AE&M 61-3 by Sociedade Brasileira de Endocrinologia e Metabologia

ISSN 2359-3997

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM Vol. 61 – No. 03 – June 2017

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. 61 – No. 03 – June 2017

memorial 203 Eulogy to Jorge Luis Gross and Mirela Jobim de Azevedo Jairo T. Hidal

Archives of Endocrinology editorials

205 Two themes in thyroid cancer: artful diagnosis and shortened lives

OFFICIAL JOURNAL 208 Gestational diabetes mellitus and type 2 diabetes: same disease in a different moment life? Maybe not OFofTHE BRAZILIAN Lenita Zajdenverg, Carlos Antonio Negrato SOCIETY OF original articles 211 Likelihood of malignancy in thyroid nodules according to a proposed Thyroid Imaging Reporting and Data System (TI-RADS) ENDOCRINOLOGY classification merging suspicious and benign ultrasound features Ricardo Luiz Costantin Delfim, Leticia Carrasco Garcez da Veiga, Ana Paula Aguiar Vidal, AND Flávia Paiva Proença Lobo Lopes, METABOLISM Cristiane Gomes Lima, Leonard Wartofsky

and Metabolism Mário Vaisman, Patrícia de Fatima dos Santos Teixeira

222 Deaths related to differentiated thyroid cancer: a rare but real event

Ana Kober N. Leite, Beatriz G. Cavalheiro, Marco Aurélio Kulcsar, Ana de Oliveira Hoff, Lenine G. Brandão, Claudio Roberto Cernea, Leandro L. Matos

228 Relation between fetal anthropometric parameters and cord blood adiponectin and high-sensitivity C-reactive protein in gestational diabetes mellitus Mohammad Reza Aramesh, Masoud Dehdashtian, Arash Malekian, Shiva ShahAli, Kobra Shojaei

233 Serum Fluorescent Advanced Glycation End (F-AGE) products in gestational diabetes patients

Archives of

João Paulo Lobo Júnior, Catiane Pompilio Brescansin, Izabella C. R. Santos-Weiss, Marciane Welter, Emanuel Maltempi de Souza, Fabiane Gomes de Moraes Rego, Geraldo Picheth, Dayane Alberton

238 Type 2 diabetes-associated genetic variants of FTO, LEPR, PPARg, and TCF7L2 in gestational diabetes in a Brazilian population Mauren Isfer Anghebem-Oliveira, Bruna Rodrigues Martins, Dayane Alberton, Edneia Amancio de Souza Ramos, Geraldo Picheth, Fabiane Gomes de Moraes Rego

Endocrinology

249 Serum levels of leptin and adiponectin and clinical parameters in women with fibromyalgia and overweight/obesity Eduardo S. Paiva, Aline Andretta, Emmanuelle Dias Batista, Márcia Maria Marques Teles Lobo, Renata Costa de Miranda, Renato Nisihara, Maria Eliana Madalozzo Schieferdecker, César L. Boguszewski

257 Applicability of predictive equations for resting energy expenditure in obese patients with obstructive sleep apnea Mariana Pantaleão del Re, Camila Maria de Melo, Marcus Vinicius dos Santos, Sergio Tufik, Marco Túlio de Mello

and Metabolism

263 Orange juice with a high-fat meal prolongs postprandial lipemia in apparently healthy overweight/obese women Raquel Cristina L. A. Coelho, Helen Hermana M. Hermsdorff, Renata S. Gomide, Raquel Duarte M. Alves, Josefina Bressan

269 Ultrasonographic assessment of thyroid volume in oldest-old individuals

Glaucia Cruzes Duarte, Lara Miguel Quirino Araujo, Felix Magalhães Filho, Clineu Mello Almada Filho, Maysa Seabra Cendoroglo

OFFICIAL JOURNAL OF THE BRAZILIAN SOCIETY OF ENDOCRINOLOGY AND METABOLISM

276 Can FIB4 and NAFLD fibrosis scores help endocrinologists refer patients with non-alcoholic fat liver disease to a hepatologist? Rodrigo Bremer Nones, Cláudia Pontes Ivantes, Maria Lucia Alves Pedroso

282 Visceral adiposity index and triglyceride/high-density lipoprotein cholesterol ratio in hypogonadism

Cem Haymana, Alper Sonmez, Aydogan Aydogdu, Serkan Tapan, Yalcin Basaran, Coskun Meric, Kamil Baskoy, Mustafa Dinc, Mahmut Yazici, Abdullah Taslipinar, Cem Barcin, Mahmut Ilker Yilmaz, Erol Bolu, Omer Azal

case reports 288 Intrathoracic stomach mimicking bone metastasis from thyroid cancer in whole-body iodine-131 scan diagnosed by SPECT/CT Francisco Javier García-Gómez, Pablo Antonio de la Riva-Pérez, Cinta Calvo-Morón, Cristina Buján-Lloret, Teresa Cambil-Molina, Juan Castro-Montaño

291 A rare case of ectopic ACTH syndrome originating from malignant renal paraganglioma Esra Tutal, Demet Yılmazer, Taner Demirci, Evrim Cakır, Salih Sinan Gültekin, Bahadır Celep, Oya Topalog#lu, Erman Çakal

reviews 296 Glycated albumin: a potential biomarker in diabetes Priscila Aparecida Correa Freitas, Lethicia Rozales Ehlert¹, Joíza Lins Camargo

305 Diagnosis and management of primary aldosteronism Leticia A. P. Vilela, Madson Q. Almeida

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: 22203.6.17 Informations:

Endocrinology

and Metabolism

Title: Suíte construtiva – SF 10116 Author: Sergio Fingermann

Technique: Acrylic paint on canvas

Measures: 0.80 mOFFICIAL X 0.70 m JOURNAL OF THE BRAZILIAN

SOCIETY OF ENDOCRINOLOGY AND METABOLISM Collection: Dan Galeria 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

INTERNATIONAL

ENDOCRINOLOGY CAMPAIGNS

President

César Luiz Boguszewski

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

Members

cesarluiz@hc.ufpr.br 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 Delano Broinstein

Members

Marcello Delano Broinstein, 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

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 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

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

ENDOCRINE DYSREGULATORS

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

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

TITLE OF SPECIALIST IN ENDOCRINOLOGY AND METABOLISM

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 (2016/2017)

President

Luiz Alberto Andreotti Turattii

Vice-Presidents

Reine Marie Chaves Fonseca Solange Travassos de Figueiredo Alves Sergio Alberto Cunha Vêncio Levimar Rocha Araujo Mauro Scharf Pinto

1ST Secretary

Domingos Augusto Malerbi

2ND Secretary

Gustavo J. P. Caldas Costa

1ST Treasurer

Antonio Carlos Lerário

2ND Treasurer

Roberto Abrao Raduan

Supervisory Board

João Eduardo Nunes Salles João Paulo Iazigi Luis Antonio de Araujo

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

memorial

Eulogy to Jorge Luis Gross and Mirela Jobim de Azevedo Jairo T. Hidal1

Hospital Israelita Albert Einstein (HIAE), São Paulo, SP, Brasil

Arch Endocrinol Metab. 2017;61/3

Correspondence to: Jairo T. Hidal hidal@einstein.br Received on June/7/2017 Accepted on June/7/2017 DOI: 10.1590/2359-3997000000277

nna Karenina (by Liev Tolstoi), probably the best novel ever written, says in its opening lines: “All happy families are alike; each unhappy family is unhappy its own way”, each one of us, colleagues and friends of Mirela & Jorge, will mourn their absence in its own way. Just now we can cry for the sudden loss of these extraordinary couple, Mirela & Jorge Luis. We can mourn them together with their families, but it will be wrong to remember them through the tragedy that took them away. We can do better, by remembering what they did, and achieved during the time it was allowed to them to be with us. I was not a personal friend of them; I never worked with them (although, I would have loved to have him as my formal mentor); we not even lived in the same city, but through the years we developed a bond, a trust, and a partnership that goes beyond anything I could write. My dealings with him were mainly when I (10 years younger than him) needed advice about the care of a certain patient or condition, and didn’t know to whom to call, but it was always to Jorge Luis, that altruistically, gently, and always friendly that I would appeal. And always, always would have the proper and kind answer. Jorge Luis was what in Yiddish (the “língua franca” of the Jews from Eastern Europe) is called: “a Mensch” (a noble and respectable person; someone to admire and emulate). Sir William Osler would have been proud of his clinical skills, and any researcher would find difficult to match his competency. His lectures and presentations were always superb (sharp, clear, and easy to understand). After finishing his PhD, at USP Ribeirão Preto, in 1975, Jorge Luis returned to his Alma Mater (UFRGS), and very fast transformed the already existing Endocrinology Unit into a leading Center, including: Clinical, Research, and Teaching. In order to fulfill the competencies required, he trained several students, recruited the best minds, and sent people to the top centers around the world. This new generation is obliged to assume the helm before it was expected, but they will do their task, because he trained them well, and it is what is expected of them.

203

Eulogy to Jorge Luis Gross and Mirela Jobim de Azevedo

People who worked with him described him as being: bright; capable of integrating several fields of knowledge; able to formulate the proper questions, which would be answered by careful research. He had no compromise with laziness, or dishonesty. At the end of the day, what counted was the greater benefit for the patient. His main research interest was about the chronic complications of diabetes mellitus, and he published several studies that range from the epidemiological; clinical and treatment of this condition. In 1983, already very successful (both in the academic and private clinic) he decided to expand his training, and expend one year in a Post-Doc at Guy’s Hospital, in London. Mirela was also a tremendous achiever, in a short span of only 18 years she graduated from Medical School and achieved the pinnacle of the academic path, her “Livre-Docência”. She became one of the youngest Full Professor at UFRGS. Her CV is impressive, even not taking in account her brief career: Over 100 published papers; over 30 PostGraduation thesis supervisions; more than 80 Examination Boards for thesis membership. Mirela was a reviewer for several leading Medical Journal, in Brazil, and abroad. Students loved her lectures, and she was an acclaimed “Honored teacher”. But besides all her professional and academic achievements, the one she cared more was her daughter, Luisa, a young medical student, for whom the loss of these wonderful people would resent more. We, the friends and admirers of Mirela and Jorge Luis have an obligation to look after her. Jorge Luis also had three other sons, from his first marriage, one of them a very well-known physician in Porto Alegre. The great American poet, Walt Whitman wrote a magnificent poem at the time of President Lincoln death, and with all due respect, I believe that “O Captain! My Captain” can be used to express our gratitude and respects for these wonderful people: O Captain! my Captain! our fearful trip is done, The ship has weather’d every rack, the prize we sought is won, The port is near, the bells I hear, the people all exulting, While follow eyes the steady keel, the vessel grim and daring; But O heart! heart! heart! O the bleeding drops of red, Where on the deck my Captain lies, Fallen cold and dead. O Captain! my Captain! rise up and hear the bells; Rise up—for you the flag is flung—for you the bugle trills, For you bouquets and ribbon’d wreaths—for you the shores a-crowding, For you they call, the swaying mass, their eager faces turning; Here Captain! dear father! This arm beneath your head! It is some dream that on the deck, You’ve fallen cold and dead.

My Captain does not answer, his lips are pale and still, My father does not feel my arm, he has no pulse nor will, The ship is anchor’d safe and sound, its voyage closed and done, From fearful trip the victor ship comes in with object won; Exult O shores, and ring O bells! But I with mournful tread, Walk the deck my Captain lies,

Fallen cold and dead.

Brazilian Endocrinologists, for one last time, lend me your attention, please stand up, Mirela and Jorge Luis have passed on. Acknowledgments: I would like to thank the important collaboration of Drs. Decio L. Eizirik, Mauro Czepielewski and Rui M. B. Maciel for this text.

204

Arch Endocrinol Metab. 2017;61/3

editorial

Two themes in thyroid cancer: artful diagnosis and shortened lives Cristiane Gomes Lima1, Leonard Wartofsky1

Arch Endocrinol Metab. 2017;61/3

MedStar Washington Hospital Center and MedStar Health Research Institute, Georgetown University School of Medicine, Washington, DC, USA

Correspondence to: Leonard Wartofsky Georgetown University School of Medicine Department of Medicine Washington Hospital Center 110 Irving Street, N.W. 20010-2975 – Washington, DC, USA Leonard.Wartofsky@Medstar.net Received on May/25/2017 Accepted on May/25/2017 DOI: 10.1590/2359-3997000000275

atients presenting with a thyroid nodule are common in the clinical practice of endocrinologists, even for those who are not thyroidologists. When seeing a patient with a thyroid nodule, the question that typically occurs first is whether the nodule could be malignant and how can the diagnosis be most efficiently and accurately determined. Then, once a diagnosis of cancer might be confirmed, and discussion turns to details about management, the next prominent question in the patient’s mind relates to their prognosis. The importance of these two questions, precise diagnosis and prognosis, forms the basis for two papers appearing in this issue of the Archives of Endocrinology and Metabolism. Detection of thyroid nodules has been increasing significantly due to the more widespread use of ultrasonography of the neck. Our professional society guidelines recommend fine-needle aspiration (FNA) as the procedure of choice for nodules > 1 cm, and the routine use of thyroid ultrasonography to characterize the nodules. Yet the subjective nature of sonogram evaluations and the lack of uniformity in the reports of the characteristics of thyroid nodules may be troublesome for management decision-making. This is the setting in which Delfim and cols. (1) propose and offer a new classification system to distinguish the ultrasound features between benign and malignant thyroid nodules. In fact, similar efforts to develop a thyroid imaging reporting and data system (TIRADS) to categorize thyroid nodules and evaluate their risk of malignancy date back to 2009 (2-6), in parallel to what has been done for breast imaging with the development of the Breast Imaging Reporting and Data System (BI-RADS) (7). The Korean Society of Thyroid Radiology has recently revised its recommendations, the K-TIRADS (8) and the American College of Radiology has just released a white paper of the TI-RADS Committee (9). The current guidelines of the American Thyroid Association (ATA) (10) recommend the use of sonographic patterns, instead of isolated sonographic features, to estimate the risk of malignancy of thyroid nodules. Each of the latter reports strive to present a standardized system for analyzing and reporting thyroid ultrasound that could result in greater diagnostic specificity. Strengths and weaknesses of the various systems relate to whether the studies are either prospective or retrospective, the readings are by single or multiple radiologist investigators, the use of different techniques for the evaluation of the nodules and varying statistical models, and correlation to different categories of the Bethesda system. Regardless of which system might be ultimately adopted, the need for a standardized terminology is rational and functional. A committee of the American College of Radiology, composed by radiologists with expertise in thyroid imaging, has developed a descriptive lexicon of the sonographic characteristics of thyroid nodules (11).

205

Issues in thyroid cancer

The Korean Society of Thyroid Radiology has also recommended terminology and defined sonographic features of nodules in its revised consensus (8). In their paper, Delfim and cols. use a well defined terminology to describe their scoring system for ultrasound features that led to their proposed TI-RADS system, based on statistical analysis and a weight conception process. In this process, certain features of the nodules, such as hypoechogenicity and microcalcifications, received a higher score than central vascularization, reinforcing the relevance of B-mode ultrasound over Doppler mode characteristics. The exclusion of indeterminate nodules from their analysis is a relative weakness of their study. However, it is generally acknowledged that imaging reporting systems are not supposed to be superior to the cytological evaluation of a thyroid nodule. Any TIRADS system should remain flexible and learn a lesson from its “older brother”, the BI-RADS system (7,12-15), insofar as being a “living” document, founded on logical and evidence-based data but open to updates as new data are acquired (16). The second paper in this issue, by Leite and cols. (17), analyzes deaths related to differentiated thyroid cancer (DTC). Current trends in management have moved us to be less aggressive in the treatment and management of DTC, practicing the so called “less is more” philosophy. However, as the authors discuss, the extremely low mortality of this type of cancer “is balanced by its high prevalence, so the number of deaths cannot be overlooked”. Indeed, endocrinologists who work in referral services of thyroid cancer and see many high risk patients appreciate the mortality risk. It is remarkable that the series of Leite and cols. included more patients with follicular cancer (with associated risk of distant metastases) than is commonly seen. Also noteworthy is the fact that 4 out of 33 patients had stage T1 disease, while 2 out of 33 patients had stage T2, i.e., low or intermediate risk patients that would be considered (by the current “less is more” philosophy) for less aggressive treatment (e.g., lobectomy instead of total thyroidectomy, and no radioiodine ablation), in keeping with the new ATA guidelines (10). Clearly, all T1 patients do not behave the same. Those low risk patients with ultimate poor outcomes could be detected by periodic risk assessment in order to detect those patients initially stratified as low risk who may develop an unexpected aggressive course of the disease. The light at end of the tunnel may derive from a beacon of promise from molecular diagnosis, either for the management of nodules or for the follow-up of 206

proven cancer patients at either low or high risk. When this diagnostic tool becomes more refined and more accessible, it will possibly identify the most suspicious nodules and those cancer patients who need a more aggressive treatment approach during follow-up. A refined and comprehensive molecular analytic approach to the thyroid nodule will provide true precision medicine for our patients and the best hope for maximizing benefit, reducing risk, and achieving good outcomes. To echo Goethe, dealing with our patients’ fear of cancer when a nodule is discovered calls forth the art of the true physician, especially for those patients whose life ultimately will be foreshortened after presentation with metastases, for we have only imprecise information on which to act and affect their outcome, and only a limited time to do so [Von Goethe, J. “Art is long, life short, judgment difficult, occasion transient.” (18)]. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Delfim RLC, Veiga LCG, Vidal APA, Lopes FPPL, Vaisman M, Teixeira PFS. Likelihood of malignancy in thyroid nodules according to a proposed Thyroid Imaging Reporting and Data System (TI-RADS) classification merging suspicious and benign ultrasound features. Arch Endocrinol Metab. 2017;61(3):211-21. 2. Horvath E, Majlis S, Rossi R, Franco C, Niedmann JP, Castro A, et al. An ultrasonogram reporting system for thyroid nodules stratifying cancer risk for clinical management. J Clin Endocrinol Metab. 2009;94(5):1748-51. 3. Park JY, Lee HJ, Jang HW, Kim HK, Yi JH, Lee W, et al. A proposal for a thyroid imaging reporting and data system for ultrasound features of thyroid carcinoma. Thyroid. 2009;19(11):1257-64. 4. Russ G, Bigorgne C, Royer B, Rouxel A, Bienvenu-Perrard M. [The Thyroid Imaging Reporting and Data System (TIRADS) for ultrasound of the thyroid]. J Radiol. 2011;92(7-8):701-13. 5. Kwak JY, Han KH, Yoon JH, Moon HJ, Son EJ, Park SH, et al. Thyroid imaging reporting and data system for US features of nodules: a step in establishing better stratification of cancer risk. Radiology. 2011;260(3):892-9. 6. Russ G, Royer B, Bigorgne C, Rouxel A, Bienvenu-Perrard M, Leenhardt L. Prospective evaluation of thyroid imaging reporting and data system on 4550 nodules with and without elastography. Eur J Endocrinol. 2013;168(5):649-55. 7. American College of Radiology. Breast Imaging Reporting and Data System (BI-RADS). Reston, VA: American College of Radiology; 1993. 8. Shin JH, Baek JH, Chung J, Ha EJ, Kim JH, Lee YH, et al. Ultrasonography Diagnosis and Imaging-Based Management of Thyroid Nodules: Revised Korean Society of Thyroid Radiology Consensus Statement and Recommendations. Korean J Radiol. 2016;17(3):370-95. 9. Tessler FN, Middleton WD, Grant EG, Hoang JK, Berland LL, Teefey SA, et al. ACRThyroid Imaging, Reporting and Data System (TI-RADS): White Paper of the ACR TI-RADS Committee. J Am Coll Radiol. 2017;14(5):587-95. 10. Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, et al. 2015 American Thyroid Association Arch Endocrinol Metab. 2017;61/3

Issues in thyroid cancer

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. 11. Grant EG, Tessler FN, Hoang JK, Langer JE, Beland MD, Berland LL, et al. Thyroid Ultrasound Reporting Lexicon: White Paper of the ACR Thyroid Imaging, Reporting and Data System (TIRADS) Committee. J Am Coll Radiol. 2015;12(12 Pt A):1272-9. 12. American College of Radiology. Breast Imaging Reporting and Data System (BI-RADS). 2nd ed. Reston, VA: American College of Radiology; 1995.

15. American College of Radiology. Breast Imaging Reporting and Data System (BI-RADS). 5th ed. Reston, VA: American College of Radiology; 2013. 16. Burnside ES, Sickles EA, Bassett LW, Rubin DL, Lee CH, Ikeda DM, et al. The ACR BI-RADS experience: learning from history. J Am Coll Radiol. 2009;6(12):851-60. 17. Leite AKN, Cavalheiro BG, Kulcsar MA, Hoff AO, Brandão LG, Cernea CR, et al. Deaths related to differentiated thyroid cancer: a rare but real event. Arch Endocrinol Metab. 2017;61(3):222-7. 18. Von Goethe JW, Blackall EA, Lange V. Wilhelm Meister’s Apprenticeship: Princeton University Press; 1995.

13. American College of Radiology. Breast Imaging Reporting and Data System (BI-RADS). 3rd ed. Reston, VA: American College of Radiology; 1998.

14. American College of Radiology. Breast Imaging Reporting and Data System (BI-RADS). 4th ed. Reston, VA: American College of Radiology; 2003.

Arch Endocrinol Metab. 2017;61/3

207

editorial

Gestational diabetes mellitus and type 2 diabetes: same disease in a different moment of life? Maybe not Lenita Zajdenverg1, Carlos Antonio Negrato2

Serviço de Nutrologia e Diabetes, Departamento de Clínica Médica, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brasil 2 Associação dos Diabéticos de Bauru, Bauru, SP, Brasil 1

Correspondence to: Lenita Zajdenverg Serviço de Nutrologia Rua Professor Rodolpho Paulo Rocco, 255, sala 9E14, 9º andar Cidade Universitária, Ilha do Fundão 21941-913 – Rio de Janeiro, RJ, Brasil lenitazaj@gmail.com Received on May/27/2017 Accepted on May/27/2017

DOI: 10.1590/2359-3997000000276

208

estational diabetes mellitus (GDM) is a medical condition that has motivated many debates in the last decades regarding its etiology, pathophysiology, diagnosis, treatment and long-term consequences to the mother and to the fetus. Women who develop GDM present a metabolic condition similar to that found in type 2 diabetes (T2D) characterized by insulin resistance associated with inadequate insulin secretion (1). Due to similar pathophysiologic mechanisms found between T2D and GDM, there is a great interest in finding markers that will lead to the understanding of a possible common origin to both diseases. Women with GDM also present an inflammatory state that, together with insulin resistance can alter placental gene transcription and many features of fetal programming, that can lead to the development of several metabolic diseases later in life such as glucose intolerance, metabolic syndrome and also a high risk of presenting cardiovascular disease. Women with GDM have sevenfold higher risk of having T2D in the future (2). Identifying risk markers for the development of GDM or for poor perinatal outcomes will allow the implementation of precocious preventive or therapeutic interventions. Recently, several biomarkers have been evaluated in order to establish this possible relationship such as cord blood adiponectin, C-reactive protein (CRP), advanced glycation end products (AGEs) and a variety of genetic polymorphisms. Adiponectin exhibits an anti-inflammatory action and may potentially play a protective role in the development of GDM and T2D. Data regarding the relationship between cord blood levels of adiponectin, newborns birth weight and children adiposity are contradictory, with some studies finding a positive correlation (3) and others not showing any correlation (4). In the present issue of the “Archives of Endocrinology and Metabolism” in a study conducted by Aramesh and cols. in Iran, 52 women with GDM and 52 with normal glucose tolerance (NGT) were evaluated regarding fetal anthropometric parameters, cord blood adiponectin and CRP. It was found that adiponectin levels were higher in the presence of GDM and was also associated with higher birth weight and later gestational ages. The levels of CRP were not different between the two groups (5). This finding contrasts with most studies associating low levels of adiponectin and increased levels of CRP with the risk of progression to T2D (6). Also published in this issue of “Archives of Endocrinology and Metabolism”, Lobo Jr. and cols. performed a study with 442 Euro-Brazilian women of which 225 had GDM and 217 presented NGT. Their study had the objective of evaluating the use of serum AGEs as a screening tool for GDM (7). It is well known that AGEs concentrations are associated with several diseases including type 1 and type 2 diabetes mainly in the presence of diabetes-related chronic complications (8). It is supposed that Arch Endocrinol Metab. 2017;61/3

Gestational and type 2 diabetes

Arch Endocrinol Metab. 2017;61/3

is named GDM, it is known that a small percentage of these patients will require further reclassification. The most common types of monogenetic diabetes are frequently diagnosed by the first time during antenatal follow-up (15). Recently, AnghebemOliveira and cols., which did not find the presence of the gen polymorphisms associated with T2D and obesity, has found in this same Brazilian population of pregnant women a higher frequency of carriers of the polimorphism of the C allele of rs780094 from the glucokinase regulatory protein (GCKR) gen in the group of women with GDM (16). The inclusion, even of a small group of pregnant women probably with monogenetic diabetes, that is not associated with the same pathophysiologic mechanisms of T2D, can interfere with the interpretation of results in studies with a small number of patients. Dysglycemia found in pregnancy can have different origins and complexities that may not be related to T2D, as proposed by the three manuscripts that were published in this periodic. Further studies are still required to find possible markers for GDM and T2D in order to discover the link that may exist or not between the two conditions. Disclosure: Lenita Zajdenverg – Advisory board of Novo Nordisk Brazil, Sanofi Brazil and Lilly Brazil. Carlos Antonio Negrato – No disclosures to declare.

REFERENCES 1. Buchanan TA, Xiang AH. Gestational diabetes mellitus. J Clin Invest. 2005;115(3):485-91. 2. Bellamy L, Casas JP, Hingorani AD, Williams D. Type 2 diabetes mellitus after gestational diabetes: a systematic review and metaanalysis. Lancet. 2009;373(9677):1773-9. 3. Mantzoros CS, Rifas-Shiman SL, Williams CJ, Fargnoli JL, Kelesidis T, Gillman MW. Cord Cord blood leptin and adiponectin as predictors of adiposity in children at 3 years of age: a prospective cohort study. Pediatrics. 2009;123(2):682-9. 4. Mantzoros C, Petridou E, Alexe D, Skalkidou A, Dessypris N, Papathoma E, et al. Serum adiponectin concentrations in relation to maternal and perinatal characteristics in newborns. Eur J Endocrinol. 2004;151(6):741-6. 5. Aramesh MR, Dehdashtian M, Malekian A, ShahAli S, Shojaei K. Relation between fetal anthropometric parameters and cord blood adiponectin and high-sensitivity C-reactive protein in gestational diabetes mellitus. Arch Endocrinol Metab. 2017;61(3):228-32. 6. Chenxiao Liu, Xiu Feng, Qi Li, Ying Wang, Qian Li, Majian Hua. Adiponectin, TNF-α and inflammatory cytokines and risk of type 2 diabetes: A systematic review and meta-analysis. Cytokine. 2016;86:100-9. 7. Lobo Jr JP, Brescansin CP, Santos-Weiss ICR, Welter M, de Souza EM, Rego FGM, et al. Serum Fluorescent Advanced Glycation

209

high oxidative stress conditions are associated with inflammation and the presence of diabetes. According to the authors, in their study, women with GDM had a good glycemic control which could have influenced the final results. They did not find different AGEs concentrations in GDM, possibly due to the mild severity and short duration of hyperglycemia. It is possible that in this environment they do not generate enough serum AGEs to make it possible discriminate between GDM and NGT groups (7). The third study published in this issue refers to a cohort of Euro-Brazilian Caucasians formed by 252 patients, 127 with GDM and 125 with NGT. AnghebemOliveira and cols. evaluated the polymorphisms of several genetic variants that are associated with T2D. The authors studied gene polymorphisms T2Drelated such as fat mass and obesity-associated (FTO), leptin receptor (LEPR), peroxisome proliferatoractivated receptor gamma (PPARγ), and transcription factor 7-like 2 (TCF7L2) (9). These polymorphisms are related to food intake, energy balance, appetite regulation, gene expression transcription, glucose and lipids metabolism, inflammation and proliferation of pancreatic beta cells. Some of TCF7L2 polymorphisms have been found to be associated with GDM in other populations (10-13). The authors have found no relation between these polymorphisms with GDM in this Brazilian studied population (9). The association between previous diagnoses of GDM with high risk of developing T2D is well known (2). Moreover, the screening of diabetes during pregnancy can lead to the discovery of an undiagnosed patient with T2D. Although the reduction in insulin sensitivity and impaired insulin secretion occur similarly in cases of GDM and T2D, the dysglycemia found in GDM is generally transitory and disappears after delivery. However, the evaluation of non-pregnant women with glucose intolerance that participated in the Diabetes Prevention Program Outcomes Study, followed for ten years, has shown that those that had a history of GDM presented an increased risk of developing T2D when compared to those without previous GDM; this finding was independent of age and BMI. Interestingly, the reduction in the progression to T2D was found only in the group that had had GDM and were treated with metformin (14). It is also important to note that although hyperglycemia that is first diagnosed during pregnancy

Gestational and type 2 diabetes

End (F-AGE) products in gestational diabetes patients. Arch Endocrinol Metab. 2017;61(3):233-7.

diabetes mellitus from the International Association of Diabetes and Pregnancy Study Groups. Diabetes. 2010;59(10):2682-9.

8. Yamagishi S. Role of advanced glycation end products (AGEs) and receptor for AGEs (RAGE) in vascular damage in diabetes. Exp Gerontol. 2011;46(4):217-24.

13. Lauenborg J, Grarup N, Damm P, Borch-Johnsen K, Jorgensen T, Pedersen O, et al. Common type 2 diabetes risk gene variants associate with gestational diabetes. J Clin Endocrinol Metab. 2009;94(1):145-50.

9. Anghebem-Oliveira MI, Martins BR, Alberton D, Ramos EAS, Picheth G, Rego FGM. Type 2 diabetes-associated genetic variants of FTO, LEPR, PPARg, and TCF7L2 in gestational diabetes in a Brazilian population. Arch Endocrinol Metab. 2017;61(3):238-48. 10. Pappa KI, Gazouli M, Economou K, Daskalakis G, Anastasiou E, Anagnou NP, et al. Gestational diabetes mellitus shares polymorphisms of genes associated with insulin resistance and type 2 diabetes in the Greek population. Gynecol Endocrinol. 2011;27(4):267-72.

14. Aroda VR, Christophi CA, Edelstein SL, Zhang P, Herman WH, Barrett-Connor E, et al.; Diabetes Prevention Program Research Group. The effect of lifestyle intervention and metformin on preventing or delaying diabetes among women with and without gestational diabetes: the Diabetes Prevention Program outcomes study 10-year follow-up. J Clin Endocrinol Metab. 2015;100(4):1646-53. 15. Chakera AJ, Spyer G, Vincent N, Ellard S, Hattersley AT, Dunne FP. The 0.1% of the population with glucokinase monogenic diabetes can be recognized by clinical characteristics in pregnancy: the Atlantic Diabetes in Pregnancy cohort. Diabetes Care. 2014;37(5):1230-6.

12. Freathy RM, Hayes MG, Urbanek M, Lowe LP, Lee H, Ackerman C, et al. Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study: common genetic variants in GCK and TCF7L2 are associated with fasting and postchallenge glucose levels in pregnancy and with the new consensus definition of gestational

16. Anghebem-Oliveira MI, Webber S, Alberton D, de Souza EM, Klassen G, Picheth G, et al. The GCKR Gene Polymorphism rs780094 is a Risk Factor for Gestational Diabetes in a Brazilian Population. J Clin Lab Anal. 2017 Mar;31(2). doi: 10.1002/jcla. 22035.

11. Shaat N, Lernmark A, Karlsson E, Ivarsson S, Parikh H, Berntorp K, et al. A variant in the transcription factor 7-like 2 (TCF7L2) gene is associated with an increased risk of gestational diabetes mellitus. Diabetologia. 2007;50(5):972-9.

210

Arch Endocrinol Metab. 2017;61/3

original article

Likelihood of malignancy in thyroid nodules according to a proposed Thyroid Imaging Reporting and Data System (TI-RADS) classification merging suspicious and benign ultrasound features Ricardo Luiz Costantin Delfim1, Leticia Carrasco Garcez da Veiga2, Ana Paula Aguiar Vidal2, Flávia Paiva Proença Lobo Lopes3, Mário Vaisman2, Patrícia de Fatima dos Santos Teixeira2

ABSTRACT Objective: The aim of this study was to describe the ultrasound features of benign and malignant thyroid nodules and evaluate the likelihood of malignancy associated with each feature according to the Bethesda System for Reporting Thyroid Cytopathology and histopathology. With this analysis, we propose a new TI-RADS classification system. Materials and methods: The likelihood of malignancy from ultrasound features were assessed in 1413 thyroid nodules according to the Bethesda System for Reporting Thyroid Cytopathology and histopathological findings. A score was established by attributing different weights to each ultrasound feature evaluated. Results: Features positively associated with malignancy in bivariate analysis received a score weight of +1. We attributed a weight of +2 to features which were independently associated with malignancy in a multivariate analysis and +3 for those associated with the highest odds ratio for malignancy (> 10.0). Hence, hypoechogenicity (graded as mild, moderate or marked, according to a comparison with the overlying strap muscle), microcalcification and irregular/microlobulated margin received the highest weights in our scoring system. Features that were negatively associated with malignancy received weights of -2 or -1. In the proposed system a cutoff score of 2 (sensitivity 97.4% and specificity 51.6%) was adopted as a transition between probably benign (TI-RADS 3) and TI-RADS 4a nodules. Overall, the frequency of malignancy in thyroid nodules according to the categories was 1.0% for TI-RADS 3, 7.8% for TIRADS 4a, 35.3% for TI-RADS 4b, and 84.7% for TI-RADS 5. Conclusion: A newly proposed TI-RADS classification adequately assessed the likelihood of malignancy in thyroid nodules. Arch Endocrinol Metab. 2017;61(3):211-21. Keywords Thyroid nodules; TI-RADS; thyroid cancer

Departamento de Endocrinologia, Universidade Federal do Rio de Janeiro (UFRJ). Clínica de Diagnóstico por Imagem (CDPI), Rio de Janeiro, RJ, Brasil 2 Departamento de Endocrinologia, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brasil 3 Departamento de Radiologia, Universidade Federal do Rio de Janeiro (UFRJ), Clínica de Diagnóstico por Imagem (CDPI), Rio de Janeiro, RJ, Brasil 1

Correspondence to: Ricardo Luiz Costantin Delfim Av. Professor Rodolpho Rocco, 255 Hospital Universitário Clementino Fraga Filho Cidade Universitária 21941-913 – Rio de Janeiro, RJ, Brasil dr.ricardodelfim@gmail.com Received on Sept/26/2016 Accepted on Jan/9//2017

INTRODUCTION

he incidence of thyroid nodules has increased 2–4-fold over the past three decades, mainly due to increased use of ultrasound and advancement in ultrasound technology (1,2). According to recent guidelines and recommendations reported by different scientific societies (3-6), ultrasound remains the most important tool in the initial evaluation of thyroid nodules since it has the ability to detect and diagnose potentially malignant thyroid nodules. Several authors (7-12) have proposed different Thyroid Imaging Reporting and Data System (TIRADS) classifications to standardize thyroid ultrasound

Arch Endocrinol Metab. 2017;61/3

reports, as demonstrated with the Breast Imaging Reporting and Data System (BI-RADS®) (13). Researchers have recently attempted to validate this type of approach as an instrument for use in clinical practice, with some authors proposing different TIRADS versions in selected populations (14-20). The first study proposing a TI-RADS classification was published by Horvath and cols. correlating 10 ultrasound patterns with the risk of malignancy in thyroid nodules (7). The study focused on relevant patterns in thyroid nodules with a low likelihood of malignancy and described important features related with benignity. Thereafter, a different classification and 211

DOI: 10.1590/2359-3997000000262

A modified TI-RADS for thyroid nodules

scoring system was proposed (8) using binary logistic regression to assess different odds ratios (OR) for each suspicious feature and generate an equation leading to a final score. The TI-RADS proposed by Kwak and cols. (9) was based on a practical and simplified scoring system to identify suspicious findings; each feature received identical weight in the proposed score, and higher scores were attributed to the occurrence of more than one suspicious ultrasound feature in the same nodule. Russ and cols. (10) proposed in the form of an atlas a classification of thyroid nodules using seven different ultrasound patterns and creating their own TI-RADS categories. A simplified version of this classification, which excluded from the assessment Doppler and elastography, was subsequently created (11). Later, Russ and cols. (12) validated their own proposed classification in 4550 nodules, which was further validated in other 242 nodules (17). In order to improve in own previous classification Kwak and cols. (21) conducted a multicenter study to develop a score attributing different values to each suspicious feature to the final score. In this proposed classification, the authors did not include benign features (21). Until now, none of the proposed TI-RADS classifications has been universally accepted. The latest guidelines on thyroid nodules and differentiated thyroid cancer developed by the ATA (4), proposes a risk classification based on different ultrasound patterns categorized into five groups. In this classification, the risk of malignancy in thyroid nodules increases from < 3% (very low suspicion) to > 70-90% (high suspicion). According to this classification, hypoechoic nodules considered as highly suspicious also display other suspicious features, such as microcalcification or irregular/microlobulated margin. This proposed approach, which is based on groups of ultrasound patterns, facilitates the clinical management of thyroid nodules. However, some nodules do not fall into any of the five proposed pattern groups in the ATA classification (4) (e.g., isoechoic nodules with micro- or macrocalcification). This fact may explain the gap seen in the risk of malignancy, from 20% in thyroid nodules with intermediate ultrasound patterns to 70% in those with a highly suspicious ultrasound pattern. A similar gap has also been reported in the guidelines proposed by the AACE/ACE/AME (5), which included three classes of ultrasound patterns categorized according to risk of malignancy into high, intermediate, and low. 212

The American College of Radiology recently assembled a committee to initiate a process to develop their own TI-RADS. The first step of the committee was to create the Thyroid Ultrasound Reporting Lexicon to describe ultrasound characteristics of thyroid nodules, providing concise written definitions and illustrations to guide practitioners (22). Also recently, the Korean Society of Thyroid Radiology proposed a modification to the TI-RADS system (K-TIRADS) using a flowchart-guided classification according to the presence or absence of different ultrasound features found in thyroid nodules (6). The aim of this study was to describe the ultrasound features of benign and malignant thyroid nodules and evaluate the likelihood of malignancy associated with each feature according to the Bethesda System for Reporting Thyroid Cytopathology (23) and histopathology. With this analysis, we propose a new TI-RADS classification system.

MATERIALS AND METHODS Study design and population We conducted a retrospective, case-control study to analyze the ultrasound features of 1413 thyroid nodules evaluated with FNAB between January 2008 and June 2013 at two institutions (CDPI – Clínica de Diagnóstico por Imagem and Labs D’or, both in Rio de Janeiro, Brazil). The criteria for the selection of the thyroid nodules were based on cytopathological features. All cytopathological samples obtained by FNAB were examined according to the Bethesda classification (23). The selected cases included thyroid nodules exhibiting suspicious or malignant cytopathology (category V or VI), which were then surgically resected and had a confirmatory histopathological report. The control sample included thyroid nodules with a benign cytopathology (category II). Most patients in the control group were followed up, and 6.5% of their nodules were evaluated with a second FNAB with a concordant cytopathology, confirming their benign nature (4,24). A benign status was also established by histopathological assessment in 2.0% of the control nodules. Nodules confirmed as benign were included in a subanalysis; those with confirmatory histopathology or a second FNAB were used as controls and compared with malignant nodules (cases). Nodules presenting any pathological divergence were excluded. Arch Endocrinol Metab. 2017;61/3

A modified TI-RADS for thyroid nodules

Table 1. Standardized definition for ultrasound features of thyroid nodules Ultrasound features Composition

Grade of echogenicity of the solid component

Solid appearance

> 90% of nodule component is solid (24)

Spongiform appearance

Predominantly cystic with multiple degenerative areas (> 50% in its composition) (25)

Hyperechogenicity

Echogenicity greater than thyroid parenchyma (10,11,25,27)

Hypoechogenicity (any degree)

Echogenicity lesser than thyroid parenchyma (28,29), including thyroid nodules with mild*, moderate** and marked *** hypoechogenicity (30)

Moderate to marked hypoechogenicity

Echogenicity similar and lesser than that of strap muscle, including thyroid nodules with moderate** and marked*** hypoechogenicity

Marked hypoechogenicity

Echogenicity lesser than that of strap muscle (30)

Absence of a halo

No identified hypoechogenic halo

Irregular thick halo

Irregular halo, ≥ 2 mm in thickness (26)

Regular thin halo

Complete and regular, < 2 mm in thickness (26)

Irregular/ microlobulated margin

Irregular or microlobulated margins (8,9,21,30)

Blurred margin

Not well defined margin (31)

Microcalcification

Peripheral and/or inner microcalcification, defined by hyperechogenic spot ≤ 2 mm, either with or without acoustic shadow (26)

Macrocalcification

Peripheral and/or inner hyperechogenic coarse or spot > 2 mm, either with or without acoustic shadow (26)

Egg shell calcification

Complete and regular calcification border (5,8,25)

Colloid Crystal

Hyperechogenic spot with comet-tail artifact (birefringence) (29,32)

Unspecific hyperechoic spots

Hyperechogenic spot without acoustic shadow or comet-tail artifact that is not well characterized as calcification or colloid crystal (29,31)

Non-ovoid shape

Anteroposterior diameter greater than its transverse or longitudinal one

Taller-than-wide shape

Anteroposterior diameter greater than its transverse diameter (30)

Any degree of central flow

Any degree of central flow (28,33)

Predominant central flow

Central flow greater than peripheral blood flow and exclusively central flow (34)

Thyroid ultrasound and FNAB evaluations Both ultrasound and FNAB were performed by the same expert radiologist with an experience of over 25 years performing ultrasound and more than 15 years performing FNAB. The ultrasound examinations were performed using different 6–15 MHz linear-array probes and one of the following equipment: HDI 5000 Ultrasound System (Philips Medical System, Bothell, WA, USA), Xario SSA-660A (Toshiba Medical System Corporation), a Logiq 5 Expert (GE Medical System, Milwaukee, WI, USA), or a Logiq E9 (GE Medical System, Milwaukee, WI, USA). After a short interview, the patients underwent a thyroid ultrasound examination followed by FNAB. All procedures were performed under realtime visualization, without an aspirator and with a similar freehand biopsy technique, independent of the institution in which the examination was performed. The ultrasound features of each lesion were meticulously classified immediately after the examination. All cytopathology reports issued prior to the Bethesda report (23) were reviewed by a single pathologist who issued a report based on the new classification system. A random subsample of 5% of the ultrasound recordings was also evaluated by an external researcher with expertise in ultrasound, without prior knowledge of cytopathological reports. A high agreement was observed between the two researchers (kappa = 0.99, p < 0.001). All nodules were evaluated and classified according to the presence of 20 predefined ultrasound features, mostly retrieved from a literature review (4,5,711,21,25-34) and detailed in Table 1. These features were included in a multiple logistic regression analysis to determine whether they were (or not) independently associated with the likelihood of malignancy (4,5,811,24-26,28-30,34). Among all ultrasound features, 13 were likely to be associated with malignancy, as previously described: (i) solid appearance, (ii) Arch Endocrinol Metab. 2017;61/3

Margins and halos

Presence of different hyperechogenic spots, including any kind of calcifications

Shape

Doppler color flow

Definition

* Echogenicity lesser than thyroid parenchyma but greater than of strap muscle (10,12); ** Echogenicity similar to the strap muscle; *** echogenic lesser than of strap muscle, characterizing a marked hypoechogenicity (30).

213

All patients had been referred for FNAB or surgery by their own physicians in an outpatient clinical setting. A minimal nodular size for enrollment in the study was not established. The study, which did not have an interventional design, was approved by the local ethics committee (053560/2012). In addition, all patients signed an informed consent form after receiving a clear explanation of the FNAB procedure, limitations, and possible complications.

A modified TI-RADS for thyroid nodules

hypoechogenicity (any degree [graded as mild, moderate or marked]), according to a comparison with the overlying strap muscle), (iii) moderate to marked hypoechogenicity, (iv) marked hypoechogenicity, (v) presence of peripheral and/or inner microcalcification, (vi) absence of a halo, (vii) irregular thick halo, (viii) irregular/microlobulated margin, (ix) blurred margin, (x) non-ovoid shape, (xi) taller-than-wide shape, (xii) presence of any degree of central blood flow, and (xiii) predominantly central blood flow (i.e., central blood flow alone or more accentuated than the peripheral one). Conversely, the following five ultrasound features were considered to be potentially associated with benign nodules (4,5,8,25,27,29,31,32): (i) a spongiform appearance, (ii) hyperechogenicity, (iii) eggshell calcification, (iv) presence of colloid crystal, and (v) thin regular halo. Indeterminate features for likelihood associations (based on disagreements in the literature) assessed and included in the analysis were (i) peripheral and/or inner macrocalcification and (ii) hyperechoic spot (5,7,8-10,29,31).

Statistical analysis We performed all statistical analyses using the Statistical Package for the Social Sciences (SPSS) for Windows, version 17.0 (IBM). Continuous variables are presented as mean ± standard deviation (SD) (median). We compared these variables between two groups using the Mann-Whitney test. For comparisons among three or more groups, we used the Kruskal-Wallis test. We expressed categorical variables as percentages and compared these variables using the chi-squared test (c2) or Fisher’s exact test in bivariate analysis. Binary logistic regression was applied to determine in a multivariate analysis which specific covariates (ultrasound features) were independently associated with malignancy.

RESULTS We evaluated 1413 thyroid nodules, of which 1174 (83.1%) were classified as category II, 155 (11.0%) as category V, and 84 (5.9%) as category VI according to the Bethesda classification criteria. Overall, 1251 (88.5%) nodules were in women. There was no statistically significant difference between the Bethesda classification and gender (categories II: 89.1%; V: 86.5%; and VI: 84.7%; p = 0.307). Patients with a Bethesda II classification were significantly older (mean ages in each category: II, 52 years; V, 44 years; and VI, 46 years; p < 0.001). 214

We obtained a histopathological analysis of all thyroid nodules with a malignant or suspicious cytopathology (n = 239). We observed a high diagnostic agreement between the cytopathological and histopathological diagnoses (kappa = 0.96; p < 0.001). The histopathological examination confirmed malignancy in 98.7% (153/155) and 98.8% (83/84) of the nodules categorized as V and VI, respectively. Among the benign nodules, a confirmatory diagnosis was obtained in a subgroup of the sample (n = 99; 8.4%) by histopathology (n = 23) or a second FNAB (n = 76).

Associations between ultrasound features and Bethesda cytopathology results Suspicious ultrasound features increased in frequency along with the degree of suspicion on cytopathology (Table 2) and the number of suspicious ultrasound features presented in the thyroid nodules was higher according to the likelihood of malignancy identified on cytopathology (Figure 1). The numbers of suspicious features were 3.7 ± 1.3 and 3.3 ± 1.2 in Bethesda VI and V nodules, respectively. These values were higher (p < 0.001) than those found in Bethesda II nodules (1.06 ± 1.4). The bivariate analysis revealed an association between each ultrasound feature and the likelihood of suspicious/malignant cytopathology (Table 2). The likelihood of confirmed malignancy, obtained by evaluating a subgroup of nodules with a confirmed diagnosis of malignancy, is also presented in Table 2. Eggshell calcification was not detected in any of the thyroid nodules removed by surgery. Table 2 also lists the results of the multivariate analysis, showing features independently associated with reported endpoints (i.e., “suspicious/malignant cytopathology” or “confirmed malignancy”). In a subanalysis including thyroid nodules with a confirmed diagnosis, the same ultrasound features were associated with either an increased or reduced likelihood of malignancy. However, five of the features (i.e., blurred margin, thick irregular halo, colloid crystal, hyperechoic spot, and macrocalcification) were no longer statistically significant. The blurred margin was the sole feature independently and negatively associated with malignancy (Table 2). Albeit, none of the spongiform nodules were malignant, this feature was not independently and negatively associated with the likelihood of malignancy (Table 2). Arch Endocrinol Metab. 2017;61/3

Arch Endocrinol Metab. 2017;61/3

0 7.1 (n = 11)

2.6 (n = 30)

21.9 (n = 256)

7.2 (n = 84)

4.2 (n = 49)

Hyperechogenicity

Thin regular halo

Hyperechoic spot

Macrocalcification

16.7 (n = 14)

9.5 (n = 8)

11.9 (n = 10)

72.3 (n = 60)

18.1 (n = 15)

19.8 (n = 18)

52.4 (n = 44)

1.2 (n = 1)

1.9 (1.3-2.8)

< 0.001

0.05

3.5 (1.0-11.3) 35.6 (22.2-57.1) 1.5 (0.9-2.4) 1.4 (0.9-2.2)

0.09 < 0.001 0.020 0.026 < 0.001 0.001

4.7 (2.9-7.8) 0.98 (0.97-0.98)

0.01

0.97 (0.96-0.98)

0.044 < 0.001

0.03

1.7 (1.1-2.7) 4.0 (2.6-6.4)

< 0.001

0.001

< 0.001

0.34 (0.2-0.55)

0.057

0.04

0.648

0.99 (0.99-1.0) 0.97 (0.96-0.99)

0.062

0.467

0.083

0.001

0.40

0.9 (0.6-1.2)

0.424

0.06

0.58

< 0.001

< 0,001

< 0.001

p value

2.5 (1.6-3.8)

< 0.001

84.5 (40.2-178)

< 0.001

13.5 (9.8-18.6)

8.8 (5.8-13.5)

< 0.001 < 0.001

26.8 (15.8-45.0)

12.3 (7.6-19.9)

Bivariate analysis*

4.01 (2.2-7.4)

2.28 (1.2-5.1)

12.6 (6.3-25.2)

< 0.001

0.045

< 0.001

0.009

< 0.001

28.8 (10.9-75.9) 0.38 (0.2-0.8)

< 0.001

p value

3.45 (2.0-9.0)

4.7 (2.4-9.4)

5.14 (2.8-9.4)

Multivariate analysis*

Benign and suspicious/malignant cytopathology (n = 1413)

2.2 (0.9-5.1)

1.22 (0.5-2.7)

0.25 (0.13-0.49)

0.98 (0.95-1.0)

0.41 (0.03-6.7)

N.E.**

0.98 (0.95-1.0)

3.4 (1.2-9.9)

0.7 (0.3-1.1)

4.1 (1.2-13.8)

2.3 (0.9-5.7)

24.8 (17.6-80.6)

1.0 (1.0-1.01)

3.2 (1.7-5.9)

1.8 (0.9-3.76)

56.1 (7.7-409)

13.2 (6.9-25.2)

9.9 (3.0-32.6)

22.3 (11.7-42.8)

10.2 (4.5-18.9)

Bivariate analysis*

0.122

0.388

0.001

0.08

0.253

N.E.**

0.08

0.02

0.10

0.015

0.07

0.001

0.628

< 0.001

0.147

0.001

< 0.001

p value

3.1 (1.1-8.3)

12.0 (2.6 -54.5)

3.43 (1.3-9.2)

4.3 (1.7-11.1)

3.9 (1.5-9.7)

Multivariate analysis*

0.03

0.01

0.02

0.04

p value

Subgroup of nodules with histopathological assessment (n = 338)

Relationship each ultrasound feature with cytopathological and histopathological assessments

< 0.001

p value

# Bethesda system categories included in analysis. * Odds ratio (confidence interval). ** N.E.: Not evaluated.

14.2 (n = 22)

12.9 (n = 20)

0.2 (n = 2)

2.3 (n = 27)

14.2 (n = 22)

Colloid Crystal

2.0 (n = 24)

Spongiform appearance

66.0 (n = 95)

39.0 (n = 60)

Eggshell calcification

3.2 (n = 37)

Predominantly central flow

9.0 (n = 14)

9.0 (n = 104)

71,1 (n = 820)

Non-ovoid shape

8.7 (n = 101)

Taller-than-wide

Any degree of central flow

8.0 (n = 14)

2.1 (n = 25)

Microcalcification

2.6 (n = 4)

86.9 (n = 73)

0.6 (n = 7)

27.4 (n = 23)

90.9 (n = 140)

40.5 (n = 34)

22.6 (n = 19)

73.5 (n = 61)

93.0 (n = 78)

16.6 (n = 25)

34.8 (n = 54)

Irregular/thick halo

0.7 (n = 8)

Irregular/microlobulated margin

26.5 (n = 41)

77.5 (n = 504)

3.7 (n = 43)

Moderate to marked hypoechogenicity

63.0 (n = 97)

Absence of a halo

12.9 (n = 149)

Marked hypoechogenicity

93.5 (n = 145)

92.0 (n = 77)

92.3 (n = 143)

11.7 (n = 136)

34.2 (n = 401)

Hypoechogenicity (any degree)

Blurred margin

48.5 (n = 567)

Bethesda system categories #

Frequency distribution of ultrasound features according to the Bethesda system (%)

Solid appearance

Ultrasound features

Table 2. Statistical analysis

A modified TI-RADS for thyroid nodules

215

A modified TI-RADS for thyroid nodules

20 p = 0.541 02

Proposed system scoring

Suspicious ultrasound features

6 4 2 0

0 p < 0.001

p < 0.001

-2

-10

II V VI Bethesda system categories

Sensibility

C 1.00 0.75 0.50 0.25 0.00

AUC = 0.921* 0.25

0.50 Specificity

0.75

1.00

Figure 1. Blox pot graphs and receiver operating characteristics (ROC). A. Distribution of suspicious ultrasound features by Bethesda system categories. B. Distribution of proposed system scoring by Bethesda system categories. C. ROC curve was applied to determine the best cut off with high sensitivity and specificity for the highest risk categories of maligancy in the proposed score system. * AUC: area under the curve.

TI-RADS scoring We developed a scoring system based on the logistic multiple regression analysis and different weights assigned to each feature according to their association with the likelihood of malignancy on cytopathology, thoroughly detailed in Table 3. Features that were positively but not independently associated with a likelihood of malignancy received a weight of +1; these features included macrocalcification, non-ovoid shape, absence of a halo, and thick irregular halo. Features independently associated with a likelihood of malignancy that received a weight of +2 included a solid appearance, predominantly central flow, hyperechoic spot, hypoechogenicity (any degree), and moderate to marked hypoechogenicity. The presence of microcalcification and an irregular/microlobulated margin received a weight of +3 since their OR were the highest (> 10.0) compared with those of other features. Blurred margin, a feature independently associated with a benign status, received a weight of -2. Features that were associated with a benign status (but 216

which the association did not emerge as independent in multivariate analysis) included a spongiform appearance, colloid crystal, hyperechogenicity, and a thin and regular halo. These last four features received a weight of -1 in our scoring system. In terms of different grades of hypoechogenicity (Figure 2), we detected that nodules with much lower echogenicity had higher scores in our proposed scoring system. Marked hypoechogenicity was comprised in categories of thyroid nodules that presented hypoechogenicity of any degree (+2) and also in categories of moderate to marked hypoechogenicity (+2), besides the addition +1 (initial own score), total was +5 for these findings. Moderate hypoechogenicity was included in any degree hypoechogenicity (+2) plus the score of moderate to marked hypoechogenicity (+2), total +4 score. Mild hypoechogenicity, was assigned a final score +2 because it was not comprised neither moderate hypoechogenicity nor marked hypoechogenicity. This conceiving process is showed on Table 3. Arch Endocrinol Metab. 2017;61/3

A modified TI-RADS for thyroid nodules

Table 3. Weight conception process for ultrasound features scoring Ultrasound features

Weight conception process

Score

Marked hypoechogenicity

This feature alone was not independently associated with a likelihood of malignancy in multivariate analysis and received initially a score weight of +1. However, the weight of this feature increased since it is also included in the feature of hypoechogenicity of any degree (+2) and in moderate to marked hypoechogenicity (+2). The sum of all these weights resulted in the value of +5, attributed here

Moderate hypoechogenicity

The presence of moderate to marked hypoechogenicity was independently associated with the likelihood of malignancy and received a weight of +2. However, the weight of this feature increased it is also included in the feature of hypoechogenicity of any degree (+2), yielding a score weight of +4. Moderate to marked hypoechogenicity term was replaced to “moderate hypoechogenicity”, as marked hypoechogenicity has its own score

Independently associated with the likelihood of malignancy (OR > 10.0)

Microcalcification Irregular/microlobu lated margin Mild hypoechogenicity

Independently associated with the likelihood of malignancy (OR > 10.0)

This degree of hypoechogenicity received a weight based only on the feature of hypoechogenicity of any degree which did not meet the criteria for moderate or marked hypoechogenicity; it was then attributed a weight of +2 since it was included in the overall group of any degree hypoechoic nodules

Independently associated with the likelihood of malignancy (OR > 1.0 and ≤ 10.0)

Solid appearance Undefined hyperechoic spot

Independently associated with the likelihood of malignancy (OR > 1.0 and ≤ 10.0)

Predominantly central flow

Independently associated with the likelihood of malignancy (OR > 1.0 and ≤ 10.0)

Ultrasound features positively associated with the likelihood of malignancy in bivariate but not multivariate analysis

Non-ovoid shape Macrocalcification Absence of a halo

Irregular/thick halo Regular thin halo Crystal colloid

+1 +1

Ultrasound features negatively associated with the likelihood of malignancy in bivariate but not multivariate analysis

Hyperechogenicity

-1 -1

Spongiform appearance Blurred margin

-1

-1 Negatively and independently associated with the likelihood of malignancy

-2

OR: odds ratio.

Hypoechogenicity (any degree)* Moderate to marked hypoechogenicity* Marked hypoechogenicity*

B C Figure 2. Hypoechogenicity gradation in thyroid nodules. Ultrasound images exemplify nodules that exhibit three grades of hypoechogenicity: (arrows). A. Mild hypoechogenicity: nodule presents echogenicity lesser than thyroid parenchyma and greater than the strap muscle (arrow). B. Moderate hypoechogenicity: nodule presents echogenicity similar to the strap muscle (arrow). C. Marked hypoechogenicity: nodule presents echogenicity lesser than the strap muscle (arrow). * Included in bivariate and multivariate analysis. Arch Endocrinol Metab. 2017;61/3

217

A modified TI-RADS for thyroid nodules

Table 4. Propose TI-RADS categories TI-RADS 1: Negative TI-RADS 2: Benign* TI-RADS 3 (final score ≤ 2): Probably benign TI-RADS 4a (final score 3–5): Low suspicion for malignancy TI-RADS 4b (final score 6–9): Moderate suspicion for malignancy TI-RADS 5 (final scores ≥ 10): Highly suggestive of malignancy Suggestion 1: Investigate initially nodules ≥ 10 mm categorized as TI-RADS 4a. For those ≥ 5–10 mm, in the highest category, consider the patient’s decision before starting to investigate the nodule. Since, a follow-up is acceptable until the nodule achieves 10 mm, when it will then require investigation (3,4) Suggestion 2: Investigate nodules with associated abnormal lymph nodes or potentially aggressive signs (paratracheal nodules, subcapsular location, or local invasion) Suggestion 3: Consider a nodule in the next superior category if its growth rate or the patient’s personal/family history suggests a high risk of malignancy (3-6,10,18) Suggestion 4: Consider the solid part of predominantly cystic nodules with an eccentric solid area as being a solid nodule and apply the score * Simple cyst (purely anechoic content with thin, regular wall), in spite of this kind of nodule was not analyzed in our sample, it is the only one related to benignity, without any need to continue diagnostic investigation.

