Organization
Founded in 1932, a bimonthly publication of the Associação Paulista de Medicina
e-mail: [email protected]
Editors: Paulo Manuel Pêgo-Fernandes and Álvaro Nagib Atallah.
Editorial advisor: Rachel Riera.
Editorial assistant: Marina de Britto.
Scientific journalist and editor: Patrícia Logullo (MTB: 2-6.152).
Editorial auxiliary: Joyce de Fátima Silva Nakamura.
Associate editors: Adriana Seber, Alexander Wagner Silva de Souza, Antonio José
Gonçalves, Aytan Miranda Sipahi, Cristina Muccioli, Delcio Matos, Domingo Marcolino
Braile, Edina Mariko Koga da Silva, Edmund Chada Baracat, Elcio dos Santos Oliveira
Vianna, Emmanuel de Almeida Burdmann, Fernando Antonio de Almeida, Fernando
Ferreira Costa, Flávio Faloppa, Heráclito Barbosa de Carvalho, José Antônio Rocha
Gontijo, José Carlos Costa Baptista-Silva, José Roberto Lapa e Silva, Júlio César Rodrigues
Pereira, Laércio Joel Franco, Marilza Vieira Cunha Rudge, Milton de Arruda Martins,
Moacir Fernandes de Godoy, Olavo Pires de Camargo, Sergio Tufik, Soubhi Kahhale,
Walter José Gomes.
Proofreading: David Elliff.
Desktop publishing: Zeppelini Editorial (www.zeppelini.com.br).
Listed in: Medline, Lilacs, SciELO, Science Citation Index Expanded and Journal Citation
Reports/Sciences Edition (impact factor 0.588) and EBSCO publishing.
International Board: Alexandre Wagner Silva de Souza (University Medical Center
Groningen, Groningen, Netherlands), Angeles R. Badell (Faculty of Medicine,
University of Barcelona, Barcelona, Spain), Charles J. Menkes (Cochin Hospital, Paris,
France), José Fragata (Hospital Cuf Infant Santo, Lisbon), Luiz Dratcu (Guy’s Hospital,
London, and Maudsley NHS Trust, York Clinic, London), Marcelo Cypel (University
Health Network, Toronto, Canada), Karla Soares-Weiser (Enhance Reviews Ltd,
Wantage, United Kingdom), Tirone E. David (Toronto General Hospital, Toronto,
Canada), Mário Viana de Queiroz (Hospital de Santa Maria, Lisbon), Wadih Arap (MD
Anderson Cancer Center, University of Texas, Houston, United States), Wellington
Cardoso (Boston University, Boston, United States).
• All articles published, including editorials and letters, represent the opinions of the
authors and do not reflect the official policy of the Associação Paulista de Medicina or the
institution with which the authors are affiliated, unless this is clearly specified.
• All rights reserved. No part of this publication may be reproduced or transmitted in any
form or by any means, electronic or mechanical, including photocopy, recording, or any
information storage and retrieval system, without permission in writing from the publisher.
Copyright © 2014 by Associação Paulista de Medicina.
• SPMJ website: access to the entire São Paulo Medical Journal/Revista Paulista de
Medicina website is free to all. We will give at least six months notice of any change in
policy. SPMJ printed version: six issues/year; 1 volume/year, beginning on first Thursday in
January.
• One-year subscription for the year 2014: individual US$ 165; institutional US$ 230.
Scientific Council
Abrão Rapoport – Hospital Heliópolis, São Paulo
Adriana Costa e Forti – Faculdade de Medicina, Universidade Federal do Ceará
Alexandre Fogaça Cristante – Faculdade de Medicina da Universidade de São Paulo
Álvaro Nagib Atallah – Escola Paulista de Medicina, Universidade Federal de São Paulo
Auro del Giglio – Faculdade de Medicina da Fundação ABC
Carlos Alberto Morais Sá – Universidade do Rio de Janeiro - UNIRIO
Carmen Cabanelas Pazos de Moura – Instituto Carlos Chagas Filho, Universidade Federal do
Rio de Janeiro
Cármino Antonio De Souza – Faculdade de Ciências Médicas, Universidade Estadual de Campinas
Dario Birolini – Faculdade de Medicina, Universidade de São Paulo
Eduardo Katchburian – Escola Paulista de Medicina, Universidade Federal de São Paulo
Eduardo Maia Freese de Carvalho – Faculdade de Medicina, Universidade Federal de
Pernambuco, Centro de Pesquisas Aggeu Magalhães - CpqAM/FIOCRUZ.
Egberto Gaspar de Moura – Instituto de Biologia Roberto Alcantara Gomes, Universidade
Estadual do Rio de Janeiro
Eliézer Silva – Hospital Israelita Albert Einstein, São Paulo
Emílio Antonio Francischetti - Faculdade de Medicina da Universidade Estadual do Rio de Janeiro
Emmanuel de Almeida Burdmann – Faculdade de Medicina de São José do Rio Preto
Fabio Bessa Lima – Instituto de Ciências Biomédicas, Universidade de São Paulo
Florence Kerr-Corrêa – Faculdade de Medicina de Botucatu, Universidade Estadual de São Paulo
Francisco José Penna – Faculdade de Medicina Universidade Federal de Minas Gerais
Geraldo Rodrigues de Lima – Escola Paulista de Medicina, Universidade Federal de São Paulo
Irineu Tadeu Velasco – Faculdade de Medicina da Universidade de São Paulo
João Renato Rebello Pinho – Instituto Adolfo Lutz, Secretaria de Estado da Saúde de São Paulo
Joel Spadaro – Faculdade de Ciências Médicas de Botucatu, Universidade Estadual de São Paulo
Jorge Pinto Ribeiro – Faculdade de Medicina, Universidade Federal do Rio Grande do Sul
Jorge Sabbaga – Hospital Alemão Oswaldo Cruz, São Paulo
José Antonio Marin-Neto – Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo
José Carlos Nicolau – Instituto do Coração, Universidade de São Paulo
José Geraldo Mill – Faculdade de Medicina, Universidade Federal do Espírito Santo
José Mendes Aldrighi – Faculdade de Saúde Pública, Universidade de São Paulo
José Roberto Lapa e Silva – Instituto de Doenças do Tórax, Universidade Federal do Rio de Janeiro
Leopoldo Soares Piegas – Instituto Dante Pazzanese de Cardiologia, São Paulo
Luiz Jacintho da Silva – Faculdade de Ciências Médicas, Universidade Estadual de Campinas
Luiz Paulo Kowalski – Hospital AC Camargo, São Paulo
Márcio Abrahão – Escola Paulista de Medicina, Universidade Federal de São Paulo
Maria Inês Schmidt – Faculdade de Medicina, Universidade Federal do Rio Grande do Sul
Maurício Mota de Avelar Alchorne – Escola Paulista de Medicina, Universidade Federal de
São Paulo
Mauro Schechter – Hospital Universitário Clementino Fraga Filho, Universidade Federal do
Rio de Janeiro
Milton de Arruda Martins – Faculdade de Medicina, Universidade de São Paulo
Moysés Mincis – Faculdade de Ciências Médicas de Santos
Nelson Hamerschlak – Hospital Israelita Albert Einstein, São Paulo
Noedir Antônio Groppo Stolf – Faculdade de Medicina, Universidade de São Paulo
Pérsio Roxo Júnior – Faculdade de Medicina de Ribeirão Preto
Raul Cutait – Hospital Sírio-Libanês, São Paulo
Raul Negrão Fleury – Instituto Lauro de Souza Lima, Coordenadoria dos Institutos de Pesquisa
da Secretaria de Saúde de São Paulo
Raul Marino Junior – Faculdade de Medicina, Universidade de São Paulo
Ricardo Brandt de Oliveira – Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo
Roberto A. Franken – Faculdade de Ciências Médicas da Santa Casa de Misericórdia de São Paulo
Ruy Laurenti – Faculdade de Saúde Pública, Universidade de São Paulo
Soubhi Kahhale – Faculdade de Medicina, Universidade de São Paulo
Wilson Roberto Catapani – Faculdade de Medicina do ABC, Santo André
Wilson Cossermelli – Reclin Reumatologia Clínica, São Paulo
Diretoria Executiva da Associação Paulista de Medicina (Triênio 2011-2014)
Presidente: Florisval Meinão
1º Vice-Presidente: Roberto Lotfi Júnior
2º Vice-Presidente: Donaldo Cerci da Cunha
3º Vice-Presidente: Paulo De Conti
4º Vice-Presidente: Akira Ishida
Secretário Geral: Paulo Cezar Mariani
1º Secretário: Ruy Y. Tanigawa
Diretor Administrativo: Lacildes Rovella Júnior
Diretor Administrativo Adjunto: Roberto de Mello
1º Diretor de Patrimônio e Finanças: Murilo Rezende Melo
2º Diretor de Patrimônio e Finanças: João Márcio Garcia
Diretor Científico: Paulo Manuel Pêgo Fernandes
Diretor Científico Adjunto: Álvaro Nagib Atallah
Diretor de Defesa Profissional: João Sobreira de Moura Neto
Diretor de Defesa Profissional Adjunto: Marun David Cury
Diretor de Comunicações: Renato Françoso Filho
Diretor de Comunicações Adjunto: Leonardo da Silva
Diretor de Marketing: Nicolau D´Amico Filho
Diretor de Marketing Adjunto: Ademar Anzai
Diretora de Eventos: Mara Edwirges Rocha Gândara
Diretora de Eventos Adjunto: Regina Maria Volpato Bedone
Diretor de Tecnologia de Informação: Desiré Carlos Callegari
Diretor de Tecnologia de Informação Adj.: Antonio Carlos Endrigo
Diretor de Previdência e Mutualismo: Paulo Tadeu Falanghe
Diretor de Previdência e Mutualismo Adjunto: Clóvis Francisco Constantino
ii
Sao Paulo Med J. 2014; 132(2):i-ii
Diretor Social: Alfredo de Freitas Santos Filho
Diretor Social Adjunto: Nelson Álvares Cruz Filho
Diretora de Responsabilidade Social: Denise Barbosa
Diretora de Responsabilidade Social Adjunta: Yvonne Capuano
Diretor Cultural: Guido Arturo Palomba
Diretor Cultural Adjunto: Carlos Alberto Monte Gobbo
Diretor de Serviços aos Associados: José Luiz Bonamigo Filho
Diretor de Serviços aos Associados Adjunto: João Carlos Sanches Anéas
Diretor de Economia Médica: Tomás Patrício Smith-Howard
Diretor de Economia Médica Adjunto: Jarbas Simas
1º Diretor Distrital: Airton Gomes
2º Diretor Distrital: Arnaldo Duarte Lourenço
3º Diretor Distrital: Lauro Mascarenhas Pinto
4º Diretor Distrital: Wilson Olegário Campagnone
5º Diretor Distrital: José Renato dos Santos
6º Diretor Distrital: José Eduardo Paciência Rodrigues
7º Diretor Distrital: José Eduardo Marques
8º Diretor Distrital: Helencar Ignácio
9º Diretor Distrital: José do Carmo Gaspar Sartori
10º Diretor Distrital: Paulo Roberto Mazaro
11º Diretor Distrital: José de Freitas Guimarães Neto 12º Diretor Distrital: Marco Antônio Caetano
13º Diretor Distrital: Marcio Aguilar Padovani
14º Diretor Distrital: Wagner de Matos Rezende
Editorial
DOI: 10.1590/1516-3180.2014.1322817
Statistical significance and clinical significance
Significância estatística e significado clínico Alessandro Wasum MarianiI, Paulo Manuel Pêgo-FernandesII
Instituto do Coração (InCor), Hospital das Clínicas (HC), Faculdade de Medicina da Universidade de São Paulo (FMUSP), São Paulo, Brazil
MD. Thoracic Surgeon, Instituto do Coração
(InCor), Hospital das Clínicas (HC), Faculdade
de Medicina da Universidade de São Paulo
(FMUSP), São Paulo, Brazil.
I
II
MD, PhD. Associate Professor, Discipline
of Thoracic Surgery, Instituto do Coração
(InCor), Hospital das Clínicas (HC), Faculdade
de Medicina da Universidade de São Paulo
(FMUSP), São Paulo, Brazil.
Medicine is evolving faster than ever and the web and communication channels, among
other technological improvements, facilitate the capacity for knowledge generation and knowledge
diffusion, not only in the field of medicine but also for science in general. The form of
knowledge diffusion that is used most is the publication of scientific articles. In this modern
scenario, the ability to read and interpret medical articles is more than desirable: it is fundamental
for up-to-date medical practice.
One important issue is the critical judgment of any study conclusion. We are already used
to looking at the methods and results in order to evaluate whether they support the conclusion.
However, even a well-designed and properly conducted study with a statistical significant P-value
does not always imply real clinical significance. At first glance, we may ask ourselves: how could
this be possible? How could our perfect positive study not be useful in clinical practice?
There are some tricks and pitfalls that could explain this:
First of all, it needs to be clear what the concepts behind these terms are. The crude statistical significance associated with the P-value means that from the statistical point of view, the
study result was not due to chance. In other words, if we replicate the study, there is a probability lower than the defined critical value (e.g. for a P-value < 0.05, the probability will be less
than 5%) that the result will not be the same. On the other hand, the concept of clinical significance, also called clinical importance, can be summarized as a difference between two therapy
results that is large enough to justify changing the standard of care.1
The pitfall could be that, even if the study outcome is statistically supported, the difference
may be too small to lead to a decision to change the current clinical practice. One hypothetical
example would be a study to test a new drug for arterial hypertension that has a statistically significant P-value, e.g. P = 0.001. However, the blood pressure reduction is only 5 mmHg, which in
clinical practice does not justify adopting the new drug. The problem in this example was not
in the statistical design but in the setup of the outcome.
If the opposite occurs, i.e. a real clinical difference between groups that a significance test
fails to identify, we have a type II error (failure to reject a false null hypothesis). However, type
II errors can be predicted by conducting a power analysis prior to conducting the investigation.
An adequate sample size reduces occurrences of type II errors. On the other hand, occurrences
of type I error (incorrect rejection of a true null hypothesis) can be diminished by lowering the
alpha (meaning the level of significance at which the P-value will be compared).
One thing that could be useful for establishing adequate clinical significance is to evaluate the confidence interval (CI), which includes all values between the limits. CIs are most
frequently reported at the 95% confidence level, which means that there is a 95% chance that
the real mean difference is encapsulated within the upper and lower limits. If the CI includes
zero, this could be interpreted as evidence that the real difference between population means
is zero and that the treatment reported is not having any effect.2 However, the Guidelines for
Reporting Statistics from the American Physiological Society state the following: “if either
bound of the confidence interval is important from a scientific perspective, then the experimental effect may be large enough to be relevant”.3
Sao Paulo Med J. 2014; 132(2):71-2
71
Editorial | Mariani AW, Pêgo-Fernandes PM
Another method, which may be helpful, is the alternative approach of the number needed to treat (NNT), which was introduced
by Laupacis et al. in 1988. This method consists of summarizing the effect of treatment in terms of the number of patients that need to
be treated with the therapy in order to expect to prevent one adverse event.4 As pointed out by Cook and Sackett, the NNT is becoming widely used as a tool for therapeutic decision-making because it is easier to interpret than the arguably less intuitive probabilities.5
Ultimately, in order to choose among different treatments, clinical physicians have to consider not only the P-value of the latest
published paper, but also the magnitude of benefit of each treatment, side-effect profiles, direct and possibly indirect costs, patients’
preferences and even their own comfort with prescribing a new therapy.
This brings us to the conclusion that the ability to understand the statistics behind articles is not enough. Having a good notion of
what real clinical significance is or could be is crucial for correct interpretation of the modern medical literature. This ability to interpret clinical significance must come from the experience of clinical practice in association with an understanding of some research
concepts like study power, type I and type II errors, bias, confidence interval, treatment effect and number needed to treat.
REFERENCES
Address for correspondence:
1. Houle TT, Stump DA. Statistical significance versus clinical
Alessandro Wasum Mariani
significance. Semin Cardiothorac Vasc Anesth. 2008;12(1):5-6.
2. Stapleton C, Scott MA, Atkinson G. The ‘so what’ factor: statistical versus
clinical [corrected] significance. Int J Sports Med. 2009;30(11):773-4.
3. Curran-Everett D, Benos DJ. Guidelines for reporting statistics in
Hospital das Clínicas - Faculdade de Medicina da Universidade de São
Paulo (HC-FMUSP)
Av. Dr Enéas de Carvalho Aguiar, 33 — Bloco II Sala 9
journals published by the American Physiological Society. Am J
Cerqueira César — São Paulo (SP) — Brasil
Physiol Regul Integr Comp Physiol. 2004;287(2):R247-9.
CEP 05403-000
4. Laupacis A, Sackett DL, Roberts RS. An assessment of clinically
useful measures of the consequences of treatment. N Engl J Med.
1988;318(26):1728-33.
5. Cook RJ, Sackett DL. The number needed to treat: a clinically useful
measure of treatment effect. BMJ. 1995;310(6977):452-4.
Sources of funding: None
Conflict of interest: None
Date of first submission: December 27, 2013
Last received: February 3, 2014
Accepted: February 12, 2014
72
Instituto do Coração (InCor)
Sao Paulo Med J. 2014; 132(2):71-2
Tel. (+55 11) 2661-5248/2661-5000
E-mail: [email protected]
ORIGINAL ARTICLE
DOI: 10.1590/1516-3180.2014.1322495
Prevalence and risk factors for scrotal lesions/anomalies
in a cohort of Brazilian men ≥ 40 years of age
Prevalência e fatores de risco de lesões escrotais em uma
população de brasileiros com idade ≥ 40 anos
Frederico Ramalho RomeroI, Antonio Wilson RomeroII, Rui Manuel de Sousa Sequeira Antunes de AlmeidaIII,
Fernando Cesar de Oliveira JúniorIV, Renato Tambara FilhoV
Instituto Curitiba de Saúde (ICS), Curitiba, Paraná, Brazil
MD, MSc, PhD, Department of Surgery,
Discipline of Urology, Universidade Federal do
Paraná (UFPR), Curitiba, Paraná, Brazil.
I
II
MD. Urologist, Urology Service, Hospital
Policlínica, Cascavel, Paraná, Brazil.
MD, MSc, PhD. Coordinator, College of
Medicine, Faculdade Assis Gurgacz (FAG),
Cascavel, Paraná, Brazil.
III
IV
MD, MSc, Coordinator, Núcleo de Atenção à
Saúde, Instituto Curitiba de Saúde (ICS), Curitiba,
Paraná, Brazil.
MD, MSc, PhD. Professor, Department of
Surgery, Discipline of Urology, Universidade
Federal do Paraná (UFPR), Curitiba, Paraná, Brazil.
V
KEY WORDS:
Epidemiology.
Etiology [subheading].
Prevalence.
Risk factors.
Genital diseases, male.
PALAVRAS-CHAVE:
Epidemiologia.
Etiologia.
Prevalência.
Fatores de risco.
Doenças dos genitais masculinos.
ABSTRACT
OBJECTIVE: To estimate the prevalence of and risk factors for cutaneous lesions of the scrotum and
intrascrotal lesions/anomalies among men included in a prostatic cancer screening program in a Brazilian
metropolitan city.
DESIGN AND SETTING: Cross-sectional study, private outpatient healthcare service.
METHODS: 1731 men aged 40 years or over, participating in a prostate cancer screening program conducted by the municipal public employees’ healthcare system, underwent systematic urological assessment by a single examiner.
RESULTS: The prevalence of scrotal diseases in our sample was 44.7% (773/1731). Tinea cruris occurred in
203 (11.7%) of the participants, with higher risk among diabetics and lower prevalence among nonwhite
individuals; scrotal tinea in eight (0.5%), with higher risk among hypertensive men; subcutaneous nodules in 12 (0.7%), especially in individuals with low schooling level; hydrocele in 283 (16.4%), with higher
frequency among participants over the age of 60 years, diabetics or individuals with previous histories of
nonspecific urethritis; spermatoceles in 174 (10.1%), with greater prevalence among individuals over the
age of 60 years or diabetics, and lower frequency among individuals who underwent vasectomy; unilateral
testicular hypotrophy/atrophy in 167 (9.7%) and bilateral hypotrophy/atrophy in 93 (5.4%), both occurring
more frequently in individuals over the age of 60 years; absence of palpable testicles due to cryptorchidism in 7 (0.4%); and epididymitis/orchitis in 5 (0.3%), with higher prevalence among diabetics. No cases
of cancer were identified in this sample.
CONCLUSIONS: Scrotal diseases were highly prevalent in this population of Brazilian men.
RESUMO
OBJETIVO: Estimar a prevalência e fatores de risco de lesões cutâneas do escroto e de lesões/anomalias
intraescrotais entre participantes de programa de rastreamento para câncer de próstata em uma cidade
metropolitana brasileira.
TIPO DE ESTUDO E LOCAL: Estudo transversal, serviço privado de atendimento ambulatorial à saúde.
MÉTODOS: 1.731 homens com idade igual ou superior a 40 anos, participantes do programa de rastreamento de câncer de próstata conduzido pelo sistema de saúde dos funcionários públicos municipais,
foram submetidos à avaliação urológica sistemática por um único examinador.
RESULTADOS: A prevalência de doenças escrotais nossa amostra foi de 44.7% (773/1731). Tinea cruris
ocorreu em 203 (11,7%) dos participantes, com maior risco em diabéticos e menor prevalência em indivíduos não brancos; tinea escrotal em oito (0,5%), com maior risco em homens hipertensos; nódulos
subcutâneos em 12 (0,7%), especialmente em indivíduos com baixa escolaridade; hidrocele em 283
(16,4%), com maior frequência nos participantes com mais de 60 anos, diabetes ou história prévia de
uretrite inespecífica; espermatoceles em 174 (10,1%), com maior prevalência acima dos 60 anos de idade
ou com diabetes, e menor frequência naqueles submetidos a vasectomia; hipotrofia/atrofia testicular unilateral em 167 (9,7%) e hipotrofia/atrofia bilateral em 93 (5,4%), ambas ocorrendo mais frequentemente
nos indivíduos com mais de 60 anos; ausência de testículos palpáveis devido à criptorquidia em 7 (0,4%);
e epididimite/orquite em 5 (0,3%), com prevalência aumentada em diabéticos. Não foram identificados
casos de câncer nesta amostra.
CONCLUSÕES: As doenças escrotais foram altamente prevalentes nesta população de homens brasileiros.
Sao Paulo Med J. 2014; 132(2):73-9
73
ORIGINAL ARTICLE | Romero FR, Romero AW, Almeida RMSSA, Oliveira Júnior FC, Tambara Filho R
INTRODUCTION
Cutaneous lesions of the scrotum and intrascrotal lesions/anomalies are common findings in the adult male population. The prevalence of these lesions is difficult to determine accurately, given
that they are frequently oligosymptomatic and are often discovered incidentally during physical examination. The results differ
according to the patients’ ages, racial influences, geographic locations, comorbidities and socioeconomic status.1
The setting in which the study is conducted (based in the
population/community or hospital/clinic setting), the type of
study (retrospective or prospective) and the type of diagnostic assessment (clinical, laboratory or imaging-based) may also
influence prevalence levels.2
Epidemiological studies are important because they contribute towards taking an appropriate approach to the conditions, improving awareness, promoting educational practices
and preventive measures and expediting treatment. They may
also enable comparisons within and between countries, allow for
temporal variations between different ages and time periods, and
guide future research to evaluate the pathogenesis, etiology
and risk factors of these diseases.
OBJECTIVE
The objective of this paper was to report on the overall prevalence and risk factors of cutaneous lesions of the scrotum and
intrascrotal lesions/anomalies collected prospectively in a crosssectional study conducted in a Brazilian metropolitan city.
METHODS
Between December 2006 and April 2011, 1731 subjects were
included in this investigation. The participants were men aged
40 years or over who were undergoing outpatient evaluation as
part of a free prostate cancer screening program conducted by
the municipal public employees’ healthcare system (Curitiba
Institute of Health), which is a private institution directed
exclusively towards workers employed by the municipality of
Curitiba. The study protocol was reviewed and approved by the
Institutional Ethics Committee for Human Research (registry
number 2253.147/2010-06).
During evaluation, the participants were classified by a single examiner as white or nonwhite (including brown or black);
they answered a general questionnaire that asked about their age,
schooling level and personal histories of diabetes, arterial hypertension, nonspecific urethritis or vasectomy (Table 1); and were
offered a complete genital-pelvic examination.
The urological examination was performed in a standardized
manner, on all subjects in the supine position by the same
examiner. The scrotum was initially inspected for cutaneous lesions,
and was then palpated to detect any intrascrotal lesions/anomalies,
74
Sao Paulo Med J. 2014; 132(2):73-9
including hydrocele, spermatocele, testicular hypotrophy/atrophy,
testicular cancer or absence of a palpable testis. Results were
collected during the first evaluation on each patient, and were not
computed more than once for each participant.
The outcomes of interest included the prevalence of scrotal
lesions/anomalies, and the relative risk (RR) and 95% confidence
intervals (95% CI) of the lesions/anomalies, according to age (≥ 60
versus < 60 years), schooling level (elementary school or lower versus high school or higher), skin color (nonwhite versus white), personal history of diabetes, arterial hypertension, nonspecific urethritis and vasectomy (yes versus no, for all of these last four categories).
Statistics for categorical variables were calculated using
Fisher’s exact test or Pearson’s chi-square test, as appropriate,
and statistical significance was defined as situations in which
P < 0.05 or when the 95% CI did not include the null hypothesis
(95% CI ≠ 1.00).
RESULTS
Cutaneous lesions of the scrotum were identified in 230 participants (13.3%, 230/1731). Mycosis was present in 211 individuals
Table 1. Demographic and clinical characteristics of study
population
Characteristic
Age
< 60 years
≥ 60 years
Total
Schooling level
Elementary school or lower
High school or higher
Missing data
Skin color
White
Nonwhite (black or brown)
Missing data
Arterial hypertension
Yes
No
Missing data
Diabetes mellitus
Yes
No
Missing data
Past history of nonspecific urethritis
Yes
No
Missing data
Past history of vasectomy
Yes
No
Missing data
Number
Percentage (%)
1352
379
1731
78.1
21.9
100.0
647
937
147
37.4
54.1
8.5
643
588
500
37.1
34.0
28.9
595
1133
3
34.4
65.4
0.2
180
1548
3
10.4
89.4
0.2
497
1012
222
28.7
58.5
12.8
210
1518
3
12.1
87.7
0.2
Prevalence and risk factors for scrotal lesions/anomalies in a cohort of Brazilian men ≥ 40 years of age | ORIGINAL ARTICLE
(91.7%, 211/230), and 203 of these cases were Tinea cruris (88.3%
of all cutaneous lesions on the scrotum [203/230] and 11.7% of all
participants [203/1731]). Fifty-nine participants (29.0%, 59/203)
with Tinea cruris presented associated balanoposthitis. Tinea
cruris occurred more commonly in participants with diabetes
(RR = 1.55, P < 0.05), and less frequently in those of nonwhite skin
color (RR = 0.68, P < 0.05) (Table 2). Scrotal tinea was identified in
eight men (3.5% [8/230] or 0.5% [8/1731]), and two of these (25%,
2/8) presented associated balanoposthitis or Tinea cruris. Scrotal
tinea was more prevalent in men with arterial hypertension than in
those without arterial hypertension (RR = 5.71, P < 0.05) (Table 2).
Other cutaneous lesions encountered on the scrotum included
subcutaneous nodules (epidermal cysts) in 12 participants
(5.2% [12/230] or 0.7% [12/1731]); psoriasis in one individual (0.4%
[1/230] or 0.06% [1/1731]); and a nonspecific papillary rash in one
individual (0.4% [1/230] or 0.06% [1/1731]). The risk-adjusted
prevalence of these lesions demonstrated that subcutaneous
nodules were more common in subjects with low schooling level
(RR = 4.34, P < 0.05), and less prevalent in those with a history of
nonspecific urethritis (RR = 0.19, P < 0.05) (Table 2).
Intrascrotal lesions/anomalies were identified in 638 participants (36.9%, 638/1731). Hydrocele was present in 283 participants (44.4% of all intrascrotal lesions [283/638], and 16.4%
overall [283/1731]). It was mild in 269 men (95.0%, 269/283),
and unilateral in 163 cases (57.6%, 163/283), of which 94 (57.7%,
94/163) were on the right side. Hydrocele was identified more frequently in participants older than 60 years (RR = 1.74, P < 0.05),
with diabetes (RR = 2.01, P < 0.05), or with a history of nonspecific urethritis (RR = 1.35, P < 0.05) (Table 2).
Spermatoceles were identified in 174 participants (27.3% of
the intrascrotal lesions [174/638], and 10.1% overall [174/1731]).
They occurred more frequently in participants older than
Table 2. Prevalence of scrotal diseases and relative risks (RR) and 95% confidence intervals (CI) of the lesions/anomalies, according to age,
schooling level, skin color and histories of diabetes mellitus, arterial hypertension, nonspecific urethritis and vasectomy
Age:
≥ 60 vs. < 60;
RR;
95% CI
Tinea cruris
Scrotal tinea
Subcutaneous
nodules
Hydrocele
Spermatocele
Unilateral
testicular
hypotrophy/atrophy
Bilateral
testicular
hypotrophy/atrophy
Epididymitis/orchitis
Cryptorchidism
13.2 vs. 11.3
1.16
0.86-1.57
0.5 vs. 0.4
1.19
0.24-5.87
0.8 vs. 0.7
1.19
0.32-4.37
24.5 vs. 14.1
(*) 1.74
1.40-2.17
15.0 vs. 8.8
(*) 1.70
1.27-2.29
14.2 vs. 9.1
(*) 1.57
1.15-2.14
8.5 vs. 5.3
(*) 1.62
1.06-2.48
0.5 vs. 0.2
2.38
0.40-14.18
0.5 vs. 0.4
1.43
0.28-7.33
Schooling level:
elementary school
vs. high school or
higher;
RR;
95% CI
11.7 vs. 11.6
1.01
0.77-1.33
0.3 vs. 0.4
0.72
0.13-3.94
1.4 vs. 0.3
(*) 4.34
1.18-15.99
17.7 vs. 15.7
1.13
0.91-1.41
11.3 vs. 9.8
1.15
0.86-1.53
10.7 vs. 9.6
1.12
0.82-1.52
6.7 vs. 5.5
1.21
0.81-1.83
0.0 vs. 0.3
0.00
0.00-NaN
0.6 vs. 0.2
2.90
0.53-15.77
Skin color:
nonwhite vs.
white;
RR;
95% CI
Diabetes:
yes vs. no;
RR;
95% CI
Arterial
hypertension:
yes vs. no;
RR;
95% CI
9.5 vs. 14.0
(*) 0.68
0.50-0.93
0.2 vs. 0.3
0.55
0.05-6.01
1.2 vs. 0.8
1.53
0.49-4.80
15.3 vs. 16.2
0.95
0.73-1.23
10.2 vs. 9.0
1.13
0.80-1.59
8.8 vs. 12.2
0.72
0.51-1.02
5.9 vs. 6.2
0.97
0.61-1.54
0.0 vs. 0.5
0.00
0.00-NaN
0.7 vs. 0.3
2.19
0.40-11.90
17.2 vs. 11.1
(*) 1.55
1.09-2.20
1.1 vs. 0.4
2.87
0.58-14.10
0.6 vs. 0.7
0.78
0.10-6.02
33.0 vs. 16.5
(*) 2.01
1.62-2.49
16.1 vs. 9.4
(*) 1.71
1.19-2.47
11.3 vs. 10.1
1.12
0.71-1.76
7.5 vs. 5.8
1.29
0.72-2.31
1.7 vs. 0.1
(*) 12.90
2.17-76.69
0.0 vs. 0.5
0.00
0.00-NaN
13.5 vs. 10.9
1.24
0.95-1.61
1.0 vs. 0.2
(*) 5.71
1.16-28.22
0.7 vs. 0.7
0.95
0.29-3.15
18.3 vs. 15.3
1.19
0.96-1.48
11.9 vs. 9.2
1.30
0.98-1.73
10.3 vs. 10.2
1.01
0.75-1.37
5.8 vs. 6.1
0.95
0.63-1.45
0.3 vs. 0.3
1.27
0.21-7.58
0.5 vs. 0.4
1.43
0.32-6.36
History of
nonspecific
urethritis;
yes vs. no;
RR;
95% CI
13.3 vs. 11.1
1.20
0.91-1.60
0.2 vs. 0.5
0.41
0.05-3.48
0.2 vs. 1.1
(*) 0.19
0.02-1.43
19.9 vs. 14.8
(*) 1.35
1.07-1.69
11.1 vs. 9.6
1.16
0.85-1.58
11.5 vs. 9.5
1.21
0.88-1.66
6.3 vs. 6.1
1.04
0.67-1.61
0.4 vs. 0.2
2.04
0.29-14.41
0.4 vs. 0.5
0.81
0.16-4.18
Vasectomy:
yes vs. no; RR;
95% CI
11.0 vs. 11.9
0.92
0.61-1.39
0.0 vs. 0.5
0.00
0.00-NaN
0.5 vs. 0.7
0.66
0.09-5.06
12.9 vs. 16.8
0.77
0.53-1.11
5.7 vs. 10.7
(*) 0.77
0.53-1.11
9.3 vs. 10.3
0.90
0.57-1.41
2.6 vs. 6.4
(*) 0.41
0.17-0.99
0.5 vs. 0.3
1.81
0.20-16.09
0.5 vs. 0.4
1.20
0.15-9.96
*Statistically significant differences (Fisher’s exact test or Pearson’s chi-square test); RR = relative risk; CI = confidence interval; NaN = not a number; vs. = versus.
