection in papillary thyroid cancer: a consensus report of the European Society of Endocrine Surgeons (ESES). Langenbecks Arch Surg. 2014;399(2):155-63. 20. Ofo E, Thavaraj S, Cope D, Barr J, Kapoor K, Jeannon JP, et al. Quantification of lymph nodes in the central compartment of the neck: a cadaveric study. Eur Arch Otorhinolaryngol. 2016;273(9):2773-8. 21. Raijmakers PG, Paul MA, Lips P. Sentinel node detection in patients with thyroid carcinoma: a meta-analysis. World J Surg. 2008;32(9):1961-7. 22. Balasubramanian SP, Harrison BJ. Systematic review and metaanalysis of sentinel node biopsy in thyroid cancer. Br J Surg. 2011;98(3):334-44. 23. Kaczka K, Celnik A, Luks B, Jasion J, Pomorski L. Sentinel lymph node biopsy techniques in thyroid pathologies – a meta-analysis. Endokrynol Pol. 2012;63(3):222-31. 24. Civantos FJ, Zitsch RP, Schuller DE, Agrawal A, Smith RB, Nason R, et al. Sentinel lymph node biopsy accurately stages the regional lymph nodes for T1-T2 oral squamous cell carcinomas: results of a prospective multi-institutional trial. J Clin Oncol. 2010;28(8):1395-400. 25. Chow TL, Lim BH, Kwok SP. Sentinel lymph node dissection in papillary thyroid carcinoma. ANZ J Surg. 2004;74(1-2):10-2. 26. Pelizzo MR, Toniato A, Sorgato N, Losi A, Torresan F, Merante Boschin I. 99Tc nanocolloid sentinel node procedure in papillary thyroid carcinoma: our mono-institutional experience on a large series of patients. Acta Otorhinolaryngol Ital. 2009;29(6):321-5.

302

Arch Endocrinol Metab. 2018;62/3

original article

Estimated costs of hospitalization due to coronary artery disease attributable to familial hypercholesterolemia in the Brazilian public health system Luciana R. Bahia1, Roger S. Rosa2, Raul D. Santos3, Denizar V. Araujo1

ABSTRACT

Arch Endocrinol Metab. 2018;62(3):303-8 Keywords Costs and cost analysis; coronary artery disease; familial hypercholesterolemia; Brazil; hospitalization

INTRODUCTION

ardiovascular diseases (CVDs) are the main cause of death in Brazil, with coronary artery disease (CAD) being the main cause among all CVDs (31%), followed by cerebrovascular disease (30%), and heart failure (18%). A significant reduction in the mortality associated with CVDs has been demonstrated over the last decade, possibly due to the control of some risk factors like smoking and greater access to health care. However, the mortality associated with other factors, such as obesity and diabetes, is still out of control or on the rise and is also influenced by population aging. In 2012, CVD was responsible for approximately 940,000 hospitalizations in the Brazilian Unified Health Care System (SUS) and, in relative terms, accounted for 8.3% of all causes of hospitalizations and 18.6% of costs reimbursed. Arch Endocrinol Metab. 2018;62/3

Departamento de Medicina Interna, Universidade do Estado do Rio de Janeiro (UERJ); Instituto de Avaliação de Tecnologia em Saúde (IATS), Rio de Janeiro, RJ, Brasil 2 Departamento de Medicina Social, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brasil 3 Unidade Clínica de Lípides, Instituto do Coração do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (InCor/HCFMUSP); Programa de Centro de Medicina Preventiva e Cardiologia, Hospital Albert Einstein (HIAE), São Paulo, SP, Brasil 1

Correspondence to: Luciana R. Bahia lucianabahia@gmail.com Received on Mar/21/2017 Accepted on Oct/31/2017 DOI: 10.20945/2359-3997000000030

The economic impact of CVDs in public and private health care systems has been the subject of some national studies (1-3). These costs are high due to requirements from patients with CVDs of hospitalization, diagnostic procedures and revascularization, medical follow-up, and chronic use of several medications. Familial hypercholesterolemia (FH) is a genetic disease that affects the lipoprotein metabolism with an autosomal dominant inheritance that is usually characterized by a two- to four-fold increase in blood low-density lipoprotein (LDL-c) concentration. FH is a serious disease that greatly increases the risk of premature CAD, accounting for 5-10% of coronary events occurring before the age of 50 years (4). In the absence of treatment, young FH carriers present a 90fold increased mortality rate (5). 303

Objective: Cardiovascular diseases are the leading cause of death in Brazil, imposing substantial economic burden on the health care system. Familial hypercholesterolemia (FH) is known to greatly increase the risk of premature coronary artery disease (CAD). This study aimed to estimate the economic impact of hospitalizations due to CAD attributable to FH in the Brazilian Unified Health Care System (SUS). Subjects and methods: Retrospective, cross-sectional study of data obtained from the Hospital Information System of the SUS (SIHSUS). We selected all adults (≥ 20 years of age) hospitalized from 2012-2014 with primary diagnoses related to CAD (ICD-10 I20 to I25). Attributable risk methodology estimated the contribution of FH in the outcomes of interest, using international data for prevalence (0.4% and 0.73%) and relative risk for events (RR = 8.56). Results: Assuming an international prevalence of FH of 0.4% and 0.73%, of the 245,981 CAD admissions/year in Brazil, approximately 7,249 and 12,915, respectively, would be attributable to an underlying diagnosis of FH. The total cost due to CAD per year, considering both sexes and all adults, was R$ 985,919,064, of which R$ 29,053,500 and R$ 51,764,175, respectively, were estimated to be attributable to FH. The average cost per FH-related CAD event was R$ 4,008. Conclusion: Based on estimated costs of hospitalization for CAD, we estimated that 2.9-5.3% are directed to FH patients. FH can require early specific therapies to lower risk in families. It is mandatory to determine the prevalence of FH and institute appropriate treatment to minimize the clinical and economic impact of this disease in Brazil.

Familial hypercholesterolemia hospitalizations

The diagnosis of FH is complex, and the exact prevalence of this disease is unclear in several populations, including the Brazilian one. The World Health Organization estimates about 10 million individuals with FH in the world; however, less than 10% of these have a known diagnosis of FH, and less than 25% receive lipid-lowering treatment (6). This study aimed to estimate the impact of FH on hospitalizations due to CAD using the population attributable risk methodology and cost data from the hospital admission system of the Brazilian SUS.

SUBJECTS AND METHODS This was a cross-sectional and descriptive study of secondary data related to hospital reimbursement obtained from the SUS database (DATASUS). The data sources studied were the abridged files of the Hospital Information System of the SUS (SIHSUS), which controls payments for services provided by hospitals (7). Data were obtained from the standard forms for hospital admission authorization (AIH), used by public managers for prospective payment based on diagnosticrelated groups (DRG), with some adaptations (8). The number of hospitalizations is slightly greater than that of hospitalized individuals, since the same individual may be hospitalized more than once for the same reason during the considered period. Hospitalizations with a primary diagnosis (reason for hospitalization) related to CAD of possible atherosclerotic origin were identified from all hospitalizations of adults (≥ 20 years) in the SUS between 2012 and 2014. The selected international codes of diseases (ICD 10) were: I20 (angina pectoris), I21 (acute myocardial infarction), I24 (other acute ischemic heart diseases), I22 (recurrent myocardial infarction), I23 (certain complications following acute myocardial infarction), and I25 (chronic ischemic heart disease). In order to stabilize annual fluctuations, we determined the average number of hospitalizations and hospital deaths in the triennium 2012-2014 by region of residence (North, Northeast, Southeast, South, and Central-West), sex, and age groups (20-44, 45-64, 6574, and 75 years or older). Costs are expressed in Brazilian reais (R$), which were converted to international dollar (Intl$) through a division by a correction factor (1.646) defined by the World Bank for the year 2013 (Purchasing Power Parity, PPP) (9). 304

Attributable fraction due to familial hypercholesterolemia The attributable risk methodology has been widely used to overcome the problem generated by the absence or underestimation of information regarding the diagnosis to be studied. This method is based on the relative risk or odds ratio for a medical condition (hospitalization for CAD) according to the presence or absence of a particular health condition (FH) and combines risk or reason with estimates of the proportion of the population with the health condition (FH prevalence) to calculate an etiological fraction. The etiologic fraction calculated in this study estimated the proportion of hospitalizations for CAD that would be attributable to the presence of FH. For calculation of the attributable fraction of the target group by sex and age group, the following formula was used (10):

RAPi = [ P x (RRi – 1) ] / [ P x (RRi – 1) + 1 ]

where RAPi is the population attributable risk fraction for the medical condition “i” due to hypercholesterolemia, P represents the prevalence of FH, and RRi is the relative risk or odds ratio of the medical condition “i” for individuals with FH compared with those without the disease.

Prevalence of familial hypercholesterolemia and risk of coronary events In the absence of national data on the prevalence of FH, we conducted a literature search and selected two population-based studies that estimated the prevalence of FH in the United States (unselected population of 36,949 individuals and prevalence of 0.4%, 95% confidence interval [95%CI] 0.32-0.48%) (11) and Denmark (unselected population of 69,106 individuals and prevalence of 0.73%) using the same diagnostic criteria (Dutch Lipid Clinic Network) (12,13). Considering that the prevalence of FH in Brazil is unknown and there is great racial and ethnic variability in the country, we opted to carry out the analyses based on the prevalence rates found in those two populations. The source of relative risk estimation was obtained from a Danish general population study (The Copenhagen General Population Study) (14). In this study, the odds ratio adjusted for other risk factors for coronary events was 13.2 (95%CI 10.0-17.4) for the group with a definitive or probable diagnosis of FH. Arch Endocrinol Metab. 2018;62/3

Familial hypercholesterolemia hospitalizations

For use in the attributable fraction formula, this value was converted into a relative risk of 8.56 (14).

(76.3%) and in the Southeast region (47.8%). Men had more hospitalizations for all diagnosis of CAD than women (60.4% vs. 40.6%). Of the total of 246,981 hospitalizations per year in individuals aged 20 years or older, 7,249 to 12,915 would be attributable to FH, assuming a prevalence of 0.4% or 0.73%, respectively. Of the total costs, 2.9-5.3% would be attributable to FH. The average cost per event was R$ 4,008 (R$ 4,254 for men and R$ 3,631 for women).

RESULTS Tables 1, 2, and 3 show the total number of hospitalizations for CAD and related costs in Brazil by sex, age group, and region of residence. Most hospitalizations (48%) occurred due to a diagnosis of angina (I20) in the age group between 45-74 years

Table 1. Hospitalizations due to coronary artery disease attributable to familial hypercholesterolemia in the population aged ≥ 20 years distributed by sex. Annual average values for the period of 2012-2014 (Brazilian Unified Health Care System, SUS) Hospitalizations

Coronary artery disease ICD-10

Men

Women

Cost per year (R$/Intl$)* Men

Cost per case (R$/Intl$)

Women

Men

Women

I20 Angina pectoris

68,745

49,863

294,160,412

171,880,995

4,278,99

3,447,04

I21 ST elevation and non-ST elevation myocardial infarction

54,787

31,576

189,979,637

100,366,337

3,467,60

3,178,53

I22 Subsequent ST elevation and non-ST elevation myocardial infarction

13,431

8,854

69,601,281

38,064,261

5,182,27

4,298,94

I23 Certain current complications following ST elevation and non-ST elevation myocardial infarction (within the 28 day period)

1,280

752

3,945,129

2,219,376

3,081,33

2,949,99

599

375

2,129,970

1,245,642

3,557,86

3,324,67

9,865

5,853

72,816,065

39,509,958

7,381,00

6,750,76

148,707

97,274

632,632,494 / 384,345,379

353,286,570 / 214,633,396

4,254 / 2,584

3,631 / 2,206

4,382

2,867

18,642,695 / 11,326,060

10,410,805 / 6,324,912

2.9

7,808

5,107

33,215,043 / 20,179,467

18,548,772 / 11,268,999

5.3

I24 Other acute ischemic heart diseases I25 Chronic ischemic heart disease Total Attributable to FH (prevalence 0.4%) (%) of total Attributable to FH (prevalence 0.73%) (%) of total

ICD-10: International Classification of Diseases, Tenth Revision; R$: Brazilian Real; Intl$: International Dollar; FH: Familial Hypercholesterolemia. * International dollar: PPP 2013; correction factor 1.646.

Table 2. Hospitalizations due to coronary artery disease attributable to familial hypercholesterolemia and distributed by age group. Annual average values for the period of 2012-2014 (Brazilian Unified Health Care System, SUS) Age groups (years)

Total

20-44

45-64

65-74

75+

18,012

123,306

64,426

40,237

245,981

48,956,113 / 29,742,474

508,160,232 / 308,724,321

286,604,042 / 174,121,532

142,198,678 / 86,390,448

985,919,064 / 598,978,775

Attributable to FH (prevalence 0.4%)

531

3,634

1,899

1,186

7,249

Coef/10,000/year

0.1

0.9

2.1

0.5

COSTS (R$/Intl$)*

1,442,660 / 876,464

14,974,691 / 9,097,625

8,445,775 / 5,131,091

4,190,374 / 2,545,792

29,053,500 / 17,650,972

946

6,474

3,383

2,113

12,915

Total CAD COSTS (R$/Intl$)*

Attributable to FH (prevalence 0.73%) Coef/10,000/year

0.1

1.6

3.7

1.0

COSTS (R$/Intl$)*

2,570,366 / 1,561,583

26,680,177 / 16,209,099

15,047,708 / 9,141,985

7,465,924 / 4,535,799

51,764,175 / 31,448,466

ICD-10: International Classification of Diseases, Tenth Revision; CAD: Coronary Artery Disease; R$: Brazilian Real; Intl$: International Dollar; FH: Familial Hypercholesterolemia; Coef/10,000/year: coefficient per 10,000 inhabitants per year. * International dollar: PPP 2013; correction factor 1.646. Arch Endocrinol Metab. 2018;62/3

305

ICD-10: I20 – I25

Familial hypercholesterolemia hospitalizations

Table 3. Hospitalizations for coronary artery disease attributable to familial hypercholesterolemia in the population aged ≥ 20 years distributed by region of residence. Annual average values for the period of 2012-2014 (Brazilian Unified Health Care System, SUS) ICD-10: I20 – I25

Brazilian Region

Total

North

Northeast

Southeast

South

Central-West

7,622

40,901

117,553

63,765

16,140

245,981

Coef/10,000/year

0.9

10.2

128.9

110.8

1.2

18.2

COSTS (R$/Intl$)**

25,708,081 / 15,618,518

151,315,900 / 91,929,465

475,074,666 / 288,623,733

271,105,593 / 164,705,706

62,714,824 / 38,101,351

985,919,064 / 598,978,775

Attributable to FH (prevalence 0.4%)

225

1,205

3,464

1,879

476

7,249

Coef/10,000/year*

0.2

0.3

0.6

0.9

0.5

COSTS (R$/Intl$)**

757,577 / 460,253

4,459,044 / 2,709,018

13,999,711 / 8,505,292

7,989,060 / 4,853,621

1,848,108 / 1,122,788

29,053,500 / 17,650,972

Attributable to FH (prevalence 0.73%)

400

2,147

6,172

3,348

847

12,915

Coef/10,000/year*

0.4

0.6

1.0

1.7

0.8

1.0

COSTS (R$/Intl$)**

1,349,764 / 820,026

7,944,610 / 4,826,616

24,943,070 / 15,153,749

14,233,985 / 8,647,622

3,292,746 / 2,000,453

51,764,175 / 31,448,466

Total CAD

ICD-10: International Classification of Diseases, Tenth Revision; CAD: Coronary Artery Disease; R$: Brazilian Real; Intl$: International Dollar; FH: Familial Hypercholesterolemia; Coef/10,000/year: coefficient per 10,000 inhabitants per year. * Not adjusted to age. ** International dollar: PPP 2013; correction factor 1.646.

DISCUSSION The present study analyzed hospitalizations in the Brazilian public health system for CAD and estimated the attributable fraction of FH in these hospitalizations. CAD was responsible for approximately 2.2% of all admissions to the public health system, with an amount paid for 245,981 hospitalizations of R$ 986 million per year (Intl$ 599 million), which is equivalent to 10.6% of all public hospitalization expenses in adults (approximately R$ 9,3 billion). Considering the prevalence of FH in American and European studies, we estimated that 2.9 to 5.3% of these hospitalizations and CAD expenses would be attributable to FH (R$ 29 to 51 million/year). As expected, we observed a higher prevalence of events and costs in males, between the ages of 45-64 years, and in the Southeast region. The largest share of costs was for the treatment of angina pectoris and acute myocardial infarction, although the cost per case was higher for individuals with recurrent myocardial infarction and chronic ischemic disease. Azambuja and cols. estimated the number of cases of severe cardiovascular disease (coronary heart disease, cerebrovascular disease, heart failure, and others) from the inpatient lethality and mortality rates and sick leaves and retirement data. Approximately 2 million cases of severe CVD were reported in 2004 in Brazil, 306

accounting for 5.2% of the population over the age of 35 years. The annual cost corresponded to 1.74% of the national gross domestic product (GDP) in 2004 and a potential loss of 0.97% of the GDP attributed to loss of productivity (15). Ribeiro and cols. demonstrated that hospitalizations for heart failure and coronary artery revascularization (angioplasty and surgeries) were responsible for the largest share of costs in the public health system in 2012 (2). Although the number of hospitalizations due to heart failure decreased between 2007 and 2012, there was a significant increase in the number of angioplasty procedures during the same period (1). Some studies on costs related exclusively to CAD have been conducted in Brazil. In 2011, Teich and Araujo estimated the costs of CAD from the perspectives of the public and private sectors, including indirect costs of productivity loss. Through an analysis of the historical series of hospitalizations for acute myocardial infarction, angioplasty, and revascularization (19982010) in the DATASUS database, the authors estimated the average costs per hospitalized patient to be R$ 5,236 and R$ 16,905 in the public and private health care systems, respectively (3). From the public perspective, the estimated values are higher than those shown in this study, possibly due to the inclusion of indirect costs attributed to loss of productivity associated with death Arch Endocrinol Metab. 2018;62/3

Familial hypercholesterolemia hospitalizations

Arch Endocrinol Metab. 2018;62/3

would underestimate the real economic impact of CAD in Brazil. The number of individuals with FH in Brazil is unknown, but is probably comparable to that in other countries, in which FH is reported to affect between 1/250 and 1/500 individuals (14). Importantly, Jannes and cols. demonstrated in a national cascade screening program (Hipercol Brasil) that early CAD was present in 12% of the index and family cases affected by FH, confirmed by molecular diagnosis. The high risk of development of ischemic events in younger age groups makes early diagnosis and treatment very important in order to minimize the great clinical and economic impact (17). Due to the high costs associated with the diagnosis and treatment of FH, international economic analyses have been conducted with the objective of estimating the cost-effectiveness of cascade genetic screening (18-20) and different therapeutic approaches for health care systems (21,22). However, it is important to consider that with the end of patents for statins and ezetimibe, which are still the treatments of choice for FH in our country, the cost of lipid-lowering treatment should reduce substantially (23). Cost estimates of hospitalizations are essential for economic evaluations and cost-effectiveness studies of several technologies for diagnosis and treatment, old or new, aimed at a more efficient management of ischemic heart disease in Brazil. Since FH is probably responsible for high costs secondary to CAD, the diagnosis and early intervention with cholesterol-lowering medications can have a strong impact in reducing the clinical and economic burden of the disease in the country. Authors’ contributions: all authors participated in the elaboration of the study protocol and discussed all stages of the analysis. LB and RSR performed the search and analysis of cost data; LB wrote the article; RSR, DVA, and RDS reviewed and approved the article. Funding: This study was funded by Sanofi. Disclosure: LB and RSR received research grants from Sanofi; DVA received consultancy and lectures fees from Sanofi, Genzyme, Roche, AstraZeneca, Novo Nordisk, Pfizer, and Teva; RDS received consulting and lectures fees from Amgen, AstraZeneca, Biolab, Boehringer-Ingelheim, Cerenis, Eli-Lilly, Genzyme, Kowa, MSD, Pfizer, Sanofi/Regeneron, Torrent, and Unilever.

REFERENCES 1. Araujo DV, Ferraz MB. Impacto econômico no tratamento da cardiopatia isquêmica crônica no Brasil: o desafio da incorporação

307

or recovery time following the event until work activity is resumed. This share of costs accounted for 73% of the total costs. Using primary collected data, Ribeiro and cols. estimated the costs of CAD in a group of 147 patients attending a referral center in public and private health care systems in a state in southern Brazil. They demonstrated annual costs of outpatient treatment of R$ 1,488 in the public and R$ 2,094 in the private health care system for the year 2002 (16). Our study has some limitations. In the absence of data on the prevalence of FH and risk of related events in our population and the fact that international data are not disaggregated by age, we used the same prevalence of FH for all age groups over 20 years, as well as the same relative risk for events. This prevented us from observing differences in the number of hospitalizations and costs between individuals in the general population and those with FH, as we had supposed by the early occurrence of events in this population. DATASUS is a database with a substantial number of national registries and coverage that reflects the amounts reimbursed to health care units but does not necessarily take into account all the resources used. This database was initially developed for administrativefinancial functions, and may not be free of coding and diagnostic errors, intentional or not. The data refer only to the hospitalizations of establishments owned or contracted by the public network (SUS), even though they represent an expressive part of the hospitalizations in the country. In addition, the unit of analysis is the hospitalization, not the patient; therefore, it does not include variables with explanatory potential such as age, regional differences, and severity of the clinical condition at the time of hospitalization. In addition, our data prevents us from evaluating the total cost attributable to FH in the entire country since we did not assess the impact of this disease on the private health care system. Chronic diseases carry high costs for the patients, their families, and the entire society due to lost productivity attributed to absenteeism, medical leave, early retirement, and premature death. For a more comprehensive and realistic estimate of the costs of a chronic disease, it is necessary to include such indirect costs, which was not done in the present analysis. In addition, the outpatient costs associated with pharmacological treatment, rehabilitation, medical examinations, and consultations were also not estimated, suggesting that the estimates obtained

Familial hypercholesterolemia hospitalizations

de novas técnicas cardiovasculares. Arq Bras Cardiol. 2005; 85:1-2. 2. Ribeiro RA, Mello RGB, Melchior R, Dill JC, Hohmann CB, Lucchese AM, et al. Custo anual do manejo da cardiopatia isquêmica crônica no Brasil: perspectiva pública e privada. Arq Bras Cardiol. 2005;85:3-8. 3. Teich V, Araujo DV. Estimativa de custo da síndrome coronariana aguda no Brasil. Rev Bras Cardiol. 2011;24(2):85-94. 4. Burnett JR, Ravine D, van Bockxmeer FM, Watts GF. Familial hypercholesterolaemia: a look back, a look ahead. Med J Aust. 2005;182(11):552-3. 5. Neil A, Cooper J, Betteridge J, Capps N, McDowell I, Durrington P, et al. Reductions in all-cause, cancer, and coronary mortality in statin-treated patients with heterozygous familial hypercholesterolaemia: a prospective registry study. Eur Heart J. 2008;29(21):2625-33. 6. World Health Organization. Familial hypercholesterolemia (FH). Report of a WHO consultation. Paris: WHO: Human Genetic Programme; 1997 October. (Report No. WHO/HGN/FH/CONS/98.7) 7. Brasil/MS/DATASUS (Brasil. Ministério da Saúde. Datasus). Transferência de arquivos. Arquivos de Dados. Seleção de Arquivos do BBS. Disponível em: http://datasus.gov.br. 8. Lessa FJD, Mendes ACG, Farias SF, de Sá DA, Duarte PO, Melo Filho DA. Novas metodologias para vigilância epidemiológica: uso do Sistema de Informações Hospitalares – SIH/SUS. Informe Epidemiológico do SUS 2000;9(S1):3-27. 9. World Bank. International Comparison Program database. Available at: http://data.worldbank.org/indicator/PA.NUS.PPP. 10. Benichou J. A review of adjusted estimators of attributable risk. Stat Methods Med Res. 2001;10(3):195-216. 11. de Ferranti SD, Rodday AM, Mendelson MM, Wong JB, Leslie LK, Sheldrick RC. Prevalence of Familial Hypercholesterolemia in the 1999 to 2012 United States National Health and Nutrition Examination Surveys (NHANES). Circulation. 2016;133(11): 1067-72. 12. Benn M, Watts GF, Tybjaerg-Hansen A, Nordestgaard BG. Familial hypercholesterolemia in the danish general population: prevalence, coronary artery disease, and cholesterol-lowering medication. J Clin Endocrinol Metab. 2012;97(11):3956-64.

15. Azambuja MIR, Foppa M, Maranhão MFC, Achutti AC. Impacto econômico dos casos de doença cardiovascular grave no Brasil: uma estimativa baseada em dados secundários. Arq Bras Cardiol. 2008;91(3):163-71. 16. Riberio RA, Mello RGB, Melchior R, Dill JC, Hohmann CB, Lucchese AM, et al. Annual Cost of Ischemic Heart Disease in Brazil. Public and Private Perspective. Arq Brasil Cardiol. 2005;85(1)3-8. 17. Jannes CE, Santos RD, de Souza Silva PR, Turolla L, Gagliardi AC, Marsiglia JD, et al. Familial hypercholesterolemia in Brazil: cascade screening program, clinical and genetic aspects. Atherosclerosis. 2015;238(1):101-7. 18. Marks D, Wonderling D, Thorogood M, Lambert H, Humphries SE, Neil HA. Screening for hypercholesterolaemia versus case finding for familial hypercholesterolaemia: a systematic review and costeffectiveness analysis. Health Technol Assess. 2000;4(29):1-123. 19. Chen CX, Hey JW. Cost-effectiveness analysis of alternative screening and treatment strategies for heterozygous familial hypercholesterolemia in the United States. Int J Cardiol. 2015;181:417-24. 20. Ademi Z, Watts GF, Pang J, Sijbrands EJ, van Bockxmeer FM, O'Leary P, et al. Cascade screening based on genetic testing is cost-effective: evidence for the implementation of models of care for familial hypercholesterolemia. Clin Lipidol. 2014;8(4):390-400. 21. Nherera L, Calvert NW, Demott K, Humphries SE, Neil HA, Minhas R, et al. Cost-effectiveness analysis of the use of a high-intensity statin compared to a low-intensity statin in the management of patients with familial hypercholesterolaemia. Curr Med Res Opin. 2010;26(3):529-36. 22. Alonso R, Fernández de Bobadilla J, Méndez I, Lázaro P, Mata N, Mata P. [Cost-effectiveness of managing familial hypercholesterolemia using atorvastatin-based preventive therapy]. Rev Esp Cardiol. 2008;61(4):382-93. 23. Santos RD, Gidding SS, Hegele RA, Cuchel MA, Barter PJ, Watts GF, et al.; International Atherosclerosis Society Severe Familial Hypercholesterolemia Panel. Defining severe familial hypercholesterolaemia and the implications for clinical management: a consensus statement from the International Atherosclerosis Society Severe Familial Hypercholesterolemia Panel. Lancet Diabetes Endocrinol. 2016 May 27. pii: S22138587(16)30041-9.

13. World Health Organization; Familial hypercholesterolaemia (FH). Report of a second WHO consultation. Geneva: World Health Organization; 1999.

14. Grant RL. Converting an odds ratio to a range of plausible relative risks for better communication of research findings. BMJ. 2014;348:7450.

308

Arch Endocrinol Metab. 2018;62/3

original article

Metabolic syndrome components are associated with oxidative stress in overweight and obese patients Nayara Rampazzo Morelli1, Bruna Miglioranza Scavuzzi1, Lucia Helena da Silva Miglioranza2, Marcell Alysson Batisti Lozovoy3, Andréa Name Colado Simão3, Isaias Dichi4

ABSTRACT Objective: The aim of this study is to evaluate the influence of the body mass index (BMI) and the metabolic syndrome (MetS) parameters on oxidative and nitrosative stress in overweight and obese subjects. Subjects and methods: Individuals were divided into three groups: the control group (G1, n = 131) with a BMI between 20 and 24.9 kg/m2, the overweight group (G2, n = 120) with a BMI between 25 and 29.9 kg/m2 and the obese group (G3, n = 79) with a BMI ≥ 30 kg/m2. Results: G3 presented higher advanced oxidation protein products (AOPPs) in relation to G1 and G2 (p = 0.001 and p = 0.011, respectively) whereas G2 and G3 had lower levels of nitric oxide (NO) (p = 0.009 and p = 0.048, respectively) compared to G1. Adjusted for the presence of MetS to evaluate its influence, the levels of AOPPs did not differ between the groups, whereas NO remained significantly lower. Data adjusted by the BMI showed that subjects with higher triacylglycerol levels had higher AOPPs (p = 0.001) and decreased total radical-trapping antioxidant parameter/uric Acid (p = 0.036). Subjects with lower high-density lipoprotein (HDL) levels and patients with higher blood pressure showed increased AOPPs (p = 0.001 and p = 0.034, respectively) and lower NO levels (p = 0.017 and p = 0.043, respectively). Subjects who presented insulin resistance had higher AOPPs (p = 0.024). Conclusions: Nitrosative stress was related to BMI, and protein oxidation and nitrosative stress were related to metabolic changes and hypertension. MetS components were essential participants in oxidative and nitrosative stress in overweight and obese subjects. Arch Endocrinol Metab. 2018;62(3):309-18 Keywords Overweight; obesity; metabolic syndrome; oxidative stress; nitrosative stress

1 Departamento de PósGraduação em Ciências da Saúde, Universidade Estadual de Londrina (UEL), Londrina, PR, Brasil 2 Departamento de Ciência e Tecnologia de Alimentos, Universidade Estadual de Londrina (UEL), Londrina, PR, Brasil 3 Departamento de Patologia, Análises Clínicas e Toxicológicas, Universidade Estadual de Londrina (UEL), Londrina, PR, Brasil 4 Departamento de Medicina Interna, Universidade Estadual de Londrina (UEL), Londrina, PR, Brasil

Correspondence to: Isaias Dichi Departamento de Medicina Interna, Universidade Estadual de Londrina Av. Robert Koch, 60 86038-440 – Londrina, PR, Brasil dichi@sercomtel.com.br Received on May/3/2017 Accepted on Nov/15/2017

INTRODUCTION

besity and overweight are chronic disorders of multifactorial origin, which can be defined as an increase in the accumulation of body fat (1). Changes in lifestyle and diet have resulted in an increased number of overweight and obese subjects in developed and developing countries (2). This trend has been verified in practically all ages, genders and ethnicities (2). Therefore, overweight and obesity have emerged as two of the largest public health problems worldwide. Excess body weight is associated with an increased risk of developing metabolic syndrome (MetS). MetS is a complex disorder that is represented by a cluster of cardiovascular risk factors that are associated with central fat deposition, abnormal plasma lipid levels, elevated blood pressure, insulin resistance and a lowArch Endocrinol Metab. 2018;62/3

grade inflammatory state. MetS has also been associated with increased oxidative and nitrosative stress (3). The harmful effects of free radicals, which are mainly represented by reactive oxygen species (ROS) or reactive nitrogen species, have been implicated in the physiopathology of overweight, obesity, hypertension, endothelial dysfunction, and MetS (4,5), suggesting that oxidative stress can be the underlying mechanism of this dysfunctional metabolic picture in obese subjects (6). In addition, high ROS production and the decrease in antioxidant capacity leads to various abnormalities. These abnormalities include endothelial dysfunction – which is characterized by a reduction in the bioavailability of vasodilators, particularly nitric oxide (NO) (7), and an increase in endothelium-derived contractile factors, favoring atherosclerotic disease (1). 309

DOI: 10.20945/2359-3997000000036

Metabolic syndrome and oxidative stress

Although markers of oxidative stress have been studied in obese and overweight patients with and without MetS, we are not aware to date of studies evaluating the influence of body weight on oxo‑nitrosative stress in the other components of MetS. In a previous study performed by our group, we verified that an increase in oxidative stress is mostly attributable to obesity in patients with MetS (8), but this is not the case in overweight subjects without MetS. It is therefore unclear whether metabolic changes of obesity, referred to as metabolic obesity, are independent risk factors for increased oxo-nitrosative stress than the other components of MetS. In order to extend the data of the mentioned study, the objective of the present study was to evaluate the influence of the body mass index (BMI) on oxidative and nitrosative stress in overweight and obese subjects and to verify whether the presence of the components of MetS would modify the results.

SUBJECTS AND MATERIALS

Subjects Patients from the Internal Medicine Ambulatory of the University Hospital of Londrina, Paraná, Brazil were chosen to participate in this cross-sectional study. Three hundred and thirty patients agreed to participate in the study. Inclusion criteria were patients (both genders) aged from 18 to 65 years. Exclusion criteria were thyroid, renal, hepatic, gastrointestinal, infectious or oncological diseases and the use of lipid-lowering drugs, drugs for hyperglycemia, anti-inflammatory drugs, hormone replacement therapy, and antioxidant supplements. For ethical reasons, patients who were taking antihypertensive drugs were not excluded and were allowed to continue taking the same dose of the drugs. The patients were divided into three groups: the control group included 131 subjects with a BMI between 20 and 24.9 kg/m2. The overweight group consisted of 120 subjects with a BMI between 25 and 29.9 kg/m2, and the group with obesity consisted of 79 subjects with a BMI ≥ 30. MetS was defined following the Adult Treatment Panel III criteria. A diagnosis of MetS was arrived at for subjects with at least three of the following five characteristics: (1) abdominal obesity, which was defined as a waist circumference (WC) ≥ 102 cm in men and ≥ 88 cm in women; (2) 310

hypertriglyceridemia, which was defined as triglycerides ≥ 150 mg/dL; (3) low levels of high-density lipoprotein (HDL) cholesterol, which was defined as HDL ≤ 40 mg/dL in men and ≤ 50 mg/dL in women; (4) high blood pressure, which was defined as blood pressure ≥ 130/85 mmHg; and (5) high-fasting glucose, which was defined as glucose ≥ 100 mg/dL.

Ethics, consent, health and safety The research was conducted in an ethical and responsible manner and is in full compliance with all relevant codes of experimentation. In addition, the Ethical Committee of the University of Londrina – Paraná, Brazil – approved all procedures involving human participants (185/2013). This clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki. Written informed consent was obtained from all the participants, who acknowledged that they cannot be identified via the paper and that they are fully anonymized. All mandatory laboratory health and safety procedures have been complied.

Anthropometric and blood pressure measurements Anthropometric measurements and laboratorial parameters were assessed. Body weight was measured to the nearest 0.1 kg in the morning through the use of an electronic scale, with individuals wearing light clothing and no shoes; height was measured to the nearest 0.1 cm through the use of a stadiometer. BMI was calculated as weight (kg) divided by height (m) squared. WC was measured on standing subjects midway between the lowest rib and the iliac crest. Three blood pressure measurements taken with a 1-min interval after the participant had been seated were recorded on the left arm. The mean of these measurements was used in the analysis. We considered the current use of antihypertensive medication as an indication of high blood pressure.

Biochemical, immunological, and hematological biomarkers After fasting for 12 hours, the subjects underwent the following laboratory blood analysis evaluated through a biochemical auto-analyzer (Dimension Dade AR Dade Behring, Deerfield, IL, USA) using Dade Behring® kits: total cholesterol, HDL, low-density Arch Endocrinol Metab. 2018;62/3

Metabolic syndrome and oxidative stress

Oxidative and nitrosative stress measurements Samples for evaluating oxidative stress and total antioxidant capacity were analyzed with e thylenediamine tetraacetic acid as an anticoagulant and antioxidant. All samples were centrifuged at 3,000 rpm for 15 minutes, and plasma aliquots were stored at -70ºC until assayed. All stress measurements were performed in triplicate.

Tert-butyl hydroperoxide-initiated chemiluminescence (CL-LOOH) CL-LOOH in plasma was evaluated as described previously by Gonzales Flecha and cols. (10). CLLOOH is considered to be much more sensitive and specific than the thiobarbituric acid reactive substances method, the usual method to determine lipid oxidation. For the CL measurement, reaction mixtures were placed in 20-mL scintillation vials (low-potassium glass) containing final concentrations of plasma (250 uL), 30 mM KH2PO4/K2HPO4 buffer (pH 7.4), and 120 mM KCl with 3 mM of LOOH in a final volume of 2 mL. CL-LOOH was measured in a Beckman LS 6000 liquid scintillation counter set to the out-of-coincidence mode, with a response of 300 to 620 nm. The vials were kept in the dark until the moment of assay, and determination was carried out in a dark room at 30°C. The results were expressed in counts per minute.

Determination of advanced oxidation protein products (AOPPs) AOPPs were determined in the plasma using the semi-automated method described by Witko-Sarsat and cols. (11). AOPP concentrations were expressed as micromoles per liter (µmol/L) of chloramines-T equivalents.

Total radical-trapping antioxidant parameter (TRAP) TRAP was determined as reported by Repetto and cols. (12). This method detects hydrosoluble and Arch Endocrinol Metab. 2018;62/3

liposoluble plasma antioxidants by measuring the chemiluminescence inhibition time induced by 2,2-azobis (2‑amidinopropane). The system was calibrated with the vitamin E analog TROLOX, and the values of TRAP were expressed in the equivalent of μM Trolox/mg UA. TRAP measurements in conditions associated with hyperuricemia, such as MetS, may be inaccurate because the UA concentration accounts for 60% of the total plasma antioxidant capacity. Some reports have verified an unexpected increase in TRAP in MetS subjects (13). Thus, a correction of TRAP based on UA concentration was performed (13).

Determination of sulfhydryl (SH) groups of proteins SH groups of proteins were evaluated in plasma samples through a spectrophotometric assay based on 2,2-dithiobisnitrobenzoic acid (DTNB), as reported previously (14), and the results are expressed in μM.

Evaluation of nitric oxide metabolites (NOx) The NO concentration in a sample was estimated by measuring the NOx in nitrites (NO2-) and nitrates (NO3-) using cadmium beads for the reduction of nitrate to nitrite. The concentrations of these metabolites were later determined according to the method proposed by Griess (15). The values were expressed in µM.

Statistical analysis Categorical data were analyzed through a chi-squared test or, when appropriate, through Fisher’s exact test, and data were expressed in absolute values. The Kolmogorov-Smirnov test was used to assess the normality of distribution. All continuous variables presented non-parametric distribution, even after logarithmic transformation. The comparisons of the three groups categorized by BMI were performed through the use of the non-parametric Kruskal-Wallis test with the post-hoc Dunn test. The variables that presented significance in the univariate analysis of variance were included in the multinomial logistic regression to verify which oxidative stress parameters were associated with BMI. The Mann-Whitney test was used to compare two groups, and a logistic binary regression analysis was performed to adjust for age, sex, ethnicity and BMI. The results were considered significant when P < 0.05. A statistical analysis program, SPSS version 20.0, was used for evaluations. 311

lipoprotein (LDL), triacylglycerol (TG), glucose and uric acid (UA). Plasma insulin level was determined by chemiluminescence microparticule immunoassay (Architect, Abbott Laboratory, Abbott Park, IL, USA). The homeostasis model assessment insulin resistance (HOMA-IR) was used as a surrogate measurement of insulin sensitivity. HOMA-IR = fasting insulin (U/ml) x fasting glucose (mmol/L)/22.5. IR was considered when HOMA-IR ≥ 2.5 (9).

Metabolic syndrome and oxidative stress

RESULTS

110%

There was no statistically significant difference in ethnicity between the three groups (Table 1). Overweight (G2) and obese subjects (G3) did not differ regarding sex and age. However, the control group (G1) had a higher frequency of women compared to the subjects in G2 (p < 0.0001) and G3 (p < 0.05), and they were younger (p < 0.0001) than those in the G2 and G3 groups. The presence of MetS was higher (p < 0.0001) in G3 compared to G1 and G2 and in G2 compared to G1 (Table 1 and Figure 1). G3 presented higher WC (p < 0.0001, p < 0.0001), glucose (p < 0.0001, p < 0.05), insulin (p < 0.001, p < 0.001), HOMA-IR (p < 0.001, p < 0.001), and TG (p < 0.001, p < 0.01) and decreased HDL–cholesterol (p < 0.001, p < 0.05) levels compared to G1 and G2, respectively (Table 1). Meanwhile, G2 had higher WC (p < 0.0001), glucose (p < 0.001), insulin, HOMA-IR, and TG and decreased (p < 0.001) HDL-cholesterol levels compared to G1. G2 and G3 showed higher total cholesterol (p < 0.05) and LDL cholesterol (p < 0.01) levels compared to those of G1 (Table 1).