Overall, the scores assigned to Bethesda category V and VI nodules were higher than those assigned to category II nodules (2.6 ± 2.5 [2.0] vs. 8.8 ± 3.18 [9.0]; p < 0.001) (Figure 1). The Receiver Operating Characteristic (ROC) curve (Figure 1) yielded an area under the curve of 0.921 (CI 95%): 0.901–0.941) and demonstrated that a score of 5 reflected the best combined sensitivity (82.0%) and specificity (87.6%), as the cutoff point between the categories of low suspicion (TI-RADS 4a) and moderate to high suspicion for malignancy. The selected cutoff score that separated the category of highly suggestive of malignancy (TI-RADS 5) from low/moderate categories (TI-RADS 4b) was 9, which was the median score obtained for Bethesda category V and VI nodules (Figure 1). On the other hand, nodules scoring 2 were classified as probably benign; this was selected as the cutoff score between TI-RADS 3 (probably benign) and TI-RADS 4a (low suspicion), as shown in Table 4, and represents a value with high sensitivity (97.4%) but reduced specificity (51.6%), as shown in (Figure 1). Overall, the frequency of malignancy in thyroid nodules according to the categories was 1.0% for TI-RADS 3, 7.8% for TI-RADS 4a, 35.3% for TI-RADS 4b, and 84.7% for TI-RADS 5. By adopting these proposed criteria for our proposed TI-RADS, the frequency of malignant or suspicious cytopathology becomes very similar to that reported by the American College of Radiology for BI-RADS and prior Thyroid Imaging Reporting and Data System researches (7-13).

DISCUSSION In this study, we observed an association between categories of a newly proposed TI-RADS and the likelihood of malignancy. This finding is similar to that 218

reported for the well-established BI-RADS concerning breast cancer. Additionally, our results are comparable to other TI-RADS classifications and are in accordance with recent guidelines classifications (4-6). Our study has quantified the ultrasound features in thyroid nodules by giving different weights to each feature positively or negatively associated with the likelihood of malignancy. We found that all nodules with echogenicity lower than or similar to that of the overlying strap muscles were independently associated with malignancy. However, those thyroid nodules with marked hypoechogenicity received higher scores in our proposed scoring system. Due to that, we divided the feature of hypoechogenicity into degrees and found that marked hypoechogenicity played an important role in our proposed scoring system (Figure 2). Comparisons between the echogenicity of the nodule with that of the overlying strap muscles can improve cancer detection, especially in the context of thyroiditis, in which the thyroid parenchyma exhibits reduced echogenicity. In support of our results, the presence of calcifications has been found to increase the likelihood of malignancy in different studies (29), particularly the presence of microcalcification. Since the size of microcalcifications has been reported to range from 0.5–3.0 mm in different studies (8-11,21;25,26,30), one should expect an overlap between micro- and macrocalcifications. However, macrocalcification as a possible suspicious feature has not been included in previous TI-RADS classifications (7,9,11,12,17). It is important to note that presence of macrocalcification is generally associated with an increased risk of malignancy (5,25,29). Additionally, it can be difficult to distinguish microcalcification from colloid crystal in the absence of Arch Endocrinol Metab. 2017;61/3

a comet-tail artifact; this prevents the identification of colloid crystal, which typically correlates with benign nodules (5,29,32). In uncertain cases, it is appropriate to use the term “hyperechoic spot”; this feature may be associated with malignancy, as observed in this study and also in other previous reports (31). A non-ovoid or nonparallel shape (i.e., a tall nodule) was also associated with the likelihood of malignancy in this study, which is consistent with previous reports (8,9,21,25,30). Furthermore, the relationship between height and longitudinal measurement, in addition to transverse measurement, was useful in this analysis. However, the taller-than-wide shape did not exhibit the same degree of association with malignancy compared with other ultrasound features proposed by Kim and cols. (30), a finding that is consistent with that reported by Russ and cols. (10). We included Doppler flow analysis in this proposed TI-RADS, as done in other studies (7,10,17). Previously, the detection of any degree of internal blood flow was positively related to an increased likelihood of malignancy (17,28,33). However, in our study, this finding was not a useful predictor of malignancy. Only predominant central blood flow was found to be an independent factor associated with the likelihood of malignancy. Similar results regarding the vascularity of thyroid nodules have been reported (26). In our sample, the presence of blurred margin was identified as an independent factor for benignity, as previously reported, based on its association with Hashimoto’s thyroiditis and benign nodules (31). These results reinforced the idea that a high number of pseudo-nodules in patients with Hashimoto’s thyroiditis may have been aspirated in the control group. Unlike blurred margins, irregular/microlobulated margins were found to be an important feature related to the likelihood of malignancy, which is consistent with findings of previous studies (3-12,21;25-30). Most suspicious features were not present in a single nodule; conversely, benign and malignant features may overlap (29). All features positively and negatively associated with the likelihood of malignancy – which may be present in the same nodule – should be evaluated to yield an overall score. Previous authors have also evaluated the benign features of thyroid nodules (7,8,10,12,17). However, we attributed different weights to benign and malignant features, which resulted in a new and unique score, unlike the risk score for malignancy created by Kwak and cols. Arch Endocrinol Metab. 2017;61/3

(21). Therefore, a separate evaluation of the findings, as done in prior studies (9,21), is a reliable and better way to predict malignancy than growth rate alone (24), in long-term follow up of thyroid nodules. In this study, as well as in others (9,21,29), a combination of suspicious findings increased the likelihood of malignancy. Moreover, a single feature with a high OR has been found to correlate more strongly with the likelihood of malignancy compared with the manifestation of two minor features (9). Likewise, the presence of features less related to malignancy should not be overlooked. In light of these considerations and our results, spongiform nodules, in the absence of other suspicious features, should not require FNAB. These nodules are associated with a very low risk for malignancy, as previously demonstrated by other researchers (7,25,27,29). A limitation of this study was the inclusion of limited Bethesda categories since we only evaluated thyroid nodules classified as Bethesda II, V, or VI. The selection criteria based on cytopathology may also have led to the exclusion of follicular carcinomas from our analyses since cytopathology alone is unable to confirm this diagnosis. Even so, a predominantly central flow was a relevant suspicious feature in our scoring system and is a useful predictor of malignancy in follicular neoplasms (5,34,35). In addition, papillary carcinomas are currently the most prevalent differentiated thyroid carcinomas (4), and cytopathology remains the most important tool in the decision to refer patients to surgery. We did not include elastography in our analysis, which may also be a limitation of this study. However, elastography was also not included in several prior classifications (7-9,11,17), or in the latest ATA guidelines (4). An additional limitation of this study was the low rate of histopathological confirmation among nodules characterized as benign on cytopathology. However, this limitation has also plagued previous studies for ethical reasons (7-12,17,18). In contrast, our subanalysis including only control thyroid nodules with a confirmed histopathology or a second FNAB strengthened our results. Nodules with two benign cytopathological results are associated with a 100% chance of benignity, as previously reported (4,24). Important strengths of our study include the fact that all examinations were conducted by a single radiologist, as reported in a previous study (10). 219

A modified TI-RADS for thyroid nodules

Moreover, we indirectly assessed reproducibility by analyzing the agreement between two ultrasound specialists in a subgroup of randomly selected nodules. In conclusion, this newly proposed TI-RADS involves the quantification of ultrasound features positively and negatively associated with malignancy, with different values attributed to each of these features. We reported the likelihood of malignancy based on cytopathology for different categories of the classification and achieved an adequate association. Additional studies are necessary to validate our findings. Financial source: we have no competing financial interests. Acknowledgements: we would like to thank the Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1.

Dean SD, Gharib H. Epidemiology of thyroid nodules. Best Pract Res Clin Endocrinol Metab. 2008;22(6):901-11.

2. Li N, Du XL, Reitzel LR, Xu L, Sturgis EM. Impact of enhanced detection on the increase in thyroid cancer incidence in the United States: review of incidence trends by socioeconomic status within the surveillance, epidemiology, and end results registry, 1980-2008. Thyroid. 2013;23(1):103-10. 3. Rosário PW, Ward LS, Carvalho GA, Graf H, Maciel RM, Maia AL, et al. Thyroid nodule and differentiated thyroid cancer: update on the Brazilian consensus. Arq Bras Endocrinol Metab. 2013;57(4):240-64. 4. Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016;26(1):1-133. 5. Gharib H, Papini E, Garber JR, Duick DS, Harrell RM, Hegedüs L, et al.; American Association of Clinical Endocrinologists, American College of Endocrinology, and Associazione Medici Endocrinologi Medical Guidelines for Clinical Practice for the Diagnosis and Management of Thyroid Nodules--2016 Update. Endocr Pract. 2016;22(5):622-39.

6. Shin JH, Baek JH, Chung J, Ha EJ, Kim JH, Lee YH, et al.; Korean Society of Thyroid Radiology (KSThR) and Korean Society of Radiology. Ultrasonography Diagnosis and Imaging-Based Management of Thyroid Nodules: Revised Korean Society of Thyroid Radiology Consensus Statement and Recommendations. Korean J Radiol. 2016;17(3):370-95. 7.

Horvath E, Majlis S, Rossi R, Franco C, Niedmann JP, Castro A, et al. An ultrasonogram reporting system for thyroid nodules stratifying cancer risk for clinical management. J Clin Endocrinol Metab. 2009;94(5):1748-51.

8. Park JY, Lee HJ, Jang HW, Kim HK, Yi JH, Lee W, et al. A proposal for a thyroid imaging reporting and data system for ultrasound features of thyroid carcinoma. Thyroid. 2009;19(11):1257-64.

220

9. Kwak JY, Han KH, Yoon JH, Moon HJ, Son EJ, Park SH, et al. Thyroid imaging reporting and data system for US features of nodules: a step in establishing better stratification of cancer risk. Radiology. 2011;260(3):892-9. 10. Russ G, Bigorgne C, Royer B, Rouxel A, Bienvenu-Perrard M. The Thyroid Imaging Reporting and Data System (TIRADS) for ultrasound of the thyroid. J Radiol. 2011;92(7-8):701-13. 11. Moifo B, Takoeta EO, Tambe J, Blanc F, Fotsin JG. Reliability of Thyroid Imaging Reporting and Data System (TIRADS) Classification in Differentiating Benign from Malignant Thyroid Nodules. Open J Radiology. 2013;3(3):103-7. 12. Russ G, Royer B, Bigorgne C, Rouxel A, Bienvenu-Perrard, Leenhardt L. Prospective evaluation of thyroid imaging reporting and data system on 4550 nodules with and without elastography. Eur J Endocrinol. 2013;168(5):649-55. 13. D’Orsi CJ, Sickles EA, Mendelson EB, Morris EA, Dershaw DD. ACR BI-RADS® Atlas, Breast Imaging Reporting and Data System. Reston, VA, American College of Radiology, 2013. 14. Na DG, Baek JH, Sung JY, Kim JH, Kim JK, Choi YJ, et al. Thyroid Imaging Reporting and Data System Risk Stratification of Thyroid Nodules: Categorization Based on Solidity and Echogenicity. Thyroid. 2016;26(4):562-72. 15. Friedrich-Rust M, Meyer G, Dauth N, Berner C, Bogdanou D, Herrmann E, et al. Interobserver agreement of Thyroid Imaging Reporting and Data System (TIRADS) and strain elastography for the assessment of thyroid nodules. PLoS One. 2013;8(10):e77927. 16. Schenke S, Rink T, Zimny M. TIRADS for sonographic assessment of hypofunctioning and indifferent thyroid nodules. Nuklearmedizin. 2015;54(3):144-50. 17. Maia FF, Matos PS, Pavin, EJ, Zantut-Wittmann DE. Thyroid imaging reporting and data system score combined with Bethesda system for malignancy risk stratification in thyroid nodules with indeterminate results on cytology. Clin Endocrinol (Oxf). 2015;82(3):439-44. 18. Zhang J, Liu BJ, Xu HX, Xu JM, Zhang YF, Liu C1, Wu J, et al. Prospective validation of an ultrasound-based thyroid imaging reporting and data system (TI-RADS) on 3980 thyroid nodules. Int J Clin Exp Med. 2015;8(4):5911-7. 19. Yoon JH, Lee HS, Kim EK, Moon HJ, Kwak JY. Malignancy risk stratification of thyroid nodules: comparison between the thyroid imaging reporting and data system and the 2014 American Thyroid Association Management Guidelines. Radiology. 2016;278(3):917-24. 20. Chandramohan A, Khurana A, Pushpa BT, Manipadam MT, Naik D, Thomas N, et al. Is TIRADS a practical and accurate system for use in daily clinical practice? Indian J Radiol Imaging. 2016;26(1):145-52. 21. Kwak JY, Jung I, Baek JH, Baek SM, Choi N, Choi YJ, et al.; Korean Society of Thyroid Radiology (KSThR); Korean Society of Radiology. Image reporting and characterization system for ultrasound features of thyroid nodules: multicentric Korean retrospective study. Korean J Radiol. 2013;14(1):110-7. 22. Grant EG, Tessler FN, Hoang JK, Langer JE, Beland MD, Berland LL, et al. Thyroid Ultrasound Reporting Lexicon: White Paper of the ACR Thyroid Imaging, Reporting and Data System (TIRADS) Committee. J Am Coll Radiol. 2015;12(12 Pt A):1272-9. 23. Cibas ES, Ali SZ. The Bethesda System for reporting thyroid cytopathology. Thyroid. 2009;19(11):1159-65. 24. Kwak JY, Koo H, Youk JH, Kim MJ, Moon HJ, Son EJ, et al. Value of US correlation of a thyroid nodule with initially benign cytologic result. Radiology. 2010 Jan;254(1):292-300. 25. Moon WJ, Jung SL, Lee JH, Na DG, Baek JH, Lee YH, et al.; Thyroid Study Group, Korean Society of Neuro- and Head and Neck Radiology. Benign and malignant thyroid nodules: US differentiation-multicenter retrospective study. Radiology. 2008;247(3):762-70. Arch Endocrinol Metab. 2017;61/3

A modified TI-RADS for thyroid nodules

26. Chammas MC, Gerhard R, De Oliveira IR, Widman A, de Barros N, Durazzo M, et al. Thyroid nodules: evaluation with power Doppler and duplex Doppler ultrasound. Otolaryngol Head Neck Surg. 2005;132(6):874-82.

31. Anderson L, Middleton WD, Teefey SA, Reading CC, Langer JE, Desser T, et al. Hashimoto Thyroiditis: Part 1, sonographic analysis of the nodular form of Hashimoto Thyroiditis. AJR Am J Roentgenol. 2010;195(1):208-15.

27. Bonavita JA, Mayo J, Babb J, Bennett G, Oweity T, Macari M, et al. Pattern recognition of benign nodules at ultrasound of the thyroid: which nodules can be left alone? AJR Am J Roentgenol. 2009;193(1):207-13.

32. Ahuja A, Chick W, King W, Metreweli C. Clinical significance of the comet-tail artifact in thyroid ultrasound. J Clin Ultrasound. 1996;24(3):129-33.

28. Papini E, Guglielmi R, Bianchini A, Crescenzi A, Taccogna S, Nardi F, et al. Risk of malignancy in nonpalpable thyroid nodules: predictive value of ultrasound and color-Doppler features. J Clin Endocrinol Metab. 2002;87(5):1941-6.

33. Remonti LR, Kramer CK, Leitão CB, Pinto LCF, Gross JL. Thyroid ultrasound features and risk of carcinoma: a systematic review and meta-analysis of observational studies. Thyroid. 2015;25(5):538-50. 34. Iared W, Shigueoka DC, Cristófoli JC, Andriolo R, Atallah AN, Ajzen SA, et al. 2010 Use of color Doppler ultrasonography for the prediction of malignancy in follicular thyroid neoplasms: systematic review and meta-analysis. J Ultrasound Med. 2010;29(3):419-25.

30. Kim EK, Park CS, Chung WY, Oh KK, Kim DI, Lee JT, et al. New sonographic criteria for recommending fine-needle aspiration biopsy of nonpalpable solid nodules of the thyroid. AJR Am J Roentgenol. 2002;178(3):687-91.

35. Nicola H, Szejnfeld J, Logullo AF, Wolosker AMB, Souza LRMF, Chiferi V. Flow pattern and vascular resistive index as predictors of malignancy risk in thyroid follicular neoplasms. J Ultrasound Med. 2005;24(7):897-904.

29. Frates MC, Benson CB, Charboneau JW, Cibas ES, Clark OH, Coleman BG, et al.; Society of Radiologists in Ultrasound. Management of thyroid nodules detected at US: Society of Radiologists in Ultrasound Consensus Conference Statement. Radiology. 2005;237(3):794-800.

Arch Endocrinol Metab. 2017;61/3

221

original article

Deaths related to differentiated thyroid cancer: a rare but real event Ana Kober N. Leite1, Beatriz G. Cavalheiro1, Marco Aurélio Kulcsar1, Ana de Oliveira Hoff2, Lenine G. Brandão3, Claudio Roberto Cernea3, Leandro L. Matos1

ABSTRACT Divisão de Cirurgia de Cabeça e Pescoço, Faculdade de Medicina da Universidade de São Paulo (FMUSP) e Instituto do Câncer do Estado de São Paulo (ICESP), São Paulo, SP, Brasil 2 Departamento de Endocrinologia, Faculdade de Medicina da Universidade de São Paulo (FMUSP) e Instituto do Câncer do Estado de São Paulo (ICESP), São Paulo, SP, Brasil 3 Divisão de Cirurgia de Cabeça e Pescoço, Faculdade de Medicina da Universidade de São Paulo (FMUSP) 1

Correspondence to: Ana Kober Nogueira Leite Av. Dr. Enéas de Carvalho Aguiar, 255, 8º andar, sala 8174 05403-000 – São Paulo, SP, Brasil kober.ana@gmail.com

Objective: The present study describes the clinical and tumor characteristics of patients that died from differentiated thyroid cancer and reports on the cause and circumstances of death in these cases. Subjects and methods: Retrospective analysis of all the differentiated thyroid cancer (DTC) related deaths at a single institution over a 5-year period, with a total of 33 patients. Results: Most of the patients were female (63.6%), with a mean age at diagnosis of 58.2 years. The most common histologic type was papillary (66.7%) and 30.3% were follicular. The distribution according to the TNM classification was: 15.4% of T1; 7.7% T2; 38.4% T3; 19.2% of T4a and 19.2% of T4b. Forty-four percent of cases were N0; 20% N1a and 36.6% of N1b. Twelve patients were considered non-responsive to radioiodine. Only one of the patients did not have distant metastases. The most common metastatic site was the lung in 69.7%. The majority of deaths were due to pulmonary complications related to lung metastases (17 patients, 51.5%), followed by post-operative complications in 5 cases, neurological disease progression in 3 cases, local invasion and airway obstruction in one patient. Median survival between diagnosis and death was reached in 49 months while between disease progression and death it was at 22 months. Conclusion: Mortality from DTC is extremely rare but persists, and the main causes of death derive from distant metastasis, especially respiratory failure due to lung metastasis. Once disease progression is established, median survival was only 22 months. Arch Endocrinol Metab. 2017;61(3):222-7. Keywords Thyroid; thyroid cancer; metastasis; prognosis; death

Received on Mai/28/2016 Accepted on Dec/23/2016 DOI: 10.1590/2359-3997000000261

INTRODUCTION

hyroid cancer (TC) is the most common endocrine malignancy and the fifth most common cancer diagnosed in women. It has been reported that its incidence has the largest annual increase in men and women amongst all cancers in the United States (1). In Brazil, it is estimated that during 2016, 1,090 new cases will be diagnosed in men and 5,870 in women, making it the 8th most common cancer in women (2). The vast majority of TCs (> 90%) originate from follicular cells and are defined as differentiated thyroid cancers (DTC) and the two histological subtypes are the papillary TC with its variants and the follicular TC (3). Differentiated thyroid cancer is usually an indolent disease that with adequate treatment has an excellent prognosis (3). The literature reports that less than 5% of patients die from the disease within 10 years (4-6). However few, there are deaths related to DTC, but the 222

rarity of this event and the long course of the disease makes it hard to analyze and determine specific risk factors for this outcome (7). The aim of the present study was to describe the clinical and tumor characteristics of patients that died from DTC and to report the cause and circumstances of death in these cases.

SUBJECTS AND METHODS We conducted a retrospective cohort study at Instituto do Câncer do Estado de São Paulo (ICESP) amongst the 1,114 patients treated for thyroid cancer from January 2009 to November 2015. The inclusion criteria were patients with DTC that died from disease progression or complications directly related to cancer treatment. Thirty-three cases (3.0%) were identified and included in this study. The study was approved by the Institutional Review Board under the number 176/15. Arch Endocrinol Metab. 2017;61/3

Differentiated thyroid cancer and death

RESULTS Amongst the 33 patients that died from DTC, the majority (63.6%) were female, and the absolute majority (94.0%) of cases were diagnosed in individuals older than 45 years, with a mean age at diagnoses of 58.2 ± 12.0 years. The most common histologic type of tumor was the papillary thyroid cancer, responsible for 22 cases and 66.7% of deaths. There were 10 cases of follicular cancer and one case of not specified differentiated thyroid cancer. Considering TNM classification, initially, most patients had T3 tumors (10 cases; 38.4%), follow by T4a and T4b with 19.2% each, four cases (15.4%) of T1 and two cases of T2 tumors. Multifocal disease was Arch Endocrinol Metab. 2017;61/3

found in 50% of patients and mean tumor size was 4.7 ± 3.4 cm (minimum of 0.5 cm and maximum of 13.5 cm). Forty-four percent were N0, 20% N1a and 36% N1b. The descriptive data of the patients included in the study are described in Table 1. Thirty-one patients were submitted to surgical treatment initially; only two cases were not because were considered inoperable from the moment of diagnosis. Thirteen patients were submitted to total thyroidectomy; eight had total thyroidectomy and central node dissection; eight had total thyroidectomy, central and lateral node dissection and two partial thyroidectomies as the initial treatment. Twenty-two patients (66.7%) received radioactive iodine (RAI) therapy, with an average of 499.4 mCi ± 264.8 in total dose (minimum of 50mCi and maximum of 1200 mCi). Twelve cases were considered unresponsive to RAI throughout the treatment. The mean stimulated thyroglobulin was 19,755 ± 5,850 ng/mL and just one case had negative thyroglobulin during the follow-up. Table 1. Descriptive data of patients with differentiated thyroid carcinoma that died from the disease (N = 33) Variable

N (%)

Age ≥ 45 years

31 (94.0)

< 45 years Mean ± SD

2 (6.0) 58.2 ± 12.0

Gender Male Female

12 (36.4) 21 (63.6)

Histologic type Papillary Follicular

22 (66.7) 10 (30.3)

T stage T1 T2 T3 T4a T4b

4 (15.4) 2 (7.7) 10 (38.4) 5 (19.2) 5 (19.2)

N stage N0 N1a N1b

From these charts data was collected and analyzed based on patient’s demographics, tumor characteristics (histologic type, size, extra-thyroid spread, recurrence, vascular invasion, node status, extra-capsular spread), treatment (surgical details, use of radioiodine, externalbeam radiation, chemotherapy), outcome (recurrence, distant metastases, cause of death and the period for each outcome). Follow-up was done according to institutional protocol: clinical examination, laboratorial analyses of thyroid hormones, thyroglobulin and anti-thyroglobulin anti-bodies every six months; neck ultrasound, chest radiography and whole body scan (selected cases) annually. Locoregional recurrence was determined when confirmed by cytological or histological analyses whereas, in distant metastasis determination, suggestive imaging studies were accepted. The disease was considered refractory to radioiodine (RAI) therapy acording to the definition in the 2015 American Thyroid Association Guidelines: the malignant or metastatic tissue did not ever concentrate RAI, the tumor tissue lost the ability to concentrate RAI following previous evidence of RAI-avid disease, the RAI was concentrated in some lesions but not in others, and the metastatic disease progressed despite significant concentration of RAI. For statistical analysis, SPSS® version 17.0 (SPSS® Inc; Illinois, USA) was used. The values obtained from the study of each continuous variable were described by means and standard-deviation (SD) and also by median and 95% confidence interval (95%CI) and relative. Absolute and relative frequencies were used to describe qualitative data. The Kaplan-Meier method was employed for survival analysis.

9 (36.0)

Multifocality Yes No

9 (50.0) 9 (50.0)

N: number of cases.

223

Differentiated thyroid cancer and death

Sixteen patients developed nodal recurrence and fourteen of these were submitted to a new surgical procedure. Thirty-two patients presented distant metastasis. Of these, in 42.4% the metastasis was diagnosed at the same time as the primary tumor. The most common site of metastasis was the lung (69.7%), followed by bone in 66.7%. Two patients had liver metastasis and two had kidney metastasis, but all of these also had lung spread as well. Forty-two percent of cases had metastasis in more than one site. During disease progression nine cases were considered to have dedifferentiation from the initial tumor. Only eight patients did not receive treatment specifically for the metastasis or its complications. Ten were submitted to surgery, six received sorafenib as part of a clinical trial, eleven underwent radiation therapy and six chemotherapy. When analyzing the specific cause that lead to their deaths, it was established that the majority of patients

0.8

0.6

Cumulative survival

B 1.0

Cumulative survival

A 1.0

0.4

0.2

0.0

0.2

60 120 180 Time between diagnosis and metastasis (months)

240

0.0

0.8

0.6

60 120 180 240 Time between diagnosis and disease progression (months)

0.4

0.2

0.0

Cumulative survival

D 1.0

Cumulative survival

C 1.0

0.4

died from respiratory failure related to their lung metastasis (17 patients, 51.5%). Other causes were post-operative complications in 5 cases, neurological disease progression in 3 cases (central nervous system invasion), local invasion and airway obstruction in one patient and other causes in 7 patients. The absolute majority of patients (85.3%) died either from progression of their metastasis or complications derived directly from its treatment. The mean time between diagnosis and death was of 71.4 months (median 49 months, minimum of 4 months and maximum of 279 months). The survival analysis (Figure 1) revealed that the great majority of patients have low cumulative survival in the first 5 years of follow-up. Median survival between diagnosis and first evidence of metastatic disease was reached in 9 months (CI95%: 0.0 â&#x20AC;&#x201C; 20.9 months), between the diagnosis and disease progression in 21 months (CI95%: 9.9 â&#x20AC;&#x201C; 32.1 months), between diagnosis and

0.2

60 120 240 180 Time between diagnosis and death (months)

300

0.0

24 48 60 72 36 84 96 106 Time between metastasis and death (months)

120

Figure 1. Kaplan-Meier curves demonstrating the evolution of cumulative survival between diagnosis and metastatic disease (A), between the diagnosis and disease progression (B), between diagnosis and death (C) and between metastasis diagnosis and death (D) in the evolution of the disease in patients that died from differentiated thyroid carcinoma. 224

Arch Endocrinol Metab. 2017;61/3

death in 49 months (CI95%: 36.8 – 61.2 months) and between first metastasis diagnosis and death in 31 months (CI95%: 24.3 – 37.6 months). Moreover, the median survival was reached in 22 months (CI95%: 14.2 – 29.8 months) from the disease progression until death (Figure 2).

DISCUSSION The present study identified a death rate of 3.0% in a cohort of 1,114 patients treated for differentiated thyroid cancer in a single institution, one of the largest casuistic in the literature. Moreover, the majority of these patients had advanced disease, almost all of them with distant metastasis, especially pulmonary. Some developed RAI therapy refractory disease, tumor dedifferentiation and most patients quickly developed disease progression, despite the therapeutic approach. We also found that once disease progression has been established, the mean survival drops significantly and within 22 months of disease progression 50% of the patients are dead. Physicians consider differentiated thyroid cancer an indolent disease that will have a good outcome if correctly treated at diagnoses. As a result, there is a global tendency to treat DTC less aggressively over the years (8-12). However, since its extremely low mortality is balanced by its high prevalence, the number of deaths cannot be overlooked. The American Cancer Society estimates over 2,800 deaths related to DTC in 2014 (13). 1.0

Cumulative survival

0.8

0.6

0.4

0.2

0.0

12 36 24 48 60 Time between disease progression and death (Months)

Figure 2. Kaplan-Meier curves demonstrating a short time between disease progression and death in patients with differentiated thyroid carcinoma (median survival reached in 22 months). Arch Endocrinol Metab. 2017;61/3

Similarly to the present study that demonstrated a death rate of 3.0%, mortality related to DTC is reported to be lower than 5% at 10 years (4-6,14). Because it is a rare outcome in a disease with long clinical course it is extremely difficult to analyze the factors that may lead to it prospectively, which makes reports like the present one become more significant. Initial reports on this subject from the 1960s show that local disease progression with airway obstruction was a significant cause of death, almost as much as distant metastasis. Tollefsen and cols. (15) in 1964 reported 40% of deaths from thyroid cancer related to local disease progression and 52% related to distant metastasis. Smith and cols. (16) at Mayo Clinic reported a similar scenario in 1988, with 36% of deaths from local disease progression and 37% from lung metastasis progression. However, this distribution changed in the series closer to 21st Century and deaths due to local disease progression became rare. In 2001, Beasley and cols. (17) showed 70% of deaths related to distant metastasis and 20% from local recurrence. Nixon and cols. (7) reported on 17 deaths from DTC with 88% of deaths from distant metastasis and two from aspiration pneumonia that could be related to local disease complications. These findings are similar to the results reported in the present study, that show 85.3% of deaths related to distant metastasis progression and only one case (3%) from local disease progression. This shift on the specific cause that leads to death in DTC patients can most certainly be explained by the change in treatment approach in the last decades, with more aggressive surgical therapy to remove all gross disease, including airway and esophagus resection if necessary. Also, radioiodine treatment is more available and used in the majority of cases, which probably contributes to this shift in disease progression. Studies associate deaths from DTC with extrathyroid extension, nodal disease, age over 45 years and distant metastasis (7,16,17). In our series the majority of patients had age over 45 years, nodal disease and all but one patient had distant metastasis. Many cases had advanced disease at presentation, with 42.5% of metastasis already at initial diagnosis and 60% of T3 and T4 tumors. The literature also shows that advanced disease is usually present at the initial diagnosis in the patients that will die from DTC (7,17). However, there are some cases that do not follow this pattern. In this series two patients presented with 225

Differentiated thyroid cancer and death

initial tumors, no nodal or distant metastasis, and during follow-up developed lung metastasis that eventually led to their deaths. One was a 63 year-old male with a 1.3 cm papillary TC that had a nodal recurrence after two years of thyroidectomy, as well as lung metastasis four years after initial diagnosis, and died within one year. The other was a 42 year-old female with a 0.5 cm unifocal folicullar carcinoma and no nodal metastasis that developed lung metastasis after four years of surgery and died six years after initial diagnosis from lung disease progression (Figure 3). Nilubol and Kebebew (13) identified, using the database of SEER (Surveillance, Epidemiology, and End Results), that amongst 1,753 deaths from DTC 12.3% were tumors with less than 2.0 cm, which brings directly into question the safety of non-operative management of initial DTC that has recently been advocated by some authors (11). Our findings of two A

cases with less than 2.0 cm in 33 deaths from DTC corroborate that the non-operative approach should be treated with caution. In conclusion, mortality from DTC is extremely rare but persists, and the main cause of death is respiratory failure due to lung metastasis. The absolute majority of patients die from distant metastasis with rapid progression and present with advanced disease, but there are cases of initial indolent disease that later develop metastasis that lead to death. The very low mortality by DTC makes this a hard subject to study. There are only small series available in the literature with distinct inclusion criteria and methodologies and, therefore, it is not possible to make any useful conclusions as to what makes some DTC behave so aggressively. Further studies in molecular biology are necessary to establish factors associated with aggressiveness in DTC so that we can select the patients at high risk for mortality and treat them accordingly. This study has the limitation of being a retrospective descriptive analyses of a cohort of patients that died from DTC, therefore preventing further analyses of factors related to this outcome. Prospective trials would bring very useful knowledge on this subject, but are very hard to do since mortality is very low and disease clinical course is long. Also, because it was not a prospective study, follow up and scans intervals may have varied between patients. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. ACS 2016 Cancer facts & figures 2016. American Cancer Society Inc., Atlanta, GA. 2. INCA 2015 Estimativa 2016: incidência de câncer no Brasil/INCA (Instituto Nacional do Câncer). INCA, Rio de Janeiro. 3. Matos LL, Suarez ER, Theodoro TR, Trufelli DC, Melo CM, Garcia LF, et al. The Profile of Heparanase Expression Distinguishes Differentiated Thyroid Carcinoma from Benign Neoplasms. PloS One. 2015;10:e0141139. 4. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med. 1994;97:418-28. 5. Hay ID, Bergstralh EJ, Goellner JR, Ebersold JR, Grant CS. Predicting outcome in papillary thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1779 patients surgically treated at one institution during 1940 through 1989. Surgery. 1993;114:1050-1057; discussion 1057-8.

Figure 3. Image of chest CT from a female patient with a pT1a follicular carcinoma of 5 mm with pulmonary disease that died from obstruction of the left bronchus (A) by a metastasis six years after initial diagnosis. The patient presented also liver and kidney metastasis (B). 226

6. Lin JD, Hsueh C, Chao TC. Early recurrence of papillary and follicular thyroid carcinoma predicts a worse outcome. Thyroid. 2009;19:1053-9. 7.

Nixon IJ, Ganly I, Palmer FL, Whitcher MM, Patel SG, Tuttle RM, et al. Disease-related death in patients who were considered free of Arch Endocrinol Metab. 2017;61/3

Differentiated thyroid cancer and death

macroscopic disease after initial treatment of well-differentiated thyroid carcinoma. Thyroid. 2011;21:501-4. 8. Adam MA, Pura J, Goffredo P, Dinan MA, Hyslop T, Reed SD, et al. Impact of extent of surgery on survival for papillary thyroid cancer patients younger than 45 years. J Clin Endocrinol Metab. 2015;100:115-21. 9. Donatini G, Castagnet M, Desurmont T, Rudolph N, Othman D, Kraimps JL. Partial Thyroidectomy for Papillary Thyroid Microcarcinoma: Is Completion Total Thyroidectomy Indicated? World J Surg. 2016;40:510-5. 10. 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;26:1-133.

13. Nilubol N, Kebebew E. Should small papillary thyroid cancer be observed? A population-based study. Cancer. 2015;121:1017-24. 14. Sciuto R, Romano L, Rea S, Marandino F, Sperduti I, Maini CL. Natural history and clinical outcome of differentiated thyroid carcinoma: a retrospective analysis of 1503 patients treated at a single institution. Ann Oncol. 2009;20:1728-35. 15. Tollefsen HR, Decosse JJ, Hutter RV. Papillary Carcinoma of the Thyroid. A Clinical and Pathological Study of 70 Fatal Cases. Cancer. 1964;17:1035-44. 16. Smith SA, Hay ID, Goellner JR, Ryan JJ, McConahey WM. Mortality from papillary thyroid carcinoma. A case-control study of 56 lethal cases. Cancer. 1988;62:1381-8. 17. Beasley NJ, Walfish PG, Witterick I, Freeman JL. Cause of death in patients with well-differentiated thyroid carcinoma. Laryngoscope. 2001;111:989-91.

11. Ito Y, Miyauchi A. Nonoperative management of low-risk differentiated thyroid carcinoma. Curr Opin Oncol. 2015;27:15-20.

12. Ito Y, Oda H, Miyauchi A. Insights and clinical questions about the active surveillance of low-risk papillary thyroid microcarcinomas [Review]. Endocr J. 2016;63:323-8.

Arch Endocrinol Metab. 2017;61/3

227

original article

Relation between fetal anthropometric parameters and cord blood adiponectin and high-sensitivity C-reactive protein in gestational diabetes mellitus Mohammad Reza Aramesh1, Masoud Dehdashtian1, Arash Malekian1, Shiva ShahAli2, Kobra Shojaei3

ABSTRACT Department of Pediatrics, Division of Neonatology, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, IR Iran 2 Student Research Committee, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, IR Iran 3 Department of Obstetrics and Gynecology, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, IR Iran 1

Correspondence to: Shiva ShahAli Ahvaz Jundishapur University of Medical Sciences, Ahvaz, IR Iran shivashahali@gmail.com Received on Jan/8/2016 Accepted on Oct/10/2016

Objectives: The objectives were to evaluate the relation between fetal anthropometric parameters and cord blood concentration of adiponectin and high sensitivity C-reactive protein (hs-CRP). Subjects and methods: A total of 104 pregnant women (52 with gestational diabetes mellitus [GDM], 52 with normal glucose tolerance (NGT) participated. Venous cord blood samples were obtained at delivery, centrifuged and the plasma was stored at -20°C. The samples were assessed for adiponectin and hs-CRP using the ELISA method. Statistical analysis was done using SPSS software. Results: The adiponectin concentration was higher in the GDM group than in the NGT group (11.05 ± 4.1 µg/ mL in GDM vs. 5.34 ± 2.63 µg/mL in NGT, p < 0.001). GDM was also higher in neonates delivered at later gestational ages (p < 0.001, Pearson correlation = 0.59). There was a positive correlation between cord blood adiponectin and birth weight in the GDM group (p < 0.001, Pearson correlation = 0.619) but not in the NGT group. There was no significant correlation between adiponectin and infant length or head circumference. There was also no significant difference in cord blood hs-CRP concentration between groups. No relation was found between hs-CRP and newborn anthropometric parameters. Conclusion: In the GDM group, adiponectin concentration was considerably higher and had a positive correlation with the ponderal index and birth weight which was not found in the NGT group. Arch Endocrinol Metab. 2017;61(3):228-32.

DOI: 10.1590/2359-3997000000235

INTRODUCTION

estational diabetes mellitus (GDM) is impaired glucose tolerance that develops or is first diagnosed during pregnancy (1-3). Recent studies have shown a 20-fold increase in the prevalence of GDM. A further 4-fold increase in prevalence is anticipated (3,4) after adoption of the new diagnostic criteria suggested by the International Association of Diabetes and Pregnancy Study Groups (IADPSG) (1,5). The presence of GDM means that many infants are exposed to hyperglycemic conditions in utero which can cause neonatal adiposity and metabolic disorders later in life. The pathogenesis of GDM and its features that result in obesity and metabolic disorders remain unclear. Recent studies have shown that the inflammatory system may play a role in the development and pathogenesis of GDM (1). Insulin resistance and inflammation are major features of maternal metabolic state in women 228

with GDM. Both of these conditions can affect fetal growth (2). The inflammatory environment alters placental gene transcription and fetal metabolic programming. Genes for lipid metabolism and those for inflammatory pathways are upregulated in the placenta of women with GDM (2). This upregulation increases or unbalances production of inflammatory cytokines and energy metabolism regulatory cytokines such as adipokines (2,6,7), which increases adiposity at birth (8,9) and predisposes the newborn to become overweight and develop metabolic diseases such as impaired glucose tolerance, metabolic syndrome, and cardiovascular disease (2,6-8). Previous studies have focused on the pathophysiologic pathways and cytokine levels in GDM and their possible effect on offspring comorbidities later in life (1,6,7,10). Adiponectin is a cytokine that is released exclusively by adipocytes in adults (9,11-13). Maternal Arch Endocrinol Metab. 2017;61/3

Fetal anthropometrics, adiponectin and hsCRP in GDM

SUBJECTS AND METHODS This analytical, case-controlled study was conducted at Imam Khomeini Hospital (a university-affiliated hospital) in the city of Ahvaz in Iran. Recruitment began in June 2014 and ended in July 2015. The study protocol was approved by the Research Ethics Committee of Ahvaz Jundishapur University of Medical Sciences. Informed consent was obtained from all patients. Prior to participation, all the participants underwent an oral glucose tolerance test (3 hour, 100 g glucose). Those who had two or more values exceeding the thresholds of the GDM diagnostic criteria as suggested by the National Diabetes Data Group (NDDG) were included in the GDM group. Women who had no values meeting a NDDG criteria were included in the NGT group. Those who had only one value exceeding the threshold of the criteria were excluded from the study (22). Inclusion criteria were a gestational age at delivery of 35 to 41 weeks, singleton pregnancy, first minute Apgar Arch Endocrinol Metab. 2017;61/3

score > 7 and uncomplicated delivery. Exclusion criteria were substance abuse by mother (including cigarettes, alcohol, etc.), chronic disease and overt diabetes in the mother and the presence of congenital anomalies. Eventually 104 pregnant women entered the study (52 women in the GDM group and 52 in the NGT group). Of the 52 women with GDM, 38 women were given a 40% carbohydrate diet to control their GDM and 14 women were treated with insulin.

Clinical and demographic data The demographic data was gathered by questionnaire. The height and weight of the pregnant women were measured using a calibrated medical scale and recorded. The body mass index (BMI) was calculated as BMI = weight (kg)/height (m2). The length of the newborns was measured using a calibrated length board and their weight with a calibrated scale. To assess fetal growth pattern, the ponderal index (PI) was calculated as Birth weight (gr) (PI = × 100). Body lenght (cm)3

Lab measurement Venous cord blood samples were obtained from the 104 full-term healthy infants. After delivery and cord clamping, 5 mL of cord blood was collected from the umbilical vein in prepared and heparinized tubes at ambient temperature using the aseptic method. The samples were centrifuged and the plasma was kept frozen and stored at -20oC until analysis. Adiponectin was assayed using an ELISA kit specific to human adiponectin (Biovendor; Laboratorni Medicina; Czech Republic). Serum adiponectin was measured as micrograms per milliliter (µg/mL). Quantitative high sensitivity C-reactive protein (hsCRP) was assayed using an i-CHROMA kit for fluorescence immunoassay specifically for determination of human hs-CRP (Boditech Med Europe; United Kingdom). Serum hs-CRP was measured as milligram per milliliter (mg/mL).

Statistics The data were analyzed by SPSS 13 software (SPSS; USA). The unpaired student T-test was utilized to assess differences between groups. Simple linear regression was employed to eliminate the influence of other predictor values. Pearson correlation was used to analyze the bivariate correlation between adiponectin levels and anthropometric parameters. 229

adiponectin has a high molecular weight; therefore, it probably does not pass through the placental barrier; its concentration in the umbilical cord is of fetal origin (14,15). The origin of adiponectin in the fetus is not clearly understood (13). In adults, adiponectin concentration is inversely related to adiposity, but its association with adiposity is poorly understood (9,11,13). Some studies indicate a positive correlation between cord plasma adiponectin concentration and adiposity in infants (8,12,14,16-20) and some have found no significant correlation (9,15). C-reactive protein (CRP) is an acute phase protein in plasma synthesized by the liver in response to pre-inflammatory cytokines. Its concentration increases 12-24 hours after the commencement of an inflammatory process and preserves its level throughout the inflammation. There is a positive correlation between CRP concentration and insulin resistance and elevated levels of CRP have been reported in GDM. A growing body of evidence suggests that CRP is related to development and progression of cardiovascular disease (21). The present study evaluated the relationship between fetal anthropometric parameters, cord blood adiponectin and hs-CRP concentration at birth and compared mean cord plasma adiponectin and hs-CRP concentrations in the GDM and normal glucose tolerant (NGT) groups.

Fetal anthropometrics, adiponectin and hsCRP in GDM

RESULTS All infants were healthy and were born at Imam Khomeini Hospital. The demographic and clinical data of the study population are shown in Table 1. Although there was no statistically significant difference in age, parity and BMI between groups. The offspring of the GDM group had higher adiposity, were more prone to be delivered by cesarean section, and were delivered one week earlier on average than the NGT group (Table 1). All of the 104 infants were included in the study (51 males and 53 females; gestational age: 35-41 weeks; birth weight: 2300-4150 g). No significant differences were observed between groups. The mean birth weight and PI were higher in the GDM group than the NGT group (birth weight: 3535.1 ± 400 g vs. 3041.7 ± 350 g, PI: 2.74 ± 0.28 g/cm3 vs. 2.40 ± 0.18 g/cm3; p < 0.001). There was no significant difference between the mean length (50.50 ± 1.96 cm in GDM group vs. 50.14 ± 1.43 cm in NGT group; p = 0.29) and head circumference (34.22 ± 0.65 cm in GDM group vs. 34.12 ± 0.79 cm in NGT group, p = 0.5) between groups.

The adiponectin present in the cord blood ranged from 0.8 to 22.30 µg/mL. Mean cord blood adiponectin was 11.05 ± 4.1 µg/mL in the GDM group and 5.34 ± 2.63 µg/mL in the NGT group and was significantly higher in the GDM group (p < 0.001). In addition, there was a significant correlation between cord blood adiponectin and birth weight in the GDM group (p < 0.001, Pearson correlation = 0.619). No such relation was found in the NGT group. There was no significant correlation between adiponectin and length or head circumference of the infants. No difference in adiponectin concentrations was observed according to gender or by mode of delivery between groups. Cord blood adiponectin concentrations were higher at later gestational ages and there was a significant correlation between gestational age and adiponectin concentration (p < 0.00; Pearson correlation = 0.59). There was no significant difference in cord blood hs-CRP concentration between groups. No significant relationship was found between hs-CRP and newborn anthropometric parameters.

Table 1. Demographic and clinical data of the study population Whole group

NGT group

GDM group

104

27.89 ± 5.63 (17-40)

26.8 ± 5.05

28.9 ± 6.02

25.17 ± 4.89

24.96 ± 5.24

25.37 ± 4.53

51/53

27/25

24/28

38.12 ± 1.46 (35.44-41)

38.65 ± 1.45

37.6 ± 1.28

< 0.001

5.24 ± 0.32

Vaginal

Caesarian

P value

Maternal characteristics: Age (years) BMI Parity

Children (boys/girls) Gestational age at delivery (weeks) HbA1c

Delivery < 0.001

Neonatal characteristics: Birth weight (kg)

3.28 ± 0.45 (2.30-4.15)

3.04 ± 0.35

3.53 ± 0.40

< 0.001

Ponderal index (gr⁄cm3)

2.57 ± 0.29 (2.04-3.39)

2.40 ± 0.18

2.74 ± 0.28

< 0.001

Birth length (cm)

50.32 ± 1.72 (47-56)

50.14 ± 1.43

50.50 ± 1.96

Head circumference (cm)

34.17 ± 0.72 (30-36)

34.12 ± 0.79

34.22 ± 0.65

8.20 ± 4.470 (0.8-22.30)

5.34 ± 2.63

11.05 ± 4.1

< 0.001

0.19 ± 0.30 (0.1-2.0)

0.19 ± 0.28

0.19 ± 0.33

Cord serum adiponectin (µg/mL) Cord serum hsCRP (mg/mL)

Data are mean ± SD; HbA1C: glycosylated hemoglobin; NS: not significant.

230

Arch Endocrinol Metab. 2017;61/3

Fetal anthropometrics, adiponectin and hsCRP in GDM

Adiponectin is a cytokine of great importance because it can cause establishment of metabolic diseases. Adiponectin concentration has a reverse correlation with inflammatory diseases such as obesity and cardiovascular disease (23,24). Studies have shown that adiponectin affects insulin sensitivity, Î˛ oxidation and inflammatory pathways. The fact that adiponectin concentrations increase at later gestational ages suggests a role for it in the early growth and development of the fetus (9,13). The present study revealed that the mean cord blood adiponectin concentration in the GDM group was significantly higher than that in the NGT group. Previous studies have reported lower cord blood adiponectin concentrations in the GDM group (11). The data showed a significant correlation between PI (Pearson correlation: 0.421), birth weight (Pearson correlation: 0.619) and cord blood adiponectin concentration. Increased adiponectin levels may play a role in increased birth weight and PI. Ballesteros and cols. (11) reported similar findings. In contrast, Lindsay and cols. (9) found no significant correlation between adiponectin concentration and anthropometric parameters. Another study suggested a reverse correlation between PI and adiponectin concentration (11,25). Some studies have shown a positive correlation between adiponectin concentration and birth weight (8,12,14,16-20), but few have failed to find a significant correlation (9,15). The current study found a positive correlation between adiponectin concentration and gestational age in the GDM group although there was no such correlation in the NGT group. Ballesteros and cols. (11) reported similar results, but Lindsay and cols. (9) found no significant correlation between gestational age and adiponectin. There was no significant correlation found for adiponectin concentrations versus delivery mode, which agrees well with the findings of Lindsay and cols. (9). The results also show no difference in adiponectin levels by gender. By contrast, Lindsay and cols. (9) reported a difference by gender for adiponectin concentration. No difference in adiponectin concentration was found between women with GDM who were treated with a 40% carbohydrate diet and those who were treated with insulin in the present study. This study found no significant difference in cord blood hs-CRP concentration between groups. These findings were similar to those of Mordwinkin and Arch Endocrinol Metab. 2017;61/3

cols. (26) and Jahromi and cols. (27). No significant relation was found between hs-CRP and newborn anthropometric parameters in the present study. To our knowledge the present study is one of the first to investigate the relation between hs-CRP and newborn anthropometric parameters. No similar studies were found to compare with the results of the present study. The study had certain limitations. Only Iranian women were involved, so the results cannot be extrapolated to other ethnicities. Moreover, financial limitations prevented adoption of IADSPG criteria for screening and diagnosing GDM. This study is one of few to assess the level of adiponectin and hs-CRP in GDM and their relation with fetal anthropometric parameters. Inflammatory pathophysiology may play a role in the development of GDM (1), but the mechanism is not fully understood. The findings of the previous studies have conflicted. Further investigation is needed to examine whether or not a significant difference exists between cord serum concentration of adiponectin and hs-CRP among women with GDM compared to healthy pregnant controls and investigate their relation to fetal anthropometric parameters. To summarize, a significantly higher adiponectin concentration was found in the GDM group and had a positive correlation with PI and birth weight which was not observed in the NGT group. Acknowledgments: the authors would like to thank the neonatal resuscitation team for cord blood sampling and the staff of the labor and delivery department of the obstetrics and gynecology ward of Imam Khomeini Hospital, affiliated with Ahvaz Jundishapur University of Medical Sciences, for their help during this research. Disclosure: this work was financially supported through grant GP93100 from the Vice-Chancellor for Research Affairs of Ahvaz Jundishapur University of Medical Sciences. This study has been extracted from the postgraduate thesis of Dr. Shiva ShahAli and was approved by the Health Research Institute of Ahvaz Jundishapur University of Medical Sciences. The authors declare that there is no conflict of interest.

REFERENCES 1. Gomes CP, Torloni MR, Gueuvoghlanian-Silva BY, Alexandre SM, Mattar R, Daher S. Cytokine levels in gestational diabetes mellitus: a systematic review of the literature. Am J Reprod Immunol. 2013;69(6):545-57. 2. Mitanchez D, Burguet A, Simeoni U. Infants born to mothers with gestational diabetes mellitus: mild neonatal effects, a long-term threat to global health. J Pediatr. 2013;164(3):445-50. 3. Aramesh MR, Dehdashtian M, Malekian A, Babapour R. comparison of the blood level of leptin in umbilical cord of newborns

231

DISCUSSION

Fetal anthropometrics, adiponectin and hsCRP in GDM

of mothers with gestational diabetes and normal mothers and its relationship with growth indices of newborns. Jentashapir J Health Res. 2015;6(5):e25797. 4. Jenum AK, Mørkrid K, Sletner L, Vange S, Torper JL, Nakstad B, et al. Impact of ethnicity on gestational diabetes identified with the WHO and the modified International Association of Diabetes and Pregnancy Study Groups criteria: a population-based cohort study. Eur J Endocrinol. 2012;166(2):317-24. 5. International Association of Diabetes and Pregnancy Study Groups Consensus Panel, Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P, et al. International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care. 2010;33(3):676-82. 6. Marseille-Tremblay C, Ethier-Chiasson M, Forest JC, Giguère Y, Masse A, Mounier C, et al. Impact of maternal circulating cholesterol and gestational diabetes mellitus on lipid metabolism in human term placenta. Mol Reprod Dev. 2008;75(6):1054-62. 7. Reece EA. The fetal and maternal consequences of gestational diabetes mellitus. J Matern Fetal Neonatal Med. 2010;23(3):199-203. 8. Josefsona JL, Zeissb DM, Rademakerd AW, Metzgere BE. Maternal leptin predicts adiposity of the neonate. Horm Res Paediatr. 2014 81(1):13-9. 9. Lindsay RS, Walker JD, Havel PJ, Hamilton BA, Calder AA, Johnstone FD. adiponectin is present in cord blood but is unrelated to birth weight. Diabetes Care. 2003;26(8):2244-9. 10. American Diabetes Association: Diagnosis and classification of diabetes mellitus. Diabetes Care. 2012;35(Suppl. 1):S64-71.

maternal and perinatal characteristics in newborns. Eur J Clin Endo. 2004(151):741-6. 16. Brynhildsen J, Sydsjo G, Blomberg M, Claesson I-M, Theodorsson E, Nystrom F, et al. Leptin and adiponectin in cord blood from children of normal weight, overweight and obese mothers. Acta Pædiatrica. 2013;102:620-24. 17. Mazaki-Tovi S, Kanety H, Pariente C, Hemi R, Schiff E, Sivan E. Cord blood adiponectin in large-for-gestational age newborns. Am J Obstet Gynecol. 2005;193 1238-42. 18. Kotani Y, Yokota I, Kitamura S, Matsuda J, Naito E, Kuroda Y. Plasma adiponectin levels in newborns are higher than those in adults and positively correlated with birth weight. Clin Endocrinol. 2004;61:418-23. 19. Kamoda T, Saitoh H, Saito M, Sugiura M, Matsui A. Serum adiponectin concentrations in newborn infants in early postnatal life. Pediatr Res. 2004(56):690-3. 20. Pardo IM, Geloneze B, Tambascia MA, Barros-Filho AA. Inverse relationship between cord blood adiponectin concentrations and the number of cigarettes smoked during pregnancy. Diab Obes Met. 2005(7):144-7. 21. Howman RA, Charles AK, Jacques A, Doherty DA, Simmer K, Strunk T, et al. Inflammatory and haematological markers in the maternal, umbilical cord and infant circulation in histological chorioamnionitis. PLoS One. 2012;7(12):1-8. 22. Plagemman A, Harder T, Kohlhoff R, Rohde W, Dorner G. Glucose tolerance and insulin secretion in children of mothers with pregestational IDDM or gestational diabetes. Diabetologia. 1997;18: 1094-100.

11. Ballesteros M, Simón I, Vendrell J, Ceperuelo-Mallafré V, Miralles RM, Albaiges G, et al. Maternal and cord blood adiponectin multimeric forms in gestational diabetes mellitus. Diabetes Care. 2011;34:2418-23.

23. Ahima R. Metabolic actions of adipocyte hormones: focus on adiponectin. Obesity 2006;14: 9S-15S.

12. Mantzoros CS, Rifas-Shiman SL, Williams CJ, Fargnoli JL, Kelesidis T, Gillman MW. Cord Cord blood leptin and adiponectin as predictors of adiposity in children at 3 years of age: a prospective cohort study. Pediatrics. 2009;123(2):682-9.

25. Tsai PJ, Yu CH, Hsu SP, Lee YH, Chiou CH, Hsu YW, et al. Cord plasma concentrations of adiponectin and leptin in healthy term neonates: positive correlation with birthweight and neonatal adiposity. Clin Endocrinol (Oxf). 2004;61(1):88-93.

13. Courville ABP. A DHA-Functional Food During Pregnancy: Impact on Maternal Dietary Intake, Endocrine Parameters and Markers of Infant Body Composition [Doctoral Dissertation]. Connecticut: University of Connecticut; 2006.

26. Mordwinkin NM, Ouzounian JG, Yedigarova L, Montoro MN, Louie SG, Rodgers KE. Alteration of endothelial function markers in women with gestational diabetes and their fetuses. J Matern Fetal Neonatal Med. 2013;26(5):507-12.

14. ChanTF,Yuan SS, Chen HS, Guu CF, Wu LC,YehYT, et al. Correlations between umbilical and maternal serum adiponectin levels and neonatal birthweights. Acta Obstet Gynecol Scand. 2004(83):165-9.

27. Jahromi BN, Ahmadi N, Cohan N, Jahromi MRN. Comparison of the umbilical artery blood gas, nucleated red blood cell, Creactive protein, and white blood cell differential counts between neonates of diabetic and nondiabetic mothers. Taiwan J Obstet Gynecol. 2011;50(3):301-5.

15. Mantzoros C, Petridou E, Alexe D, Skalkidou A, Dessypris N, Papathoma E, et al. Serum adiponectin concentration in relation to

24. Matsuzawa. The metabolic syndrome and adipocytokines. FEBS Letters. 2006;580:2917-21.

232

Arch Endocrinol Metab. 2017;61/3

original article

Serum Fluorescent Advanced Glycation End (F-AGE) products in gestational diabetes patients João Paulo Lobo Júnior1, Catiane Pompilio Brescansin1, Izabella C. R. SantosWeiss1, Marciane Welter1, Emanuel Maltempi de Souza2, Fabiane Gomes de Moraes Rego1, Geraldo Picheth1, Dayane Alberton1

Objectives: Advanced glycation end products (AGEs) are involved in the pathogenesis and complications of diabetes mellitus (DM). Gestational DM (GDM) is characterized by increased glycemia and oxidative stress, which are factors associated with high serum AGE concentrations. The aim of this study was to evaluate the utility of a serum fluorescence AGE (F-AGE) method as a screening tool for gestational diabetes. Subjects and methods: Serum samples from 225 GDM patients and 217 healthy pregnant women (healthy controls) were diluted 50-fold in phosphate-buffered saline, and the AGEs were estimated by fluorometric analysis (λEx 350 nm/ λEm 440 nm). Results: No significant (P > 0.05) differences in AGE concentrations, expressed in Arbitrary Units (UA/mL × 104), were observed in the women with GDM or in the healthy controls. Furthermore, F-AGE concentrations did not change significantly during the pregnancy (12-32 weeks of gestation). Only the GDM group had a positive correlation (r = 0.421; P < 0.001) between F-AGEs and serum creatinine concentrations. Conclusion: It was not possible to distinguish women with gestational diabetes from the healthy controls on the basis of serum F-AGE concentrations. Arch Endocrinol Metab. 2017;61(3):233-7. Keywords Diabetes screening; gestational diabetes; fluorescent advanced glycation end products

INTRODUCTION

dvanced glycation end products (AGEs) are generated by the non-enzymatic reaction of a sugar ketone or aldehyde group with the free amino groups of proteins, amino acids, lipids, and nucleic acids under conditions of hyperglycemia and oxidative stress (1,2). AGEs may cause tissue injury both directly, through phenomena such as trapping and cross-linking, and indirectly, by binding to specific receptors such as receptors for AGE (RAGE), which is expressed on the surface of numerous cell types, such as macrophages, monocytes, endothelial cells, neurons, and smooth muscle cells (3,4). The AGE-RAGE interaction can lead to oxidative stress, production of growth factors and cytokines, chronic inflammatory responses, and cellular and vascular dysfunction (5,6). Elevated AGEs concentrations are associated with several diseases, including diabetes mellitus (DM) (5,7,8). DM is a pathology characterized by hyperglycemia, oxidative stress, inflammation, and consequently, the AGE-RAGE interaction is enhanced (1,5). While some studies have shown that Arch Endocrinol Metab. 2017;61/3

1 Pós-Graduação em Ciências Farmacêuticas, Departamento de Análise Clínica, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brasil 2 Departamento de Bioquímica e Biologia Molecular, UFPR, Curitiba, PR, Brasil

Correspondence to: Dayane Alberton Departamento de Análises Clínicas, Universidade Federal do Paraná Rua Prefeito Lothário Meissner, 632 80210-170 – Curitiba, PR, Brasil dayanealberton@ufpr.br Received on Mar/3/2016 Accepted on Nov/11/2016 DOI: 10.1590/2359-3997000000238

AGE concentrations are higher in type 1 (T1D) and type 2 (T2D) diabetic patients than in healthy subjects, especially in diabetes with secondary complications (9,10), others have shown that AGE concentrations are also elevated in gestational DM (GDM) (11,12), and still other studies have demonstrated that AGEs concentrations were not significantly different between women with GDM and healthy pregnant women (13,14). However, a standard method to measure AGEs has not yet been established, making it difficult to compare results (15). The absence of a universal method to measure AGEs is largely due to the characteristics of these compounds. AGEs constitute a large, complex, and heterogeneous group of molecules, and only some structures have been identified (1,16). εN-carboxymethyl-lysine (CML), pentosidine, and methylglyoxal derivatives are examples of well-characterized AGEs (4,16). High performance liquid chromatography (HPLC), enzyme-linked immunosorbent assay (ELISA), immunohistochemistry and fluorescence spectroscopy have been used to measure the concentrations of the different types of AGEs (17-19). 233

ABSTRACT

Fluorescent-AGEs in GDM

Most AGEs have a characteristic fluorescence with an excitation maximum approximately at 370 nm and an emission maximum around 445 nm (20). Unlike other methods, fluorescence spectroscopy is rapid, cost effective and sample preparation is simple. In this study, the fluorescence method was applied to measure the AGEs in the serum of pregnant women with GDM and healthy pregnant women in order to evaluate the screening capacity of this method and to examine the relationship AGEs concentration to other biochemistry parameters.