Sao Paulo Med J. 2014; 132(2):73-9
75
ORIGINAL ARTICLE | Romero FR, Romero AW, Almeida RMSSA, Oliveira Júnior FC, Tambara Filho R
60 years (RR = 1.70, P < 0.05), with diabetes (RR = 1.71, P < 0.05),
or with no history of vasectomy (RR = 0.77, P < 0.05) (Table 2).
Unilateral testicular hypotrophy/atrophy was present in 167
participants (26.2% of the intrascrotal anomalies [167/638], and
9.7% overall [167/1731]), of which 92 cases (55.1%, 92/167)
were on the left side. Bilateral testicular hypotrophy/atrophy was
identified in 93 participants (14.6% [93/638] or 5.4% [93/1731]),
with complete bilateral atrophy in 13 of them (14.0%, 13/93)
(2.0% [13/638] or 0.8% [13/1731]). Hypotrophy/atrophy was
identified more frequently in participants older than 60 years.
The risk-adjusted prevalence for hypotrophy/atrophy is summarized in Table 2.
Absence of a palpable testis either due to prior orchiectomy
or due to cryptorchidism was observed in 20 participants (3.1% of
the intrascrotal anomalies [20/638], and 1.2% overall [20/1731]).
Prior orchiectomy was reported by 13 men (2.0% [13/638] or
0.8% [13/1731]), of which seven cases (53.8%, 7/13) were on the
right side. Cryptorchidism occurred in seven men (1.1% [7/638]
or 0.4% [7/1731]), more commonly on the left side (71.4%, 5/7).
Other intrascrotal lesions/anomalies encountered included
epididymitis/orchitis in five participants (0.8% [5/638] or
0.3% [5/1731]), with higher prevalence among diabetics than
among non-diabetics (RR = 12.90, P < 0.05) (Table 2); postoperative adhesion/hematoma in two cases (0.3% [2/638] or
0.1% [2/1731]); benign paratesticular masses in two cases (0.3%
[2/638] or 0.1% [2/1731]); and scrotal calculi in one case (0.2%
[1/638] or 0.06% [1/1731]). No testicular cancer was identified
in this cohort.
Subcutaneous scrotal nodules
Most cases of multiple nodules on the scrotum are due to calcified or noncalcified epidermal cysts, steatocystomas (sebaceous cysts) or other benign tumors.6 The term scrotal calcinosis can be used in the presence of calcification, but it has been
used inappropriately by many authors even in the absence of
calcification.6,7
The prevalence of subcutaneous nodules in our cohort was
0.7%, with a 334% higher risk among men with elementary
school or lower schooling level.
Scrotal mycosis
In the present study, scrotal mycosis was identified in 12.1% of all
the participants: Tinea cruris in 96.7%, and scrotal tinea in 3.8%.
Tinea of the crural region is an exceedingly common pruritic
superficial fungal infection of the groin and adjacent skin including the scrotum.3 Tinea cruris is more frequent in men than in
women, with a 4:1 ratio, and it accounts for 13.9% of all superficial mycotic infections.4 It can occur at any age, but is much
Hydrocele
The most common cause of scrotal swelling is an accumulation
of fluid between the parietal and visceral layers of the tunica
vaginalis of the scrotum, surrounding one or both of the testes.2
Congenital hydroceles result from a patent processus vaginalis
that allows entry of peritoneal fluid into the scrotum. In adults,
hydroceles are frequently secondary to trauma, infections
(Bancroftian filariasis, tuberculosis and sexually transmitted diseases [STDs]), regional or systemic diseases, radiotherapy, inguinal or scrotal surgery or neoplasms, but they are most commonly
idiopathic in origin.2,8
The prevalence of hydroceles differs according to the age
group, etiology and evaluation method, with greater frequency
of results within the first two years of life, in areas that are
endemic for Bancroftian filariasis and through routine ultrasound examination. In individuals living in endemic areas of filariasis in the northeastern region of Brazil, hydrocele has been
detected through physical examination in 2.2-5.4% of children
and adults,9-11 and 40-50% of men undergoing ultrasound.12 In
non-endemic areas, the clinical prevalence of hydrocele ranges
from 0.6 to 2.9%,10 and the ultrasound prevalence is around
more common after adolescence due to the hormone-dependent
growth of testes and scrotum, enlargement of sweat glands and
increasing body weight. High environmental temperature, sweating, prolonged wearing of wet bathing suits and obesity also play
a very important role in its causation and repeated relapses.3
In our study, the prevalence of Tinea cruris was 11.7%, with a
55% higher risk among men with diabetes, and a 32% lower risk
among those of nonwhite skin color.
Superficial dermatophytic infection of the scrotum is rare
even in the presence of severe infection of the groin and thigh,
25-35%.13,14
In our study, although a small amount of hydrocele was
detected through meticulous scrotal palpation in 15.5% of the
participants > 40 years of age in a non-endemic region, moderate
to severe hydrocele was identified through inspection or palpation in only 0.8%.
The risk of hydrocele was 35% higher in subjects with histories
of nonspecific urethritis. STDs may be associated with secondary involvement of the epididymis, which may potentially cause
defective drainage or obstruction of the venous/lymphatic vessels
DISCUSSION
76
probably due to the higher pH and the presence of capric acid,
one of the fatty acids of the epidermal barrier.3,5 In our study, the
adjusted risk of scrotal tinea was 187% higher among diabetics, and almost six times higher among participants with arterial
hypertension, but these results were not significant, most likely
due to the low prevalence of this condition.
Sao Paulo Med J. 2014; 132(2):73-9
Prevalence and risk factors for scrotal lesions/anomalies in a cohort of Brazilian men ≥ 40 years of age | ORIGINAL ARTICLE
of the tunica vaginalis.7,15 In these cases, the resultant hydrocele
may persist even after the triggering factor has been treated.15
The risk-adjusted prevalence of hydroceles among diabetics was twice as high among the participants of our study. In
the presence of diabetes mellitus, development of hydroceles
may result from microvascular disease and occlusion, or from
increased susceptibility to epididymitis.16
Spermatoceles
Spermatoceles are the most common cystic condition encountered within the scrotum. They are usually situated in the head of
the epididymis, frequently asymptomatic, occasionally bilateral
and multiple, and often smaller than one centimeter.17
It has been estimated that about 30% of asymptomatic men
undergoing scrotal ultrasound have one or more of these cysts.8,18
Many urologists are comfortable with making a diagnosis based
on history and physical examination alone, while others often use
ultrasound to confirm it.8 However, imaging studies do not provide additional information in men with spermatocele in whom
the testes are palpably normal.19
Cystic dilatations of the tubules of the efferent ductules of
the epididymis probably result from obstruction, and are more
frequently encountered in cases of cystic fibrosis, von HippelLindau disease, maternal exposure to diethylstilbestrol and polycystic kidney disease.7,17
In our cohort, spermatoceles were clinically detected in
around 10% of the subjects, with a 70% higher prevalence in men
> 60 years of age, and a similarly increased risk among diabetics.
Although diabetes mellitus is not usually reported as a risk for
spermatoceles, diabetes is an acknowledged risk factor for renal,
pancreatic and ovarian cysts, and it is also more frequently correlated with cystic fibrosis and polycystic kidney disease, which are
often associated with spermatoceles.7,17
We also identified an inverse relationship between spermatoceles and vasectomy. However, since most participants with
histories of vasectomy were < 60 years of age, and spermatoceles
were more prevalent in men > 60 years, it may be possible that
the inverse relationship may have been biased by the lower age
of the participants with histories of vasectomy.
Testicular hypotrophy/atrophy
Hypotrophy (hypoplasia) and atrophy of the testes refers to partial
or complete shrinking of testicular volume. The prevalence of testicular hypotrophy/atrophy is dependent on the underlying cause
of the condition.20-22 In our cohort of men attending a prostate cancer screening program, the prevalences of unilateral and bilateral
testicular hypotrophy/atrophy were, respectively, 9.7% and 5.4%.
A variety of reports have suggested that testicular volume
declines with advancing age.23 In the present study, the prevalence
of both unilateral and bilateral hypotrophy/atrophy was roughly
60% higher among men > 60 years of age. Other independent
risk factors for unilateral testicular hypotrophy/atrophy include
testicular torsion, cryptorchidism, nonspecific orchitis, mumps
orchitis, genital trauma, varicocele, surgical injury, or a combination of these factors. On the other hand, bilateral testicular
hypotrophy/atrophy may be caused by malnutrition, alcohol and
drug abuse and numerous chronic illnesses, although most cases
are idiopathic, and several causes of unilateral atrophy may result
in contralateral testicular injury.20,21,23-25
The subjects with histories of vasectomy showed a 59% lower
risk of bilateral hypotrophy/atrophy, compared with those without a history of vasectomy. Although this result was statistically
significant and it may imply that vasectomy protects against
potential risk factors for bilateral testicular hypotrophy/atrophy,
it is possible that this was confounded by uncontrolled factors. It is
plausible, for example, that men who developed bilateral testicular failure already had a history of infertility or subfertility when
they were younger and, therefore, were not considered eligible
for vasectomy.
Epididymitis/orchitis
Epididymitis, the most common cause of acute scrotal pain in all
age groups, is usually caused by bacterial reflux from the bladder
or the prostate gland through the vas deferens to the epididymis,
and it often spreads to the testes (epididymo-orchitis).2 Isolated
orchitis is rare and it is generally associated with viral infections.26
Epididymitis/orchitis occurs in approximately 0.7% of
men aged 18 to 50 years,26 and it occurred in 0.3% of the men
> 40 years of age in our study. A bimodal distribution has been
observed, with peak incidences occurring among men aged 16 to
30 years and among those aged 51 to 70 years.26 In sexually active
men younger than 35 years of age, chlamydia and gonorrhea are
the most common causes. In men older than 35 or those who
practice insertive anal intercourse, enteric Gram-negative bacilli
are the most common causative pathogens.2,26
The risk factors include sexual activity, bladder outlet obstruction, recent urinary tract surgery or instrumentation, anatomical anomalies, strenuous physical activity, bicycle or motorcycle
riding, and prolonged periods of sitting.2,26 We also identified a
higher risk of epididymitis/orchitis among participants with diabetes mellitus, who presented a nearly 13-fold higher risk.
Cryptorchidism
Testicular ectopy comprises failure of the testes to descend completely, unilaterally or bilaterally into the scrotum, and it is the
most prevalent congenital anomaly at birth. The prevalence of
cryptorchidism is age-dependent, with a 9.2-30% prevalence
among premature infants, 3.4-5.8% among full-term infants,
Sao Paulo Med J. 2014; 132(2):73-9
77
ORIGINAL ARTICLE | Romero FR, Romero AW, Almeida RMSSA, Oliveira Júnior FC, Tambara Filho R
0.8-1.82% at one year of age, and 0.8-1% at puberty and into
adulthood.27 Nonpalpable testes accounts for only 20% to 30% of
the cases.2 In the present study, nonpalpable testes due to cryptorchidism were registered in 0.5% of the men > 40 years of age, with
an almost threefold higher prevalence (but without statistical significance) among participants with low schooling level.
4. Chimelli PAV, Sofiatti AA, Nunes RS, Martins JEC. Dermatophyte
Limitations of the study
This epidemiological study only involved men > 40 years of age
within an established private healthcare system, but it covers a
specific age range of adults that provides important information
about the prevalence and risk factors of several scrotal diseases
in this group.
The participants were examined in the supine position. By
doing so, we failed to detect prevalent intrascrotal lesions, including varicoceles and inguinal hernias, which are better evaluated
in the standing-up position. Furthermore, experimental and
clinical observations have shown substantial imprecision in clinical evaluations and measurements on the testes.23
7. Noël B, Bron C, Künzle N, De Heller M, Panizzon RG. Multiple nodules
agents in the city of São Paulo, from 1992 to 2002. Rev Inst Med Trop
Sao Paulo. 2003;45(5):259-63.
5. Romano C, Ghilardi A, Papini M. Nine male cases of tinea genitalis.
Mycoses. 2005;48(3):202-4.
6. Dubey S, Sharma R, Maheshwari V. Scrotal calcinosis: idiopathic or
dystrophic? Dermatol Online J. 2010;16(2):5.
of the scrotum: histopathological findings and surgical procedure.
A study of five cases. J Eur Acad Dermatol Venereol. 2006;20(6):707-10.
8. Rubenstein RA, Dogra VS, Seftel AD, Resnick MI. Benign intrascrotal
lesions. J Urol. 2004;171(5):1765-72.
9. Braga C, Albuquerque MFM, Schindler HR, et al. Perfil epidemiológico
da filariose linfática em crianças residentes em áreas endêmicas
[Epidemiological pattern of lymphatic filariasis in children living in
endemic areas]. J Pediatr (Rio J). 1997;73(2):95-100.
10. Bonfim C, Lessa F, Oliveira C, et al. Situação da filariose bancroftiana
na Região Metropolitana do Recife: estudo em uma área endêmica
no Município de Jaboatão dos Guararapes, Pernambuco, Brasil
[The occurrence and distribution of lymphatic filariasis in Greater
Implications for clinical practice and future research
The results from our study are important as an epidemiological
resource for patients and healthcare providers, and as a means
of comparison with other populations. Our results have established associations between several epidemiological risk factors
that were hitherto poorly evaluated in the literature and a variety
of scrotal lesions/anomalies routinely seen in consultation office
practice. Future studies should validate the consistency of the
associations identified in our study.
Metropolitan Recife: the case of an endemic area in Jaboatão
dos Guararapes, Pernambuco, Brazil]. Cad Saude Publica. 2003;
19(5):1497-505.
11. Albuquerque MF, Marzochi MC, Sabroza PC, et al. Bancroftian
filariasis in two urban areas of Recife, Brazil: pre-control
observations on infection and disease. Trans R Soc Trop Med
Hyg. 1995;89(4):373-7.
12. Rocha A, Lima G, Medeiros Z, et al. Circulating filarial antigen in the
hydrocele fluid from individuals living in a bancroftian filariasis area –
Recife, Brazil: detected by the monoclonal antibody Og4C3-assay.
CONCLUSIONS
We estimated the prevalence of scrotal diseases commonly seen
in consultation office practice, in an adult population of Brazilian
men. The prevalence of cutaneous lesions was 13.3%, and mycosis was present in most of these (91.7%). We found significant
correlations linking Tinea cruris with diabetes and white skin
color; scrotal tinea with arterial hypertension; subcutaneous
nodules with low schooling level and no history of nonspecific
urethritis; hydrocele with age older than 60 years, diabetes and
histories of nonspecific urethritis; spermatoceles with age older
than 60 years, diabetes and no history of vasectomy; testicular
hypotrophy/atrophy with age older than 60 years; and epididymitis/orchitis with diabetes.
Mem Inst Oswaldo Cruz. 2004;99(1):101-5.
13. Pepe F, Pepe P. Incidenza di patologia ecografica scrotale in 60 pazienti
asintomatici di età superiore ai 70 anni [Incidence of scrotal disease
diagnosed with ultrasonography in 60 asymptomatic patients over
70 years of age]. Minerva Urol Nefrol. 1994;46(2):101-3.
14. Mahmood T, Farooq K, Asghar J, Rashid A. Evaluation of scrotal
pathology on ultrasonography. Pak J Med Health Sci. 2011;5(2):3413. Available from: http://www.pakmedinet.com/17855. Accessed in
2013 (May 15).
15. Wesson MB. Traumatic hydrocele: with an analysis of thirty cases. Cal
West Med. 1929;31(2):127-33.
16.
Tooke
JE.
Microvasculature
in
diabetes.
Cardiovasc
Res.
1996;32(4):764-71.
17.Edington GH. Cysts of the Epididymis. Postgrad Med J.
REFERENCES
1. Duke GA. Common scrotal lesions. Can Fam Physician. 1983;29:1331-4.
18. Vissamsetti B, O’Flynn KJ, Pearce I. Diagnosis and treatment of benign
2. Wampler SM, Llanes M. Common scrotal and testicular problems.
scrotal swellings. Trends in Urology & Men’s Health. 2011;2(3):27-30.
Prim Care. 2010;37(3):613-26.
3. Kumar B, Talwar P, Kaur S. Penile tinea. Mycopathologia. 1981;75(3):169-77.
78
1936;12(127):184-91.
Sao Paulo Med J. 2014; 132(2):73-9
Available from: http://onlinelibrary.wiley.com/doi/10.1002/tre.200/
pdf. Accessed in 2013 (May 15).
Prevalence and risk factors for scrotal lesions/anomalies in a cohort of Brazilian men ≥ 40 years of age | ORIGINAL ARTICLE
19.London NJ, Smart JG, Kinder RB, et al. Prospective study of
routine scrotal ultrasonography in urological practice. Br J Urol.
1989;63(4):416-9.
20. Davis NF, McGuire BB, Mahon JA, et al. The increasing incidence of
mumps orchitis: a comprehensive review. BJU Int. 2010;105(8):1060-5.
21. Pinto KJ, Kroovand RL, Jarow JP. Varicocele related testicular atrophy
and its predictive effect upon fertility. J Urol. 1994;152(2 Pt 2):788-90.
22. Bujan L, Mieusset R, Mansat A, et al. Testicular size in infertile men:
relationship to semen characteristics and hormonal blood levels. Br J
Urol. 1989;64(6):632-7.
23.Handelsman DJ, Staraj S. Testicular size: the effects of aging,
malnutrition, and illness. J Androl. 1985;6(3):144-51.
24.Hanley HG. The surgery of male subfertility; Hunterian lecture
delivered at the Royal College of Surgeons of England on 24th May
1955. Ann R Coll Surg Engl. 1955;17(3):159-83.
25.Osegbe DN, Amaku EO. The causes of male infertility in 504
consecutive Nigerian patients. Int Urol Nephrol. 1985;17(4):349-58.
26.Trojian TH, Lishnak TS, Heiman D. Epididymitis and orchitis: an
overview. Am Fam Physician. 2009;79(7):583-7.
27. Sociedade Brasileira de Urologia. Colégio Brasileiro de Radiologia.
Projeto diretrizes. Afecções testiculares: diagnóstico e tratamento.
Available from: http://www.projetodiretrizes.org.br/6_volume/01AfeccoesDiagn.pdf. Accessed in 2013 (May 15).
Sources of funding: None
Conflict of interest: None
Date of first submission: March 31, 2012
Last received: May 20, 2013
Accepted: May 28, 2013
Address for correspondence:
Frederico Ramalho Romero
Rua Mato Grosso, 1.979
Centro — Cascavel (PR) — Brasil
CEP 85812-020
Tel. (+55 45) 9995-3003
E-mail: [email protected]
Sao Paulo Med J. 2014; 132(2):73-9
79
ORIGINAL ARTICLE
DOI: 10.1590/1516-3180.2014.1322572
Adaptation to prostheses among patients with
major lower-limb amputations and its association
with sociodemographic and clinical data
Adaptação a prótese em pacientes com amputações maiores de membros
inferiores e sua associação com os dados sociodemográficos e clínicos
Marco Antonio NunesI, Ivo Campos-NetoII, Leonardo Costa FerrazII, Camilla Andrade LimaII,
Tâmara Oliviera RochaII, Thaisa Fátima RochaII
Universidade Federal de Sergipe, Aracaju, Sergipe, Brazil
PhD. Adjunct Professor and Head of
Department of Medicine, Universidade Federal
de Sergipe (UFS), Aracaju, Sergipe, Brazil.
I
II
MD. Researcher and Student in the Department
of Medicine, Universidade Federal de Sergipe
(UFS), Aracaju, Sergipe, Brazil.
KEY WORDS:
Amputation.
Lower extremity.
Prosthesis fitting.
Primary health care.
Health surveys.
PALAVRAS-CHAVE:
Amputação.
Extremidade inferior.
Ajuste de prótese.
Atenção primária à saúde.
Inquéritos epidemiológicos.
ABSTRACT
CONTEXT AND OBJECTIVE: Lower-limb amputation compromises patients’ independence and autonomy, and therefore they should be referred for rehabilitation in order to adapt to prostheses and regain
autonomy. The aim here was to assess adaptation to prostheses among patients with major lower-limb
amputations and its association with sociodemographic and clinical data.
DESIGN AND SETTING: This was a cross-sectional study in the city of Aracaju, Brazil.
METHODS: The patients were identified by primary healthcare teams. The inclusion criterion was
that these should be patients who underwent major lower-limb amputations of any etiology.
Associations between sociodemographic and clinical variables and the adaptation to lower-limb
prostheses were assessed.
RESULTS: 149 patients were examined. Adaptation to the prosthesis occurred in 40% (60/149) of them,
but only 62% (37/60) were using it. Adaptation occurred more often among male patients (P = 0.017)
and among those who had a higher educational level (P = 0.013), with a longer time since amputation
(P = 0.049) and when the etiology was trauma (P = 0.003). The result from logistic regression analysis
showed that only patients with low education (P = 0.031) were significantly associated with a lower frequency of adaptation to prostheses.
CONCLUSION: It was found that patients with a low educational level became adapted to the prosthesis
less frequently.
RESUMO
CONTEXTO E OBJETIVO: A amputação de membros inferiores compromete a independência e a autonomia dos pacientes, por isso, eles devem ser encaminhados para a reabilitação para a adaptação das
próteses e assim viabilizar a recuperação da autonomia. O objetivo foi avaliar a adaptação de prótese em
pacientes com amputações maiores de membros inferiores e sua associação com dados sócio-demográficos e clínicos.
TIPO DE ESTUDO E LOCAL: Estudo transversal realizado na cidade de Aracaju, Brasil.
MÉTODO: Os pacientes foram identificados pelas equipes de atenção primária à saúde. O critério de inclusão foi pacientes submetidos a amputações de membros inferiores principais de todas as etiologias.
Foram avaliadas as associações entre variáveis sociodemográficas e clínicas e a adaptação de próteses de
membros inferiores.
RESULTADOS: Foram examinados 149 pacientes. A adaptação da prótese ocorreu em 40% (60/149) deles,
mas apenas 62% (37/60) a utilizavam. Adaptação ocorreu mais frequentemente em pacientes do sexo
masculino (P = 0,017) e naqueles que tinham maior nível de escolaridade (P = 0,013), com maior tempo de
amputação (P = 0,049) e quando a etiologia (P = 0,003) foi o trauma. O resultado da análise de regressão
logística mostrou que apenas a com baixa escolaridade (P = 0,031) foi significativamente associada com
uma menor frequência na adaptação de próteses.
CONCLUSÃO: Verificou-se que pacientes com baixa escolaridade tiveram menor frequência de ajuste
para a prótese.
80
Sao Paulo Med J. 2014; 132(2):80-4
Adaptation to prostheses among patients with major lower-limb amputations and its association with sociodemographic and clinical data | ORIGINAL ARTICLE
INTRODUCTION
Amputation of a limb is not only an esthetic loss: it also compromises autonomy and self-esteem, leaving the patient helpless and dependent.1 Therefore, autonomy and independence
need to be preserved, and such patients should be encouraged
to undertake self-care, recognize their limits and return to their
usual activities.
These patients should be referred to a rehabilitation program. This is a fundamental part of care and essential for a
good functional outcome, because it enables recovery of independence after returning to the community, such that these
individuals can gain the ability to perform their usual activities as independently as possible in order to achieve optimal social participation.2-5 The main goal of rehabilitation is
to allow integration into the community as a productive and
independent member.
The rehabilitation process firstly involves careful selection of
patients for use of lower-limb prostheses.3 The healthcare team
needs to recognize the patients’ clinical and social problems and
understand the factors associated with successful outcomes from
this process.
OBJECTIVE
The aim of this study was to assess the adaptation to prostheses
among patients with major lower-limb amputations and its association with sociodemographic and clinical data.
METHODS
A cross-sectional study was conducted from May 12 to June 30,
2011. The research was planned in accordance with the Declaration
of Helsinki and was approved by the Research Ethics Committee of
Universidade Federal de Sergipe on May 6, 2011. The patients
were identified by primary healthcare teams in the city of
Aracaju, in Brazil and, after informed consent had been obtained
from the patients, data were gathered through visits to patients’
homes. Relevant data were recorded on a standardized form.
Sample
The inclusion criteria were that the subjects should be patients
who had undergone unilateral or bilateral lower-limb amputation
performed at levels above the ankle joint (major amputation);
and that the etiologies should relate to trauma, diabetes mellitus,
infection, ischemia or cancer. The criterion for exclusion was the
presence of mental impairment that precluded participation.
To calculate the sample size, it was assumed that the variable
that contained the response of interest had a population prevalence of 50% adaptation to lower-limb prostheses,6 maximum
error of estimate of 8% and significance level of 5%. Thus, the
sample size was calculated as 151 individuals.
Variables and instruments
Sociodemographic data such as age, gender, marital status
and education (which was classified as low when the subjects
had not completed primary education), and clinical variables
such as etiology, length of time since amputation and number
of associated morbid conditions, were gathered on the survey
form. The subjects were also asked about occurrences of adjustment of the lower-limb prosthesis, which were defined as fitting
the prosthesis for use with or without external support. Walking
within the community was not differentiated from walking only
at home, and patients who used the prosthesis for cosmetic purposes or for transfers were recorded as nonusers.
Statistical analysis
We performed descriptive analyses to examine the clinical and
demographic data. Continuous variables were presented as mean
values and 95% confidence intervals, whereas categorical variables were presented as absolute and relative frequencies. The
relationships between sociodemographic and clinical variables
and adaptation to prostheses were tested through using contingency tables and calculation of Pearson’s chi-square test or
Fisher’s exact test. Differences between the means of pairs of
groups were analyzed using Student’s t test. We then performed
logistic regression analysis to control for confounding variables
using those that showed associations with P < 0.20, with the aim
of exploring the magnitude of associations between the sociodemographic and clinical variables and the results from prosthesis
fitting. P values < 0.05 were considered statistically significant.
RESULTS
Since two amputees refused to sign the informed consent statement, we interviewed 149 patients. They had undergone major
lower-limb amputations at a mean age of 60.2 years (95% CI:
from 57.3 to 63.2); 62% (92/149) were males, 50% (74/149) were
married and 65% (97/149) reported a low educational level. The
mean length of time since amputation was 76.8 months (95% CI:
from 61.9 to 91.8). The amputation had been bilateral in 18%
(27/149) of the patients; 45% (67/149) reported that diabetic foot
was the cause of amputation; and in 26% (38/149), the procedure
related to trauma. Regarding associated diseases, 23% (34/149)
reported none, 53% (79/149) had diabetes mellitus and 57%
(85/149) had hypertension (Table 1). Adaptation to the prosthesis had occurred in 40% (60/149) of the patients with major
amputations, but only 62% (37/60) of these patients were using
their prostheses daily or occasionally.
Regarding evaluation of adaptation to the prosthesis (Table 2), this
occurred among 28% (16/57) of the women and 48% (44/92) of the
men (P = 0.017), and among 33% (32/97) of the patients with low education and 54% (28/52) of those with higher education (P = 0.013).
Sao Paulo Med J. 2014; 132(2):80-4
81
ORIGINAL ARTICLE | Nunes MA, Campos-Neto I, Ferraz LC, Lima CA, Rocha TO, Rocha TF
Table 1. Frequencies of sociodemographic and clinical variables
relating to patients with major lower-limb amputations
Variable
n
%
Female
57
38
Male
92
62
Married
74
50
Not married
75
50
Low level
97
65
High level
52
35
Diabetic foot
67
45
Trauma
38
26
Ischemia
19
13
Infections
17
11
Others
8
5
Yes
60
40
No
89
60
Yes
37
62%
No
23
38%
Gender
Marital status
Education
Etiology
Adaptation to prosthesis
Use of prosthesis
Table 2. Relationship between adaptation to the prosthesis and the
sociodemographic and clinical variables, among patients with major
lower-limb amputations
Variable
Adaptation to prosthesis
Yes (n %)
No (n %)
Total (n %)
Female
16 (28%)
41 (72%)
57 (100%)
Male
44 (48%)
48 (52%)
92 (100%)
Low level
32 (33%)
65 (67%)
97 (100%)
High level
28 (54%)
24 (46%)
52 (100%)
Married
33 (45%)
41 (55%)
74 (100%)
Not married
27 (36%)
48 (64%)
75 (100%)
Yes
30 (38%)
49 (62%)
79 (100%)
No
30 (43%)
40 (57%)
70 (100%)
Yes
33 (39%)
52 (61%)
85 (100%)
No
27 (42%)
37 (58%)
64 (100%)
Yes
25 (37%)
42 (63%)
67 (100%)
No
35 (43%)
47 (57%)
82 (100%)
Yes
23 (61%)
15 (39%)
38 (100%)
No
37 (33%)
74 (67%)
111 (100%)
Total
60 (40%)
89 (60%)
149 (100%)
P-value
Gender
0.017
Education
0.013
Marital status
0.285
Diabetes mellitus
0.544
Hypertension
0.679
Diabetic foot
0.506
Trauma
82
Sao Paulo Med J. 2014; 132(2):80-4
0.003
Moreover, there was no influence from being married or having a
partner (P = 0.285). There was also no significant differences between
age groups (P = 0.146) and numbers of associated diseases (P = 0.307)
among these patients (Table 3). However, those with a longer time
since amputation were associated with a greater chance of successful
prosthesis fit (P = 0.049).
When the associated diseases were taken into account,
although those who reported diabetes mellitus (P = 0.679) and
hypertension (P = 0.544) showed lower frequencies of achieving prosthesis fitting, this was not statistically significant, which
likewise occurred among patients who reported that diabetic foot
(P = 0.506) was the cause of amputation. However, when the cause
was trauma, 61% (23/38) achieved prosthesis fitting (P = 0.003).
