CLINICS September 2017 by Clinics

Volume 72 Number 9 - September/2017

CLINICS Editor Edmund Chada Baracat

Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Area Editors Ana Maria de Ulhoa Escobar Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Anna Sara Shafferman Levin Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Antonio Egidio Nardi Universidade Federal do Rio de Janeiro Rio de Janeiro, RJ, Brazil Anuar Ibrahim Mitre Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Berenice Bilharinho Mendonca Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Bruno Zilberstein Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Carlos Serrano Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Carmen Silvia Valente Barbas Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Claudia Regina Furquim de Andrade Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Claudio Roberto Cernea Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Emilia Inoue Sato Universidade Federal de São Paulo São Paulo, SP, Brazil Flair José Carrilho Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Fulvio Alexandre Scorza Universidade Federal de São Paulo São Paulo, SP, Brazil Geraldo Busatto Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Heitor Franco de Andrade Jr. Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Heloisa de Andrade Carvalho Hospital das Clı´nicas da Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Jesus Paula Carvalho Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Joaquim Prado Moraes-Filho Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil José Guilherme Cecatti Universidade Estadual de Campinas Campinas, SP, Brazil José Maria Soares Júnior Hospital das Clı´nicas da Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Luiz Eugenio Garcez-Leme Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Luíz Fernando Onuchic Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Lydia Masako Ferreira Universidade Federal de São Paulo São Paulo, SP, Brazil Marcos Intaglietta University of California, San Diego San Diego, CA, USA Maria José Carvalho Carmona Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Mauricio Etchebehere Universidade Estadual de Campinas Campinas, SP, Brazil Michele Correale University of Foggia Foggia, Italy Naomi Kondo Nakagawa Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Nelson Wolosker Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Newton Kara-Junior Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Olavo Pires de Camargo Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Paulo Hoff Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Paulo Pêgo-Fernandes Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Raul Coimbra University of California, San Diego La Jolla, CA, USA Renato Delascio Lopes Duke University Medical Center Durham, NC, USA Ricardo Bassil Lasmar Universidade Federal Fluminense Niterói, RJ, Brazil Ricardo Nitrini Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Rosa Maria Rodrigues Pereira Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Rossana Francisco Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Rubens Belfort Jr. Universidade Federal de São Paulo São Paulo, SP, Brazil Ruth Guinsburg Universidade Federal de São Paulo São Paulo, SP, Brazil Ruy Jorge Cruz Junior University of Pittsburgh Pittsburgh, PA, USA Sandro Esteves ANDROFERT - Andrology & Human Reproduction Clinic Campinas, SP, Brazil Sergio Paulo Bydlowski Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Sigmar de Mello Rode Universidade Estadual Paulista Júlio de Mesquita Filho São José dos Campos, SP, Brazil Simone Appenzeller Universidade Estadual de Campinas Campinas, SP, Brazil Valeria Aoki Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Editorial Board Abhijit Chandra King George’s Medical College Lucknow, India Adamastor Humberto Pereira Universidade Federal do Rio Grande do Sul Porto Alegre, RS, Brazil Adauto Castelo Universidade Federal de São Paulo São Paulo, SP, Brazil Ademar Lopes Fundação Antônio Prudente, Hospital do Câncer São Paulo, SP, Brazil Alberto Azoubel Antunes Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Alexandre Roberto Precioso Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Andrea Schmitt University of Goettingen Goettingen, Germany

Arnaldo Valdir Zumiotti Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Artur Brum-Fernandes Université de Sherbrooke Québec, Canadá Carmita Helena Najjar Abdo Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Cesar Gomes Victora Faculdade de Medicina da Universidade Federal de Pelotas Pelotas, RS, Brasil Daniel Romero Muñoz Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Eduardo Ferreira Borba Neto Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Edmund Neugebauer Witten/Herdecke University Witten, North Rhine - Westphalia, Germany

Egberto Gaspar de Moura Jr. Universidade do Estado do Rio de Janeiro Rio de Janeiro, RJ, Brazil Ernest Eugene Moore University of Colorado Denver Denver, CO, USA Euclides Ayres Castilho Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Fábio Biscegli Jatene Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Francisco Laurindo Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Hiroyuki Hirasawa Chiba University School of Medicine Chiba, Japan Irismar Reis de Oliveira Faculdade de Medicina da Universidade Federal da Bahia Salvador, BA, Brasil

Irshad Chaudry University of Alabama Birmingham, AL, USA Ivan Cecconello Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Ke-Seng Zhao Southern Medical University Guangzhou, China Laura Cunha Rodrigues London School of Hygiene and Tropical Medicine - University of London London, UK Marcelo Zugaib Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Marco Martins Amatuzzi Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Maria Aparecida Shikanai Yasuda Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Mauro Perretti William Harvey Research Institute London, UK

Noedir Antonio Groppo Stolf Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Michael Gregory Sarr Mayo Clinic Rochester, MN, USA

Pedro Puech-Leão Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Peter Libby Brigham and Women’s Hospital Boston, Boston, MA, USA Philip Cohen University of Houston Health Center Houston, Texas, USA Rafael Andrade-Alegre Santo Tomás Hospital Republic of Panamai, Panamá Ricardo Antonio Reﬁnetti Faculdade de Medicina da Universidade Federal do Rio de Janeiro Rio de Janeiro, RJ, Brazil Roberto Chiesa San Raffaele Hospital Milan, Italy

Milton de Arruda Martins Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Mitchell C. Posner The University of Chicago Medical Center Chicago, IL, USA Moyses Szklo Johns Hopkins Bloomberg School of Public Health Baltimore, USA Naomi Kondo Nakagawa Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Navantino Alves Faculdade de Ciências Médicas de Minas Gerais Belo Horizonte, MG, Brazil

Samir Rasslan Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Tarcisio Eloy Pessoa de Barros Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Valentim Gentil Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Wagner Farid Gattaz Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Board of Governors Alberto José da Silva Duarte Aluisio Augusto Cotrim Segurado Ana Claudia Latronico Xavier Berenice Bilharinho de Mendonça Carlos Alberto Buchpiquel Carlos Roberto Ribeiro de Carvalho Clarice Tanaka Claudia Regina Furquim de Andrade Cyro Festa Daniel Romero Muñoz Edivaldo Massazo Utiyama Edmund Chada Baracat Eduardo Massad Eloisa Silva Dutra de Oliveira Bonfá Euripedes Constantino Miguel Fábio Biscegli Jatene Flair José Carrilho Geraldo Busatto Gerson Chadi Gilberto Luis Camanho Giovanni Guido Cerri Irene de Lourdes Noronha Irineu Tadeu Velasco

Ivan Cecconello Jorge Elias Kalil José Antonio Franchini Ramires José Antonio Sanches José Eduardo Krieger José Otávio Costa Auler José Ricardo de Carvalho Mesquita Ayres Linamara Rizzo Battistella Luiz Augusto Carneiro D’Albuquerque Luiz Fernando Onuchic Magda Maria Sales Carneiro-Sampaio Manoel Jacobsen Teixeira Marcelo Zugaib Miguel Srougi Milton de Arruda Martins Mirian Nacagami Sotto Nelson de Luccia Olavo Pires de Camargo Paulo Andrade Lotufo Paulo Hilário Nascimento Saldiva Paulo Manuel Pêgo Fernandes Paulo Marcelo Gehm Hoff Paulo Rossi Menezes

Editorial Director

Kavita Kirankumar Patel-Rolim Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Pedro Puech-Leão Remo Susanna Ricardo Ferreira Bento Ricardo Nitrini Roberto Kalil Roberto Zatz Roger Chammas Rolf Gemperli Rosa Maria Rodrigues Perreira Sandra Joseﬁna Ferraz Ellero Grisi Selma Lancman Tarcísio Eloy Pessoa de Barros Uenis Tannuri Umbertina Conti Reed Valentim Gentil Vanderson Geraldo Rocha Venâncio Avancini Ferreira Alves Vicente Odone Wagner Farid Gattaz Werther Brunow de Carvalho William Carlos Nahas Wilson Jacob

Editorial Assistants

Nair Gomes Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil Daniela Aquemi Higa Faculdade de Medicina da Universidade de São Paulo São Paulo, SP, Brazil

Editorial Office: Rua Dr. Ovídio Pires de Campos, 225 - 6˚ Andar CEP 05403-010 São Paulo/SP Tel.: +55-11-2661-6235 Email: clinics@hc.fm.usp.br Website: www.scielo.br/clinics Submission: http://mc04.manuscriptcentral.com/clinics-scielo Indexations: LILACS; MEDLINE; PubMed; PubMed Central; SciELO; Science Citation Index Expanded (ISI Web of Knowledge); Scopus; Ulrich’s Periodical Directory; Qualis/Capes - Classified as an International Circulation Journal in Medicine. Clinics. São Paulo: Scientific Journal of Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, 2005Monthly Periodical: January to December ISSN 1807-5932 printed version ISSN 1980-5322 online version Formerly Revista do Hospital das Clínicas da FMUSP, 1946–2004. 1. Medicine-scientific production. 2.Medical Sciences I. Hospital das Clínicas da Faculdade deMedicina da Universidade de São Paulo. CDD 610

PUBLICATION INFORMATION AND EDITORIAL POLICIES CLINICS publishes peer-reviewed articles of interest to clinicians and researchers in the medical sciences. CLINICS is registered with PubMed Central (http://www.ncbi.nlm.nih.gov/pmc/journals/ 795/) and SciELO (http://www.scielo.br/clinics) and complies with the policies of funding agencies which request or require deposition of the published articles that they fund into publicly available databases. CLINICS supports the position of the International Committee of Medical Journal Editors (ICMJE) on trial registration. Visit http://www.icmje.org for further details and http://www.who.int/ictrp/network/en/ for the WHOâ&#x20AC;&#x2122;s list of approved registries. Please make sure to submit your manuscript in the exact format that is described below. Failure to do so will cause the submission to be returned to you during the preliminary examination by the Editorial OfďŹ ce. - Manuscripts involving human subjects or the use of laboratory animals must clearly state adherence to appropriate guidelines and approval of protocols by their institutional review boards. Photographs that may identify patients or other human participants of studies shall be acceptable only when a legally valid consent form is signed by the participating patient, other human participant, or his/her legally constituted representative. - Authors are strongly advised to use abbreviations sparingly whenever possible to avoid jargon and improve readability of the manuscript. All abbreviations must be defined the first time that they are used. Only terms or expressions that are used at least 5 times throughout the text should be abbreviated. Never use abbreviations that spell common English words, such as FUN, PIN, SCORE, and SUN.

clinical research. Original studies must conform to the following format: Title: Up to 250 characters. Running title: Up to 40 characters. Authors' full names and contribution: Authors should

PROFESSIONAL ENGLISH REVISION Please keep in mind that ALL accepted manuscripts must be sent by the authors to the company American Journal Experts (https://www.aje.com) for premium editing prior to publication.

Publication FEES CLINICS uses a business model whereby expenses are partly recovered by charging the authors, or research sponsors, a publication fee for each published article. We have updated our 2016 prices, effective for articles submitted from 05/01/2016 onwards, as follows: original articles, review articles and rapid communications: US$ 1,500.00 (International authors) and R$ 1.500,00 (Brazilian authors). There are no charges for Invited reviews, editorials and letters to the editors. Waiver Policy: Most institutions and funding agencies have set up funding sources to cover Article Publication Costs. Unfortunately CLINICS cannot grant fee waivers due to its 2016 policies.

MANUSCRIPT CATEGORIES ORIGINAL STUDY: Complete original studies should be submitted in this category. Two sections are offered: basic and

have participated sufficiently in the work to take public responsibility for appropriate portions of the content. Such participation must be declared in this section of the manuscript. Any financial or other relationships that may lead to a conflict of interest. Abstract: Abstracts are limited to 250 words and structured into objectives, method, results, and conclusions. Citations or abbreviations (except internationally recognized abbreviations, such as weights, measures, and physical or chemical abbreviations) are not permitted. Authors are strongly encouraged not to display numerical statistical information but to merely state what is significantly different (or not) between the described parameters. Please make sure to always include your Abstract in the main document file. Keywords: For keywords, 3â&#x20AC;&#x201C;6 items from the Medical Subject Headings (MeSh) should be used. Introduction: The introduction should set the purpose of the study, provide a brief summary (not a review) of previous relevant studies, and state the new advances in the current investigation. The introduction should not include data or conclusions from the work being reported. A final sentence summarizing the novel finding to be presented is permissible. Materials and Methods: This section should briefly give clear and sufficient information to permit the study to be repeated by others. Standard techniques only need to be referenced. Previously published methods may be briefly described following the reference. Ethics: When reporting experiments on human subjects, indicate whether the procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional or regional) and with the Helsinki Declaration of 1975, which was revised in 1983. When reporting experiments on animals, indicate whether the institutionâ&#x20AC;&#x2122;s guide, a national research councilâ&#x20AC;&#x2122;s guide, or any national law on the care and use of laboratory animals was followed. Results: The results section should be a concise account of the new information that was discovered, with the least personal judgment. Do not repeat in text all the data in the tables and illustrations but briefly describe what these data comprise. Discussion: The discussion should include the significance of the new information and relevance of the new findings in light of existing knowledge. Only unavoidable citations should be included. Citation to review articles are not encouraged in this section. Acknowledgments: This section should be short, concise, and restricted to acknowledgments that are necessary.

References in text: CLINICS adopts the Vancouver

format. Cite references in the text using Arabic numerals in the order of appearance, within parentheses, (1) after the previous word, with spacing as in this example: â&#x20AC;&#x2DC;â&#x20AC;&#x2DC;Diabetes (2), hypertension (3,4) and alcoholism (5â&#x20AC;&#x201C;9) are complex medical problems (10).â&#x20AC;&#x2122;â&#x20AC;&#x2122; Under exceptional circumstances, authorsâ&#x20AC;&#x2122; names may appear in text: Single author: â&#x20AC;&#x2DC;â&#x20AC;&#x2DC;Einstein (11) proposed a new theory â&#x20AC;Śâ&#x20AC;&#x2122;â&#x20AC;&#x2122;, Two authors: â&#x20AC;&#x2DC;â&#x20AC;&#x2DC;Watson & Crick (12) reported on the structure of â&#x20AC;Śâ&#x20AC;&#x2122;â&#x20AC;&#x2122;, or Three or more authors: â&#x20AC;&#x2DC;â&#x20AC;&#x2DC;Smith et al. (13) described â&#x20AC;Śâ&#x20AC;&#x2122;â&#x20AC;&#x2122; Reference List: Only citations that appear in the text should be referenced. References must be restricted to directly relevant published works, papers, or abstracts. Unpublished papers, unless accepted for publication, should not be cited. Work that is accepted for publication should be referred to as â&#x20AC;&#x2DC;â&#x20AC;&#x2DC;in pressâ&#x20AC;&#x2122;â&#x20AC;&#x2122; and a letter of acceptance of the journal must be provided. Authors are responsible for the accuracy and completeness of their references and for correct text citation. Usually the total number of references should not exceed 35. For up to 6 authors, list all authors. For more than 6 authors, list first 6 authors followed by â&#x20AC;&#x2DC;â&#x20AC;&#x2DC;et al.â&#x20AC;&#x2122;â&#x20AC;&#x2122;. Tables and Figures: Tables: Should be constructed using the table feature in your word processor or using a spreadsheet program such as Excel. The tables should be numbered in order of appearance in the text, using Arabic numerals. Each table should have a title and an explanatory legend, if necessary. All tables must be referenced and succinctly described in the text. Under no circumstances should a table repeat data that are presented in an illustration. Statistical measures of variation (i.e., standard deviation or standard error) should be identified, and decimal places in tabular data should be restricted to those with mathematical and statistical significance. Figures: Photographs, illustrations, charts, drawings, line graphs, etc. are all defined as figures. Number figures consecutively using Arabic numerals in order of appearance. Figure legend(s) should be descriptive and should allow examination of the figure without reference to text. Images must be of professional quality and uploaded as *.tiff files. Generally, figures will be reduced to fit one column of text. The actual magnification of all photomicrographs should be provided, preferably by placing a scale bar on the print. Line graphs and charts should never be sent as *.jpeg illustrations. We recommend preparing line graphs and charts as ExcelH files and copying these files into a Word *.doc sheet. REVIEW ARTICLES: Review articles should cover themes that are relevant to medical practice. Spontaneously submitted reviews are welcome; however, potential authors should bear in mind that they are expected to have expertise in the reviewed ďŹ eld. The sections should be arranged as follows: Title: Up to 250 characters. Running title: Up to 40 characters. Authors' full names and contribution: Authors should

have participated sufficiently in the work to take public

responsibility for appropriate portions of the content. Such participation must be declared in this section of the manuscript. Any financial or other relationships that may lead to a conflict of interest. Abstract, keywords and text should be arranged to cover the subject that is being reviewed. If appropriate, the method of reference collection should be described. The use of headings, subheadings, and paragraph titles is encouraged to improve clarity. Abbreviations, acknowledgments, tables and figures should be formatted as described in the Original Study section. The number of references is at the discretion of the authors. No publication fee discount is allowed for spontaneously submitted review articles that are accepted for publication.

RAPID COMMUNICATIONS: Title: Up to 250 characters. Running title: Up to 40 characters. Authors' full names and contribution: Authors should

LETTERS TO THE EDITOR: Letters to the editor expressing comments or dissenting opinions concerning articles that have been recently published in CLINICS are published at the discretion of the editor. A letter is a single section containing untitled text concerning the article under discussion, followed by references. No publication fee is charged for this class of manuscripts. EDITORIAL: Editorials should cover broad aspects of medical or biological sciences. Such manuscripts are published at the discretion of the editor. No publication fee is charged for this class of manuscripts. COMMENTARY: A commentary is an invited text with respect to an article that is being published by CLINICS. No publication fee is charged for this class of manuscripts.

INVITED REVIEW: These reviews are by invitation only and follow the format proposed for general reviews. No publication fee is charged for this class of manuscripts. SPECIAL ISSUE ARTICLE: Special issue articles are by invitation only and follow a speciﬁc format that is set by the editor in charge of the collection. Currently CLINICS does not accept: case reports, technical notes, translations and validations of questionnaires, and articles referring to ﬁrst demonstration in Brazil.

SUBMISSION A copyright transfer form, signed by all authors, must be uploaded as a complementary ﬁle as soon as the manuscript is submitted.

When the authors are satisfied that the manuscript complies with the journal format, our site should be accessed using the website www.clinics.org.br. The system will guide authors through the manuscript submission process and will prompt authors to input information into specific fields as they are submitting their manuscript. The editorial office and authors will be automatically notified of the submission. Progress of the manuscript through the Editorial Office’s procedures will be available to authors at all times.

PEER REVIEW Manuscripts are reviewed by at least two expert consultants. Accepted manuscripts are edited to comply with the journal’s format, remove redundancies, and improve clarity and understanding without altering meaning. The edited text will be presented to authors for approval.

ISSN-1807-5932

CLINICS CONTENTS Clinics 2017 72(9)516–587

CLINICAL SCIENCE

Accuracy of transient elastography-FibroScanH, acoustic radiation force impulse (ARFI) imaging, the enhanced liver ﬁbrosis (ELF) test, APRI, and the FIB-4 index compared with liver biopsy in patients with chronic hepatitis C Taisa Grotta Ragazzo, Denise Paranagua-Vezozzo, Fabiana Roberto Lima, Daniel Ferraz de Campos Mazo, Mário Guimarães Pessoa, Claudia Pinto Oliveira, Venancio Avancini Ferreira Alves, Flair José Carrilho . . . . . . . . . . 516

Cytogenomic assessment of the diagnosis of 93 patients with developmental delay and multiple congenital abnormalities: The Brazilian experience Évelin Aline Zanardo, Roberta Lelis Dutra, Flavia Balbo Piazzon, Alexandre Torchio Dias, Gil Monteiro Novo-Filho, Amom Mendes Nascimento, Marília Moreira Montenegro, Jullian Gabriel Damasceno, Fabrícia Andreia Rosa Madia, Thaís Virgínia Moura Machado da Costa, Maria Isabel Melaragno, Chong Ae Kim, Leslie Domenici Kulikowski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 526

Experimental study of histological changes in vascular loops according to the duration of the postoperative period: Application in reconstructive microsurgery Renata Gregorio Paulos, Bruno Alves Rudelli, Renee Zon Filippe, Gustavo Bispo dos Santos, Ana Abarca Herrera, Andre Araujo Ribeiro, Marcelo Rosa de Rezende, Teng Hsiang-Wei, Rames Mattar-Jr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 538

Long-term efﬁciency of cataract surgery with hydrophilic acrylic loﬂex intraocular lens Priscilla A. Jorge, Camila Ribeiro Koch, Delano Jorge, Newton Kara-Junior. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543

Early menarche and teenager pregnancy as risk factors for morbid obesity among reproductive-age women: A case-control study Amanda Gonc¸alves Neves, Karina Tamy Kasawara, Ana Carolina Godoy-Miranda, Flávio Hideki Oshika, Elinton Adami Chaim, Fernanda Garanhani Surita. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547

Endoscopic endonasal approach for mass resection of the pterygopalatine fossa Jan Plzák, Vít Kratochvil, Adam Kešner, Pavol Šurda, Aleš Vlasák, Eduard Zveˇ ˇr ina . . . . . . . . . . . . . . . . . . . . . . . . 554

The value of high-ﬂow nasal cannula oxygen therapy after extubation in patients with acute respiratory failure Hong-Zhuan Song, Juan-Xian Gu, Hui-Qing Xiu, Wei Cui, Gen-Sheng Zhang. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562

Caring for critically ill patients outside intensive care units due to full units: a cohort study Fabiane Urizzi, Marcos T. Tanita, Josiane Festti, Lucienne T.Q. Cardoso, Tiemi Matsuo, Cintia M.C. Grion . . . . . . 568

BASIC RESEARCH

An activating mutation in the CRHR1 gene is rarely associated with pituitary-dependent hyperadrenocorticism in poodles Viviani De-Marco, Luciani R. Carvalho, Mariana F. Guzzo, Paulo S.L. Oliveira, Larissa G. Gomes, Berenice B. Mendonca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 575

A Biomechanical Analysis of the Interlock Suture and a ModiďŹ ed Kessler-Loop Lock Flexor Tendon Suture Wenfeng Yang, Dan Qiao, Yuanfei Ren, Yvjin Dong, Yaohua Shang, Tiehui Zhang . . . . . . . . . . . . . . . . . . . . . . . . 582

CLINICAL SCIENCE

Accuracy of transient elastography-FibroScans, acoustic radiation force impulse (ARFI) imaging, the enhanced liver fibrosis (ELF) test, APRI, and the FIB-4 index compared with liver biopsy in patients with chronic hepatitis C Taisa Grotta Ragazzo,I,* Denise Paranagua-Vezozzo,I Fabiana Roberto Lima,II Daniel Ferraz de Campos Mazo,I,II Ma´rio Guimara˜es Pessoa,I Claudia Pinto Oliveira,I Venancio Avancini Ferreira Alves,III Flair Jose´ CarrilhoI I

Divisao de Gastroenterologia e Hepatologia Clinica, Departamento de Gastroenterologia, Hospital das Clinicas HCFMUSP, Faculdade de Medicina, Universidade de Sao Paulo, Sao Paulo, SP, BR. II Divisao de Gastroenterologia, Faculdade de Ciencias Medicas, Universidade Estadual de Campinas, Campinas, SP, BR. III Departamento de Patologia, Faculdade de Medicina FMUSP, Universidade de Sao Paulo, Sao Paulo, SP, BR.

OBJECTIVES: Although liver biopsy is the gold standard for determining the degree of liver fibrosis, issues regarding its invasiveness and the small amount of liver tissue evaluated can limit its applicability and interpretation in clinical practice. Non-invasive evaluation methods for liver fibrosis can address some of these limitations. The aim of this study was to evaluate the accuracy of transient elastography-FibroScans, acoustic radiation force impulse (ARFI), enhanced liver fibrosis (ELF), the aspartate aminotransferase-to-platelet ratio index (APRI), and the FIB-4 index compared with liver biopsy in hepatitis C. METHODS: We evaluated chronic hepatitis C patients who were followed at the Division of Clinical Gastroenterology and Hepatology, Hospital das Clıńicas, Department of Gastroenterology of University of Saõ Paulo School of Medicine, Saõ Paulo, Brazil, and who underwent liver biopsy. The accuracy of each method was determined by a receiver operating characteristic (ROC) curve analysis, and fibrosis was classified as significant fibrosis (XF2), advanced fibrosis (XF3), or cirrhosis (F4). The Obuchowski method was also used to determine the diagnostic accuracy of each method at the various stages of fibrosis. In total, 107 FibroScans, 51 ARFI, 68 ELF, 106 APRI, and 106 FIB-4 analyses were performed. RESULTS: A total of 107 patients were included in the study. The areas under the ROC curve (AUROCs) according to fibrosis degree were as follows: significant fibrosis (XF2): FibroScans: 0.83, FIB-4: 0.76, ELF: 0.70, APRI: 0.69, and ARFI: 0.67; advanced fibrosis (XF3): FibroScans: 0.85, ELF: 0.82, FIB-4: 0.77, ARFI: 0.74, and APRI: 0.71; and cirrhosis (F4): APRI: 1, FIB-4: 1, FibroScans: 0.99, ARFI: 0.96, and ELF: 0.94. The accuracies of transient elastography, ARFI, ELF, APRI and FIB-4 determined by the Obuchowski method were F0-F1: 0.81, 0.78, 0.44, 0.72 and 0.67, respectively; F1-F2: 0.73, 0.53, 0.62, 0.60, and 0.68, respectively; F2-F3: 0.70, 0.64, 0.77, 0.60, and 0.67, respectively; and F3-F4: 0.98, 0.96, 0.82, 1, and 1, respectively. CONCLUSION: Transient elastography remained the most effective method for evaluating all degrees of fibrosis. The accuracy of all methodologies was best at F4. KEYWORDS: Hepatitis C Chronic; Liver Cirrhosis; Elastography; Biomarkers/Blood; Disease Progression; Data Accuracy. Ragazzo TG, Paranagua-Vezozzo D, Lima FR, Mazo DF, Pessoa MG, Oliveira CP, et al. Accuracy of transient elastography-FibroScans, acoustic radiation force impulse (ARFI) imaging, the enhanced liver fibrosis (ELF) test, APRI, and the FIB-4 index compared with liver biopsy in patients with chronic hepatitis C. Clinics. 2017;72(9):516-525 Received for publication on December 15, 2016; First review completed on February 9, 2017; Accepted for publication on March 15, 2017 *Corresponding author. E-mail: taisagr@usp.br

’ INTRODUCTION There are approximately 130 to 150 million carriers of the hepatitis C virus (HCV) worldwide. Approximately 55-85% of patients with HCV develop the chronic form of the disease, and approximately 15-30% of these patients are at risk of developing cirrhosis within 20 years of diagnosis. Approximately 300,000-500,000 people die each year as a result of

Copyright & 2017 CLINICS – This is an Open Access article distributed under the terms of the Creative Commons License (http://creativecommons.org/licenses/by/ 4.0/) which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is properly cited. No potential conflict of interest was reported. DOI: 10.6061/clinics/2017(09)01

516

Accuracy of transient elastrography Ragazzo TG et al.

CLINICS 2017;72(9):516-525

pathologist from the Department of Pathology of HC-FMUSP to avoid operator bias in the analysis. The selected samples had a minimum of 5 portal spaces (a mean of 13 portal spaces: 43 biopsies had between 5 and 10 portal spaces and 64 had between 11 and more than 35 portal spaces) and a mean size of 15 mm (varying between 7 and 23 mm). The classification score used for analysis of the biopsies was the METAVIR score (12). The inclusion criteria were as follows: 1. HCV PCR-RNA positivity for at least 6 months and a clinical or histopathological diagnosis of treatmentnaive HCV; 2. negative serological test for hepatitis B or HIV co-infection; and 3. representative liver biopsy (minimum of 5 portal spaces, non-subcapsular fragment) performed 60 days prior to the exams. The exclusion criteria were as follows: 1. refusal to provide informed consent; 2. patient under 18 or over 70 years of age; 3. unavailability of liver biopsy (contraindication); 4. biopsies performed more than 60 days before the evaluation; 5. non-representative liver biopsy; 6. clinical suspicion or image evidence of HCC; 7. ascites; 8. body mass index (BMI) X30kg/m2 and 9. previous treatment for HCV. 10. unreliable FibroScans results (as described later). The following examinations were utilized for fibrosis evaluation: TE-FibroScans, ARFI, ELF, APRI, and FIB-4. FibroScans was developed by Echosens (Paris, France) in 2003 (13). The equipment consists of a 5-MHz US transducer coupled to a base with a vibratory axis, and it uses VCTE technology (elastography technology with velocity-controlled impulses). The vibration emitted by the transducer is of medium amplitude (2 mm) and low frequency (50 Hz). It produces a wave through the liver tissue, which allows estimation of the elasticity of the hepatic parenchyma through the equation E=3pV2, where: E=elasticity, p=density (a constant in tissue), and V=velocity of wave propagation. The denser the tissue, the faster the wave propagates. The results of liver elasticity measurements are expressed in kilopascals (kPa) in an interval of 2.5-75 kPa. The section analyzed corresponds to a cylinder 1 cm in diameter and 4 cm in length, corresponding to a volume 100-fold larger than that obtained with liver biopsy (11,13-15). A liver stiffness assessment is generally considered reliable when the following criteria are fulfilled: 10 valid measurements, success rate 460%, and ratio of the interquartile range to the median (IQR/M) p30% (16-18). Patients with invalid/ unreliable measurements were excluded from the study. The TE examinations in this study were performed by the same experienced and highly trained operator, thus eliminating the risk of inter-observer bias. All patients fasted for at least three hours prior to examination because the FibroScans reading (in kPa) varies immediately following food consumption, altering the result (19). The inclusion criteria of BMI p30 kg/m2 and skin-liver distance are factors that affect the choice of transducer used. The M transducer was used for all TE examinations in this study to avoid potential bias in interpreting the results in kPa because when both the M, and XL transducers are used, different results can occur (20-22). ARFI imaging (Siemens Acuson S2000, Virtual Touchä tissue quantification; Siemens, Erlangen, Germany) is another tool for evaluating liver stiffness that can be incorporated into a conventional US machine. Transducers sensitive to the propagation of acoustic pulses of a frequency of 2.67 Hz generate shear waves that propagate in the tissue perpendicular to the direction of the acoustic impulse. The shear waves are then tracked using US based on correlation in a

complications arising from HCV (1). In Brazil, hepatocellular carcinoma (HCC) is one of the most frequent complications of cirrhosis caused by HCV (2,3). Liver biopsy is still considered the gold standard for liver tissue evaluation, allowing the ascertainment not only of the degree of fibrosis but also of other important parameters, such as inflammation, necrosis, steatosis, and the presence of hepatic iron in the sample obtained (4). However, liver biopsy is an invasive procedure that carries risks, sometimes causing pain, hemorrhage, and even death, among other complications (5). Issues regarding the quality of the liver samples and interpretation of the results can also occur. The quality of a liver biopsy is generally related to the length and number of portal spaces evaluated (6,7). The results of the pathological anatomy can vary according to the subjective interpretation of the individual pathologist (8). Due to these limitations, non-invasive methods of liver fibrosis evaluation have been studied intensely and have improved over recent decades. These methods can be divided into two categories, namely, indirect markers, which can be assessed by routine clinical exams (e.g., aminotransferases and platelet count) (9), and direct markers, which include serum levels of substances involved in the molecular pathogenesis of fibrosis, such as matrix metalloproteinases, hyaluronic acid, and cytokines [tumor necrosis factor a (TNF-a) and transforming growth factor b (TGF-b)] (10). Elastography methods use an existing modality, such as ultrasound (US), to observe the internal tissue deformations that occur in response to an applied force and convert the resulting information to a suitable form for display. A wide variety of approaches have evolved, both for applying the force and for measuring and displaying the tissue’s response after applying a force that is either dynamic (e.g., by thumping or vibrating) or that varies so slowly that it is considered "quasi-static" (e.g., by probe palpation). The deformation may be represented in an elasticity image (elastogram) or as a local measurement of tissue displacement that may be detected and displayed directly [Acoustic Radiation Force Impulse (ARFI) imaging]. Another form of representation is through the regional values of their speed (without making images) using methods referred to herein as transient elastography (TE) (11). The aim of this study was to evaluate the accuracy of TE-FibroScans, ARFI imaging, enhanced liver fibrosis (ELF), the aspartate aminotransferase-to-platelet ratio index (APRI), and FIB-4 compared with liver biopsy in chronic hepatitis C patients.

’ PATIENTS AND METHODS We performed a prospective study evaluating treatmentnaive patients chronically infected with HCV, who were on the waiting list for liver biopsy at the outpatient clinic of the Division of Clinical Gastroenterology and Hepatology, Hospital das Clínicas, Department of Gastroenterology of University of São Paulo School of Medicine, São Paulo, Brazil (HC-FMUSP), from August 3, 2012, to May 31, 2014. A total of 107 patients were included, with 107 liver biopsies, 107 TE exams, 106 APRI and FIB-4 exams, 68 ELF exams and 51 ARFI exams performed. The liver biopsies were performed with a 14-G Tru-Cutt needle (15cm) (Medical Technology, Gainesville, FL, USA). Liver histology was examined by the same experienced liver

517

Accuracy of transient elastrography Ragazzo TG et al.

CLINICS 2017;72(9):516-525

Statistical analyses

small region of interest (ROI) of 5 mm x 10 mm, which the operator can easily locate in the field of vision up to a depth of 8 cm with a convex catheter. This allows the detection of liver stiffness through the velocity of propagation of the shear wave, which is linearly correlated with the stiffness of the liver (11,23,24). The results are expressed in meters per second (m/s) with an interval scale of 0.5-4.4 m/s and precision of ±20% related to the interval (25). A result is considered high quality when 10 valid measurements are recorded with an IQR/M ratio o30% and a success rate of X60% (26). The examination is performed at the height of the medial axillary line and the xiphoid appendage with a projection of S8. As a direct method utilizing biomarkers in the blood, ELF is a serum test that yields a single value, combining quantitative measures of hyaluronic acid (HA), pro-peptide amino-terminal of pro-collagen type III (PIIINP), and tissue inhibitor metalloproteinase 1 (TIMP-1) in human serum in an algorithm (10,27). The ADVIA Centaurs CP was used for ELF, following formula was applied to yield results: 2.494+0.846 In (C HA) + 0.735 In (C PIIINP) + 0.391 In (C TIMP-1). Regarding the methods using indirect biomarkers, APRI and FIB-4 were calculated through the following scores: APRI score= [(AST/ULN) 100]/platelet count 109/L. (28) (ULN: upper limit of normal) FIB-4 score= {[age (yr) x AST (U/L)] / [platelet count (109/L) x ALT (U/L)]} (29) The study was approved by the Ethics Committee of HCFMUSP. Of the 250 patients eligible for the study, 143 were excluded for various reasons, as described in Figure 1. Among the patients with liver biopsy, 20 were excluded due to nonrepresentative liver histology (o5 portal spaces and/or nonsubcapsular fragment), and 5 patients were excluded because of unreliable FibroScans results (one patient presented with narrow intercostal space, 3 patients had IQR/M 430% and one patient had an exam success rate o60%). A total of 107 patients were included, as shown in Figure 1.

The R package version 3.2.1(R Core Team, Vienna, Austria) was used for statistical and graph analyses. The performance of the non-invasive methods was estimated using ROC curves (using the package pROC version 1.8) by identifying the optimal cut-off points of different degrees of liver fibrosis in terms of sensitivity and specificity. The area under the ROC curve (AUROC) indicates the accuracy of the studied methods. The Obuchowski method was used to determine the accuracy of the non-invasive methods of liver fibrosis evaluation (30,31) Comparative analyses of more than two groups were performed using analysis of variance (ANOVA), the Levene statistic and the Kruskal-Wallis test. For multiple analyses, or analyses of two groups with more than two other groups, the Tukey and non-parametric Tukey tests were used. For qualitative (categorical) variables, Fisher’s exact test and the chi-squared test were used. Statistical significance was defined as po0.05.

’ RESULTS A total of 107 patients were included. The population’s gender was 50.4% (n=54) female, with 67.2% (n=72) of white ethnicity and 31.7% (n=34) of black ethnicity. In terms of exposure to HCV, 27.2% (n=29) had a history of blood transfusion and 25.2% (n=27) contracted the virus through tattooing. Genotype 1 was the most common, found in 81.9% of cases. One of the assessed patients had genotype 1 and genotype 2 co-infection. The degree of fibrosis according to the METAVIR scale was as follows: F0=7.4% (n=8), F1=40.1% (n=43), F2=28.9% (n=31), F3=21.5% (n=23), and F4=1.8% (n=2). Table 1 shows anthropometric and laboratory data according to the degree of fibrosis, demonstrating the importance of clinical and laboratory parameters according to liver disease severity. Figures 2, 3 and 4 and Tables 2, 3 and 4 present ROC curves and statistical characteristics for significant fibrosis (XF2), advanced fibrosis (XF3) and cirrhosis (F4), respectively.

Figure 1 - Flowchart of study population enrollment. HCV= hepatitis C virus, HCC= hepatocellular carcinoma.

518

Accuracy of transient elastrography Ragazzo TG et al.

CLINICS 2017;72(9):516-525

Table 1 - Correlation of anthropometric and laboratory data with fibrosis stage according to METAVIR score (means±standard deviation). Degree of Fibrosis

Age (years) BMI WC (cm) SLD (cm) AST (U/L) ALT (U/L) AP (U/L) gGT(U/L) ALB (g/dL) INR PTL (/mm3)

F0(n=8)

F1(n=43)

F2(n=31)

F3 (n=23)

F4(n=2)

40.6±9.8 25.5±2.7 94.3±8.1 1.6±0.1 25.8±7 32.1±16.3 66.7±16.1 56.1±69.8 4.2±0.2 1.0±0.1 237x103±36x103

44.8±10.9 24.7±2.8 92.1±6.7 1.5±0.2 35.9±13.2 47.8±24.4 66.3±17.9 54.4±41.6 4.4±0.3 1.0±0.1 234x103±57x103

49.9±9.9 24.9±3.0 92.1±8.3 1.5±0.3 46.5±31.3 65.9±67.7 71.7±22.1 142.2±169.3 4.5±0.2 1.0±0.1 214x103±46x103

53.8±8.9 25.6±2.6 94.5±7.8 1.6±0.4 57.2±39.7 75.0±57.6 71.2±23.3 105.4±95.9 4.3±0.4 1.0±0.1 194x103±66x103

56.5±3.5 23.9±5.5 92±15.5 1.6±0.7 181±15.5 136.5±16.2 227.5±136.4 249±16.9 4.05±0.2 1.1±0.1 118x103±35x103

BMI=body mass index, WC=waist circumference, SLD=skin-liver distance, AST=aspartate aminotransferase, ALT=alanine aminotransferase, FA=alkaline phosphatase, gGT=gamma glutamyltransferase, ALB=albumin; INR=international normalized ratio, PTL= platelet count.

Figure 2 - ROC curve for significant fibrosis (XF2).

influence the success and reliability of FibroScans (32,33). TE examinations were performed taking into account these variables, as described in the Methods section. Two factors can influence the results of ARFI. High BMI can cause underestimation of fibrosis, and a skin-liver distance 42.5 cm increases the discrepancy compared with liver biopsy (34,35). Gender can also be a confounding factor in ARFI. Male patients typically have higher ARFI values than female

Table 5 shows the accuracy of the non-invasive methods of liver fibrosis evaluation.

’ DISCUSSION Several factors can influence the results of non-invasive methods of liver fibrosis evaluation. Regarding TE, fasting, operator bias, and anthropometric characteristics can

519

Accuracy of transient elastrography Ragazzo TG et al.

CLINICS 2017;72(9):516-525

Figure 3 - ROC curve for advanced fibrosis (XF3).

Sp: 85; F4 AUROC: 0.97, Se: 86, Sp: 96). The specificity and sensitivity of FibroScans were highest at XF4 (40). FibroScans has also proven to be an effective method of assessing fibrosis progression for a variety of other pathologies, such as hepatic steatosis (41). It is also applicable in the evaluation of fibrosis progression and complications, such as portal hypertension (42,43), and in the clinical treatment of HCV and other chronic liver diseases (44,45). Gara et al. (46) showed that, despite the high sensitivity and specificity of FibroScans and ARFI in the diagnosis of cirrhosis, due to the possibility of false-positives, it is always necessary to view the results in the context of clinical exams or an imaging exam. ARFI had AUROCs of 0.67 for XF2 (Se: 64, Sp: 69), 0.74 for XF3 (Se: 57, Sp: 74), and 0.97 for F4 (Se: 100, Sp: 97) in this study. ARFI showed less accuracy than that found in the literature. In the meta-analysis by Friedrich-Rust et al. (21), the mean AUROCs for ARFI were 0.87, 0.91, and 0.93 for XF2, XF3 and F4, respectively. A meta-analysis by Nierhoff et al. (47) yielded AUROCs of 0.84, 0.89 and 0.91 for XF2, XF3, and F4 respectively, indicating that FibroScans is a good tool for diagnosing significant fibrosis, and an excellent tool for diagnosing advanced fibrosis or cirrhosis. Regarding the findings related to ARFI and FibroScans, this study revealed differences from the meta-analysis of Bota et al. (48), who found FibroScans and ARFI to be equally accurate in diagnosing significant fibrosis and cirrhosis. Our study revealed

patients (23). ELF can be affected by age; the frequencies of cardiovascular disease and chronic inflammatory diseases are higher in older populations (36,37). With significant fibrosis defined as XF2, FibroScans showed a higher degree of accuracy than the other methods assessed. An AUROC of 0.836 is considered an extremely good result (38). ELF also showed good results, with an AUROC of 0.707. ARFI and APRI both showed sufficiently satisfactory results, with AUROC values of 0.672 and 0.691, respectively. In the ROC curve for APRI, it was not possible to distinguish a cut-off point between F1 and F2; therefore, it was not possible to separate F1/F2 in the analysis of group F2F3F4 (XF2). FibroScans and ELF were extremely effective in identifying advanced fibrosis (XF3), with AUROCs of 0.85 and 0.82, respectively. FIB-4, ARFI, and APRI all yielded results that were considered good, with AUROCs of 0.77, 0.74, and 0.71, respectively. All of the methods were excellent in identifying cirrhosis (F4) with AUROCs of 0.94 for ELF, 0.969 for ARFI, 0.995 for FibroScans, 1 for APRI and 1 for FIB-4. In this study, FibroScans was consistently accurate in classifying degrees of fibrosis as XF2, XF3, or F4 with AUROCs of 0.83, 0.85, and 0.99 respectively, which was consistent with the findings of Castera et al. (16) (XF2 AUROC 0.83, Se: 67, Sp: 89; XF3 AUROC: 0.90, Se: 73, Sp: 91; and F4 AUROC: 0.95, Se: 87, Sp: 91) and of Ziol et al. (39) (XF2 AUROC: 0.79, Se: 56, Sp: 91; XF3 AUROC: 0.96, Se: 86,

520

Accuracy of transient elastrography Ragazzo TG et al.

CLINICS 2017;72(9):516-525

Figure 4 - ROC curve for cirrhosis (F4).