100%

Absence of MetS Presence of MetS

90%

Prevalence (%)

80% 70% 60% 50% 40% 30% 20% 10% 0%

BMI < 25

BMI ≥ 30

25 < BMI < 30

Body Mass Index (BMI)

Figure 1. Prevalence of the metabolic syndrome across body mass index categories. Normal weight, < 25 kg/m2; overweight, 25-30 kg/m2; and obese, ≥ 30 kg/m2.

Table 2 shows the results of oxidative stress in the three studied groups with p values adjusted for sex and age. G3 presented higher AOPP values in relation to

Table 1. Clinical and laboratory characteristics of controls (G1), overweight (G2) and obese subjects (G3) Gender (% men) Ethnicity (% caucasians) MetS (Y/N)

G1 (n = 131)

G2 (n = 120)

G3 (n = 79)

G1 X G2

G1 X G3

G2 X G3

< 0.0001

< 0.05

0.1231

0.8420

0.8525

0.6180

12/119

65/55

72/7

< 0.0001

32.0 (25.0-43.0)

43.0 (34.5-53.0)

43.0 (34.0-50.0)

< 0.001

22.04 (20.90-23.57)

27.12 (25.98-28.33)

32.25 (31.04-34.99)

<0.001

< 0.001

WC (cm)

82.0 (77.0-88.0)

97.0 (91.0-101.0)

108.0 (103.0-115.0)

< 0.0001

Fating glucose (mg/dL)

86.0 (83.0-92.0)

92.0 (85.0-98.0)

98.0 (88.0-107.0)

< 0.001

< 0.05

Insulin (U/mL)

6.4 (4.65-8.80)

8.10 (5.90-12.90)

14.5 (10.9-17.4)

< 0.001

HOMA-IR

1.338 (1,020-1,940)

1.989 (1.285-3.238)

3.234 (2.696-4.781)

< 0.001

Total cholesterol (mg/dL)

186.0 (152.5-209,0)

197.0 (168.0-226.0)

198.0 (180.0-224.0)

< 0.050

< 0.010

Age BMI (kg/m2)

HDL-cholesterol (mg/dL)

56.5 (48.5-67.0)

46.5 (38.0-59.5)

42.0 (37.0-51.0)

< 0.001

< 0.05

LDL-cholesterol (mg/dL)

109.8 (83.8-130.2)

121.0 (95.2-141.1)

125.3 (100.5-142.0)

< 0.050

< 0.010

Triacylglycerol (mg/dL)

74.5 (48.5-107.5)

124.0 (90.0-183.0)

175.0 (127.0-231.0)

< 0.001

< 0.01

MetS: metabolic syndrome; BMI: body mass index; WC: waist circumference; IR: insulin resistance; HOMA: homeostasis model assessment; HDL: high-density lipoprotein; LDL: low-density lipoprotein; NS: nonsignificant; WC: waist circumference.

Table 2. Oxidative stress evaluation in controls (G1), overweight (G2) and obese subjects (G3)

Hydroperoxides (cpm)

G1 (n = 131)

G2 (n = 120)

G3 (n = 79)

G1 X G2*

G1 X G3*

G2 X G3*

13900 (10740-17010)

14120 (10950-17350)

13540 (10250-16260)

AOPP (µmol/L)

127.2 (98.2-174.4)

159.5 (124.7-230.3)

195.2 (157.3.257.1)

0.001

0.011

NO (μM)

25.67 (13.67-40.45)

13.83 (8.17-42.84)

12.10 (7.96-27.63)

0.009

0.048

TRAP/UA (μM Trolox/mg/dL)

177.1 (147.2-207.5)

158.8 (126.9-190.0)

138.0 (115.3-164.9)

AOPP: advanced oxidation protein products; NO: nitric oxide; TRAP: total radical-trapping antioxidant parameter; UA: uric acid; NS: nonsignificant. * Adjusted p value for sex and age. AOPP was not significant after adjusting for the presence of MetS, whereas NO maintained its significance.

312

Arch Endocrinol Metab. 2018;62/3

Metabolic syndrome and oxidative stress

Table 3. Stepway analysis for oxidative stress evaluation in controls (G1), overweight (G2) and obese subjects (G3) BMI Parameters

Wald

Odds-Ratio

Confidence interval

Gender

G1 x G2 G1 x G3 G2 x G3

14.294 0.944 6.174

< 0.0001 0.331 0.013

3.635 1.501 2.422

1.862 – 7.098 0.662 – 3.404 1.205 – 4.868

Age

G1 x G2 G1 x G3 G2 x G3

20.998 5.46 3.211

< 0.0001 0.019 0.073

1.067 1.040 1.026

1.038 – 1.097 1.006 – 1.074 0.998 – 1.056

MetS

G1 x G2 G1 x G3 G2 x G3

13.538 53.367 21.823

< 0.0001 < 0.0001 < 0.0001

4.308 29.759 0.145

1.979 – 9.378 11.974 – 73.956 0.064 – 0.326

AOPP

G1 x G2 G1 x G3 G2 x G3

0.080 0.102 0.491

0.777 0.750 0.483

0.999 1.001 0.999

0.995 – 1.004 0.996 – 1.005 0.995 – 1.002

G1 x G2 G1 x G3 G2 x G3

7.377 3.440 0.065

0.007 0.064 0.799

1.018 1.016 1.002

1.005 – 1.031 0.99 – 1.032 0.987 – 1.017

MetS: metabolic syndrome; AOPP: advanced oxidation protein products; NO: nitric oxide. * Adjusted p value for sex, age and presence of MetS. AOPP was not significant after adjusting for sex and age, whereas NO maintained its significance when comparing controls and overweight subjects.

Arch Endocrinol Metab. 2018;62/3

and ethnicity; the results are shown in Table 4. Subjects with higher TG levels had higher AOPPs (p = 0.001) and decreased TRAP/UA levels (p = 0.036) compared to individuals without hypertriacylglycerolemia. Subjects with lower HDL cholesterol and patients with higher blood pressure levels showed increased AOPPs (p = 0.001 and p = 0.034, respectively) and lower NO levels (p = 0.017 and p = 0.043, respectively) compared to individuals without low HDL-cholesterol levels and with normal blood pressure. Subjects who presented insulin resistance had higher AOPP levels (p = 0.024) compared to those without insulin resistance. Multiple regressive stepwise analyses were performed (Table 5). After adjusting for sex, age and the presence of MetS, AOPPs maintained their significance for TG, HDL, blood pressure and HOMA-IR (< 0.0001, < 0.0001, 0.017, and 0.024, respectively); NO maintained its significance for HDL (0.025), and TRAP maintained its significance for TG (0.034).

DISCUSSION The redox state was similar in healthy, overweight and obese subjects when controlled for the presence of MetS, and therefore, the principal finding of the present study was that oxidative stress evaluated through lipid and protein oxidation in patients with obesity is mainly related to the presence of MetS and is less related to BMI. However, nitrosative stress with decreased NO bioavailability was associated with BMI, independently of the presence of MetS. In addition, 313

G1 and G2 (p = 0.001 and p = 0.011, respectively), whereas significantly lower NOx values were found in G2 and G3 when compared to those of G1 (p = 0.009 and p = 0.048, respectively). The groups were then adjusted for the presence of MetS to evaluate its influence on the results. In this new analysis, AOPPs did not differ between the groups, whereas significant lower NO maintained its significance. Lipid hydroperoxides and TRAP/UA did not have any significant change in the groups. Table 3 shows the results obtained after performing multiple regressive stepwise analyses to clarify the importance of body weight on oxidative and nitrosative stress; AOPPs were not significant after adjusting for sex, age and the presence of MetS when comparing controls and overweight, controls and obese and overweight and obese (0.777; 0.750; 0.483, respectively). NO maintained its significance when comparing controls and overweight subjects (0.007; 0.064; 0.799, respectively). To verify the association between oxidative stress biomarkers and the presence of MetS, a binary logistic regression was performed and was adjusted for sex and age. The AOPP levels were directly associated with the presence of MetS (Wald = 16.039, df = 1, OR = 1.009, 95% CI = 1.005-1.009, p < 0.0001), and the NO values were inversely associated (Wald = 18.941, df = 1, OR = 0.958, 95% CI = 0.940-0.977, p < 0.0001) with the presence of MetS (data not shown). The association between the oxidative stress parameters and the individual components of MetS was measured, and the values were adjusted by BMI, sex, age

Metabolic syndrome and oxidative stress

Table 4. Oxidative stress evaluation according to the components of the metabolic syndrome Gender (%men)

Ethnicity (% cauc)

Age

BMI (Kg/m2)

LOOH (cpm)

AOPP (µmol/L)

NO (μM)

TRAP/UA (μM Trolox/mg/dL)

TG < 150 mg/dL n = 218

36.0 (28.0-47.0)

24.42 (21.74-27.68)

14120 (10900-17730)

134.2 (101.2-174.7)

23.72 (12.29-42.84)

170.7 (138.4-204.1)

TG ≥ 150 mg/dL n = 108

46.0 (37.0-53.0)

29.75 (26.63-32.10)

13570 (8549-18750)

225.1 (160.1-275-89)

11.19 (6.60-27.63)

137.8 (118.0-173.0)

< 0.001

0.6321

< 0.0001

----

0.001

0.036

24.36 (21.91-28.04)

14290 (10800-18110)

136.8 (102.9-181.2)

25.78 (12.38-43.93)

163.9 (138.0-205.1)

p * Adjusted p Normal HDL n = 186

38.5 (30.0-47.0)

Reduced HDL n = 137

42.0 (30.5-50.0)

27.99 (25.44-31.60)

13650 (11480-16720)

183.8 (131.4-256.5)

12.37 (7.53-29.91)

149.6 (122.4-186.3)

< 0.0001

0.0011

* Adjusted p

----

0.001

0.017

Normotensive n = 217

35.0 (27.0-44.0)

24.80 (21.83-28.04)

13750 (11130-17650)

134.4 (100.4-183.8)

22.84 (11.90-40.23)

166.3 (137.8-200.9)

Hypertensive n = 112

47.0 (39.0-55.0)

28.60 (26.17-31.70)

14550 (10010-18190)

195.7 (154.3-274.2)

11.95 (6.80-33.73)

145.7 (119.5-184.8)

0.1414

0.9926

< 0.0001

0.7971

< 0.0001

0.0010

----

0.034

0.043

Without IR n = 163

38 (29.0-47.0)

24.22 (21.76-26.67)

14200 (11070-17890)

137.4 (104.0-184.9)

23.5 (11.91-41.00)

169.6 (138.0-203.2)

With IR n = 115

42 (30.0-53.0)

30.05 (27.21-32.92)

13550 (9291-16690)

182 (127.9-256.5)

11.78 (7.07-27.74)

146.3 (116.3-179.6)

0.0781

0.023

< 0.0001

----

0.024

p * Adjusted p

Cauc: Caucasian; BMI: body mass index; IR: insulin resistance; LOOH: hydroperoxides; AOPP: advanced oxidation protein products; NO: nitric oxide; TRAP: total radical-trapping antioxidant parameter; UA: uric acid; TG: tryacylglycerols; NS: nonsignificant. * Binary logistic regression adjusted for sex, age, ethnicity and BMI.

this study verified that protein oxidation was associated with several individual components of MetS, including insulin resistance. The present data are partially in agreement with our previous study, which showed that increases in oxidative stress markers in overweight subjects were only verified in the presence of MetS (8). However, in that study, obese subjects and nitrosative stress were not evaluated, which differs from the conditions of the present study. Thus, the impact of this original work lays in the observation that only nitrosative stress was related to BMI, whereas protein oxidation was related to each component of MetS. Although an increase in oxidative stress in patients with obesity is an undisputed issue and can be caused by several factors (16), the present study is in line with others, which pointed out the utmost importance of the presence of MetS to reinforce this association. Skalicky and cols. (17) verified in obese subjects and Krzystek314

Korpacka and cols. (18) verified in overweight and obese adolescents that oxidative stress seemed to be increased through a combination of risk factors associated with MetS rather than by obesity per se. Fujita and cols. (19) demonstrated that values of oxidative stress increased with the number of components of MetS. Taken together, these data suggest that – although weight gain or visceral fat may contribute, to some extent, to an increase in oxidative stress – the presence of MetS is fundamental in showing ROS augmentation in overweight and obese subjects. Our data are also in accordance with that of previous studies, which showed that hypertriacylglycerolemia, hypertension, lower HDL cholesterol values and insulin resistance are essential factors in provoking oxidative stress (13,19,20). Obesity and IR are considered key factors for the development of MetS. There is mounting evidence that oxidative stress is involved in the development of insulin resistance and that, once IR Arch Endocrinol Metab. 2018;62/3

Metabolic syndrome and oxidative stress

Table 5. Multiple regressive stepwise analysis for oxidative stress evaluation in different components of the metabolic syndrome TG

HDL

Blood Pressure

HOMA-IR

Parameters

Wald

Odds-Ratio

Confidence interval

Sex

8.613

0.003

0.289

0.126 – 0.662

Age

0.098

0.755

1.005

0.972 – 1.040

BMI

21.695

< 0.0001

1.212

1.118 – 1.314

AOPP

20.380

< 0.0001

1.011

1.006 – 1.016

1.082

0.298

0.992

0.977 – 1.007

0.104

0.748

1.949

0.033 – 113.506

TRAP-UA

4.495

0.034

0.990

0.980 – 0.999

BMI

13.434

< 0.0001

1.113

1.051 – 1.179

AOPP

14.431

< 0.0001

1.006

1.003 – 1.010

5.006

0.025

0.987

0.975 – 0.998

TRAP-UA

0.062

0.803

0.999

0.993 – 1.005

Age

20.787

< 0.0001

1.086

1.048 – 1.125

BMI

16.929

< 0.0001

1.167

1.084 – 1.256

AOPP

5.683

0.017

1.005

1.001 – 1.009

3.509

0.061

1.013

0.999 – 1.028

0.737

0.391

4.993

0.127 – 196.458

TRAP-UA

1.586

0.208

0.995

0.987 – 1.003

Age

4.753

0.029

0.969

0.0942 – 0.997

BMI

57.168

< 0.0001

1.463

1.325 – 1.614

AOPP

5.092

0.024

1.004

1.001 – 1.008

1.574

0.210

0.991

0.977 – 1.005

TRAP-UA

0.370

0.543

0.998

0.990 – 1.005

is acquired, all the other components of MetS could be developed as a result (21,22). Hypertriacylglycerolemia and hypertension lead to an increased production of superoxide anion (O2-) via the nicotinamide adenosine diphosphate oxidase pathway. This anion reacts rapidly with NO to form peroxynitrite (ONOO-) – thus inactivating NO and leading to endothelial dysfunction, one of the mechanisms responsible for hypertension in these patients (23) – whereas HDL cholesterol antioxidant activity, a major mechanism mediating its cardioprotective effect, is impaired (20). Of note, in the current study, hypertriacylglycerolemia showed the highest degree of redox imbalance, as it was the only MetS component, and it concomitantly increased protein oxidation and decreased antioxidant capacity. Our results show an association between the presence of insulin resistance and increased levels of protein oxidation. Although several reports have established the importance of insulin resistance and Arch Endocrinol Metab. 2018;62/3

oxidative stress in the development of both diabetes and cardiovascular disease (24,25), the precise role of oxidative stress as a cause or consequence of insulin resistance is still debated. Furukawa and cols. (26) demonstrated in cultured adipocytes that elevated levels of fatty acids increased oxidative stress via NADPH oxidase activation, and oxidative stress caused dysregulated production of adipocytokines – including adiponectin, plasminogen activator inhibitor-1, IL6, and monocyte chemotactic protein-1. In addition, in mice with obesity, treatment with an NADPH oxidase inhibitor reduced ROS production in adipose tissue; attenuated the dysregulation of adipocytokines; and improved diabetes, hyperlipidemia, and hepatic steatosis. NADPH oxidase inhibitors could improve insulin sensitivity via the suppression of the effects induced through chronic exposure to ROS. These results suggested that increased oxidative stress in accumulated fat is an early instigator of MetS and that 315

TG: triacylglycerol; HDL: high-density lipoprotein; HOMA-IR: homeostasis model assessment-insulin resistance; BMI: body mass index; AOPP: advanced oxidation protein products; NO: nitric oxide; Sulfhydryl (SH) groups of proteins; TRAP-UA: total radical-trapping antioxidant parameter-uric acid. After adjusting for sex and age AOPP maintained its significance for TG, HDL, hypertension and HOMA-IR, NO maintained its significance for HDL and TRAP-AU maintained its significance for TG. Whereas the other oxidative and nitrosative stress markers lost significance.

Metabolic syndrome and oxidative stress

the redox state in adipose tissue is a potentially useful therapeutic target for obesity-associated MetS. In addition, hydrogen peroxide impairs insulin signaling and inhibits glucose transport, two cardinal features of insulin resistance (27). On the other hand, insulin itself promotes hydrogen peroxide formation in human fat cells (4). Altogether, it is tempting to speculate that oxidative stress can be both cause and consequence of insulin resistance (16,28). It has been suggested that AOPPs are an early marker of MetS and are the most appropriate parameter for the determination of oxidative stress in MetS patients (29). AOPPs are formed during oxidative stress through the action of chlorinated oxidants, mainly hypochlorous acid and chloramines, produced by myeloperoxidase in activated neutrophils (11). AOPPs are structurally similar to advanced glycation end products (AGEs) and exert similar biological activities to those of AGEs – i.e., induction of pro-inflammatory cytokines and adhesion molecules (11). The present study showing that AOPPs were associated with metabolic changes and hypertension is in line with the importance of protein oxidation in patients with these components of MetS, independent of BMI, and confirm our previous finding that protein oxidation is more related to MetS parameters than lipid oxidation (30). Of note, BMI does not participate in the definition of the MetS and does not have a robust association verified between WC and insulin resistance. Although some studies have reported an association between AOPPs and BMI (13,30), the current study only presented this finding when the groups were not adjusted for the presence of MetS. Changes in NOx levels have been linked with disorders of metabolic and cardiovascular homeostasis that can culminate in obesity-induced IR and endothelial dysfunction. Serum NOx were shown to be increased in MetS and be associated with other metabolic components such as BP, BMI, waist-to-hip ratio, and fasting plasma glucose in cluster analyses (31). In addition to overproduced NO, reduced levels of NO may also be a risk factor for the development of cardiometabolic disease (32). Although endothelial dysfunction has been considered an important issue in patients with obesity, the results of studies on NOx levels have been contradictory. Whereas some reports have shown higher NO levels (33), others have found the opposite results, similarly to the present study (34). NO is synthesized in endothelial cells by 316

endothelial nitric oxide synthase (eNOS) activity, and it is responsible for vasodilatation and the maintenance of endothelial function; eNOS is expressed constitutively and synthesizes NO in only small amounts under basal conditions. In contrast, oxidative stress provokes inducible nitric oxide synthase (iNOS) expression even in low-grade inflammatory conditions, such as obesity, and consequently increases NO – which would be consumed in a reaction with superoxide anion, yielding peroxynitrite (35). This hypothesis is supported by some authors who demonstrated an increase in nitrotyrosine, a marker of endogenous peroxinitrite generation (36). Thus, the balance between eNOS and iNOS could explain NO increases or decreases in obese subjects. Although oxidative stress may induce NO production, the NO decrease associated with BMI found in the present study is probably related to higher NO consumption through oxidative stress, reducing NO bioavailability. Furthermore, previous investigations have shown that HDL induces a variety of signaling events – involving scavenger receptor B type I, cholesterol efflux and the stimulation of the phosphorylation of eNOS – that lead to the activation and increased expression of eNOS and the subsequent production of NO. Thus, the decreased NO levels in the present study may be related to the HDL cholesterol reduction (37,38). Overweight is highly associated with arterial hypertension, independently from the occurrence of MetS, and a BMI of 25 kg/m2 or greater accounted for approximately 34% and 62% of hypertension in men and in women, respectively (35). NO plays a major role in regulating blood pressure, and its deficient bioactivity is an important component of hypertension (37). Hypertensive subjects have increased generation of ROS – which scavenge NO, thereby reducing NO bioavailability (23). This study confirms the well-established relationship between NO decreases and hypertension, independent of whether BMI is considered. The following limitations have to be considered in the present study. The first limitation is the small number of participants. Second, the food pattern and physical activities of the individuals were not measured. Third, the antihypertensive drugs the patients were taking, such as angiotensin-converting enzyme inhibitors, may elevate plasma adiponectin levels, which in turn can increase NO levels (23). Fourth, additional tests to evaluate oxidative and, especially, Arch Endocrinol Metab. 2018;62/3

Metabolic syndrome and oxidative stress

9. Oliveira EP; Lima MD, Souza MLA. Síndrome metabólica, seus fenótipos e resistência à insulina pelo HOMA-RI. Arq Bras Endocrinol Metab. 2007;51:1506-15. 10. Gonzalez Flecha B, Llesuy S, Boveris A. Hydroperoxide-initiated chemiluminescence: an assay for oxidative stress in biopsies of heart, liver, and muscle. Free Radic Biol Med. 1991;10:93-100. 11. Witko-Sarsat V, Friedlander M, Capeillère-Blandin C, NguyenKhoa T, Nguyen AT, Zingraff J, et al. Advanced oxidation protein products as a novel marker of oxidative stress in uremia. Kidney Int. 1996;49:1304-13. 12. Repetto M, Reides C, Gomez Carretero ML, Costa M, Griemberg G, Llesuy S. Oxidative stress in blood of HIV infected patients. Clin Chim Acta. 1996;255:107-17. 13. Simão ANC, Dichi JB, Barbosa DS, Cecchini R, Dichi I. Influence of uric acid and gamma-glutamyltransferase on total antioxidant capacity and oxidative stress in patients with metabolic syndrome. Nutrition. 2008;24:675-81. 14. Hu ML. Measurement of protein thiol groups and glutathione in plasma. In: Abelson JN, Simon MI, editors. Methods in enzymology. San Diego, CA: Academic Press; 1994. p. 380-2. 15. Navarro-Gonzálvez JA, García-Benayas C, Arenas J. Semiautomated measurement of nitrate in biological fluids. Clin Chem. 1998;44:679-81. 16. Simão ANC, Lovozoy MAB, Dichi I. Oxidative stress in overweight and obesity. In: Dichi I, Breganó JW, Simão ANC, Cecchini R (eds). Oxidative Stress in Chronic Diseases. Boca Raton: CRC Press; 2014. p. 121-36.

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

17. Skalicky J, Muzakova V, Kandar R, Meloun M, Rousar T, Palicka V. Evaluation of oxidative stress and inflammation in obese adults with metabolic syndrome. Clin Chem Lab Med. 2008;46:499-505.

REFERENCES

18. Krzystek-Korpacka M, Patryn E, Boehm D, Berdowska I, Zielinski B, Noczynska A. Advanced oxidation protein products (AOPPs) in juvenile overweight and obesity prior to and following weight reduction. Clin Biochem. 2008;41:943-9.

1. Fernández-Sánchez A, Madrigal-Santillán E, Bautista M, EsquivelSoto J, Morales-González A, Esquivel-Chirino C, et al. Inflammation, oxidative stress, and obesity. Int J Mol Sci. 2011;12:3117-32. 2. Wang Y, Monteiro C, Popkin BM. Trends of obesity and underweight in older children and adolescents in the United States, Brazil, China, and Russia. Am J Clin Nutr. 2002;75:971-7. 3. Wildman RP, McGinn AP, Kim M, Muntner P, Wang D, Cohen HW, et al. Empirical derivation to improve the definition of the Metabolic Syndrome in the evaluation of cardiovascular disease risk. Diabetes Care. 2011;34(3):746-8. 4. Ohmori K, Ebihara S, Kuriyama S, Ugajin T, Ogata M, Hozawa A, et al. The relationship between body mass index and plasma lipid peroxidation biomarker in an older, healthy Asian community. Ann Epidemiol. 2005;15:80-4.

19. Fujita K, Nishizawa H, Funahashi T, Shimomura I, Shimabukuro M. Systemic oxidative stress is associated with visceral fat accumulation and the metabolic syndrome. Circ J. 2006;70:1437-42. 20. Hansel B, Giral P, Nobecourt E, Chantepie S, Bruckert E, Chapman MJ, et al. Metabolic syndrome is associated with elevated oxidative stress and dysfunctional dense high-density lipoprotein particles displaying impaired antioxidative activity. J Clin Endocrinol Metab. 2004;89:4963-71. 21. Barnard RJ, Roberts CK, Varon SM, Berger JJ. Diet-induced insulin resistance precedes other aspects of the metabolic syndrome. J Appl Physiol (1985). 1998;84(4):1311-5. 22. Simão ANC, Lozovoy MAB, Dichi I. Oxidative Stress in Metabolic Syndrome. Role of Oxidative Stress in Chronic Diseases, 1st ed., vol. 1. CRC Press; 2014. p. 246-59.

5. Van Guilder GP, Hoetzer GL, Greiner JJ, Stauffer BL, Desouza CA. Influence of metabolic syndrome on biomarkers of oxidative stress and inflammation in obese adults. Obesity (Silver Spring). 2006;14:2127-31.

23. Simão ANC, Lozovoy MA, Bahls LD, Morimoto HK, Simão TN, Matsuo T, et al. Blood pressure decrease with ingestion of a soy product (kinako) or fish oil in women with metabolic syndrome: role of adiponectin and nitric oxide. Br J Nutr. 2012;108:1435-42.

6. Ceriello A, Motz E. Is oxidative stress the pathogenic mechanism underlying insulin resistance, diabetes, and cardiovascular disease? The common soil hypothesis revisited. Arterioscler Thromb Vasc Biol. 2004;24:816-23.

24. Stern MP. Diabetes and cardiovascular disease. The “common soil” hypothesis. Diabetes. 1995;44:369-74.

7. Singh VP, Aggarwal R, Singh S, Banik A, Ahmad T, Patnaik BR, et al. Metabolic syndrome is associated with increased oxo-nitrative stress and asthma-like changes in lungs. PLoS One. 2015;10(6):e0129850. 8. Venturini D, Simão ANC, Scripes NA, Bahls LD, Melo PA, Belinetti FM, et al. Oxidative stress evaluation in overweight subjects with or without metabolic syndrome. Obesity (Silver Spring). 2012;20:2361-6. Arch Endocrinol Metab. 2018;62/3

25. Kip KE, Marroquin OC, Kelley DE, Johnson BD, Kelsey SF, Shaw LJ, et al. Clinical importance of obesity versus the metabolic syndrome in cardiovascular risk in women. A report from the Women’s Ischemia Syndrome evaluation (WISE) Study. Circulation. 2004;109:706-13. 26. Furukawa S, Fujita T, Shimabukuro M, Iwaki M, Yamada Y, Nakajima Y, et al. Increased oxidative stress in obesity and its impact on metabolic syndrome. J Clin Invest. 2004;114:1752-61. 27. Keaney JF Jr, Larson MG, Vasan RS, Wilson PW, Lipinska I, Corey D, et al. Obesity and systemic oxidative stress: clinical correlates

317

nitrosative stress would make our data more consistent. Nevertheless, the present study also has several strengths. First, to our knowledge, this is the first study to evaluate concomitantly oxidative and nitrosative stress in overweight and obese subjects. Second, we adjusted the results of oxidative stress measurements for the presence of MetS to evaluate its influence on the results. In conclusion, only nitrosative stress was related to BMI, whereas protein oxidation was related to each component of MetS. In addition, both NO and advanced oxidative protein products were related to hypertension. In general, MetS components were essential participants in overweight and obese subjects, but hypertriacylglycerolemia was the parameter that showed the highest degree of redox imbalance. Although more studies are warranted to confirm the present data, this study reinforces the importance of concomitantly analyzing oxidative and nitrosative stress to obtain a more complete picture of overweight, obesity and associated conditions.

Metabolic syndrome and oxidative stress

of oxidative stress in the Framingham Study. Arterioscler Thromb Vasc Biol. 2003;23:434-9. 28. Simão ANC, Lozovoy MA, Dichi I. Metabolic syndrome: new targets for an old problem. Expert Opin Ther Targets. 2012;16:147-50. 29. Zurawska-Płaksej E, Grzebyk E, Marciniak D, SzymańskaChabowska A, Piwowar A. Oxidatively modified forms of albumin in patients with risk factors of metabolic syndrome. J Endocrinol Invest. 2014;37:819-27. 30. Venturini D, Simão ANC, Dichi I. Advanced oxidation protein products are more related to metabolic syndrome components than biomarkers of lipid peroxidation. Nutr Res. 2015;35:759-65.

33. Elizalde M, Rydén M, van Harmelen V, Eneroth P, Gyllenhammar H, Holm C, et al. Expression of nitric oxide synthases in subcutaneous adipose tissue of nonobese and obese humans. J Lip Res. 2000;41:1244-51. 34. Esposito K, Ciotola M, Schisano B, Misso L, Giannetti G, Ceriello A, et al. Oxidative stress in the metabolic syndrome. J Endocrinol Invest. 2006;29:791-5. 35. Mineo C, Deguchi H, Griffin JH, Shaul PW. Endothelial and antithrombotic actions of HDL. Circ Res. 2006;98:1352-64. 36. Wilson PW, D’Agostino RB, Sullivan L, Parise H, Kannel WB. Overweight and obesity as determinants of cardiovascular risk: the Framingham experience. Arch Intern Med. 2002;162:1867-72. 37. Hermann M, Flammer A, Lüscher TF. Nitric oxide in hypertension. J Clin Hypertens (Greenwich). 2006;8:17-29.

32. Bahadoran Z, Mirmiran P, Ghasemi A, Azizi F. Serum nitric oxide metabolites are associated with the risk of hypertriglyceridemicwaist phenotype in women: Tehran lipid and glucose study. Nitric Oxide. 2015;50:52-7.

38. Besler C, Heinrich K, Rohrer L, Doerries C, Riwanto M, Shih DM, et al. Mechanisms underlying adverse effects of HDL on eNOSactivating pathways in patients with coronary artery disease. J Clin Invest. 2011;121:2693-708.

31. Zahedi Asl S, Ghasemi A, Azizi F. Serum nitric oxide metabolites in subjects with metabolic syndrome. Clin Biochem. 2008;41:1342-7.

318

Arch Endocrinol Metab. 2018;62/3

original article

Self-report of psychological symptoms in hypoparathyroidism patients on conventional therapy Amanda J. Arneiro1, Bianca C. C. Duarte1, Rodrigo M. Kulchetscki1, Victor B. S. Cury1, Maicon P. Lopes1, Breno S. Kliemann1, Ileana B. Bini1, Marcos A. Assad1, Gleyne L. K. Biagini2, Victoria Z. C. Borba1,3, Carolina A. Moreira1,3,4

ABSTRACT Objective: Hypoparathyroidism is characterized by parathyroid hormone deficiency and hypocalcemia. It has been demonstrated that these patients may also present psychiatric symptoms and decrease of quality of life. The aims of this study were to evaluate the presence of psychopathological symptoms in a cohort of patients with hypoparathyroidism and compare to a control group. Subjects and methods: Patients were submitted to a cross-sectional Symptom Checklist-90-R (SCL-90-R) questionnaire that evaluates psychopathological symptoms by means of the Global Severity Index (GSI), Positive Symptoms Total (PST) and Positive Symptom Distress Index (PSDI). A score based in the positive symptoms was calculated (T-score). The test group was composed of patients with hypoparathyroidism, and control by thyroidectomized patients without hypoparathyroidism. A correlation between the presence of psychological symptoms and clinical features was analyzed. Results: The study included 57 patients with a mean age of 51.1 ± 16.4 years; 20 as a control and 37, test group. There were no differences between groups regarding gender, mean age and age at diagnose. Hypoparathyroidism patients presented higher GSI index than the control group (p = 0.038). Mean T-score of the test group was as elevated as 58.2 ± 5.3 (reference range < 55). No correlation of the number of psychological symptoms to clinical and laboratorial parameters was observed. Conclusion: Patients with hypoparathyroidism attending our outpatient clinics presented an increase in the number of self-report of psychological symptoms when compared with a control group. However, no correlation with hypocalcemia and clinical parameters was observed. Future studies are necessary to evaluated if the absence of PTH play a role on it. Arch Endocrinol Metab.

Divisão de Endocrinologia, Universidade Federal do Paraná (UFPR), Serviço de Endocrinologia e Metabologia (SEMPR), Curitiba, PR, Brasil 2 Divisão de Endocrinologia, Hospital Universitário Evangélico de Curitiba, Curitiba, PR, Brasil 3 Departamento de Medicina Interna, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brasil 4 Laboratório PRO, Setor de Histomorfometria Óssea, Fundação Pró-Renal, Curitiba, PR, Brasil 1

Correspondence to: Carolina A. Moreira carolina.aguiar.moreira@gmail.com Received on Aug/20/2017 Accepted on Dec/20/2017 DOI: 10.20945/2359-3997000000041

2018;62(3):319-24

INTRODUCTION

arathyroid hormone (PTH) is the main regulator of calcium metabolism. It maintains calcium homeostasis by directly regulating bone remodeling and renal excretion, and indirectly regulating the production of 1,25-dihydroxyvitamin D. Deficiency of PTH leads to hypoparathyroidism, a metabolic disorder in which the diagnosis is confirmed by the occurrence of hypocalcemia (1). Although several etiologies may be associated with hypoparathyroidism, in approximately 75% of the patients the disease occurs after surgery in the anterior cervical region (thyroidectomy and parathyroidectomy) (2). Hypoparathyroidism is a rare disorder, and only a few studies have directly assessed its prevalence. Estimates point out to approximately 77,000 cases (3) Arch Endocrinol Metab. 2018;62/3

and 37/100,000 person-years (4) in the United States, and 22/100,000 person-years in Denmark (5). Treatment of hypoparathyroidism includes supplemental calcium at high doses and calcitriol or other vitamin D analogs. The main morbidities affecting these patients are due to the occurrence of hypocalcemia and hyperphosphatemia and the adverse effects of long-term use of calcium and vitamin D treatment such as nephrolithiasis, nephrocalcinosis, and renal function impairment (1). Studies have shown that patients with hypoparathyroidism present a variety of psychological and neurological manifestations (6-9) and a high prevalence of psychiatric morbidities (7,8), including an increased incidence of depression and bipolar disorder (7) and a greater risk of hospitalization due to neuropsychiatric conditions (8). Decreased 319

Keywords Hypoparathyroidism; psychological distress; psychopathological symptoms; SCL-90-R; parathyroid hormone

Psychological symptoms in hypoparathyroidism

cognitive function has also been demonstrated in these patients (9). In an online interview with over 300 adults with hypoparathyroidism, more than 50% of the respondents reported cognitive and emotional symptoms (6), including mental lethargy (72%), memory loss (65%), sleep disturbances (57%), anxiety (59%), and depression (53%). Although evidence shows frequent psychological and neurological manifestations interfering with the quality of life of patients with hypoparathyroidism, these manifestations have not been carefully detailed, and are not usually identified by physicians caring for these patients. In a US study including patients with permanent hypoparathyroidism following thyroidectomy, 47% of the respondents reported feeling that their health was “much worse” after surgery (10). In contrast, only 16% of the surgeons who had operated on these patients estimated that the patients were dissatisfied after surgery. The authors then concluded that the impact of postoperative hypoparathyroidism on a patient’s quality of life is actually underestimated (10). The aims of this study were to evaluate the presence of psychopathological symptoms, which is related to well being and mood, in a cohort of patients with hypoparathyroidism undergoing conventional treatment with calcium and vitamin D, and analyze the association between these symptoms to other clinical factors including disease duration and serum calcium levels.

SUBJECTS AND METHODS In this cross-sectional, case-control study, we conducted an active search of medical records of patients attending the Thyroid Cancer and Bone Metabolism outpatient clinics at the Endocrine Division at Federal University of Paraná (SEMPR) to identify patients with hypoparathyroidism and controls without hypoparathyroidism. We identified 57 patients who were further contacted by phone and invited to participate in the study. All patients who agreed to participate signed an informed consent form. The study was approved by our local Ethics Committee. The study groups consisted of 37 patients with postsurgical or autoimmune hypoparathyroidism and 20 control patients without hypoparathyroidism who had undergone total thyroidectomy. The majority of the patients were underwent to surgery in the last 23 320

years by the head and neck surgery team of the Hospital de Clínicas, Federal University of Parana. Exclusion criteria comprised a duration of hypoparathyroidism shorter than 6 months and a history of psychiatric or neoplastic diseases, except for thyroid neoplasms. We performed a detailed review of the medical records of the participants to gather information regarding their age, gender, etiology, disease duration, age at diagnosis, and doses of medications in use. In addition, we obtained the patients’ serum levels of calcium, phosphate, 25-hydroxyvitamin D, and PTH, collected on average 6 months after surgery, and calculated their mean levels and standard deviations. To calculate the mean levels of serum calcium, phosphate, and PTH, we analyzed the first and last visits of each patient, as well as three intermediate visits (a total of five time points). We have previously published the mean treatment doses of calcium carbonate, calcitriol, and cholecalciferol used by the patients at five time points during their follow-up (11). Patients in both groups filled out the Symptom Checklist-90-R (SCL-90-R) questionnaire, an internationally and nationally validated (12) tool. The SCL-90-R was chosen, since it is the most complete questionnaire to evaluated psychopathological symptoms and also has been previously used in wellrecognized study in patients with hypoparathyroidism receiving conventional therapy (13). The questionnaire comprises 90 items, each item graded on a scale from 0 (none) to 4 (extreme). The patient must choose the grade (number) that best corresponds to how much they have worried or how much distress they have experienced related to that item over the previous 7 days. After receiving instructions, the patients filled out the questionnaires individually. The SCL-90-R assesses presence of psychopathological symptoms of an individual by evaluating three global indices of distress: the Global Severity Index (GSI), the Positive Symptom Total (PST), and the Positive Symptom Distress Index (PSDI) (12,14,15). GSI is considered the best indicator to assess the current level of the disorder, and is calculated by adding up the obtained values and dividing the result by the number of items answered, thus reflecting the quality of life in general. The PST assesses the number of symptoms, and is calculated by the number of items scored above zero. The PSDI measures the intensity of symptoms and is calculated by dividing the sum of the items answered by the Arch Endocrinol Metab. 2018;62/3

Psychological symptoms in hypoparathyroidism

Statistical analysis We analyzed the data with the software R, version 3.2.3 (16). The normality of the data was evaluated with the Shapiro-Wilk test. To evaluate statistical differences between groups of data following a normal distribution (serum calcium), we used the parametric Student’s t test, and for data without a normal distribution (current age, age at diagnosis, disease duration, etiology, phosphate, vitamin D, and PTH), we used the nonparametric Wilcoxon-Mann-Whitney test. To evaluate whether the proportion of men and women between the groups differed statistically, we used Fisher’s exact test. Student’s t test was also used to analyze the sum of the items of the questionnaire, as well as the scores obtained for each of the indices – GSI, PST, and PSDI. To assess the correlation between quality of life and disease duration, age at diagnosis, etiology, and serum calcium and phosphate, we used Spearman’s rank correlation coefficient, because the data followed a non-normal distribution. We considered p values < 0.05 as statistically significant.

hypoparathyroidism. There were no significant differences between groups regarding gender, mean age, mean age at diagnosis, and education level. In general, patients with hypoparathyroidism were asymptomatic except by presenting occasional paresthesia and cramps. Table 1 presents the laboratory parameters in both groups. As expected, the serum levels of calcium, phosphate, and PTH differed significantly between the groups, with patients with hypoparathyroidism presenting lower calcium and higher phosphate levels when compared with those in the control group. Patients in the hypoparathyroidism group received conventional replacement therapy with calcium and vitamin D. The mean daily doses prescribed were 1.321 ± 705 mg of calcium carbonate, 0.5 ± 0.28 mcg of calcitriol, and 25500 ± 15000 IU of cholecalciferol. Table 2 shows the SCL-90-R scores. When compared with control patients, those with hypoparathyroidism had greater GSI scores (1.1 ± 0.5 and 0.8 ± 0.4, respectively, p = 0.03). The significance was maintained when only the post surgical hypoparathyroidism patients (N:31) were compared to the control group (0.8 ± 0.4 vs. 1.1 ± 0.6; p = 0.04). The mean number of the questionnaire items filled out by the patients was higher in the hypoparathyroidism group (93.5 ± 47.3) compared with the control group (84.3 ± 45.2, p = 0.036). No differences were observed between groups in regard to the PST and PSDI subscales (Table 2). The mean GSI T-score in patients with hypoparathyroidism was 58.1 ± 5.6. Overall, 27 patients with hypoparathyroidism (72.9%) presented impaired quality of life (T-scores above 55). There was Table 1. Characteristics of the study population regarding clinical and laboratory parameters Parameters

Arch Endocrinol Metab. 2018;62/3

Hypoparathyroidism

P value

34 women 3 men

Current age (years)2

52.7 ± 14.9

50.2 ± 17.3

0.535

Age at diagnosis (years)2

43.0 ± 17.1

38.0 ± 18.6

0.425

Disease duration (years)

8.4 ± 7.3

13.1 ± 5.9

0.025*

Calcium (mg/dL)3

9.3 ± 7.3

8.4 ± 1.0

0.003*

Phosphate (mg/dL)

3.6 ± 0.4

8.4 ± 1.0

< 0.001*

Vitamin D (mg/dL)2

33.6 ± 11.1

50.2 ± 22.2

0.131

PTH (pg/mL)

49.5 ± 15.7

8.9 ± 4.7

< 0.001*

RESULTS The study included 57 patients with a mean age of 51.1±16.4 years. In total, 20 patients were allocated to the control group and 37 to the hypoparathyroidism group. In the latter, 31 patients had postsurgical hypoparathyroidism, and six had autoimmune

Control 19 women 1 man

Gender1

Statistical analysis with Fisher’s exact test. Mann-Whitney statistical analysis. 3 Student’s t test statistical analysis. * Statistically significant difference. The values are expressed as mean ± standard deviation. 1 2

321

number of items scored above zero. The latter is also used to assess the patient’s response style, or in other words, if they tend to “augment” or “attenuate” their symptoms (12,14,15). Lastly, the SCL-90-R evaluates nine dimensions of psychological symptoms, including somatization, obsessive-compulsive traits, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism (12,14,15). These nine dimensions of psychological symptoms of the SCL-90-R were not evaluated in the present study. We converted the GSI responses to T-scores in order to compare these results with those in the literature. The classification is based on T-score values, as follows: T-scores between 40 and 55 are within the normal range, those greater than 55 indicate already significant impairment of mood, while those above 60 are considered pathological, equivalent to severe impairment of well-being and mood (12,14,15).