Given these results, logistic regression analysis was applied
to the data in order to assess the likelihood of an association
between multiple independent and dependent variables represented by the result relating to adaptation to the prosthesis.
Thus, among the variables of age, gender, low education, trauma
and length of time since amputation, selection for entry into the
model was based on their individual effect on the dependent
variable, at a predetermined significance level, which in this case
was chosen to be alpha equals 0.20. The result from this analysis is shown in Table 4. The association between the observed
dependent and independent variables provided by logistic
regression analysis showed that only the variable of low education (P = 0.031) was significant (Table 4).
DISCUSSION
Our results regarding adaptation to prostheses were very similar to those obtained by other authors,3,6,7 although only 62% of
our patients were using their prosthesis daily or occasionally and
thus a significant number of the amputees were unable to regain
the ability to walk even with rehabilitation. Despite technological
advances in scientific knowledge, the results regarding prosthesis use have varied,6,8,9 and therefore the predictors for prosthetic
rehabilitation remain unknown. The purpose of using a prosthesis is to compensate for the functional loss,10 because although
locomotion with a wheelchair can be a good alternative due to
its energy consumption, mobility becomes very limited. In addition, prostheses have an important social and cosmetic influence
and help to avoid disturbances to body image. Thus, they have a
significant influence on psychosocial adaptation to amputation11
and on performance in activities of daily living.12
Therefore, since social restrictions have an impact on patients’
lives, rehabilitation should be an ongoing, individualized and
planned process, from before the operation until the definitive
prosthesis is inserted, in order to allow patients to recover normal
life and perform all their basic activities. However, dissatisfaction with and low utilization of physiotherapy and occupational
Adaptation to prostheses among patients with major lower-limb amputations and its association with sociodemographic and clinical data | ORIGINAL ARTICLE
Table 3. Relationship between adaptation to the prosthesis,
the mean age, the length of time since amputation and the
number of associated diseases, among patients with major
lower-limb amputations
Adaptation to
prosthesis
n
Mean
SD
Yes
60
57.6
18.7
No
89
62.0
18.0
Length of
time since
amputation
Yes
60
96.3
111.7
No
89
63.7
74.2
Associated
diseases
Yes
60
1.2
0.9
No
89
1.4
0.9
Age
P-value
0.145
0.049
0.307
SD = standard deviation.
Table 4. Result from logistic regression analysis
Variable
Age
Gender
Education
Trauma
Length of time
since amputation
Estimate
-0.0071
0.5327
0.8153
-0.9012
Standard error
0.0120
0.3958
0.3782
0.5097
W test
0.3570
1.8114
4.6489
3.1263
P-value
0.5502
0.1783
0.0311
0.0770
-0.0019
0.0021
0.7689
0.3806
therapy services have been reported, which may be related to
cost, difficulty of access or lack of availability of services and their
potential benefits.13
We observed that patients with low education less frequently
adapted to the prosthesis. One explanation for this finding may
be what has been reported previously, i.e. that those with higher
educational levels use rehabilitative care more often.14 Moreover,
not every amputee is a candidate for a prosthesis and therefore
it is necessary to know how to recognize their problems, including social and economic issues. In addition, just as in our study,
females have been associated with a worse outcome from the
rehabilitation process.15
Longer time since amputation was also associated with a
greater chance of success. However, delays in referral to rehabilitation services are sometimes unavoidable, and this can lead
to development of joint contractures and local complications,
thereby making adaptation to the prosthesis more difficult.16
Factors relating to the prosthesis have been most correlated with
the outcome from rehabilitation.17
Our results revealed that individuals who reported diabetes mellitus and hypertension showed a slight tendency towards
a lower frequency of adaptation to the prosthesis, although this
was not statistically significant. Nonetheless, it has been shown
that associated diseases compromise amputees’ independence.3,18
Moreover, when amputation occurs among elderly people, at a
level above the knee or bilaterally, those who previously could
not walk and had multiple associated morbid conditions are
found to fail to adapt to the prosthesis or present a lower chance
of using it.6,8,19 We also observed that when the etiology of amputation was trauma, 61% were able to adapt to the prosthesis, perhaps because most of these patients were young and had few
diseases. In addition to hypertension and diabetes, peripheral
arterial disease is also associated more frequently with non-traumatic amputations.12
One of the limitations of this study was that no inference
could be made in relation to causality, because the research
design was cross-sectional. Moreover, the morbid conditions
were self-reported. However, despite the number of studies
already available, many questions still remain to be answered in
this field and prospective studies are needed on the predictors
for a proper fit and for achievement of walking ability among
patients, in order to guide appropriate indications for the rehabilitation process.
Discussion on the evolution of patients undergoing lowerlimb amputation is necessary because this should not be considered to be the end of the therapeutic procedure, but a new stage.
Thus, these patients need help to cope with limb loss and reorganize their lives in view of the new reality. For this reason, surgeons should provide guidance for patients, including in relation
to period of rehabilitation, thus establishing a proper physicianpatient relationship. However, this may become a problem when
the physician has not had adequate preparation, and therefore
simply operating these patients is not enough.
CONCLUSION
Therefore, considering the importance of orientation among
healthcare professionals towards better care for these patients
and appropriate referral to rehabilitation services, we found
that the prevalence of adaptation to lower-limb prostheses was
38% and that patients with a low educational level less frequently adapted to the prosthesis. Since such adaptation and
regaining the ability to walk are the main objectives of the
rehabilitation process, knowledge of these factors can help
the healthcare team to provide better care for patients with
such characteristics. In this manner, patients can regain function and quality of life after amputation, through encouragement of better self-care, lower dependence, greater range of
social interactions and less isolation, and also through promotion of preventive actions.
REFERENCES
1. Unwin J, Kacperek L, Clarke C. A prospective study of positive adjustment
to lower limb amputation. Clin Rehabil. 2009;23(11):1044-50.
2. Bussmann JB, Grootscholten EA, Stam HJ. Daily physical activity and
heart rate response in people with a unilateral transtibial amputation
for vascular disease. Arch Phys Med Rehabil. 2004;85(2):240-4.
Sao Paulo Med J. 2014; 132(2):80-4
83
ORIGINAL ARTICLE | Nunes MA, Campos-Neto I, Ferraz LC, Lima CA, Rocha TO, Rocha TF
3. Lim TS, Finlayson A, Thorpe JM, et al. Outcomes of a contemporary
amputation series. ANZ J Surg. 2006;76(5):300-5.
Sources of funding: none
Conflict of interest: none
4. Zidarov D, Swaine B, Gauthier-Gagnon C. Life habits and
prosthetic profile of persons with lower-limb amputation during
Date of first submission: August 4, 2012
rehabilitation and at 3-month follow-up. Arch Phys Med Rehabil.
Last received: February 5, 2013
2009;90(11):1953-9.
Accepted: May 28, 2013
5. Burger H, Marincek C. Return to work after lower limb amputation.
Disabil Rehabil. 2007;29(17):1323-9.
6. Nehler MR, Coll JR, Hiatt WR, et al. Functional outcome in a
Marco Antonio Nunes
contemporary series of major lower extremity amputations. J Vasc
Rua Cláudio Batista, s/no
Surg. 2003;38(1):7-14.
Santo Antônio — Aracaju (SE) — Brasil
7. van der Schans CP, Geertzen JH, Schoppen T, Dijkstra PU. Phantom
CEP 49060-100
pain and health-related quality of life in lower limb amputees. J Pain
Tel. (+55 79) 2105-1811
Symptom Manage. 2002;24(4):429-36.
E-mail: [email protected]
8. Wan-Nar Wong M. Changing dynamics in lower-extremity
amputation in China. Arch Phys Med Rehabil. 2005;86(9):1778-81.
9. Ephraim PL, Wegener ST, MacKenzie EJ, Dillingham TR, Pezzin LE.
Phantom pain, residual limb pain, and back pain in amputees: results
of a national survey. Arch Phys Med Rehabil. 2005;86(10):1910-9.
10. van Velzen JM, van Bennekom CA, Polomski W, et al. Physical capacity
and walking ability after lower limb amputation: a systematic review.
Clin Rehabil. 2006;20(11):999-1016.
11. Gallagher P, Horgan O, Franchignoni F, Giordano A, MacLachlan
M. Body image in people with lower-limb amputation: a Rasch
analysis of the Amputee Body Image Scale. Am J Phys Med Rehabil.
2007;86(3):205-15.
12. Mac Neill HL, Devlin M, Pauley T, Yudin A. Long-term outcomes
and survival of patients with bilateral transtibial amputations after
rehabilitation. Am J Phys Med Rehabil. 2008;87(3):189-96.
13. Whyte AS, Carroll LJ. A preliminary examination of the relationship
between employment, pain and disability in an amputee population.
Disabil Rehabil. 2002;24(9):462-70.
14. Pezzin LE, Dillingham TR, MacKenzie EJ. Rehabilitation and the longterm outcomes of persons with trauma-related amputations. Arch
Phys Med Rehabil. 2000;81(3):292-300.
15. Horgan O, MacLachlan M. Psychosocial adjustment to lower-limb
amputation: a review. Disabil Rehabil. 2004;26(14-15):837-50.
16. Kauzlarić N, Kauzlarić KS, Kolundzić R. Prosthetic rehabilitation of
persons with lower limb amputations due to tumour. Eur J Cancer
Care (Engl). 2007;16(3):238-43.
17. Matsen SL, Malchow D, Matsen FA 3rd. Correlations with patients’
perspectives of the result of lower-extremity amputation. J Bone
Joint Surg Am. 2000;82-A(8):1089-95.
18.Stineman MG, Kurichi JE, Kwong PL, et al. Survival analysis in
amputees based on physical independence grade achievement.
Arch Surg. 2009;144(6):543-51; discussion 552.
19.Moore TJ, Barron J, Hutchinson F 3rd, et al. Prosthetic usage
following major lower extremity amputation. Clin Orthop Relat Res.
1989;(238):219-24.
84
Address for correspondence:
Sao Paulo Med J. 2014; 132(2):80-4
ORIGINAL ARTICLE
DOI: 10.1590/1516-3180.2014.1322544
Incidence and risk factors for retinopathy
of prematurity in a Brazilian reference service
Incidência e fatores de risco para retinopatia da prematuridade
em um serviço de referência brasileiro
Eduardo GonçalvesI, Luciano Sólia NásserII, Daniella Reis MartelliIII, Isadora Ramos AlkmimIV, Thalita Veloso MourãoV,
Antônio Prates CaldeiraVI, Hercílio Martelli-JúniorVI
Universidade Estadual de Montes Claros (Unimontes), Montes Claros, Minas Gerais, Brazil
I
MD. Doctoral Student and Professor,
Postgraduate Health Science Program,
Universidade Estadual de Montes Claros
(Unimontes), and Faculdades Integradas
Pitágoras (FIPMoc), Montes Claros, Minas
Gerais, Brazil.
MD. Master’s Student, Postgraduate Health
Science Program, Universidade Estadual de
Montes Claros (Unimontes), Montes Claros,
Minas Gerais, Brazil.
II
MD. Doctoral Student and Professor,
Postgraduate Health Science Program,
Universidade Estadual de Montes Claros
(Unimontes), Montes Claros, Minas Gerais, Brazil.
III
IV
Medical Student, Universidade Estadual de
Montes Claros (Unimontes), Montes Claros,
Minas Gerais, Brazil.
V
Medical Student, Faculdades Integradas Pitágoras
(FIPMoc), Montes Claros, Minas Gerais, Brazil.
MD, PhD. Professor, Postgraduate Health
Science Program, Universidade Estadual de
Montes Claros (Unimontes), Montes Claros,
Minas Gerais, Brazil.
ABSTRACT
CONTEXT AND OBJECTIVE: Retinopathy of prematurity (ROP) is a known cause of blindness in which
diagnosis and timely treatment can prevent serious harm to the child. This study aimed to evaluate the
incidence of ROP and its association with known risk factors.
DESIGN AND SETTING: Longitudinal incidence study in the neonatal intensive care unit (NICU) of Universidade Estadual de Montes Claros.
METHODS: Newborns admitted to the NICU with gestational age less than 32 weeks and/or birth weight
less than 1,500 grams, were followed up over a two-year period. The assessment and diagnosis of ROP
were defined in accordance with a national protocol. The chi-square test or Fisher’s exact test were used
to determine associations between independent variables and ROP. Analysis on the independent effect of
the variables on the results was performed using multiple logistic regression.
RESULTS: The incidence of ROP was 44.5% (95% confidence interval, CI = 35.6-46.1) in the study population. The risk factors associated with the risk of developing the disease were: birth weight less than
1,000 grams (odds ratio, OR = 4.14; 95% CI = 1.34-12.77); gestational age less than 30 weeks (OR = 6.69;
95% CI = 2.10-21.31); use of blood derivatives (OR = 4.14; 95% CI = 2.99-8.99); and presence of sepsis
(OR = 1.99; 95% CI = 1.45-2.40).
CONCLUSIONS: The incidence of ROP was higher than that found in the literature. The main risk factors
were related to extreme prematurity.
VI
KEY WORDS:
Retinopathy of prematurity.
Incidence.
Infant, premature.
Risk factors.
Gestational age.
PALAVRAS-CHAVE:
Retinopatia da prematuridade.
Incidência.
Prematuro.
Fatores de risco.
Idade gestacional.
RESUMO
CONTEXTO E OBJETIVO: A retinopatia da prematuridade (ROP) é causa conhecida de cegueira e diagnóstico e tratamento oportunos podem evitar graves danos à criança. Este estudo objetivou avaliar a
incidência da ROP e sua associação com fatores de risco conhecidos.
TIPO E LOCAL DE ESTUDO: Estudo longitudinal de incidência na Unidade de Terapia Intensiva Neonatal
(UTIN) da Universidade Estadual de Montes Claros.
MÉTODOS: Foram acompanhados neonatos admitidos na UTIN, com idade gestacional menor que 32
semanas e/ou peso ao nascimento inferior a 1.500 gramas, por um período de dois anos. Avaliação e diagnóstico de ROP foram definidos conforme protocolo nacional. Utilizou-se o teste qui-quadrado ou teste
exato de Fisher para determinar a associação entre as variáveis independentes e ROP. A análise do efeito
independente das variáveis sobre o resultado (ROP) foi realizada por meio de regressão logística múltipla.
RESULTADOS: A incidência de ROP foi de 44,5% (intervalo de confiança, IC 95% = 35,6-46,1) entre a
população estudada. Os fatores de risco pesquisados que apresentaram relação de risco para o desenvolvimento da doença foram: peso de nascimento menor que 1000 gramas (odds ratio, OR = 4,14;
IC 95% = 1,34;12,77), idade gestacional menor que 30 semanas (OR = 6,69; IC 95% = 2,10;21,31), uso
de hemoderivados (OR = 4,14; IC 95% = 2,99;8,99) e presença de sepse (OR = 1,99; IC 95% = 1,45;2,40).
CONCLUSÃO: A incidência de ROP foi maior que a encontrada na literatura. Os principais fatores de risco
estão relacionados à prematuridade extrema.
Sao Paulo Med J. 2014; 132(2):85-91 85
ORIGINAL ARTICLE | Gonçalves E, Násser LS, Martelli DR, Alkmim IR, Mourão TV, Caldeira AP, Martelli-Júnior H
INTRODUCTION
Retinopathy of prematurity (ROP) is a vasoproliferative eye disease of multifactorial etiology that affects the retinal vascularization of premature infants.1,2 The importance of ROP lies in its
frequency and in prevention of blindness due to this condition,
given that, once diagnosed and treated, it is unlikely to develop
into complete loss of vision.3-6 The incidence of blindness varies
between countries, and it is influenced by the level of perinatal
care and the existence of screening programs for early diagnosis.7
In the United States, around 0.12% of all live births develop
ROP, or one case for every 820 newborns, and there are an estimated 300 new cases of blindness annually due to ROP in that
country.3 In Brazil, studies have shown increased numbers of
ROP cases, especially in large centers.7 It is known that ROP is
multifactorial, due to the immaturity of the preterm retina,3,8 and
that the risk factors are: prematurity, low birth weight, oxygen
therapy, intracranial hemorrhage and persistent ductus arteriosus, among others.8-10
OBJECTIVE
The aim of this study was to evaluate the incidence of ROP at
a Brazilian reference service, along with the main known risk
factors for this important condition and its association with
morbidity.
METHODS This was a prospective longitudinal incidence study and it
was approved by the Ethics Committee for Institutional
Research of Montes Claros State University (Universidade
Estadual de Montes Claros, Unimontes), in Minas Gerais,
Brazil (protocol 2013/10).
In this study, a convenience sample limited to a period of
time was used. The study included preterm infants who had
been admitted to the neonatal intensive care unit (NICU) of the
Unimontes university hospital, between May 2009 and April
2011. The inclusion criteria were that the birth weight should be
≤ 1,500 g and/or the gestational age should be ≤ 32 weeks, with
survival at least until the sixth week of life;6 and that the infants
were treated as outpatients for follow-up care in the eye clinic.
The study excluded infants born with ocular malformation or
ocular genetic alterations and those who did not survive past the
sixth week of life.
There was no sample calculation. Over the period from May
2009 to April 2011, 124 patients who met the inclusion criteria
were admitted. However, 12 patients died and two did not return
to the clinic for monitoring. The eye examination took place in
the sixth week of life, or between 32 and 36 weeks of corrected
gestational age. It consisted of using a binocular indirect ophthalmoscope (BIO) under pupil dilation produced by the eye drops
86
Sao Paulo Med J. 2014; 132(2):85-91
0.5% tropicamide (Midriacyl, Alcon) and 2.5% phenylephrine
(Fenilefrina, Allergan), which were applied twice with an interval of ten minutes. The test was completed one hour afterwards
and, when necessary, anesthetic eye drops (Anestalcon, Alcon)
were also used. A 28-diopter lens (Nikon, Melville, NY, USA) and
lid speculum (Alfonso Eye Speculum, Storz, Bausch & Lomb Inc.,
San Dimas, CA, USA) were used. Scleral indentation was used
when necessary.
Children who did not present any degree of retinopathy of prematurity were not considered to be ROP patients. Children who
had some degree of ROP were considered to have ROP. The degree
of retinopathy assigned to each patient was that of the severest
degree seen in the eyes of the infant under examination.6
The evaluations were repeated periodically, in accordance
with the procedures laid out in the Brazilian Examination and
Treatment Guidelines for individuals with ROP.6 The ophthalmological examinations were always performed by the same
ophthalmologist, who had had specialized retinal training, in
the neonatology center. The ophthalmologist did not have any
prior knowledge of the medical histories of any of the patients.
The examinations were conducted at the neonatal unit and
the patients were monitored until reaching 42 weeks of corrected gestational age, or until complete retinal stabilization
had been achieved.
The ophthalmological monitoring of the patients was performed based on the stage of the disease. Patients with incomplete vascularization of the retina were monitored at intervals
of one to three weeks until vascularization had been completed.
Infants with ROP of degrees I, II or III4 (excluding threshold disease) underwent weekly monitoring until complete vascularization of the retina had been achieved. Those with threshold
disease underwent treatment for retinal photo-coagulation or
cryotherapy. Infants with ROP of degree IV underwent evaluation of scleral explant with or without associated photo-coagulation or cryotherapy or posterior vitrectomy. For infants with
ROP of degree V, possible surgical treatment needed to be discussed. Because of technical limitations, patients requiring treatment by means of photo-coagulation, cryotherapy or another
surgical procedure were referred to another unit of the hospital.
The assessed risk factors for ROP were categorized based on
the characteristics of the newborns, the types of therapies used
and the diseases detected. The characteristics of the newborns
were evaluated according to gestational age (calculated by the
unit’s neonatologist using data on the last menstrual period as
the first option, followed by ultrasound data and data obtained
using the New Ballard method),11 birth weight, presence of multiple gestations, Apgar score in the first and fifth minutes and the
severity score, SNAPPE II (Score for Neonatal Acute Physiology
Perinatal Extension).12,13
Incidence and risk factors for retinopathy of prematurity in Brazilian reference service | ORIGINAL ARTICLE
In relation to the therapies used in this study, the following factors were taken into consideration: maximum fraction of
inspired oxygen (FiO2) level; use of oxygen therapy by means of
continuous positive airway pressure (CPAP) or mechanical ventilation; use of indomethacin; use of a surfactant, aminophylline
or caffeine; phototherapy use; blood transfusions; use of diuretics; and corticosteroid use.
For diseases detected during hospitalization, presence of the
following was recorded: initial respiratory distress, bronchopulmonary dysplasia (use of oxygen for 28 days or more), sepsis (clinical or laboratorial diagnosis), patent ductus arteriosus (diagnosis by Doppler echocardiography) and intracranial
hemorrhage (transfontanellar ultrasound between five and ten
days of life).
For statistical analysis, the study population was classified as
individuals with ROP and those without ROP, i.e. patients who
did not show the illness. The descriptive analysis was performed
using absolute numbers and percentages of the qualitative variables and of the central trend averages: means and their respective standard deviations (SD) of the quantitative variables. In further analysis, all the variables were dichotomized.
In the univariate analysis, the chi-square test and the Fisher
exact test were used to determine any associations between the
independent variables and the outcome (ROP). The magnitude of
the effect of the risk factors on the outcome was expressed as the
odds ratio (OR), with its respective 95% confidence interval (CI).
The analysis on the independent effects of the intervening variables on the outcome (ROP) was performed by means of multiple logistic regression, using forward modeling. All variables
that showed P < 0.25 in the association test were included in the
modeling process. To build the database and to do the statistical analysis, the Statistical Package for the Social Sciences (SPSS)
software (SPSS 18.0 for Windows, SPSS Inc., Chicago, IL, USA)
and the Epi Info software (CDC Epi Info 3.5.4, Atlanta, Georgia,
USA) were used.
RESULTS
Between May 2009 and April 2011, 124 children weighing less
than 1,500 grams were born, of which 12 died before the first
evaluation and 2 did not return for examination after being discharged. Thus, 110 infants were evaluated. The population studied was classified as individuals with ROP, i.e. those who showed
some degree of retinopathy of prematurity, and individuals without ROP, i.e. those who did not show retinopathy of prematurity. Among these 110 infants, the prevalence of ROP was 44.5%
(95% CI = 35.6-46.1). Considering all the births that took place
in the department during the study period, the incidence of ROP
was 1.1%. Stage I ROP was the form with the highest incidence
among these infants (Table 1), affecting 19 children (17.3% of
those with ROP), followed by stage III with 16 children (14.5%)
and stage II with 14 children (12.7%). There were no patients
with stages IV or V. Two premature infants (1.8%) needed laser
treatment because they showed threshold disease. The average weights and gestational ages of all the patients are shown
in Table 2.
In the study population, 49.1% of the patients were male.
Seventy-four percent of the mothers reported that they had
attended prenatal care consultations, 72% presented some complications during pregnancy and 43% had some complications
during childbirth. The most common form of delivery was cesarean (65.5%). The SNAPPE II score was obtained for all the preterm infants after they had reached an average of 12 hours of life,
ranging from 0 to 88. About half of the population studied was
small for the gestational age (49.8%).
Among the risk factors studied, the following correlated significantly (P < 0.05) with the development of ROP, in univariate
analysis: gestational age less than 30 weeks, Apgar scores at the
first and fifth minute less than 7, SNAPPE II score less than 12,
use of blood transfusions, use of diuretics, use of aminophylline
or caffeine, use of surfactants, presence of sepsis and presence of
bronchopulmonary dysplasia (Table 3).
Table 1. Incidence of retinopathy of prematurity (ROP) and disease stage
ROP stage
Without ROP
ROP
Total
I
II
III
IV
V
Total
Incidence
n
%
61
49
110
19
14
16
0
0
49
55.5
44.5
100
17.3
12.7
14.5
0
0
44.5
Table 2. Means and standard deviations of the birth weight and gestational age risk factors of the different stages of retinopathy of
prematurity (ROP)
Risk factors
Weight (grams)
Gestational age (weeks)
Without ROP
(n = 61)
1153.43 ± 241.48
30.24 ± 2.58
ROP I
(n = 19)
1012.13 ± 183.70
28.39 ± 2.77
ROP II
(n = 14)
1097.30 ± 213.43
27.96 ± 3.01
ROP III
(n = 16)
814.35 ± 159.39
26.32 ± 1.69
P-value
0.004
< 0.001
Sao Paulo Med J. 2014; 132(2):85-91 87
ORIGINAL ARTICLE | Gonçalves E, Násser LS, Martelli DR, Alkmim IR, Mourão TV, Caldeira AP, Martelli-Júnior H
Table 3. Risk factors for development of retinopathy of prematurity (ROP) in infants with weight ≤ 1,500 grams and/or gestational age
≤ 32 weeks; univariate analysis
Risk factors
With ROP
(n = 49)
n (%)
Without ROP
(n = 61)
n (%)
P-value
OR (CI)
Male gender
22 (44.9)
32 (52.5)
0.430
0.74 (0.35-1.57)
Gestational age less than 30 weeks
36 (73.5)
16 (26.2)
< 0.001
Birth weight less than 1,000 grams
14 (28.6)
10 (16.4)
0.124
Presence of multiple pregnancies
5 (10.2)
4 (6.6)
0.508
0.62 (0.16-2.44)
Use of prenatal corticosteroids
13 (26.5)
10 (16.4)
0.194
1.84 (0.73-4.66)
Apgar score at one minute less than 7
38 (77.6)
35 (57.4)
0.026
2.57 (1.11-5.95)
Apgar score at five minutes less than 7
17 (34.7)
9 (14.8)
0.014
3.07 (1.22-7.70)
Initial respiratory discomfort
46 (93.9)
58 (95.1)
1.000
0.79 (0.15-4.12)
SNAPPE II score less than 12
34 (69.4)
21 (34.4)
< 0.001
4.32 (1.93-9.66)
Oxygen therapy using headpiece (HOOD) for more than 5 days
12 (24.5)
10 (16.4)
0.290
1.65 (0.65-4.23)
Oxygen therapy using CPAP for more than 5 days
13 (26.5)
8 (13.1)
0.075
2.39 (0.90-6.36)
Oxygen therapy using mechanical ventilation for more than 5 days
29 (59.2)
21 (34.4)
0.001
2.76 (1.27-6.01)
Use of blood transfusions
27 (55.1)
13 (21.3)
< 0.001
Use of diuretics
30 (61.2)
20 (32.8)
0.003
Use of indomethacin
10 (20.4)
8 (13.1)
0.304
1.69 (0.61-4.69)
Use of aminophylline/caffeine
36 (73.5)
33 (54.1)
0.037
2.35 (1.05-5.28)
Use of surfactant
29 (59.2)
20 (32.8)
0.006
2.97 (1.36-6.49)
Presence of sepsis
48 (98.0)
50 (82.0)
0.001
1.98 (1.63-2.41)
Presence of bronchopulmonary dysplasia
26 (53.1)
15 (24.6)
0.002
3.47 (1.54-7.78)
Presence of intracranial hemorrhage
10 (20.4)
10 (16.4)
0.589
1.31 (0.49-3.45)
Presence of patent ductus arteriosus (diagnosed using Doppler
echocardiography)
14 (28.6)
13 (21.3)
0.379
1.48 (0.62-3.53)
Use of phototherapy
45 (91.8)
60 (98.4)
0.170
1.10 (0.02-1.73)
7.79 (3.32-18.28)
2.04 (0.81-5.11)
4.53 (1.97-10.41)
3.24 (1.48-7.09)
OR = odds ratio; CI = confidence interval; CPAP = continuous positive airway pressure; SNAPPE = Score for Neonatal Acute Physiology and SNAP Perinatal Extension.
The multivariate analysis using hierarchical logistic regression (the forward method) is represented in Table 4. All of
the variables that showed P < 0.25 in the association test were
included in the modeling process. Among the risk factors investigated, we found that the following showed a risk of developing ROP: birth weight less than 1,000 grams (OR = 4.14; 95%
CI = 1.34-12.77); gestational age less than 30 weeks (OR = 6.69;
95% CI = 2.10-21.31); use of blood transfusions (OR = 4.14;
95% CI = 2.99-8.99); and presence of sepsis (OR = 1.99; 95%
CI = 1.45-2.40). There was no verified association with any risk
factors, except for the significant trend of the gestational age factor (P = 0.06).
DISCUSSION
Over the period of this study, 124 infants with a birth weight of
< 1,500 grams were born; however, only 110 entered our department. Twelve children with birth weight < 1,500 grams died
(10.9%) and two children (1.8%) did not return to the clinic
for follow-up. Thus, the mortality rate was much lower than the
national mortality rate of 60% for children born at this weight.13
88
Sao Paulo Med J. 2014; 132(2):85-91
Table 4. Risk factors for retinopathy of prematurity (ROP)
development in infants with weight < 1,500 grams
Risk factors
Gestational age less than 30 weeks
Birth weight less than 1,000 grams
Use of prenatal corticosteroids
Apgar score at one minute less than 7
Apgar score at five minutes less than 7
SNAPPE II score less than 12
Oxygen therapy using CPAP for more
than 5 days
Oxygen therapy using mechanical
ventilation for more than 5 days
Use of blood transfusions
Use of diuretics
Use of aminophylline/caffeine
Use of surfactant
Presence of sepsis
Presence of bronchopulmonary dysplasia
Use of phototherapy
OR*
6.69
4.14
1.61
1.71
1.32
1.91
95% CI† P-value
2.10-21.31 < 0.001
1.34-12.77
0.014
0.45-5.79
0.464
0.59-5.00
0.319
0.36-4.82
0.670
0.62-5.88
0.260
0.92
0.26-3.28
0.899
0.79
0.21-2.92
0.720
4.14
1.02
0.64
1.25
1.99
1.92
0.27
2.99-8.99
0.24-4.55
0.17-2.38
0.39-3.96
1.45-2.40
0.68-5.46
0.69-1.07
0.012
0.956
0.500
0.701
< 0.001
0.219
0.062
OR = odds ratio; †CI = confidence interval; SNAPPE = Score for Neonatal
Acute Physiology and SNAP Perinatal Extension; CPAP = continuous positive
airway pressure.
*
Incidence and risk factors for retinopathy of prematurity in Brazilian reference service | ORIGINAL ARTICLE
This includes all preterm infants assessed. In the Brazilian literature, ROP incidence rates of 29.1%,13 28.5%9 and 27.2%14 have
been described. These studies fell within a similar range of incidence rates and showed the highest rate for stage I ROP, which
was about the same as what was found in the present study. This
could explain the higher ROP incidence rate found.15
The Cryotherapy for Retinopathy of Prematurity (CRYOROP) study showed an incidence of 65.8%.16 However, the weight
criterion required for inclusion in the study was lower than that
of the present study (weight < 1,250 grams). Another important American study, the Early Treatment for Retinopathy of
Prematurity Study (ETROP), also included infants with weights
of less than 1,250 grams and showed an incidence of 68%.17
Although gestational age can sometimes be hard to pinpoint as it is often imprecise and sometimes unknown, gestational age and birth weight were still used as the criteria for this
study. These criteria were set because the Brazilian guidelines for
screening ROP, published in 2007,14 define the inclusion criteria as birth weight < 1,500 g and/or a gestational age < 32 weeks.
Additionally, low gestational age and low birth weight have been
associated with and have the same consequence as immature retinal tissue.9,18,19 The results from these studies showed that both
gestational age and birth weight were associated with ROP.