FibroScans to be superior to ARFI in diagnosing significant fibrosis. Perhaps the predominance in this study of fibrosis at stages F0, F1, and F2 could account for this divergence from the literature. However, there are studies that found equivalency between FibroScans and ARFI for diagnosing advanced fibrosis and cirrhosis but that revealed FibroScans as the better choice for identifying significant fibrosis (49). In two comparative analyses (50,51), both ARFI and APRI demonstrated the ability to evaluate the progression of advanced fibrosis and cirrhosis in HCV, with ARFI exhibiting slightly better accuracy (50,51). For ELF, we found AUROCs of 0.71 for XF2 fibrosis (Se: 82, Sp: 56), 0.82 for XF3 fibrosis (Se: 83, Sp: 69), and 0.94 for F4 fibrosis (Se: 100, Sp: 100). Despite presenting inferior accuracy for XF2 fibrosis compared with other studies (10,52), ELF produced results similar to those of Parkes et al. (53), with AUROCs of 0.74, 0.84 and 0.90 for XF2, XF3, and F4 fibrosis, respectively, in a specific HCV population. Fagan et al. (54) found that a cut-off score for ELF of X9.8 had a sensitivity of 74.4% and a specificity of 92.4% for advanced fibrosis. In this study, the cut-off for XF3 fibrosis was 9.4 with Se=83 and Sp=69. In the present study, APRI had a XF3 AUROC: 0.71 and an F4 AUROC of 1. This result differ from the original work, which revealed a XF3 AUROC of 0.88 and an F4 AUROC of 0.94 (24). APRI could not identify the individual stages of

fibrosis, and the fibrosis of some patients remained unclassified when the initial cut-off was applied. Furthermore, the appropriate definition of the limits of normal AST remains uncertain. Each laboratory establishes a different value for the upper limit of normal (55). APRI and FIB-4 are excellent for cirrhosis, but care should be taken with regard to the few patients who have test results close to normal levels. These patients are at risk for false-negative results; therefore, the tests should be performed alongside imaging examinations (56). Among the various comparative studies of non-invasive methods of liver fibrosis assessment, Crisan et al. (57) showed good results for the accuracy of APRI, FIB-4, and FibroScans in diagnosing XF3 fibrosis. Peterson et al. (58) evaluated APRI and ELF in case of significant fibrosis. That study differed from the others in that it did not use the METAVIR classification, but it demonstrated that one methodology can complement the others in some cases. Poynard et al. (59) conducted a meta-analysis of a variety of chronic liver diseases and confirmed the accuracy of FIB-4 and APRI in identifying advanced fibrosis and cirrhosis in HCV. The results using the Obuchowski method revealed the degree of fibrosis, and it was impossible to distinguish between significant fibrosis, advanced fibrosis, and cirrhosis. In almost all stages of fibrosis, FibroScans was the most

521

6.5 kPa 0.71 0.92 0.91 0.75 9.11 0.31 0.83 0.58 0.81 0.79 0.62 3.5 0.2 0.75

0.83 0.98 0.95 0.92 23.67 0.47 0.91

95% CI 1.22m/s2 0.64 0.69 0.67 0.67 2.08 0.52 0.67

Estimate

ARFI

0.43 0.48 0.45 0.45 1.09 0.29 0.51

0.82 0.86 0.84 0.84 3.97 0.93 0.82

95% CI 8.98 0.83 0.56 0.59 0.81 1.9 0.31 0.70

Estimate

ELF

0.64 0.4 0.42 0.62 1.28 0.13 0.58

0.94 0.72 0.83 0.9 2.81 0.71 0.83

95% CI 0.67 0.48 0.86 0.79 0.6 3.44 0.6 0.69

Estimate

APRI

0.35 0.73 0.63 0.46 1.65 0.46 0.59

0.62 0.94 0.87 0.8 7.21 0.79 0.79

95% CI 1.29 0.68 0.76 0.76 0.68 2.83 0.42 0.76

Estimate

FIB4

0.54 0.62 0.62 0.54 1.67 0.28 0.66

0.8 0.87 0.85 0.82 4.78 0.64 0.85

95% CI

522

7.1kPa 0.8 0.79 0.54 0.93 3.86 0.25 0.85 0.59 0.69 0.41 0.83 2.42 0.11 0.77

0.93 0.87 0.8 0.96 6.15 0.56 0.92

95% CI 1.41m/s2 0.57 0.84 0.57 0.84 3.52 0.51 0.74

Estimate

ARFI

0.29 0.68 0.35 0.61 1.49 0.27 0.57

0.82 0.94 0.82 0.94 8.34 0.95 0.90

95% CI 9.47 0.83 0.7 0.37 0.95 2.75 0.24 0.82

Estimate

ELF

0.52 0.56 0.25 0.81 1.71 0.07 0.71

0.98 0.81 0.85 0.97 4.39 0.86 0.93

95% CI

0.67 0.6 0.77 0.44 0.86 2.56 0.52 0.71

Estimate

APRI

0.39 0.66 0.32 0.72 1.54 0.32 0.59

0.79 0.85 0.66 0.92 4.25 0.86 0.83

95% CI

1.22 0.84 0.6 0.4 0.92 2.13 0.26 0.77

Estimate

FIB4

0.64 0.49 0.29 0.81 1.55 0.11 0.66

0.95 0.71 0.72 0.95 2.93 0.66 0.89

95% CI

Se=sensitivity, Sp=specificity, PPV=positive predictive value, NPV=negative predictive value, dlr.positive=likelihood ratio positive, dlr.negative=likelihood ratio negative, AUROC area under the ROC curve.

cut-off Se Sp PPV NPV dlr.positive dlr.negative AUROC

Estimate

FibroScans

Table 3 - Statistical characteristics of advanced fibrosis (XF3).

cut-off Se Sp PPV NPV dlr.positive dlr.negative AUROC

Estimate

FibroScans

Table 2 - Statistical characteristics of significant fibrosis (XF2).

Accuracy of transient elastrography Ragazzo TG et al. CLINICS 2017;72(9):516-525

Accuracy of transient elastrography Ragazzo TG et al.

1 1

f 1

6.51 1 1 1 1 Inf 0 1 1 1 1 1 1 1 4.3 1 1 1 1 I0 1 0.64 1 1 0.98

11 0.2 1 1 0.94 Inf 0.8 0.94 1 0.99 1 1 48.89 1 1 27kPa 1 0.99 0.67 1 105 0 0.99

0.16 0.95 0.26 0.91 14.93 0 0.98

1 1 1 1 738.46 NaN 1

2.37m/s 1 0.94 0.4 1 16.33 0 0.96

cut-off Se Sp PPV NPV dlr.positive dlr.negative AUROC

95% CI Estimate

FibroScans

Table 4 - Statistical characteristics of cirrhosis (F4).

Estimate

ARFI

0.16 0.83 0.18 0.81 5.46 0 0.90

95% CI

Estimate

ELF

0.25 0.93 0.17 0.85

0.80 0.91 -

95% CI

Estimate

APRI

0.16 0.97 0.35 0.91 0 0 1

95% CI

Estimate

FIB4

0.16 0.97 0.35 0.91

01-

95% CI

1 1 1 1

CLINICS 2017;72(9):516-525

Table 5 - Accuracy according to sequential pairs using the Obuchowski method. Biopsy

FibroScans

ARFI

ELF

APRI

FIB4

vs vs vs vs

0.81 0.73 0.70 0.98

0.78 0.53 0.64 0.96

0.44 0.62 0.77 0.82

0.72 0.60 0.60 1

0.67 0.68 0.67 1

F0 F1 F2 F3

F1 F2 F3 F4

accurate method. It was most accurate at the extremes of F1 and F4, with values of 0.81 and 0.98, respectively, and demonstrated lower accuracy at intermediate stages. ELF showed exactly the same accuracy at F4 in this study as the only other published study that used the Obuchowski method (60). A limitation of our study is the small sample size, especially in patients at the extremes of the classification of hepatic fibrosis, F0 and F4. This may have had an impact on the results, as previously discussed. FibroScans was the most accurate method in diagnosing significant fibrosis (XF2) and advanced fibrosis (XF3) (with AUROCs of 0.84 and 0.85 respectively). APRI and FIB-4 were also very accurate in identifying cirrhosis. The accuracy of all the methodologies was best at F4, but the TE remained the most effective method for evaluating all degrees of fibrosis.

’ AUTHOR CONTRIBUTIONS Ragazzo TG designed the study, analyzed the data and helped drafting the manuscript. Paranagua-Vezozzo D designed the study and helped drafting the manuscript. Lima FR analyzed the data and helped drafting the manuscript. Mazo DF and Pessoa MG designed the study and helped drafting the manuscript. Oliveira CP designed the study and analyzed the data. Alves VA and Carrilho FJ designed the study, analyzed the data and helped drafting the manuscript.

’ REFERENCES 1. World Health Organization. Hepatitis C. Avaiable: http://www.who.int/ mediacentre/factsheets/fs164/en/ 2. Carrilho FJ, Kikuchi L, Branco F, Gonc¸alves CS, Mattos AA, Brazilian HCC Study Group. Clinical and epidemiological aspects of hepatocellular carcinoma in Brazil. Clinics. 2010;65(12):1285-90, http://dx.doi.org/ 10.1590/S1807-59322010001200010. 3. Paranaguá-Vezozzo DC, Ono SK, Alvarado-Mora MV, Farias AQ, CunhaSilva M, Franc¸a JI, et al. Epidemiology of HCC in Brazil: incidence and risk factors in a ten-year cohort. Ann Hepatol. 2014;13(4):386-93. 4. Sebastiani G, Alberti A. Non invasive fibrosis biomarkers reduce but not substitute the need for liver biopsy. World J Gastroenterol. 2006;12(23): 3682-94, http://dx.doi.org/10.3748/wjg.v12.i23.3682. 5. Rustagi T, Newton E, Kar P. Percutaneous liver biopsy. Trop Gastroenterol. 2010;31(3):199-212, http://dx.doi.org/10.1111/j.1365-2893.2012.01608.x. 6. Bravo AA, Sheth SG, Chopra S. Liver biopsy. N Engl J Med. 2001; 344(7):495-500, http://dx.doi.org/10.1056/NEJM200102153440706. 7. Al Knawy B, Shiffman M. Percutaneous liver biopsy in clinical practice. Liver Int. 2007;27(9):1166-73, http://dx.doi.org/10.1111/j.1478-3231.2007. 01592.x. 8. Regev A, Berho M, Jeffers LJ, Milikowski C, Molina EG, Pyrsopoulos NT, et al. Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection. Am J Gastroenterol. 2002;97(10)20: 2614-8, http://dx.doi.org/10.1111/j.1572-0241.2002.06038.x. 9. 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 NAFDL. Hepatology. 2007;45(4):846-54, http:// dx.doi.org/10.1002/hep.21496. 10. Rosenberg WM, Voelker M, Thiel R, Becka M, Burt A, Schuppan D, et al. Serum markers detect the presence of liver fibrosis: a cohort study. Gastroenterology. 2004;127(6):1704-13, http://dx.doi.org/10.1053/j.gastro. 2004.08.052. 11. Bamber J, Cosgrove D, Dietrich CF, Fromageau J, Bojunga J, Calliada F, et al. EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography. Part 1: Basic principles and technology. Ultraschall Med. 2013;34(2):169-84, http://dx.doi.org/10.1055/s-0033-1335205.

523

Accuracy of transient elastrography Ragazzo TG et al.

CLINICS 2017;72(9):516-525

33. Perazzo H, Fernandes FF, Gomes A, Terra C, Perez RM, Figueiredo FA. Interobserver variability in transient elastography analysis of patients with chronic hepatitis C. Liver Int. 2015;35(5):1533-9, http://dx.doi.org/ 10.1111/liv.12551. 34. Nishikawa T, Hashimoto S, Kawabe N, Harata M, Nitta Y, Murao M, et al. Factors correlating with acoustic radiation force impulse elastography in chronic hepatitis C. World J Gastroenterol. 2014;20(5):1289-97, http://dx. doi.org/10.3748/wjg.v20.i5.1289. 35. Nadebaum D, Gibson R, Howell J, Christie M, Gorelik A, Liew D, et al. Central obesity negatively impacts the performance of acoustic radiation force impulse imaging (ARFI) in the assessment of liver fibrosis. American Association for the Study of Liver Diseases. The Liver Meeting; 2014; November 7th-11th, Boston. 36. Pinzani M. The ELF panel: a new crystal ball in hepatology? Gut. 2010; 59(9):1165-7, http://dx.doi.org/10.1136/gut.2010.214932. 37. Parkes J, Roderick P, Harris S, Day C, Mutimer D, Collier J, et al. Enhanced liver fibrosis test can predict clinical outcomes in patients with chronic liver disease. Gut. 2010;59(9):1245-51, http://dx.doi.org/ 10.1136/gut.2009.203166. 38. Šimundić AM. Measures of diagnostic accuracy: basic definitions. EJIFCC. 2009;19(4):203-11. 39. Ziol M, Handra-Luca A, Kettaneh A, Christidis C, Mal F, Kazemi F, et al. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with chronic hepatitis C. Hepatology. 2005;41(1):48-54, http://dx. doi.org/10.1002/hep.20506. 40. Castera L, Forns X, Alberti A. Non-invasive evaluation of liver fibrosis using transient elastography. J Hepatol. 2008;48(5):835-47, http://dx.doi. org/10.1016/j.jhep.2008.02.008. 41. You SC, Kim KJ, Kim SU, Kim BK, Park JY, Kim DY, et al. Factors associated with significant liver fibrosis assessed using transient elastography in general population. World J Gastroenterol. 2015;21(4):1158-66, http:// dx.doi.org/10.3748/wjg.v21.i4.1158. 42. Bureau C, Metivier S, Peron JM, Selves J, Robic MA, Gourraud PA, et al. Transient elastography accurately predicts presence of significant portal hypertension in patients with chronic liver disease. Aliment Pharmacol Ther. 2008;27(12):1261-8, http://dx.doi.org/10.1111/j.1365-2036.2008. 03701.x. 43. Hong WK, Kim MY, Baik SK, Shin SY, Kim JM, Kang YS, et al. The usefulness of non-invasive liver stiffness measurements in predicting clinically significant portal hypertension in cirrhotic patients: Korean data. Clin Mol Hepatol. 2013;19(4):370-5, http://dx.doi.org/10.3350/cmh.2013. 19.4.370. 44. Bonder A, Afdhal N. Utilization of FibroScan in clinical practice. Curr Gastroenterol Rep. 2014;16(2):372, http://dx.doi.org/10.1007/s11894014-0372-6. 45. Serejo F, Marinho R, Velosa J, Costa A, Carneiro de Moura M. Elastografia Hepática Transitória, um método não invasivo para avaliac¸ão da fibrose em doentes com hepatite C crônica. J Port Gastrenterol. 2007;14(1):8-14. 46. Gara N, Zhao X, Kleiner DE, Liang TJ, Hoofnagle JH, Ghany MG. Discordance among transient elastography, aspartate aminotransferase to platelet ratio index, and histologic assessments of liver fibrosis in patients with chronic hepatitis C. Clin Gatroenterol Hepatol. 2013;11(3):303-8, http://dx.doi.org/10.1016/j.cgh.2012.10.044. 47. Nierhoff J, Chávez Ortiz AA, Herrmann E, Zeuzem S, Friedrich-Rust M. The efficiency of acoustic radiation force impulse imaging for the staging of liver fibrosis: a meta-analysis. Eur Radiol. 2013;23(11):3040-53, http:// dx.doi.org/10.1007/s00330-013-2927-6. 48. Bota S, Herkner H, Sporea I, Salzl P, Sirli R, Neghina AM, et al. Metaanalysis: ARFI elastography versus transient elastography for the evaluation of liver fibrosis. Liver Int. 2013;33(8):1138-47, http://dx.doi.org/ 10.1111/liv.12240. 49. Sporea I, Badea R, Sirli R, Lupsor M, Popescu A, Danila M, et al. How efficient is acoustic radiation force impulse elastography for the evaluation of liver stiffness? Hepat Mon. 2011;11(7):532-8. 50. Fierbinteanu-Braticevici C, Andronescu D, Usvat R, Cretoiu D, Baicus C, Marinoschi G. Acoustic radiation force imaging sonoelastography for noninvasive staging of liver fibrosis. World J Gastroenterol. 2009;15(44): 5525-32, http://dx.doi.org/10.3748/wjg.15.5525. 51. Li SM, Li GX, Fu DM, Wang Y, Dang LQ. Liver fibrosis evaluation by ARFI and APRI in chronic hepatitis C. World J Gastroenterol. 2014;20(28): 9528-33, http://dx.doi.org/10.3748/wjg.v20.i28.9528. 52. Wahl K, Rosenberg W, Vaske B, Manns MP, Schulze-Osthoff K, Bahr MJ, et al. Biopsy-controlled liver fibrosis staging using the enhanced liver fibrosis (ELF) score compared to transient elastography. PLoS ONE. 2012;7(12):e51906, http://dx.doi.org/10.1371/journal.pone.0051906. 53. Parkes J, Guha IN, Roderick P, Harris S, Cross R, Manos MM, et al. Enhanced Liver Fibrosis (ELF) test accurately identifies liver fibrosis in patients with chronic hepatitis C. J Viral Hepat. 2011;18(1):23-31, http:// dx.doi.org/10.1111/j.1365-2893.2009.01263.x. 54. Fagan KJ, Pretorius CJ, Horsfall LU, Irvine KM, Wilgen U, Choi K, et al. ELF score X9.8 indicates advanced hepatic fibrosis and is influenced by age, steatosis and histological activity. Liver Int. 2015;35(6):1673-81, http://dx.doi.org/10.1111/liv.12760.

12. The French METAVIR Cooperative Study Group. Intraobserver and interobserver variations in liver biopsy interpretation in patients with chronic hepatitis C. Hepatology. 1994;20(1 Pt 1):15-20, http://dx.doi.org/ 10.1002/hep.1840200104. 13. Castera L. Use of elastometry (FibroScan) for the non-invasive staging of liver fibrosis. Gastroenterol Clin Biol. 2007;31(5):524-30, http://dx.doi. org/10.1016/S0399-8320(07)89422-X. 14. de Lédinghen V, Vergniol J. Transient elastography (FibroScan). Gastroenterol Clin Biol. 2008;32(6 Suppl 1):58-67, http://dx.doi.org/10.1016/ S0399-8320(08)73994-0. 15. Foucher J, Chanteloup E, Vergniol J, Castéra L, Le Bail B, Adhoute X, Bertet J, Couzigou P, de Lédinghen V. Diagnosis of cirrhosis by transient elastography (Fibroscan): a prospective study. Gut. 2006;55(3):403-8, http://dx.doi.org/10.1136/gut.2005.069153. 16. Castera L, Vergniol J, Foucher J, Le Bail B, Chanteloup E, Haaser M, et al. Prospective comparison of transient elastography, Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology. 2005;128(2):343-50, http://dx.doi.org/10.1053/j.gastro.2004. 11.018. 17. Bousier J, Zarski JP, de Ledinghen V, Rousselet MC, Sturm N, Lebail B, et al. Determination of reliability criteria for liver stiffness evaluation by transient elastography. Hepatology. 2013;57(3):1182-91, http://dx.doi. org/10.1002/hep.25993. 18. Castera L, Foucher J, Bernard PH, Carvalho F, Allaix D, Merrouche W, et al. Pitfalls of liver stiffness measurement: a 5-year prospective study of 13,369 examinations. Hepatology. 2010;51(3):828-35, http://dx.doi.org/ 10.1002/hep.23425. 19. Mederacke I, Wursthorn K, Kirschner J, Rifai K, Manns MP, Wedemeyer H, et al. Food intake increases liver stiffness in patients with chronic or resolved hepatitis C virus infection. Liver Int. 2009;29(10):1500-6, http:// dx.doi.org/10.1111/j.1478-3231.2009.02100.x. 20. Myers RP, Pomier-Layrargues G, Kirsch R, Pollett A, Beaton M, Levstik M, et al. Discordance in fibrosis staging between liver biopsy and transient elastography using the FibroScan XL probe. J Hepatol. 2012;56(3): 564-70, http://dx.doi.org/10.1016/j.jhep.2011.10.007. 21. Myers RP, Crotty P, Pomier-Layrargues G, Ma M, Urbanski SJ, Elkashab M. Prevalence, risk factors and causes of discordance in fibrosis staging by transient elastography and liver biopsy. Liver Int. 2010;30(10):1471-80, http://dx.doi.org/10.1111/j.1478-3231.2010.02331.x. 22. Herrero JI, Inarrairaegui M, D’Avola D, Sangro B, Prieto J, Quiroga J. Comparison of the M and XL FibroScan(s) probes to estimate liver stiffness by transient elastography. Gastroenterol Hepatol. 2014;37(4):233-9, http://dx.doi.org/10.1016/j.gastrohep.2013.10.009. 23. Liao LY, Kuo KL, Chiang HS, Lin CZ, Lin YP, Lin CL. Acoustic radiation force impulse elastography of the liver in healthy patients: test location, reference range and influence of gender and body mass index. Ultrasound Med Biol. 2015;41(3):698-704, http://dx.doi.org/10.1016/j.ultrasmedbio. 2014.09.030. 24. Piscaglia F, Marinelli S, Bota S, Serra C, Venerandi L, Leoni S, et al. The role of ultrasound elastographic techniques in chronic liver disease: current status and future perspectives. Eur J Radiol. 2014;83(3):450-5, http://dx.doi.org/10.1016/j.ejrad.2013.06.009. 25. Friedrich-Rust M, Nierhoff J, Lupsor M, Sporea I, Fierbinteanu-Braticevici C, Strobel D, et al. Performance of Acoustic Radiation Force Impulse imaging for the staging of liver fibrosis: a pooled meta-analysis. J Viral Hepat. 2012;19(2):e212-9, http://dx.doi.org/10.1111/j.1365-2893.2011.01537.x. 26. Bota S, Sporea I, Sirli R, Popescu A, Gradinaru-Tascau O. How useful are ARFI elastography cut-off values proposed by meta- analysis for predicting the significant fibrosis and compensated liver cirrhosis. Med Ultrason. 2015;17(2):200-5, http://dx.doi.org/10.11152/mu.2013.2066.172.arf. 27. Lichtinghagen R, Pietsch D, Bantel H, Manns MP, Brand K, Bahr MJ. The Enhanced Liver Fibrosis (ELF) score: normal values, influence factors and proposed cut-off values. J Hepatol. 2013;59(2):236-42, http://dx.doi.org/ 10.1016/j.jhep.2013.03.016. 28. 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, http://dx.doi.org/10.1053/jhep.2003.50346. 29. Vallet-Pichard A, Mallet V, Nalpas B, Verkarre V, Nalpas A, DhalluinVenier 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, http://dx.doi.org/10.1002/hep.21669. 30. Obuchowski NA. An ROC-type measure of diagnostic accuracy when the gold standard is continuous-scale. Stat. Med. 2006;25(3):481-93, http:// dx.doi.org/10.1002/sim.2228. 31. Obuchowski NA, Goske MJ, Applegate KE. Assessing physicians’ accuracy in diagnosing paediatric patients with acute abdominal pain: measuring accuracy for multiple diseases. Statist Med. 2001;20(21):3261-78, http://dx.doi.org/10.1002/sim.944. 32. Juárez-Hernández E, Uribe-Ramos MH, Ramos-Ostos MH, LópezRamírez AY, Ornelas-Arroyo S, Romero-Flores JL, et al. Factors associated with the quality of transient elastography. Dig Dis Sci. 2015;60(7):2177-82, http://dx.doi.org/10.1007/s10620-015-3611-2.

524

Accuracy of transient elastrography Ragazzo TG et al.

CLINICS 2017;72(9):516-525

58. Petersen JR, Stevenson HL, Kasturi KS, Naniwadekar A, Parkes J, Cross R, et al. Evaluation of the aspartate aminotransferase/platelet ratio index and enhanced liver fibrosis tests to detect significant fibrosis due to chronic hepatitis C. J Clin Gastroenterol. 2014;48(4):370-6, http://dx.doi. org/10.1097/MCG.0b013e3182a87e78. 59. Poynard T, Ngo Y, Perazzo H, Munteanu M, Lebray P, Moussalli J, et al. Prognostic value of liver fibrosis biomarkers: a meta-analysis. Gastroenterol Hepatol (N Y). 2011;7(7):445-54. 60. GuĂŠchot J, TrocmĂŠ C, Renversez JC, Sturm N, Zarski J P; ANRS HC EP 23 Fibrostar Study Group. Independent validation of the Enhanced Liver Fibrosis (ELF) score in the ANRS HC EP 23 Fibrostar cohort of patients with chronic hepatitis C. Clin Chem Lab Med. 2012;50(4):693-9, http:// dx.doi.org/10.7326/0003-4819-158-11-201306040-00005.

55. Silva Jr RG, Fakhouri R, Nascimento TV, Santos IM, Barbosa LM. Aspartate aminotransferase-to-platelet ratio index for fibrosis and cirrhosis prediction in chronic hepatitis C patients. Braz J Infect Dis. 2008;12(1): 15-9, http://dx.doi.org/10.1590/S1413-86702008000100005. 56. Martin J, Khatri G, Gopal P, Singal AG. Accuracy of ultrasound and noninvasive markers of fibrosis to identify patients with cirrhosis. Dig Dis Sci. 2015;60(6):1841-7, http://dx.doi.org/10.1007/s10620-0153531-1. 57. Crisan D, Radu C, Lupsor M, Sparchez Z, Grigorescu MD, Grigorescu M. Two or more synchronous combination of noninvasive tests to increase accuracy of liver fibrosis assessement in chronic hepatitis C; results from a cohort of 446 patients. Hepat Mon. 2012;12(3):177-84, http://dx.doi.org/ 10.5812/hepatmon.5106.

525

CLINICAL SCIENCE

Cytogenomic assessment of the diagnosis of 93 patients with developmental delay and multiple congenital abnormalities: The Brazilian experience E´velin Aline Zanardo,I,* Roberta Lelis Dutra,I Flavia Balbo Piazzon,I Alexandre Torchio Dias,I Gil Monteiro Novo-Filho,I Amom Mendes Nascimento,I Marı´ lia Moreira Montenegro, I Jullian Gabriel Damasceno,I Fabrı´ cia Andreia Rosa Madia, I Thaı´ s Virgı´ nia Moura Machado da Costa, I Maria Isabel Melaragno, II Chong Ae Kim, III Leslie Domenici Kulikowski I I Laboratorio de Citogenomica, Departamento de Patologia, Faculdade de Medicina FMUSP, Universidade de Sao Paulo, Sao Paulo, SP, BR. II Departamento de Morfologia e Genetica, Universidade Federal de Sao Paulo, Sao Paulo, SP, BR. III Unidade de Genetica, Departamento de Pediatria, Instituto da Crianca, Hospital das Clinicas HCFMUSP, Faculdade de Medicina, Universidade de Sao Paulo, Sao Paulo, SP, BR.

OBJECTIVE: The human genome contains several types of variations, such as copy number variations, that can generate specific clinical abnormalities. Different techniques are used to detect these changes, and obtaining an unequivocal diagnosis is important to understand the physiopathology of the diseases. The objective of this study was to assess the diagnostic capacity of multiplex ligation-dependent probe amplification and array techniques for etiologic diagnosis of syndromic patients. METHODS: We analyzed 93 patients with developmental delay and multiple congenital abnormalities using multiplex ligation-dependent probe amplifications and arrays. RESULTS: Multiplex ligation-dependent probe amplification using different kits revealed several changes in approximately 33.3% of patients. The use of arrays with different platforms showed an approximately 53.75% detection rate for at least one pathogenic change and a 46.25% detection rate for patients with benign changes. A concomitant assessment of the two techniques showed an approximately 97.8% rate of concordance, although the results were not the same in all cases. In contrast with the array results, the MLPA technique detected B70.6% of pathogenic changes. CONCLUSION: The obtained results corroborated data reported in the literature, but the overall detection rate was higher than the rates previously reported, due in part to the criteria used to select patients. Although arrays are the most efficient tool for diagnosis, they are not always suitable as a first-line diagnostic approach because of their high cost for large-scale use in developing countries. Thus, clinical and laboratory interactions with skilled technicians are required to target patients for the most effective and beneficial molecular diagnosis. KEYWORDS: Cytogenomic Techniques; MLPA; Array; Developmental Delay; Multiple Congenital Abnormalities. Zanardo EA, Dutra RL, Piazzon FB, Dias AT, Novo-Filho GM, Nascimento AM, et al. Cytogenomic assessment of the diagnosis of 93 patients with developmental delay and multiple congenital abnormalities: The Brazilian experience. Clinics. 2017;72(9):526-537 Received for publication on December 14, 2016; First review completed on February 21, 2017; Accepted for publication on May 4, 2017 *Corresponding author. E-mail: evelinzanardo@yahoo.com.br

’ INTRODUCTION

CNVs are the most prevalent type of structural variation in the human genome and can affect the transcription rate, sequence, structure, and function of genes. These genomic variations include a range of deletions and duplications larger than 1 kb and up to several Mb (1,2). Although these variations often represent only small genomic segments, they can generate several specific clinical abnormalities, such as developmental delay (DD) and multiple congenital abnormalities (MCAs) (1-4). However, the etiology of these disorders is not well understood, making genetic counseling and treatment difficult (1,2,5). Different cytogenomic techniques have been used to detect these changes, including the MLPA (multiplex ligation-dependent probe amplification) and array techniques (1,6,7).

The human genome contains several types of structural variations that contribute to genetic diversity and disease susceptibility (1,2). These structural variations include single nucleotide alterations, such as point mutations or SNPs (single nucleotide polymorphisms), small InDels, and copy number variations (CNVs) (1,3). Copyright & 2017 CLINICS – This is an Open Access article distributed under the terms of the Creative Commons License (http://creativecommons.org/licenses/by/ 4.0/) which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is properly cited. No potential conflict of interest was reported. DOI: 10.6061/clinics/2017(09)02

526

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

CLINICS 2017;72(9):526-537

and counseling in the national health network (http:// bvsms.saude.gov.br/bvs/publicacoes/diretrizes_atencao_ integral_pessoa_doencas_raras_SUS.pdf). Thus, genetic services must study the best strategies for molecular assessment to diagnose each patient referred with DD and MCA, as the introduction of a single molecular diagnostic method, such as array technology, as a first-line assessment method for patients with DD and MCA is impractical in Brazil due to insufficient public investment in the health care system and because low-income patients cannot afford such tests. In this study, we report our experience with the implementation and assessment of MLPA using different kits, array platforms (Affymetrix, Agilent and Illumina), and probe densities for the molecular diagnostic and scientific analysis of 93 Brazilian patients with DD and MCA.

MLPA is a technique that is used to detect deletions and duplications in genetic diseases of interest, such as the most common microdeletion/microduplication syndromes and subtelomeric regions (8,9). This method is considered a faster alternative and is more economically viable than other molecular techniques (3,10), and it allows quantitative genomic screening of target-specific sequences through simultaneous hybridization and amplification via polymerase chain reaction (PCR) using more than 50 different probes in a single reaction (3,8,11,12). The screening of specific submicroscopic changes via MLPA detects abnormalities in 5 to 10% of patients with a normal conventional karyotype (13-15). Thus, in a single test, the MLPA evaluates patients with characteristics of microdeletion/microduplication syndromes and/or patients with suspected subtelomeric abnormalities (9,15-18). Although MLPA allows the evaluation of multiple different genomic regions, the main limitation of this technique is the need for a clinical hypothesis to direct the selection of a specific kit for analysis (3,8). In contrast, the array technique does not require a specific clinical diagnosis before use. The array technique permits the assessment of the CNVs present in the whole genome of a patient in a single reaction with a high level of resolution (B0.7 kb), depending on the platform, types of probes and how they are distributed in the genome, thus increasing the detection rate of complex imbalances (4,19,20). This technique involves the hybridization of probes to complementary DNA (genomic sequence segments) on a slide or chip array and subsequent analysis of the fluorescence annealed to the target DNA sequences using specific software (7,21). Currently, there are several companies that offer this technology on different platforms, offering slides or chips with a high density or coverage of the genome. However, these platforms vary in the number of probes used, and several of them can interrogate millions of regions in a single sample (4,7,20,22,23). The main advantage of the array technique is the ability to investigate the entire genome in a single experiment with higher resolution and accuracy compared with traditional and molecular cytogenetics, as this allows the investigation of small changes that may have an impact on the phenotype of patients without a definitive clinical diagnosis (19,22,24). Thus, arrays have been employed to diagnose patients with DD and MCAs as well as normal karyotypes, increasing the detection rate of small genomic imbalances and the diagnosis of patients with clinical phenotypes of unknown etiology (22,25). The main limitations of the array technique are the high cost of large-scale application for developing countries, the experimental time required (3-5 days), and the expertise required for classification of the results (CNVs), which can only be interpreted by a highly qualified professional (25-27). An unequivocal diagnosis is fundamental to providing suitable answers regarding the prognosis and risk of recurrence and can contribute to improving public health policy (2,25,28). In developed countries, the array technique is already being used as the first-line molecular diagnostic test in patients with MCA (28,29). Recently, Brazil has modified its policies in the field of genetics, including the clinical genetics policy guidelines of the Sistema Único de Saúde (SUS), and has provided financial incentives to cover the costs of genetic testing

’ MATERIALS AND METHODS This study involved 93 patients who were evaluated using MLPA and array techniques. The patients presented with DD and MCAs, such as minor facial anomalies, including a high forehead, frontal bossing, broad nasal bridge, low-set ears, ocular hypertelorism, and abnormalities of the eyes, as well as major congenital defects, such as skeletal and genital malformations, heart defects, and structural brain abnormalities. All patients were previously assessed through conventional cytogenetic analysis to identify their numerical and structural chromosomal abnormalities; metaphase chromosomes were obtained from peripheral blood lymphocyte samples the patients, and G-banding analysis was performed using standard procedures. In each case, twenty metaphase chromosomes were analyzed at a 550-chromosome band resolution (X5 Mb) and then classified according to the International System for Human Cytogenetic Nomenclature 2013 (ISCN) guidelines. Genomic DNA was isolated from 3 mL of peripheral whole blood from patients using a commercially available DNA isolation kit (QIAamp DNA Blood Mini Kits, Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The quality and quantity of the DNA samples were determined using a Qubits 2.0 Fluorometer (Invitrogen, Carlsbad, California, USA), and the integrity of the DNA was ascertained via agarose gel electrophoresis analysis. All of the genomic DNAs were screened with the following three MLPA kits: for the most common microdeletion/microduplication syndromes, the SALSA MLPA probemix P064-B2 Mental Retardation-1 kit was employed, which includes probes for the 1p36 deletion, Williams-Beuren, Smith-Magenis, Miller-Dieker, 22q11.2 deletion, Prader-Willi/ Angelman, Alagille, Saethre-Chotzen, and Sotos syndromes; for subtelomeric imbalances, the SALSA MLPA probemix P036-E1 Human Telomere-3 and SALSA MLPA probemix P070-B2 Human Telomere-5 kits were used, which include subtelomeric probes for all chromosomes (MRC-Holland, Amsterdam, Netherlands). In several cases, the patients’ genomic DNA samples were also assessed using specific MLPA kits to confirm the observed changes. The kits used in these cases were the SALSA MLPA probemix P250-B1 DiGeorge and SALSA MLPA probemix P356-A1 Chromosome 22q kits, which are specific for chromosome 22, and the SALSA MLPA probemix P029-A1 Williams-Beuren Syndrome kit, which is specific for

527

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

CLINICS 2017;72(9):526-537

The SNP and bead arrays supply the B allele frequency (BAF), which represents the proportion of B alleles in the genotype. A region without evidence of CNVs should show a log2 ratio near zero and three BAF clusters of 0, 0.5, and 1, corresponding to the AA, AB, and BB genotypes, respectively. All samples were evaluated and were found to be in accordance with the quality standards. The results were analyzed according to the American College of Medical Genetics guidelines (30) using independent tests and were compared with the following databanks of CNVs and classified as benign, pathogenic or VOUS (variants of uncertain clinical significance): the Database of Genomic Variants (DGV – http://projects.tcag.ca/variation/), the Database of Chromosomal Imbalance and Phenotype in Humans Using Ensembl Resources (DECIPHER – http:// decipher.sanger.ac.uk/) and the UCSC Genome Bioinformatics database (http://genome.ucsc.edu). The genomic positions are reported according to their mapping on the GRCh37/hg19 genome build.

changes in chromosome 7q11 (MRC-Holland, Amsterdam, Netherlands). DNA denaturation, hybridization of probes, ligation, and PCR were performed according to the manufacturer’s instructions, as described by Schouten et al. (11). Separation of the amplification products via electrophoresis was performed using an ABI 3500 Genetic Analyzer (Thermo Fisher Scientific, Waltham, Massachusetts, USA), and the data were analyzed using GeneMarker software, version 1.6 (www. softgenetics.com-Softgenetics, State College, Pennsylvania, USA). The peak area of each fragment was compared with that of a control sample, and the results were considered abnormal when the relative peak-height ratio was less than 0.75 (deletion) or greater than 1.25 (duplication). The details of the regions and probes detected by each kit can be found at www.mlpa.com. The arrays were employed on three different platforms, from Agilent Technologies (Santa Clara, California, USA), Affymetrix (Santa Clara, California, USA) and Illumina (San Diego, California, USA), which differ in the technology used. On the Agilent platform, we used the Human Genome CGH Microarray 2x105K slide, containing 105,750 probes with an average spacing of 22 kb, the SurePrint G3 Human CGH Microarray 4x180K slide, containing 180,880 probes distributed throughout the genome with an average spacing of 13 kb, and the SurePrint G3 Human CGH Microarray 8x60K slide, containing 62,976 probes with an average spacing of 41 kb. On the Affymetrix platform, we used the Affymetrix Genome-Wide Human SNP Array 6.0 chip (1.8 million genetic markers), which contains 906,600 single-nucleotide polymorphism (SNP) probes and over 946,000 probes for the detection of CNVs, with a median physical inter-marker distance of 1-5 kb, as well as the CytoScan HD chip, which contains 2,696,550 CNV probes and 749,157 SNP probes, with an average spacing of 1.1 kb. On the Illumina platform, we employed the HumanCytoSNP-12 BeadChip, with 300,000 oligonucleotide probes and an average spacing of 9.7 kb, and the CytoSNP-850K, with 843,888 markers and an average probe spacing of 1.8 kb across the whole array. In all samples, amplification, hybridization, staining and washing were performed according to the manufacturers’ protocols, and the data were extracted by a specific scanner. The CGH arrays are based on the principle of comparison between the signal intensities of a sample and commercially acquired human male control DNA (Promega Corporation, Madison, Wisconsin, USA). For the SNP arrays (Affymetrix) and bead arrays (Illumina), only a single hybridization is performed for the patient DNA, and the signal intensities are then compared with a reference dataset based on pre-run reference samples. The raw data were analyzed using Feature Extraction v9.5, Affymetrix Chromosome Analysis Suite (ChAS) v.1.2, or KaryoStudio v1.4.3.0 Build 37 software. The data were normalized, and log2 ratios were calculated by dividing the normalized intensity of the sample by the mean intensity across the reference sample. The criteria used to determine a CNV included the involvement of at least five consecutive probes sets in a region and log2 ratio cut-offs of -0.41 and +0.32 for loss and gain, respectively. The software produced graphical representations of CNV breakpoints for each sample.

Ethics The Research Ethics Committee of the Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HC-FMUSP) approved this study, and written informed consent for publication was obtained from the parents of the patients (CAPPesq no 0619/11).

’ RESULTS In this study, we assessed 93 patients with DD and MCAs via the MLPA and array techniques. The patients showed either a normal karyotype or a karyotype with an undetermined abnormality according to G-banding, which made it impossible to obtain a conclusive diagnosis. We found that B97.8% (91/93) of the results from the two methods were consistent with each other (all results are described in Table 1). Among the evaluated patients, B13.2% (12/91) showed no alterations according to either technique; B54.9% (50/91) only showed changes in the array analysis; and B39.9% (29/91) of the patients showed CNVs according to both techniques (Figure 1). One case with inconclusive results was found in our cohort, and further evaluation using other molecular techniques should be performed to definitively diagnose this patient. Although the changes observed using both techniques were consistent, the breakpoint determined by the array did not correspond exactly to the genomic localization of the MLPA probe, and there were several array probes between these two probes. The MLPA results were inconsistent with the array results in two cases. We found a duplication in the FZD9 gene in one case (P064 and P029), and in the other, we identified two alterations (del 16p13.3 with the P036 kit and del 19p13.3 with the P070 kit) using MLPA, which were confirmed via independent reactions. However, these alterations were not identified with the array because none of the array probes are located at exactly the same position as the MLPA probe. Several of the MLPA results were inconclusive, but this did not affect the comparison of the techniques because the regions targeted by MLPA were repeated in several of the kits used in this study. Thus, the results were concordant, and although the results were not the same in all cases, the

528

529

32 33

29 30

26 27

20 21

18 19

15 16 17

08 09 10 11 12 13 14

dup dup dup dup dup

04 05 06 07

del 6q24.3-q25.1 del 17q23.3 del Xq22.1 del 4q34.3-q35.2 dup 5q34-q35.3 dup 12q24.32q24.33 dup 15q11.1-q21.2 del 9p24.3-p24.2 dup 18q12.3-q23 dup 5p15.33-p13.3 dup 14q11.2-q12 dup 8p23.2 dup 9p13.1-p12 dup 11q24.2 del 4p16.3-p16.1 dup 16p13.11 dup 10q11.22 del 16p12.2 del 7q11.23 dup Xq24 LOH 4q24-q26 LOH 4q32.3-q34.1 LOH 17p13.2-p12 LOH 17q21.2-q22 del 3p13-p12.1 del Xp11.23 del 7p21.1-p15.3 del 22q13.2 del 8q24.23 dup Xp22.33

dup Xq22.2 del 1q21.1-q21.2 dup 12q13.11 del 8p23.2 del 16p11.2 del 6q25.2-q27 dup Xp22.33 dup 22q13.31 dup 22q11.22 del 22q11.21 dup Xq28

7q11.23-q22.1 22q11.21 18q22.2-q22.3 17q21.31 22q11.21

del 17p11.2 del 19p13.12 del 22q11.21

CNVs

01 02 03

148,971,363 61,947,000 99,904,100 179,962,284 160,148,716 126,850,508

20,375,156 199,953 39,129,720 37,692 19,361,358 2,310,313 40,294,324 126,501,321 48,283 15,052,746 46,947,635 21,599,125 72,722,981 117,394,974 102,641,428 166,848,001 6,004,639 38,640,744 74,143,047 47,871,775 20,703,948 41,036,329 137,730,280 93,118

52,129,171 4,366,197 78,012,829 33,434,546 25,127,451 2,581,969 42,374,011 126,671,287 6,471,246 16,289,532 47,741,321 21,740,231 74,138,121 117,742,647 118,463,264 175,764,593 12,043,573 54,902,055 85,618,308 48,001,226 21,582,516 41,640,297 137,850,011 506,344

149,820,948 61,977,500 99,905,800 190,790,881 180,712,253 133,819,092

74,480,670 - 99,700,362 18,844,632 - 18,979,405 68,090,674 - 68,756,043 44,204,373 - 44,788,310 18,877,787 - 19,008,108 No change 103,111,457 - 103,303,968 146,516,199 - 147,828,939 47,608,167 - 47,740,591 4,814,896 - 5,044,296 32,502,868 - 32,951,981 153,258,023 - 165,115,007 1,957,876 - 2,065,015 47,327,892 - 47,675,283 22,314,463 - 22,580,314 18,877,787 - 21,462,353 152,667,088 - 153,878,001 No change No change

17,626,111 - 17,640,000 14,729,069 - 14,768,462 21,034,808 - 21,572,202

Start - End

Array results

31,754,015 4,166,244 38,883,109 33,396,854 5,766,093 271,656 2,079,687 169,966 6,423,143 1,236,786 793,686 141,106 1,415,140 347,673 15,821,836 8,916,592 6,038,934 16,261,311 11,475,261 129,451 878,568 603,968 119,731 413,226

849,585 30,500 1,700 10,828,597 20,563,537 6,968,584

192,511 1,312,740 132,424 229,400 449,113 11,856,984 107,139 347,391 265,851 2,584,566 1,210,913

25,219,692 134,773 665,369 583,937 130,321

13,889 39,393 537,394

Size (pb)

Pathogenic Benign VOUS Pathogenic Benign

VOUS Pathogenic Pathogenic VOUS 4 regions â&#x20AC;&#x201C; LOH

Benign Pathogenic VOUS Pathogenic

Pathogenic

VOUS Benign

VOUS Pathogenic Benign Benign Benign Pathogenic Benign Benign Benign Pathogenic

Pathogenic VOUS VOUS Pathogenic VOUS

Pathogenic Benign VOUS

Classification

nml nml

nml

nml del 7q11.23 typical

nml

nml nml

nml

dup 15q11.12 typical

dup 5q35.3 typical

nml dup 7q11.23 atypical (FZD9) nml nml

nml nml del 22q11.21 typical

nml nml nml nml nml nml nml

nml nml del 22q11.21 atypical (SNAP29) nml nml nml nml

Kit P064

Table 1 - Description of cytogenomic results obtained via the MLPA and array techniques.

nml nml

nml

nml nml

del 4p16.3

del 9p24.3; dup 18q23 dup 5p15.33; dup 14q11.2-cen nml inconclusive

dup 5q35.3; del 4q35.2 dup 12q24.33; dup 15q11.2-cen

nml nml

nml nml nml

nml nml nml nml nml nml nml

nml nml nml nml

nml nml nml

Kit P036

nml nml

nml

nml nml

del 4p16.3

del 9p24.3; dup 18q23 dup 5p15.33; dup 14q11.2-cen nml nml

dup 5q35.3; del 4q35.2 dup 12q24.33; dup 15q11.2-cen

nml nml

nml nml nml

nml nml nml nml nml nml nml

nml nml nml nml

nml nml nml

Kit P070

Kit P250

nml -

nml

nml nml

inconclusive

inconclusive inconclusive del 22q11.21 typical

del 22q11.21 atypical (SNAP29 and LZTR1) nml nml nml

MLPA results

nml -

nml

nml nml del 22q11.21 typical

inconclusive dup 22q11.21 atypical (PRODH)

nml

Kit P356

dup 7q11.23 atypical (FZD9) nml -

nml -

Kit P029

CLINICS 2017;72(9):526-537 Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

530

41 42

38 39

Array results

71,904 19,571,776 174,466,591 204,104 148,092 74,307,023 203,291,000 189,360,000 78,667,293 180,300,936 129,676,581 32,562,410 41,692,304 44,727,846 44,864,687 564,620

2,473,257 3,474,630 137,730,280 71,021,037 137,730,280 20,213,937 28,698,698 8,105,359 114,783,837 137,900,733 115,745,240 53,011

del 1p36.32 dup 1p36.32 del 8q24.23 dup 7q11.22 del 8q24.23 dup 14q11.2 LOH 7p15.1-p12.1 LOH 8p23.1-p22 LOH 8q23.3-q24.23 LOH 8q24.23-q24.3 LOH 9q32-q34.11 LOH 17p13.3-13.1

6,143,107 45,212,898 9,353,507 11,047,140 100,351,154 41,897,482 93,632,889 107,328,319 156,586,155 122,697,234 95,842,069 106,386,553 11,761,688 8,386,306 10,082,476 161,623,467 13,798,819 90,027,810 239,550,182 172,176,461 47,087,371 47,330,328 13,468,616 45,130 116,878,379

3,446,813 3,641,681 137,850,011 71,272,257 137,850,011 20,379,392 52,857,194 18,289,407 137,679,805 146,293,086 130,633,433 9,193,945

2,425,306 59,311,250 197,845,254 11,659,355 2,310,571 80,943,189 203,312,000 189,364,000 82,400,000 180,394,157 129,896,364 32,615,311 44,244,868 44,824,251 45,723,022 2,456,203

6,248,244 45,359,483 9,546,184 11,608,207 100,589,056 90,442,925 97,474,630 149,605,182 170,898,549 131,869,597 102,302,850 115,106,996 27,853,219 16,372,158 15,254,051 190,880,409 14,177,667 90,247,720 243,029,573 180,705,539 47,756,480 47,675,283 14,566,406 1,691,646 133,819,092

Start - End

del 8p23.2-p23.1 dup 10q11.21 dup Xp22.31-p22.2 dup Xp22.2 dup 15q26.3 LOH 3p22.1-p11.1 LOH 3q11.1-q11.2 LOH 6q21-q25.1 LOH 6q25.3-q27 LOH 10q26.12-q26.3 LOH 13q32.1-q33.1 LOH 13q33.2-q34 LOH 16p13.13-p12.1 LOH 19p13.2-p13.11 LOH 20p12.2-p12.1 del 4q32.1-q35.2 dup 5p15.2 del 2p11.2 del 2q37.3 dup 5q35.1-q35.3 dup 10q11.22 dup 22q13.31 del 9p23-p22.3 del 7p22.3 dup 12q24.22q24.33 dup 5p15.33 del Yq11.221-q12 dup 3q26.31-q29 del 9p24.3-p23 del 17p13.3 dup 17q25.1-q25.3 del 2q33.1 del 3q28 LOH Xq21.1 del 1q25.3 dup 3q22.1 del 9p21.1 dup 9p11.2 del 9p11.2 dup 9p11.2 del 1p36.33-p36.32

CNVs

Table 1 - Continued.