MATERIALS AND METHODS Subjects A total of 442 unrelated Euro-Brazilian pregnant women were examined. Healthy pregnant women were classified as controls (n = 217). Women with gestational diabetes (GDM, n = 225) were classified by the criteria of the Brazilian Diabetes Society – 2009 (21). Briefly, fasting plasma glucose ≥ 6.1 mmol/L and glycemia 2 h after 75 g oral glucose ≥ 7.8 mmol/L at 24th – 28th weeks of gestation. Patients with overt renal failure and cardiovascular disease were not included in the study. The study was approved by the Federal University of Paraná’s Ethics Committee according to the Declaration of Helsinki, and all subjects gave written consent before measurements.

Clinical and laboratory data

Clinical and anthropometric data were collected from patient files or from electronic patient registers. Fasting (8 h) blood was collected in Ethylenediaminetetraacetic Acid tripotassium salt tubes (K3EDTA, Vacutainer®, Becton Dickinson, New Jersey, USA) and Serum Separator tubes (Gel SST® II Advance, Becton Dickinson, New Jersey, USA). The plasma and serum obtained were stored at -20 °C. The biochemical blood parameters were determined using an Architect Ci8200 system (Abbott Diagnostic Laboratory, Illinois, USA) with reagents, calibrators, and controls from the manufacturer (Table 1).

Fluorescent AGE assay Measurement of fluorescent AGEs (F-AGEs) concentrations was based on the spectrofluorimetric detection (22). Serum was diluted 50-fold with phosphate-buffered saline (KH2PO4 1.06 mmol/L, 234

NaCl 155.10 mmol/L and Na2HPO4.7H2O 2.97 mmol/L, pH 7.4) and homogenized with vortex mixer for 10 seconds. The diluted serum (300 µL) was transferred into black 96-well plates. The excitation and emission wavelengths were 350 nm and 440 nm, respectively (Spectrofluorimeter Infinitive M200, TECAN, Mannedorf, Switzerland). PBS solution was used as blank. The fluorescence intensity was expressed in arbitrary units per milliliter of serum (AU/mL) and in AU/g of total protein. The total protein was measured by the biuret method (Architect Ci8200 system, Abbott Diagnostic Laboratory, Illinois, USA). The analytical coefficient of variation (CVa) was determined intra-assay as 5.1% (n = 15) and inter-assay as CVa = 7.9% (n = 22). Table 1. Anthropometric and laboratory characteristics of the study groups Parameters

Control (n = 217)

GDM (n = 225)

Age (years)

29 (27–33)

32 (28–36)

< 0.001*

Weight (kg)

66.1 (58.5–73.8)

80.3 (70–93)

< 0.001*

Height (m)

1.61 ± 0.06

1.60 ± 0.07

0.003

25.4 (22.5–28.3)

32.0 (27.7–36.4)

< 0.001*

Fasting glucose (mmol/L)

BMI (kg/m )

4.7 (4.4–4.9)

4.8 (4.6–5.4)

< 0.001*

Glucose 2h-75g (mmol/L)

4.8 (4.5-5.6)

9.0 (8.2–10.0)

< 0.001*

5.6 (5.3–6.1)

69 ± 7

63 ± 5

< 0.001

HbA1C (%) Total Protein (g/L) Albumin (g/L) Creatinine (µmol/L)

43 (38–46)

34 (32–36)

< 0.001*

70.7 (61.9–79.6)

61.9 (53.0–70.7)

< 0.001*

AGE (AU/mL × 104)

2.50 ± 0.86

2.42 ± 0.72

0.262

AGE (AU/g × 105)

3.65 ± 1.15

3.84 ± 1.36

0.114

Values are presented as mean ± SD, median (interquartile range); - no information available. Control, healthy pregnant women; GDM: gestational diabetes mellitus. P values, Student t-test (two-sided) or * Mann–Whitney U test.

Statistical analysis The Kolmogorov-Smirnov test was applied to test the data for a normal distribution. Variables with a normal distribution were reported as mean ± SD and those with non-normal distribution as median (interquartile range, 25-75%). Comparisons between groups with continuous variables were tested with Student’s t test (independent) or the Mann–Whitney U test, as appropriate. ANOVA was used to compare more than two groups with normal distributions. The correlation analyses were carried out with Pearson’s correlation test. A P-value < 0.05 was accepted as the threshold for defining statistical Arch Endocrinol Metab. 2017;61/3

Fluorescent-AGEs in GDM

significance. Statistical evaluation was performed with the Statistica software for Windows, version 8.0 (StatSoft Inc., Tulsa, OK, USA). ROC (Receiver Operating Characteristics) curves, cut-off points and the area under the curve (AUC) were calculated using MedCalc ver 12.2.1.0 (Mariakerke, Belgium).

significant positive correlation (r = 0.421; P < 0.001) between F-AGE concentration (UA/g protein) and serum creatinine concentrations in the GDM group (Figure 3). Healthy pregnant women showed no correlation between F-AGE concentration and creatinine concentrations (r = 0.124; P = 0.049). 6 P = 0.227

RESULTS

P = 0.821

P = 0.276

29-32

> 32

5 F-AGEs (AU/g protein)x105

The clinical and laboratory characteristics of the healthy women and women with GDM pregnant patients are shown in Table 1. The low fasting glucose and HbA1C concentrations suggested that the GDM patients had good glycemic control. Additionally, the low creatinine concentrations (< 106 µmol/L) indicate an absence of kidney damage. The F-AGEs concentrations expressed as arbitrary units per unit volume (AU/mL) and per mass of protein (AU/g) were not statistically significant (P > 0.05) between the two groups (Table 1). ROC curve analysis (Figure 1), with the area under the curve (AUC) of 0.537 (P = 0.188), showed that the fluorescence assay did not have sufficient specificity (67.3%) and sensibility (48.7%) to classify the groups. The F-AGEs were also not significantly different in four gestational periods (Figure 2). There was a

P = 0.508

4 3 2 1

12-23

24-28

Weeks of gestation

Figure 2. F-AGEs expressed in AU/g of protein were compared in four gestational periods. The results for healthy pregnant woman are shown as black circles and those for the GDM patients are represented by grey squares. The vertical bars represent 1-standard deviation. The P-values (Student’s t test) compared the F-AGE concentrations in the same gestational period. Variance analysis (ANOVA) did not show a significant difference for controls (P = 0.076) or GDM (P = 0.928). 9

F-AGEs

100

F-AGE, (UA/g protein)x105

Sensitivity

4 3

1 30

0 0

Criterion >3.9 Sensitivity 48.7% Specificity 67.3%

100

60 70 Creatinine, µmol/L

100

Figure 3. Linear correlation between serum creatinine and F-AGEs for the GDM group. Pearson’s Correlation (r = 0.421; P < 0.001) between F-AGEs and serum creatinine. The regression (solid) and 95% confidence interval (dotted) lines are shown.

Figure 1. Receiver operating characteristic (ROC) curve for F-AGEs values in pregnant women with and without gestational diabetes. The AUC is 0.537 ± 0.036 (P = 0.188). The black circle (arrow) indicates the cut-off point (> 3.9 AU/g of protein) with sensitivity and specificity values of 48.7% and 67.3%, respectively. The dotted lines delimit the 95% confidence interval and the straight line is the line of equality. Arch Endocrinol Metab. 2017;61/3

DISCUSSION Serum AGEs can be detected by many analytical methods, such as ELISA, radioimmunoassay, radioreceptor assay, fluorescence spectroscopy, and HPLC (18,20,23,24). 235

100-Specificity

Fluorescent-AGEs in GDM

Fluorescence spectroscopy is an easy and rapid method (22). Different studies have shown that the fluorescence assay (detecting F-AGEs) can be used to distinguish type 1 (T1D) and type 2 (T2D) diabetic patients from healthy subjects (19,22). The high oxidative stress conditions associated with diabetes likely play a more important role in AGE formation, in particular in type 2 diabetes, than the hyperglycemic state (22). We postulated that GDM-induced mild hyperglycemia combined with oxidative stress could promote a significant increase in AGE concentrations when compared to that observed in healthy pregnant women. Therefore, we decided to evaluate the utility of a simple, fast, and inexpensive fluorimetric method to screen for GDM, where 96 samples could be processed in a short interval of time and with acceptable analytical performance (CVa < 8%; inter-assay). Our results showed that the proposed method could not be used to distinguish between the healthy patients and the GDM patients (Table 1). The ROC curve analysis (Figure 1) confirms that fluorescent AGEs were not able to efficiently discriminate the studied groups by the low sensibility and specificity observed. These results are also consistent with a previous study, which showed that the skin autofluorescence AGE, measured using the AGE-Reader (DiagnOptics Technologies BV, Groningen, The Netherlands), also failed to distinguish GDM patients from healthy pregnant women (13). The authors justified this result due to mild severity and short duration of hyperglycemia in GDM at diagnosis. In our study, the good glycemic control observed in the GDM group (HbA1c 5.6%) likely explains the inability of fluorescence spectroscopy method to distinguish the GDM group from the healthy pregnant women. Therefore, we hypothesize that the presence of the mild hyperglycemia and oxidative stress in our GDM patients did not generate serum F-AGEs concentration enough to discriminate the groups studied. Buongiorno and cols. (11) differentiated GDM patients without adequate glycemic control from the control group by quantifying the AGE concentrations using the ELISA method, but could not differentiate women who previously had DM pregnancies and good glycemic control from healthy pregnant women. In addition, in our population studied, no difference in F-AGEs was observed in the four major periods of gestation between the healthy women and women with gestational diabetes (Figure 2). On the other hand, AGE concentrations measured by the fluorescence 236

method in the serum of Chinese GDM pregnant women in mid-gestational and later gestational periods were also similar, but higher when compared to those of healthy pregnant women in the same gestational periods (25). Of the correlations tested between F-AGE concentrations and biochemical blood parameters, the F-AGE was positively correlated with serum creatinine concentrations only in the GDM group (Figure 3). Similar results were described for T1D and T2D patients (r = 0.84; P < 0.001) (26,27). In contrast, a significant positive correlation between low molecular weight serum F-AGEs and serum creatinine was shown in individuals with only minimal renal disturbance or with normal creatinine concentrations (28). Pentosidine (free form), an F-AGE, and possibly other AGEs are filtered through the glomeruli and reabsorbed in the proximal tubules (29). Therefore, decreased glomerular filtration rate and tubule cell damage could also be involved in AGE accumulation, as suggested by Wagner and cols. (27), who showed that patients with impaired renal function presented with increased serum CML and F-AGE concentrations and decreased creatinine clearance. In the present study, it is not clear why GDM pregnant women with normal serum creatinine concentrations presented with increased AGE concentrations, and more studies are necessary to determine the extent to which these findings are repeated elsewhere. In summary, serum F-AGEs concentrations measured by fluorescence spectroscopy were not able to distinguish women with gestational diabetes from the healthy pregnant controls in our population. Acknowledgments: this study was supported by CNPq and Fundação Araucária. The authors would also like to thank the Academic Publishing Advisory Center (Centro de Assessoria de Publicação Acadêmica, CAPA – www.capa.ufpr.br) of the Federal University of Paraná for assistance with English language editing. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Bierhaus A, Hofmann MA, Ziegler R, Nawroth PP. AGEs and their interaction with AGE-receptors in vascular disease and diabetes mellitus. I. The AGE concept. Cardiovasc Res. 1998;37:586-600. 2. Vlassara H, Uribarri J. Advanced Glycation End Products (AGE) and Diabetes: Cause, Effect, or Both? Curr Diabetes Rep. 2014;14. 3. Schmidt AM, Yan SD, Wautier JL, Stern D. Activation of receptor for advanced glycation end products – A mechanism

Arch Endocrinol Metab. 2017;61/3

Fluorescent-AGEs in GDM

for chronic vascular dysfunction in diabetic vasculopathy and atherosclerosis. Circ Res. 1999;84:489-97. 4. Singh R, Barden A, Mori T, Beilin L. Advanced glycation endproducts: a review. Diabetologia. 2001;44:129-46. 5. Ramasamy R, Vannucci SJ, Yan SS, Herold K, Yan SF, Schmidt AM. Advanced glycation end products and RAGE: a common thread in aging, diabetes, neurodegeneration, and inflammation. Glycobiology. 2005;15:16R-28R. 6. Schmidt AM, Yan SD, Yan SF, Stern DM. The biology of the receptor for advanced glycation end products and its ligands. Bba-Mol Cell Res. 2000;1498:99-111. 7. Goldin A, Beckman JA, Schmidt AM, Creager MA. Advanced glycation end products: sparking the development of diabetic vascular injury. Circulation. 2006;114:597-605. 8. Raposeiras-Roubin S, Rodino-Janeiro BK, Paradela-Dobarro B, Grigorian-Shamagian L, Garcia-Acuna JM, Aguiar-Souto P, et al. Fluorescent advanced glycation end products and their soluble receptor: the birth of new plasmatic biomarkers for risk stratification of acute coronary syndrome. Plos One. 2013;8:e74302. 9. Schalkwijk CG, Ter Wee PM, Stehouwer CD. Plasma levels of AGE peptides in type 1 diabetic patients are associated with serum creatinine and not with albumin excretion rate: possible role of AGE peptide-associated endothelial dysfunction. Annals of the New York Academy of Sciences. 2005;1043:662-70. 10. Yamagishi S. Role of advanced glycation end products (AGEs) and receptor for AGEs (RAGE) in vascular damage in diabetes. Exp Gerontol. 2011;46:217-24. 11. Buongiorno AM, Morelli S, Sagratella E, Castaldo P, Di Virgilio A, Maroccia E, et al. Levels of advanced glycosylation endproducts (AGE) in sera of pregnant diabetic women: comparison between type 1, type 2 and gestational diabetes mellitus. Annali dellâ&#x20AC;&#x2122;Istituto superiore di sanita. 1997;33:375-8. 12. Pertynska-Marczewska M, Glowacka E, Sobczak M, Cypryk K, Wilczynski J. Glycation endproducts, soluble receptor for advanced glycation endproducts and cytokines in diabetic and non-diabetic pregnancies. Am J Reprod Immunol. 2009;61:175-82. 13. de Ranitz-Greven WL, Bos DC, Poucki WK, Visser GH, Beulens JW, Biesma DH, et al. Advanced glycation end products, measured as skin autofluorescence, at diagnosis in gestational diabetes mellitus compared with normal pregnancy. DiabetesTechnolTher. 2012a;14:43-9. 14. de Ranitz-Greven WL, Kaasenbrood L, Poucki WK, Hamerling J, Bos DC, Visser GH, et al. Advanced glycation end products, measured as skin autofluorescence, during normal pregnancy and pregnancy complicated by diabetes mellitus. Diabetes Technol Ther. 2012b;14:1134-9.

17. Ikeda K, Higashi T, Sano H, Jinnouchi Y, Yoshida M, Araki T, et al. N-epsilon-(carboxymethyl)lysine protein adduct is a major immunological epitope in proteins modified with advanced glycation end products of the Maillard reaction. Biochemistry. 1996;35:8075-83. 18. Munch G, Keis R, Wessels A, Riederer P, Bahner U, Heidland A, et al. Determination of advanced glycation end products in serum by fluorescence spectroscopy and competitive ELISA. Eur J Clin Chem Clin. 1997;35:669-77. 19. Wrobel K, Wrobel K, GaraySevilla ME, Nava LE, Malacara JM. Novel analytical approach to monitoring advanced glycosylation end products in human serum with on-line spectrophotometric and spectrofluorometric detection in a flow system. Clin Chem. 1997;43:1563-9. 20. Zilin S, Naifeng L, Bicheng L, Jiping W. The determination of AGEpeptides by flow injection assay, a practical marker of diabetic nephropathy. Clin Chim Acta. 2001;313:69-75. 21. Diabetes SBD. Diretrizes da Sociedade Brasileira de Diabetes. 2009. 22. Kalousova M, Skrha J, Zima T. Advanced glycation end-products and advanced oxidation protein products in patients with diabetes mellitus. Physiol Res. 2002;51:597-604. 23. Hayase F, Nagaraj RH, Miyata S, Njoroge FG, Monnier VM. Aging of proteins: immunological detection of a glucose-derived pyrrole formed during maillard reaction in vivo. J Biol Chem. 1989;264:3758-64. 24. Makita Z, Vlassara H, Cerami A, Bucala R. Immunochemical detection of advanced glycosylation end products in vivo. The Journal of biological chemistry. 1992;267:5133-8. 25. Guosheng L, Hongmei S, Chuan N, Haiying L, Xiaopeng Z, Xianqiong L. The relationship of serum AGE levels in diabetic mothers with adverse fetal outcome. J Perinatol. 2009;29:483-8. 26. Makita Z, Radoff S, Rayfield EJ, Yang Z, Skolnik E, Delaney V, et al. Advanced Glycosylation End-Products in Patients with Diabetic Nephropathy. New Engl J Med. 1991;325:836-42. 27. Wagner Z, Wittmann I, Mazak I, Schinzel R, Heidland A, KientschEngel R, et al. N-epsilon-(carboxymethyl)lysine levels in patients with type 2 diabetes: Role of renal function. Am J Kidney Dis. 2001;38:785-91. 28. Sharp PS, Rainbow S, Mukherjee S. Serum levels of low molecular weight advanced glycation end products in diabetic subjects. Diabetic Med. 2003;20:575-9. 29. Miyata T, Ueda Y, Horie K, Nangaku M, Tanaka S, de Strihou CV, et al. Renal catabolism of advanced glycation end products: The fate of pentosidine. Kidney Int. 1998;53:416-22.

15. Mitsuhashi T, Vlassara H, Founds HW, Li YM. Standardizing the immunological measurement of advanced glycation endproducts using normal human serum. J Immunol Methods. 1997;207:79-88.

16. Jaisson S, Gillery P. Evaluation of nonenzymatic posttranslational modification-derived products as biomarkers of molecular aging of proteins. Clin Chem. 2010;56:1401-12.

Arch Endocrinol Metab. 2017;61/3

237

original article

Type 2 diabetes-associated genetic variants of FTO, LEPR, PPARg, and TCF7L2 in gestational diabetes in a Brazilian population Mauren Isfer Anghebem-Oliveira1,2, Bruna Rodrigues Martins1, Dayane Alberton1, Edneia Amancio de Souza Ramos3, Geraldo Picheth1, Fabiane Gomes de Moraes Rego1

ABSTRACT Departamento de Análises Clínicas, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brasil 2 Escola de Ciências da Vida, Pontifícia Universidade Católica do Paraná (PUCPR), Curitiba, PR, Brasil 3 Departamento de Patologia Básica, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brasil 1

Correspondence to: Fabiane Gomes de Moraes Rego Departamento de Análises Clínicas, Universidade Federal do Paraná Rua Prefeito Lothário Meissner, 632 80210-170 – Curitiba, PR, Brasil rego@ufpr.br Received on Mai/5/2016 Accepted on Oct/18/2016 DOI: 10.1590/2359-3997000000258

Objective: Gestational diabetes mellitus (GDM) is a metabolic disorder that shares pathophysiologic features with type 2 diabetes mellitus. The aim of this study was to investigate the association of the polymorphisms fat mass and obesity-associated (FTO) rs1421085, leptin receptor (LEPR) rs1137100, rs1137101, peroxisome proliferator-activated receptor gamma (PPARg) rs1801282, and transcription factor 7-like 2 (TCF7L2) rs7901695 with GDM. Subjects and methods: 252 unrelated Euro-Brazilian pregnant women were classified into two groups according to the 2015 criteria of the American and Brazilian Diabetes Association: healthy pregnant women (n = 125) and pregnant women with GDM (n = 127), matched by age. The polymorphisms were genotyped using fluorescent probes (TaqMan®). Results: All groups were in Hardy-Weinberg equilibrium. The genotype and allele frequencies of the studied polymorphisms did not show significant differences between the groups (P > 0.05). In the healthy and GDM groups, the C allele frequencies (95% CI) of the FTO rs1421085 polymorphism were 36.8% [31–43%] and 35.0% [29–41%]; the G allele frequencies (95% CI) of the LEPR rs1137100 polymorphism were 24.8% [19–30%] and 22.8% [18–28%]; the G allele frequencies (95% CI) of the LEPR rs1137101 polymorphism were 43.6% [37–50%] and 42.9% [37–49%]; the G allele frequencies (95% CI) of the PPARg rs1801282 polymorphism were 7.6% [4–11%] and 8.3% [5–12%]; and the C allele frequencies (95% CI) of the TCF7L2 rs7901695 polymorphism were 33.6% [28–39%] and 39.0% [33–45%], respectively. Conclusion: The studied polymorphisms were not associated with GDM in a Brazilian population. Arch Endocrinol Metab. 2017;61(3):238-48. Keywords Diabetes; gestational diabetes mellitus; polymorphisms; genetic variants; genotype

INTRODUCTION

estational diabetes mellitus (GDM) is a complex metabolic disorder defined as glucose intolerance diagnosed in the second or third trimester of pregnancy (1). In Brazil, about 7% of pregnant women exhibit GDM and this prevalence is increasing in parallel with the obesity epidemic (2). As for type 2 diabetes mellitus (T2DM), the pathogenesis of GDM is associated with insulin resistance owing to a reduction of beta cell function. In GDM, pancreatic beta cells are unable to produce enough insulin to compensate for the insulin resistance that commonly occurs during pregnancy (3,4). GDM and T2DM have similar pathophysiologic features, suggesting that GDM is also a polygenic disease (5). Therefore, studies on the etiology of GDM have 238

primarily been based on T2DM-associated genetic variants (6). To further elucidate the genetic mechanisms underlying GDM, we selected several gene polymorphisms previously associated with T2DM: rs1421085 (fat mass and obesity associated; FTO), rs1137100 and rs1137101 (leptin receptor; LEPR), rs1801282 (peroxisome proliferator-activated receptor gamma; PPARg), and rs7901695 (transcription factor 7-like 2; TCF7L2) and investigated their association with GDM. To our knowledge, this is the first study involving these genetic variants and GDM in a Brazilian population. Figure 1 illustrates the genes and polymorphisms studied. FTO is a protein-coding gene located at the chromosome region 16q12.2 and associated with the Arch Endocrinol Metab. 2017;61/3

T2DM-associated genetic variants and GDM

control of food intake and energy balance (7). Because of the relationship between FTO and obesity, several studies have been conducted to verify the association of FTO polymorphisms and T2DM (8,9). In certain populations, FTO variants increase susceptibility to T2DM independent of their effect on weight gain, suggesting that changes in the environment or other genetic factors may contribute to the different associations observed between different ethnic groups (10). Leptin is a hormone produced by adipocytes and other tissues such as the gastric mucosa, which acts as a signaling molecule in the regulation of body fat mass by negative feedback. Reaching the brain via the bloodstream, leptin acts on hypothalamic receptors thereby reducing appetite, stimulating energy consumption and the loss of body mass as well as the sympathetic nervous system (11). Leptin exerts its function by binding to leptin receptors (Ob-R). The soluble isoform of the Ob-R receptor, called Ob-Re, is generated by alternative splicing or by proteolytic Arch Endocrinol Metab. 2017;61/3

cleavage of the membrane isoform. Leptin molecules may circulate as the free form or linked to Ob-R. This binding appears to postpone the action of leptin as the leptin-soluble receptor complex increases the half-life of leptin and modulates its action on target cells (12). Leptin receptors are encoded by the LEPR gene (13). Leptin receptors are located mainly in the hypothalamus, but are also found in tissues and cells that regulate glucose homeostasis such as pancreatic beta cells. Therein, leptin receptors effect the inhibition of insulin secretion mediated by leptin. This behavior suggests LEPR as a candidate gene for association with DM (14). Consistent with this, LEPR was recently identified as a strong candidate for GDM (15) and was therefore selected for our study. Peroxisome proliferator activated receptors, known as PPARs, belong to the superfamily of hormonal nuclear receptors that act as transcription factors regulating gene expression. PPARs play a key role in regulating cell differentiation, the metabolism of 239

Figure 1. Genomic structure of the studied genes and the location of the selected polymorphisms. rs: reference sequence. (A) FTO: fat mass and obesity associated gene. (B) LEPR: leptin receptor gene. (C) PPARg: peroxisome proliferator-activated receptor gamma gene. (D) TCF7L2: transcription factor 7-like 2 gene.

T2DM-associated genetic variants and GDM

glucose and lipids, and inflammation (16). In long periods of hypoglycemia, PPARs are involved in the supply of fatty acids and ketone bodies from adipose tissue as a source of energy. There are 3 subtypes of PPARs encoded by distinct genes: α, β/δ, and γ (Figure 1). PPARγ (PPARg) acts as a mediator of the link between lipid and glucose metabolism (17). Accordingly, PPARg agonists have been used in the treatment of dyslipidemia and hyperglycemia (18). TCF7L2 is a transcription factor involved in stimulating the proliferation of pancreatic beta cells and the production of GLP-1, which stimulates insulin secretion (19). The TCF7L2 gene has been associated with T2DM in different populations (19-21). Common variants in intronic region of TCF7L2 have been identified as strong predictors of T2DM genetic risk (22). TCF7L2 polymorphisms modulate blood glucose and insulin secretion and the association of rs7901695 with GDM has previously been described in Caucasians (19). Another TCF7L2 polymorphism (rs7903146), which is in linkage disequilibrium with rs7901695, has been associated with GDM in Danish (23), Australian (21), Greek (5), and Swedish (20) populations.

SUBJECTS AND METHODS This cross-sectional study was conducted with unrelated Euro-Brazilian pregnant women who attended a Public Hospital in Southern Brazil. The ethnic group classiﬁcation was based on phenotype (pigmentation of the abdomen, hair color, and type and conformation of the nose and lips). Participants having white skin and European physical characteristics and/or self-reported the European ancestry was classified as Euro-Brazilians (24). The Ethics Committee of the Federal University of Parana approved the study and written informed consent was obtained from the subjects. Euro-Brazilians pregnant women with any gestational period were included in the study. Subjects with kidney (preeclampsia) disease, cardiovascular disease or infection disease were excluded. The sample (n = 252) was classified into two groups according to the 2015 criteria of the American Diabetes Association (1) and Brazilian Diabetes Association (2): healthy pregnant women (Control; n = 125) and patients with GDM (n = 127). The groups were matched by age. Clinical and anthropometric data were obtained from all subjects. Hypertension was defined according 240

to the VI Brazilian Hypertension Guidelines (25), which consider hypertension in pregnancy when the systolic blood pressure > 140 mmHg and/or diastolic blood pressure > 90 mmHg. No subjects were under antihypertensive treatment. Biochemical parameters were determined by routine laboratory methods using the automated system Architect ci8200 (Abbott Laboratórios do Brasil, Curitiba, PR, Brazil) with reagents, calibrators, and controls as recommended by the manufacturer. 1,5 anhydroglucitol concentrations were measured enzymatically (GlycoMark, Inc., New York, NY, USA). Glycated hemoglobin (HbA1c) was measured using Cobas Integra® 400 Plus (Roche Diagnostica Brasil Ltda., São Paulo, SP, Brazil) with the A1C-2 Tina-Quant® Hemoglobin A1c Gen 2 reagent method certified by the National Glycohemoglobin Standardization Program. DNA was extracted from whole blood using a modified “salting out” method (26) and the concentrations were normalized to 20 ng/μL for subsequent assays. Only samples with absorbance ratios (280/260) between 1.8 and 2.0 (NanoDrop, ThermoScientific, Waltham, MA, USA) were used in this study. The polymorphisms were genotyped using real-time polymerase chain reaction (PCR) amplification with fluorescent probes as described in Table 1. Genotyping experiments were carried out using a 7500 Fast™ Real-Time PCR System (Life Technologies/Applied Biosystems Foster City, CA, USA). The manufacturer provided the reagents (Master Mix® and Genotyping Assay® SNPs) and other realtime PCR materials (Applied Biosystems). The reaction mixture (8 μL final volume) contained 3.0 μL Master Mix (DNA polymerase, Mg2+, buffer, and additives), 0.1 μL SNP Genotyping Assay (40X), 1.9 μL ultrapure water, and 3.0 μL genomic DNA at 20 ng/μL. The cycle sequencing conditions were: 1 cycle of 1 min at 60°C (pre-PCR), 1 cycle of 10 min at 95°C, 45 cycles of 15 s at 95°C followed by 60°C for 90 s, and 1 final cycle of 30 s at 60°C (final extension). The quality of the genotyping was 98% or higher. Normality was tested with the Kolmogorov-Smirnov test. Comparisons of parameters with normal distribution were performed using the Student t-test for independent samples or the Mann-Whitney U test for non-normally distributed variables. Categorical variables were compared using the Fisher exact test (two-tailed) or the Chi-square test, as appropriate. Allele frequencies and HardyWeinberg (HW) equilibrium were evaluated by the Chisquare test (http://ihg.gsf.de/cgi-bin/hw/hwa1.pl). Arch Endocrinol Metab. 2017;61/3

T2DM-associated genetic variants and GDM

Table 1. Characteristics of the studied polymorphisms OMIM number

Gene

Chromosome position

Location

Polymorphism

TaqMan® probe

610966

FTO

16q12.2

Intron 1

rs1421085

C>T

C___8917103_10

601007

LEPR

1p31.3

Exon 4 Exon 6

rs1137100 rs1137101

A>G A>G

C____518168_20 C___8722581_10

601487

PPARg

3p25.2

Exon B

rs1801282

C>G

C___1129864_10

602228

TCF7L2

10q25.2– 25.3

Intron 3

rs7901695

C>T

C___384583_10

OMIM: Online Mendelian Inheritance in Man®. References: (46,47).

RESULTS Anthropometric and laboratorial data are presented in Table 2. The GDM group presented a higher body mass index (32.7 ± 5.0 vs 26.9 ± 6.3 kg/m2; P < 0.001). The mean of systolic and diastolic blood pressure (mmHg) for control and GDM groups were 108.8 ± 13.3/68.7 ± 9.2 vs 118.4 ± 12.9/73.8 ± 10.5 (P < 0.001). The frequency of hypertension in subjects with GDM differed from the control group, being higher (14.9% vs 4.8%, respectively, P = 0.007). GDM group also had a higher family history of DM (approximately 70%) when compared with Euro-descendant healthy pregnant women from Portugal (16%) (27) and France (19%) (28). The median (control vs GDM) of fasting glucose and 2-h, 75g glucose, 84 (79–88) vs 88 (81– 95) mg/dL and 86.2 (79–102) vs 161.0 (149–176) mg/dL, showed higher concentration in GDM group (P < 0.001). In GDM group no one of the patients were under insulin therapy. The lipid profile showed no difference between the groups (P > 0.05), but triglycerides was higher in GDM group, 124.0 (96– 171) vs 221.0 (175–270) mg/dL (P < 0.001) (Table 2). The other parameters were within reference range for both groups. The genotype and allele frequencies of the studied polymorphisms were not significantly different (P > 0.05) between the groups (Table 3). All genotypes were in Hardy-Weinberg equilibrium for both groups (P > 0.05). One-way ANOVA and correlation analysis (Pearson) were applied to identify association and correlation between laboratory biomarkers described in Table 2 and the studied polymorphisms. Genotypes for all polymorphisms, coded 1 (common homozygous), Arch Endocrinol Metab. 2017;61/3

2 (heterozygous) and 3 (rare homozygous), showed no significance (P > 0.05) for all analysis.

DISCUSSION Elevated BMI is a strong predictor of GDM and insulin resistance (29). In the present study, pregnant women with GDM were found to be heavier than healthy pregnant women (Table 2). The rate of hypertension in the GDM group was higher than that reported in the literature, which is 5–10% (30,31). However, GDM group showed good glycemic control as assessed by a HbA1c median value of 5.6%. Notably, pregnant women with a family history for DM are at increased risk for developing GDM and for giving birth to macrosomic children (32,33). GDM also induces a dyslipidemia state that is consistent with insulin resistance (34); triglycerides concentrations in particular were higher in the GDM group (P < 0.001). Finally, although total protein, albumin, creatinine, urea and uric acid levels differed between the groups (P < 0.001), they were within the reference range for these parameters and none of the subjects exhibited clinical symptoms of kidney disease or others pathologies. Pregnancy affects essentially all aspects of kidney physiology. Glomerular filtration rate (GFR) increases 50% as compared with nonpregnant levels (35) with subsequent decrease in serum creatinine, urea, and uric acid values (36). Also, increases in glomerular filtration rate and minor increases in urinary albumin excretion have been reported early in the course of diabetes (37,38). These events, could explain the lower total proteins, albumin, creatinine and urea levels found in GDM compared to control group. The uric acid concentration in GDM was approximately 0.7 mg/dL higher than in control group (4.3 vs 3.6 mg/dL, p < 0.001) (Table 2). This observations has been described in other studies and associated with insulin resistance and hypertension, which predominates in GDM group (39-41). 241

Data analysis was performed using Statistica for Windows 8.0 software (StatSoft Inc., Tulsa, OK, USA), and a probability less than 5% (P < 0.05) was considered significant for all analyses.

T2DM-associated genetic variants and GDM

Table 2. Anthropometric and laboratory characteristics of the studied groups Characteristics

Control (n = 125)

GDM (n = 127)

Age, years

30.6 ± 4.7

31.9 ± 6.4

0.070

Body mass indexa, kg/m2

26.9 ± 5.0

32.7 ± 6.3

< 0.001

4.8

14.9

0.007*

–

70.1

–

Hypertension, % Family history for diabetes, %

Fasting glucose, mg/dL

84.0 (79–88)

88 (81–95)

< 0.001**

2-h 75g glucose, mg/dL

86.2 (79–102)

161.0 (149–176)

< 0.001**

HbA1c, %

–

5.6 (5.3–5.9)

–

11.7 ± 6.9

9.8 ± 5.1

0.060

Total cholesterol, mg/dL

213.9 ± 50.0

224.7 ± 45.6

0.074

HDL-cholesterol, mg/dL

55.4 ± 15.8

56.8 ± 12.5

0.451

LDL-cholesterol, mg/dL

130.9 ± 41.5

123.5 ± 38.9

0.146

124.0 (96–171)

221.0 (175–270)

< 0.001**

Total protein, g/dL

6.9 ± 0.8

6.4 ± 0.5

< 0.001

Albumin, g/dL

4.2 ± 0.6

3.4 ± 0.4

< 0.001

0.80 (0.7–0.9)

0.70 (0.60–0.72)

< 0.001**

1,5 anhydroglucitol, µg/mL

Triglycerides, mg/dL

Creatinine, mg/dL Urea, mg/dL Uric acid, mg/dL

20.4 ± 5.3

16.1 ± 4.8

< 0.001

3.6 (3.0–3.9)

4.3 (3.7–4.9)

< 0.001**

Values are presented as mean ± SD, median (interquartile range) or %; “ –”: no information available. Control: healthy pregnant women; GDM: gestational diabetes. a Pregnant BMI calculated at blood collection time. P-values, Student’s t-test (independent variables), * Chi-square test or ** Mann-Whitney U test.

Table 3. Genotype and allele frequencies of the genetic variants studied in the Control and GDM groups Gene/SNP

Genotypes

Control n = 125

GDM n = 127

FTO rs1421085 (C>T)

53 (42.4)

52 (40.9)

0.420*

LEPR rs1137100 (A>G)

LEPR rs1137101 (A>G)

PPARg rs1801282 (C>G)

TCF7L2 rs7901695 (C>T)

52 (41.6)

61 (48.0)

20 (16.0)

14 (11.1)

C-allele [95% CI]

36.8 [31–43]

35.0 [29–41]

0.680

70 (56.0)

73 (57.5)

0.687*

48 (38.4)

50 (39.4)

7 (5.6)

4 (3.1)

G-allele [95% CI]

24.8 [19–30]

22.8 [18–28]

0.604

43 (34.4)

38 (29.9)

0.228*

55 (44.0)

69 (54.3)

27 (21.6)

20 (15.8)

G-allele [95% CI]

43.6 [37–50]

42.9 [37–49]

0.876

107 (85.6)

108 (85.0)

0.851*

17 (13.6)

17 (13.4)

1 (0.8)

2 (1.6)

G-allele [95% CI]

7.6 [4–11]

8.3 [5–12]

0.782

52 (41.6)

44 (34.6)

0.413*

62 (49.6)

67 (52.8)

11 (8.8)

16 (12.6)

C-allele [95% CI]

33.6 [28–39]

39.0 [33–45]

0.209

Genotypes depicted as number (%). 95% CI: Confidence interval of 95%. All polymorphisms were in Hardy-Weinberg equilibrium. P: probability. Chi-square test or * Two tailed Fisher’s Exact test.

242

Arch Endocrinol Metab. 2017;61/3

T2DM-associated genetic variants and GDM

The FTO rs9939609 predisposes to childhood, adult (42,43) and pregnancy obesity (44). The FTO rs1421085 polymorphism is associated with obesity in adults and in European and Chinese children (45) and with obesity and T2DM in African Americans (46). No information is available regarding its frequency in pregnant women or its association with GDM, underlying why this polymorphism was selected for analysis in the present study. However, the FTO rs1421085 polymorphism was not associated with GDM in the current population studied, nor with the parameters analyzed. The effect of sample size might have contributed to this result. Furthermore, the physiological effect of the presence of this intronic variant and the reported increased risk of obesity and DM need to be elucidated. The frequency of the C allele on healthy pregnant women in this study was approximately two to threefold higher than the frequency reported in European, Chinese, and Japanese populations, whereas it was approximately 5 times lower than that reported in an African populations (HapMap â&#x20AC;&#x201C; YRI). Brazilians are an admixture population. This genetic background could explain the differences of alleles frequencies when compared to others populations, even with populations with more similar ancestors such as Caucasians. Table 4 compares the reported risk allele frequencies in healthy subjects, including the current cohort of pregnant women, from different populations.

LEPR rs1137100 and rs1137101 polymorphisms The LEPR rs1137100 polymorphism (K109R; Lys109Arg; 326A > G) is characterized by an A > G substitution resulting in a change of lysine to arginine at codon 109 (Figure 1). Caucasians with the AA genotype are 2-fold more likely to develop T2DM than those with other genotypes (47). The LEPR rs1137101 polymorphism (Q223R; Gln223Arg; 668A > G) is an A> G substitution resulting in the exchange of glutamine for arginine at position 668 of LEPR (Figure 1). The presence of this variant was considered as an independent risk factor for T2DM in Malaysian subjects (48). Finns with glucose intolerance and the presence of the GG genotype (Finnish Diabetes Prevention Study) showed a higher risk for T2DM compared to those carrying the A allele. The rs1137100 and rs1137101 variants are each located in the region encoding the extracellular Arch Endocrinol Metab. 2017;61/3

domain of the leptin receptor. The exchange of amino acids generated by the variants affects all receptor isoforms and might change the action of leptin toward insulin (49). The allele frequency for rs1137100 G reported in the present studied was higher than the rate reported for Africans and significantly lower than that for Asians. The ethnic composition of the sample of this study (Euro-Brazilians) is the probable reason for the similarity with the frequencies reported for the English, French, and Europeans in general (Table 4). Elevated leptin concentrations are associated with adiposity and insulin resistance and it has been reported that individuals with the AA genotype of LEPR rs1137100 showed higher concentrations than the G allele carriers. Although the function of this variant has not been elucidated, it was associated with the promotion of a change in the extracellular domain of the receptor that affects the leptin binding affinity (50). In this study it was not possible to verify the association of the polymorphism with obesity, BMI, or lipid or glucose changes (data not showed). The frequency of G allele for rs1137101 variant was similar to those reported for Europeans and lower than those of Africans (HapMap). Asian subjects presented a higher frequency of the G allele for rs1137101 than other populations, as shown in Table 4. The variants rs1137100 and rs1137101 of LEPR were not associated with GDM in the studied population, nor with the analyzed parameters described in Table 2 (P > 0.05).

PPARg rs1801282 polymorphism The PPARg rs1801282 polymorphism results in a substitution of proline for alanine at codon 12 of exon B (C > G; Pro12Ala; Figure 1). This polymorphism causes a conformational change in the protein, and the presence of the minor allele is associated with a reduction in the activity of PPARg (50). The rs1801282 C allele is associated with increased transcriptional activity of PPARg and, consequently, increased sensitivity to insulin. The association with T2DM is controversial, as some studies have found a positive association while others showed that the presence of the variant conferred protection against T2DM (51). Similarly, some studies found no association of rs1801282 with GDM (5,23). In a study conducted in Turkey, this polymorphism had no effect on the prevalence of GDM or glucose concentrations in pregnant women, but its presence 243

FTO rs1421085 polymorphism

T2DM-associated genetic variants and GDM

impacted the weight of pregnant women with GDM (52). The presence of the rs1801282 GG genotype was associated with a higher BMI before pregnancy and a

higher pre-pregnancy obesity rate, but was related to a 50% reduction in the risk of developing GDM in a French population (53).

Table 4. Comparison between the allele frequencies of healthy pregnant women from the current study with those obtained from healthy subjects in other studies FTO rs1421085 polymorphism Population

Characteristics

Genotype (%)

Allele (%)

Reference

127 125

40.9 42.4

48.0 41.6

11.1 16.0

35.0 36.8

Present work

Chinese Han

70.7

24.4

4.9

11.6

HapMap-HCB

European

180

27.4

53.1

19.5

16.7

HapMap-CEU

67.1

28.2

4.7

18.8

HapMap-JPT

190 97

30.5 28.8

51.1 52.6

18.4 18.6

43.9 44.8

(48)

29.5

30.5

55.2

(49)

86.7

13.3

6.6

HapMap-YRI

Allele (%)

Reference

Euro-Brazilian

GDM Controls

Japanese Turkish

Obeses Controls

Caucasian

Obeses

African

LEPR rs1137100 polymorphism Population Euro-Brazilian

Characteristics GDM Controls

African

Genotype (%)

N AA

127 125

57.5 56.0

39.4 38.4

3.1 5.6

22.8 24.8

Present work

69.6

27.7

2.7

16.5

HapMap-YRI

French

Obeses Controls

877 877

56.7 53.4

37.4 39.2

5.9 7.4

24.6 27.0

(50)

British

Obeses Controls

190 132

54.0 55.0

38.0 39.0

8.0 6.0

27.4 25.8

(51)

180

49.6

42.5

8.0

29.2

HapMap-CEU

470

42.8

46.8

10.4

33.8

(52)

Chinese Han

2.5

27.5

70.0

83.7

HapMap-HCB

Japanese

2.5

40.2

57.3

97.4

HapMap-JPT

Allele (%)

Reference

European Brazilians

Hypertension

LEPR rs1137101 polymorphism

Population

Characteristics

Genotype (%)

N AA

Euro-Brazilian

GDM Controls

127 125

29.9 34.4

54.3 44.0

15.8 21.6

42.9 43.6

Present work

French

Obeses Controls

877 877

31.6 31.6

50.0 47.1

18.4 21.3

43.4 44.9

(50)

British

Obeses Controls

190 132

29.0 28.0

53.0 58.0

18.0 14.0

44.6 42.8

(51)

European

180

25.9

53.6

20.5

47.3

HapMap-CEU

African

11.1

58.3

30.6

59.7

HapMap-YRI

Japanese

1.2

30.5

68.3

83.5

HapMap-JPT

Chinese Han

2.2

17.8

80.0

88.9

HapMap-HCB

244

Arch Endocrinol Metab. 2017;61/3

T2DM-associated genetic variants and GDM

PPARg rs1801282 polymorphism Population Euro-Brazilian

Characteristics

Genotype (%)

Allele (%)

Reference

GDM Controls

127 125

85 85.6

13.4 13.6

1.6 0.8

8.3 7.6

Present work

French

Mothers with glucose tolerance

1708

10.4

(42)

Danish

GDM Women with glucose tolerance

283 2446

75.8 75.2

22.6 22.7

1.6 2.1

12.8 13.5

(22)

Danish

T2DM Controls

1461 4986

76 75

22 23

2 2

13.0 13.9

(53)

Italian

Peripheral arterial disease Controls

201

74.1

23.9

2.0

14.0

(54)

201

84.6

14.9

0.5

8.0

Scandinavian

GDM Controls

637 1232

73.5 74.5

24.8 24.2

1.7 1.3

14.1 13.4

(19)

Scandinavian

GDM Controls

400 428

71.5 74.1

27.7 24.5

0.8 1.4

14.6 13.7

(44)

Arabs

GDM Controls

100 122

91 86.9

9 12.3

0 0.8

4.5 7.0

Turkish

GDM Controls

62 100

80.7 84

19.3 16

0 0

19.4 16.0

(41)

Korean

GDM Controls

94 41

94.6 82.9

5.4 17.1

0 0

2.7 8.5

(55)

Greek

GDM Controls

148 107

96.6 93.5

3.4 6.5

0 0

3.0 2.0

(5)

Chinese

GDM Controls

55 173

94.5 90.8

5.5 9.2

0 0

3.0 5.0

(56)

Korean

GDM Controls

869 632

91.7 89.7

8.2 10.0

0.1 0.3

4.0 5.0

(43)

TCF7L2 rs7901895 polymorphism

Euro-Brazilian

Characteristics GDM Controls

Chinese Han

Genotype (%)

Allele (%)

Reference

127 125

34.6 41.6

52.8 49.6

12.6 8.8

39.0 33.6

Present work

95.3

4.7

2.3

HapMap-HCB

Swedish women

GDM Controls

1102 794

51.2 34.2

34.2 43.1

7.6 11.5

26.5 34.4

(45)

Swedish men

T2DM Controls

825 793

52.6 59.2

39.9 36.0

7.5 4.8

27.5 22.8

(57)

180

54.5

34.8

10.7

28.1

HapMap-CEU

European Japanese

94.2

4.7

0.1

3.5

HapMap-JPT

GDM GDM

30 40

(18)

GDM Controls

45 25

38 40

44 52

18 8

40 34

(58)

27.4

56.5

15.9

44.2

HapMap-YRI

Caucasian African-American Spanish African

Allele frequencies are presented as % [95% confidence interval]. The frequencies of the minor allele (T) that differ from the confidence interval (95%) for the healthy group of the current study are highlighted in bold. Arch Endocrinol Metab. 2017;61/3

245

Population

T2DM-associated genetic variants and GDM

PPARg rs1801282 was not associated with GDM in the current studied population, nor with the analyzed parameters described in Table 2 (P > 0.05). In accordance with the results of our study, no association of the variant with GDM was found in Korean (54), Scandinavian, or Arab pregnant women (55). The G allele frequencies observed in this study were similar to the French and greater than those reported for the Arab, Greek, Korean, and Chinese populations, whereas Danish, Scandinavian and Turkish populations showed a higher frequency of the G allele (Table 4). Regional population differences might explain these findings.

TCF7L2 rs7901695 polymorphism The T allele of rs7901695 conferred increased risk for GDM in American Caucasians, with an odds ratio of 1.98 (19). In Swedish patients with GDM the CT and CC genotypes of rs7901695 showed a strong association with GDM even after adjusting for maternal age, number of pregnancies, and family history of DM and HLA-DQ genotypes (56). Based on this information, we expected to find an association between the rs7901695 polymorphism and GDM in the current studied population, but this was not observed. The rs7901895 C allele is more common in African populations. The frequency found in healthy pregnant women in the present study is similar to those reported for Europeans in general and for Swedish and Spanish pregnant women. In this study, pregnant women with GDM showed allele frequencies similar to those of Spanish, Caucasian, and African American women with GDM, and slightly higher than those of Swedish women with GDM (Table 4). In contrast, less than 5% of the Asian population carries the C allele. Ethnicity-specific factors might be responsible for these differences. In conclusion, the FTO rs1421085, LEPR rs1137100 and rs1137101, PPARg rs1801282, and TCF7L2 rs7901695 polymorphisms were not associated with GDM in a Brazilian population, nor with the other parameters analyzed. The data from this study will likely contribute to the understanding the potential roles of these variants across populations; however, further research is required to identify the underlying factors influencing the risk of GDM in the Brazilian white population. Disclosure: no potential conflict of interest relevant to this article was reported. 246

REFERENCES 1. ADA. (2) Classification and diagnosis of diabetes. Diabetes Care. 2015;38 Suppl:S8-16. 2. SBD. Diretrizes SBD. Sociedade Brasileira de Diabetes. 2015. Available from: <http://www.diabetes.org.br/sbdonline/images/ docs/DIRETRIZES-SBD-2015-2016.pdf>. Accessed on: Apr. 20, 2016. 3. Buchanan TA, Xiang AH, Page KA. Gestational diabetes mellitus: risks and management during and after pregnancy. Nat Rev Endocrinol. 2012;8(11):639-49. 4.

Baz B, Riveline JP, Gautier JF. ENDOCRINOLOGY OF PREGNANCY: Gestational diabetes mellitus: definition, aetiological and clinical aspects. Eur J Endocrinol. 2016;174(2):R43-51.

5. Pappa KI, Gazouli M, Economou K, Daskalakis G, Anastasiou E, Anagnou NP, et al. Gestational diabetes mellitus shares polymorphisms of genes associated with insulin resistance and type 2 diabetes in the Greek population. Gynecol Endocrinol. 2011;27(4):267-72. 6. Robitaille J, Grant AM. The genetics of gestational diabetes mellitus: evidence for relationship with type 2 diabetes mellitus. Genet Med. 2008;10(4):240-50. 7. Merkestein M, Laber S, McMurray F, Andrew D, Sachse G, Sanderson J, et al. FTO influences adipogenesis by regulating mitotic clonal expansion. Nat Commun. 2015;6:6792. 8. Field SF, Howson JM, Walker NM, Dunger DB, Todd JA. Analysis of the obesity gene FTO in 14,803 type 1 diabetes cases and controls. Diabetologia. 2007;50(10):2218-20. 9. Legry V, Cottel D, Ferrieres J, Arveiler D, Andrieux N, Bingham A, et al. Effect of an FTO polymorphism on fat mass, obesity, and type 2 diabetes mellitus in the French MONICA Study. Metabolism. 2009;58(7):971-5. 10. Larder R, Cheung MK, Tung YC, Yeo GS, Coll AP. Where to go with FTO? Trends Endocrinol Metab. 2011;22(2):53-9. 11. Oswal A, Yeo G. Leptin and the control of body weight: a review of its diverse central targets, signaling mechanisms, and role in the pathogenesis of obesity. Obesity (Silver Spring). 2010;18(2):221-9. 12. Cammisotto P, Bendayan M. A review on gastric leptin: the exocrine secretion of a gastric hormone. Anat Cell Biol. 2012;45(1):1-16. 13. Winick JD, Stoffel M, Friedman JM. Identification of microsatellite markers linked to the human leptin receptor gene on chromosome 1. Genomics. 1996;36(1):221-2. 14. Emilsson V, Liu YL, Cawthorne MA, Morton NM, Davenport M. Expression of the functional leptin receptor mRNA in pancreatic islets and direct inhibitory action of leptin on insulin secretion. Diabetes. 1997;46(2):313-6. 15. Zhang Q, He M, Wang J, Liu S, Cheng H, Cheng Y. Predicting of disease genes for gestational diabetes mellitus based on network and functional consistency. Eur J Obstet Gynecol Reprod Biol. 2015;186:91-6. 16. Varga T, Czimmerer Z, Nagy L. PPARs are a unique set of fatty acid regulated transcription factors controlling both lipid metabolism and inflammation. Biochim Biophys Acta. 2011;1812(8):1007-22. 17. Janani C, Ranjitha Kumari BD. PPAR gamma gene--a review. Diabetes Metab Syndr. 2015;9(1):46-50. 18. Li Y, Qi Y, Huang TH, Yamahara J, Roufogalis BD. Pomegranate flower: a unique traditional antidiabetic medicine with dual PPAR-alpha/-gamma activator properties. Diabetes Obes Metab. 2008;10(1):10-7. 19. Stuebe AM, Wise A, Nguyen T, Herring A, North KE, Siega-Riz AM. Maternal genotype and gestational diabetes. Am J Perinatol. 2014;31(1):69-76. 20. Shaat N, Lernmark A, Karlsson E, Ivarsson S, Parikh H, Berntorp K, et al. A variant in the transcription factor 7-like 2 (TCF7L2) Arch Endocrinol Metab. 2017;61/3

T2DM-associated genetic variants and GDM

21. Freathy RM, Hayes MG, Urbanek M, Lowe LP, Lee H, Ackerman C, et al. Hyperglycemia and Adverse Pregnancy Outcome (HAPO) study: common genetic variants in GCK and TCF7L2 are associated with fasting and postchallenge glucose levels in pregnancy and with the new consensus definition of gestational diabetes mellitus from the International Association of Diabetes and Pregnancy Study Groups. Diabetes. 2010;59(10):2682-9. 22. Cauchi S, El Achhab Y, Choquet H, Dina C, Krempler F, Weitgasser R, et al. TCF7L2 is reproducibly associated with type 2 diabetes in various ethnic groups: a global meta-analysis. J Mol Med (Berl). 2007;85(7):777-82. 23. Lauenborg J, Grarup N, Damm P, Borch-Johnsen K, Jorgensen T, Pedersen O, et al. Common type 2 diabetes risk gene variants associate with gestational diabetes. J Clin Endocrinol Metab. 2009;94(1):145-50. 24. Krieger H, Morton NE, Mi MP, Azevedo E, Freire-Maia A, Yasuda N. Racial admixture in north-eastern Brazil. Ann Hum Genet. 1965;29(2):113-25. 25. Sociedade Brasileira de Cardiologia; Sociedade Brasileira de Hipertensão; Sociedade Brasileira de Nefrologia. VI Brazilian Guidelines on Hypertension. Arq Bras Cardiol. 2010;95(1 Suppl):1-51. 26. Lahiri DK, Nurnberger JI, Jr. A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res. 1991;19(19):5444. 27. Ribeiro AMC, Nogueira-Silva C, Melo-Rochae G, Pereira ML, Rocha A. Diabetes gestacional: determinação de fatores de risco para diabetes mellitus. Rev Port Endocrinol Diabetes Metab. 2015;10(1):6.

39. Seghieri G, Breschi MC, Anichini R, De Bellis A, Alviggi L, Maida I, et al. Serum homocysteine levels are increased in women with gestational diabetes mellitus. Metabolism. 2003;52(6):720-3. 40. Yoo TW, Sung KC, Shin HS, Kim BJ, Kim BS, Kang JH, et al. Relationship between serum uric acid concentration and insulin resistance and metabolic syndrome. Circ J. 2005;69(8):928-33. 41. Santos IC, Frigeri HR, Rea RR, Almeida AC, Souza EM, Pedrosa FO, et al. The glucokinase gene promoter polymorphism -30G>A (rs1799884) is associated with fasting glucose in healthy pregnant women but not with gestational diabetes. Clin Chim Acta. 2010;411(11-12):892-3. 42. Frayling TM, Timpson NJ, Weedon MN, Zeggini E, Freathy RM, Lindgren CM, et al. A common variant in the FTO gene is associated with body mass index and predisposes to childhood and adult obesity. Science. 2007;316(5826):889-94. 43. Dina C, Meyre D, Gallina S, Durand E, Korner A, Jacobson P, et al. Variation in FTO contributes to childhood obesity and severe adult obesity. Nat Genet. 2007;39(6):724-6. 44. Martins MC, Trujillo J, Farias DR, Struchiner CJ, Kac G. Association of the FTO (rs9939609) and MC4R (rs17782313) gene polymorphisms with maternal body weight during pregnancy. Nutrition. 2016;32(11-12):1223-30. 45. Wang L, Yu Q, Xiong Y, Liu L, Zhang X, Zhang Z, et al. Variant rs1421085 in the FTO gene contribute childhood obesity in Chinese children aged 3-6 years. Obes Res Clin Pract. 2013;7(1):e14-22. 46. Bressler J, Kao WH, Pankow JS, Boerwinkle E. Risk of type 2 diabetes and obesity is differentially associated with variation in FTO in whites and African-Americans in the ARIC study. PloS One. 2010;5(5):e10521.

28. Miailhe G, Kayem G, Girard G, Legardeur H, Mandelbrot L. Selective rather than universal screening for gestational diabetes mellitus? Eur J Obstet Gynecol Reprod Biol. 2015;191:95-100.

47. Salopuro T, Pulkkinen L, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, et al. Genetic variation in leptin receptor gene is associated with type 2 diabetes and body weight: The Finnish Diabetes Prevention Study. Int J Obes (Lond). 2005;29(10): 1245-51.

29. Basraon SK, Mele L, Myatt L, Roberts JM, Hauth JC, Leveno KJ, et al. Relationship of Early Pregnancy Waist-to-Hip Ratio versus Body Mass Index with Gestational Diabetes Mellitus and Insulin Resistance. Am J Perinatol. 2016;33(1):114-21.

48. Etemad A, Ramachandran V, Pishva SR, Heidari F, Aziz AF, Yusof AK, et al. Analysis of Gln223Agr polymorphism of Leptin Receptor Gene in type II diabetic mellitus subjects among Malaysians. Int J Mol Sci. 2013;14(9):19230-44.