Several studies in the literature have shown that the lower
the birth weight and the lower the gestational age are, the higher
the chance of developing ROP also is.20 Moreover, lower birth
weight and lower gestational age are associated with the development of more serious forms of ROP. In the current study, stage
I of the disease was the most frequently observed form, accounting for 17.3% of the cases when all infants with birth weight
< 1,500 grams are included. This proportion rises to 25.8% if only
the infants with birth weight < 1,250 grams are included. These findings were not reported in the CRYO-ROP or ETROP studies, which
showed higher rates of the severer forms of ROP in infants with
lower birth weights and in those with lower gestational age. By analyzing the average birth weight and gestational age among individuals with varying stages of ROP in the present study, it was noted that
there was a statistically significant difference between the preterm
infants with stage III ROP and those without ROP, thus confirming the association between the immature retina and the ROP stage.
Just 1.8% of the premature infants in the present study were
treated. This small number can be explained by the decrease in
stage III cases. It has been questioned whether there is any association between the severity of the SNAPPE II score and retinopathy of prematurity.14 The present study showed that although a
significant association was observed in univariate analysis, it did
not remain in the final model of the multivariate analysis.
The use of oxygen therapy in this study was evaluated in
terms of number of days and form of administration. After
multivariate analysis, none of the forms of oxygen administration (mechanical ventilation, CPAP or HOOD) showed
any statistical association with ROP. Other studies have confirmed the association between the risk of ROP and the use
of mechanical ventilation 21 and CPAP. Oxygenation, regardless of the form of use, has been implicated in causing ROP. 22
The less mature the preterm infant is, the greater the need
for mechanical ventilation and CPAP is, which explains
this finding.
The treatment for bronchopulmonary dysplasia includes use
of diuretics and corticosteroids. The present study showed that
there was an association between the risk of ROP and the use
of diuretics, but only in univariate analysis. In the multicenter
STOP-ROP study,23 it was found that supplemental use of oxygen
therapy for preterm infants with pre-threshold disease increased
the risk of chronic pulmonary diseases and also increased the use
of diuretics and length of hospital stay. Use of prenatal corticosteroids did not show any statistical impact on development of ROP,
which has not been shown in other studies.24
The presence of sepsis in our study was considered to be a
risk factor for ROP and showed significance. In the literature, this
association has already been described.25 A study by Chen et al.
showed that sepsis can be considered to be an important risk factor for ROP, in screening for preterm infants weighing between
1,501 and 2,000 grams.26
The presence of blood transfusions was another risk factor
that showed a significant relationship between the groups. This
finding has also been reported by several authors in the literature.5,9,19 It is believed that fetal hemoglobin has a greater affinity to oxygen than does adult hemoglobin. Thus, a transfusion
of adult hemoglobin could generate possible hyperoxia due to
increased oxygen delivery to tissues.27 Another theory is that
there could be an increase in free radicals after the transfusions
due to an increase in plasma free iron.28-30
Logistic regression takes into account the relative contributions of various factors and makes it clear that the true risk
factor is low gestational age and the consequential immaturity of the premature infant’s various tissues. In other words,
the true risk factor is prematurity and the other factors (low
SNAPPE II score at birth, bronchopulmonary dysplasia, use
of oxygen in mechanical ventilation, use of diuretics and the
necessity for blood transfusions) are just consequences of this
prematurity and are driven towards the same statistical significance observed in isolated tests. Healthcare professionals should be alert to the risk of ROP among preterm infants,
and such awareness should guide preventive and timely care.
Further studies should take into consideration the risk factors
identified in this study, as well as the possibility of new variables that imply risks for premature infants.
Sao Paulo Med J. 2014; 132(2):85-91 89
ORIGINAL ARTICLE | Gonçalves E, Násser LS, Martelli DR, Alkmim IR, Mourão TV, Caldeira AP, Martelli-Júnior H
CONCLUSIONS
The observed incidence was higher than that found in the literature, thus showing that occurrences of retinopathy of prematurity remain high among infants with very low birth weight. The
development of ROP was inversely proportional to the weight
and gestational age at birth. Prevention of prematurity and caution in using oxygen in neonatal intensive care units may help
reduce the future incidence of retinopathy of prematurity.
10. Sears NC, Sears JE. Oxygen and retinopathy of prematurity. Int
Ophthalmol Clin. 2011;51(1):17-31.
11. Ballard JL, Khoury JC, Wedig K, et al. New Ballard Score, expanded to
include extremely premature infants. J Pediatr. 1991;119(3):417-23.
12. Vanpée M, Walfridsson-Schultz U, Katz-Salamon M, et al. Resuscitation
and ventilation strategies for extremely preterm infants: a comparison
study between two neonatal centers in Boston and Stockholm. Acta
Paediat. 2007;96(1):10-6; discussion 8-9.
13. Branco de Almeida MF, Guinsburg R, Martinez FE, et al. Fatores
REFERENCES
perinatais associados ao óbito precoce em prematuros nascidos nos
1. Fortes Filho JB, Valiatti FB, Eckert GU, et al. Ser pequeno para a idade
centros da rede brasileira de pesquisas neonatais [Perinatal factors
gestacional é um fator de risco para a retinopatia da prematuridade?
associated with early death in preterm infants born in the centers
Estudo com 345 pré-termos de muito baixo peso [Is being small for
of the Brazilian Neonatal Research Network]. Rev Soc Boliv Pediatr.
gestational age a risk factor for retinopathy of prematurity? A study
with 345 very low birth weight preterm infants]. J Pediatr (Rio J).
2009;85(1):48-54.
2. Fortes Filho JB, Barros CK, Costa MC, Procianoy RS. Resultados
de um programa de prevenção da cegueira pela retinopatia da
Physiology and Perinatal Extension II as a predictor of retinopathy
of prematurity: study in 304 very-low-birth-weight preterm infants.
Ophthalmologica. 2009;223(3):177-82.
prematuridade na Região Sul do Brasil [Results of a program for the
15. Hård AL, Löfqvist C, Fortes Filho JB, et al. Predicting proliferative
prevention of blindness caused by retinopathy of prematurity in
retinopathy in a Brazilian population of preterm infants with the
southern Brazil]. J Pediatr (Rio J). 2007;83(3):209-16.
screening algorithm WINROP. Arch Ophthalmol. 2010;128(11):1432-6.
3. Lad EM, Nguyen TC, Morton JM, Moshfeghi DM. Retinopathy of
prematurity in the United States. Br J Ophthalmol. 2008;92(3):320-5.
16. Palmer EA, Flynn JT, Hardy RJ, et al. Incidence and early course of
retinopathy of prematurity. The Cryotherapy for Retinopathy of
4. Shinsato RN, Paccola L, Gonçalves WA, et al. Frequência de retinopatia
Prematurity Cooperative Group. Ophthalmology. 1991;98(11):1628-40.
da prematuridade em recém-nascidos no Hospital das Clínicas da
17. Good WV, Hardy RJ, Dobson V, et al. The incidence and course of
Faculdade de Medicina de Ribeirão Preto da Universidade de São
retinopathy of prematurity: findings from the early treatment for
Paulo [Frequency of retinopathy of prematurity at newborns at the
retinopathy of prematurity study. Pediatrics. 2005;116(1):15-23.
Clinical Hospital, Ribeirão Preto Medical School, University of São
Paulo]. Arq Bras Oftalmol. 2010;73(1):60-5.
5. Fortes Filho JB, Eckert GU, Valiatti FB, et al. The influence of gestational
age on the dynamic behavior of other risk factors associated with
retinopathy of prematurity (ROP). Graefes Arch Clin Exp Ophthalmol.
2010;248(6):893-900.
6. Zin A, Florêncio T, Fortes Filho JB, et al. Proposta de diretrizes
18. Bonotto LB. Workshop de Retinopatia da Prematuridade em
Fortaleza.
Oftalmopediatria.
Available
from:
http://www.
oftalmopediatria.com/texto.php?cs=17&n=1&ct=64&ano=2010.
Accessed in 2013 (May 14).
19. Hellström A, Ley D, Hansen-Pupp I, et al. New insights into the
development of retinopathy of prematurity--importance of early
weight gain. Acta Paediatr. 2010;99(4):502-8.
brasileiras do exame e tratamento de retinopatia da prematuridade
20. Liu L, Tian T, Zheng CX, et al. Risk factors and laser therapy for
(ROP) [Brazilian guidelines proposal for screening and treatment
retinopathy of prematurity in neonatal intensive care unit. World J
of retinopathy of prematurity (ROP)]. Arq Bras Oftalmol. 2007;
Pediat. 2009;5(4):304-7.
70(5):875-83.
7. Fortes Filho JB, Eckert GU, Procianoy L, Barros CK, Procianoy RS.
21. Al-Amro SA, Al-Kharfi TM, Thabit AA, Al-Mofada SM. Risk factors for
acute retinopathy of prematurity. Compr Ther. 2007;33(2):73-7.
Incidence and risk factors for retinopathy of prematurity in very low
22. Karkhaneh R, Mousavi SZ, Riazi-Esfahani M, et al. Incidence and risk
and in extremely low birth weight infants in a unit-based approach
factors of retinopathy of prematurity in a tertiary eye hospital in
in southern Brazil. Eye (Lond). 2009;23(1):25-30.
90
2010;49(1):48-57.
14. Fortes Filho JB, Dill JC, Ishizaki A, et al. Score for Neonatal Acute
Tehran. Br J Ophthalmol. 2008;92(11):1446-9.
8. Jalali S, Matalia J, Hussain A, Anand R. Modification of screening
23. Supplemental Therapeutic Oxygen for Prethreshold Retinopathy of
criteria for retinopathy of prematurity in India and other middle-
Prematurity (STOP-ROP), a randomized, controlled trial. I: primary
income countries. Am J Ophthalmol. 2006;141(5):966-8.
outcomes. Pediatrics. 2000;105(2):295-310.
9. Lermann VL, Fortes Filho JB, Procianoy RS. The prevalence of
24. Hagadorn JI, Richardson DK, Schmid CH, Cole CH. Cumulative illness
retinopathy of prematurity in very low birth weight newborn infants.
severity and progression from moderate to severe retinopathy of
J Pediatr (Rio J). 2006;82(1):27-32.
prematurity. J Perinatol. 2007;27(8):502-9.
Sao Paulo Med J. 2014; 132(2):85-91
Incidence and risk factors for retinopathy of prematurity in Brazilian reference service | ORIGINAL ARTICLE
25. Chen ML, Guo L, Smith LE, Dammann CE, Dammann O. High or low
oxygen saturation and severe retinopathy of prematurity: a metaanalysis. Pediatrics. 2010;125(6):e1483-92.
26. Yanovitch TL, Siatkowski RM, McCaffree M, Corff KE. Retinopathy
of prematurity in infants with birth weight > or = 1250 gramsincidence, severity, and screening guideline cost-analysis. J AAPOS.
2006;10(2):128-34.
27. Vinekar A, Dogra MR, Sangtam T, Narang A, Gupta A. Retinopathy of
prematurity in Asian Indian babies weighing greater than 1250 grams
at birth: ten year data from a tertiary care center in a developing
country. Indian J Ophthalmol. 2007;55(5):331-6.
28. Tlucek PS, Corff KE, Bright BC, et al. Effect of decreasing target oxygen
saturation on retinopathy of prematurity. J AAPOS. 2010;14(5):406-11.
29. York JR, Landers S, Kirby RS, Arbogast PG, Penn JS. Arterial oxygen
fluctuation and retinopathy of prematurity in very-low-birth-weight
infants. J Perinatol. 2004;24(2):82-7.
30. Saugstad OD. Oxygen and retinopathy of prematurity. J Perinatol.
2006;26 Suppl 1:S46-50; discussion S63-4.
Sources of funding: None
Conflict of interest: None
Date of first submission: June 25, 2012
Last received: June 6, 2013
Accepted: June 14, 2013
Address for correspondence:
Eduardo Gonçalves
Rua Gabriel Passos, 116 — apto 201
Centro — Montes Claros (MG) — Brasil
CEP 39400-112
Tel. (+55 38) 8822-2575
Fax. (+55 38) 3224-8372
E-mail: [email protected]
Sao Paulo Med J. 2014; 132(2):85-91 91
ORIGINAL ARTICLE
DOI: 10.1590/1516-3180.2014.1322579
A retrospective study on cervical intraepithelial lesions
of low-grade and undetermined significance: evolution,
associated factors and cytohistological correlation
Estudo retrospectivo sobre lesões intraepiteliais cervicais de baixo grau e de
significado indeterminado: evolução, fatores associados e correlação citohistológica
Criseide SilvaI, Elia Cláudia Souza AlmeidaII, Eliângela de Castro CôboIII, Valéria Fátima Machado ZeferinoIV,
Eddie Fernando Cândido MurtaV, Renata Margarida EtchebehereVI
Universidade Federal do Triângulo Mineiro (UFTM), Uberaba, Minas Gerais, Brazil
MSc. Biomedic, Postgraduate Course on
Pathology, Universidade Federal do Triângulo
Mineiro (UFTM), Uberaba, Minas Gerais, Brazil.
I
II
MSc, PhD. Dentist, Discipline of Histology,
Universidade Federal do Triângulo Mineiro
(UFTM), Uberaba, Minas Gerais, Brazil.
MSc. Biomedic, Discipline of Special Pathology,
Universidade Federal do Triângulo Mineiro
(UFTM), Uberaba, Minas Gerais, Brazil.
III
IV
MSc. Pharmacist and Nursing Assistant,
Universidade Federal do Triângulo Mineiro
(UFTM), Uberaba, Minas Gerais, Brazil.
MD, PhD. Coordinator of the Postgraduate
Course on Health Sciences, Universidade Federal
do Triângulo Mineiro (UFTM), Uberaba, Minas
Gerais, Brazil.
V
VI
MD, PhD. Surgical Pathology Service,
Universidade Federal do Triângulo Mineiro
(UFTM), Uberaba, Minas Gerais, Brazil.
KEY WORDS:
Papillomaviridae.
Risk factors.
Uterine cervical neoplasms.
Vaginal smears.
Uterine neoplasms.
PALAVRAS-CHAVE:
Papillomaviridae.
Fatores de risco.
Neoplasias do colo do útero.
Esfregaço vaginal.
Neoplasia uterinas.
92
Sao Paulo Med J. 2014; 132(2):92-6
ABSTRACT
CONTEXT AND OBJECTIVES: Cervical cancer is an important cause of morbidity and mortality throughout the world. There is some controversy about the factors that may be associated with infection by the
human papillomavirus (HPV) that may favor or protect against evolution from a low-grade intraepithelial
lesion to a high-grade intraepithelial lesion or invasive neoplasia. The objective here was to evaluate the
evolution of low-grade intraepithelial lesions and squamous or glandular lesions of undetermined significance, the associated factors and cytohistological correlations.
DESIGN AND SETTING: Retrospective study conducted in a public tertiary-level university hospital.
METHODS: Information was obtained by reviewing patient records and/or colposcopy reports. A statistical analysis was performed using logistic regression, calculating the odds ratio and applying chisquare tests.
RESULTS: Of the 3390 patients, 409 evolved to high-grade intraepithelial lesions, of which 354 had an
initial diagnosis of HPV infection, 27 of squamous atypia of undetermined significance, 22 of low-grade
intraepithelial lesions with or without cytological diagnosis of infection by associated HPV and six of glandular cell atypia of undetermined significance.
CONCLUSIONS: Lactobacillus sp and bacterial vaginosis on the smears, smoking and immunodepression
were factors associated with evolution. A single partner, use of hormonal contraceptives, lower parity,
age and a cytological diagnosis of cytolytic vaginosis, T. vaginalis, Candida sp or cocci were factors associated with protection. With regard to cytohistological correlation, there was a 74.08% agreement among
patients with high-grade lesions and a biopsy obtained during the same period.
RESUMO
CONTEXTO E OBJETIVOS: O câncer do colo uterino é importante causa de morbidade e mortalidade no
mundo. Existem controvérsias sobre fatores que podem estar associados à infecção por papilomavírus
humano (HPV) e favorecer ou proteger contra a evolução de lesão intraepitelial de baixo grau até lesão
intraepitelial de alto grau ou neoplasia invasiva. O objetivo aqui foi de avaliar a evolução de lesões intraepiteliais de baixo grau e de significado indeterminado, escamosas ou glandulares, os fatores associados e
as correlações citohistológicas.
TIPO DE ESTUDO E LOCAL: Estudo retrospectivo realizado em hospital universitário público terciário.
MÉTODOS: As informações foram obtidas por meio de revisão de prontuários e/ou fichas de colposcopia.
A análise estatística foi realizada por meio de regressão logística, calculando-se o odds ratio e aplicando-se
o teste qui-quadrado.
RESULTADOS: Das 3.390 pacientes, 409 evoluíram para lesão intraepitelial de alto grau, das quais 354
tinham diagnóstico inicial de infecção por HPV, 27 de atipias escamosas de significado indeterminado, 22
de lesão intraepitelial de baixo grau com ou sem diagnóstico citológico de infecção por HPV associado e
6 de atipias em células glandulares de significado indeterminado.
CONCLUSÕES: Lactobacillus sp e vaginose bacteriana nos esfregaços, tabagismo e imunodepressão foram
fatores associados com a evolução. Único parceiro, uso de contraceptivo hormonal, baixa paridade, idade
e diagnóstico citológico de vaginose citolítica, T. vaginalis, Candida sp ou cocos foram fatores associados a
proteção. Com relação à correlação citohistológica, encontramos concordância de 74,08% nas pacientes
com lesão de alto grau e biópsia no mesmo período.
A retrospective study on cervical intraepithelial lesions of low-grade and undetermined significance:
evolution, associated factors and cytohistological correlation | ORIGINAL ARTICLE
INTRODUCTION
The human papillomavirus (HPV) plays a central role in cervical
carcinogenesis, and around it revolve various factors that directly
or indirectly influence whether or not changes in the cervical squamous epithelium occur that can evolve into cancer.1 Among the
factors most studied are immunological factors, smoking, age,
pregnancy, use of hormonal contraceptives, color and microbiota. All these, to a greater or lesser extent, appear to boost the
virus’s action in host cells and to facilitate carcinogenesis. The high
rate of spontaneous regression in cases of HPV infection and the
small percentage that evolve into invasive neoplasia suggest that
viral infection by itself is not sufficient and that other variables are
involved in this process.1
Information about the prevalence of HPV is most frequently
obtained at the start of an individual’s sex life, i.e. during adolescence or at around 20 years of age.2 This infection is fleeting in
most cases, without clinical manifestation and can become spontaneously cured. There is very large variability in the incidence of
infection among white and black women within the same population, although this pattern seems to have become more frequent over recent years.3
In the vagina’s normal microbiota, Lactobacillus sp produces
acidic pH (3.8 to 4.5), which inhibits the growth of various other
kinds of bacteria. On the other hand, vaginal content in which
there is an absence or low concentration of Lactobacillus sp may
be associated with pathogenic processes.4 Therefore, this bacterium has an important role in infection control, in cytolytic vaginosis and in maintenance of a healthy genital tract.
Smoking can also be correlated with higher incidence and persistence of HPV infection and evolution to dysplasia/carcinoma
in situ and invasive neoplasia.5 Its importance in oncogenesis is
already well known, and high concentrations of tobacco derivatives such as nicotine have been observed in the cervical area.6-8
Women with multiple sexual partners, who start sex activity early
and who are smokers or the partners of smokers also present
higher risk of developing cervical intraepithelial neoplasias.3
The greater frequency of HPV infection among pregnant
women than among non-pregnant women suggests that pregnancy is a risk factor for this infection. The maximum clinical expression of infection occurs during gestation, with rapid
regression during the puerperium. This increase in incidence
may be explained by immunological modulation or by the influence of hormonal factors during gestation. It is known that gestation gives rise to imbalance in the vaginal microbiota, thus favoring infections such as HPV, as well as other infectious agents.9
OBJECTIVES
To analyze the evolution to high-grade intraepithelial lesion,
related risk or protection factors and cytohistological correlations
among patients with cytological diagnoses of HPV infection,
low-grade intraepithelial lesion with or without an association
with HPV, squamous cell atypia of undetermined significance or
glandular cell atypia of undetermined significance.
METHODS
This study was approved by the university’s Ethics Committee on
March 28, 2008 (protocol number 1032). Among all the patients
followed up within Gynecology and Obstetrics Department at a
public tertiary-level university hospital between 1995 and 2000,
we reviewed the records and/or colposcopy reports of 3,390
patients with diagnosis of low-grade intraepithelial lesions or of
lesions of undetermined significance: 1,398 (41.24%) had a cytological diagnosis of HPV infection; 73 (2.15%) had low-grade
intraepithelial lesions with or without an associated diagnosis
of HPV infection; 1,689 (49.83%) had squamous cell atypia; and
230 (6.78%) had glandular cell atypia of undetermined significance. We sought information about age, color, age when sexual activity started, number of partners, number of pregnancies,
smoking, use of hormonal contraceptives, immunosuppression, other infections or associated changes (cytolytic vaginosis, Trichomonas vaginalis, bacterial vaginosis, Candida sp and
cocci), by reviewing the records and/or colposcopy reports.
We performed a statistical analysis using logistic regression,
odds ratio calculations and chi-square tests, seeking to evaluate
whether these factors were or were not associated with a risk of
or protection against evolution of the lesions. Of the 409 patients
who evolved cytologically to high-grade intraepithelial lesions,
only 297 (72.62%) had undergone a biopsy at our service during
the same period.
RESULTS
Among the 409 patients who evolved to high-grade intraepithelial lesions, 354 (86.55%) had an initial diagnosis of HPV infection, 22 (5.38%) had low-grade intraepithelial lesions with or
without a diagnosis of HPV infection, 27 (6.60%) had squamous
cell atypias of undetermined significance and 6 (1.47%) had glandular cell atypias of undetermined significance. The average age
of the patients with a cytological diagnosis of HPV infection was
28.52 years (± 11.04); the average age of those with a diagnosis
of HPV infection that evolved into a high-grade intraepithelial
lesion was 33.80 years (± 12.95); and the average age of those who
evolved and did not have a diagnosis of HPV infection was 30.23
years (± 11.30).
Among the 1,432 patients (42.24%) with a diagnosis of HPV
infection or low-grade intraepithelial lesions associated with a
diagnosis of HPV infection, 336 (23.46%) presented evolution
and 1,096 (76.54%) did not. Among the 1,958 patients (57.76%)
who did not have a cytological diagnosis of HPV infection,
Sao Paulo Med J. 2014; 132(2):92-6 93
ORIGINAL ARTICLE | Silva C, Almeida ECS, Côbo EC, Zeferino VFM, Murta EFC, Etchebehere RM
73 (3.73%) evolved and 1,885 (96.27%) did not. The difference in
evolution between the two groups proved to be of extremely high
statistical significance (P < 0.0001).
When the risk factors evaluated were compared with whether
or not the patient evolved to a high-grade intraepithelial lesion,
we found that among the 509 smokers, 140 (27.50%) evolved
and among the 107 immunodepressed individuals, 14 (13.08%)
did. There was a cytological diagnosis of bacterial vaginosis
for 711 patients, of whom 93 (13.08%) evolved to a high-grade
intraepithelial lesion; 2,184 patients had a cytological diagnosis
of Lactobacillus sp, of whom 214 (9.80%) evolved. The statistical analysis showed that smoking, immunosuppression, bacterial vaginosis and Lactobacillus sp were risk factors. Moreover,
as expected, HPV infection was considered to be a risk factor
for evolution. In evaluating the numbers of patients who used
hormonal contraceptives (552; 84.40%), had a single partner
(621; 86.01%), were infected with Candida sp (560; 89.03%),
were infected with cocci (269; 89.67%), were infected with
Trichomonas vaginalis (135; 88.82%) and presented cytolytic
vaginosis (175; 92.11%), but who did not evolve to a high-grade
intraepithelial lesion, we observed that these factors were associated with protection.
Parity and age were univariate analyses and were not included
in the logistic regression. The statistical analysis showed that neither
white skin nor pregnancy interfered with evolution. The factors studied and their associations with a risk of or protection against evolution to high-grade intraepithelial lesions are summarized in Table 1.
Only 297 of the 409 patients who evolved to high-grade
intraepithelial lesions had undergone a biopsy performed within
our service, with histological concordance of 74.07%.
Table 1. Factors evaluated among patients with or without evolution
from HPV, low-grade intraepithelial lesions or squamous or glandular
lesions of undetermined significance to high-grade intraepithelial
lesion, between 1995 and 2000
Factor
Human papillomavirus
Smoking
Immunosuppression
Bacterial vaginosis
Lactobacillus sp
Hormonal contraceptive
Single partner
Candida sp
Cocci
Trichomonas vaginalis
Cytolytic vaginosis
White color
Pregnancy
94
Evolution
n (%)
336 (23.46%)
140 (27.50%)
14 (13.08%)
93 (13.08%)
214 (9.80%)
102 (15.60%)
101 (13.99%)
69 (10.97%)
31 (10.33%)
17 (11.18%)
15 (7.89%)
267 (11.07%)
70 (12.89%)
Sao Paulo Med J. 2014; 132(2):92-6
Without
Total
evolution
n (%)
n (%)
1,096 (76.54%) 1,432 (42.24%)
369 (72.50%)
509 (15.01%)
93 (86.92%)
107 (3.16%)
618 (86.92%)
711 (20.97%)
1,970 (90.20%) 2,184 (64.42%)
552 (84.40%)
654 (19.29%)
621 (86.01%)
722 (21.30%)
560 (89.03%)
629 (18.55%)
269 (89.67%)
300 (8.85%)
135 (88.82%)
152 (4.48%)
175 (92.11%)
190 (5.60%)
2,145 (88.93%) 2,412 (71.15%)
473 (87.11%)
543 (16.02%)
DISCUSSION
The average age of the patients with diagnoses of HPV infection,
low-grade intraepithelial lesion, atypias in squamous or glandular cells of undetermined significance, which evolved into highgrade intraepithelial lesions, was 33.80 years (± 12.95). Among
the patients who did not evolve, the average age was 32.36 years
(± 12.26). Evolution of the lesions is slow when it occurs, and the
average time that elapses between infection and manifestation of
a high-grade lesion or cancer is up to 15 years,10 thus explaining
the higher average age of our patients who evolved.
Our data and that of similar studies11-14 confirm that HPV
infection is fundamental to the development of high-grade
intraepithelial lesions, thus making it a veritable precursor
for cervical cancer. Even among the 73 patients (3.73%) who
did not have a cytological diagnosis of HPV infection, we cannot completely discount this association, since we did not perform molecular biology tests, which are considered to be more
sensitive and specific. Another study conducted in our region,
also on patients with a cytological diagnosis of HPV infection,
found a probability of evolution to high-grade intraepithelial lesions of 0.4%, over a four-year period. Nevertheless, the
authors of that study were unable to determine the risk factors
for persistence or evolution of the infection.15 Other authors
have reported that around 10% of patients with the HPV infection present evolution.13,16,17
It is believed that the relationship between evolution to a
high-grade epithelial lesion and presence of Lactobacillus sp
stems from the high incidence of this microbiota, which is predominant in the vaginal environment,18 and not its actual interference in lesion evolution. In relation to smoking, 27.5% of the
patients who reported this habit evolved, which was a statistically significant result. The importance of smoking to oncogenesis is already well known, and a high concentration of tobacco
derivatives such as nicotine has been isolated in the cervical area.
Smoking has been correlated with the incidence and persistence
of HPV infection and its evolution to dysplasia, carcinoma in situ
and invasive neoplasia.6
Cellular immune response seems to play an important role
in curing HPV infection. There is a high prevalence of infection
or pre-neoplastic lesions in people with compromised immune
systems, such as those with renal transplants and individuals
with acquired immunodeficiency syndrome.19-21 Another study
conducted at our service demonstrated that there was strong
expression of CD3+ lymphocytes in patients colonized by cervical intraepithelial neoplasia grade III (CIN III) who presented
recurrence of the lesion, thus suggesting that these lymphocytes
are of key importance in lesion evolution.22
We also observed that there was a positive association between
bacterial vaginosis and the patients’ evolution. In the literature, a
A retrospective study on cervical intraepithelial lesions of low-grade and undetermined significance:
evolution, associated factors and cytohistological correlation | ORIGINAL ARTICLE
significant association between HPV DNA and a microbiota indicative of bacterial vaginosis is shown. Some authors have suggested
that bacterial vaginosis has an important role in the development
of cervical neoplasia because of production of oncogenic nitrosamines from anaerobic bacteria, and also through stimulation of
production of cytokines such as interleukin 1B. Another possibility that might favor evolution of lesions associated with bacterial
vaginosis would be changes to vaginal pH, which our study was
unable to evaluate, since it was retrospective and our service did
not routinely perform evaluations on vaginal pH.23
With regard to other factors, hormonal contraceptives have
been described in the literature as being associated with cell
transformation and progression from low-grade to high-grade
lesions.24 This was contrary to what our study observed, in which
use of hormonal contraceptives was a protective factor. Having
a single partner probably functions as a protective factor, as
observed in our study. However, the behavioral patterns of these
partners and their ages in relation to those of the women perhaps
also need to be observed. Some authors have taken the view that
these factors are as important as the number of partners.25-27
Infection with Candida sp also proved to be a factor protecting against evolution. Other authors have proposed that candidiasis might activate latent HPV infection.28,29 We believe that a
very low pH, which favors infection by Candida sp, is one of the
factors associated with this protection. A reduction in pH makes
the vagina inhospitable to certain bacterial species30 and probably
makes it difficult for HPV infection to evolve.
The presence of cocci in the vaginal microbiota also appears to
confer protection against evolution. Their presence appears to be
related mainly to inadequate hygiene habits and not to changes in
vaginal pH.31 Utagawa et al. have suggested that socioeconomic
status and inadequate hygiene are key factors for HPV infection.32
Despite this claim, we did not find any studies correlating the presence of cocci in cervicovaginal cytological samples with HPV
infection and its evolution.
Defining factors that can enhance or minimize cervical viral
carcinogenesis is very important for clinical practice because
these have an impact on development of the precursor lesions.
However, further studies are needed to clarify the mechanisms of
action of these factors.
with protection. With regard to cytohistological correlation, there
was a 74.08% agreement among patients with high-grade lesions
who had undergone a biopsy during the same period.
REFERENCES
1. Pinto AP, Túlio S, Cruz OR. Co-fatores do HPV na oncogênese cervical. [HPV
cofactors in cervical carcinogenesis]. Rev Assoc Med Bras. 2002;48(1):73-8.
2. Schiffman MH. Recent progress in defining the epidemiology of
human papillomavirus infection and cervical neoplasia. J Natl Cancer
Inst. 1992;84(6):394-8.
3. Murta EFC, Franca HG, Carneiro MC, et al. Câncer do colo uterino:
correlação com o início da atividade sexual e paridade. [Cervical
cancer: analysis of first sexual intercourse and parity]. Rev Bras
Ginecol Obstet. 1999;21(9):555-9.
4. Soper DE. Gynecologic Complications of Bacterial Vaginosis: Fact or
Fiction? Curr Infect Dis Rep. 1999;1(4):393-7.
5. Murta EFC, Souza MAH, Adad SJ, Araújo Júnior E. Infecção pelo
papilomavírus humano em adolescentes: relação com o método
anticoncepcional, gravidez, fumo e achados citológicos [Human
papillomavirus infection in adolescents: relation to contraceptive
method, pregnancy, smoking, and cytologic findings]. Rev Bras
Ginecol Obstet. 2001;23(4):217-21.
6. Sasson IM, Haley NJ, Hoffmann D, et al. Cigarette smoking and
neoplasia of the uterine cervix: smoke constituents in cervical mucus.
N Engl J Med. 1985;312(5):315-6.
7. Barton SE, Maddox PH, Jenkins D, et al. Effect of cigarette smoking
on cervical epithelial immunity: a mechanism for neoplastic change?
Lancet. 1988;2(8612):652-4.