Psychological symptoms in hypoparathyroidism

Table 2. Indices obtained with the SCL-90-R questionnaire in the controls and hypoparathyroidism groups Control (n = 20)

Hypoparathyroidism (n = 37)

Sum of items answered1

67.4 ± 36.2

93.5 ± 47.3

0.036

Number of Items Answered²

87.7 ± 10.0

87.9 ± 9.4

0.252

SCL-90-R

0.8 ± 0.4

1.1 ± 0.5

0.038*

PST (Positive Symptom Total)¹

34.2 ± 16.1

43.3 ± 18.4

0.069

PSDI (Positive Symptom Distress Index)1

1.9 ± 0.7

2.1 ± 0.5

0.324

GSI (Global Severity Index)1

Student’s t test statistical analysis. Mann-Whitney statistical analysis. * Statistically significant difference. The values are expressed as mean±standard deviation.

1 2

no correlation between GSI scores with age, disease duration, etiology, or levels of serum calcium and phosphate. No correlation between quality of life and disease duration was found.

DISCUSSION This study demonstrated that patients with hypoparathyroidism presented status of more psychological distress compared with a control group of individuals without hypoparathyroidism, albeit a correlation with the clinical factors investigated was not confirmed. The significant difference in GSI between the two groups showed that this impairment of mood and in some cases, well-being, might have affected several dimensions of physical and mental symptoms and therefore, quality of life. As demonstrated by the PST scores, patients with hypoparathyroidism showed a tendency to present more symptoms. Regarding the PSDI, the difference between the two groups was also not significant, showing that in spite of the global decrease in well-being, patients with hypoparathyroidism do not tend to amplify the intensity of their symptoms. A few studies have evaluated of psychological distress and well-being in patients with hypoparathyroidism. Arlt and cols. have published a case-control study with the application of questionnaires assessing a broad-spectrum of well-being parameters in patients with hypoparathyroidism undergoing conventional treatment in Germany (13). In their study, 25 patients with hypoparathyroidism following thyroidectomy were evaluated with the SCL-90-R and other questionnaires. In line with our findings, 322

the authors also observed a significant impairment of mood with a pattern usually seen in anxiety disorders among patients with hypoparathyroidism when compared with thyroidectomized controls without hypoparathyroidism. In both groups, the authors found no correlation between neuropsychiatric assessment scores with serum calcium or disease duration. The mean T-score related to the GSI was only slightly higher than that in our cohort, but it was higher than 55 in both studies, reflecting a significant impairment of mood among patients with hypoparathyroidism across different cultural contexts. Other studies also assessing patients with hypoparathyroidism undergoing replacement with calcium and vitamin D had different methodologies than ours, including the absence of a control group and lack of a detailed description of treatment type or use of T-scores. These differences hinder a direct comparison of these studies with ours. Despite that, these studies also demonstrated neurological and psychological impairments in this group of patients (6-9). The causes of the neuropsychiatric changes in patients with hypoparathyroidism are still unclear. Both our study and the one by Arlt and cols. (13) have shown a negative correlation between quality of life assessment scores with serum calcium and disease duration. In addition, most patients undergoing conventional treatment managed to maintain calcium levels within the normal range. Mitchel and cols. periodically measured the serum calcium levels of 120 patients with hypoparathyroidism undergoing calcium and vitamin D replacement for 7 years, and demonstrated that 71% of the patients achieved an acceptable range of calcemia for the disease, considered by the authors as between 7.5 and 9.5 mg/dL (17). Considering that the high prevalence of neuropsychiatric manifestations coexists with normal serum calcium levels in large part of the patients, this has led to the question of whether the presence of psychological symptoms and therefore, decrease in quality of life, might not be another consequence of chronic hypocalcemia. Interestingly, it has been described that the PTH receptor PTH2R is more common in the brain than in other tissues (18). One study has used in situ hybridization to study the distribution of this receptor in the brain of primates and humans, revealing that its location suggests that it may be related to areas regulating fear, anxiety, reproductive behavior, and Arch Endocrinol Metab. 2018;62/3

Psychological symptoms in hypoparathyroidism

Arch Endocrinol Metab. 2018;62/3

the next 5 years (22). In fact, more evidence is necessary to determine the relation between these patients’ wellbeing and their biochemical parameters, as well as their response to different treatments. In conclusion, patients with hypoparathyroidism attending our outpatient clinics presented an increase in the number of self-report of psychological symptoms when compared with a control group, which was similar to results found in other studies. However, no correlation with hypocalcemia and clinical parameters was observed. Clinicians should recognize the presence of this psychological distress since hypoparathyroidism is a long-term disease that often develops early in life. Future studies are necessary to improve the understanding of the etiology and pathophysiology of the psychiatric changes associated with this disease. In addition, it is important to elucidate if the absence of PTH play a role on it. Disclosure: the authors declare that there are no conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.

REFERENCES 1. Shoback DM, Bilezikian JP, Costa AG, Dempster D, Dralle H, Khan AA, et al. Presentation of hypoparathyroidism: etiologies and clinical features. J Clin Endocrinol Metab. 2016;101(6):2300-12. 2. Asari R, Passler C, Kaczirek K, Scheuba C, Niederle B. Hypoparathyroidism after total thyroidectomy: a prospective study. Arch Surg. 2008;143(2):132-7. 3. Powers J, Joy K, Ruscio A, Lagast H. Prevalence and incidence of hypoparathyroidism in the United States using a large claims database. J Bone Miner Res. 2013;28(12):2570-6. 4.

Clarke BL, Leibson C, Emerson J, Ransom JC, Lagast H. Co-morbid-medical conditions associated with prevalent hypoparathyroidism: a population-based study. J Bone Miner Res. 2011;26:S182

5. Underbjerg L, Sikjaer T, Mosekilde L, Rejnmark L. Cardiovascular and renal complications to postsurgical hypoparathyroidism: a Danish nationwide controlled historic follow-up study. J Bone Miner Res. 2013;28(11):2277-85. 6. Hadker N, Egan J, Sanders J, Lagast H, Clarke BL. Understanding the burden of illness associated with hypoparathyroidism reported among patients in the PARADOX study. Endocr Pract. 2014;20(7):671-9. 7. Underbjerg L, Sikjaer T, Mosekilde L, Rejnmark L. Postsurgical hypoparathyroidism – risk of fractures, psychiatric diseases, cancer, cataract, and infections. J Bone Miner Res. 2014;29(11):2504-10. 8.

Underbjerg L, SikjaerT, Mosekilde L, Rejnmark L.The Epidemiology of Nonsurgical Hypoparathyroidism in Denmark: A Nationwide Case Finding Study. J Bone Miner Res. 2015;30(9):1738-44.

9. Aggarwal S, Kailash S, Sagar R, Tripathi M, Sreenivas V, Sharma R, et al. Neuropsychological dysfunction in idiopathic hypoparathyroidism and its relationship with intracranial

323

nociception. Moreover, the location of the receptor in the brain in primates and humans is very similar (19). These findings suggest that PTH action in the central nervous system may be important in neuromodulating several behaviors related to general well-being. Some studies have evaluated patients receiving synthetic PTH replacement, a treatment option that became available in the United States in 2015. Two publications (20,21) have shown improvements in quality of life following replacement with this type of hormone. However, these studies have not approached the relation between serum calcium and well-being scores, as done in our study and in that by Arlt and cols. (13). Vescini and cols. (20) used the 36-item shortform health survey (SF-36) questionnaire to evaluate 42 patients with hypoparathyroidism before and after PTH (1-34) replacement for 6 months. The authors observed a significant improvement in all domains assessed with the questionnaire, including physical and mental symptoms. The positive effects of the treatment with PTH (184) were confirmed in another study that evaluated 69 patients during 5 years, as part of a cohort study of patients with postsurgical and idiopathic hypoparathyroidism assessed before, during, and after discontinuation of hormone therapy, using the SF-36 questionnaire. After 2 months of treatment with PTH (1-84), the patients showed improvements in quality of life indices, a result that persisted in the last evaluation performed 5 years after treatment start. In addition, 27 patients who discontinued treatment due to side effects experienced decreased well-being at the end of that period when compared with those who maintained the therapy. These patients showed significant improvement in T-scores related to social interaction, vitality, mental health, limitations caused by physical health problems, and general health perception (21). In summary, treatment with PTH replacement was associated with improved quality of life in some studies with a longitudinal patient follow-up model. Clearly, patients with hypoparathyroidism present some degree of impaired quality of life. The neuropsychiatric complications of the disease are not well recognized and are often underestimated. Nevertheless, at the First International Conference on “The Diagnosis, Management, and Treatment of Hypoparathyroidism” in 2015, researchers have emphasized the quality of life as one of the areas requiring additional research within

Psychological symptoms in hypoparathyroidism

calcification and serum total calcium. Eur J Endocrinol. 2013;168(6):895-903.

Vienna, Austria. Disponível Acesso em: 7 out. 2016.

em:

http://www.R-project.org/.

10. Cho NL, Moalem J, Chen L, Lubitz CC, Moore FD Jr, Ruan DT. Surgeons and patients disagree on the potential consequences from hypoparathyroidism. Endocr Pract. 2014;20(5):427-46.

17. Mitchell DM, Regan S, Cooley MR, Lauter KB, Vrla MC, Becker CB, et al. Long-term follow-up of patients with hypoparathyroidism. J Clin Endocrinol Metab. 2012;97(12):4507-14.

11. Lopes MP, Kliemann BS, Bini IB, Kulchetscki R, Borsani V, Savi L, et al. Hypoparathyroidism and pseudohypoparathyroidism: etiology, laboratory features and complications. Arch Endocrinol Metabol. 2016;60(6):532-6.

18. Usdin TB, Gruber C, Bonner TI. Identification and functional expression of a receptor selectively recognizing parathyroid hormone, the PTH2 receptor. J Biol Chem. 1995;270(26):15455-8.

12. Laloni DT. Escala de Avaliação de Sintomas-90-R-SCL-90-R: adaptação, precisão e validade [tese]. Pontifícia Universidade Católica de Campinas, SP; 2001. Disponível em: http://tede. bibliotecadigital.puc-campinas.edu.br:8080/jspui/handle/ tede/389. Acesso em: 7 out. 2016. 13. Arlt W, Fremerey C, Callies F, Reincke M, Schneider P,Timmermann W, et al. Well-being, mood and calcium homeostasis in patients with hypoparathyroidism receiving standard treatment with calcium and vitamin D. Eur J Endocrinol. 2002;146(2):215-22. 14. Franke G. SCL-90-R. Die Symptom-Checkliste von Derogatis – Deutsche Version – Manual. Goettingen: Beltz Test GmbH; 1995. 15. Derogatis LR. SCL-90-R Administration, Scoring & Procedures Manual-II.Towson, MD: Clinical Psychometric Research; 1983. p. 14-5.

20. Vescini F, Santonati A, Palermo A, Maddaloni E, Bosco D, Spada A, et al. PTH(1-34) for surgical hypoparathyroidism: a prospective, open-label investigation of efficacy and quality of life. J Clin Endocrinol Metab. 2015;100(9):3590-7. 21. Cusano NE, Rubin MR, McMahon DJ, Irani D, Anderson L, Levy E, et al. PTH(1-84) is associated with improved quality of life in hypoparathyroidism through 5 years of therapy. J Clin Endocrinol Metab. 2014;99(10):3694-9. 22. Brandi ML, Bilezikian JP, Shoback D, Bouillon R, Clarke BL, Thakker RV, et al. Management of Hypoparathyroidism: Summary Statement and Guidelines. J Clin Endocrinol Metab. 2016;101(6):2273-83.

16. R Development Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing,

19. Bagó AG, Dimitrov E, Saunders R, Seress L, Palkovits M, Usdin TB, et al. Parathyroid hormone 2 receptor and its endogenous ligand tuberoinfundibular peptide of 39 residues are concentrated in endocrine, viscerosensory and auditory brain regions in macaque and human. Neuroscience. 2009;162(1):128-47.

324

Arch Endocrinol Metab. 2018;62/3

original article

Bouts of exercise elicit discordant testosterone: cortisol ratios in runners and non-runners Thiago Paes de Barros De Luccia1, José Eduardo Soubhia Natali1, Alexandre Moreira2, José Guilherme Chaui-Berlinck1, José Eduardo Pereira Wilken Bicudo1,3

Objective: The testosterone:cortisol ratio (T:C) is suggested to be used in order to examine whether physical exercise generates either a “catabolic environment” or an “anabolic environment”. The present study aims to evaluate the acute time-course profile of cortisol and testosterone due to an episode of physical exercise. A biphasic profile in the T:C ratio response was hypothesized. Materials and methods: Morning sessions of treadmill running at two different intensities (Heart Rate at 65% and 80% of the maximum cardiac reserve) were performed by 6 male non-runners (NR) and 12 trained male runners (subdivided into trained runners T1 and T2). Cortisol and testosterone were measured in saliva. NR and T1 ran for 30 minutes at both intensities, and T2 ran for 46 minutes (± 4.1) at 65% and 42 minutes (± 3.5) at 80%. Results: In the 80% heart rate target, both groups of runners showed the biphasic time-profile, while the non-runners group did not. However, at the 65% level, none of the groups presented the hypothesized biphasic response. Conclusions: A biphasic timeprofile in the testosterone:cortisol ratio can be seen in short-bout, high intensity exercise (treadmill running) during the morning in men trained for this specific physical activity. Arch Endocrinol Metab. 2018;62(3):325-31 Keywords Testosterone; cortisol; physical endurance

Departamento de Fisiologia, Instituto de Biociências, Universidade de São Paulo (USP), São Paulo, SP, Brasil 2 Departamento de Esporte, Escola de Educação Física e Esporte, Universidade de São Paulo (USP), São Paulo, SP, Brasil 3 School of Biological Sciences, University of Wollongong, NSW, Australia 1

ABSTRACT

Correspondence to: Thiago Paes de Barros De Luccia Departamento de Fisiologia, Instituto de Biociências, Universidade de São Paulo Rua do Matão, 101 05508-090 – São Paulo, SP, Brasil tpbl78@gmail.com Received on Oct/31/2017 Accepted on Jan/19/2018

INTRODUCTION

rom the classic fight-or-flight reaction to subtle dominance relationships in groups, changes are observed in specific hormonal values (1,2). Part of these changes can be understood in terms of their direct metabolic consequences, and both physical and psychological factors seem to play a role as causal factors (1). Although physical exercises are beneficial to health, such activities can generate harmful effects in both men (3) and women (4), a situation that is aggravated by excessive physical exercise (5,6). However, it is not fully known how the beneficial health and fitness-related effects of exercise come to end, subsequently becoming harmful to the human body. The study of hormonal changes related to physical exercise, which have been the focus of much research in the fields of physiology and health, may contribute to this subject (1). Although there is still no isolated marker capable of diagnosing training problems and/or overtraining (7-9), several indicators have been proposed in recent

Arch Endocrinol Metab. 2018;62/3

decades in search of this supposed turning point. One of these markers is the ratio between testosterone, considered an anabolic hormone, and cortisol, considered a catabolic hormone (10,11). The testosterone:cortisol ratio has been reported as a sign of anabolic status in athletes before competing (12), acute training response (13,14), psychophysiological responses to competition venue (15), training effects (16) and training motivation (12,17). An imbalance between the anabolic and catabolic milieu of the metabolism could be associated with certain components of the prescribed exercise (e.g. training volume and intensity) that should be monitored with the intention of improving sports and exercise performance whilst avoiding any deleterious effects from such activity (18). With regards to acute hormonal responses related to exercise, the testosterone: cortisol ratio (T:C) has been suggested to be used in order to examine whether physical exercise generates either a “catabolic 325

DOI: 10.20945/2359-3997000000042

Exercise and testosterone: cortisol ratio

326

biphasic response

T:C ratio (arbitrary units)

environment” or an “anabolic environment” (19). In a long-term training process, monitoring T:C has been advocated in order to verify the hormonal responses to a given training load, which could aid in preventing nonfunctional overreaching or overtraining syndrome (20). In this case, is believed that a catabolic environment might be persistently present (with elevated cortisol and low testosterone concentration) (21), associated with decline in performance, psychological changes, and neuroendocrine disorders. In general, a low testosterone concentration may be indicative of poor health in men (22), and extreme levels of circulating androgens, whether high or low, can have negative effects on women’s health (23). In women, chronic hypercortisolism has been associated with exercise-induced amenorrhea (24). In fact, the analysis of bone mineral density in this group changed the idea of amenorrhea in athletes being a benign event, and linked this phenomenon to premature bone loss associated with a decline in the levels of progesterone and estradiol (25). In a prospective, population-based study of chronic heart disease related to “stress” in men, a low ratio between testosterone and cortisol showed a strong positive association with the components of insulin resistance syndrome (26). According to some studies, the onset of physical activity stimulates the production of both cortisol and testosterone (27,28). However, cortisol remains elevated in the circulation following the episode of exercise, negatively affecting the synthesis of testicular testosterone (29). These events could generate a biphasic time-profile in the T:C ratio, a possible occurrence that has not yet been well studied. These events, which are considered in our hypotheses, are pictorially illustrated in Figure 1 (References used to build the chart (3032), in which we show the expected circadian variation in T:C and the biphasic response elicited by the physical activity described above. From this point of view, it is possible to observe that the added effect of acute exercise and the circadian rhythms of testosterone and cortisol obscures our understanding of the biphasic time-profile. Therefore, it is important to analyze the dynamics of these hormones throughout the day. A single isolated measurement might provide a poor picture depending on the sampling phase in relation to the action of cortisol. In fact, this putative biphasic time-profile could play a role in the contradictory data presented in the literature regarding the T:C ratio and acute physical activity (33).

08:00

circadian

09:00

10:00

11:00

Hour

Figure 1. Graphical representation of the T:C ratio throughout the morning. The dashed line shows the expected physiological circadian rhythm of T:C. The dotted line represents the hypothesized biphasic response due to a bout of exercise.

Questions in the study The present study aims to evaluate the acute timecourse profile of cortisol and testosterone due to an episode of physical exercise, and to answer the following questions: 1) Is there a biphasic profile in the response of the T:C ratio? 2) Is there a relationship between this profile and the intensity of acute physical exercise? 3) Is there a relationship between this profile and the duration of acute physical exercise? 4) Is there a difference in this profile between runners and non-runners?

MATERIALS AND METHODS Subjects Volunteers were selected and divided into two groups in accordance to their previous habitual physical exercise profile (the trained runner group (T, n = 12) and non-runner group (NR, n = 6)). The trained volunteers ran an average of 42 (± 19.2) km/week. They were randomly assigned to two subgroups, T1 and T2, described below. The non-runner volunteers had varied physical activities but they were not runners. Only male volunteers were selected for the study (mean age = 36 years (± 8.32); mean body mass = 68.8 kg (± 8.32). All participants had a clinical follow-up and signed an informed consent form (the study was approved by the Ethics Committee of the Bioscience Institute – University of Sao Paulo, CAAE 12937713.0.0000.5464). Arch Endocrinol Metab. 2018;62/3

Exercise and testosterone: cortisol ratio

Volunteers visited the laboratory for the first time in order to be physically evaluated and receive instructions about the experimental procedures. Subsequently, each volunteer performed a running session on the treadmill to establish the individual speed corresponding to the target heart rates (see below).

of beats to be attained was 6,300 (this number of beats resulted in sessions that lasted longer than the 30-min ones completed by the NR and T1 groups, but not so long as to compromise the similarity in the time of day for hormonal collection). From the preliminary data collection, the individual running time at R65 and R80 were computed for each volunteer as the session time = 6,300 / Rp. As such, the mean running time at R65 was 46 minutes (± 4.1) min and 42 minutes (± 3.5) at R80.

Exercise intensity zones – target heart rates – intensity versus demand

Running sessions

The exercise intensity was determined by the heart rate reserve method (34): Rp = Rrest + p × (Rmax – Rrest) (1) Where Rrest is the resting heart rate, Rmax is the inferred maximum heart rate of the subject (220 – age in years), p is the desired percentage of the maximum intensity (heart rate reserve) and Rp is the computed target heart rate. The protocols used were differentiated by the exercise intensity, the training level of the subjects, and a controlling variable (time or heartbeat) that limits the duration of the exercise. The protocols for each subset are summarized in Table 1. Table 1. Schematics of the experiments

Intensity

Fitness

Non-runners

Trained

Protocol

Fixed time

Fixed heartbeats

65%

NR R65

T1 R65

T2 R65

80%

NR R80

T1 R80

T2 R80

The experimental groups were separated into subsets with regards to training level (trained runners vs. non-runners) and the protocol (fixed time vs. fixed heartbeats). Individuals from these subsets performed the exercise at two intensities (65% vs. 80%).

The two exercise intensities selected were 65% and 80% of the heart rate reserve (R65 and R80, respectively). During the experimental procedures, treadmill speed was adjusted every 5 min based on the mean heart rate observed within this 5 min period in order to keep the heart rate as close as possible to the desired target zone. Each running session for volunteers belonging to the NR or T1 groups comprised a 30-min run at the individually predetermined R65 and R80. Therefore, these groups performed the exercise at fixed intensities. The T2 group performed the exercise in a fixed final energy demand. To impose such a similar final demand within the T2 group, the total number of heartbeats for a running session was fixed. The selected number Arch Endocrinol Metab. 2018;62/3

The experiments were conducted at a controlled room temperature of 21 (± 0.5) degrees Celsius. The average relative air humidity was 66% (± 10%). Estimated altitude of the laboratory: 785 meters above sea level. All running sessions were performed in the morning, beginning between 08:00 and 09:00. There were two running sessions, one day at 65% of the heart rate reserve (R65) and the other at 80% of the heart rate reserve (R80). The order of the sessions was randomly assigned to each volunteer. During the running session, the volunteer’s heart rate was continuously monitored by surface ECG. The electrodes were attached to the thorax and abdomen (CM5 – a modified Lead I configuration). The ECG was acquired using a sampling rate of 1000 Hz and standard filters with a MP30 interface and the Biopac Student Lab Pro software (Biopac Systems Inc., Goleta, CA, USA).

Hormone collection and analysis The saliva samples for the hormonal analysis were obtained by direct salivation into plastic tubes, stored in a thermal bag at 10° Celsius and then in a freezer (at - 20° Celsius). Samples were taken at three different occasions, at approximately the same time of day, in order to avoid circadian variations. Basal hormones. Basal samples were collected at 08:00 and 11:00 on days when the volunteer did not perform any physical exercise. The volunteers were required to eat a small meal with around 40 grams of carbohydrate 1 hour before the first saliva sample (07:00). Experimental hormonal data. On the experimental days, the saliva was collected prior to physical activity (around 08:00 – PRE; the volunteers were instructed to maintain a similar routine pattern as on the day of basal collection), immediately after the bout of exercise 327

Preliminary data collection

Exercise and testosterone: cortisol ratio

(around 08:45 – POST), and later on after the activity (around 11:00 – LATE). The volunteers conducted their normal daily activities between the POST and the LATE samples, but were oriented not to ingest large amounts of food. Salivary cortisol and testosterone were measured using a specific enzyme-linked immunosorbent assay (ELISA) for each hormone (DiaMetra – Salivary Steroid Hormones, Italy; intra-assay variation ≤ 10% and interassay variation ≤ 8.3% for the cortisol kit, and intraassay variation ≤ 8.0% and inter-assay variation ≤ 13.2% for the testosterone kit). Testosterone was measured in picograms per milliliter, and cortisol in nanograms per milliliter.

Statistical approaches First stage: In order to characterize the primary differences between the groups, a one-way ANOVA or two-way ANOVA with repeated measures was employed, as indicated in the results section. Statistical significance was set at p £ 0.05 in this stage. Second stage: Heart rate. The question of whether the NR group has similar heart rates as the T1 and T2 groups as a whole was addressed by a t-test using the Holm-Bonferroni method for post-hoc significance. A priori, there were four potential comparisons: NR vs. T1; NR vs. T2, NR vs. (T1 and T2), T1 vs. T2. Therefore, the adjusted significance levels (critical p-values) were 0.013, 0.017, 0.025, 0.050 for the ordered p-values obtained in the comparisons. Time-profile of the T:C ratio. In this case, for each group, there were three potential comparisons: PRE vs. POST; PRE vs. LATE; POST vs. LATE. In a similar manner to the heart rate comparisons, the differences were addressed by a t-test using the Holm-Bonferroni method for post-hoc significance. The adjusted significance levels were 0.017, 0.025, 0.050 for the ordered p-values obtained in the comparisons.

RESULTS The first step in the analysis was the characterization of the groups based on their heart rates in each running section (Table 2). A one-way ANOVA comparison of the heart rates showed an F-critical (2.15) of 3.68 and the F values obtained for the R65 and R80 were 5.53 and 3.72, respectively (p-values of 0.015 and 0.048), leading to the conclusion that the groups are dissimilar in their heart rates. 328

Comparisons between groups (see statistical approaches section) revealed that NR is different from the T group as a whole (p = 0.004 for R65, and p = 0.013 for R80), while T1 and T2 are not different (p = 0.596 for R65, and p = 0.887 for R80). Therefore, despite the fact that the individuals in the NR group are not sedentary, their heart rates are higher than the those of individuals in the T1 and T2 groups, who had trained for running. Table 3 contains the raw T:C ratio data for all the volunteers in the present study. A two-Way ANOVA with repeated measures detected that the 80% intensity zone generates a difference among the groups (detected as the groups themselves, p = 0.00004, and in the sample time, p = 0.01), while at the 65% level these differences are not significant. To complete the analysis and remain on track with the focus of the present study, i.e. whether there is a biphasic time-profile in the T:C ratio elicited by physical activity, the time-profile (i.e. PRE, POST and LATE) within each group was characterized as described in the statistical analysis section: • Non-runner group: there were no differences over time both at the 65% and 80% intensity zones. • T1 runner group: in the 80% intensity zone, POST > PRE (p = 0.0059). • T2 runner group: in the 65% intensity zone, POST > PRE (p = 0.0231) and LATE > PRE (p = 0.0021), although POST and LATE were not different. In the 80% intensity zone, POST > PRE (p = 0.0083). Figures 2A and 2B are graphical representations of these results superimposed on the circadian and biphasic response hypothesis depicted in Figure 1.

DISCUSSION A biphasic time-profile in the testosterone:cortisol ratio can be seen in short-bout, high intensity exercise (treadmill running) during the morning in men trained for this specific physical activity. Irrespective of the level of training and intensity, it seems that a session of exercise in the morning causes a disruption in the circadian rhythm of the T:C ratio. How long such a disruption remains during the rest of the day deserves further investigations. Moreover, whether the modification in the T:C ratio promotes a catabolic milieu or not remains an open question, as well as the use of this ratio to evaluate the performance of athletes in acute physical exercise. Arch Endocrinol Metab. 2018;62/3

Exercise and testosterone: cortisol ratio

Table 2. Heart rates Group NR

Subject

R80

R65 bpm

min

thb

bpm

min

thb

150

4500

164

4920

159

4770

169

5070

145

4350

161

4830

154

4620

163

4890

140

4200

153

4590

155

4650

160

4800

146

4380

157

4710

139

4170

156

4680

133

3990

151

4530

128

3840

141

4230

123

3690

141

4230

135

4050

156

4680

140

6300

161

6300

128

6300

140

6300

146

6300

157

6300

119

6300

131

6300

140

6300

150

6300

150

6300

158

6300

Heart rates (bpm), running time (min) and total heartbeats (thb) for each individual in each protocol (65% and 80%).

Table 3. T:C data

Subject

R80

R65 PRE

POST

LATE

PRE

POST

LATE

11.40

17.67

13.41

13.58

13.17

25.38

15.12

14.75

13.59

9.68

13.85

14.76

17.14

12.11

18.71

13.25

8.48

22.50

11.72

15.60

11.88

18.94

23.63

15.16

20.80

19.31

22.58

16.87

23.46

23.64

9.19

12.26

15.85

21.00

21.34

14.26

10.47

15.68

9.14

12.92

14.69

10.00

19.46

11.84

14.74

10.20

17.37

17.30

14.89

18.06

13.00

13.13

16.45

13.51

16.96

17.77

17.23

11.26

15.67

15.81

15.92

20.05

22.39

13.35

15.27

18.13

12.30

14.03

15.15

13.05

16.62

15.03

9.09

13.32

12.36

21.40

23.23

15.84

10.92

11.21

14.48

15.38

26.89

21.15

9.31

17.20

16.60

13.37

23.43

24.07

19.76

22.11

26.52

17.64

23.11

23.57

9.67

13.37

18.43

18.47

22.15

23.32

12.81

14.85

16.47

17.88

23.77

15.69

Group

PRE: prior to running; POST: immediately after running; LATE: around 2.5 hours after the running section.

Arch Endocrinol Metab. 2018;62/3

329

Exercise and testosterone: cortisol ratio

T:C ratio (arbitrary units)

NR and T1

08:00

09:00

10:00

11:00

Hour

T:C ratio (arbitrary units)

T1 and T2 NR

08:00

09:00

10:00

11:00

Hour

Figure 2. Graphical representation of the T:C results shown in Table 2. (A) Target heart rate of 65% (R65). (B) Target heart rate of 80% (R80). The curves shown in Figure 1 are in faded gray (circadian and biphasic).

The results of the present study suggest that biphasic behavior in the T:C profile seems to depend on the energetic demand of the exercise, as well as the training status of the individuals. Therefore, in the 80% heart rate target, both groups of runners show the biphasic time-profile, while the non-runner group does not. However, at the 65% level, none of the groups present the hypothesized biphasic response. This also answers our questions regarding whether there is an association between the biphasic time-profile and the intensity or duration of the acute physical exercise. The answer is yes, there is such an association between the biphasic time-profile and both intensity and duration, in the sense that biphasic behavior of the T:C ratio could only be detected in more elevated levels of demand, i.e. at the 80% heart rate target. Our final query was in respect to the putative biphasic time-profile and the individual’s training status. The results shown in Figure 2B indicate that training specificity plays a role in eliciting a biphasic response. 330

Here, in a biphasic mode or otherwise, and despite the specific training level, it seems that a bout of acute exercise is able to perturb the circadian rhythm of T:C, reducing the ratio found a few hours after physical activity. An exception for that was the T2 group in the 65% target, and we have no hypothesis or explanation for this result. Tremblay and cols. (35) describe a significant increase in cortisol only after 120 minutes of running at 55% of O2max in trained runners (contrasting with running sessions of 40 and 80 minutes). Jacks and cols. (36) evaluated the salivary cortisol responses in 60-minute exercise sessions on a cycle ergometer at three intensities (44.5, 62.3 and 76% of O2max) in 10 untrained men. The hormone fell in the low and moderate intensities and increased in the high intensity. Despite the intensity, cortisol sampled at 40 minutes during the bouts of exercise showed no significant difference (36). Hoffman (37) describes a dual response of testosterone depending on the duration of the physical activity, with an increase in short bouts (less than 2 ½ hours) and a reduction in exercises lasting longer than 3 hours. Therefore, the picture of a “rise and fall” in cortisol and testosterone level is not evident prima facie. Actually, many of these studies do not consider the biphasic time-profile and, as a consequence, additional data acquisition (e.g. at different post-exercise times) could lead to different results. In a recent meta-analysis (33), the authors concluded that salivary cortisol, and particularly testosterone, are highly influenced by both the study design and sampling time. The present results seem to support this, as described above. In addition, the time (period of the day) of sampling the hormones should be taken into account due to the circadian variation of cortisol and testosterone, both in their circulating levels and in their response curves. Financial support: this study was supported by a research grant from Fundação de Amparo à Pesquisa do Estado de São Paulo (Fapesp – grant #2014/14296-1). Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1.

Mastorakos G, Pavlatou M, Diamanti-Kandarakis E, Chrousos GP. Exercise and the stress system. Hormones (Athens). 2005;4(2): 73-89. Arch Endocrinol Metab. 2018;62/3

Exercise and testosterone: cortisol ratio

3. Gouarné C, Groussard C, Gratas-Delamarche A, Delamarche P, Duclos M. Overnight urinary cortisol and cortisone add new insights into adaptation to training. Med Sci Sports Exerc. 2005;37(7):1157-67. 4. Mountjoy M, Sundgot-Borgen J, Burke L, Carter S, Constantini N, Lebrun C, et al. The IOC consensus statement: beyond the Female Athlete Triad--Relative Energy Deficiency in Sport (RED-S). Br J Sports Med. 2014;48(7):491-7. 5. Kim YJ, Lee YH, Park Y, Jee H. Relationship of exercise habits and cardiac events to resting blood pressure in middle-aged amateur marathoners. Gazz Med Ital. 2014;173:273-81. 6. Möhlenkamp S, Lehmann N, Breuckmann F, Bröcker-Preuss M, Nassenstein K, Halle M, et al.; Marathon Study Investigators; Heinz Nixdorf Recall Study Investigators. Running: the risk of coronary events: Prevalence and prognostic relevance of coronary atherosclerosis in marathon runners. Eur Heart J. 2008;29(15):1903-10. 7. Armstrong LE, VanHeest JL. The unknown mechanism of the overtraining syndrome: clues from depression and psychoneuroimmunology. Sports Med. 2002;32(3):185-209. 8. Borresen J, Lambert MI. Autonomic control of heart rate during and after exercise: measurements and implications for monitoring training status. Sports Med. 2008;38(8):633-46. 9. Smith TB, Hopkins WG, Lowe TE. Are there useful physiological or psychological markers for monitoring overload training in elite rowers? Int J Sports Physiol Perform. 2011;6(4):469-84. 10. Kuhn C. Anabolic steroids. Recent Prog Horm Res. 2002;57:411-34. 11. Christiansen JJ, Djurhuus CB, Gravholt CH, Iversen P, Christiansen JS, Schmitz O, et al. Effects of cortisol on carbohydrate, lipid, and protein metabolism: studies of acute cortisol withdrawal in adrenocortical failure. J Clin Endocrinol Metab. 2007;92(9):3553-9. 12. Cook CJ, Crewther BT, Kilduff LP. Are free testosterone and cortisol concentrations associated with training motivation in elite male athletes? Psychol Sport Exercise. 2013;14(6):882-5. 13. Beaven CM, Gill ND, Cook CJ. Salivary testosterone and cortisol responses in professional rugby players after four resistance exercise protocols. J Strength Cond Res. 2008;22(2):426-32. 14. Beaven CM, Gill ND, Ingram JR, Hopkins WG. Acute salivary hormone responses to complex exercise bouts. J Strength Cond Res. 2011;25(4):1072-8. 15. Arruda AF, Aoki MS, Miloski B, Freitas CG, Moura NR, Moreira A. Playing match venue does not affect resting salivary steroids in elite Futsal players. Physiol Behav. 2016;155:77-82. 16. Gomes RV, Moreira A, Lodo L, Nosaka K, Coutts AJ, Aoki MS. Monitoring training loads, stress, immune-endocrine responses and performance in tennis players. Biol Sport. 2013;30(3):173-80. 17. Cook CJ, Crewther BT. The effects of different pre-game motivational interventions on athlete free hormonal state and subsequent performance in professional rugby union matches. Physiol Behav. 2012;106(5):683-8. 18. Kraemer WJ, Ratamess NA. Hormonal responses and adaptations to resistance exercise and training. Sports Med. 2005;35(4):339-61. 19. Hayes LD, Bickerstaff GF, Baker JS. Interactions of cortisol, testosterone, and resistance training: influence of circadian rhythms. Chronobiol Int. 2010;27(4):675-705.

Arch Endocrinol Metab. 2018;62/3

20. Meeusen R, Duclos M, Foster C, Fry A, Gleeson M, Nieman D, et al.; European College of Sport Science; American College of Sports Medicine. Prevention, diagnosis, and treatment of the overtraining syndrome: joint consensus statement of the European College of Sport Science and the American College of Sports Medicine. Med Sci Sports Exerc. 2013;45(1):186-205. 21. Adlercreutz H, Härkönen M, Kuoppasalmi K, Näveri H, Huhtaniemi I, Tikkanen H, et al. Effect of training on plasma anabolic and catabolic steroid hormones and their response during physical exercise. Int J Sports Med. 1986;7 Suppl 1:27-8. 22. Araujo AB, Dixon JM, Suarez EA, Murad MH, Guey LT, Wittert GA. Clinical review: endogenous testosterone and mortality in men: a systematic review and meta-analysis. J Clin Endocrinol Metab. 2011;96(10):3007-19. 23. Davis SR, Wahlin-Jacobsen S. Testosterone in women – the clinical significance. Lancet Diabetes Endocrinol. 2015;3(12):980-92. 24. Olson BR. Exercise-induced amenorrhea. Am Fam Physician. 1989;39(2):213-21. 25. Drinkwater BL, Nilson K, Ott S, Chesnut CH 3rd. Bone mineral density after resumption of menses in amenorrheic athletes. JAMA. 1986;256(3):380-2. 26. Smith GD, Ben-Shlomo Y, Beswick A, Yarnell J, Lightman S, Elwood P. Cortisol, testosterone, and coronary heart disease: prospective evidence from the Caerphilly study. Circulation. 2005;112(3):332-40. 27. Bloom SR, Johnson RH, Park DM, Rennie MJ, Sulaiman WR. Differences in the metabolic and hormonal response to exercise between racing cyclists and untrained individuals. J Physiol. 1976;258(1):1-18. 28. Wilkerson JE, Horvath SM, Gutin B. Plasma testosterone during treadmill exercise. J Appl Physiol Respir Environ Exerc Physiol. 1980;49(2):249-53. 29. Cumming DC, Quigley ME, Yen SS. Acute suppression of circulating testosterone levels by cortisol in men. J Clin Endocrinol Metab. 1983;57(3):671-3. 30. Bird SP, Tarpenning KM. Influence of circadian time structure on acute hormonal responses to a single bout of heavy-resistance exercise in weight-trained men. Chronobiol Int. 2004;21(1):131-46. 31. Hayes LD, Grace FM, Kilgore JL, Young JD, Baker J. Diurnal variation of cortisol, testosterone, and their ratio in apparently healthy males. Sport SPA. 2012;9:5-13. 32. Hayes LD, Grace FM, Kilgore J, Young JD, Baker J. Salivary hormone response to maximal exercise at two time points during the day. Sport SPA. 2013;10:25-30. 33. Hayes LD, Grace FM, Baker JS, Sculthorpe N. Exercise-induced responses in salivary testosterone, cortisol, and their ratios in men: a meta-analysis. Sports Med. 2015;45(5):713-26. 34. ACSM. ACSM’s Guidelines for Exercise Testing and Prescription. J Can Chiropr Assoc. 2013;9:456. 35. Tremblay MS, Copeland JL, Van Helder W. Influence of exercise duration on post-exercise steroid hormone responses in trained males. Eur J Appl Physiol. 2005;94(5-6):505-13. 36. Jacks DE, Sowash J, Anning J, McGloughlin T, Andres F. Effect of exercise at three exercise intensities on salivary cortisol. J Strength Cond Res. 2002;16(2):286-9. 37. Hoffman J. Physiological aspects of Sport training and performance. Florida: Human Kinetics; 2014.