973,556 167,051 119,731 251,220 119,731 165,455 24,158,496 10,184,048 22,895,968 8,392,353 14,888,193 9,140,934

2,353,402 39,739,474 23,378,663 11,455,251 2,162,479 6,636,166 21,000 4,000 3,732,707 93,221 219,783 52,901 2,552,564 96,405 858,335 1,891,583

105,137 146,585 192,677 561,067 237,902 48,545,443 3,841,741 42,276,863 14,312,394 9,172,363 6,460,781 8,720,443 16,091,531 7,985,852 5,171,575 29,256,942 378,848 219,910 3,479,391 8,529,078 669,109 344,955 1,097,790 1,646,516 16,940,713

Size (pb)

8 regions – LOH

Benign

Pathogenic Benign VOUS Pathogenic

1 region – LOH VOUS

Benign

Pathogenic

Pathogenic Benign Benign Pathogenic

Benign Pathogenic

Pathogenic

Benign 10 regions – LOH

Benign

Classification

nml

del 1p36 atypical (TP73 nml)

nml

del 17p13.3 atypical (HIC and METTL16) nml

nml

nml nml

nml

nml dup 5q35.3 typical

nml

Kit P064

nml

del 1p36.33

nml

dup 5p15.33; del Yq12 dup 3q29; del 9p24.3 del 17p13.3; dup 17q25.3 nml

nml del 7p22.3; dup 12q24.33

nml del 2q37.3; dup 5q35.3 nml

del 4q35.2

nml

Kit P036

nml

del 1p36.33

nml

dup 5p15.33; del Yq12 dup 3q29; del 9p24.3 del 17p13.3; dup 17q25.3 nml

nml del 7p22.3; dup 12q24.33

nml del 2q37.3; dup 5q35.3 nml

del 4q35.2

nml

Kit P070

MLPA results

nml -

nml

Kit P250

nml -

nml

Kit P356

Kit P029

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al. CLINICS 2017;72(9):526-537

531

64 65 66 67 68 69 70 71 72 73

dup 10q11.21 del Xp11.23 dup 9p13.1-p12

dup 10q11.22 del 8p21.3-p21.2

LOH 22q12.3-q13.1 LOH 22q13.31-q13.33 dup 4q28.3 del 22q11.23-q12.1 dup Xq22.2 del Xp22.13p22.12 dup 14q32.33 dup 7q11.23 del 5q12.1 del 4q35.1-q35.2 dup Xq27.1-q28 dup Xp22.33 dup Xq28 dup 9p24-p23 del 18q22-q23 dup Xp22.31 del 2q37.3 dup 5q35.1-q35.3 dup 18q12.1 del 4p16.3-p16.1 dup 8p23.3-p23.1 dup 4q26-q35.2 dup 6q27 del 7p21.2 del 7q36.3 dup 6p22.3-p12.3 dup 2q22.2-q22.3 dup 10q11.22 dup 2p25.3-p24.3 del 4q35.1-q35.2 dup 6q27 del 7q11.23 del 7q11.23 del 7q11.23 dup Xp22.33 dup Yp11.31-q11.23 dup 7p14.3 del 13q31.3 dup 16q24.1-q24.3 dup 14q11.2 del 16p13.3 dup 22q11.22 del 16p13.3 dup Xq22.2

CNVs

Table 1 - Continued.

40,864,782 51,169,045 132,305,574 25,910,879 103,303,968 19,719,264

106,067,618 -106,823,886 76,143,705 - 76,615,349 59,209,183 - 59,522,613 185,821,036 - 190,880,409 139,513,770 - 154,929,412 2,139,005 - 2,319,653 154,939,018 - 155,235,833 46,587 - 13,014,232 70,657,389 - 78,014,582 7,811,750 - 8,115,453 239,550,182 - 243,029,573 172,246,068 - 180,705,539 27,778,530 - 28,050,968 48,283 - 9,370,908 176,818 - 6,974,050 118,777,687 - 190,880,409 168,329,404 - 168,612,631 14,436,385 - 14,737,999 158,498,994 - 159,119,486 24,247,896 - 50,203,633 143,387,612 - 145,082,658 46,972,140 - 47,681,957 72,184 - 14,844,939 186,468,992 - 190,880,409 168,336,052 - 168,596,251 72,569,012 - 72,685,658 73,082,174 - 74,267,872 74,298,092 - 74,601,104 192,991 - 2,693,037 0 - 28,800,000 33,134,410 - 33,193,210 94,422,000 - 94,480,000 85,817,324 - 90,148,796 20,213,937 - 20,425,051 105,320 - 203,254 22,314,463 - 22,573,637 227,406 - 828,466 103,173,049 - 103,303,968 No change No change 47,084,916 - 47,741,321 23,148,930 - 23,310,904 No change 45,212,898 - 45,359,483 47,871,775 - 47,985,557 40,294,324 - 42,374,011 No change No change

33,850,168 45,136,360 131,880,992 25,732,697 103,179,170 18,179,714

Start - End

Array results Size (pb)

146,585 113,782 2,079,687

656,405 161,974

756,268 471,644 313,430 5,059,373 15,415,642 180,648 296,815 12,967,645 7,357,193 303,703 3,479,391 8,459,471 272,438 9,322,625 6,797,232 72,102,722 283,227 301,614 620,492 25,955,737 1,695,046 709,817 14,772,755 4,411,417 260,199 116,646 1,185,698 303,012 2,500,046 28,800,000 58,805 58,000 4,331,472 211,114 97,934 259,174 601,060 130,919

7,014,614 6,032,685 424,582 178,182 124,798 1,539,550

nml

Benign Pathogenic

Benign Benign Benign

Benign Benign

VOUS

Pathogenic Benign

Benign

VOUS Pathogenic

Pathogenic

Pathogenic VOUS

nml nml inconclusive nml nml nml nml nml nml nml

nml

del 7q11.23 atypical (FZD9 nml)

nml

dup 5q35.3 typical

Benign Pathogenic

Pathogenic Benign VOUS

nml

Pathogenic

Benign

nml

VOUS Pathogenic Pathogenic Benign VOUS Pathogenic

nml

Kit P064

Benign

Classification

nml inconclusive inconclusive nml nml del 16p13.3 nml nml nml nml

del 16p13.3; dup 16q24.3

dup Yp11.32; dup Yq12

inconclusive

nml

del 4p16.3; dup 8p23.3 dup 4q35.2; del 7q36.3

del 2q37.3; dup 5q35.3

dup 9p24.3; del 18q23

del 4q35.2; dup Xq28

nml

Kit P036

nml nml nml nml nml del 19p13.3 nml nml nml nml

del 16p13.3; dup 16q24.3

dup Yp11.32; dup Yq12

dup 2p25.3; del 4q35.2

nml

del 4p16.3; dup 8p23.3 dup 4q35.2; del 7q36.3

del 2q37.3; dup 5q35.3

dup 9p24.3; del 18q23

del 4q35.2; dup Xq28

nml

Kit P070

nml

Kit P250

inconclusive

del 4q35 (KLKB1)

MLPA results

nml

Kit P356

del 7q11.23 atypical (FKBP6, FZD9 and TBL2 nml)

Kit P029

CLINICS 2017;72(9):526-537 Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

532

del 5q14.3-q15

del 8p23.1 dup 2q22.3-q23.1 dup 5p13.2 dup 6p21.1 dup 8q22.2 dup 11p15.2 dup 13q31.3 dup 17q11.2 dup 7q11.1-q11.21 dup 12p11.1 dup 17q11.2 dup 17q11.2 del 13q12.12 del 22q11.21 dup 17q21.31 del 22q11.21 dup 1p21.1 dup 5p13.2 dup 17q11.2 del 22q11.21 del 15q11.2 del 16p12.2 dup 7p22.3-p21.1 dup 7p21.1-p15.2 dup 2q24.3 dup 5p13.2 dup 9p24.1 dup 22q11.21 dup 7q21.3 del 9p23-p22.3 del 9p24.1 del Xq25 dup Xp22.33 del 9p23-p22.3 dup 22q11.23-q12.1

Array results

27,266,362 - 28,693,558 No change No change 90,027,810 - 90,247,720 14,687,571 - 15,214,708 27,151,611 - 27,337,941 110,863,908 - 110,982,530 95,467,621 - 96,178,713 13,468,616 - 14,566,406 25,732,697 - 25,910,879 No change 6,982,980 - 12,483,094 148,649,175 - 148,956,584 36,902,936 - 37,159,877 44,810,418 - 45,334,537 100,111,153 - 100,528,645 14,504,463 - 14,906,450 92,492,127 - 92,815,210 29,574,712 - 29,699,649 61,074,194 - 62,403,985 34,362,752 - 34,853,011 29,444,844 - 29,562,294 29,574,712 - 29,699,649 23,548,470 - 24,960,000 18,886,915 - 21,463,730 44,246,211 - 44,580,136 18,889,490 - 20,312,668 103,155,605 - 103,510,258 36,816,661 - 37,158,123 29,479,196 - 29,697,251 18,886,915 - 20,312,668 24,357,212 - 24,472,002 21,578,388 - 21,839,340 44,935 - 19,155,339 19,159,422 - 26,403,574 166,821,406 - 166,939,893 36,877,640 - 37,158,123 5,112,844 - 5,252,074 18,886,915 - 19,008,108 95,467,621 - 96,178,713 13,468,616 - 14,566,406 8,012,608 - 8,227,101 126,923,848 - 127,145,037 1,921,638 - 2,065,015 13,466,329 - 14,566,406 25,650,648 - 25,910,879 No change 90,124,906 - 94,954,205

152,667,088 - 153,903,395 No change

Start - End

4,829,299

5,500,114 307,409 256,941 524,119 417,492 401,987 323,083 124,937 1,329,791 490,259 117,450 124,937 1,411,530 2,576,815 333,925 1,423,178 354,653 341,462 218,055 1,425,753 114,790 260,952 19,110,404 7,244,152 118,487 280,483 139,230 121,193 711,092 1,097,790 214,493 221,189 143,377 1,100,077 260,231

219,910 527,137 186,330 118,622 711,092 1,097,790 178,182

1,427,196

1,236,307

Size (pb)

Pathogenic

Benign

Pathogenic

nml nml

nml

dup 7p typical

del 22q11 atypical

Benign Pathogenic Benign

Pathogenic Benign

del 22q11 typical

nml

nml nml

nml nml nml nml

nml nml

Kit P064

Pathogenic

Benign

Pathogenic Benign

Benign Benign

Benign

VOUS

Pathogenic

Classification

nml nml

nml

dup 7p22.3

nml

nml nml

nml nml nml nml

nml nml

Kit P036

Abbreviations: Nml, normal; dup, duplication; del, deletion; VOUS, variant of uncertain clinical significance; pb, base pairs.

92 93

82 83

80 81

dup Yq11.23

76 77 78 79

del 2p11.2 dup 21q11.2 del Xp21.3 dup 2q13 dup 7q21.3 del 9p23-p22.3 dup 22q11.23-q12.1

dup Xq28

CNVs

74 75

Table 1 - Continued.

nml nml

nml

dup 7p22.3

nml

nml nml

nml nml nml nml

nml nml

Kit P070

nml

nml nml -

nml

Kit P250

del 22q11 atypical

del 22q11 typical

nml

del 8p23 typical

MLPA results

del 22q11 atypical

inconclusive

nml

nml nml -

nml

Kit P356

Kit P029

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al. CLINICS 2017;72(9):526-537

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

CLINICS 2017;72(9):526-537

Figure 1 - Cytogenomic map of the raw data of all alterations identified via the MLPA and array techniques. The gray circles represent the locations of the breakpoints of the alterations identified by both techniques, in which the center circle corresponds to the MLPA results and the middle circle to the array results. Each bar refers to the position of each identified copy number change: the red bar refers to deletions, the blue to duplications, and the green to loss of heterozygosity. The genomic positions are reported according to their mapping on the GRCh38/hg38 genome build from the UCSC Genome Browser.

MLPA technique detected B70.6% of the pathogenic CNVs detected using the array.

deletion in chromosome 8p23 (three probes) was detected with the P250 kit.

ARRAY Analysis

MLPA Analysis

The array technique was applied to all patients using different platforms (Agilent, Affymetrix or Illumina) and chip densities. The results showed that B14% (13/93) of the patients did not exhibit CNVs, while B86% (80/93) exhibited several different genomic alterations, including deletions, duplications and loss of heterozygosity (LOH). These changes were classified as pathogenic, benign or VOUS. Among the patients showing changes in the genome, we observed a 46.25% (37/80) detection rate for patients with benign and/or VOUS CNVs and a 53.75% (43/80) rate for patients with at least one pathogenic change (Figure 3). Among the patients with pathogenic CNVs, B51.2% (22/ 43) exhibited only one alteration that was considered pathogenic, while B44.2% (19/43) showed at least two changes with important clinical significance, and B4.6% (2/43) of patients exhibited three or more pathogenic CNVs, possibly due to complex rearrangements. In several cases, these patients with pathogenic changes also displayed concomitant benign changes or VOUS. Regarding the size of the changes, the majority of patients exhibited benign CNVs or VOUS ranging from 100 to 500 kb and pathogenic CNVs that were larger than 1 Mb.

The MLPA technique was employed to diagnose all patients using several different kits. No changes were detected in B66.7% (62/93) of the patients, and in four cases, one or two kits showed inconclusive results; however, these cases did not influence the assessment and interpretation of the results. CNVs were detected with at least one of the kits in B33.3% (31/93) of patients (Figure 2). Approximately 22.6% (7/31) of these changes were detected by the P064 kit, corresponding to one deletion typical of the Williams-Beuren syndrome, one duplication in chromosome 7q11, and five deletions of 22q11.2, which were atypical in three patients and typical in the other two patients. All alterations were confirmed by the specific P029, P250 and/or P356 kits. We also detected subtelomeric alterations in B45.2% (14/ 31) of the patients. One deletion was detected in two patients; two duplications in different chromosomes were detected in one patient; two deletions were found in another patient, one of which was detected with the P036 kit and the other with the P070 kit; and the remaining 10 patients showed concomitant deletions and duplications, all of which were present in the subtelomeric regions of different chromosomes. The MLPA test also allowed us to simultaneously detect CNVs with all of the main kits used in this study (P064, P036 and P070); these changes were identified in B25.8% (8/31) of the patients. One atypical duplication (in the PRODH gene) was only detected by the P356 kit, specific for chromosome 22, and one

â&#x20AC;&#x2122; DISCUSSION Establishing an unequivocal clinical and molecular diagnosis for patients with DD and MCA is essential for correlating genotypes and phenotypes and making genetic counseling more effective.

533

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

CLINICS 2017;72(9):526-537

Figure 2 - The results of MLPA. The blue bar indicates the number of duplications; the red bar indicates deletions; and the green bar indicates the number of normal results detected via MLPA.

Figure 3 - The number of CNVs identified on each chromosome via the array technique. The red bar indicates pathogenic CNVs; the blue bar indicates benign CNVs; the gray bar indicates VOUS; and the green bar indicates LOH.

syndrome, whereas for others it is possible to determine alterations in an increasing number of critical regions associated with specific clinical characteristics (1,6).

With advances in cytogenomic techniques, different syndromes can be better evaluated. Thus, for certain changes, specific genes are now highlighted as being responsible for most of the clinical features of a defined

534

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

CLINICS 2017;72(9):526-537

These alterations are rare and difficult to detect because they involve specific genes or exons that are associated with few clinical characteristics, or a phenotype present in most patients, making it difficult to determine the correct kit to use. An important limitation of MLPA is that the signal intensity of the probes varies according to DNA characteristics, including those associated with the extraction method, storage time, elution solution, degree of degradation (if present), and the presence of several types of contaminants, such as extraction reagents, proteins, RNAs, and salts. These influences can be minimized if all samples are prepared by the same technician using the same method. However, it is not always possible to eliminate this bias because samples may be sent from other locations, and storage times and DNA extraction methods may differ from the standard, which can cause artifacts during analysis that only a specialist can identify (8,18). In our analyses using the MLPA technique, 4 patients showed inconclusive results with one or two of the kits, but none of these findings limited the detection of changes because the surveyed regions were represented in the other kits used in this study. These data highlight the importance of using different combinations of kits because one kit can act as a control for another, confirming the alterations detected and excluding false positive and negative results (10,32). In a study performed by Marenne et al. (2), MLPA was used to validate data from arrays. DNA from 56 patients were analyzed via MLPA in two independent reactions, providing a concordance rate of 97.25%. Therefore, MLPA is a reproducible technique. The sizes and breakpoints of chromosomal abnormalities can currently be determined with greater precision, accuracy and sensitivity using array techniques (6,19). All of the patients included in our study were assessed using the array technique according to the availability of platforms or slides/chips in the laboratory (Agilent, Affymetrix or Illumina). The slides/chips differ in the technologies involved (CGH, oligonucleotides or beads) and in the number and spacing of probes distributed throughout the genome. Technologies with higher genome coverage provide more accurate breakpoint data and can be used to diagnose micro changes or several CNVs that were previously considered a single alteration (e.g., a normal region interposed by two affected regions). In these cases, the low coverage of several arrays may determine those changes to be a single deletion and not a complex rearrangement that may reflect a change in the patientâ&#x20AC;&#x2122;s phenotype (4,19,33). A total of 93 samples were evaluated, and all of the different technologies employed proved to be satisfactory for detecting variations in the genome, which in most cases corroborated the clinical characteristics of each patient. The data included results that were considered normal (without changes) for B14% of the patients. This rate is much lower than that described in the literature. In 2013, VallespĂn et al. (27) evaluated 540 samples (patients with learning disabilities, autism and/or multiple congenital malformations) using a customized array with an average coverage of B43 kb and showed that no CNVs were detectable in 31.85% of the patients. In this study, the samples that were considered normal were assessed using Agilent 180K (2/13 patients), Agilent 60K (1/13 patients) and Illumina (10/13 patients) arrays, all of which exhibit a high rate of genome coverage. The results (particularly those from the Illumina platform; 65 samples), were considered normal because the majority of the evaluated patients had not received a

Currently, the MLPA technique has become very useful for the detection of the main microdeletion/microduplication syndromes and subtelomeric imbalances, as it is a rapid technique that is able to detect typical changes correlated with specific phenotypes (e.g., Williams-Beuren syndrome or deletion of 22q11.2), in addition to being detecting small and/or atypical deletions and duplications in target regions (9,15,16). MLPA has the ability to assess more than 45 target regions in a single reaction without cell culture, making it a cost-effective and widely used technique for the validation of other methods, such as array-based analysis (12,15). In this study, MLPA analysis using the P064 and/or P036 and P070 kits detected alterations in approximately 33.3% of patients. Using the same combination of MLPA kits, Jehee et al. (31) identified pathogenic changes in 21.8% of 261 patients with DD and MCA. In a study performed on 258 patients with intellectual disabilities and dysmorphisms in 2007, the rate of the detection of alterations using several kits was 10.1%, among which only 5.8% were changes in regions correlated with syndromes, and 5.0% were associated with subtelomeric regions (15). In the patients included in the present study, the changes identified with a specific kit for the main microdeletion/ microduplication syndromes (P064) corresponded to B7.5% of all samples, or B22.6% of all changes, representing Williams-Beuren syndrome, duplications of chromosome 7q11 and deletions of chromosome 22q11.2. In addition, subtelomeric changes were found in B15.1% of the samples evaluated via MLPA, or B45.2% of the patients with copy number changes. In a similar study, the detection rate for alterations in the regions of the main microdeletion/microduplication syndromes was 6.6%, and the detection rate for subtelomeric alterations was 7.3% (10). The percentage of copy number changes detected in the genome via MLPA depends on the criteria used to select patients, and the data obtained in this study corroborate the data reported in the literature for the regions corresponding to the main syndromes. However, the obtained values for subtelomeric regions were higher than those previously described by several authors. A subtelomeric analysis conducted by Koolen et al. (14) detected changes in 6.7% of 210 patients with idiopathic intellectual disabilities. Two years later, Palomares et al. (32) detected alterations in 10% of patients with the same phenotypic characteristics using subtelomeric kits. With the exception of two cases, all of the patients who presented only subtelomeric abnormalities exhibited two changes: one deletion associated with one duplication on different chromosomes, or two deletions or duplications. This set of changes in the same patients may result from complex rearrangements and translocations between chromosomes or regions of instability that are susceptible to rearrangements via DNA repair mechanisms. We also detected changes with the three main kits used in this study (P064, P036 and P070) accounting for B25.8% of the CNVs identified among the abnormal results. These alterations may result from a microdeletion syndrome located near the telomere of a chromosome, such as 1p36 deletion syndrome, or complex rearrangements between different regions of chromosomes due to instability and microhomology. In addition to the changes detected by the main kits used in this study, we were able to identify an atypical change involving a single gene (2 exons evaluated) using the P356 kit and a deletion in 8p23 (3 genes evaluated) using the P250 kit.

535

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

CLINICS 2017;72(9):526-537

All of the changes detected in the present study were checked against several international databases, including the DGV, Decipher and UCSC databases. Nevertheless, a more appropriate assessment of the changes identified in our patients would result in the creation of a database with information specifically from Brazilian people. Most of the obtained results (B97.8%) were concordant with each other for the regions investigated. However, not all of the results were in agreement, as the MLPA technique covers approximately 45 specific regions of the genome in each available kit, and this technique therefore depends on a clinical features and direction toward a specific target. Approximately 54.9% of the CNVs were not detected via MLPA compared with array analysis, and higher rates for this comparison (72-81%) are reported in the literature (2). Despite the presence of the same alteration, one case was discordant in relation to the breakpoints detected via array analysis and the position of the MLPA probe. Therefore, to obtain a conclusive molecular diagnosis, other techniques should be applied to reevaluate the exact breakpoints involved. All of the techniques employed in this study have advantages and disadvantages depending on the application and could potentially be applied together to obtain a complete molecular diagnosis. Our findings showed that the interpretation of genotypephenotype correlations in patients with complex genomic rearrangements is very difficult, but these results can directly contribute to the elucidation of new syndromes. Arrays are a powerful tool for the identification and characterization of genomic abnormalities and can provide accurate diagnoses of previously unidentified or unexplained diseases that are suspected to have a genetic cause, contributing to appropriate clinical management of the affected patients. When an array is not available, MLPA with a combination of three kits (P064, P036 and P070) is a remarkable tool that can detect abnormalities in patients with DD and MCA (10,15,31). Clinical and laboratory interactions with skilled technicians are required to target a patient for the most effective and beneficial molecular diagnosis, in which an appropriate clinical hypothesis is crucial for the successful detection of changes. Patients exhibiting normal results or benign alterations may present a clinical phenotype due to balanced rearrangements with disruptions in several genes or mutations in specific genes. In this case, other molecular techniques are required to achieve a complete diagnosis, such as exome sequencing, which can detect changes in 80% of patients with developmental delays of unknown cause, and analysis using normal arrays (4,20).

suspected clinical diagnosis. These patients should be further evaluated and subjected to exome sequencing or targeted tests searching for mutations in specific genes or gene disruptions due to unbalanced translocations (4,20). Among the patients who presented alterations in the genome, the array technique showed that 46.25% of the patients presented benign changes or changes of uncertain clinical significance, while 53.75% of the patients presented at least one pathogenic change. Among the patients exhibiting alterations of clinical significance, the majority of patients presented only one or two pathogenic changes in the genome, which were or were not combined with other alterations, corresponding to B51.2% and B44.2% of the patients, respectively. Complex alterations with three or more pathogenic CNVs in different regions were observed in approximately 4.6% of the patients. The detection rate of pathogenic alterations visualized in this study was much higher than the rates previously reported in several articles. Rosenberg et al. (34) investigated 81 patients with intellectual disabilities and facial dysmorphisms via the CGH array technique and concluded that 16% of the patients exhibited a pathogenic chromosomal imbalance related to their phenotype, while 4% of the patients exhibited changes of uncertain clinical significance. Gijsbers et al. (25) used several SNP array platforms to investigate patients with intellectual disabilities and multiple congenital abnormalities and detected alterations in 22.6% of 318 evaluated patients. Therefore, array analysis was considered the most appropriate test for the initial molecular investigation of patients with these characteristics and normal karyotypes. Hochstenbach et al. (28) also recommended arrays as the first diagnostic test in this patient group. Based on analyzing many studies, they concluded that the rate of detection using arrays would correspond to at least 19% of pathogenic changes. Other studies have shown similar rates, regardless of the platform selected to diagnose patients with intellectual disabilities, malformations and/or neurological disorders and normal karyotypes (20,27,28). Regarding the size of the observed changes, we identified the greatest number of patients with pathogenic CNVs that were larger than 1 Mb. These large changes usually involve more causative genes of a disease. However, the severity of the clinical manifestations in patients is not necessarily directly correlated with the size of the change but is correlated with location and gene content. Therefore, a small change can potentially reflect a more severe phenotype due to the pathogenicity of the altered gene (1,35). With the implementation of SNP arrays, it has become possible to identify changes that were previously undiagnosed using CGH arrays. In this study, we identified four patients with LOH or UPD regions that can be correlated with recessive disorders (20,24,25). The main challenge in analyzing the results of the arrays is determining which changes are significant for each patient, as it is common to identify more than one change per patient, and all of the changes could potentially influence the phenotype in many cases. The identification of benign and VOUS changes is associated with the increased array density used for diagnosis, as arrays with a greater number of probes are able to identify a greater number of microalterations and determine the breakpoints of these changes with higher accuracy. However, the identification of regions involving genes without an established function or regions that do not contain well-described genes will also increase (24,27,29).

’ ACKNOWLEDGMENTS We thank all of the children who participated in this study and their parents. This study was supported by grants from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP).

’ AUTHOR CONTRIBUTIONS Zanardo EA wrote the paper and performed cytogenomic analysis. Dutra RL performed cytogenomic analysis and genotype-phenotype correlations. Piazzon FB performed the clinical evaluation and cytogenomic analysis. Dias AT, Novo-Filho GM and Montenegro MM performed molecular

536

Cytogenomic assessment of 93 patients with DD/MCA Zanardo EA et al.

CLINICS 2017;72(9):526-537

analysis and classical cytogenetic analysis; Nascimento AM prepared the samples and performed DNA extraction; Damasceno JG created the graphics and images. Madia FA and Costa TV discussed the results. Melaragno MI and Kim CA provided the samples and clinically assessed the patients; Kulikowski LD designed and coordinated the study. All authors read and approved the ﬁnal manuscript.

18. Ahn JW, Ogilvie CM, Welch A, Thomas H, Madula R, Hills A, et al. Detection of subtelomere imbalance using MLPA: validation, development of an analysis protocol, and application in a diagnostic centre. BMC Med Genet. 2007;8:9, http://dx.doi.org/10.1186/1471-2350-8-9. 19. Manning M, Hudgins L, Professional Practice and Guidelines Committee. Array-based technology and recommendations for utilization in medical genetics practice for detection of chromosomal abnormalities. Genet Med. 2010;12(11):742-5, http://dx.doi.org/10.1097/GIM.0b013e3181f8baad. 20. Siggberg L, Ala-Mello S, Linnankivi T, Avela K, Scheinin I, Kristiansson K, et al. High-resolution SNP array analysis of patients with developmental disorder and normal array CGH results. BMC Med Genet. 2012;13:84, http://dx.doi.org/10.1186/1471-2350-13-84. 21. Salman M, Jhanwar SC, Ostrer H. Will the new cytogenetics replace the old cytogenetics? Clin Genet. 2004;66(4):265-75, http://dx.doi.org/ 10.1111/j.1399-0004.2004.00316.x. 22. Edelmann L, Hirschhorn K. Clinical utility of array CGH for the detection of chromosomal imbalances associated with mental retardation and multiple congenital anomalies. Ann N Y Acad Sci. 2009;1151:157-66, http://dx.doi.org/10.1111/j.1749-6632.2008.03610.x. 23. Alkan C, Coe BP, Eichler EE. Genome structural variation discovery and genotyping. Nat Rev Genet. 2011;12(5):363-76, http://dx.doi.org/ 10.1038/nrg2958. 24. Bruno DL, Ganesamoorthy D, Schoumans J, Bankier A, Coman D, Delatycki M, et al. Detection of cryptic pathogenic copy number variations and constitutional loss of heterozygosity using high resolution SNP microarray analysis in 117 patients referred for cytogenetic analysis and impact on clinical practice. J Med Genet. 2009;46(2):123-31, http://dx.doi. org/10.1136/jmg.2008.062604. 25. Gijsbers AC, Lew JY, Bosch CA, Schuurs-Hoeijmakers JH, van Haeringen A, den Hollander NS, et al. A new diagnostic workflow for patients with mental retardation and/or multiple congenital abnormalities: test arrays first. Eur J Hum Genet. 2009;17(11):1394-402, http://dx.doi.org/10.1038/ ejhg.2009.74. 26. Bi W, Borgan C, Pursley AN, Hixson P, Shaw CA, Bacino CA, et al. Comparison of chromosome analysis and chromosomal microarray analysis: what is the value of chromosome analysis in today’s genomic array era? Genet Med. 2013;15(6):450-7, http://dx.doi.org/10.1038/ gim.2012.152. 27. Vallespín E, Palomares Bralo M, Mori MÁ, Martín R, García-Miñaúr S, Fernández L, et al. Customized high resolution CGH-array for clinical diagnosis reveals additional genomic imbalances in previous well-defined pathological samples. Am J Med Genet A. 2013;161A(8):1950-60, http:// dx.doi.org/10.1002/ajmg.a.35960. 28. Hochstenbach R, van Binsbergen E, Engelen J, Nieuwint A, Polstra A, Poddighe P, et al. Array analysis and karyotyping: workflow consequences based on a retrospective study of 36,325 patients with idiopathic developmental delay in the Netherlands. Eur J Med Genet. 2009; 52(4):161-9, http://dx.doi.org/10.1016/j.ejmg.2009.03.015. 29. Miller DT, Adam MP, Aradhya S, Biesecker LG, Brothman AR, Carter NP, et al. Consensus statement: chromosomal microarray is a first-tier clinical diagnostic test for individuals with developmental disabilities or congenital anomalies. Am J Hum Genet. 2010;86(5):749-64, http://dx.doi. org/10.1016/j.ajhg.2010.04.006. 30. Kearney HM, Thorland EC, Brown KK, Quintero-Rivera F, South ST, Working Group of the American College of Medical Genetics Laboratory Quality Assurance Committee. American College of Medical Genetics standards and guidelines for interpretation and reporting of postnatal constitutional copy number variants. Genet Med. 2011;13(7):680-5, http://dx.doi.org/10.1097/GIM.0b013e3182217a3a. 31. Jehee FS, Takamori JT, Medeiros PF, Pordeus AC, Latini FR, Bertola DR, et al. Using a combination of MLPA kits to detect chromosomal imbalances in patients with multiple congenital anomalies and mental retardation is a valuable choice for developing countries. Eur J Med Genet. 2011;54(4):e425-32, http://dx.doi.org/10.1016/j.ejmg.2011.03.007. 32. Palomares M, Delicado A, Lapunzina P, Arjona D, Amiñoso C, Arcas J, et al. MLPA vs multiprobe FISH: comparison of two methods for the screening of subtelomeric rearrangements in 50 patients with idiopathic mental retardation. Clin Genet. 2006;69(3):228-33, http://dx.doi.org/ 10.1111/j.1399-0004.2006.00567.x. 33. Shaffer LG, Bejjani BA. Medical applications of array CGH and the transformation of clinical cytogenetics. Cytogenet Genome Res. 2006; 115(3-4):303-9, http://dx.doi.org/10.1159/000095928. 34. Rosenberg C, Knijnenburg J, Bakker E, Vianna-Morgante AM, Sloos W, Otto PA, et al. Array-CGH detection of micro rearrangements in mentally retarded individuals: clinical significance of imbalances present both in affected children and normal parents. J Med Genet. 2006;43(2):180-6, http://dx.doi.org/10.1136/jmg.2005.032268. 35. Feuk L, Marshall CR, Wintle RF, Scherer SW. Structural variants: changing the landscape of chromosomes and design of disease studies. Hum Mol Genet. 2006;15 Spec No 1:R57-66, http://dx.doi.org/10.1093/hmg/ ddl057.

’ REFERENCES 1. Feuk L, Carson AR, Scherer SW. Structural variation in the human genome. Nat Rev Genet. 2006;7(2):85-97, http://dx.doi.org/10.1038/nrg1767. 2. Marenne G, Rodríguez-Santiago B, Closas MG, Pérez-Jurado L, Rothman N, Rico D, et al. Assessment of copy number variation using the Illumina Infinium 1M SNP-array: a comparison of methodological approaches in the Spanish Bladder Cancer/EPICURO study. Hum Mutat. 2011;32(2): 240-8, http://dx.doi.org/10.1002/humu.21398. 3. Shen Y, Wu BL. Designing a simple multiplex ligation-dependent probe amplification (MLPA) assay for rapid detection of copy number variants in the genome. J Genet Genomics. 2009;36(4):257-65, http://dx.doi.org/ 10.1016/S1673-8527(08)60113-7. 4. Vissers LE, de Vries BB, Veltman JA. Genomic microarrays in mental retardation: from copy number variation to gene, from research to diagnosis. J Med Genet. 2010;47(5):289-97, http://dx.doi.org/10.1136/jmg. 2009.072942. 5. Connolly JJ, Glessner JT, Almoguera B, Crosslin DR, Jarvik GP, Sleiman PM, et al. Copy number variation analysis in the context of electronic medical records and large-scale genomics consortium efforts. Front Genet. 2014;5:51, http://dx.doi.org/10.3389/fgene.2014.00051. 6. Feenstra I, Brunner HG, van Ravenswaaij CM. Cytogenetic genotypephenotype studies: improving genotyping, phenotyping and data storage. Cytogenet Genome Res. 2006;115(3-4):231-9, http://dx.doi.org/10.1159/ 000095919. 7. Emanuel BS, Saitta SC. From microscopes to microarrays: dissecting recurrent chromosomal rearrangements. Nat Rev Genet. 2007;8(11): 869-83, http://dx.doi.org/10.1038/nrg2136. 8. Kozlowski P, Jasinska AJ, Kwiatkowski DJ. New applications and developments in the use of multiplex ligation-dependent probe amplification. Electrophoresis. 2008;29(23):4627-36, http://dx.doi.org/10.1002/ elps.200800126. 9. Cho EH, Park BY, Cho JH, Kang YS. Comparing two diagnostic laboratory tests for several microdeletions causing mental retardation syndromes: multiplex ligation-dependent amplification vs fluorescent in situ hybridization. Korean J Lab Med. 2009;29(1):71-6, http://dx.doi.org/ 10.3343/kjlm.2009.29.1.71. 10. Pohovski LM, Dumic KK, Odak L, Barisic I. Multiplex ligation-dependent probe amplification workflow for the detection of submicroscopic chromosomal abnormalities in patients with developmental delay/intellectual disability. Mol Cytogenet. 2013;6(1):7, http://dx.doi.org/10.1186/17558166-6-7. 11. Schouten JP, McElgunn CJ, Waaijer R, Zwijnenburg D, Diepvens F, Pals G. Relative quantification of 40 nucleic acid sequences by multiplex ligationdependent probe amplification. Nucleic Acids Res. 2002;30(12):e57, http://dx.doi.org/10.1093/nar/gnf056. 12. Jennings LJ, Yu M, Fitzpatrick C, Smith FA. Validation of multiplex ligation-dependent probe amplification for confirmation of array comparative genomic hybridization. Diagn Mol Pathol. 2011;20(3):166-74, http://dx.doi.org/10.1097/PDM.0b013e31820b2517. 13. De Vries BB, Winter R, Schinzel A, van Ravenswaaij-Arts C. Telomeres: a diagnosis at the end of the chromosomes. J Med Genet. 2003;40(6): 385-98, http://dx.doi.org/10.1136/jmg.40.6.385. 14. Koolen DA, Nillesen WM, Versteeg MH, Merkx GF, Knoers NV, Kets M, et al. Screening for subtelomeric rearrangements in 210 patients with unexplained mental retardation using multiplex ligation dependent probe amplification (MLPA). J Med Genet. 2004;41(12):892-9, http://dx.doi.org/ 10.1136/jmg.2004.023671. 15. Kirchhoff M, Bisgaard AM, Bryndorf T, Gerdes T. MLPA analysis for a panel of syndromes with mental retardation reveals imbalances in 5.8% of patients with mental retardation and dysmorphic features, including duplications of the Sotos syndrome and Williams-Beuren syndrome regions. Eur J Med Genet. 2007;50(1):33-42, http://dx.doi.org/10.1016/ j.ejmg.2006.10.002. 16. Fernández L, Lapunzina P, Arjona D, López Pajares I, García-Guereta L, Elorza D, et al. Comparative study of three diagnostic approaches (FISH, STRs and MLPA) in 30 patients with 22q11.2 deletion syndrome. Clin Genet. 2005;68(4):373-8, http://dx.doi.org/10.1111/j.1399-0004.2005. 00493.x. 17. Vorstman JA, Jalali GR, Rappaport EF, Hacker AM, Scott C, Emanuel BS. MLPA: a rapid, reliable, and sensitive method for detection and analysis of abnormalities of 22q. Hum Mutat. 2006;27(8):814-21, http://dx.doi. org/10.1002/humu.20330.

537

CLINICAL SCIENCE

Experimental study of histological changes in vascular loops according to the duration of the postoperative period: Application in reconstructive microsurgery Renata Gregorio Paulos, * Bruno Alves Rudelli, Renee Zon Filippe, Gustavo Bispo dos Santos, Ana Abarca Herrera, Andre Araujo Ribeiro, Marcelo Rosa de Rezende, Teng Hsiang-Wei, Rames Mattar-Jr. Departamento de Ortopedia e Traumatologia, Instituto de Ortopedia e Traumatologia (IOT), Hospital das Clinicas HCFMUSP, Faculdade de Medicina, Universidade de Sao Paulo, Sao Paulo, SP, BR.

OBJECTIVES: To analyze the histological changes observed in venous grafts subjected to arterial blood flow as a function of the duration of the postoperative period to optimize their use in free flap reconstructions. METHOD: Twenty-five rats (7 females and 18 males) underwent surgery. Surgeries were performed on one animal per week. Five weeks after the first surgery, the same five animals were subjected to an additional surgery to assess the presence or absence of blood flow through the vascular loop, and samples were collected for histological analysis. This cycle was performed five times. RESULTS: Of the rats euthanized four to five weeks after the first surgery, no blood flow was observed through the graft in 80% of the cases. In the group euthanized three weeks after the first surgery, no blood flow was observed in 20% of the cases. In the groups euthanized one to two weeks after the first surgery, blood flow through the vascular loop was observed in all animals. Moreover, intimal proliferation tended to increase with the duration of the postoperative period. Two weeks after surgery, intimal proliferation increased slightly, whereas strong intimal proliferation was observed in all rats evaluated five weeks after surgery. CONCLUSION: Intimal proliferation was the most significant change noted in venous grafts as a function of the duration of the postoperative period and was directly correlated with graft occlusion. In cases in which vascular loops are required during free flap reconstruction, both procedures should preferably be performed during the same surgery. KEYWORDS: Vascular Loop; Intimal Hyperplasia; Vascular Remodeling. Paulos RG, Rudelli BA, Filippe RZ, Santos GB, Herrera AA, Ribeiro AA, et al. Experimental study of histological changes in vascular loops according to the duration of the postoperative period: Application in reconstructive microsurgery. Clinics. 2017;72(9):538-542 Received for publication on December 19, 2016; First review completed on May 10, 2017; Accepted for publication on May 10, 2017 *Corresponding author. E-mail: mailto:renata.paulos@hc.fm.usp.br

’ INTRODUCTION

arteriovenous anastomoses one year after the procedure varied between 20% and 50%. One of the factors that may lead to such complications is intimal hyperplasia. This cellular proliferation phenomenon, which is also known as graft arterialization, is due to the increased pressure gradient in the graft after anastomosis. The aim of this study was to analyze the histological changes observed in venous grafts subjected to arterial blood flow as a function of the duration of the postoperative period and therefore optimize their use in free flap reconstructions.

Arteriovenous anastomosis is a procedure performed by various medical specialties and represents an optional treatment for different diseases. This procedure is used for the creation of fistulas in patients with chronic renal failure who require dialysis (1), in coronary artery bypass graft procedures, in free flap reconstruction (2,3), and for the repair of traumatic arterial lesions in general, including reimplantation and emergency revascularization. One of the most important complications of this procedure is graft thrombosis, which may lead to treatment failure (4). Sterpetti et al. (4) reported that the rates of thrombosis of

’ MATERIALS AND METHODS A total of 25 Wistar rats underwent surgery (7 females and 18 males), with a mean weight of 344 (248–440) grams immediately before the procedure. Anastomoses were performed by the same surgeon in a microsurgery laboratory under the supervision of technicians. One animal per week underwent surgery. Five weeks after the first surgery, the same five animals were subjected to an additional surgery to assess the presence or absence of blood

538

Histological changes in vascular loops Paulos RG et al.

CLINICS 2017;72(9):538-542

was reached and connected at both ends. The distal end of the mandibular branch was sectioned, and the vessel lumen was irrigated with saline to prevent thrombus formation in the lumen. On the contralateral side, the left carotid artery was identified and dissected. A surgical drape was placed under the left carotid artery, and a double clamp was placed on this vessel. The proximal end of the venous graft (the mandibular branch of the right external jugular vein) was sectioned and fully freed. The direction of the graft was reversed, and the graft was placed on the surgical drape next to the left carotid artery. Two arteriotomies were conducted, and the vessel was flushed with saline. The venous graft was sutured to the arteriotomies using two end-to-side anastomoses with 10.0 nylon sutures, and six sutures were used for each anastomosis. Subsequently, an additional suture was used to connect the carotid artery to the two end-to-side anastomoses (Figure 1). The clamp was removed, and anastomosis patency was evaluated. The wound was closed using simple non-absorbable wire sutures (nylon 4-0).

flow through the vascular loop and collect samples for histological analysis. The rats were euthanized after this second surgical procedure. This cycle was performed five times to obtain five distinct animal groups, each containing five animals. The first, second, third, fourth, and fifth groups were sacrificed one, two, three, four, and five weeks after the anastomosis, respectively. The guidelines for euthanasia outlined by the Brazilian College of Animal Experimentation (Colégio Brasileiro de Experimentac¸ão Animal–COBEA) (1999) were followed. At the end of the surgical procedure and while the animals were still anesthetized, thiopental was administered intraperitoneally to induce deep anesthesia followed by euthanasia through intracardiac administration of potassium chloride. The samples collected for histological analysis included the left carotid artery containing the venous graft and the deep branch of the left jugular vein (the vein contralateral to that used as the graft served as a control for the evaluation of histological changes in the graft). The samples obtained from each animal were identified, transferred to a test tube containing formalin, and used to prepare the histological slides. The samples were fixed in buffered 10% formalin and subjected to histological analysis. The vascular segment was fixed without sectioning. Fivemicrometer histological sections were prepared and stained with hematoxylin and eosin using standard protocols. The histological analysis involved the semi-quantitative evaluation of the following variables: inflammatory infiltrates, vascular proliferation, fibrosis, intimal proliferation, and foreign body giant cell reactions. The semi-quantitative evaluation involved scoring the intensity of these variables using a scale ranging from 0 to 3, such that 0 indicated no intensity, 1 indicated mild intensity (1%–30%), 2 indicated moderate intensity (30%–60%), and 3 indicated severe intensity (greater than 60%). The presence of thrombi and pseudoaneurysms was also analyzed in the anastomosis area. The inflammatory infiltrates had a mixed pattern and primarily contained neutrophils, lymphocytes, and plasma cells. Quantification, which was classified as mild, moderate, or severe, was determined by measuring the percentage of inflammatory infiltrates present in the anastomosis area. The degree of intimal proliferation was assessed according to the degree of intimal hyperplasia and the degree of protrusion into the vascular lumen. Foreign body giant cell reactions, i.e., reactions around foreign bodies that involve histiocytes and giant cells and that from granulomas, was evaluated using the same criteria used for the inflammatory infiltrates. The quantification of fibrosis (hyalinized fibrosis and reactive fibroplasia) and vascular proliferation (proliferation of the capillaries in the wall of the animal) was assessed by calculating the percentage of fibrosis and proliferation in the wall of the anastomosis area.