30. Cho GJ, Park JH, Lee H, Yoo S, Shin SA, Oh MJ. Prepregnancy Factors as Determinants of the Development of Diabetes Mellitus After First Pregnancy. J Clin Endocrinol Metab. 2016;101(7): 2923-30.

49. Tabassum R, Mahendran Y, Dwivedi OP, Chauhan G, Ghosh S, Marwaha RK, et al. Common variants of IL6, LEPR, and PBEF1 are associated with obesity in Indian children. Diabetes. 2012;61(3):626-31.

31. Shahbazian H, Nouhjah S, Shahbazian N, Jahanfar S, Latifi SM, Aleali A, et al. Gestational diabetes mellitus in an Iranian pregnant population using IADPSG criteria: Incidence, contributing factors and outcomes. Diabetes Metab Syndr. 2016;10(4):242-6.

50. Meirhaeghe A, Amouyel P. Impact of genetic variation of PPARgamma in humans. Mol Genet Metab. 2004;83(1-2):93-102.

32. Zhang C, Ning Y. Effect of dietary and lifestyle factors on the risk of gestational diabetes: review of epidemiologic evidence. Am J Clin Nutr. 2011;94(6 Suppl):1975S-9S.

52. Tok EC, Ertunc D, Bilgin O, Erdal EM, Kaplanoglu M, Dilek S. PPAR-gamma2 Pro12Ala polymorphism is associated with weight gain in women with gestational diabetes mellitus. Eur J Obstet Gynecol Reprod Biol. 2006;129(1):25-30.

33. Ben-Haroush A, Yogev Y, Hod M. Epidemiology of gestational diabetes mellitus and its association with Type 2 diabetes. Diabet Med. 2004;21(2):103-13. 34. Ryckman KK, Spracklen CN, Smith CJ, Robinson JG, Saftlas AF. Maternal lipid levels during pregnancy and gestational diabetes: a systematic review and meta-analysis. BJOG. 2015;122(5): 643-51. 35. Dunlop W. Serial changes in renal haemodynamics during normal human pregnancy. Br J Obstet Gynaecol. 1981;88(1):1-9. 36. Cheung KL, Lafayette RA. Renal physiology of pregnancy. Adv Chronic Kidney Dis. 2013;20(3):209-14. 37. Mogensen CE. Microalbuminuria as a predictor of clinical diabetic nephropathy. Kidney Int. 1987;31(2):673-89. 38. Parving HH, Viberti GC, Keen H, Christiansen JS, Lassen NA. Hemodynamic factors in the genesis of diabetic microangiopathy. Metabolism. 1983;32(9):943-9. Arch Endocrinol Metab. 2017;61/3

51. Groop L, Pociot F. Genetics of diabetes--are we missing the genes or the disease? Mol Cell Endocrinol. 2014;382(1):726-39.

53. Heude B, Pelloux V, Forhan A, Bedel JF, Lacorte JM, Clement K, et al. Association of the Pro12Ala and C1431T variants of PPARgamma and their haplotypes with susceptibility to gestational diabetes. J Clin Endocrinol Metab. 2011;96(10):E1656-60. 54. Cho YM, Kim TH, Lim S, Choi SH, Shin HD, Lee HK, et al. Type 2 diabetes-associated genetic variants discovered in the recent genome-wide association studies are related to gestational diabetes mellitus in the Korean population. Diabetologia. 2009;52(2):253-61. 55. Shaat N, Ekelund M, Lernmark A, Ivarsson S, Nilsson A, Perfekt R, et al. Genotypic and phenotypic differences between Arabian and Scandinavian women with gestational diabetes mellitus. Diabetologia. 2004;47(5):878-84. 56. Papadopoulou A, Lynch KF, Shaat N, Hakansson R, Ivarsson SA, Berntorp K, et al. Gestational diabetes mellitus is associated

247

gene is associated with an increased risk of gestational diabetes mellitus. Diabetologia. 2007;50(5):972-9.

T2DM-associated genetic variants and GDM

with TCF7L2 gene polymorphisms independent of HLADQB1*0602 genotypes and islet cell autoantibodies. Diabet Med. 2011;28(9):1018-27. 57. NCBI. dbSNP Short Genetic Variantions 2013 [Available from: http:// www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?searchType=adhoc_ search&type=rs&rs=rs5393. 58. OMIM. Online Mendelian Inheritance in Man 2014 [Available from: http://www.omim.org/. 59. Solak M, Ozdemir Erdogan M, Yildiz SH, Ucok K, Yuksel S, Arikan Terzi ES, et al. Association of obesity with rs1421085 and rs9939609 polymorphisms of FTO gene. Molecular biology reports. 2014;41(11):7381-6. 60. Harbron J, van der Merwe L, Zaahl MG, Kotze MJ, Senekal M. Fat mass and obesity-associated (FTO) gene polymorphisms are associated with physical activity, food intake, eating behaviors, psychological health, and modeled change in body mass index in overweight/obese Caucasian adults. Nutrients. 2014;6(8):3130-52. 61. Phillips CM, Goumidi L, Bertrais S, Field MR, Ordovas JM, Cupples LA, et al. Leptin receptor polymorphisms interact with polyunsaturated fatty acids to augment risk of insulin resistance and metabolic syndrome in adults. J Nutr. 2010;140(2):238-44. 62. Gotoda T, Manning BS, Goldstone AP, Imrie H, Evans AL, Strosberg AD, et al. Leptin receptor gene variation and obesity: lack of association in a white British male population. Hum Mol Genet. 1997;6(6):869-76.

64. Hansen SK, Nielsen EM, Ek J, Andersen G, Glumer C, Carstensen B, et al. Analysis of separate and combined effects of common variation in KCNJ11 and PPARG on risk of type 2 diabetes. J Clin Endocrinol Metab. 2005;90(6):3629-37. 65. Catalano M, Cortelazzo A, Santi R, Contino L, Demicheli M, Yilmaz Y, et al. The Pro12Ala polymorphism of peroxisome proliferatoractivated receptor-gamma2 gene is associated with plasma levels of soluble RAGE (Receptor for Advanced Glycation Endproducts) and the presence of peripheral arterial disease. Clin Biochem. 2008;41(12):981-5. 66. Chon SJ, Kim SY, Cho NR, Min DL, Hwang YJ, Mamura M. Association of variants in PPARgamma(2), IGF2BP2, and KCNQ1 with a susceptibility to gestational diabetes mellitus in a Korean population. Yonsei Med J. 2013;54(2):352-7. 67. Cheng Y, Ma Y, Peng T, Wang J, Lin R, Cheng HD. [Genotype discrepancy between maternal and fetal Pro12Ala polymorphism of PPARG2 gene and its association with gestational diabetes mellitus]. Zhonghua Fu Chan Ke Za Zhi. 2010;45(3):170-3. 68. Mayans S, Lackovic K, Lindgren P, Ruikka K, Agren A, Eliasson M, et al. TCF7L2 polymorphisms are associated with type 2 diabetes in northern Sweden. Eur J Hum Genet. 2007;15(3):342-6. 69. Pagan A, Sabater-Molina M, Olza J, Prieto-Sanchez MT, BlancoCarnero JE, Parrilla JJ, et al. A gene variant in the transcription factor 7-like 2 (TCF7L2) is associated with an increased risk of gestational diabetes mellitus. Eur J Obstet Gynecol Reprod Biol. 2014;180:77-82.

63. Pena G, Guimaraes AL, Veloso RR, Reis TC, Gomes CS, Neto JF, et al. Leptin Receptor Gene Gln223Arg Polymorphism Is Not

Associated with Hypertension: A Preliminary Population-Based Cross-Sectional Study. Cardiol Res Pract. 2014;2014:879037.

248

Arch Endocrinol Metab. 2017;61/3

original article

Serum levels of leptin and adiponectin and clinical parameters in women with fibromyalgia and overweight/obesity Eduardo S. Paiva¹, Aline Andretta², Emmanuelle Dias Batista², Márcia Maria Marques Teles Lobo², Renata Costa de Miranda3, Renato Nisihara4, Maria Eliana Madalozzo Schieferdecker2, César L. Boguszewski5

ABSTRACT Objectives: The objectives of this study were to evaluate the serum levels of adipokines in women with fibromyalgia with and without overweight/obesity, and to correlate the adipokines levels with clinical parameters associated with fibromyalgia and adipose tissue mass (body fat). Subjects and methods: The study included 100 women divided into four groups: (a) fibromyalgia and overweight/obesity; (b) fibromyalgia and normal weight; (c) controls and overweight/obesity; and (d) controls and normal weight. Patients and controls were evaluated for clinical, anthropometric, and fibromyalgia-related parameters. Assessments included serum levels of leptin, adiponectin, monocyte chemoattractant protein-1 (MCP-1), and C-reactive protein (CRP). Levels of adipokines were further adjusted for fat mass. Results: Fibromyalgia patients with overweight/obesity or normal weight had no differences in clinical parameters. Unadjusted leptin levels were lower in fibromyalgia patients than controls, a finding that was more remarkable in fibromyalgia patients with overweight/obesity. Leptin levels had no correlation with clinical parameters of fibromyalgia or inflammation markers (MCP-1 and CRP), and adiponectin levels showed no difference between groups. Conclusions: No correlation was observed between adjusted leptin levels and clinical parameters of fibromyalgia. Patients with fibromyalgia and overweight/obesity presented lower levels of leptin than controls with overweight/ obesity. Arch Endocrinol Metab. 2017;61(3):249-56. Keywords Fibromyalgia; obesity; leptin; adiponectin

1 Departamento de Medicina Interna, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brasil 2 Departamento de Nutrição, UFPR, Curitiba, PR, Brasil 3 UFPR, Curitiba, PR, Brasil, University of Rome Tor Vergata, Rome, Italy 4 Laboratório de Imunopatologia, Hospital de Clínicas, UFPR, Curitiba, PR, Brasil 5 Serviço de Endocrinologia e Metabologia (SEMPR), Departamento de Medicina Interna, UFPR, Curitiba, PR, Brasil

Correspondence to: Eduardo S. Paiva Departamento de Medicina Interna, Universidade Federal do Paraná, Hospital de Clínicas Rua General Carneiro, 181 80060-900 – Curitiba, PR, Brasil eduevicky@terra.com.br Received on Oct/3/2016 Accepted on Nov/25/2016

INTRODUCTION

ibromyalgia is a clinical syndrome characterized by diffuse muscle pain and hyperalgesia on muscle palpation (1) in the absence of articular or muscle inflammation. Several other symptoms may be present, such as fatigue, non-restorative sleep, and cognitive changes represented by memory and concentration problems. Fibromyalgia is a common syndrome, with a worldwide prevalence of 2 to 5%, (2) and the most accepted pathophysiology explaining its occurrence involves a sensitization of the central nervous system to pain (1). Even though fibromyalgia is not an inflammatory disease, there is great interest in the study of cytokines in its pathophysiology, since these mediators are related to nociceptive sensitization (3). Adipocytes produce

Arch Endocrinol Metab. 2017;61/3

several substances that act as inflammatory mediators (4), including adipokines (or adipocytokines), which are cytokine-like substances mainly produced by the adipose tissue, such as leptin, adiponectin, and resistin (5). Adipokines act mainly in satiety mechanisms and body weight maintenance, but also present pro- and anti-inflammatory actions, as well as pro- and antinociceptive properties which modulate pain perception (5). Leptin tends to increase the levels of C-reactive protein (CRP) and may have a role in maintaining neuropathic pain (6) and pain related to knee osteoarthritis (7). Adiponectin is related to decreased inflammatory activity and leads to decreased levels of TNF-alpha, IL-6, and IL-1 and increased levels of IL10 (an anti-inflammatory cytokine) (8). 249

DOI: 10.1590/2359-3997000000248

Leptin in obese fibromyalgia patients

According to the literature, the occurrence of overweight and obesity is associated with the worsening of fibromyalgia symptoms (9,10). However, little is known about the role of adipokines in modulating fibromyalgia symptoms. There are also no studies correlating the levels of adipokines with the actual adipose tissue mass in patients with fibromyalgia. The main objective of this cross-sectional study was to evaluate the levels of leptin and adiponectin in patients with fibromyalgia with and without overweight/obesity. The secondary objective was to analyze the correlation of the adipokines levels with clinical fibromyalgia criteria and the impact of overweight/obesity on these same parameters.

SUBJECTS AND METHODS Women with a clinical diagnosis of fibromyalgia were recruited from the Fibromyalgia Outpatient Clinic at Hospital de Clínicas, Universidade Federal do Paraná (UFPR). The reason for including only women in the study was the low prevalence of fibromyalgia among men. The inclusion criterion was a diagnosis of fibromyalgia according to the 1990 classification criteria of the American College of Rheumatology (ACR) (11). Patients with fibromyalgia and depression or anxiety were only included if their treatment had remained unchanged for 3 months. The exclusion criteria were medication change over the previous 4 weeks, use of corticosteroids or anticytokine agents, pregnancy, lactation, and a diagnosis of diabetes, decompensated endocrine diseases, infectious diseases (over the previous 4 weeks), demyelinating neurological diseases, peripheral neuropathies, inflammatory articular diseases, systemic autoimmune diseases, severe cardiovascular diseases, malignancy (over the previous year), and severe psychiatric diseases (substance abuse, schizophrenia, psychosis). Nonsteroidal anti-inflammatory agents were suspended 48 hours before collection of blood samples, and all other medications were required to remain unchanged for at least 30 days. Additionally, we excluded patients with class II and III obesity (body mass index [BMI] ≥ 35 kg/m2 and > 40 kg/m², respectively) or low weight (BMI ≤ 18.5 kg/m2). The control group comprised employees of the Hospital de Clínicas at UFPR, matched by age and BMI with the fibromyalgia patients. Assessments of clinical and laboratory parameters and body composition by bone densitometry were 250

performed on the same day for each participant. The blood samples for laboratory tests were collected after a 12-hour fast and immediately placed on ice, centrifuged at 4°C, and stored at -80°C. Measurement of serum leptin, adiponectin, and monocyte chemoattractant protein-1 (MCP-1) were performed by enzyme immunoassay (ELISA, Quantikine RDSY-DLP00 and RDSY-DRP300, R & D Systems, Minneapolis, MN, USA) and the results were adjusted according to the participants’ fat mass, assessed by dual-energy X-ray absorption (DXA; Lunar Prodigy Advance, GE Healthcare, Pittsburgh, PA, USA) (12). Additionally, we measured CRP levels by nephelometry (Siemens BNII, Munich, Germany; minimum detection level 0.1 mg/dL). After a small meal offered by the investigators, the participants underwent a clinical evaluation that included tender points (TPs) count and evaluation of pain threshold in the trapezius muscle using a Fischer algometer (model FDK 20, Wagner Instruments, Greenwich, CT, USA). Each TP was manually palpated with a strength of 4 kg/cm2 and the response was recorded as positive (with pain) or negative (without pain). In order to measure the pain threshold in the TP located in the right trapezius, the algometer was placed against the skin of the participant and pressed with a strength of 1 kg/sec until pain onset. Then, the pressure (in kg/cm2) was recorded and defined as the pain threshold. All participants filled out the Patient Health Questionnaire-9 (PHQ-9) to screen for depression and anxiety; (13) a final score above nine in the PHQ9 indicates the occurrence of a mood disorder. We used the Fibromyalgia Impact Questionnaire (FIQ) validated for Brazilian Portuguese (14) to measure the degree of impact of fibromyalgia on the patients’ quality of life. The FIQ comprises 10 questions with a maximum score of 100 points; the higher the score, the greater the impact of fibromyalgia on the individual’s quality of life. After that, the participants underwent anthropometric assessment for BMI calculation and DXA evaluation for analysis of body composition (total and compartmental body fat, lean mass, and bone mineral content) (15).

Statistical analysis The required number of participants was inferred by results of studies analyzing adipokines in painful Arch Endocrinol Metab. 2017;61/3

Leptin in obese fibromyalgia patients

conditions, such as rheumatoid arthritis (16) and headache (17). We concluded that 25 patients would be required in each group to detect a 20% difference in adipokines and cytokines levels. In the fibromyalgia and normal weight group, two samples were destroyed; therefore, only 23 patients were evaluated in this group. The statistical analysis was performed with the software JMP 7.0 (SAS Institute Inc., Cary, NC, USA). To compare continuous variables, we used Student’s t and Wilcoxon-Mann tests for parametric and nonparametric data, respectively. For correlations, we used Pearson’s test and Spearman’s correlation for parametric and nonparametric data, respectively. To test ratios, we used the chi-square test, and although we used median values to calculate nonparametric data, we present the data as mean and standard deviation (SD) values for the purpose of clarity (18).

RESULTS The study included 100 women categorized into one of the following groups: (a) fibromyalgia and overweight/

obesity I (n = 27); (b) fibromyalgia and normal weight (n = 23); (c) controls with overweight/obesity I (n = 25); and controls with normal weight (n = 25). The occurrence of overweight/obesity I was defined by a BMI ≥ 25 kg/m² and ≤ 35 kg/m², respectively, while normal weight was determined by a BMI between 18.5 and 24.9 kg/m². Table 1 compares the data of the participants with fibromyalgia and controls. The mean age was similar in both groups (47.92 ± 8.22 years versus 47.14 ± 9.93 years, respectively) and the mean BMI (26.40 ± 3.85 kg/m2 versus 25.56 ± 3.61 kg/m2, respectively) indicated slight overweight in both groups. The mean FIQ score in patients with fibromyalgia was 70.62 ± 17.73, reflecting a great impact of the disease on the participants’ quality of life. As for the group of patients with fibromyalgia as a whole, as well as for the groups of fibromyalgia with overweight/obesity and normal weight, we found no correlation between the BMI with measures related to the impact of fibromyalgia, such as the number of TPs, pain threshold, FIQ score, and PHQ-9 results (Table 2).

Table 1. General characteristics of the participants with fibromyalgia and controls Mean values (±SD)

Fibromyalgia n = 50

Controls n = 50

Age (years)

47.92 (±8.22)

47.14 (±9.93)

WC (cm)

92.80 (±9.48)

90.52 (±8.83)

BMI (kg/m²)

26.40 (±3.85)

25.56 (±3.61)

FM (kg)

25.30 (±7.63)

25.69 (±7.31)

TP (n)

16.1 (±1.97)

4.72 (±4.04)

< 0.0001

Pain threshold (kg/m²)

2.90 (±0.75)

5.46 (±1.93)

< 0.0001

FIQ (AU)

70.62 (±17.73)

10.64 (±12.32)

< 0.0001

PHQ-9 (AU)

16.28 (±5.69)

3.76 (±4.31)

< 0.0001

12,812.2 (±8,619.3)

18,316.3 (±10,190.2)

< 0.005

531.63 (±365.13)

684.05 (±300.27)

< 0.05

16,396 (±10,088.2)

15,316.5 (±7,188.5)

441.04 (±184.74)

487 (±267.04)

0.31 (±0.34)

0.31 (±0.44)

Leptin (pg/mL) Adjusted leptin (pg/mL/kg) Adiponectin (ng/mL) MCP-1 (pg/mL) CRP (mg/dL)

Table 2. Correlation of clinical fibromyalgia parameters with body mass index (r Spearman correlation) in patients with fibromyalgia Tender Points

Pain Threshold

PHQ-9

FIQ

MCP-1

CRP

BMI

0.09 (NS)

0.11 (NS)

-0.06 (NS)

0.04 (NS)

-0.03 (NS)

0.46 (p < 0.005)

Adjusted leptin

-0.10 (NS)

-0.04 (NS)

-0.07 (NS)

-0.10 (NS)

-0.03 (NS)

0.002 (NS)

FM: fibromyalgia; BMI: body mass index; FM: fat mass; TP: tender points; FIQ: Fibromyalgia Impact Questionnaire; PHQ-9: Patient Health Questionnaire-9; MCP-1: monocyte chemoattractant protein-1; CRP: C-reactive protein; NS: statistically not significant. Arch Endocrinol Metab. 2017;61/3

251

FM: fibromyalgia; WC: waist circumference; BMI: body mass index; FM: fat mass; TP: tender points; FIQ: Fibromyalgia Impact Questionnaire; AU: arbitrary units; PHQ-9: Patient Health Questionnaire-9; MCP-1: monocyte chemoattractant protein-1; CRP: C-reactive protein; NS: statistically not significant.

Leptin in obese fibromyalgia patients

Among patients with fibromyalgia, there was no difference between overweight/obesity subjects versus those with normal weight in regards to fibromyalgia parameters, including TPs count, pain threshold, FIQ scores, and PHQ-9 (Table 3).

Leptin Since fat tissue is the main source of serum leptin, before analyzing the results of the study, we correlated the levels of this adipokine with the participants’ BMI. We found that leptin levels correlated positively and significantly with the BMI in the entire cohort (r = 0.47, p < 0.0001), in the fibromyalgia group (r = 0.36, p = 0.009), and in the control group (r = 0.70, p < 0.0001). Compared with the control group, the fibromyalgia group had lower unadjusted leptin levels (12,812.2 ± 8,619.3 pg/mL versus 18,316.3 ± 10,190.2 pg/mL, respectively; p < 0.005). A comparison of the leptin levels adjusted by fat mass showed similar results, with lower levels present in the group with fibromyalgia when compared with controls (531.63 ± 365.13 pg/mL/kg versus 684.05 ± 300.27 pg/mL/kg, respectively; p < 0.05) (Table 1). This finding occurred mainly due to the fact that the unadjusted leptin levels were lower in patients with fibromyalgia and overweight/obesity than in controls with overweight/obesity (14,715.00 ± 8,884.65 pg/mL versus 23,903.00 ± 9,788.74 pg/mL; p < 0.005). The same finding was observed regarding the leptin values adjusted by fat mass (Table 4).

Although the leptin levels in patients with fibromyalgia and overweight/obesity were lower than those in participants in the control group with overweight/obesity, when we compared patients with fibromyalgia with overweight/obesity with those with normal weight we found a trend towards higher unadjusted leptin levels in the overweight/ obesity group (14,715.00 ± 8,884.65 pg/mL versus 10,578.00 ± 7,906.84 pg/mL; p = 0.05). This suggests that patients with fibromyalgia and overweight/obesity produce more leptin than those with fibromyalgia and normal weight. Still, the leptin levels were well below those produced by controls with overweight/obesity matched for BMI, waist circumference (WC), and fat mass. In none of the general groups (fibromyalgia patients and controls) there was a correlation between serum levels of leptin with clinical parameters related to fibromyalgia, such as FIQ scores, PHQ-9 results, pain threshold, or TPs count. Also, there was no correlation between levels of leptin and inflammation markers (MCP-1 and CRP) in patients with fibromyalgia (Table 2).

Adiponectin There was no difference in adiponectin levels between patients and controls. However, in the group with fibromyalgia and overweight/obesity, a slightly positive correlation was found between adiponectin levels and TPs count (r = 0.38; p = 0.04).

Table 3. Comparison between fibromyalgia groups (overweight/obesity versus normal weight) Mean values (±SD)

Fibromyalgia overweight/obesity

Fibromyalgia normal weight

n = 27

n = 23

Age (years)

50.25 (±6.76)

45.17 (±9)

< 0.05

WC (cm)

99.50 (±5.43)

84.93 (±6.37)

< 0.0001

BMI (kg/m²)

29.44 (±2.1)

22.82 (±1.76)

< 0.0001

FM (kg)

30.90 (±4.97)

18.73 (±4.14)

< 0.0001

TP (n)

16.4 (±1.9)

15.73 (±2)

Pain threshold (kg/m²)

2.85 (±0.55)

2.95 (±0.94)

FIQ (AU)

71.94 (±17.9)

69.09 (±17.8)

PHQ (AU)

Leptin (pg/mL) Adjusted leptin (pg/mL/kg) Adiponectin (ng/mL) MCP-1 (pg/mL) CRP (mg/dL)

16.07 (±5.74)

16.52 (±5.75)

14,715 (±8,884.65)

10,578 (±7,906.84)

p = 0.05

476.13 (±276.9)

596.78 (±445.1)

15,140 (±8,936.5)

17,870 (±11,317.6)

425.26 (±171.8)

459.56 (±202.17)

0.42 (±0.41)

0.19 (±0.18)

< 0.05

252

Arch Endocrinol Metab. 2017;61/3

Leptin in obese fibromyalgia patients

Table 4. Comparison between groups with overweight/obesity with and without fibromyalgia Mean values (±SD)

Fibromyalgia overweight/obesity

Control overweight/obesity

n = 27

n = 25

Age (years)

50.25 (±6.76)

49.92 (±8.64)

WC (cm)

99.50 (±5.43)

96.43 (±6.63)

BMI (kg/m²)

29.44 (±2.1)

28.49 (±2.44)

FM (kg)

30.90 (±4.97)

31.41 (±4.74)

TP (n)

16.4 (±1.9)

5.16 (±4.16)

< 0.0001

Pain threshold (kg/m²)

2.85 (±0.55)

6.12 (±1.95)

< 0.0001

FIQ (AU)

71.94 (±17.9)

9.9 (±12.46)

< 0.0001

PHQ (AU)

16.07 (±5.74)

3.08 (±3.51)

< 0.0001

14,715 (±8,884.65)

23,903 (±9,788.74)

< 0.005

476.13 (±276.9)

758.39 (±305.3)

< 0.005

15,140 (±8,936.5)

17,407 (±8,222.6)

425.26 (±171.8)

578.08 (±287.17)

= 0.08

0.42 (±0.41)

0.29 (±0.28)

Leptin (pg/mL) Adjusted leptin (pg/mL/kg) Adiponectin (ng/mL) MCP-1 (pg/mL) CRP (mg/dL)

MCP-1 and CRP No difference in MCP-1 and CRP levels was observed between the fibromyalgia and control groups. In a comparison between the group of fibromyalgia and overweight/obesity versus that of fibromyalgia with normal weight, the CRP levels were higher in the former (0.42 ± 0.41 mg/dL versus 0.19 ± 0.18 mg/dL, respectively; p < 0.05). In the group of patients with fibromyalgia and overweight/obesity, the CRP levels correlated directly with values of WC, BMI, and fat mass, demonstrating an inflammatory component associated with overweight and obesity.

DISCUSSION The aim of this study was to evaluate the levels of the adipokines leptin and adiponectin in patients with fibromyalgia with and without overweight/obesity. According to the literature, leptin levels may be increased in patients with fibromyalgia and overweight/ obesity, since leptin is considered a pro-nociceptive adipokine. The findings of this study demonstrated that patients with fibromyalgia and overweight/obesity had decreased leptin levels and that the levels did not correlate with clinical parameters associated with fibromyalgia. The association between overweight and obesity with musculoskeletal pain is evident in some diseases such as osteoarthritis, in which the joints that support Arch Endocrinol Metab. 2017;61/3

the excess weight are more symptomatic, and a loss in weight leads to symptom improvement (19). However, the relationship between pain and fat seems to be more complex. In migraine, for instance, the patient’s BMI has been associated with the occurrence of symptoms, which clearly cannot be explained by mechanical overload (20). In a cross-sectional study of 470 patients aged ≥ 70 years, WC was one of the strongest variables associated with the occurrence of chronic pain. Even though the occurrence of fibromyalgia was not evaluated in that study, there was a trend among participants with abdominal obesity to present a greater number of painful areas (21). The occurrence of overweight and obesity is common in fibromyalgia, where prevalence rates range from 50% to 70% (22). There may be several hypotheses to explain the contribution of obesity in worsening the symptoms of fibromyalgia, including worse physical conditioning (23), changes in sleep quality related to obesity (22), and a strong association between fibromyalgia and depression, which in turn can be associated with weight disorders and an increased risk of other musculoskeletal injuries. However, studies on the impact of excess weight in patients with fibromyalgia have shown conflicting results. In patients with systemic lupus erythematosus and fibromyalgia, increased BMI has been linked with an increased risk of the concurrent presence of fibromyalgia (24). Another study including 36 patients with fibromyalgia 253

Leptin in obese fibromyalgia patients

(with half of the participants being obese and 21% overweight) found no correlation between BMI and TPs count, FIQ score, or depression and anxiety levels (22). A cross-sectional study conducted in Spain including 177 female patients with fibromyalgia (25), of whom 70% were overweight or obese, has shown that patients with a BMI ≥ 25 kg/m² had higher levels of pain and fatigue when evaluated with the FIQ and the Short Form-36 Health Survey (SF-36); however, no correlation was found between BMI and pain threshold. The impact of the BMI on fibromyalgia may be greater in severely obese patients, as a study conducted at Mayo Clinic (9) has shown that individuals with severe obesity had worse FIQ scores and results in several SF-36 domains. In another study (10) including 179 patients with fibromyalgia, only the group with severe obesity showed worse scores in the physical function domain of the SF-36. This relationship could not be observed in the present study since it did not include severely obese patients. The adipose tissue is an actively secretory organ that sends and responds to signals that modulate appetite, energy expenditure, and insulin sensitivity, as well as the endocrine and reproductive systems, bone metabolism, and inflammatory and immune responses (5). Patients with overweight and obesity are commonly described as having a low but constant level of inflammation. For example, macrophages residing in the white adipose tissue produce 30% of the circulating IL-6 (5). In the adipose tissue, these cytokines may be regulated by adipokines. Leptin is considered a proinflammatory adipokine since it increases the number and survival of T lymphocytes and directs them to a Th1 profile (26). In addition, leptin may attract macrophages to the adipose tissue, especially by inducing the production of MCP-1 (7). In the case of adiponectin, an experimental evaluation indicates that it may have an anti-inflammatory mechanism of action (8). Evidence shows that adipokines also participate in pain perception (5). Among animals with ligation of the sciatic nerve, only those producing leptin manifest tactile allodynia (leptin-deficient ob/ob mice do not manifest allodynia) (27). In pain associated with knee osteoarthritis, leptin secreted in the infrapatellar (Hoffa’s) fat can contribute to nociception (7). In addition, adiponectin is associated with decreased baseline inflammatory activity, reduction in TNF-alpha, IL-6, and IL-1, in addition to an increase 254

in IL-10 (anti-inflammatory), which correlates with lower nociception levels (8). In the present study, fibromyalgia patients with overweight/obesity presented lower levels of leptin than controls, a finding that differs from others in the literature. This was mainly due to the fact that patients with fibromyalgia and overweight/obesity showed lower leptin levels compared with the participants in the control group with overweight/obesity. In addition, there was no correlation between leptin levels and clinical fibromyalgia parameters. Leptin levels in musculoskeletal conditions and fibromyalgia have shown contradicting results. In patients with knee osteoarthritis assessed before and after bariatric surgery (28), the surgery led to a significant decrease in BMI (20%) with improvement in pain standards and knee function, and a reduction in levels of markers of articular destruction, serum IL-6, CPR, and leptin. Before the procedure, there was a correlation between pain and knee function questionnaires with the levels of inflammatory markers and articulation destruction, but not with leptin. Following surgery and clinical improvement, there was no longer a correlation of these questionnaires with any serum marker, which led the authors to conclude that the improvement in low-degree inflammation associated with obesity and the reduction in leptin had little or no importance in the clinical improvement associated with the weight loss. A small cross-sectional study including 16 patients with fibromyalgia and 21 controls matched for BMI found no difference in leptin levels measured by ELISA. Additionally, there was also no association between leptin levels with clinical parameters of fibromyalgia, such as FIQ scores and the mental and physical components of the SF-36 questionnaire (29). The study that mostly resembled ours examined 50 women with fibromyalgia and 50 controls matched by BMI. However, such study failed to adjust the leptin levels by fat mass or stratify them by BMI (30). These authors demonstrated that leptin levels were significantly lower in patients with fibromyalgia than in controls. In addition, leptin levels correlated negatively with several clinical fibromyalgia parameters, such as pain intensity, fatigue and anxiety, and with the quality of life scores, depression, sleep, and FIQ. The authors suggested that the lack of leptin or resistance to its action could favor a depressive status, and this factor could lead to worsening of fibromyalgia. The leptin levels Arch Endocrinol Metab. 2017;61/3

Leptin in obese fibromyalgia patients

Arch Endocrinol Metab. 2017;61/3

of not including these patients aimed to avoid the presence of mechanical articular problems, especially osteoarthritis of the spine and knees, which could impair the evaluation. The correlation of adipokines with visceral fat mass (measured by computed tomography, for example) could represent better the source of these substances; the measurement of adipokines in the cerebrospinal fluid could have been used to check their action in the central nervous system. In contrast, a strong point of our study was the inclusion of a correlation between leptin and adiponectin levels with the patients’ total fat mass. The role of overweight/obesity and adipokines in fibromyalgia is more complex than a linear relationship between obesity, leptin production, and more severe pain. The leptin levels were lower in patients with fibromyalgia and overweight/obesity when compared with control participants with overweight/obesity, which may indicate that the occurrence of fibromyalgia in overweight patients leads to decreased production of leptin. We did not observe a correlation between leptin levels and clinical parameters of fibromyalgia among patients with this condition. Nor was there a correlation between this adipokine with inflammation markers, although patients with fibromyalgia and overweight have a greater inflammatory status indicated by the presence of higher CRP levels. Regarding adiponectin, the group with fibromyalgia and overweight presented a negative correlation between this adipokine and BMI, a paradoxical fact that has also been verified in other studies. There was no effect of overweight and obesity on the impact of the disease as measured by the FIQ in patients with fibromyalgia, just as in those observed with depression. Funding: partial funding was provided by the Associação SEMPR Amigos and Fundo de Apoio à Pesquisa – Sociedade Brasileira de Reumatologia (FAP-SBR). Acknowledgements: the authors would like to thank the Internal Medicine Department-UFPR and the Nutrition Department for their support and Robert M. Bennett MD for his input in the initial concept. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Clauw DJ. Fibromyalgia: update on mechanisms management. J Clin Rheumatol. 2007;13(2):102-9.

and

255

are decreased in depression, and the administration of leptin may improve depressive symptoms (31). In the present study, mood changes, as determined by the PHQ-9, showed no difference between the two fibromyalgia groups. The lower levels of leptin found in patients with fibromyalgia and overweight/obesity in our study could be related to changes in the hypothalamicpituitary-adrenal (HPA) axis. Most studies have shown that the baseline functioning of the HPA axis is normal in fibromyalgia (32), but under chronic stress, there is an increase in cortisol levels with a “flattened” daytime pattern (33). Since leptin has a suppressive action in the HPA axis (34), its lower levels in patients with fibromyalgia could collaborate with a hyperactivity of the axis, but we cannot rule out that a variability in leptin levels in humans could hinder the evaluation of this adipokine. A study on metabolic syndrome and leptin has shown a wide variation in leptin levels, even among obese patients (almost seven times between the lowest and highest values in patients with BMI > 40 kg/m2) (12). We found no correlation between adiponectin levels with clinical or functional fibromyalgia parameters, except for a slight positive correlation with the TPs count. The role of cytokines in fibromyalgia has been the target of several studies (3). Patients with fibromyalgia have high serum levels of IL-1RA and IL-6. IL-8 inflammatory chemokines (cytokines related to chemotaxis) such as MCP-1 have also been described as being increased in patients with fibromyalgia. A study of 92 patients with fibromyalgia has shown that these patients have higher MCP-1 levels than controls and that their sera induce increased migration of macrophages. Interestingly, the levels of these cytokines did not differ when obese and non-obese patients with fibromyalgia were compared, a finding similar to the present study (35). We observed that patients with fibromyalgia and overweight/obesity had increased CRP levels and that this group was the only one with a correlation, albeit weak, between CRP and FIQ scores. This finding could suggest that this group of patients has two sources of inflammation, the adipose tissue and the fibromyalgia itself (22). The present study has limitations. For example, we did not include patients with severe obesity, which could have led to more pronounced results. The choice

Leptin in obese fibromyalgia patients

2. Wolfe F, Ross K, Anderson J, Russell IJ, Hebert L. The prevalence and characteristics of fibromyalgia in the general population. Arthritis Rheum. 1995;38(1):19-28. 3. Paiva ES, da Costa ED, Scheinberg M. Fibromyalgia: an update and immunological aspects. Curr Pain Headache Rep. 2008;12(5): 321-6. 4. Berry PA, Jones SW, Cicuttini FM, Wluka AE, Maciewicz RA. Temporal relationship between serum adipokines, biomarkers of bone and cartilage turnover, and cartilage volume loss in a population with clinical knee osteoarthritis. Arthritis Rheum. 2011;63(3):700-7.

19. Felson DT. Weight and osteoarthritis. J Rheumatol Suppl. 1995;43:7-9. 20. Bigal ME, Lipton RB, Holland PR, Goadsby PJ. Obesity, migraine, and chronic migraine: possible mechanisms of interaction. Neurology. 2007;68(21):1851-61. 21. Ray L, Lipton RB, Zimmerman ME, Katz MJ, Derby CA. Mechanisms of association between obesity and chronic pain in the elderly. Pain. 2011;152(1):53-9. 22. Okifuji A, Bradshaw DH, Olson C. Evaluating obesity in fibromyalgia: neuroendocrine biomarkers, symptoms, and functions. Clin Rheumatol. 2009;28(4):475-8.

5. Fantuzzi G. Adipose tissue, adipokines, and inflammation. J Allergy Clin Immunol. 2005;115(5):911-9; quiz 20.

23. Ursini F, Naty S, Grembiale RD. Fibromyalgia and obesity: the hidden link. Rheumatol Int. 2011;31(11):1403-8.

6. Lim G, Wang S, Zhang Y, Tian Y, Mao J. Spinal leptin contributes to the pathogenesis of neuropathic pain in rodents. J Clin Invest. 2009;119(2):295-304.

24. Chaiamnuay S, Bertoli AM, Fernandez M, Apte M, Vila LM, Reveille JD, et al. The impact of increased body mass index on systemic lupus erythematosus: data from LUMINA, a multiethnic cohort (LUMINA XLVI) [corrected]. J Clin Rheumatol. 2007;13(3):128-33.

7. Clockaerts S, Bastiaansen-Jenniskens YM, Runhaar J, Van Osch GJ, Van Offel JF, Verhaar JA, et al. The infrapatellar fat pad should be considered as an active osteoarthritic joint tissue: a narrative review. Osteoarthritis Cartilage. 2010;18(7):876-82. 8.

Neumann E, Frommer K, Vasile M, Muller-Ladner U. Adipocytokines as driving forces in rheumatoid arthritis and related inflammatory diseases? Arthritis Rheum. 2011;63(5):1159-69. Kim CH, Luedtke CA, Vincent A, Thompson JM, Oh TH. Association of body mass index with symptom severity and quality of life in patients with fibromyalgia. Arthritis Care Res (Hoboken). 2012;64(2):222-8.

25. Aparicio VA, Ortega FB, Carbonell-Baeza A, Gatto-Cardia C, Sjostrom M, Ruiz JR, et al. Fibromyalgia’s key symptoms in normal-weight, overweight, and obese female patients. Pain Manag Nurs. 2013;14(4):268-76. 26. Farooqi IS, Matarese G, Lord GM, Keogh JM, Lawrence E, Agwu C, et al. Beneficial effects of leptin on obesity, T cell hyporesponsiveness, and neuroendocrine/metabolic dysfunction of human congenital leptin deficiency. J Clin Invest. 2002;110(8):1093-103.

10. Timmerman GM, Calfa NA, Stuifbergen AK. Correlates of body mass index in women with fibromyalgia. Orthop Nurs. 2013;32(2):113-9.

27. Maeda T, Kiguchi N, Kobayashi Y, Ikuta T, Ozaki M, Kishioka S. Leptin derived from adipocytes in injured peripheral nerves facilitates development of neuropathic pain via macrophage stimulation. Proc Natl Acad Sci U S A. 2009;106(31):13076-81.

11. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Goldenberg DL, et al. The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum. 1990;33(2):160-72.

28. Richette P, Poitou C, Garnero P, Vicaut E, Bouillot JL, Lacorte JM, et al. Benefits of massive weight loss on symptoms, systemic inflammation and cartilage turnover in obese patients with knee osteoarthritis. Ann Rheum Dis. 2011;70(1):139-44.

12. Paz-Filho GJ, Volaco A, Suplicy HL, Radominski RB, Boguszewski CL. Decrease in leptin production by the adipose tissue in obesity associated with severe metabolic syndrome. Arq Bras Endocrinol Metabol. 2009;53(9):1088-95.

29. Ablin JN, Aronov N, Shimon I, Kanety H, Pariente C, Aloush V, et al. Evaluation of leptin levels among fibromyalgia patients before and after three months of treatment, in comparison with healthy controls. Pain Res Manag. 2012;17(2):89-92.

13. de Lima Osorio F, Vilela Mendes A, Crippa JA, Loureiro SR. Study of the discriminative validity of the PHQ-9 and PHQ-2 in a sample of Brazilian women in the context of primary health care. Perspect Psychiatr Care. 2009;45(3):216-27.

30. Olama SM, Elsaid TO, El-Arman M. Serum leptin in Egyptian patients with fibromyalgia syndrome: relation to disease severity. Int J Rheum Dis. 2013;16(5):583-9.

14. Marques AP, Santos AMB, Assumpção A, Matsutani LA, Lage LV, Pereira CAB. Validation of the Brazilian Version of the Fibromyalgia Impact Questionnaire (FIQ). Rev Bras Reumatol. 2006;46(1):24-31. 15. Lobo MM, Paiva Edos S, Andretta A, Schieferdecker ME. [Body composition by dual-energy x-ray absorptiometry in women with fibromyalgia]. Rev Bras Reumatol. 2014;54(4):273-8. 16. Popa C, Netea MG, de Graaf J, van den Hoogen FH, Radstake TR, Toenhake-Dijkstra H, et al. Circulating leptin and adiponectin concentrations during tumor necrosis factor blockade in patients with active rheumatoid arthritis. J Rheumatol. 2009;36(4):724-30. 17. Peterlin BL, Alexander G, Tabby D, Reichenberger E. Oligomerization state-dependent elevations of adiponectin in chronic daily headache. Neurology. 2008;70(20):1905-11.

32. Adler GK, Geenen R. Hypothalamic-pituitary-adrenal and autonomic nervous system functioning in fibromyalgia. Rheum Dis Clin North Am. 2005;31(1):187-202, xi. 33. Crofford LJ. The hypothalamic-pituitary-adrenal axis in the pathogenesis of rheumatic diseases. Endocrinol Metab Clin North Am. 2002;31(1):1-13. 34. Roubos EW, Dahmen M, Kozicz T, Xu L. Leptin and the hypothalamo-pituitary-adrenal stress axis. Gen Comp Endocrinol. 2012;177(1):28-36. 35. Zhang Z, Cherryholmes G, Mao A, Marek C, Longmate J, Kalos M, et al. High plasma levels of MCP-1 and eotaxin provide evidence for an immunological basis of fibromyalgia. Exp Biol Med (Maywood). 2008;233(9):1171-80.

18. Lydersen S. Statistical review: frequently given comments. Ann Rheum Dis. 2015;74(2):323-5.

31. Wedrychowicz A, Zajac A, Pilecki M, Koscielniak B, Tomasik PJ. Peptides from adipose tissue in mental disorders. World J Psychiatry. 2014;4(4):103-11.

256

Arch Endocrinol Metab. 2017;61/3

original article

Applicability of predictive equations for resting energy expenditure in obese patients with obstructive sleep apnea Mariana Pantaleão del Re1, Camila Maria de Melo1, Marcus Vinicius dos Santos2, Sergio Tufik1, Marco Túlio de Mello3

ABSTRACT Objective: To investigate the applicability of predictive equations for resting energy expenditure (REE) in obese individuals with obstructive sleep apnea (OSA) and the effects of OSA severity on REE. Materials and methods: Twenty-nine obese men, 41.5 ± 7 years old, with moderate and severe OSA were recruited. All subjects were submitted to a clinical polysomnography, body composition, and indirect calorimetry measurements. REE was also predicted by three different equations: Harris and Benedict (1919), Cunningham (1990), and DRI (2002). Results: No effects of OSA severity on REE were found. The measured REE (2416.0 ± 447.1 kcal/day) and the REE predicted by equations were different from each other (F = 2713.88; p < 0.05): Harris and Benedict (2128.0 ± 245.8 kcal/day), Cunningham (1789.1 ± 167.8 kcal/day) and DRI (2011.1 ± 181.4 kcal/ day). Pearson correlations showed a moderate positive correlation between the REE measured and predicted by all equations. Conclusion: Our findings suggest that predictive equations for REE underestimate the energy expenditure in obese patients with sleep apnea. Also, no effects of OSA severity on REE were found. Arch Endocrinol Metab. 2017;61(3):257-62. Keywords Obesity; obstructive sleep apnea; energy expenditure; resting energy expenditure; predictive equations

1 Universidade Federal de São Paulo (Unifesp), Psicobiologia, São Paulo, SP, Brasil 2 Universidade Federal de São Paulo (Unifesp), Biociências, São Paulo, SP, Brasil 3 Faculdade de Educação Física, Fisioterapia e Terapia Ocupacional, Universidade Federal de Minas Gerais (UFMG), Belo Horizonte, MG, Brasil

Correspondence to: Camila Maria de Melo Rua Botucatu, 862, 1º andar 04023-900 – São Paulo, SP, Brasil camariamelo@gmail.com Received on Mar/1//2016 Accepted on Sep/12/2016

INTRODUCTION

bstructive sleep apnea (OSA) is characterized by repeated episodes of total or partial upper airway obstruction during sleep (1). A bidirectional relationship between OSA and obesity/weight gain is already described (2). While OSA can develop as a consequence of obesity, the disease per se might result in energy expenditure changes and excessive daytime sleepiness, leading to weight gain (2,3). The total energy expenditure (TEE) is determined by different components: resting energy expenditure (REE), thermic effect of food (TEF), and physical activity-related energy expenditure (PAEE) (4). OSA might modulate energy expenditure in all components of TEE. Studies suggest that REE is increased in OSA patients. Ucok and cols. (5) showed higher REE in OSA patients compared with a group of snorers, even after corrections for body mass and fat-free mass. Arch Endocrinol Metab. 2017;61/3

Oxyhemoglobin desaturation, successive arousals, and continuous sympathetic activation might be responsible for enhanced metabolism in OSA patients (5,6). OSA also leads to elevated diurnal somnolence and fatigue, which contributes to a sedentary lifestyle and decreased PAEE. From the clinical practice point of view, it’s important that clinicians accurately estimate energy expenditure in their patients so that higher success rates of nutritional management could be achieved. Many methods for energy expenditure measurements are available nowadays, such as indirect calorimetry, accelerometers, and doubly labeled water (4,7). Indirect calorimetry is a method commonly used by researchers and physicians to measure REE, although it is not available in many hospitals and outpatient health care clinics. In the absence of more accurate methods, many predictive equations were developed during past 257

DOI: 10.1590/2359-3997000000228

Resting energy expenditure in obstrictive sleep apnea patients

decades for distinct populations and physiological conditions (7-9). Considering the disturbance in oxygen saturation and energy expenditure found in OSA patients, we tested the hypothesis that predictive equations currently available might not be accurate for this particular population. The aim of the present study was to compare the REE measured by indirect calorimetry with prediction equations in obese OSA patients and, as a secondary aim, to compare the effects of OSA severity on REE.

MATERIALS AND METHODS The present study was approved by Ethics Committee of Universidade Federal de São Paulo (#142861) and registered in Clinical Trials (#NCT01985035). Twenty-nine adult obese men (body mass index between [BMI] between 30 and 40 kg/m²) between 30 and 50 years old were included in the study. All subjects were submitted to clinical polysomnography for diagnosis and classification of OSA severity. Apneas and hypopneas were scored using standard American Academy of Sleep Medicine guidelines (1), and apneahypopnea index (AHI) was used for OSA severity classification. REE was measured by indirect calorimetry (Quark CEPET, COSMED) (10). The measurement was made after an overnight fast. The measurement lasted for 30 minutes with the patients in the supine position and in a temperature- and luminosity-controlled room. Body composition was evaluated by plethysmography (BOD POD) (11). REE and body composition measurements, and the clinical polysomnography, were all performed with the maximum of one week between procedures. The measurement of REE by indirect calorimetry was compared with three prediction equations: Harris and Benedict (8), Cunningham (9), and the equations recommended by the Food and Nutrition Board (DRI) (12), as demonstrated in Table 1.

Table 1. Resting energy expenditure predictive equations compared with indirect calorimetry Author (Year) Harris e Benedict (8) Cunningham (9) DRI (12)

258

Equation REE = 66,5 + (13,8 x Weight (kg)) + (5 x Height (m)) + (6,8 x Age) REE = 370 + (21,6 x Fat Free Mass (kg)) REE = 293 - 3,8 x Age + 456,4 x Height (m) + 10,12 x Weight (kg)

Statistical analysis was performed using SPSS software (IBM Corp., Armonk, NY-version 19.0). Shapiro-Wilk test was used for normality. To observe the effect of OSA severity on REE subjects, the subjects were distributed into three groups based on OSA severity: AHI > 15 ≤ 30 events/h (Group 1), AHI > 30 ≤ 50 events/h (Group 2) and AHI > 50 events/h (Group 3). For comparison between measured REE and REE predicted by equations, and for group comparisons, ANOVA tests were used to make repeated measures. Pearson correlations were also performed between predicted and measured REE values. Bland– Altman agreements plots were made using the MedCalc software (Medcalc Software, Ostend, Belgium) to determine the agreement between measured and predicted REE, and p < 0.05 was adopted for statistical significance.

RESULTS Twenty-nine adult men, 41.53 ± 7.26 years, obese, and diagnosed with mild or severe OSA were included in this study. Subjects’ characteristics are described in Table 2. Besides a difference in body mass between Group 2 and Group 3, no other significant differences were found in anthropometric and body composition assessments. As expected, some differences were found in sleep parameters: Group 3 showed higher AHI, respiratory disturbance index (RDI), NREM sleep stage 1 (%), spent more time in SaO2 lower than 90%, less NREM sleep stage 3, and less REM sleep than Group 1 and 2. The measured REE by indirect calorimetry showed higher values of energy expenditure (2416.0 ± 447.1 kcal/day) than that predicted by all equations: Harris and Benedict (2128.0 ± 245.8 kcal/day), Cunningham (1789.1 ± 167.8 kcal/day) and DRI (2011.1 ± 181.4 kcal/day), as demonstrated in Figure 1. Pearson correlations were made for agreement between different methods and all equations showed moderate correlation with REE measured by indirect calorimetry, as showed in Figure 2. Between groups divided by AHI, no differences were found in total REE or REE corrected by body mass or fat-free mass (Table 3). Figure 3 represents the Bland–Altman agreement plots. There were agreements in the measured REE and the values predicted by equations. However, in the resulting the agreements, substantial differences were observed between the methods. Arch Endocrinol Metab. 2017;61/3

Resting energy expenditure in obstrictive sleep apnea patients

Table 2. Anthropometrics and sleep characteristics of obese patients with sleep apnea All participants (Mean ± SD)

Group 1 N=8 (Mean ± SD)

Group 2 N=8 (Mean ± SD)

Group 3 N = 13 (Mean ± SD)

Body mass (kg)

106.4 ± 15.2

106.5 ± 11.8

95.5 ± 7.9

113.2 ± 17.1b

Height (m)

1.75 ± 0.07

1.76 ± 0.07

1.71 ± 0.06

1.76 ± 0.08

Variables

Body mass index (kg/m²)

34.7 ± 4.2

34.1 ± 3.8

32.6 ± 2.8

36.5 ± 4.6

Body fat (%)

37.7 ± 5.2

35.2 ± 3.3

37.5 ± 6.0

39.3 ± 5.5

Fat-free mass (%)

62.1 ± 5.5

64.8 ± 3.3

62.5 ± 6.0

60.1 ± 5.8

Fat-free mass (kg)

65.7 ± 7.7

68.9 ± 5.6

59.7 ± 7.1

67.4 ± 7.7

Sleep latency (min)

7.0 ± 7.5

8.1 ± 6.2

5.4 ± 6.1

8.0 ± 9.1

280.4 ± 158.5

386.6 ± 33.7

343.5 ± 129.9

342.7 ± 53.8

Sleep eficiency (%)

Total sleep time (min)

86.4 ± 10.6

89.3 ± 7.1

92.9 ± 3.2

81.2 ± 12.7

NREM Sleep stage 1 (%)

18.9 ± 13.1

11.4 ± 5.9

10.4 ± 4.6

28.6 ± 13.4a

NREM Sleep stage 2 (%)

45.6 ± 8.4

46.2 ± 10.2

44.4 ± 6.3

46.5 ± 8.8

NREM Sleep stage 3 (%)

16.9 ± 8.5

22.5 ± 67.0

19.1 ± 7.2

9.1 ± 8.0a

REM Sleep (%)

19.4 ± 6.5

19.9 ± 7.1

26.1 ± 5.2a

15.8 ± 3.0b

101.4 ± 108.2

141.8 ± 56.7

162.7 ± 58.9

345.4 ± 115.7

AHI (events/h)

Total Arousals (n)

51.7 ± 24.7

23.6 ± 2.8

40.0 ± 5.9a

76.8 ± 10.4a,b

RDI (Events/h)

54.0 ± 22.9

28.9 ± 3.6

42.4 ± 6.4

77.2 ± 10.1a,b

SaO2 (%)

93.4 ± 1.5

93.5 ± 1.3

93.5 ± 0.6

93.1 ± 1.6b

Min SaO2 (%)

74.4 ± 8.9

80.2 ± 3.9

77.7 ± 6.0

68.7 ± 9.5a,b

TRT SaO2 < 90% (%)

18.7 ± 19.1

9.0 ± 9.0

7.4 ± 4.9

32.6 ± 21.3a,b

Resting energy expenditure (kcal/day)

4000

3000

2000

1000

DRI

Cunninghan

Figure 1. Resting energy expenditure measured by indirect calorimetry and estimated by different equations. São Paulo, 2015. IC: indirect calorimetry; HB: Harris & Benedict equation; DRI: DRI’s equation. Measured repeated ANOVA, F = 49.992 p ≤ 0,05. Values presented in Mean ± Standard Deviation; * p < 0.05.

DISCUSSION The principal finding of the present study is that the predictive equations underestimate the REE measured Arch Endocrinol Metab. 2017;61/3

by indirect calorimetry in obese individuals with OSA. The relationship between OSA and obesity has been extensively studied in recent years (2,3), and the effects of OSA on energy expenditure have been a target of some studies (5,6,13). In contrast with previous studies that suggested that REE is increased in OSA patients, our study doesn’t confirm any effect of OSA severity on REE. In a pioneer study, Ryan and cols. (6) compared 14 eutrophic subjects with moderate or severe OSA with 14 healthy control subjects and found increased REE in OSA patients, but failed to find differences when REE values were corrected by fat-free mass values. Kezirian and cols. (13) observed a positive correlation between REE and AHI in a sample of 212 individuals with OSA. It’s important to note that in these studies, the REE values were not corrected for body mass and body composition. Recently, Fekete and cols. (14), in a study with 92 individuals with OSA and 19 control subjects, also found a higher REE in patients with OSA. In two other studies, the authors failed to find differences in REE between patients with OSA and healthy subjects (15,16). 259

Group 1: AHI > 15 ≤ 30 ev/h; Group 2: AHI > 30 ≤ 50 ev/h; Group 3: AHI > 50 ev/h; a Different from Group 1 (p < 0.05); b Different from Group 2 (p < 0.05); Mild OSA: AHI ≥ 15 events/h; Severe OSA: AHI ≥ 30 events/h; NREM: Non-rapid eye movement sleep; REM: Rapid eye movement sleep; SaO2: Average Oxyhemoglobin saturation; Min SaO2: Minimal Oxyhemoglobin saturation; TRT SaO2 < 90%: Total Recording Time with SaO2 < 90%.

3000

2600

r = 0.563; p < 0.05 DRI equation (kcal/day)

Haris & Benedict equation (kcal/day)

Resting energy expenditure in obstrictive sleep apnea patients

2500

2000

1500 1000

2000 3000 REE by indirect calorimetry (kcal/day)

Cunninghan equation (kcal/day)

2400 2200 2000 1800 1600 1000

4000

2400

r = 0.568; p < 0.05

2000 3000 REE by indirect calorimetry (kcal/day)

4000

r = 0.504; p < 0.05

2200 2000 1800 1600 1400 1200 1000

2000

3000

4000

REE by indirect calorimetry (kcal/day)

A 1200 1000 800 600 400 200 0 -200 -400

B + 1.96 SD 1012.3

Mean 288.0

-1.96 SD -436.4

-600 1600 1800 2000 2200 2400 2600 2800 3000 3200

Indirect calorimetry – Cunningham equation

Mean of indirect calorimetry and HB equation

Indirect calorimetry – DRI equation

Indirect calorimetry – HB equation

Figure 2. Pearson correlation between measured resting energy expenditure and prediction equations. São Paulo, 2015.

1400 + 1.96 SD 1200 1140.1 1000 800 600 Mean 400 404.9 200 0 -1.96 SD -200 -400 -330.3 -600 1600 1800 2000 2200 2400 2600 2800 3000 Mean of indirect calorimetry and DRI equation

C 2000 1500

+ 1.96 SD 1392.0

1000 500 0

Mean 627.0 -1.96 SD -138.0

-500 1400 1600 1800 2000 2200 2400 2600 2800 3000 Mean of indirect calorimetry and Cunningham equation

Figure 3. Bland-Altman plots of resting energy expenditure measured by indirect calorimetry and predicted by different equations. São Paulo, 2015. HB: Harris & Benedict equation; DRI: DRI’s equation. 260

Arch Endocrinol Metab. 2017;61/3

Resting energy expenditure in obstrictive sleep apnea patients

Table 3. Resting energy expenditure in all participant and subgroups of AHI. São Paulo, 2015 Characteristics

All participants (Mean ± SD)

Group 1 N = 8 (Mean ± SD)

Group 2 N = 8 (Mean ± SD)

Group 3 N = 13 (Mean ± SD)

REE (kcal/day)

2416.0 ± 447.1

2539.5 ± 400.6

2072.7 ± 424.3

2551.3 ± 399.0

REE/kg (kcal/day)

22.79.8 ± 3.42

23.8 ± 2.4

21.8 ± 4.7

22.7 ± 3.1

36.9 ± 6.0

36.9 ± 5.0

35.0 ± 7.5

38.0 ± 5.7

REE/kg of FFM (kcal/day)

Group 1: AHI > 15 ≤ 30 ev/h; Group 2: AHI > 30 ≤ 50 ev/h; Group 3: AHI > 50 ev/h; REE: resting energy expenditure; FFM: Fat-free mass.