8. Duggan-Keen MF, Brown MD, Stacey SN, Stern PL. Papillomavirus
vaccines. Front Biosci. 1998:D1192-208.
9. Faro S. Bacterial vaginitis. Clin Obstet Gynecol. 1991;34(3):582-6.
10. Fedrizzi EN, Schlup CG, Menezes ME, Ocampos M. Infecção pelo
Papilomavírus humano (HPV) em mulheres de Florianópolis, Santa
Catarina [Human Papillomavirus (HPV) infection in women of Florianópolis,
Santa Catarina, Brazil]. DST J Bras Doenças Sex Transm. 2008;20(2):73-9.
11. Muñoz N, Bosch FX, de Sanjosé S, Shah KV. The role of HPV in the
etiology of cervical cancer. Mutat Res. 1994;305(2):293-301.
12. Schiffman MH. New epidemiology of human papillomavirus infection
and cervical neoplasia. J Natl Cancer Inst. 1995;87(18):1345-7.
13. Brasil. Ministério da Saúde. Secretaria de Atenção à Saúde. Instituto
Nacional de Câncer. Coordenação de Prevenção e Vigilância.
CONCLUSIONS
According to our study, we can conclude that Lactobacillus sp and
bacterial vaginosis in smears, smoking and immunodepression
were factors associated with evolution of low-grade intraepithelial lesions or lesions of undetermined significance to high-grade
intraepithelial lesions. A single partner, use of hormonal contraceptives, lower parity, age and a cytological diagnosis of cytolytic
vaginosis, T. vaginalis, Candida sp or cocci were factors associated
Nomenclatura brasileira para laudos cervicais e condutas
preconizadas: recomendações para profissionais de saúde. Rio de
Janeiro: INCA; 2006. Available from: http://bvsms.saude.gov.br/bvs/
publicacoes/Nomenclaturas_2_1705.pdf. Accessed in 2013 (May 3).
14. Brasil. Ministério da Saúde. DST-AIDS Hepatites Virais. Departamento
de DST, Aids e Hepatites Virais. Condiloma acuminado (HPV). Available
from:
http://www.aids.gov.br/pagina/condiloma-acuminado-hpv.
Accessed in 2013 (May 3).
Sao Paulo Med J. 2014; 132(2):92-6 95
ORIGINAL ARTICLE | Silva C, Almeida ECS, Côbo EC, Zeferino VFM, Murta EFC, Etchebehere RM
15. Murta EFC, Souza MAH, Lombardi W, Lombardi B, Borges LS.
Aspectos epidemiológicos da infecção pelo papilomavírus humano.
[Epidemiologic aspects of human papilomavirus infection]. J Bras
27. Koutsky LA, Galloway DA, Holmes KK. Epidemiology of genital human
papillomavirus infection. Epidemiol Rev. 1988;10:122-63.
28. Voog E, Ricksten A, Löwhagen GB. Prevalence of Epstein-Barr virus and
human papillomavirus in cervical samples from women attending an
Ginecol. 1997;107(4):95-9.
16. Lungu O, Sun XW, Felix J, et al. Relationship of human papillomavirus
type to grade of cervical intraepithelial neoplasia. JAMA.
STD-clinic. Int J STD AIDS. 1995;6(3):208-10.
29. Murta EF, Souza MA, Araújo Júnior E, Adad SJ. Incidence of Gardnerella
vaginalis, Candida sp and human papilloma virus in cytological
1992;267(18):2493-6.
17. Passos MRL. HPV - Que bicho é esse? DST J Bras Doenças Sex Transm.
smears. Sao Paulo Med J. 2000;118(4):105-8.
30. Santana AP, Gomes AC, Marques KJF, et al. Afecções ginecológicas
2001;13(5):3.
18. Donders GG, Bosmans E, Dekeersmaecker A, et al. Pathogenesis
evidenciadas no Papanicolaou em uma unidade de Saúde da Família
of abnormal vaginal bacterial flora. Am J Obstet Gynecol.
na cidade de João Pessoa. In: Anais Eletrônicos do X Encontro de
2000;182(4):872-8.
Extensão da UFPB-RAC. João Pessoa: UFPB; 2008. Available from:
19. Camara GNNL, Cruz MR, Veras VS, Martins CRF. Os papilomavírus
humanos - HPV: carcinogênese e imunogênese. Universitas
http://www.prac.ufpb.br/anais/xenex_xienid/x_enex/ANAIS/
Area6/6CCSETSOUT_01.pdf. Accessed in 2013 (May 3).
www.
31. Nai GA, Mello ALP, Ferreira AD, Barbosa RL. Frequência de
publicacoesacademicas.uniceub.br/index.php/cienciasaude.
Gardnerella vaginalis em esfregaços vaginais de pacientes
Accessed in 2013 (May 3).
histerectomizadas [Frequency of Gardnerella Vaginalis in vaginal
Ciências
Saúde.
2003;1(1):159-68.
Available
from:
20. Ferreira H, Lala ERP, Borba KP. Rastreamento de neoplasia intracervical
em um grupo de mulheres portadoras de vírus da imunodeficiência
smears of hysterectomized women]. Rev Assoc Med Bras (1992).
2007;53(2):162-5.
humana. Procura por neoplasia intracervical em mulheres
32. Utagawa ML, Pereira SMM, Cavalierere MJ, Shirata NK. Lesões
soropositivas [A search for intracervical neoplasia in a group of
precursoras de câncer do colo uterino em adolescentes: impacto em
women bearing the human immunodeficiency virus. A search for
saúde pública [Precursor lesions of the cervical cancer in adolescents:
intracervical neoplasia in women with the HIV virus]. Publicatio
public health impact]. Folha Méd. 2000;119(4):55-8.
UEPG: Ciências Biológicas e da Saúde. 2006;12(3):7-12. Available
from:
http://www.revistas2.uepg.br/index.php/biologica/article/
view/434/435. Accessed in 2013 (May 3).
21. Cameron JE, Hagensee ME. Human papillomavirus infection and
Dissertation presented to the Postgraduate Course on Pathology,
Universidade Federal do Triângulo Mineiro, Uberaba, Minas Gerais, Brazil,
on August 23, 2010
disease in the HIV+ individual. Cancer Treat Res. 2007;133:185-213.
22. Maluf PJ, Michelin MA, Etchebehere RM, Adad SJ, Murta EF.
Sources of funding: None
T lymphocytes (CD3) may participate in the recurrence of
Conflict of interest: None
cervical intraepithelial neoplasia grade III. Arch Gynecol Obstet.
2008;278(6):525-30.
23. Discacciati MG, Rabelo-Santos SH, Campos EA, et al. Vaginose
bacteriana e DNA de papilomavírus humano de alto risco oncogênico
Date of first submission: August 16, 2012
Last received: May 21, 2013
Accepted: June 14, 2013
em mulheres submetidas a conização com alça diatérmica para
tratamento de neoplasia intra-epitelial cervical de alto grau [Bacterial
Address for correspondence:
vaginosis and high-risk HPV-DNA in women submitted to diathermic
Renata Margarida Etchebehere
conization for the treatment of high-grade cervical intra-epithelial
Serviço de Patologia Cirúrgica, HC-UFTM
neoplasia]. Rev Bras Ginecol Obstet. 2004;26(9):721-5.
Rua Getúlio Guaritá, 130
24. Pater A, Bayatpour M, Pater MM. Oncogenic transformation by
human papillomavirus type 16 deoxyribonucleic acid in the presence
CEP 38025-440
of progesterone or progestins from oral contraceptives. Am J Obstet
E-mail: [email protected]
Gynecol. 1990;162(4):1099-103.
25. Okada MMK, Gonçalves MAG, Geraldo PC. Epidemiologia e
patogênese do papilomavírus humano (HPV). In: Oyakama N,
Carvalho JJM. I Consenso Brasileiro de HPV: Papilomavirus humano.
São Paulo: BG Cultural; 2000. p. 1-6.
26. Baseman JG, Koutsky LA. The epidemiology of human papillomavirus
infections. J Clin Virol. 2005;32 Suppl 1:S16-24.
96
Abadia — Uberaba (MG) — Brasil
Sao Paulo Med J. 2014; 132(2):92-6
ORIGINAL ARTICLE
DOI: 10.1590/1516-3180.2014.1322490
Relationship between cardiovascular risk
factors and the echogenicity and pattern
of the carotid intima-media complex in men
Relação entre os fatores de risco cardiovascular e a ecogenicidade
e o padrão do complexo íntima-média carotídeo em homens
Priscilla Lopes da Fonseca Abrantes SarmentoI, Frida Liane PlavnikII, Andrea ScaciotaIII,
Joab Oliveira LimaIV, Robson Barbosa MirandaV, Sergio Aron AjzenVI
Department of Diagnostic Imaging and Department of Nephrology, Universidade Federal de São Paulo (Unifesp), São Paulo, Brazil
MD, MSc, PhD. Doctoral Student of Radiology,
Department of Diagnostic Imaging,
Universidade Federal de São Paulo (Unifesp), São
Paulo, Brazil.
I
II
MD, MSc, PhD. Affiliate Professor, Department of
Nephrology, Universidade Federal de São Paulo
(Unifesp), São Paulo, Brazil.
BSc. Computer Graphics Specialist, Coordinator
of Teaching and Research Support for Residents
and Postgraduate Students, Department of
Diagnostic Imaging, Universidade Federal de
São Paulo (Unifesp), São Paulo, Brazil.
III
IV
MD, PhD. Adjunct Professor, Department
of Statistics, Universidade Federal da Paraíba
(UFPB), João Pessoa, Paraíba, Brazil.
MD. Assistant Professor, Department of Surgery,
Faculdade de Medicina do ABC (FMABC), Santo
André, São Paulo, Brazil.
V
VI
MD, MSc, PhD. Titular Professor, Head of
the Department of Diagnostic Imaging,
Universidade Federal de São Paulo (Unifesp), São
Paulo, Brazil.
KEY WORDS:
Ultrasonography.
Carotid intima-media thickness.
Atherosclerosis.
Risk factors.
Metabolic syndrome X.
PALAVRAS-CHAVE:
Ultrassonografia.
Espessura intima-média carotídea.
Aterosclerose.
Fatores de risco.
Síndrome X metabólica.
ABSTRACT
CONTEXT AND OBJECTIVE: The thickness of the carotid intima-media complex (C-IMC) is considered to
be a marker of early atherosclerosis, but visual and echogenic changes to the C-IMC can also be noted.
The objective here was to evaluate the relationship between cardiovascular risk factors and the echogenicity of the C-IMC and identify those most associated with an “abnormal” C-IMC.
DESIGN AND SETTING: Cross-sectional study in the ultrasound sector of the Department of Diagnostic
Imaging, Universidade Federal de São Paulo.
METHODS: Eighty men were evaluated. Measurements of arterial blood pressure, waist circumference (WC), lipid profile, fasting glucose, uric acid and high-sensitivity C-reactive protein were obtained.
The thickness of the C-IMC was measured by means of B-mode ultrasound, and the intima-media grayscale mean (IM-GSM) and standard deviation (IM-SD) were calculated.
RESULTS: The following were discriminating variables: fasting glucose (r2 = 0.036; P = 0.013), uric acid
(r2 = 0.08; P = 0.03), IM-SD (r2 = 0.43; P < 0.001), IM-GSM (r2 = 0.35; P < 0.001) and thickness of the C-IMC
(r2 = 0.29; P < 0.001). IM-GSM showed significant correlations with WC (r = -0.22; P = 0.005), fasting glucose
(r = -0.24; P = 0.002) and high-density lipoprotein cholesterol (HDL-C) (r = 0.27; P = 0.0007).
CONCLUSION: IM-GSM showed correlations with WC, fasting glucose and HDL-C. However, uric acid and
IM-SD presented the greatest discriminating impact. These results suggest that visual changes in C-IMC
may help identify patients with potential cardiovascular risk, independently of the thickness of the C-IMC.
RESUMO
CONTEXTO E OBJETIVO: A espessura do complexo íntima-média carotídeo (CIM-C) é considerada um
marcador da aterosclerose precoce, mas alterações visuais e da ecogenicidade do CIM-C também podem
ser observadas. O objetivo foi avaliar a relação entre os fatores de risco cardiovascular e a ecogenicidade
do CIM-C e identificar aqueles mais relacionados com o CIM-C “alterado”.
TIPO DE ESTUDO E LOCAL: Estudo transversal no setor de ultrassonografia do Departamento de Diagnóstico por Imagem, Universidade Federal de São Paulo.
MÉTODOS: Oitenta homens foram avaliados. Aferição da pressão arterial, medida da circunferência abdominal (CA), perfil lipídico, glicemia de jejum (GLI), ácido úrico (AU) e proteína C-reativa de alta sensibilidade
foram obtidos. A espessura do CIM-C foi medida por ultrassom modo B e a média da escala de cinza (GSM)
e do desvio padrão do CIM (DPIM) foram calculados.
RESULTADOS: As variáveis discriminantes foram GLI (r2 = 0,036; P = 0,013), AU (r2 = 0,08; P = 0,03), DPIM
(r2 = 0,43; P < 0,001), GSM (r2 = 0,35; P < 0,001) e espessura do CIM-C (r2 = 0.29; P < 0,001). Houve correlação
significativa entre GSM e CA (r = -0,22; P = 0,005), GLI (r = -0,24; P = 0,002) e lipoproteína de alta densidade
do colesterol (HDL-C) (r = 0,27; P = 0,0007).
CONCLUSÃO: A GSM teve correlação com CA, GLI, HDL-C. Entretanto, AU e DPIM apresentaram maior
impacto discriminante, sugerindo que alterações visuais do CIM-C, independentemente da espessura,
podem auxiliar na identificação de pacientes com potencial risco cardiovascular.
Sao Paulo Med J. 2014; 132(2):97-104
97
ORIGINAL ARTICLE | Sarmento PLFA, Plavnik FL, Scaciota A, Lima JO, Miranda RB, Ajzen SA
INTRODUCTION
Atherosclerosis is a chronic inflammatory disease that occurs
mainly in large and medium-sized elastic and muscular arteries.1
The characteristic changes of atherosclerosis, such as inflammatory lesions or endothelial dysfunction, represent different stages
of the disease.1
Clinically evident atherosclerosis is preceded by subclinical
changes in the arterial wall.2 The morphological characteristics
of the carotid artery, like the intima-media thickness, correspond
to the existence of vascular diseases.3,4 Carotid intima-media
thickness is a commonly used measurement of atherosclerosis,5
and major traditional cardiovascular risk factors, like hypertension, diabetes, obesity and hypercholesterolemia, are associated
with increased carotid intima-media thickness and with the risk
of cardiovascular events.6 Since atherosclerosis is an inflammatory disease, there is a growing amount of evidence that biomarkers like high-sensitivity C-reactive protein present increased levels in individuals with cardiovascular disease.2
In evaluating carotid arteries by means of ultrasound, it
becomes evident from visual inspection of the intima-media
complex that, even in intima-media complexes of normal thickness, there is great variation in echogenicity, texture and intimamedia pattern.5,7 It is possible that such changes may precede
development of a significant increase in the thickness of the
carotid intima-media complex.
The variations in texture and echogenicity observed by a
physician during ultrasound examinations can be objectively
evaluated from the visual pattern of the carotid intima-media
complex and from the gray-scale mean, respectively.5,8 The
echogenicity of the intima-media complex may also be related
to risk factors other than the carotid intima-media thickness,
as can be seen in the brachial artery.5 These changes have also
been correlated with age and risk factors for atherosclerosis.9
In older subjects, the intima-media layer is more discontinuous and irregular, and these changes could account for the leukocytes and fatty streaks clustered along the endothelium of
atherosclerotic vessels that are seen on electron micrographs.9
Subclinical atherosclerotic lesions typically begin as endothelial
damage with gradual intima-media thickening and development
of a more granular appearance on ultrasound, which represents
a more advanced stage of atherosclerosis and precedes development of significant thickening.9
OBJECTIVE
The objectives of this study were to evaluate the relationship
between cardiovascular risk factors and the echogenicity of the
carotid intima-media complex and to identify the factors most
associated with recognition of an “abnormal” carotid intimamedia complex.
98
Sao Paulo Med J. 2014; 132(2):97-104
METHODS
Population
Eighty male volunteers aged 30 to 60 years were included
in this study, from March to October 2010. We used a convenience sample size to test our hypothesis in this initial study. All the volunteers were selected and referred
from the nephrology outpatient clinic of Hospital do Rim e
Hipertensão, Universidade Federal de São Paulo (Unifesp).
They had signed a consent form, approved by the Ethics
Committee of Unifesp.
The criteria for exclusion included insulin-dependent
diabetes, history of smoking or use of statins. Any presence
of plaque in the common carotid artery was also a criterion
for exclusion.
Anthropometric measurements such as weight (kg),
height (m), and waist circumference (cm) were made with the
volunteer wearing light clothes and no shoes. Body mass index
was calculated by dividing weight (kg) by height squared (m2),
and waist circumference was measured at the level of the umbilicus with the volunteers in the supine position.
A fasting blood sample was collected in order to assess
lipid profile (total cholesterol, low-density lipoprotein-cholesterol, high-density lipoprotein-cholesterol, and triglycerides), triglyceride/high-density lipoprotein-cholesterol ratio
(atherogenic index), fasting glucose, C-reactive protein and
uric acid.
Systolic and diastolic blood pressures were measured in the
volunteers in the seated position, after five minutes of resting,
using a calibrated sphygmomanometer. Three consecutive measurements were made and the average of the last two measurements was used for the analysis.
The volunteers were divided into two groups: with and
without metabolic syndrome, which was defined in accor­
dance with the National Cholesterol Education Program/Adult
Treatment Panel III (NCEP/ATP III) criteria.10
Duplex scanning
The carotid intima-media thickness was measured by means of
B-mode ultrasound (Accuvix V10, Medison) with a 7.5 MHz to
10 MHz linear array transducer. All the examinations were performed in the ultrasound sector of the Department of Diagnostic
Imaging, Unifesp. The volunteers were examined in the supine
position, by a physician who was blind to the patients’ cardiovascular risk factors. Three equidistant measurements were made on
the posterior wall of the right and left common carotid arteries,
at a distance of 1 cm to 1.5 cm from the bulb. All the images were
produced using the same time gain compensation; the depth and
the B-mode gain could be altered, since these make no difference
to the results.
ORIGINAL ARTICLE | Sarmento PLFA, Plavnik FL, Scaciota A, Lima JO, Miranda RB, Ajzen SA
Significantly higher values for waist circumference, C-reactive
protein, uric acid, systolic blood pressure and triglycerides, and
significantly lower values for high-density ­lipoprotein-cholesterol,
were seen among the metabolic syndrome patients (Table 1).
The values for the carotid intima-media thickness were within
what are considered to be the normal values on the right or left
sides, but the carotid intima-media thickness was significantly
higher in the individuals with metabolic syndrome (0.59 ± 0.14
versus 0.68 ± 0.16; P < 0.0007). The intima-media grayscale mean
and standard deviation were different between the groups with
and without metabolic syndrome, but not significantly. Table 2
shows the ultrasound analysis.
Plaque was observed in 42 bulbs (26.25%) promoting < 50%
stenosis in internal carotid arteries, which was in accordance
with the criteria established by the University of Washington.13
No significant stenosis was measured, and there was no significant differences between the groups.
To analyze technical influences, images of the same carotid artery
were produced using the same time gain compensation but changing
the depth, B-mode gain and type of insonation (anterior or posterior
to sternocleidomastoid muscle). No significant changes in the intimamedia grayscale mean or standard deviation values were noted.
Visual classification of the arteries
There was no significant difference between the two clinical groups
in relation to the intima-media grayscale means and standard
deviations, from analysis on the entire intima-media complex
of 160 carotid arteries. 43.2% (69/160) of the common carotid
arteries were visually classified as normal and 56.8% (91/160) as
abnormal. The interobserver agreement was significant (K = 1;
P = 0.0000* in the group with metabolic syndrome; and K = 0.79;
P = 0.0014* in the group without metabolic syndrome).
Correlations between 160 arteries and the clinical and
ultrasound variables
In the “abnormal” arteries, the carotid intima-media thickness was
greater (0.68 mm ± 0.15) than in the arteries that were considered
“normal” (0.55 ± 0.11 mm) (P < 0.0001). The correlations between
the carotid intima-media thickness and the variables available
from the 160 arteries were significant in relation to age (r = 0.21;
P = 0.007); presence of metabolic syndrome (r = 0.23; P = 0.003);
fasting glucose (r = 0.20; P = 0.01); triglycerides (r = 0.39; P = 0.000);
and low-density lipoprotein-cholesterol (r = 0.29; P = 0.002).
Regarding changes in echogenicity, the intima-media grayscale mean presented significant inverse correlations with waist
circumference (r = -0.22; P = 0.005) and with fasting glucose
(r = -0.24; P = 0.002); and a positive association with ­high-density
lipoprotein-cholesterol (r = 0.27; P = 0.0007). The textural
changes relating to the intima-media standard deviation only
showed a significant inverse correlation with the presence of metabolic syndrome when evaluated qualitatively (Spearman’s correlation: r = -0.14; P = 0.049).
Table 1. Baseline clinical data
Variable
Without metabolic syndrome
(n = 45)
With metabolic syndrome
(n = 35)
P-value
Age (y)
48 ± 7
50 ± 7
> 0.05
WC (cm)
93 ± 9
107 ± 9
0.0001
SBP (mmHg)
133 ± 15
142 ± 16
0.02
DBP (mmHg)
92 ± 12
93 ± 9
0.47
HDL-C (mg/dl)
47 ± 10
37 ± 7
0.0001
TRI (mg/dl)
129 ± 65
194 ± 98
0.0007
Glucose (mg/dl)
88 ± 14
93 ± 15
0.15
Uric acid (mg/dl)
6.5 ± 1.2
7.3 ± 1.6
0.02
0.39 (0.02-2.79)
0.71 (0.03-6.26)
0.002
2.9 ± 1.6
5.4 ± 3.0
0.00001
hs-CRP*
Atherogenic index
WC = waist circumference; SBP = systolic blood pressure; DBP = diastolic blood pressure; HDL-C = high-density lipoprotein cholesterol;
TRI = triglycerides; hs-CRP = high-sensitivity C-reactive protein; *value expressed as median (min-max).
Table 2. Ultrasound measurements
Variable
Intima-media thickness (mm)
Intima-media gray-scale mean (IM-GSM)
Intima-media standard deviation (IM-SD)
100
Sao Paulo Med J. 2014; 132(2):97-104
Without metabolic syndrome
(n = 90)
0.59 ± 0.14
38 ± 17
17.3 ± 5.5
With metabolic syndrome
(n = 70)
0.68 ± 0.16
32 ± 15
17.8 ± 5.3
P-value
0.0007
0.08
0.09
Relationship between cardiovascular risk factors and the echogenicity and pattern of the carotid intima-media complex in men | ORIGINAL ARTICLE
Multiple linear regression analysis considering the visual classification of 160 arteries and only the clinical variables showed
that the presence of diabetes mellitus was the only variable to have
a statistically significant impact (r2 = 0.036; P = 0.016), in distinguishing between “normal” and “abnormal” arteries. In analyzing the laboratory variables, it became evident that increased levels of both fasting glucose (r2 = 0.036; P = 0.013) and uric acid
(r2 = 0.08; P = 0.03) had significant impacts. Considering the
intima-media variables, the intima-media standard deviation
was the factor of greatest impact (r2 = 0.43; P < 0.001) followed
by the intima-media grayscale mean (r2 = 0.35; P < 0.001) and
intima-media thickness (r2 = 0.29; P < 0.001).
DISCUSSION
According to the World Health Organization, there are approximately one billion people worldwide who are overweight.
Of these, 300 million are obese.14 High serum triglyceride and
low serum high-density lipoprotein-cholesterol levels and glucose abnormalities associated with obesity are also risk factors
predictive of cardiovascular disease. These metabolic changes are
components of the so-called metabolic syndrome.15 Metabolic
syndrome is not clearly defined in the literature, and this definition has been continuously modified over the last few years.10
Despite the link between metabolic syndrome and obesity, not all
obese patients have the metabolic abnormalities of metabolic syndrome. These individuals are described as “metabolically healthy
obese patients”.16 Although clinical events occur after such individuals reach their fifties, early signs of atherosclerosis can be
detected in their twenties and thirties.17,18 The risk of future cardiovascular events and even death becomes greater with the presence of a single component of metabolic syndrome.19
In our study, the most prevalent cardiovascular risk factors
were hypertension, obesity shown by waist circumference and
elevated triglycerides. Uric acid can also be considered to be an
independent risk factor for cardiovascular disease, since it causes
hypertension.20 Increasing evidence is suggesting that uric acid
may play a role in metabolic syndrome, given that hyperuricemia
is present in individuals with metabolic syndrome who are not
overweight or obese.20 We found a correlation between uric acid
and intima-media echogenicity that accounted for about 8% of
the change in the visual pattern.
Age was correlated with the intima-media thickness, but not
with intima-media echogenicity, perhaps because of the homogeneity of the patients’ ages and the small sample size for the
age group.
In the overall assessment of carotid intima-media complex patterns, systolic blood pressure did not show any significant impact. One of the arguments to explain this might be the
fact that all the hypertensive patients were being treated with
antihypertensive medication that was not suspended at any time
during the study.
Kablak-Ziembicka et al. showed that both the intima-media
thickness and the C-reactive protein and tumor necrosis factoralpha levels stratify cardiovascular events, thereby independently
contributing towards the classic risk factors. Thus, these authors
showed that in situations with elevated levels of these markers,
the probability of event-free survival decreased.2 Other studies
have concluded that increases in C-reactive protein levels are
associated with higher cardiovascular risk only in the presence
of carotid atherosclerosis.5,7,8 The C-reactive protein values were
similar in the two groups of patients studied and showed no
difference in relation to intima-media echogenicity.
Dyslipidemia plays an important role in the progression
of atherosclerosis and thickness of the intima-media complex.
Statin administration has the aim of modifying the lipid profile, and the patients in this study had their medication withdrawn thirty days before the beginning of the evaluations. The
ultrasound variables (intima-media gray-scale mean, intimamedia standard deviation and carotid intima-media thickness)
were correlated with lipid profile, such that the triglyceride levels correlated positively and significantly with carotid intimamedia thickness and high-density lipoprotein-cholesterol with
the intima-media grayscale mean.
Therefore, we believe that the changes in echogenicity, texture and intima-media pattern occur earlier than or occur simultaneously with the increase in intima-media thickness, given that
abnormal arteries had thicker intima-media complexes, although
still within the normal limits.
Similar to what was observed with the carotid intima-media
thickness, the visual changes to the intima-media complex also
correlated with the presence of metabolic syndrome and with
increased fasting glucose levels. Other components of metabolic
syndrome such as increased waist circumference and decreased
levels of high-density lipoprotein-cholesterol also correlated
with an abnormal visual pattern. Thus, the artery may already be
abnormal in metabolically impaired patients, but without sufficient criteria for diagnosing metabolic syndrome.
It has been suggested that the distinction between subjects
with metabolic syndrome and those with metabolically impaired
obesity has important implications for therapeutic medical
decision-making.19
Patients with metabolic syndrome present greater numbers of risk factors. On the other hand, in the present study,
those without metabolic syndrome also presented some risk
factors like increased waist circumference and triglyceride levels. On the whole, the inflammatory and atherogenic factors are
determinant in causing arterial changes and increasing the cardiovascular risk. The pattern of the arteries was significantly
Sao Paulo Med J. 2014; 132(2):97-104
101
ORIGINAL ARTICLE | Sarmento PLFA, Plavnik FL, Scaciota A, Lima JO, Miranda RB, Ajzen SA
influenced by metabolic variables (uric acid and fasting glucose)
and ultrasound variables (intima-media standard deviation).
Many investigators have regarded determining the intimamedia thickness as a routine clinical practice for reclassification
of cardiovascular risk in population-based studies.21,22 Intimamedia thickness is widely used as a marker for early carotid atherosclerosis.4,23 Knowing the intima-media thickness is considered to improve cardiovascular risk assessments particularly
among patients with intermediate Framingham scores.22
In type 2 diabetic patients, the extent of the carotid intimamedia complex can help predict coronary events in the same way
as the Framingham score, but using these two indices together
significantly improves the risk prediction.23 Higher fasting glucose levels explained 28% of the increased thickness of the carotid
intima-media complex in the patients with metabolic syndrome
and about 3% of the change in echogenicity and intima-media
pattern, in general. Regarding the echogenicity of the intimamedia complex, its association with cardiovascular risk stratification has not yet been described.
Although the intima-media thickness in the metabolic syndrome group (0.59 ± 0.14) was significantly higher than in the
non-metabolic syndrome group (0.68 ± 0.16) (P = 0.0007),
the average values were within the reference limits for normal
values. Regarding the changes to the intima-media pattern, some
of them that were previously considered to be “normal” could be
classified as either “normal” or “abnormal”. Carotid arteries with
low values for intima-media thickness but with changes to the
intima layer should be noted in order to improve the risk stratification and medical therapy.
Different longitudinal studies that aimed to estimate the predictive value of increased intima-media thickness have used different methods and different populations,24 and thus there is no
consensus as to which values are the best.25 Perhaps the basis for
defining an abnormal intima-media thickness need to be more
effective:21 some studies have suggested that patients with values > 1.0 mm should be treated more aggressively,21 even including use of lipid-lowering therapies. All statins have demonstrated regression in carotid intima-media thickness.26 Although
no study has evaluated the effect of statins on the intima-media
gray-scale mean, patients using statins were excluded from the
present study, or their treatment was interrupted thirty days
before the evaluation.
The carotid intima-media echogenicity (intima-media grayscale mean) has been correlated with risk factors differing from
those used in relation to carotid intima-media thickness.7 In our
opinion, although the intima-media echogenicity is less commonly used,6 it should also be taken into consideration.
Although numerical values for the difference between “normal” and “abnormal” artery grayscale means were not shown,
102
Sao Paulo Med J. 2014; 132(2):97-104
lower means values for the intima-media grayscale were correlated with metabolic syndrome components such as increased
waist circumference, higher blood glucose levels and lower levels
of high-density lipoprotein-cholesterol, and hence with increased
cardiovascular risk.
Andersson et al.7 showed that echogenicity was lower in subjects who had suffered a stroke than in stroke-free subjects, and
that there was a non-significant trend towards a more echolucent intima-media complex in patients who had had myocardial
infarction. Wohlin et al.8 described the intima-media grayscale
mean at around the age of 50 years as a preliminary cutoff for
increased mortality among elderly men. Loizou et al.27 observed
that there was a decrease in the media layer grayscale mean with
increasing age, and gave the explanation for this that in the initial stages of atherosclerotic disease, lipid concentrations increase
and hyperplasia of muscle fibers occurs in the media layer, which
produces hypoechoic (echolucent) structures. With aging and
increasing of the media layer grayscale mean, changes to the
intima-media pattern are observed; the hyperechogenic intima
layer becomes less bright and the double-line pattern becomes
less evident,9,27 or even unrecognizable.11 At the same time, the
intima layer can present irregularities and acoustic holes.8,9
The intimal reflection line is included in the measurement
of the intima-media grayscale mean.5,8 The echogenicity of this
reflection line is usually related to the echogenicity of the darker
space below.5,8
The intimal reflection line comprises only 10-20% of all the
pixels in the region of interest,5,8 but we believe that inclusion of
this line may modify the intima-media grayscale mean. For this
reason, we also evaluated the intima-media standard deviation.