Ehics This study was approved by the Research Ethics Committee (number 043/14), and all animals were maintained in an animal facility with water and food ad libitum.

’ RESULTS In the first procedure, after the release of the clamp, adequate blood flow was observed in all cases. In the group of rats sacrificed five weeks after the first surgery, blood flow was not observed through the graft in four animals (80%). A similar result was observed in the rats sacrificed four weeks after the initial surgery. In animals sacrificed three weeks after the initial surgery, blood flow

Surgical procedure Anesthesia was performed through intramuscular administration of 10 mg/kg xylazine and 50–75 mg/kg ketamine. After the animals were anesthetized, they were placed in the supine position, and then the skin was shaved and prepped with a local antiseptic and anesthetic. A transverse incision at the base of the neck (shoulder to shoulder) was used as the access route. The trifurcation of the right external jugular vein was identified, and then its deep branch (mandibular) was dissected until the bifurcation

Figure 1 - Final aspect of the surgical procedure. C – left carotid; E – venous graft; a – end-to-side anastomoses; small arrow – blood flow direction; large arrow – suture connecting the carotid artery to the end-to-side anastomoses.

539

Male Female Female Male Female Male Male

Male Male Male Male Female Female Female Female Male Male Male Male Male Male Male Female Female Male Male

Male

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Sex

1 2 3 4 5 6 7

Rat No.

540

330

415 435 381 440 295 280 293 272 367 355 380 340 330 405 370 300 295 355 347

350 304 248 397 238 372 400

Initial weight

4 3 2 1 5 4 3 Died 1 5 4 3 2 2 1 5 4 3 2

5 Died 3 2 1 5 4

Period until euthanasia

Table 1 - Animals and changes observed.

Present

Present Absent Absent Absent Present Present Present Absent Absent Present Present

1 2 1 2 1 1 0 1 1 1 Slide not evaluated 1

1 1 1 2 Slide not evaluated 1 1 0 1 1 2 1

Present Present Present Absent Absent Present Present Present Present Absent Absent Present

Inflammatory infiltrates

Present

Blood flow through the vascular loop during the second surgery

1 1 2 1 1 1 1 1 1 1

1 1 1 1 1 2 2

2 2 1 2

Granulomatous reaction to foreign bodies

0 1 1 0 0 1 1 2 2 1

1 1 1 0 2 1 1

1 1 1 2

Fibrosis

1 2 1 1 1 1 1 2 2 1

2 2 1 1 2 2 1

1 2 1 2

Vascular proliferation

0 2 2 2 1 1 1 2 3 1

2 1 1 0 2 2 2

1 1 1 2

Intimal proliferation

Absent

Absent Present Present Present Absent Absent Absent Present Present Present

Absent Absent Absent Absent Present Present Absent

Absent Absent Absent Present

Absent

Thrombus

Absent

Absent Absent Absent Absent Absent Absent Absent Absent Absent Absent

Absent Absent Absent Absent Absent Absent Absent

Absent Absent Absent Absent

Absent

Pseudoaneurysm

Histological changes in vascular loops Paulos RG et al. CLINICS 2017;72(9):538-542

Histological changes in vascular loops Paulos RG et al.

CLINICS 2017;72(9):538-542

Similar results were observed by Wong et al. (1), who conducted an experimental study where arteriovenous fistulas were created in rats and then collected for analysis at 7, 14, and 28 days postoperatively. In their study, significant changes were observed in the intima after seven days. Furthermore, significant hyperplasia was observed two weeks after surgery. Other changes observed included inflammatory infiltrates, granulomatous reactions to foreign bodies, fibrosis, and vascular proliferation. Among these changes, vascular proliferation also increased during the postoperative period. In the rats with 4 and 5 weeks of postoperative recovery, 4 of the 5 (80%) grafts were occluded in each group. However, inflammatory infiltrates and fibrosis were more evident in the 5-week group. This result suggests that graft occlusion may increase inflammation and fibrosis, but the opposite effect (i.e., that inflammation and fibrosis would contribute to graft failure) does not occur, as would be expected. In our study, of the seven female and 18 male rats evaluated, four females (57.15%) and five males (27.78%) experienced complications in the vascular loop. In this respect, previous studies have demonstrated that women are more likely to develop an occlusion of saphenous vein grafts (11,12). Several studies have evaluated the use of arteriovenous loops in free flap reconstructions (2,3,13-20). These loops can be used at a stage in which the loop and free flap reconstruction are conducted simultaneously or in two stages, where the arteriovenous loop is conducted first, and free flap reconstruction is performed during a second surgery.

was absent in only one animal (20%). In addition, blood flow was observed in the vascular loop in all animals in which the second surgery was conducted one to two weeks after the first surgery (Table 1). In rat No. 25, blood flow was observed in the venous graft during the second surgery. However, the histological analysis indicated the presence of a thrombus in the vascular loop. In the remaining rats for which a thrombus was visualized during the histological examination, blood flow through the graft was not observed during the second procedure (Table 2). In the group sacrificed five weeks after the first surgery, the increase in vascular proliferation and intimal proliferation was moderate to severe in all animals evaluated (Figure 2). This group also presented the most significant changes in inflammatory infiltrates and fibrosis. No pseudoaneurysms occurred in any of the cases evaluated, and granulomatous reactions to foreign bodies did not correlate with the duration of the postoperative period, i.e., they occurred at random. The analysis by sex indicated that of the seven females and 18 males with follow-up, four females (57.15%) and five males (27.78%) exhibited no blood flow in the venous graft during the second procedure.

’ DISCUSSION Intimal hyperplasia is observed in venous segments after arteriovenous anastomoses. It is one of the leading complications associated with stenosis during coronary surgery (5) and leads to decreased blood flow and possible thrombosis (6). Intimal hyperplasia is considered the main cause of failure of arteriovenous fistulas. Studies have been developed to elucidate the mechanism of venous graft wall thickening after the passage of arterial blood through the graft (a phenomenon known as arterialization). Numerous authors have designated the progressive changes that lead to obstruction as ‘‘venous graft disease.’’ These studies have primarily been conducted by cardiac surgery teams to prevent bypass graft loss (7-10). The results observed in our study corroborate the findings of previous studies. We observed an increase in intimal proliferation in seven of the eight histological analyses of rats with failure of the vascular loop (one rat without blood flow could not be evaluated histologically due to a problem during the preparation of the slide). We also noted that intimal proliferation tended to increase as the postoperative period increased. In two weeks, only a slight increase in the intima was observed, whereas intimal proliferation increased in all rats evaluated after five weeks.

Figure 2 - Slide showing strong intimal proliferation, which led to luminal stenosis (rat No. 24).

Table 2 - Histological evaluation of rats with occlusion of the vascular loop. Rat No.

6 12 13 17 18 19 23 24

Sex

Initial weight

Period until euthanasia

Blood flow through the vascular loop during the second surgery

Inflammatory infiltrates

Fibrosis

Vascular proliferation

Intimal proliferation

Male Female Female Male Male Male Female Female

372 295 280 355 380 340 300 295

5 5 4 5 4 3 5 4

Absent Absent Absent Absent Absent Absent Absent Absent

2 1 2 2 1 2 1 1

2 2 1 1 1 0 2 2

2 2 2 2 1 1 2 2

2 2 2 2 2 2 2 3

541

Histological changes in vascular loops Paulos RG et al.

CLINICS 2017;72(9):538-542

At present, a tendency exists for the use of a single surgery (3) because an increased rate of complications, including venous graft occlusions, has been observed when both procedures are performed separately (17,21). Our study demonstrated complete patency of the venous grafts in the first two weeks after surgery, and the histological analysis indicated that significant intimal proliferation started to become apparent three weeks after surgery. These results corroborate the preference for the simultaneous performance of vascular loop and free flap reconstruction. We observed that the failure rate of venous grafts related to intimal proliferation increased with the duration of the postoperative period. Therefore, performing definitive surgery without changes in the intima is safer. Overall, autonomization of flaps was observed two to three weeks after surgery, which is the period necessary for the release of pedicle flaps, such as cross-leg flaps (22). Therefore, in cases of venous graft occlusion, flap loss would not occur after this period in single-stage procedures (loop and free flap reconstruction). It was not our aim to determine the period of venous graft occlusion in medical practice because the time intervals observed in previous studies did not correspond to the intervals found in humans. However, our study suggests that the simultaneous execution of both procedures is safer because the vessel that receives the flap does not exhibit histological alterations that cause occlusions over time. We noted that intimal proliferation was the most significant change in venous grafts during the postoperative period and was directly correlated with graft occlusion. Therefore, our results suggest that in cases in which vascular loops are required for free flap reconstruction, both procedures should be performed during the same surgery.

10.

11.

12.

13. 14.

’ AUTHOR CONTRIBUTIONS 15.

Paulos RG designed the study and performed the surgeries. Rudelli BA analyzed the data and prepared the manuscript. Filippe RZ is the pathologist responsible for all the histopathological examinations. Santos GB ensured that the study followed the guidelines of the Brazilian College of Animal Experimentation (1999) for approval by the Research Ethics Committee. Herrera AA edited the manuscript. Ribeiro AA collected the data and created the tables. Rezende MR revised the manuscript. HsiangWei T supervised the surgical technique and prepared the manuscript. Mattar-Jr R prepared and revised the manuscript.

16.

17.

18.

’ REFERENCES 19.

1. Wong CY, de Vries MR, Wang Y, van der Vorst JR, Vahrmeijer AL, Zonneveld AJ, et al. Vascular remodeling and intimal hyperplasia in a novel murine model of arteriovenous fistula failure. J Vasc Surg. 2014; 59(1):192-201.e1, http://dx.doi.org/10.1016/j.jvs.2013.02.242. 2. Oswald TM, Stover SA, Gerzenstein J, Lei MP, Zhang F, Muskett A, et al. Immediate and delayed use of arteriovenous fistulae in microsurgical flap procedures: a clinical series and review of published cases. Ann Plast Surg. 2007;58(1): 61-3, http://dx.doi.org/10.1097/01.sap.0000250743. 78576.35. 3. Shipkov H, Traikova N, Voinov P, Boucher F, Braye F, Mojallal A. Vascularloops in reconstructive microsurgery: A review of the literature.

20. 21. 22.

542

Ann Chir Plast Esthet. 2014;59(1):1-8, http://dx.doi.org/10.1016/j.anplas. 2013.06.009. Sterpetti AV, Cucina A, Lepidi S, Randone B, Corvino V, D’Angelo LS, et al. Formation of myointimal hiperplasia and cytokine production in experimental vein graft. Surgery. 1998;123(4):461-9, http://dx.doi.org/ 10.1016/S0039-6060(98)70169-2. Schleimer K, Grommes J, Greiner A, Jalaie H, Kalder J, Langer S, et al. Training a sophisticated microsurgical technique: interposition of external jugular vein graft in the common carotid artery in rats. J Vis Exp. 2012; (69). pii:4124, http://dx.doi.org/10.1016/S0039-6060(98)70169-2. Norberto JJ, Sidawy AN, Trad KS, Jones BA, Neville RF, Najjar SF, et al. The protective effect of vein cuffed anastomoses is not mechanical in origin. J Vasc Surg. 1995;21(4):558-64, http://dx.doi.org/10.1016/S07415214(95)70187-7. Perek B, Malinska A, Stefaniak S, Ostalska-Nowicka D, Misterski M, Zabel M, et al. Predictive factors of late venous aortocoronary graft failure: ultrastructural studies. PLoS One. 2013;8(8):e70628, http://dx.doi. org/10.1371/journal.pone.0070628. Berard X, Déglise S, Alonso F, Saucy F, Meda P, Bordenave L, et al. Role of hemodynamic forces in the ex vivo arterialization of human saphenous veins. J Vasc Surg. 2013;57(5):1371-82, http://dx.doi.org/10.1016/j.jvs. 2012.09.041. Yayla Ç, Canpolat U, Akyel A, Yayla KG, Yilmaz S, Ac¸ikgöz SK, et al. Association Between Platelet to Lymphocyte Ratio and Saphenous Vein Graft Disease. Angiology. 2016;67(2):133-8, http://dx.doi.org/10.1177/ 0003319715578258. Dog˘an M, Akyel A, Cimen T, Öksüz F, Celik IE, Aytürk M, et al. Relationship between neutrophil-to-lymphocyte ratio and saphenous vein graft disease in patients with coronary bypass. Clin Appl Thromb Hemost. 2015;21(1):25-9, http://dx.doi.org/10.1177/1076029613488935. Une D, Kulik A, Voisine P, Le May M, Ruel M. Correlates of saphenous vein graft hyperplasia and occlusion 1 year after coronary artery bypass grafting: analysis from the CASCADE randomized trial. Circulation. 2013;128(11 Suppl 1):S213-8, http://dx.doi.org/10.1016/S0039-6060(98) 70169-2. Loop FD, Golding LR, MacMillan JP, Cosgrove DM, Lytle BW, Sheldon WC. Coronary artery surgery in women compared with men: analyses of risks and long-term results. J Am Coll Cardiol. 1983;1(2 Pt 1):383-90, http://dx.doi.org/10.1016/S0039-6060(98)70169-2. Zhang F, Oliva A, Kao SD, Newlin L, Buncke HJ. Microvascular vein grafts in the rat cutaneous free-flap model. J Reconstr Microsurg. 1994; 10(4):229-33, http://dx.doi.org/10.1055/s-2007-1006589. Reichenberger MA, Harenberg PS, Pelzer M, Gazyakan E, Ryssel H, Germann G, et al. Arteriovenous loops in microsurgical free tissue transfer in reconstruction of central sternal defects. J Thorac Cardiovasc Surg. 2010;140(6):1283-7, http://dx.doi.org/10.1016/j.jtcvs.2010.05.019. Brüner S, Bickert B, Sauerbier M, Germann G. Concept of arteriovenous loupes in high-risk free-tissue transfer: history and clinical experiences. Microsurgery. 2004;24(2):104-13, http://dx.doi.org/10.1002/micr.20006. Engel H, Pelzer M, Sauerbier M, Germann G, Heitmann C. An innovative treatment concept for free flap reconstruction of complex central chest wall defects--the cephalic-thoraco-acromial (CTA) loop. Microsurgery. 2007;27(5): 481-6, http://dx.doi.org/10.1002/micr.20391. Lin CH, Mardini S, Lin YT, Yeh JT, Wei FC, Chen HC. Sixty-five clinical cases of free tissue transfer using long arteriovenous fistulas or vein grafts. J Trauma. 2004;56(5):1107-17, http://dx.doi.org/10.1097/01.TA. 0000114637.29779.AB. Angel MF, Chang B, Clark N, Wong L, Ringelman P, Manson PN. Further clinical use of the interposition arteriovenous loop graft in free tissue transfers. Microsurgery. 1993;14(8):479-81, http://dx.doi.org/10.1002/micr. 1920140803. Lind B, McCarthy W, Derman G, Jacobs C. Arteriovenous loop grafts for free tissue transfer. Vasc Endovascular Surg. 2012;46(1):30-3, http:// dx.doi.org/10.1177/1538574411418843. Freedman AM, Meland NB. Arteriovenous shunts in free vascularized tissue transfer for extremity reconstruction. Ann Plast Surg. 1989;23(2): 123-8, http://dx.doi.org/10.1097/00000637-198908000-00005. Cavadas PC. Arteriovenous vascular loops in free flap reconstruction of the extremities. Plast Reconstr Surg. 2008;121(2):514-20, http://dx.doi. org/10.1097/01.prs.0000297634.53915.e5. Georgescu AV, Irina C, Ileana M. Cross-leg tibial posterior perforator flap. Microsurgery. 2007;27(5):379-83, http://dx.doi.org/10.1002/micr.20375.

CLINICAL SCIENCE

Long-term efficiency of cataract surgery with hydrophilic acrylic Ioflex intraocular lens Priscilla A. Jorge,I,* Camila Ribeiro Koch,I Delano Jorge,II Newton Kara-JuniorI I Departamento de Oftalmologia, Faculdade de Medicina FMUSP, Universidade de Sao Paulo, Sao Paulo, SP, BR. Universidade Federal de Sao Paulo, Sao Paulo, SP, BR.

Departamento de Oftalmologia,

OBJECTIVE: To evaluate the efficiency of long-term cataract surgery using low-cost intraocular lens implantation in community campaigns. METHODS: Fifty-eight randomly selected patients were evaluated four years after phacoemulsification and Ioflex intraocular lens implantation. Causes of low visual acuity related to the intraocular lens were evaluated, and treatment costs were calculated. RESULTS: The mean age of patients was 72±10.2 years. Four years after surgery, 25 eyes (43.0%) had decreased visual acuity related to the intraocular lens: posterior capsule opacification was noted in 24 eyes (41.3%), and intraocular lens opacification was noted in one eye (1.7%). The total cost of the post-surgical complication treatments represented 6.3% of the initial budget of the entire surgical patient group. CONCLUSIONS: The efficiency of cataract surgery with low-cost Ioflex intraocular lens implantation was significantly reduced in a long-term follow-up study because postoperative complications related to intraocular lenses emerged at higher rates than when the gold-standard treatment was used. KEYWORDS: Cataract; Lenses; Intraocular; Economics; Postoperative Complications; Cost and Cost Analysis. Jorge PA, Koch CR, Jorge D, Kara-Junior N. Long-term efficiency of cataract surgery with hydrophilic acrylic Ioflex intraocular lens. Clinics. 2017;72(9): 543-546 Received for publication on January 12, 2017; First review completed on February 20, 2017; Accepted for publication on May 16, 2017 *Corresponding author. E-mail: prialmeida.j@gmail.com

’ INTRODUCTION

actions on clinical evidence of its efficiency (6). Thus, for the first time, this study evaluates the long-term efficiency of cataract surgery using the Ioflex IOL.

Phacoemulsification surgery is the most widely used technique for cataract surgery due to its good visual outcomes (1). However, long-term postoperative complications due to different intraocular lens (IOL) materials and designs can affect budgets and impair the efficiency of implementing this procedure in public healthcare systems (2,3). Given the large number of cataract surgeries being conducted, even small differences in costs per surgery add up to larger amounts in total cost to the healthcare system (4). Ioflex (Mediphacos, Belo Horizonte, MG, Brazil) is a lowcost hydrophilic acrylic IOL commonly used in community programs for cataract surgery for underprivileged people (2). This IOL has been marketed for more than 15 years and has been approved by the national health agencies in several countries in Europe, Asia, and Latin America. An efficient IOL should produce good results with low cost. Reports indicate post-surgical complications related to the Ioflex IOL (2,5) and suggest that health managers must base their

’ MATERIALS AND METHODS A sample of 102 subjects was randomly selected from a population of 192 patients who underwent cataract surgery and Ioflex IOL implantation within the past four years without intraoperative complications as part of a community campaign for underprivileged people in Pernambuco, Brazil. The follow-up time ranged from 42 to 48 months postoperatively. The mean follow-up time was 45.6 months. A total of 31 patients could not be located due to either communication difficulties related to geographical distance or patient death. Of the 71 patients located, 13 (18.3%) were not present for examination. Thus, 58 patients were available for examination. To avoid bias, only the right eye was included in the study for patients who underwent cataract surgery in both eyes. The same surgeon performed all surgical procedures. In each case, a 2.8-mm incision was made under peribulbar anesthesia using a balanced salt solution (BSSs, Alcon Labs, Fort Worth, TX, USA) and 2% methylcellulose viscoelastic (Ophthalmus, São Paulo, Brazil). Continuous curvilinear capsulorhexis approximately 1 mm smaller than the optical part of the IOL and cortical clean-up procedures were used in all cases. Data were collected from the medical records and

543

Long-term efficiency of cataract surgery Jorge PA et al.

CLINICS 2017;72(9):543-546

Table 1 - Surgical and post-surgical procedures expenses refund based on Brazilian Health System guidelines on the 58 eyes evaluated.

included information about the patient’s ocular health, presurgical corrected visual acuity (VA), and postoperative corrected VA after one month. Examination of the corrected VA, a biomicroscopy slit lamp examination, and fundoscopy were performed at the evaluation four years after the surgery. Posterior capsule opacification (PCO) was diagnosed during the slit lamp examination, and the decision regarding laser capsulotomy was based on VA deterioration. Patients with worsened VA (one line or more on Snellen card notation) compared to that measured one month after surgery, without improvement with glasses, that was related to PCO were referred for neodymium:yttrium-aluminum-garnet (Nd:YAG) laser treatment. Surgical and post-surgical procedure costs were based on Brazilian Health System guidelines. All values were converted to US dollars (USD) at a rate of one dollar to R$ 3.1 (Brazilian currency). The postoperative complications of the IOL were compared with those of a gold-standard hydrophobic IOL (AcrySof SN60 one-piece, Alcon, Fort Worth, TX, USA) based on a study by Cullin et al. (7), who evaluated PCO rates and related costs for three different IOLs in a retrospective study of 1,527 patients with a mean follow-up time of 41.5 months, ranging from 19 to 76 months. The mean follow-up time of patients with the hydrophobic AcrySof IOL was 41.4 months. The study included Nd:YAG capsulotomy patients with development of postoperative PCO and clinical signs of secondary visual impairment, such as reduced VA or glare. Both our study and that by Cullin et al. (7) detected PCO at an outpatient visit. The criterion that we used for PCO was visual loss after primary cataract surgery as subjective data to suggest the necessity of Nd:YAG laser capsulotomy. In addition, none of the studies described the use of Trypan blue during the surgeries.

Cataract surgery Nd:YAG laser IOL exchange

Value/unit USD $

N of Procedures

Total cost USD $

142.90 14.51 175.76

58 24 1

8,288.20 348.24 175.76

in a final corrected VA recovery to 0 logMAR (20/20 in Snellen card notation). Two additional patients with PCO were diagnosed with glaucoma and posterior segment disease. The cost of each Ioflex IOL was USD $50.00, and the cost of each gold-standard hydrophobic IOL was USD $99.59. The costs of the cataract surgery, Nd:YAG laser capsulotomy procedures, and IOL exchange are listed in Table 1. The additional cost to maintain a good visual outcome 4 years after surgery was USD $9.03 per patient with Ioflex IOL implantation, whereas the estimated cost of a gold-standard IOL implant would have been USD $1.07 per patient. Thus, the low-cost IOL is USD $7.96 greater than the gold standard IOL in the initial surgery budget (5.5%). The cost of the postsurgical procedures using the Ioflex IOL represents 6.3% (USD $524.00) of the initial budget of the entire surgical patient group.

’ DISCUSSION The results revealed high rates of late postoperative complications related to the IOL, with significant PCO in 41.3% of patients and IOL opacification in 1.7% of patients who underwent Ioflex hydrophilic acrylic IOL implantation. For the gold-standard hydrophobic acrylic IOL, 7.4% of patients had PCO, and no IOL opacification was reported (7). Hydrophilic acrylic lenses are associated with higher rates of PCO than hydrophobic IOLs. Few papers report PCO evaluation more than four years after surgery. Ronbeck et al. (8) reported that hydrophobic acrylic IOLs delayed PCO development, with a median time of survival without Nd:YAG capsulotomy of 108 months. PCO rates were reported for approximately 50% of the patients approximately 10 years after surgery. We found similar rates approximately 4 years after surgery. Chang et al. (9) reported a rate of 10% for Nd:YAG laser capsulotomy for a one-piece hydrophobic IOL and 22% for a 3-piece hydrophobic IOL 5 to 7 years after surgery. These results demonstrate that hydrophobic acrylic IOLs may delay the need for laser capsulotomy for several years in eyes submitted to cataract surgery and exhibit lower occurrence rates than hydrophilic acrylic PCO rates. Given that the life expectancy of people worldwide is increasing, it is important to delay the occurrence of postoperative complications as much as possible. Vasavada et al. (10) reported Nd:YAG laser capsulotomy cases in 12.9% and 16% of the hydrophilic IOLs evaluated and none in the hydrophobic group 3 years after surgery. Kugelberg et al. (11) reported capsulotomy in 42% of a hydrophilic IOL group compared to 10% in a hydrophobic group at 2 years after surgery. Gauthier et al. (12) reported an increasing number of Nd:YAG laser treatments over time, with 8.8% in the hydrophobic IOL group and 37.2% in the hydrophilic lens group with a follow-up of 24 months. The results from these studies are consistent with our results.

Ethics This study was approved by the Ethics Committee of the Altino Ventura Foundation (054/2011) in accordance with the Helsinki Declaration.

’ RESULTS The mean age of the patients who received an Ioflex IOL was 72.0±10.2 years. The mean preoperative corrected VA was 0.79±0.41 logMAR (20/125 in Snellen card notation). One month after surgery, the mean corrected VA improved to 0.25±0.20 logMAR (20/35 in Snellen card notation), except for one patient who experienced a posterior segment problem. Four years after surgery, 39 (67.0%) of the 58 evaluated eyes had detectable PCO upon biomicroscopy evaluation, and 24 eyes (41.3%) had decreased VA due to PCO. Patients with slight PCO underwent a new refractive examination for updated glasses and exhibited a mean corrected VA of 0.13±0.03 logMAR (20/26 in Snellen card notation), whereas the mean corrected VA of the 24 patients with decreased VA due to PCO was 0.53±0.21 logMAR (20/67 in Snellen card notation). These patients were referred for Nd:YAG laser capsulotomy and consequently demonstrated a corrected VA that had recovered to 0.05±0.06 logMAR (20/22 in Snellen card notation). One patient (1.7%) exhibited decreased VA from 0.09 logMAR at one month after surgery to 0.6 logMAR 4 years after surgery (20/25 to 20/80 in Snellen card notation). This result was due to IOL opacification, and the patient was referred for IOL exchange, resulting

544

Long-term efficiency of cataract surgery Jorge PA et al.

CLINICS 2017;72(9):543-546

post-surgical complications with poor visual outcomes are related to low-cost IOLs (2,25). This study revealed that use of the Ioflex acrylic hydrophilic IOL may cause the public health system to spend almost 6.3% more than the initial budget on post-surgical treatments. Therefore, the results of this paper should make managers aware that they should schedule an outpatient visit at least 4 years after surgery if the low-cost hydrophilic IOL is used and reserve a budget for Nd:YAG laser capsulotomy. In contrast, for hydrophobic IOLs, the visit could be scheduled 10 years after surgery. The current study had some important limitations. The comparison was performed with another study, specifically a retrospective study with a different population and different surgeons. The follow-up time of PCO evaluation was similar but was not identical. None of the studies determined a criterion for software for image analysis to evaluate the severity of PCO. However, we believe that despite this limitation, the results were interesting and worth reporting. In conclusion, the material and design of the Ioflex IOL afford it short-term effectiveness, but reduced efficiency for long-term cataract surgery is noted upon follow-up. After surgery, an increased number of patients underwent laser capsulotomy in a shorter period of time than with the highcost IOLs. We have highlighted the social impact of a costeffective IOL, particularly for public healthcare policies in developing countries.

This information could provide a basis for discussions about the long-term economic consequences of choosing an IOL with a design that affects the risk of PCO development. It is not yet known whether the occurrence of PCO in hydrophobic IOLs is only 50% or continues to increase after 12 years post-surgery (8). Similarly, it is not known whether hydrophilic IOLs will continue to exhibit greater PCO rates than hydrophobic IOLs. More studies are needed to better explain PCO evolution. The results of our study allows us to estimate the future trends to at least better prepare managers and public health programs. We have previously reported the occurrence of Ioflex IOL opacification, which exhibited and incidence rate of 7% (5). The current finding of a 1.7% IOL opacification rate was due to the selection criterion, which included only one eye of each patient. This criterion was not used in previous research. Although opacification is not a common complication, it is the main cause of hydrophilic acrylic IOL extraction. The causes of this complication are not completely understood, but the literature indicates that it is caused by calcification due to calcium and/or phosphate in the IOL (13). Due to technological advances, the IOL design has been modified from a round edge to include a square edge, which has resulted in a reduction in the incidence of PCO (12). However, studies suggest that square-edged IOLs should completely encompass the 360 degrees around the IOL optic to provide an effective barrier (14). Furthermore, another study demonstrated differences between brands of squareedged lenses, suggesting that variations in their edge profiles may account for clinical differences in postoperative PCO rates (15). Acrylic IOLs also appear to lose their PCO preventive effect despite their sharp optic edges (8,16). Our study suggests that the initial savings of USD $49.59 per patient with the use of the hydrophilic IOL compared to the hydrophobic IOL is reduced when considering the additional cost of treatments for avoidable postoperative complications. We estimate that the Ioflex IOL loses its efficiency, as it costs USD $7.96 more than the gold standard IOL in the initial surgical budget (5.5%). However, even with the additional cost of postoperative treatments, the total cost was still lower when using the low-cost IOL. Community cataract programs are performed worldwide in developing countries (2,17). Reviews of blindness in regions such as Latin America, Nigeria, Cameroon, and India reported a need to improve the implementation of social action programs to prevent blindness and improve cataract surgery outcomes (18-23). In a long-term postoperative follow-up study, De Senne et al. (24) demonstrated that 75.6% (Project A) and 64.7% (Project B) of patients experienced low VA due to postoperative complications in a community cataract program in Brazil (mainly PCO and refractive errors). This finding indicates that long-term postsurgical complications can be as harmful as failing to have the cataract extracted. Data from this study highlight the need for community cataract programs to provide easy access to specialized services over a long follow-up period to ensure the initial efficiency of the cataract treatment. This notion is particularly important when hydrophilic acrylic IOLs are used, especially the Ioflex IOL, which is widely used in community campaigns in Brazil given its low cost. Long-term postoperative complications pose a risk for the efficiency of cataract treatment. This risk could be partially avoided by using better IOLs with proven clinical results given that the majority of

’ AUTHOR CONTRIBUTIONS Jorge PA provided substantial contributions to the conception and design of the study, acquisition, analysis and interpretation of data, drafting and critical revision of the manuscript for important intellectual content and approval of final version of the manuscript. Koch CR provided substantial contributions to the conception and design of the study, drafting of the manuscript and approval of the final version of the manuscript. Jorge D provided substantial contributions to the acquisition of data, critical revision of the manuscript for important intellectual content and approval of the final version of the manuscript. Kara-Junior N provided substantial contributions to the conception and design of the study, analysis and interpretation of data, drafting and critical revision of the manuscript for important intellectual content and approval of the final version of the manuscript.

’ REFERENCES 1. Behndig A, Montan P, Stenevi U, Kugelberg M, Lundstrom M. One million cataract surgeries: Swedish National Cataract Register 1992–2009. J Cataract Refract Surg. 2011;37(8):1539-45, http://dx.doi.org/10.1016/ j.jcrs.2011.05.021. 2. Jorge Pde A, Jorge D, Ventura CV, Ventura BV, Lira W, Ventura CM, et al. Incidence of posterior capsule opacification following the implantation of a foldable hydrophilic acrylic intraocular lens: a 4 year follow-up study. Arq Bras Oftalmol. 2014;77(4):222-4, http://dx.doi.org/10.5935/00042749.20140057. 3. Kara-Jr N, Sirtoli MG, Santhiago MR, Parede TR, Espindola RF, Carvalho Rde S. Phacoemulsification versus extracapsular extraction: governmental costs. Clinics. 2010;65(4):357-61, http://dx.doi.org/10.1590/S1807-593220 10000400002. 4. Hiratsuka Y, Yamada M, Murakami A, Okada AA, Yamashita H, Ohashi Y, et al. Cost-effectiveness of cataract surgery in Japan. Jpn J Ophthalmol. 2011;55(4):333-42, http://dx.doi.org/10.1007/s10384-011-0041-3. 5. Jorge Pde A, Jorge D, Ventura CV, Ventura BV, Lira W, Ventura MC, et al. Late opacification in hydrophilic acrylic intraocular lenses: analysis of 87 eyes in a random sample of 102 patients. J Cataract Refract Surg. 2013;39(3):403-7, http://dx.doi.org/10.1016/j.jcrs.2012.09.023. 6. Kara-Junior N, José NK. The occurrences of consecutive infections after cataract surgeries: random events or a product of mistaken public politics? Clinics. 2016;71(6):295-6, http://dx.doi.org/10.6061/clinics/2016(06)01. 7. Cullin F, Busch T, Lundstrom M. Economic considerations related to choice of intraocular lens (IOL) and posterior capsule opacification frequency – a comparison of three different IOLs. Acta Ophthalmol. 2014; 92(2):179-83, http://dx.doi.org/10.1111/aos.12026.

545

Long-term efficiency of cataract surgery Jorge PA et al.

CLINICS 2017;72(9):543-546

17. Vock L, Menapace R, Stifter E, Georgopoulos M, Sacu S, Buehl W. Posterior capsule opacification and neodymium:YAG laser capsulotomy rates with a round-edged silicone and a sharp-edged hydrophobic acrylic intraocular lens 10 years after surgery. J Cataract Refract Surg. 2009; 35(3):459-65, http://dx.doi.org/10.1016/j.jcrs.2008.11.044. 18. Lindfield R, Vishwanath K, Ngounou F, Khanna RC. The challenges in improving outcome of cataract surgery in low and middle income countries. Indian J Ophthalmol. 2012;60(5):464-9, http://dx.doi.org/10.4103/ 0301-4738.100552. 19. Limburg H, Barria von-Bischhoffshausen F, Gomez P, Silva JC, Foster A. Review of recent surveys on blindness and visual impairment in Latin America. Br J Ophthalmol. 2008;92(3):315-9, http://dx.doi.org/ 10.1136/bjo.2007.125906. 20. Arieta CE, de Oliveira DF, Lupinacci AP, Novaes P, Paccola M, Jose NK, et al. Cataract remains an important cause of blindness in Campinas, Brazil. Ophthalmic Epidemiol. 2009;16(1):58-63, http://dx.doi.org/10.1080/ 09286580802575032. 21. Rabiu MM, Kyari F, Ezelum C, Elhassan E, Sanda S, Murthy GV, et al. Review of the publications of the Nigeria national blindness survey: methodology, prevalence, causes of blindness and visual impairment and outcome of cataract surgery. Ann Afr Med. 2012;11(3):125-30, http://dx. doi.org/10.4103/1596-3519.96859. 22. Oye JE, Kuper H. Prevalence and causes of blindness and visual impairment in Limbe urban area, South West Province, Cameroon. Br J Ophthalmol. 2007;91(11):1435-9, http://dx.doi.org/10.1136/bjo.2007.115840. 23. Nirmalan PK, Thulasiraj RD, Maneksha V, Rahmathullah R, Ramakrishnan R, Padmavathi A, et al. A population based eye survey of older adults in Tirunelveli district of south India: blindness, cataract surgery, and visual outcomes. Br J Ophthalmol. 2002;86(5):505-12, http://dx.doi.org/ 10.1136/bjo.86.5.505. 24. De Senne FM, Cardillo JA, Rocha EM, Kara-José N. Long-term visual outcomes in the Cataract-Free Zone Project in Brazil. Acta Ophthalmol Scand. 2002;80(3):262-6, http://dx.doi.org/10.1034/j.1600-0420.2002.800306.x. 25. Johansson B. Clinical consequences of acrylic intraocular lens material and design: Nd:YAG-laser capsulotomy rates in 3 x 300 eyes 5 years after phacoemulsification. Br J Ophthalmol. 2010;94(4):450-5, http://dx.doi. org/10.1136/bjo.2009.166181.

8. Rønbeck M, Kugelberg M. Posterior capsule opacification with 3 intraocular lenses: 12-year prospective study. J Cataract Refract Surg. 2014; 40(1):70-6, http://dx.doi.org/10.1016/j.jcrs.2013.07.039. 9. Chang A, Behndig A, Rønbeck M, Kugelberg M. Comparison of posterior capsule opacification and glistenings with 2 hydrophobic acrylic intraocular lenses: 5- to 7-year follow-up. J Cataract Refract Surg. 2013;39(5): 694-8, http://dx.doi.org/10.1016/j.jcrs.2012.11.032. 10. Vasavada AR, Raj SM, Shah A, Shah G, Vasavada V, Vasavada V. Comparison of posterior capsule opacification with hydrophobic acrylic and hydrophilic acrylic intraocular lenses. J Cataract Refract Surg. 2011; 37(6):1050-9, http://dx.doi.org/10.1016/j.jcrs.2010.12.060. 11. Kugelberg M, Wejde G, Jayaram H, Zetterström C. Two-year follow-up of posterior capsule opacification after implantation of a hydrophilic or hydrophobic acrylic intraocular lens. Acta Ophthalmol. 2008;86(5):533-6, http://dx.doi.org/10.1111/j.1600-0420.2007.01094.x. 12. Gauthier L, Lafuma A, Laurendeau C, Berdeaux G. Neodymium:YAG laser rates after bilateral implantation of hydrophobic or hydrophilic multifocal intraocular lenses: Twenty-four month retrospective comparative study. J Cataract Refract Surg. 2010;36(7):1195-200, http://dx.doi.org/10.1016/ j.jcrs.2010.01.027. 13. Ventura BV, Ventura M, Werner L, Santhiago MR. Calcium deposits on hydrophilic acrylic intraocular lenses. J Cataract Refract Surg. 2013; 39(1):142-3, http://dx.doi.org/10.1016/j.jcrs.2012.10.025. 14. Maddula S, Werner L, Ness PJ, Davis D, Zaugg B, Stringham J, et al. Pathology of 157 human cadaver eyes with round-edged or modern square-edged silicone intraocular lenses: analyses of capsule bag opacification. J Cataract Refract Surg. 2011;37(4):740-8, http://dx.doi.org/ 10.1016/j.jcrs.2010.10.058. 15. Werner L, Mamalis N, Pandey SK, Izak AM, Nilson CD, Davis BL, et al. Posterior capsule opacification in rabbit eyes implanted with hydrophilic acrylic intraocular lenses with enhanced square edge. J Cataract Refract Surg. 2004;30(11):2403-9, http://dx.doi.org/10.1016/j.jcrs.2004. 02.085. 16. Werner L, Tetz M, Feldmann I, Bucker M. Evaluating and defining the sharpness of intraocular lenses: microedge structure of commercially available square-edged hydrophilic intraocular lenses. J Cataract Refract Surg. 2009;35(3):556-66, http://dx.doi.org/10.1016/j.jcrs.2008.11.042.

546

CLINICAL SCIENCE

Early menarche and teenager pregnancy as risk factors for morbid obesity among reproductive-age women: A case-control study Amanda Gonc¸alves Neves,I Karina Tamy Kasawara,I Ana Carolina Godoy-Miranda,I Fla´vio Hideki Oshika,I Elinton Adami Chaim,II Fernanda Garanhani SuritaI,* I Departamento de Tocoginecologia, Faculdade de Cieˆncias Me´dicas, Universidade Estadual de Campinas – UNICAMP, Campinas, SP, BR. II Departamento de Cirurgia, Faculdade de Medicina, Universidade de Campinas, Campinas, SP, BR.

OBJECTIVES: The aim of this study was to evaluate potential risk factors, including non-communicable diseases, for morbid obesity in women between 20 and 49 years of age. METHODS: We performed a case-control study with 110 morbidly obese women and 110 women with adequate weight who were matched by age and with a 1:1 case to control ratio. All women were between 20 to 49 years old and non-menopausal. Possible risk factors were evaluated through a self-report questionnaire assessing socio-demographic, obstetric and gynecological characteristics, presence of non-communicable diseases and habits. Multiple logistic regression was used to estimate the odds ratio with respective confidence intervals. RESULTS: Menarche under 12 years old, teenage pregnancy and lower educational level were shown to be risk factors for morbid obesity among women of reproductive age. Incidences of non-communicable diseases (diabetes, hypertension, dyslipidemia, liver disease, lung disease, thyroid dysfunction, and joint pain) were increased in women with morbid obesity. CONCLUSIONS: Early menarche, teenage pregnancy and low education level are risk factors for the occurrence of morbid obesity in women of reproductive age. Some non-communicable diseases were already more prevalent in women with morbid obesity even before 50 years of age. KEYWORDS: Women’s Health; Morbid Obesity; Reproductive Factors; Non-communicable Diseases; Pregnancy in Adolescence. Neves AG, Kasawara KT, Godoy-Miranda AC, Oshika FH, Chaim EA, Surita FG. Early menarche and teenager pregnancy as risk factors for morbid obesity among reproductive-age women: A case-control study. Clinics. 2017;72(9):547-553 Received for publication on February 1, 2017; First review completed on May 7, 2017; Accepted for publication on June 30, 2017 *Corresponding author. E-mail: surita@unicamp.br

’ INTRODUCTION

Health Organization considers overweight and obesity the fifth main cause of death worldwide and indicates that at least 2.8 million adults die per year due to these conditions (3). Body mass index (BMI) is one of the simplest and most commonly used methods to determine obesity in adults. BMI variations from 30 to 34.9 kg/m2 and 35.0 to 39.9 kg/m2 define obesity degrees I and II, respectively. Morbid obesity or Grade III is defined by a BMI equal to or higher than 40 kg/m2. Due to the increase in the maximum degree of obesity, morbid obesity was recently and additionally categorized into super obesity (BMI 50-59 kg/m2) and supersuper obesity (BMIX60 kg/m2) (4). Although it is not the only parameter used to diagnose obesity, BMI is considered a predictor of mortality, showing a direct relationship. BMI ranging from 30 to 35 kg/m2, for example, reduces an individual’s survival by 2 to 4 years, while a BMI from 40 to 45 kg/m2 reduces survival by 8 to 10 years (3). Currently, 8% of women of childbearing age are morbidly obese (4). With regard to women’s reproductive health, obese women are more likely to undergo infertility treatment.

The increase in overweight and obesity rates has reached global epidemic proportions, leading to a major public health problem. The number of overweight and obese individuals rose from 857 million in 1980 to 2.1 billion in 2013, and the proportion rose from 28.8% to 36.9% in men and from 29.8% to 38% in women (1,2). Obesity is a chronic and multifactorial disease in which there is an imbalance between calories consumed and spent, which increases the risk of chronic non-communicable diseases, such as cardiovascular and endocrine disorders, some types of cancers and musculoskeletal disorders. The World

547

Risk factors of morbid obesity Neves AG et al.

CLINICS 2017;72(9):547-553

The data collected were transferred to Excels and later described by means, standard deviation (SD), median (M), and frequencies. In this study, early menarche was considered menarche at an age under 12 years old (9). Alcohol consumption was considered any alcohol use reported by the participant, and physical activity was considered any body movement produced by the skeletal muscles that resulted in increased energy expenditure compared to the rest position and was performed on a regular basis (five times per week). The contraceptive method used currently at the time of the interview and in the past, non-communicable diseases and occurrence of urinary continence were collected according to the women’s report. To assess the association between categorical variables and the outcome, we used the McNemar test (two classes) and Bowker symmetry test (three or more classes). To assess the quantitative variables according to the outcome groups, we performed Wilcoxon signed rank tests. Multivariate analysis was performed using multiple conditional logistic regression univariate and multivariate analyses to estimate odds ratio (OR) adjusted with confidence intervals (CI) and stepwise variable selection to evaluate the factors associated with morbid obesity. All analyses were performed by matching the case and control participants’ age. The significance level was 5%, and the software used for analysis was SAS version 9.4 for Windows.

Regarding perinatal outcomes, previous research has identified an association, especially in more obese women, between obesity and the inability to initiate and maintain breastfeeding. In addition, there may be an association between the prevalence of obesity in the female population of childbearing age and obstetric variables such as parity (5). Excessive gestational weight gain and postpartum weight retention also represent risks for obesity in women (6,7). Morbid obesity in women of reproductive age is the result of a cluster of interrelated factors. In addition, reproductive factors exert some influence on this condition. These relationships have not been sufficiently described in the literature. The aim of this study was to evaluate factors that could be a risk factor for morbid obesity in women aged 20 to 49, including the occurrence of non-communicable diseases associated with this condition.

’ MATERIAL AND METHODS This study was approved by the Institutional Review Board of the University of Campinas, Brazil (CAAE report: 32924114.9.0000.5404). All items of the Strengthening the Reporting of Observational Studies in Epidemiology consensus were followed (8). A case-control study was performed from November 2014 to September 2015 in a tertiary referral center at the University of Campinas, Brazil. We defined ‘‘cases’’ as women with morbid obesity according to the World Health Organization classification (2). Cases were selected from individuals seeking bariatric surgery in a gastrosurgery outpatient clinic. Participants considered ‘‘controls’’ were women with adequate BMI who were selected at the family planning outpatient clinic. Individuals in both groups were between 20 and 49 years old and not menopausal. All participants were matched by age. For each individual case, a control with no more than two years difference in age was included (1:1 ratio). The sample size was calculated by using the differences in obesity prevalence rates for different ages of menarche and the number of pregnancies as a reference (9). A previous study in Brazil showed an obesity prevalence of 28.6% in women with menarche between 8 and 11 years old and 13.3% in women with menarche at older than 14 years old. According to the number of pregnancies, there is a prevalence of 7.2% obesity in nulliparous women and 37.9% in women with five or more children (9). Considering these differences, a 5% significance level and a ratio of 1 case to 1 control, the total estimated sample size was 220 subjects for the association with age at menarche and 66 women for the association with the number of pregnancies. At the highest value (n=220), 110 women would be necessary in each group. Women were selected through medical records, before a medical visit, to determine the BMI of the potential participants. The cases and controls that met the inclusion criteria were invited to participate and sign the informed consent form before being enrolled. Women with communication difficulties and any other condition that might lead to misunderstanding the questions were also excluded from the study. Then, a 40-minute interview was conducted in which the women answered objective questions in a data collection form, developed specifically for this study, on socio-demographics, reproductive and obstetric history, habits, and comorbidities. The confidentiality of data were preserved and the patients were identified only with a number.