Arch Endocrinol Metab. 2017;61/3

Based on our results from Person correlations and Bland–Altman plots, it’s possible to say that besides the equations underestimating the measured REE, all three equations had at least moderate agreements with the measured values. The Bland–Altman analysis showed some important dispersion of the data. Even though it is the oldest equation, the Harris and Benedict equation showed the biggest agreement with the measured REE. This equation has been considered a reasonable equation for obese individuals (20,21). It’s important to say that it was not developed using data from obese individuals; therefore, some concerns about accuracy in this population persist (20). Although the equations underestimate the REE of OSA patients, it should be taken into consideration that the elevation in energy expenditure due to OSA is small compared with the rise in energy intake that a sleeprestricted person might experience, which culminates in an increased predisposition to gain weight. Based on the data from the present study and the aforementioned literature, we believe that the energy expenditure equations commonly used in clinical practice may have significant biases, considering that OSA can disrupt energy expenditure behavior during sleep, and this change might persist during the day. Clinical physicians and nutritionist should recognize the presence of OSA in obese patients and take into consideration this fact when weight loss is desired. In conclusion, prediction equations for REE can underestimate the REE measured by indirect calorimetry in obese patients with OSA, despite the agreement between methods and that the severity of OSA had no effect on REE. Acknowledgements: the authors thank the financial support of Fundação de Amparo à Pesquisa do Estado de São Paulo (Fapesp – 2012/09700-2), Associação Fundo de Incentivo à Pesquisa (AFIP) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). Disclosure: no potential conflict of interest relevant to this article was reported. 261

The abovementioned studies highlight that OSA has direct effects on energy expenditure. The intermittent hypoxia and sleep fragmentation might lead to a continuous sympathetic activation, increasing energy expenditure during the night (17). Furthermore, this nocturnal sympathetic hyperactivity could be extended during the day, increasing the REE (18,19). While REE seems to be enhanced in individuals with OSA, TEE could be compromised. Excessive diurnal somnolence and lower physical activity levels observed in these population might contribute to lowering TEE. Major and cols. (17) concluded that the more time spent in low saturation rates (below 90%) during the night, the lower the 24 h energy expenditure, which could be a possible explanation for the relationship between OSA and weight gain. In patients with OSA, the nocturnal desaturation and the arousals during the night culminate in higher sympathetic activity and higher energy expenditure. However, in the long term, this mechanism could lead to an adaptive decrease in adrenoceptors activation (17), resulting in decreased energy expenditure in OSA patients in the long term. Besides the controversy, the fact is that one way or another, energy expenditure could be affected by OSA, which is an important issue for obesity management in patients with OSA. Due to the limited availability of the indirect calorimetry method in clinical practice, it’s important that physicians recognize these metabolic disruptions and choose the best method to predict energy expenditure of obese patients with OSA (19). The prediction equations were described as able to provide an easy way for energy expenditure estimation. In our study, three equations were used to compare the measurement of REE: the Harris and Benedict equation, the Cunningham equation, and the DRI equation. The first two were developed using indirect calorimetry data from healthy individuals while the last one used doubly labeled water data from obese individuals (8,9,12).

Resting energy expenditure in obstrictive sleep apnea patients

REFERENCES 1. Berry RB, Budhiraja R, Gottlieb DJ, Gozal D, Iber C, Kapur VK, et al.; American Academy of Sleep Medicine. Rules for scoring respiratory events in sleep: update of the 2007 AASM Manual for the Scoring of Sleep and Associated Events. Deliberations of the Sleep Apnea Definitions Task Force of the American Academy of Sleep Medicine. J Clin Sleep Med. 2012;8(5):597-619. 2. Ong CW, O’Driscoll DM, Truby H, Naughton MT, Hamilton GS. The reciprocal interaction between obesity and obstructive sleep apnoea. Sleep Med Rev. 2013;17(2):123-31. 3. Bonsignore MR, Borel AL, Machan E, Grunstein R. Sleep apnoea and metabolic dysfunction. Eur Respir Rev. 2013;22(129):353-64. 4.

Melo CM, Tirapegui J, Ribeiro SM. [Human energetic expenditure: concepts, assessment methods and relationship to obesity]. Arq Bras Endocrinol Metabol. 2008;52(3):452-64.

5. Ucok K, Aycicek A, Sezer M, Fidan F, Akgun L, Akkaya M, et al. Resting metabolic rate and anthropometric measurements in male sleep apnea patients. Intern Med. 2011;50(8):833-8. 6. Ryan CF, Love LL, Buckley PA. Energy expenditure in obstructive sleep apnea. Sleep. 1995;18(3):180-7. 7. Trumbo P, Schlicker S, Yates AA, Poos M; Food and Nutrition Board of the Institute of Medicine, The National Academies. Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein and amino acids. J Am Diet Assoc. 2002;102(11):1621-30. 8. Harris JA, Benedict FG. A Biometric Study of Human Basal Metabolism. Proc Natl Acad Sci U S A. 1918;4(12):370-3.

12. Institute of Medicine FaNB. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids (Macronutrients). National Academies Press, Washington, DC, 2005. 13. Kezirian EJ, Kirisoglu CE, Riley RW, Chang E, Guilleminault C, Powell NB. Resting energy expenditure in adults with sleep disordered breathing. Arch Otolaryngol Head Neck Surg. 2008;134(12):1270-5. 14. Fekete K, Boutou AK, Pitsiou G, Chavouzis N, Pataka A, Athanasiou I, et al. Resting energy expenditure in OSAS: the impact of a single CPAP application. Sleep Breath. 2016;20(1):121-8. 15. Lin CC, Chang KC, Lee KS. Effects of treatment by laser-assisted uvuloplasty on sleep energy expenditure in obstructive sleep apnea patients. Metabolism. 2002;51(5):622-7. 16. Beitler JR, Awad KM, Bakker JP, Edwards BA, DeYoung P, Djonlagic I, et al. Obstructive sleep apnea is associated with impaired exercise capacity: a cross-sectional study. J Clin Sleep Med. 2014;10(11):1199-204. 17. Major GC, Series F, Tremblay A. Does the energy expenditure status in obstructive sleep apnea favour a positive energy balance? Clin Invest Med. 2007;30(6):E262-8. 18. Dempsey JA, Veasey SC, Morgan BJ, O’Donnell CP. Pathophysiology of sleep apnea. Physiol Rev. 2010;90(1):47-112. 19. Quiroz-Olguin G, Serralde-Zuniga AE, Saldana-Morales MV, Gulias-Herrero A, Guevara-Cruz M. Validating an energy expenditure prediction equation in overweight and obese Mexican patients. Nutr Hosp. 2014;30(4):749-55.

10. Ashcraft CM, Frankenfield DC. Validity Test of a New OpenCircuit Indirect Calorimeter. JPEN JPEN J Parenter Enteral Nutr. 2015;39(6):738-42.

20. Parra-Carriedo A, Cherem-Cherem L, Galindo-De Noriega D, Diaz-Gutierrez MC, Perez-Lizaur AB, Hernandez-Guerrero C. [Comparison of resting energy expenditure determined by indirect calorimetry and estimated by predictive formulas in women with obesity degrees I to III]. Nutr Hosp. 2013;28(2): 357-64.

11. Wells JC, Fuller NJ. Precision of measurement and body size in whole-body air-displacement plethysmography. Int J Obes Relat Metab Disord. 2001;25(8):1161-7.

21. Livingston EH, Kohlstadt I. Simplified resting metabolic ratepredicting formulas for normal-sized and obese individuals. Obes Res. 2005;13(7):1255-62.

9. Cunningham JJ. A reanalysis of the factors influencing basal metabolic rate in normal adults. Am J Clin Nutr. 1980;33(11):2372-4.

262

Arch Endocrinol Metab. 2017;61/3

original article

Orange juice with a high-fat meal prolongs postprandial lipemia in apparently healthy overweight/obese women Raquel Cristina L. A. Coelho1, Helen Hermana M. Hermsdorff1, Renata S. Gomide1, Raquel Duarte M. Alves1, Josefina Bressan1

ABSTRACT Objective: We investigated the postprandial response of lipid markers to a high-fat meal (HFM) with two different beverages in apparently healthy normal-weight and overweight/obese women. Subjects and methods: This crossover, randomized study enrolled 36 women, of whom 21 had normal weight (body mass index [BMI] 22 ± 1.8 kg/m2) and 15 had overweight/obesity (BMI 31 ± 3.7 kg/m2). In two different test days, the participants ingested a HFM (37% of energy as saturated fat) with 500 mL of water (HFM-W) or 500 mL of orange juice (HFM-OJ). Blood samples were collected at baseline (12-hour fasting), and at 2, 3, and 5 hours postprandial. The analysis included fasting and postprandial total cholesterol, HDL-c, LDL-c, triglycerides (TG), uric acid, and complement C3. Brazilian Clinical Trials Registry (ReBEC); Primary Identification Number: RBR-2h3wjn (www. ensaiosclinicos.gov.br). Results: TG levels increased at 3 hours with HFM-OJ in normal-weight women (p = 0.01) and returned to normal levels at 5h. TG increased at 3 hours with HFM-W (p = 0.01) and HFM-OJ (p = 0.02), and remained high at 5 hours (p = 0.03) in overweight/obese women. Complement C3 remained unchanged, but showed different responses between meals (p = 0.01 for positive incremental area under the curve [piAUC] HFM-OJ vs. HFM-W, respectively). Conclusions: In apparently healthy overweight/obese women compared with normal-weight ones, the concomitant intake of orange juice with a HFM prolonged postprandial lipemia but had no effect on postprandial complement C3 concentrations. Arch Endocrinol Metab. 2017;61(3):263-8.

1 Departamento de Nutrição e Saúde, Universidade Federal de Viçosa (UFV), Viçosa, MG, Brasil

Correspondence to: Helen Hermana M. Hermsdorff Av. P. H. Rolfs, s/n, Departamento de Nutrição e Saúde, Universidade Federal de Viçosa, 36570-900 – Viçosa, MG, Brasil helenhermana@ufv.br Received on Oct/8/2015 Accepted on Sep/12/2016 DOI: 10.1590/2359-3997000000229

Keywords

INTRODUCTION

ostprandial lipemia (PPL) refers to the dynamic changes in serum lipids and lipoproteins that occur after a fat load or meal. These responses are reflected mainly in changes in plasma triglycerides (TG). TGrich lipoproteins and their remnants are known as risk predictors of coronary heart disease (1-2). Indeed, PPL has gained interest since recent reports have demonstrated that increases in postprandial TG levels are possibly even stronger independent predictors of cardiovascular diseases than fasting TG (3). Moreover, complement C3 has been positively associated with obesity, insulin resistance, metabolic syndrome features, and fasting and postprandial TG (4). In this sense, the Western dietary pattern, characterized by a consumption of high-energy density diets and refined foods, contributes to a positive energy Arch Endocrinol Metab. 2017;61/3

balance leading to weight gain and obesity and may trigger low-grade systemic inflammation and metabolic syndrome abnormalities (5,6). Western individuals remain in a postprandial state for most of the day (7). Consequently, repeated acute dietary stressors induced by a high-fat meal (HFM) could trigger a large increase of most risk factors for cardiovascular diseases related to obesity, such as increased circulating cholesterol, TG, and glucose (8). In turn, fruit intake has been associated with an improvement in lipid profile and reduction in concentrations of inflammatory and oxidative stress markers (9,10). We have previously published a review of studies assessing the anti-inflammatory properties of orange juice (OJ), which appears to mediate the plasma levels and gene expression of factors involved in metabolic and inflammatory responses in postprandial and chronic (≥ 7 consecutive days) periods (11). 263

Orange juice; high-fat meal; postprandial lipemia; obesity

Orange juice with HFM prolongs lipemia

In the present study, we investigated the postprandial response of lipid markers to a HFM with two different beverages (water and OJ) in apparently healthy normalweight and overweight/obese women.

SUBJECTS AND METHODS Subjects Recruitment was conducted through the university website, posters, and active search in clinical and medical service centers. In total, 74 women were recruited and 45 were enrolled in the study. Six women failed to complete the study claiming lack of time and three presented technical problems in blood collection. The participants comprised normal-weight (n = 21) and overweight/obese (n = 15) women. They were apparently healthy with no recent acute or chronic inflammatory diseases and/or use of anti-inflammatory or immunosuppressive drugs and steroids. They were non-smokers and, to be enrolled, could not be pregnant or nursing. Subjects were excluded if they had any past or present cardiovascular disease, diagnosed diabetes or any inflammatory condition, or used medications known to affect the study outcomes. Approval for the study was obtained from the Ethics Committee for Human Research at Universidade Federal de Viçosa (Of. Ref. Nº 184/2011) and all procedures involving human subjects complied with the Declaration of Helsinki, as revised in 2000. All participants signed a written informed consent form. The study is registered at the Brazilian Clinical Trials Registry (ReBEC; www.ensaiosclinicos.gov. br) with the primary identification number RBR-2h3wjn.

Study design The dietary intervention followed a randomized crossover design, with a washout period of at least 7 days between the test days. Two days prior to each test day, the subjects followed a low-antioxidant diet (washout) by avoiding olive and fish oils, fresh fruits and vegetables, tea, coffee, fruit juices, and wine. The subjects were then randomly assigned to either a HFM plus 500 mL of water (HFM-W) or a HFM plus 500 mL OJ (HFM-OJ) group. On the morning of each test day, the participants arrived at the laboratory at 7:00-7:30 am. Body weight, height, and blood pressure were measured, and a fasting blood sample was taken before the test meal. Body weight and height were measured using standard procedures according 264

to previously described protocols (12). Body mass index (BMI) was calculated using the equation: BMI = weight (kg)/height2 (m2). The percentage of body fat was estimated by bioelectrical impedance analysis (Biodynamics 310e, Chicago, USA) using standard protocols (12). Blood pressure was measured using a standard mercury sphygmomanometer, with the patient in the seated position. The test meals were served at 7:30-8:00 am and consumed within 30 min. Postprandial blood samples were obtained at 2, 3, and 5 hours after the beginning of the test meals. The subjects remained in the laboratory and were not allowed to consume any additional foods or beverages, except for water.

Composition of the meals The HFM consisted of two muffins with bacon and cheese (90g each), providing 1010 kcal, with 78% of energy as fat (37% as saturated fat), 16% as carbohydrates, and 6% as protein. The muffins were accompanied by 500 mL of water (HFM-W) or OJ (HFM-OJ) in different test meal days. Concentrated integral sugarfree OJ (100% OJ) was provided by Fast Fruit® in 1-liter packages; an amount of 500 mL provides 215 kcal from 50g of carbohydrates (information provided by the manufacturer). The juice package was opened at the time of the consumption.

Assessment of metabolic markers Fasting and postprandial blood samples were collected in EDTA tubes using a 21G butterfly needle and were centrifuged immediately at 2,200 x g at 5ºC for 15 min. The plasma was then separated and stored at -80ºC. The analyses were performed in the semiautomatic analyzer BS200 (Mindray, Nanshan, China). Plasma concentrations of TG, total cholesterol, HDL-c, LDL-c, uric acid, and glucose were measured using colorimetric enzymatic commercial kits (Bioclin, Belo Horizonte, Brazil). Plasma concentrations of complement C3 were measured by turbidimetric methods using commercially available kits (Bioclin, Belo Horizonte, Brazil).

Statistical analysis The results are presented as mean ± standard deviation (SD). Age, BMI, body weight and composition, and plasma baseline metabolic biomarkers were compared between groups using Student’s t test or Mann-Whitney test, as appropriate. As the groups (normal-weight and overweight/obese) differed in age, the analyses were Arch Endocrinol Metab. 2017;61/3

Orange juice with HFM prolongs lipemia

RESULTS Baseline A total of 36 women completed the study and served as controls for themselves. Baseline values of weight, BMI, body fat, glucose, HDL-c, and uric acid differed between groups (normal weight vs. overweight/obese), as presented in Table 1. Table 1. Baseline characteristics of the participants Participants (n) Age (y)

Normal-weight women

Overweight/ Obese women

24 ± 4

31 ± 8

P value

0.022

Weight (kg)

58 ± 5

81.4 ± 13

< 0.001

BMI (kg/m2)

22 ± 1.8

31.1 ± 3.7

< 0.001

Body fat (%)

25.8 ± 3.2

37 ± 3.2

< 0.001

SBP (mmHg)

103.3 ± 7.2

110.6 ± 8.8

0.200

DBP (mmHg)

64.8 ± 6.7

69.4 ± 8.7

0.103

Glucose (mg/dL)

88.2 ± 6.5

97.9 ± 7

0.004

TC (mg/dL)

168.5 ± 31.6

168.21 ± 26.4

0.793

HDL-c (mg/dL)

67.2 ± 17.2

50.8 ± 7.1

0.005

LDL-c (mg/dL)

81.4 ± 24

87 ± 16.5

0.672

96.8 ± 32.8

136.3 ± 65.8

0.176

3.7 ± 0.7

4.4 ± 0.6

0.004

137.5 ± 29.3

142.9 ± 25.9

0.436

TG (mg/dL) Uric acid (mg/dL) C3 (mg/dL)

Values are expressed as mean ± standard deviation (SD). BMI: body mass index; MBR: basal metabolic rate; SBP: systolic blood pressure; DBP: diastolic blood pressure; TC: total cholesterol; TG: triglycerides. P values for comparisons between groups using Student’s t test or Mann-Whitney test. The results remained after adjustment by age (Tukey-Kramer test). Arch Endocrinol Metab. 2017;61/3

Metabolic postprandial response After consumption of the HFM-W meal, TG levels tended to increase in normal-weight volunteers at 3 hours postprandial relative to fasting (p = 0.07). However, this increase at the third hour was only significant when these participants consumed the HFM-OJ meal (p = 0.01). At 5 hours, there was no difference between baseline and postprandial TG in normal-weight women (p = 0.99). Overweight/obese women presented increased TG in relation to fasting at 3 hours postprandial both after the HFM-W (p = 0.01) and HFM-OJ (p = 0.02) meals. Furthermore, the increase in TG compared with fasting remained at 5 hours after consumption of the HFM-OJ meal in overweight/obese volunteers (p = 0.03) (Figure 1). At 5 hours postprandial, there was a trend towards a difference in TG levels between the groups of normal-weight and overweight/obese volunteers (p = 0.06). Complement C3 presented a significantly higher piAUC (p = 0.03) in the normal-weight group. Total cholesterol, HDL-c, and LDL-c did not vary statistically over time or between diets.

DISCUSSION In this study, PPL occurred 3 hours after the ingestion of a HFM. Studies have shown that circulating TG presents a pronounced increase (i.e., PPL) 1 hour after consumption of a typical fat-containing meal (30–60g of fat) and may remain high for 5–8 hours (13,14). Recent epidemiological studies have clearly evidenced the predictive relationship between the extent of postprandial hypertriglyceridemia and a relative risk of cardiovascular events (15,16). This finding is relevant concerning the analyses of lipids as biomarkers in chronic diseases. The relevant finding of this study was that overweight/obese women consuming a HFM-OJ meal had a prolonged TG increase, with higher TG levels at 5 hours postprandial. Peairs and cols. reported increased postprandial TG in obese patients (17). The first point of the discussion is the difference in PPL between normalweight and overweight/obese women. Normal-weight women presented a TG increase only after the HFMOJ meal, while overweight/obese volunteers had a TG increase after both meals. In fact, the amplitude and duration of the PPL are related to the meal composition and the physiopathological condition of the subjects, 265

adjusted for this variable (Tukey-Kramer test). A mixed model using a three-way repeated-measures ANOVA was applied to test the differences between test meals throughout the test day for postprandial metabolic and inflammatory variables with test meals, groups, and time as repeated factors. Post hoc testing was performed using the Tukey-Kramer test. The statistical analyses were performed using the SAS statistical package (version 9.2; SAS Institute Inc, Cary, NC, USA). The rejection level of significance used was 5%. The incremental area under the curve (piAUC) was calculated using GraphPad Prism (Version 5; GraphPad Software Inc., USA). Power analysis, calculated with the analyst procedure of SAS, indicated that a sample of 15 subjects per group would allow the detection of a treatment effect accounting for 5% of the within-subject variance in TG with more than 99% of power at the 5% level of probability.

Orange juice with HFM prolongs lipemia

180

160

140 120 *

100 80

Water Orange juice

Test C meal 160 155 150 145 140 135 130 125

120

Water Orange juice

100

2 3 Time (Hours)

Test meal

140

C3 (mg/dL)

2 3 Time (Hours)

Triglycerides (mg/dL)

Test meal

Test D meal 160 155 150 145 140 135 130 125

Water Orange juice 2 3 Time (Hours)

Figure 1. Line plots showing the changes (as mean ± standard error) in plasma triglycerides and C3 after a high-fat meal plus water (HFM-W) and a high-fat meal plus orange juice (HFM-OJ) in normal-weight (A and C, respectively) and overweight (B and D, respectively) participants. Mixed model using three-way repeated-measures ANOVA followed by Tukey-Kramer post hoc analysis: * p < 0.05; § p < 0.01. Difference between beverages at a single point.

including obesity (18,19). PPL in overweight/obese women disclosed a lipid intolerance state that could not be detected in fasting. In obese humans, fasting plasma lipids can be normal but postprandial lipid metabolism is abnormal with an accumulation of TGrich remnant lipoproteins. In addition, the catabolism of their chylomicrons (CM) remnants was markedly decreased when compared with lean women (20). The decreased clearance of CM remnants in overweight/ obese subjects may be explained by a competition between these remnants and the increased hepatic production of VLDL for clearance by low-density lipoprotein receptors. The second point of the discussion is the role of OJ on PPL in normal-weight and overweight/obese women. Studies about the effect of fruit-derived antioxidant and fructose intake on fatty meal-induced metabolic changes have reported contradictory results. We observed an enhanced PPL when OJ was added to the HFM. Similar to our findings, Stanhope and cols. observed that diets rich in highly digestible carbohydrates can lead to higher levels of fasting plasma TG as a result of hepatic VLDL and CM remnant accumulation due to altered lipoprotein secretion and/ 266

or clearance (21). Cerletti and cols. found that the concomitant intake of OJ resulted in a reduced increase in plasma TG and reduction in total cholesterol (22). These findings altogether indicate that postprandial metabolism resulting from the digestion and absorption of available nutrients is a highly complex process involving numerous potential interactions. In addition, other studies have shown that the amount or type of carbohydrate in a meal alter the postprandial lipid metabolism (23). However, data obtained after addition of glucose (50 to 100g) to a HFM have not provided reproducible findings in healthy subjects (24). Thus, despite epidemiological data suggesting an inverse association between citrus fruits intake and cardiovascular disease risk, our understanding of the mechanisms by which flavonones potentially reduce this risk remains unclear. As a limitation of the study, the total energy intake in the meal with OJ was 215 kcal higher than that in the meal with water. The addition of OJ increased the amount of carbohydrates in the meal (40.7g for HFM-W vs. 91g for HFM-OJ), without contributing to fat content. Given its fructose content, OJ may have altered the PPL by increasing hepatic fat synthesis (25). Arch Endocrinol Metab. 2017;61/3

Orange juice with HFM prolongs lipemia

Arch Endocrinol Metab. 2017;61/3

to clarify the role of OJ or citrus juice intake in the lipid metabolism and in the prevention of chronic diseases. Acknowledgments: we wish to thank all volunteers who participated in this study, the nursing staff for their excellent technical assistance, and all students who helped in the study fieldwork. This work was supported by the Foundation for Research Support of the State of Minas Gerais (Fapemig – CDS – APQ-00474-12) and by CNPq (474679/2013-6). The Capes Foundation and CNPq provided research grants to RCLA Coelho and RDM Alves, respectively. HHM Hermsdorff and J Bressan are CNPq fellows. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Kolovou G, Ooi TC. Postprandial lipaemia and vascular disease. Curr Opin Cardiol. 2013;28(4):446-51. 2. Sarwar N, Sandhu MS, Ricketts SL, Butterworth AS, Di Angelantonio E, Boekholdt SM, et al. Triglyceride-mediated pathways and coronary disease: collaborative analysis of 101 studies. Lancet. 2010;375(9726):1634-9. 3. Nordestgaard BG, Benn M, Schnohr P, Tybjaerg-Hansen A. Nonfasting triglycerides and risk of myocardial infarction, ischemic heart disease, and death in men and women. JAMA. 2007;298(3):299-308. 4. Klop B, Proctor SD, Mamo JC, Botham KM, Castro Cabezas M. Understanding postprandial inflammation and its relationship to lifestyle behaviour and metabolic diseases. Int J Vasc Med. 2012:947417. 5. Bressan J, Hermsdorff HH, Zulet MA, Martinez JA. Hormonal and inflammatory impact of different dietetic composition: emphasis on dietary patterns and specific dietary factors. Arq Bras Endocrinol Metabol. 2009;53(5):572-81. 6. Van Oostrom AJ, Alipour A, Plokker TW, Sniderman AD, Cabezas MC.The metabolic syndrome in relation to complement component 3 and postprandial lipemia in patients from an outpatient lipid clinic and healthy volunteers. Atherosclerosis. 2007;190(1):167-73. 7. Nakajima K, Nakano T, Tokita Y, Nagamine T, Inazu A, Kobayashi J, et al. Postprandial lipoprotein metabolism: VLDL vs chylomicrons. Clin Chim Acta. 2011;412(15-16):1306-18. 8. Sharrett AR, Heiss G, Chambless LE, Boerwinkle E, Coady SA, Folsom AR, et al. Metabolic and lifestyle determinants of postprandial lipemia differ from those of fasting triglycerides: The Atherosclerosis Risk In Communities (ARIC) study. Arterioscler Thromb Vasc Biol. 2001;21(2):275-81. 9. Hermsdorff HH, Zulet MA, Puchau B, Martinez JA. Fruit and vegetable consumption and proinflammatory gene expression from peripheral blood mononuclear cells in young adults: a translational study. Nutr Metab (Lond). 2010;7:42. 10. Cocate PG, Natali AJ, Oliveira AD, Longo GZ, Alfenas Rde C, Peluzio Mdo C, et al. Fruit and vegetable intake and related nutrients are associated with oxidative stress markers in middleaged men. Nutrition. 2014;30(6):660-5. 11. Coelho RC, Hermsdorff HH, Bressan J. Anti-inflammatory properties of orange juice: possible favorable molecular and metabolic effects. Plant Foods Hum Nutr. 2013;68(1):1-10. 12. de Oliveira FC, Alves RD, Zuconi CP, Ribeiro AQ, Bressan J. Agreement between different methods and predictive equations for resting energy expenditure in overweight and obese Brazilian men. J Acad Nutr Diet. 2012;112(9):1415-20.

267

However, Stookey and cols. have shown that the addition of OJ to a meal with 12g of fat limited fat oxidation in the postprandial period (26). These results suggest that reduced fat oxidation might mediate the effects of caloric beverages on weight gain, independent of energy excess. In adults, reduced fat oxidation predicts weight gain, independent of metabolic rate (27). Another point is the 48-hour washout diet applied before the test meal days. This low-antioxidant diet aimed to minimize the variability of the biochemical and inflammatory markers analyzed, which was important in view of the physiological approach in the current study. However, this diet could hamper the application of the results in free-living conditions. Regarding complement C3, some studies have shown a correlation between fasting C3 and TG (28,29). However, we did not find correlations between fasting C3 and metabolic syndrome parameters, probably because of the small number of individuals assessed. Furthermore, Halkes and cols. have found plasma C3 levels significantly higher than fasting at 2, 4, and 6 hours after fat ingestion (50g/m2 of fat) by normolipidemic subjects with coronary artery disease and healthy controls (30). We did not find a significant postprandial difference in plasma C3. Possible causes for this negative outcome are the different populations assessed, different fat overload, and the addition of OJ. Charlesworth and cols. also found no significant postprandial changes in C3 after a mixed meal, although their study lacked information about the amount of fat (31). Van Oostrom and cols., while studying the relationship of PPL with meal composition in healthy subjects, observed that the addition of glucose to a fat overload decreased PPL and prevented a fat-specific increase in C3 (32). For Schär and cols., OJ or hesperidin supplement did not acutely affect cardiovascular risk biomarkers (33). Because of existing controversial results concerning postprandial response of complement C3 to meal composition, more studies are necessary to investigate the biochemical and dietary factors related to variations in circulating C3 in healthy and obese people. In conclusion, our study has shown that overweight/ obese women have enhanced PPL when compared with normal-weight women, and that the increase in TG was more prolonged when these overweight/ obese volunteers consumed a HFM with OJ. Since the findings of our present study and previously published ones are still controversial, more studies are necessary

Orange juice with HFM prolongs lipemia

13. Lairon D, Defoort C (2011) Effects of nutrients on postprandial lipemia. Curr Vasc Pharmacol. 2011;9(3):309-12.

25. Lairon D. Macronutrient intake and modulation on chylomicron production and clearance. Atheroscler Suppl. 2008;9(2):45-8.

14. Wierzbicki AS, Clarke RE, Viljoen A, Mikhailidis DP. Triglycerides: a case for treatment? Curr Opin Cardiol. 2012;27(4):398-404. 15. Goldberg IJ, Eckel RH, McPherson R.Triglycerides and heart disease: still a hypothesis? Arterioscler Thromb Vasc Biol. 2011;31(8):1716-25.

26. Stookey JD, Hamer J, Espinoza G, Higa A, Ng V, Tinajero-Deck L, et al. Orange juice limits postprandial fat oxidation after breakfast in normal-weight adolescents and adults. Adv Nutr. 2012;3(4):629S-635S.

16. Freiberg JJ, Tybjaerg-Hansen A, Jensen JS, Nordestgaard BG. Nonfasting triglycerides and risk of ischemic stroke in the general population. JAMA. 2008;300(18):2142-52.

27. Melanson EL, MacLean PS, Hill JO. Exercise improves fat metabolism in muscle but does not increase 24-h fat oxidation. Exerc Sport Sci Rev. 2009;37:93-101.

17. Peairs AD, Rankin JW, Lee YW. Effects of acute ingestion of different fats on oxidative stress and inflammation in overweight and obese adults. Nutr J. 2011;10:122.

28. Muscari A, Massarelli G, Bastagli L, Poggiopollini G, Tomassetti V, Drago G, et al. Relationship of serum C3 to fasting insulin, risk factors and previous ischaemic events in middle-aged men. Eur Heart J. 2000;21(13):1081-90.

18. Herieka M, Erridge C. High-fat meal induced postprandial inflammation. Mol Nutr Food Res. 2014;58(1):136-46. 19. Tam CS, Viardot A, Clement K, Tordjman J, Tonks K, Greenfield JR, et al. Short-term overfeeding may induce peripheral insulin resistance without altering subcutaneous adipose tissue macrophages in humans. Diabetes. 2010;59(9):2164-70. 20. Lopez-Miranda J, Williams C, Lairon D. Dietary, physiological, genetic and pathological influences on postprandial lipid metabolism. Br J Nutr. 2007;98(3):458-73. 21. Stanhope KL, Schwarz JM, Keim NL, Griffen SC, Bremer AA, Graham JL, et al. Consuming fructose-sweetened, not glucosesweetened, beverages increases visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese humans. J Clin Invest. 2009;119(5):1322-34. 22. Cerletti C, Gianfagna F, Tamburrelli C, De Curtis A, D’Imperio M, Coletta W, et al. Orange juice intake during a fatty meal consumption reduces the post-prandial low grade inflammatory response in healthy subjects. Thromb Res. 2015;135(2):255-9. 23. Parks EJ, Krauss RM, Christiansen MP, Neese RA, Hellerstein MK. Effects of a low-fat, high-carbohydrate diet on VLDL-triglyceride assembly, production, and clearance. J Clin Invest. 1999;104(8):1087-96.

30. Halkes CJ, van Dijk H, de Jaegere PP, Plokker HW, van Der Helm Y, Erkelens DW, et al. Postprandial increase of complement component 3 in normolipidemic patients with coronary artery disease: effects of expanded-dose simvastatin. Arterioscler Thromb Vasc Biol. 2001;21(9):1526-30. 31. Charlesworth JA, Peake PW, Campbell LV, Pussel BA, O’Grady S, Tzilopoulos T. The influence of oral lipid loads on acylation stimulating protein (ASP) in healthy volunteers. Int J Obes Relat Metab Disord. 1998;22(11):1096-102. 32. Van Oostrom AJ, Dijk HV, Verseyden C, Sniderman AD, Cianflone K, Rabelink T, et al. Addition of glucose to an oral fat load reduces postprandial free fatty acids and prevents the postprandial increase in complement component 3. Am J Clin Nutr. 2004;79(3):510-5. 33. Schär MY, Curtis PJ, Hazim S, Ostertag LM, Kay CD, Potter JF, et al. Orange juice derived flavanone and phenolic metabolites do not acutely affect cardiovascular risk biomarkers: a randomized, placebo controlled, crossover trial in men at moderate risk of cardiovascular disease. Am J Clin Nutr. 2015;101(5):931-8.

24. Roche HM, Gibney MJ. Long-chain n-3 polyunsaturated fatty acids and triacylglycerol metabolism in the postprandial state. Lipids. 1999;34 Suppl:S259-65.

29. Hermsdorff HHM, Puchau B, Zulet MA, Martínez JA. Association of body fat distribution with proinflammatory gene expression in peripheral blood mononuclear cells from young adult subjects. OMICS. 2010;14(3):297-307.

268

Arch Endocrinol Metab. 2017;61/3

original article

Ultrasonographic assessment of thyroid volume in oldest-old individuals Glaucia Cruzes Duarte1, Lara Miguel Quirino Araujo1, Felix Magalhães Filho1, Clineu Mello Almada Filho1, Maysa Seabra Cendoroglo1

ABSTRACT Objective: The aim of this study was to describe the relationship between thyroid volume and age, gender, anthropometric characteristics, and echogenicity in oldest-old subjects in an iodine-sufficient area. Subjects and methods: The study included 81 independent elderly individuals aged ≥ 80 years (65 [80.2%] women). We determined these individuals’ anthropometric characteristics, body mass index (BMI), and lean body mass, as well as thyroid volume and echogenicity by ultrasonography. Results: We observed that octogenarians and nonagenarians had different profiles of thyroid echogenicity. The volume of the thyroid was smaller in nonagenarians than octogenarians (p = 0.012, r = 0.176), and subjects aged 80–89 years had more often hypoechoic glands than those aged ≥ 90 years (p = 0.01 versus 0.602). Conclusion: The identification of ultrasonographic differences in oldestold individuals will contribute to establishing preclinical markers, such as echogenicity, to identify individuals at risk of developing autoimmune thyroid disease. Future prospective studies should identify if 80–89-year-old individuals with hypoechoic glands progress to hypothyroidism, and if the absence of changes in echogenicity (i.e. a normal thyroid parenchyma) would have a positive impact on longevity among nonagenarians. Arch Endocrinol Metab. 2017;61(3):269-75.

Disciplina de Geriatria e Gerontologia, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil

Correspondence to: Glaucia Cruzes Duarte Rua Professor Francisco de Castro, 105 04020-050 – São Paulo, SP, Brasil endocrino.draglauciaduarte@gmail.com Received on Mar/14/2016 Accepted on Aug/10/2016 DOI: 10.1590/2359-3997000000223

INTRODUCTION

ltrasonography is widely used in clinical practice as the most reliable method to determine the volume (1,2) and structure of the thyroid gland. The volume of the thyroid is influenced by age, gender, body mass index (BMI), lean body mass, iodine intake, and genetic factors (3-5). Post-mortem thyroid examination of individuals aged ≥ 50 years (6) and confirmed in centenarians (7) has identified progressive atrophy, fibrosis, increased adipose tissue, and decreased follicles and colloid, contributing to a decrease in the volume of the gland with aging (8). However, studies evaluating the dimensions of the thyroid in elderly individuals have not included many participants aged ≥ 80 years and have failed to report the biometric characteristics of the individuals in this specific age group, despite the fact that these characteristics are known to influence the volume of the thyroid in children and adults (9). Thyroid ultrasonography is considered an auxiliary method to identify the occurrence and prognosis of autoimmune thyroid diseases (10-12), and has an important role in identifying individuals at risk of these Arch Endocrinol Metab. 2017;61/3

conditions in epidemiological studies (13,14). Thyroid glands affected with autoimmune disorders may show a hypoechoic pattern (15,16) caused by increased cellularity and a variable degree of lymphocytic infiltration (17). These structural changes usually precede the detection of autoantibodies in the serum and other laboratory abnormalities (18-20). Measurement of thyroid-stimulating hormone (TSH) levels is a common screening method to identify thyroid dysfunction. In 15% of the individuals above the age of 70 years, TSH levels may be elevated, suggesting that the superior limit of the normal range of this hormone may change with aging (21). However, serum TSH is not a sensitive marker in old individuals (22); therefore, thyroid ultrasonography may bring additional information and help predict the progression to thyroid diseases. A better understanding of the ultrasonographic features predicting the development of thyroid diseases, in addition to TSH measurement, could have a large impact on clinical practice guidelines in the geriatric population. In this study, we analyzed by ultrasonography the thyroid 269

Keywords Thyroid diseases; thyroiditis, autoimmune; ultrasonography; oldest old

Thyroid volume in the oldest old

volume of oldest-old individuals, and the relationship of the thyroid volume with age, gender, anthropometric characteristics (weight, height, BMI, lean body mass), and echogenicity.

SUBJECTS AND METHODS A total of 81 independent subjects (65 women, 16 men) aged ≥ 80 years and living in São Paulo, Brazil, were recruited from the geriatrics clinic at Universidade Federal de São Paulo between August, 2012, and February, 2014. The participants were included in the study after signing an informed consent form. The study received approval of the Ethics Committee at Universidade Federal de São Paulo and was conducted according to the Declaration of Helsinki. After recruitment, the individuals were allocated to a “normal TSH group” or an “increased TSH” group; this last included individuals with serum TSH levels > 4.5 mIU/L. The exclusion criteria were cognitive impairment; renal, hepatic or hematological diseases; and history of radioiodine therapy or thyroidectomy. Serum TSH values (normal range 0.5–4.5 mIU/L) and free thyroxine (FT4; normal range 0.83–1.7 ng/dL) were obtained from medical records. Lean body mass was assessed by bioelectrical impedance analysis (Biodynamics-310, Model A, Biodynamics Corp., Seattle, USA) according to the device manufacturers’ instructions. Thyroid ultrasonography was performed by the same physician (GD) to avoid interobserver variation. The evaluations were conducted on a LOGIC e (GE) equipment attached with a 7.5 MHz linear transducer. The gain was adjusted to minimize the echoes in the carotid artery and jugular vein, limiting variations in brightness. During the examination, the subjects rested in a supine position with their necks slightly hyperextended. We obtained images in the transverse and longitudinal planes, and measured each lobe at its maximum transverse, longitudinal, and anteroposterior diameters (height, width, and depth) to calculate the volume of the thyroid according to Brunn and cols. (23). For the purpose of this study, we considered as “appropriate” those thyroid volumes between 6 and 20 milliliters (mL) (24). We determined the echogenicity of the thyroid parenchyma using a grayscale analysis, comparing the parenchyma with adjacent structures. The observed echogenicity was then categorized into one of three classes: isoechoic (when the echogenicity of the 270

parenchyma was similar to that of the submandibular gland), mildly hypoechoic (when the parenchyma was hypoechoic compared with the submandibular gland, but hyperechoic in relation to the cervical muscles), or hypoechoic (when the parenchyma was isoechoic or hypoechoic when compared with the cervical muscles).

Statistical analysis We used independent samples t test or MannWhitney test to evaluate the relationship between anthropometric features and thyroid function according to age group or gender. Chi-square test or Fisher’s exact test was used to analyze qualitative variables. Analysis of variance (ANOVA) compared the volume of the thyroid according to age group (octogenarians, 80–89 years; and nonagenarians, ≥ 90 years), and levels of echogenicity. Spearman’s correlation coefficient or multiple linear regression were used to correlate the volume of the thyroid with age, weight, height, BMI, lean body mass, and echogenicity. P values < 0.05 were considered significant. All analyses were performed with the statistical software R, version 2.15.2 and/or NCSS.

RESULTS A total of 81 patients (65 women, 16 men) were allocated to the normal TSH group (n = 52) or increased TSH group (n = 29). Overall, there were 54 (66.7%) individuals aged 80–89 years and 27 (33.3%) aged ≥ 90 years (Table 1). As expected, the average TSH level found in elderly individuals in the normal TSH group was lower than that in individuals in the increased TSH group, with a significant difference for both octogenarians (2.95 versus 5.69 mIU/L; p = 0.027) and nonagenarians (3.17 versus 5.56 mIU/L; p = 0.013) (Table 1). There were no differences regarding FT4 values between both groups (normal TSH versus increased TSH). All anthropometric characteristics and thyroid function tests are shown in Table 1. Among individuals aged 80–89 years, 35 were in the normal TSH group, and 19 were in the increased TSH group. There were no differences in thyroid volumes (10.0 ± 3.60 mL versus 9.18 ± 6.59 mL, p = 0.105) between individuals in the normal TSH and increased TSH groups. In contrast, hypotrophic glands (< 6 mL) were observed in six elderly individuals in the normal TSH group (17%) and eight in the increased Arch Endocrinol Metab. 2017;61/3

Thyroid volume in the oldest old

in the increased TSH group. We did not observe any patient with goiter among nonagenarians, or differences in echogenicity pattern between nonagenarians patients in the normal TSH and increased TSH groups (p = 0.602). Thyroid nodules were present in 30 octogenarians and 21 nonagenarians, totaling 62.9% of the sample. Six patients with nodules greater than 1.0 cm and suspicious ultrasonographic features (solid and hypoechoic, with microcalcifications and irregular borders) were referred to fine-needle aspiration biopsy, which excluded malignancy. Mean thyroid volumes, echogenicity, and presence or absence of nodules in both groups and subgroups are shown in Table 2.

TSH group (42%). Goiter (> 20 mL) was found in one oldest-old individual and was associated with nodules, but the TSH level in this individual was in the normal range. When we compared octogenarians allocated to the normal TSH group with those in the increased TSH group, we observed an increased frequency of isoechoic glands in the first group and of hypoechoic glands in the second one (p = 0.001). Among individuals aged ≥ 90 years, the mean thyroid volume was significantly different between individuals in the normal TSH and increased TSH groups, also as expected (11.50 ± 4.13 mL versus 7.37 ± 3.29 mL, respectively, p = 0.012). There were six hypotrophic (< 6 mL) glands, two in the normal TSH group and four

Table 1. Anthropometric characteristics and thyroid function in oldest-old individuals 80-89 years

Normal TSH

Increased TSH

Mean ± SD

(n = 35)

(n = 19)

Weight (kg)

63.78 ± 14.06

57.77 ± 10.84

Height (m)

1.55 ± 0.09

BMI (kg/m2) Lean body mass (kg)

≥ 90 years P

Normal TSH

Increased TSH

(n = 17)

(n = 10)

0.112*

60.25 ± 10.15

61.40 ± 9.22

0.841a

1.52 ± 0.06

0.118*

1.54 ± 0.07

1.47 ± 0.03

0.010a

26.19 ± 4.22

25.05 ± 4.31

0.384a

25.28 ± 3.72

28.36 ± 4.22

0.040a

40.11 ± 9.42

34.28 ± 5.18

0.027a

37.89 ± 7.64

35.59 ± 4.73a

0.353*

TSH mIU/mL

2.95 ± 1.81

5.69 ± 4.78

0.031a

3.17 ± 1.90

5.56 ± 2.39

0.013a

T4L ng/dL

1.19 ± 0.23

1.20 ± 0.24

0.911*

1.26 ± 0.25

1.72 ± 1.18

0.414a

Characteristics

* Student’s t test; a Mann-Whitney test.

Table 2. Thyroid volume, echogenicity, and presence of nodules in oldest old individuals 80-89 years

Thyroid volume (mL) – Mean ± SD Total

≥ 90 years

Normal TSH

Increased TSH

Normal TSH

Increased TSH

(n = 35)

(n = 19)

(n = 17)

(n = 10)

10.07 ± 3.60

9.18 ± 6.59

0.105a

11.50 ± 4.13

7.37 ± 3.29

0.012*

< 6 mL (n; %)

6; 17.1%

8; 42.1%

0.058b

2; 11.8%

4; 40.0%

0.154b

> 20 mL (n; %)

0; 0.0%

1; 5.3%

0.352b

0; 0.0%

0;0.0%

ECO1

22; 62.9%

4; 21.1%

8; 47.1%

3; 30%

ECO2

11; 31.4%

7; 36.8%

4; 23.5%

2; 20%

ECO3

2; 5.7%

8; 42.1%

5; 29.4%

5; 50%

23; 65.7%

7; 36.8%

14; 82.4%

7; 70%

12; 34.3%

12; 63.2%

3; 17.6%

3; 30%

Echogenicity – (n; %) 0.001c

0.602b

Yes No

0.080c

> 0.999b

* Student’s t test; Mann-Whitney test; Fisher’s exact test; chi-square test. a

ECO1: isoechoic (the echogenicity of the parenchyma was similar to that of the submandibular gland); ECO2: mildly hypoechoic (the parenchyma was hypoechoic compared with the submandibular gland, but hyperechoic in relation to the cervical muscles); ECO3: hypoechoic (the parenchyma was isoechoic or hypoechoic when compared with the cervical muscles). Arch Endocrinol Metab. 2017;61/3

271

Nodules – (n; %)

Thyroid volume in the oldest old

Concerning gender, there were 16 men and 65 women (19.7% and 80.3%, respectively) in the overall cohort. The p values for the statistical tests are not shown for men, only the descriptive results in each group. This approach was preferred due to the fact that there were only two men in the increased TSH group. The thyroid volume of men in the normal TSH group was 11.66 ± 3.40 mL compared with 12.86 ± 4.78 mL in those in the increased TSH group. Among women, there were significant differences regarding TSH levels in the normal TSH and increased

TSH groups, as expected (normal TSH group, 3.01 ± 1.74 mIU/L; increased TSH group, 5.67 ± 4.02 mIU/L; p = 0.03) and thyroid volume (10.13 ± 3.90 mL versus 8.24 ± 5.68 mL, respectively; p = 0.015). There were eight (21.1%) women in the normal TSH group with atrophic glands, and no cases of goiter. In the increased TSH group, 12 individuals had thyroid volumes smaller than 6 mL, and one female patient had a volume greater than 20 mL. Regarding qualitative variables, only the echogenicity was statistically significant (p = 0.001), as shown in Tables 3 and 4.

Table 3. Anthropometric characteristics and thyroid function by gender in oldest-old individuals Male

Female

Normal TSH

Increased TSH

Normal TSH

Increased TSH

(n = 14)

(n = 2)

(n = 38)

(n = 27)

Weight (kg)

69.76 ± 12.56

54.00 ± 6.08

60.00 ± 12.20

59.40 ± 10.52

0.905a

Height (m)

0.124a

Characteristics Mean ± SD

1.63 ± 0.08

1.57 ± 0.08

1.52 ± 0.07

1.50 ± 0.05

BMI (kg/m )

26.24 ± 3.05

22.15 ± 4.88

25.76 ± 4.39

26.49 ± 4.42

0.513*

Lean body mass (kg)

47.76 ± 9.66

38.45 ± 0.49

36.04 ± 5.90

34.49 ± 5.04

0.305*

TSH mIU/mL

3.07 ± 2.11

5.05

3.01 ± 1.74

5.67 ± 4.02

0.003a

FT4 ng/dL

1.21 ± 0.22

1.00

1.22 ± 0.25

1.43 ± 0.80

0.074a

* Student’s t test; a Mann-Whitney test. BMI: body mass index; FT4: free thyroxine.

Table 4. Thyroid volume, echogenicity, and presence of nodules by gender in oldest-old individuals Men

Women

Normal TSH

Increased TSH

Normal TSH

Increased TSH

(n = 35)

(n = 19)

(n = 17)

(n = 10)

11.66 ± 3.40

12.86 ± 4.78

10.13 ± 3.90

8.24 ± 5.68

0.015a

< 6 mL (n; %)

0; 0.0%

8; 21.1%

12; 44.4%

0.082c

> 20 mL (n; %)

0; 0.0%

1;3.7%

0.415b

ECO1

10; 71.4%

2; 100.0%

20; 52.6%

5; 18.5%

ECO2

1; 7.1%

0; 0.0%

14; 36.8%

9; 33.3%

ECO3

3; 21.4%

0; 0.0%

4; 10.5%

13; 48.1%

8; 57.1%

1; 50.0%

29; 67.4%

14; 32.6%

12; 63.2%

3; 17.6%

3; 30%

Thyroid volume (mL) – Mean ± SD Total

Echogenicity – (n; %) 0.001c

Nodules – (n; %) Yes No

12; 34.3%

0.074c

* Student’s t test; Mann-Whitney test; Fisher’s exact test; Chi-square test. a

272

Arch Endocrinol Metab. 2017;61/3

Thyroid volume in the oldest old

Arch Endocrinol Metab. 2017;61/3

2 Echogenicity

Figure 1. Total thyroid volume (mL) versus echogenicity in oldest-old individuals. 273

Although there has been an increasing prevalence of thyroid disorders in elderly individuals (25,26), most population studies have failed to dedicate exclusive attention to individuals aged 80 years or older. In our study, we compared the thyroid volume of octogenarians and nonagenarians with and without evidence of thyroid disorder, since it is not clear in the literature if the volume of this gland in elderly men and women are predictive of thyroid disease. In a clinical context, overt hypothyroidism is preceded by a period of subclinical thyroid dysfunction, with a range of nonspecific symptoms that can be confounded with other geriatric syndromes (27) and may, as a consequence, be undertreated (28,29). According to some authors, the volume of the thyroid decreases with aging (6-8,24). Although the mean thyroid volume in the overall cohort was deemed appropriate according to the criteria adopted in our study, we observed that 20.5% of our oldest-old individuals had a reduced thyroid volume and 1.2% had goiter (> 20 mL). The finding of a prevalence of goiter below 5% was already expected in our population, since the city of São Paulo, where our study was conducted, is considered an iodine-sufficient area (9,30,31). Anthropometric characteristics are known to affect the volume of the thyroid. We found no difference in anthropometric characteristics in our octogenarian and nonagenarian subjects (p = 0.301, r = 0.087), although there were significant differences between individuals in the normal TSH and increased TSH groups regarding lean body mass (p = 0.027, r = 0.486) in individuals aged 89–89 years, and regarding height (p = 0.01, r = 0.303) and BMI (p = 0.04, r = -0.114) in those aged ≥ 90 years. The volume of the thyroid was reduced in nonagenarians (p = 0.012, r = 0.176). Elderly individuals are known to have a progressive decrease in height with increasing age, which in turn impacts their BMI, changing the correlation between BMI and thyroid volume. Men are described as having larger thyroids than women (32-34). Our study showed that oldest-old men had slightly larger thyroids than women, although we were unable to conclude this finding with statistical tests due to the limited number of elderly men in the increased TSH group, which prevented the comparison between gender and thyroid volume. In contrast, our results indicated a significant number of women with a

small thyroid volume (< 6 mL) and more hypoechoic glands (p = 0.001, r = -0.380) when compared with men in our cohort. Our observation that the thyroid volume correlated inversely with thyroid echogenicity (p = 0.001, r = -0.424; Figure 1) in the elderly population has also been shown in children and adults (15,16). The echogenicity of the thyroid changed from isoechoic in individuals allocated to the normal group, to moderately or markedly hypoechoic in octogenarians and women in the increased TSH group. This corroborates previous reports that the hypoechogenicity of the gland is linked to the presence of circulating antithyroid antibodies (35), reflecting intraglandular inflammatory activity and thyroiditis (36). Even though our objective was to describe thyroid ultrasonographic features and factors influencing the variation in thyroid volume, a limitation of our study was the fact that we lacked information regarding antithyroid antibody concentrations in these patients, especially antithyroperoxidase. This prevented a correlation between ultrasonographic findings and the occurrence of autoimmune thyroid disease (15,16,35). However, our finding that certain groups with TSH > 4.5 mIU/L had smaller and more hypoechoic glands could be a sign of autoimmune thyroid disease in these individuals. There were differences in echogenicity patterns between octogenarians and nonagenarians. Among

Thyroid volume (ml)

DISCUSSION

Thyroid volume in the oldest old

individuals allocated to the increased TSH group, subjects aged 80–89 years had more often hypoechoic glands than those aged ≥ 90 years. It is possible that the decreased echogenicity in octogenarians glands, if properly followed up until they reach the age of 90 years or more, could reflect a higher rate of progression to hypothyroidism. Also, the absence of echogenicity abnormalities in individuals aged ≥ 90 years could potentially be associated with longevity. As expected and previously described, the occurrence of nodules increased progressively with age and affected 62.9% of the individuals in our cohort, confirming findings from the literature (37). Thyroid abnormalities are commonly found in elderly individuals. TSH measurements alone may not identify if these abnormalities represent physiological changes in thyroid hormone levels with advancing age or subclinical diseases (22,38). The fact that we searched for ultrasonographic elements that could help decide when levothyroxine replacement should be started is a strength of our study. Although the size of our cohort cannot be characterized as representative of an entire population, our study adds important information for the management of oldest-old patients, since there is a literature gap on specific data about this long-lived population. Although our cross-sectional study does not add definitive information, it is still relevant to enhance the therapeutic planning of very old patients. Thyroid ultrasonography has become a low-cost method to support diagnostic and therapeutic decisions in thyroid disorders by considering parameters such as gland volume and echogenicity. It is important to know that the inverse correlation between volume and echogenicity that we found in our cohort could represent a sign of significant thyroid failure. Each geographic region should have its own reference regarding normal thyroid volume, taking into account nutritional variations (including iodine intake) and genetic differences, although there are many logistical challenges in individualizing the ultrasonographic findings in each elderly population. In conclusion, this study included thyroid ultrasonographic evaluation of oldest-old individuals in São Paulo, Brazil, offering specific thyroid volume values which correlated inversely with echogenicity. Future prospective studies should demonstrate if hypoechoic glands in individuals aged 80–89 evolve into hypothyroidism, and if the absence of echogenicity changes would be associated with longevity in 274

individuals aged ≥ 90 years. Ultrasonographic followup of patients older than 80 years could contribute to establishing a preclinical marker of autoimmune thyroid diseases in predisposed individuals. Authors contributions: all authors contributed to the conception and design of the study; data acquisition, analysis, and interpretation; drafting, revising, and approving the manuscript for submission. Funding: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes) and Fundação de Amparo à Pesquisa do Estado de São Paulo (Fapesp). Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Gray SL, O’Neill G, McGarry G. The predictive value of structured ultrasonographic staging for thyroid nodules. J Laryngol Otol. 2014;128(10):914-21. 2. Kaniuka-Jakubowska S, Piskunowicz M, Zapasnik A, Lewczuk A, Kaniuka A, Mizan-Gross K, et al. US not bright but right method of thyroid volume estimation even in large and substernal extended goitres. Comparison of US and three methods of CT thyroid evaluation – prospective study. Clin Endocrinol (Oxf). 2014;83(3):412-9. 3. Nafisi Moghadam R, Shajari A, Afkhami-Ardekani M. Influence of physiological factors on thyroid size determined by ultrasound. Acta Med Iran. 2011;49(5):302-4. 4. Sari R, Balci MK, Altunbas H, Karayalcin U. The effect of body weight and weight loss on thyroid volume and function in obese women. Clin Endocrinol (Oxf). 2003;59(2):258-62. 5. Zou Y, Ding G, Lou X, Zhu W, Mao G, Zhou J, et al. Factors influencing thyroid volume in Chinese children. Eur J Clin Nutr. 2013;67(11):1138-41. 6. Brown RA, Al-Moussa M, Beck J. Histometry of normal thyroid in man. J Clin Pathol. 1986;39(5):475-82. 7. Andersen-Ranberg K, Jeune B, Hoier-Madsen M, Hegedus L. Thyroid function, morphology and prevalence of thyroid disease in a population-based study of Danish centenarians. J Am Geriatr Soc. 1999;47(10):1238-43. 8. Gonczi J, Szabolcs I, Kovacs Z, Kakosy T, Goth M, Szilagyi G. Ultrasonography of the thyroid gland in hospitalized, chronically ill geriatric patients: thyroid volume, its relationship to age and disease, and the prevalence of diffuse and nodular goiter. J Clin Ultrasound. 1994;22(4):257-61. 9. Zimmermann MB, Hess SY, Molinari L, De Benoist B, Delange F, Braverman LE, et al. New reference values for thyroid volume by ultrasound in iodine-sufficient schoolchildren: a World Health Organization/Nutrition for Health and Development Iodine Deficiency Study Group Report. Am J Clin Nutr. 2004;79(2):231-7. 10. Duarte GC, Tomimori EK, de Camargo RY, Catarino RM, Ferreira JE, Knobel M, et al. Excessive iodine intake and ultrasonographic thyroid abnormalities in schoolchildren. J Pediatr Endocrinol Metab. 2009;22(4):327-34. 11. Rosario PW, Bessa B, Valadao MM, Purisch S. Natural history of mild subclinical hypothyroidism: prognostic value of ultrasound. Thyroid. 2009;19(1):9-12. Arch Endocrinol Metab. 2017;61/3

Thyroid volume in the oldest old

12. Shin DY, Kim EK, Lee EJ. Role of ultrasonography in outcome prediction in subclinical hypothyroid patients treated with levothyroxine. Endocr J. 2010;57(1):15-22. 13. Tomimori EK, Bisi H, Medeiros-Neto G, Camargo RY. Ultrasonographic evaluation of thyroid nodules: comparison with cytologic and histologic diagnosis. Arq Bras Endocrinol Metabol. 2004;48(1):105-13. 14. Raber W, Gessl A, Nowotny P, Vierhapper H. Thyroid ultrasound versus antithyroid peroxidase antibody determination: a cohort study of four hundred fifty-one subjects. Thyroid. 2002;12(8): 725-31. 15. Gutekunst R, Hafermann W, Mansky T, Scriba PC. Ultrasonography related to clinical and laboratory findings in lymphocytic thyroiditis. Acta Endocrinol (Copenh). 1989;121(1):129-35. 16. Nordmeyer JP, Shafeh TA, Heckmann C. Thyroid sonography in autoimmune thyroiditis. A prospective study on 123 patients. Acta Endocrinol (Copenh). 1990;122(3):391-5. 17. Mokshagundam S, Barzel US. Thyroid disease in the elderly. J Am Geriatr Soc. 1993;41(12):1361-9. 18. Namba H, Yamashita S, Kimura H, Yokoyama N, Usa T, Otsuru A, et al. Evidence of thyroid volume increase in normal subjects receiving excess iodide. J Clin Endocrinol Metab. 1993;76(3):605-8. 19. Mazziotti G, Sorvillo F, Iorio S, Carbone A, Romeo A, Piscopo M, et al. Grey-scale analysis allows a quantitative evaluation of thyroid echogenicity in the patients with Hashimoto’s thyroiditis. Clin Endocrinol (Oxf). 2003;59(2):223-9. 20. Marcocci C, Vitti P, Cetani F, Catalano F, Concetti R, Pinchera A. Thyroid ultrasonography helps to identify patients with diffuse lymphocytic thyroiditis who are prone to develop hypothyroidism. J Clin Endocrinol Metab. 1991;72(1):209-13. 21. Boucai L, Hollowell JG, Surks MI. An approach for development of age-, gender-, and ethnicity-specific thyrotropin reference limits. Thyroid. 2011;21(1):5-11. 22. Laurberg P, Andersen S, Carle A, Karmisholt J, Knudsen N, Pedersen IB. The TSH upper reference limit: where are we at? Nat Rev Endocrinol. 2011;7(4):232-9. 23. Brunn J, Block U, Ruf G, Bos I, Kunze WP, Scriba PC. Volumetric analysis of thyroid lobes by real-time ultrasound (author’s transl). Dtsch Med Wochenschr. 1981;106(41):1338-40. 24. Tramalloni J, Monpeyssen H. Tireóide normal e variantes. Variantes anatômicas. In: Tramalloni J, Monpeyssen H. Ultra-Sonografia da Tireóide. Rio de Janeiro, RJ: Ed. Revinder, 2009. p. 23-8.