A carotid artery with a normal double-line pattern may, when
the intima layer is brighter and the media layer is hypoechogenic
(darker), has the same value as a carotid without the double line
and with a hyperechogenic intima-media complex. In this case,
the distinction is made using the intima-media standard deviation, which is higher in carotid arteries with a double-line pattern. This suggests that the intima-media grayscale mean alone
is unable to distinguish between “normal” and “abnormal” arteries. The visual criteria for normal arteries correspond to textural
changes, higher entropy and higher intima-media standard deviation. Abnormal intima-media layers with atherosclerotic disease
present lower values for the intima-media standard deviation.
Studies on the echogenicity of the intima-media complex
have only reported isolated values for the intima-media grayscale mean. We did not find any reports on the extent of use of the
intima-media standard deviation as a cofactor in distinguishing
between “normal” and “abnormal” carotid intima-media complexes with analysis on the texture of the carotid intima-media
complex. According to our results, the intima-media standard
Relationship between cardiovascular risk factors and the echogenicity and pattern of the carotid intima-media complex in men | ORIGINAL ARTICLE
deviation was the variable with the greatest impact, accounting
for 45% of the change in the intima-media complex pattern.
In order to test the power of our findings, given that there
is a scarcity of such data in the literature, we used the difference
between the mean values for the intima-media standard deviation and the grayscale mean, comparing normal and abnormal
arteries. This test made it possible to establish the sensitivity
of each variable, i.e. the intima-media grayscale mean and the
intima media-standard deviation.
Since waist circumference was the most significant risk factor in diagnosing metabolic syndrome, and this factor was present in 100% of such patients, and given that the echogenicity of
the carotid intima-media complex was measured from the grayscale mean, we conducted an inverse sample size calculation using
Pockok’s test. This test indicated the minimum sample that would
be needed for this study on 69 arteries for each group (with or
without metabolic syndrome). In this analysis, 90 arteries in the
group without metabolic syndrome and 70 arteries in the group
with metabolic syndrome were evaluated. Thus, the sample size
was found to be fully coherent, adjusted and appropriate.
The main limitation of this study was the lack of a control group. There was some difficulty in selecting healthy men
between their forties and sixties. Many were considered healthy
prior to clinical evaluation and laboratory tests. Even though
none were diagnosed with any classical cardiovascular risk factors, it still remains questionable whether they might have had
other, less prevalent factors, such as metabolic defects or genetic
polymorphisms.9
REFERENCES
1. Ross R. Atherosclerosis--an inflammatory disease. N Engl J Med.
1999;340(2):115-26.
2. Kablak-Ziembicka A, Przewlocki T, Sokolowski A, Tracz W,
Podolec P. Carotid intima-media thickness, hs-CRP and TNF-α
are independently associated with cardiovascular event risk in
patients with atherosclerotic occlusive disease. Atherosclerosis.
2011;214(1):185-90.
3. Graf IM, Schreuder FH, Hameleers JM, et al. Wall irregularity
rather than intima-media thickness is associated with nearby
atherosclerosis. Ultrasound Med Biol. 2009;35(6):955-61.
4. Chambless LE, Folsom AR, Davis V, et al. Risk factors for progression
of common carotid atherosclerosis: the Atherosclerosis Risk in
Communities Study, 1987-1998. Am J Epidemiol. 2000; 155(1):38-47.
5. Lind L, Andersson J, Rönn M, et al. Brachial artery intima-media
thickness and echogenicity in relation to lipids and markers of
oxidative stress in elderly subjects: the prospective investigation
of the vasculature in Uppsala Seniors (PIVUS) Study. Lipids.
2008;43(2):133-41.
6. Lind L, Andersson J, Rönn M, Gustavsson T. The echogenicity
of the intima-media complex in the common carotid artery is
closely related to the echogenecity in plaques. Atherosclerosis.
2007;195(2):411-4.
7. Andersson J, Sundström J, Gustavsson T, et al. Echogenecity of the
carotid intima-media complex is related to cardiovascular risk factors,
dyslipidemia, oxidative stress and inflammation: the Prospective
Investigation of the Vasculature in Uppsala Seniors (PIVUS) study.
Atherosclerosis. 2009;204(2):612-8.
8. Wohlin M, Sundström J, Andrén B, Larsson A, Lind L. An echolucent
CONCLUSION
The classification of arteries into “normal” or “abnormal” patterns significantly influenced variables such as uric acid and
metabolic syndrome components (presence of diabetes, high
fasting blood glucose levels and low high-density lipoproteincholesterol levels) and the three variables of the intima-media
complex (intima-media standard deviation, intima-media
grayscale mean and intima-media thickness). Considering all
these variables, the intima-media standard deviation was the
most relevant factor. The values for the intima-media grayscale mean showed statistically significant correlations with
clinically variable components of metabolic syndrome, such
as high fasting glucose levels, low high-density lipoproteincholesterol and increased waist circumference. The intimamedia standard deviation had a significant correlation with
the presence of metabolic syndrome.
Prospective studies are necessary in order to confirm the
importance of these results in relation to normal values for
the intima-media thickness and evaluate whether these changes
have any clinical impact on predictions of cardiovascular events.
carotid artery intima-media complex is a new and independent
predictor of mortality in an elderly male cohort. Atherosclerosis.
2009;205(2):486-91.
9. Labropoulos N, Leon LR Jr, Brewster LP, et al. Are your arteries older
than your age? Eur J Vasc Endovasc Surg. 2005;30(6):588-96.
10. Expert Panel on Detection, Evaluation, and Treatment of High Blood
Cholesterol in Adults. Executive Summary of the Third Report of the
National Cholesterol Education Program (NCEP) Expert Panel on
Detection, Evaluation, And Treatment of High Blood Cholesterol In
Adults (Adult Treatment Panel III). JAMA. 2001;285(19):2486-97.
11. Pignoli P, Tremoli E, Poli A, Oreste P, Paoletti R. Intimal plus medial
thickness of the arterial wall: a direct measurement with ultrasound
imaging. Circulation. 1986;74(6):1399-406.
12. Belcaro G, Barsotti A, Nicolaides AN. “Ultrasonic biopsy” --a noninvasive screening technique to evaluate the cardiovascular risk and
to follow up the progression and the regression of arteriosclerosis.
Vasa. 1991;20(1):40-50.
13. Grant EG, Benson CB, Moneta GL, et al. Carotid artery stenosis: grayscale and Doppler US diagnosis--Society of Radiologists in Ultrasound
Consensus Conference. Radiology. 2003;229(2):340-6.
Sao Paulo Med J. 2014; 132(2):97-104
103
ORIGINAL ARTICLE | Sarmento PLFA, Plavnik FL, Scaciota A, Lima JO, Miranda RB, Ajzen SA
14. Sánchez-Castillo CP, Pichardo-Ontiveros E, López-R P. Epidemiología
Sources of funding: Coordination Office for the Improvement of Higher
de la obesidad. [The epidemiology of obesity]. Gac Med Mex.
Education Personnel (Coordenação de Aperfeiçoamento de Pessoal de
2004;140(Suppl 2):S3-20.
Nível Superior, Capes) for scholarship no. 33009015.029
15. Ribeiro Filho FF, Mariosa LS, Ferreira SR, Zanella MT. Gordura visceral e
Conflict of interest: None
síndrome metabólica: mais que uma simples associação [Visceral fat
and metabolic syndrome: more than a simple association]. Arq Bras
Date of first submission: March 23, 2012
Endocrinol Metabol. 2006;50(2):230-8.
Last received: May 30, 2013
16. Sims EA. Are there persons who are obese, but metabolically healthy?
Accepted: June 27, 2013
Metabolism. 2001;50(12):1499-504.
17. Stary HC. Evolution and progression of atherosclerotic lesions in
coronary arteries of children and young adults. Arteriosclerosis..
Priscilla Lopes da Fonseca Abrantes Sarmento
1989;9(1 Suppl):I19-32.
Av. Esperança, 189/602
18. Strong JP, Restrepo C, Guzmán M. Coronary and aortic atherosclerosis
Manaíra — João Pessoa (PB) — Brasil
in New Orleans. II. Comparison of lesions by age, sex, and race. Lab
CEP 58038-280
Invest. 1978;39(4):364-9.
Tel. (+55 11) 5908-7911
19. Arnlöv J, Ingelsson E, Sundström J, Lind L. Impact of body mass index
and the metabolic syndrome on the risk of cardiovascular disease
and death in middle-aged men. Circulation. 2010;121(2):230-6.
20. Feig DI, Kang DH, Johnson RJ. Uric acid and cardiovascular risk. N Engl
J Med. 2008;359(17):1811-21.
21. Bard RL, Kalsi H, Rubenfire M, et al. Effect of carotid atherosclerosis
screening on risk stratification during primary cardiovascular disease
prevention. Am J Cardiol. 2004;93(8):1030-2.
22. Bernard S, Sérusclat A, Targe F, et al Incremental predictive value
of carotid ultrasonography in the assessment of coronary risk in a
cohort of asymptomatic type 2 diabetic subjects. Diabetes Care.
2005;28(5):1158-62.
23.Mackinnon AD, Jerrard-Dunne P, Sitzer M, et al. Rates and
determinants of site-specific progression of carotid artery intimamedia thickness: the carotid atherosclerosis progression study.
Stroke. 2004;35(9):2150-4.
24. Lorenz MW, Markus HS, Bots ML, Rosvall M, Sitzer M. Prediction of
clinical cardiovascular events with carotid intima-media thickness: a
systematic review and meta-analysis. Circulation. 2007;115(4):459-67.
25. Tosetto A, Prati P, Baracchini C, Manara R, Rodeghiero F. Age-adjusted
reference limits for carotid intima-media thickness as better indicator
of vascular risk: population-based estimates from the VITA project. J
Thromb Haemost. 2005;3(6):1224-30.
26. Riccioni G. Statins and carotid intima-media thickness reduction: an
up-to-date review. Curr Med Chem. 2009;16(14):1799-805.
27.Loizou CP, Pantziaris M, Pattichis MS, Kyriacou E, Pattichis CS.
Ultrasound image texture analysis of the intima and media layers of
the common carotid artery and its correlation with age and gender.
Comput Med Imaging Graph. 2009;33(4):317-24.
Acknowledgements: Coordination Office for the Improvement of
Higher Education Personnel (Coordenação de Aperfeiçoamento de
Pessoal de Nível Superior, Capes)
104
Address for correspondence:
Sao Paulo Med J. 2014; 132(2):97-104
E-mail: [email protected]
ORIGINAL ARTICLE
DOI: 10.1590/1516-3180.2014.1322604
Mortality due to cardiovascular diseases
in the Americas by region, 2000-2009
Mortalidade por doenças cardiovasculares nas Américas segundo região, 2000-2009
Vilma Pinheiro GawryszewskiI, Maria de Fatima Marinho de SouzaII
Communicable Diseases and Health Analysis, Pan-American Health Organization, World Health Organization, Washington, United States
MD, MSc, PhD. Advisor, Health Information and
Analysis, Health Information and Analysis Unit,
Pan-American Health Organization, Washington
DC, United States.
I
II
MD, MSc, PhD. Advisor, Regional Health
Observatory, Pan-American Health Organization.
KEY WORDS:
Mortality.
Cardiovascular disease.
Trends [subheading].
Inequality.
Americas.
PALAVRAS-CHAVE:
Mortalidade.
Doenças cardiovasculares.
/tendências.
Desigualdades em saúde.
Américas.
ABSTRACT
CONTEXT AND OBJECTIVE: Cardiovascular diseases are the leading cause of death worldwide. The
aim here was to evaluate trends in mortality due to cardiovascular diseases in three different regions
of the Americas.
DESIGN AND SETTING: This was a time series study in which mortality data from three different regions
in the Americas from 2000 to the latest year available were analyzed.
METHODS: The source of data was the Mortality Information System of the Pan-American Health Organization (PAHO). Data from 27 countries were included. Joinpoint regression analysis was used to
analyze trends.
RESULTS: During the study period, the age-adjusted mortality rates for men were higher than those of
females in all regions. North America (NA) showed lower rates than Latin America countries (LAC) and the
Non-Latin Caribbean (NLC). Premature deaths (30-69 years old) accounted for 22.8% of all deaths in NA,
38.0% in LAC and 41.8% in NLC. The trend analysis also showed a significant decline in the three regions.
NA accumulated the largest decline. The average annual percentage change (AAPC) and 95% confidence
interval was -3.9% [-4.2; -3.7] in NA; -1.8% [-2.2; -1.5] in LAC; and -1.8% [-2.7; -0.9] in NLC.
CONCLUSION: Different mortality rates and reductions were observed among the three regions.
RESUMO
CONTEXTO E OBJETIVO: As doenças cardiovasculares são as causas principais de morte em todo o mundo. O objetivo do estudo foi avaliar as tendências na mortalidade decorrente das doenças cardiovasculares
em três diferentes regiões das Américas.
TIPO DE ESTUDO E LOCAL: Este é um estudo de série temporal que analisa dados de mortalidade em três
diferentes regiões das Américas, de 2000 até o último ano disponível.
MÉTODOS: A fonte de dados foi a Sistema de Informação de Mortalidade da Organização Pan-Americana da Saúde (OPAS). Dados de 27 países foram incluídos. Utilizou-se joinpoint regression para analisar
as tendências.
RESULTADOS: Durante o período de estudo, as taxas ajustadas por idade padronizadas de mortalidade
dos homens foram mais altas que a das mulheres em todas as regiões. As taxas da América do Norte
(AN) foram inferiores que as dos países da América Latina (AL) e do Caribe Não Latino (CNL). As mortes
prematuras (30-69 anos) foram 22,8% do total de mortes na AN, 38,0% na AL e 41,8% no CNL. A análise
das tendências mostrou diminuição significativa nas três regiões. A AN acumulou a maior diminuição.
A porcentagem média de mudança anual (AAPC) e respectivos intervalos de confiança de 95% foram
-3,9% [-4,2; -3,7] na AN; na AL foi -1.8% [-2.2; -1.5]; e -1,8% [-2,7; -0,9] no CNL.
CONCLUSÕES: Foram observadas diferentes taxas de mortalidade e diferentes reduções nas três regiões.
Sao Paulo Med J. 2014; 132(2):105-10
105
ORIGINAL ARTICLE | Gawryszewski VP, Souza MFM
INTRODUCTION
Cardiovascular diseases (CVDs) are the leading cause of death
worldwide, accounting for 30% of the total number of annual
deaths globally.1 They comprise major disorders of the heart and
blood vessels, which include heart coronary disease and stroke.
Nowadays, this group of diseases is not considered to be only a
problem of developed countries, since estimates for 2010 showed
that low and middle-income countries were more affected, since
they accounted for 80% of these deaths.1
Globally, ageing and unhealthy behavioral changes like
tobacco smoking, physical inactivity and unhealthy diets,
have become important contributors to the increased prevalence of intermediate risk factors such as obesity, dyslipidemia, raised blood pressure and raised blood sugar. Age is an
unavoidable risk factor, but avoidance of premature mortality
should be a public health concern; the percentage of premature
deaths ranges from 4% in high income countries to 42% in lowincome countries.2
This group of diseases is important to public health not only
because of their magnitude but also because of the possibility of
intervention. Prevention strategies at population level to reduce
circulatory system mortality rates can be classified as high-risk
and community-based.3 The first of these relates to identifying
individuals who are at high risk, through screening, and referring them for treatment. The second of these comprises implementation of programs at population level aimed at modifying
one or more scientifically well established risk factors such as use
of tobacco, physical inactivity, etc. As a result of implementation of
such programs and policies, declines in mortality due to CVDs
have been documented, mainly in developed countries but also
in some Latin America countries.4-8
Socioeconomic indicators have been associated with differences in mortality, incidence and prevalence of cardiovascular
risk factors,1,7-8 and considerable disparities among countries in
the Americas regarding CVD death rates have also been found.8
OBJECTIVE
The objective of this study was to evaluate trends in mortality due
to CVD in three different regions of the Americas.
METHODS
Case definition and source of data
Data were obtained from the Mortality Information System
(updated in August 2012), which comprises mortality databases provided to the Pan-American Health Organization
(PAHO) by the countries in the Americas. To ensure comparability among countries with different qualities of information,
we used the data corrected for under-reporting and ill-defined
deaths, in accordance with published methodology.9 This study
106
Sao Paulo Med J. 2014; 132(2):105-10
included all the deaths for which the underlying cause of mortality was coded in Chapter IX, Diseases of the Circulatory
System (I00-I99), of the International Statistical Classification of
Diseases and Related Health Problems, Tenth Revision (ICD-10).
Premature mortality was defined as deaths that occurred
among people aged 30-69 years old. For this definition, the life
expectancy in the countries of the region of the Americas10 and
a publication from the World Health Organization (WHO)11
were taken into account.
The criterion for including countries in this study was
that they needed to have a time series available for the study
period. The countries were distributed geographically into
three regions: 1) North America (2 countries): Canada and
the United States; 2) Latin America (13 countries): Argentina,
Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador,
Mexico, Nicaragua, Panama, Paraguay, Peru and Venezuela;
3) Non-Latin Caribbean countries (12 countries): Antigua &
Barbuda, Aruba, Belize, Bermuda, Guyana, Montserrat, Saint
Kitts & Nevis, Saint Vincent and the Grenadines, Suriname,
Trinidad & Tobago, Turks & Caicos Islands and Virgin Islands
(US). In addition to geographical location, these sub-regions
have some economic, social and cultural similarity. The period
of study included data from 2000 to the latest available year
(LAY). For North America and Latin America, the LAY was
2009, and for the Non-Latin Caribbean it was 2008.
Age-adjusted rates were calculated using estimates from the
United Nations Population Division12 and the WHO Standard
Population13 (direct standardization). This study analyzed anonymous secondary data on mortality and therefore no ethics
approval was required.
Data analysis
This was a time-series study. We began with a descriptive analysis on 16,940,728 deaths due to CVD registered in the system,
focusing on sex, age, sub-region of residence and year of occurrence. To explain the trend patterns, a subsequent analysis was
performed using the Joinpoint Regression Program, version 4.0.1
(January 2013), from the National Cancer Institute (http://surveillance.cancer.gov/joinpoint/). The dependent variable was the
age-adjusted rates and the independent variable was the year.
This software identified points at which the slope of the linear
trend changed significantly,5 by using the age-adjusted rates from
2000 to LAY to fit a regression line to the natural logarithm of
the rates. It calculated the average annual percentage change
(AAPC), which is a summary measurement of the trend over a
fixed time period, taking into account the trend transitions.14 The
AAPC was considered to be significant when the slope was different from “zero” at alpha = 0.05. The significance tests used a
Monte Carlo permutation method.
Mortality due to cardiovascular diseases in the Americas by region, 2000-2009 | ORIGINAL ARTICLE
Trend analysis
The results from the trend analysis (Table 1 and Figures 1, 2
and 3) showed that CVD mortality has declined in the three
regions of the Americas. In North America, the overall ageadjusted rates per 100,000 dropped from 192.3 in 2000 to 135.5
in 2009, a decrease of 29.5%. The rates were higher among
males but the decrease was greater among females. In Latin
America, mortality has also been declining: the overall ageadjusted rates per 100,000 dropped from 229.9 in 2000 to 191.4
in 2009, a decrease of 14.1%. Compared with North America,
the rates were higher and the decrease was smaller. In the NonLatin Caribbean Region the adjusted rates per 100,000 dropped
from 296.4 in 2000 to 264.1 in 2008, a decrease of 10.9%. The
percentage decline was lower than in North America and Latin
America. Although this region showed the smallest number of
Table 1. Mortality due to cardiovascular diseases (ICD-10 I00-I99)
according to sex and region. North America, Latin America and
Non-Latin Caribbean, 2000-LAY
North America
Latin America
Non-Latin Caribbean
Deaths/year (average)
Female
491,475
366,302
3,535
Male
445,775
383,784
3,952
Total
937,249
750,086
7,486
31.3
36.0
% proportion of deaths 30-69
Female
14.6
Male
31.8
44.3
46.9
Total
22.8
38.0
41.8
Age-adjusted rates/100,000
2000
Female
159.9
193.4
243.3
Male
229.9
257.6
366.7
Total
192.3
222.9
296.4
109.3
163.1
216.4
LAY
Female
Male
165.8
224.9
325.1
Total
135.5
191.4
264.1
Female
-31.6%
-15.6%
-11.0%
Male
-27.8%
-12.7%
-11.3%
% of change 2000-LAY
Total
-29.5%
-14.1%
-10.9%
LAY = latest available year; North America (2009), Latin America (2009) and NonLatin Caribbean (2008).
250.0
200.0
Age-adjusted rates/100,000
RESULTS
During the study period, in all the countries included in the
study, females accounted for 51.0% and males for 49.0% of
all CVD deaths. Table 1 presents some descriptive characteristics of this mortality in each region. The highest annual
average of deaths was seen in North America, where almost
one million deaths occurred annually (55.3% overall), followed by Latin America (around 800,000 deaths; 44.3% overall) and then Non-Latin Caribbean (around 7,500 deaths;
0.4% overall). The proportion of premature mortality varied
widely regarding sex and region. Males showed a higher proportion of premature mortality in all sub-regions, compared
with females. North America presented the lowest proportion
of premature deaths (22.8% overall; 14.6% among females
and 31.8% among males), in comparison with the other subregions. In Latin America, these percentages were: 38.0%
overall; 31.3% among females and 44.3% among males. In the
Non-Latin Caribbean countries, almost half of one percent of
the deaths due to circulatory system diseases were considered
to be premature (41.8% overall; 36.0% among females and
46.9% among males).
Although the average number of deaths was greater among
females, the age-adjusted rates were not. In all regions, the rate
among males was higher than the rate among female. Although
the average number of deaths was greater in North America, the
age-adjusted rates were not. Both in 2000 and LAY, the North
American selected countries presented lower age-adjusted rates
than the Latin American selected countries and Non-Latin
Caribbean selected countries. Using the mortality rates for the
latest year available, the risk of dying due to a CVD presented
by a person who lived in the Non-Latin Caribbean region was
1.9 times the risk presented by a person who lived in the North
American region.
150.0
100.0
50.0
0.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Male
Female
Total
Figure 1. Trends in mortality due to cardiovascular diseases (ICD-10
I00-I99) (age-adjusted rates/100,000), in selected countries in North
America (Canada and the USA), 2000-2009.
Sao Paulo Med J. 2014; 132(2):105-10
107
ORIGINAL ARTICLE | Gawryszewski VP, Souza MFM
deaths, its age-adjusted rates were the highest. The trend curves
for males and females followed the same pattern as shown by
the total population curve (Figures 1, 2 and 3).
300.0
Age-adjusted rates/100,000
250.0
200.0
150.0
100.0
50.0
0.0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Female
Male
Total
Figure 2. Trends in mortality due to cardiovascular diseases (ICD-10
I00-I99) (age-adjusted rates/100,000), in selected countries in Latin
America, 2000-2009.
The results from the joinpoint analysis (Table 2) showed that
CVD mortality in North America declined significantly from
2000 to 2009 and the corresponding AAPC was -3.9% (-4.3%
among females and -3.7% among males). In Latin America, the
AAPC was -1.8% over the entire period (-2.0% among females
and -1.6% among males). In the Non-Latin Caribbean, the
AAPC from 2004 to 2008 (the last 5 observations) was -1.8%
(-1.8 % among females and -1.8% among males).
DISCUSSION
Our findings showed that mortality due to CVD has been
decreasing in the three regions of the Americas, which is consistent with previous studies.4-8 However, there were marked
differences among these regions. The North American countries, which have more favorable socioeconomic indicators,9
showed lower mortality rates, lower proportions of premature
mortality and higher declines, in comparison with the selected
countries in Latin American and the Non-Latin Caribbean.
The question that arises is what the reasons might be for
the differences in the decline of CVD mortality. In Canada and
the USA, mortality rates peaked in the mid-1960s and then
experienced a long term decline.4,5 In the USA, approximately
half of the decline in coronary disease mortality from 1980 to
2000 was attributable to clinical treatment (revascularization,
initial treatment for acute myocardial infarction or angina and
Table 2. Joinpoint analysis for cardiovascular diseases (ICD10 I00-I99), according to sex and region. North America, Latin
America and Non-Latin Caribbean, 2000-LAY
400.0
Age-adjusted rates/100,000
350.0
AAPC [Confidence
Interval]
Lower
Endpoint
Upper
Endpoint
North America
2000
2009
-3.9* [-4.2; -3.7]
200.0
Latin America
2000
2009
-1.8* [-2.2; -1.5]
150.0
Non-Latin
Caribbean
2004
2008
-1.8* [-2.7; -0.9]
North America
2000
2009
-4.3* [-4.5; -4.0]
Latin America
2000
2009
-2.0* [-2.4; -1.6]
Non-Latin
Caribbean
2004
2008
-1.8* [-3.0; -0.7]
North America
2000
2009
-3.7* [-3.9; -3.4]
Latin America
2000
2009
-1.6* [-2.0; -1.3]
300.0
Total
250.0
Female
100.0
50.0
0.0
2000
2001
Male
2002
2003
Female
2004
2005
2006
2007
2008
Total
Figure 3. Trends in mortality due to cardiovascular diseases (ICD-10
I00-I99) (age-adjusted rates/100,000), in selected countries in NonLatin Caribbean, 2000-2008.
108
Years
Sao Paulo Med J. 2014; 132(2):105-10
Male
Non-Latin
2004
2008
-1.8* [-2.6; -1.0]
Caribbean
LAY = last available year; *The AAPC (average annual percentage
change) is significantly different from “zero” at alpha = 0.05.
Mortality due to cardiovascular diseases in the Americas by region, 2000-2009 | ORIGINAL ARTICLE
other treatments) and approximately half to changes in risk
factors (reductions in total cholesterol, blood pressure, smoking and physical inactivity).15 For some Latin America countries for which published trend series studies are available, the
decline took place recently, or there was no decline or even
an increase.4-5,8 These countries have probably implemented
community-based prevention programs more recently.
In addition, populations that live in countries with more
disadvantaged incomes might face greater difficulties in
obtaining access to treatment, compared with those who live
in North American countries. Some countries in the Americas,
such as Argentina, Brazil, Cuba and Canada, have public universal health coverage. Despite this, in Brazil, a study found
that premature mortality due to circulatory system diseases
was 2.6 times higher in poor areas in Porto Alegre than in
rich areas.16 In São Paulo, the decline in the risk of death due
to heart disease was faster for people living in the wealthiest areas and slower for people living in lower-resourced
neighborhoods.17 In the United States, which has a different
health system and the percentage health insurance coverage
is around 83%, a study showed racial disparities in use of coronary artery bypass grafting and percutaneous transluminal
coronary angioplasty among elderly women and men, which
was probably due to access to care and financial barriers.18
Regarding risk factors, a lot of work has to be done in the
region. In the USA, around 90% of the people consume sodium
at levels above the recommended guidelines.19 The prevalence
of an abnormally large abdominal waist circumference among
people aged 20 years and over was around 70% in San Salvador,
El Salvador (52% among males and 79% among females), and
64% in Belize (20% among the white population and 63% among
the black and mixed population). 20 In Brazil, the prevalence
of medical diagnoses of hypertension among the adult population reached 23% in 2010 (26% among females and 21% among
males).21 On the positive side, prevention activities have been
implemented in many countries. Argentina, Barbados, Bolivia,
Brazil, Chile, Cuba, Dominica, Mexico, Panama and Venezuela
have implemented CVD and hypertension control programs.22
Tobacco policies have been implemented in 13 countries in
Latin America. Mexico, Brazil, Chile, Costa Rica, Dominica,
Paraguay and Peru have implemented initiatives to regulate the
marketing of foods to children.23
disability and deaths was around US$ 161 billion, which raised
the total cost of these diseases to US$ 475 billion.23
The main limitations of the present study are the following: 1) although the rates were corrected for under-registration of deaths and ill-defined causes, it is possible that these
problems might affect the rates in some countries, especially
those with lower resources; and 2) some lower-resourced
countries have made efforts to improve the quality of data,
including decreasing the numbers of ill-defined deaths, which
might lead to an increase in circulatory system disease rates.
Since disease is a great contributor to poverty, the Americas
should strive to bridge the gap in treatment and preventive
control of CVD among its regions. Gold-standard prevention
requires complementary clinical and community approaches
and monitoring information regarding mortality, morbidity and
prevalence of risk factors. Eliminating health disparities
should be a goal in relation to CVD mortality reduction in
the region.
It should be noted that the CVD burden goes beyond mortality. It has social and economic impacts such as increased
costs of healthcare, disabilities and losses in productive years
of life due to premature mortality. For example, in the United
States, the estimated cost of medical care for CVDs in 2009
was around US$ 314 billion, and the indirect costs due to
5. Rodríguez T, Malvezzi M, Chatenoud L, et al. Trends in mortality from
CONCLUSION
Mortality rates due to CVD have been decreasing since 2000 in
the North American, Latin American and Non-Latin Caribbean
regions. Disparities in risk, premature mortality and trends
were seen across these three different regions. REFERENCES 1. Ordúñez García P, Campillo Artero C. Regional Consultation.
Priorities for cardiovascular health in the Americas. Key messages for
policymakers. Washington: Pan-American Health Organization; 2011.
Available from: http://www1.paho.org/priorities/pdf-en/book.pdf.
Accessed in 2013 (May 23).
2. Mendis S, Puska P, Norrving B. Global Atlas on cardiovascular
disease
prevention
and
control.
Geneva:
World
Health
Organization; 2011. Available from: http://whqlibdoc.who.int/
publications/2011/9789241564373_eng.pdf.
Accessed
in
2013
(May 23).
3. Newschaffer CJ, Longjian L, Sim A. Cardiovascular diseases. In:
Remington PL, Brownson RC, Wegner MV, eds. Chronic disease
epidemiology and control. Washington: American Public Health
Association; 2010. p. 383-428.
4. Mirzaei M, Truswell R, Taylor, Leeder SR. Coronary heart disease
epidemics: not all the same. Heart. 2009;95(9):740-6.
coronary heart and cerebrovascular diseases in the Americas: 19702000. Heart. 2006;92(4):453-60.
6. Schmidt MI, Duncan BB, Azevedo e Silva G, et al. Chronic noncommunicable diseases in Brazil: burden and current challenges.
Lancet. 2011;377(9781):1949-61.
Sao Paulo Med J. 2014; 132(2):105-10
109
ORIGINAL ARTICLE | Gawryszewski VP, Souza MFM
7. Frieden TR; Centers for Disease Control and Prevention (CDC).
Available from: http://www.paho.org/hon/?cx=01428377084524
Forward: CDC Health Disparities and Inequalities Report - United
0200164%3A4igtso_4aue&q=Encuesta+de+diabetes%2C+hipert
States, 2011. MMWR Surveill Summ. 2011;60 Suppl:1-2.
ensión+y+factores+de+riesgo+de+enfermedades+crónicas&sea
8. de Fatima Marinho de Souza M, Gawryszewski VP, Orduñez P,
rchword=Encuesta+de+diabetes%2C+hipertensión+y+factores+
Sanhueza A, Espinal MA. Cardiovascular disease mortality in the
de+riesgo+de+enfermedades+crónicas&sa=Buscar...&cof=FORID
Americas: current trends and disparities. Heart. 2012;98(16):1207-12.
%3A0&searchphrase=all&ie=iso-8859-1&scope=1&option=com_
9. Ahmad OB, Boschi-Pinto C, Lopez AD, et al. Age standardization of
search&Itemid=1&siteurl=www.paho.org%2Fhon%2F&ref=www.
rates: a new WHO standard. GPE Discussion Paper Series: No. 31.
paho.org%2F&ss=32j1024j2#gsc.tab=0&gsc.q=encuesta%20
World Health Organization 2001. Available from: http://www.who.
int/healthinfo/paper31.pdf. Accessed in 2013 (May 23).
diabete&gsc.page=1. Accessed in 2013 (May 23).