331

2. Czoty PW, Gould RW, Nader MA. Relationship between social rank and cortisol and testosterone concentrations in male cynomolgus monkeys (Macaca fascicularis). J Neuroendocrinol. 2009;21(1):68-76.

original article

Maternal hypothyroxinemia in the first trimester of gestation and association with obstetric and neonatal outcomes and iron deficiency: a prospective Brazilian study Pedro Weslley Rosario1, Luis Fernando Faria Oliveira2, Maria Regina Calsolari1

ABSTRACT Programa de Pós-Graduação e Serviço de Endocrinologia, Santa Casa de Belo Horizonte, Belo Horizonte, MG, Brasil 2 Fundação Universidade de Itaúna, Itaúna, MG, Brasil 1

Correspondence to: Pedro Weslley Rosario Instituto de Ensino e Pesquisa, Santa Casa de Belo Horizonte Rua Domingos Vieira, 590 30150-240 – Belo Horizonte, MG, Brasil pedrowsrosario@gmail.com Received on Sept/14/2017 Accepted on Jan/19/2018 DOI: 10.20945/2359-3997000000043

Objective: To evaluate the association of isolated hypothyroxinemia in the first trimester with obstetric and neonatal outcomes and iron deficiency. Subjects and methods: The study was prospective. Women who had become pregnant spontaneously were initially selected. Next, antithyroid peroxidase antibodies (TPOAb), free T4 (FT4), total T4 (TT4), TSH, and ferritin were measured. TPOAb-positive women were excluded. The final sample consisted of 596 women with serum TSH between 0.1 and 2.5 mIU/l. Hypothyroxinemia was defined as FT4 < 0.86 ng/dL and < 0.92 ng/dL, corresponding to the 5th and 10th percentiles, respectively, and TT4 < 7.8 ng/dL. None of the pregnant women was treated with levothyroxine until the end of pregnancy. Results: The women ranged in age from 18 to 36 years, with a median gestation of 9 weeks. T4 levels were not correlated with BMI or maternal TSH. Isolated hypothyroxinemia was observed in 4.3% (FT4 < 0.86 ng/dL), 9% (FT4 < 0.92 ng/dL), and 7% (TT4 < 7.8 ng/dL) of the pregnant women. The frequencies of obstetric and neonatal outcomes were similar in women with versus without hypothyroxinemia. In women without iron deficiency, 8.4%, 3.9%, and 6.5% had FT4 < 0.92 ng/dl, FT4 < 0.86 ng/dL and TT4 < 7.8 mg/ dL, respectively. These frequencies of hypothyroxinemia were significantly higher among women with iron deficiency (20.7%, 14.8% and 17.2%, respectively). Conclusions: This prospective Brazilian study found no association between isolated hypothyroxinemia in the first trimester of gestation and obstetric or neonatal outcomes, but an association was demonstrated with iron deficiency. Arch Endocrinol Metab. 2018;62(3):332-6 Keywords Pregnancy; first trimester; hypothyroxinemia; iron deficiency; obstetric and neonatal outcomes

INTRODUCTION

hen screening for thyroid dysfunction in pregnancy is indicated, it should be performed in the first prenatal visit, which usually occurs in the first trimester. Possible consequences of subclinical hypothyroidism and isolated maternal hypothyroxinemia also appear to be more worrisome when they occur in the first trimester of gestation, a period when the fetus still does not produce thyroid hormone and depends on maternal thyroxine (T4) (1). It is therefore more important to know the impact of these conditions when diagnosed in the early stages of pregnancy (1,2). Isolated hypothyroxinemia is diagnosed in the presence of serum TSH within the normal range associated with T4 concentrations below the 2.5th, 5th or 10th percentile (2-4). Regarding TSH, except when 332

the assay- and population-specific references values are available, the range of 0.1 to 2.5 mIU/l has been accepted for the definition of maternal euthyroidism in the first trimester (2,3,5-7). It is worth noting that this range does also seem to be adequate for our population (8). Iodine deficiency appears to the most important factor in the etiology of isolated hypothyroxinemia, but other factors have also been implicated, such as environmental disruptors, obesity, and iron deficiency (4). No consensus exists regarding the consequences and need for levothyroxine (L-T4) replacement therapy in isolated hypothyroxinemia (4). The American College of Obstetricians and Gynecologists (ACOG) does not consider this condition in its guidelines (9). The American Thyroid Association (ATA) does not recommend its treatment (3), while the Endocrine Arch Endocrinol Metab. 2018;62/3

Hypothyroxinemia in pregnant women

SUBJECTS AND METHODS The study was prospective and was approved by the local Research Ethics Committee (8).

Selection The population studied was from the metropolitan region of Belo Horizonte (Minas Gerais, Brazil). One thousand and eighty two pregnant women who underwent prenatal exams at a clinical analysis laboratory and who had become pregnant spontaneously were initially interviewed and examined (8). Women who met the clinical criteria shown in Table 1 (n = 748) were first selected. Next, anti-thyroid peroxidase antibodies (TPOAb), free T4 (FT4), total T4 (TT4), TSH, and ferritin were measured. TPOAb-positive pregnant women (n = 38) were excluded. Women with hyperemesis gravidarum, twin pregnancy or trophoblastic disease (n = 50) were also excluded. The sample consisted of 660 women and 596 with serum TSH between 0.1 and 2.5 mIU/l (2,3,5-7,8) were selected.

Protocol At the time of the study, and still today, T4 measurement is not necessary in women with TSH ≤ 2.5 mIU/L without TPOAb. There was also, and continues to be, no consensus recommendation regarding the treatment of hypothyroxinemia. Hence, in order to avoid interference with the results, the protocol was approved Arch Endocrinol Metab. 2018;62/3

foreseeing the non-repetition of thyroid function tests during pregnancy and was thus executed. None of the pregnant women was treated with L-T4 until the end of pregnancy. Regarding iron supplementation, the women only received the recommended daily doses in all pregnant women without additional supplementation in those with iron deficiency.

Outcomes Obstetric and neonatal outcomes were compared between women with FT4 < 0.86 ng/dL and < 0.92 ng/dL and those with FT4 ≥ 0.92 ng/dL. These values correspond, respectively, to the 5th and 10th percentiles of FT4 concentrations obtained for the initially selected 660 TPOAb-negative women (8). We also compared women with TT4 < 7.8 ng/dL and ≥ 7.8 ng/dL. This value is provided by the assay as the lower limit of normal for the first trimester of gestation and corresponds exactly to 1.5 times the lower limit of the reference range for non-pregnant healthy adults (12). The outcomes (reported in Table 2) were predefined before the start of the study based on those used in the largest series that evaluated the association between hypothyroxinemia and pregnancy outcomes (13). The definitions of hypothyroxinemia were also pre-established.

Laboratory Iron deficiency was defined as a ferritin concentration < 15 ng/mL (14). TSH, FT4, TT4 and TPOAb were measured with a chemiluminescent assay (Diagnostic Products Corporation, Los Angeles, CA). Serum samples were obtained from the women in the morning (at about 8 a.m.) after an 8- to 10-h fast (8). Table 1. Inclusion criteria Clinical criteria Absence of thyroid disease, no current or previous treatment with antithyroid drugs or L-T4, no history of 131I therapy or thyroidectomy No use of potentially interfering medications such as dopaminergic agonists or antagonists, neuroleptics, corticosteroids, estrogen, amiodarone, interferon, lithium, anticonvulsants, metformin, octreotide; or recent (in the past 8 weeks) exposure to iodinated contrast agents No history of head and neck external radiotherapy Absence of type 1 diabetes or other autoimmune diseases

Society proposes partial replacement therapy (5). According to the European Thyroid Association (ETA), treatment can be considered when hypothyroxinemia is detected in the first trimester, but not when it is detected in the second or third trimester (2). A large recent study found no benefit of L-T4 replacement therapy in isolated hypothyroxinemia, but one limitation of that study is that this therapy was generally initiated only in the 18th week of gestation (10,11). In view of the lack of consensus regarding the management of isolated hypothyroxinemia, notably when detected in the first trimester, the objective of this study was to evaluate the association of this condition with obstetric and neonatal outcomes. The association with iron deficiency was also explored. To our knowledge, no Brazilian study has so far evaluated these associations.

No family history of thyroid disease Absence of goiter or any palpable thyroid anomaly Absence of ophthalmopathy Complementary criteria: ≤ 12 weeks gestation

333

Hypothyroxinemia in pregnant women

Table 2. Obstetric outcomes Obstetric outcomes

FT4 (ng/dL)

TT4 (mg/dL)

< 0.86* (n = 26)*

< 0.92** (n = 54)**

≥ 0.92* ** (n = 542)* **

< 7.8*** (n = 42)***

≥ 7.8*** (n = 554)***

Hypertensive disease of pregnancy

2 (7.7%)*

5 (9.2%)**

47 (8.6%)* **

4 (9.5%)***

50 (9%)***

Gestational mellitus diabetes

3 (11.5%)*

6 (11.1%)**

56 (10.3%)* **

5 (12%)***

60 (10.8%)***

Placental abruption

0**

2 (0.3%)* **

0***

2 (0.3%)***

Delivery before 34 weeks of gestation

1 (1.8%)**

5 (0.9%)* **

1 (2.4%)***

5 (0.9%)***

Delivery before 37 weeks of gestation

1 (3.8%)*

2 (3.7%)**

15 (2.7%)* **

2 (4.7%)***

16 (2.8%)***

2 (3.7%)**

16 (2.9%)* **

2 (4.7%)***

16 (2.8%)***

Fetal loss (after initial evaluation)

FT4: free T4; TT4: total T4. * FT4 < 0.86 ng/dL versus ≥ 0.92 ng/dL: p ≥ 0.5 in all comparisons. ** FT4 < 0.92 ng/dL versus ≥ 0.92 ng/dL: p ≥ 0.65 in all comparisons. *** TT4 < 7.8 mg/dL versus TT4 ≥ 7.8 mg/dL: p ≥ 0.35 in all comparisons.

Statistical The Spearman or Pearson correlation test was used to analyze the correlation between T4 and body mass index (BMI) and TSH. The t-test or Mann-Whitney test and Fisher’s exact test or the c2 test were used for comparison between groups. A P value < 0.05 was considered statistically significant.

RESULTS The characteristics of the 596 patients are shown in Table 3. FT4 levels were not correlated with BMI (p = 0.6) or maternal TSH (p = 0.1). TT4 levels were also not correlated with BMI (p = 0.7) or maternal TSH (p = 0.15). Tables 2 and 4 show the obstetric and neonatal outcomes, respectively, according to maternal T4 concentrations in the first trimester of gestation. The groups were similar in terms of age, BMI and maternal TSH levels. No difference in the frequencies of obstetric or neonatal outcomes was observed, regardless of the definition of isolated hypothyroxinemia. Table 3. Characteristics of the 596 patients Age [range, median (years)]

18-36 (27)

Body mass index [range, median (kg/m )]

18-30 (22.6)

Gestational age [range, median (weeks)]

7-12 (9)

Primigravidae TSH [range, median (mIU/L)] Hypothyroxinemia

Iron deficiency

334

312 (52.3%) 0.1-2.5 (0.98) Free T4 < 0.86 ng/dL: 26 (4.3%) Free T4 < 0.92 ng/dL: 54 (9%) Total T4 < 7.8 mg/dL: 42 (7%) 29 (4.8%)

Iron deficiency was diagnosed in 29 (4.8%) pregnant women. This frequency was 4.2% in women with FT4 ≥ 0.92 ng/dL. The frequencies of iron deficiency were significantly higher in women with FT4 < 0.86 ng/dL and < 0.92 ng/dL (hypothyroxinemia), 15.4% (p = 0.03) and 11.1% (p = 0.04), respectively. Women with TT4 < 7.8 mg/dL (hypothyroxinemia) also exhibited a higher frequency of iron deficiency than women with TT4 ≥ 7.8 mg/dL (12% versus 4.3%, p = 0.045). In women without iron deficiency, 8.4%, 3.9% and 6.5% had FT4 < 0.92 ng/dL, FT4 < 0.86 ng/dL and TT4 < 7.8 mg/dL respectively. These frequencies of hypothyroxinemia were significantly higher among women with iron deficiency, 20.7% (p = 0.038), 14.8% (p = 0.032) and 17.2% (p = 0.045), respectively.

DISCUSSION Some characteristics of the present study should be highlighted. This was a prospective study. Hypothyroxinemia was defined using three different criteria of T4 concentration. Regarding TSH, a cutoff of 2.5 mIU/l was used since it is still the most accepted value for the first trimester of gestation (2,3,5-7) and is compatible with the normal range for our population (8). In addition, an overestimated TSH cutoff may compromise the conclusions because of the inclusion of women with subclinical hypothyroidism. Women with TPOAb were excluded to rule out the possible interference of thyroid autoimmunity with the outcomes. The women were evaluated in the first trimester since eventual repercussions of isolated hypothyroxinemia would be greater during this period (1,2,11) and treatment is particularly considered in this stage of pregnancy (2). None of Arch Endocrinol Metab. 2018;62/3

Hypothyroxinemia in pregnant women

Table 4. Neonatal outcomes Neonatal outcomes

FT4 (ng/dL)

Birth weight < 2,500 g

TT4 (mg/dL)

< 0.86* (n = 26)*

< 0.92** (n = 54)**

≥ 0.92* ** (n = 542)* **

< 7.8*** (n = 42)***

≥ 7.8*** (n = 554)***

1 (3.8%)*

2 (3.7%)**

32 (5.9%)* **

2 (4.7%)***

33 (5.9%)***

Birth weight < 1,500 g

1 (1.8%)**

8 (1.4%)* **

1 (2.3%)***

8 (1.6%)***

Need for intensive therapy

1 (1.8%)**

11 (2%)* **

1 (2.3%)***

11 (2%)***

Need for mechanical ventilation > 24 h

1 (1.8%)**

11 (2%)* **

1 (2.3%)***

11 (2%)***

Necrotizing enterocolitisa

0**

1 (0.2%)* **

0***

1 (0.2%)***

Intraventricular hemorrhage grade 3 or 4

0**

0* **

0***

Major congenital malformations

1 (1.8%)**

5 (0.9%)* **

0***

6 (1%)***

Neonatal death (< 28 days)

0**

3 (0.5%)* **

0***

3 (0.5%)***

FT4: free T4; TT4: total T4. With need for surgery. * FT4 < 0.86 ng/dL versus ≥ 0.92 ng/dL: p ≥ 0.8 in all comparisons. ** FT4 < 0.92 ng/dL versus ≥ 0.92 ng/dL: p ≥ 0.45 in all comparisons. *** TT4 < 7.8 mg/dL versus TT4 ≥ 7.8 mg/dL: p ≥ 0.5 in all comparisons.

the women was treated with L-T4 during pregnancy. Finally, few studies have investigated the relationship between hypothyroxinemia and iron deficiency. To our knowledge, this is the first Brazilian study analyzing the association of hypothyroxinemia with obstetric and neonatal outcomes and iron deficiency. Regarding the association of hypothyroxinemia with iron deficiency, our results confirm the data of a Chinese observational study, which also evaluated women in the first trimester of gestation from an iodine-sufficient area (15). Since it would be more than a coincidence that iron and iodine deficiencies are the consequence of broad nutritional deficiency or that isolated hypothyroxinemia (in the absence of hypothyroidism) causes iron deficiency, iron deficiency is more likely to contribute to hypothyroxinemia (15). It is possible that iodine sufficiency influences the consequence of iron deficiency in terms of thyroid dysfunction, whether it is predominantly hypothyroxinemia (without TSH elevation) in an iodine-sufficient population (15) like ours, or subclinical hypothyroidism in an iodinedeficient population (16). However, further clinical studies involving pregnant women are necessary to consistently confirm this association. We found no association between isolated hypothyroxinemia in the first trimester of gestation and obstetric or neonatal outcomes. Casey and cols., comparing 16,011 women with normal TSH and T4 and 233 with hypothyroxinemia before 20 weeks of gestation, also observed no association between this alteration and any of the obstetric or neonatal outcomes analyzed (13). Cleary-Goldman and cols. Arch Endocrinol Metab. 2018;62/3

evaluated 10,021 pregnant women with normal TSH and T4 and 232 with hypothyroxinemia (17). In the first trimester, hypothyroxinemia was associated with preterm labor (adjusted odds ratio [aOR] 1.62; 95% confidence interval [CI] 1.00-2.62) and macrosomia (aOR 1.97; 95% CI 1.37-2.83) (17). In the second trimester, it was associated with gestational diabetes (aOR 1.7; 95% CI 1.02-2.84) (17). Finally, a third study involving more than 5,000 women revealed an association of hypothyroxinemia with preterm labor (18), but not with hypertensive disease in pregnancy (19). A meta-analysis that included these and other smaller studies found only a statistically significant increased risk of placental abruption (OR 2.3; 95% CI: 1.1–4.8) compared to euthyroid controls (20). Taken together, these findings show that the association of hypothyroxinemia with obstetric and neonatal outcomes is still inconsistent (4). In fact, there is no reproducibility of the results in the different studies and no cause-effect relationship could be established. We recognize some limitations of the study. First, the number of women with hypothyroxinemia may have been insufficient to obtain statistical significance of small differences compared to euthyroid women. Second, since this was not an intervention study, the objective was limited to demonstrating the association between iron deficiency and hypothyroxinemia, but we did not evaluate the effect of iron supplementation on T4 concentrations. Finally, we analyzed obstetric and neonatal outcomes but not parameters of fetal brain development, which could also be compromised by isolated hypothyroxinemia (3). Although an 335

Hypothyroxinemia in pregnant women

association between hypothyroxinemia at the beginning of pregnancy and some impairment of fetal brain development has been reported in some series (3,20), two randomized studies found no beneficial effect of L-T4 therapy on this aspect (10,21). In conclusion, this prospective Brazilian study found no association of isolated hypothyroxinemia in the first trimester of gestation with obstetric or neonatal outcomes. Despite the need for confirmation in larger series, this result weakens the concern regarding serum T4 in pregnant women with normal TSH, i.e., with isolated hypothyroxinemia. In contrast, an association was demonstrated between hypothyroxinemia and iron deficiency. This finding suggests the importance of iron and possible repercussions of its deficiency on thyroid function and the need for intervention studies evaluating the effects of iron supplementation on thyroid status in pregnant women. Funding: this work was supported by “National Council for Scientific and Technological Development – CNPq”. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Haddow JE, Palomaki GE, Allan WC, Williams JR, Knight GJ, Gagnon J, et al. Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child. N Engl J Med. 1999;341(8):549-55. 2. Lazarus J, Brown RS, Daumerie C3 Hubalewska-Dydejczyk A, Negro R, Vaidya B. 2014 European thyroid association guidelines for the management of subclinical hypothyroidism in pregnancy and in children. Eur Thyroid J. 2014;3(2):76-94. 3. Alexander EK, Pearce EN, Brent GA, Brown RS, Chen H, Dosiou C, et al. 2017 Guidelines of the American Thyroid Association for the Diagnosis and Management of Thyroid Disease During Pregnancy and the Postpartum. Thyroid. 2017;27(3):315-89. 4. Dosiou C, Medici M. Management of Endocrine Disease: Isolated maternal hypothyroxinemia during pregnancy: knowns and unknowns. Eur J Endocrinol. 2017;176(1):R21-R38.

5. De Groot L, Abalovich M, Alexander EK, Amino N, Barbour L, Cobin RH, et al. Management of Thyroid Dysfunction during Pregnancy and Postpartum: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2012;97(8):2543-65. 6. Brenta G, Vaisman M, Sgarbi JA, Bergoglio LM, Andrada NC, Bravo PP, et al.; Task Force on Hypothyroidism of the Latin American Thyroid Society (LATS). Clinical practice guidelines for the management of hypothyroidism. Arq Bras Endocrinol Metabol. 2013;57(4):265-91.

336

7. Sgarbi JA, Teixeira PF, Maciel LM, Mazeto GM, Vaisman M, Montenegro Junior RM, et al.; Brazilian Society of Endocrinology and Metabolism. The Brazilian consensus for the clinical approach and treatment of subclinical hypothyroidism in adults: recommendations of the thyroid Department of the Brazilian Society of Endocrinology and Metabolism. Arq Bras Endocrinol Metabol. 2013;57(3):166-83. 8. Rosario PW, Carvalho M, Calsolari MR. TSH reference values in the first trimester of gestation and correlation between maternal TSH and obstetric and neonatal outcomes: a prospective Brazilian study. Arch Endocrinol Metab. 2016;60(4):314-8. 9. American College of Obstetricians and Gynecologists. Practice Bulletin No. 148: Thyroid disease in pregnancy. Obstet Gynecol. 2015;125(4):996-1005. 10. Casey BM, Thom EA, Peaceman AM, Varner MW, Sorokin Y, Hirtz DG, et al.; Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal–Fetal Medicine Units Network. Treatment of Subclinical Hypothyroidism or Hypothyroxinemia in Pregnancy. N Engl J Med. 2017;376(9):81525. 11. Cooper DS, Pearce EM. Subclinical Hypothyroidism and Hypothyroxinemia in Pregnancy – Still No Answers. N Engl J Med. 2017;376(9):876-7. 12. IMULITE® and IMMULITE® 2000 Reference Range Compendium. 13. Casey BM, Dashe JS, Spong CY, McIntire DD, Leveno KJ, Cunningham GF. Perinatal significance of isolated maternal hypothyroxinemia identified in the first half of pregnancy. Obstet Gynecol. 2007;109(5):1129-35. 14. Daru J, Allotey J, Peña-Rosas JP, Khan KS. Serum ferritin thresholds for the diagnosis of iron deficiency in pregnancy: a systematic review. Transfus Med. 2017;27(3):167-74. 15. Yu X, Shan Z, Li C, Mao J, Wang W, Xie X, et al. Iron deficiency, an independent risk factor for isolated hypothyroxinemia in pregnant and nonpregnant women of childbearing age in China. J Clin Endocrinol Metab. 2015;100(4):1594-601. 16. Zimmermann MB, Burgi H, Hurrell RF. Iron deficiency predicts poor maternal thyroid status during pregnancy. J Clin Endocrinol Metab. 2007;92(9):3436-40. 17. Cleary-Goldman J, Malone FD, Lambert-Messerlian G, Sullivan L, Canick J, Porter TF, et al. Maternal thyroid hypofunction and pregnancy outcome. Obstet Gynecol. 2008;112(1):85-92. 18. Korevaar TI, Schalekamp-Timmermans S, de Rijke YB, Visser WE, Visser W, de Muinck Keizer-Schrama SM, et al. Hypothyroxinemia and TPO-antibody positivity are risk factors for premature delivery: the generation R study. J Clin Endocrinol Metab. 2013;98(11):4382-90. 19. Medici M, Korevaar TI, Schalekamp-Timmermans S, Gaillard R, de Rijke YB, Visser WE, et al. Maternal early-pregnancy thyroid function is associated with subsequent hypertensive disorders of pregnancy: the generation R study. J Clin Endocrinol Metab. 2014;99(12):E2591-8. 20. Chan S, Boelaert K. Optimal management of hypothyroidism, hypothyroxinaemia and euthyroid TPO antibody positivity preconception and in pregnancy. Clin Endocrinol (Oxf). 2015;82(3):313-26. 21. Lazarus JH, Bestwick JP, Channon S, Paradice R, Maina A, Rees R, et al. Antenatal thyroid screening and childhood cognitive function. N Engl J Med. 2012;366:493-501.

Arch Endocrinol Metab. 2018;62/3

original article

Effectiveness and safety of carbohydrate counting in the management of adult patients with type 1 diabetes mellitus: a systematic review and meta-analysis Eliege Carolina Vaz1, Gustavo José Martiniano Porfírio2, Hélio Rubens de Carvalho Nunes3, Vania dos Santos Nunes-Nogueira1 Departamento de Clínica Médica, Faculdade de Medicina de Botucatu, Universidade Estadual de São Paulo (Unesp), Botucatu, SP, Brasil 2 Centro Cochrane do Brasil, Disciplina de Medicina de Urgência e Medicina Baseada em Evidências, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil 3 Departamento de Saúde Pública, Faculdade de Medicina de Botucatu, Universidade Estadual de São Paulo (Unesp), Botucatu, SP, Brasil 1

Objective: This study aimed to evaluate the effectiveness and safety of carbohydrate counting (CHOC) in the treatment of adult patients with type 1 diabetes mellitus (DM1). Materials and methods: We performed a systematic review of randomized studies that compared CHOC with general dietary advice in adult patients with DM1. The primary outcomes were changes in glycated hemoglobin (HbA1c), quality of life, and episodes of severe hypoglycemia. We searched the following electronic databases: Embase, PubMed, Lilacs, and the Cochrane Central Register of Controlled Trials. The quality of evidence was analyzed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE). Results: A total of 3,190 articles were identified, and two reviewers independently screened the titles and abstracts. From the 15 potentially eligible studies, five were included, and 10 were excluded because of the lack of randomization or different control/ intervention groups. Meta-analysis showed that the final HbA1c was significantly lower in the CHOC group than in the control group (mean difference, random, 95% CI: -0.49 (-0.85, -0.13), p = 0.006). The meta-analysis of severe hypoglycemia and quality of life did not show any significant differences between the groups. According to the GRADE, the quality of evidence for severe hypoglycemia, quality of life, and change in HbA1c was low, very low, and moderate, respectively. Conclusion: The meta-analysis showed evidence favoring the use of CHOC in the management of DM1. However, this benefit was limited to final HbA1c, which was significantly lower in the CHOC than in the control group. Arch Endocrinol Metab. 2018;62(3):337-45 Keywords Type 1 diabetes mellitus; carbohydrate counting; quality of life; systematic review; meta-analysis

INTRODUCTION

iabetes mellitus (DM) comprises a heterogeneous group of metabolic disorders that commonly feature hyperglycemia, which results from disturbances in insulin secretion, insulin action, or both (1). In most cases, type 1 DM (DM1) is an autoimmune disease characterized by the destruction of insulin-producing beta cells, accounting for 5% to 10% of all DM cases (1). In Brazil, eight of every 100,000 people under the age of 20 have DM1 (2). The therapeutic treatment and control of DM1 includes the use of insulin for glycemic control, balanced diet, and regular physical activity. Daily insulin requirements vary based on age, diet, patient self-monitoring of blood glucose and daily routines.

Arch Endocrinol Metab. 2018;62/3

Correspondence to: Vania dos Santos Nunes Nogueira Departamento de Clínica Médica, Faculdade de Medicina de Botucatu, Universidade Estadual de São Paulo Av. Prof. Mário Rubens Guimarães Montenegro, s/nº Unesp, Campus de Botucatu 18618-687 – Botucatu, SP, Brasil vsnunes@fmb.unesp.br

Received on Jun/5/2017 Accepted on Feb/7/2018 DOI: 10.20945/2359-3997000000045

Glycemic control of patients with DM is important because it impacts the development of diabetic complications (3). Diabetes control is evaluated mainly according to the levels of HbA1c, fasting blood glucose, and postprandial blood glucose (blood glucose measured two hours after meal consumption). Borderline normal values without the risk of hypoglycemia, impaired mental status, and patient welfare indicate good glycemic control (4). The American Diabetes Association recommends the following levels for nonpregnant adults: HbA1c < 7%, preprandial capillary plasma glucose between 80 mg/dL and 130 mg/dL, and peak postprandial capillary plasma glucose < 180 mg/dL (4). The Diabetes Control and Complications Trial (DCCT) 337

ABSTRACT

Carbohydrate counting in the type 1 diabetes

showed that adequate glycemic control in patients with DM1 (e.g., fasting blood glucose levels up to 110 mg/dL, postprandial glucose levels lower than 180 mg/dL, and HbA1c < 6.5%) delays the onset and progression of microvascular complications, such as retinopathy, nephropathy, and neuropathy, and reduces the risk of any cardiovascular event by 42% and that of nonfatal infarction, stroke, and death by 57% (3). The treatment of patients with DM1 facilitates proper development in children and adolescents and improves the quality of life (QOL) of patients in general (5). DM1 control cannot be achieved solely via regular insulin use. Combining insulin use with diet and physical activity is important. In particular, adjusting insulin therapy to an individualized food plan is key to proper metabolic control (3). Conventional nutritional advice for patients with DM1 is the same as for the general population. Specifically, a balanced nutrition with appropriate concentrations of macroand micronutrients should be based on the goals of treatment (i.e., total carbohydrate (CHO), 45%-60% of total energy intake (VET); protein, 15%-20% of VET; total fat (GT), up to 30% of VET; and minimum dietary fiber, 20 g/day or 14 g/1000 kcal) (6). In addition to conventional nutritional DM1 treatments, carbohydrate counting (CHOC) is a meal planning tool that allows for great variation in food choices among individuals with DM (7), with the main objective of providing flexibility in food intake (8). Few dietary restrictions and the option to decide the number of meals (traditional treatment plans recommend eating six meals per day) may improve acceptance of the disease and overall QOL (9). CHOC consists of measuring the amount of carbohydrates to be eaten during every meal in grams. Based on that count and preprandial blood glucose levels, the patient calculates the dose of fast or regular insulin they need before each meal (10,11). This method can be used for any patient with diabetes in combination with the use of varying doses of rapidacting insulin or continuous subcutaneous insulin infusion (12). Two CHOC methods are widely used: listing carbohydrate equivalents (A) and measuring the carbohydrate in grams (B). In method A, foods are grouped so that each food portion chosen by the patient corresponds to 15 g of carbohydrate, classifying them as equivalents. Method B consists of the sum of carbohydrate grams in each food per meal based on information in food labels and tables (13). 338

To improve glycemic control and decrease the frequency of acute and chronic complications, CHOC is now recommended as another nutritional tool (3,14). Regarding the efficacy of the CHOC method in metabolic DM1 control in the DCCT study, individuals who adjusted their pre-meal insulin doses based on carbohydrate counts had a 0.5% decrease in HbA1c compared to the group that used a fixed dose (15). Dias and cols. (16) showed that HbA1c levels were reduced in a group of 55 adult patients, and although the total daily dose of insulin increased, no weight gain was observed. Waller and cols. (17) also evaluated CHOC in children and adolescents with DM1 and reported no changes in HbA1c, body mass index (BMI), or frequency of hypoglycemic episodes. However, the children and their parents showed an improvement in QOL. We hypothesized that the CHOC method in adult individuals with DM1 may be more effective and efficient for glycemic control and better improve QOL compared to conventional nutritional guidance. This study aimed to evaluate the effectiveness and safety of CHOC in the treatment of adult patients with DMI using a systematic literature review.

MATERIALS AND METHODS This review was performed according to Cochrane Methodology (18) and reported according to the PRISMA Statement (19).

Eligibility criteria We included randomized controlled trials with at least three months of follow-up, and evaluation of outcomes in which patients were randomly divided into two groups, intervention or comparison. Data were interpreted based on patient-characteristics, intervention, comparison, and outcomes (PICO) as described below.

Patients Patients consisted of men and women aged over 18 years old who had been diagnosed with DM1 for at least six months and were not in the “honeymoon period”, in which the pancreas can produce small amounts of insulin that can be enough to achieve adequate glycemic control at a daily dose of less than 0.5 IU insulin/kg in 24 hours. Patients had standard Arch Endocrinol Metab. 2018;62/3

Carbohydrate Counting in the type 1 diabetes

Intervention Individuals in the intervention group had nutritional counseling for CHOC to determine the amount of fast or regular insulin that they would need before each main meal.

Comparison The comparison group included individuals who had conventional nutritional advice and used fixed doses of fast or regular insulin before meals.

Outcomes Assessed outcomes were reduction in HbA1c, frequency of severe hypoglycemia, improved QOL, body weight or BMI gain, lipid profile, and total daily dose of insulin. Validated questionnaires were used to evaluate QOL: Audit of Diabetes-Dependent Quality of Life (ADDQoL), Diabetes Treatment Satisfaction Questionnaire (DTSQ), and Diabetes Quality of Life Measure (DQoL).

Search strategy and selection No language restriction was imposed. We searched the following electronic databases through November 30, 2016 to identify randomized clinical trials involving CHOC versus conventional nutritional advice in the treatment of DM1 patients: Embase (1980-2016), PubMed (1966-2016), Lilacs (1982-2016), and the Cochrane Central Register of Controlled Trials (CENTRAL, the Cochrane Library, issue 2016). We also searched for ongoing clinical trials on the clinicaltrials. gov website. Medical Subject Heading terms used included “Type 1 Diabetes Mellitus”, “Carbohydrates”, “Nutrition Therapy”, and “Randomized Controlled Trial”. Two reviewers (ECV and VSNN) independently screened the titles and abstracts identified in the literature search. Studies potentially eligible for inclusion in the review were selected for complete reading. Arch Endocrinol Metab. 2018;62/3

Data extraction and risk of bias Both reviewers assessed the study quality and extracted data using an extraction template. For each trial, we assigned the risk of bias considering the quality scores for random sequence generation, allocation concealment, blinding of outcome assessment, and incomplete outcome data. We used the criteria described in the Cochrane Reviewer’s Handbook (18) to classify these scores as adequate (low risk of bias), unclear, and inadequate (high risk of bias).

Data synthesis and analysis We performed the meta-analysis by using a randomeffects model in Review Manager 5.3 software. For dichotomous outcomes, the relative risk was calculated with a 95% confidence interval and continuous variables were expressed as a weighted mean difference with 95% confidence intervals. Potential causes of heterogeneity among studies were also analyzed. The I2 statistic was used to measure the impact of heterogeneity for each outcome (where an I2 ≥ 50 indicates a considerable level of heterogeneity) (18). When we found heterogeneity, we attempted to determine possible reasons for it via subgroup analysis or by examining individual studies.

Quality assessment The quality of evidence per outcome measurement was graded according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group. The confidence of the GRADE system decreases if randomized studies have major limitations that may interfere with treatment effect estimates (20). These limitations include risk of bias for each study, inconsistency, indirectness, imprecision, and publication bias of each evaluated outcome per GRADE considerations.

RESULTS From the database searches, 3190 articles were identified (Figure 1). Fifteen articles were potentially eligible for inclusion in the analysis and were selected for full review. Five of the 15 studies were included for analysis (7,10,15,21,22). Of the 10 excluded studies, three were not randomized (23-25), three compared two different methods for mealtime insulin dosing; no group had conventional nutritional advice using a fixed dose of fast or regular insulin before meals (8,26,27). In three studies, patients were children or adolescents (28-30), 339

nutritional counseling with a professional nutritionist and took slow-acting or intermediate and multiple fast or regular insulin doses before meals (breakfast, lunch, and dinner) or continuous subcutaneous insulin infusion (CSII). Studies that included pregnant women, individuals with a BMI > 40 kg/m², kidney failure, or HbA1c >14% were excluded from analysis.

Carbohydrate counting in the type 1 diabetes

and one study compared three different possibilities of insulin self-adjustments, without a group using a fixed dose of fast or regular insulin before meals (31). The baseline characteristics of study participants and eligibility criteria of the included studies are presented in Tables 1 and 2, respectively. P values < 0.05 were considered statistically significant. Dafne and cols. (7) performed a single-center study in England. A total of 169 patients with DM1 who had been diagnosed more than two years prior without chronic complications and intensive insulin therapy were randomized to CHOC or conventional nutritional treatment. The main outcome measures after a 1 of additional records identified through other sources Conferences 0 Hand-searches 1 Specialist in the field 0

3246 of records identified through databases searching PubMed 1586 Embase 1480 Lilacs and Central Cochrane 180

3190 of records after duplicates removed

53 of records screened

15 of full-text articles assessed for eligibility

5 of studies included in quantitative synthesis (Meta-analysis)

33 of records excluded 10 of full-text articles excluded, with reasons: 03 The groups were different from proposed PICO 03 Non-randomized studies 01 Control group was different from proposed PICO 03 Patients were children or adolescents

Figure 1. Flowchart for identifying eligible studies

six-month follow-up were: HbA1c, severe hypoglycemia, and the impact of diabetes on QOL as assessed using the ADDQoL questionnaire. Laurenzi and cols. (10) recruited patients from a clinic in Milan, Italy. A total of 61 adult patients with DM1 who had been treated with CSII were randomly assigned to learn CHOC in the intervention group or to estimate pre-meal insulin doses empirically for six months. The main outcome measures were: HbA1c, fasting glucose, BMI, waist circumference, daily insulin dose, hypoglycemic events, and analysis of QOL through the Diabetes-Specific Quality-of-life Scale, which evaluates individual treatment goals in patients with DM1. In the study of Scavone and cols. (Italy) (21), 256 patients with DM1 who had been diagnosed for more than five years were randomized to a CHOC group or a control group. Weight, BMI, HbA1c, lipid profile, uric acid, creatinine, microalbuminuria, daily insulin requirements, and number of episodes of hypoglycemia (blood glucose < 70 mg/dL) were the main outcomes evaluated. Schmidt and cols. (15) recruited patients from two centers in Denmark. The authors randomized 63 adults with DM1 and poor metabolic control (HbA1c: 8.0% to 10.5%) to the CHOC or control groups for more than 12 months using analogues of basal and fast insulin. The main outcome measures were: change in HbA1c, weight, satisfaction with the treatment of diabetes, and perceived frequency of hyper- and hypoglycemia. The parameters were measured according to the Diabetes Treatment Satisfaction status version and version change questionnaires (DTSQs and DTSQc, respectively). QOL was analyzed using the ADDQoL questionnaire.

Table 1. Baseline characteristics of patients in each included study Study

Dafne, 2002

Number of randomized patients

Male/Female

Fasting Glucose nmol/L (SD)

BMI or weight (kg) (SD)

G1- 9.4 (1.2) G2 = 9.3 (1.1)

G1 = 80.5 (1.7) G2 = 77.4 (13.4)

Age (SD)

HbA1C (SD)

Insulin dose (SD)

G1 = 84 G2 = 85

Laurenzi, 2011

G1 = 28 G2 = 28

G1 = 15/13 G2 = 09/19

G1 = 41.2 (10.0) G2 = 39.8 (9.8)

G1 = 7.9 (0.9) G2 = 8.1 (1.5)

G1 = 23.7 (21-25.2) G2 = 23.8 (20.8-26.8)

**G1 = 36 (24.5-49) **G2 = 33 (28.5-39.5)

Scavone, 2010

G1 = 100 G 2 = 156

G1 = 49/51 G2 = 74/82

G1 = 39 (11) G2 = 39 (11)

G1 = 7.8 (1.3) G2 = 7.5 (0.8)

Schmidt, 2012

G1 = 26 G2 = 09

G1 = 10/11 G2 = 06/02

G1= 41 (10) G2 = 46 (09)

G1 = 9.2 (0.6) G2 = 9.1 (0.7)

*G1 = 0.6 (0.2) *G2 = 0.7 (0.17)

Trento, 2009

G1 = 27 G2 = 29

G1= 18/9 G2 = 12/17

G1 = 37.33 (12.6) G2 = 36.76 (7.9)

G1 = 7.6 (1.3) G2 = 7.7 (1.24)

G1 = 9.64 (5.17) G2 = 9.05 (5.08)

G1 = 24.4 (2.6) G2 = 23.5 (3.3)

**G1 = 47.9 (10.6) **G2 = 45.7 (12.6)

G1: intervention group; G2: control group. * Daily insulin dose per kg; ** Total insulin dose (basal and bolus). - No information provided.