’ RESULTS A total of 357 women were invited to participate in this study; however, 137 participants were excluded for different reasons: refused to participate, were younger than 20 years old or older than 49 years old and were excluded for not having a morbid obese BMI (cases) or adequate BMI (controls). Therefore, 220 women were included, with 110 morbid obesity cases and 110 non-obese controls with adequate BMI. The socio-demographic and anthropometric characteristics and habits of both groups are described in Table 1. Educational level (elementary school; OR 3.82 95% CI 1.52-9.59) and the mean household income of women with morbid obesity were lower compared to the control group. Physical activity was more prevalent (po0.001) among women with morbid obesity. Alcohol consumption was also more prevalent among women with morbid obesity. Regarding reproductive variables, early menarche (under 12 years old; OR 2.06, 95% CI 1.14-3.75) teenage pregnancy (15-19 years old; OR 2.36, 95% CI 1.23-4.56) and nulliparity (OR 2.67, 95% CI 1.09-6.52) were associated with the occurrence of morbid obesity (Table 2). The mean BMI was higher among women with morbid obesity, regardless of the number of pregnancies. However, as the number of cases decreased with increasing gestations, the power of the sample became limited to evaluate this association after the 2nd pregnancy (Table 2). Furthermore, women with morbid obesity had a higher prevalence of some non-communicable diseases, such as diabetes, hypertension, dyslipidemia, liver disease, pulmonary disease, thyroid dysfunction, and joint pain, compared with controls (Table 3). The results of multivariate analysis with stepwise selection criteria of variables showed that education, age at menarche and age at 1st pregnancy were significantly associated with morbid obesity. Women at higher risk of morbid obesity were those with less education (2.6-fold increased risk for

548

Risk factors of morbid obesity Neves AG et al.

CLINICS 2017;72(9):547-553

Table 1 - Socio-demographic and anthropometric characteristics of women with morbid obesity and adequate BMI. Characteristics

BMI average (SD)* Mean Weight (SD)*, kg Mean Household Income* (SD) (n=141)c

Educational level** Elementary High school College or University Employed** (n=214)b Yes No Physical activity practice*** Yes No Weekly frequency of physical activity (days/week)** (n=216)d Not practicing p4 X5 Daily duration of physical activity (min/day)** (n=208)e Not practicing o 60 X 60 Smoking** Yes No Alcohol consumption** Yes No Illegal drugs** Yes No

Morbid obesity (n=110)

Control group (n=110)

44.96 (5.33) 116.62 (15.86) 609.81 (327.54)

23.33 (1.76) 62.03 (6.49) 909.14 (647.24)

22 59 29

20.00 53.64 26.36

12 45 53

10.91 40.91 48.18

75 32

70.09 29.9

72 35

67.29 32.71

76 34

69.09 30.91

41 69

37.27 62.73

34 31 43

31.48 28.70 39.81

67 23 18

62.04 21.30 16.67

33 34 37

31.73 32.69 35.58

66 9 29

63.46 8.65 27.88

24 86

21.82 78.18

27 83

24.55 75.45

14 96

12.73 87.27

6 104

5.45 94.55

0.0 110

0.0 100

2 108

1.82 98.1

p-value

95% CI

3.82 2.50 1.00

(1.52-9.59) (1.33-4.70) ---

--o0.001 0.009

0.655

o0.001

0.033

o0.001

0.631

0.033

0.157

(*) Wilcoxon signed rank test (**) McNemar Test; a: 1 missing/group; b: 3 missing/group; c: 39 missing/group; d: 2 missing/group; e: 6 missing/group. Mean Household Income in US Dollars. Abbreviations: Body Mass Index (BMI), Years (y), Standard Deviation (SD), Kilograms (kg).

socio-economic conditions (11). However, high-income countries have shown little or no change in these values over time. Improved sanitation and access to healthcare, better nutrition, and better socio-economic conditions may be related to this stabilization (12). Conversely, in low- and middle-income countries with unfavorable socio-economic conditions, women who live in urban areas with higher socio-economic profiles were shown to have a lower age at menarche than those who were living in rural and low-income areas. Importantly, girls who are in disadvantaged situations are unable to obtain adequate nutrition for proper growth and development (12). Moreover, it is important to emphasize the difference in dietary habits among settings with different levels of development regarding nutrition and age at menarche. A recent systematic review covering the period between 1980 and 2013 summarized non-genetic factors influencing age at menarche, but these findings were inconsistent. These factors have increasing relevance because they could be modified to improve womenâ&#x20AC;&#x2122;s health. For example, interventions including altering animal protein intake and physical activity could be implemented (13). The decline in age at menarche has been associated with increased BMI and insulin resistance and thus negative changes in lipid profile, which can lead to higher risks of cardiovascular and metabolic diseases in women (14). This

high school and 5.2-fold higher risk for elementary education), those with early menarche (2.1-times increased risk) and those who were adolescents at the 1st pregnancy (2.4-times increased risk for 15-19 years old). The cesarean delivery route showed a higher risk for morbid obesity (2.8-times higher risk) (Table 4).

â&#x20AC;&#x2122; DISCUSSION Our data showed an association of early menarche (less than 12 years old), teenage pregnancy (15 to 19 years old) and low education with morbid obesity in women aged 20 to 49. This study also confirms the strong relationship between morbid obesity and some non-communicable diseases, even during reproductive ages. Similar results were found in a cross-sectional study of 1,273 Iranian girls (10) that found an association of early menarche with greater adiposity and body fat. Another study found that girls who had early menarche had a decreased height, higher BMI, higher fat percentage and larger waist circumference compared to other girls (11). However, the association between early menarche and fat distribution pattern has not been clearly identified yet. The age of menarche is decreasing due to several factors, including genetics, environmental factors, educational level, nutritional status, sedentary lifestyle, life habits, and

549

Risk factors of morbid obesity Neves AG et al.

CLINICS 2017;72(9):547-553

Table 2 - Gynecological and obstetric characteristics of women with morbid obesity and adequate BMI. Characteristics

Morbid obesity (n=110)

Age of menarche (y)* (n=220) o12 X12 Age at 1st pregnancy (y)** (n=208)a o15 15-19 X 20 Nulliparous Number of pregnancies** (n=218)b 0 1 2 3 4 or + Parity** (n=214)c 0 1 2 3 4 or + C-section delivery* (n=74)d Yes No Regular menstrual cycle* (n=220) Yes No Use of any contraceptive method* (n=218)b Yes No Types of Contraceptive method* " Intrauterine device Combined oral contraception Combined injectable contraception Injectable depot medroxyprogesterone Others Interval between pregnancies (y)**

Mean BMI per pregnancy order** BMI 1st pregnancy(**) BMI 2nd pregnancy (**)

Control group (n=110)

95% CI

42 68

38.18 61.82

25 85

22.73 77.27

2.06 1.00

(1.14-3.75) ---

5 40 35 25

4.81 38.46 33.65 23.08

4 28 56 16

3.85 26.92 53.85 15.38

2.07 2.36 1.00 3.51

(0.43-9.97) (1.23-4.56) --(1.33-9.28)

26 28 30 14 11

23.85 25.69 27.52 12.84 10.09

16 41 30 14 8

14.68 37.61 27.52 12.84 7.84

2.67 1.00 1.42 1.33 1.79

(1.09-6.52) --(0.69-2.95) (0.53-3.32) (0.63-5.04)

25 36 29 9 8

23.36 33.64 27.10 8.41 7.48

16 46 35 8 2

14.95 42.99 32.71 7.48 1.87

2.25 1.00 1.08 1.45 4.33

(0.93-5.45) --(0.53-2.19) (0.47-4.42) (0.87-21.56)

51 23

68.92 31.08

40 34

54.05 45.95

1.86 1.00

(0.99-3.48) ---

60 50

54.55 45.45

66 44

60.00 40.00

1.00 1.23

--(0.73-2.07)

104 5

95.41 4.59

99 10

90.83 9.17

2.25 1.00

(0.69-7.31) ---

20 72 21 14 43

18.35 66.06 19.27 12.96 39.45

42 66 20 16 40

38.53 60.55 18.35 14.81 36.70

3.00 1.27 1.06 1.18 1.13

(1.52-5.94) (0.73-2.23) (0.54-2.10) (0.53-2.64) (0.65-1.95)

Mean

4.81

3.61

5.55

4.73

Mean

48 16

34.61 34.71

8.82 5.48

48 16

26.24 26.45

3.83 3.54

1.30 1.19

(1.12-1.51) (1.02-1.39)

5 missing/group; b: 1 missing/group; c: 3 missing/group; d: 36 missing/group. " : more than one could be used. (*) McNemar test; (**) Wilcoxon signed rank test. Abbreviations: Body Mass Index (BMI), Mean (M), Standard Deviation (SD).

this population. Girls who stay in school longer are less likely to become pregnant. In addition, when pregnant, many of them end up quitting their studies and risk their own economic prospectives and other opportunities because they must temporarily set aside their own life plan to care for the baby (16). In our sample, educational level was the major sociodemographic risk factor for the occurrence of morbid obesity, showing an association between a lower level of education and the development of this condition (OR 4.25; 95% CI 1.7310.48). A recent study in Brazil revealed that the percentage of overweight or obesity among women with 8 years of schooling was 58.3% and dropped to 36.6% among women with at least 12 years of study. A higher educational level could be related to better eating habits and a healthier lifestyle, which may contribute to maintaining the ideal weight for women (17).

requires special attention to improve health care for women and girls and promote increased survival into adulthood (12,14). Occurrence of the first pregnancy during adolescence was another reproductive variable associated with morbid obesity in this sample (OR 2.11; 95% CI 1.09-4.10). Pregnant adolescents are still developing and tend to continue accumulating fat instead of using their reserves, as occurs in adult women during pregnancy. Adolescents also have a greater increase in global and central adiposity compared with adult mothers. Throughout adulthood, the BMI tends to increase; thus, the association between teenage pregnancy and obesity may be due to the fact that women accumulate weight sooner than expected, with weight retention beginning during adolescence (15). Teenage pregnancy can also have consequences on health, education, and income, which could influence the weight in

550

Risk factors of morbid obesity Neves AG et al.

CLINICS 2017;72(9):547-553

Table 3 - Non-communicable diseases and urinary incontinence according to the occurrence or not of morbid obesity. Variable

Diabetes No Yes Hypertension No Yes Dyslipidemia No Yes Ischemic heart disease No Yes Liver Disease No Yes Pulmonary Disease No Yes Thyroid Dysfunction No Yes Stroke No Yes Thrombosis No Yes Urinary Incontinence No Yes Joint pain Yes No Pain intensity (n=210)a No pain p5 X6 a:

Morbid obesity (n=110)

Control group (n=110)

95% CI

96.36 3.64

1.00 5.67

--(1.66-19.34)

98 12

89.09 10.91

1.00 23.50

--(5.71-96.75)

76.36 23.64

98 12

89.09 10.91

1.00 2.75

--(1.22-6.18)

104 6

94.55 5.45

108 2

98.18 1.82.

1.00 3.00

--(0.61-14.86)

101 9

91.82 8.18

110 .

100.00 .

1.00 12.49

--(1.97-999.99)

87 23

79.09 20.91

105 5

95.45 4.55

1.00 5.50

--(1.90-15.96)

90 20

81.82 18.8

107 3

97.27 2.73

1.00 6.67

--(1.98-22.44)

109 1

99.09 0.91

109 1

99.09 0.91

1.00 1.00

--(0.06-15.99)

106 4

96.36 3.64

109 1

99.09 0.91

1.00 4.00

--(0.45-35.79)

77 33

70.00 30.00

89 21

80.91 19.09

1.00 1.71

--(0.94-3.10)

58 52

52.73 47.27

33 77

30.00 70.00

2.56 1.00

(1.44-4.57) ---

52 15 38

49.52 14.29 36.19

76 14 15

72.38 13.33 14.29

1.00 1.51 3.31

--(0.66-3.46) (1.65-6.66)

92 18

83.64 16.36

106 4

53 57

48.18 51.82

84 26

5 missing/group

sample was limited to identify an association in this study. Further studies with large sample sizes for the number of previous pregnancies or longitudinal studies are needed to better assess this risk factor. An increased incidence of C-section was found among women with morbid obesity (p=0.054). The association with maternal obesity has been linked to increased complications during pregnancy, labor, delivery, and postpartum (21). Maternal obesity may also increase the risks for C-section. A recent study suggested that in obese and overweight women, the progression of labor is significantly slower, which contributes to an increased labor induction rate and obstetric interventions, such as a C-section delivery (22). Among the comorbidities evaluated in this study, joint pain was more prevalent among the obese group. Musculoskeletal changes related to obesity are the result of body mechanic adaptations to maintain the balance due to increased body mass. This condition functionally limits the ability to perform tasks and exercise that could contribute to weight loss, aggravating the problem and contributing to the reduction in quality of life and life expectancy. Regular physical activity could contribute to weight loss, decreased joint problems and consequent improvement in quality of life (23,24).

Our data add to those presented in previous studies to contribute to health policies for adolescents that promote healthy pubertal development in relation to nutritional status and weight gain. Thus, public policies are needed to provide better care for women, prioritizing their sustainable development in the social, economic and environmental spheres, as advocated by the United Nations (16,18) Another factor considered in this study was the relationship between obesity and parity. Some researchers have shown that pregnancy and postpartum are a risk period for the development of excess weight. During pregnancy, women may gain more weight than recommended and consequently have more difficulty in postpartum weight loss (19). Similarly, pre-gestational BMI was related to gestational weight gain, differentiated between nulliparous and multiparous individuals (20). In this study, we did not observe an increase in BMI associated with an increased parity. However, the current research is limited as the type of study chosen for this analysis may have masked this result. Adoption of BMI as the main criterion for the inclusion of cases and controls may have masked the effect of this variable. A prospective cohort study would be ideal for this evaluation; since the number of cases decreases with the increase in parity, the power of the

551

Risk factors of morbid obesity Neves AG et al.

CLINICS 2017;72(9):547-553

Table 4 - Results of univariate and multivariate analyses for morbid obesity in women 20-49 years old.

Variable Age of menarche (years) Age at 1st pregnancy (years)

Educational level

Number of pregnancies

Number of deliveries

Mode of delivery

Categories X12 (ref.) o12 X20 (ref.) 15-19 o15 Nulliparous College/University (ref.) High School Elementary School 1 (ref.) 2 3 X4 Nulliparous 1 (ref.) 2 3 X4 Nulliparous Vaginal (ref.) C-section Nulliparous

Univariate

Multivariate

n=110 eutrophic and n=110 morbid obese

n=107 eutrophic and n=107 morbid obese

O.R.* 1.00 2.06 1.00 2.36 2.07 3.51 1.00 2.50 3.82 1.00 1.42 1.33 1.79 2.67 1.00 1.08 1.45 4.33 2.25 1.00 1.86 3.45

CI 95% O.R.*

O.R.*

CI 95% O.R.*

--1.14-3.75 --1.23-4.56 0.43-9.97 1.33-9.28 --1.32-4.70 1.52-9.59 --0.69-2.95 0.53-3.32 0.63-5.04 1.09-6.52 --0.53-2.19 0.47-4.42 0.87-21.56 0.93-5.45 --0.99-3.48 1.29-9.24

1.00 2.33 1.00 2.90 2.25 10.20 1.00 3.11 6.93 -

--1.05-5.18 --1.34-6.28 0.32-15.72 2.49-41.82 --1.43-6.75 2.08-23.11 -

1.00 2.77 10.20

--1.22-6.29 2.49-41.82

* OR (Odds Ratio) = hazard ratio for morbid obesity; 95% CI OR = 95% confidence interval for the hazard ratio. (Ref).: reference level. Stepwise criteria for selecting variables.

The results presented in our study may contribute to other analyses showing that morbidly obese women of reproductive age have similar risk factors as the overweight and obese population. However, these data may help to clarify this issue, contributing to actions aimed at better monitoring of different stages of women’s lives to establish potential risk factors that may lead to an increase in obesity (morbid obesity) and its complications. This is one more step toward providing education, gender equality and access to quality health care to ensure the physical, mental, and social wellbeing and a better quality of life for this population.

This study also evaluated the physical activity among women and noted that in the morbidly obese group, this habit was more prevalent (po0.001), both with regard to weekly frequency of practice and daily duration of this activity. However, it is important to note that these women were supported by a multidisciplinary team at the hospital and thus were encouraged to engage in physical activity according to the World Health Organization and/or the Institute of Medicine recommendations (23,24). Few studies have addressed the relationship between the use of alcohol and morbid obesity; in the present study, there was higher consumption of alcohol among women with morbid obesity compared with non-obese women identified in the bivariate analysis but not in the multivariate analysis. A previous study (25) showed increased alcohol consumption among morbidly obese subjects while awaiting bariatric surgery, while another (26) identified obesity as a protective factor for women with excessive alcohol consumption. These are isolated data, and studies designed specifically to assess this issue should be conducted to prevent misleading conclusions. The increasing incidence of obesity and the associated risk of chronic non-communicable diseases, mortality and poor quality of life have made this an important public health issue, given the complications caused by obesity and the costs related to its treatment (1). Thus, identifying the risk factors involved in the development of morbid obesity among women of reproductive age will contribute to the planning and implementation of educational programs to prevent obesity in general and improve quality of life among this population. Another limitation of this study is that the questionnaire developed by our research group has not been validated before and the information gathered by the questionnaire may be compromise due to recall bias.

’ ACKNOWLEDGMENTS We wish to thank the statistical ofﬁce of CAISM-UNICAMP for performing the statistical analysis and the SAR3HAS (Reproductive Health and Healthy Habits) Group for the valuable discussion regarding data analysis.

’ AUTHOR CONTRIBUTIONS Neves AG, Surita FG and Chaim EA developed the research and study design. Neves AG, Godoy-Miranda AC and Oshika FH were responsible for the data collection. Neves AG, Surita FG and Kasawara KT performed the data analysis and interpretation. Neves AG, Surita FG and Kasawara KT wrote the ﬁrst draft of the manuscript. All authors reviewed and approved the ﬁnal version of the manuscript.

’ REFERENCES 1. Ng M, Fleming T, Robinson M, Thomson B, Graetz N, Margono C, et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384(9945):766–81, http://dx. doi.org/10.1016/S0140-6736(14)60460-8. 2. World Health Organization (WHO). Obesity and overweight. WHO [Internet]. World Health Organization; 2016 [cited 2016 Sep 4];

552

Risk factors of morbid obesity Neves AG et al.

CLINICS 2017;72(9):547-553

3. 4.

10.

11. 12.

13. 14.

Available from: http://www.who.int/gho/ncd/risk_factors/overweight_text/en/. World Health Organization (WHO). Obesity - Situation and trends. WHO [Internet]. World Health Organization; 2015 [cited 2016 Oct 10]; Available from: http://www.who.int/gho/ncd/risk_factors/obesity_text/en/. Prospective Studies Collaboration. Whitlock G, Lewington S, Sherliker P, Clarke R, Emberson J, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet. 2009;373(9669):1083–96, http://dx.doi.org/10.1016/S0140-6736 (09)60318-4. Lutsiv O, Mah J, Beyene J, McDonald SD. The effects of morbid obesity on maternal and neonatal health outcomes: a systematic review and meta-analyses. Obes Rev. 2015;16(7):531–46, http://dx.doi.org/10.1111/ obr.12283. Nohr EA, Timpson NJ, Andersen CS, Davey Smith G, Olsen J, Sørensen TI. Severe obesity in young women and reproductive health: the Danish National Birth Cohort. PLoS One. 2009;4(12):e8444, http://dx.doi.org/ 10.1371/journal.pone.0008444. Correia LL, da Silveira DM, e Silva AC, Campos JS, Machado MM, Rocha HA, et al. [Prevalence and determinants of obesity and overweight among reproductive age women living in the semi-arid region of Brazil]. Cien Saude Colet. 2011;16(1):133–45, http://dx.doi.org/10.1590/S1413-8123 2011000100017. The EQUATOR Network. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies [Internet]. Available from: http://www. strobe-statement.org/index.php?id=strobe-home. Teichmann L, Olinto MT, Costa JS, Ziegler D. Fatores de risco associados ao sobrepeso e a obesidade em mulheres de São Leopoldo, RS. Rev Bras Epidemiol. 2006;9(3):360–73, http://dx.doi.org/10.1590/S1415-790X2006 000300010. Al-Awadhi N, Al-Kandari N, Al-Hasan T, Almurjan D, Ali S, Al-Taiar A. Age at menarche and its relationship to body mass index among adolescent girls in Kuwait. BMC Public Health. 2013;13:29, http://dx.doi. org/10.1186/1471-2458-13-29. Pathak PK, Tripathi N, Subramanian SV. Secular trends in menarcheal age in India-evidence from the Indian human development survey. PLoS One. 2014;9(11):e111027, http://dx.doi.org/10.1371/journal.pone.0111027. Ameade EP, Garti HA. Age at Menarche and Factors that Influence It: A Study among Female University Students in Tamale, Northern Ghana. PLoS One. 2016;11(5):e0155310, http://dx.doi.org/10.1371/journal.pone. 0155310. Yermachenko A, Dvornyk V. Nongenetic determinants of age at menarche: a systematic review. Biomed Res Int. 2014;2014:371583, http://dx. doi.org/10.1155/2014/371583. Lakshman R, Forouhi NG, Sharp SJ, Luben R, Bingham SA, Khaw KT, et al. Early age at menarche associated with cardiovascular disease and

15.

16. 17.

18. 19.

20.

21. 22.

23.

24. 25.

26.

553

mortality. J Clin Endocrinol Metab. 2009;94(12):4953–60, http://dx.doi. org/10.1210/jc.2009-1789. Tanaka T, Ashihara K, Nakamura M, Kanda T, Fujita D, Yamashita Y, et al. Associations between the pre-pregnancy body mass index and gestational weight gain with pregnancy outcomes in Japanese women. J Obstet Gynaecol Res. 2014;40(5):1296–303, http://dx.doi.org/10.1111/jog.12353. UNFPA. Motherhood in Childhood: Facing the challenge of adolescent pregnancy. [cited 2016 Oct 10]; Available from: https://www.unfpa.org/ sites/default/files/pub-pdf/EN-SWOP2013-final.pdf. Malta DC, Stopa SR, Iser BP, Bernal RT, Claro RM, Nardi AC, et al. Risk and protective factors for chronic diseases by telephone survey in capitals of Brazil, Vigitel 2014. Rev Bras Epidemiol. 2015;18(Suppl 2):238–55, http://dx.doi.org/10.1590/1980-5497201500060021. United Nations. The Millennium Development Goals Report. United Nations [Internet]. 2015;72. Available from: https://visit.un.org/millenniumgoals/2008highlevel/pdf/MDG_Report_2008_Addendum.pdf. Institute of Medicine, National Research Council (US) Committee to Reexamine IOM Pregnancy Weight Guidelines. Weight gain during pregnancy: reexamining the guidelines. Washington (DC): National Academies Press (US). 2009. Paulino DS, Surita FG, Peres GB, do Nascimento SL, Morais SS. Association between parity, pre-pregnancy body mass index and gestational weight gain. J Matern Fetal Neonatal Med. 2016;29(6):880–4, http://dx. doi.org/10.3109/14767058.2015.1021674. Berendzen JA, Howard BC. Association between cesarean delivery rate and body mass index. Tenn Med. 2013;106(1):35–7. Vinturache A, Moledina N, McDonald S, Slater D, Tough S. Pre-pregnancy Body Mass Index (BMI) and delivery outcomes in a Canadian population. BMC Pregnancy Childbirth. 2014;14:422, http://dx.doi.org/10.1186/ s12884-014-0422-y. Berry PA, Wluka AE, Davies-Tuck ML, Wang Y, Strauss BJ, Dixon JB, et al. The relationship between body composition and structural changes at the knee. Rheumatology. 2010;49(12):2362–9, http://dx.doi.org/10.1093/ rheumatology/keq255. World Health Organization (WHO). Global Recommendations on Physical Activity for Health. [cited 2016 Oct 10]; Available from: http://apps. who.int/iris/bitstream/10665/44399/1/9789241599979_eng.pdf. Backman O, Stockeld D, Rasmussen F, Näslund E, Marsk R. Alcohol and substance abuse, depression and suicide attempts after Roux-en-Y gastric bypass surgery. Br J Surg. 2016;103(10):1336–42, http://dx.doi. org/10.1002/bjs.10258. Pickering RP, Goldstein RB, Hasin DS, Blanco C, Smith SM, Huang B, et al. Temporal relationships between overweight and obesity and DSMIV substance use, mood, and anxiety disorders: results from a prospective study, the National Epidemiologic Survey on Alcohol and Related Conditions. J Clin Psychiatry. 2011;72(11):1494–502, http://dx.doi.org/ 10.4088/JCP.10m06077gry.

CLINICAL SCIENCE

Endoscopic endonasal approach for mass resection of the pterygopalatine fossa Jan Plza´k,I,* Vı´t Kratochvil,I Adam Kesˇner,I Pavol Sˇurda,II Alesˇ Vlasa´k,III Eduard ZveˇrˇinaI I Department of Otorhinolaryngology and Head and Neck Surgery, 1st Faculty of Medicine, Charles University, University Hospital Motol, V U´valu 84, 150 06, Prague 5, Czech Republic. II Guy’s and St Thomas’ NHS Foundation Trust, Great Maze Pond, SE1 9RT London, UK. III Department of Neurosurgery, ´ valu 84, 150 06, Prague 5, Czech Republic. 2nd Faculty of Medicine, Charles University, University Hospital Motol, V U

OBJECTIVES: Access to the pterygopalatine fossa is very difficult due to its complex anatomy. Therefore, an open approach is traditionally used, but morbidity is unavoidable. To overcome this problem, an endoscopic endonasal approach was developed as a minimally invasive procedure. The surgical aim of the present study was to evaluate the utility of the endoscopic endonasal approach for the management of both benign and malignant tumors of the pterygopalatine fossa. METHOD: We report our experience with the endoscopic endonasal approach for the management of both benign and malignant tumors and summarize recent recommendations. A total of 13 patients underwent surgery via the endoscopic endonasal approach for pterygopalatine fossa masses from 2014 to 2016. This case group consisted of 12 benign tumors (10 juvenile nasopharyngeal angiofibromas and two schwannomas) and one malignant tumor. RESULTS: No recurrent tumor developed during the follow-up period. One residual tumor (juvenile nasopharyngeal angiofibroma) that remained in the cavernous sinus was stable. There were no significant complications. Typical sequelae included hypesthesia of the maxillary nerve, trismus, and dry eye syndrome. CONCLUSION: The low frequency of complications together with the high efficacy of resection support the use of the endoscopic endonasal approach as a feasible, safe, and beneficial technique for the management of masses in the pterygopalatine fossa. KEYWORDS: Endoscopic Endonasal Approach; Pterygopalatine Fossa; Skull Base; Tumor. Plza´k J, Kratochvil V, Kesˇner A, Sˇurda P, Vlasa´k A, Zveˇrˇina E. Endoscopic endonasal approach for mass resection of the pterygopalatine fossa. Clinics. 2017;72(9):554-561 Received for publication on April 18, 2017; First review completed on May 29, 2017; Accepted for publication on July 7, 2017 *Corresponding author. E-mail: jan.plzak@lf1.cuni.cz

’ INTRODUCTION

Vidian nerve and artery. The palatovaginal (pharyngeal) canal is located medially to the Vidian canal, and the sphenopalatine foramen is a communication medial to the nasal cavity via the palatine bone. The descending palatine canal continues inferiorly to the apex of the PPT and opens into the oral cavity via the greater and lesser palatine foramina. The foramen rotundum, containing V2, is a communication that is located cranially to the middle crania fossa. The inferior orbital fissure is open anterosuperiorly to the orbit. The PPF is a very complex anatomical area of the skull base with interconnections to the head and neck (Figure 1) (1–3). Accessing the PPF is difficult, which is traditionally approached via an open method, such as lateral rhinotomy, midfacial degloving, facial translocation, transantral maxillectomy, and the Fisch C and D procedures. Although these procedures provide good exposure of the PPF, they are often complicated by unacceptable facial scaring and deformity as well as dysfunction of the facial and infraorbital nerves. Endoscopic sinus surgery is now a standard procedure for inflammatory sinonasal disease, and more recently, it has been adopted as a basic approach for the treatment of benign sinonasal tumors, such as inverted papilloma or juvenile

The pterygopalatine fossa (PPF) is a narrow, inverted, cone-shaped space that is localized posteriorly to the dorsal wall of the maxillary sinus and consists of fat, the pterygopalatine ganglion, the Vidian nerve, the maxillary nerve (V2), and terminal branches of the maxillary artery. The anteroposterior order of the layers is as follows: fat, blood vessels, and neural structures. The posterior wall of the PPF is formed by the base of the pterygoid process and is bordered by the palatine bone medially and the middle cranial fossa cranially. Laterally, the PPF is open to the infratemporal fossa (ITF) via the pterygomaxillary fissure. The Vidian canal opens into the posterior part of the PPF and contains the

554

Endoscopy of the pterygopalatine fossa PlzaÂ´k J et al.

CLINICS 2017;72(9):554-561

Figure 1 - CT anatomy of the pterygopalatine fossa (asterisk). (A) An axial image showing a communication to the ITF via the pterygomaxillary fissure (long arrow) to the middle cranial fossa via the foramen rotundum (dashed arrow) and to the nasal cavity via the sphenopalatine foramen (short arrow). (B) An axial image in a different plane showing the Vidian canal running to the foramen lacerum (short arrow). (C) A coronal image showing a communication to the nasal cavity via the sphenopalatine foramen (short arrow) to the orbit via the infraorbital fissure (dashed arrow) and to the ITF via the pterygomaxillary fissure (long arrow). (D) A sagittal image showing the greater palatine canal (short arrow) running to the oral cavity with a communication to the orbit via the infraorbital fissure (dashed arrow) and to the middle cranial fossa via the foramen rotundum (long arrow).

â&#x20AC;&#x2122; MATERIAL AND METHODS

nasopharyngeal angiofibroma (JNA) (4,5). Advances in endoscopic endonasal surgery, such as interventional radiology with preoperative embolization, improved instrumentation, computer-based navigation, and hemostatic materials, as well as a two-surgeon technique, have enabled endoscopic endonasal access to tumors of the PPF (6). Primary tumors of the PPF, such as schwannomas of V2, are rare, while sinonasal tumors more frequently extend into the PPF (7). Hence, relatively few studies (most of which were case series involving low numbers of patients) have investigated the endoscopic endonasal approach (EEA) to access tumors of the PPF (8â&#x20AC;&#x201C;11). Our study follows the recommendations of previous authors, who called for the publication of more cases. Therefore, the surgical aim of the present study was to evaluate the utility of the EEA for the management of both benign and malignant tumors of the PPF. The present study included the highest number of cases per year among related published reports. Adequate follow-up and appropriate postoperative care (a novelty of trismus physiotherapy) were emphasized.

The surgical records of 13 patients who underwent resection of a tumor of the PPF via the EEA at the Department of Otorhinolaryngology and Head and Neck Surgery, 1st Faculty of Medicine, Charles University, University Hospital Motol, Prague, Czech Republic from April 2014 to March 2016 were retrospectively reviewed. All patients who underwent resection via the EEA for a tumor of the PPF during this period were enrolled. The study protocol was approved by the institutional review board of our hospital, and written informed consent was obtained from all patients. Diagnosis was based on patient history and the results of complete clinical otorhinolaryngological examinations, which included nasal endoscopy, computed tomography (CT), and magnetic resonance imaging (MRI). No biopsy was performed for 10 patients with clinical and imaging findings that were consistent with a diagnosis of JNA. For confirmed cases of JNA, preoperative angiography with embolization

555

Endoscopy of the pterygopalatine fossa Plza´k J et al.

CLINICS 2017;72(9):554-561

lateral pterygoid muscle. Navigation with CT/MRI fusion and perioperative frozen sections were used to ensure safe radical resection. The final step consisted of drilling around the Vidian canal to prevent JNA recurrence. In case of malignancy, the pterygoid root and plates were also resected.

was performed one day before surgery. Two patients had schwannomas of V2: one was biopsied from a portion of the tumor in the lateral part of the choana at an otorhinolaryngology department in another hospital, and the second was diagnosed with a high probability based on preoperative assessment and was subsequently confirmed by perioperative biopsy. The last patient in this cohort was the only case with malignancy. In this patient, undifferentiated sinonasal carcinoma (cT4b cN0) infiltrated the bilateral nasal cavities, bilateral ethmoids, unilateral maxillary and sphenoid sinuses, unilateral PPF, and the brain in the anterior cranial fossa. This was the only case to undergo combination surgery via the EEA for the management of the sinonasal and pterygopalatine portions of the tumor and bifrontal craniotomy for the intracranial portion. This patient also received adjuvant chemoradiotherapy. Postoperative followup of all cases included clinical examination with an emphasis on endoscopic evaluation and radiological workup according to European guidelines (4). All surgeries were performed under general anesthesia. In 11 cases of sinonasal tumors extending to the PPF (10 cases of JNA and one case of undifferentiated sinonasal carcinoma), the tumor in the nasal cavity blocking the approach to the posterior wall of the maxillary sinus was resected using a piecemeal technique to expose the surgical intranasal corridor as the first step. The nasal septum was resected in cases of undifferentiated sinonasal carcinomas due to tumor infiltration and in schwannoma cases with large extensions to the middle cranial fossa, ITF, and cheek to safely achieve a larger working space. The second step (the ethmoid step) involved anteroposterior ethmoidectomy, sphenoidectomy, extensive supraturbinal antrostomy, and partial resection of the middle (dorsal part) and superior turbinates. Antrostomy was extended to the posterior wall of the maxillary sinus. In cases of infiltration of undifferentiated sinonasal carcinoma, complete medial maxillectomy was performed. Complete medial maxillectomy was also performed in cases of giant schwannoma (P11) extending from the PPF through the ITF laterally from the sagittal plane passing at the lateral border of the maxillary sinus to ensure sufficient lateral access. The sphenopalatine foramen and sphenopalatine artery entering the nasal cavity near the tip of the middle turbinate was identified as an important surgical landmark. The posterior wall of the maxillary sinus was then removed while preserving the periosteum of the PPF as the third step. Preservation of the periosteum surfacing the PPF was helpful for subsequent identification of the Vidian nerve medially and the maxillary nerve superolaterally. The extent of the surgical window in the posterior wall was dependent on the exact location and extent of the tumor in the PPF. The extent of drilling around the sphenopalatine foramen was also determined based on the extent of PPF involvement. The fourth step was appropriate resection of the PPF tumor. The first structure encountered after incision of the PPF periosteum was fat tissue, followed by blood vessels. The distal branches of the maxillary artery (Vidian, descending palatal, and palatovaginal branches) and its main trunk were identified and cauterized. The neural structures all lie deeper than this plexus of arteries. During blunt dissection of the tumor, maximal effort was taken to preserve the maxillary and infraorbital nerves, the greater and lesser palatine nerves, and the Vidian nerve (if they were not involved in the tumor) to minimize neurological morbidity. The posterolateral border of dissection was the anterior surface of the

’ RESULTS The group of 13 patients with tumors of the PPF who were treated via the EEA included 11 males and 2 females with a mean age of 21.8 (range, 15–56) years (Table 1). The following symptoms were observed among these 13 patients: unilateral nasal obstruction in 12 patients, epistaxis in six patients, and pain in four patients. One patient (P12) with a smaller schwannoma had no clinical symptoms with incidental finding on MRI. A total of 12 patients had benign tumors: 10 cases of JNA (3x IIa, 1x IIb, 5x IIc, and 1x IIIb according to the Radkowski grading system, i.e., six cases with extension of the tumor out of the PPF to the ITF – grade IIc and/or intracranially – grade IIIb) and two with schwannomas arising from the maxillary nerve. Representative cases of JNA (P5) and schwannoma (P11) are shown in Figures 2 and 3, respectively. One malignant tumor was undifferentiated sinonasal carcinoma (cT4b cN0) that infiltrated the bilateral nasal cavities, bilateral ethmoids, unilateral maxillary and sphenoid sinuses, and unilateral PPF, including the intracranial portion with brain infiltration (Figure 4). This patient was the only one that underwent combination surgery, i.e., EEA for the management of the sinonasal and pterygopalatine portions of the tumor and bifrontal craniotomy for the intracranial portion. This patient was also the only one to receive adjuvant chemoradiotherapy using proton beam therapy. The optimal management of sinonasal undifferentiated carcinoma remains unclear. Treatment strategies vary among institutions, and no single approach has demonstrated a clear therapeutic advantage. Surgery as the primary mode of treatment with postoperative chemoradiotherapy is acceptable for patients with resectable disease and improves oncologic control (4,13). Ten tumors (all JNAs) originated in the nasopharynx with extension to the PPF (in six cases, more laterally to the ITF). Both schwannomas of V2 were primary tumors of the PPF. Seven patients with JNA in this cohort underwent primary surgery, and three were revision surgeries. Case P2 underwent external lateral rhinotomy 2 years before endoscopic revision for a growing residual tumor in the PPF and ITF. Case P7 was a giant JNA that was endoscopically resected three times in 2 years at another otorhinolaryngology department. The last surgery was performed without preoperative embolization due to the blood supply from the internal carotid artery. This surgery was stopped prematurely due to uncontrollable bleeding from the tumor. Subsequent multiple attempts for nasal tamponade removal were unsuccessful due to continued uncontrollable bleeding, and the patient was transported to the Department of Otorhinolaryngology and Head and Neck Surgery, 1st Faculty of Medicine, Charles University with a residual JNA in the PPF and ITF. Subsequent R0 endoscopic resection was performed without preoperative embolization. JNA case P8 underwent endoscopic resection twice in 2 years at another otorhinolaryngology department. A residual tumor remained in the PPF and ITF; therefore, the patient underwent revision surgery at the Department of Otorhinolaryngology and Head and Neck Surgery, 1st Faculty of Medicine, Charles University.

556

557

56 44

P10

P11

P12 P13

F F

Sex

Schwannoma SNUC T4bN0 M0

JNA Radkowski 2c JNA Radkowski 2c JNA Radkowski 2a JNA Radkowski 2a JNA Radkowski 2c JNA Radkowski 3b JNA Radkowski 2c JNA Radkowski 2c JNA Radkowski 2b JNA Radkowski 2a Schwannoma

Diagnosis

PPF SN

PPF

Origin

PPF, ITF, cheek, MCF PPF, ITF PPF

PPF

PPF, ITF

PPF, ITF, MCF PPF, ITF

PPF, ITF

PPF

PPF, ITF

Extension

32 21

Maximal laterolateral size in PPF /+ITF/(mm)

0 0

3x endoresection at another department 2x endoresection at another department 0

lat. rhinotomy 2 yrs ago 0

Previous treatment

0 0

embolization

0 (ICA supply)

embolization

Preoperative management

R0 R0

R1 (cavernous sinus) R0

Resection

0 CHRT (proton beam therapy)

Adjuvant therapy

19 36

Followup (mo)

V2 hypesthesia trismus

V2 hypesthesia, transient trismus

transient V2 hypesthesia

V2 hypesthesia, trismus

dry eye syndrome, transient V2 hypesthesia V2 hypesthesia

transient V2 hypesthesia

Sequelae

Abbreviations: M = male; F = female; JNA = juvenile nasopharyngeal angiofibroma; SNUC = sinonasal undifferentiated carcinoma; NP = nasopharynx; PPF = pterygopalatine fossa; SN = sinonasal; ITF = infratemporal fossa; MCF = middle cranial fossa; ICA = internal carotid artery; CHRT = chemoradiotherapy

Age (yr)

Case

Table 1 - Summary of the clinical characteristics of patients undergoing an endonasal endoscopic approach to the pterygopalatine fossa.

CLINICS 2017;72(9):554-561 Endoscopy of the pterygopalatine fossa PlzaÂ´k J et al.

Endoscopy of the pterygopalatine fossa Plza´k J et al.

CLINICS 2017;72(9):554-561

Figure 2 - Juvenile nasopharyngeal angiofibroma, case P5, Radkowski grade 2c. Axial CT (A) and MRI T2-weighted scans (B) showing a mass in the dorsal two thirds of the left nasal cavity and in the sphenoid sinus (marked by S) with a bulky extension to the pterygopalatine fossa and the ITF. Extensive anteroposterior widening of the sphenopalatine foramen is marked by arrows. (C) An axial MRI T2-weighted scan one year after surgery demonstrating no evidence of disease.

cavities to the PPF and ITF (12) and is indicated not only for inflammatory disease but also for benign and even malignant tumors that are very difficult to reach in this anatomical region (13). The use of a natural pathway via the nasal and paranasal cavities reduces morbidity associated with the access route compared to the external approach. Cosmesis, especially external scarring, is also an important factor in favor of the EEA. Angled endoscopes provide much better visualization of the PPF, which is difficult to access. A profound knowledge of the PPF and ITF and their neighboring spaces is essential for performing safe surgery. The creation of an appropriate endoscopic surgical corridor is crucial to allow safe manipulation, tissue dissection, and hemostasis. It is more difficult to manage minor bleeding using the EEA than when using an open approach. In addition, a narrower variety of instruments are usable for mass dissection in the EEA. Biopsy as well as radical resection of PPF tumors via the EEA have been reported in the literature (14-16). The case series reported by Battaglia et al. (11) describing 37 patients who were treated during a 13-year period is among the most extensive. Our study follows the recommendations of this group, which called for the publication of more cases. The present study included the highest number of cases per year among related published reports. The PPF is one of the most anatomically complex regions of the human body because it contains a range of vital neural and vascular structures with interconnections to the orbit, middle cranial fossa, nasal cavity, and oral cavity (17,18). All of these structures must be carefully considered when planning surgery and especially during the surgical procedure. Preservation of the neural structures in the PPF is

Of the 13 patients, 12 had no microscopic residual disease (R0). In case of JNA P6 (stadium Radkowski grade 3b – intracranial extension with cavernous sinus invasion), a residual tumor of approximately 1 cm in diameter was left in the cavernous sinus area. Repetitive postoperative MRI at 6-month intervals showed that the residual tumor was stable. No development of a new residual or recurrent tumor was observed in any of the other cases during the follow-up period. The mean blood loss during surgery was 550 (range, 100– 1,700) mL. Ten patients received intraoperative transfusions. No significant intraoperative neuro/vascular complications occurred. The mean duration of postoperative hospitalization was 5.8 (range, 4–12) days. There were no postoperative infections or systemic complications. Postoperative neural dysfunction was observed in eight cases: four cases of transient V2 hypesthesia and four cases of permanent V2 hypesthesia (two cases of V2 schwannoma; JNA Radkowski grade 3b in case P6 and JNA Radkowski grade 2c after three previous non-radical surgeries in case P7). Dry eye syndrome due to pterygopalatine ganglion dysfunction occurred in one case. Permanent trismus was present in two cases: one with a JNA (case P7) and one with a carcinoma (case P13), in which deterioration mainly occurred after chemoradiotherapy.

’ DISCUSSION The management of PPF lesions represents one of the areas in which open surgery has been shifted back to the stage of the last decade. Endoscopic endonasal surgery is performed for lesions with boundaries that extend beyond the paranasal

558

Endoscopy of the pterygopalatine fossa Plza´k J et al.