21.Brasil. Ministério da Saúde. Secretaria de Vigilância em Saúde.
10.Pan American Health Organization. Health situation in the
Secretaria de Gestão Estratégica e Participativa. Vigitel Brasil 2010:
Americas: basic indicators 2012. Washington: Pan American
vigilância de fatores de risco e proteção para doenças crônicas por
Health Organization; 2012. Available from: http://ais.paho.org/chi/
inquérito telefônico. Brasília: Ministério da Saúde; 2011. Available
brochures/2012/BI_2012_ENG.pdf. Accessed in 2013 (May 23).
from:
11.Health Organization. WHO Discussion Paper: A comprehensive
http://portal.saude.gov.br/portal/arquivos/pdf/vigitel_2010_
preliminar_web.pdf. Accessed in 2013 (May 23)
global monitoring framework and voluntary global targets for the
22. Pan-American Health Organization. Non communicable diseases
prevention and control of NCDs. Available from: http://www.who.
project & Health information and analysis project & Sustainable
int/nmh/events/2011/consultation_dec_2011/WHO_Discussion_
development and environment area. Non communicable diseases in
Paper_FINAL.pdf. Accessed in 2013 (May 23).
the Americas: Basic indicators on non communicable diseases, 2011.
12. Organización Panamericana de la Salud. Organización Mundial de la
Salud. Estadísticas de salud de las Américas. Edición 2006. Available
Washington; 2011.
23. Labarthe DR. Cardiovascular diseases: a global public health challenge.
from: http:www.paho.org/ESA2006. Accessed in 2013 (May 23).
In: Labarthe DR. Epidemiology and prevention of cardiovascular
13.United Nations, Department of Economic and Social Affairs,
diseases: a global challenge. 2nd edition. Massachusetts: Jones and
Population Division. World population prospects: The 2010 Revision,
Bartlett Publishers; 2011. p. 3-18. Available from: http://samples.
Highlights and Advance Tables. New York: United Nations; 2011.
jbpub.com/9780763746896/46894_fm_00i_xviii.pdf. Accessed in
Available
2013 (May 23).
from:
http://esa.un.org/wpp/documentation/pdf/
WPP2010_Highlights.pdf. Accessed in 2013 (May 23).
14. Clegg LX, Hankey BF, Tiwari R, Feuer EJ, Edwards BK. Estimating
Sources of funding: None
average annual per cent change in trend analysis. Stat Med.
Conflicts of interest: None
2009;28(29):3670-82.
15.Ford ES, Ajani UA, Croft JB, et al. Explaining the decrease in
Date of first submission: September 25, 2012
U.S. deaths from coronary disease, 1980-2000. N Engl J Med.
Last received: June 18, 2013
2007;356(23):2388-98.
Accepted: July 5, 2013
16. Bassanesi SL, Azambuja MI, Achutti A. Premature mortality due to
cardiovascular disease and social inequalities in Porto Alegre: from
Address for correspondence:
evidence to action. Arq Bras Cardiol. 2008;90(6):370-9.
Vilma Pinheiro Gawryszewski
17. Lotufo PA, Fernandes TG, Bando DH, Alencar AP, Benseñor IM. Income
and heart disease mortality trends in Sao Paulo, Brazil, 1996 to 2010.
Pan-American Health Organization
Int J Cardiol. 2012;6. [Epub ahead of print].
525 23rd St.
18. Gillum RF. Coronary revascularization in older women and men
Washington DC 20037
in the United States: trends in ethnic differences. Am Heart J.
United States
2004;147(3):418-24.
Tel. 202-492-7492
19. Centers for Disease Control and Prevention (CDC). Million hearts:
strategies to reduce the prevalence of leading cardiovascular disease
risk factors--United States, 2011. MMWR Morb Mortal Wkly Rep.
2011;60(36):1248-51.
20. Organización Panamericana de la Salud. Iniciativa Centro Americana
de Diabetes (CAMDI). Encuesta de diabetes, hipertensión y factores
de riesgo de enfermedades crónicas. Tegucigalpa, Honduras; 2009.
110
Health Analysis and Information Project
Sao Paulo Med J. 2014; 132(2):105-10
Fax. 202-974-3674
E-mail: [email protected], [email protected]
ORIGINAL ARTICLE
DOI: 10.1590/1516-3180.2014.1322666
Association between sex hormone-binding
globulin (SHBG) and metabolic syndrome among men
Associação entre globulina de ligação a hormônio
sexual (SHBG) e síndrome metabólica em homens
Emmanuela Quental Callou de SáI, Francisco Carleial Feijó de SáII, Kelly Cristina OliveiraIII,
Fausto FeresIV, Ieda Therezinha Nascimento VerreschiV
Escola Paulista de Medicina — Universidade Federal de São Paulo (EPM-Unifesp), São Paulo, and Instituto Dante Pazzanese de
Cardiologia, São Paulo, São Paulo, Brazil
MD, PhD. Endocrinologist, Universidade Federal
do Ceará (UFC), Barbalha, Ceará, Brazil.
I
II
MD. Cardiologist and Hemodynamicist,
Universidade Federal do Ceará (UFC), Barbalha,
Ceará, Brazil.
MSc. Pharmacist, Universidade Federal de São
Paulo (Unifesp), São Paulo, Brazil.
III
IV
MD, PhD. Cardiologist and Hemodynamicist,
Instituto Dante Pazzanese de Cardiologia, São
Paulo, Brazil.
MD, PhD. Endocrinologist, Universidade Federal
de São Paulo, (Unifesp), São Paulo, Brazil.
V
KEY WORDS:
Sex hormone-binding globulin.
Metabolic syndrome X.
Men.
Coronary artery disease.
Coronary angiography.
PALAVRAS-CHAVE:
Globulina de ligação a hormônio sexual.
Síndrome X metabólica.
Homens.
Doença da artéria coronariana.
Angiografia coronária.
ABSTRACT
CONTEXT AND OBJECTIVE: Metabolic syndrome consists of a set of factors that imply increased risk of
cardiovascular diseases. The objective here was to evaluate the association between sex hormone-binding
globulin (SHBG), sex hormones and metabolic syndrome among men.
DESIGN AND SETTING: Retrospective analysis on data from the study “Endogenous oestradiol but not
testosterone is related to coronary artery disease in men”, conducted in a hospital in São Paulo.
METHODS: Men (aged 40-70) who underwent coronary angiography were selected. The age, weight,
height, waist circumference, body mass index and prevalence of dyslipidemia, hypertension and diabetes of each patient were registered. Metabolic syndrome was defined in accordance with the criteria of
the Third Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation
and Treatment of High Blood Cholesterol in Adults (NCEP-ATPIII). Serum samples were collected to assess
the levels of glucose, total cholesterol, HDL-cholesterol (high density lipoprotein), triglycerides, albumin,
SHBG, estradiol and total testosterone (TT). The levels of LDL-cholesterol (low density lipoprotein) were
calculated using Friedewald’s formula and free testosterone (FT) and bioavailable testosterone (BT) using
Vermeulen’s formula.
RESULTS: 141 patients were enrolled in the study. The prevalence of metabolic syndrome was significantly
higher in the first SHBG tercile than in the second and third terciles. A statistically significant positive association between the SHBG and TT values was observed, but no such association was seen between
SHBG, BT and FT.
CONCLUSION: Low serum levels of SHBG are associated with higher prevalence of metabolic syndrome
among male patients, but further studies are required to confirm this association. RESUMO
CONTEXTO E OBJETIVO: A síndrome metabólica (SM) consiste em um conjunto de fatores que implicam
risco elevado para doenças cardiovasculares. O objetivo foi avaliar a associação entre a globulina ligadora
de esteroides sexuais (SHBG), hormônios sexuais e a SM em homens.
TIPO DE ESTUDO E LOCAL: Análise retrospectiva de dados do estudo “Estradiol mas não testosterona
se correlaciona com doença arterial coronariana em homens”, conduzido em um hospital em São Paulo.
MÉTODOS: Foram selecionados pacientes do sexo masculino com idade entre 40 e 70 anos, submetidos
a angiografia coronária. A idade, a prevalência de dislipidemia, hipertensão e diabetes, o peso, a altura,
cintura e o índice de massa corpórea de cada paciente foram coletados. A definição de SM seguiu os
critérios do NCEP-ATPIII. Amostras séricas foram coletadas para análises da glicose, colesterol total, colesterol-HDL (high density lipoprotein), triglicerídeos, albumina, SHBG, estradiol e testosterona total (TT). O
colesterol-LDL (low density lipoprotein) foi calculado pela fórmula de Friedewald e as testosteronas livre (TL)
e biodisponível (TB) pela fórmula de Vermeulen.
RESULTADOS: Entraram no estudo 141 pacientes. A prevalência de SM foi significativamente maior no primeiro tercil de SHBG em comparação ao segundo e terceiro tercis. Foi verificada uma associação positiva
e significativa ente os valores de SHBG e TT, porém essa associação não foi verificada entre SHBG e TB e TL.
CONCLUSÃO: Baixos níveis séricos de SHBG estiveram associados com alta prevalência da SM em pacientes do sexo masculino. Faz-se necessário que estudos avaliem essa associação.
Sao Paulo Med J. 2014; 132(2):111-5
111
ORIGINAL ARTICLE | Callou de Sá EQ, Sá FCF, Oliveira KC, Feres F, Verreschi ITN
INTRODUCTION
Metabolic syndrome consists of a set of factors that confer
increased risk of cardiovascular diseases, including obesity (especially abdominal obesity), insulin resistance (regardless of the
presence of diabetes mellitus), dyslipidemia (increased triglyceride levels and reduced HDL cholesterol levels) and systemic arterial hypertension (SAH). The primary abnormality relating to
metabolic syndrome appears to be insulin resistance in peripheral tissues.1
Sex hormone-binding protein (SHBG), which is produced
by the liver, binds to testosterone with high affinity and to estradiol (E2) with lower affinity. Insulin is an important regulator of
SHBG production in the liver. In vitro studies have shown that
physiological concentrations of insulin inhibit SHBG production in cultured hepatoma cells.2 Pasquali et al.3 showed that
inhibition of insulin secretion by means of diazoxide induces
an increase in SHBG levels, both in obese men and in men with
normal body weights. In addition, men who present low SHBG
concentrations are at increased risk of developing metabolic syndrome.4 The prevalence of concomitant dyslipidemia, hypertension and diabetes suggests that insulin resistance may be a determinant of SHBG levels.
OBJECTIVE
This study assessed the association between SHBG, sex hormones and prevalence of metabolic syndrome.
METHODS
Study design
This study was conducted by performing a retrospective analysis
on the data from a previous study, “Endogenous oestradiol, but
not testosterone, is related to coronary artery disease in men”.5
Study population
We selected male patients aged 40-70 years who were admitted to
hospital in order to undergo coronary angiography for investigation and/or staging of ischemic heart disease, at Hospital Dante
Pazzanese de Cardiologia. Patients who were smokers, were
using anti-androgenic drugs (such as ketoconazole, cimetidine,
spironolactone, androcur or finasteride) or had a previous history of myocardial infarction, stroke and/or major surgery within
the past six months were excluded from the study. In addition,
individuals with a body mass index (BMI, defined as weight in kg
divided by the square of height in meters) ≥ 40 kg/m2 or serum
creatinine levels > 2.0 mg/dl, and patients who showed evidence
of major liver disease during clinical examinations were also
excluded from the study.
After verifying the study inclusion criteria, the patients were
divided into three groups based on SHBG terciles.
112
Sao Paulo Med J. 2014; 132(2):111-5
Informed consent was obtained from all subjects, and the
project had previously been approved by the Ethics Committees
of the two participating institutions.
Assessments
Data on age, prevalence of components of metabolic syndrome
(obesity, dyslipidemia, SAH and diabetes), current medications,
weight, height and waist circumference measurements and BMI
were gathered in relation to each patient through standard questionnaires immediately after the individual had been given explanations about the study and had consented to participation in it.
Metabolic syndrome was defined in accordance with the criteria
recommended by the Third Report of the National Cholesterol
Education Program Expert Panel on Detection, Evaluation and
Treatment of High Blood Cholesterol in Adults (NCEP-ATPIII).
These criteria state that three or more of the following criteria
should be present: abdominal obesity (waist circumference > 102
cm in men and > 88 cm in women); triglycerides ≥ 150 mg/dl;
HDL cholesterol < 40 mg/dl in men and < 50 mg/dl in women;
blood pressure ≥ 130/85 mmHg; and fasting glucose ≥ 110 mg/dl.6
Laboratory analysis
Serum samples were collected between 8:00 AM and 10:00
AM, after overnight fasting, before the coronary angiography.
Glucose, total cholesterol, HDL cholesterol, triglyceride and
albumin levels were measured in the laboratory of Hospital
Dante Pazzanese de Cardiologia. LDL cholesterol was calculated
using Friedewald’s formula.7
Blood samples for hormone determinations were taken to the
Steroid Laboratory of Escola Paulista de Medicina, Universidade
Federal São Paulo (EPM-Unifesp), immediately after collection. The samples were centrifuged and frozen at -21 °C for no
longer than six months. All determinations were performed in
duplicate. SHBG was measured by means of the immunofluorometric assay (IFMA; Delfia Perkin Elmer, São Paulo, Brazil),
with a detection limit of 0.5 nmol/l. The intra-assay coefficient
of variation (CV) was 3.9, 4.9 and 3.3% and the inter-assay CV
was 2.3, 3.0 and 2.4%, for 25.5, 63.9 and 138.0 nmol/l, respectively. Testosterone was measured by means of radioimmunoassay with local historical controls, with a detection limit of
0.35 nmol/l, intra-assay CV of 7.5 and 13.2% for 16.7 and 3.86
nmol/l, respectively, and inter-assay CV of 15.5 and 17.6% for
22.87 and 4.97 nmol/l, respectively. The anti-serum used was
the anti-testosterone 3-(O-carboxymethyl)-oxime-BSA, and
the radioactive standard was (1,2,6,73H)-testosterone (250 µCi)
(Amersham Biosciences, Uppsala, Sweden). E2 was measured by
means of IFMA (Delfia Perkin Elmer, São Paulo, Brazil), with a
detection limit of 0.05 nmol/l, intra-assay CV 6.9% and interassay CV 9.7%. Free and bioavailable testosterone were evaluated
Association between sex hormone-binding globulin (SHBG) and metabolic syndrome among men | ORIGINAL ARTICLE
from serum total testosterone (TT), SHBG and serum albumin in
accordance with the formula of Vermeulen et al.8
males with SHBG levels > 28.3 nmol/l presented a 52% lower
risk of having diabetes, compared with men with SHBG levels
≤ 28.3 nmol/l. Colangelo et al.11 observed a significant inverse
Statistical analysis
Analysis of variance (ANOVA) was used to determine the statistical significance of numerical variables, and Fisher’s exact
test was used for categorical variables. Data were analyzed using
GraphPad Prism 5.0 (San Diego, CA, USA). Relative risk (RR)
evaluations were used to assess the prevalence of metabolic syndrome according to SHBG tercile groups. Statistical significance
was set at P < 0.05.
association between SHBG levels and abnormal fasting glucose
levels and type 2 diabetes among 3,156 men of various ethnicities. Muller et al.12 also reported negative correlations between
SHBG levels and risk factors for metabolic syndrome. A crosssectional study by Gannagé-Yared,13 which included 201 young
males aged 18-30 years, identified significant negative correlations between serum SHBG levels and the levels of triglycerides,
HOMA-IR and C-reactive protein (CRP). Finally, in cross-sectional and longitudinal studies on the second and third-generation populations of the Framingham Heart Study, Bhasin et al.14
observed that SHBG is an independent predictor for incidence
of metabolic syndrome. The cross-sectional evaluations in that
study revealed that TT and FT were associated with the prevalence of metabolic syndrome, but a stronger association was
observed for TT than for FT. Nevertheless, neither TT nor FT
was associated with the prevalence of metabolic syndrome in the
longitudinal evaluations. Consequently, these authors concluded
RESULTS
A total of 141 patients were eligible for the study. Increased SHBG
levels were negatively associated with BMI, abdominal circumference and prevalence of diabetes (Table 1). No inverse associations between SHBG levels and fasting serum triglyceride levels
(P = 0.06) or between SHBG levels and SAH prevalence (P = 0.08)
were observed (Table 1). The prevalence of metabolic syndrome
was significantly higher in the first (lowest) SHBG tercile than in
the second (middle) and third (highest) SHBG terciles (Table 2).
A significant positive association between SHBG and TT levels was observed, but no significant associations were observed
between SHBG and bioavailable testosterone (BT) or between
SHBG and free testosterone (FT) (Table 1). Similarly, no statistically significant association was observed between SHBG and E2
levels (Table 1).
DISCUSSION
In this sample of 141 male patients admitted to undergo coronary angiography in order to diagnose coronary artery disease, at
Hospital Dante Pazzanese de Cardiologia, SHBG levels presented
significant inverse associations with BMI, abdominal circumference and prevalence of type 2 diabetes (Table 1). However, no
inverse associations between SHBG levels and fasting serum triglyceride levels (P = 0.06) or between SHBG levels and SAH prevalence (P = 0.08) were observed (Table 1). Furthermore, no significant association between the serum levels of SHBG and HDL
cholesterol was observed (Table 1). A significant negative association between SHBG levels and the prevalence of metabolic syndrome was found for patients in the second and third SHBG terciles, with a RR of metabolic syndrome of 0.60 for the second
tercile (95% CI: 0.06 - 0.47) and 0.29 for the third tercile (95% CI:
0.16 - 0.54), in comparison with the first tercile (Table 2).
There is increasing evidence in the literature to suggest that
low SHBG levels are correlated with components of metabolic
syndrome. In a recently published paper, Ding et al.9 concluded
that low SHBG levels are a strong risk predictor for type 2 diabetes. Similarly, in a meta-analysis, Ding et al.10 found that white
Table 1. Laboratory and clinical characteristics of patients according to
their sex hormone-binding globulin (SHBG) terciles
SHBG tercile
P
1 (lowest)
2
3 (highest)
SHBG mean (nmol/l)
27.20
43.11
72.01
< 0.001
SHBG variation (nmol/l)
12.2-35.5 36.1-49.9 50.5-159
Age (years)*
57.0
56.73
58.88
0.33
28.76
27.46
25.13
< 0.001
BMI (kg/m2)*
Abdominal circumference (cm)*
103.49
97.66
93.53
< 0.001
Diabetes (%)
38.30
21.28
8.51
0.002
Dyslipidemia (%)
70.21
63.83
57.45
0.44
Hypertension (%)
95.74
80.85
87.23
0.08
Blood glucose (mg/dl)*
111.04
107.49
99.07
0.42
Total cholesterol (mg/dl)*
183.72
180.30
168.38
0.22
LDL-cholesterol (mg/dl)*
107.24
105.34
98.70
0.51
HDL-cholesterol (mg/dl)*
40.30
41.98
43.34
0.35
Triglycerides (mg/dl)*
174.38
171.28
131.79
0.06
Total testosterone (nmol/l)*
13.63
16.48
20.84
< 0.001
Bioavailable testosterone (nmol/l)*
7.45
7.23
6.61
0.55
Free testosterone (nmol/l)*
0.32
0.31
0.28
0.50
Estradiol (pmol/l)*
68.72
79.14
74.25
0.29
BMI = body mass index; LDL = low-density lipoprotein; HDL = high density
lipoprotein. *Data are expressed as means.
Table 2. Prevalence of metabolic syndrome according to sex hormonebinding globulin (SHBG) terciles
SHBG terciles
Metabolic syndrome %
Relative risk
Confidence interval (95%)
1 (lowest)
70.21
1
2
44.68
0.60
0.06-0.47
3 (highest)
19.15
0.29
0.16-0.54
Sao Paulo Med J. 2014; 132(2):111-5
113
ORIGINAL ARTICLE | Callou de Sá EQ, Sá FCF, Oliveira KC, Feres F, Verreschi ITN
that their data did not corroborate the hypothesis that low TT
levels were independently associated with the prevalence of
metabolic syndrome.
The primary abnormality observed in metabolic syndrome
appears to be insulin resistance in peripheral tissues. Because
insulin is a potent inhibitor of SHBG production in the liver, it is
possible that decreased levels of SHBG could be an early marker
for metabolic syndrome. Similarly, both Heald et al.,15 in a study
examining European, Pakistani and Afro-Caribbean populations, and Chubb et al.,16 in a population-based study, suggested
that SHBG is a potential marker for metabolic syndrome. In a
recent non-interventional study examining 80 patients with metabolic syndrome, it was reported that an increase of one unit in
insulin levels resulted in a decrease of 0.25 units in SHBG levels.17
Treatment with the PPARγ agonist rosiglitazone has been
shown to increase the blood levels of SHBG, especially in patients
with polycystic ovarian syndrome.18
Insulin resistance can be defined as a subnormal state of
biological responses to circulating insulin. Recent evidence has
shown that inflammatory serum mediators may induce insulin resistance.19 CRP is the best studied and best characterized
inflammatory marker. Kupelian et al.,20 in a recent non-interventional population-based study examining 2,301 men aged 30-79
years, and Gannagé-Yared,13 in a cross-sectional study on young
men, reported an inverse association between the levels of SHBG
and CRP, thus corroborating the inverse association between
SHBG and insulin resistance.
The current study was not intended to evaluate potential
causal relationships between SHBG and metabolic syndrome. In
fact, the existing data in the literature are still insufficient to confirm whether SHBG is an early marker or whether it is a component of metabolic syndrome.
In this study, a significant positive association between
SHBG and TT was observed (P < 0.001), but no significant
associations were observed between SHBG and BT, FT or E2
(Table 1). In plasma, it is known that only approximately 2%
of all testosterone circulates in the free form (i.e. the fraction
known as FT).21 In contrast, 44% is bound to SHBG, and 54%
binds to albumin.21 Both the FT and the albumin-bound fractions are readily available to tissues. The sum of these two fractions is referred to as the BT level. Therefore, we hypothesized that a significant positive association would be observed
between TT and SHBG levels.
Epidemiological studies have shown that decreased TT levels are associated with an increased risk of developing metabolic
syndrome.22,23 These studies have suggested that low testosterone
levels could contribute towards the physiopathology of metabolic syndrome and that androgen replacement therapy should
be used for males with metabolic syndrome and testosterone
114
Sao Paulo Med J. 2014; 132(2):111-5
deficiency.24,25 However, in a Brazilian study conducted by Callou
de Sá et al.,5 no significant differences in the prevalence of components of metabolic syndrome (as defined according to the NCEPATPIII criteria)6 between TT terciles were observed. Similarly, in
the study conducted by Bhasin et al.14 examining the second and
third generations of the Framingham Heart Study, no significant
associations between the prevalence of metabolic syndrome and
TT or FT were observed in the longitudinal evaluations. Thus,
these authors concluded that their data did not corroborate the
hypothesis that low TT levels were independently associated with
the prevalence of metabolic syndrome.
Evaluation of SHBG as a marker for metabolic syndrome among
males is important because this syndrome consists of a number of
factors that confer increased risk of cardiovascular diseases, which
are the main group of diseases causing death in Brazil.
CONCLUSION
We conclude that low serum levels of SHBG were associated
with higher prevalence of metabolic syndrome in our sample of
adult Brazilian males aged 40-70 years. Our data should not be
extrapolated to females or to other ethnic groups or age groups.
Additional prospective studies to assess this association directly
are required, especially with regard to the potential causal relationship between SHBG and metabolic syndrome.
REFERENCES
1. Reusch JE. Current concepts in insulin resistance, type 2 diabetes mellitus,
and the metabolic syndrome. Am J Cardiol. 2002;90(5A):19G-26G.
2. Peiris AN, Stagner JI, Plymate SR, et al. Relationship of insulin
secretory pulses to sex hormone-binding globulin in normal men. J
Clin Endocrinol Metab. 1993;76(2):279-82.
3. Pasquali R, Casimirri F, De Iasio R, et al. Insulin regulates testosterone
and sex hormone-binding globulin concentrations in adult normal
weight and obese men. J Clin Endocrinol Metab. 1995;80(2):654-8.
4. Laaksonen DE, Niskanen L, Punnonen K, et al. Sex hormones,
inflammation and the metabolic syndrome: a population-based
study. Eur J Endocrinol. 2003;149(6):601-8.
5. Callou de Sá EQ, Feijó de Sá FC, e Silva R de S, et al. Endogenous
oestradiol but not testosterone is related to coronary artery disease
in men. Clin Endocrinol (Oxf ). 2011;75(2):177-83.
6. Expert Panel of Detection, Evaluation and Treatment of High Blood
Cholesterol in Adults. Executive Summary of the Third Report of the
National Cholesterol Education Program (NCEP) Expert Panel on
Detection, Evaluation, And Treatment of High Blood Cholesterol in
Adults (Adult Treatment Panel III). JAMA. 2001;285(19):2486-97.
7. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the
concentration of low-density lipoprotein cholesterol in plasma,
without use of the preparative ultracentrifuge. Clin Chem.
1972;18(6):499-502.
Association between sex hormone-binding globulin (SHBG) and metabolic syndrome among men | ORIGINAL ARTICLE
8. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple
22. Zmuda JM, Cauley JA, Kriska A, et al. Longitudinal relation between
methods for the estimation of free testosterone in serum. J Clin
endogenous testosterone and cardiovascular disease risk factors in
Endocrinol Metab. 1999;84(10):3666-72.
middle-aged men. A 13-year follow-up of former Multiple Risk Factor
9. Ding EL, Song Y, Manson JE, et al. Sex hormone-binding globulin
and risk of type 2 diabetes in women and men. N Engl J Med.
2009;361(12):1152-63.
10. Ding EL, Song Y, Malik VS, Liu S. Sex differences of endogenous sex
hormones and risk of type 2 diabetes: a systematic review and metaanalysis. JAMA. 2006;295(11):1288-99.
11. Colangelo LA, Ouyang P, Liu K, et al. Association of endogenous sex
hormones with diabetes and impaired fasting glucose in men: multiethnic study of atherosclerosis. Diabetes Care. 2009;32(6):1049-51.
intervention Trial participants. Am J Epidemiol. 1997;146(8):609-17.
23. Simon D, Charles MA, Nahoul K, et al. Association between plasma
total testosterone and cardiovascular risk factors in healthy adult
men: The Telecom Study. J Clin Endocrinol Metab. 1997;82(2):682-5.
24.Howell SJ, Radford JA, Adams JE, et al. Randomized placebocontrolled trial of testosterone replacement in men with mild Leydig
cell insufficiency following cytotoxic chemotherapy. Clin Endocrinol
(Oxf ). 2001;55(3):315-24.
25.Haddad RM, Kennedy CC, Caples SM, et al. Testosterone and
12. Muller M, Grobbee DE, den Tonkelaar I, Lamberts SW, van der Schouw
cardiovascular risk in men: a systematic review and meta-analysis of
YT. Endogenous sex hormones and metabolic syndrome in aging
randomized placebo-controlled trials. Mayo Clin Proc. 2007;82(1):29-39.
men. J Clin Endocrinol Metab. 2005;90(5):2618-23.
13. Gannagé-Yared MH, Chedid R, Abs L. Relation between androgens
This article was presented in defense of the doctoral thesis of Emmanuela
and cardiovascular risk factors in a young population. Clin Endocrinol
Quental Callou, within the Endocrinology Program of Universidade Federal
(Oxf ). 2011;74(6):720-5.
de São Paulo (Unifesp), in São Paulo on May 26, 2010. It was also presented
14. Bhasin S, Jasjua GK, Pencina M, et al. Sex hormone-binding globulin,
as a poster at the IV Brazilian Congress on Updates in Endocrinology and
but not testosterone, is associated prospectively and independently
Metabolism and IX Paulista Congress of Endocrinology and Metabolism,
with incident metabolic syndrome in men: the framingham heart
on August 24 to 27, 2011
study. Diabetes Care. 2011;34(11):2464-70.
15. Heald AH, Anderson SG, Ivison F, et al. Low sex hormone binding
Sources of funding: None
globulin is a potential marker for the metabolic syndrome in different
Conflict of interest: None
ethnic groups. Exp Clin Endocrinol Diabetes. 2005;113(9):522-8.
16. Chubb SA, Hyde Z, Almeida OP, et al. Lower sex hormone-binding
Date of first submission: February 5, 2013
globulin is more strongly associated with metabolic syndrome than
Last received: July 14, 2013
lower total testosterone in older men: the Health in Men Study. Eur J
Accepted: July 16, 2013
Endocrinol. 2008;158(6):785-92.
17. Caldas ADA, Porto AL, Motta LDC, Casulari LA. Relação entre insulina e
Address for correspondence:
hipogonadismo em homens com síndrome metabólica [Relationship
Emmanuela Quental Callou de Sá
between insulin and hypogonadism in men with metabolic
Rua Divino Salvador, s/no
syndrome]. Arq Bras Endocrinol Metabol. 2009;53(8):1005-11.
Centro — Barbalha (CE) — Brasil
18. Azziz R, Ehrmann D, Legro RS, et al. Troglitazone improves ovulation
CEP 63180-000
and hirsutism in the polycystic ovary syndrome: a multicenter,
Tel. (+55 88) 3312-5001
double blind, placebo-controlled trial. J Clin Endocrinol Metab.
E-mail: [email protected]
2001;86(4):1626-32.
19. de Carvalho MH, Colaço AL, Fortes ZB. Citocinas, disfunção endotelial
e resistência à insulina [Cytokines, endothelial dysfunction, and
insulin resistance]. Arq Bras Endocrinol Metabol. 2006;50(2):304-12.
20. Kupelian V, Chiu GR, Araujo AB, et al. Association of sex hormones
and C-reactive protein levels in men. Clin Endocrinol (Oxf ). 2010;
72(4):527-33.
21. Dunn JF, Nisula BC, Rodbard D. Transport of steroid hormones:
binding of 21 endogenous steroids to both testosterone-binding
globulin and corticosteroid-binding globulin in human plasma. J Clin
Endocrinol Metab. 1981;53(1):58-68.
Sao Paulo Med J. 2014; 132(2):111-5
115
SHORT COMMUNICATION
DOI: 10.1590/1516-3180.2014.1322640
Prevalence of Candida albicans and non-albicans isolates
from vaginal secretions: comparative evaluation of
colonization, vaginal candidiasis and recurrent vaginal
candidiasis in diabetic and non-diabetic women
Prevalência de Candida albicans e não albicans isoladas de secreção vaginal:
avaliação comparativa entre colonização, candidíase vaginal e candidíase
vaginal recorrente em mulheres diabéticas e não diabéticas
Luciene Setsuko Akimoto GuntherI, Helen Priscila Rodrigues MartinsII, Fabrícia GimenesIII,
André Luelsdorf Pimenta de AbreuIV, Marcia Edilaine Lopes ConsolaroV, Terezinha Inez Estivalet SvidzinskiV
Department of Clinical Analysis and Biomedicine, Universidade Estadual de Maringá (UEM), Maringá, Paraná, Brazil
MSc. Professor, Department of Clinical Analysis
and Biomedicine, Universidade Estadual de
Maringá (UEM), Maringá, Paraná, Brazil.
I
II
MSc. Nurse, Municipal Health Department of
Curitiba, Curitiba, Paraná, Brazil.
III
PhD. Postdoctoral Student, Department
of Clinical Analysis and Biomedicine,
Universidade Estadual de Maringá (UEM),
Maringá, Paraná, Brazil.
IV
MSc. Postgraduate Doctoral Student,
Department of Clinical Analysis and
Biomedicine, Universidade Estadual de Maringá
(UEM), Maringá, Paraná, Brazil.