340

Arch Endocrinol Metab. 2018;62/3

Carbohydrate Counting in the type 1 diabetes

Table 2. Length of follow-up, inclusion criteria, and outcomes of included studies Follow-up

Inclusion criteria

DAFNE, 2002

6 months

> 18 years of HbA1c from 7.5% to 12% and diagnosis greater than 2 years without advanced complications

Change in HbA1C (HPLC), severe hypoglycemia and hyperglycemia, quality of life (ADDQoL, DTSQ, and W-BQ12), weight, blood pressure, lipid profile, injections, glucose monitoring, and daily total dose of insulin

Laurenzi, 2011

3 and 6 months

Age between 18 and 65 years and treatment with continuous insulin infusion pump for more than 3 months

Change in HbA1C (HPLC), hypoglycemia, quality of life (DSQOLS), BMI, waist, fasting glucose, and daily insulin dose

Scavone, 2010

9 months

Diagnosis of type 1 diabetes mellitus over 5 years

Changes in HbA1c, hypoglycemia, daily insulin dose, weight, lipid profile, creatinine, and microalbuminuria

Schmidt, 2012

4 months

Age between 18 and 65 years, poor metabolic control, diabetes duration over 12 months, and use of basal and fast analogue insulin.

Changes in HbA1C, severe hypoglycemia, treatment satisfaction and perceived frequency of hypo- and hyperglycemia (DTSQs and DTSQc), quality of life (ADDQoL), change in the perception of problem areas (PAID), and change in fear of hypoglycemia (HFS)

Trento, 2009

30 months

Age <70 years, onset of diabetes before 30 years of age, and onset of insulin use within the first year of the diagnosis

Changes in HbA1C (HPLC), severe hypoglycemia and hyperglycemia, quality of life (DQOL, GISED, CSI), BMI, and lipid profile and fasting glucose

Trento and cols. (22) included 56 patients with DM1 who had all been diagnosed before age 30 years. Twenty-seven subjects were randomized to a CHOC program and the remaining patients were assigned to the control group. Body weight, fasting glucose, HbA1c, total cholesterol, high-density lipoprotein cholesterol, triglycerides and creatinine, frequency of hypoglycemia, and QOL were the main outcome measures.

Risk of bias Dafne and cols. (7) and Laurenzi and cols. (10) randomized patients using a computer-generated random number. Schmidt and cols. (15) performed the random distribution with a 1: 3: 3 ratio in blocks of 14 with sealed, opaque envelopes containing group assignments. Scavone and cols. (21) and Trento and cols. (22) did not describe how the randomization sequence was generated. Only Dafne, Laurenzi, and Schmidt described allocation concealment and as such were classified as low risk of bias. The other two studies did not provide any information regarding the allocation process. Most of the studies included did not report blinding for outcome evaluation. However, except for severe hypoglycemia, most were laboratory assessments, which were not susceptible to bias. QOL questionnaires were self-applied and could not be blinded. Only Laurenzi and cols. reported that patients who did not complete the treatment regimen were included in the final analysis (low risk) (10). Trento and cols. (22) reported that all participants completed the Arch Endocrinol Metab. 2018;62/3

Outcomes

treatment (low risk). In the study of Dafne and cols. (7), 28 patients were lost to follow-up and were not included in the final analysis, although the number of patients who were lost was not significantly different between the groups (15 in the intervention group and 13 in the control group) (low risk). Scavone and cols. (21) had a 27% loss of patients in the intervention group, and they were not included in the final analysis (high risk). Schmidt and cols. (15) had a 19% loss, and these patients were not included in the final analysis (high risk).

Meta-analysis of outcomes The five studies included analyzed changes in HbA1c levels at the end of the study. Meta-analysis showed that the final HbA1c was significantly lower in the CHOC group than in the control group (mean difference, random, 95% CI: -0.49 (-0.85, -0.13), p = 0.006, I2 = 72%) (Figure 2). In the four trials, the number of patients who experienced at least one episode of severe hypoglycemia can be assessed (7,15,21,22). The meta-analysis of this outcome was not significantly different between groups (risk ratio, random, 95% CI: 0.94 (0.55, 1.6), p = 0.82, I2 0%) (Figure 3). Regarding QOL, two studies (7,15) used the ADDQoL instrument, but no difference was noted between groups (mean difference, random, 95% CI: -0.23 (-1.4, 0.94), p = 0.7, I2 = 84%). The same studies also used the DTSQs questionnaire and found no difference between groups (mean difference, random, 95% CI: 3.53 (-7.11, 14.16), p = 0.52, I2 = 95%). 341

Study

Carbohydrate counting in the type 1 diabetes

Figure 2. Meta-analysis of change in HbA1c.

Figure 3. Meta-analysis of episodes of severe hypoglycemia.

We plotted the QOL outcomes using different questionnaires (DQOL and DTSQs) cited in three studies (7,15,22,23) and found no significant differences between groups (std. mean difference, random, 95% CI: 0.64 (-0.7, 1.98), p = 0.35, I2 = 94%). Meta-analysis of total cholesterol, HDL-C, and triglycerides could only be performed based on the results reported in the study of Dafne and cols. (7) and Trento and cols. (22); no significant differences were noted among groups. According to the GRADE, the quality of evidence of the primary outcomes was moderate for changes in HbA1c, low for episodes of severe hypoglycemia, and very low for QOL (Table 3).

DISCUSSION Most individuals with DM1 have a hard time managing fasting and postprandial blood glucose levels. In addition, many patients with this disease have poor compliance to dietary advice. Poor disease control can increase the risk of complications, such as retinopathy and other microvascular conditions (3). Ahola and cols. (32) reported that only one-third of patients maintained controlled blood glucose levels after a meal and that approximately 40% experienced frequent hyperglycemia despite having seemingly normal metabolic control. As such, the search for tools to improve these health issues

Table 3. Quality of evidence of primary outcomes according to GRADE approach Outcomes

Risk of bias

Inconsistency

Indirectness

Imprecision

Publication bias

Intervention vs comparator 95% CI

Participants (studies)

Quality of evidence

Severe hypoglycemia

Serious* (-1)

Serious (-1)***

Unlikely

RR 0.92 (0.54 a 1.56)

453 (4)

+Low

Quality of Life (ADDQoL)

Serious* (-1)

Serious (-1)**

Serious (-1)****

Unlikely

MD 3.53 (-7,11 a 14.16)

168 (2)

+++Very low

Change in HbA1c

Serious* (-1)

Unlikely

MD -0.45 (-0.77, -0.13)

535 (5)

+ Moderate

* Most of the included studies did not report about allocation concealment, and they did not perform an intention-to-treat analysis. ** Presence of statistical heterogeneity (I > 75%). *** 95% CI overlaps no effect but includes important benefit or important harm. **** Optimal information size criterion was not meet. ADDQOL: Audit of Diabetes â&#x20AC;&#x201C; Dependent Quality of Life. RR: Relative risk. MD: Mean difference. ++ Low evidence: The authors are not confident in the effect estimate, and the true value may be substantially different from it. ++ Very low evidence: The authors do not have any confidence in the estimate, and it is likely that the true value is substantially different from it. + Moderate evidence: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. 2

342

Arch Endocrinol Metab. 2018;62/3

has increased, and CHOC may be the only effective option for adherence to dietary requirement in patients with variable dietary habits. To reduce postprandial blood glucose, protocols from the DAPHNE program and the Diabetes Teaching and Treatment have used the CHOC method for nutritional counseling (33). Some studies reported that CHOC can provide better glycemic control and lead to an improved QOL for patients (3,17). Patients using this method have greater flexibility in food choices without the concern of postprandial hyperglycemia given that the amount of carbohydrates ingested is considered when computing the amount of insulin to be administered before meals. In daily clinical practice, the goal is to maintain good long-term disease control, prevent chronic complications from DM, and reduce the frequency of hypoglycemia to improve overall QOL. We performed a systematic review focusing on the efficacy and safety of the CHOC method in the management of patients with DM1. We included randomized trials that compared the CHOC method with conventional nutritional guidance in the treatment of patients with DM1. Five studies met the established inclusion criteria and were included in qualitative and quantitative analyses. Most of the studies assessed changes in HbA1c, frequency of hypoglycemia, and QOL as primary endpoints. Meta-analysis showed a significant difference in final HbA1c favoring the intervention group. A criticism of HbA1c is that even though levels are associated with the frequency of chronic complications and rate of morbidity and mortality, the value of this laboratory outcome is often discussed without considering glycemic variability. Although it is important for HbA1c levels to be lower than the cutoff values that indicate disease control, blood glucose levels can range from high to low. An association between glycemic variability and development of micro diabetes-related complications has been shown in type 2 DM and has also been studied as a possibility in DM1 (34). If confirmed, HbA1c values in DM1 would be inadequate to determine the superiority of one treatment to another. However, in this present review, the frequency of hypoglycemia was the same between the groups, which means that the relevance of lowering HbA1c would not be reduced. Regarding QOL outcomes, several different instruments were used in the studies included, which Arch Endocrinol Metab. 2018;62/3

negatively affected the single meta-analysis of this parameter. However, independent of the instrument used, an improvement in QOL from baseline compared with the final visit in most of the studies was noted, although no difference was observed between the intervention and control groups. Improvement in QOL can be more associated with follow-up programs and nutritional guidance than the initial methods evaluated. Applying the GRADE approach for the outcomes “change in HbA1c” and “severe hypoglycemia,” it was necessary to rate down for the risk of bias because five out of the six studies lost patients to follow-up without an intention to treat analysis. In addition, concealment allocation and randomization processes were unclear in two of the studies. Similarly, imprecision was rated down for both, because optimal information size criterion was not met and 95% CI overlaps no effect but includes important benefit or important harm, respectively. Rating down for indirectness and publication bias was unnecessary. The quality of evidence for “change in HbA1c” and “severe hypoglycemia” was moderate and low, respectively, indicating that further research is likely to have an important impact on our confidence and authors are not confident in the effect estimate. The quality of the evidence regarding QOL outcomes was very low, and any estimate of its effect is uncertain. Bell and cols. (35) recently published a similar systematic review. They included a study that was not included in our analysis because the comparison groups were different from the proposed PICO (31) They also included another study that we excluded because patients in the control group were predominantly children, and they were provided nutritional guidance of low glycemic index (36). Finally, Bell and cols. did not use the GRADE. The results of the previous study favored the intervention group, and the authors interpreted the results in support of recommending CHOC instead of general dietary advice in patients with DM1. Considering the studies included in the present systematic review, the meta-analysis showed evidence favoring the use of CHOC in the management of adult patients with DM1. However, this benefit was limited to final HbA1c, which was significantly lower in the CHOC group than in the control group. Therefore, new randomized trials with greater internal and external validation and long-term outcomes are needed to analyze whether or not a significant difference exists between these two nutritional guidance tools in terms 343

Carbohydrate Counting in the type 1 diabetes

Carbohydrate counting in the type 1 diabetes

of other important diabetes-related outcomes, such as mortality, QOL, and diabetes complications. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2014;37 Suppl 1:S81-90. 2. DIAMOND Project Group. Incidence and trends of childhood type 1 diabetes worldwide 1990-1999. Diabet Med. 2006;23(8):857-66. 3.

Diabetes Control and Complications Trial Research Group, Nathan DM, Genuth S, Lachin J, Cleary P, Crofford O, Davis M, et al. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329(14):977-86.

4. American Diabetes Association. 6. Glycemic Targets. Diabetes Care. 2017;40(Suppl 1):S48-S56. 5. Diet, nutrition and the prevention of chronic diseases. World Health Organ Tech Rep Ser. 2003;916:i-viii, 1-149, backcover. 6. Evert AB, Boucher JL, Cypress M, Dunbar SA, Franz MJ, MayerDavis EJ, et al. Nutrition therapy recommendations for the management of adults with diabetes. Diabetes Care. 2014;37 Suppl 1:S120-43. 7. DAFNE Study Group. Training in flexible, intensive insulin management to enable dietary freedom in people with type 1 diabetes: dose adjustment for normal eating (DAFNE) randomised controlled trial. BMJ. 2002;325(7367):746. 8. Rabasa-Lhoret R, Garon J, Langelier H, Poisson D, Chiasson JL. Effects of meal carbohydrate content on insulin requirements in type 1 diabetic patients treated intensively with the basalbolus (ultralente-regular) insulin regimen. Diabetes Care. 1999;22(5):667-73. 9. Rossi MC, Nicolucci A, Di Bartolo P, Bruttomesso D, Girelli A, Ampudia FJ, et al. Diabetes Interactive Diary: a new telemedicine system enabling flexible diet and insulin therapy while improving quality of life: an open-label, international, multicenter, randomized study. Diabetes Care. 2010;33(1):109-15. 10. Laurenzi A, Bolla AM, Panigoni G, Doria V, Uccellatore A, Peretti E, et al. Effects of carbohydrate counting on glucose control and quality of life over 24 weeks in adult patients with type 1 diabetes on continuous subcutaneous insulin infusion: a randomized, prospective clinical trial (GIOCAR). Diabetes Care. 2011;34(4):823-7. 11. Hegar K, Heiber S, Brandle M, Christ E, Keller U. Carbohydrate counting of food. Swiss Med Wkly. 2011;141:w13224. 12. Bell KJ, King BR, Shafat A, Smart CE. The relationship between carbohydrate and the mealtime insulin dose in type 1 diabetes. J Diabetes Complications. 2015;29(8):1323-9.

13. Nutricionistas Membros do Departamento de Nutrição da SBD. Manual de contagem de carboidratos para pessoas com diabetes. Sociedade Brasileira de Diabetes, 2016.

17. Waller H, Eiser C, Knowles J, Rogers N, Wharmby S, Heller S, et al. Pilot study of a novel educational programme for 11-16 year olds with type 1 diabetes mellitus: the KICk-OFF course. Arch Dis Child. 2008;93(11):927-31. 18. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011] [Internet]. 2011. 19. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med. 2009;151(4):W65-94. 20. Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ; GRADE Working Group. What is “quality of evidence” and why is it important to clinicians? BMJ. 2008;336(7651):995-8. 21. Scavone G, Manto A, Pitocco D, Gagliardi L, Caputo S, Mancini L, et al. Effect of carbohydrate counting and medical nutritional therapy on glycaemic control in Type 1 diabetic subjects: a pilot study. Diabet Med. 2010;27(4):477-9. 22. Trento M, Trinetta A, Kucich C, Grassi G, Passera P, Gennari S, et al. Carbohydrate counting improves coping ability and metabolic control in patients with Type 1 diabetes managed by Group Care. J Endocrinol Invest. 2011;34(2):101-5. 23. Bao J, Gilbertson HR, Gray R, Munns D, Howard G, Petocz P, et al. Improving the estimation of mealtime insulin dose in adults with type 1 diabetes: the Normal Insulin Demand for Dose Adjustment (NIDDA) study. Diabetes Care. 2011;34(10):2146-51. 24. Chiesa G, Piscopo MA, Rigamonti A, Azzinari A, Bettini S, Bonfanti R, et al. Insulin therapy and carbohydrate counting. Acta Biomed. 2005;76 Suppl 3:44-8. 25. Dubé MC, Lavoie C, Galibois I, Weisnagel SJ. Nutritional strategies to prevent hypoglycemia at exercise in diabetic adolescents. Med Sci Sports Exerc. 2012;44(8):1427-32. 26. Rossetti P, Ampudia-Blasco FJ, Laguna A, Revert A, Vehì J, Ascaso JF, et al. Evaluation of a novel continuous glucose monitoringbased method for mealtime insulin dosing – the iBolus – in subjects with type 1 diabetes using continuous subcutaneous insulin infusion therapy: a randomized controlled trial. Diabetes Technol Ther. 2012;14(11):1043-52. 27. Gilbertson HR, Brand-Miller JC, Thorburn AW, Evans S, Chondros P, Werther GA. The effect of flexible low glycemic index dietary advice versus measured carbohydrate exchange diets on glycemic control in children with type 1 diabetes. Diabetes Care. 2001;24(7):1137-43. 28. Gökşen D, Atik Altınok Y, Ozen S, Demir G, Darcan S. Effects of carbohydrate counting method on metabolic control in children with type 1 diabetes mellitus. J Clin Res Pediatr Endocrinol. 2014;6(2):74-8. 29. Albuquerque IZ, Stringhini SMF, Marques RMB, Mundim CA, Rodrigues MLD, Campos MRH. Carbohydrate counting, nutritional status and metabolic profile of adolescents with type 1 diabetes mellitus. Sci Med. 2014;24(4).

14. Diretrizes da Sociedade Brasileira de Diabetes (2015-2016). Rio de Janeiro: AC Farmacêutica Ltda.; 2016. Princípios para Orientação Nutricional no Diabetes Mellitus; p. 91-110.

30. Enander R, Gundevall C, Strömgren A, Chaplin J, Hanas R. Carbohydrate counting with a bolus calculator improves postprandial blood glucose levels in children and adolescents with type 1 diabetes using insulin pumps. Pediatr Diabetes. 2012;13(7):545-51.

15. Schmidt S, Meldgaard M, Serifovski N, Storm C, Christensen TM, Gade-Rasmussen B, et al. Use of an automated bolus calculator in MDI-treated type 1 diabetes: the BolusCal Study, a randomized controlled pilot study. Diabetes Care. 2012;35(5):984-90.

31. Kalergis M, Pacaud D, Strychar I, Meltzer S, Jones PJ, Yale JF. Optimizing insulin delivery: assessment of three strategies in intensive diabetes management. Diabetes Obes Metab. 2000;2(5):299-305.

16. Dias VM, Pandini JA, Nunes RR, Sperandei SL, Portella ES, Cobas RA, et al. Effect of the carbohydrate counting method on glycemic control in patients with type 1 diabetes. Diabetol Metab Syndr. 2010;2:54.

32. Ahola AJ, Mäkimattila S, Saraheimo M, Mikkilä V, Forsblom C, Freese R, Groop PH; FinnDIANE Study Group. Many patients with type 1 diabetes estimate their prandial insulin need inappropriately. J Diabetes. 2010;2(3):194-202.

344

Arch Endocrinol Metab. 2018;62/3

Carbohydrate Counting in the type 1 diabetes

33. Speight J, Amiel SA, Bradley C, Heller S, Oliver L, Roberts S, et al. Long-term biomedical and psychosocial outcomes following DAFNE (Dose Adjustment For Normal Eating) structured education to promote intensive insulin therapy in adults with sub-optimally controlled type 1 diabetes. Diabetes Res Clin Pract. 2010;89(1):22-9.

36. Gilbertson HR, Thorburn AW, Brand-Miller JC, Chondros P, Werther GA. Effect of low-glycemic-index dietary advice on dietary quality and food choice in children with type 1 diabetes. Am J Clin Nutr. 2003;77(1):83-90.

34. American Diabetes Association. Executive summary: standards of medical care in diabetes â&#x20AC;&#x201C; 2011. Diabetes Care. 2011;34 Suppl 1:S4-10.

35. Bell KJ, Barclay AW, Petocz P, Colagiuri S, Brand-Miller JC. Efficacy of carbohydrate counting in type 1 diabetes: a systematic review and meta-analysis. Lancet Diabetes Endocrinol. 2014;2(2):133-40.

Arch Endocrinol Metab. 2018;62/3

345

original article

Long term maintenance of glucose and lipid concentrations after Roux-en-Y gastric bypass Fernanda Cristina Carvalho Mattos Magno1, Priscila Alves Medeiros de Sousa1, Marcelo Paiva Rodrigues2, Lícia Lopes Pio Pereira2, José Egídio Paulo de Oliveira1, Eliane Lopes Rosado1, João Régis Ivar Carneiro1 1 Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brasil 2 Instituto Fernando Luiz Barroso, Rio de Janeiro, RJ, Brasil

Correspondence to: alves_priscila@hotmail.com Received on Jul/29/2017 Accepted on Feb/17/2018 DOI: 10.20945/2359-3997000000047

ABSTRACT Objective: Roux-en-Y gastric bypass (RYGB) reduces body weight and the comorbidities associated with obesity. The aim of this study was to evaluate whether glucose and lipid profiles were maintained during a 5-year follow-up period after RYGB. Subjects and methods: Anthropometric and laboratory data from 323 patients who had undergone this operation were analyzed. Differences in laboratory variables between the baseline and 12, 24, 36, 48 and 60 months postoperatively (PO) were assessed using a one-way ANOVA test to compare the three groups. Delta significance using one-way ANOVA was performed to assess anthropometric variable in the postoperative period (p < 0.05). Results: 77 patients (24%) were included in Group 1 (G1), 101 (32%) in Group 2 (G2), and 141 (44%) in Group 3 (G3). The majority of patients, 71.7% in G1, 82.8% in G2, and 70% in G3, showed high triglycerides (TG) before surgery. A decrease in weight loss was observed in all groups followed by an increase in body weight in G2 and G3 at 36, 48 and 60 months. Laboratory results for G1, G2 and G3 showed no significant differences between groups at baseline and during the post-operative period. Conclusion: Our results suggest that weight regain after RYGB has no significant impact on the long-term evolution of the lipid profile and glycemia. Arch Endocrinol Metab. 2018;62(3):346-51 Keywords Bariatric surgery; weight regain; lipemia; glycemia

INTRODUCTION

besity is considered a risk factor for the development of various comorbidities, including impaired glucose homeostasis and cardiovascular disease (1). Roux-en-Y gastric bypass (RYGB) is the most commonly performed surgical treatment for obesity; however, although it is well documented that RYGB results in weight loss and metabolic improvements, some degree of weight regain has been observed 2–3 years after surgery (2), supporting the hypothesis that obesity is a chronic, incurable and progressive disease (3). RYGB includes the anatomical diversion of the upper gastrointestinal tract. These changes in intestinal anatomy are thought to alter the function of the nutrientdriven enteroendocrine system, causing beneficial effects on glucose homeostasis that are independent of weight loss itself (4). In addition, weight loss elicits metabolic and cardiovascular benefits after RYGB, but evidence suggests that surgery might also activate 346

weight-independent lipid amelioration mechanisms (1). Garcia-Marirrodriga and cols. (5) reported significant associations between lipid subfractions and excess weight loss after RYGB. However, a recent study conducted by Bradley and cols. (6) does not support the notion of clinically meaningful, weight lossindependent effects of RYGB surgery on β cell function and insulin sensitivity in non diabetic obese subjects after they have experienced considerable weight loss. Weight loss results in improvements in glycemia and lipemia, and this is most evident in patients who have undergone RYGB (7). However, an elegant review by Shah and cols. (8) showed that the initial improvements seen following surgery decreased over a long-term follow-up period, possibly because of weight regain or unsatisfactory weight loss. Since weight gain can be considered a risk factor for dyslipidemia and changes in glucose levels, the aim of this study was to evaluate whether glucose and lipid profiles were maintained during a 5-year follow-up period after RYGB. Arch Endocrinol Metab. 2018;62/3

Glucose and lipid concentrations after Roux-em-Y gastric bypass

Subjects This retrospective study was conducted by examining the medical records of 323 patients who underwent RYGB in a private clinic in Rio de Janeiro from 2000 to 2007 and maintained regular postoperative followup visits for at least 24 months after surgery. Personal, anthropometric and laboratory data were obtained from the medical records. This study was approved by the Research and Ethics Committee of the Clementino Fraga Filho University Hospital (176/09). Three groups were classified according to the rate of inclination of each individual weight over time using linear regression slope for subjects with at least 36 months follow-up. The slope was obtained from the first 24 months after surgery, when patients’ body weight is expected to stabilize. Negative slopes were considered G1 (patients without weight regain). Positive slopes (variation between 0 and 2 kg) were considered G2 (patients who might or might not regain weight). Positive slopes (over to 2 kg) were considered G3 (patients with weight regain).

Anthropometric and laboratory assessment Anthropometric variables which were assessed included height and weight measured during the preoperative period and at 12, 24, 36, 48 and 60 months PO. Body mass index (BMI) was calculated by dividing the body weight in kilograms (kg) by the square of the height in meters (m2) (9). Total body mass and stature were measured following Gibson, using a mechanical anthropometric balance (Welmy®) with a maximum capacity of 300 kg, divided by 100 g, and the stadiometer from the anthropometric balance with a scale of 0.1 cm (10). The patients were barefoot and wore minimal clothing when weighed. Laboratory parameters assessed included total cholesterol (TC), low-density lipoprotein cholesterol (LDLc), high-density lipoprotein cholesterol (HDLc), triglycerides (TG) and glucose. These were measured during the preoperative period and at 12, 24, 36, 48 and 60 months PO. TC, HDLc, TG and glucose were determined by the enzymatic-colorimetric method (11). Concentrations of LDLc were calculated based on Friedwald’s equation (12). Concentrations of serum lipids that were considered normal were those provided in the guidelines of the National Cholesterol Education Program – Adult Treatment Panel III (NCEP/ATPIII): CT < 200 mg/dL, TG < Arch Endocrinol Metab. 2018;62/3

150 mg/dL, HDLc > 40 mg/dL e LDL-c < 130 mg/dL (13). A concentration of < 100 mg/dL was considered normal for fasting blood glucose (14).

Statistical analysis Means and standard deviations (SD) were determined for quantitative data. All data were assessed for normality using the Kolmogorov-Smirnov test. The Chi-square test was used for the comparison of proportions of abnormal values in TC, LDLc, HDLc, TG and glucose to baseline. Differences in laboratory variables between the baseline and at 12, 24, 36, 48 and 60 months PO were assessed using a one-way ANOVA test to compare the groups. Delta significance using one-way ANOVA was used to assess anthropometric variables PO. Statistical significance was defined as two-tailed p < 0.05. Statistical analysis was performed using SPSS (Statistical Package for the Social Sciences) software, version 21.0 for Windows.

RESULTS Of the 323 patients included in this study, 77 were included in G1 (24%), 101 in G2 (32%), and 141 in G3 (44%). Demographic data and the proportion of patients with abnormal laboratory values at baseline are shown in Table 1. Table 2 showed no difference in the proportion of groups with abnormal values for the lipid subtractions and glucose concentration at baseline. Most patients from the 3 groups showed high TG before surgery. Figure 1 shows weight loss for G1, G2 and G3 during the postoperative period. Comparing these groups, a decrease in weight loss was observed in all groups followed by an increase in body weight for G2 and G3 at 36, 48, and 60 months (p < 0.01). Table 1. Baseline characteristics of the patients who received RYGB over 5 years of follow-up Number of patients (n = 323) Female

72% (233/323)

Male

29% (90/323)

Age (years)

38 ± 10

Height (m)

1,65 ± 0,86

Baseline Weight (kg) Baseline BMI (Kg/m ) 2

123,91 ± 21,9 45,4 ± 6,1

TC > 200 mg/dL

40% (86/217)

LDLc > 130 mg/dL

31% (64/206)

BMI: body mass index, TC: total cholesterol; LDLc: Low-density lipoprotein cholesterol; HDLc: high density lipoprotein cholesterol; TG: triglycerides.

347

SUBJECTS AND METHODS

Glucose and lipid concentrations after Roux-em-Y gastric bypass

Table 2. Laboratorial profile with abnormal levels at baseline Laboratorial data

p value

Percentage

TC > 200 mg/dL

36.5%

43.1%

39%

0.76

LDLc > 130 mg/dL

32%

32.8%

31.1%

0.90

HDLc < 40 mg/dL

32%

34.4%

30.9%

0.90

TG > 100 mg/dL

71.7%

82.8%

70%

0.17

Glucose > 100 mg/dL

28.8%

48.6%

41.7%

0.08

TC: total cholesterol; LDLc: low-density lipoprotein cholesterol; HDLc: high-density lipoprotein cholesterol; TG: triglycerides.

Figure 1. Weight loss (mean) in kilograms per year in patients who underwent RYGB. M: months; kg: kilograms; G1: group 1; G2: group 2; G3: group 3.

Laboratory results for G1, G2 and G3 are shown in Table 3. No differences between groups at baseline and during the post-operative period were found.

DISCUSSION This study examines the impact of weight regain on plasma glucose concentrations and lipid profiles in obese subjects during a 5-year follow-up period after marked RYGB-induced weight loss. The three groups of patients were comparable in terms of age, gender, anthropometric characteristics and laboratory results at baseline. After the second or third year PO, long-term studies have shown a tendency for a weight increase in gastric bypass patients (15-20). In this study, weight loss did not differ significantly over the first 24 months after surgery. Weight reduction can improve the lipid profile, alter carbohydrate metabolism, reduce morbidity and mortality, and generally enhance the quality of life of 348

morbidly obese patients (21). Upper gastrointestinal tract diversion results in weight loss, but also may have weight loss-independent metabolic effects. Studies have reported contradictory findings regarding weight-loss dependent and independent effects of gastric bypass on glucose homeostasis and lipid profiles (1,8). The RYGB technique may itself improve total cholesterol and LDLc levels (22). An observational study in obese patients confirmed that BMI is correlated with levels of TG and HDLc, but not with LDLc levels (23). In our study, glucose and lipid profiles were lower in all groups after surgery and this was not related to weight. Further studies analyzing laboratory parameters and weight dynamics after RYGB might contribute to clarifying whether the metabolic improvements seen after RYGB rely on weight loss or are an effect of the surgical technique itself. The proportion of patients with abnormal total cholesterol concentrations prior to surgery was high in all groups but decreased during the long-term outcomes. Brolin and cols. (24) found that the greatest Arch Endocrinol Metab. 2018;62/3

Glucose and lipid concentrations after Roux-em-Y gastric bypass

Table 3. Lipid profiles and glucose concentrations (means and standard deviation) of patients who underwent RYGB Variable TC

LDL

HDL

Glucose

Groups

Baseline

12 mPO

24 mPO

36 mPO

48 mPO

60 mPO

Mean/SD

196±43

164 ± 33

170 ± 30

169 ± 43

164 ± 31

171 ± 32

200 ± 46

162 ± 28

170 ± 32

170 ± 34

170 ± 35

179 ± 32

100

200 ± 47

127

170 ± 34

107

170 ± 32

176 ± 36

180 ± 34

174 ± 41

p value

0.85

0.18

0.99

0,48

0.12

0.77

120 ± 35

93 ± 24

93 ± 22

84 ± 27

89 ± 28

89 ± 17

124 ± 43

90 ± 23

90 ± 27

90 ± 23

91 ± 27

89 ± 22

119 ± 36

125

95 ± 28

105

93 ± 24

95 ± 28

97 ± 30

p value

0.70

0.45

0.66

0.07

0.38

88 ± 24 0.98

46 ± 10

56 ± 17

57 ± 14

65 ± 20

64 ± 15

67 ± 17

47 ± 15

56 ± 12

63 ± 17

64 ± 16

68 ± 23

46 ± 11

123

55 ± 11

106

62 ± 15

67 ± 18

65 ± 20

61 ± 25

p value

0.90

0.94

0.06

0.68

0.85

0.35

147 ± 70

87 ± 34

85 ± 37

83 ± 46

84 ± 35

89 ± 46

160 ± 93

78 ± 31

77 ± 29

78 ± 35

83 ± 41

84 ± 34

100

143 ± 63

125

86 ± 35

109

78 ± 32

78 ± 29

82 ± 34

86 ± 24

p value

0.38

0.18

0.29

0.70

0.95

0.89

99 ± 20

83 ± 12

85 ± 8

83 ± 8

85 ± 8

89 ± 13

110 ± 36

85 ± 9

84 ± 8

87 ± 12

89 ± 10

103

104 ± 29

127

83 ± 12

106

84 ± 6

89 ± 30

90 ± 32

86 ± 8

p value

0.12

0.56

0.51

0.43

0.65

0.35

postoperative reductions in cholesterol and triglycerides generally occur in patients with the highest preoperative elevations, which is in accordance with our results for TG. Previous studies have shown a decline in total cholesterol and TG concentrations, and an elevation in HDLc concentrations, during the PO period. These changes were sustained over a long period of time in those patients who kept lost weight off as well as in those who experienced weight regain over time. Jamal and cols. (25) reported that the amelioration of the overall lipid profile was sustained over a 6-year PO follow-up. Laguna and cols also showed that TG and LDL-C levels decreased 30% with respect to preoperative levels, while HDL-C increased 97% over initial values. Both TC:HDL-C and TG:HDL-C ratios normalized after RYGB and values were sustained over the weight regain period until the end of the study (26). Bradley and cols. (6) support the idea that weight loss itself is primarily responsible for the clinical effects of RYGB on insulin sensitivity, β cell function, and oral glucose tolerance in non diabetic obese adults. Arch Endocrinol Metab. 2018;62/3

On the other hand, Reed and cols. (27) showed improved fasting insulin and glucose levels a week after RYGB, despite persistent insulin resistance. Peterli and cols. (28) also reported early and augmented insulin responses as early as a week PO, potentially mediating improved early glycemic control. In our study, improved blood glucose levels were maintained throughout the postoperative period, regardless of weight regain. This may be associated with the enteroendocrine effects of RYGB, and suggests that glucose homeostasis after RYGB is weight-independent. Postoperative reduction in the levels of gastric inhibitory protein (GIP), and enhanced production of the antidiabetic hormone glucagon-like peptide (GLP-1), may underlie the changes in glucose seen after surgery (29). In addition, patients who participated in a prospective human study of gastric bypass had increased intestinal peptide YY (PYY) and GLP-1 responses, which were sustained even more than a year PO (30). Numerous other factors have been implicated as potential contributors to the metabolic improvement observed after bariatric surgery, including other 349

TC: total cholesterol; LDLc: low-density lipoprotein cholesterol; HDLc: high-density lipoprotein cholesterol; TG: triglycerides; 12 mPO: 12 months post-operative; 24 mPO: 24 months post-operative; 36 mPO: 36 months post-operative; 48 mPO: 48 months post-operative; 60 mPO: 60 months post-operative.

Glucose and lipid concentrations after Roux-em-Y gastric bypass

intestinal gut hormones (GLP-2, PYY), ghrelin (an anorexic hormone secreted by the gastric fundus), adipokines, the increased energy expenditure following surgery, changes in the gut microbiome, and bile acid metabolism (31). Weight regain after gastric bypass has been attributed to multiple factors, including dilation of the gastric reservoir and gastrojejunostomy site, increased calorie intake of high glycemic index carbohydrates and fats, intolerance to red meat, hormonal alterations resulting from the adaptive process, binge eating, alcohol and drug consumption (32,33), difficulty promoting behavior changes, and lack of physical exercise. However, low adherence to follow-up with the multidisciplinary team PO is notoriously prevalent and problematic following bariatric surgery (34). Wardé-Kamar and cols. (35) suggest that such low adherence may occur due to underestimation of, or misinformation about, the surgery’s consequences after long periods of time. Patients who regain weight may also feel embarrassed to seek help from the care service center where they had the operation. Reasons such as loss or change of health insurance, job loss, moving to another city or simply choosing not to continue the treatment after surgery may also reduce follow-up participation. The limitations of our study include the retrospective nature of the analysis, incomplete followup, reduced sample size at the end of the study, and difficulty contacting patients who did not attend their scheduled follow-up appointments. These limitations are commonly reported in the literature for this kind of study. As laboratory results might change with further increases in body weight, a follow-up period of longer than 5 years would be necessary to establish whether they eventually return to preoperative levels (impairing the surgery’s success and the patient’s health). However, in spite of these limitations, longterm studies are crucial to determine long-term patient outcomes after bariatric surgery. In conclusion, our results suggest that weight regain has no significant impact in the long-term evolution of the lipid profile and glycemia after RYGB. This study highlights the need for closer follow-up of RYGB patients, and for the implementation of actions aimed at improving adherence to the treatment protocols established by multidisciplinary teams. Additional longterm studies are required to establish whether a weight regain threshold might affect laboratory results, and to 350

elucidate the exact mechanisms leading to these effects and their clinical consequences for the patient’s health. Acknowledgements: we thank Fernando Luiz Barroso. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Stefater M, Kohli R, Inge T. Advances in the surgical treatment of morbid obesity. Mol Aspects Med. 2013;34(1):84-94. 2.

Christou N, Look D, Maclean L. Weight gain after short- and longlimb gastric bypass in patients followed for longer than 10 years. Ann Surg. 2006;244(5):734-40.

3. Malinowski S. Nutritional and metabolic complications of bariatric surgery. Am J Med Sci. 2006;331(4):219-25. 4. Buchwald H, Estok R, Fahrbach K, Banel D, Jensen MD, Pories WJ, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and meta-analysis. Am J Med. 2009;122(3):248-256.e5. 5. Garcia-Marirrodriga I, Amaya-Romero C, Ruiz-Diaz GP, Férnandez S, Ballesta-López C, Pou JM, et al. Evolution of lipid profiles after bariatric surgery. Obes Surg. 2012;22(4):609-16. 6. Bradley D, Conte C, Mittendorfer B, Eagon JC, Varela JE, Fabbrini E, et al. Gastric bypass and banding equally improve insulin sensitivity and β cell function. J Clin Invest. 2012;122(12):4667-74. 7.

Schauer PR, Burguera B, Ikramuddin S, Cottam D, Gourash W, Hamad G, et al. Effect of laparoscopic Roux-en Y gastric bypass on type 2 diabetes mellitus. Ann Surg. 2003 Oct;238(4):467-84.

8. Shah M1, Simha V, Garg A. Review: Review: long-term impact of bariatric surgery on body weight, comorbidities, and nutritional status. J Clin Endocrinol Metab. 2006;91(11):4223-31. 9. World Health Organization: Physical Status: the use and interpretation of anthropometry: Technical report series 854. Geneva; 1998. p. 1-452. 10. Gibson S. Principles of nutritional dymatic. New York: Oxford; 1990. p. 691. 11. McGowan MW, Artiss JD, Strandbergh DR, Zak B. A peroxidasecoupled method for the colorimetric determination of serum triglycerides. Clin Chem. 1983;29(3):538-42. 12. Friedwald W, Levy R, Fredrickson D. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499-502. 13. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA. 2001;285(19):2486-97. 14. American Diabetes Association. Standards of medical care in diabetes--2012. Diabetes Care. 2012;35:11-63. 15. Dalcanale L, Oliveira CP, Faintuch J, Nogueira MA, Rondó P, Lima VM, et al. Long-term nutritional outcome after gastric bypass. Obes Surg. 2010;20(2):181-7. 16. Faria SL, de Oliveira Kelly E, Lins RD, Faria OP. Nutritional management of weight regain after bariatric surgery. Obes Surg. 2010;20(2):135-9. 17. Meguid MM, Glade MJ, Middleton FA. Weight regain after Rouxen-Y: a significant 20% complication related to PYY. Nutrition. 2008;24(9):832-42. 18. Pories WJ. Bariatric surgery: risks and rewards. J Clin Endocrinol Metab. 2008 Nov;93(11 Suppl 1):S89-96. Arch Endocrinol Metab. 2018;62/3

Glucose and lipid concentrations after Roux-em-Y gastric bypass

19. Lopez P, Patel N, Koche L. Outpatient complications encountered following Roux-en-Y gastric bypass. Med Clin North Am. 2007;91(3):471-83, xii. 20. Christou N, Look D, Maclean L. Weight gain after short- and longlimb gastric bypass in patients followed for longer than 10 years. Ann Surg. 2006;244(5):734-40. 21. Francischi R, Pereira L, Freitas C, Klopfer M, Santos RC, Vieira P, et al. Obesidade: atualização sobre sua etiologia, morbidade e tratamento. Rev Nutr. 2000;13(1):17-28. 22. Benaiges D, Goday A, Ramon JM, Hernandez E, Pera M, Cano JF; Obemar Group. Laparoscopic sleeve gastrectomy and laparoscopic gastric bypass are equally effective for reduction of cardiovascular risk in severely obese patients at one year of follow-up. Surg Obes Relat Dis. 2011;7(5):575-80. 23. Sundquist J, Winkleby M, Pudaric S. Cardiovascular disease risk factors among older black, Mexican-American, and white women and men: an analysis of NHANES III, 1988-1994. Third National Health and Nutrition Examination Survey. Third National Health and Nutrition Examination Survey. J Am Geriatr Soc. 2001;49(2):109-16. 24. Brolin RE, Bradley LJ, Wilson AC, Cody RP. Lipid risk profile and weight stability after gastric restrictive operations for morbid obesity. J Gastrointest Surg. 2000;4(5):464-9. 25. Jamal M, Wegner R, Heitshusen D, Liao J, Samuel I. Resolution of hyperlipidemia follows surgical weight loss in patients undergoing Roux-en-Y gastric bypass surgery: a 6-year analysis of data. Surg Obes Relat Dis. 2011;7(4):473-9.