CLINICS 2017;72(9):554-561

Figure 3 - Schwannoma of V2, case P11. Coronal (A) and axial CT (B) scans showing considerable deformation of the splanchno and neurocranium on the right side. A mass (74 x 65 x 61 mm) arising from V2 in the pterygopalatine fossa with bulky extension to the ITF, cheek, and middle cranial fossa via the foramen rotundum, which is extremely enlarged (long arrow) – see side difference. The schwannoma pushed the posterior wall of the maxillary sinus forward (short arrow) and the pterygoid process backward (dashed arrow) – also see side differences. (C) An axial T1-weighted MRI scan with gadolinium contrast showing the margin between the tumor and soft tissue. (D) A sagittal T2-weighted MRI scan showing that the tumor is pressing on the orbital apex via the infraorbital fissure (long arrow) with an intracranial portion (24 x 21 x 20 mm) in the middle cranial fossa (dotted line). Postoperative T2-weighted MRI scans in the sagittal (E) and axial (F) planes demonstrate no evidence of disease 21 months after surgery. A nasoseptal flap (long arrow) was used to cover exposed dura mater of the middle cranial fossa, although no clear leakage of cerebrospinal fluid was observed intraoperatively. Considerable retraction of soft tissue into the postresection empty space of the cheek and ITF is visible on the axial MRI – compare (C) and (F).

symptom. All other patients did not develop trismus, presumably because of early rehabilitation of mouth opening starting on the 2nd to 5th postoperative days. The patients were instructed during self-rehabilitation to perform a daily exercise of mouth opening. The use of total endoscopic endonasal management of residual/recurrent disease located in the PPF remains controversial. There were three such cases in our study, all of which were JNAs. Cases P7 and P8 were treated first by repeated endonasal endoscopy via the EEA at other departments. After careful evaluation of images, we found no resections of the posterior wall of the maxillary sinus. Therefore, we expected no previous manipulation in the PPF and no scarring. Our presumptions were confirmed during the surgeries; thus, radical endonasal endoscopic resections were performed. The third patient (P2) was the last who underwent external surgery for JNA at our Department of Otorhinolaryngology. Additionally, in this case, there was no intervention in the PPF through the posterior wall of the maxillary sinus during the primary external surgery; thus, subsequent revision endoscopic surgery for a growing mass in the PPF was successful. The low frequency of complications together with the high efficacy of resection (no recurrence and only one stable persistent residuum in the cavernous sinus) support

necessary to limit surgical morbidity. Vascular complications are mostly caused by damage to branches of the maxillary artery. The management of massive blood loss is one of the limiting steps of the EEA to access the PPF. No major intraoperative complications occurred among the 13 patients in this cohort, while 9 (69.2%) developed minor complications, which included transient complications in three cases (V2 hypesthesia and trismus). Permanent postoperative sequelae were present in 6 (46.1%) patients, which included dry eye syndrome in one case, V2 hypesthesia in three cases, V2 hypesthesia with trismus in one case, and trismus in one case. V2 damage was associated with extensive PPF involvement in advanced JNA and with schwannomas arising from V2. Resection extending to the pterygoid muscle is thought to induce postoperative trismus. Therefore, we recommend trismus physiotherapy for all PPF surgeries. Despite this precaution, permanent trismus was observed in two cases. Case P7 developed slight permanent trismus after surgery for JNA, which might have resulted from the three previous non-radical surgeries. One case of carcinoma (case P13) developed permanent trismus after postoperative chemoradiotherapy. In case P11 with giant trigeminal schwannoma originating from the pterygopalatine fossa, we succeeded with the treatment of postoperative trismus by physiotherapy, and the patient is now free of this

559

Endoscopy of the pterygopalatine fossa PlzaÂ´k J et al.

CLINICS 2017;72(9):554-561

Figure 4 - Sinonasal undifferentiated carcinoma T4b, case P13. Axial PET/CT (A) and axial (B) and coronal (D) MRI T1-weighted with gadolinium contrast images showing a mass in the bilateral nasal cavities, bilateral ethmoid sinuses, right maxillary and sphenoid sinuses, and right pterygopalatine fossa (arrows) with an intracranial portion with brain infiltration. Axial (C) and coronal (E) MRI T1weighted scans with gadolinium contrast demonstrating no evidence of disease two years after surgery and postoperative chemoradiotherapy.

(2) the availability of adequate technical equipment (endoscopes, endoscopic skull base instruments, a navigation system, hemostatic facilities, etc.), and (3) adequate experience and skill of the surgeon in the use of advanced endoscopic techniques.

endoscopic endonasal surgery as a feasible, safe, and beneficial technique for the management of masses in the PPF. The following are fundamental prerequisites to safely and effectively performing this approach: (1) thorough preoperative evaluation of imaging examinations (MRI/CT),

560

Endoscopy of the pterygopalatine fossa Plza´k J et al.

CLINICS 2017;72(9):554-561

’ ACKNOWLEDGMENTS This study was supported by Progres Q28, UNCE 204013 and the Ministry of Health of the Czech Republic (grant Nos. 15- 28933A and 16-29032A).

8. 9.

’ AUTHOR CONTRIBUTIONS

10.

Plzák J and Zveˇ rˇ ina E were responsible for the study design, surgeons, manuscript writing and supervision of the study. Kratochvil V, Kešner A, Šurda P and Vlasák A were responsible for the data collection and processing, surgeons, and manuscript revision

11.

’ REFERENCES 12.

1. Alfieri A, Jho HD, Schettino R, Tschabitscher M. Endoscopic endonasal approach to the pterygopalatine fossa: anatomic study. Neurosurgery. 2003; 52(2):374-78, http://dx.doi.org/10.1227/01.NEU.0000044562.73763.00. 2. Cavallo LM, Messina A, Gardner P, Esposito F, Kassam AB, Cappabianca P, et al. Extended endoscopic endonasal approach to the pterygopalatine fossa: anatomical study and clinical considerations. Neurosurg Focus. 2005;19(1):E5. 3. Solari D, Magro F, Cappabianca P, Cavallo LM, Samii A, Esposito F, et al. Anatomical study of the pterygopalatine fossa using an endoscopic endonasal approach: spatial relations and distances between surgical landmarks. J Neurosurg. 2007;106(1):157-63, http://dx.doi.org/10.3171/ jns.2007.106.1.157. 4. Lund VJ, Stammberger H, Nicolai P, Castelnuovo P, Beal T, Beham A, et al. European position paper on endoscopic management of tumours of the nose, paranasal sinuses and skull base. Rhinol Suppl. 2010;22:1-143. 5. Nicolai P, Villaret AB, Farina D, Nadeau S, Yakirevitch A, Berlucchi M, et al. Endoscopic surgery for juvenile angiofibroma: a critical review of indications after 46 cases. Am J Rhinol Allergy. 2010;24(2):e67-72, http:// dx.doi.org/10.2500/ajra.2010.24.3443. 6. Tomazic PV, Gellner V, Koele W, Hammer GP, Braun EM, Gerstenberger C, et al. Feasibility of piezoelectric endoscopic transsphenoidal craniotomy: a cadaveric study. Biomed Res Int. 2014;2014:341876, http://dx. doi.org/10.1155/2014/341876. 7. Karligkiotis A, Turri-Zanoni M, Sica E, Facco C, Freguia S, Mercuri A, et al. Role of endoscopic surgery in the management of sinonasal and

13.

14. 15.

16. 17. 18.

561

skull base schwannomas. Head Neck. 2016;38(Suppl 1):E2074-82, http:// dx.doi.org/10.1002/hed.24383. Klossek JM, Ferrie JC, Goujon JM, Fontanel JP. Endoscopic approach of the pterygopalatine fossa: report of one case. Rhinology. 1994;32(4):208-10. DelGaudio JM. Endoscopic transnasal approach to the pterygopalatine fossa. Arch Otolaryngol Head Neck Surg. 2003;129(4):441-6, http://dx. doi.org/10.1001/archotol.129.4.441. Hofstetter CP, Singh A, Anand VK, Kacker A, Schwartz TH. The endoscopic, endonasal, transmaxillary transpterygoid approach to the pterygopalatine fossa, infratemporal fossa, petrous apex, and the Meckel cave. J Neurosurg. 2010;113(5):967-74, http://dx.doi.org/10.3171/2009. 10.JNS09157. Battaglia P, Turri-Zanoni M, Lepera D, Sica E, Karligkiotis A, Dallan I, et al. Endoscopic transnasal approaches to pterygopalatine fossa tumors. Head Neck. 2016;38(Suppl 1):E214-20, http://dx.doi.org/10.1002/hed. 23972. Battaglia P, Turri–Zanoni M, Dallan I, Gallo S, Sica E, Padoan G, et al. Endoscopic endonasal transpterygoid transmaxillary approach to the infratemporal and upper parapharyngeal tumors. Otolaryngol Head Neck Surg. 2014;150(4):696-702, http://dx.doi.org/10.1177/0194599813 520290. Castelnuovo P, Turri–Zanoni M, Battaglia P, Bignami M, Bolzoni Villaret A, Nicolai P. Endoscopic endonasal approaches for malignant tumours involving the skull base. Curr Otorhinolaryngol Rep. 2013;1:197-205, http://dx.doi.org/10.1007/s40136-013-0028-3. Lane AP, Bolger WE. Endoscopic transmaxillary biopsy of pterygopalatine space masses: a preliminary report. Am J Rhinol. 2002;16(2): 109-12. Jian XC, Liu JP. A new surgical approach to extensive tumors in the pterygomaxillary fossa and the skull base. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2003;95(2):156-62, http://dx.doi.org/10.1067/ moe.2003.41. Oakley GM, Harvey RJ. Endoscopic resection of pterygopalatine fossa and infratemporal malignancies. Otolaryngol Clin North Am. 2017; 50(2):301-13, http://dx.doi.org/10.1016/j.otc.2016.12.007. Abuzayed B, Tanriover N, Gazioglu N, Cetin G, Akar Z. Extended endoscopic endonasal approach to the pterygopalatine fossa: anatomic study. J Neurosurg Sci. 2009;53(2):37-44. Tashi S, Purohit BS, Becker M, Mundada P. The pterygopalatine fossa: imaging anatomy, communications, and pathology revisited. Insights Imaging. 2016;7(4):589-99, http://dx.doi.org/10.1007/s13244-016-0498-1.

CLINICAL SCIENCE

The value of high-flow nasal cannula oxygen therapy after extubation in patients with acute respiratory failure Hong-Zhuan Song,I,# Juan-Xian Gu,I,# Hui-Qing Xiu,II Wei Cui,II Gen-Sheng ZhangII,* I Department of Critical Care Medicine, Haining People’s Hospital, Zhejiang, China. II Department of Critical Care Medicine, Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China.

OBJECTIVE: To investigate the value of high-flow nasal cannula oxygen therapy after extubation in patients with acute respiratory failure. METHODS: A single-center, prospective, randomized, controlled pilot trial was conducted between January 2013 and December 2014. Sixty enrolled patients were randomized immediately after extubation into either a high-flow nasal cannula group (n=30) or an air entrainment mask group (n=30) at a fixed inspired oxygen fraction (40%). The success rate of oxygen therapy, respiratory and hemodynamic parameters and subjective discomfort (using a visual analogue scale) were assessed at 24h after extubation. RESULTS: The two groups were comparable at extubation. A total of 46 patients were successfully treated including 27 patients in the high-flow nasal cannula group and 19 patients in the air entrainment mask group. Compared to the air entrainment mask group, the success rate of oxygen therapy and the partial pressure of arterial oxygen were significantly higher and the respiratory rate was lower in the high-flow nasal cannula group. In addition, less discomfort related to interface displacement and airway dryness was observed in the high-flow nasal cannula group than in the air entrainment mask group. CONCLUSIONS: At a fixed inspired oxygen fraction, the application of a high-flow nasal cannula after extubation achieves a higher success rate of oxygen therapy and less discomfort at 24h than an air entrainment mask in patients with acute respiratory failure. KEYWORDS: Oxygen Therapy; Acute Respiratory Failure; Extubation; Noninvasive Ventilation; High-Flow Nasal Cannula. Song HZ, Gu JX, Xiu HQ, Cui W, Zhang GS. The value of high-flow nasal cannula oxygen therapy after extubation in patients with acute respiratory failure. Clinics. 2017;72(9):562-567 Received for publication on March 10, 2017; First review completed on April 21, 2017; Accepted for publication on July 13, 2017 *Corresponding author. E-mail: genshengzhang@zju.edu.cn #

Contributed equally to this work.

’ INTRODUCTION

Oxygen therapy is crucial to maintain the oxygen demand and to prevent the recurrence of ARF. A variety of conventional oxygen therapy (COT) systems that are classified as either fixed or variable performance are available, with various limitations (4). The air entrainment mask, as a fixed type, can increase dryness of the respiratory tract, influence daily patient activity, and cause patient discomfort. In addition, a nasal cannula cannot provide a constant oxygen concentration. Meanwhile, noninvasive ventilation (NIV) might not be suitable for all extubated patients due to their poor tolerance and cooperation (5). Recently, the high-flow nasal cannula (HFNC) oxygen system (Optiflow, Fisher & Paykel Healthcare, Auckland, New Zealand), delivering heated and humidified highflow (up to 60 L/min) gas, was introduced and has become widely used in clinical practice (6). It has several advantages, such as efficient humidification and airway mucociliary clearance, accurate fraction of inspired oxygen (FiO2) delivery and a low-level positive airway pressure (7). The concept of HFNC initially originated from the treatment of

Acute respiratory failure (ARF) is the most common cause of admission to the intensive care unit (ICU) and often requires endotracheal intubation and mechanical ventilation (1). After improvements, these ventilated ARF patients should be considered for weaning and extubation. However, extubation remains a challenge in the critical care field, as a high reintubation rate exists (19%) (2). Patients are still pathophysiologically unstable after extubation, with symptoms such as incomplete recovery of primary diseases, oxygen deficit, upper airway obstruction, excess respiratory secretions, inadequate cough, or respiratory muscle weakness (3). Copyright & 2017 CLINICS – This is an Open Access article distributed under the terms of the Creative Commons License (http://creativecommons.org/licenses/by/ 4.0/) which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is properly cited. No potential conflict of interest was reported. DOI: 10.6061/clinics/2017(09)07

562

High-flow nasal cannula oxygen after extubation Song et al.

CLINICS 2017;72(9):562-567

number sequence. In both groups, the FiO2 was set at 40%. The flow level of the HFNC began at 60 L/min and was adjusted downward in 5- to 10-L/min decrements as the target oxygenation improved or stabilized, while the flow rate of the air entrainment mask was set at 10 L/min. The target oxygenation in our study was a pulse oxygen saturation (SpO2) of 94–98% for most patients (hypoxic respiratory failure) or 88–92% for those with hypercapnic respiratory failure. Demographic data such as age, gender, and Acute Physiology and Chronic Health Evaluation II (APACHE-II) score were recorded at baseline. Electrocardiography variables, heart rate, blood pressure, mean arterial pressure, respiratory rate, SpO2, and arterial blood gas values were monitored at baseline and 24h after extubation. A visual analog scale (20) [from 0 (no discomfort) to 10 (maximum discomfort)] was used to assess patient discomfort related to airway dryness or the interface. In addition, adverse events, including replacement of the original oxygen system (shifting to an HFNC from the air entrainment mask in this study) and requiring NIV or endotracheal intubation, were also recorded. Decisions to apply NIV or endotracheal intubation were based on previous criteria (21,22). The ventilator mode of NIV was set to pressure support mode or pressure controlled mode. The criteria for shifting to an HFNC from the air entrainment mask were based on a previous study (23) and were modified as follows: respiratory rate 430 breaths/min, SpO2 o90%, or intolerance to the air entrainment mask. The primary objective of this study was to evaluate the success rate of oxygen therapy at 24h after extubation. In the present study, successful oxygen therapy was defined as not requiring a replacement oxygen device, NIV, or reintubation within 24h after extubation. The secondary objectives of the study were to investigate respiratory variables, hemodynamic variables, and patient discomfort at 24h after extubation. The results of this study were presented as absolute numbers and percentages or means and standard deviations for continuous data if normally distributed and medians and ranges if not normally distributed. Comparisons between the two groups were performed using a t test or Mann-Whitney U test for metric data and the chi-square test for categorical data. A two-sided p-value of less than 0.05 was considered significant. IBM SPSS Statistics for Windows software (IBM Corp. Released 2011. Version 20.0. Armonk, NY, USA) was used for data analysis.

preterm infants as an alternative to nasal continuous positive airway pressure (CPAP) (8). The feasibility, efficacy, and tolerance of the HFNC have been tested in neonatal and pediatric care (9). More recently, the HFNC has been widely used in adults with hypoxemic ARF of different etiologies (10), in do-not-intubate patients (11), after cardiac surgery (12), in invasive practices such as bronchoscopy (13) and in other applications. Compared to COT, the HFNC can enhance patient comfort and tolerance, improve oxygenation, and reduce the reintubation rate (14,15). However, in some cases, the HFNC did not improve oxygenation or reduce the need for escalation of respiratory support (16). Thus, the exact efficacy of HFNC is still controversial or uncertain. Two recent randomized clinical trials have shown that use of an HFNC after extubation can decrease the need for reintubation compared with COT (15) and is not inferior to NIV for preventing reintubation and postextubation respiratory failure (17). However, in these two studies, the FiO2 was adjustable, and the studies included patients who were at either low risk (15) or high risk for reintubation (17). Given that the HFNC has the intrinsic advantages of efficient humidification and dynamic positive airway pressure (7), we hypothesized that at a fixed FiO2 and use of an HFNC after extubation could achieve a higher success rate of oxygen therapy than an air entrainment mask in ARF patients.

’ MATERIALS AND METHODS This study was approved by the Ethics Committee of Haining People’s Hospital, Zhejiang, China. Written informed consent was obtained from included patients or their surrogates. Mechanically ventilated ARF patients who were admitted to the 24-bed adult comprehensive ICU at Haining People’s Hospital between January 2013 and December 2014 were prospectively included. The criteria for ARF were based on the conventional definition: partial pressure of arterial oxygen (PaO2) o60 mmHg, partial pressure of arterial carbon dioxide (PaCO2) 445 mmHg, or both (18). Patients were included in this study if they had undergone mechanical ventilation for at least 48h and were ready for tracheal extubation after clinical weaning assessments, according to the international consensus conference on weaning (19); the criteria for weaning assessments in this study included evidence of clinical improvement of the original pathologic process leading to ARF, relative cardiovascular stability with (at most) a minimum requirement for vasopressors, adequate mentation, efforts at spontaneous ventilation, and adequate oxygenation (defined as a PaO2/FiO2 of at least 150 mmHg with FiO2p0.4 and PEEPp8 cmH2O). Patients were excluded from this study if they had poor cooperation, a tracheostomy, or a decreased level of consciousness (Glasgow Coma Scale score of 12 points or less); were younger than 18 years old or pregnant; or did not sign the informed consent form. The patients who met the weaning criteria and successfully passed the spontaneous breathing trial with 7 cmH2O of pressure support for 30 to 120 min were eligible and ready for tracheal extubation. The baseline time point was defined as the end of the spontaneous breathing test and immediately before extubation. Immediately after extubation, the patients were randomized into two groups: oxygen treatment by HFNC (PT101AZ, Fisher & Paykel Healthcare, Auckland, New Zealand) or air entrainment mask (Jinlin Medical Appliances Factory, Hangzhou, China). Randomization was accomplished using a computer-generated random

’ RESULTS Between January 2013 and December 2014, 68 patients met the inclusion criteria and were eligible to participate in this study (flow chart in Figure 1). Eight patients were excluded from the study: three patients declined informed consent, two patients removed invasive arterial monitoring catheters during the study period, and three other patients were transferred out of the ICU on the day of extubation. Eventually, 30 patients were included in the HFNC group, and 30 patients were included in the air entrainment mask group. The demographic and clinical characteristics of the two groups at baseline were comparable, as shown in Table 1. Among the 60 patients, 46 were successfully treated by initial oxygen therapy within 24h after extubation, including 27 in the HFNC group and 19 in the air entrainment mask group. The success rate of oxygen therapy by HFNC (27/30, 90%) was significantly higher than that by the air entrainment mask (19/30, 63.3%) (p=0.01) (Table 2). Of the 11 patients

563

High-flow nasal cannula oxygen after extubation Song et al.

CLINICS 2017;72(9):562-567

Figure 1 - Flow diagram of participants enrolled in the study and associated clinical outcomes after extubation.ICU: Intensive care unit; HFNC: high-flow nasal cannula; NIV: noninvasive ventilation; IV: invasive ventilation.

Table 1 - Baseline patient characteristics.

Male, n (%) Age, years APACHE-II Causes of ARF, n (%) Pneumonia COPD exacerbation Cardiogenic pulmonary edema Multiple trauma Cardiac arrest Others* Length of MV before inclusion (days) Heart rate (beats/min) Mean arterial pressure (mmHg) PaCO2 (mmHg) PaO2 (mmHg) SpO2 (%)

HFNC (n=30)

Air entrainment mask (n=30)

116 (53.3) 66±14 12.87±3.0

18 (60) 71±13 12.36±3.29

12 (40) 7 (23.3) 5 (16.7) 1 (3.3) 2 (6.7) 3 (10) 5.5±3.4 82.80±9.85 83±9 41.5±6.7 82.8±11.0 96.2±2.3

13 (43.3) 6 (20) 6 (20) 2 (6.7) 1 (3.3) 2 (6.7) 5.4±2.8 81.53±8.92 83±10 42.3±7.1 81.7±11.6 95.1±2.9

* Others included psychiatric drug poisoning (n=1), severe acidosis (n=3), and epileptic seizures (n=1). HFNC: High-flow nasal cannula; APACHE-II: Acute Physiology and Chronic Health Evaluation II score; ARF: Acute respiratory failure; COPD: Chronic obstructive pulmonary disease; MV: Mechanical ventilation.

who failed treatment with the air entrainment mask, 3 received NIV, 3 required reintubation, and 5 were shifted to the HFNC (these patients were all successfully treated by HFNC). Although no significant difference was found, the rate of ventilator support (3/30, 10%) in the HFNC group was lower than that in the air entrainment mask group (6/30, 20%). A total of four patients required reintubation: one in the HFNC group (1/30, 3.33%) and three in the air entrainment mask group (3/30, 10%); the difference was not statistically significant (Table 2).

At 24 h after extubation, the average flow rate in the HFNC group was 36.8±2.8 L/min. In addition, the HFNC significantly improved the PaO2 and SpO2 at 24h after extubation compared to the air entrainment mask (83.2± 10.5 mmHg vs. 74.5±13.1 mmHg, p=0.016; 98.0±1.3% vs. 96.9±1.4%, p=0.011, respectively). The PaCO2 values were similar between the two groups (p=0.591, Table 2). The respiratory rate was significantly lower for the HFNC group than for the air entrainment mask group (22±3.6 breaths/min vs. 26±4.3 breaths/min, p=0.003). No significant differences in

564

High-flow nasal cannula oxygen after extubation Song et al.

CLINICS 2017;72(9):562-567

Table 2 - Primary and secondary outcomes in the two treatment groups.

Primary outcomes Success rate of oxygen therapy NIV Endotracheal intubation Replacement of oxygen device Secondary outcomes PaO2 (mmHg) SpO2 (%) PaCO2 (mmHg) Respiratory rate (breaths/min) Heart rate (beats/min) Mean arterial pressure (mmHg)

HFNC

Air entrainment mask

n=30 27/30 (90%) 2 (6.67%) 1 (3.33%) 0 (0%) n=27 83.2±10.5 98.0±1.3 41.4±6.5 22±4 85±9 85±8

n=30 19/30 (63.33%) 3 (10%) 3 (10%) 5 (16.67%) n=19 74.5±13.1 96.9±1.4 42.2±13.1 26±4 87±14 86±9

p 0.012 0.639 0.290 0.062 0.016 0.011 0.591 0.003 0.598 0.824

HFNC: High-flow nasal cannula; NIV: Noninvasive ventilation.

namely, that HFNC was not inferior to NIV in terms of reintubation or postextubation failure. The HFNC has been increasingly used in patients after extubation, although its efficacy remains unclear. Tiruvoipati et al. (24) showed that the HFNC was better tolerated than a high-flow face mask in 50 extubated patients, despite a similar effectiveness of oxygen delivery. In 2014, Rittayamai et al. (25) found that the HFNC improved dyspnea and the respiratory and heart rates compared to a non-rebreathing mask in 17 extubated patients. However, these two studies observed only the short-term efficacy of the HFNC (24,25). The relative long-term efficacy of the HFNC in extubated patients was investigated in two recent well-designed studies (14,15). Maggiore et al. (14) found that HFNC oxygen therapy resulted in significantly better oxygenation, better comfort, and a lower reintubation rate (3.8%) during the 48-h study period than COT in 105 critically ill patients. Similarly, Hernández et al. (15) also showed that HFNC application resulted in a significantly lower reintubation rate (4.9%) than COT (12.2%) within 72 h in 527 adult patients at low risk for reintubation. However, the influence of the FiO2 on the efficacy of the HFNC for oxygen therapy was not excluded in these two studies. Our study extended the application of the HFNC to use in patients after extubation, even for those with a fixed FiO2. The exact mechanism of the beneficial effects observed from the application of the HFNC remains unclear. The following factors should be considered. (a) The high-flow oxygen (up to 60 L/min) delivered by the HFNC meets or exceeds the patient’s peak inspiratory demand, which might deliver a more accurate FiO2 with a fixed FiO2. (b) The delivery of high-flow oxygen flushes the anatomical dead space of the upper airway, creating a reservoir of fresh gas available for every breath to minimize re-breathing of CO2, which would improve the efficiency of ventilation and oxygen delivery. The PaCO2 level at 24 h after oxygen therapy was similar between the two groups, which suggested that the HFNC had no significant effect on the PaCO2 values in our study. (c) In contrast to COT, the HFNC system can generate positive airway pressure to increase airway compliance and reduce the breathing work (26). (d) The sufficiently heated humidified air produced by the HFNC facilitates secretion clearance and decreases bronchial hyperresponse symptoms (27). (e) Better subjective comfort generally results in better subject compliance and a better outcome (28). Our study confirmed that HFNC oxygen therapy led to reduced patient discomfort compared to the

Figure 2 - High-flow nasal cannula (HFNC) oxygen therapy improved the discomforts related to the interface and symptoms of airway dryness compared to the air entrainment mask. Patients’ discomforts were assessed by scores related to the interface (panel A) or airway dryness (panel B) via a visual analog scale [from 0 (no discomfort) to 10 (maximum discomfort)] (*po0.001 compared to the air entrainment mask).

heart rate and mean arterial pressure were observed between the two groups (Table 2). Discomforts related to the interface and symptoms of airway dryness were significantly lower in the HFNC group than those in the air entrainment mask group [3(3-4.5) vs. 7(6-8); 3(2-3.5) vs. 5(4.7-6), both po0.001] (Figure 2).

’ DISCUSSION The main findings of this study were as follows: after extubation of mechanically ventilated ARF patients, the application of an HFNC achieved a higher success rate of oxygen therapy within 24 h than the air entrainment mask. In addition, use of the HFNC resulted in improved oxygenation, a decreased respiratory rate, and reduced patient discomfort compared to the air entrainment mask. After failure with the air entrainment mask, five patients were shifted to the HFNC group and ultimately achieved success with oxygen therapy, thus avoiding the introduction of NIV or reintubation. Consistent with other studies (14,15), our study also confirmed that the HFNC is a promising oxygen therapy device for patients with ARF after extubation. Moreover, this procedure could even be used as an alternative after the failure of other oxygen devices to reduce the need for NIV or reintubation; these findings indirectly supported the conclusion obtained by Hernández et al. (17),

565

High-flow nasal cannula oxygen after extubation Song et al.

CLINICS 2017;72(9):562-567

the benefit of HFNC oxygen treatment were unrelated to the amount of oxygen, and we used a fixed FiO2. In summary, HFNC oxygen therapy after extubation in mechanically ventilated ARF patients can achieve a higher success rate of oxygen therapy, improved oxygenation, and a lower occurrence of discomfort than an air entrainment mask. Thus, HFNC oxygen therapy might be a promising treatment for ARF patients after extubation.

air entrainment mask. We also showed that the rate of ventilatory support (NIV or reintubation) in the air entrainment mask group tended to be higher than that in the HFNC group, and this trend might be associated with the mismatch between the oxygen flow and the patient’s inspiratory demand, the patients’ intolerance to the mask, the instability of the delivered FiO2, and/or insufficiency of heating and humidification (4,7). An oxygen target (a SpO2 of 94–98% for most patients) and a fixed FiO2 of 40% were set in the present study, which could raise the question of how to improve oxygenation. Indeed, we performed other measures, such as maintenance of a high flow rate and/or sputum removal; NIV and reintubation were also considered if necessary. Most patients in the current study met the target oxygenation at a fixed FiO2 of 40%, and only 6.67% (4/60) required reintubation, mainly due to respiratory muscle weakness, cardiac failure, excess respiratory secretions, and/or changes in mental status. Our study has several limitations. (a) The small number of patients and the pilot study design resulted in a relatively low statistical power. Based on a superior design according to the acute sample size using SAS software (Cary, North Carolina USA), the sample size of 30 in each group would have a power of 0.807, with a level of significance of 0.05 (the success rates of oxygen therapy were 90% and 63% in the HFNC and air entrainment mask groups, respectively) (Table 2). (b) In this study, we considered the 5 patients with successful shifts to the HFNC oxygen treatment as treatment failures to their initial oxygen therapy by air entrainment mask (Figure 1). To some extent, it might be reasonable to crossover these 5 patients because this finding further suggests that HFNC oxygen treatment might be superior to use of an air entrainment mask. As this study was open label by nature, it was difficult to blind participants and clinicians to the allocated oxygen therapy, which is a problem inherent to many studies of medical devices. This problem was relatively weakened by the randomization techniques and the comparability of the groups at baseline in the current study. (c) The APACHE-II scores in both groups were approximately 12; thus, the severity of the population in this study was relatively minor. Whether HFNC therapy is suitable for more serious patients with respiratory failure after extubation requires further investigation. (d) Although a fixed FiO2 setting was used in both groups in our study, the actual FiO2 delivered to the patient might be inconsistent with the set FiO2 in some cases. For example, the actual FiO2 delivered to the patient could be lower than 40% with both the air entrainment mask (in case of a higher respiratory rate or tidal volume) and the HFNC (when lower flow rates were used while the patient’s peak inspiratory flow was higher). The use of greater flow rates with the HFNC could allow more stable delivery of the FiO2. Unfortunately, we did not measure the actual FiO2 delivered to the patient. (e) The potential beneficial effects of the HFNC on sputum clearance and bronchial hyper-response symptoms were not assessed, which might also contribute to the higher success rate of initial oxygen treatment by improving mucociliary function and reducing the work of breathing. (f) The assessment of patient discomfort was subjective. However, this method is widely used in the measurement of breathlessness and other symptoms (20). (g) Indeed, we also clinically titrated the amount of oxygen according to the patients’ needs, but not with a fixed FiO2. However, the aim of the current study was to investigate whether the potential mechanisms underlying

’ ACKNOWLEDGMENTS This work was supported in part by grants from the National Natural Science Foundation of China (81570017, GS Zhang), the Medical and Health Research Program of Zhejiang Province (Core Talents Plan) (No. 2016RCA014, GS Zhang) and the Medical and Health Research Program of Zhejiang Province (2017KY371, GS Zhang).

’ AUTHOR CONTRIBUTIONS Song HZ, Gu JX and Zhang GS designed the study. Song HZ, Gu JX and Xiu HQ were responsible for the patient care, technical advice, data acquisition, statistical analysis, interpretation of the data and manuscript writing.Zhang GS and Cui W revised the manuscript. All authors read and approved the ﬁnal version of the manuscript.

’ REFERENCES 1. Vincent JL, Akc¸a S, De Mendonc¸a A, Haji-Michael P, Sprung C, Moreno R, et al. The epidemiology of acute respiratory failure in critically ill patients. Chest. 2002;121(5):1602-9, http://dx.doi.org/10.1378/chest.121.5.1602. 2. Miu T, Joffe AM, Yanez ND, Khandelwal N, Dagal AH, Deem S, et al. Predictors of reintubation in critically ill patients. Respir Care. 2014; 59(2):178-85, http://dx.doi.org/10.4187/respcare.02527. 3. Rothaar RC, Epstein SK. Extubation failure: magnitude of the problem, impact on outcomes, and prevention. Curr Opin Crit Care. 2003;9(1): 59-66, http://dx.doi.org/10.1097/00075198-200302000-00011. 4. Williams AB, Jones PL, Mapleson WW. A comparison of oxygen therapy devices used in the postoperative recovery period. Anaesthesia.1988; 43(2):131-5, http://dx.doi.org/10.1111/j.1365-2044.1988.tb05483.x. 5. Esteban A, Frutos-Vivar F, Ferguson ND, Arabi Y, Apezteguía C, González M, et al. Noninvasive positive-pressure ventilation for respiratory failure after extubation. N Engl J Med. 2004;350(24):2452-60, http:// dx.doi.org/10.1056/NEJMoa032736. 6. Ward JJ. High-flow oxygen administration by nasal cannula for adult and perinatal patients. Respir Care. 2013;58(1):98-122, http://dx.doi.org/ 10.4187/respcare.01941. 7. Spoletini G, Alotaibi M, Blasi F, Hill NS. Heated Humidified High-Flow Nasal Oxygen in Adults: Mechanisms of Action and Clinical Implications. Chest. 2015;148(1):253-61, http://dx.doi.org/10.1378/chest.14-2871. 8. Manley BJ, Owen LS, Doyle LW, Andersen CC, Cartwright DW, Pritchard MA, et al. High-flow nasal cannulae in very preterm infants after extubation. N Engl J Med. 2013;369(15):1425-33, http://dx.doi.org/10.1056/ NEJMoa1300071. 9. Manley BJ, Dold SK, Davis PG, Roehr CC. High-flow nasal cannulae for respiratory support of preterm infants: a review of the evidence. Neonatology. 2012;102(4):300-8, http://dx.doi.org/10.1159/000341754. 10. Frat JP, Thille AW, Mercat A, Girault C, Ragot S, Perbet S, et al. High-Flow oxygen through nasal cannula in acute hypoxemic respiratory failure. N Engl J Med. 2015;372(23):2185-96, http://dx.doi.org/10.1056/NEJMoa 1503326. 11. Peters SG, Holets SR, Gay PC. High-flow nasal cannula therapy in do-notintubate patients with hypoxemic respiratory distress. Respir Care. 2013;58(4):597-600, http://dx.doi.org/10.4187/respcare.01887. 12. Parke R, McGuinness S, Dixon R, Jull A. Open-label, phase II study of routine high-flow nasal oxygen therapy in cardiac surgical patients. Br J Anaesth. 2013;111(6):925-31, http://dx.doi.org/10.1093/bja/aet262. 13. Simon M, Braune S, Frings D, Wiontzek AK, Klose H, Kluge S. High-flow nasal cannula oxygen versus non-invasive ventilation in patients with acute hypoxaemic respiratory failure undergoing flexible bronchoscopy- a prospective randomised trial. Crit Care. 2014;18(6):712, http://dx.doi. org/10.1186/s13054-014-0712-9. 14. Maggiore SM, Idone FA, Vaschetto R, Festa R, Cataldo A, Antonicelli F, et al. Nasal high-flow versus Venturi mask oxygen therapy after extubation. Effects on oxygenation, comfort, and clinical outcome. Am J Respir Crit Care Med. 2014;190(3):282-8, http://dx.doi.org/10.1164/rccm. 201402-0364OC.

566

High-flow nasal cannula oxygen after extubation Song et al.

CLINICS 2017;72(9):562-567

15. Hernรกndez G, Vaquero C, Gonzรกlez P, Subira C, Frutos-Vivar F, Rialp G, et al. Effect of Postextubation High-Flow Nasal Cannula vs Conventional Oxygen Therapy on Reintubation in Low-Risk Patients: A Randomized Clinical Trial. JAMA. 2016;315(13):1354-61, http://dx.doi.org/10.1001/ jama.2016.2711. 16. Corley A, Bull T, Spooner AJ, Barnett AG, Fraser JF. Direct extubation onto high-flow nasal cannulae post-cardiac surgery versus standard treatment in patients with a BMI X30: a randomised controlled trial. Intensive Care Med. 2015;41(5):887-94, http://dx.doi.org/10.1007/s00134-0153765-6. 17. Hernรกndez G, Vaquero C, Colinas L, Cuena R, Gonzรกlez P, Canabal A, et al. Effect of Postextubation High-Flow Nasal Cannula vs Noninvasive Ventilation on Reintubation and Postextubation Respiratory Failure in High-Risk Patients: A Randomized Clinical Trial. JAMA. 2016;316(15): 1565-74, http://dx.doi.org/10.1001/jama.2016.14194. 18. Roussos C, Koutsoukou A. Respiratory failure. Eur Respir J Suppl. 2003;47: 3s-14s, http://dx.doi.org/10.1183/09031936.03.00038503. 19. Boles JM, Bion J, Connors A, Herridge M, Marsh B, Melot C, et al. Weaning from mechanical ventilation. Eur Respir J. 2007;29(5):1033-56, http://dx.doi.org/10.1183/09031936.00010206. 20. Bausewein C, Farquhar M, Booth S, Gysels M, Higginson IJ. Measurement of breathlessness in advanced disease: a systematic review. Respir Med. 2007;101(3):399-410, http://dx.doi.org/10.1016/j.rmed. 2006.07.003. 21. Ferrer M, Valencia M, Nicolas JM, Bernadich O, Badia JR, Torres A. Early noninvasive ventilation averts extubation failure in patients at risk:

22.

23. 24. 25.

26. 27.

28.

567

a randomized trial. Am J Respir Crit Care Med. 2006;173(2):164-70, http://dx.doi.org/10.1164/rccm.200505-718OC. Davidson AC, Banham S, Elliott M, Kennedy D, Gelder C, Glossop A, et al. British Thoracic Society/Intensive Care Society Guideline for the ventilator management of acute hypercapnic respiratory failure in adults. BMJ Open Resp Res. 2016;3:e000133, http://dx.doi.org/10.1136/bmjresp2016-000133. Sotello D, Rivas M, Mulkey Z, Nugent K. High-Flow nasal cannula oxygen in adult patients: a narrative review. Am J Med Sci. 2015;349(2): 179-85, http://dx.doi.org/10.1097/MAJ.0000000000000345. Tiruvoipati R, Lewis D, Haji K, Botha J. High-flow nasal oxygen vs high-flow face mask: a randomized crossover trial in extubated patients. J Crit Care. 2010;25(3):463-8, http://dx.doi.org/10.1016/j.jcrc.2009.06.050. Rittayamai N, Tscheikuna J, Rujiwit P. High-flow nasal cannula versus conventional oxygen therapy after endotracheal extubation: a randomized crossover physiologic study. Respir Care. 2014;59(4):485-90, http://dx.doi.org/10.4187/respcare.02397. Groves N, Tobin A. High flow nasal oxygen generates positive airway pressure in adult volunteers. Aust Crit Care. 2007;20(4):126-31, http://dx. doi.org/10.1016/j.aucc.2007.08.001. Hasani A, Chapman TH, McCool D, Smith RE, Dilworth JP, Agnew JE. Domiciliary humidification improves lung mucociliary clearance in patients with bronchiectasis. Chron Respir Dis. 2008;5(2): 81-6, http://dx. doi.org/10.1177/1479972307087190. Roca O, Riera J, Torres F, Masclans JR. High-flow oxygen therapy in acute respiratory failure. Respir Care. 2010;55(4): 408-13.

CLINICAL SCIENCE

Caring for critically ill patients outside intensive care units due to full units: a cohort study Fabiane Urizzi,I Marcos T. Tanita,II Josiane Festti,III Lucienne T.Q. Cardoso,III Tiemi Matsuo,IV Cintia M.C. GrionIII,* I Pos-graduacao, Hospital Universitario, Universidade Estadual de Londrina, Londrina, PR, BR. II Unidade de Terapia Intensiva Adulto, Hospital Universitario, Universidade Estadual de Londrina, Londrina, PR, BR. III Departamento de Medicina Interna, Hospital Universitario, Universidade Estadual de Londrina, Londrina, PR, BR. IV Departamento de Estatistica, Universidade Estadual de Londrina, Londrina, PR, BR.

OBJECTIVES: This study sought to analyze the clinical and epidemiologic characteristics of critically ill patients who were denied intensive care unit admission due to the unavailability of beds and to estimate the direct costs of treatment. METHODS: A prospective cohort study was performed with critically ill patients treated in a university hospital. All consecutive patients denied intensive care unit beds due to a full unit from February 2012 to February 2013 were included. The data collected included clinical data, calculation of costs, prognostic scores, and outcomes. The patients were followed for data collection until intensive care unit admission or cancellation of the request for the intensive care unit bed. Vital status at hospital discharge was noted, and patients were classified as survivors or non-survivors considering this endpoint. RESULTS: Four hundred and fifty-four patients were analyzed. Patients were predominantly male (54.6%), and the median age was 62 (interquartile range (ITQ): 47 - 73) years. The median APACHE II score was 22.5 (ITQ: 16 - 29). Invasive mechanical ventilation was used in 298 patients (65.6%), and vasoactive drugs were used in 44.9% of patients. The median time of follow-up was 3 days (ITQ: 2 - 6); after this time, 204 patients were admitted to the intensive care unit and 250 had the intensive care unit bed request canceled. The median total cost per patient was US$ 5,945.98. CONCLUSIONS: Patients presented a high severity in terms of disease scores, had multiple organ dysfunction and needed multiple invasive therapeutic interventions. The study patients received intensive care with specialized consultation during their stay in the hospital wards and presented high costs of treatment. KEYWORDS: Intensive Care Units; Bed Occupancy; Health Services Accessibility; Direct Service Costs; Costs and Cost Analysis. Urizzi F, Tanita MT, Festti J, Cardoso LT, Matsuo T, Grion CM. Caring for critically ill patients outside intensive care units due to full units: a cohort study. CLINICS. 2017;72(9):568-574 Received for publication on March 17, 2017; First review completed on June 10, 2017; Accepted for publication on July 13, 2017 *Corresponding author. E-mail: cintiagrion@hotmail.com

’ INTRODUCTION

may be rejected from the ICU. Some patients, however, may be denied admission to the ICU due to a lack of available beds (2), and this delay in admission has been associated with increased mortality (3). The concept of rapid and early care has been established in various fields of medicine (4-6). Considering that most acute illnesses develop in stages of physiological and organ dysfunction, the logical step would certainly be to provide specialized care for any critically ill patient within the hospital, regardless of the location. This approach has been described as an "intensive care system without walls’’ (7). In an attempt to provide intensive care outside the ICU environment, rapid response teams (RRTs) have been created with the goal of early identification of the signs and symptoms of physiological worsening in patients, thereby reducing the risk of adverse events in inpatient units. This strategy consists of a bedside intensive therapy system formed usually by a doctor, nurse and physiotherapist (8). The benefits that this system have provided to hospitalized

The intensive care unit (ICU) provides a continuous monitoring system for critically ill patients who have the potential for recovery or are in a life-threatening situation. In recent decades, this sector has observed a significant increase in the demand for beds associated with the reduced mortality among patients admitted. Additionally, an increase in the complexity of diseases and number of chronic health conditions has been reported (1). A patient who may not benefit from treatment due to either a very good or very poor prognosis

568

Caring for patients outside intensive care units Urizzi F et al.

CLINICS 2017;72(9):568-574

condition, presence of comorbidities, length of stay before ICU admission, data for the calculation of costs and prognostic scores and outcomes. The source of the data was the patient’s records, which were consulted during their stay in the hospital, and the data were then transcribed using instruments designed to carry out the present research. Each patient was followed until one of the following primary outcomes occurred: ICU admission, cancellation of the request for the ICU due to clinical improvement or limitation of therapeutic support, transfer to another hospital or death. Data collection began on the day of refusal of admission to the ICU and continued until a primary outcome occurred. Patients were classified as survivors and non-survivors considering their vital status at hospital discharge. The Acute Physiology and Chronic Health Evaluation (APACHE II), Sequential Organ Failure Assessment (SOFA) and Therapeutic Intervention Scoring System (TISS-28) scores were collected. The APACHE II score was calculated to characterize the severity of the study population. The SOFA score was used to observe variations in organ dysfunction/ failure. The TISS-28 score was used to quantify therapeutic interventions carried out during treatment outside the ICU. Each of these scores was calculated according to their original descriptions (17-19). In an interim analysis, a smaller sample of patients estimated to be representative was used to perform a cost analysis. For this purpose, data were collected from patients included in the study period from February to July 2012. The model adopted to collect costs applied the "bottom-up" approach, a methodology considered the gold standard, in which the goal is to estimate costs by individual patient or by a group of patients (20). The measurement of direct costs is then generated, and the sum of the costs can provide a conservative estimate of the true value of treating a health problem (21). For the present study, the direct costs generated for the treatment of patients were divided into four categories: clinical support, consumer items, human resources and hospital fees.

patients have been described by several authors (9-15). Most studies have adopted ‘‘before and after’’ quasi-experimental designs, demonstrating a reduction in adverse events in hospitalized patients and hospital mortality. However, there are few studies describing the care costs of critically ill patients outside of the intensive care environment. Research in this area is important for the financial planning of actions in health care. The aim of the present study was to analyze the clinical and epidemiological characteristics of critically ill patients who were denied ICU admission due to the unavailability of beds and to estimate the direct costs of treatment during this period.