27. Bensenor IM, Olmos RD, Lotufo PA. Hypothyroidism in the elderly: diagnosis and management. Clin Interv Aging. 2012;7:97-111. 28. Somwaru LL, Rariy CM, Arnold AM, Cappola AR. The natural history of subclinical hypothyroidism in the elderly: the cardiovascular health study. J Clin Endocrinol Metab. 2012;97(6):1962-9. 29. de Jongh RT, Lips P, van Schoor NM, Rijs KJ, Deeg DJ, Comijs HC, et al. Endogenous subclinical thyroid disorders, physical and cognitive function, depression, and mortality in older individuals. Eur J Endocrinol. 2011;165(4):545-54. 30. Duarte GC, Tomimori EK, Camargo RY, Rubio IG, Wajngarten M, Rodrigues AG, et al. The prevalence of thyroid dysfunction in elderly cardiology patients with mild excessive iodine intake in the urban area of Sao Paulo. Clinics (Sao Paulo). 2009;64(2):135-42. 31. Camargo RY, Tomimori EK, Neves SC, Knobel M, Medeiros-Neto G. Prevalence of chronic autoimmune thyroiditis in the urban area neighboring a petrochemical complex and a control area in Sao Paulo, Brazil. Clinics (Sao Paulo). 2006;61(4):307-12. 32. Wesche MF, Wiersinga WM, Smits NJ. Lean body mass as a determinant of thyroid size. Clin Endocrinol (Oxf). 1998;48(6):701-6. 33. Gomez JM, Maravall FJ, Gomez N, Guma A, Soler J. Determinants of thyroid volume as measured by ultrasonography in healthy adults randomly selected. Clin Endocrinol (Oxf). 2000;53(5): 629-34. 34. Turcios S, Lence-Anta JJ, Santana JL, Pereda CM, Velasco M, Chappe M, et al. Thyroid volume and its relation to anthropometric measures in a healthy Cuban population. Eur Thyroid J. 2015;4(1):55-61. 35. Miranda DM, Massom JN, Catarino RM, Santos RT, Toyoda SS, Marone MM, et al. Impact of nutritional iodine optimization on rates of thyroid hypoechogenicity and autoimmune thyroiditis: a cross-sectional, comparative study. Thyroid. 2015;25(1):118-24. 36. Willms A, Bieler D, Wieler H, Willms D, Kaiser KP, Schwab R. Correlation between sonography and antibody activity in patients with Hashimoto thyroiditis. J Ultrasound Med. 2013;32(11): 1979-86. 37. Popoveniuc G, Jonklaas J. Thyroid nodules. Med Clin North Am. 2012;96(2):329-49. 38. Peeters RP. Thyroid function and longevity: new insights into an old dilemma. J Clin Endocrinol Metab. 2009;94(12):4658-60.

25. Gesing A, Lewinski A, Karbownik-Lewinska M. The thyroid gland and the process of aging; what is new? Thyroid Res. 2012;5(1):16.

26. Bremner AP, Feddema P, Leedman PJ, Brown SJ, Beilby JP, Lim EM, et al. Age-related changes in thyroid function: a longitudinal study of a community-based cohort. J Clin Endocrinol Metab. 2012;97(5):1554-62.

Arch Endocrinol Metab. 2017;61/3

275

original article

Can FIB4 and NAFLD fibrosis scores help endocrinologists refer patients with non-alcoholic fat liver disease to a hepatologist? Rodrigo Bremer Nones1, Cláudia Pontes Ivantes1, Maria Lucia Alves Pedroso2

1 Serviço de Gastroenterologia, Hospital Nossa Senhora das Graças, Curitiba, PR, Brasil 2 Unidade de Clínica Médica, Hospital de Clínicas, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brasil

Correspondence to: Maria Lucia Alves Pedroso Universidade Federal do Paraná, Hospital de Clínicas, Unidade de Clínica Médica, Curitiba, PR, Brasil Av. Manoel Ribas, 985, sala 64 80810-000 – Curitiba, PR, Brasil malu.ap@uol.com.br Received on March/18/2016 Accepted on Sep/26/2016 DOI: 10.1590/2359-3997000000233

ABSTRACT Objective: The objective of this study is to evaluate the performance of mathematical models used in non-invasive diagnosis of liver fibrosis in nonalcoholic fatty liver disease (NAFLD) patients to determine when the patient needs to be referred to a hepatologist. Subjects and methods: Patients referred by endocrinologists to the liver outpatient departments in two hospitals in Curitiba, Brazil, over a 72-month period were analyzed. The results calculated using the APRI, FIB 4, FORNS and NAFLD Fibrosis Score non-invasive liver fibrosis assessment models were analyzed and compared with histological staging of this population. Results: Sixty-seven patients with NAFLD were analyzed. Forty-two of them (62.68%) were female, mean age was 54.76 (±9.63) years, mean body mass index 31.42 (±5.64) and 59 (88.05%) of the 67 cases had glucose intolerance or diabetes. A diagnosis of steatohepatitis was made in 45 (76.27%) of the 59 biopsied patients, and advanced liver fibrosis (stages 3 and 4) was diagnosed in 18 (26.86%) of the 67 patients in the study population. The FIB 4 and NAFLD Fibrosis Score models had a high negative predictive value (93.48% and 93.61%, respectively) in patients with severe liver fibrosis (stages 3 and 4). Conclusion: In conclusion, use of the FIB 4 and NAFLD Fibrosis Score models in NAFLD patients allows a diagnosis of severe liver disease to be excluded. Arch Endocrinol Metab. 2017;61(3):276-81. Keywords NAFLD; nonalcoholic steatohepatitis; mathematical model; liver fibrosis; specialist

INTRODUCTION

onalcoholic fatty liver disease (NAFLD) is of considerable interest to endocrinologists because of the high prevalence of this condition in diabetic and obese patients (1). The prevalence of NAFLD in the western population is estimated to be 30% (2), a figure similar to that reported in Brazilian epidemiologic studies (3,4). The condition is defined as fat deposits in the liver (hepatic steatosis) similar to those found in alcohol abusers but in patients who neither consume significant amounts of alcohol nor use other substances that are a secondary cause of steatosis (5,6). NAFLD is classified as simple hepatic steatosis or nonalcoholic steatohepatitis (NASH). While the former accounts for the majority of cases and has a benign course (2,7,8), the latter affects 10% of patients and is characterized by steatosis accompanied by signs of cell injury (hepatocellular ballooning) and liver inflammation. In 20% of these cases, it can progress to cirrhosis and hepatocellular carcinoma (9). Distinguishing between the two conditions is a 276

major challenge as patients are usually asymptomatic (10-12) with normal liver enzyme levels, and imaging tests can fail to identify the steatosis (2,7). A liver biopsy is the only gold-standard diagnostic test for NASH (9,13). However, routine biopsies are not risk-free and occasionally cannot be performed with an adequate sample size. Moreover, interobserver agreement for evaluation of histological criteria of NASH may be low (14,15). In recent years, there has been a search for non-invasive diagnostic methods to assess liver damage, i.e., methods for identifying liver fibrosis that can indicate possible development of advanced liver fibrosis or even cirrhosis without the need for a liver biopsy. These include a) laboratory tests used in mathematical models or diagnostic algorithms, such as the ELF panel, FibroMeter, FibroTest, NAFLD Fibrosis Score, FIB 4, FORNS and BARD (16), and (b) imaging tests, such as elastography, which assesses the elasticity of the liver (17). Non-invasive methods allow examinations to be performed in sequence to assess the course of the disease (7). However, there is a dearth of studies investigating Arch Endocrinol Metab. 2017;61/3

When refer NAFLD patient to a hepatologist?

SUBJECTS AND METHODS The study population consisted of patients seen at the liver outpatient departments in two hospitals in Curitiba, Brazil (Hospital de Clínicas da Universidade Federal do Paraná and Hospital Nossa Senhora das Graças) between March 2005 and January 2011, after referral by endocrinologists. Patients with an echographic diagnosis of hepatic steatosis who agreed to have a percutaneous liver biopsy during this period were selected. Patients with liver cirrhosis secondary to NAFLD, in whom the diagnosis was based on clinical, endoscopic and/or echographic findings and who had metabolic syndrome, were also included. Other etiologies of liver disease, including alcohol abuse, hepatitis B and C infections, autoimmune hepatitis, hereditary hemochromatosis, α-1 antitrypsin deficiency, Wilson’s disease, primary biliary cirrhosis and primary sclerosing cholangitis were excluded in all the patients in the study. It was established in direct patient interviews that none of the patients selected had a history of hepatic steatosis-inducing drug use or alcohol consumption in excess of 20 g per day. Anthropometric data (weight, height and waist circumference) were collected, and body mass index [weight in kg / (height in m)2] (BMI) was calculated. Overweight and obesity were defined as a BMI ≥ 25 and 30, respectively. Diagnosis of glucose intolerance and diabetes followed the American Diabetes Association criteria (18), while diagnosis of metabolic syndrome was based on the NECP ATP III guidelines (19). Of the various non-invasive models for evaluating liver fibrosis, several that could be easily performed using simple demographic and laboratory data and were part of routine patient follow-up were selected. The models used were APRI ([AST level / upper limit of normal AST] x 100 / platelet count (109/L)) (20), FIB 4 (age x AST / [platelet count (109/L) x (ALT)1/2]) (21), FORNS (7.811 – 3.131 x ln [platelet count (109/L)] + 0.78 x ln [GGT] + 4.367 x ln [age] – 0.014 x [total cholesterol]) (22) and NAFLD score (-1.675 + 0.037 x age + 0.094 x BMI (kg/m2) + 1.13 x diabetes/ glucose intolerance [yes = 1, no = 0] + 0.99 x ALT/ AST – 0.013 x platelet count (109/L) – 0.66 x albumin (g/dL)) (23). Values for all the models were calculated for all the patients selected, apart from those for whom Arch Endocrinol Metab. 2017;61/3

laboratory data were not available. In the case of patients who did not have a biopsy, the results of laboratory tests at the time of the echographic examination were used. The liver biopsies were always examined by the same pathologist. The criterion for diagnosis of steatohepatitis was the concomitant presence of hepatic steatosis, hepatocellular ballooning and lobular inflammation. At the same time, the presence and extent of liver fibrosis were also evaluated and classified as follows: stage 1, zone 3 perisinusoidal fibrosis; stage 2, portal fibrosis in addition to stage 1; stage 3, bridging fibrosis in addition to stage 2; stage 4, cirrhosis (13). The results obtained using the non-invasive models were compared with the findings of histological staging of the study population. For each model an ROC curve was fitted, the optimal cut-off point (best sensitivity and specificity) was estimated and the area under the curve (AUROC), sensitivity and specificity were calculated with a 95% confidence interval. The negative predictive value (NPV) and positive predictive value (PPV) were calculated using the data for the prevalence of liver fibrosis in each of the groups in the study population. A significance level of p < 0.05 was used. The study protocol was approved by the Ethics Committee at the Hospital de Clínicas, Universidade Federal do Paraná.

RESULTS In all, 195 patients were evaluated and 67 selected. Of these, 59 had a liver biopsy and 8 were diagnosed with cirrhosis based on the clinical, endoscopic and/or echographic findings. Mild or no fibrosis was present in 55% of patients, stage 2 or higher fibrosis in 45% and stage 3 or 4 fibrosis in 27%. Of the patients biopsied, 45 (76.3%) were diagnosed with steatohepatitis (p < 0.0001). Table 1 summarizes the demographic, laboratory and histological data and the results obtained using the non-invasive models to evaluate fibrosis in all the patients. The laboratory data available was used in the APRI, FIB 4, FORNS and NAFLD Score mathematical models. All the results for the non-invasive models were higher in patients with significant fibrosis. Table 2 shows the performance of each of the models for patients with stage 2 or higher liver fibrosis. The estimated AUROC as well as the cut-off points and positive and negative predictive values are also shown. Figure 1 shows the ROC curves for the non-invasive models for stage 2 or higher liver fibrosis. The best diagnostic accuracy was achieved with the FIB 4 model (AUROC = 0.83). 277

the performance of these methods in Brazilian NAFLD patients. This study therefore sought to evaluate the results of non-invasive laboratory tests for diagnosing liver fibrosis in patients with NAFLD.

When refer NAFLD patient to a hepatologist?

Sensitivity and specificity varied between 50.0% and 68.2% and 79.3% and 94.6%, respectively. The best specificity was achieved with the FIB 4 model. The PPV of this model (90.47%) was also better than that of any of the other models. The NPVs for all the models were similar and varied between 68.0% and 76.1%.

NAFLD Score 100

Sensitivity

Age (years) (n = 67)

54.76 ± 9.63

Females (n = 67)

42 / 67 (62.68%)

p = 0.0498

BMI > 25 (n = 67)

57 / 67 (85.06%)

p < 0.0001

100 80

59 / 67 (88.05%)

p < 0.0001

Metabolic syndrome (n = 59)

46 / 59 (68.65%)

p = 0.0031

Glucose (mg/dL) (n = 65)

134.02 ± 49.82

Total cholesterol (mg/dL) (n = 59)

192.37 ± 53.31

Platelets (10 /L) (n = 67)

227.94 ± 82.50

AST (UI/L) (n = 67)

42.48 ± 30.45

ALT (UI/L) (n = 67)

54.61 ± 41.34

GGT (mg/dL) (n = 58)

40 60 100-Specificity

100

60 40

Sensitivity: 63.3 Specificity: 94.6 Criterion: > 1,7432

127.74 ± 205.09

Albumin (g/dL) (n = 56)

FIB 4

Sensitivity

102.99 ± 12.88

Glucose intolerance or DM (n = 67)

4.46 ± 0.61

100

100-Specificity

Size of the liver biopsy (cm) (n = 59)

2.11 ± 1.00

Portal spaces analyzed (n = 59)

13.56 ± 5.09

AUROC: area under the ROC curve; PPV: positive predictive value; NPV: negative predictive value; CI: confidence interval.

0 and 1

37 (55.22%)

12 (17.91%)

Figure 1. ROC curves NAFLD Score and FIB 4 models in patients in stage 2 or higher liver fibrosis.

Liver fibrosis (n = 67)

4 (5.97%)

14 (20.89%)

APRI (n = 67)

0.57 ± 0.54

FIB4 (n = 67)

1.72 ± 1.43

FORNS (n = 51)

5.17 ± 1.90

NAFLD Score (n = 51)

-1.05 ± 1.63

Table 3. NPV of the different non-invasive models for evaluation of liver fibrosis in patients with fibrosis stage 3 and 4

Table 2. Comparison of the performance of the different non-invasive models for evaluating liver fibrosis in patients with stage 2, 3 or 4 fibrosis APRI

Level of significance

Waist circumference (cm) (n = 67)

Sensitivity: 50.0 Specificity: 86.2 Criterion: >-0.054

Table 1. Demographic, laboratory and histological data and the results obtained using the non-invasive models to evaluate liver fibrosis in all the patients Results

FIB 4

FORNS

NPV

Patients that could avoid a liver biopsy

Estimated false negative

APRI

0,4467

90,48

32/67 (52,23%)

3 (9,52%)

FIB 4

1,7432

93,48

45/67 (67,16%)

3 (6,52%)

FORNS

6,6024

89,73

37/51 (72,54%)

4 (10,27%)

NAFLD Score

-0,037

93,61

36/51 (70,58%)

2 (6,39%)

NAFLD Score

AUROC

0.705

0.830

0.765

0.674

95% CI

0.58-0.81

0.718-0.910

0.625-0.872

0.525-0.823

p value

0.002

0.0001

0.035

Cut-off

0.518

1.7432

5.3097

-0.054

Sensitivity (%)

50.00

63.33

68.18

50.00

Specificity (%)

89.19

94.59

79.31

86.21

PPV

78.94

90.47

72.76

74.61

NPV

68.76

76.09

75.46

68.02

AUROC: area under the ROC curve; PPV: positive predictive value; NPV: negative predictive value; CI: confidence interval. 278

Cut-off

Table 4. NPV of the different non-invasive models for evaluation of liver fibrosis in patients with fibrosis stage 2, 3 and 4 Cut-off

NPV

Patients that could avoid a liver biopsy

Estimated false negative

APRI

0,518

78,94

26/67 (38,80%)

14 (21,06%)

FIB 4

1,7432

90,47

22/67 (32,83%)

6 (9,53%)

FORNS

5,3097

72,76

21/51 (41,17%)

14 (27,24%)

NAFLD Score

-0,054

74,61

16/51 (31,37%)

13 (25,39%)

Arch Endocrinol Metab. 2017;61/3

DISCUSSION In addition to being one of the main current causes of liver disease, NAFLD can be expected to be the main liver disease in the future, given the increasing prevalence of obesity and diabetes in the adult and pediatric populations (1,24). Diagnosis of this condition will therefore be more important for endocrinologists and for public health systems (6) as it may increase direct and indirect health costs (7), generate referrals to specialists (12) and alter morbidity and mortality of patients, due to increased risk of cardiovascular events (25,26), and progression of the liver disease (27). Hepatocellular carcinoma rates are expected to increase in the future (28,29), and NAFLD is predicted to be the main reason for liver transplants in 2020 (27). Differentiating between simple hepatic steatosis and steatohepatitis and, in particular, diagnosing the presence of significant liver fibrosis in NAFLD patients is of enormous importance for prognosis of the disease. Although liver biopsy is considered the gold standard for diagnosing and staging patients with possible NAFLD, its routine use is questionable in overweight and obese individuals, in whom the procedure may be technically more difficult and there is a higher risk of the liver fragment not being suitable for analysis (15,17,30-32). Furthermore, the benign progression of NAFLD in most individuals with this condition and the lack of effective treatment for steatohepatitis makes patients reluctant to undergo a biopsy (8,9,11,33). In addition, liver biopsies are not readily available to most of the population in Brazil, who depend on the public health service (34). These difficulties were also encountered in the present study. Assessment of the different clinical parameters analyzed in this population, such as gender, BMI, waist circumference, glucose intolerance, diabetes and metabolic syndrome, failed to identify the presence of liver fibrosis or steatohepatitis. These parameters were found in similar proportions in all the groups, as reported in other studies (35). Only patient age varied significantly, as patients with more advanced disease stage were older. This finding could be explained by NAFLD having a longer course and progressing silently with greater distortion of normal liver architecture in this population. Furthermore, the use of simple laboratory parameters, such as AST and ALT levels, did not help stage NAFLD in the patients evaluated, as has already been described in other populations and studies (16,36-40). Of the 45 Arch Endocrinol Metab. 2017;61/3

patients with a histological diagnosis of steatohepatitis, 17 (37.77%) had normal levels of both transaminases, and of the 14 cirrhotic patients analyzed, four also had normal transaminase levels (28.57%). Mathematical models based on simple demographic and laboratory data are cheap, practical, easy to reproduce and allow liver fibrosis stages to be determined non-invasively in NAFLD patients. Nevertheless, in this study, these models had high NPVs in patients with advanced liver fibrosis or cirrhosis (liver fibrosis stage 3 and 4). The best NPVs were observed for the FIB 4 and NAFLD Score models (93.48% and 93.61%, respectively) using cut-off values low of 1.743 and -0.037, respectively. In other words, when these tests are carried out, 93% of cases without advanced liver fibrosis or cirrhosis would be identified, as shown in Table 3. Similar performances for these two models were already reported in other populations (16,31,39,41-44) and despite their varied cut-off levels, the bulk of evidence gathered highlights the ability of these tests to indicate reliably the absence of advanced fibrosis. These indirect markers models have a high NPV, so that liver biopsies can be indicated only in cases in which there is diagnostic uncertainty about the severity of the disease. In contrast, the diagnostic performance of the models analyzed in patients with moderate or advanced liver fibrosis or cirrhosis (liver fibrosis stage 2, 3 and 4) was not uniform. The other models could not be used for this purpose as their performance was inferior. This study has shown that a greater understanding of this subject is required. Further research should therefore be undertaken with larger study populations and, if possible, the mathematical models should be used in association with other methods for non-invasive evaluation of fibrosis, such as elastography, as proposed by other authors (16,27). Another question that remains to be elucidated is what advantages the sequential use of these markers to monitor the progress of NAFLD in these patients may offer. In conclusion, in this Brazilian population of NAFLD patients, referred by endocrinologists, the FIB 4 and NAFLD score mathematical models used were able to identify which patients had the greater likelihood of not having advanced fibrosis or cirrhosis. Further studies with larger populations and more cirrhotic patients should be carried out so that the findings can be compared with the results of this study. Disclosure: no potential conflict of interest relevant to this article was reported. 279

When refer NAFLD patient to a hepatologist?

REFERENCES 1. Wree A, Kahraman A, Gerken G, Canbay A. Obesity affects the liver - the link between adipocytes and hepatocytes. Digestion. 2011;83(1-2):124-33. 2. Bellentani S, Scaglioni F, Marino M, Bedogni G. Epidemiology of non-alcoholic fatty liver disease. Dig Dis. 2010;28(1):155-61. 3. Cotrim HP, Parise ER, Oliveira CP, Leite N, Martinelli A, Galizzi J, S, et al. Nonalcoholic fatty liver disease in Brazil. Clinical and histological profile. Ann Hepatol. 2011;10(1):33-7. 4. IBGE. POF 2008-2009 - Antropometria e estado nutricional de crianças, adolescentes e adultos no Brasil. Comunicação Social do Instituto Brasileiro de Geografia e Estatística. Available at: http://www.ibge.gov.br/home/presidencia/noticias/noticia_ visualiza.php?id_noticia=1699&id_pagina=1. Published 2010. Accessed on: Mar. 11, 2012. 5. Lonardo A, Lombardini S, Ricchi M, Scaglioni F, Loria P. Review article: hepatic steatosis and insulin resistance. Aliment Pharmacol Ther. 2005;22 Suppl 2:64-70. 6. Loria P, Lonardo A, Carulli N. Should nonalcoholic fatty liver disease be renamed? Dig Dis. 2005;23(1):72-82. 7. Ratziu V, Bellentani S, Cortez-Pinto H, Day C, Marchesini G. A position statement on NAFLD/NASH based on the EASL 2009 special conference. J Hepatol. 2010;53(2):372-84. 8. Preiss D, Sattar N. Non-alcoholic fatty liver disease: an overview of prevalence, diagnosis, pathogenesis and treatment considerations. Clin Sci (Lond). 2008;115(5):141-50. 9. Brunt EM. Pathology of nonalcoholic fatty liver disease. Nat Rev Gastroenterol Hepatol. 2010;7(4):195-203. 10. Vuppalanchi R, Chalasani N. Nonalcoholic fatty liver disease and nonalcoholic steatohepatitis: selected practical issues in their evaluation and management. Hepatology. 2009;49(1):306-17. 11. Bayard M, Holt J, Boroughs E. Nonalcoholic fatty liver disease. Am Fam Physician. 2006;73(11):1961-8. 12. Farrell GC, Larter CZ. Nonalcoholic fatty liver disease: from steatosis to cirrhosis. Hepatology. 2006;43(2 Suppl 1):S99-112. 13. Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology. 2005;41(6):1313-21. 14. Takahashi Y, Fukusato T. Histopathology of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. World J Gastroenterol. 2014;20(42):15539-48. 15. Sumida Y, Nakajima A, Itoh Y. Limitations of liver biopsy and noninvasive diagnostic tests for the diagnosis of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. World J Gastroenterol. 2014;20(2):475-85.

16. Dowman JK, Tomlinson JW, Newsome PN. Systematic review: the diagnosis and staging of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis. Aliment Pharmacol Ther. 2011;33(5):525-40.

21. Vallet-Pichard A, Mallet V, Nalpas B, Verkarre V, Nalpas A, Dhalluin-Venier V, et al. FIB-4: an inexpensive and accurate marker of fibrosis in HCV infection. Comparison with liver biopsy and fibrotest. Hepatology. 2007;46(1):32-6. 22. Forns X, Ampurdanès S, Llovet JM, Aponte J, Quintó L, MartínezBauer E, et al. Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology. 2002;36(4 Pt 1):986-92. 23. Angulo P, Hui JM, Marchesini G, Bugianesi E, George J, Farrell GC, et al. The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology. 2007;45(4):846-54. 24. Widhalm K, Ghods E. Nonalcoholic fatty liver disease: a challenge for pediatricians. Int J Obes (Lond). 2010;34(10):1451-67. 25. Hu KC, Wang HY, Liu SC, Liu CC, Hung CL, Bair MJ, et al. Nonalcoholic fatty liver disease: updates in noninvasive diagnosis and correlation with cardiovascular disease. World J Gastroenterol. 2014;20(24):7718-29. 26. Arulanandan A, Ang B, Bettencourt R, Hooker J, Behling C, Lin GY, et al. Association Between Quantity of Liver Fat and Cardiovascular Risk in Patients With Nonalcoholic Fatty Liver Disease Independent of Nonalcoholic Steatohepatitis. Clin Gastroenterol Hepatol. 2015;13(8):1513-20.e1. 27. Musso G, Gambino R, Cassader M, Pagano G. Meta-analysis: natural history of non-alcoholic fatty liver disease (NAFLD) and diagnostic accuracy of non-invasive tests for liver disease severity. Ann Med. 2011;43(8):617-49. 28. Starley BQ, Calcagno CJ, Harrison SA. Nonalcoholic fatty liver disease and hepatocellular carcinoma: a weighty connection. Hepatology. 2010;51(5):1820-32. 29. Sun B, Karin M. Obesity, inflammation, and liver cancer. J Hepatol. 2012;56(3):704-13. 30. Byrne CD, Olufadi R, Bruce KD, Cagampang FR, Ahmed MH. Metabolic disturbances in non-alcoholic fatty liver disease. Clin Sci (Lond). 2009;116(7):539-64. 31. Shah AG, Lydecker A, Murray K, Tetri BN, Contos MJ, Sanyal AJ. Comparison of noninvasive markers of fibrosis in patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol. 2009;7(10):1104-12. 32. Estep JM, Birerdinc A, Younossi Z. Non-invasive diagnostic tests for non-alcoholic fatty liver disease. Curr Mol Med. 2010;10(2):166-72. 33. Krawczyk M, Bonfrate L, Portincasa P. Nonalcoholic fatty liver disease. Best Pract Res Clin Gastroenterol. 2010;24(5):695-708. 34. Lembrança L, Medina J, Portugal M, Almeida D, Solla J, Gadelha R, et al. Epidemiological assessment of liver disease in northeastern Brazil by means of a standardized liver biopsy protocol. Ann Hepatol. 2011;10(1):43-9. 35. Uslusoy HS, Nak SG, Gülten M. Noninvasive predictors for liver fibrosis in patients with nonalcoholic steatohepatitis. World J Hepatol. 2011;3(8):219-27.

17. Clark PJ, Patel K. Noninvasive tools to assess liver disease. Curr Opin Gastroenterol. 2011;27(3):210-6.

36. Uslusoy HS. Non-alcoholic steatohepatitis with normal aminotransferase values. World J Gastroenterol. 2009;15(15):1863-8.

18. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2012;35 Suppl 1:S64-71.

37. Fierbinteanu-Braticevici C. Noninvasive investigations for non alcoholic fatty liver disease and liver fibrosis. World J Gastroenterol. 2010;16(38):4784-91.

19. 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. 20. Wai CT, Greenson JK, Fontana RJ, Kalbfleisch JD, Marrero JA, Conjeevaram HS, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology. 2003;38(2):518-26.

280

38. Neuschwander-Tetri BA, Clark JM, Bass NM, Van Natta ML, UnalpArida A, Tonascia J, et al. Clinical, laboratory and histological associations in adults with nonalcoholic fatty liver disease. Hepatology. 2010;52(3):913-24. 39. Sumida Y, Yoneda M, Hyogo H, Itoh Y, Ono M, Fujii H, et al. Validation of the FIB4 index in a Japanese nonalcoholic fatty liver disease population. BMC Gastroenterol. 2012;12(1):2. 40. Yoneda M, Imajo K, Eguchi Y, Fujii H, Sumida Y, Hyogo H, et al. Noninvasive scoring systems in patients with nonalcoholic fatty Arch Endocrinol Metab. 2017;61/3

When refer NAFLD patient to a hepatologist?

advanced fibrosis in patients with non-alcoholic fatty liver disease. Gut. 2010;59(9):1265-9.

41. Shah AG, Lydecker A, Murray K, Tetri BN, Contos MJ, Sanyal AJ; Nash Clinical Research Network. Use of the Fib4 Index for Non-Invasive Evaluation of Fibrosis in Nonalcoholic Fatty Liver Disease. Clin Gastroenterol Hepatol. 2009;7(10):1104-12.

43. Adams LA, George J, Bugianesi E, Rossi E, De Boer WB, van der Poorten D, et al. Complex non-invasive fibrosis models are more accurate than simple models in non-alcoholic fatty liver disease. J Gastroenterol Hepatol. 2011;26(10):1536-43.

42. McPherson S, Stewart SF, Henderson E, Burt AD, Day CP. Simple non-invasive fibrosis scoring systems can reliably exclude

44. Cheung O, Sanyal AJ. Recent advances in nonalcoholic fatty liver disease. Curr Opin Gastroenterol. 2010;26(3):202-8.

liver disease with normal alanine aminotransferase levels. J Gastroenterol. 2013;48(9):1051-60.

Arch Endocrinol Metab. 2017;61/3

281

original article

Visceral adiposity index and triglyceride/high-density lipoprotein cholesterol ratio in hypogonadism Cem Haymana1, Alper Sonmez1, Aydogan Aydogdu1, Serkan Tapan2, Yalcin Basaran1, Coskun Meric1, Kamil Baskoy3, Mustafa Dinc1, Mahmut Yazici1, Abdullah Taslipinar1, Cem Barcin4, Mahmut Ilker Yilmaz5, Erol Bolu6, Omer Azal1

Department of Endocrinology and Metabolism, Gulhane School of Medicine, Ankara, Turkey 2 Department of Biochemistry, Gulhane School of Medicine, Ankara, Turkey 3 Department of Endocrinology and Metabolism, Gulhane School of Medicine, Haydarpasa Training Hospital, Istanbul, Turkey 4 Department of Cardiology, Gulhane School of Medicine, Ankara, Turkey 5 Department of Nephrology, Gulhane School of Medicine, Ankara, Turkey 6 Department of Endocrinology and Metabolism, Memorial Atasehir Hospital, Istanbul, Turkey 1

Correspondence to: Cem Haymana Department of Endocrinology and Metabolism, Gulhane School of Medicine Tevfik Saglam Street, 06010, Etlik-Ankara, Turkey cemhaymana@hotmail.com Received on Jan/21/2016 Accepted on Oct/10/2016

ABSTRACT Background: Cardiometabolic risk is high in patients with hypogonadism. Visceral adiposity index (VAI) and triglyceride/high-density lipoprotein cholesterol (TG/HDL-C) ratio are the practical markers of atherosclerosis and insulin resistance and independent predictors of cardiaovascular risk. To date, no study has evaluated VAI levels and TG/HDL-C ratio in hypogonadism. Subjects and methods: A total of 112 patients with congenital hypogonadotrophic hypogonadism (CHH) (mean age, 21.7 ± 2.06 years) and 124 healthy subjects (mean age, 21.5 ± 1.27 years) were enrolled. The demographic parameters, VAI, TG/HDL-C ratio, asymmetric dimethylarginine (ADMA), high-sensitivity C-reactive protein (hs-CRP), and homeostatic model assessment of insulin resistance (HOMA-IR) levels were measured for all participants. Results: The patients had higher total cholesterol (p = 0.04), waist circumference, triglycerides, insulin, and HOMA-IR levels (p = 0.001 for all) than the healthy subjects. VAI and ADMA and TG/HDL-C levels were also higher in patients than in healthy subjects (p < 0.001 for all). VAI was weakly correlated with ADMA (r = 0.27, p = 0.015), HOMA-IR (r = 0.22, p = 0.006), hs-CRP (r = 0.19, p = 0.04), and total testosterone (r = −0.21, p = 0.009) levels, whereas TG/HDL-C ratio was weakly correlated weakly with ADMA (r = 0.30, p = 0.003), HOMA-IR (r = 0.22, p = 0.006), and total testosterone (r = −0.16, p = 0.03) levels. Neither VAI nor TG/HDL-C ratio determined ADMA, HOMA-IR, and hs-CRP levels. Conclusions: The results of this study demonstrate that patients with hypogonadism have elevated VAI and TG/HDL-C ratio. These values are significantly correlated with the surrogate markers of endothelial dysfunction, inflammation, and insulin resistance. However, the predictive roles of VAI and TG/HDL-C ratio are not significant. Prospective follow-up studies are warranted to clarify the role of VAI and TG/HDL-C ratio in predicting cardiometabolic risk in patients with hypogonadism. Arch Endocrinol Metab. 2017;61(3):282-7. Keywords Hypogonadism; cardiometabolic risk; visceral adiposity index; triglyceride/high-density lipoprotein cholesterol ratio

DOI: 10.1590/2359-3997000000236

INTRODUCTION

ypogonadism is a syndrome characterized by low testosterone levels and a clinical spectrum of poor libido, energy loss, muscle atrophy, and depression. In addition to fertility disturbance, cardiometabolic risk is increased in patients with hypogonadism (1,2). The prevalence of cardiac and metabolic disorders, such as type 2 diabetes mellitus, hypertension, dyslipidemia, and obesity, are significantly increased in these patients (2,3). However, the mechanism by which cardiometabolic risk increases in patients with hypogonadism remains to be completely elucidated. Inflammation, insulin resistance, and endothelial dysfunction are the major contributors to increased cardiometabolic risk in hypogonadism (4-8). In our

282

previous studies, we observed that even young patients with hypogonadism exhibit endothelial dysfunction, inflammation, and insulin resistance (9-11). However, none of the surrogate markers of endothelial dysfunction, inflammation, and insulin resistance are sufficiently robust to be used as prognostic tools. Thus, a simple, widely available, relatively inexpensive, and generally reproducible marker to predict cardiometabolic risk in patients with hypogonadism is needed. Visceral adiposity index (VAI) is a mathematical model based on simple anthropometric [body mass index (BMI) and waist circumference (WC)] and metabolic [triglycerides (TG) and high-density lipoprotein cholesterol (HDL-C)] parameters and is considered as a simple surrogate marker of visceral adipose dysfunction (12). VAI is strongly associated with Arch Endocrinol Metab. 2017;61/3

VAI and TG/HDL ratio in hypogonadism

SUBJECTS AND METHODS This retrospective analysis was performed by evaluating the database of the Department of Endocrinology and Metabolism, Gulhane Military Medical Academy School of Medicine, Ankara, Turkey. Military service is compulsory for all young men in Turkey, and Gulhane Military Medical Academy School of Medicine is the tertiary medical center for military recruits. Patients with hypogonadism are referred to the Department of Endocrinology and Metabolism for both treatment and follow-up. Some of these patients, generally those living in rural regions, have never received treatment. A total of 273 young patients with hypogonadism were registered between 2007 and 2012. Of these, 64 patients were excluded because of a history of androgen replacement; 27 because of high testosterone levels (200–300 ng/dL); 16 because of liver, kidney, or pulmonary disease; 22 because of a diagnosis other than hypogonadotropic hypogonadism (CHH; i.e., primary hypogonadism, panhypopituitarism, or pituitary adenoma); and 32 because of incomplete data on demographic and metabolic parameters. A total of 112 treatment-naive patients (mean age, 21.7 ± 2.06 years) with congenital CHH were included. The control group included 124 age- and BMI-matched healthy subjects (mean age, 21.5 ± 1.27 years). Arch Endocrinol Metab. 2017;61/3

A portion of the data for this study population was previously published (9-11). None of the control subjects had a chronic disorder or used any medications, including over-the-counter drugs. All subjects provided informed consent, and the Local Ethical Committee of Gulhane School of Medicine approved the study. This study has been registered with Clinicaltrials.gov (NCT02111434). Detailed medical histories of all patients were obtained before the study. Height, weight, and WC were measured with the subjects in their underwear. BMI was computed as the ratio of weight to the square of height (kg/m2). WC was measured on the line between the iliac crest and the lower costal margin parallel to the ground after subjects exhaled. The pubertal developments of the subjects were assessed according to the Tanner stages. CHH diagnosis was based on a history of failure to undergo spontaneous puberty before 18 years of age and was confirmed with tests demonstrating low serum total testosterone and normal or low gonadotropin levels. Pituitary hormones were evaluated in all patients to exclude panhypopituitarism, and pituitary or hypothalamic mass lesions were excluded by evaluation with magnetic resonance imaging.

Sample collection and laboratory measurements For biochemical analyses, all blood samples were collected from the antecubital veins between 08:00 and 09:00 h after overnight fasting. The samples were centrifuged for 15 min at 4,000 × g, aliquoted, and immediately frozen at −80°C for analyses. Fasting plasma glucose, total cholesterol, TG, and HDL-C levels were measured by the enzymatic colorimetric method using an Olympus AU-2700 autoanalyzer with reagents from Olympus Diagnostics (GmbH, Hamburg, Germany). Low-density lipoprotein cholesterol level was calculated using Friedewald’s formula (22). Serum basal insulin, total testosterone, follicle-stimulating hormone, and luteinizing hormone levels were measured by the chemiluminescence method using a UniCel DxI 800 Access Immunoassay System (Miami, FL, USA). Complete blood count was obtained using the Olympus AU-2700 autoanalyzer (GmbH). Insulin sensitivity was calculated by the homeostatic model assessmentinsulin resistance (HOMA-IR) using the following formula: HOMA-IR = (insulin × glucose)/405 (23). Plasma high-sensitivity C-reactive protein (hs-CRP) levels were determined in 58 patients and 69 control subjects. Plasma asymmetric dimethylarginine (ADMA) 283

visceral adiposity measured using magnetic resonance imaging and cardiovascular and cerebrovascular events (12). Recent data also indicate that hypertriglyceridemia and low HDL-C are key components of the metabolic syndrome and are strongly predictive of coronary artery disease (13,14). The TG/HDL-C ratio is another practical marker of atherosclerosis and insulin resistance and an independent predictor of cardiovascular risk (15-17). The role of TG/HDL-C ratio in predicting cardiometabolic risk has been tested in several metabolic disorders, such as diabetes mellitus, hypertension, chronic kidney disease, and nonalcoholic fatty liver disease (18-21). To date, no studies have investigated VAI and TG/HDL-C ratio in patients with hypogonadism. Therefore, we designed the present study to answer the following questions: 1. Do VAI and TG/HDL-C ratio differ between patients with hypogonadism and healthy subjects? 2. Are insulin resistance, inflammation, and endothelial dysfunction associated with VAI and TG/ HDL-C ratio?

VAI and TG/HDL ratio in hypogonadism

levels were determined in 59 patients and 33 control subjects using the enzyme-linked immunosorbent assay kit (Immundiagnostik, Bensheim, Germany). The minimum detectable concentration of ADMA was 0.05 µmol/L. The hs-CRP level was determined in serum by the immunoturbidimetric fixed rate method using the Olympus AU-2700 autoanalyzer (GmbH). The intra- and inter-assay coefficients of variation were 5.8% and 3.1%, respectively. The minimum detectable concentration of hs-CRP was 0.07 mg/L. VAI was calculated using the following sex-specific equation (13): Males : VAI =

Healthy controls (n = 124)*

WC × 39.68 + (1.88 × BMI)

TG 1.03

21.5 ± 1.27

21.7 ± 2.06

0.31

22.8 ± 2.11

22.17 ± 3.26

0.09

WC (cm)

79.16 ± 6.24

83.61 ± 11.14

0.001

SBP (mmHg)

115 ± 10.7

117.2 ± 13.2

0.451

DBP (mmHg)

68.84 ± 5.88

72.07 ± 8.73

0.093

FBG (mg/dL)

84.08 ± 1.0

85.7 ± 7.62

0.16

HDL-chol (mg/dL) LDL-chol (mg/dL)

160.8 ± 26.4

0.04

89.0 (63.0–130.5)

0.001

51.7 ± 15.0

47.1 ± 10.0

0.007

84.7 ± 30.7

91.6 ± 22.3

0.06

35.3 ± 42.4

< 0.001

Insulin (µU/mL)a

6.59 (4.96–8.7)

8.49 (5.65–12.83)

0.001

HOMA-IRa

1.39 (0.92–1.84)

1.8 (1.18–2.76)

0.001

VAI

2.17 ± 1.17

3.22 ± 2.31

< 0.001

TG/HDL-C

1.69 ± 0.95

2.35 ± 1.5

< 0.001

hs-CRP (mg/L)

0.87 ± 1.21 n = 58

1.18 ± 1.1 n = 69

0.13

ADMA (µmol/L)

0.33 ± 0.6 n = 59

0.66 ± 0.17 n = 33

< 0.001

T. testosterone (ng/mL)

All data were recorded in a computer database and analyzed using the SPSS 18.0 program (SPSS, Inc., Chicago, IL, USA). Results are expressed as means ± standard deviation. The variables were assessed for normality using the Kolmogorov–Smirnov test, and the equality of variance was evaluated using the Levene’s test. Inter-group differences were analyzed using the Student’s t-test and Mann–Whitney U test as appropriate. The correlations were performed using the Pearson’s or Spearman’s correlation tests. Differences were considered significant at a p value of < 0.05.

284

151.1 ± 34.1 74.0 (54.7–97.0)

550.8 ± 127.7

Statistical analysis

The demographic and biochemical characteristics of the patients and control subjects are given in Table 1. Compared with healthy controls, patients had significantly higher total cholesterol levels (p = 0.04); WC, TG, insulin, and HOMA-IR levels (p = 0.001 for all); and ADMA and TG/HDL-C levels and VAI (p < 0.001 for all). VAI were weakly correlated with ADMA (r = 0.27, p = 0.015), HOMA-IR (r = 0.22, p = 0.006), hs-CRP (r = 0.19, p = 0.04; Figure 1A), and total testosterone (r = −0.21, p = 0.009) levels, whereas TG/HDL-C ratio was weakly correlated with ADMA (r = 0.30, p = 0.003), HOMA-IR (r = 0.22, p = 0.006; Figure 1B), and total testosterone (r = −0.16, p = 0.03) levels. In a stepwise linear regression analysis, neither VAI nor TG/HDL ratio remained in the model to be independent determinants of ADMA or HOMA-IR levels. Total testosterone level was the only significant independent determinant of ADMA level, whereas WC was an independent determinant of HOMA-IR level.

BMI (kg/m²)

TG (mg/dL)a

1.31 HDL

Patients (n = 112)*

Age (year)

T. Chol (mg/dL)

( ) ( ) ( )

RESULTS

Table 1. Demographic and metabolic parameters of patients with congenital hypogonadotropic hypogonadism and healthy control subjects in this study

Mann-Whitney U test. Results are given as means (25% – 75%). p: Student’s t-test; comparison of parameters between healthy control subjects and patients. * hs-CRP and ADMA levels are provided separately. ADMA: asymmetric dimethylarginine; BMI: body mass index; DBP: diastolic blood pressure; FBG: fasting blood glucose; HDL-chol: high-density lipoprotein cholesterol; HOMA-IR: homeostatic model assessment-insulin resistance; hs-CRP: high-sensitivity C-reactive protein; LDL-C: low-density lipoprotein cholesterol; SBP: systolic blood pressure; T. chol: total cholesterol; TG: triglycerides; TG/HDL-C: triglycerides/high-density lipoprotein cholesterol ratio; T. testosterone: total testosterone; VAI: visceral adiposity index; WC: waist circumference.

DISCUSSION The results of the present study show that patients with hypogonadism have significantly higher VAI and TG/HDL-C ratio than control subjects. Moreover, VAI and TG/HDL-C ratio are significantly correlated with endothelial dysfunction, insulin resistance, and inflammation (VAI only). However, neither VAI nor TG/HDL-C ratio is applicable as the independent predictor of endothelial dysfunction, inflammation, or insulin resistance in patients with CHH. Cardiovascular and metabolic disorders, such as obesity, dyslipidemia, hypertension, and type 2 diabetes mellitus, are prevalent in patients with hypogonadism (24-27), but the mechanism by which cardiovascular and metabolic risk increase in hypogonadism is unknown. Inflammation, oxidative stress, insulin resistance, and endothelial dysfunction are the Arch Endocrinol Metab. 2017;61/3

fundamental contributors to increased cardiometabolic risk (4-8). In our previous studies, we reported metabolic derangements and endothelial dysfunction, inflammation, and insulin resistance in young and treatment-naĂŻve patients with hypogonadism (9-11). Patients with hypogonadism have increased fat mass and visceral adiposity, and the latter is associated with an increased risk of developing diabetes, hypertension, dyslipidemia, and atherosclerosis (28,29). We hypothesized that VAI, as an applicable marker for evaluating visceral adipose function, is useful for assessing increased cardiometabolic risk in patients with hypogonadism. A 8,00

ADMA hsCRP HOMA-IR

6,00

* 4,00

* *

* HOMA-IR * * r=0.27 p=0.015 * * *** * * * *** * * * * ** * * hsCRP * *** **** ** ** p=0.04 *r=0.19 * * * * * *ADMA * * * ** *** * * *** ** * p=0.006 r=0.22 * *

2,00

0,00 0,00

2,00

4,00

VAI

6,00

8,00

10,00 ADMA hsCRP HOMA-IR

8,00

6,00

4,00 HOMA-IR r=0.22 p=0.006 hsCRP r=0.16 p=0.07

2,00

ADMA r=0.30 p=0.003

0,00 0,00

2,00

4,00 TG/HDL-C

6,00

Figure 1. (A) Scatter plot diagram of the correlation between visceral adiposity index (VAI) and asymmetric dimethylarginine (ADMA), highsensitivity C-reactive protein (hs-CRP), and homeostatic model assessment-insulin resistance (HOMA-IR) levels. (B) Scatter plot diagram of the correlation between triglycerides/high-density lipoprotein cholesterol ratio (TG/HDL-C) and asymmetric dimethylarginine (ADMA), highsensitivity C-reactive protein (hs-CRP), and homeostatic model assessment-insulin resistance (HOMA-IR) levels. Arch Endocrinol Metab. 2017;61/3

VAI has been recently developed as a novel sexspecific index based on WC, BMI, TG, and HDL-C (12). VAI is a marker of visceral adipose dysfunction and is strongly associated with cardiovascular events and type 2 diabetes (12,30,31). VAI is significantly correlated with inflammation and insulin resistance in patients with type 2 diabetes and cardiovascular disorders (12,32). VAI is also a practical measure for assessing cardiometabolic risk in patients with polycystic ovary syndrome (33). TG/HDL-C ratio is another clinical indicator of insulin resistance and has been evaluated as a predictor of diabetes and coronary heart disease (3436). This ratio may serve as a simpler method for identifying insulin-resistant individuals with increased cardiometabolic risk (37). TG/HDL-C ratio is also correlated with endothelial dysfunction. In our previous study, we showed that TG/HDL-C ratio is a significant determinant of endothelial dysfunction and a simple predictor of cardiovascular outcomes in patients with chronic kidney disease (38). Low HDL-C and increased TG levels are also well-established features of patients with hypogonadism (9,39,40). ADMA is an endogenous inhibitor of nitric oxide synthase and a well-known surrogate marker of endothelial dysfunction (41). Elevated ADMA levels in chronic metabolic diseases, such as type 2 diabetes, hypertension, dyslipidemia, and chronic kidney disease predict cardiovascular morbidity and mortality (42). We have previously reported elevated ADMA levels in patients with hypogonadism (10,11), which implies increased endothelial dysfunction in this patient group. Most of the markers, such as ADMA, used to define increased cardiometabolic risk in patients with hypogonadism are time-consuming or costly, which precludes their use in routine daily practice. To our knowledge, the present study is the first to measure VAI and TG/HDL-C ratio in patients with hypogonadism. Our results show that VAI and TG/HDL-C ratio are significantly increased in patients with hypogonadism and are related to markers of inflammation, insulin resistance, and endothelial dysfunction. However, the roles of VAI and TG/HDL-C ratio in predicting endothelial dysfunction, inflammation, and insulin resistance are not sufficiently robust for these parameters to be applicable in clinical practice to predict cardiometabolic risk in patients with hypogonadism. According to the results, total testosterone level and WC 285

VAI and TG/HDL ratio in hypogonadism

are the only independent determinants of endothelial dysfunction and insulin resistance, respectively. This study has both limitations and advantages. The study population comprising young, treatment-naïve patients with CHH may not be representative of the general population of patients with hypogonadism. Our small sample size may be another limitation. However, because few patients with hypogonadism reach adulthood without receiving treatment, we believe that the number of the patients in our study is adequate because of the unique conditions of the study population. The advantages of the present study are its homogeneous study population and the lack of confounding factors, such as chronic metabolic disorders and concomitant medications. In conclusion, the present study shows that patients with hypogonadism have elevated VAI and TG/ HDL-C ratio, which are significantly correlated with the surrogate markers of endothelial dysfunction, inflammation, and insulin resistance. However, the predictive roles of VAI and TG/HDL-C ratio for endothelial dysfunction, inflammation, and insulin resistance are not significant. Prospective follow-up studies are warranted to clarify the role of VAI and TG/HDL-C ratio in determining cardiometabolic risk in patients with hypogonadism.

7. Yialamas MA, Dwyer AA, Hanley E, Lee H, Pitteloud N, Hayes FJ. Acute sex steroid withdrawal reduces insulin sensitivity in healthy men with idiopathic hypogonadotropic hypogonadism. J Clin Endocrinol Metab. 2007;92(11):4254-9.

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

16. Jeppesen J, Hein HO, Suadicani P, Gyntelberg F. Low triglycerideshigh high-density lipoprotein cholesterol and risk of ischemic heart disease. Arch Intern Med. 2001;161(3):361-6.

REFERENCES 1. Traish AM, Saad F, Feeley RJ, Guay A.The dark side of testosterone deficiency: III Cardiovascular disease. J Androl. 2009;30:477-94. 2. Kupelian V, Page ST, Araujo AB, Travison TG, Bremner WJ, McKinlay JB. Low sex hormone-binding globulin, total testosterone, and symptomatic androgen deficiency are associated with development of the metabolic syndrome in nonobese men. J Clin Endocrinol Metab. 2006;91(3):843-50. 3. Stellato RK, Feldman HA, Hamdy O, Horton ES, McKinlay JB. Testosterone, sex hormone-binding globulin, and the development of type 2 diabetes in middle-aged men: prospective results from the Massachusetts male aging study. Diabetes Care. 2000;23(4):490-4.

4. Akishita M, Hashimoto M, Ohike Y, Ogawa S, Iijima K, Eto M, et al. Low testosterone level is an independent determinant of endothelial dysfunction in men. Hypertens Res. 2007;30(11):102934. 5. Laaksonen DE, Niskanen L, Punnonen K, Nyyssönen K, Tuomainen TP, Salonen R, et al. Sex hormones, inflammation and the metabolic syndrome: a population-based study. Eur J Endocrinol. 2003;149(6):601-8. 6. Traish AM, Saad F, Guay AT. The dark side of testosterone deficiency: II. Type 2 diabetes and insulin resistance. J Androl. 2009;30(1):23-32.

286

8. Fu L, Gao QP, Shen JX. Relationship between testosterone and indexes indicating endothelial function in male coronary heart disease patients. Asian J Androl. 2008;10(2):214-8. 9. Sonmez A, Haymana C, Bolu E, Aydogdu A, Tapan S, Serdar M, et al. Metabolic syndrome and the effect of testosterone treatment in young men with congenital hypogonadotropic hypogonadism. Eur J Endocrinol. 2011;164(5):759-64. 10. Meric C, Sonmez A, Aydogdu A, Tapan S, Haymana C, Basaran Y, et al. Osteoprotegerin, fibroblast growth factor 23, and vitamin D3 levels in male patients with hypogonadism. Horm Metab Res. 2014;46(13):955-8. 11. Sonmez A, Haymana C, Aydogdu A, Tapan S, BasaranY, Meric C, et al. Endothelial dysfunction, insulin resistance and inflammation in congenital hypogonadism, and the effect of testosterone replacement. Endocr J. 2015;62(7):605-13. 12. Amato MC, Giordano C, Galia M, Criscimanna A, Vitabile S, Midiri M, et al. Visceral Adiposity Index. Diabetes Care. 2010;33(4):920-2. 13. Carey VJ, Bishop L, Laranjo N, Harshfield BJ, Kwiat C, Sacks FM. Contribution of high plasma triglycerides and low high-density lipoprotein cholesterol to residual risk of coronary heart disease after establishment of low-density lipoprotein cholesterol control. Am J Cardiol. 2010;106(6):757-63. 14. Sarwar N, Danesh J, Eiriksdottir G, Sigurdsson G, Wareham N, Bingham S, et al. Triglycerides and the risk of coronary heart disease: 10158 incident cases among 262525 participants in 29 Western prospective studies. Circulation. 2007;115(4):450-8. 15. Dobiasova M, Frohlich J. The plasma parameter log (TG/HDL-C) as an atherogenic index: correlation with lipoprotein particle size and esterification rate in apoB-lipoprotein-depleted plasma (FERHDL). Clin Biochem. 2001;34(7):583-8.

17. McLaughlin T, Reaven G, Abbasi F, Lamendola C, Saad M, Waters D, et al. Is there a simple way to identify insulin-resistant individuals at increased risk of cardiovascular disease? Am J Cardiol. 2005;96(3):399-404. 18. Vega GL, Barlow CE, Grundy SM, Leonard D, DeFina LF. Triglyceride-to-high-density-lipoprotein-cholesterol ratio is an index of heart disease mortality and of incidence of type 2 diabetes mellitus in men. J Investig Med. 2014;62(2):345-9. 19. Onat A, Can G, Kaya H, Hergenç G. “Atherogeni index of plasma” (log10 triglyceride/high-density lipoprotein-cholesterol) predicts high blood pressure, diabetes, and vascular events. J Clin Lipidol. 2010;4(2):89-98. 20. Sung KC, Ryan MC, Kim BS, Cho YK, Kim BI, Reaven GM. Relationships between estimates of adiposity, insulin resistance, and nonalcoholic fatty liver disease in a large group of nondiabetic Korean adults. Diabetes Care. 2007;30(8):2113-8. 21. Sonmez A, Yilmaz MI, Saglam M, Unal HU, Gok M, Cetinkaya H, et al. The role of plasma triglyceride/high-density lipoprotein cholesterol ratio to predict cardiovascular outcomes in chronic kidney disease. Lipids Health Dis. 2015;14:29. 22. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of lowdensity lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499-502. 23. Wallace TM, Matthews DR. The assessment of insulin resistance in man. Diabet Med. 2002;19(7):527-34. Arch Endocrinol Metab. 2017;61/3

VAI and TG/HDL ratio in hypogonadism

24. Kupelian V, Page ST, Araujo AB, Travison TG, Bremner WJ, McKinlay JB. Low sex hormone-binding globulin, total testosterone, and symptomatic androgen deficiency are associated with development of the metabolic syndrome in nonobese men. J Clin Endocrinol Metab. 2006;91(3):843-50. 25. Laaksonen DE, Niskanen L, Punnonen K, Nyyssonen K, Tuomainen TP, Valkonen VP, et al. Testosterone and sex hormone binding globulin predict the metabolic syndrome and diabetes mellitus in middle-aged men. Diabetes Care. 2004;27(5):1036-41. 26. Caldas AD, Porto AL, Motta LD, Casulari LA. Relationship between insulin and hypogonadism in men with hypogonadism. Arq Bras Endocrinol Metabol. 2009;53(8):1005-11. 27. Stellato RK, Feldman HA, Hamdy O, Horton ES, McKinlay JB. Testosterone, sex hormone-binding globulin, and the development of type 2 diabetes in middle-aged men: prospective results from the Massachusetts Male Aging Study. Diabetes Care. 2000;23(4):490-4. 28. Rader DJ. Effect of insulin resistance, dyslipidemia, and intraabdominal adiposity on the development of cardiovascular disease and diabetes mellitus. Am J Med. 2007;120(3 Suppl 1):S12-8. 29. Despre’s JP. Intra-abdominal obesity: an untreated risk factor for type 2 diabetes and cardiovascular disease. J Endocrinol Invest. 2006;29(3 Suppl):77-82. 30. Chen HY, Chiu YL, Chuang YF, Hsu SP, Pai MF, Yang JY, et al. Visceral adiposity index and risks of cardiovascular events and mortality in prevalent hemodialysis patients. Cardiovasc Diabetol. 2014;13:136. 31. Wang Y, He S, He J, Wang S, Liu K, Chen X. Predictive value of visceral adiposity index for type 2 diabetes mellitus: A 15 year prospective cohort study. Herz. 2015;40 Suppl 3:277-81. 32. Amato MC, Pizzolanti G, Torregrossa V, Misiano G, Milano S, Giordano C. Visceral adiposity index (VAI) is predictive of an altered adipokine profile in patients with type 2 diabetes. PLoS One. 20;9(3):e91969.

34. Gaziano JM, Hennekens CH, O’Donnell CJ, Breslow JL, Buring JE. Fasting triglycerides, high-density lipoprotein, and risk of myocardial infarction. Circulation. 1997;96(8):2520-5. 35. Hadaegh F, Hatami M,Tohidi M, Sarbakhsh P, Saadat N, Azizi F. Lipid ratios and appropriate cut off values for prediction of diabetes: a cohort of Iranian men and women. Lipids Health Dis. 2010;9:85. 36. Kannel WB, Vasan RS, Keyes MJ, Sullivan LM, Robins SJ. Usefulness of the triglyceride-high-density lipoprotein versus the cholesterol-high-density lipoprotein ratio for predicting insulin resistance and cardiometabolic risk (from the Framingham Offspring Cohort). Am J Cardiol. 2008;101(4):497-501. 37. McLaughlin T, Reaven G, Abbasi F, Lamendola C, Saad M, Waters D, et al. Is there a simple way to identify insulin-resistant individuals at increased risk of cardiovascular disease? Am J Cardiol. 2005;96(3):399-404. 38. Sonmez A, Yilmaz MI, Saglam M, Unal HU, Gok M, Cetinkaya H, et al. The role of plasma triglyceride/high-density lipoprotein cholesterol ratio to predict cardiovascular outcomes in chronic kidney disease. Lipids Health Dis. 2015;14:29. 39. Tsai EC, Matsumoto AM, Fujimoto WY, Boyko EJ. Association of bioavailable, free, and total testosterone with insulin resistance: influence of sex hormone-binding globulin and body fat. Diabetes Care. 2004;27(4):861-8. 40. Zmuda JM, Cauley JA, Kriska A, Glynn NW, Gutai JP, Kuller LH. Longitudinal relation between endogenous testosterone and cardiovascular disease risk factors in middle-aged men. A 13year follow-up of former Multiple Risk Factor Intervention Trial participants. Am J Epidemiol. 1997;146(8):609-17. 41. Böger RH, Maas R, Schulze F, Schwedhelm E. Asymmetric dimethylarginine (ADMA) as a prospective marker of cardiovascular disease and mortality--an update on patient populations with a wide range of cardiovascular risk. Pharmacol Res. 2009;60(6):481-7. 42. Chan NN, Chan JC. Asymmetric dimethylarginine (ADMA): a potential link between endothelial dysfunction and cardiovascular diseases in insulin resistance syndrome? Diabetologia. 2002;45(12):1609-16.