28. Peterli R, Wölnerhanssen B, Peters T, Devaux N, Kern B, Christoffel-Courtin C, et al. Improvement in glucose metabolism after bariatric surgery: comparison of laparoscopic Roux-en-Y gastric bypass and laparoscopic sleeve gastrectomy: a prospective randomized trial. Ann Surg. 2009;250(2):234-41. 29. Korner J, Bessler M, Inabnet W, Taveras C, Holst JJ. Exaggerated glucagon-like peptide-1 and blunted glucose-dependent insulinotropic peptide secretion are associated with Roux-en-Y gastric bypass but not adjustable gastric banding. Surg. Surg Obes Relat Dis. 2007;3(6):597-601. 30. Pournaras DJ, Osborne A, Hawkins SC, Mahon D, Ghatei MA, Bloom SR, et al. The gut hormone response following Roux-en-Y gastric bypass: cross-sectional and prospective study. Obes Surg. 2010;20(1):56-60. 31. Corcelles R, Daigle CR, Schauer PR. Management of endocrine disease: metabolic effects of bariatric surgery. Eur J Endocrinol. 2016;174(1):R19-28. 32. Barhouch AS, Zardo M, Padoin AV, Colossi FG, Casagrande DS, Chatkin R, et al. Excess weight loss variation in late postoperative period of gastric bypass. Obes Surg. 2010;20(11):1479-83. 33. Odom J, Zalesin KC, Washington TL, Miller WW, Hakmeh B, Zaremba DL, et al. Behavioral predictors of weight regain after bariatric surgery. Obes Surg. 2010;20(3):349-56. 34. Silver HJ, Torquati A, Jensen GL, Richards WO. Weight, dietary and physical activity behaviors two years after gastric bypass. Obes Surg. 2006;16(7):859-64. 35. Wardé-Kamar J, Rogers M, Flancbaum L, Laferrère B. Calorie intake and meal patterns up to 4 years after Roux-en-Y gastric bypass surgery. Obes Surg. 2004;14(8):1070-9.

26. Laguna S, Andrada P, Silva C, Rotellar F, Valenti V, Gil MJ, et al. Body weight-independent variations in HDL-cholesterol following gastric bypass. In: Anales del Sistema Sanitario De Navarra. Navas Tolosa 21, Pamplona, 31002, Spain: Gobierno de Navarra, 2016. p. 23-33.

27. Reed MA, Pories WJ, Chapman W, Pender J, Bowden R, Barakat H, et al. Roux-en-Y gastric bypass corrects hyperinsulinemia implications for the remission of type 2 diabetes. J Clin Endocrinol Metab. 2011;96(8):2525-31.

Arch Endocrinol Metab. 2018;62/3

351

review

An update of genetic basis of PCOS pathogenesis Raiane P. Crespo1, Tania A. S. S. Bachega1, Berenice B. Mendonça1, Larissa G. Gomes1

Divisão de Endocrinologia e Laboratório de Hormônios e Genética Molecular (LIM-42), Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HCFMUSP), São Paulo, SP, Brasil

Correspondence to: Larissa Garcia Gomes Av. Dr. Enéas de Carvalho Aguiar, 155, 2º andar, Bloco 6 05403-000 – São Paulo, SP, Brasil larissag.gomes@gmail.com Received on Dec/13/2017 Accepted on May/30/2018 DOI: 10.20945/2359-3997000000049

ABSTRACT Polycystic ovary syndrome (PCOS) is a common and complex endocrine disorder that affects 5-20% of reproductive age women. PCOS clinical symptoms include hirsutism, menstrual dysfunction, infertility, obesity and metabolic syndrome. There is a wide heterogeneity in clinical manifestations and metabolic complications. The pathogenesis of PCOS is not fully elucidated, but four aspects seem to contribute to the syndrome to different degrees: increased ovarian and/or adrenal androgen secretion, partial folliculogenesis arrest, insulin resistance and neuroendocrine axis dysfunction. A definitive etiology remains to be elucidated, but PCOS has a strong heritable component indicated by familial clustering and twin studies. Genome Wide Association Studies (GWAS) have identified several new risk loci and candidate genes for PCOS. Despite these findings, the association studies have explained less than 10% of heritability. Therefore, we could speculate that different phenotypes and subphenotypes are caused by rare private genetic variants. Modern genetic studies, such as whole exome and genome sequencing, will help to clarify the contribution of these rare genetic variants on different PCOS phenotypes. Arch Endocrinol Metab. 2018;62(3):352-61 Keywords Polycystic ovary syndrome; PCOS; genetics; pathogenesis; complex genetic trait

INTRODUCTION

olycystic ovary syndrome (PCOS) is a common and complex endocrine disorder that affects 5 to 20% of reproductive age women, and it is a major cause of hirsutism and anovulatory infertility (1). However, PCOS is more than a reproductive disease; it is associated with a wide range of metabolic disorders, such as glucose intolerance, diabetes, dyslipidemia, hypertension, hepatic steatosis, and increased cardiovascular surrogate markers. An increased prevalence of cardiovascular outcomes and mortality has not yet been demonstrated in this population (2). PCOS diagnosis is currently based on a combination of the following signs and symptoms: hyperandrogenism and/or hyperandrogenemia, ovulatory dysfunction and polycystic ovarian morphology (PCOM). In the first attempt to define diagnostic criteria, the National Institutes of Health (NIH) Conference proposed in 1990 that the presence of both hyperandrogenism and chronic anovulation were mandatory for diagnosis (3) (Table 1). In 2003, the Rotterdam Consensus included PCOM finding as a criterion, and defined that the presence of at least two out of the three main characteristics were necessary to confirm the diagnosis 352

(Table 1) (4). Therefore, the Rotterdam Consensus expanded the possibilities of combinations of the three classic manifestations, allowing the characterization of four main phenotypes (Table 2). In 2006, the Androgen Excess Society proposed a modification, in which hyperandrogenism would be an essential condition for diagnosis, reducing the number of phenotypic possibilities (5). More recently, a Consensus by NIH tried to unify the recommendations, suggesting the use of the Rotterdam Consensus for PCOS diagnosis in order to include the four phenotypes possibilities, and thus expanding the population with PCOS to its largest number (6).

CLINICAL ASPECTS: WIDE HETEROGENEITY There is wide clinical heterogeneity, in terms of not only the main components of the syndrome, but also of the metabolic aspects. Cross-sectional studies have shown that metabolic abnormalities are more common in the classic and NIH phenotypes, often classified together with the denomination of classic (also adopted in this current review), compared to the other Rotterdam phenotypes, namely, ovulatory and normoandrogenic (Table 2). Moghetti and cols. found that the prevalence Arch Endocrinol Metab. 2018;62/3

of each phenotype is about 70% for classic, 15% for ovulatory and 15% for normoandrogenic phenotypes (7). Other epidemiological studies in unselected populations reported the prevalence of 40–45% for classic phenotype, ~35% for ovulatory phenotype and ~20% for normoandrogenic phenotype (8). The prevalence of insulin resistance in PCOS patients is around 70%; however, insulin resistance differs among phenotypes: 80% in the classic, 65% in the ovulatory and only 38% in the normoandrogenic phenotype (7). The presence of metabolic syndrome also differs among these groups. Çelik and cols. compared 183 normoandrogenic PCOS patients with 504 classic/ ovulatory PCOS patients and found a significantly higher prevalence of metabolic syndrome (OR 2.95) in the classic/ovulatory group (9). This clinical and metabolic heterogeneity could be explained by distinct pathogenic causes. Table 1. Diagnostic criteria for PCOS NIH 1990

Rotterdam 2003

AE-PCOS Society 2006

• Chronic anovulation • Clinical and/or biochemical signs of hyperandrogenism (Both criteria needed)

• Oligo- and/or anovulation • Clinical and/or biochemical signs of hyperandrogenism • Polycystic ovaries (Two of three criteria needed)

• Clinical and/or biochemical signs of hyperandrogenism • Ovarian dysfunction (Oligo-anovulation and/or polycystic ovarian morphology) (Both criteria needed)

Table 2. PCOS phenotypes according to each diagnostic criteria Phenotypes

Classic

NIH

Hyperandrogenism

Yes

Chronic anovulation

Yes

Polycystic ovaries

Yes

Ovulatory Normoandrogenic Yes

Yes

Yes X

NIH 1990

Rotterdam 2003

AE-PCOS Society 2006

PCOS PATHOGENESIS ASPECTS Ovarian and adrenal hyperandrogenism There are considerable evidences, an that PCOS is an intrinsic disorder of the ovaries and that the primary defect resides in the increased androgen biosynthesis. Normally, ovarian theca cells produce androgens in response to LH. The theca cells express the CYP17A1 gene, which encodes the P450c17 enzyme, catalyzing Arch Endocrinol Metab. 2018;62/3

both 17α-hydroxylase and 17,20-lyase activity, the rate-limiting step in sex steroid synthesis. Androgen production is cyclical and modulated by intraovarian and extra-ovarian mechanisms. As LH rises, a downregulation of LH receptors and a decrease in CYP17A1 expression occur, minimizing the androgenic production. Estrogen and androgen inhibit 17α-hydroxylase and 17,20-lyase activity in a paracrine and autocrine negative feedback loop. In contrast, insulin and IGFs stimulate the P450c17 enzyme and up-regulate LH receptor sites (10) (Figure 1). PCOS ovaries are typically hypersensitive to LH, which is caused by partial escape from LH receptor downregulation. The imbalance of the intra-ovarian regulatory system seems to play a role in this increased sensitivity to LH. Exogenous factors, such as insulin resistance and hyperinsulinemia, are examples of extraovarian modulators that disrupt normal intra-ovarian regulatory mechanisms (10) (Figure 1). In vitro studies have demonstrated that isolated theca cells or tissues from PCOS women secrete more androgen than cells from normal controls. The increased androgen synthesis in PCOS theca cells results from increased expression of 17α-hydroxylase/17,20-lyase and the cholesterol side-chain cleavage enzyme, suggesting an intrinsic theca cell defect (11). Taken together, these data indicate that ovarian cells from PCOS patients have intrinsic and stable characteristics that contribute to their phenotype, even after exclusion of gonadotrophin, hyperinsulinemia and hyperandrogenism stimulus. Although the ovaries are considered the main source of androgens in PCOS, increased adrenal androgens, mainly DHEA and DHEAS, are present in around 20–30% of PCOS women (12). Normal adrenal zona reticularis steroidogenic enzymes resemble theca cell enzymes, favoring the formation of DHEA, which is rapidly sulfated to DHEAS by sulfotransferase 2A1 (SULT2A1). DHEAS is the most abundant adrenal precursor, but it is an inert terminal product. However, DHEAS can be converted to DHEA, which can be metabolized into more potent androgens. Genetic variants in SULT2A1 have been shown to alter the DHEAS/DHEA ratio in PCOS (13). Increased adrenal P450c17 activity is also suggested in PCOS women with adrenal hyperandrogenism. This adrenal hyperandrogenism does not seem to depend on an increased hypothalamic–pituitary-adrenal drive, but it reflects a generalized adrenal hyper-responsiveness for androgenic biosynthesis (12). 353

An update of genetic basis of PCOS pathogenesis

Figure 1. Ovarian steroidogenesis. (A) Normal ovarian steroid synthesis. (B) PCOS ovarian steroid synthesis. In comparison to normal theca cells, PCOS show increased expression of LH receptor and increased expression of CYP17A1 gene, leading to enhance of 17α-hydroxylase and 17,20-lyase activity, and amplifying androgen synthesis. Exogenous factors, such as hyperinsulinemia and IGFs are modulatory hormones that can disrupt normal intra-ovarian regulatation of steroidogenesis.

theca cells

354

Arch Endocrinol Metab. 2018;62/3

An update of genetic basis of PCOS pathogenesis

Genetic and epigenetic factors are probably implicated in the theca cell and adrenal steroidogenesis dysfunction. An example of a genetic cause of adrenal PCOS is cortisone reductase deficiency. The original study described mutations in 11β-hydroxysteroid dehydrogenase (11βHSD) type I and hexose-6phosphate dehydrogenase in a digenic triallelic mode of inheritance. In11β-HSD type I deficiency, cortisone is not convert into cortisol, with a consequent elevation of ACTH, which stimulates androgen production, leading to the PCOS phenotype (14).

Ovarian folliculogenesis The beginning of follicular recruitment is an intrinsic ovarian process, independent of gonadotropins. The primordial follicles are maintained in the quiescence state by epithelial-mesenchymal interactions, inhibitory transcription factors (LKB1/STK11/ BMP4) and proapoptotic factors (FOX) secreted by oocytes. A set of local follicular factors regulates follicular recruitment (GDF9, BMP6 and BMP9) (10), follicle growth and development. For example, BMP15 acts synergistically with GDF9 proliferating the granulosa cells (10). Another important modulator of folliculogenesis is AMH, a member of the GFR-beta superfamily that is produced by the granulosa cells of small growing follicles. The highest expression of AMH is detected in preantral and small antral follicles of ≤ 4 mm. Once the follicles reach > 8 mm, AMH expression is reduced, and

Gonadotropin-independent folliculogenesis

these follicles become more sensitive to FSH action, culminating with increased follicle growth and estrogen production, selection of the dominant follicle and subsequent ovulation in the normal ovary (15). AMH acts to prevent the initial FSH-independent recruitment of primary follicles from the primordial follicle pools, inhibits FSH-dependent follicular maturation and the selection of the dominant follicle (1) (Figure 2). Several studies have shown that serum AMH levels are significantly higher in patients with PCOS (15-17), and this finding is compatible with the large number of follicles in the preantral and antral stages, both of which are stages of maximum AMH production. Pellatt and cols. demonstrated that the production of AMH per cell is 75 times greater in the granulosa cells of anovulatory PCOS patients (16). This finding suggests that high serum AMH levels are not only a consequence of the increased granulosa cell mass in anovulatory patients, but are also a potential causal factor for anovulation in PCOS (15,16). However, the cause for the increased production of AMH per granulosa cell remains unknown. Several other signals produced by granulosa cells and oocytes modulate (positively and negatively) the ovarian environment for oocyte maturation. These factors include inhibin B, TGF-beta superfamily members (ex. BMPs), cytokines (ex. TNF-alpha), and microRNAs (10).

Gonadotropin-dependent folliculogenesis

Granulosa cell

Primary follicle

Preantral follicle

Early antral follicle

Late antral follicle

Preovulatory follicle

Cystic follicle

P450aro

Figure 2. Role of AMH in folliculogenesis. AMH has an inibitory effect on initial recruitment of primary follicles, on FSH-dependent follicular maturation and selection of dominant follicle, and on FSH-induced aromatase expression on granulosa cells, reducing the conversion of testosterone into estradiol. Higher AMH levels in PCOS patients turns the follicles more resistant to FSH action, culminating in inhibition of follicular maturarion and ovulation, and in inhibiton of aromatase expression, and consequently, leading to hyperandrogenism. Adapted from Azziz and cols. 2016 (1). Arch Endocrinol Metab. 2018;62/3

355

Primordial follicle

Thecal cell

An update of genetic basis of PCOS pathogenesis

Regulatory folliculogenesis factors, such as GDF9 and BMP15, have already been studied as possible causes of PCOS, but there is still no evidence that these molecules are responsible for abnormal folliculogenesis (18). Mutations of BMP15 have been implicated in early ovarian failure caused by ovarian dysgenesis (19). FOX transcription factors have also demonstrated a potential role in PCOS pathogenesis. Mikaeili and cols. correlated FOX03 expression and activation in granulosa cells with apoptosis rates, and found a positive correlation between them, making this transcription factor a potential candidate target (20).

amplifies the response of theca cells to LH and increases the expression of P450c17 and 3Î˛-HSD2, leading to increased androgen production (29). This pathway is well illustrated by studies in rats presenting IR induced by an obesogenic diet; in these studies, the knock-out of insulin receptors improved hyperandrogenism and anovulation (30). Reaffirmation of the role of insulin in steroidogenesis is observed in the treatment of PCOS. All treatments that reduce serum insulin levels, whether weight loss due to lifestyle changes, bariatric surgery, metformin or thiazolidinedione, significantly diminish anovulation and hyperandrogenemia (31,32).

Insulin resistance

Neuroendocrine axis

In 1970, rare forms of severe insulin resistance (IR), acanthosis nigricans and hyperandrogenism, were described (Insulin Resistance type A, RabsonMendenhall Syndrome and Leprechaunism) (21). The molecular mechanism of insulin resistance in these diseases was explained by a reduction in the binding of insulin to its receptor and by a defect in insulin receptor autophosphorylation (22,23). Family forms of lipodystrophy with severe IR were also associated with hyperandrogenism at the same time (24). Remarkable hyperinsulinemia, as a common feature of these syndromes, suggested for the first time that insulin could directly stimulate androgen production (24). In 1980, Burghen and cols. reported that women with PCOS had increased insulin levels in response to the oral glucose tolerance test that were not justified only by obesity (25). In addition, women with classical PCOS had acanthosis nigricans, which gave them a phenotypic picture similar to the rare IR syndromes described previously (25). Observational studies suggest that in obese PCOS women, metabolic abnormalities related to IR and obesity have a greater impact on anovulation mechanisms than excess androgens (26). However, IR in PCOS patients is not necessarily associated with obesity. Some lean patients with PCOS present with IR, and even obese women have IR disproportionate to the degree of adiposity (27). Therefore, insulin resistance in PCOS is an intrinsic characteristic aggravated by obesity and not simply a consequence of obesity. In PCOS syndrome, IR is not generalized in all tissues. Defects in metabolic actions of insulin occur in muscle and adipose tissues, but mitogenic and steroidogenic actions of insulin in ovaries are preserved (28). By acting on the ovarian insulin receptor, insulin

Since ovarian steroidogenesis is directly related to the gonadotropic stimulus, alterations of LH are indicated as one of the factors involved in the development of PCOS. Progesterone is the primary regulator of GnRH pulsatility. It is known that PCOS patients present a GnRH-generating pulse resistance to negative feedback by progesterone, which results in a higher LH pulses frequency and/or amplitude (33). This phenomenon has already been attributed to hyperandrogenemia in several studies, in which treatment with flutamide for 4 weeks restored hypothalamic sensitivity to ovarian steroids (34). At this point, alterations of the gonadotropic axis appear to be secondary to hyperandrogenemia. Moreover, LH excess is an inconstant feature of PCOS, thus, it is difficult to attribute a primary role to LH excess in the pathogenesis of all PCOS patients. However, in patients with congenital virilization, changes in LH pulsatility appear to be determinant for the PCOS phenotype (10). Therefore, ovarian and adrenal hyperandrogenism, altered folliculogenesis, increased insulin resistance and neuroendocrine axis dysfunction could have a greater or lesser role in PCOS pathogenesis (Figure 3). The classification of specific clinical patterns in patients with PCOS could improve the search for genetic determinants and potentially the development of specific treatment regimens.

356

PCOS GENETIC HERITABILITY Since 1968, studies have suggested an important genetic role contributing to the etiology of PCOS (35). First-degree relatives of PCOS patients have a higher Arch Endocrinol Metab. 2018;62/3

An update of genetic basis of PCOS pathogenesis

Classic phenotype

Ovulatory phenotype

Normoandrogenic phenotype

Figure 3. Schematic components of PCOS pathogenesis. Four main physiologic mechanisms contribute to PCOS pathogenesis: hyperandrogenism (HY), insulin resistance (IR), folliculogenesis dysfunction (FC), and neuroendocrine axis dysfunction (ND). These mechanisms contribute to each phenotype in different degrees. The classical phenotype shows alterations in all four mechanism, being HY and IR the main ones (A). In the ovulatory phenotype, HY and ND predominate and folliculogenesis seems not compromised (B). In contrast, in the normoandrogenic phenotype, FC seems to play the most important role in the pathogenesis, while HY is not present (C).

Arch Endocrinol Metab. 2018;62/3

for PCOS (41,42) (Table 3). These data were validated in Caucasians, confirming the presence of 12 out of the 17 variants with the same risk effect, indicating a similar genetic risk profile in different populations (43-48). However, the frequency of risk alleles differs significantly among populations (43). Table 3. PCOSGWAS susceptibility gene markers Han Chinese: 2011 (41), 2012 (42)

European 1 2015 (47)

European 2 2015 (48)

LHCGR

GATA4/NEIL2

ERBB4

FSHR

FSHB

C9orf3

RAD50

THADA

DENND1A

KRR1

YAP1

RAB5B5 HMGA2 TOX3 INSR

GWAS results have provided new insights into the biological pathways that may be involved in PCOS pathogenesis. The DENND1A (Differentially Expressed in Normal and Neoplastic Development isoform A1) gene was identified as a potential risk marker. DENND1A encodes a protein (connecdenn 1) associated with clathrin-coated pits where cell-surface receptors reside. The DENND1A protein is located in the cytoplasm and in the nuclei of theca cells (10). This gene has two major transcripts: DENND1A variant 1 (DENND1A.V1), which encodes a 1,009-aa protein with a C-terminal proline-rich domain, and 357

risk of being affected by the syndrome compared to the general population. Kahsar-Miller and cols. studied first-degree relatives of 93 PCOS patients and found that 35% of non-menopausal mothers and 40% of sisters were also affected (36). Examining a large twin cohort of 1332 monozygotic and 1873 dizygotic twin sisters, Vink and cols. found a higher correlation for PCOS in monozygotic compared to dizygotic twins. This study concluded that the genetic component contributes to over 70% of PCOS pathogenesis (37). Therefore, PCOS is a complex genetic disease with high inheritance rates and heterogeneous phenotypes. Studies based on candidate genes for PCOS have not been successful. Studies involving more than 100 candidate genes, especially those related to reproductive axis, insulin resistance and chronic inflammation have not shown reproducible results (38,39). The lack of consistency across studies may stem from the small sample size, phenotypic heterogeneity among patients, an inadequate control group, the lack of comorbidity matching (obesity, insulin resistance), or the limitation to only one or two variants genotyped in each gene of interest instead of the whole gene (40). The genome-wide association study (GWAS) is a more recent genomic approach to understanding the genetics of complex diseases. This method searches the genome for single nucleotide variants that occur more frequently in people with a particular disease than in people without the disease. The goal is to identify genetic risk factors for common diseases. Two GWAS conducted in Han Chinese women identified 11 loci, accounting for 17 SNPs, with a strong association risk

An update of genetic basis of PCOS pathogenesis

DENND1A variant 2 (DENND1A.V2), which encodes a truncated 559-aa protein that contains the DENN domain and a clathrin-binding domain; however, the protein lacks the proline-rich domain and includes a C-terminal 33-aa sequence that is not found in the larger connecdenn variant 1 (49). McAllister’s group showed that DENND1A.V2 protein is more expressed in PCOS theca cells compared to normal theca cells. They also forced overexpression of DENND1A.V2 in normal theca cells, resulting in increased CYP17A1 and CYP11A1 expression and increased androgen biosynthes, compatible with PCOS theca cells profile. Additionally, knock-out of DENND1A.V2 in PCOS theca cells reduced CYP17A1 and CYP11A1 expression and androgen biosynthesis (10,49). Recently, DENND1A expression in the adrenal zona reticularis was shown (50). Taking all these data into consideration, the DENND1A variant 2 is potentially one of the mechanisms involved with the intrinsic abnormality in ovarian theca cells steroidogenesis in PCOS. Some other genes (LHCGR, FSHR, INRS, THADA, HMGA2, RAB5B, SUOX, YAP, ZNF217) located in susceptibility loci indicated by GWAS studies were related to the gonadotropic regulatory axis, glucose and lipid metabolism, and cell cycle regulation. • The LHCGR (luteinizing hormone/ choriogonadotropin receptor) gene is a G protein-coupled receptor, which is expressed in granulosa cells, mainly in the preovulatory follicles. Increased LHCGR expression in granulosa cells in preovulatory follicles allows these follicles to respond to the LH peak in the middle of the cycle, resulting in ovulation. Inactivating mutations of this gene result in increased LH levels, amenorrhea or oligomenorrhea and infertility (51,52). In contrast, activating mutations of this gene are associated with hyperandrogenism (53). • The FSHR (follicle stimulating hormone receptor) gene is related to the ovarian response to FSH and appears to be a highly plausible candidate gene for PCOS. Inactivating mutations of FSHR lead to hypergonadotropic hypogonadism and follicles stagnation in the preantral stage (50). Polymorphic variations of this gene in patients with PCOS have been associated with increased FSH concentrations (44) and resistance to exogenous ovarian 358

stimulation with gonadotrophins and clomiphene (54). • The INRS gene is the most prominent gene associated with insulin resistance indicated by GWAS. Mutations in the tyrosine kinase domain of the insulin receptor are known causes of insulin resistance and severe hyperinsulinemia (55). Polymorphic variants in different exons have been reported as risk factors for PCOS, but the results were generally not consistent in subsequent studies (50). • THADA and HMGA2 genes have also been described in the GWAS studies involving patients with DM2 (56). RAB5B and SUOX genes are located at the susceptibility locus for DM1 (57). YAP and ZNF217 genes do not have a known ovarian function but they are related to cell proliferation and apoptosis (58). The wide genetic possibilities indicated by GWAS could be related to the phenotypic heterogeneity. However, these loci identified in GWAS so far explain less than 10% of PCOS heritability. The rationale for GWAS large-scale sequencing studies is “common disease is associated with common variants”, postulating that common diseases are related to allelic variants present in more than 1-5% of the population. From this principle, GWAS studies allow the identification of an enormous number of genetic variants associated with complex diseases; however, most of the variants individually or in combination confer small increments in risk (1.1-1.5 fold). Variants with lower frequency and larger effect size that are not captured by GWAS may account for the deficit in heritability found using this method (59). Moreover, the associations of the PCOS risk variants and some PCOS features, such as testosterone levels, hyperinsulinemia, and ovarian morphology could not be clearly established. The next step in genomics era is understanding how these variants contribute to disease’s pathogenesis as discussed by Jones and Goodarzi (60).

GENETIC APPROACH IN PCOS POS-GWAS DISCOVERIES Currently, GWAS have provided insights into the genetics of PCOS, pointing to some susceptible loci, usually in a non-coding region associated with the disease, but at the same time, these studies have shown that this complex disease cannot be explained Arch Endocrinol Metab. 2018;62/3

An update of genetic basis of PCOS pathogenesis

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

REFERENCES 1. Azziz R, Carmina E, Chen Z, Dunaif A, Laven JS, Legro RS, et al. Polycystic ovary syndrome. Nat Rev Dis Primers. 2016;2:16057.

4. Group REA-SPCW. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril. 2004;81(1):19-25. 5. Azziz R, Carmina E, Dewailly D, Diamanti-Kandarakis E, EscobarMorreale HF, Futterweit W, et al. Positions statement: criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Society guideline. J Clin Endocrinol Metab. 2006;91(11):4237-45. 6. Legro RS, Arslanian SA, Ehrmann DA, Hoeger KM, Murad MH, Pasquali R, et al. Diagnosis and treatment of polycystic ovary syndrome: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2013;98(12):4565-92. 7.

Moghetti P, Tosi F, Bonin C, Di Sarra D, Fiers T, Kaufman JM, et al. Divergences in insulin resistance between the different phenotypes of the polycystic ovary syndrome. J Clin Endocrinol Metab. 2013;98(4):E628-37.

8. Lizneva D, Suturina L, Walker W, Brakta S, Gavrilova-Jordan L, Azziz R. Criteria, prevalence, and phenotypes of polycystic ovary syndrome. Fertil Steril. 2016;106(1):6-15. 9. Çelik E, Türkçüoğlu I, Ata B, Karaer A, Kırıcı P, Eraslan S, et al. Metabolic and carbohydrate characteristics of different phenotypes of polycystic ovary syndrome. J Turk Ger Gynecol Assoc. 2016;17(4):201-8. 10. Rosenfield RL, Ehrmann DA.The Pathogenesis of Polycystic Ovary Syndrome (PCOS):The Hypothesis of PCOS as Functional Ovarian Hyperandrogenism Revisited. Endocr Rev. 2016;37(5):467-520. 11. Nelson VL, Legro RS, Strauss JF, McAllister JM. Augmented androgen production is a stable steroidogenic phenotype of propagated theca cells from polycystic ovaries. Mol Endocrinol. 1999;13(6):946-57. 12. Goodarzi MO, Carmina E, Azziz R. DHEA, DHEAS and PCOS. J Steroid Biochem Mol Biol. 2015;145:213-25. 13. Louwers YV, de Jong FH, van Herwaarden NA, Stolk L, Fauser BC, Uitterlinden AG, et al. Variants in SULT2A1 affect the DHEA sulphate to DHEA ratio in patients with polycystic ovary syndrome but not the hyperandrogenic phenotype. J Clin Endocrinol Metab. 2013;98(9):3848-55. 14. Draper N, Walker EA, Bujalska IJ, Tomlinson JW, Chalder SM, Arlt W, et al. Mutations in the genes encoding 11beta-hydroxysteroid dehydrogenase type 1 and hexose-6-phosphate dehydrogenase interact to cause cortisone reductase deficiency. Nat Genet. 2003;34(4):434-9. 15. Garg D,Tal R.The role of AMH in the pathophysiology of polycystic ovarian syndrome. Reprod Biomed Online. 2016;33(1):15-28. 16. Pellatt L, Hanna L, Brincat M, Galea R, Brain H, Whitehead S, et al. Granulosa cell production of anti-Müllerian hormone is increased in polycystic ovaries. J Clin Endocrinol Metab. 2007;92(1):240-5. 17. Fallat ME, SiowY, Marra M, Cook C, Carrillo A. Müllerian-inhibiting substance in follicular fluid and serum: a comparison of patients with tubal factor infertility, polycystic ovary syndrome, and endometriosis. Fertil Steril. 1997;67(5):962-5. 18. Wei LN, Huang R, Li LL, Fang C, Li Y, Liang XY. Reduced and delayed expression of GDF9 and BMP15 in ovarian tissues from women with polycystic ovary syndrome. J Assist Reprod Genet. 2014;31(11):1483-90. 19. Dixit H, Rao LK, Padmalatha VV, Kanakavalli M, Deenadayal M, Gupta N, et al. Missense mutations in the BMP15 gene are associated with ovarian failure. Hum Genet. 2006;119(4):408-15.

2. Gunning MN, Fauser BCJM. Are women with polycystic ovary syndrome at increased cardiovascular disease risk later in life? Climacteric. 2017;20(3):222-7.

20. Mikaeili S, Rashidi BH, Safa M, Najafi A, Sobhani A, Asadi E, et al. Altered FoxO3 expression and apoptosis in granulosa cells of women with polycystic ovary syndrome. Arch Gynecol Obstet. 2016;294(1):185-92.

3. JK Z. Diagnostic criteria of polycystic ovary syndrome; towards a rational approach. In: Dunaif A, editor. Polycystic Ovary Syndrome ed. Boston, MA: Blackwell Scientific; 1992.

21. Kahn CR, Flier JS, Bar RS, Archer JA, Gorden P, Martin MM, et al.The syndromes of insulin resistance and acanthosis nigricans. Insulinreceptor disorders in man. N Engl J Med. 1976;294(14):739-45.

Arch Endocrinol Metab. 2018;62/3

359

by a limited number of variants with a low effect on phenotype. Much of the speculation about missing heritability from PCOS GWAS studies has focused on the role of rare genetic variants with substantial effect on phenotype. It is in this context that massive parallel sequencing, which can process millions of sequences in parallel, provides a mechanism to screen the whole exome or to sequence the whole genome for rare variants. Whole exome sequencing (WES) has been able to advance the understanding of complex diseases via two approaches: searching for lower frequency coding variants that may only be present in subsets of the population and sequencing individuals with a severe phenotype. Rare variants with large effects in a specific gene may be found in extreme phenotypes, which can provide insights into the underlying pathophysiology of the common disorder (61). Recently, Gorsic and cols. identified 18 rare variants in PCOS patients using WES, and 17 of 18 (94%) of the variants were associated with reduced AMH signaling. They hypothesize that these AMH mutations lead to the PCOS phenotype by disrupting AMH’s transcriptional inhibition of CYP17A1, leading to increased androgen biosynthesis (62). As demonstrated, WES seems promising in identifying rare genetic variants that contribute to PCOS pathogenesis and in mapping genetic variants that contribute to different aspects of each phenotype of the syndrome. In conclusion, major advances in the genetics knowledge, combining with functional data and epigenetic studies in PCOS might increase our understanding of the etiology of this disease, with possible translational implications. The determination of genetic markers may improve the diagnosis of the syndrome and its phenotypes, allowing early intervention in associated co-morbidities and adequating treatment in a more individualized way.

An update of genetic basis of PCOS pathogenesis

22. Musso C, Cochran E, Moran SA, Skarulis MC, Oral EA, Taylor S, et al. Clinical course of genetic diseases of the insulin receptor (type A and Rabson-Mendenhall syndromes): a 30-year prospective. Medicine (Baltimore). 2004;83(4):209-22. 23. Taylor SI, Cama A, Accili D, Barbetti F, Quon MJ, de la Luz Sierra M, et al. Mutations in the insulin receptor gene. Endocr Rev. 1992;13(3):566-95.

41. Chen ZJ, Zhao H, He L, Shi Y, Qin Y, Li Z, et al. Genome-wide association study identifies susceptibility loci for polycystic ovary syndrome on chromosome 2p16.3, 2p21 and 9q33.3. Nat Genet. 2011;43(1):55-9.

24. Garg A, Peshock RM, Fleckenstein JL. Adipose tissue distribution pattern in patients with familial partial lipodystrophy (Dunnigan variety). J Clin Endocrinol Metab. 1999;84(1):170-4.

42. Shi Y, Zhao H, Cao Y, Yang D, Li Z, Zhang B, et al. Genome-wide association study identifies eight new risk loci for polycystic ovary syndrome. Nat Genet. 2012;44(9):1020-5.

25. Burghen GA, Givens JR, Kitabchi AE. Correlation of hyperandrogenism with hyperinsulinism in polycystic ovarian disease. J Clin Endocrinol Metab. 1980;50(1):113-6.

43. Louwers YV, Stolk L, Uitterlinden AG, Laven JS. Cross-ethnic meta-analysis of genetic variants for polycystic ovary syndrome. J Clin Endocrinol Metab. 2013;98(12):E2006-12.

26. Robinson S, Kiddy D, Gelding SV, Willis D, Niththyananthan R, Bush A, et al. The relationship of insulin insensitivity to menstrual pattern in women with hyperandrogenism and polycystic ovaries. Clin Endocrinol (Oxf). 1993;39(3):351-5.

44. Mutharasan P, Galdones E, Peñalver Bernabé B, Garcia OA, Jafari N, Shea LD, et al. Evidence for chromosome 2p16.3 polycystic ovary syndrome susceptibility locus in affected women of European ancestry. J Clin Endocrinol Metab. 2013;98(1):E185-90.

27. Dunaif A, Segal KR, Futterweit W, Dobrjansky A. Profound peripheral insulin resistance, independent of obesity, in polycystic ovary syndrome. Diabetes. 1989;38(9):1165-74. 28. 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. 29. Miller WL, Tee MK. The post-translational regulation of 17,20 lyase activity. Mol Cell Endocrinol. 2015;408:99-106. 30. Wu S, Divall S, Nwaopara A, Radovick S, Wondisford F, Ko C, et al. Obesity-induced infertility and hyperandrogenism are corrected by deletion of the insulin receptor in the ovarian theca cell. Diabetes. 2014;63(4):1270-82. 31. Nestler JE, Jakubowicz DJ. Decreases in ovarian cytochrome P450c17 alpha activity and serum free testosterone after reduction of insulin secretion in polycystic ovary syndrome. N Engl J Med. 1996;335(9):617-23. 32. Eid GM, Cottam DR, Velcu LM, Mattar SG, Korytkowski MT, Gosman G, et al. Effective treatment of polycystic ovarian syndrome with Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2005;1(2):77-80. 33. Pastor CL, Griffin-Korf ML, Aloi JA, Evans WS, Marshall JC. Polycystic ovary syndrome: evidence for reduced sensitivity of the gonadotropin-releasing hormone pulse generator to inhibition by estradiol and progesterone. J Clin Endocrinol Metab. 1998;83(2):582-90. 34. Eagleson CA, Gingrich MB, Pastor CL, Arora TK, Burt CM, Evans WS, et al. Polycystic ovarian syndrome: evidence that flutamide restores sensitivity of the gonadotropin-releasing hormone pulse generator to inhibition by estradiol and progesterone. J Clin Endocrinol Metab. 2000;85(11):4047-52. 35. Cooper HE, Spellacy WN, Prem KA, Cohen WD. Hereditary factors in the Stein-Leventhal syndrome. Am J Obstet Gynecol. 1968;100(3):371-87. 36. Kahsar-Miller MD, Nixon C, Boots LR, Go RC, Azziz R. Prevalence of polycystic ovary syndrome (PCOS) in first-degree relatives of patients with PCOS. Fertil Steril. 2001;75(1):53-8. 37. Vink JM, Sadrzadeh S, Lambalk CB, Boomsma DI. Heritability of polycystic ovary syndrome in a Dutch twin-family study. J Clin Endocrinol Metab. 2006;91(6):2100-4. Copyright© AE&M all rights reserved.

40. Mykhalchenko K, Lizneva D, Trofimova T, Walker W, Suturina L, Diamond MP, et al. Genetics of polycystic ovary syndrome. Expert Rev Mol Diagn. 2017;17(7):723-33.

38. Urbanek M, Legro RS, Driscoll DA, Azziz R, Ehrmann DA, Norman RJ, et al. Thirty-seven candidate genes for polycystic ovary syndrome: strongest evidence for linkage is with follistatin. Proc Natl Acad Sci U S A. 1999;96(15):8573-8. 39. Talbot JA, Bicknell EJ, Rajkhowa M, Krook A, O’Rahilly S, Clayton RN. Molecular scanning of the insulin receptor gene in women with polycystic ovarian syndrome. J Clin Endocrinol Metab. 1996;81(5):1979-83.