’ METHODS This was a prospective cohort study of critically ill patients who were treated outside the ICU due to a full unit and attended by the RRT in a university hospital from February 2012 to February 2013. The study was conducted in adult medical-surgical admission wards at the University Hospital of the Londrina State University (HU/UEL). The HU/UEL is a supplementary service of the State University of Londrina and is characterized as a university public hospital, with 330 beds, including 20 adult ICU beds. The RRT in this hospital is composed of an intensivist physician and a physiotherapist who are assigned exclusively to the tasks of this team. The RRT responds to yellow and blue codes, assists in the care of all critically ill patients denied ICU beds and evaluates patients post-discharge from the ICU to prevent early readmissions. For a yellow or blue code, the nurse caring for the patient on the ward triggers a call for RRT consultation and participates in the care of the patient together with the two members of the RRT. During this evaluation, if there is a need to transfer the patient to the ICU, the RRT accompanies this intra-hospital transportation. Patients denied ICU beds remain under the care of the local staff with the support of the RRT. In these cases, the RRT performs two scheduled daily evaluations of these patients to assist with medical prescriptions, clinical decisions, therapeutic interventions, and checking of laboratory and other exam results. If additional evaluations are needed, the local staff calls the RRT. Triage decisions when there is a lack of a sufficient number of ICU beds are made according to the Society of Critical Care Medicine’s guidelines (16). The ICU request is performed in the electronic hospital system that provides the diagnostic and clinical data of the patient. If there is no ICU bed immediately available, the intensivist on duty with the RRT evaluates the request and classifies the patient according to the prioritization model (16). If there is more than one patient in the first level of prioritization and only one bed is available, then the length of time waiting for the ICU bed is taken into consideration, and the ‘‘first come, first served’’ rule is applied. All adult ICU beds are made available for ICU requests in the electronic system. A convenience sample was obtained from all adult patients admitted in the study period who presented with a critical condition requiring admission to a monitored ICU bed and who were refused admission due to lack of availability. Patients under the age of 18 years and those who had a waiting time for admission to the ICU of less than 24 hours were excluded. The data collected consisted of demographic data (age and sex) and clinical data, including the diagnosis of a critical

1. Clinical support: costs related to pharmacy needs (ointments, body oil, strips for verification of blood sugar and items for pressure ulcer prevention), renal support, laboratory, laboratory tests, imaging and complementary examinations. 2. Consumer items: costs of medication, nutrition, blood and its derivatives. 3. Human resources: medical procedures and physical therapists. 4. Hospital fees: considered a fixed daily rate, independent of the type of disease. Items from the clinical support category were analyzed according to the medical prescription. For the calculation of medication costs, a standard dose was considered, calculated as the mean daily prescription for a patient of 70 kg body weight. In the category of human resources, values were attributed to medical procedures performed as well as the physiotherapy service, which was divided into motor and respiratory therapy, as noted in the medical record. In the category of hospital fees, the values of the daily hospital fees and intensivist doctor on duty with the RRT were computed. Costs related to the use of equipment, infrastructure, electricity, security systems, information technology, and non-clinical support and indirect costs (lost productivity, etc.) were not analyzed.

569

Caring for patients outside intensive care units Urizzi F et al.

CLINICS 2017;72(9):568-574

221 patients excluded as follows: 60 patients were under 18 years of age, 102 patients spent less than 24 hours waiting for the ICU bed, and 59 patients were considered losses due to a lack of sufficient information in the medical records to complete the case report form. In total, 454 patients were evaluated. Among the patients studied, data collection on direct costs was performed for 151 during the period from February to July 2012. Of the 454 patients included in the study, 54.6% were male, and the median age was 62 (47-73) years. The median APACHE II severity score was 22.5 (16-29). Organ dysfunction measured by the SOFA score at study entry presented a median of 8 (4-10), and the SOFA score at the time of primary outcome was 8 (4-13). At this point, 159 patients (35%) presented a SOFA outcome lower than the SOFA score at study entry; for 53 patients (11.7%), this score did not change during the analysis, and for 242 patients (53.3%), the SOFA outcome score was higher in relation to study entry. The median score for therapeutic interventions (TISS-28) at study entry was 27 (21-30). Regarding the sector of origin, 65% of patients were cared for in the emergency room, 34.8% in the general wards and 0.2% in the operating room. According to the diagnostic categories of the APACHE II system, 92.3% of patients were considered clinical, and 7.7% were surgical. For the diagnosis at the time of requesting an ICU bed, according to the standardization of the APACHE II system, sepsis was the most frequent diagnosis (62.2%), followed by the principal "cardiovascular" system (12.8%), the principal "neurological" system (10%), postoperative conditions (3.2%) and other diagnoses (11.8%). A diagnosis of infection was present in 366 (80.6%) patients at some time during the study period, and the identified sources of infection were the lungs (76.5%), urinary system (9%), abdominal area (6%), skin and soft tissue (5.2%), and the bloodstream, surgical site or other source (3.3%). Regarding the classification of infections, 198 patients presented septic shock, 158 had sepsis, and 10 had a localized infection. In 39% of cases, the reason for ICU admission was respiratory failure, followed by hemodynamic instability (36.3%), neurological monitoring (14.5%), cardiac monitoring (7.0%) and postoperative care (2.9%). The most frequent comorbidities were hypertension (19.3%), diabetes mellitus (18.5%), congestive heart failure (14.5%), cirrhosis (8.4%), chronic renal failure (6.6%), immunosuppression (6.1%) and other (26.6%). Mechanical ventilation was required in 298 patients (65.6%) at some point while they waited for an ICU vacancy, and of the 156 patients who were on spontaneous ventilation, 91 required respiratory support with oxygen therapy. When mechanical ventilation was initiated, an intensivist physiotherapist was assigned to care for the patient and consult in cases of any difficulties regarding this intervention. The need for therapeutic interventions was associated with higher mortality (Table 1). In relation to the treatment of hemodynamic instability, 44.9% of patients used vasopressors or inotropic agents, including norepinephrine (36.1%), adrenaline (4.2%), dobutamine (3.1%) and dopamine (1.5%). On the other hand, 5.9% of patients required the use of vasodilators, including sodium nitroprusside (3.3%) and nitroglycerin (2.6%), and 14.9% of patients required the use of multiple vasoactive drugs at some point during the study period.

After data collection, a value was assigned to all items. The values were obtained from standard tables and index values for medical procedures outlined by the Brazilian Medical Association (BMA) (22). For items of hospital consumption, medications and solutions, the Brasíndice price list (23) version 799, year 2014, was used. Subsequently, the values were translated into US dollars (US$) based on the average price of currency for the year 2013. Comorbidities were defined according to the criteria published in the Charlson comorbidity index (24). The need for ICU admission was classified as one of the following: respiratory failure; hemodynamic instability; metabolic disorder, postoperative; cardiac monitoring; neurological monitoring; or other. The diagnosis of infection was based on clinical, microbiological and imaging results, and the source of infection was classified as lung, urinary tract, bloodstream, abdominal, surgical site or other. The diagnosis and classification of sepsis used the Third International Consensus Definitions for sepsis and septic shock (25). Sepsis was defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, and septic shock was defined as a subset of sepsis in which particularly profound circulatory, cellular, and metabolic abnormalities were associated with a greater risk of mortality than with sepsis alone.

Statistical analysis Data were analyzed using SPSS version 19.0 (Armonk, NY: IBM Corp.) and MedCalc version 15 (MedCalc Software, Mariakerke, Belgium); the significance level was 5%, and the confidence interval was 95%. Continuous quantitative variables were described after the normality of the distribution was verified with the Shapiro-Wilk test. For variables that presented a normal distribution, the mean and standard deviation were calculated; otherwise, the median and interquartile range (ITQ) (25th percentile 75th percentile) were calculated. The nominal categorical variables were described as absolute and relative frequencies (%) of each variable. Categorical variables were compared using Pearson’s chi-square test (w2) or Fisher’s exact test, where over 20% of the expected frequencies in the tables were lower than five. Continuous variables were compared using Student’s t-test or the Mann-Whitney test according to the data distribution. The area under the receiver operating characteristic (ROC) curve was calculated to evaluate the accuracy and compare the performance of the APACHE II, SOFA and TISS-28 to discriminate survivors and nonsurvivors. The areas under the ROC curve of the indices were compared in pairs using a non-parametric approach, based on the difference between the areas and standard error.

Ethical approval and consent to participate This study was approved by the Ethics Committee for Research Involving Human Beings, State University of Londrina/Northern Paraná Regional University Hospital, as Opinion No. 281/2010, December 17, 2010, CAAE No. 0255.0.268.000-10. The ethics committee waived the need for informed consent.

’ RESULTS During the study period, 675 critically ill patients had ICU bed requests denied and were cared for in the hospital wards with intensivist consultation by the RRT. There were

570

Caring for patients outside intensive care units Urizzi F et al.

CLINICS 2017;72(9):568-574

Table 1 - Demographic and clinical characteristics of the study patients. All Age (years) Males (%) Time before ICU (days)= Length of hospital stay (days) ICU indication Clinical (%) Surgery (%) Reason for ICU admission Respiratory failure Hemodynamic instability Neurological monitoring Cardiac monitoring Postoperative care Other APACHE II SOFA TISS-28 VM (%) Vasoactive drugs (%)

62 248 3 16

Survivors (n=154)

(47-73) (54.6) (2-6) (7-31)

419 (92.3) 35 (7.7) 177 165 66 32 13 1 22.5 8 27 298 204

(39) (36.3) (14.5) (7) (2.9) (0.2) (16-29) (4-10) (21-30) (65.6) (44.9)

56.5 69 2 14

(37-67) (44.8) (2-4) (8-28)

136 (88.3) 18 (11.7) 43 46 30 27 7 1 16 4 22 69 35

(27.9) (29.9) (19.5) (17.5) (4.5) (0.6) (9-21) (2-7) (16-28) (23.2) (17.2)

Non-survivors (n=300) 64 179 2 2

(49.5-75) (59.6) (2-3) (2-3)

283 (94.3) 17 (5.7) 134 119 36 5 6 0 25 9 29 229 169

(44.7) (39.7) (12) (1.7) (2) (0) (20-31) (6-11) (24-32) (76.8) (82.8)

P value o0.001* 0.002w 0.43* o0.001* 0.023w

o0.001w

o0.001* o0.001* o0.001* o0.001w o0.001w

Legend: ICU - intensive care unit; APACHE II - Acute Physiology and Chronic Health Evaluation; SOFA - Sequential Organ Failure Assessment; TISS-28 - Therapeutic Intervention Scoring System - 28; VM - mechanical ventilation. * Mann-Whitney test; w Fisher’s exact test; = Period between ICU request and admission or cancellation.

total costs, with a median of US$ 6,159.26 (US$ 4,038.63 - US$ 11,257.26) in non-survivors and US$ 5,015.29 (US$ 2,845.04 US$ 8,207.94) in survivors, p=0.041 (Table 3). To compare the day-by-day costs between survivors and non-survivors, the results for the first seven days of observation are presented in Figure 1. The results showed that the median daily cost was higher in non-survivors.

Of the therapeutic interventions in these patients, 51 patients required procedures that are usually performed in an ICU but needed to be carried out in the hospital wards, such as tracheal intubation, insertion of a transvenous pacemaker, hemodialysis, and insertion of a chest tube. In addition, 150 patients required intra- or inter-hospital transportation for the performance of diagnostic or therapeutic procedures. Of the 454 patients analyzed, 204 (44.9%) were admitted to the ICU after a waiting period of 3 days (2-6). Two patients (0.4%) were transferred to another hospital, and 25 patients (5.5%) had therapeutic support limited after agreement by the head doctor and family because they presented an irreversible clinical condition and therapeutic interventions were considered futile. While waiting for an ICU bed, 101 patients (22.3%) died and 122 patients (26.9%) had their ICU admission requests canceled due to clinical improvement. The demographic characteristics and prognostic scores of the group of patients whose data were collected for the calculation of direct costs were no different from those of the other patients in the study. Patients not included in the cost analysis stayed longer in the hospital, but the period of cost analysis was similar. Patients included in cost analysis also required vasoactive drugs more frequently (Table 2). The median total cost per patient was US$ 5,945.98 (US$ 3,831.98 - US$ 10,073.41), while the median daily cost was US$ 1,618.51 (US$ 1,235.29 - US$ 2,032.40). Considering the blocks of direct costs, the median total cost of laboratory tests was US$ 502.82 (US$ 255.61 - US$ 778.37); medical procedures, US$ 81.81 (US$ 0.00 - US$ 525.28); medications, US$ 4,342.51 (US$ 2,452.17 - US$ 7,713.01); and clinical support, US$ 298.71 (US$ 45.64 - US$ 551.78). In addition, the costs in accordance with outcomes were analyzed (survivors and non-survivors), and there was a higher resource consumption of medicines in patients who died (US$ 4,532.12 [US$ 2,698.91 - US$ 9,022.49]) than among those who were discharged alive (US$ 3,594.37 [US$ 1,231.11 - US$ 6,136.93]), p=0.009. The same finding was observed in relation to the

’ DISCUSSION The present study describes the care of critically ill patients outside the ICU in the hospital wards, which is becoming a common reality for hospitals around the world. Caring for these patients with daily intensivist consultations and the aid of an RRT was a local solution to increase safety for these hospitalized patients. These patients required mechanical ventilation, vasoactive drugs and invasive procedures, and they received such assistance outside of a monitored ICU bed. This situation is associated with high costs of care and possibly with an increase in adverse events. There are few studies in the literature evaluating the level of care and outcomes of critically ill patients treated outside the ICU. Previous studies (26,27) have described the benefits of transfer to an ICU bed in a period of up to three days and have shown that patients with an APACHE II score above 16 benefit the most from early admission (up to 3 days) to an ICU (27). Usually, the critically ill patient is identified and treated by the RRT and promptly transferred to an ICU to receive continued care. However, it is also possible that the RRT continues to provide critical care for patients who are denied ICU admission and remain in the hospital wards awaiting ICU bed availability. Unfortunately, this situation is becoming increasingly frequent, particularly in low- and middle-income countries. Despite the attempt to provide intensive care for patients outside the ICU in this study institution, a small proportion of the patients became refractory to treatment after hours or days of waiting for an ICU bed. It is possible that these patients’ clinical conditions

571

Caring for patients outside intensive care units Urizzi F et al.

CLINICS 2017;72(9):568-574

Table 2 - Comparison of demographic and clinical characteristics of the study patients who were or were not included in the analysis of direct costs. In cost analysis (n=151) Age (years)* Males (%) Time before ICU (days)y Length of hospital stay (days) ICU indication Clinical (%) Surgery (%) Reason for ICU admission Respiratory failure Hemodynamic instability Neurological monitoring Cardiac monitoring Postoperative care Other APACHE II* SOFA (day 1)* TISS (day 1)* VM (%) Vasoactive drugs (%)

64 83 3 21

Not in cost analysis (n=304)

(49-72) (55.0) (2-6) (11-38.75)

60 165 4 14

(43-75) (54.5) (2-6) (7-28)

P value 0.32w 0.92= 0.32w o0.001w

140 (92.7) 11 (7.3)

279 (92.1) 24 (7.9)

0.81=

67 49 23 10 2 0 23 8 27 106 78

110 116 43 22 11 1 22 7 27 192 126

0.40=

(44.4) (32.5) (15.2) (6.6) (1.3) (0) (16-29) (4-11) (21-30) (70.2) (51.7)

(36.3) (38.3) (14.2) (7.3) (3.6) (0.3) (15-28) (4-10) (21-30) (63.4) (41.6)

0.11w 0.16w 0.43w 0.15= 0.04=

Legend: * Median (interquartile range); w Mann-Whitney test; = Chi-square; y Period between ICU request and admission or cancellation (period of cost analysis) SOFA (day 1): Sequential Organ Failure Assessment at study entry APACHE II: Acute Physiology and Chronic Health Evaluation TISS (day 1): Therapeutic Intervention Scoring System at study entry.

Table 3 - Median cost (US$) of critically ill patients treated outside intensive care units according to the hospital outcome. Variable Laboratory tests Medical procedures Medications Clinical support Total cost Daily cost

Total (n=151) 502.82 81.81 4342.51 298.71 5945.98 1618.51

(255.61-778.37) (0.00-525.38) (2452.17-7713.01) (45.65-551.78) (3831.98-10073.41) (1235.29-2032.40)

Survivors (n=53) 505.54 9.26 3594.37 275.89 5015.29 1269.71

(241.74-753.81) (0.00-598.86) (1231.11-6136.93) (45.65-565.69) (2845.04-8207.94) (847.76-1653.87)

Non-survivors (n=98) 499.52 81.81 4532.12 315.13 6159.26 1787.36

(272.00-821.14) (0.00-528.26) (2698.91-9022.49) (39.70-549.67) (4038.63-11257.26) (1441.40-2280.60)

P value* 0.685 0.551 0.009 0.881 0.041 o0.001

Median data (interquartile deviation) * Mann-Whitney test.

than that of survivors, with a daily average of h1,380.00 for all patients (30). When analyzing diagnostic groups in a multicenter study in Italian ICUs, the authors found that patients with multiple trauma (h4,717.00), acute abdominal issues (h3,529.00) and pneumonia or acute respiratory distress syndrome (h3,946.00) presented the highest levels of resource consumption (31). In the present study, it was observed that 80.6% of the diagnoses were related to infection and its complications. The first and largest multicenter study conducted in Brazil on this issue noted that the cost of treatment of septic patients was high, and the median daily cost was higher in non-survivors (US$ 1,094.00) compared to survivors (US$ 826.00) (32). In an evaluation of the economic impact of implementing a hospital protocol for sepsis, Koening et al. also noted that the mean overall costs were higher for non-survivors (US$ 27,308.00) than for survivors (US$ 20,021.00) (33). In other countries, such as the United States (34), Italy (35), Germany (36) and France (37), it has also been shown that resource consumption was higher in non-surviving septic patients than among survivors. Regarding this point, it is worth noting that each country is unique in relation to its

became refractory due to sub-optimal treatment provided outside the ICU, and it is conceivable that patients may die due to the lack of ICU beds (3). Since most critical illnesses are time sensitive, delaying ICU admission can lead to a worse prognosis if patients do not have access to adequate care, but there are conflicting data in the literature regarding this issue. Some authors describe a 1.5% increase in the risk of death for each hour of delay in ICU admission (3) for patients treated in hospital wards, while others have described no change in prognosis for surgical patients cared for in the post-anesthesia care unit (28). The direct costs associated with the care of these patients were high and comparable with the costs described in the literature regarding ICU patients (26,27,29-32). In the present study, the cost of non-surviving patients was higher than that of survivors. Jacobs et al., in an analysis of the daily cost of ICU patients in the UK, found that the cost of non-survivors was higher than that of survivors, with a daily consumption of a mean of d578.00 for all patients and d748.00 for nonsurvivors (29). Other factors that determined costs included mechanical ventilation and the APACHE II score. It was also noted in France that the cost of non-survivors was higher

572

Caring for patients outside intensive care units Urizzi F et al.

CLINICS 2017;72(9):568-574

Figure 1 - Median daily costs and 95% confidence intervals for the survivors and non-survivors in the first week of the study period.

health system, management models, reimbursement rates and cost components; therefore, such comparisons should be interpreted with caution. However, this finding occurred independent of national differences and characteristics. The findings of the present study contribute to the planning of resource allocation, admission protocols with a prioritization model, investment in staff, and training courses. There are some limitations to the present study that should be considered. The first is related to the analysis of a single center; thus, interpretation of the results must be performed with caution and be limited to institutions with similar characteristics. The second refers to the observational design, which may be prone to selection bias and could have affected the results. This study is novel in its analysis of the costs of critically ill patients treated outside the ICU, and this contribution is the main strength of this study. Another strength of this study was the rigor used in the cost analysis, which applied the methodology considered the "gold standard". Moreover, the results are relevant, comparable and reproducible for other institutions with the same characteristics. Critically ill patients treated outside the ICU presented prognostic scores that demonstrated a high degree of organ dysfunction and required a large number of therapeutic interventions, including vasoactive drugs and invasive mechanical ventilation. The direct costs of treatment of these critically ill patients were high and associated with a poor prognosis.

2. Robert R, Reignier J, Tournoux-Facon C, Boulain T, Lesieur O, Gissot V, et al. Refusal of intensive care unit admission due to a full unit: impact on mortality. Am J Respir Crit Care Med. 2012;185(10):1081-7, http://dx.doi. org/10.1164/rccm.201104-0729OC. 3. Cardoso LT, Grion CM, Matsuo T, Anami EH, Kauss IA, Seko L, et al. Impact of delayed admission to intensive care units on mortality of critically ill patients: a cohort study. Crit Care. 2011;15(1):R28, http://dx.doi.org/10.1186/cc9975. 4. Nardi G, Riccioni L, Cerchiari E, De Blasio E, Gristina G, Oransky M, et al. Impact of an integrated treatment approach to the severely injured patients (ISS =/4 16) on hospital mortality and quality of care. Minerva Anestesiol. 2002;68(1-2):25–35. 5. Fresco C, Carinci F, Maggioni AP, Ciampi A, Nicolucci A, Santoro E, et al. Very early assessment of risk for in-hospital death among 11,483 patients with acute myocardial infarction. GISSI investigators. Am Heart J. 1999; 138(6 Pt 1):1058–64, http://dx.doi.org/10.1016/S0002-8703(99)70070-0. 6. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–77, http://dx.doi.org/10.1056/ NEJMoa010307. 7. Tee A, Calzavacca P, Licari E, Goldsmith D, Bellomo R. Bench-to-bedside review: The MET syndrome--the challenges of researching and adopting medical emergency teams. Crit Care. 2008;12(1):205, http://dx.doi.org/ 10.1186/cc6199. 8. Lee A, Bishop G, Hillman KM, Daffurn K. The Medical Emergency Team. Anaesth Intensive Care. 1995;23(2):183–6. 9. Bellomo R, Goldsmith D, Uchino S, Buckmaster J, Hart G, Opdam H, et al. Prospective controlled trial of effect of medical emergency team on postoperative morbidity and mortality rates. Crit Care Med. 2004; 32(4):916-21, http://dx.doi.org/10.1097/01.CCM.0000119428.02968.9E. 10. Pittard AJ. Out of our reach? Assessing the impact of introducing a critical care outreach service. Anaesthesia. 2003;58(9):882–5, http://dx.doi.org/ 10.1046/j.1365-2044.2003.03331.x. 11. Karpman C, Keegan MT, Jensen JB, Bauer PR, Brown DR, Afessa B. The impact of rapid response team on outcome of patients transferred from the ward to the ICU: a single-center study. Crit Care Med. 2013; 41(10):2284–91, http://dx.doi.org/10.1097/CCM.0b013e318291cccd. 12. Laurens N, Dwyer T. The impact of medical emergency teams on ICU admission rates, cardiopulmonary arrests and mortality in a regional hospital. Resuscitation. 2011;82(6):707–12, http://dx.doi.org/10.1016/ j.resuscitation.2010.11.031. 13. Boniatti MM, Azzolini N, Viana MV, Ribeiro BS, Coelho RS, Castilho RK, et al. Delayed medical emergency team calls and associated outcomes. Crit Care Med. 2014;42(1):26–30, http://dx.doi.org/10.1097/CCM.0b013e 31829e53b9. 14. Gonc¸alves PD, Polessi JA, Bass LM, Santos Gde P, Yokota PK, Laselva CR, et al. Reduced frequency of cardiopulmonary arrests by rapid response teams. Einstein. 2012;10(4):442–8, http://dx.doi.org/10.1590/S1679-4508 2012000400009. 15. Lopes JL, Silva RC, Quilici AP, Palomo JS, Gonzalez MM, Ferreira FG. Implementac¸ão dos Times de Resposta Rápida: experiência de um hospital de alta complexidade em cardiopneumologia. Rev Bras Clin Med. 2012;10(5):394–7. 16. Nates JL, Nunnally M, Kleinpell R, Blosser S, Goldner J, Birriel B, et al. ICU Admission, Discharge, and Triage Guidelines: A Framework to

’ AUTHOR CONTRIBUTIONS Urizzi F, Matsuo T and Grion CM participated in the study concept and design. Urizzi F, Tanita MT, Festti J and Cardoso LT carried out the acquisition of data and participated in the analysis and interpretation of data. Urizzi F and Grion CM drafted the manuscript. Matsuo T performed the statistical analysis. All authors participated in the critical revision of the manuscript for intellectual content and read and approved the ﬁnal version of the manuscript.

’ REFERENCES 1. Zimmerman JE, Kramer AA, Knaus WA. Changes in hospital mortality for United States intensive care unit admissions from 1988 to 2012. Crit Care. 2013;17(2):R81, http://dx.doi.org/10.1186/cc12695.

573

Caring for patients outside intensive care units Urizzi F et al.

17. 18.

19.

20. 21. 22.

23. 24.

25.

26.

27.

CLINICS 2017;72(9):568-574

Enhance Clinical Operations, Development of Institutional Policies, and Further Research. Crit Care Med. 2016;44(8):1553-602, http://dx.doi.org/ 10.1097/CCM.0000000000001856. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: A severity of disease classification system. Crit Care Med. 1985;13(10): 818–29, http://dx.doi.org/10.1097/00003246-198510000-00009. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22(7):707–10, http://dx.doi.org/ 10.1007/BF01709751. Miranda DR, de Rijk A, Schaufeli W. Simplified Therapeutic Intervention Scoring System: The TISS-28 items–results from a multicenter study. Crit Care Med. 1996;24(1):64-73, http://dx.doi.org/10.1097/00003246-19960 1000-00012. Reis Miranda D, Jegers M. Monitoring costs in the ICU: a search for a pertinent methodology. Acta Anaesthesiol Scand. 2012;56(9):1104–13, http://dx.doi.org/10.1111/j.1399-6576.2012.02735.x. Goddeeris JH. Theoretical considerations on the cost of illness. J Health Econ. 1983;2(2):149–59, http://dx.doi.org/10.1016/0167-6296(83)90004-8. Associação Medica Brasileira [Internet]. Classificação Brasileira Hierarquizada de Procedimentos Médicos (CBHPM). 2012. [cited 2014 May 7]. Available from: http://www.amb.org.br/_arquivos/_downloads/cbhpm_ 2012.pdf. Brasíndice. [Internet]. Guia Farmacêutico Brasíndice. 2014. [cited 2014 Apr 10]. Available from: http://www.brasindice.com.br/brasindice/index.php. Charlson ME, Pompei P, Ales KL, Mackenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373-83, http://dx.doi.org/ 10.1016/0021-9681(87)90171-8. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801-10, http://dx.doi. org/10.1001/jama.2016.0287. Simchen E, Sprung CL, Galai N, Zitser-Gurevich Y, Bar-Lavi Y, Gurman G, et al. Survival of critically ill patients hospitalized in and out of intensive care units under paucity of intensive care unit beds. Crit Care Med. 2004;32(8):1654–61, http://dx.doi.org/10.1097/01.CCM.0000133021. 22188.35. Simchen E, Sprung CL, Galai N, Zitser-Gurevich Y, Bar-Lavi Y, Levi L, et al. Survival of critically ill patients hospitalized in and out of intensive

28.

29.

30. 31.

32.

33.

34.

35. 36.

37.

574

care. Crit Care Med. 2007;35(2):449–57, http://dx.doi.org/10.1097/01. CCM.0000253407.89594.15. Zhou JC, Pan KH, Huang X, Yu WQ, Zhao HC. Delayed admission to ICU does not increase the mortality of patients post neurosurgery. Int J Neurosci. 2015;125(6):402-8, http://dx.doi.org/10.3109/00207454.2014. 943370. Jacobs P, Edbrooke D, Hibbert C, Fassbender K, Corcoran M. Descriptive patient data as an explanation for the variation in average daily costs in intensive care. Anaesthesia. 2001;56(7):643–7, http://dx.doi.org/10.1046/ j.1365-2044.2001.02052.x. Sznajder M, Aegerter P, Launois R, Merliere Y, Guidet B, CubRea. A costeffectiveness analysis of stays in intensive care units. Intensive Care Med. 2001;27(1):146-53, http://dx.doi.org/10.1007/s001340000760. Rossi C, Simini B, Brazzi L, Rossi G, Radrizzani D, Iapichino G, et al. Variable costs of ICU patients: a multicenter prospective study. Intensive Care Med. 2006;32(4):545–52, http://dx.doi.org/10.1007/s00134-0060080-2. Sogayar AM, Machado FR, Rea-Neto A, Dornas A, Grion CM, Lobo SM, et al. A multicentre, prospective study to evaluate costs of septic patients in Brazilian intensive care units. Pharmacoeconomics. 2008;26(5):425–34, http://dx.doi.org/10.2165/00019053-200826050-00006. Koening, A, Picon PD, Feijó J, Silva E, Westphal GA. Estimate of the economic impact of implementing an in hospital protocol for the early detection and treatment of severe sepsis in public and private hospitals in southern Brazil. Rev Bras Ter Intensiva. 2010;22(3):213–9, http://dx.doi. org/10.1590/S0103-507X2010000300001. Angus, DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001;29 (7):1303-10, http://dx.doi.org/10.1097/00003246-200107000-00002. Lucioni C, Mazzi S, Currado I. Sepsis costs in Italy. Intensive Care Med. 2001;27(Suppl.2):S284. Moerer O, Schmid A, Hofmann M, Herklotz A, Reinhart K, Werdan K, et al. Direct costs of severe sepsis in three German intensive care units based on retrospective electronic patient record analysis of resource use. Intensive Care Med. 2002;28(10):1440-6, http://dx.doi.org/10.1007/ s00134-002-1429-9. Adrie C, Alberti C, Chaix-Couturier C, Azoulay E, De Lassence A, Cohen Y, et al. Epidemiology and economic evaluation of severe sepsis in France: age, severity, infection site, and place of acquisition (community, hospital, or intensive care unit) as determinants of workload and cost. J Crit Care. 2005;20(1):46–58, http://dx.doi.org/10.1016/j.jcrc.2004.10.005.

BASIC RESEARCH

An activating mutation in the CRHR1 gene is rarely associated with pituitary-dependent hyperadrenocorticism in poodles Viviani De-Marco,I,II,* Luciani R. Carvalho,I Mariana F. Guzzo,I Paulo S.L. Oliveira,III Larissa G. Gomes,I Berenice B. MendoncaI,* I Unidade de Endocrinologia do Desenvolvimento, Laboratorio de Hormonios e Genetica Molecular LIM/42, Faculdade de Medicina FMUSP, Universidade de Sao Paulo, Sao Paulo, SP, BR. II Curso de Medicina Veterinaria, Universidade de Santo Amaro, Sao Paulo, SP, BR. III Laboratorio Nacional de Biociencias – LNBio, Campinas, SP, BR.

OBJECTIVES: Pituitary-dependent hyperadrenocorticism is the most common cause of naturally occurring hypercortisolism in dogs. CRHR1 expression in human and dog corticotrophinomas suggested that this gene affects pituitary tumorigenesis. The present study aimed to investigate mutations in the CRHR1 coding region in poodles with pituitary-dependent hyperadrenocorticism. METHODS: Fifty poodles with pituitary-dependent hyperadrenocorticism and 50 healthy poodles were studied. Genomic DNA was amplified by PCR and analyzed by Sanger sequencing. RESULTS: The novel CRHR1 p.V97M mutation was identified in one dog. This valine residue, located in the aminoterminal extracellular domain, exhibits high affinity for its corticotropin-releasing hormone (CRH) ligand. Bioinformatic analysis revealed structural rearrangements in the mutant protein, with a 17% increase in the surface binding affinity between CRHR1 and CRH. In vitro functional studies showed that mutant CRHR1 induced higher ACTH secretion than the wild type after stimulation with human CRH. CONCLUSION: These results suggest that germline activating mutations in CRHR1 may be a rare cause of pituitary hyperadrenocorticism in poodles. KEYWORDS: Hyperadrenocorticism; CRHR1; Mutation; Cushing’s Disease; Dogs. De-Marco V, Carvalho LR, Guzzo MF, Oliveira PS, Gomes LG, Mendonca BB. An activating mutation in the CRHR1 gene is rarely associated with pituitary-dependent hyperadrenocorticism in poodles. Clinics. 2017;72(9):575-581 Received for publication on December 16, 2017; First review completed on February 17, 2017; Accepted for publication on May 16, 2017 *Corresponding author. E-mail: vivianidemarco@gmail.com / beremen@usp.br

’ INTRODUCTION

(POMC) to release ACTH. The expression of these factors is regulated by glucocorticoids. Other hypothalamic compounds, such as vasopressin, can also stimulate ACTH release but with much lower potency (6). The finding that CRHR1 is expressed in corticotroph adenoma cells suggests that this receptor might play a role in pituitary tumorigenesis. Human patients with Cushing’s disease exhibit ACTH hypersecretion in response to a CRH stimulation test, despite the autonomous secretion of ACTH (7, 8). Similarly, in vitro studies have shown that adenomatous corticotroph cells exhibit a dose-dependent increase in ACTH levels after CRH stimulation (9, 10). Additionally, rat models treated with high doses of CRH developed corticotroph hyperplasia (11, 12). Therefore, multiple lines of evidence indicate that chronic and excessive hypothalamic hormone stimulation can result in increased pituitary hormone secretion and cell proliferation, leading to tumor development and neoplastic transformation in the adenohypophysis (13). In dogs with PDH, the findings have been somewhat conflicting, with different studies reporting hyposecretion, normal secretion or hypersecretion of ACTH in response to CRH. However, dogs with PDH generally exhibit a persistent response to CRH, with elevated ACTH (14-16).

Cushing’s disease is a common endocrinopathy in dogs and is characterized by polydipsia, polyuria, polyphagia, abdominal enlargement and alopecia (1-3). Pituitary-dependent hyperadrenocorticism (PDH) is the most common cause of naturally occurring hypercortisolism in dogs, accounting for 80 to 85% of all cases. PDH mainly affects dogs aged 6 years and older, and poodles, dachshunds, shih tzus and various terrier breeds appear to be at the greatest risk (4, 5). More than 90% of dogs with PDH have pituitary tumors, most of which are microadenomas (o10 mm in diameter) (1, 3). The secretion of ACTH is mainly controlled by corticotropin-releasing hormone (CRH), which binds to its receptor CRHR1 and activates the transcription of proopiomelanocortin

575

CRHR1 mutation causing hypercortisolism De-Marco V et al.

CLINICS 2017;72(9):575-581

performed. Plasma samples were collected over 18 months and measured using 10 assays. Plasma ACTH concentrations were measured using an immunoradiometric assay (IRMA) with a commercial kit (ELSA-ACTH, CisBio International, Codolet, France). The ACTH intra-assay and inter-assay coefficients of variation according to the manufacturer were o6.1% and o5.3%, respectively; the intra-assay and interassay coefficients obtained in the laboratory were o7.8% and o9.9%, respectively. Plasma ACTH measurements were performed at the Hormone and Molecular Genetics Laboratory of Clinics Hospital, School Medicine, University of São Paulo, Brazil. We used a human ACTH kit due to the lack of a specific kit for dogs and because the ACTH molecule is highly homologous between these two species. The control group comprised 50 healthy poodles, including 32 females and 18 males with a mean age of 9.4± 2.8 years (range, 6 to 16 years), which presented to the Veterinary Teaching Hospital for ophthalmopathies, skin lesions or elective castration. Control dogs were matched to PDH dogs according to age (X6 years) and breed. The criteria for inclusion in the control group were that each dog was healthy, not obese, and presenting with no clinical signs suggestive of hyperadrenocorticism and with normal routine biochemical tests (urea, creatinine, alanine aminotransferase, alkaline phosphatase) and hemogram.

Considering the above evidence of the role of CRH and its receptor CRHR1 in pituitary tumor development and the high incidence of PDH in poodle dogs, our aim was to screen for CRHR1 germline mutations in poodles with PDH.

’ MATERIALS AND METHODS Animals This prospective observational study included dogs that were initially evaluated at the Veterinary Medical Teaching Hospital of the University of Guarulhos from April 2006 through September 2007. All dogs were enrolled in the study with the informed consent of their owners. The possibility of PDH was based on historical data and a physical examination. The inclusion criteria were that each dog must have had at least two clinical signs identified by the owner (polyuria, polydipsia, polyphagia or abdominal enlargement) and at least 2 of the following 6 clinicopathological findings: high serum alkaline phosphatase activity, high serum alanine aminotransferase activity, hypercholesterolemia, hypertriglyceridemia, urine specific gravity o1.020, and thrombocytosis (1, 4). All dogs underwent abdominal ultrasonography. For each dog, the results of a low-dose dexamethasone suppression test (LDDST) were consistent with hyperadrenocorticism (serum cortisol concentrations 8 hours after dexamethasone treatment X1.4 mg/dL). The diagnostic criteria for PDH included the following: an LDDST result indicative of spontaneous hyperadrenocorticism, ultrasonographic evidence of bilateral adrenal enlargement showing 2 relatively equal-sized, homogeneous adrenal glands and the absence of an adrenal mass, and plasma concentration of endogenous ACTH 417 pg/mL (17, 18). Computed tomography (CT) and magnetic resonance imaging (MRI) were not performed due to the cost of these examination, the need for anesthesia and because none of the dogs had signs of neurological conditions or suspicion of pituitary macroadenoma. All dogs with concurrent disorders were excluded from the study, as were those with an incomplete diagnosis of PDH, those with adrenal-dependent hyperadrenocorticism suspicion, and those that recently underwent PDH treatment.

CRHR1 sequencing and analysis. Genomic DNA was extracted from peripheral blood leukocytes that were obtained from the 50 PDH and 50 control poodles using standardized protocols (20). The CRHR1 coding region and intron-exon junctions (GenBank NC_006591.3) were amplified by PCR in an automated thermal cycler (Applied Biosystems PCR System 9700) using specific primers. The PCRs were performed with 2 mL of genomic DNA in a 50-mL final volume that contained 200 mM dNTPs, 0.5 mM (20 pmol) primers, 1.5 U of GoTaq Polymerase (Promega Corporation, Madison, WI, USA), 10 mL of 5x PCR reaction buffer (Promega), 1.5 mM MgCl2 and 0.4 M betaine. The PCR programs included an initial activation at 98oC for 5 min followed by 40 cycles of 45 s at 98oC, 30 s at 50-58oC (depending on the primers) and 45 s at 72oC, with a final extension at 72oC for 5 min. The final extension step for larger fragments, namely exons 6-8 (993 bp) and 9-12 (1,235 bp), was 72oC for 3 min. The products were visualized on a 1.5% agarose gel containing ethidium bromide. Eight pairs of intron-flanking primers were designed based on the intron/exon structure of the CRHR1 gene using the Primer3 core program. Exons 6 to 8 and exons 9 to 12 were amplified as single fragments (Table 1). The resulting amplified products were purified and subjected to capillary electrophoresis in an automatic ABI3100 Genetic Analyzer (Applied Biosystems). The resulting sequences were compared with the CRHR1 reference sequence and published on the NCBI website under Ensembl accession number ENSCAFT00000021552.

Endocrine tests and hormone assays For the LDDST, blood samples (2 mL each) were collected before and 8 hours after IV administration of 0.01 mg/kg dexamethasone to determine the serum concentrations of cortisol. A serum cortisol concentration X1.4 mg/dL at 8 hours after dexamethasone administration was considered consistent with naturally occurring hyperadrenocorticism (2). Serum cortisol measurements were performed at the PROVET Hormone Laboratory in São Paulo, Brazil. A radioimmunoassay was performed using a solid-phase commercial diagnostic kit for cortisol (Coat-A-Count, Siemens Medical Solutions Diagnostics, Malvern, PA, USA) that was previously validated for canine cortisol (19). The intra-assay and inter-assay coefficients of variation for cortisol were o5.1% and o6.4%, respectively. Plasma samples for ACTH measurements were collected in plastic tubes containing EDTA and immediately refrigerated. The samples were transported to the laboratory inside a Styrofoam container with ice for a maximum period of 1h before subsequent centrifugation at 1600 g for 15 min at 4oC. The samples were stored at -70oC until the assay was

CRHR1 allelic variant analysis by restriction enzyme digestion. The allelic variant in exon 4 (codon 97) of the CRHR1 gene eliminates a BstUI cleavage site. To identify this variant in the control group, we performed enzymatic digestion. Digestion reactions were performed on all samples in the control group.

576

CRHR1 mutation causing hypercortisolism De-Marco V et al.

CLINICS 2017;72(9):575-581

Table 1 - Forward (F) and reverse (R) primers for canine CRHR1 amplification, with product sizes (bp) and annealing temperatures (Ta). Primer

Sequence

1F 1R 2F 2R 3F 3R 4F 4R 5F 5R 6-8F 6-8R 9-12F 9-12R 13F 13R

50 50 50 50 50 50 50 50 50 50 50 50 50 50 50 50

CTGAGTCAGGAGACGGCGCA 30 TCATTGTGGCGAAGCTGCTG 30 TTGGGATCCTAGGCTTGATG 30 GCAAGATTCCAGGTCTCCAG 30 GAACGAATGCTGGATCCCTA 30 TCTCGTGGATACAACCAC 30 AGATGGACGAACCAATGGAC 30 AGTCTCCTTCCCACCCTGTC 30 GGGAAGGGGAATAACTACCG 30 GTCTCTTGGAACCTCTGGCA 30 TCCTCAGTTTCCCCATCCATA 30 TGAGGTACAGGCTCTCAGCC 30 GAGGCATTGTGTTGGGATCT 30 CAGGTCCCTAATGAGGATGC 30 ACGTGTACTGCTGCTTGTGG 30 GAGAGCAGCCATAGTCTGGG 30

Product size (bp)

culture incubator for 24h. The cells were then washed 3 times with PBS, and 100 nM human CRH (hCRH; Ferring) was added to each well. Aliquots (10 mL) of the supernatant media were collected from transfected AtT20 cells for ACTH measurement 6, 12 and 24h after hCRH addition. The assays were performed twice in triplicate. The ACTH levels in media from transiently transfected mutant and wild-type cells were expressed in terms of the fold change over the basal ACTH level in non-stimulated cells that were transfected with empty vector.

Ta (oC)

422

413

456

581

401

993

1,235

419

Statistical analysis Analysis of variance (ANOVA) followed by Bonferroni’s adjustment was used to compare the normalized fold change in ACTH among groups at each time point. The data were analyzed using Sigma Stat 3.5 software. Statistical significance was set at po0.05.

’ RESULTS Demographic and general characteristics of the animals

Molecular modeling of CRHR1. A homology model for the mutant canine CRHR1 was constructed based on an available crystal structure (PDB ID: 3EHT), which includes the human extracellular domain of CRHR1 bound to the CRH peptide template (21), using YASARA (22). The p.V97M mutant complex was constructed by superimposing the modeled structure of the mutant canine CRHR1 over the structure of the wild-type human CRHR1-CRH complex. To assess whether the mutation interferes with ligandreceptor binding, structure optimization was performed by energy minimization to estimate the binding energy between the hormone and the receptor. The binding energy was calculated using YASARA (See: http://www.yasara.org) with the YAMBER3 force field (22).

The study included 50 consecutively selected poodles with PDH, including 33 females and 17 males with a mean age of 8.7±2.8 years (range, 1.5 to 14 years), which presented at the Veterinary Hospital of Guarulhos University. Among the 50 dogs, 3 had family members that were also affected by PDH. The initial laboratory screening tests consisted of determining the serum levels of cholesterol, triglycerides, alkaline phosphatase, and alanine aminotransferase as well as tests for glycemia, urinary density, dexamethasone suppression, and abdominal ultrasonography. Hypercholesterolemia and hypertriglyceridemia were found in 78% (n=39/50) and 74% (n=37/50) of the cases, respectively, with mean cholesterol values of 396 ± 156 mg/dL and triglycerides of 169±96 mg/dL. Elevated blood levels of alkaline phosphatase and alanine aminotransferase were observed in 78% (n=39/50) of the cases (mean value 459± 401 IU/L) and in 70% (n=35/50) of the cases (mean value 204±158 IU/L), respectively. Urinary densities below the reference values (1.025–1.045) were identified in 80% of the animals (n=40/50), of which 20% (n=10/50) showed hyposthenuria (densityp1008). Hyperglycemia was identified in six animals (12%), two of which were insulin-dependent diabetics. No animals showed suppression of serum cortisol levels 8 hours after the intravenous application of dexamethasone (mean values of cortisol=5.4±3.3 mg/dL), confirming endogenous hypercortisolism. All cases were diagnosed with PDH based on ACTH baseline plasma concentrations above 17 pg/mL (mean values of ACTH=42±30 pg/mL). Abdominal ultrasound showed symmetric adrenal glands with regular contours, homogeneous echotextures, and the absence of nodulations and calcifications. The mean values of the caudal pole thickness of the right and left adrenal glands were 0.73±0.14 and 0.74±0.11 cm, respectively. Dogs with pituitary-dependent hypercortisolism have, on average, a caudal pole thickness greater than or equal to 0.7 cm (23).

In vitro studies. Plasmid construction: The human CRHR1 cDNA sequence in the pcDNA plasmid was purchased from LabLife (Cambridge, MA, USA). The plasmid was mutagenized using specific primers (sense, 50 -CTGGG CCGCCCGCATGAATTACTCCG-30 and anti-sense, 50 -CGGA GTAATTCATGCGGGCGGCCCAG-30 ) with the QuikChange Mutagenesis Kit (Stratagene, La Jolla, CA, United States), according to the manufacturer’s instructions. The presence of the mutagenic insertion was confirmed by automated sequencing using a pcDNA primer (50 -TAGAAGGCACAGTCGAGG-30 ). Cell culture: AtT20 cells were grown in DMEM (Invitrogen) supplemented with 10% fetal bovine serum (Invitrogen) and 1x antibiotics (Invitrogen) at 37oC with 5% CO2. The AtT20 cells were seeded at a density of 5x104 cells per well in 24-well plates 1 day before transfection. The cells were transfected with 1,000 ng/well of plasmid DNA. The empty pcDNA vector was used to reach the total amount of plasmid per well. Transient transfection: In each well, 1 mg of DNA was diluted in 150 mM NaCl to a final volume of 50 mL, and 2 mL of jetPEI reagent was diluted separately in 150 mM NaCl to a final volume of 50 mL. These 2 reagents were mixed and incubated for 20 min at room temperature. The jetPEI/DNA mixture (100 mL) was added to AtT20 cells that were seeded in 500 mL of serum-containing medium; the transfection reagent and medium were homogenized by gently swirling the plate. The transfected cells were incubated in a cell

Mutation analysis We identified only one heterozygous variant in one of the dogs with PDH, in exon 4 of the CRHR1 gene. This variant results from a guanine-to-adenine (G - A) change in codon

577

CRHR1 mutation causing hypercortisolism De-Marco V et al.