33. Amato MC, Verghi M, Galluzzo A, Giordano C.The oligomenorrhoic phenotypes of polycystic ovary syndrome are characterized by a

high visceral adiposity index: a likely condition of cardiometabolic risk. Hum Reprod. 2011;26(6):1486-94.

Arch Endocrinol Metab. 2017;61/3

287

case report

Intrathoracic stomach mimicking bone metastasis from thyroid cancer in whole-body iodine-131 scan diagnosed by SPECT/CT Francisco Javier García-Gómez1, Pablo Antonio de la Riva-Pérez1, Cinta Calvo-Morón1, Cristina Buján-Lloret1, Teresa Cambil-Molina1, Juan Castro-Montaño1

SUMMARY The whole-body iodine-131 scintigraphy is an imaging technique in monitoring patients with a history of thyroid cancer. Although the rate of false positives is negligible, it is not nonexistent. We report the case of an intervened and treated patient for thyroid cancer with good clinical and biochemical response. Scintigraphic findings were consistent with unsuspected bone metastasis. Fused SPECT/CT data allowed accurate diagnosis of giant diaphragmatic hernia associated with intrathoracic stomach, a very rare pathology that can lead to false positive results. Arch Endocrinol Metab. 2017;61(3):288-90.

1 Department of Nuclear Medicine, Virgen Macarena University Hospital, Sevilla, Spain

Correspondence to: Francisco Javier García-Gómez javier191185@gmail.com Received on Jan/8/2016 Accepted on Oct/18/2016 DOI: 10.1590/2359-3997000000243

CASE REPORT

e report a 76 year old female with history of T1N0M0 well differentiated papillary thyroid cancer being operated and treated with ablative dose of iodine-131 (131I). During regular followup, a whole-body 131I scintigraphy was performed 12 months after treatment ends in order to evaluate the persistence of thyroid remainders. Whole-body 131I scan (Figure 1) revealed high intense hypermetabolic uptake foci, located in the midline. This finding was observed with greater intensity of uptake in the posterior projection (Figure 1, right panel; black arrow) being consistent with disease progression as spine bone metastasis. Intense back pain was referred when the patient was purposely questioned. Hormone levels at moment of scan were stimulated Tg < 0.2 ngr/mL, nonstimulated Tg < 0.2 ngr/mL and antithyroglobulin antibodies (TgAB) 349.4 U/L (Reference range: 0.0200.0 U/L). Subsequently, a single photon emission computed tomography/computed tomography (SPECT/CT) was performed due to discordance between biochemical findings (low hormone levels) and the clinical picture of intense back pain and functional imaging consistent with spine bone metastasis. A severe thoracolumbar 288

scoliosis that determines a giant diaphragmatic hernia was revealed thanks to multimodality scan (Figure 2). Thereby, high intense hypermetabolic foci corresponded to physiological activity of the improperly positioned gastric mucosa mimicking bone metastasis.

Figure 1. Whole-body iodine-131 scintigraphy. 150 x 252 mm (96 x 96 DPI). Arch Endocrinol Metab. 2017;61/3

Thyroid cancer imaging

disease (6-10). It can also be seen in healthy tissue, such as thymus, breast, liver and gastrointestinal tract (11) being able to be causes of false positive (12). In this case, 131I SPECT/CT allowed reaching an accurate diagnosis, by discarding unsuspected metastatic bone disease and changing significantly the management and prognosis of the patient. Laparotomy or laparoscopic surgical repair is still the treatment of choice for giant hiatal hernia (13).

Imaging findings

DISCUSSION The giant diaphragmatic hernia associated with intrathoracic stomach is a very rare entity (1) that seems to be related to increased intra-abdominal pressure and diaphragmatic laxity or kyphoscoliosis deviation in obese patients (2). Usually it corresponds to a type III hiatal hernia, with both sliding and paraesophageal components, wherein at least 30% of the stomach is in intrathoracic situation (3). Diaphragmatic hernias remain a diagnostic and surgical challenge, in which imaging techniques have become the cornerstone. In another vein, whole-body 131I scintigraphy has proven to be a minimally invasive and safe technique that allows the diagnosis metastasis and recurrence after thyroidectomy when performed at 6-12 months of thyroid remainders ablation with radioiodine, as well as in monitoring during the long-term follow-up. However, the ATA and ETA guidelines state that lowrisk individuals who have had a first post-radioactive iodine remnant ablation whole-body scan with an undetectable Tg level and a negative anti-Tg antibody level as well as a negative neck ultrasound do not require routine follow-up whole-body scan. Even if false positives in 131I body scans are extremely rare, they have been reported due to contamination by body fluids, ectopic thyroid tissue, infectious and inflammatory processes, benign and malignant tumors, serous cysts and even pulmonary bronchiolitis (4,5). Few cases of false negatives by intrathoracic hernia have been described, because the strange location of the physiological uptake of radioiodine by the gastric mucosa and the limited resolution of planar imaging may suggest that results from a recurrence of the Arch Endocrinol Metab. 2017;61/3

Whole-body scintigraphy after intravenous injection of 187 MBq (5 mCi) of 131I (Figure 1) demonstrated physiological uptake of the radiotracer as well as a pathological uptake deposit which was located in the midline. This finding was present in both anterior and posterior projections, presenting much greater intensity in the latter (black arrow). Due to lack of precise anatomical definition of this technique, it was compatible with spinal bone metastases as a result of the central location, greater in backplanes and high intensity of uptake. A 131I SPECT/CT (128x128 matrix; step and shoot mode; 360 degrees orbit; 25 seconds per image; lowdose CT: 140 kV, 2.5 mA) was performed in order to clarifying our clinical and radiological inconsistency. Fused images, SPECT alone and CT alone (top-down, respectively) revealed severe kyphoscoliosis associated with giant diaphragmatic hernia and intrathoracic stomach. In this way, hypermetabolic foci corresponded to physiological radiotracer uptake of the gastric mucosa. It is known that SPECT/CT was highly accurate in patients who underwent a single challenge of radioiodine therapy (9). In our case, SPECT/CT helped clarify our diagnostic doubt and substantially modified the management and prognosis of the patient. Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Grazia KJA, Godoy ZM, Cavallo BI, Cortés AC. Hernia hiatal gigante con estómago intratorácico: Reporte de un caso y revisión de la literatura. Rev Chil Radiol. 2012;18:179-83. 2. Schuchert MJ, Adusumilli PS, Cook CC, Colovos C, Kilic A, Nason KS, et al. The impact of scoliosis among patients with giant paraesophageal hernia. J Gastrointest Surg. 2011;15:23-8. 3. Mitiek MO, Andrade RS. Giant hiatal hernia. Ann Thorac Surg. 2010;89:S2168-73.

289

Figure 2. Axial, sagittal and coronal views (left, center and right columns) of hydrib iodine-131 SPECT/CT. Fused images are exposed in the upper row, the SPECT alone in the center row and CT alone in the lower row. 88 x 60 mm (300 x 300 DPI).

Thyroid cancer imaging

4. García Alonso MP, Balsa Bretón MA, Paniagua Correa C, Castillejos Rodríguez L, Penín González FJ, Elviro Peña R, et al. Iodine uptake in the chest in the follow-up of well-differentiated thyroid cancer. Rev Esp Med Nucl. 2011;30:24-8. 5. Thientunyakit T. False-positive 131I whole-body scan in well-differenciated thyroid cáncer patient with respiratory bronchiolitis. Clin Nucl Med. 2013;38:730-4. 6. Ho Y, Hicks R. Hiatus hernia: a potential cause of false-positive iodine-131 scan in thyroid carcinoma. Clin Nucl Med. 1998;23:621-2. 7. Schneider JA, Divgi CR, Scott AM, Macapinlac HA, Sonenberg M, Goldsmith SJ, et al. Hiatal hernia on whole-body radioiodine survey mimicking metastatic thyroid cancer. Clin Nucl Med. 1993;18:751-3.

10. Willis LL, Cowan RJ. Mediastinal uptake of I-131 in a hiatal hernia mimicking recurrence of papillary thyroid carcinoma. Clin Nucl Med. 1993;18:961-3. 11. Oh JR, Ahn BC. False-positive uptake on radioiodine whole-body scintigraphy: physiologic and pathologic variants unrelated to thyroid cancer. Am J Nucl Med Mol Imaging. 2012;2:362-85. 12. Buton L, Morel O, Gault P, Illouz F, Rodien P, Rohmer V. False-positive iodine-131 whole-body scan findings in patients with differentiated thyroid carcinoma: report of 11 cases and review of the literature. Ann Endocrinol (Paris). 2013;74:221-30. 13. Zhou Y, Du H, Che G. Giant congenital diaphragmatic hernia in an adult. J Cardiothorac Surg. 2014;9:31.

8. Lee YS, Ryu YH, Chang HS, Park CS. Hiatal hernia detected by iodine-131 scan for ablation of thyroid carcinoma. ANZ J Surg. 2010;80:198.

9. Ceylan Gunay E, Erdogan A. Mediastinal radioiodine uptake due to hiatal hernia: a false-positive reason in 131I scan. Rev Esp Med Nucl. 2010;29:95.

290

Arch Endocrinol Metab. 2017;61/3

case report

A rare case of ectopic ACTH syndrome originating from malignant renal paraganglioma Esra Tutal1, Demet Yılmazer1, Taner Demirci2, Evrim Cakır2, Salih Sinan Gültekin3, Bahadır Celep4, Oya Topalog#lu2, Erman Çakal2

SUMMARY Ectopic adrenocorticotropic hormone (ACTH) syndrome is characterized by hypercortisolism due to the hypersecretion of a non-pituitary ACTH-secreting tumor leading to Cushing’s syndrome. Only a few cases have been reported previously as causing ectopic ACTH related to paraganglioma. Herein, we present a case of Cushing’s syndrome, in who was proved to be attributable to an ACTHsecreting renal malignant paraganglioma. A 40-year-old woman presented with a five-month history of newly diagnosed hypertension and diabetes, weakness, hyperpigmentation, oligomenorrhea, hirsutism, and acneiform lesions. She showed cushingoid features, including moon face, facial hirsutism, facial and truncal acne, hyperpigmentation, and severe muscle weakness of the limbs. She did not have other findings such as striae, supraclavicular fat accumulation, and buffalo hump. Laboratory examination showed the presence of hypopotasemia, hyperglycemia, hyperthyroidism, and leukocytosis. The serum levels of ACTH, cortisol, and urine-free cortisol were markedly elevated. Results of an overnight 2-mg dexamethasone suppression test included a basal serum cortisol of 61.1 mcg/dL (normal range: 4.6-22.8 mcg/dL) and a cortisol value of 46.1 mcg/dL after dexamethasone administration. There was no suppression found after 2-day 8-mg dexamethasone administration. Magnetic resonance imaging (MRI) of the pituitary gland indicated two microadenomas. An abdominal MRI scan revealed horseshoe kidney, bilateral adrenal hyperplasia, and masses with dimensions of 35 x 31 mm in the left kidney. Inferior petrosal sinus sampling showed no evidence of a central-to-peripheral gradient of ACTH. A positron emission tomography/computed tomography scan showed intense increased activity in the lower pole of the left kidney. Left adrenalectomy and left partial nephrectomy were performed. The resected tumor was diagnosed as the ACTH-secreting paraganglioma in the pathological examination, which was confirmed by immunohistochemical studies with chromogranin A, synaptophysin, and ACTH. Only a few cases of paragangliomas as a cause of ectopic ACTH syndrome have been reported. To our knowledge, this is the first case of renal paraganglioma resulting in Cushing’s syndrome due to ectopic ACTH hypersecretion. Arch Endocrinol

1 Diskapi Yildirim Beyazit Teaching and Research Hospital, Department of Endocrinology and Metabolism, Irfan Bastug Caddesi, Ankara, Turkey 2 Department of Endocrinology and Metabolism, Diskapi Teaching and Research Hospital, Irfan Bastug Caddesi, Ankara, Turkey 3 Department of Nuclear Medicine, Diskapi Teaching and Research Hospital, Irfan Bastug Caddesi, Ankara, Turkey 4 Department of General Surgery, Diskapi Teaching and Research Hospital, Irfan Bastug Caddesi, Ankara, Turkey

Correspondence to: Esra Tutal Diskapi Hospital, Irfan Bastug Caddesi, Ankara, Turkey akkaymakesra@yahoo.com Received on Apr/28/2016 Accepted on Oct/10/2016 DOI: 10.1590/2359-3997000000240

INTRODUCTION

ctopic adrenocorticotropic hormone (ACTH) syndrome is characterized by hypercortisolism with bilateral adrenocortical hyperplasia and hyperfunction due to the hypersecretion of nonpituitary ACTH-secreting tumor, which leads to Cushing’s syndrome. Ectopic ACTH syndrome appears in approximately 10-15% of adult patients with Cushing’s syndrome (1). Most cases of ectopic ACTH syndrome are caused by malignancies, including the small-cell type of lung carcinomas, thymic carcinoids, islet cell tumors of the pancreas, medullary carcinoma of the thyroid, and bronchial adenomas or carcinoids. Paragangliomas are rare tumors that arise from Arch Endocrinol Metab. 2017;61/3

neural crest cells and are associated with autonomic ganglia. Pheochromocytomas that cause ectopic ACTH syndrome are very rare (2). A few cases with ACTH-secreting paragangliomas have been previously reported, which have been localized in the paranasal sinus (3,4), cervical (5), mediastinal/thoracic (6-8), and retroperitoneal (9) regions. To the best of our knowledge, there hasn’t been any published report in the literature about ACTH-secreting renal malignant paraganglioma. In this report, we present a case of a 40-year-old woman diagnosed with Cushing’s syndrome, which proved to be attributable to an ACTH-secreting renal malignant paraganglioma. 291

Metab. 2017;61(3):291-5.

A case of ectopic ACTH syndrome

CASE REPORT A 40-year-old woman presented with a five-month history of newly diagnosed hypertension and diabetes, weakness, hyperpigmentation, oligomenorrhea, hirsutism, and acneiform lesions. Physical examination revealed a blood pressure of 140/95 mmHg, a heart rate of 82 beats/min, weight of 58 kg, and height of 155 cm. She showed cushingoid features including moon face, facial hirsutism, facial and truncal acne, hyperpigmentation, and severe muscle weakness of the limbs. She did not show findings such as striae, supraclavicular fat accumulation, and buffalo hump. The thyroid examination revealed a 2 cm diameter nodule. Her mood was not depressed. She did not give a special medical problem history in her family. Laboratory examination showed the presence of hypopotasemia, hyperglycemia, hyperthyroidism, and leukocytosis (Table 1). The serum levels of ACTH, cortisol, and urinefree cortisol were markedly elevated (Table 2). Results

of an overnight 2-mg dexamethasone suppression test included a basal serum cortisol of 61.1 mcg/dL (normal range: 4.6-22.8 mcg/dL) and a cortisol value of 46.1 mcg/dL after the dexamethasone administration. There was no suppression after 2-day 8-mg dexamethasone administration (Table 3). We didn’t find any elevation of urinary metanephrine and nometanephrine levels. The patient gave her written informed consent. Thyroid scintigraphy showed a hyperactive nodule, which was localized in the right lobe of the thyroid gland (Figure 1). Magnetic resonance imaging (MRI) of the pituitary gland indicated two microadenomas at the mid-anterior and left-posterior sites. An abdominal MRI scan revealed horseshoe kidney, bilateral adrenal hyperplasia, and masses with dimensions of 35 x 31 mm in the left kidney (Figure 2). Thoracic MRI findings were normal. Table 3. Results of the 2-mg and 8-mg dexametasone supression tests

Table 1. Baseline laboratory values of the patient

Results

Normal range

WBC

11.17

5.2-11.4 10^3/µL

11.7

12-18 g/dL

Plt

218

130-400 10^3/µL

Neutrophil

9.45

1.9-8 10^3/µL

Eosinophil

0-0.8 10^3/µL

Lymphocyte

1.11

0.9-5.2 10^3/µL

Glucose

133

70-100 mg/dL

143

136-145 mmol/L

2.8

3.5-5.1 mmol/L

Basal

ACTH (pg/mL)

Cortisol (mcg/dL)

Urinary free cortisol (ug/d)

679

61.1

2032, 1850*

2 mg DST 8 mg DST

46.1 856

46.8

1030

* Urinary free cortisol were measured two times.

WBC: white blood cell; Hg: hemoglobin; Plt: platelet; Na: sodium; K: potassium.

Table 2. Baseline hormonal values of the patient

Normal range

ACTH (basal)

679

0-46 pg/mL

Cortisol (basal)

61.1

4.6-2.28 ug/dL

2032.5

3.5-5.5 ug/24 hr

1.14

0.74-1.52 ng/dL

Urinary free cortisol Free T4

Results

Free T3

2.7

2.3-4.2 pg/mL

TSH

0.07

0.64-6.27 muI/L

FSH

4.9

IU/L

1.0

IU/L

Estradiol

57.8

pg/mL

PRL

4.4

3.4-29.8 ng/mL

1.05

0-10 ng/mL

Total testosterone

98.45

14.2-73.1 ng/dL

ACTH: adrenocorticotropic hormone; TSH: thyroid-stimulating hormone; FSH: follicle-stimulating hormone; LH: luteinizing hormone; PRL: prolactin; GH: growth hormone.

292

Figure 1. PET/CT scan. (A) Shows a mild focal uptake in the pituitary gland region. (B) Demonstrates horseshoe kidney deformity and intense increased activity in the lower pole of the left kidney (thick black arrow). (C, D) Show heterogeneous increased uptake in the right adrenal gland (thick white arrow), intense increased uptake in the left adrenal gland (black arrowheads), and a focal increased uptake in a lymph node located in the inferior adjacent to the left adrenal gland (thin yellow arrows). Arch Endocrinol Metab. 2017;61/3

A case of ectopic ACTH syndrome

We did not found any increment after desmopressin injection. Inferior petrosal sinus sampling showed that there was no evidence of a central-to-peripheral gradient of ACTH at baseline or after administration of ovine corticotropin-releasing hormone (CRH) (Table 4). A positron emission tomography/computed tomography (PET/CT) scan was performed using 18 F-fluorodeoxyglucose (FDG). The images (Figure 1) showed several pathological uptakes. A mild focal uptake with the maximum standardized uptake value (SUVmax: 2.76) was observed in the pituitary gland Table 4. Results of Inferior Petrosal Sinus Sampling at baseline and after administration of ovine CRH ACTH (right)

ACTH (left )

ACTH (peripheral)

-2 min

848

832

812

0 min

866

904

814

2 min

916

948

846

5 min

858

818

795

10 min

866

837

839

Arch Endocrinol Metab. 2017;61/3

region. The horseshoe kidney deformity, in addition to an intense increased activity in the lower pole of the left kidney (SUVmax: 9.87), was detected. There was a bilateral increased uptake in both adrenal glands with the right one having a heterogeneous character (left SUVmax: 7.21 and right SUVmax: 6.21). A focal increased uptake (SUVmax: 4.31) in a lymph node located in the inferior adjacent to the left adrenal gland was observed. Left adrenalectomy and left partial nephrectomy were performed. The resected tumor was diagnosed as the ACTH-secreting paraganglioma in the pathological examination, which was confirmed by immunohistochemical studies with chromogranin A, synaptophysin, and ACTH (Figure 2). The periadrenal lymph node was evaluated as a metastatic lymph node (Figure 3). The left adrenal mass was assessed as compatible with the adrenal hyperplasia. After surgical resection of the paraganglioma, the patient’s blood glucose and potassium levels have gradually returned to the near normal ranges without medication. Postoperative levels of plasma ACTH and cortisol returned to the normal ranges. She needed a 7.5-mg 293

Figure 2. Hematoxylin and eosin (H&E) stain of the tumor in the kidney (A), immunohistochemistry with antibodies specific for ACTH (B), chromogranin A (C), and synaptophysin (D).

A case of ectopic ACTH syndrome

Figure 3. Hematoxylin and eosin (H&E) stain of the metastatic periadrenal lymph node.

dose of prednisolone per day. In addition, one month after surgery, the patient was treated with radioiodine 131 I in a dose of 740 MBq (20 mCi) due to clinically apparent hyperthyroidism.

DISCUSSION Paragangliomas are rare tumors arising from chromaffin tissue cells derived from the embryonic neural crest. Paragangliomas may be located between the cervical region and the lower pelvis cavity (wherever sympathetic or parasympathetic ganglia are present). Most of these tumors arise sporadically in the later period of life, especially after the sixth decade. Approximately 20% of these tumors are malignant (10). In the majority of cases, paragangliomas of the head and neck are benign. Typically, these tumors are asymptomatic, but sometimes, catecholamine excess symptoms and signs, including hypertension, diabetes, and hypermetabolism, may be seen. Hypersecretion of catecholamines were not found in our patient. However, 15-35% of abdominal paragangliomas are malignant, especially in patients who have mutations of the gene encoding the B subunit of the mitochondrial complex II enzyme succinate dehidrogenase enzyme subunit B (SDHB) (11). We could not evaluate SDHB mutation in our patient for financial reasons. Malignancy is defined by the presence of metastases, and the most common sites for metastases of malignant paragangliomas are lymphatic nodes (local or distant), as in our patient. The other common sites for metastases are bones, lungs, and liver (12). Although there is no 294

consensus about long-term postoperative follow-up, these patients should be monitored for recurrence (13). During preoperative workup for neuroendocrine disorders such as paragangliomas, which may have adrenal or extra-adrenal localizations, an accurate definition of the disease or disease extension, and the nature of the lesions, can be possible by means of whole-body PET/CT imaging due to the simultaneous assessment of functional and anatomical information and the obtaining of standard uptake values of the lesions (14,15). It may be challenging to make differential diagnosis between the benign and malignant pathologies in a case with bilateral increased FDG uptake of adrenal glands (16). However, in the current patient, final assessment was decided in accordance with benign hyperplasia for adrenal uptakes and as a malignant tumor with regional metastasis for the renal lesion due to the more intense FDG uptake in the left renal mass observed by MRI and CT with its higher density (> 10 HU), and the presence of increased FDG uptake in the adjacent lymph node involvement. Another recommended functional imaging modality is 123I or 131I-metaiodobenzylguanidine (MIBG) scintigraphy, which has been used widely for the assessment of patients with paragangliomas. 18 F-labelled fluoro-deoxy-glucose (18F-FDG) PET and somatostatin analogues labelled with gallium-68 may be used for detection of small lesions and metastatic lesions (17). The primary aim of the treatment of malignant paragangliomas is surgical removal of the primary tumor and, if possible, the resection of the metastatic foci. In our patient, after complete surgical removal of the tumor and the metastatic lymph node, clinical and biochemical improvement was found. For inoperable tumors, radioactive isotop treatment with 131I-MIBG may provide symptomatic relief and some tumor shrinkage. Only a few cases of paragangliomas as the cause of ectopic ACTH syndrome have been reported. To our knowledge, this is the first case of metastatic renal paraganglioma resulting in Cushingâ&#x20AC;&#x2122;s syndrome due to ectopic ACTH hypersecretion. However, we could not perform a genetic analysis of the patient using succinate dehidrogenase enzyme, particularly for SDHB mutations. Disclosure: no potential conflict of interest relevant to this article was reported. Arch Endocrinol Metab. 2017;61/3

A case of ectopic ACTH syndrome

REFERENCES 1. Ilias I, Torpy DJ, Pacak K, Mullen N, Wesley RA, Nieman LK. Cushing’s syndrome due to ectopic corticotropin secretion: twenty years’ experience at the National Institutes of Health. J Clin Endocrinol Metab. 2005;90(8):4955-62. 2. Aniszewski JP,Young WF Jr,Thompson GB, Grant CS, van Heerden JA. Cushing syndrome due to ectopic adrenocorticotropic hormone secretion. World J Surg. 2001;25(7):934-40. 3. Apple D, Kreines K. Cushing’s syndrome due to ectopic ACTH production by a nasal paraganglioma. Am J Med Sci. 1982;283(1):32-5. 4. Lieberum B, Jaspers C, Münzenmaier R. ACTH-producing paraganglioma of the paranasal sinuses. HNO. 2003;51(4):328-31. 5. Omura M, Sato T, Cho R, Iizuka T, Fujiwara T, Okamoto K, et al. A patient with malignant paraganglioma that simultaneously produces adrenocorticotropic hormone and interleukin-6. Cancer. 1994;74(5):1634-9. 6. Park HK, Park CM, Ko KH, Rim MS, Kim YI, Hwang JH, et al. A case of Cushing’s syndrome in ACTH-secreting mediastinal paraganglioma. Korean J Intern Med. 2000;15(2):142-6. 7. Hashimoto K, Suemaru S, Hattori T, Sugawara M, Ota Z, Takata S, et al. Multiple endocrine neoplasia with Cushing’s syndrome due to paraganglioma producing corticotropin-releasing factor and adrenocorticotropin. Acta Endocrinol (Copenh). 1986;113(2):189-95.

10. Andersen KF, Altaf R, Krarup-Hansen A, Kromann-Andersen B, Horn T, Christensen NJ, et al. Malignant pheochromocytomas and paragangliomas - the importance of a multidisciplinary approach. Cancer Treat Rev. 2011;37(2):111-9. 11. Chrisoulidou A, Kaltsas G, Ilias I, Grossman AB. The diagnosis and management of malignant phaeochromocytoma and paraganglioma. Endocr Relat Cancer. 2007;14(3):569-85. 12. Zelinka T, Timmers HJ, Kozupa A, Chen CC, Carrasquillo JA, Reynolds JC, et al. Role of positron emission tomography and bone scintigraphy in the evaluation of bone involvement in metastatic pheochromocytoma and paraganglioma: specific implications for succinate dehydrogenase enzyme subunit B gene mutations. Endocr Relat Cancer. 2008;15(1):311-23. 13. Plouin PF, Amar L, Dekkers OM, Fassnacht M, Gimenez-Roqueplo AP, Lenders JW, et al.; Guideline Working Group. European Society of Endocrinology Clinical Practice Guideline for long-term follow-up of patients operated on for a phaeochromocytoma or a paraganglioma. Eur J Endocrinol. 2016;174(5):G1-10. 14. Taïeb D, Neumann H, Rubello D, Al-Nahhas A, Guillet B, Hindié E. Modern nuclear imaging for paragangliomas: beyond SPECT. J Nucl Med. 2012;53(2):264-74. Review. 15. Taïeb D, Sebag F, Barlier A, Tessonnier L, Palazzo FF, Morange I, et al. 18F-FDG avidity of pheochromocytomas and paragangliomas: a new molecular imaging signature? J Nucl Med. 2009;50(5):711-7. 16. Balasubramaniam S, Fojo T. Practical considerations in the evaluation and management of adrenocortical cancer. Semin Oncol. 2010;37(6):619-26.

9. Willenberg HS, Feldkamp J, Lehmann R, Schott M, Goretzki PE, Scherbaum WA. A case of catecholamine and glucocorticoid excess syndrome due to a corticotropin-secreting paraganglioma. Ann N Y Acad Sci. 2006;1073:52-8.

17. Parenti G, Zampetti B, Rapizzi E, Ercolino T, Giachè V, Mannelli M. Updated and new perspectives on diagnosis, prognosis, and therapy of malignant pheochromocytoma/paraganglioma. J Oncol. 2012;2012:872713.

8. Dahir KM, Gonzalez A, Revelo MP, Ahmed SR, Roberts JR, Blevins LS Jr. Ectopic adrenocorticotropic hormone hypersecretion due to a primary pulmonary paraganglioma. Endocr Pract. 2004;10(5):424-8.

Arch Endocrinol Metab. 2017;61/3

295

review

Glycated albumin: a potential biomarker in diabetes Priscila Aparecida Correa Freitas1, Lethicia Rozales Ehlert¹, Joíza Lins Camargo1,2 Programa de Pós-graduação em Ciências Médicas – Endocrinologia, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brasil 2 Serviço de Endocrinologia, Hospital de Clínicas de Porto Alegre (HCPA), Porto Alegre, RS, Brasil 1

Correspondence to: Joíza Lins Camargo Serviço de Endocrinologia, Hospital de Clínicas de Porto Alegre Rua Ramiro Barcellos, 2350 Prédio 12, CPE, 4º andar 90035-903 – Porto Alegre, RS, Brasil jcamargo@hcpa.edu.br Received on Aug/2/2016 Accepted on Feb/13/2017 DOI: 10.1590/2359-3997000000272

ABSTRACT Diabetes mellitus (DM) is a chronic and metabolic disease that presents a high global incidence. Glycated hemoglobin (A1C) is the reference test for long-term glucose monitoring, and it exhibits an association with diabetic chronic complications. However, A1C is not recommended in clinical situations which may interfere with the metabolism of hemoglobin, such as in hemolytic, secondary or iron deficiency anemia, hemoglobinopathies, pregnancy, and uremia. The glycated albumin (GA) is a test that reflects short-term glycemia and is not influenced by situations that falsely alter A1C levels. GA is the higher glycated portion of fructosamine. It is measured by a standardized enzymatic methodology, easy and fast to perform. These laboratory characteristics have ensured the highlight of GA in studies from the last decade, as a marker of monitoring and screening for DM, as well as a predictor of long-term outcomes of the disease. The aim of this review was to discuss the physiological and biochemistry characteristics of the GA, as well as its clinical utility in DM. Arch Endocrinol Metab. 2017;61(3):296-304. Keywords Diabetes mellitus; glycated albumin; glycated proteins

INTRODUCTION

iabetes mellitus (DM) is a chronic metabolic disease caused by diminished or absent secretion of insulin or even by reduced tissue sensitivity to insulin (1,2). Presently, DM is a worldwide epidemic and a great challenge to health care systems everywhere. The International Diabetes Federation (IDF) estimates that one in eleven adults have DM, totalizing approximately 415 million people, and 193 million of them have not yet been diagnosed (1). Chronic hyperglycemia is a common feature in all subtypes of this disease and is associated with long-term damage, which increases the morbidity and mortality rates and causes dysfunction of different organs, such as kidney failure, blindness, and amputation of limbs (2). These chronic complications are costly to the health care systems and reduce the life expectancy of diabetic patients (1,2). Currently, the laboratory tests used to diagnose DM are glycated hemoglobin (A1C), fasting plasma glucose (FG) and two-hour plasma glucose (2hG) after a 75g oral glucose tolerance test (OGTT) (2,3). A1C is also the reference test for glycemic monitoring since it directly reflects mean glycemia (5) and is strongly

296

correlated to the long-term complications of DM (4,5). However, the use of A1C is not recommended in some clinical situations that influence the hemoglobin metabolism (3,6,7), due to the possible interference in its results making them misinterpreted. Moreover, recent studies have shown a disagreement between the A1C levels in different ethnic groups for equal levels of glycemia (8,9), but the reasons for these disparities have not yet been well explained. Glycated albumin (GA) is a laboratory test that has gained some importance for glycemic monitoring in DM in the last decades (10,11). GA is one of the fructosamines, but it has the advantage of not being influenced by the concentration of other serum proteins since it is specific to the albumin glycation rates (12). Further, GA does not require fasting for its measurement and reflects short-term glycemia due to the half life time of the albumin, which is approximately 3 weeks. Compared to A1C, GA is not affected by the presence of hemolytic processes and abnormal Hb (13). Besides, in conditions such as anemia, pregnancy, postprandial hyperglycemia and DM using insulin, GA seems to be a better glycemic marker than A1C (11) and also it is especially indicated for diabetic patients on Arch Endocrinol Metab. 2017;61/3

Glycated albumin in diabetes mellitus

BIOCHEMICAL CHARACTERISTICS OF GA AND BIOLOGICAL IMPACT OF GLYCATION Albumin is a high molecular weight protein with 66.7 kDa, composed of a single polypeptide chain which contains 585 amino acids, 17 disulfide bridges and 3 homologous domains that are connected in a helical structure (18). It is the main plasma protein, representing about 60% of the total proteins in the blood, with concentrations between 3.0 and 5.0 g/dL and a half-life of 14 to 20 days (18,19). Albumin structure makes it easier to perform its physiological functions, such as maintaining pH and blood osmotic pressure. Also, albumin acts as a powerful antioxidant and as the main transporter of metabolic products, ions, nutrients, drugs, hormones and fatty acids (20). Similar to the other proteins, albumin also goes through the physiologic process of glycation (21). By definition, glycation is a non-enzymatic spontaneous reaction in which a reducing sugar is added to a free amino group, typically lysine or arginine present within proteins, also called as Maillard reaction (Figure 1) (18-20). The first step of this reaction involves the formation of an unstable and reversible product known as Schiff base, formed by the bonding of a carbonyl group of an acyclic carbohydrate with the N-terminal amino acid (19). This intermediate product can suffer a change in its conformation and result in a stable and irreversible ketamine, known as the Amadori product (22). The main Amadori adduct formed is fructoselysine, a reaction between glucose and lysine, which may occur on 59 lysine sites present in albumin (18). However, lysine 525 has been identified as the largest albumin glycation site, which is evidenced both in vivo and in vitro experiments (23,24). The set of ketamines formed by non-enzymatic glycation of Arch Endocrinol Metab. 2017;61/3

proteins is chemically called “fructosamine”. Among the serum fructosamines, GA is the main constituent, representing about 80% of the total of glycations in plasma (18). The glucose concentration and time of exposure between protein and sugar are the determining factors for the glycations performed during the life of the protein. In other words, glycation depends on the degree and duration of hyperglycemia (22). Extracellular proteins, such as albumin, may be more susceptible to Amadori rearrangements than intracellular proteins as Hb (18). This is due to plasmatic proteins being directly exposed to plasma glucose. These features could justify the differences in the rates of albumin glycation that are about 9 to 10 times greater than those of hemoglobin (25). However, in the in vitro experiment by Ueda and Matsumoto, it was evidenced that GA production was about 4.5 times greater than A1C after adding known and equal concentrations of glucose in previously treated samples from healthy volunteers. These findings showed that even in identical in vitro glycation conditions, GA is produced faster than A1C (20). In advanced glycation stages, additional oxidative and irreversible events occur regarding the glycated proteins, producing stable and heterogeneous compounds known as advanced glycation end products (AGEs – Figure 1). Although the formation of AGEs

Reducing sugar

Protein

Protein +

 NH2

NH2 Reducing sugar

Advanced glycation end products (AGEs)

Shiff base

Protein Reactive dicarbonyl compounds

NH2 Amadori product Reducing sugar

Figure 1. Maillard reaction illustration. In the first glycation stage, there is production of Shiff base by a reaction between a reducing sugar and a free amine group present into the polypeptide chain of plasma proteins and, subsequently, a rearrangement yield the Amadori product. In the following stages, the degradation of the Shiff base and Amadori products, as well as sugar autoxidation are responsible for forming reactive dicarbonyl compounds, known as AGEs’ precursors. 297

hemodialysis (14,15). Recently, studies on type 1 (16) and type 2 DM patients (17) reported an association of GA with the chronic complications of the disease. Although GA is being studied in the last few years, this test is not yet widely used in laboratory routine, and few commercial reagents are available on the market to its analysis. However, the results of clinical investigations make GA a promising marker in DM. In this context, the proposal of this review is to present the physiological and laboratory characteristics of GA, and discuss its clinical usefulness in the diagnosis and management of DM.

Glycated albumin in diabetes mellitus

is a normal process, conditions of typical hyperglycemia in patients with DM increase their production rates (26). AGEs receptors are present in cells of different tissues, such as macrophages, muscle, endothelial and glial cells (27). They are expressed as membrane molecules, constituents of immunoglobulin superfamily and act as signal transduction receptors, inducing oxidative stress and starting an inflammatory cascade by activation of the nuclear factor-κB (NF-κB). NF-κB modulates the gene transcription of pro-inflammatory molecules such as interleukins 1, 6 and 8 and tumor necrosis factor-α, and also the vascular cell adhesion molecule-1 and intercellular adhesion molecule-1 (26). As a consequence of this cascade, there is an increased production of reactive oxygen species, which is directly associated with the pathogenesis and longterm complications in DM (21,27). Kisugi and cols. evaluated samples from a patient with DM during one month of hospitalization due to hyperglycemia symptoms and evidenced that the formation of AGEs was drastically reduced with the concomitant reduction of the GA levels (24).

LABORATORY MEASUREMENT OF GA Historically, fructosamine has been used in clinical practice when a short-term glycemia evaluation is needed (12,28). However, this test presents low accuracy since it is influenced by all plasma proteins and also by other molecules present in the blood, such as bilirubin, uric acid and low molecular weight substances (12). Further, fructosamine is not available at all laboratories (18,29) and there are no well-established international standards for its use. Methods for the evaluation of GA have been developed since the 1980s using serum or plasma samples (28). The older methods presented many disadvantages due to the techniques complexity or the high costs and/or lack of precision. Besides, the non-standardization of these assays corroborated to the unpopularity of GA, and all attention were directed to A1C (30). GA can be measured by ion-exchange highperformance liquid chromatography (HPLC), boronate affinity chromatography, immunoassays (radioimmunoassay and Enzyme Linked Immuno Sorbent Assay), colorimetric method with thiobarbituric acid and enzymatic methods using proteinase and ketamine oxidase (12,28,29,31), however these 298

methods are currently not available in the laboratory routine (32). The reference intervals described for GA depend on the method used since GA levels may vary according to the glycation sites analyzed by the assay employed, and also if the method of analysis considers the GA molecule for measurement and not its glycated amino acids (31). For instance, the immunoassay techniques, colorimetric methods with thiobarbituric acid and enzymatic methods consider the glycated amino acids as the reference for the GA levels. On the other hand, the HPLC techniques and other chromatography methods consider the GA molecule to define their levels. Despite this difference, all methods available agree that the proportion of GA in patients with DM increases 2 to 5 fold compared to normoglycemic patients (18). An enzymatic methodology with a shorter operational time and easier to perform both manually and automatically was proposed to evaluate the GA levels in order to overcome the limitations of the previously existing techniques (12). This method presents three steps (Figure 2), using specific proteinase for albumin and ketamine oxidase, besides the bromocresol green reagent for albumin determination and later calculation of %GA. In the validation performed to introduce the test on the market, the analytic performance was excellent and the assay was not influenced by bilirubin and glucose, but a slight interference in the GA levels in the presence of Hb and ascorbic acid was reported (12). Other studies described similar results, concluding that the new enzymatic methodology, known as “Lucica GA-L®” (Asahi Kasei Pharma Corporation, Tokyo, Japan) showed reproducibility, accuracy (31) and a good correlation with A1C (30). Subsequently, other manufacturers have released similarly methodologies for GA analysis, but instead of a specific measurement of its levels, these assays employ math equations to obtain %GA levels (33,34). In addition, the biological variation of GA measured by Lucica GA-L® is lower when compared to fructosamine and A1C (1.7%, 2.8% and, 2.4%, respectively) (35). GA presents good stability when frozen at very low temperatures. In the study by Kohzuma and cols. frozen samples at -80ºC maintained the GA levels stable for 4 years (31). Watano and cols. found similar results for storage of serum samples at -70ºC. However, they observed a considerable increase in GA levels frozen at -20°C after 6 months (36). Nathan and cols. measured the GA levels in samples of participants Arch Endocrinol Metab. 2017;61/3

Glycated albumin in diabetes mellitus

in the study The Diabetes Control and Complications Trial Research Group (DCCT) frozen at -70ÂşC 23 years ago and concluded that the stability of this analyte remained adequate (16). However, despite all the characteristics cited, the GA test is not yet regularly available in laboratory practice (3), but it has been much used in DM clinical research in the last decade. The factor responsible for the increased number of studies on GA was the consolidation, although without a defined international consensus, of Lucica GA-LÂŽ enzymatic assay for GA determination. Even though others enzymatic assays for GA have been launched into the market, currently there are only foreign suppliers available (34). It makes the GA a costlier test than A1C in Brazil. Recently, we compared two different assays for GA and the price per test was around U$ 4 to 6, in contrast with A1C test that is around U$ 2 to 3 in Brazil (34). However, this outlook is likely to change in a near future.

controversies that limit it use. They are related to certain clinical situations or to the analytical methods employed (3,6). These conditions may yield false results for A1C that are not truly correlated with the mean glycemia (28), and directly affecting the identification and management of patients with DM. In such cases, GA may be an adequate alternative for A1C in the glycemic control (37).

GA and presence of alterations in Hb GA can be used as an alternative to A1C in any hematological alteration that interferes in the halflife of red blood cells and/or in the structure or chemical characteristics of Hb (28). Hemolytic anemias and bleeding episodes reduce the A1C values, while iron deficiency anemias, thalassemias, and hemoglobinopathies may elevate its results (6,11,38). During the fetal period, the main type of Hb in the red blood cells is the fetal Hb (HbF), which is gradually replaced by HbA after birth. Since A1C is a glycation product of the HbA, neonates tend to have falsely diminished levels (39). However, the interferences with A1C measurements are method-dependent. Some analytical methodologies may not be affected by common interferences such as hemoglobin variants. The National Glycohemoglobin Standardization Program

USE OF GA UNDER CONDITIONS THAT AFFECT A1C In clinical practice, A1C is used as a reference test for glucose monitoring in DM, and it is also a diagnostic tool (2). However, there are some punctual disadvantages and

Peroxidase

Pigment

First step H2O + O2 H2O2 + chromogem

Protease Glycated albumin

Ketoamine oxidase Glycated amino acids

Amino acids + Glucosone

Second step HO

Br Br

OH Br

Total albumin

O S O O

Pigment

Bromocresol green

GA% =

Glycated albumin Total albumin

X 100

Figure 2. Enzymatic reaction for GA determination. First step: glycated amino acids are released from the GA molecule through an albumin-specific protease. A ketoamine oxidase separates the free amino acids and glucosone, this last an intermediate product of Amadori reaction. The final pigment is proportional to the amount of GA in the sample; Second step: plasma albumin reacts with bromocresol green into an acid environment, resulting in a colored compound that is related to total albumin concentration; Third step: the percentage of GA is obtained by a math calculation considering the two previous reactions. Arch Endocrinol Metab. 2017;61/3

299

Third step

Glycated albumin in diabetes mellitus

(NGSP) provides detailed information regarding interference in A1C assays by manufacturers (7).

GA and pregnancy During pregnancy, it is recommended that women who already have DM and those who develop gestational DM be followed by glucose self-monitoring and by the A1C levels (2). However, it has been well established that during the last months of pregnancy there is an increased demand for iron, which directly reflects on changes in the A1C throughout the pregnancy (40). In a prospective study by Hashimoto and cols. conducted in pregnant Japanese women with DM, a significant elevation of A1C was found at the end of pregnancy, inversely to the ferritin levels and transferrin saturation. On the other hand, GA remained stable through this period, because it did not suffer interference from the physiological changes characteristic of pregnancy (41).

GA and chronic kidney disease (CKD) In patients with DM and CKD, A1C may not be a reliable marker of glycemic control (42). Patients with CKD generally present erythropoietin deficiency and, consequently, they develop anemia. Thus it is necessary to use exogenous erythropoietin to compensate the diminished endogenous synthesis by the kidney, and also iron, which falsely alters the levels of A1C. Further, these patients may need blood transfusion frequently and, when on hemodialysis, they present a 20-50% diminished lifetime of the erythrocytes, also contributing to false values of A1C (43). The increased uremia in CKD results in the production of carbamylated Hb, an in vitro interfering factor in some analytic methodologies for A1C (6). Some studies have demonstrated that GA provides a more precise control of glycemia in patients with

advanced stages of CKD (14,15,42). However, in the presence of massive proteinuria with diminished serum albumin, the GA levels can also be falsely altered (42,44), and it is necessary to perform a critical evaluation and adequately choose the best glycemic marker in this condition.

GA IN THE DIAGNOSIS OF DM Despite the evident informative value of A1C in monitoring DM, some authors have questioned the cutoff point used for this test in diagnosing the disease. This is because the current criteria adopted show a discrepancy between the proportion and profile of patients identified as having DM by the A1C, compared to the tests based on glycemia (45,46). In addition, the patients who present special conditions that interfere with A1C results should be screened for DM with alternative markers. Since the enzymatic method for GA was recently developed, few diagnostic accuracy studies of GA for DM have been published (Table 1). In 2006, the Japan Diabetes Society (JDS) established a reference interval for GA from 12.3% to 16.9% (47). Years later, in a larger study (N = 1.575), Furusyo and cols. published a reference interval for GA from 12.2% to 16.5%, corroborating with that reported by the JDS. Further, this study found that the cutoff point of GA ≥ 15.5% presented a good sensitivity and specificity (both 83.3%) to identify DM, using FG and/or A1C (≥ 126 mg/dL and ≥ 6.5%, respectively) as reference tests (48). In 2015, the same group evaluated 176 residents of Japan diagnosed with DM by the OGTT, according to WHO criteria. ROC curve analysis showed that GA presented significant differences in the area under the curve (AUC) for DM diagnosis, and these values increased when combining GA with FG or 2hG than GA when isolated (AUC: 0.863, 0.968 and 0.672, respectively) (49).

Table 1. Diagnostic accuracy studies of GA and the cutoff points found to screening DM

Study

Country

Male

RI GA (%)

DM cutoff

SxS

Tominaga and cols. 2006

699

Japan

52%

12.3 - 16.9

Paroni and cols. 2007

Italy

37%

11.7 - 16.9

Kohzuma and cols. 2011

201

USA

47%

11.9 - 15.8

Furusyo and cols. 2011

1.575

Japan

30%

12.2 - 16.5

15.5%

83.3 x 83.3

Hwang and cols. 2014

852

Korean

58%

14.3%

66.4 x 52.5

Ikezaki and cols. 2015

176

Japan

46%

15.2%

62.1 x 61.9

Hsu and cols. 2015

2.192

Taiwan

50%

14.9%

78.5 x 80.0

Testa and cols. 2017

252

Italy

38%

9.0 - 16.0

RI GA: reference interval for GA; SxS: sensitivity and specificity to the cutoff points found.

300

Arch Endocrinol Metab. 2017;61/3

Glycated albumin in diabetes mellitus

GA IN GLUCOSE MONITORING Differently from A1C long-term formation (about 120 days, mean life of the erythrocytes), GA is formed in a period of approximately 2 to 4 weeks (Figure 3) (37). This feature enhances GA sensitivity to the rapid alterations in glucose levels, which may not be efficiently identified with an isolated measure of plasma glucose (13,19). 0

A1c

50%

25%

50%

120 days

25%

Figure 3. Glycation rates of GA e A1C. GA is produced over the life span of albumin of approximately 8 weeks, however, the first 2 weeks account for half of its production. Differently, due to the life span of erythrocytes, that is around 120 days, A1C takes approximately 4 months to be completely produced, and the first month is responsible for half of its glycation. Arch Endocrinol Metab. 2017;61/3

Compared to A1C, GA is more suitable to monitor the beginning of drug therapy in DM (55), and also to control the dose and change of medication (51), since its levels diminish faster than A1C in intensive treatment (19). Paroni and cols. evidenced that GA was a better marker to evaluate the responses to treatment with insulin in type 2 DM patients with inadequate glycemic control, and also that GA presented a greater correlation with FG than A1C (R = 0.75 versus R = 0.54, respectively) (30). Moreover, Yoon and cols. reported that the worsening of the beta-pancreatic cell function was associated with the time of duration of DM, and also with increased GA and GA/A1C ratio, but not with A1C alone (56). In general, GA can be employed to show mean glycemia and also to evaluate the glycemic variability and postprandial glucose levels more adequately than A1C (37,57). Elevations in postprandial glycemia are associated with the increased risk of cardiovascular diseases and microangiopathy, thus the detection of these glucose variations is important (37). The reasons why GA is better related to postprandial glycemia have not yet been elucidated (11).

GA AS A PREDICTOR OF LONG TERM COMPLICATION IN DM The chronic hyperglycemia considerably increases the risk of developing micro and macrovascular diseases over time (1,2). A1C is a marker that has been strongly explored in clinical research and much evidence has supported its use as a predictor for these complications in DM (4,5). However, there is still controversy if mean glycemia itself or glycemic variability is the main determining factor for chronic damage in DM (16). Recent studies have evaluated the predictive potential value of tests that are more associated with the short term glycemia and that can be used as alternative markers for A1C, such as GA (8,16,17,58). Selvin and cols. cross-sectionally evaluated 1,600 individuals recruited for the Atherosclerosis Risk in Communities (ARIC) study conducted in the USA. They observed that in participants with type 2 DM, both GA and fructosamine were significantly associated with the prevalence of albuminuria, CKD, and retinopathy (8). In a longitudinal study, the same group evaluated 12,306 participants from ARIC, who were followed for over 20 years and demonstrated that both GA and fructosamine were similarly associated with A1C to predict retinopathy and CKD in DM. These findings were confirmed in 301

Hwang and cols. assessed different cutoff points of GA for DM and pre-DM diagnosis in 852 Korean adults, using the ADA criteria to classify the disease. The study reported a cutoff point of 12.5% for pre-DM and 14.3% for DM. GA presented greater sensitivity than A1C (66.4% GA versus 52.5% A1C), but less specificity (88.3% GA versus 95.1% A1C) to predict 2 hG ≥ 200 mg/dL. When the values of 14.3% of GA were associated with FG ≥ 126 mg/dL, higher sensitivity was obtained (77.5%, CI: 72.17–82.0) to diagnose DM (50). Hsu and cols. described a cutoff point of GA ≥ 14.9% for DM (sensitivity: 78.5%; specificity: 80.0%), evaluating 2,192 adult individuals in Taiwan. Also, when the values of 5.7% and 6.5% of A1C were considered, the corresponding GA was 14.5% and 16.5%, respectively (51). Smaller studies described GA intervals in individuals without DM ranging from 11.9 to 15.8% (N = 201 residents of North Carolina, USA) (31); 10.2 to 16.1% (N = 217 African immigrants in America) (52); 10.5 to 17.5% (N = 44 volunteers of a Canadian study) (33); and 9.0 to 16.0% (N = 252 European persons) (53). In young obese persons aged from 10 to 18 years, the value of GA to diagnose DM was ≥ 12% when 2 hG was used as the reference test, and ≥ 14% when A1C was used as reference diagnostic criterion (54).

Glycated albumin in diabetes mellitus

patients diagnosed with DM during the baseline period and in those who developed DM during the follow-up. The odds ratio (OR) observed for the onset of retinopathy in patients with DM and GA levels between 15.7% and 23.0%, was lower than when GA > 23.0% (OR > 8 and OR > 15, respectively), even in a statistical model adjusted for the A1C levels (17). In addition, Selvin and cols. have also demonstrated a similar association between GA and A1C regarding coronary heart disease, ischemic stroke, heart failure, and death (59). Nathan and cols. used data from the DCCT and Epidemiology of Diabetes Interventions and Complications (EDIC) studies to evaluate the correlation between GA and chronic complications in type 1 DM. They showed that GA, as well as A1C, were strongly associated with the onset of retinopathy and nephropathy after a mean follow-up time of 6.5 years, but none evidence was seen for 7-points glucose profile. Only A1C was associated with cardiovascular disease (16). In another smaller study with 154 type 1 DM patients followed over 2.8 years, the progression to nephropathy was associated only with GA and not with A1C. The authors did not find association between these two glycemic markers and the cardiovascular outcomes (58). Apparently, GA is predictive for microvascular complications both in type 1 and type 2 DM. However, regarding macrovascular outcomes, GA seems to be a good marker only in type 2 DM. The mechanisms involved in the development of atherosclerosis and cardiovascular diseases in type 1 DM might explain these findings.

LIMITATIONS OF GA Some situations that interfere with albumin metabolism may also influence GA values. Theoretically, GA is not altered by the serum albumin levels, since its values are corrected for the total albumin, but low levels of this protein are associated with increased glycation rates. On the other hand, increased protein metabolism implicates in lower GA levels (60). Therefore, in conditions as hyperthyroidism, hypothyroidism, liver cirrhosis, nephrotic syndrome with massive proteinuria, or other specific disorders, the use of GA may be misleading and should be avoided (32). However, because this test is relatively new, few studies have been carried out to verify interfering factors in GA levels. Other interfering situations on GA levels already described are age, obesity and inflammatory conditions 302

(observed by the increase of C-reactive protein), smoking, and hypertriglyceridemia (11,32,37). There is little evidence regarding the interpretation of GA in different ethnic groups. However, Selvin and cols. analyzed 1,376 persons without DM and 343 with DM, and found that both GA and A1C are significantly elevated in Blacks compared to Whites (8). Thus, the data presented here show the necessity of being careful when interpreting GA levels in some clinical situations.

CONCLUSIONS GA is a short-term marker of glycemia that has been evaluated as an alternative test to A1C in patients with DM. If compared to A1C, GA is more reliable to evaluate glycemic variability. Also, it is especially indicated for patients on hemodialysis and its levels are not affected in the presence of anemias or hemolytic processes. Compared to the fructosamine test, GA is more advantageous, since it is not influenced by other serum proteins. The enzymatic methodology for its analysis is easy and quick to implement, and highly efficient analytically and with greater standardization. As previously described, in clinical situations that falsely alter A1C levels, the measurement of GA may assign a reliable result for monitoring DM. However, the physiology of the formation of these two glycated proteins ensures advantages to GA compared to A1C in access glucose control, even in the absence of interfering factors. Finally, many studies have shown that GA has good diagnostic accuracy and is strongly associated with the diabetic microvascular complications. Despite all benefits of GA, it does not replace the use of A1C, once each test has its advantages and limitations. The choice regarding which test to use should be guided by the clinical patient features and tests availability. Further, it is necessary an international consensus about laboratory issues and clinical use of GA, to guarantee its inclusion in the routine of clinical laboratory worldwide, thus improving the future screening and management of DM patients. Acknowledgments: this work was supported by Research Incentive Fund (FIPE) from the Hospital de Clínicas de Porto Alegre (HCPA). L.R.E. is recipient of a scholarship from Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul (Fapergs). Fundings: no competing financial interests exist. Disclosure: no potential conflict of interest relevant to this article was reported. Arch Endocrinol Metab. 2017;61/3

Glycated albumin in diabetes mellitus

1. International Diabetes Federation: IDF Diabetes Atlas. 7th ed. Belgium: International Diabetes Federation, 2015. 2. American Diabetes Association: Standards of Medical Care in Diabetes â&#x20AC;&#x201C; 2016. Diabetes Care 2016;39(Supplement1):S1-S112. 3. Sacks DB, Bruns DE, Goldstein DE, Maclaren NK, McDonald JM, Parrott M. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Clin Chem. 2011;57(6):e1-e47. 4. The Diabetes Control and Complications Trial Research Group (DCCT). The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl JMed. 1993;30;329(14): 977-86. 5. U.K. Prospective Diabetes Study (UKPDS) Group: Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998;352(9131):837-53. 6. Cavagnolli G, Pimentel AL, Freitas PA, Gross JL, Camargo JL. Factors affecting A1C in non-diabetic individuals: Review and metaanalysis. Clin Chim Acta. 2015;445:107-14. 7. National Glycohemoglobin Standardization Program (NGSP): factors that interfere with HbA1c test results. 2015. Available from: <http://www.ngsp.org/factors.asp>. Accessed on: Dec 4, 2015. 8. Selvin E, Steffes MW, Ballantyne CM, Hoogeveen RC, Coresh J, Brancati FL. Racial differences in glycemic markers: a crosssectional analysis of community-based data. Ann Intern Med. 2011;154(5):303-9. 9. Ziemer DC, Kolm P, Weintraub WS, Vaccarino V, Rhee MK, Twombly JG, et al. Glucose-independent, black-white differences in hemoglobin A1c levels: a cross-sectional analysis of 2 studies. Ann Intern Med. 2010;152(12):770-7. 10. Kohzuma T, Koga M. Lucica GA-L glycated albumin assay kit: a new diagnostic test for diabetes mellitus. Mol Diagn Ther. 2010;14(1):49-51. 11. Koga M, Kasayama S. Clinical impact of glycated albumin as another glycemic control marker. Endocr J. 2010;57(9):751-62. 12. Kouzuma T, Usami T, Yamakoshi M, Takahashi M, Imamura S. An enzymatic method for the measurement of glycated albumin in biological samples. Clin Chim Acta. 2002;324(1-2):61-71. 13. Kim C, Bullard KM, Herman WH, Beckles GL. Association between iron deficiency and A1C Levels among adults without diabetes in the National Health and Nutrition Examination Survey, 19992006. Diabetes Care. 2010;33(4):780-5. 14. Freedman BI, Shenoy RN, Planer JA, Clay KD, Shihabi ZK, Burkart JM, et al. Comparison of glycated albumin and hemoglobin A1c concentrations in diabetic subjects on peritoneal and hemodialysis. Perit Dial Int. 2010;30(1):72-9. 15. Sany D, Elshahawy Y, Anwar W. Glycated albumin versus glycated hemoglobin as glycemic indicator in hemodialysis patients with diabetes mellitus: variables that influence. Saudi J Kidnei Dis Transpl. 2013;24(2):260-73. 16. Nathan DM, McGee P, Steffes MW, Lachin JM; DCCT/EDIC Research Group. Relationship of glycated albumin to blood glucose and HbA1c values and to retinopathy, nephropathy, and cardiovascular outcomes in the DCCT/EDIC study. Diabetes. 2014;63(1):282-90. 17. Selvin E, Rawlings AM, Grams M, Klein R, Sharrett AR, Steffes M, et al. Fructosamine and glycated albumin for risk stratification and prediction of incident diabetes and microvascular complications: a prospective cohort analysis of the Atherosclerosis Risk in Communities (ARIC) study. Lancet Diabetes Endocrinol. 2014;2(4):279-88. Arch Endocrinol Metab. 2017;61/3

18. Anguizola J, Matsuda R, Barnaby OS, Hoy KS, Wa C, DeBolt E, et al. Glycation of human serum albumin. Clin Chim Acta. 2013;425:64-76. 19. Danese E, Montagnana M, Nouvenne A, Lippi G. Advantages and pitfalls of fructosamine and glycated albumin in the diagnosis and treatment of diabetes. J Diabetes Sci Technol. 2015;9(2):169-76. 20. UedaY, Matsumoto H. Recent topics in chemical and clinical research on glycated albumin. J Diabetes Sci Technol. 2015;9(2):177-82. 21. Arasteh A, Farahi S, Habibi-Rezaei M, Moosavi-Movahedi AA. Glycated albumin: an overview of the In Vitro models of an In Vivo potential disease marker. J Diabetes Metab Disord. 2014;13:49. 22. Cohen MP. Intervention strategies to prevent pathogenetic effects of glycated albumin. Arch Biochem Biophys. 2003;419(1):25-30. 23. Garlick RL, Mazer JS. The principal site of nonenzymatic glycosylation of human serum albumin in vivo. J Biol Chem. 1983;258:6142-6. 24. Kisugi R, Kouzuma T, Yamamoto T, Akizuki S, Miyamoto H, Someya Y, et al. Structural and glycation site changes of albumin in diabetic patient with very high glycated albumin. Clin Chim Acta. 2007;382(1-2):59-64. 25. Rondeau P, Bourdon E. The glycation of albumin: structural and functional impacts. Biochimie. 2011;93(4):645-58. 26. Khan MS, Tabrez S, Rabbani N, Shah A. Oxidative Stress Mediated Cytotoxicity of Glycated Albumin: Comparative Analysis of Glycation by Glucose Metabolites. J Fluoresc. 2015;25(6):1721-6. 27. Singh VP, Bali A, Singh N, Jaggi AS. Advanced Glycation End Products and Diabetic Complications. Korean J Physiol Pharmacol. 2014;18(1):1-14. 28. Raghav A, Ahmad J. Glycated serum albumin: a potential disease marker and an intermediate index of diabetes control. Diabetes Metab Syndr. 2014;8(4):245-51. 29. Roohk HV, Zaidi AR. A review of glycated albumin as an intermediate glycation index for controlling diabetes. J Diabetes Sci Technol. 2008;2(6):1114-21. 30. Paroni R, Ceriotti F, Galanello R, Battista Leoni G, Panico A, Scurati E, et al. Performance characteristics and clinical utility of an enzymatic method for the measurement of glycated albumin in plasma. Clin Biochem. 2007;40(18):1398-405. 31. Kohzuma T, Yamamoto T, Uematsu Y, Shihabi ZK, Freedman BI. Basic Performance of an Enzymatic Method for Glycated Albumin and Reference Range Determination. J Diabetes Sci Technol. 2011;5(6):1455-62. 32. Furusyo N, Hayashi J. Glycated albumin and diabetes mellitus. Biochim Biophys Acta. 2013;1830(12):5509-14. 33. Rodriguez-Capote K, Tovell K, Holmes D, Dayton J, Higgins TN. Analytical evaluation of the Diazyme glycated serum protein assay on the siemens ADVIA 1800: comparison of results against HbA1c for diagnosis and management of diabetes. J Diabetes Sci Technol. 2015;9(2):192-9. 34. Freitas PAC, Ehlert LR, Camargo JL. Comparison between two enzymatic methods for glycated albumin. Anal Methods. 2016;8:8173-8. 35. Montagnana M, Paleari R, Danese E, Salvagno GL, Lippi G, Guidi GC, et al. Evaluation of biological variation of glycated albumin (GA) and fructosamine in healthy subjects. Clin Chim Acta. 2013;423:1-4. 36. Watano T, Sasaki K, Omoto K, Kawano M. Stability of stored samples for assays of glycated albumin. Diabetes Res Clin Pract. 2013;101(1):e1-2. 37. Koga M. Glycated albumin; clinical usefulness. Clin Chim Acta. 2014;433:96-104. 38. Silva JF, Pimentel AL, Camargo JL. Effect of iron deficiency anaemia on HbA1c levels is dependent on the degree of anaemia. Clin Biochem. 2016;49(1):117-20.