360

45. Goodarzi MO, Jones MR, Li X, Chua AK, Garcia OA, Chen YD, et al. Replication of association of DENND1A and THADA variants with polycystic ovary syndrome in European cohorts. J Med Genet. 2012;49(2):90-5. 46. Welt CK, Styrkarsdottir U, Ehrmann DA, Thorleifsson G, Arason G, Gudmundsson JA, et al. Variants in DENND1A are associated with polycystic ovary syndrome in women of European ancestry. J Clin Endocrinol Metab. 2012;97(7):E1342-7. 47. Hayes MG, Urbanek M, Ehrmann DA, Armstrong LL, Lee JY, Sisk R, et al. Genome-wide association of polycystic ovary syndrome implicates alterations in gonadotropin secretion in European ancestry populations. Nat Commun. 2015;6:7502. 48. Day FR, Hinds DA, Tung JY, Stolk L, Styrkarsdottir U, Saxena R, et al. Causal mechanisms and balancing selection inferred from genetic associations with polycystic ovary syndrome. Nat Commun. 2015;6:8464. 49. McAllister JM, Modi B, Miller BA, Biegler J, Bruggeman R, Legro RS, et al. Overexpression of a DENND1A isoform produces a polycystic ovary syndrome theca phenotype. Proc Natl Acad Sci U S A. 2014;111(15):E1519-27. 50. McAllister JM, Legro RS, Modi BP, Strauss JF. Functional genomics of PCOS: from GWAS to molecular mechanisms. Trends Endocrinol Metab. 2015;26(3):118-24. 51. Toledo SP, Brunner HG, Kraaij R, Post M, Dahia PL, Hayashida CY, et al. An inactivating mutation of the luteinizing hormone receptor causes amenorrhea in a 46,XX female. J Clin Endocrinol Metab. 1996;81(11):3850-4. 52. Latronico AC, Chai Y, Arnhold IJ, Liu X, Mendonca BB, Segaloff DL. A homozygous microdeletion in helix 7 of the luteinizing hormone receptor associated with familial testicular and ovarian resistance is due to both decreased cell surface expression and impaired effector activation by the cell surface receptor. Mol Endocrinol. 1998;12(3):442-50. 53. Latronico AC, Lins TS, Brito VN, Arnhold IJ, Mendonca BB. The effect of distinct activating mutations of the luteinizing hormone receptor gene on the pituitary-gonadal axis in both sexes. Clin Endocrinol (Oxf). 2000;53(5):609-13. 54. Overbeek A, Kuijper EA, Hendriks ML, Blankenstein MA, Ketel IJ, Twisk JW, et al. Clomiphene citrate resistance in relation to follicle-stimulating hormone receptor Ser680Ser-polymorphism in polycystic ovary syndrome. Hum Reprod. 2009;24(8):2007-13. 55. Moller DE, Yokota A, White MF, Pazianos AG, Flier JS. A naturally occurring mutation of insulin receptor alanine 1134 impairs tyrosine kinase function and is associated with dominantly inherited insulin resistance. J Biol Chem. 1990;265(25):14979-85. 56. Voight BF, Scott LJ, Steinthorsdottir V, Morris AP, Dina C, Welch RP, et al. Twelve type 2 diabetes susceptibility loci identified through large-scale association analysis. Nat Genet. 2010;42(7):579-89. Arch Endocrinol Metab. 2018;62/3

An update of genetic basis of PCOS pathogenesis

57. Barrett JC, Clayton DG, Concannon P, Akolkar B, Cooper JD, Erlich HA, et al. Genome-wide association study and meta-analysis find that over 40 loci affect risk of type 1 diabetes. Nat Genet. 2009;41(6):703-7. 58. Li T, Zhao H, Zhao X, Zhang B, Cui L, Shi Y, et al. Identification of YAP1 as a novel susceptibility gene for polycystic ovary syndrome. J Med Genet. 2012;49(4):254-7.

61. de Bruin C, Dauber A. Insights from exome sequencing for endocrine disorders. Nat Rev Endocrinol. 2015;11(8):455-64. 62. Gorsic LK, Kosova G, Werstein B, Sisk R, Legro RS, Hayes MG, et al. Pathogenic Anti-MĂźllerian Hormone Variants in Polycystic Ovary Syndrome. J Clin Endocrinol Metab. 2017;102(8):2862-72.

59. Manolio TA, Collins FS, Cox NJ, Goldstein DB, Hindorff LA, Hunter DJ, et al. Finding the missing heritability of complex diseases. Nature. 2009;461(7265):747-53.

60. Jones MR, Goodarzi MO. Genetic determinants of polycystic ovary syndrome: progress and future directions. Fertil Steril. 2016;106(1):25-32.

Arch Endocrinol Metab. 2018;62/3

361

brief report

Serum TSH level stability after 5 years in euthyroid adults at low risk for thyroid dysfunction Pedro Weslley Rosario1, Maria Regina Calsolari1

Programa de Pós-Graduação e Serviço de Endocrinologia, Santa Casa de Belo Horizonte, Belo Horizonte, MG, Brasil 1

Correspondence to: Pedro Weslley Rosario Instituto de Ensino e Pesquisa, Santa Casa de Belo Horizonte. Rua Domingos Vieira, 590 30150-240 – Belo Horizonte, MG, Brasil pedrowsrosario@gmail.com Received on Set/09/2016 Accepted on Apr/25/2017 DOI: 10.20945/2359-3997000000037

ABSTRACT Objective: To evaluate changes in thyroid function after 5 years, the interval proposed for new assessment, in initially euthyroid adults. Subjects and methods: Initially, 1,426 apparently healthy adults considered low risk for thyroid dysfunction, were evaluated by measurement of TSH. After 5 years, 1,215 (85.2%) subjects were reevaluated. Results: After 5 years, four subjects were receiving levothyroxine (L-T4) replacement therapy and 25 others had TSH > 4 mIU/L, only two of them with TSH > 10 mIU/L. All of these subjects had TSH > 3 mIU/L in the initial evaluation. During reassessment, none of the subjects had been or was treated for hyperthyroidism and 22 had TSH < 0.4 mIU/L (none of them < 0.1 mIU/L). Nineteen of these subjects had TSH ≤ 0.6 mIU/L in the initial evaluation. Among the 1,098 subjects with TSH between 0.6 and 3 mIU/L in the initial evaluation, reassessment showed that none of the subjects was using L-T4; only three had TSH > 4 mIU/L (none of them > 10 mIU/L); none had been or was treated for hyperthyroidism, and only three had TSH < 0.4 mIU/L (none of them < 0.1 mIU/L). These results did not differ between men and women or between subjects ≤ 60 and > 60 years. Conclusion: RepeatTSH measurement within an interval of only 5 years would not be cost-effective in adults without known thyroid disease or risk factors for dysfunction who exhibit TSH between 0.6 and 3 mIU/L. Arch Endocrinol Metab. 2018;62(3):362-5 Keywords Serum TSH; adults; screening; repetition; 5 years

INTRODUCTION

lthough no consensus exists regarding age, some societies including the American Thyroid Association (1) and Brazilian Society of Endocrinology and Metabolism [SBEM in the Portuguese acronym] (2) recommend screening for thyroid dysfunction by serum TSH measurement in asymptomatic adults (1,2). In the case of individuals that exhibit normal TSH concentrations in the initial evaluation, a new measurement is recommended after 5 years (1-3). This interval was defined more because it is the recommended interval for periodic assessments of general health rather than resulting from a specific analysis of changes in thyroid function (1). We previously evaluated thyroid function in healthy adults considered low risk for thyroid dysfunction (4,5). These adults were reevaluated after 5 years, the interval proposed for repetition in initially euthyroid adults (13), in order to verify changes in thyroid function in these subjects.

population. The initial results of those studies were published previously (4,5). Initially, 2,532 apparently healthy adults (age > 18 years), not including pregnant women, were interviewed (4,5). Of these, 2,327 subjects agreed to participate in the study. Nine hundred and one subjects were excluded because of high-risk factors for thyroid disease or conditions potentially interfering with thyroid function detected upon clinical evaluation (Table 1) (4,5). The final sample consisted of 1,426 apparently healthy adults considered low risk for Table 1. Inclusion criteria Absence of thyroid disease, no current or previous treatment with antithyroid drugs or L-T4, no history of 131I therapy or thyroidectomy No use of potentially interfering medications such as dopaminergic agonists or antagonists, neuroleptics, corticosteroids, estrogen, amiodarone, interferon, lithium, anticonvulsants, metformin, octreotide; or recent (in the past 8 weeks) exposure to iodinated contrast agents No history of head and neck external radiotherapy Absence of type 1 diabetes or other autoimmune diseases No family history of thyroid disease

SUBJECTS AND METHODS

Absence of goiter or any palpable thyroid anomaly

This study was a reevaluation of two previous studies that evaluated the TSH reference values for Brazilian

Absence of ophthalmopathy

362

Normal free T4 Arch Endocrinol Metab. 2018;62/3

Serum TSH after 5 years

thyroid dysfunction without known interfering factors (4,5). TSH was obtained in the initial evaluation. After 5 years, these subjects were invited for a new clinical evaluation and TSH measurement. Of the 1,426 subjects included, 1,215 (85.2%) were reevaluated (605 men and 610 women, 915 with age ≤ 60 years and 300 with age > 60 years). Reassessment was not possible in the remaining subjects because of failure to contact them, change of city, and refusal to participate. The study was prospective and was approved by the local Research Ethics Committee. Serum samples were obtained from the subjects in the morning (at about 8 a.m.) after an 8- to 10-h fast. TSH was measured with a chemiluminescent assay (Immulite 2000, Diagnostic Products Corporation, Los Angeles, CA), with functional sensitivity of 0.02 mIU/L, and intra- and interassay coefficients of variation < 7% for values ranging from 0.1 to 40 mIU/L. The Fisher’s exact test was used for statistical analysis (comparison between the two groups). A P-value less than 0.05 was considered significant.

0.4 IUI/L (none of them with TSH < 0.1 mIU/L). It is worth noting that 19 subjects with TSH < 0.4 mIU/L had levels ≤ 0.6 mIU/L in the initial evaluation. Among the 1,098 subjects with TSH between 0.6 and 3 mIU/L in the initial evaluation, reassessment revealed that none of them was receiving L-T4; only three had TSH > 4 mIU/L (none of them with TSH > 10 mIU/L); none of the subjects had been or was treated for hyperthyroidism, and only three had TSH < 0.4 mIU/L (none of them with TSH < 0.1 mIU/L). Among the subjects with initial TSH between 0.6 and 3 mIU/L, the proportions of subjects with TSH of 0.4-4 mIU/L, < 0.4 mIU/L or > 4 mIU/L after 5 years were the same in men (n = 546: 0.18%, 99.45%, and 0.37%, respectively) and women (n = 552: 0.36%, 99.46%, and 0.18%, respectively) and in subjects £ 60 years (n = 822: 0.24%, 99.51%, and 0.24%, respectively) and > 60 years (n = 276: 0.36%, 99.28%, and 0.36%, respectively). The results of reassessment after 5 years according to initial TSH are shown in Table 2.

DISCUSSION

RESULTS After 5 years, four subjects were receiving levothyroxine (L-T4) replacement therapy initiated because of confirmed subclinical hypothyroidism and 25 others had TSH > 4 mIU/L, only two of them with TSH > 10 mIU/L. It is worth noting that all subjects receiving L-T4 therapy and 22 subjects with TSH > 4 mIU/L (including the two with TSH > 10 mIU/L) had TSH > 3 mIU/L in the initial evaluation. During reassessment, none of the subjects had been or was treated for hyperthyroidism and 22 had TSH <

The results of the present study confirm that in subjects without known thyroid disease, but with a slightly reduced TSH measurement (0.2-0.4 mIU/L), spontaneous TSH normalization in the subsequent measurement is common (6). Even if normalization does not occur, persistence of the initial finding is more common than progression to TSH < 0.1 mIU/L or clinical hyperthyroidism (6,7). The same applies to adults with slightly elevated TSH (4-6 mIU/L) (6,8,9), with the observation of TSH normalization in approximately 40% and persistence of slightly elevated

Table 2. Results of reassessment after 5 years according to initial serum TSH Reassessment after 5 years Treatment for hyperthyroidism

TSH < 0.1 mIU/L

TSH 0.1-0.4 mIU/L

TSH 0.4-4 mIU/L

TSH 4-10 mIU/L

TSH > 10 mIU/L

Under L-T4 therapy

< 0.4 (0.2-0.38) (n = 25)

13 (52%)

12 (48%)

0.4-0.6 (n = 25)

6 (24%)

19 (76%)

0.6-3 (n = 1,098)

3 (0.27%)

1092 (99.45%)

3 (0.27%)

3-4 (n = 38)

30 (79%)

5 (13%)

1 (2.6%)

2 (5.2%)

> 4* (up to 5.84) (n = 29)

11 (37.9%)

15 (51.7%)

1 (3.4%)

2 (6.9%)

* None of the 24 patients without TPOAb in the initial assessment was treated with L-T4, 11 had normal TSH, and 13 continued to have elevated TSH but < 10 mIU/L. Among the 5 patients with TSH > 4 mIU/L and positive TPOAb in the initial assessment, 2 had elevated TSH < 10 mIU/L, one had TSH > 10 mIU/L, and 2 were taking L-T4. Arch Endocrinol Metab. 2018;62/3

363

Initial TSH (mIU/L)

Serum TSH after 5 years

TSH in 50%, while progression to TSH > 10 mIU/L or need for L-T4 after 5 years is observed in only 10%. The consensual recommendation is to repeat the test in these individuals with slightly altered TSH before any treatment decision (10-12). Another implication of this observation is that in population studies evaluating the association between thyroid function and certain outcome(s), but obtaining only a single TSH measurement, a large proportion of the subjects with slightly altered TSH may, in fact, not have true thyroid dysfunction. The most relevant finding of the study was that, among subjects without known thyroid disease or risk factors for thyroid dysfunction and with initially normal TSH (0.4-4 mIU/L), none had TSH < 0.1 mIU/L and only 0.25% had TSH > 10 mIU/L after 5 years. Most subjects who progressed to reduced (but > 0.1 mIU/L) or elevated TSH (but < 10 mIU/L) had initial TSH < 0.6 and > 3 mIU/L, respectively. With initial TSH between 0.6 and 3 mIU/L, 99.5% of the subjects remained euthyroid after 5 years. Thus, repetition of TSH in this subgroup detected alterations in only one of every 200 retested subjects and even these few individuals with altered TSH did not require treatment. A known previous study demonstrated that 98% of individuals with initially normal TSH continued to present normal TSH when retested within an interval of 5 years (6). However, important limitations of that study were (i) its retrospective design, (ii) only patients with known thyroid disease were excluded but not individuals at high risk for dysfunction, and (iii) the second TSH was obtained on average 19 months after the first measurement (6). Furthermore, that study was conducted exclusively in the Israeli population (6). We highlight some characteristics of our study. This was a prospective study and the reassessment rate was high (85%). The interval to reassessment was exactly that recommended by some societies (1-3), including SBEM (2). Because this is a current recommendation in Brazil (2), we believe it is important to conduct a study in our country that evaluates this interval for TSH repetition. To our knowledge, no such study exists. On the other hand, it is important to note that the results reported here apply to individuals without known thyroid disease or without the risk factors for dysfunction listed in Table 1. Although measurement of antithyroid peroxidase antibodies (TPOAb) or ultrasonography (US) is not indicated in these individuals, the results 364

could be different if TPOAb are positive or US is altered. The great stability in thyroid function observed after 5 years was limited to subjects with initial TSH between 0.6 and 3 mIU/L. Finally, the results permit us to conclude that the recommended interval of 5 years (1-3) seems to be too short to reevaluate thyroid function, but the appropriate interval cannot be defined and it cannot be stated whether repetition would only be indicated if any clinical suspicion appears. In conclusion, in this prospective Brazilian study performed in adults without known thyroid disease or risk factors for thyroid dysfunction (Table 1) who exhibit TSH between 0.6 and 3 mIU/L, TSH repetition within the recommended interval of 5 years (1-3) was not cost-effective. Two hundred reassessments were necessary to detect a single case of altered TSH, which even did not result in immediate treatment. Disclosure: This work was supported by the Brazilian National Council for Scientific and Technological Development (CNPq) through a productivity fellowship granted to P.W.R. The initial evaluation of one of the studies (4) was partly supported by Sanofi-Aventis.

REFERENCES 1. Ladenson PW, Singer PA, Ain KB, Bagchi N, Bigos ST, Levy EG, et al. American Thyroid Association guidelines for detection of thyroid dysfunction. Arch Intern Med. 2000;160:1573-5. 2. Carvalho GA, Perez CL, Ward LS. The clinical use of thyroid function tests. Arq Bras Endocrinol Metabol. 2013;57:193-204. 3. Glenn GC, Laboratory Testing Task Force of the College of American Pathologists. Practice parameter on laboratory panel testing for screening and case finding in asymptomatic adults. Arch Pathol Lab Med. 1996;120:929-43. 4. Rosario PW, Xavier AC, Calsolari MR. TSH reference values for adult Brazilian population. Arq Bras Endocrinol Metabol. 2010;54:603-6. 5. Rosario PW, Calsolari MR. TSH reference range in older adults: a Brazilian study. Arq Bras Endocrinol Metabol. 2014;58:389-93. 6. Meyerovitch J, Rotman-Pikielny P, Sherf M, Battat E, Levy Y, Surks MI. Serum thyrotropin measurements in the community: fiveyear follow-up in a large network of primary care physicians. Arch Intern Med. 2007;167:1533-8. 7. Rosario PW. Natural history of subclinical hyperthyroidism in elderly patients with TSH between 0.1 and 0.4 mIU/l: a prospective study. Clin Endocrinol (Oxf). 2010;72:685-8. 8. Huber G, Staub JJ, Meier C, Mitrache C, Guglielmetti M, Huber P, et al. Prospective study of the spontaneous course of subclinical hypothyroidism: prognostic value of thyrotropin, thyroid reserve, and thyroid antibodies. J Clin Endocrinol Metab. 2002;87:3221-6. 9. Rosario PW, Carvalho M, Calsolari MR. Natural history of subclinical hypothyroidism with TSH â&#x2030;¤ 10 mIU/l: a prospective study. Clin Endocrinol (Oxf). 2016;84:878-81. 10. Sgarbi JA, Teixeira PF, Maciel LM, Mazeto GM, Vaisman M, Montenegro Junior RM, et al.; Brazilian Society of Endocrinology and Metabolism. The Brazilian consensus for the clinical approach and treatment of subclinical hypothyroidism in adults: recomArch Endocrinol Metab. 2018;62/3

Serum TSH after 5 years

mendations of the thyroid Department of the Brazilian Society of Endocrinology and Metabolism. Arq Bras Endocrinol Metabol. 2013;57:166-83.

12. Ross DS, Burch HB, Cooper DS, Greenlee MC, Laurberg P, Maia AL, et al. 2016 American Thyroid Association Guidelines for Diagnosis and Management of Hyperthyroidism and Other Causes of Thyrotoxicosis. Thyroid 2016;26:1343-421.

11. Brenta G, Vaisman M, Sgarbi JA, Bergoglio LM, Andrada NC, Bravo PP, et al.; Task Force on Hypothyroidism of the Latin American Thyroid Society (LATS). Clinical practice guidelines for the

management of hypothyroidism. Arq Bras Endocrinol Metabol. 2013;57:265-91.

Arch Endocrinol Metab. 2018;62/3

365

brief report

Graves’ ophthalmopathy: low-dose dexamethasone reduces retinoic acid receptor-alpha gene expression in orbital fibroblasts Divisão de Endocrinologia, Departamento de Medicina Interna, Faculdade de Medicina de Botucatu, Universidade Estadual de São Paulo (Unesp), Botucatu, SP, Brasil 2 Instituto Federal de Educação, Ciência e Tecnologia do Estado de São Paulo (IFSP), São Roque, SP, Brasil 3 Departamento de Oftalmologia, Otorrinolaringologia e Cirurgia de Cabeça e Pescoço, Faculdade de Medicina de Botucatu, Universidade Estadual de São Paulo (Unesp), Botucatu, SP, Brasil 1

Correspondence to: Gláucia Maria Ferreira da Silva Mazeto Departamento de Medicina Interna, Universidade Estadual de São Paulo 18618-970 – Botucatu, SP, Brasil gmazeto@fmb.unesp.br Received on Aug/31/2017 Accepted on Jan/23/2018 DOI: 10.20945/2359-3997000000044

Sarah Santiloni Cury1, Miriane Oliveira1, Maria Teresa Síbio1, Sueli Clara1, Renata De Azevedo Melo Luvizotto1, Sandro Conde2, Edson Nacib Jorge3, Vania dos Santos Nunes1, Célia Regina Nogueira1, Gláucia Maria Ferreira da Silva Mazeto1

ABSTRACT Objective: Graves’ ophthalmopathy (GO) is an autoimmune disease that leads to ocular proptosis caused by fat accumulation and inflammation, and the main treatment is corticosteroid therapy. Retinoid acid receptor-alpha (RARα) seems to be associated with inflammation and adipocyte differentiation. This study aimed to assess the effect of glucocorticoid treatment on orbital fibroblasts of GO patient treated or not with different glucocorticoid doses. Materials and methods: Orbital fibroblasts collected during orbital decompression of a female patient with moderately severe/severe GO were cultivated and treated with 10 nM and 100 nM dexamethasone (Dex). RARα gene expression in the treated and untreated cells was then compared. Results: Fibroblast RARα expression was not affected by 100 nM Dex. On the other hand, RARα expression was 24% lower in cells treated with 10 nM Dex (p < 0.05). Conclusions: Orbital fibroblasts from a GO patient expressed the RARα gene, which was unaffected by higher, but decreased with lower doses of glucocorticoid. Arch Endocrinol Metab. 2018;62(3):366-9 Keywords Receptors; retinoic acid; gene expression; Graves ophthalmopathy

INTRODUCTION

raves’ ophthalmopathy (GO) is characterized by expansion of the orbital tissue due to immunemediated inflammation and adipocyte proliferation secondary to orbital fibroblast (OF) differentiation (1). The main treatment for moderate to severe cases is corticosteroid therapy, which carries some risks (2,3), while other therapeutic approaches have provided with limited results in a reasonable number of patients (4). Thus, new GO treatment options are being sought. Retinoic acid (RA) is the biologically active form of vitamin A. It is widely used in clinical practice, especially in the form of isotretinoin for the treatment of acne (5). Nonetheless, studies indicate that RA may also play a role in adipogenesis (6,7) and inflammatory/ autoimmune processes (8). The effects of RA depend on nuclear receptors from two families of nuclear 366

transcription regulators: RA receptor (RAR) and retinoid X receptor (RXR) (9,10). RAR, but not RXR, has affinity for all forms of RA (11). RARα stands out as it is the most frequently expressed RAR in cells (12). The OF of GO patients express RA receptors, including RARα, and retinoids could inhibit adipocyte growth and differentiation, and induce apoptosis in these cells, consequently presenting therapeutic potential in GO (13,14). However, the development of severe GO has been associated with the simultaneous use of RA, recombinant thyroidstimulating hormone (TSH), and radioactive iodine (15). Hence, the use of RA in GO may or may not have satisfactory results, depending on patient clinical condition, associated treatments, and RAR existence/ functionality/response. Other studies on the use of RA or the expression of its nuclear receptors in GO have not been found. Arch Endocrinol Metab. 2018;62/3

RARα in Graves’ ophthalmopathy

The present study, approved by the Research Ethics Committee of Botucatu Medical School under process number 4037-2011), assessed RARα gene expression in cultured OF treated or not with 10 nM or 100 nM dexamethasone (Dex). The cells were initially obtained during orbital decompression performed at the Clinics Hospital (HC) of the Botucatu Medical School (FMB) – Unesp, placed in a falcon tube containing the medium 199 (LCG®) and antibiotic, and transported to the experimental laboratory of medical clinic, where the cells were cultivated, as previously described (16,17). Once the fibroblasts in the wells reached 80% confluence, they were treated with 10 nM or 100 nM Dex in biological triplicates, and their RARα gene expression was compared with that of untreated fibroblasts (control group, C) (16,17). After treatment, the culture medium 199 was removed, and cells were collected from the plates with 400 µL of TRIzol® (Life Tech, USA) for RNA extraction, also perfomed with TRIzol®. Sample integrity was verified by running them for 30 minutes in 1% agarose gel at a voltage of 80 mV. The RNA samples were then analyzed by spectrophotometry. Samples with 260:280 ratios below 1.6 were discarded because of protein contamination. Complementary deoxyribonucleic acid (cDNA) was synthesized from 1 µg of RNA by reverse transcription using the High-capacity cDNA Reverse Transcription Kit (Applied Biosystems, CA, USA). In short, the following were added to the RNA sample: 2 µL of 10x buffer, 0.8 µL of 100 mM dNTP mix, 2 µL of random primer, 1 µL of RNase inhibitor, 1 µL of reverse transcriptase, and 12.2 µL of nucleasefree water. Samples were incubated at 25ºC for 10 minutes, at 37ºC for 120 minutes, at 85ºC for 5 seconds, and maintained at 4ºC. Each cDNA sample Arch Endocrinol Metab. 2018;62/3

RESULTS Orbital fibroblasts expressed the RARα gene. Gene expression in cells treated with 10 nM Dex was 24% lower than the control group (p < 0.05). However, gene expression in cells treated with 100 nM Dex was not significantly different from controls or the 10 nM Dex group (Figure 1). ns 1,20

p < 0,05 ns

1,00 0,80 0,60 0,40 0,20 0,00

Without treatment

10 nM dexamethasone

100 nM dexamethasone

Figure 1. Effect of 10 nM and 100 nM dexamethasone on retinoic acid receptor alpha (RARα) mRNA in orbital fibroblasts from a patient with Graves’ ophthalmopathy. The experiment was performed in triplicate. Data were reported as mean and standard deviation. ANOVA was used in conjunction with Tukey’s test (p < 0.05). ns: not significant. 367

MATERIALS AND METHODS

was analyzed by TaqMan Assays (Life Technologies, USA) containing specific primers (ESR1: estrogen receptor alpha and cyclophilin: internal control) for the target RNAs. Each reaction used 10 μL of TaqMan® Universal PCR Master Mix (Life Technologies, USA) and 3 μL of the reverse transcription product. The final volume was adjusted to 20 μL with nuclease-free water. Quantitative reverse transcription polymerase chain reaction (RT-qPCR) was performed as instructed by the Minimum Information for Publication of Quantitative Real-Time PCR Experiment (MIQE) (18). Samples were normalized by the internal control (cyclophilin), and gene expression was quantified by the 2-DDCT method (19). The control group was adjusted for comparison between groups (16,17). All data were expressed as mean ± standard deviation. Intragroup treatments were compared by two-factor analysis of variance (ANOVA) complemented by Tukey’s multiple comparison test. The significance level was set at 5% (p < 0.05).

RARα gene expression (relative value)

We recently reported the case of a 62-year-old female patient with inactive, moderately severe to severe GO, not previously treated with glucocorticoid, whose fibroblasts expressed the peroxisome proliferatoractivated receptor-gamma (PPARg) and estrogen receptor-alpha (ERα) genes. Interestingly, this gene expression responded to glucocorticoids in a doserelated manner (16,17). Thus, the aim of this study was to evaluate the RARα gene expression by cultured OF from GO patient, which were treated or not with different glucocorticoid doses.

RARα in Graves’ ophthalmopathy

DISCUSSION This study confirmed that the OF of GO patients expressed the RARα gene. Lower doses of glucocorticoid reduced RARα gene expression while higher doses did not change it significantly. RA plays an important role in adipogenesis. Low RA concentrations promote adipocyte differentiation (7), but supraphysiological concentrations inhibit it (6). These findings indicate the importance of studying RA receptors in disorders where adipocytes differentiate and proliferate as a pathophysiological mechanism. Normal human fibroblasts from the lungs and skin express RARα, and its expression is modulated in certain situations, such as types of cell growth conditions, and treatment with ascorbic acid, for example (20,21). Likewise, the OF of patients with GO express RARα, which respond differently, depending on the concomitant medication used (13,15). An impact of RAR in GO physiopathology could be its role in adipogenesis. Wang and cols. found that mouse adipocytes treated with all-trans-retinoic acid (ATRA) had higher RAR expression, RARg more so than RARα, with consequent decrease of PPARg expression (14). It is known that PPARg plays a role in GO development, and recently, we reported that OF from a GO patient treated with small doses of corticosteroid had higher PPARg gene expression. This could lead to higher differentiation of fibroblasts into adipocytes and speculation that higher glucocorticoid doses would better treat this disease (16). Yet, in addition to adipogenesis, inflammation also has an important role in GO pathophysiology (1). Thus, vitamin A could act on immunity and inflammation. In autoimmune diseases, RA helps T-cell induction and gene regulation via RAR, which behaves as a transcription factor (8), making RAR essential for preventing and maintaining the tolerance to autoimmune/inflammatory diseases. In light of this, RA treatment could drive distinct pathogenesis evolution by its effects through RAR-modulated pathways in OF, mainly by changing cellular proliferation, differentiation, and apoptosis (13). Indeed, OF from GO patients treated with RA presented morphological alterations, and a decrease in cell growth/proliferation accompanied by the expression of RAR subtypes (13). Moreover, RARα agonist inhibited 36% of the TGF-β1 stimulatory effect on extracellular matrix remodeling and human Tenon fibroblast contractility (22). 368

Despite this, there is a lack of knowledge regarding the molecular mechanisms of RAR activation in GO. In addition, the OF from GO patients have a hyperresponsive phenotype and consist of several cell subsets, such as Thy1+/Thy1-, fibrocytes/CD34+ (23). Perhaps RAR stimulation in different types of fibroblasts could result in different responses, aimed at adipogenesis or fibrogenesis. Our results show that different doses of glucocorticoid, the main treatment for moderate to severe cases of GO, can lead to a differential RARα expression. This finding suggests that corticotherapy could alter the OF phenotype for RAR, and adds another variable to the pathophysiology of OG since it could result in a different response from these receptors to eventual stimuli. To summarize, in the GO treatment, the appropriate RAR expression, modulated by glucocorticoid therapy, would help to reduce adipocyte proliferation and inflammation. As the smaller Dex dose reduced RARα expression, and the higher dose did not affect it, as observed with PPARg (16), one could speculate that lower glucocorticoid doses would have lesser effects, both anti-inflammatory and in inhibiting adipocyte differentiation. Considering the eventual therapeutic role of retinoids in GO, and the dynamic and sensitive mechanism that modifies RARα mRNA levels in fibroblasts, higher doses of glucocorticoids would be more appropriate if combined with RA. This study has some limitations, such as using OF from a single patient and not investigating other RA receptors or their protein expression. However, to our knowledge, this was the first study assessing RARα gene expression by the OF from a GO patient treated with different glucocorticoid doses, given that glucocorticoids are currently the main treatment for this ocular disorder. In conclusion, the orbital fibroblasts from a GO patient expressed the RARα gene, which was not affected by higher, but decreased with lower, glucocorticoid doses. Other studies with more patients are needed to confirm these results. Acknowledgment: the authors are grateful to the Research Support Foundation of the State of Sao Paulo (Fapesp) for financial support (grant #:03/03651-0 and 2011/18464-8). Disclosure: no potential conflict of interest relevant to this article was reported. Arch Endocrinol Metab. 2018;62/3

RARα in Graves’ ophthalmopathy

REFERENCES 1. Iyer S, Bahn R. Immunopathogenesis of Graves’ ophthalmopathy: the role of the TSH receptor. Best Pract Res Clin Endocrinol Metab. 2012;26(3):281-9. 2. Bahn R. High-dose intravenous glucocorticoid therapy for Graves’ ophthalmopathy: where are we now? Thyroid. 2012;22(1):1-2. 3. Philip R, Saran S, Gutch M, Agroyia P, Tyagi R, Gupta K. Pulse dexamethasone therapy versus pulse methylprednisolone therapy for treatment of Graves’s ophthalmopathy. Indian J Endocrinol Metab. 2013;17(Suppl 1):S157-9. 4. Bartalena L, Baldeschi L, Boboridis K, Eckstein A, Kahaly GJ, Marcocci C, et al.; European Group on Graves’ Orbitopathy (EUGOGO). The 2016 European Thyroid Association/European Group on Graves’ Orbitopathy Guidelines for the Management of Graves’ Orbitopathy. Eur Thyroid J. 2016;5(1):9-26. 5. Owen CE. Treating acne with high-dose isotretinoin. JAMA. 2014;311(20):2121-2. 6. Sato M, Hiragun A, Mitsui H. Preadipocytes possess cellular retinoid binding proteins and their differentiation is inhibited by retinoids. Biochem Biophys Res Commun. 1980;95(4):1839-45. 7. Safonova I, Darimont C, Amri EZ, Grimaldi P, Ailhaud G, Reichert U, et al. Retinoids are positive effectors of adipose cell differentiation. Mol Cell Endocrinol. 1994;104(2):201-11. 8. Kim CH. Retinoic acid, immunity, and inflammation. Vitam Horm. 2011;86:83-101. 9. Evans RM.The steroid and thyroid hormone receptor superfamily. Science. 1988;240(4854):889-95. 10. Green S, Chambon P. Nuclear receptors enhance our understanding of transcription regulation. Trends Genet. 1988;4(11):309-14. 11. Mangelsdorf DJ, Ong ES, Dyck JA, Evans RM. Nuclear receptor that identifies a novel retinoic acid response pathway. Nature. 1990;345(6272):224-9. 12. Elder JT, Fisher GJ, Zhang QY, Eisen D, Krust A, Kastner P, et al. Retinoic acid receptor gene expression in human skin. J Invest Dermatol. 1991;96(4):425-33.

14. Wang X, Yang P, Liu J, Wu H, Yu W, Zhang T, et al. RARγ-C-FosPPARγ2 signaling rather than ROS generation is critical for alltrans retinoic acid-inhibited adipocyte differentiation. Biochimie. 2014;106:121-30. 15. Berg G, Andersson T, Sjödell L, Jansson S, Nyström E. Development of severe thyroid-associated ophthalmopathy in a patient with disseminated thyroid cancer treated with recombinant human thyrotropin/radioiodine and retinoic acid. Thyroid. 2005;15(12):1389-94. 16. Cury SS, Oliveira M, Síbio MT, Clara S, Luvizotto RD, Conde S, et al. Graves ophthalmopathy: low-dose glucocorticoid increases peroxisome proliferator-activated receptor-gamma gene expression. Arq Bras Oftalmol. 2014;77(5):339-340. 17. Cury SS, Oliveira M, Síbio MT, Clara S, Luvizotto Rde A, Conde S, et al. Gene expression of estrogen receptor-alpha in orbital fibroblasts in Graves’ ophthalmopathy. Arch Endocrinol Metab. 2015;59(3):273-6. 18. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009;55(4):611-22. 19. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402-8. 20. Jaworski J, Klapperich CM. Fibroblast remodeling activity at twoand three-dimensional collagen-glycosaminoglycan interfaces. Biomaterials. 2006;27(23):4212-20. 21. Duarte TL, Cooke MS, Jones GD. Gene expression profiling reveals new protective roles for vitamin C in human skin cells. Free Radic Biol Med. 2009;46(1):78-87. 22. Liu Y, Kimura K, Orita T, Suzuki K, Teranishi S, Mori T, Sonoda K. Inhibition by a retinoic acid receptor γ agonist of extracellular matrix remodeling mediated by human Tenon fibroblasts. Mol Vis. 2015;21:1368-77. 23. Dik WA, Virakul S, van Steensel L. Current perspectives on the role of orbital fibroblasts in the pathogenesis of Graves’ ophthalmopathy. Exp Eye Res. 2016;142:83-91.

13. Pasquali D, Bellastella A, Colantuoni V, Vassallo P, Bonavolontà G, Rossi V, et al. All-trans retinoic acid- and N-(4-hydroxyphenil)-

retinamide-induced growth arrest and apoptosis in orbital fibroblasts in Graves’ disease. Metabolism. 2003;52(11):1387-92.

Arch Endocrinol Metab. 2018;62/3

369

case report

Endocrinologia, Instituto do Câncer do Estado de São Paulo, Faculdade de Medicina da Universidade de São Paulo (FMUSP), São Paulo, SP, Brasil 2 Laboratório de Endocrinologia Celular e Molecular (LIM25), Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo (HCFMUSP), São Paulo, SP, Brasil 3 Oncologia Clínica, Instituto do Câncer do Estado de São Paulo (ICESP), Faculdade de Medicina da Universidade de São Paulo (FMUSP), São Paulo, SP, Brasil 4 Cirurgia de Cabeça e Pescoço, Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo (HCFMUSP), São Paulo, SP, Brasil 5 Cirurgia de Cabeça e Pescoço, Instituto do Câncer do Estado de São Paulo (ICESP), Faculdade de Medicina da Universidade de São Paulo (FMUSP), São Paulo, SP, Brasil 6 Radiologia, Instituto do Câncer do Estado de São Paulo (ICESP), Faculdade de Medicina da Universidade de São Paulo (FMUSP), São Paulo, SP, Brasil 7 Medicina Nuclear, Instituto do Câncer do Estado de São Paulo (ICESP), Faculdade de Medicina da Universidade de São Paulo (FMUSP), São Paulo, SP, Brasil 1

Correspondence to: Debora Lucia Seguro Danilovic Laboratório de Endocrinologia Celular e Molecular (LIM 25), Faculdade de Medicina da Universidade de São Paulo Av. Dr. Arnaldo, 455, 4° andar, sala 4305 01246-903 – São Paulo, SP, Brasil deboradanilovic@usp.br

Potential role of sorafenib as neoadjuvant therapy in unresectable papillary thyroid cancer Debora L. S. Danilovic1,2, Gilberto Castro Jr.3, Felipe S. R. Roitberg3, Felipe A. B. Vanderlei4, Fernanda A. Bonani5, Ricardo M. C. Freitas6, George B. Coura-Filho7, Rosalinda Y. Camargo2, Marco A. Kulcsar5, Suemi Marui2, Ana O. Hoff1

SUMMARY Total thyroidectomy, radioiodine (RAI) therapy, and TSH suppression are the mainstay treatment for differentiated thyroid carcinomas (DTCs). Treatments for metastatic disease include surgery, external-beam radiotherapy, RAI, and kinase inhibitors for progressive iodine-refractory disease. Unresectable locoregional disease remains a challenge, as standard therapy with RAI becomes unfeasible. We report a case of a young patient who presented with unresectable papillary thyroid carcinoma (PTC), and treatment with sorafenib allowed total thyroidectomy and RAI therapy. A 20-year-old male presented with severe respiratory distress due to an enlarging cervical mass. Imaging studies revealed an enlarged multinodular thyroid gland, extensive cervical adenopathy, severe tracheal stenosis, and pulmonary micronodules. He required an urgent surgical intervention and underwent tracheostomy and partial left neck dissection, as the disease was deemed unresectable; pathology revealed PTC. Treatment with sorafenib was initiated, resulting in significant tumor reduction allowing near total thyroidectomy and bilateral neck dissection. Postoperatively, the patient underwent radiotherapy for residual tracheal lesion, followed by RAI therapy for avid cervical and pulmonary disease. The patient’s disease remains stable 4 years after diagnosis. Sorafenib has been approved for progressive RAI-refractory metastatic DTCs. In this case report, we describe a patient with locally advanced PTC in whom treatment with sorafenib provided sufficient tumor reduction to allow thyroidectomy and RAI therapy, suggesting a potential role of sorafenib as an induction therapy of unresectable DTC. Arch Endocrinol Metab. 2018;62(3):370-5

Received on Jan/11/2017 Accepted on Jan/17/2018 DOI: 10.20945/2359-3997000000046

INTRODUCTION

he treatment of differentiated thyroid carcinomas (DTCs) involves a combination of total thyroidectomy, radioactive iodine (RAI) therapy, and TSH suppression. On the other hand, treatment of metastatic DTC is more complex and includes surgery when feasible, 131I therapy, external beam radiotherapy, and other localized treatment procedures such as

370

thermal ablation or stereotactic radiation. However, a few cases of metastatic carcinomas progress despite RAI treatment, and therapy with multikinase inhibitors (MKIs) becomes the alternative to control systemic disease. The MKIs can inhibit tumor growth and angiogenesis. They block molecular targets involved in the pathogenesis of thyroid cancer, such as BRAF, RET, vascular endothelial growth factor receptors (VEGFRs), Arch Endocrinol Metab. 2018;62/3

platelet-derived growth factor receptors (PDGFRs), and fibroblast growth factor receptors (FGFRs). Due to clinical benefit demonstrated in phase-III clinical trials, the FDA has recently approved the use of sorafenib and lenvatinib in treatment of progressive iodine-refractory metastatic DTC (1,2). Locally aggressive papillary thyroid carcinoma is as an unusual presentation, considering the characteristic slow growth of papillary carcinomas. Near total or total thyroidectomy is advised and necessary for effective RAI therapy; therefore, when invasive neck disease is unresectable, traditional therapy, especially RAI, becomes unfeasible. We present a case of a patient with unresectable papillary thyroid carcinoma at presentation in whom preoperative treatment with sorafenib allowed surgical intervention and RAI therapy. This case report was approved by the local Institutional Review Board.