CLINICS 2017;72(9):575-581

choice of drug in canine PDH therapy, was not available in Brazil. The animal showed partial improvement, with several adjustments of the mitotane dose being necessary during the 10-months of follow-up therapy. For personal and financial reasons, the owner discontinued the treatment. The animal died 6 months after the treatment was discontinued, likely due to pulmonary embolism. The affected dog’s male offspring is not a carrier of the p.V97M mutation and did not shown symptoms of Cushing’s disease at 7 years of age.

97 that causes a valine-to-methionine substitution (p.V97M) (Figure 1). Although both valine (encoded by GTG) and methionine (encoded by ATG) are nonpolar amino acids, the Val97 residue is highly conserved among different species. According to BLAST (https://blast.ncbi.nlm.nih.gov/Blast. cgi), the coding region of CRHR1 is 98% similar between dogs and humans, and the protein shares the same amino acid, valine, at position 97. Moreover, this change occurred in the amino-terminal portion of the extracellular domain of the CRHR1 receptor, which has a high affinity for the ligand (22). The p.V97M allelic variant was not found among the 100 poodle alleles in the control group or in the normal siblings of the disease cases. The animal with the p.V97M allelic variant was 9 years old and had a history suggestive of hypercortisolism for at least 12 months. The dog showed clinical features of hyperadrenocorticism, suggesting chronic and progressive disease (Figure 2). The serum cortisol level after an intravenous injection of dexamethasone was 7.8 mg/dL, and his baseline plasma ACTH level (45 pg/mL) confirmed the presence of ACTH-dependent hypercortisolism. An abdominal ultrasound examination revealed thickening of the caudal poles of the right (1.0 cm) and left (0.92 cm) adrenal glands, both of which showed regular contours, homogeneous echotexture and reduced echogenicity. The animal was treated with mitotane at a dose of 25 mg/kg every 12h for 7 days (induction phase); subsequently, it was treated with 25 mg/kg twice a week (maintenance phase). At this time, trilostane, which is currently considered the first

Mutant protein modeling study The allelic variant p.V97M is located in the extracellular amino terminus of the CRHR1 protein, which is considered the primary region for receptor-ligand binding (21, 24). Protein modeling revealed that relative to the wild-type protein, the CRHR1 mutant showed a 17% increase in its binding affinity for CRH. Changes in the 3-dimensional structure of the extracellular domain of CRHR1 affect receptor activation and affinity for the ligand (21, 25). The quaternary structures of the mutant and wild-type proteins were examined, and the binding energy between the hormone and its receptors was evaluated to better understand the involvement of this variant in the pathogenesis of ACTH-dependent hyperadrenocorticism. A comparison of the 3-dimensional structures of the complexes formed with the mutant (p.V97M) and wild-type proteins revealed a structural rearrangement in the contact surface between the p.V97M receptor and the ligand (Figure 3). To assess whether the mutation interferes with receptorligand binding, optimization of the mutant complex structure was performed using energy minimization. The binding energy between the hormone and the receptor was then estimated. The binding enthalpy was 1,893 kJ/mol for the wild-type complex and 2,134 kJ/mol for the p.V97M complex; the mutant CRHR1 exhibited a 17% increase in its binding affinity for CRH, therefore suggesting a clear gain of function for the p.V97M variant.

Figure 1 - The wild-type sequence of exon 4 of the CRHR1 gene from a control dog (upper panel) and the heterozygous G-to-A substitution causing the amino acid change (p.V97M) in a poodle with ACTH-dependent hypercortisolism (lower panel).

Functional characterization of the CRHR1 mutant protein The fold changes in the ACTH levels after stimulation relative to the basal levels in AtT20 cells transfected with empty, wild type and mutant vectors were 5.01±0.65, 5.41±0.57 and 11.57±1.45, respectively, at 12 h and 14.0± 1.72, 21±3.6 and 38.9±0.45, respectively, at 24h. There was a significant increase in the ACTH levels in the mutant relative to the wild-type at 12 h (*po0.05) and 24 h (**po0.01) (Figure 4).

’ DISCUSSION The expression of CRHR1 in human corticotrophinomas suggests that hormone receptors play a role in pituitary tumorigenesis, promoting sustained cell stimulation even in the absence of hypothalamic hormones (26-28). Corticotroph tumors in dogs and humans exhibit CRHR1 overexpression and greater sensitivity to CRH (13, 29, 30). In the present study, we screened the CRHR1 gene for mutations in poodles with PDH and identified 1 dog with a novel p.V97M activating heterozygous allelic mutation that was located at the amino-terminal portion of the extracellular domain of CRHR1. The p.V97M CRHR1 mutant has a larger

Figure 2 - The dog with ACTH-dependent hypercortisolism and carrying the allelic variant p.V97M, showing marked abdominal distension, alopecia, skin atrophy and telangiectasia.

578

CRHR1 mutation causing hypercortisolism De-Marco V et al.

CLINICS 2017;72(9):575-581

Figure 3 - The structural rearrangement in the mutant protein (red) and the resulting changes in the contact surface between the ligand and the mutant receptor (containing methionine). The wild-type receptor is shown for comparison.

Figure 4 - The observed fold changes in ACTH relative to the basal ACTH level in cells transfected with empty vector were 5.01±0.65 (empty vector, EV), 5.41±0.57 (wild-type, WT) and 11.57±1.45 (mutant, MUT) at 12h and 14.0±1.72 (EV), 21±3.6 (WT) and 38.9±0.45 (MUT) at 24h. There were significant differences in the ACTH levels of cells transfected with the mutant plasmid compared with cells transfected with the wild-type plasmid or with empty vector at 24h (**po0.01, *po0.05).

structural rearrangement due to an altered contact surface between the mutated receptor and the ligand. In vitro studies revealed that AtT20 cells transfected with the mutant plasmid over-secreted ACTH compared relative to cells transfected with the wild-type plasmid. The maximal effects

apolar surface created by the methionine side chain, which may increase the van der Waals contacts with the Ile41 residue of CRH and thus favor ligand binding. Optimization of the molecular and quaternary structure of this mutated protein by energy minimization indicated the existence of a

579

CRHR1 mutation causing hypercortisolism De-Marco V et al.

CLINICS 2017;72(9):575-581

stimulus that causes hyperplasia of specific cell types in the pituitary, itself giving rise to several distinct clones with variable potential to develop into tumors. Such stimuli may include hypothalamic trophic factors, intrapituitary growth factors or pituitary-specific oncogenes (43). Another hypothesis is that the affected dog, by carrying a CRHR1 activating mutation, developed only pituitary hyperplasia and not an adenoma. Pituitary hyperplasia and pituitary adenoma are both possible explanations for the ACTH over-secretion and chronic hypercortisolism observed in this dog. This study aimed to investigate whether Cushing’s disease in poodles has a genetic origin. We found only one dog with an activating mutation among 50 poodles with hyperadrenocorticism, which indicated that in most of cases, the mutation did not cause Cushing’s disease. However, taking into account the high similarity of CRHR1 between humans and dogs, our data suggest that this mutation may also be involved in some human cases of Cushing’s disease; thus, these results may be clinically relevant. The limitations of this study included the unavailability of tumor tissues and the availability of only one male offspring from the dog with the allelic variant p.V97M to test for segregation. In conclusion, we reported a mutation in CRHR1 that leads to persistent activation of the receptor in a poodle with PDH, indicating that this is a rare cause of ACTH-dependent hypercortisolism. This new finding, although rare, suggests that other germline mutations that activate the hypothalamuspituitary-adrenal axis may be associated with this phenotype in dog breeds commonly affected by ACTH-dependent hyperadrenocorticism.

of the mutant and wild-type plasmids on hormone release were observed after 24 h of hCRH stimulation, which caused approximately 38.9- and 21-fold increases in ACTH levels, respectively, over the basal levels in cells transfected with empty vectors. This result confirmed that the variant is an activating mutation. Changes in the 3-dimensional structure of the extracellular domain of CRHR1 affect the activation of the receptor and its affinity for the ligand (21, 24, 31, 32). The binding of CRH to CRHR1 ligand-binding sites has been extensively investigated in studies of chimeric receptors. These studies have indicated that residues 1-118 of the N-terminal domain (specifically, amino acid residues 43-50 and 76-84 in humans and 68-109 in mice) are the most important regions for agonist binding (33-36). Pioszak, Parker (21) reported the importance of valine 97 (Val97) for interactions between CRHR1 and CRH. This interaction occurs through 2 intermolecular hydrogen bonds between the oxygen and nitrogen atoms of the amide C-terminal ligand (CHR) and the amidic nitrogen atoms and carbonyl oxygen of the main chain of the receptor (residue Val97). Optimization of the molecular and quaternary structure of the mutant protein by energy minimization indicated the occurrence of a structural rearrangement due to an altered contact surface between the mutant receptor and the ligand. The p.V97M CRHR1 mutant has a larger apolar surface that is created by the methionine side chain, which may increase van der Waals contacts with the Ile41 residue of CRH to favor binding. Additionally, the prediction that the surface binding energy is increased between the ligand (CRH) and the mutant receptor (CRHR1m) supports the hypothesis that this activating mutation in CRHR1 is involved in the pathogenesis of PDH in this poodle. The late onset of PDH in this animal may contradict the notion that the disease has a genetic etiology; however, other model germline disorders that present later in life, including retinitis pigmentosa and multiple endocrine neoplasia, are well accepted (37, 38). In the affected poodle, the increased ACTH stimulation and secretion caused by the increased affinity of the mutant CRHR1 for CRH might have taken several years to promote corticotroph hyperplasia. Unfortunately, this possibility remains untested, as neither skull magnetic resonance (to visualize the pituitary mass) nor animal necropsy (to identify corticotroph hyperplasia or neoplasia) was performed. Thyrotropin receptors (TSHRs) have been widely studied as models for activating mutations, which are located not only in the transmembrane region but also in the amino-terminal extracellular domain (39). Cases of non-autoimmune congenital hyperthyroidism have been reported as secondary to de novo germline mutations in the TSHR extracellular domain in humans (40). In addition to the influence of the extracellular amino-terminal domain of GPCRs on receptor-ligand binding affinity, this domain might also act as an internal antagonist that is capable of having agonist or antagonist properties depending on whether it is bound to the ligand (39). Although the monoclonal origin of pituitary adenomas has been well established since 1990, there have been instances in which the monoclonal character was acquired at a later stage of polyclonal growth (41). Schulte, Oldfield (42) demonstrated that corticotroph adenomas of the pituitary gland might arise from a single cell or from more than one cell, making them monoclonal or polyclonal in origin. These observations are compatible with the idea of an initiating

’ ACKNOWLEDGMENTS We thank Vivian Moura Massari and Ricardo Paranhos for preparing the plasmids and the staff from hormone laboratory for handling the samples.

’ AUTHOR CONTRIBUTIONS De Marco V performed clinical evaluations and genetic studies. Guzzo MF and Gomes LG performed in vitro studies. Oliveira PS performed bioinformatic analysis. Gomes LG, Carvalho LR and Mendonca BB supervised the work.

’ REFERENCES 1. Feldman EC, Nelson RW, Reusch CE, Scott-Moncrieff JCR, Behrend EN. Canine and feline endocrinology and reproduction, 4rd ed. St. Louis: Elsevier Science, 2015. 2. Behrend EN, Kooistra HS, Nelson R, Reusch CE, Scott-Moncrieff JC. Diagnosis of spontaneous canine hyperadrenocorticism: 2012 ACVIM consensus statement (small animal). J Vet Intern Med. 2013;27(6):1292-304, http://dx.doi.org/10.1111/jvim.12192. 3. Peterson ME. Diagnosis of hyperadrenocorticism in dogs. Clin Tech Small Anim Pract. 2007;22(1):2-11, http://dx.doi.org/10.1053/j.ctsap. 2007.02.007. 4. Kooistra HS, Galac S. Recent advances in the diagnosis of Cushing’s syndrome in dogs. Vet Clin North Am Small Anim Pract. 2010;40(2): 259-67, http://dx.doi.org/10.1016/j.cvsm.2009.10.002. 5. Ling GV, Stabenfeldt GH, Comer KM, Gribble DH, Schechter RD. Canine hyperadrenocorticism: pretreatment clinical and laboratory evaluation of 117 cases. J Am Vet Med Assoc. 1979;174(11):1211-5. 6. Seasholtz A. Regulation of adrenocorticotropic hormone secretion: lessons from mice deficient in corticotropin-releasing hormone. J Clin Invest. 2000;105(9):1187-8, http://dx.doi.org/10.1172/JCI10002. 7. Hermus AR, Pieters GF, Pesman GJ, Smals AG, Benraad TJ, Kloppenborg PW. Responsivity of adrenocorticotropin to corticotropin-releasing hormone and lack of suppressibility by dexamethasone are related phenomena in Cushing’s disease. J Clin Endocrinol Metab. 1986;62(4):634-9, http://dx.doi.org/10.1210/jcem-62-4-634.

580

CRHR1 mutation causing hypercortisolism De-Marco V et al.

CLINICS 2017;72(9):575-581

26. Chen R, Lewis KA, Perrin MH, Vale WW. Expression cloning of a human corticotropin-releasing-factor receptor. Proc Natl Acad Sci U S A. 1993; 90(19):8967-71, http://dx.doi.org/10.1073/pnas.90.19.8967. 27. Suda T, Tozawa F, Dobashi I, Horiba N, Ohmori N, Yamakado M, et al. Corticotropin-releasing hormone, proopiomelanocortin, and glucocorticoid receptor gene expression in adrenocorticotropin-producing tumors in vitro. J Clin Invest. 1993;92(6):2790-5, http://dx.doi.org/10.1172/JCI116898. 28. Sakai Y, Horiba N, Sakai K, Tozawa F, Kuwayama A, Demura H, et al. Corticotropin-releasing factor up-regulates its own receptor gene expression in corticotropic adenoma cells in vitro. J Clin Endocrinol Metab. 1997;82(4):1229-34, http://dx.doi.org/10.1210/jcem.82.4.3883. 29. Dieterich KD, Gundelfinger ED, Lüdecke DK, Lehnert H. Mutation and expression analysis of corticotropin-releasing factor 1 receptor in adrenocorticotropin-secreting pituitary adenomas. J Clin Endocrinol Metab. 1998;83(9):3327-31. 30. Teshima T, Hara Y, Takekoshi S, Teramoto A, Osamura RY, Tagawa M. Expression of genes related to corticotropin production and glucocorticoid feedback in corticotroph adenomas of dogs with Cushing’s disease. Domest Anim Endocrinol. 2009;36(1):3-12, http://dx.doi.org/10.1016/ j.domaniend.2008.08.002. 31. Dautzenberg FM, Kilpatrick GJ, Wille S, Hauger RL. The ligand-selective domains of corticotropin-releasing factor type 1 and type 2 receptor reside in different extracellular domains: generation of chimeric receptors with a novel ligand-selective profile. J Neurochem. 1999;73(2):821-9, http://dx. doi.org/10.1046/j.1471-4159.1999.0730821.x. 32. Perrin MH, Grace CR, Riek R, Vale WW. The three-dimensional structure of the N-terminal domain of corticotropin-releasing factor receptors: sushi domains and the B1 family of G protein-coupled receptors. Ann N Y Acad Sci. 2006;1070:105-19, http://dx.doi.org/10.1196/annals.1317.065. 33. Liaw CW, Grigoriadis DE, Lovenberg TW, De Souza EB, Maki RA. Localization of ligand-binding domains of human corticotropin-releasing factor receptor: a chimeric receptor approach. Mol Endocrinol. 1997; 11(7):980-5, http://dx.doi.org/10.1210/mend.11.7.9946. 34. Wille S, Sydow S, Palchaudhuri MR, Spiess J, Dautzenberg FM. Identification of amino acids in the N-terminal domain of corticotropin-releasing factor receptor 1 that are important determinants of high-affinity ligand binding. J Neurochem. 1999;72(1):388-95, http://dx.doi.org/10.1046/j.14714159.1999.0720388.x. 35. Hillhouse EW, Grammatopoulos DK. The molecular mechanisms underlying the regulation of the biological activity of corticotropin-releasing hormone receptors: implications for physiology and pathophysiology. Endocr Rev. 2006;27(3):260-86, http://dx.doi.org/10.1210/er.2005-0034. 36. Grace CR, Perrin MH, Gulyas J, Digruccio MR, Cantle JP, Rivier JE, et al. Structure of the N-terminal domain of a type B1 G protein-coupled receptor in complex with a peptide ligand. Proc Natl Acad Sci U S A. 2007;104(12):4858-63, http://dx.doi.org/10.1073/pnas.0700682104. 37. Wu DM, Khanna H, Atmaca-Sonmez P, Sieving PA, Branham K, Othman M, et al. Long-term follow-up of a family with dominant X-linked retinitis pigmentosa. Eye (Lond). 2010;24(5):764-74, http://dx.doi.org/10.1038/ eye.2009.270. 38. Hoff AO, Hauache OM. [Multiple endocrine neoplasia type 1 (MEN 1): clinical, biochemical and molecular diagnosis and treatment of the associated disturbances]. Arq Bras Endocrinol Metabol. 2005;49(5):735-46, http://dx.doi.org/10.1590/S0004-27302005000500014. 39. Duprez L, Parma J, Costagliola S, Hermans J, Van Sande J, Dumont JE, et al. Constitutive activation of the TSH receptor by spontaneous mutations affecting the N-terminal extracellular domain. FEBS Lett. 1997; 409(3):469-74, http://dx.doi.org/10.1016/S0014-5793(97)00532-2. 40. Gruters A, Schoneberg T, Biebermann H, Krude H, Krohn HP, Dralle H, et al. Severe congenital hyperthyroidism caused by a germ-line neo mutation in the extracellular portion of the thyrotropin receptor. J Clin Endocrinol Metab. 1998;83(5):1431-6, http://dx.doi.org/10.1210/jcem.83.5.4776. 41. Herman V, Fagin J, Gonsky R, Kovacs K, Melmed S. Clonal origin of pituitary adenomas. J Clin Endocrinol Metab. 1990;71(6):1427-33, http://dx.doi.org/10.1210/jcem-71-6-1427. 42. Schulte HM, Oldfield EH, Allolio B, Katz DA, Berkman RA, Ali IU. Clonal composition of pituitary adenomas in patients with Cushing’s disease: determination by X-chromosome inactivation analysis. J Clin Endocrinol Metab. 1991;73(6):1302-8, http://dx.doi.org/10.1210/jcem-73-6-1302. 43. Clayton RN, Farrell WE. Clonality of pituitary tumours: more complicated than initially envisaged? Brain Pathol. 2001;11(3):313-27, http://dx. doi.org/10.1111/j.1750-3639.2001.tb00402.x.

8. Takeda R, Ito T, Kawato M, Nakabayashi H, Miyamori I, Morise T, et al. ACTH secretory responsiveness of pituitary adrenotroph cell tumor to adrenocorticotropin-releasing factor in Cushing’s disease and Nelson’s syndrome. Exp Clin Endocrinol. 1986;87(1):31-7, http://dx.doi.org/10.1055/ s-0029-1210519. 9. Grino M, Boudouresque F, Conte-Devolx B, Gunz G, Grisoli F, Oliver C, et al. In vitro corticotropin-releasing hormone (CRH) stimulation of adrenocorticotropin release from corticotroph adenoma cells: effect of prolonged exposure to CRH and its interaction with cortisol. J Clin Endocrinol Metab. 1988;66(4):770-5, http://dx.doi.org/10.1210/jcem66-4-770. 10. Horvath SE, Asa SL, Kovacs K, Adams LA, Singer W, Smyth HS. Human pituitary corticotroph adenomas in vitro: morphologic and functional responses to corticotropin-releasing hormone and cortisol. Neuroendocrinology. 1990;51(3):241-8, http://dx.doi.org/10.1159/000125345. 11. Gertz BJ, Contreras LN, McComb DJ, Kovacs K, Tyrrell JB, Dallman MF. Chronic administration of corticotropin-releasing factor increases pituitary corticotroph number. Endocrinology. 1987;120(1):381-8, http://dx. doi.org/10.1210/endo-120-1-381. 12. McNicol AM, Kubba MA, McTeague E. The mitogenic effects of corticotrophin-releasing factor on the anterior pituitary gland of the rat. J Endocrinol. 1988;118(2):237-41, http://dx.doi.org/10.1677/joe.0.1180237. 13. Asa SL. The role of hypothalamic hormones in the pathogenesis of pituitary adenomas. Pathol Res Pract. 1991;187(5):581-3, http://dx.doi. org/10.1016/S0344-0338(11)80149-4. 14. Orth DN, Peterson ME, Drucker WD. Plasma immunoreactive proopiomelanocortin peptides and cortisol in normal dogs and dogs with Cushing’s syndrome: diurnal rhythm and responses to various stimuli. Endocrinology. 1988;122(4):1250-62, http://dx.doi.org/10.1210/endo122-4-1250. 15. Peterson ME, Kemppainen RJ, Orth DN. Effects of synthetic ovine corticotropin-releasing hormone on plasma concentrations of immunoreactive adrenocorticotropin, alpha-melanocyte-stimulating hormone, and cortisol in dogs with naturally acquired adrenocortical insufficiency. Am J Vet Res. 1992;53(3):421-5. 16. van Wijk PA, Rijnberk A, Croughs RJ, Wolfswinkel J, Selman PJ, Mol JA. Responsiveness to corticotropin-releasing hormone and vasopressin in canine Cushing’s syndrome. Eur J Endocrinol. 1994;130(4):410-6, http:// dx.doi.org/10.1530/eje.0.1300410. 17. Feldman EC. Distinguishing dogs with functioning adrenocortical tumors from dogs with pituitary-dependent hyperadrenocorticism. J Am Vet Med Assoc. 1983;183(2):195-200. 18. Gould SM, Baines EA, Mannion PA, Evans H, Herrtage ME. Use of endogenous ACTH concentration and adrenal ultrasonography to distinguish the cause of canine hyperadrenocorticism. J Small Anim Pract. 2001;42(3):113-21, http://dx.doi.org/10.1111/j.1748-5827.2001.tb02007.x. 19. Watson AD, Church DB, Emslie DR. Plasma cortisol concentrations in dogs given cortisone or placebo by mouth. Res Vet Sci. 1993;55(3):379-81, http://dx.doi.org/10.1016/0034-5288(93)90112-S. 20. Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res. 1988; 16(3):1215, http://dx.doi.org/10.1093/nar/16.3.1215. 21. Pioszak AA, Parker NR, Suino-Powell K, Xu HE. Molecular recognition of corticotropin-releasing factor by its G-protein-coupled receptor CRFR1. J Biol Chem. 2008;283(47):32900-12, http://dx.doi.org/10.1074/jbc.M805 749200. 22. Krieger E, Darden T, Nabuurs SB, Finkelstein A, Vriend G. Making optimal use of empirical energy functions: force-field parameterization in crystal space. Proteins. 2004;57(4):678-83, http://dx.doi.org/10.1002/ prot.20251. 23. Barthez PY, Nyland TG, Feldman EC. Ultrasonographic evaluation of the adrenal glands in dogs. J Am Vet Med Assoc. 1995;207(9):1180-3. 24. Perrin MH, Grace CR, Digruccio MR, Fischer WH, Maji SK, Cantle JP, et al. Distinct structural and functional roles of conserved residues in the first extracellular domain of receptors for corticotropin-releasing factor and related G-protein-coupled receptors. J Biol Chem. 2007;282(52): 37529-36, http://dx.doi.org/10.1074/jbc.M703748200. 25. Teli T, Markovic D, Hewitt ME, Levine MA, Hillhouse EW, Grammatopoulos DK. Structural domains determining signalling characteristics of the CRH-receptor type 1 variant R1beta and response to PKC phosphorylation. Cell Signal. 2008;20(1):40-9, http://dx.doi.org/10.1016/ j.cellsig.2007.08.014.

581

BASIC RESEARCH

A Biomechanical Analysis of the Interlock Suture and a Modified Kessler-Loop Lock Flexor Tendon Suture Wenfeng Yang, Dan Qiao, Yuanfei Ren, Yvjin Dong, Yaohua Shang, Tiehui Zhang * Dalian Municipal Central Hospital, Dalian, China.

OBJECTIVE: In this work, we attempted to develop a modified single-knot Kessler-loop lock suture technique and compare the biomechanical properties associated with this single-knot suture technique with those associated with the conventional modified Kessler and interlock suture techniques. METHODS: In this experiment, a total of 18 porcine flexor digitorum profundus tendons were harvested and randomly divided into three groups. The tendons were transected and then repaired using three different techniques, including modified Kessler suture with peritendinous suture, interlock suture with peritendinous suture, and modified Kessler-loop lock suture with peritendinous suture. Times required for suturing were recorded and compared among groups. The groups were also compared with respect to 2-mm gap load, ultimate failure load, and gap at failure. RESULTS: For tendon repair, compared with the conventional modified Kessler suture technique, the interlock and modified Kessler-loop lock suture techniques resulted in significantly improved biomechanical properties. However, there were no significant differences between the interlock and modified Kessler-loop lock techniques with respect to biomechanical properties, gap at failure, and time required. CONCLUSIONS: The interlock and modified Kessler-loop lock techniques for flexor tendon sutures produce similar mechanical characteristics in vitro. KEYWORDS: Tendon Repair; Animal Experimentation; Flexor Tendon. Yang W, Qiao D, Ren Y, Dong Y, Shang Y, Zhang T. A Biomechanical Analysis of the Interlock Suture and a Modified Kessler-Loop Lock Flexor Tendon Suture. Clinics. 2017;72(9):582-587 Received for publication on February 23, 2017; First review completed on June 3, 2017; Accepted for publication on July 13, 2017 *Corresponding author. E-mail: tiehuizhang@126.com

â&#x20AC;&#x2122; INTRODUCTION

to reduce clinical complications associated with flexor tendon suture. The modified Kessler suture is regarded as the most widely used clinical suture technique for tendon repair. However, the optimal surgical approach for achieving a successful outcome remains controversial (6). In the interlock stitch suture technique, locking loops produce improved strength and gap resistance after flexor tendon repair compared with the grasping loops used in the modified Kessler suture technique (7). Additionally, various cross-locked cruciate tendon sutures (8) with diverse biomechanical properties have been developed from interlock stitch sutures and have become common alternatives for flexor tendon repair. However, the interlock stitch suture technique is time-consuming and relatively complex, hindering its wide application in clinical practice. Liu et al. (9) modified the Kessler-loop lock suture to enhance resistance strength. However, the use of two knots in the repaired area could lead to a high risk of early failure, suture slippage, and tendon torsion (10). It is widely accepted that the number of suture passages at the injury site is an extremely important factor in the biomechanical outcomes for different types of tendon sutures. Four-strand suture techniques (such as the Strickland and interlock approaches) have been the most widely investigated techniques in recent decades. In the present study, we modified the Kessler-loop lock suture technique by changing

Flexor tendon injury is a common clinical trauma. Early postoperative active functional exercise can not only effectively reduce the formation of local adhesions and edema but also facilitate sliding function of the tendon, thereby promoting endogenous tendon healing (1,2). The choice of tendon repair technique plays an important role in tendon healing and postoperative functional exercise. The ideal repair should be easy to perform; provide sufficient strength for healing (over 30 N) (3), which can ensure minimal interference with tendon vascularity; and involve secure suture knots and smooth junction of tendon ends. Currently, various suture techniques (4,5) have been proven to exhibit sufficient resistance to gap formation. However, many techniques are demanding with respect to surgical experience and suturing skills. It is essential to explore a relatively simple tenorrhaphy with good mechanical strength

Copyright & 2017 CLINICS â&#x20AC;&#x201C; This is an Open Access article distributed under the terms of the Creative Commons License (http://creativecommons.org/licenses/by/ 4.0/) which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is properly cited. No potential conflict of interest was reported. DOI: 10.6061/clinics/2017(09)10

582

A modified Kessler-loop lock suture technique Yang W et al.

CLINICS 2017;72(9):582-587

the double-knot suture to a single-knot suture and compared our modified technique with the conventional modified Kessler suture and interlock suture techniques with respect to biomechanical properties and suturing time.

After suturing was completed, the repaired tendons were covered with wet gauze (wet with normal saline), stored at 4oC overnight, and subjected to biomechanical testing the following morning.

â&#x20AC;&#x2122; METHODS

Load-to-failure Tests The tendon samples were 50 mm in length with the repair site in the middle. The ends of the tendons were clamped 15 mm from the repair site in a testing machine (Model 5567A, Instron, UK). Bluehill software (Instron, UK) was used to set and record the test parameters (Figure 2). Each tendon was subjected to a preload of 1.0 N and then elongated at a constant velocity of 10 mm/min until the suture site had completely ruptured. The load-to-failure tests were recorded using a camera. During the test, the gap at the repair site was measured using a vernier caliper. In this study, 2-mm gap load was defined as the tensile force needed to produce a 2-mm gap at the repair site. The ultimate load was recorded as the peak force during the stretching process. A computer was used to draw load-displacement curves.

Tendon Harvest and Repair Porcine flexor tendons are similar in structure and diameter to human flexor tendons and are thus appropriate for use in biomechanical studies. Eighteen porcine flexor digitorum profundus tendons (circumference: 8-10 mm; approximate weight: 300-400 g) were harvested from the hind legs of adult pigs (Dalian Chuming Co., Ltd.). This study was approved by the Institutional Animal Care and Use Committee of Dalian Medical University. All operations were performed in accordance with international guidelines regarding the care and treatment of experimental animals. The tendons were randomly divided into three groups (with six tendons in each group). In this experiment, 4-0 nylon (Weigao, China) was used for core sutures, and 6-0 nylon (Weigao, China) was used for peritendinous sutures. The middle of each tendon was transected with a scalpel and then repaired using one of the following techniques: (A) modified Kessler suture with peritendinous suture (MK); (B) interlock suture with peritendinous suture (IS); and (C) modified Kessler-loop lock suture with peritendinous suture (MKL). Details regarding these techniques are presented in Figure 1. During the Kessler suture technique, sutures were tightened to provide proper tension in the first locking loop. A final range of 10-12 mm was maintained for stitches on both sides. Three surgical knots were used for all suture knots, which consisted of six alternating positive and negative single knots. All tendon repairs were performed by a single experienced surgeon within one day. To minimize tissue injury, the surgeon performed the tendon repairs under an operating microscope with a magnification of 10 . Times required for suturing were recorded for each group.

Statistical Analysis Stata 12.0 (Computer Sciences Corporation, US) was used for statistical analysis. ANOVA followed by post hoc Tukey tests was used to evaluate differences in2-mm gap load, ultimate failure load, gap at failure, and suture time among the three groups. po0.05 was regarded as indicative of statistical significance.

â&#x20AC;&#x2122; RESULTS Two-millimeter Gap Load Two-millimeter gap load is an important index for evaluating resistance strength after tendon repair. During tendon healing, the formation of a gap of 2 mm or more increases the risks of tendon adhesion and impacts tendon healing. As shown in Tables 1 and 3 and Figure 3, compared with the MK technique, both the IS technique (p=0.011) and the

Figure 1 - Various suture techniques used for mechanical testing. A. Modified Kessler suture with peritendinous suture (MK); B. Interlock suture with peritendinous suture (IS); C. Modified Kessler-loop lock suture with peritendinous suture (MKL).

583

A modified Kessler-loop lock suture technique Yang W et al.

CLINICS 2017;72(9):582-587

Figure 2 - Porcine tendons repaired using various suture techniques. A. Modified Kessler with peritendinous suture (MK); B. Interlock suture with peritendinous suture (IS); C. Modified Kessler-loop lock suture with peritendinous suture (MKL); D. Repaired porcine tendon in mechanical testing; E. Suture failure in mechanical testing; F. Load-deformation curve.

Failure Profile

Table 1 - The 2-mm gap loads and ultimate failure loads for three suture techniques. Group

MK IS MKL

2-mm Gap Load (N)

As shown in Table 4, suture breakage always occurred in the IS group. The breakage locations were usually at the gap site in the peritendinous suture, although a small portion of ruptures occurred at the root of a knot. Compared with the MK and MKL groups, the IS group had less suture pullout and knot failure but a higher risk of suture breakage.

Ultimate Failure Load (N)

Mean

39.1 58.8 59.4

11.6 15.9 5.8

43.4 72.6 63.4

9.6 7.7 7.2

Gap at Failure As shown in Tables 2 and 3 and Figure 4, the largest gap at failure was observed in the MKL group, whereas statistically similar findings were obtained for the MK and IS groups (p=0.428).

Table 2 - The gaps at failure and times required for three suture techniques. Group

MK IS MKL

Gap at Failure (mm)

Time Required (min)

Mean

Time Required for the 3Suture Techniques

9.8 10.7 12.4

1.0 0.7 1.5

11.2 15.4 14.2

2.5 1.3 1.8

As shown in Tables 2 and 3 and Figure 5, the MK technique was faster to perform than the IS technique (p=0.011). There was no significant difference in suturing time between the IS and MKL groups (p=0.577).

â&#x20AC;&#x2122; DISCUSSION

MKL technique (p=0.028) could produce significantly improved 2-mm gap loads. The IS and MKL techniques did not significantly differ with respect to 2-mm gap load (p=0.851).

The early mobilization of reconstructed flexor tendons is highly significant for tendon recovery. By establishing stress orientation, early active movement can promote the regular arrangement of collagen fibers and break adhesion between the impaired tendon and surrounding tissues. Various suture techniques, such as the MK, double-strand Kessler suture, Silfverskiold (11), Strickland (12), cross-locked cruciate suture (13), Tsuge (14), and combined Kessler-Tsuge (15) techniques, have been used in clinical practice. However, research has demonstrated that these approaches produce different outcomes. Early active mobilization after tendon repair imposes high load at the repair site and requires resistance strength against a 2-mm gap load of more than

Ultimate Failure Load Ultimate failure load is another essential indicator for describing the biomechanical properties of tendon repair. As shown in Tables 1 and 3 and Figure 3, compared with the MK technique, both the IS technique (po0.001) and the MKL technique (p=0.006) could produce significantly improved ultimate failure loads. However, the IS and MKL techniques did not significantly differ with respect to ultimate failure load (p=0.220).

584

A modified Kessler-loop lock suture technique Yang W et al.

CLINICS 2017;72(9):582-587

Table 3 - p-values for comparisons between flexor tendon suture techniques. Comparison

2-mm Gap Load

Ultimate Failure Load

Gap at Failure

Time Required

MK vs. IS MK vs. MKL IS vs. MKL

0.011 0.028 0.851

o0.001 0.006 0.220

0.428 0.008 0.075

0.011 0.070 0.577

Figure 3 - Ultimate failure loads and 2-mm gap loads for different suture techniques. Compared with the MK technique, the IS and MKL techniques resulted in significantly greater 2-mm gap loads and ultimate failure loads. Error bars represent the SD from five porcine flexor tendons. *po0.05 vs. MK.

Table 4 - Failure modes for three suture techniques. Group MK (n=6) IS (n=6) MKL (n=6)

Suture Breakage

Suture Pullout

Knot Failure

n=4 n=6 n=4

n=4 n=2 n=4

n=3 n=1 n=2

stitches running through the tendon fibers, the IS technique is based on the grasping loops used in the Kessler suture technique combined with locking loops in the tendonâ&#x20AC;&#x2122;s core. However, the crossing shear force in the locking suture of the IS pattern increases the risk of suture breakage. If preload is insufficient during tendon suturing, the rotational stress caused by loose sutures typically leads to twisting and instability of the injured tendon ends; this phenomenon impacts endogenous tendon healing. Therefore, the IS technique is more suitable for experienced surgeons and requires the use of high-quality sutures to withstand tensile forces. In the Kessler-loop lock suture technique, the tendon is locked by friction between loop sutures and the tendon, with the objective of withstanding greater tensile force and preventing gap formation (20). This approach requires fewer knots than the MK technique. The cross-linked collagen fibers in the tendon generate upward force on the suture, reducing tendon slippage and splitting. The low tension between the injured ends of the tendon may be associated with better tendon healing, although further research is necessary to confirm this possibility. Therefore, we altered the double-strand Kessler-loop lock suture approach by utilizing single-strand sutures and performed suturing under an operating microscope with a magnification of 10 . This approach is easier for less experienced surgeons to understand and accept. Since this approach does not require special sutures, it is relatively readily applied in clinical practice. Nevertheless, there are certain limitations of our study. This investigation involved the use of devitalized porcine

30 N (16). Certain of the most commonly used suture methods (17) do not allow the repaired tendon to withstand over 30 N of tension. Although other repair techniques (18) can provide adequate strength, these techniques require complex surgical procedures or affect blood supply and tendon healing. Suture methods for flexor tendons mainly involve grasping loops and locking loops. Compared with grasping loops, locking loops improve the strength and gap resistance of the tendon repair by tightly holding tendon bands together. It has also been reported that a grasping loop-locking loop combination can maximize the advantages of both suture techniques and dramatically improve the biomechanical properties achieved in tendon repair (19). In the current work, both the IS and MKL suture techniques exhibited the advantages of locking and grasping loops and produced significantly improved 2-mm gap loads and ultimate tensile strengths relative to the MK technique. The IS and MKL approaches did not significantly differ with respect to 2-mm gap load, ultimate failure load, gap at failure, or time required for suturing. However, different biomechanical properties are observed for the IS and MKL techniques. To withstand greater tensile force with an intra-tendinous locking structure and

585

A modified Kessler-loop lock suture technique Yang W et al.

CLINICS 2017;72(9):582-587

Figure 4 - Gaps at failure for different suture techniques. The largest gap at failure was seen in the MKL group (12.4±1.5 mm), whereas gaps at failure were similar for the MK and IS techniques (9.8±1.0 mm vs. 10.7±0.7 mm, respectively). *po0.05 vs. MK.

Figure 5 - Times required for different suture techniques. Less time was required for the MK technique (11.2±2.5 min) than for the IS technique. There was no significant difference between the times required for the IS and MKL techniques. *po0.05 vs. MK.

flexor tendons that may have different biomechanical properties than living tissues. However, in the future, our results could be evaluated in animal models and via clinical studies. Our approach was compared with only the Kessler and interlock suture techniques; this comparison could be extended to other representative 4-strand suture techniques, such as the double-strand Kessler and Strickland approaches.

fundus tendons. Dong Y and Shang Y contributed to writing the manuscript. Zhang T directed the work. All authors read and approved the ﬁnal version of the manuscript.

’ REFERENCES 1. Gelberman RH, Amifl D, Gonsalves M, Woo S, Akeson WH. The influence of protected passive mobilization on the healing of flexor tendons: a biochemical and microangiographic study. Hand.1981;13(2):120-8, http:// dx.doi.org/10.1016/S0072-968X(81)80051-4. 2. Singh R, Rymer B, Theobald P, Thomas PB. A Review of Current Concepts in Flexor Tendon Repair: Physiology, Biomechanics, Surgical Technique and Rehabilitation. Orthop Rev (Pavia). 2015;7(4):6125, http://dx.doi. org/10.4081/or.2015.6125. 3. Lawrence TM, Davis TR. A biomechanical analysis of suture materials and their influence on a four-strand flexor tendon repair. J Hand Surg Am. 2005;30(4):836-41, http://dx.doi.org/10.1016/j.jhsa.2005.03.011. 4. Diao E, Hariharan JS, Soejima O, Lotz JC. Effect of peripheral suture depth on strength of tendon repairs. J Hand Surg Am. 1996;21(2):234-9, http:// dx.doi.org/10.1016/S0363-5023(96)80106-7.

’ ACKNOWLEDGMENTS The authors thank Jianfeng Li for helping to perform mechanical tests in the School of Chemical Engineering (Dalian University of Technology).

’ AUTHOR CONTRIBUTIONS Yang W performed biomechanical testing, analyzed data and wrote the manuscript. Qiao D and Ren Y harvested porcine ﬂexor digitorum pro-

586

A modified Kessler-loop lock suture technique Yang W et al.

CLINICS 2017;72(9):582-587

13. Wada A, Kubota H, Hatanaka H, Hotokezaka S, Miura H, Iwamoto Y. The mechanical properties of locking and grasping suture loop configurations in four-strand core suture techniques. J Hand Surg Br. 2000;25(6):548-51, http://dx.doi.org/10.1054/jhsb.2000.0441. 14. Hatanaka H, Manske PR. Effect of suture size on locking and grasping flexor tendon repair techniques. Clin Orthop Relat Res. 2000;(375): 267-74. 15. Renner C, Corella F, Fischer N. Biomechanical evaluation of 4-strand flexor tendon repair techniques, including a combined Kessler-Tsuge approach. J Hand Surg Am. 2015;40(2):229-35, http://dx.doi.org/10.1016/ j.jhsa.2014.10.055. 16. Gan AW, Neo PY, He M, Yam AK, Chong AK, Tay SC. A biomechanical comparison of 3 loop suture materials in a 6-strand flexor tendon repair technique. J Hand Surg Am. 2012;37(9):1830-4, http://dx.doi.org/10.1016/ j.jhsa.2012.06.008. 17. Jiahao L, Haisheng Q, Jianwei W, Chuanpu L. Biomechanical Study on a new method of suturing the tendon. Zhejiang Journal of Traumatic Surgery. 2015;20(4):653-5. 18. Lister GD, Kleinert HE, Kutz JE, Atasoy E. Primary flexor tendon repair followed by immediate controlled mobilization. J Hand Surg Am. 1977; 2(6):441-51, http://dx.doi.org/10.1016/S0363-5023(77)80025-7. 19. Nelson GN, Potter R, Ntouvali E, Silva MJ, Boyer MI, Gelberman RH, et al. Intrasynovial flexor tendon repair: a biomechanical study of variations in suture application in human cadavera. J Orthop Res. 2012;30(10): 1652-9, http://dx.doi.org/10.1002/jor.22108. 20. Lotz JC, Hariharan JS, Diao E. Analytic model to predict the strength of tendon repairs. J Orthop Res. 1998;16(4):399-405, http://dx.doi.org/10.1002/ jor.1100160402.

5. Wade PJ, Muir IF, Hutcheon LL. Primary flexor tendon repair: the mechanical limitations of the modified Kessler technique. J Hand Surg Br. 1986;11(1):71-6, http://dx.doi.org/10.1016/0266-7681(86)90018-5. 6. Lee H. Double loop locking suture: a technique of tendon repair for early active mobilization. Part I: Evolution of technique and experimental study. J Hand Surg Am.1990;15(6): 945-52, http://dx.doi.org/10.1016/ 0363-5023(90)90021-I. 7. Robertson GA, al-Qattan MM. A biomechanical analysis of a new interlock suture technique for flexor tendon repair. J Hand Surg Br. 1992;17(1): 92–3, http://dx.doi.org/10.1016/0266-7681(92)90020-3. 8. Tahmassebi R, Peltz TS, Haddad R, Scougall P, Gianoutsos M, Walsh W. The Interlocking Modification of the Cross Locked Cruciate Tendon Repair (Modified Adelaide Repair): A Static and Dynamic Biomechanical Assessment. J Hand Microsurg. 2015;7(1):6–12, http://dx.doi.org/10.1007/ s12593-014-0144-4. 9. Qinghe L, Zuoshan Z, Kai Z. Semi-Knot Locking Suture (SKLS) and It Experimental Study. The Journal of Bone and Joint Injury. 2003;18(10):692-4. 10. Le SV, Chiu S, Meineke RC, Williams P, Wongworawat MD. Number of suture throws and its impact on the biomechanical properties of the fourstrand cruciate locked flexor tendon repair with FiberWire. J Hand Surg Eur. 2012;37(9):826–31, http://dx.doi.org/10.1177/1753193412447503. 11. Dwyer CL, Dominy DD, Cooney TE, Englund R, Gordon L, Lubahn JD. Biomechanical comparison of double grasping repair versus cross-locked cruciate flexor tendon repair. Hand (N Y). 2015;10(1):16-22. 12. Iamaguchi RB, Villani W, Rezende MR, Wei TH, Cho AB, Santos GB, et al. Biomechanical comparison of the four-strand cruciate and Strickland techniques in animal tendons. Clinics. 2013;68(12):1543-7, http://dx.doi. org/10.6061/clinics/2013(12)11.

587

No Rebouças, pequenos eventos

tornam-se grandes.

Seja pequeno, médio ou grande, o Rebouças tem o espaço ideal para todo o tipo de evento. A melhor logística e infraestrutura de apoio maximizam os recursos operacionais e potencializam os encontros científicos, promocionais, institucionais ou culturais. O Centro de Convenções Rebouças é isso: um dos mais bem planejados e tradicionais espaços receptivos do País, dimensionado para sediar congressos, convenções, exposições, seminários, simpósios, cursos, lançamento de produtos, entre outros, bem no coração da cidade de São Paulo.

Atualmente possui 8 ambientes climatizados e modernizados que atendem até 1.200 participantes, além de uma equipe preparada e empenhada em acompanhar cerca de 280 eventos por ano. EM BREVE, estará ampliando suas instalações, que permitirá o dobro da capacidade de público.

Av. Rebouças, 600 - 05402-000 - São Paulo - Brasil - Tel.: 55 11 3898-7850 / Fax: 55 11 3898-7878 - reboucas@hcnet.usp.br