303

REFERENCES

Glycated albumin in diabetes mellitus

39. Suzuki S, Koga M, Amamiya S, Nakao A, Wada K, Okuhara K, et al. Glycated albumin but not HbA1c reflects glycaemic control in patients with neonatal diabetes mellitus. Diabetologia. 2011;54(9):2247-53. 40. Hashimoto K, Koga M. Indicators of glycemic control in patients with gestational diabetes mellitus and pregnant women with diabetes mellitus. World J Diabetes. 2015;6(8):1045-56. 41. Hashimoto K, Osugi T, Noguchi S, Morimoto Y, Wasada K, Imai S, et al. A1C but not serum glycated albumin is elevated because of iron deficiency in late pregnancy in diabetic women. Diabetes Care. 2010;33(3):509-11. 42. Zheng CM, Ma WY, Wu CC, Lu KC. Glycated albumin in diabetic patients with chronic kidney disease. Clin Chim Acta. 2012;413(1920):1555-61. 43. Vos FE, Schollum JB, Walker RJ. Glycated albumin is the preferred marker for assessing glycaemic control in advanced chronic kidney disease. NDT Plus. 2011;4(6):368-75. 44. Harada K, Sumida K, Yamaguchi Y, Akai Y. Relationship between the accuracy of glycemic markers and the chronic kidney disease stage in patients with type 2 diabetes mellitus. Clin Nephrol. 2014;82(2):107-14. 45. Carson AP, Reynolds K, Fonseca VA, Muntner P. Comparison of A1c and fasting glucose criteria to diagnose diabetes among US adults. Diabetes Care. 2010;33(1):95-7. 46. Cavagnolli G, Comerlato J, Comerlato C, Renz PB, Gross JL, Camargo JL. HbA1c measurement for the diagnosis of diabetes: is it enough? Diabet Med. 2011;28(1):31-5. 47. Tominaga M, Makino E, Yoshino G, et al.: Report of the committee on standardization of laboratory testing related to diabetes mellitus of Japan Diabetes Society: determination of reference intervals of hemoglobin A1c (IFCC) and glycoalbumin in the Japanese population. J Japan Diab Soc. 2006;49:825-33. 48. Furusyo N, Koga T, Ai M, Otokozawa S, Kohzuma T, Ikezaki H, et al. Utility of glycated albumin for the diagnosis of diabetes mellitus in a Japanese population study: results from the Kyushu and Okinawa Population Study (KOPS). Diabetologia. 2011;54(12):3028-36. 49. Ikezaki H, Furusyo N, Ihara T, Hayashi T, Ura K, Hiramine S, et al. Glycated albumin as a diagnostic tool for diabetes in a general Japanese population. Metabolism. 2015;64(6):698-705.

51. Hsu P, Ai M, Kanda E, Yu NC, Chen HL, Chen HW, et al. A comparison of glycated albumin and glycosylated hemoglobin for the screening of diabetes mellitus in Taiwan. Atherosclerosis. 2015;242(1):327-33. 52. Sumner AE, Duong MT, Aldana PC, Ricks M, Tulloch-Reid MK, Lozier JN, et al. A1C Combined With Glycated Albumin Improves Detection of Prediabetes in Africans: The Africans in America Study. Diabetes Care. 2016;39(2):271-7. 53. Testa R, Ceriotti F, Guerra E, Bonfigli AR, Boemi M, Cucchi M, et al. Glycated albumin: correlation to HbA1c and preliminary reference interval evaluation. Clin Chem Lab Med. 2017;55(2):e31-e33. 54. Chan CL, Pyle L, Kelsey M, Newnes L, Zeitler PS, Nadeau KJ. Screening for type 2 diabetes and prediabetes in obese youth: evaluating alternate markers of glycemia - 1,5-anhydroglucitol, fructosamine, and glycated albumin. Pediatric Diabetes. 2016;17(3):206-11. 55. Yoshiuchi K, Matsuhisa M, Katakami N, Nakatani Y, Sakamoto K, Matsuoka T, et al. Glycated albumin is a better indicator for glucose excursion than glycated hemoglobin in type 1 and type 2 diabetes. Endocr J. 2008;55(3):503-7. 56. Yoon HJ, Lee YH, Kim KJ, Kim SR, Kang ES, Cha BS, et al. Glycated Albumin Levels in Patients with Type 2 Diabetes Increase Relative to HbA1c with Time. Biomed Res Int. 2015;2015:1-8. 57. Suwa T1, Ohta A, Matsui T, Koganei R, Kato H, Kawata T, et al. Relationship between clinical markers of glycemia and glucose excursion evaluated by continuous glucose monitoring (CGM). Endocr J. 2010;57(2):135-40. 58. Yoon HJ, Lee YH, Kim SR, Rim TH, Lee EY, Kang ES, et al. Glycated albumin and the risk of micro- and macrovascular complications in subjects with Type 1 Diabetes. Cardiovasc Diabetol. 2015;14:53. 59. Selvin E, Rawlings AM, Lutsey PL, Maruthur N, Pankow JS, Steffes M, et al. Fructosamine and Glycated Albumin and the Risk of Cardiovascular Outcomes and Death. Circulation. 2015;132(4):269-77. 60. Bhonsle HS, Korwar AM, Kote SS, Golegaonkar SB, Chougale AD, Shaik ML, et al. Low plasma albumin levels are associated with increased plasma protein glycation and HbA1c in diabetes. J Proteome Res. 2012;11(2):1391-6.

50. Hwang YC, Jung CH, Ahn HY, Jeon WS, Jin SM, Woo JT, et al. Optimal glycated albumin cutoff value to diagnose diabetes in Korean

adults: a retrospective study based on the oral glucose tolerance test. Clin Chim Acta. 2014;437:1-5.

304

Arch Endocrinol Metab. 2017;61/3

review

Diagnosis and management of primary aldosteronism Leticia A. P. Vilela1, Madson Q. Almeida1,2

ABSTRACT Primary aldosteronism (PA) is the most common form of secondary hypertension (HTN), with an estimated prevalence of 4% of hypertensive patients in primary care and around 10% of referred patients. Patients with PA have higher cardiovascular morbidity and mortality than age- and sexmatched patients with essential HTN and the same degree of blood pressure elevation. PA is characterized by an autonomous aldosterone production causing sodium retention, plasma renin supression, HTN, cardiovascular damage, and increased potassium excretion, leading to variable degrees of hypokalemia. Aldosterone-producing adenomas (APAs) account for around 40% and idiopathic hyperaldosteronism for around 60% of PA cases. The aldosterone-to-renin ratio is the most sensitive screening test for PA. There are several confirmatory tests and the current literature does not identify a “gold standard” confirmatory test for PA. In our institution, we recommend starting case confirmation with the furosemide test. After case confirmation, all patients with PA should undergo adrenal CT as the initial study in subtype testing to exclude adrenocortical carcinoma. Bilateral adrenal vein sampling (AVS) is the gold standard method to define the PA subtype, but it is not indicated in all cases. An experienced radiologist must perform AVS. Unilateral laparoscopic adrenalectomy is the preferential treatment for patients with APAs, and bilateral hyperplasia should be treated with mineralocorticoid antagonist (spironolactone or eplerenone). Cardiovascular morbidity caused by aldosterone excess can be decreased by either unilateral adrenalectomy or mineralocorticoid antagonist. In this review, we address the most relevant issues regarding PA screening, case confirmation, subtype classification, and treatment. Arch Endocrinol Metab.

1 Unidade de Suprarrenal, Endocrinologia do Desenvolvimento, Laboratório de Hormônios e Genética Molecular – LIM42, Divisão de Endocrinologia e Metabologia, Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo (HCFMUSP), São Paulo, SP, Brasil 2 Instituto do Câncer do Estado de São Paulo (Icesp), FMUSP, São Paulo, SP, Brasil

2017;61(3):305-12.

Correspondence to: Madson Q. Almeida Unidade de Suprarrenal, Laboratório de Hormônios e Genética Molecular LIM-42, Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo Av. Dr. Enéas de Carvalho Aguiar, 155, 2º andar, Bloco 6 05403-900 – São Paulo, SP, Brasil madsonalmeida@usp.br

Keywords Primary aldosteronism; resistant hypertension; diagnosis; aldosterone; renin

Received on Sept/11/2016 Accepted on May/4/2017 DOI: 10.1590/2359-3997000000274

ypertension (HTN) affects between 10 to 40% of the general population and is the leading risk fator for premature death in the world (1,2). Robust experimental and clinical evidence implicates mineralocorticoids in the pathogenesis of HTN (3). Most monogenic forms of HTN in humans can be associated to defects in renal sodium balance (4). Several studies demonstrated that elevated aldosterone levels are predictors of adverse outcome in HTN (5), heart failure (6,7), myocardial infarction (8), and renal insufficiency (9). Primary aldosteronism (PA) is the most common form of secondary HTN with an estimated prevalence of 4% of hypertensive patients in primary care and around 10% of referred patients (10,11). PA is particularly common in patients with resistant HTN, with a prevalence of 14 to 21% (12,13). Resistant HTN is defined as office, or clinic, systolic blood pressure (BP) of ≥ 140 mmHg, diastolic BP of ≥ 90 mmHg, or an elevation of both, on at least 3 antihypertensive medications from different drug classes, preferably including a diuretic (14). Arch Endocrinol Metab. 2017;61/3

PA is the most common curable form of HTN. Because of the adverse cardiovascular effects of excess aldosterone that are independent of high BP levels, patients with PA have higher cardiovascular morbidity and mortality than age- and sex-matched patients with essential HTN and the same degree of BP elevation (15,16) (Table 1). Table 1. Cardiovascular and metabolic complications in primary aldosteronism (PA) compared to essential hypertension (EH) PA (%)

EH (%)

Atrial fibrilation (17)

3.9

1.1

0.001

Coronary artery disease (17)

5.7

2.8

0.03

Heart failure (17)

4.1

1.2

0.003

Nonfatal myocardial infarction (17)

4.4

1.7

0.01

Stroke (18)

7.4

3.5

0.006

41.1

29.6

0.05

22.4

16.8

0.04

Cardiovascular events

Metabolic alterations Metabolic syndrome (19) Abnormal glucose metabolism (20)

Meta analysis.

305

Primary aldosteronism diagnosis

PA is characterized by an autonomous aldosterone production which is inappropriately high for sodium state and is not regulated by angiotensina II or plasma potassium concentrations. This autonomous aldosterone production causes sodium retention, plasma renin supression, HTN, cardiovascular damage, and increased potassium excretion, leading to variable degree of hypokalemia (21). In the largest study of PA prevalence, hypokalemia was identified in 48% of aldosterone-producing adenomas (aldosteronomas) and in 17% of idiopathic hyperaldosteronism (bilateral adrenal hyperplasia) (22). The frequency of PA subtypes and hypokalemia in different cohorts depends on whether PA is routinely screened among hypertensive patients and if adrenal vein sampling (AVS) is available in the specialized center. In general, aldosteroneproducing adenomas (APAs) account for around 40% and idiopathic hyperaldosteronism for around 60% of PA cases. APAs are small benign tumors (1-3 cm) originating from the glomerulosa zone, but in few cases can be smaller than 1 cm and diagnosed only if the AVS shows a lateralized aldosterone production. If a patient with PA has an adrenal tumor larger than 4 cm, we should consider the rare possibility of an aldosterone-producing adrenal carcinoma. Other rare causes of PA, accounting for less than 1%, are unilateral adrenal hyperplasia and familial PA, which is discussed elsewhere in this review. In our cohort of 104 PA cases from the Clinics Hospital of Sao Paulo University Medical School, aldosteronomas were diagnosed in 79% and idiopathic hyperaldosteronism in 20% of our cases, suggesting that idiopathic hyperaldosteronism is probably underdiagnosed at our institution. Familial PA type I (glucocorticoid remediable form of PA) was diagnosed in a single patient. Hypokalemia was identified in 79% of patients with aldosteronomas and in 68% of those with idiopathic hyperaldosteronism. The case detection of PA is recommended in the conditions

listed in Table 2, as addressed in the 2016 Endocrine Society Clinical Practice Guideline of PA (23).

SCREENING AND CASE CONFIRMATION The aldosterone-to-renin ratio is the most sensitive test that screens for PA. Immunometric assays can be employed to measure renin either by testing for plasma renin activity (PRA) or for direct renin concentration (DRC). Currently, most comercial kits measure direct renin concentration; however, the aldosterone-to-renin ratios used in PA screening were determined using PRA. Because the aldosterone-to-renin ratio is more dependent on renin, assays should be sensitive enough to detect PRA levels of 0.2-0.3 ng/mL/h (or DRC of 2 mU/L). When measuring plasma aldosterone (A) in ng/dL and PRA in ng/mL/h, the most commonly adopted A/PRA cut-off for PA screening is 30 (21). In the presence of low renin levels, the aldosterone-torenin ratio may be elevated even when aldosterone is not high. Then, a minimum aldosterone concentration of 15 ng/dL has been proposed as part of the screening criteria. In our institution, 6% of PA patients had aldosterone levels between 12.5 and 15 ng/dL. Because of that, we use the minimum aldosterone level of 12.5 ng/dL to proceed with PA investigation (Figure 1). In a recently introduced and already commonly used automated DRC assay (DiaSorin, LIAISON XL instrument), we can use the conversion factor of 12 (DRC/12 = PRA) to calculate the A/PRA ratio (21). A recent study validated this automated chemiluminescent assay for DRC and aldosterone. Using DRC (mU/L), the A/DRC ratio of 2.1 had a sensitivity of 92%, a specificity of 92%, and negative predictive value of 99% to PA diagnosis. When using the A/DRC of 3.3, sensitivity was 84% and specificity was 96% for PA diagnosis (24). Although we need more studies to validate the aldosterone-to-renin ratio using

Table 2. Recommendations for primary aldosteronism (PA) screening in clinical setting Hypertension and spontaneous or diuretic-induced hypokalemia Hypertension and adrenal incidentaloma

BP above 150/100 on three separate measurements obtained on different days Hypertension (BP â&#x2030;Ľ 140/90) resistant to three conventional antihypertensive drugs (preferably including a diuretic) Controlled BP (< 140/90) on four or more antihypertensive drugs Hypertension and sleep apnea Hypertension and a family history of early onset hypertension or cerebrovascular accident at a young age (< 40 years) All hypertensive first-degree relatives of patients with PA BP: blood pressure.

306

Arch Endocrinol Metab. 2017;61/3

Primary aldosteronism diagnosis

Positive screening for primary aldosteronism: Aldosterone ≥ 12 ng/dL and A/DRC ratio ≥ 2.5 or A/PRA ratio ≥ 30 (correct low K+) A ≥ 20 ng/dL, supressed DRC and A/DRC ratio ≥ 8.8 (A/PRA ratio ≥ 100)

women have a positive screening for PA, we can repeat the test without oral contraceptives (if is the case) or during the follicular phase, or even proceed with confirmatory testing. Table 3. Factors that may interfere with screening for primary aldosteronism Aldosterone

DRC

Aldosterone/ DRC

β-Adrenergic blockers



 (FP)

Central agonists (clonidine, α-methyldopa)



 (FP)

Diuretics





 (FN)

Ca2 blockers (DHPs)





 (FN)

ACE inhibitors, ARBs





 (FN)



 (FP)





 (FP)

To perform furosemide test or saline infusion test No need to perform confirmatory tests Positive test: Furosemide test: DRC < 24 mUI/mL (PRA < 2 ng/mL/h) Saline infusion test: Aldosterone > 10 ng/dL

Adrenal CT scan

Figure 1. Algorithm for the detection and confirmation of primary aldosteronism. A: aldosterone; DRC: direct renin concentration; PRA: plasma renin activity; CT: computed tomography.

In order to increase sensitivity, aldosterone and renin samples should be collected in the morning after patients have been out of bed for at least 2 hours, and usually after they have been seated for 5-15 minutes. Ideally, patients should have unrestricted dietary salt intake before testing and should be potassium-replete (21). It should be emphasized that mineralocorticoid antagonists (spironolactone or eplerenone) and other types of diuretics should be withdrawn for at least 4 weeks before testing. When it is not clinically possible to withdraw mineralocorticoid antagonists or diuretics, suppressed renin levels associated with high aldosterone levels strongly suggest a PA diagnosis. In many cases (excluding mineralocorticoid antagonist or diuretic treatment), the aldosteroneto-renin ratio can be confidently interpreted despite the effect of continued medications, thus avoiding delay and allowing the patient to proceed directly to confirmatory testing. Very often, a washout of all interfering antihypertensive medications is not feasible in patients with severe HTN. Then, the aldosterone-torenin ratio should be interpreted in light of the potential confounding factors (Table 3) (21). In premenopausal ovulating women, false positives can occur during the luteal phase, but only if renin is measured as DRC. Similarly, the use of oral contraceptives is associated with a false positive screening for PA, but only if renin is measured as DRC (25). Then, if premenopausal Arch Endocrinol Metab. 2017;61/3

Advancing age Premenopausal women (vs. Male)

DRC: direct renin concentration; FP: false positive; FN: false negative; ARBs: angiotensin II type 1 receptor blockers; DHP: dihydropyridines.

In the Endocrine Division of the Clinics Hospital of University of Sao Paulo Medical School, we follow the algorithm for detection and confirmation of PA shown in Figure 1. First, potassium levels should be normal to adequately interpret the aldosterone-to-renin ratio. If a patient has aldosterone levels ≥ 12.5 ng/dL and A/ PRA (ng/mL/h) ratio ≥ 30 or A/DRC (mU/L) ratio ≥ 2.5, the PA screening is considered positive. In our cohort, all PA patients had the A/PRA ratio above 30 or A/DRC ratio above 2.5 when DRC was determined. As mentioned before, we always recommend the withdrawn of mineralocorticoid antagonists and diuretics for at least 4 weeks when clinically feasible. Otherwise, we perform the PA screening with all other antihypertensive medications. If aldosterone levels are < 20 ng/dL and PRA or DRC are not suppressed (but within the lower normal range), we recommend the replacement of the antihypertensive medications with verapamil, hydralazine, and α-1 blockers (doxazosin or prazosin). Clonidine can be added if a fourth medication is needed. If a patient has aldosterone levels ≥ 20 ng/dL and suppressed PRA or DRC levels, there is no need for further confirmatory testing and we can proceed with a computed tomography (CT) scan to investigate PA etiology (26). In other cases, we perform confirmatory testing to exclude false positives (Figure 1). There are several confirmatory tests and the current literature 307

DRC, it seems that A/DRC between 2 and 3 should be equivalent to A/PRA ratio of 30.

Primary aldosteronism diagnosis

does not identify a “gold standard” confirmatory test for PA (21). Specialized centers choose their preferred confirmatory test based on their own experience. We suggest starting case confirmation with the furosemide test. In our experience, the furosemide test had an accuracy higher than 90% in diagnosing PA. If the furosemide test is not conclusive and the patient doesn’t have contraindication for sodium loading, we perform the saline infusion test. The main problem with the saline infusion test is the volume overload in a short period performed in patients with severe/ resistant HTN or congestive heart failure. Currently, we only use the captopril test as a first option in patients with very impaired renal function, because of its too low reproducibility. Before performing any confirmatory test, we should check to see if potassium levels are normal and should correct hypokalemia. Below is detailed description of the three confirmatory tests mentioned above: • Furosemide test: Patients receive furosemide 40 mg iv and stay in an upright position for 2h, starting at 8-9.30 AM. Blood samples for PRA or DRC, aldosterone, and potassium are drawn at time zero and after 2 h. PRA < 2 ng/mL/h (or DRC < 24 mU/L) confirms the PA diagnosis (26). Although patients with essential hypertension can present low renin levels, plasma renin activity (PRA) increases above 2 ng/mL/h after furosemide injection. • Saline infusion test: Patients stay in the recumbent position for at least 1 h before and during the infusion of 2L of 0.9% saline iv over 4h, starting at 8-9.30 AM. Blood samples for aldosterone, PRA or DRC, and potassium are drawn at time zero and after 4 h. Aldosterone levels > 10 ng/dL confirm the diagnosis of PA, and aldosterone < 5 ng/dL excludes the diagnosis. Aldosterone levels between 5 and 10 ng/dL are considered indeterminate, although a cutoff of 6.8 ng/dL has been found to have the best trade-off between sensitivity and specificity (27). • Captopril test: Patients receive 50 mg of captopril orally after sitting or standing for at least 1 h. Blood samples are drawn for measurement of PRA or DRC, plasma aldosterone, and cortisol at time zero and at 1h and 2h after captopril, with the patient remaining seated during this period. Plasma aldosterone is normally suppressed by captopril (> 30%). In patients with PA, aldosterone level remains elevated and renin remains suppressed (26). APAs can be 308

abnormally regulated by ACTH. Then, if cortisol levels decrease during the test, we should diminish the percentage of cortisol variation from aldosterone variation to analyze only the captopril effect.

SUBTYPE CLASSIFICATION All patients with PA should undergo adrenal CT as the initial study in subtype testing to exclude adrenocortical carcinoma. Magnetic resonance imaging has no advantage over CT in subtype evaluation of PA (21). Proper distinction between APAs and bilateral hyperplasia is crucial, because the former is treated by adrenalectomy and the latter by mineralocorticoid receptor antagonists. For the diagnosis of these two subtypes, adrenal CT scan or bilateral adrenal vein sampling (AVS) is used. Adrenal CT has several limitations, because its accuracy for diagnosing APAs is limited. CT scan can reveal normal-appearing adrenals, but the patient can have a very small APA below the detection limit of CT. Moreover, nonfunctioning unilateral adrenal adenomas are not uncommon, especially in older patients (> age 40 years), and an APA can be incorrectly diagnosed in a patient with bilateral adrenal hyperplasia and normal-appearing adrenal (28). In a different situation, bilateral adrenal nodules might be interpreted as bilateral hyperplasia on the basis of CT findings, but the patient can have an APA and an adrenal incidentaloma in the contralateral side (Figure 2). The 2016 Endocrine Society Clinical Practice Guideline of PA recommends not to perform AVS in younger patients (< 35 years) with spontaneous hypokalemia, marked aldosterone excess, and unilateral adrenal lesions with radiological features consistent with a cortical adenoma on adrenal CT scan (21). In this situation, the probability of an adrenal incidentaloma is very low. At our institution, we don’t consider the age of PA diagnosis to indicate AVS but the age of HTN diagnosis, because the median time of HTN before PA diagnosis is 14 years. In other words, most of patients with APA had been diagnosed with HTN before 35 or 40 years, but PA was only diagnosed after 40 years. This trend reflects the fact that PA screening is very delayed in our country. Based on this observation, we do not recommend AVS in patients with severe PA (A ≥ 20 ng/dL, hypokalemia and suppressed DRC or PRA) with HTN diagnosed before 40 years and unilateral adrenal lesion (> 1cm) and normal contralateral adrenal on adrenal CT scan. Arch Endocrinol Metab. 2017;61/3

Primary aldosteronism diagnosis

based approaches, we need to reanalyze this issue in larger cohorts. AVS should be performed under a condition of suppressed renin, because increased renin levels by diuretics or mineralocorticoid antagonists can lead to the stimulation of the contralateral adrenal to an APA, and unilateral PA might be misclassified as bilateral. Therefore, it is important to check if DRC or PRA is suppressed before performing AVS. An experienced radiologist is required to perform AVS. AVS includes catheterization and collecting blood samples from the right and left adrenal veins and from the inferior vena cava to measure cortisol and aldosterone. The right adrenal vein may be especially difficult to catheterize because it is short and enters the inferior vena cava at an acute angle (30). We recently started to measure serum cortisol during the AVS procedure to evaluate the successful catheterization of the right adrenal vein. Currently, our rate of successful catheterization is 80%. This strategy has been previously employed to improve successful catheterization (31).

Right adrenal vein B

Left adrenal vein

• Right adrenal vein Cortisol: 701 µg/dL Aldosterone: 647 ng/dL A/C ratio: 0.923 • Left adrenal vein Cortisol: 335.2 µg/dL Aldosterone: 1890 ng/dL A/C ratio: 5.6 • Inferior vena cava Cortisol: 30.3 µg/dL Aldosterone: 41.6 ng/dL A/C ratio: 1.37 • Cortisol right adrenal vein/Inferior vena cava: 23.1 • Cortisol left adrenal vein/Inferior vena cava: 11.06 • A/C Left/Right (lateralization): 6.1 • A/C Right adrenal vein/Inferior vena cava (contralateral supression) = 0.67

Figure 2. A male patient, 69 years, with resistent hypertension for 30 years, diagnosed with PA. Hormonal data: aldosterone (A) = 24.4 ng/dL; DRC < 1.6 mU/mL (PRA 0.13); A/DRC ratio = 15.25; A/PRA ratio = 188. (A) Adrenal CT showing a 1.8 cm nodule at right adrenal gland and a 1.2 cm nodule at left adrenal gland. (B) Fluoroscopic imaging from AVS. (C) Analysis of AVS sampling showed lateralization of aldosterone production to the left side. Then, this patient had an adrenal incidentaloma on the right side and an aldosterone-producing adenoma on the left side. The patient underwent left adrenal adrenalectomy and had biochemical cure of PA. A: aldosterone; C: cortisol; A/C: aldosterone to cortisol ratio; PA: primary aldosteronism; AVS: adrenal vein sampling. Arch Endocrinol Metab. 2017;61/3

309

In a recent randomised controlled trial (SPARTACUS TRIAL), patients with PA were randomly assigned to undergo either adrenal CT or AVS to determine the presence of APA (with subsequent treatment consisting of adrenalectomy) or bilateral adrenal hyperplasia (subsequent treatment with mineralocorticoid receptor antagonists) (29). Of the 184 patients that completed follow-up, 92 received CT-based treatment (46 adrenalectomy and 46 mineralocorticoid receptor antagonist) and 92 received AVS-based treatment (46 adrenalectomy and 46 mineralocorticoid receptor antagonist). The persistence of PA was higher in patients who received CT-based treatment (20%) compared to patients who received AVS-based treatment (11%), but this difference was not statistically significant (29). This important trial showed that AVS should not be recommended for all PA patients, but it did not rule out the importance of AVS in PA subtype classification, particularly in cases with bilateral lesions or normalappearing adrenal. Although the SPARTACUS trial is the first randomized study to compare CT- vs. AVS-

Primary aldosteronism diagnosis

We perform AVS under cosyntropin infusion with sequential bilateral AVS. The use of continuous cosyntropin infusion during AVS minimizes stressinduced fluctuations in aldosterone secretion (since aldosterone can be regulated by ACTH in APAs). In addition, the cosyntropin infusion maximizes the gradient of cortisol from adrenal vein to inferior vena cava, confirming the successful sampling of the adrenal vein (32). To determine the aldosterone lateralization and contralateral suppression, we use the cortisolcorrected aldosterone ratio (aldosterone divided by cortisol level in its respective vein; A/C) to correct for dilution effects. To evaluate if the nondominant adrenal (not oversecreting aldosterone) is suppressed, we calculate the contralateral suppression (defined as the A/C ratio from the low-side to inferior vena cava). See the AVS protocol in the Clinics Hospital of Sao Paulo University Medical School below: 1) Dilute cosyntropin (250 µg) in 250 mL of saline 0.9% and start 50 mL/h (30 min before AVS); 2) Collect blood for aldosterone and cortisol measurements from right and left adrenal veins and from inferior vena cava; 3) Determine cortisol ratio between adrenal veins and the periphery (if the ratio is > 5 on each side, the catheterization was successful); 4) Determine A/C ratio to evaluate lateralization. We should calculate the A/C ratio from high-side to low-side. A ratio of more than 4:1 indicates unilateral aldosterone excess. A ratio of less than 3:1 suggests bilateral aldosterone hypersecretion. With these cutoffs, AVS for detecting unilateral aldosterone hypersecretion has a sensitivity of 95% and specificity of 100%; 5) If the lateralization ratio is between 3:1 and 4:1, the AVS is undetermined. But if contralateral suppression is < 0.67, it suggests a lateralization to the higher side (33). In Figure 2, we demonstrate a clinical case in which AVS was recommended and very useful in the subtype definition. If a patient with PA has an APA ≥ 2.5 cm or macronodular bilateral hyperplasia, screening for subclinical Cushing is recommended: 1 mg dexamethasone suppression test, ACTH, 24h urinary free cortisol, salivar cortisol and dehydroepiandrosterone sulfate.

FAMILIAL PRIMARY ALDOSTERONISM In patients with an onset of confirmed PA earlier than 20 years of age, and in those who have a family history of PA or stroke at a young age (< 40 years), the Endocrine 310

Society Clinical Practice Guideline of PA suggests genetic testing for familial PA type I (glucocorticoid remediable aldosteronism) and type III (caused by KCNJ5 germline mutations) (21) (Table 4). Table 4. Familial forms of primary aldosteronism Type I

Type III

Cause

Hybrid CYP11B1/ CYP11B2

Unknown

Germline KCNJ5

Transmission

Autosomal dominant

Genetic diagnosis

Long PCR

KCNJ5 sequencing

Hypertension onset

Very often < 20 years

Adulthood

Very often < 10 years

Hypertension severity

Severe to resistant hypertension (normal BP is rare)

Stage 1 to resistant hypertension (normal BP is not often)

Stage 3 to resistant hypertension

Hypokalemia

Rare

Not often

Very often

< 4 ng/dL

> 4 ng/dL

Adrenal CT

Normal

Unilateral or bilateral lesions

Bilateral macronodular hyperplasia

Treatment

Dexamethasone or mineralocorticoid antagonist

Unilateral adrenalectomy or mineralocorticoid antagonist

Bilateral adrenalectomy or mineralocorticoid antagonist

Aldosterone after dexamethasone

BP: blood pressure. CT: computed tomography.

Familial PA type I is caused by an unequal crossover between the genes CYP11B1 (which encodes steroid 11α-hydroxylase) and CYP11B2 (which encodes aldosterone synthase) (34,35). The resulting hybrid gene encodes an enzyme chimera with aldosterone synthase activity that is expressed in the adrenal zona fasciculata under control of ACTH instead of angiotensin II. Patients with familial PA type I show suppressed plasma aldosterone levels (< 4 ng/dL) after dexamethasone (0.5 mg every 6h during 48 or 72h) (36). Familial PA type I is treated with dexamethasone in adults (0.125-0.25 mg/d). If additional drugs are necessary to control BP or normalize renin levels, mineralocorticoid antagonist can be added (37,38). Familial PA type II is clinically and biochemically indistinguishable from sporadic forms. Prevalence was reported to be as high as 6% in a large population with PA (39). Familial PA type II is diagnosed when at least two first-degree members of the same family have confirmed PA (APA or bilateral hyperplasia). The molecular basis of FH-II is still unknown, but genetic analyses demonstrated a link with chromosome 7p22 (40). Arch Endocrinol Metab. 2017;61/3

Primary aldosteronism diagnosis

TREATMENT Cardiovascular morbidity caused by aldosterone excess can be decreased by either unilateral adrenalectomy or mineralocorticoid antagonist (43). Overall reduction of left ventricular mass has been demonstrated to be similar in unilateral adrenalectomy or mineralocorticoid antagonist treatment at the end of a 6.4 year follow-up (44). However, a study comparing both treatments in terms of cardiovascular mortality remains to be conducted. Unilateral laparoscopic adrenalectomy is indicated for patients with APAs. If the patient with an APA is unable or unwilling to undergo surgery, medical treatment including a mineralocorticoid antagonist is recommended. Before unilateral adrenalectomy, the patient should be treated with mineralocorticoid antagonist in order to normalize potassium levels and renin levels, avoiding hyporeninemic hypoaldosteronism in the postoperative period. In addition, we should measure sodium, potassium, aldosterone, and renin levels in the first week after surgery to monitor treatment response. After unilateral adrenalectomy, HTN is cured in about 50% (range of 35-80%) of patients with APA (21). Primary HTN in first-degree relatives and a longer duration of HTN before diagnosing PA are associated with low rates of HTN cure after unilateral adrenalectomy (45,46). Bilateral hyperplasia should be treated with mineralocorticoid antagonist (spironolactone or eplerenone). In Brazil, only spironolactone is available for treatment. Spironolactone (50-400 mg/d) has been the agent of choice in the medical treatment of PA, reducing BP levels as well as the need for antihypertensive drugs (47). The starting dose for spironolactone is 50 mg in a single dose. The dose should be increased by 50 mg each 3-4 weeks. During treatment, we aim to control blood pressure and to normalize potassium and renin levels. Since spironolactone antagonizes androgen receptor and inhibits androgen production, it promotes dose-dependent gynecomastia and loss of libido in Arch Endocrinol Metab. 2017;61/3

men. In females, spironolactone can cause menstrual irregularities and breast tenderness and enlargement (48). To avoid or decrease side effects, amiloride or a small dose of thiazide diuretic can be used to avoid a higher dose of spironolactone. In patients with stage III chronic kidney disease or in older patients, mineralocorticoid antagonist should be administered with caution because of the risk of hyperkalemia and worsening renal function. Eplerenone is a selective mineralocorticoid antagonist without antiandrogen effects and, therefore, is less associated with endocrine side effects and given twice daily (49). Despite its better tolerability, eplerenone has a higher cost and can be less effective than spironolactone to lower BP in the medical treatment of PA (50). In the next years, it is likely that new mineralocorticoid antagonists and selective aldosterone synthase inhibitors will be available to treat PA (21). Disclosure: no potential conflict of interest relevant to this article was reported.

REFERENCES 1. Murray CJ, Lopez AD. Measuring the global burden of disease. N Engl J Med. 2013;369(5):448-57. 2. Egan BM, Zhao Y, Axon RN. US trends in prevalence, awareness, treatment, and control of hypertension, 1988-2008. JAMA. 2010;303(20):2043-50. 3. Stewart PM. Mineralocorticoid hypertension. Lancet. 1999; 353(9161):1341-7. 4. Lifton RP. Molecular genetics of human blood pressure variation. Science. 1996;272(5262):676-80. 5. Vasan RS, Evans JC, Larson MG, Wilson PW, Meigs JB, Rifai N, et al. Serum aldosterone and the incidence of hypertension in nonhypertensive persons. N Engl J Med. 2004;351(1):33-41. 6. Guder G, Bauersachs J, Frantz S, Weismann D, Allolio B, Ertl G, et al. Complementary and incremental mortality risk prediction by cortisol and aldosterone in chronic heart failure. Circulation. 2007;115(13):1754-61. 7. Tsutamoto T, Sakai H, Tanaka T, Fujii M, Yamamoto T, Wada A, et al. Comparison of active renin concentration and plasma renin activity as a prognostic predictor in patients with heart failure. Circ J. 2007;71(6):915-21. 8. Beygui F, Collet JP, Benoliel JJ, Vignolles N, Dumaine R, Barthelemy O, et al. High plasma aldosterone levels on admission are associated with death in patients presenting with acute STelevation myocardial infarction. Circulation. 2006;114(24):2604-10. 9. Tylicki L, Larczynski W, Rutkowski B. Renal protective effects of the renin-angiotensin-aldosterone system blockade: from evidence-based approach to perspectives. Kidney Blood Press Res. 2005;28(4):230-42. 10. Hannemann A, Wallaschofski H. Prevalence of primary aldosteronism in patientâ&#x20AC;&#x2122;s cohorts and in population-based studies--a review of the current literature. Horm Metab Res. 2012;44(3):157-62. 11. Plouin PF, Amar L, Chatellier G. Trends in the prevalence of primary aldosteronism, aldosterone-producing adenomas, and surgically correctable aldosterone-dependent hypertension. Nephrol Dial Transplant. 2004;19(4):774-7.

311

Familial PA type III is caused by mutations in the KCNJ5 gene encoding the potassium channel Kir 3.4, resulting in increased sodium conductance and cell depolarization (41). Familial PA type III is characterized by severe HTN in early childhood associated with PA, hypokalemia, and macronodular bilateral hyperplasia (42). Because of the HTN severity, bilateral adrenalectomy may be needed to control BP (36).

Primary aldosteronism diagnosis

12. Douma S, Petidis K, Doumas M, Papaefthimiou P, Triantafyllou A, Kartali N, et al. Prevalence of primary hyperaldosteronism in resistant hypertension: a retrospective observational study. Lancet. 2008;371(9628):1921-6. 13. Calhoun DA, Nishizaka MK, Zaman MA, Thakkar RB, Weissmann P. Hyperaldosteronism among black and white subjects with resistant hypertension. Hypertension. 2002;40(6):892-6. 14. Calhoun DA, Jones D, Textor S, Goff DC, Murphy TP, Toto RD, et al. Resistant hypertension: diagnosis, evaluation, and treatment: a scientific statement from the American Heart Association Professional Education Committee of the Council for High Blood Pressure Research. Circulation. 2008;117(25):e510-26. 15. Milliez P, Girerd X, Plouin PF, Blacher J, Safar ME, Mourad JJ. Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism. J Am Coll Cardiol. 2005;45(8):1243-8. 16. Stowasser M, Sharman J, Leano R, Gordon RD, Ward G, Cowley D, et al. Evidence for abnormal left ventricular structure and function in normotensive individuals with familial hyperaldosteronism type I. J Clin Endocrinol Metab. 2005;90(9):5070-6. 17. Savard S, Amar L, Plouin PF, Steichen O. Cardiovascular complications associated with primary aldosteronism: a controlled cross-sectional study. Hypertension. 2013;62(2):331-6. 18. Mulatero P, Monticone S, Bertello C, Viola A, Tizzani D, Iannaccone A, et al. Long-term cardio- and cerebrovascular events in patients with primary aldosteronism. J Clin Endocrinol Metab. 2013;98(12):4826-33. 19. Fallo F, Veglio F, Bertello C, Sonino N, Della Mea P, Ermani M, et al. Prevalence and characteristics of the metabolic syndrome in primary aldosteronism. J Clin Endocrinol Metab. 2006;91(2):454-9. 20. Chen W, Li F, He C, Zhu Y, Tan W. Elevated prevalence of abnormal glucose metabolism in patients with primary aldosteronism: a meta-analysis. Ir J Med Sci. 2014;183(2):283-91. 21. Funder JW, Carey RM, Mantero F, Murad MH, Reincke M, Shibata H, et al. The Management of Primary Aldosteronism: Case Detection, Diagnosis, and Treatment: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2016;101(5):1889-916. 22. Rossi GP, Bernini G, Caliumi C, Desideri G, Fabris B, Ferri C, et al. A prospective study of the prevalence of primary aldosteronism in 1,125 hypertensive patients. J Am Coll Cardiol. 2006;48(11):2293-300. 23. Funder JW, Carey RM, Fardella C, Gomez-Sanchez CE, Mantero F, Stowasser M, et al. Case detection, diagnosis, and treatment of patients with primary aldosteronism: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2008;93(9):3266-81. 24. Rossi GP, Ceolotto G, Rossitto G, Seccia TM, Maiolino G, Berton C, et al. Prospective validation of an automated chemiluminescencebased assay of renin and aldosterone for the work-up of arterial hypertension. Clin Chem Lab Med. 2016;54(9):1441-50. 25. Ahmed AH, Gordon RD, Taylor PJ, Ward G, Pimenta E, Stowasser M. Effect of contraceptives on aldosterone/renin ratio may vary according to the components of contraceptive, renin assay method, and possibly route of administration. J Clin Endocrinol Metab. 2011;96(6):1797-804. 26. Nanba K, Tamanaha T, Nakao K, Kawashima ST, Usui T, Tagami T, et al. Confirmatory testing in primary aldosteronism. J Clin Endocrinol Metab. 2012;97(5):1688-94. 27. Rossi GP, Belfiore A, Bernini G, Desideri G, Fabris B, Ferri C, et al. Prospective evaluation of the saline infusion test for excluding primary aldosteronism due to aldosterone-producing adenoma. J Hypertens. 2007;25(7):1433-42. 28. Kloos RT, Gross MD, Francis IR, Korobkin M, Shapiro B. Incidentally discovered adrenal masses. Endocr Rev. 1995;16(4):460-84. 29. Dekkers T, Prejbisz A, Kool LJ, Groenewoud HJ, Velema M, Spiering W, et al. Adrenal vein sampling versus CT scan to determine treatment in primary aldosteronism: an outcome-based randomised diagnostic trial. Lancet Diabetes Endocrinol. 2016;4(9):739-46. 30. Doppman JL, Gill JR, Jr. Hyperaldosteronism: sampling the adrenal veins. Radiology. 1996;198(2):309-12. 31. Yoneda T, Karashima S, Kometani M, Usukura M, Demura M, Sanada J, et al. Impact of New Quick Gold Nanoparticle-Based Cortisol Assay During Adrenal Vein Sampling for Primary Aldosteronism. J Clin Endocrinol Metab. 2016;101(6):2554-61.

312

32. Rossi GP, Barisa M, Allolio B, Auchus RJ, Amar L, Cohen D, et al. The Adrenal Vein Sampling International Study (AVIS) for identifying the major subtypes of primary aldosteronism. J Clin Endocrinol Metab. 2012;97(5):1606-14. 33. El Ghorayeb N, Mazzuco TL, Bourdeau I, Mailhot JP, Zhu PS, Therasse E, et al. Basal and Post-ACTH Aldosterone and Its Ratios Are Useful During Adrenal Vein Sampling in Primary Aldosteronism. J Clin Endocrinol Metab. 2016;101(4):1826-35. 34. Sutherland DJ, Ruse JL, Laidlaw JC. Hypertension, increased aldosterone secretion and low plasma renin activity relieved by dexamethasone. Can Med Assoc J. 1966;95(22):1109-19. 35. Lifton RP, Dluhy RG, Powers M, Rich GM, Cook S, Ulick S, et al. A chimaeric 11 beta-hydroxylase/aldosterone synthase gene causes glucocorticoid-remediable aldosteronism and human hypertension. Nature. 1992;355(6357):262-5. 36. Mulatero P, Monticone S, Rainey WE, Veglio F, Williams TA. Role of KCNJ5 in familial and sporadic primary aldosteronism. Nat Rev Endocrinol. 2013;9(2):104-12. 37. Dluhy RG, Anderson B, Harlin B, Ingelfinger J, Lifton R. Glucocorticoid-remediable aldosteronism is associated with severe hypertension in early childhood. J Pediatr. 2001;138(5):715-20. 38. Dluhy RG, Lifton RP. Glucocorticoid-remediable aldosteronism. J Clin Endocrinol Metab. 1999;84(12):4341-4. 39. Pallauf A, Schirpenbach C, Zwermann O, Fischer E, Morak M, HolinskiFeder E, et al. The prevalence of familial hyperaldosteronism in apparently sporadic primary aldosteronism in Germany: a single center experience. Horm Metab Res. 2012;44(3):215-20. 40. Lafferty AR, Torpy DJ, Stowasser M, Taymans SE, Lin JP, Huggard P, et al. A novel genetic locus for low renin hypertension: familial hyperaldosteronism type II maps to chromosome 7 (7p22). J Med Genet. 2000;37(11):831-5. 41. Scholl UI, Nelson-Williams C, Yue P, Grekin R, Wyatt RJ, Dillon MJ, et al. Hypertension with or without adrenal hyperplasia due to different inherited mutations in the potassium channel KCNJ5. Proc Natl Acad Sci U S A. 2012;109(7):2533-8. 42. Geller DS, Zhang J, Wisgerhof MV, Shackleton C, Kashgarian M, Lifton RP. A novel form of human mendelian hypertension featuring nonglucocorticoid-remediable aldosteronism. J Clin Endocrinol Metab. 2008;93(8):3117-23. 43. Rossi GP, Bolognesi M, Rizzoni D, Seccia TM, Piva A, Porteri E, et al. Vascular remodeling and duration of hypertension predict outcome of adrenalectomy in primary aldosteronism patients. Hypertension. 2008;51(5):1366-71. 44. Catena C, Colussi G, Lapenna R, Nadalini E, Chiuch A, Gianfagna P, et al. Long-term cardiac effects of adrenalectomy or mineralocorticoid antagonists in patients with primary aldosteronism. Hypertension. 2007;50(5):911-8. 45. Celen O, O’Brien MJ, Melby JC, Beazley RM. Factors influencing outcome of surgery for primary aldosteronism. Arch Surg. 1996;131(6):646-50. 46. Sawka AM, Young WF, Thompson GB, Grant CS, Farley DR, Leibson C, et al. Primary aldosteronism: factors associated with normalization of blood pressure after surgery. Ann Intern Med. 2001;135(4):258-61. 47. Wambach G, Helber A, Bonner G, Hummerich W, Meurer KA, Kaufmann W. [Spironolactone in essential hypertension associated with abnormal aldosterone regulation and in Conn’s syndrome (author’s transl)]. Dtsch Med Wochenschr. 1980;105(18):647-51. 48. Jeunemaitre X, Chatellier G, Kreft-Jais C, Charru A, DeVries C, Plouin PF, et al. Efficacy and tolerance of spironolactone in essential hypertension. Am J Cardiol. 1987;60(10):820-5. 49. Burgess ED, Lacourciere Y, Ruilope-Urioste LM, Oparil S, Kleiman JH, Krause S, et al. Long-term safety and efficacy of the selective aldosterone blocker eplerenone in patients with essential hypertension. Clin Ther. 2003;25(9):2388-404. 50. Parthasarathy HK, Menard J, White WB, Young WF, Jr., Williams GH, Williams B, et al. A double-blind, randomized study comparing the antihypertensive effect of eplerenone and spironolactone in patients with hypertension and evidence of primary aldosteronism. J Hypertens. 2011;29(5):980-90. Arch Endocrinol Metab. 2017;61/3

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

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.

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

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

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.

XigDuo XRTM (dapagliflozina + cloridrato de metformina) comprimidos revestidos de liberação prolongada. Indicações: XIGDUO XR é indicado como adjuvante à dieta e exercícios para melhorar o controle glicêmico em adultos com diabetes mellitus tipo 2 quando o tratamento com ambos, dapagliflozina e metformina, é apropriado. XIGDUO XR não é indicado para uso em pacientes com diabetes tipo 1. XIGDUO XR não deve ser usado para o tratamento da cetoacidose diabética. Contraindicações: doença ou disfunção renal moderada a grave (p.ex., níveis de creatinina sérica ≥1,5 mg/dL [homens], ≥1,4 mg/dL [mulheres] ou TFGe <60 mL/min/1,73 m2 ou ClCr <60 mL/min pelo Cockcroft-Gault), inclusive secundária a condições como choque, IAM e septicemia; acidose metabólica aguda ou crônica, incluindo cetoacidose diabética, com ou sem coma, que deve ser tratada com insulina; história de reação de hipersensibilidade grave à substância ativa ou a qualquer um dos excipientes; disfunção hepática. Cuidados e Advertências: acidose láctica (metformina plasmática > 5 µg/mL - maior risco em idosos, disfunção renal, doença hepática, insuficiência cardíaca congestiva, hipoxemia, desidratação, sepse, ingestão excessiva de álcool e uso de contraste intravascular), disfunção renal, disfunção hepática, ingestão excessiva de álcool, cetoacidose (maior risco em disfunções pancreáticas como DM1, pancreatite, cirurgia pancreática, redução da dose de insulina, redução da ingestão calórica, infecções, cirurgias, doenças concomitante e abuso de álcool), níveis de vitamina B12 (risco de redução em pacientes susceptíveis), procedimentos cirúrgicos, alterações no estado clínico, medicações concomitantes que afetem a função renal ou a hemodinâmica ou a eliminação da metformina, administração de meio de contraste intravascular iodado (aumento do risco de insuficiência renal aguda), estados de hipóxia (choque, ICC, IAM, insuficiência renal pré-renal), mau controle glicêmico secundário a febre, trauma, infecção ou cirurgias, pacientes sob risco de depleção de volume intravascular (idosos, uso de diuréticos), uso concomitante com medicamentos que causam hipoglicemia (insulina e sulfonilureias), sepse urinária e pielonefrite, uso em idosos, gravidez, lactação, uso pediátrico, câncer de bexiga ativo. Categoria de Risco na Gravidez: C. Interações Medicamentosas: com dapagliflozina (sem alterações clínicas relevantes, sem necessidade de ajuste de dose): bumetanida, sinvastatina, rifampicina, ácido mefenâmico; com metformina: medicamentos catiônicos (cimetidina), glibenclamida, furosemida, nifedipino; outros medicamentos hiperglicemiantes (tiazidas e outros diuréticos, corticosteroides, fenotiazinas, produtos da tireoide, estrógenos, contraceptivos orais, fenitoína, ácido nicotínico, simpatomiméticos, medicamentos bloqueadores do canal de cálcio e isoniazida). Interferência com teste do 1,5-anidroglucitol (1,5.AG). Reações Adversas: infecção genital, infecção do trato urinário, poliúria, dor nas costas, dor de cabeça, hipoglicemia, desidratação, hipovolemia ou hipotensão, diarreia, náuseas, vômitos, erupção cutânea, redução dos níveis séricos de vitamina B12, aumento do hematócrito. Posologia: deve ser individualizada com base no regime atual do paciente, desde que não exceda a dose máxima recomendada de 10 mg de dapagliflozina e de 2000 mg de cloridrato de metformina de liberação prolongada. XIGDUO XR deve, de modo geral, ser administrado uma vez ao dia com a refeição da noite. Apresentações: XigDuo XR comprimidos revestidos de liberação prolongada de: 5 mg/1000 mg em embalagens com 14 e 60 comprimidos; 10 mg/500 mg em embalagens com 14 comprimidos e 10 mg/1000 mg em embalagens com 14 e 30 comprimidos. USO ADULTO. USO ORAL. VENDA SOB PRESCRIÇÃO MÉDICA. SE PERSISTEREM OS SINTOMAS, O MÉDICO DEVERÁ SER CONSULTADO. Para maiores informações, consulte a bula completa do produto. www.astrazeneca.com.br. Reg. MS – 1.0180.0407 (XIG006_min).

FORXIGA® – (dapagliflozina) comprimidos revestidos. Indicações: FORXIGA é indicado como adjuvante a dieta e exercícios para melhora do controle glicêmico em pacientes com diabetes mellitus tipo 2 em monoterapia ou em combinação com metformina; tiazolidinediona; sulfonilureia; inibidor da DPP4 (com ou sem metformina); ou insulina (isolada ou com até duas medicações antidiabéticas orais), quando a terapia existente juntamente com dieta e exercícios não proporciona controle glicêmico adequado. Indicado em combinação inicial com metformina quando ambas as terapias são apropriadas. FORXIGA não é indicado para uso por pacientes com diabetes tipo 1 e não deve ser utilizado para o tratamento de cetoacidose diabética Contraindicações: hipersensibilidade a dapagliflozina ou aos outros componentes da fórmula. Advertências e Precauções: Foram reportados alguns relatos pós-comercialização de cetoacidose em pacientes diabéticos tipo 1 e tipo 2 em uso de FORXIGA. Embora uma relação causal ainda não tenha sido estabelecida, recomenda-se que pacientes que apresentem sinais de cetoacidose incluindo náusea, vômitos, dor abdominal, prostração ou dispneia sejam avaliados quanto a presença de cetoacidose, mesmo que sua glicemia esteja menor que 250 mg/dL. FORXIGA® deve ser usado com cautela ou ser temporariamente suspenso em pacientes sob risco de depleção de volume, pacientes com hipertensão ou outra doença cardiovascular, infecções do trato urinário, incluindo urosepse e pielonefrite, uso concomitante com medicamentos que podem causar hipoglicemia, gravidez, lactação uso pediátrico, uso geriátrico. Categoria de Risco na Gravidez: C. Reações Adversas: infecção genital, infecção do trato urinário, dor nas costas, poliúria e erupção cutânea. Interações Medicamentosas: (sem alterações clínicas relevantes, sem necessidade de ajuste de dose) metformina, pioglitazona, sitagliptina, glimepirida, voglibose, hidroclorotiazida, bumetanida, valsartana, sinvastatina, rifampicina, ácido mefenâmico. Outras interações: os efeitos da dieta, tabagismo, produtos à base de plantas e uso de álcool sobre a farmacocinética da dapagliflozina não foram especificamente estudados. Interferência com o teste 1,5-anidroglucitol (1,5-AG). Posologia: a dose recomendada de FORXIGA, em monoterapia ou terapia combinada, é 10 mg, uma vez ao dia, a qualquer hora do dia, independentemente das refeições. Para pacientes em risco de depleção de volume devido a condições coexistentes, uma dose inicial de 5 mg de FORXIGA pode ser apropriada. Não são necessários ajustes de dose de FORXIGA com base na função renal ou hepática. Apresentações: embalagens com 30 comprimidos revestidos de 5 mg e embalagens com 14 ou 30 comprimidos revestidos de 10 mg. USO ORAL. USO ADULTO. VENDA SOB PRESCRIÇÃO MÉDICA. SE PERSISTIREM OS SINTOMAS, O MÉDICO DEVERÁ SER CONSULTADO. Para maiores informações, consulte a bula completa do produto. Reg. MS - 1.0180.0404 (FRX013_min).

CONTRAINDICAÇÕES: FORXIGA® é contraindicado a pacientes com conhecida hipersensibilidade à dapagliflozina ou aos outros componentes da fórmula. INTERAÇÕES MEDICAMENTOSAS: em estudos realizados em indivíduos sadios, a farmacocinética CONTRAINDICAÇÕES: doença renal ou disfunção renal da dapagliflozina não foi alterada pela metformina, moderada a grave. INTERAÇÃO MEDICAMENTOSA: pioglitazona, sitagliptina, glimepirida, voglibose, hidroclorotiazida, bumetanida, valsartana ou sinvastatina. cimetidina.

REFERÊNCIA BIBLIOGRÁFICA: 1. XIGDUO® (dapagliflozina+metformina XR) comprimidos [bula do medicamento]. São Paulo, SP. Bristol-Myers Squibb Farmacêutica S.A.; 2016. 2. Forxiga® (dapagliflozina) comprimidos [bula do medicamento]. São Paulo, SP. BristolMyers Squibb Farmacêutica S.A.; 2016. 3. Del Prato S, Nauck M, Durán-Garcia S, Maffei L, Rohwedder K, Theuerkauf A, Parikh S. Long-term glycaemic response and tolerability of dapagliflozin versus a sulphonylurea as add-on therapy to metformin in patients with type 2 diabetes: 4-year data. Diabetes, Obesity and Metabolism 17: 581–590, 2015. 4. Bolinder J, Lunggren O, Johansson L, Wildong J, Langkilde AM, Sjoström CD, Sugg J, Parikh S. Dapagliflozin maintains glycaemic control while reducing weight and body fat mass over 2 years in patients with type 2 diabetes mellitus inadequately controlled on metformin. Diabetes Obes Metab. 2014;16(2):159-169. 5. Parikh S, Wilding J, Jabbour S, Hardy E. Dapagliflozin in type 2 diabetes: effectiveness across the spectrum of disease and over time. Int J Clin Pract, February 2015, 69, 2, 186–198. 6. Bailey CJ, Gross JL, Pieters A, Bastien A, List JF. Effect of dapagliflozin in patients with type 2 diabetes who have inadequate glycaemic control with metformin: a randomized, double-blind, placebo-controlled trial. Lancet. 2010;375(9733):2223-2233. 7. Bailey CJ, Gross JL, Hennicken D, Iqdal N, Mansfield TA, List JF. Dapagliflozin add-on to metformin in type 2 diabetes inadequately controlled with metformin: randomized, double-blind, placebo-controlled 102-week trial. BMC Med. 2013;11:43. 8. Henry RR, Murray AV, Marmolejo MH, Hennicken D, Ptaszynska A, List JF. Dapagliflozin, metformin XR, or both: initial pharmacotherapy for type 2 diabetes, a randomised controlled trial. Int J Clin Pract, May 2012, 66, 5, 446–456. 9. Rosenstock J, Vico M, Wei L, Salsali A, List JF. Effects of dapagliflozin, an SGLT2 inhibitor, on HbA(1c), body weight, and hypoglycemia risk in patients with type 2 diabetes inadequately controlled on pioglitazone monotherapy. Diabetes Care. 2012;35(7):1473-1478. 10. Resolução - RE No- 2.234, de 26 de Outubro de 2015. XIGDUO XR: registro de medicamento novo. Diario Oficial da União 2015:Supl(204):pp43. 11. Bangalore S, Kamalakkannan G, Parkar S, Messerli FH. Fixed-Dose Combinations Improve Medication Compliance: A Meta-Analysis. Am J of Med 2007;120:713-719 Material destinado ao profissional de saúde. 1629647 – Produzido em março/2017