CASE REPORT A 20-year-old male presented at the emergency room with severe respiratory distress. He had a 4-year history of an enlarging cervical mass and dyspnea for 6 months. Imaging studies revealed a large and nodular heterogeneous thyroid with calcifications causing severe tracheal stenosis and compression of internal jugular veins, in addition to conglomerates of cervical lymph nodes and multiple, up to 1-cm, pulmonary nodules (Figure 1A, Figure 2A). The severe respiratory distress required urgent surgical intervention. During surgery, the right thyroid lobe was completely adhered to the trachea, as were metastatic right neck lymph nodes to the thyroid. Due to extensive bleeding and functional impairment of the right recurrent laryngeal nerve during intraoperative neuromonitoring, total thyroidectomy was abandoned and only left neck dissection and tracheostomy were performed; an intraoperative frozen section biopsy revealed papillary thyroid carcinoma. External beam radiotherapy is usually the first palliative choice for unresectable thyroid carcinoma, especially in older patients, but the presence of bulky disease would not only demand high cumulative doses but also result in high toxicity and even increased risk of secondary malignancies (3). Additionally, RAI therapy would not be feasible without Arch Endocrinol Metab. 2018;62/3

thyroidectomy. Therefore, after multidisciplinary discussion, the recommendation was to proceed with sorafenib treatment. The patient tolerated sorafenib at a dose of 400 mg bid with manageable side effects of fatigue, hand and foot skin reaction, and diarrhea. Reduction of the cervical mass, particularly metastatic lymph nodes, and pulmonary nodules was observed within 6 months of therapy, resulting in removal of the tracheostomy tube. Sorafenib was maintained for 13 months until significant tumoral reduction to surgery (Figure 1B, Figure 2B). Near total thyroidectomy and bilateral neck dissection were performed after discontinuation of sorafenib for 2 weeks with no complications. Pathology revealed a 7.2-cm multifocal, diffuse sclerosing papillary carcinoma with gross extrathyroidal extension, positive margins, and involvement of 13 out of 38 metastatic lymph nodes (pT4pN1bM1, stage II) (4). Genetic analysis of tumoral tissue was negative for the p.V600E BRAF mutation. The patient was placed on TSH suppression therapy; suppressed thyroglobulin was 1.5 ng/mL but with positive anti-thyroglobulin antibody (103 UI/mL). He underwent external beam radiotherapy, 70 Gy delivered in 35 fractions, for control of residual macroscopic tracheal lesion. A whole-body scan (131I-WBS) performed 3 months after radiotherapy revealed an uptake of 5.8% in the cervical area and iodine-avid pulmonary nodules (Figure 3A), which allowed him to be treated with 100 mCi of 131I. Postdose WBS with SPECT/CT confirmed radioiodine avid lesions in the thyroid bed (with persistent but smaller tracheal compression and lumen reduction), lymph nodes of the left retropharyngeal space, right neck level III, and left neck level V and bilateral pulmonary nodules (Figures 3B and 3C). Imaging studies 16 months after RAI therapy revealed a persistent but stable lesion in the thyroid bed, neck metastatic lymph nodes up to 1.4 cm, and stable pulmonary nodules up to 1.1 cm. Suppressed thyroglobulin was undetectable (< 0.2 ng/mL), but levels of anti-thyroglobulin antibody increased (944 UI/mL). He received a second dose of RAI, 200 mCi, confirming persistent radioiodine avid lesions. Patient remains asymptomatic with stable metastatic disease after 52 months of diagnosis (Figure 2C). His suppressed thyroglobulin level is undetectable with decreasing levels of anti-thyroglobulin antibodies (455 UI/mL). 371

Neoadjuvant use of sorafenib in thyroid cancer

Figure 1. (A) Initial CT scans revealing thyroid enlargement with tracheal stenosis and bilateral conglomerates of cervical lymph nodes with compression of adjacent structures. (B) CT scans after 12 months of sorafenib therapy, demonstrating significant reduction of neck disease.

Figure 2. CT imaging demonstrating improvement of tracheal compression and lumen reduction. (A) Initial scan. (B) Scan after 12 months of sorafenib. (C) Last scan after 4 years of diagnosis.

DISCUSSION Extensive invasive thyroid carcinoma is an uncommon initial presentation of DTC (5); however, its aggressive behavior significantly endangers prognosis. Cancer-specific mortality is 3-fold higher 372

in the T4 stage according to the AJCC classification system (6). The 10-year disease-specific survival of thyroid carcinomas with extensive invasion of soft tissues, larynx, trachea, esophagus, or recurrent laryngeal nerve (T4a) reduces from 73% to 7.7% Arch Endocrinol Metab. 2018;62/3

Neoadjuvant use of sorafenib in thyroid cancer

specifically when the tumor invades the prevertebral fascia or mediastinal vessels or encases the carotid artery (T4b) (7). Surgical removal is essential to eradicate the tumor. Residual microscopic disease can be managed with RAI therapy and TSH suppression, but macroscopic persistent disease may require other adjuvant therapy, usually external beam radiotherapy, to improve local control (8). Indication of induction chemotherapy in inoperable DTC is not well established (9). Retrospective nonrandomized studies reported the benefit of cytotoxic agents in locally invasive follicular and papillary carcinomas prior to surgery (10,11). The studies observed tumor reduction after preoperative use of vinblastine alone, vinblastine with doxorubicin, and other regimens, including radiotherapy in 14% and 25% of the cases. Ito and cols. reported significant tumor reduction in 1 out of 2 papillary thyroid carcinomas with a squamous cell carcinoma component treated with paclitaxel before surgery (12). None of the studies reported toxicity. This is the first description of the off-label induction use of sorafenib in locally invasive papillary Arch Endocrinol Metab. 2018;62/3

thyroid carcinoma. The phase-III trial for RAIrefractory differentiated thyroid cancer (DECISION trial) included 7 patients (3.8%) with locally advanced thyroid cancer in the sorafenib arm (1), however, no specific information regarding this group of patients was provided. As inclusion criteria required RAI-refractory disease, one could assume that these patients likely had undergone thyroidectomy prior to a whole-body iodine scan and/or RAI therapy. Therefore, despite having locally advanced disease, they were likely different from our case report. The preoperative use of MKI has been described in other cancers, alone or combined with other chemotherapies (13-15). Sunitinib has been used in an unresectable medullary thyroid cancer (16). Particularly in renal cell carcinomas, the neoadjuvant use of MKI provided 9.6% to 28.3% reduction in renal tumor diameter, and it changed unresectable to resectable tumors in around 20% of cases (13). Preferentially, an induction therapy should have high rates of tumor reduction. Despite a significant tumor shrinkage with sorafenib in our case, the objective response rate in the sorafenib phase-III trial was 373

Figure 3. (A) Diagnostic whole-body scan after 3 months of adjuvant external beam radiotherapy. (B) Post-therapy whole-body scan revealing radioiodineavid cervical lesions and pulmonary nodules. (C) Concomitant post-therapy single photon emission computed tomographyâ&#x20AC;&#x201C;computed tomography (SPECT/CT) images.

Neoadjuvant use of sorafenib in thyroid cancer

only 12.2%. (1). On the contrary, a phase-III trial of lenvatinib in RAI-refractory DTC led to complete response in 15% and partial response in 63% of treated patients. The median time to objective response with lenvatinib was only 2 months (17). Possibly, lenvatinib would be a more effective alternative for induction therapy of locally advanced DTC. Clinical trials are necessary to define the most suitable drug. A major criticism of preoperative use of MKIs is wound complications. Inhibition of VEGFR impairs angiogenesis and granulation tissue formation (18). There are potential risks of severe bleeding and poor wound healing. Tracheoesophageal fistula could occur in a cervical invasive carcinoma. After a withdrawal period of 2 weeks, our patient had adequate wound healing. During previous experiences of patients on sorafenib with renal carcinomas, no major complications were observed, despite discontinuing the drug only a median of 3 days before surgery. The authors attributed the safety of preoperative use of sorafenib to its short half-life (25-48 hours) (19). There are only a few reports of wound complications with MKIs (18). Therefore, more data are necessary to establish the adequate time to withhold the drug before a surgical procedure. It is possible that the use of more selective inhibitors would avoid wound complications, such as vemurafenib for BRAF-mutated papillary thyroid carcinomas, which is under investigation (NCT01709292). In a phase-II trial of vemurafenib in patients with BRAFV600E-positive metastatic or unresectable papillary thyroid cancer refractory to radioactive iodine, researchers observed a partial response in 38.5% of treated patients (20). Although the objective response was smaller than that observed with lenvatinib (2), the drug does not present an anti-angiogenic effect that favors aerodigestive fistulas and impairs wound healing. In this case report, we present 2 important observations. The first one is the potential use of sorafenib preoperatively for tumor reduction, permitting surgery. Surgical resection is more effective in the local control of the disease, preventing precocious airway obstruction and definite tracheostomy and reducing mortality related to invasive disease. We indeed improved the prognosis of a young patient with radioiodine-avid distant metastases. The other observation is the effective use of RAI after sorafenib and radiation 374

therapy. The patientâ&#x20AC;&#x2122;s response to RAI probably was independent of sorafenib, as a previous study could not demonstrate the effect of sorafenib on the reinduction of RAI uptake despite control of tumor progression (21). Moreover, radioiodine uptake was not impaired by the prior radiotherapy. In summary, we describe the first case of unresectable papillary thyroid carcinoma in which sorafenib was used as an induction therapy. Sorafenib treatment resulted in improvement of respiratory symptoms and in sufficient reduction of the tumor mass to enable total thyroidectomy and radioactive iodine treatment, providing long-term control of the disease. Acknowledgments and disclosure: D.L.S.D. received research support from Astra Zeneca and Eisai and lectures fees from Bayer. A.O.H. received research support from Eisai, AstraZeneca, Exelixis and Bayer and consultation and lectures fees from Bayer and Genzyme. G.C. received consultation and lectures fees from Bayer. F.S.R.R., G.B.C. and M.A.K. received lectures fees from Bayer. The other authors have nothing to disclosure.

REFERENCES 1.

Brose MS, Nutting CM, Jarzab B, Elisei R, Siena S, Bastholt L, et al.; DECISION investigators. Sorafenib in radioactive iodine-refractory, locally advanced or metastatic differentiated thyroid cancer: a randomised, double-blind, phase 3 trial. Lancet. 2014;384(9940):319-28.

2. Schlumberger M, Tahara M, Wirth LJ, Robinson B, Brose MS, Elisei R, et al. A phase 3, multicenter, double-blind, placebo-controlled trial of lenvatinib (E7080) in patients with 131I-refractory differentiated thyroid cancer (SELECT). J Clin Oncol. 2014;32:abstr LBA6008. 3. Kiess AP, Agrawal N, Brierley JD, Duvvuri U, Ferris RL, Genden E, et al. External-beam radiotherapy for differentiated thyroid cancer locoregional control: A statement of the American Head and Neck Society. Head Neck. 2016;38(4):493-8. 4. 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. 5. Nixon IJ, Wang LY, Migliacci JC, Eskander A, Campbell MJ, Aniss A, et al. An International Multi-Institutional Validation of Age 55 Years as a Cutoff for Risk Stratification in the AJCC/UICC Staging System for Well-Differentiated Thyroid Cancer. Thyroid. 2016;26(3):373-80. 6. Orosco RK, Hussain T, Brumund KT, Oh DK, Chang DC, Bouvet M. Analysis of age and disease status as predictors of thyroid cancer-specific mortality using the Surveillance, Epidemiology, and End Results database. Thyroid. 2015;25(1):125-32. 7.

Wada N, Nakayama H, Suganuma N, Masudo Y, Rino Y, Masuda M, et al. Prognostic value of the sixth edition AJCC/UICC TNM classification for differentiated thyroid carcinoma with extrathyroid extension. J Clin Endocrinol Metab. 2007;92(1):215-8.

8. Ark N, Zemo S, Nolen D, Holsinger FC, Weber RS. Management of locally invasive well-differentiated thyroid cancer. Surg Oncol Clin N Am. 2008;17(1):145-55, ix. Arch Endocrinol Metab. 2018;62/3

Neoadjuvant use of sorafenib in thyroid cancer

9. Dang RP, McFarland D, Le VH, Camille N, Miles BA, Teng MS, et al. Neoadjuvant therapy in differentiated thyroid cancer. Int J Surg Oncol. 2016;2016:3743420.

16. Cleary JM, Sadow PM, Randolph GW, Palmer EL, Lynch TP, Nikiforov YE, et al. Neoadjuvant treatment of unresectable medullary thyroid cancer with sunitinib. J Clin Oncol. 2010;28(23):e390-2.

10. Besic N, Auersperg M, Gazic B, Dremelj M, Zagar I. Neoadjuvant chemotherapy in 29 patients with locally advanced follicular or Hürthle cell thyroid carcinoma: a phase 2 study. Thyroid. 2012;22(2):131-7.

17. Schlumberger M, Tahara M, Wirth LJ, Robinson B, Brose MS, Elisei R, et al. Lenvatinib versus placebo in radioiodine-refractory thyroid cancer. N Engl J Med. 2015;372:621-30.

11. Besic N, Auersperg M, Dremelj M, Vidergar-Kralj B, Gazic B. Neoadjuvant chemotherapy in 16 patients with locally advanced papillary thyroid carcinoma. Thyroid. 2013;23(2):178-84.

18. Shah DR, Dholakia S, Shah RR. Effect of tyrosine kinase inhibitors on wound healing and tissue repair: implications for surgery in cancer patients. Drug Saf. 2014;37(3):135-49.

12. Ito Y, Higashiyama T, Hirokawa M, Fukushima M, Kihara M, Takamura Y, et al. Clinical trial of weekly paclitaxel chemotherapy for papillary thyroid carcinoma with squamous cell carcinoma component. Endocr J. 2012;59(9):839-44.

19. Cowey CL, Amin C, Pruthi RS, Wallen EM, Nielsen ME, Grigson G, et al. Neoadjuvant clinical trial with sorafenib for patients with stage II or higher renal cell carcinoma. J Clin Oncol. 2010;28(9):1502-7.

13. Borregales LD, Adibi M, Thomas AZ, Wood CG, Karam JA. The role of neoadjuvant therapy in the management of locally advanced renal cell carcinoma. Ther Adv Urol. 2016;8(2):130-41.

20. Brose MS, Cabanillas ME, Cohen EE, Wirth LJ, Riehl T, Yue H, et al. Vemurafenib in patients with BRAF(V600E)-positive metastatic or unresectable papillary thyroid cancer refractory to radioactive iodine: a non-randomised, multicentre, open-label, phase 2 trial. Lancet Oncol. 2016;17(9):1272-82.

14. Loibl S, Rokitta D, Conrad B, Harbeck N, Wüllner M, Warm M, et al. Sorafenib in the Treatment of Early Breast Cancer: Results of the Neoadjuvant Phase II Study – SOFIA. Breast Care (Basel). 2014;9(3):169-74.

21. Hoftijzer H, Heemstra KA, Morreau H, Stokkel MP, Corssmit EP, Gelderblom H, et al. Beneficial effects of sorafenib on tumor progression, but not on radioiodine uptake, in patients with differentiated thyroid carcinoma. Eur J Endocrinol. 2009;161(6):923-31.

15. Barbier L, Muscari F, Le Guellec S, Pariente A, Otal P, Suc B. Liver resection after downstaging hepatocellular carcinoma with Sorafenib. Int J Hepatol. 2011;2011:791013.

Arch Endocrinol Metab. 2018;62/3

375

case report

Inflammatory myopathy in the context of an unusual overlapping laminopathy UETeM – Molecular Pathology Group. IDIS-CIMUS, University of Santiago de Compostela, Spain 2 Division of Rheumatology, University Clinical Hospital of Santiago de Compostela, Spain 3 Division of Pathology, University Clinical Hospital of Santiago de Compostela, Spain. 4 Division of Endocrinology and Nutrition, University Clinical Hospital of Santiago de Compostela, Spain 5 División of Neurology, University Clinical Hospital of Santiago de Compostela, Spain 6 CIBER Fisiopatología de la Obesidad y la Nutrición (CIBERobn), Madrid, Spain 1

Correspondence to: David Araújo-Vilar UETeM, Department of Medicine, IDIS CIMUS, Facultade de Medicina, University of Santiago de Compostela Avda de Barcelona s/n 15707 – Santiago de Compostela, Spain david.araujo@usc.es Received on Jul/19/2017 Accepted on Feb/7/2018 DOI: 10.20945/2359-3997000000048

Cristina Guillín-Amarelle1, Sofía Sánchez-Iglesias1, Antonio Mera2, Elena Pintos3, Ana Castro-Pais4, Leticia Rodríguez-Cañete1, Julio Pardo5, Felipe F. Casanueva4,6, David Araújo-Vilar1,4

SUMMARY Laminopathies are genetic disorders associated with alterations in nuclear envelope proteins, known as lamins. The LMNA gene encodes lamins A and C, and LMNA mutations have been linked to diseases involving fat (type 2 familial partial lipodystrophy [FPLD2]), muscle (type 2 Emery– Dreifuss muscular dystrophy [EDMD2], type 1B limb-girdle muscular dystrophy [LGMD1B], and dilated cardiomyopathy), nerves (type 2B1 Charcot–Marie–Tooth disease), and premature aging syndromes. Moreover, overlapping syndromes have been reported. This study aimed to determine the genetic basis of an overlapping syndrome in a patient with heart disease, myopathy, and features of lipodystrophy, combined with severe metabolic syndrome. We evaluated a 54-year-old woman with rheumatoid arthritis, chronic hypercortisolism (endogenous and exogenous), and a history of cured adrenal Cushing syndrome. The patient presented with a complex disorder, including metabolic syndrome associated with mild partial lipodystrophy (Köbberling-like); mild hypertrophic cardiomyopathy, with Wolff–Parkinson– White syndrome and atrial fibrillation; and limb-girdle inflammatory myopathy. Mutational analysis of the LMNA gene showed a heterozygous c.1634G>A (p.R545H) variant in exon 10 of LMNA. This variant has previously been independently associated with FPLD2, EDMD2, LGMD1B, and heart disease. We describe a new, LMNA-associated, complex overlapping syndrome in which fat, muscle, and cardiac disturbances are related to a p.R545H variant. Arch Endocrinol Metab. 2018;62(3):376-82

INTRODUCTION

utations in the LMNA gene (NM_170707.2) have been associated with a broad spectrum of diseases (1), including type 2 familial partial lipodystrophy (FPLD2), LMNA-related metabolic syndrome, type 2 Emery–Dreifuss muscular dystrophy (EDMD2), type 1B limb-girdle muscular dystrophy (LGMD1B), conduction-system diseases and dilated cardiomyopathy (DCM1A), and progeroid syndromes. FPLD2 begins in women during puberty, with a phenotype of fat loss in the limbs and buttocks, fat accumulation in the face and neck, well-defined musculature, phlebomegaly, insulin resistance, atherogenic dyslipidemia, and high cardiovascular risk (2). The differential diagnosis includes Cushing’s syndrome and truncal obesity. Previous studies have described a LMNA-associated metabolic syndrome with a Köbberling-like fat distribution (3,4). EDMD2 is characterized as a progressive skeletal muscle weakness associated with early joint contractures. 376

LGMD1B causes muscular weakness in the spine and pelvic girdle (1). Cardiac muscle laminopathies can present as nonspecific alterations in electrical conduction, or they can cause malignant arrhythmias and sudden death. Two of the most intriguing features of laminopathies are their clinical heterogeneity and the prevalence of overlapping syndromes. We investigated a female patient with a double Cushing syndrome (endogenous and exogenous), metabolic syndrome, cardiomyopathy, and limb-girdle muscular dystrophy. We aimed to determine whether this syndrome was related to a LMNA mutation.

CASE AND METHODS This study was approved by the Ethics Review Panel of the Xunta de Galicia. The patient and her relatives provided informed consent for participation in the study and for publication of their clinical, biochemical, and genetic information. Arch Endocrinol Metab. 2018;62/3

Inflammatory myopathy and laminopathy

Patient clinical history The patient (Table 1) was a 45-year-old female diagnosed with rheumatoid arthritis at age 40 y and treated with corticosteroids from that point. She was referred to the Endocrinology Division due to high blood pressure, mixed dyslipidemia, and newly diagnosed diabetes (glycated hemoglobin: 9.6%; systolic blood pressure: 150 mmHg; diastolic blood pressure: 90 mmHg; low-density lipoprotein cholesterol: 5.2 mmol/L; high-density lipoprotein cholesterol: 0.65 mmol/L; triglycerides: 3.42 mmol/L). The patient had had an android appearance since puberty, with mild lipoatrophy in the limbs and buttocks and fat accumulation in the abdomen and face (Figure 1). The patient reported that her deceased father had striking hypermuscular limbs, prominent abdominal fat, and diabetes. He had undergone a pacemaker implantation because of a third-degree atrioventricular block. At age 46 y, the patient displayed poor glycemic control, and glucocorticoids were discontinued for several months, with no improvement. She was then

diagnosed with adrenal Cushing’s syndrome, based on a 34-mm right adrenal adenoma (urinary free cortisol: 1100 µg/ 24 h), which was histologically confirmed after a laparoscopic adrenalectomy. Additional hydrocortisone replacement was necessary, and the patient currently continues this treatment. After the adrenal tumor was resected, glycemic control, blood pressure, and lipid levels improved. She continued pharmacological treatment with antihypertensive drugs and metformin, but she was able to discontinue insulin. Some months after the adrenalectomy, the patient was diagnosed with Wolff–Parkinson–White syndrome associated with atrial fibrillation, due to preexcitation, which was successfully ablated with radiofrequency. Transthoracic ultrasonography revealed a slightly hypertrophic non-dilated left ventricle, with preserved systolic function, type 1 diastolic dysfunction, and mild mitral regurgitation. No coronary lesions were detected with cardiac catheterization. At age 50 y, the patient complained of muscular weakness without pain or contractures.

Table 1. Chronological evolution of clinical data Age (years)

Diabetes. Start

Cushing.

Adrenal

antidiabetic

Start statins

surgery,

Myopathy.

Event

49-52

Adrenal

and

and insulin;

start

Stop statins

antihypertensi

intensive

hydrocortiso

at 49

ve drugs

antihypertensi

ve therapy BMI

30.5

31.7

30.8

36.5

25-26

BP (mmHg)

150/90

165/109

130/9

140/93

125/80

CPK (UI/L)

176

157

378-2500

LDH (U/L)

250

331

236

137

65-91

9.6

10.2

8.1

6-6.1

HbA1C (%) Triglycerides (mg/ dL)

164

122

146

254

202-214

HDL (mg/dL)

40-40

LDL (mg/dL)

161

126

63-139

Leptin (µg/L)

12.7

9.9

1-8.4

Insulin* (mUI/L)

46.5

43.9

UFC (Ug/24h)

838

1014

27.4

2.2

1.6

Cortisol** (ug/dL) ACTH

Glucose (mg/dL)

556-860

BP: blood pressure; UFC: urinary free cortisol; *: without exogenous insulin: **: stopping oral corticoid almost 24h before.

Arch Endocrinol Metab. 2018;62/3

377

Inflammatory myopathy and laminopathy

Creatine kinase levels ranged from 321 to 2525 IU/L, and high levels persisted after discontinuation of statins. On physical examination, muscle weakness was evident, predominantly in the pelvic girdle. She was unable to get up from the ground and had marked difficulty in rising from a chair without hand support. Electromyoneurography revealed myopathy changes and spontaneous activity (fibrillation and positive waves) in proximal muscles, without polyneuropathy.

Body composition Skinfolds (Table 2) were measured in triplicate on the dominant extremity with a Lange skinfold caliper (Cambridge Scientific Industries, MD, USA). Segmental body fat distribution was assessed with whole-body dual-energy X-ray absorptiometry, performed with a Lunar model DPX apparatus (GE Medical Systems, Milwaukee, WI, USA). Figure 1. Photographs of the patient show body morphology due to a LMNA variant. (A) Before receiving a cure for Cushing’s syndrome (46 years old). (B) Six years after receiving a cure for Cushing’s syndrome (54 years old).

Molecular analyses DNA was prepared from peripheral white blood cells following standard procedures (5). LMNA exons 1–12

Table 2. Changes in body composition before and after Cushing cure evaluated by anthropometry and DXA Age Weight (kg)

Cured cushing

48.6

82.2

65.3

Obese control*

Non obese control*

Height (cm)

161

BMI (kg/m2)

31.7

25.21

38.1 ± 6.5

26.6 ± 2.8

Tricipital skinfold (mm)

38.8 ± 10.0

29.4 ± 11.7

Bicipital skinfold (mm)

32.4 ± 11.6

21.5 ± 11.8

Suprailiac skinfold (mm)

52.2 ± 14.7

30.6 ± 14.4

Subescapular skinfold (mm)

44.2 ± 12.3

28.4 ± 16.4

Thigh skinfold (mm)

41.1 ± 14.5

22.7 ± 8.6

Calf skinfold (mm)

Active cushing

18.9 ± 14.2

11.2 ± 4.8

WHR

1.0

0.93

0.89 ± 0.08

0.88 ± 0.06

Total fat (kg)

37.5

20.06

43.8 ± 11.7

26.4 ± 7.7

Total fat (%)

46.1

30.6

48.2 ± 6.4

39.3 ± 7.0

Upper limbs fat (kg)

3.48

2.26

4.44 ± 1.1

3.1 ± 1.0

Upper limbs fat (%)

45.3

31.6

47.7 ± 5.7

42.1 ± 7.3

Lower limbs fat (kg)

9.69

6.16

12.4 ± 3.8

8.4 ± 2.2

Lower limbs fat (%)

40.8

30.5

44.6 ± 6.4

39.8 ± 6.4

Trunk fat (kg)

23.2

10.84

22.9 ± 5.1

14.0 ± 5.7

Trunk fat (%)

51.7

31.8

52.5 ± 7.6

40.8 ± 10.2

Visceral fat (g)

764

2012 ± 894

992 ± 693

2.39

1.75

1.61

1.66

Trunkal/Lower limbs fat ratio (kg) *: Normal ranges from ref. 3; WHR: waist to hip ratio.

378

Arch Endocrinol Metab. 2018;62/3

Inflammatory myopathy and laminopathy

Histological muscle studies Deltoid muscle biopsy samples were snap frozen in liquid nitrogen, and cryostat sections were processed and stained to exclude dystrophies and other pathological conditions (e.g., mitochondrial myopathies, metabolic diseases) as follows: PAS, modified Gomori trichrome, Oil Red O, enzyme histochemistry (muscle phosphorylase, muscle phosphofructokinase, lactate dehydrogenase, AMP deaminase, cytochrome-C oxidase, SDH, and NADH), and immunohistochemistry (beta-spectrin to check preservation of the plasma membrane; slow, fast, and fetal myosin; utrophin; alpha, beta, gamma, and delta sarcoglycan; dystrophin fractions: N- terminal, C-terminal, and rod domain; NOS; alpha and beta dystroglycan; dysferlin, caveolin-3, collagen type VI, laminin alpha2 chain of merosin (laminin-2), telethonin, myotilin, desmin, calpain-3, MHC class I antigen, and inflammatory

markers (CD68/KP1, CD3 and CD20 for histiocytes and lymphocytes). The only commercial antibodies available for Lamin A/C are not helpful for diagnostic purposes (here it was used as a positive control for emerin antibody). Immunostains were performed after antigen retrieval using a standard avidin–biotin immunoperoxidase detection technique (EnVision Systems, Dako, Glostrup, Denmark).

RESULTS We identified a heterozygous c.1634G>A (p.R545H) variant of LMNA, located in exon 10. This variant had not been probed for pathogenicity with in silico approaches (PolyPhen and SIFT). Its allelic frequency was 0.00019 (1000 Genomes Project). Of the patient’s relatives whom we studied (mother, sisters, offspring), none carried this variant. Histological findings of muscle samples were consistent with acute and chronic, nonspecific, inflammatory myopathy (Figure 2).

Figure 2. (1) Snap frozen cryostat sections of deltoid muscle showing endomisial inflammatory infiltrates composed mainly of hystiocytes (asterisks) (Hematoxylin and eosin, 400x). (2) MHC class I antigen is upregulated in all fibers, with immunolabeling at plasma membrane (arrow) and sarcoplasm (asterisk). See also “pseudovacuoles” secondary to “ice crystals” snap frozen artifact (arrow head) (Immunoperoxidase reaction, diaminobenzidine brown staining chromogen, 400x). (3a) Immunostaining of CD68 (KP1) antibody remark histiocityc inflammatory component between myofibers, with (3b) a minor population of CD3 antibody positive T lymphocytes (asterisks) (400x). Arch Endocrinol Metab. 2018;62/3

379

and the surrounding intronic sequences were amplified by PCR. Primers and conditions have been previously described (6).

Inflammatory myopathy and laminopathy

DISCUSSION We investigated an unusual case of laminopathy and found that it was due to a p.R545H LMNA variant that overlapped with chronic hypercortisolism. Thus, after chronic hypercortisolism had been diagnosed and cured, the patient exhibited atypical partial lipodystrophy, idiopathic inflammatory myopathy, and cardiomyopathy with conduction disturbances. Although this variant has not been shown to be pathogenic with in silico approaches, it has been associated with FPLD2 (7), EDMD2 (8), LGMD1B (9), and heart disease (10). In FPLD2, more than 80% of cases are due to missense mutations in exon 8; however, atypical phenotypes have been related to mutations in other exons (11). Moreover, several patients with mutations outside the immunoglobulin-like fold of lamin A have lacked the typical FPLD2 phenotype but experienced insulin resistance (4). In addition, some cases of FPLD2 have been associated with heart conduction disorders, valvulopathies, and cardiomyopathy (12-14). Other authors have reported LMNA-associated complex phenotypes, including heart failure and limb-girdle muscular dystrophy, due to a Ser334del variant (15,16); muscular dystrophy, lipodystrophy, and cardiac rhythm disturbances related to a R527P variant (17); or FPLD, early heart failure, first-degree atrioventricular block, and late proximal muscle weakness due to a R28W variant (12). Exon 10 of LMNA corresponds to the C-terminal domain, common to both lamins A and C, which forms an immunoglobulin-like, three-dimensional structure (1). The conformation of this domain is well defined for residues 430–545 (18). Arginine 545 is at the external surface of the structure. Mutations in the Ig-fold can affect either head-to-tail polymerization, which destabilizes the three-dimensional structure of the C-terminal domain, or lamin A/C interactions with other proteins (1). At least 13 missense mutations at the C-terminal domain are related to EDMD2 (18), including the nearby R541H. Although the R545 residue has not been specifically studied, it does not seem to influence the stability of the 3D structure. However, the exchange of arginine (positive polar) for histidine (neutral polar) could alter interactions with other proteins at the nuclear lamina. This study was particularly challenging because of the presence of Cushing’s syndrome and the 380

autoimmune background. Chronic hypercortisolism causes a characteristic fat distribution, in particular, excess abdominal fat (19); however, no particular changes in limb fat have been reported (20). Strikingly, in this patient, once hypercortisolism was cured, an abnormal fat distribution became more evident, although it was not as severe as observed in classical Dunnigan disease. This change in fat distribution was particularly intriguing because it highlighted important differences among LMNA mutations that cause FPLD (13). The R545H variant, previously associated with FPLD (7), caused severe metabolic syndrome with android fat distribution in this patient. The patient was initially diagnosed with polymyositis, based on girdle weakness, high creatine kinase, and muscle lymphocyte infiltration. Because of her autoimmune background, this diagnosis was probably the most parsimonious. However, the R545H variant has been associated with LGMD1B (9), which clinically overlaps with polymyositis. Other LMNA mutations that cause muscular and/or cardiac laminopathies have been related to inflammatory changes in muscle specimen biopsies. For example, in biopsies from patients with infantile- onset LMNAassociated myopathy, Komaki and cols. (21) reported mononuclear cell infiltrations that were positive for lymphocyte markers CD4, CD8, or CD20, active necrosis, and regeneration. Additionally, they observed elevated sarcolemmal HLA staining in many fibers. Similarly, HLA class I antigens were upregulated in our samples and in samples described previously in a study on LGMD1B (15,16). It cannot be ruled out that chronic hypercortisolism had influenced the myopathy. However, there are some clues to differences between the corticoid-myopathy case and the case described here. In the first case, serum values of muscle enzymes are typically normal or slightly elevated, electromyoneurography is usually normal, and there are no inflammatory infiltrates or necrosis on muscle biopsy samples (22). Although unusual, polymyositis has been associated with cardiac involvement, manifested as rhythm disturbances, conduction defects, and heart failure (23). Similarly, laminopathies, like FPLD2 or LGMD1B (9,14), and particularly the R545H variant, have also been related to heart disease (10). Patients with cardiac compromise related to LMNA typically show initial signs of nonspecific rhythm disturbances after age 30 y (24), and they frequently need a Arch Endocrinol Metab. 2018;62/3

Inflammatory myopathy and laminopathy

Acknowledgments: we are indebted to the patients of this study for their collaboration. This work was funded by the Instituto de Salud Carlos III (grant number: PI081449) and the European Regional Development Fund, FEDER. In addition, SRG was awarded a Research Fellowship granted by the Asociación Española de Familiares y Afectados de Lipodistrofias (AELIP). None of the authors has any conflict of interest to disclose. Ethical Publication Statement: we confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Broers JL, Ramaekers FC, Bonne G, Yaou RB, Hutchison CJ. Nuclear lamins: laminopathies and their role in premature ageing. Physiol Rev. 2006;86(3):967-1008. 2. Brown RJ, Araujo-Vilar D, Cheung PT, Dunger D, Garg A, Jack M, et al. The Diagnosis and Management of Lipodystrophy Syndromes: A Multi-Society Practice Guideline. J Clin Endocrinol Metab. 2016;101(12):4500-11. 3. Guillin-Amarelle C, Sanchez-Iglesias S, Castro-Pais A, RodriguezCanete L, Ordonez-Mayan L, Pazos M, et al. Type 1 familial partial lipodystrophy: understanding the Kobberling syndrome. Endocrine. 2016;54(2):411-21. 4. Decaudain A, Vantyghem MC, Guerci B, Hecart AC, Auclair M, Reznik Y, et al. New metabolic phenotypes in laminopathies: LMNA mutations in patients with severe metabolic syndrome. J Clin Endocrinol Metab. 2007;92(12):4835-44. 5. Sambrook J, Russell DW. The Condensed Protocols from Molecular Cloning: A Laboratory Manual: Cold Spring Harbor Laboratory Press; 2006. Arch Endocrinol Metab. 2018;62/3

6. Araujo-Vilar D, Loidi L, Dominguez F, Cabezas-Cerrato J. Phenotypic gender differences in subjects with familial partial lipodystrophy (Dunnigan variety) due to a nuclear lamin A/C R482W mutation. Horm Metab Res. 2003;35(1):29-35. 7. Chan D, McIntyre AD, Hegele RA, Don-Wauchope AC. Familial partial lipodystrophy presenting as metabolic syndrome. J Clin Lipidol. 2016;10(6):1488-91. 8. Huong S. Molecular genetic studies in hereditary laminopathies of man.: Ernst- Moritz-Arndt-Universität, Greifswald; 2010. 9. Maggi L, Carboni N, Bernasconi P. Skeletal Muscle Laminopathies: A Review of Clinical and Molecular Features. Cells. 2016;5(3):pii: E33. 10. van Rijsingen IA, Nannenberg EA, Arbustini E, Elliott PM, Mogensen J, Hermans-van Ast JF, et al. Gender-specific differences in major cardiac events and mortality in lamin A/C mutation carriers. Eur J Heart Fail. 2013;15(4):376-84. 11. Garg A, Vinaitheerthan M, Weatherall PT, Bowcock AM. Phenotypic heterogeneity in patients with familial partial lipodystrophy (dunnigan variety) related to the site of missense mutations in lamin a/c gene. J Clin Endocrinol Metab. 2001;86(1): 59-65. 12. Garg A, Speckman RA, Bowcock AM. Multisystem dystrophy syndrome due to novel missense mutations in the amino-terminal head and alpha-helical rod domains of the lamin A/C gene. Am J Med. 2002;112(7):549-55. 13. Araujo-Vilar D, Lado-Abeal J, Palos-Paz F, Lattanzi G, Bandin MA, Bellido D, et al. A novel phenotypic expression associated with a new mutation in LMNA gene, characterized by partial lipodystrophy, insulin resistance, aortic stenosis and hypertrophic cardiomyopathy. Clin Endocrinol (Oxf). 2008;69(1):61-8. 14. Panikkath R, Panikkath D, Sanchez-Iglesias S, Araujo-Vilar D, Lado-Abeal J. An Uncommon Association of Familial Partial Lipodystrophy, Dilated Cardiomyopathy, and Conduction System Disease. J Investig Med High Impact Case Rep. 2016;4(3):2324709616658495. 15. Madej-Pilarczyk A, Niezgoda A, Janus M, Wojnicz R, Marchel M, Fidzianska A, et al. Limb-girdle muscular dystrophy with severe heart failure overlapping with lipodystrophy in a patient with LMNA mutation p.Ser334del. J Appl Genet 2017;58(1):87-91. 16. Madej-Pilarczyk A, Niezgoda A, Janus M, Wojnicz R, Marchel M, Fidziańska A, et al. Limb-girdle muscular dystrophy with severe heart failure overlapping with lipodystrophy in a patient with LMNA mutation p.Ser334del. J Appl Genet. 2017;58(1):87-91. 17. van der Kooi AJ, Bonne G, Eymard B, Duboc D, Talim B, Van der Valk M, et al. Lamin A/C mutations with lipodystrophy, cardiac abnormalities, and muscular dystrophy. Neurology. 2002;59(4):620-3. 18. Krimm I, Ostlund C, Gilquin B, Couprie J, Hossenlopp P, Mornon JP, et al. The Ig-like structure of the C-terminal domain of lamin A/C, mutated in muscular dystrophies, cardiomyopathy, and partial lipodystrophy. Structure. 2002;10(6):811-23. 19. Garrapa GG, Pantanetti P, Arnaldi G, Mantero F, Faloia E. Body composition and metabolic features in women with adrenal incidentaloma or Cushing’s syndrome. J Clin Endocrinol Metab. 2001;86(11):5301-6. 20. Rockall AG, Sohaib SA, Evans D, Kaltsas G, Isidori AM, Monson JP, et al. Computed tomography assessment of fat distribution in male and female patients with Cushing’s syndrome. Eur J Endocrinol. 2003;149(6):561-7. 21. Komaki H, Hayashi YK, Tsuburaya R, Sugie K, Kato M, Nagai T, et al. Inflammatory changes in infantile-onset LMNA-associated myopathy. Neuromuscul Disord. 2011;21(8):563-8. 22. Gupta A, Gupta Y. Glucocorticoid-induced myopathy: Pathophysiology, diagnosis, and treatment. Indian J Endocrinol Metab. 2013;17(5):913-6. 23. Eisen A, Arnson Y, Dovrish Z, Hadary R, Amital H. Arrhythmias and conduction defects in rheumatological diseases--a comprehensive review. Semin Arthritis Rheum. 2009;39(3):145-56.

381

permanent pacemaker. Later, cardiac compromise may be further complicated with cardiomyopathy (25). Our patient was diagnosed with Wolff–Parkinson–White syndrome, atrial fibrillation, mitral regurgitation, and mild hypertrophic cardiomyopathy. Taken together, the diagnoses of severe metabolic syndrome, girdle weakness appearing in the fourth decade, and heart involvement in a patient that carried the R545H variant in LMNA resembled a complex laminopathic disorder. However, the presence of chronic hypercortisolism represented a confounding factor, and rheumatoid arthritis prevented us from definitely ruling out a random association of different unrelated autoimmune entities. In summary, this case emphasizes the need for actively searching for LMNA mutations in patients with clinical features compatible with partial lipodystrophy, cardiac conduction abnormalities, cardiomyopathy, and/or certain types of myopathy. A correct diagnosis can facilitate early interventions to prevent the consequences of these pathologies, which can be lethal.

Inflammatory myopathy and laminopathy

25. Carboni N, Mateddu A, Marrosu G, Cocco E, Marrosu MG. Genetic and clinical characteristics of skeletal and cardiac muscle in patients with lamin A/C gene mutations. Muscle Nerve. 2013;48(2):161-70.

24. van Berlo JH, de Voogt WG, van der Kooi AJ, van Tintelen JP, Bonne G, Yaou RB, et al. Meta-analysis of clinical characteristics of 299 carriers of LMNA gene mutations: do lamin A/C mutations portend a high risk of sudden death? J Mol Med (Berl). 2005;83(1):79-83.

382

Arch Endocrinol Metab. 2018;62/3

Instructions for authors GENERAL INFORMATION

should be 2,000 words or less, with no more than four figures and tables, and no more than 30 references.

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. Manuscripts 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. 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).

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 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.

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.

Case Report 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 delineation of the natural history or management of the condition described. Case Reports

GENERAL FORMAT 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.

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. 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

Structured Abstracts 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.

MANUSCRIPT CATEGORIES

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.

Results and Discussion 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.

Units of Measure 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.

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.

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.

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 of reproducing color figures in print (the publisher will provide price quotes upon acceptance of the manuscript).

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