PhD. Professor, Department of Clinical Analysis
and Biomedicine, Universidade Estadual de
Maringá (UEM), Maringá, Paraná, Brazil.
V
KEY WORDS:
Candida.
Candidiasis, vulvovaginal.
Diabetes mellitus.
Fluconazole.
Therapeutics.
PALAVRAS-CHAVE:
Candida.
Candidíase vulvovaginal.
Diabetes mellitus.
Fluconazol.
Terapêutica.
116
Sao Paulo Med J. 2014; 132(2):116-20
ABSTRACT
CONTEXT AND OBJECTIVE: Vulvovaginal candidiasis (VVC) is caused by abnormal growth of yeast-like
fungi on the female genital tract mucosa. Patients with diabetes mellitus (DM) are more susceptible to
fungal infections, including those caused by species of Candida. The present study investigated the frequency of total isolation of vaginal Candida spp., and its different clinical profiles — colonization, VVC and
recurrent VVC (RVVC) — in women with DM type 2, compared with non-diabetic women. The cure rate
using fluconazole treatment was also evaluated.
DESIGN AND SETTING: Cross-sectional study conducted in the public healthcare system of Maringá,
Paraná, Brazil.
METHODS: The study involved 717 women aged 17-74 years, of whom 48 (6.7%) had DM type 2 (mean
age: 53.7 years), regardless of signs and symptoms of VVC. The yeasts were isolated and identified using
classical phenotypic methods.
RESULTS: In the non-diabetic group (controls), total vaginal yeast isolation occurred in 79 (11.8%) women,
and in the diabetic group in 9 (18.8%) (P = 0.000). The diabetic group showed more symptomatic (VVC +
RVVC = 66.66%) than colonized (33.33%) women, and showed significantly more colonization, VVC and
RVVC than seen among the controls. The mean cure rate using fluconazole was 75.0% in the diabetic
group and 86.7% in the control group (P = 0.51).
CONCLUSION: We found that DM type 2 in Brazilian women was associated with yeast colonization, VVC
and RVVC, and similar isolation rates for C. albicans and non-albicans species. Good cure rates were obtained using fluconazole in both groups.
RESUMO
CONTEXTO E OBJETIVO: Candidíase vulvovaginal (CVV) é causada pelo crescimento anormal de fungos
do tipo leveduras na mucosa do trato genital feminino. Pacientes com diabetes mellitus (DM) são mais
susceptíveis a infecções fúngicas, incluindo por espécies de Candida. O presente estudo investigou a frequência de isolamento total de Candida spp. vaginal, e diferentes quadros clínicos (CVV e CVV recorrenteCVVR) em mulheres com DM tipo 2 comparadas às não diabéticas. A razão de cura do tratamento com
fluconazol também foi avaliada.
TIPO DE ESTUDO E LOCAL: Estudo transversal realizado no sistema público de saúde de Maringá, Paraná, Brazil.
MÉTODO: O estudo envolveu 717 mulheres de 17-74 anos de idade e, destas, 48 (6,7%) tinham DM 2
(média de 53,7 anos), independentemente de sinais e sintomas de CVV. As leveduras foram isoladas e
identificadas por métodos fenotípicos clássicos.
RESULTADOS: No grupo de não diabéticas (controle), leveduras vaginais totais foram isoladas em 79
(11,8%) mulheres, e no grupo de diabéticas, em 9 (18,8%) (P = 0,000). O grupo de diabéticas mostrou mais
mulheres sintomáticas (CVV + CVVR = 66,66%) do que colonizadas (33.33%), e significativamente mais colonização, CVV e CVVR, que as controle. A razão média de cura com fluconazol foi de 75.0% no grupo
diabéticas e 86.7% no controle (P = 0.51).
CONCLUSÃO: Nós encontramos que DM 2 em mulheres brasileiras associou-se com colonização vaginal
por leveduras, CVV e CVVR, razão similar de isolamento de C. albicans e espécies não albicans. Boa taxa de
cura foi obtida com fluconazol em ambos os grupos.
Prevalence of Candida albicans and non-albicans isolates from vaginal secretions: comparative evaluation of colonization, vaginal candidiasis and recurrent vaginal
candidiasis in diabetic and non-diabetic women | SHORT COMMUNICATION
INTRODUCTION
Vulvovaginal candidiasis (VVC) is classified by the World Health
Organization (WHO) as a pathological condition that is frequently sexually transmitted (STD).1 Because it affects millions
of women annually, thereby causing great discomfort, interfering
with sexual and affective relations and impairing work performance, VVC has been considered to be an important worldwide
public health problem.2
VVC is caused by abnormal growth of yeast-like fungi on
the mucosa of the female genital tract. It is clinically characterized by occurrences of intense vulvar itching, leucorrhea, dyspareunia, dysuria, edema and vulvovaginal erythema.2,3 Vaginal
yeasts become pathogenic when the colonization site on the host
is favorable to their development. Several factors may increase
this risk, such as previous colonization by the yeast, immunosuppressive diseases, diabetes mellitus (DM) and other factors.4,5
Patients with DM are more susceptible to bacterial and fungal infections, including those caused by species of Candida.6
Some investigators have suggested that VVC occurs more frequently in diabetic women, and others that a correlation exists
between hyperglycemia and VVC.2,7 However, few studies have
addressed the problem of VVC among Brazilian diabetic women.
OBJECTIVE
The objective of this study was to determine the frequency of
total isolation of vaginal Candida spp., and the clinical profiles,
VVC and recurrent VVC (RVVC), in women with DM type 2,
compared with non-diabetic women. The cure rate from fluconazole treatment was also evaluated.
METHODS
This experimental study involved women aged between 17 and
74 years who participated in the Cervical Cancer Triage Program,
regardless of signs and symptoms of VVC, between January 1 and
December 31, 2010, in the public healthcare system of Maringá,
Paraná, Brazil. Six healthcare centers participated in the study.
This study was approved by the Ethics Committee for Research on
Humans (COPEP) at Universidade Estadual de Maringá (UEM)
(185/2007). The exclusion criteria were pregnancy or a history
of immunosuppressive disease, including AIDS. The women
answered a standardized questionnaire that sought information
regarding VVC symptoms. Subjects were identified as affected
by DM type 2 according to the American Diabetes Association
(ADA) definition, if their fasting serum glucose was 7 mmol/l
(126 mg/dl) or more, as reported in the patients’ medical records.
A vaginal sample was collected using a sterile swab, inoculated in sterile saline and sent to the Medical Mycology
Laboratory at UEM, where it was immediately seeded onto plates
containing Sabouraud dextrose agar (SDA) (Difco, United States)
with 100 mg/ml of chloramphenicol, and incubated at 25 °C for
five days. A pool of the colonies grown on each plate was subcultured in CHROMagar Candida medium (Probac, France).
Beginning with the pure culture, the yeasts were identified using
classical phenotypic methods.8
The clinical profiles of the women with positive culture for
yeasts were classified into three types: colonized, but without
symptoms of VVC; with VVC, presenting an acute episode with
at least two of the following symptoms: discharge, itching, dysuria and dyspareunia; and with recurrent VVC (RVVC), presenting two or more of these symptoms and at least four episodes in
twelve months.9 Women who had RVVC were counted only once.
All the women in the diabetic group and 28 in the non-diabetic group with cultures positive for yeasts were treated, independent of clinical profile, using oral fluconazole (Neoquímica,
Brazil) at a single dose of 150 mg weekly for two weeks. They
were instructed to return 20 days after the end of treatment, so
that material for a new yeast culture could be collected.
The data were analyzed by means of the chi-square test, using
the STATA for Statistics and Data Analysis 9.1 software. All variables were expressed as absolute and relative frequencies. P values < 0.05 were considered significant.
RESULTS
Figure 1 shows an overview of the study. A total of 717 women
were included, of whom 48 (6.7%) had DM type 2 (mean age:
53.7 years). In the control group (mean age: 43.3 years), total vaginal yeast isolation occurred in 79 (11.8%) women, distributed
as C. albicans (n = 43; 54.4%) and non-albicans species (n = 36;
45.6%) (odds ratio [OR] = 7; 2.2-11.5; P = 0.02). In no case was
more than one yeast species isolated.
In the diabetic group, total yeast isolation occurred in 9
women (18.8%), which was a higher proportion than among
the controls (OR = 7.77; 3.88-15.56; P = 0.000), distributed as
C. albicans (n = 5; 55.6%) and non-albicans species (n = 4; 44.4%)
(P = 0.2) (Table 1).
Among the non-albicans species, C. glabrata was the most
frequent isolate in both the controls (n = 23/79; 29.1%) and the
diabetic group (n = 3/9; 33.3%).
With regard to clinical profiles, in the control group 24/79
women (30.4%) were colonized or had VVC or RVVC (55/79;
69.9%) (P = 0.411), with no significant difference between
VVC (44/79; 55.7%) and RVVC (11/79, 13.9%) (P = 0.201). In
the diabetic group, more women were symptomatic (VVC +
RVVC = 3/9, 66.66%) than colonized (3/9; 33.33%) (OR = 0.5;
0.125-1.999; P = 0.005) (Table 1).
The diabetic women showed a significantly higher proportion
of colonization (OR = 5; 2.08-23.46; P = 0.005), VVC + RVVC
(OR = 5; 2.50-10.44; P = 0.004), VVC (OR = 13; 4.23-49.13;
Sao Paulo Med J. 2014; 132(2):116-20
117
SHORT COMMUNICATION | Gunther LSA, Martins HPR, Gimenes F, Abreu ALP, Consolaro MEL, Svidzinski TIE
717 women screened using vaginal
culturing for Candida spp.
Colonization:
33.33%
C. albicans:
33.33%
Diabetic group: 178 diabetic women
Control group: 669 non-diabetic women
18.8% with vaginal Candida spp.
11.8% with vaginal Candida spp.
VVC: 33.33%
C. albicans:
100.0%
RVVC: 33.33%
C. albicans:
33.33%
Colonization:
30.4%
C. albicans:
41.7%
VVC: 55.7%
C. albicans:
54.5%
RVVC: 13.9%
C. albicans:
66.7%
Control group: 28 women treated with
oral fluconazole, in a single dose of
150 mg weekly, for 2 weeks
Diabetic group: 9 women treated with
oral fluconazole, in a single dose of
150 mg weekly, for 2 weeks
Return between 20 and
30 days after the end of
the complete treatment
Mean cure rate for diabetic
women: 75.0%
Mean cure rate for non-diabetic
women: 86.7%
Figure 1. An overview of the study and its results.
Table 1. Frequency of total isolation of vaginal Candida spp. and the clinical conditions of colonization, vulvovaginal candidiasis (VVC)
and recurrent VVC (RVVC) among diabetic and non-diabetic women in the public healthcare system of Maringá, Paraná, Brazil
Diabetic group
Non-diabetic group
(n = 48)
(n = 669)
Total isolation* Colonization*
VVC*
RVVC*
Total isolation
Colonization
VVC
RVVC
n
%
n
%
n
%
n
%
n
%
n
%
n
%
n
%
54.4
10
41.7
24
54.5
9
66.7
C. albicans
5
55.6
1
33.3
3
100.0
1
33.3
43†
Non-albicans
4
44.4
2
66.7
2
66.7
36
45.6
14
58.3
20
45.5
2
33.3
33.3
3‡
33.3
79
11.8
24§
30.4
44§
55.7
11§
13.9
Total
9
18.8
3
33.3
3‡
n = number; VVC = vulvovaginal candidiasis; RVVC = recurrent vulvovaginal candidiasis.
*
Total Candida spp. isolation (clinical conditions: colonization, VVC and RVVC) was significantly greater in the diabetic group (P = 0.000); †C. albicans was the most
frequent species isolated in the non-diabetic group (P = 0.02) considering only total Candida spp. isolation, but not for different clinical conditions (colonization,
VVC and RVVC) (P > 0.05); ‡Diabetic group showed more symptomatic women (VVC and RVVC) than colonized women (odds ratio, OR = 0.5; 0.125-1.999;
P = 0.005). §Control group was equally colonized or had VVC or RVVC (P = 0.411), without significant difference between VVC and RVVC (P = 0.201).
Diabetic group showed more colonization (OR = 5; 2.08-23.46; P = 0.005), VVC + RVVC (OR = 5; 2.50-10.44; P = 0.004), VVC (OR = 13; 4.23-49.13; P = 0.000) and
RVVC (OR = 2.6; 0.70-10.05; P = 0.000) than seen in the controls.
Candida
species
118
Sao Paulo Med J. 2014; 132(2):116-20
Prevalence of Candida albicans and non-albicans isolates from vaginal secretions: comparative evaluation of colonization, vaginal candidiasis and recurrent vaginal
candidiasis in diabetic and non-diabetic women | SHORT COMMUNICATION
P = 0.000) and RVVC (OR = 2.6; 0.70-10.05; P = 0.000) than seen
in the controls.
A total of 8 women in the diabetic group and 15 in the nondiabetic group correctly concluded the treatment with fluconazole, with mean cure rates of 75.0% and 86.7% respectively
(P = 0.51). In both the diabetic and the control group, C. glabrata
and C. tropicalis showed resistance to fluconazole.
trivial infection by some, the increasing incidence of VVC associated with diabetes raises additional issues regarding prevention
and patient management.
REFERENCES
1. Reese RE, Betts RF. Antibiotic use. In: Reese RE, Betts RF, editors. A
practical approach to infectious disease. 3rd ed. Boston: Little, Brown
and Company; 1991. p. 821-1007.
DISCUSSION
We found that DM type 2 in Brazilian women was associated
with Candida spp. colonization, VVC and RVVC; and that the
cure rate with fluconazole was satisfactory. Other investigators
have described associations between DM and VVC, colonization and RVVC, in different countries.6,10 This is very worrisome
because VVC afflicts millions of both non-diabetic and diabetic
women, causing great discomfort and interfering with sexual and
affective relations.2,3 Uncontrolled DM causes metabolic alterations, such as increased levels of glycogen, which can significantly
increase colonization and infection by Candida.11 The increased
glycogen level lowers the vaginal pH, thereby facilitating development of VVC.12
The women with DM studied here showed similar rates of C.
albicans and non-albicans species, thus differing from the control
group. The rates were similar to those reported by Lattif et al.10
and Faraji et al.13 In some populations, there has been an increase
in the isolation of vaginal non-albicans species, but most investigators agree that this does not seem to be a general trend.10,13
The relatively high cure rate with fluconazole in both diabetic
and non-diabetic women shows that this is a good therapeutic option even for Brazilian women with diabetes, who require
attention to treatment of non-albicans species, particularly C.
glabrata, which are intrinsically less susceptible to azole antifungals.3,14 Similarly to our results, Brumar et al.7 also reported a high
cure rate with fluconazole (85.71%) in diabetic women with VVC.
We acknowledge that the number of diabetic women in our
study was small, and that this group may not fully represent populations of diabetic women. Nevertheless, the number of diabetics (n = 48) was relatively high in this population of 717 women,
and we believe that this study contributes important information for management of diabetic women with vaginal Candida
spp. However, the interaction between DM type 2 and vaginal
Candida species merits further evaluation in relation to glycemic
control, in a larger sample of diabetic women in Brazil.
2. Sobel JD. Vulvovaginal candidosis. Lancet. 2007;369(9577):1961-71.
CONCLUSION
We found that diabetes in Brazilian women was associated with
yeast colonization, VVC and RVVC, with similar isolation rates
for C. albicans and non-albicans species; and that the cure rate
with fluconazole was relatively high. Although regarded as a
13. Faraji R, Rahimi MA, Rezvanmadani F, Hashemi M. Prevalence of
3. Wei Y, Feng J, Luo ZC. Isolation and genotyping of vaginal nonalbicans Candida spp. in women from two different ethnic groups in
Lanzhou, China. Int J Gynaecol Obstet. 2010;110(3):227-30.
4. Souza PC, Storti-Filho A, Souza RJ, et al. Prevalence of Candida sp. in
the cervical-vaginal cytology stained by Harris-Shorr. Arch Gynecol
Obstet. 2009;279(5):625-9.
5. Hetticarachchi N, Ashbee HR, Wilson JD. Prevalence and management
of non-albicans vaginal candidiasis. Sex Transm Infect. 2010;86(2):99-100.
6. Antony G, Saralaya V, Gopalkrishna-Bhat K, et al. Effect phenotypic
switching on expression of virulence factors by Candida albicans
causing candidiasis in diabetic patients. Rev Iberoam Micol.
2009;26(3):202-5.
7. Brumar A, Rosu AF, Calina D, et al. Study concerning vulvovaginal
candidiasis in women with diabetes. European Journal of Hospital
Pharmacy Science and Practice. 2012;19(2):213. Available from:
http://ejhp.bmj.com/content/19/2/213.1.short. Accessed in 2013
(Jun 11).
8. Kurtzman CP, Fell JW. The yeasts. A taxonomic study. 4th ed.
Amsterdam: Elsevier; 1998.
9. Chassot F, Camacho DP, Patussi EV, et al. Can Lactobacillus
acidophilus influence the adhesion capacity of the Candida albicans
on the combined contraceptive vaginal ring? Contraception.
2010;81(4):331-5.
10.Lattif AA, Mukhopadhyay G, Banerjee U, Goswami R, Prasad R.
Molecular typing and in vitro fluconazole susceptibility of Candida
species isolated from diabetic and nondiabetic women with
vulvovaginal candidiasis in India. J Microbiol Immunol Infect.
2011;44(1):166-71.
11. Corrêa PR, David PRS, Peres NP, et al. Caracterização fenotípica
de leveduras isoladas da mucosa vaginal em mulheres adultas
[Phenotypic characterization of yeasts isolated from the vaginal
mucosa of adult women]. Rev Bras Ginecol Obstet. 2009;31(4):177-81.
12. Carrara MA, Bazotte RB, Donatti L, et al. Effect of experimental
diabetes on the development and maintenance of vulvovaginal
candidiasis in female rats. Am J Obstet Gynecol. 2010;200(6):659.e1-4.
vaginal candidiasis infection in diabetic women. African Journal of
Microbiology Research. 2012;6(11):2773-8. Available from: http://
www.academicjournals.org/ajmr/pdf/Pdf2012/23%20March/
Faraji%20et%20al.pdf. Accessed in 2013 (Jun 11).
Sao Paulo Med J. 2014; 132(2):116-20
119
SHORT COMMUNICATION | Gunther LSA, Martins HPR, Gimenes F, Abreu ALP, Consolaro MEL, Svidzinski TIE
14. Dalben-Dota KF, Faria MG, Bruschi ML, et al. Antifungal activity of
propolis extract against yeasts isolated from vaginal exudates. J
Altern Complement Med. 2010;16(3):285-90.
Sources of funding: The present study was supported by a grant from
Fundação Araucária de Apoio ao Desenvolvimento Tecnológico e
Científico do Paraná, Brazil, protocol number 15.025/2009
Conflict of interest: None
Date of first submission: December 3, 2012
Last received: April 18, 2013
Accepted: June 19, 2013
Address for correspondence:
Marcia Edilaine Lopes Consolaro
Departamento de Análises Clínicas e Biomedicina
Universidade Estadual de Maringá
Av. Colombo, 5.790
Zona 07 — Maringá (PR) — Brasil
CEP 87020-900
Tel. (+55 44) 3011- 4795/3011-5996
E-mail: [email protected]
120
Sao Paulo Med J. 2014; 132(2):116-20
CASE REPORT
DOI: 10.1590/1516-3180.2014.1322635
Brainstem abscess of undetermined origin:
microsurgical drainage and brief antibiotic therapy
Abscesso do tronco encefálico de origem indeterminada:
drenagem microcirúrgica e antibioticoterapia curta
Pedro Tadao Hamamoto FilhoI, Marco Antonio ZaniniII
Faculdade de Medicina de Botucatu (FMB), Universidade Estadual Paulista (Unesp), Botucatu, São Paulo, Brazil
MD. Medical Resident, Department of
Neurology, Psychology and Psychiatry,
Faculdade de Medicina de Botucatu (FMB),
Universidade Estadual Paulista (Unesp),
Botucatu, São Paulo, Brazil.
I
II
MD, PhD. Associate Professor, Department
of Neurology, Psychology and Psychiatry,
Faculdade de Medicina de Botucatu (FMB),
Universidade Estadual Paulista (Unesp),
Botucatu, São Paulo, Brazil.
KEY WORDS:
Brain stem.
Abscess.
Central nervous system bacterial infections.
Neurosurgical procedures.
Anti-bacterial agents.
PALAVRAS-CHAVE:
Tronco encefálico.
Abscesso.
Infecções bacterianas do sistema nervoso
central.
Procedimentos neurocirúrgicos.
Antibacterianos.
ABSTRACT
CONTEXT: Solitary brainstem abscesses are rare and they are usually associated with other infections. They
are severe conditions with high morbidity and mortality. The surgical options are stereotactic aspiration
and microsurgical drainage. Systemic antibiotic therapy is used for more than six weeks.
CASE REPORT: We present the case of a young man with a solitary abscess at the pons, without other
systemic infections. The patient was treated by means of microsurgical drainage and antibiotic therapy for
three weeks. His postoperative recovery was good.
CONCLUSIONS: A microsurgical approach may be considered to be an important option for large abscesses that are multiloculated, close to the surface or contain thick fluid. Complete emptying of the purulent accumulation may diminish the required duration of antibiotic therapy.
RESUMO
CONTEXTO: Abscessos isolados do tronco encefálico são raros e geralmente associados a outras infecções. Trata-se de condição grave, com grande morbidade e mortalidade. Opções cirúrgicas são aspiração
com estereotaxia e drenagem microcirúrgica. Antibioticoterapia sistêmica tem sido usada por mais de
seis semanas.
RELATO DE CASO: Apresentamos o caso de um jovem com abscesso pontino sem outras infecções sistêmicas. O paciente foi tratado com drenagem microcirúrgica e antibioticoterapia por três semanas. Houve
boa evolução pós-operatória.
CONCLUSÕES: Acesso microcirúrgico pode ser considerado uma opção importante no tratamento
de grandes abscessos do tronco encefálico, que são multiloculados, próximos da superfície ou que
contenham líquido espesso. Drenagem completa do material purulento pode diminuir o período de
antibioticoterapia.
Sao Paulo Med J. 2014; 132(2):121-4
121
CASE REPORT | Hamamoto Filho PT, Zanini MA
Table 1. Results from our literature review of medical databases for case reports and case series of brainstem abscesses that were treated
surgically. Search conducted on February 14, 2013
Database
Medline (via PubMed)
Embase (via Elsevier)
Lilacs
Search strategies
(((brain stem) AND abscess) AND surgery) AND “case reports”
[Publication Type]
(((brain stem) AND abscess) AND surgery) AND “case reports”
[Publication Type]
abscesso [Palavras] AND tronco [Palavras] AND humanos [Limites]
We found that few cases have been published. Abstracts or full
texts were analyzed and it was seen that less than 30 reports were
similar to ours. Moreover, only one of these published cases was
from Brazil.13
Papers found
Papers related
88
14
47
11
8
2
10. Ghannane H, Laghmari M, Aniba K, Lmejjati M, Benali SA. Diagnostic
and management of pediatric brain stem abscess, a case-based
update. Childs Nerv Syst. 2011;27(7):1053-62.
11. Suzer T, Coskun E, Cirak B, Yagci B, Tahta K. Brain stem abscesses in
childhood. Childs Nerv Syst. 2005;21(1):27-31.
CONCLUSION
A microsurgical approach may be considered to be an important option for large abscesses that are multiloculated, close to the
surface or contain thick fluid. Complete emptying of the purulent accumulation may diminish the required duration of antibiotic therapy.
12. Nakajima H, Iwai Y, Yamanaka K, Kishi H. Successful treatment
REFERENCES
Sources of funding: None
1. Arzoglou V, D’Angelo L, Koutzoglou M, Di Rocco C. Abscess of
Conflict of interest: None
of brainstem abscess with stereotactic aspiration. Surg Neurol.
1999;52(5):445-8.
13. Hermes de N Jr, Rodrigues Pereira EL, Castro Ribeiro DE, et al.
Staphylococcus aureus brainstem abscess in a Brazilian Amazon
man. Case report. J Neurosurg Sci. 2011;55(4):383-5.
the medulla oblongata in a toddler: case report and technical
considerations based on magnetic resonance imaging tractography.
Date of first submission: November 27, 2012
Neurosurgery. 2011;69(2):E483-6; discussion E486-7.
Last received: September 5, 2013
2. Nathoo N, Nadvi SS, Narotam PK, van Dellen JR. Brain abscess:
Accepted: September 9, 2013
management and outcome analysis of a computed tomography era
experience with 973 patients. World Neurosurg. 2011;75(5-6):716-26;
Address for correspondence:
discussion 612-7.
Pedro Tadao Hamamoto Filho
3. Danziger J, Allen KL, Bloch S. Brain-stem abscess in childhood. Case
report. J Neurosurg. 1974;40(3):391-3.
4. Jamjoom ZA. Solitary brainstem abscess successfully treated by
microsurgical aspiration. Br J Neurosurgery. 1992;6(3):249-53.
5. Rosenblum ML, Hoff JT, Norman D, Weinstein PR, Pitts L. Decreased
Depto Neurologia, Psicologia e Psiquiatria
Distrito de Rubião Jr, s/no
Botucatu (SP) — Brasil
CEP 18618-970
mortality from brain abscesses since the advent of computerized
Tel. (+55 14) 3880-1220
tomography. J Neurosurg. 1978;49(5):658-68.
Fax. (+55 14) 3815-5965
6. Wait SD, Beres EJ, Nakaji P. Bacterial abscess of the medulla oblongata.
J Clin Neurosci. 2009;16(8):1082-4.
7. Kirchhoff DC, Kirchhoff DFB, Muoio V. Abscessos do tronco cerebral:
apresentação de seis casos [Brain stem abscess: a study of six cases].
Arq Bras Neurocir. 2008;27(1):30-4.
8. Fulgham JR, Wijdicks EF, Wright AJ. Cure of a solitary brainstem abscess
with antibiotic therapy: case report. Neurology. 1996;46(5):1451-4.
9. Lai PH, Li KT, Hsu SS, et al. Pyogenic brain abscess: findings from in
vivo 1.5-T and 11.7-T in vitro proton MR spectroscopy. AJNR Am J
Neuroradiol. 2005;26(2):279-88.
124
Unesp - Campus de Botucatu
Sao Paulo Med J. 2014; 132(2):121-4
E-mail: [email protected]
LETTER TO EDITOR
DOI: 10.1590/1516-3180.2014.1322655
Screening for 22q11 deletion syndrome
among patients with congenital heart defects
Triagem para a síndrome de deleção 22q11 entre pacientes com cardiopatia congênita
Rafael Fabiano Machado RosaI, Rosana Cardoso Manique RosaII, Patrícia TrevisanII, Carla GraziadioIII, Marileila Varella-GarciaIV,
Giorgio Adriano PaskulinV, Paulo Ricardo Gazzola ZenVI
Universidade Federal de Ciências da Saúde de Porto Alegre (UFCSPA) and Complexo Hospitalar Santa Casa de Porto Alegre (CHSCPA),
Porto Alegre, Rio Grande do Sul, Brazil
PhD. Professor of Postgraduate Program on
Pathology, Universidade Federal de Ciências da
Saúde de Porto Alegre (UFCSPA), and Clinical
Geneticist, Universidade Federal de Ciências
da Saúde de Porto Alegre (UFCSPA), Complexo
Hospitalar Santa Casa de Porto Alegre (CHSCPA)
and Hospital Materno Infantil Presidente Vargas
(HMIPV), Porto Alegre, Rio Grande do Sul, Brazil.
I
MD. Postgraduate Student, Postgraduate
Program on Pathology, Universidade Federal de
Ciências da Saúde de Porto Alegre (UFCSPA),
Porto Alegre, Rio Grande do Sul, Brazil.
II
MD. Postgraduate Student, Postgraduate
Program on Pathology and Assistant Professor
of Clinical Genetics, Universidade Federal de
Ciências da Saúde de Porto Alegre (UFCSPA);
and Clinical Geneticist, Universidade Federal de
Ciências da Saúde de Porto Alegre (UFCSPA) and
Complexo Hospitalar Santa Casa de Porto Alegre
(CHSCPA), Porto Alegre, Rio Grande do Sul, Brazil.
III
IV
PhD. Professor of Medical Oncology and
Director of the Cancer Center Cytogenetics Core,
University of Colorado Denver (UCD), Aurora,
Colorado, United States.
PhD. Associate Professor of Clinical Genetics
and Coordinator of the Postgraduate Program
on Pathology, Universidade Federal de Ciências
da Saúde de Porto Alegre (UFCSPA); and Clinical
Geneticist, Universidade Federal de Ciências da
Saúde de Porto Alegre (UFCSPA) and Complexo
Hospitalar Santa Casa de Porto Alegre (CHSCPA),
Porto Alegre, Rio Grande do Sul, Brazil.
V
VI
PhD. Adjunct Professor of Clinical Genetics and
Professor of the Graduate Program in Pathology,
Universidade Federal de Ciências da Saúde de
Porto Alegre (UFCSPA); and Clinical Geneticist,
Universidade Federal de Ciências da Saúde de
Porto Alegre (UFCSPA) and Complexo Hospitalar
Santa Casa de Porto Alegre (CHSCPA), Porto
Alegre, Rio Grande do Sul, Brazil.
The 22q11 deletion syndrome (22q11DS), or velocardiofacial/DiGeorge syndrome, is considered to be the second most known genetic cause of congenital heart disease (CHD).1 Our
aim was to evaluate the effectiveness of different screening methods for 22q11DS in patients
with CHD. Our study evaluated a consecutive sample of patients with CHD hospitalized for the
first time in a pediatric and cardiac intensive care unit of a referral hospital in southern Brazil.
All of them underwent the examination through fluorescent in situ hybridization for 22q11DS.
These patients were part of the study by Rosa et al.2 CHDs were classified by a cardiologist as
conotruncal or non-conotruncal. We excluded patients with other chromosomal abnormalities.
Three different approaches composed the screening:
(1) Testing suggested by Tobias et al.3 The clinical findings are divided into three categories:
A) conotruncal CHD;
B) abnormalities common in 22q11DS, such as hypocalcemia and non-conotruncal CHD; and
C) abnormalities less common in 22q11DS, such as short stature and hypotonia. Patients
that have an alteration in group A, two findings in group B, or one finding in group B
plus one in group C are tested;
(2)Testing suggested by the American Heart Association (AHA) Congenital Cardiac Defects
Committee,4 which consists of testing all newborns/infants with interrupted aortic arch (IAA),
truncus arteriosus (TA), tetralogy of Fallot (TOF), ventricular septal defect (VSD) (perimembranous conoseptal hypoplasia or malalignment) with aortic arch anomaly (AAA), AAA
alone and discontinuous branch pulmonary arteries. The screening also includes any newborn/infant/child with CHD associated with another feature of 22q11DS (such as hypocalcemia, facial dysmorphia and palate abnormality); newborns/infants with VSD, and any child/
adolescent/adult with TOF, TA, IAA, VSD or AAA who has one other feature of 22q11DS;
(3) Testing performed at some centers,5 where only patients with conotruncal heart defects are tested.
For all these approaches, we calculated the sensitivity, specificity, positive predictive value
(PPV), negative predictive value (NPV) and receiver operating characteristic (ROC) curves. The
significance level used was 5% (P ≤ 0.05). The total sample consisted of 170 patients (93 males),
with ages ranging from 1 to 4934 days (mean of 847.7 days, standard deviation of 1225.1).
22q11DS was identified in four patients (2.4%): two newborns with TOF, one newborn with
VSD associated with AAA, and one adolescent with atrial septal defect. One hundred and eleven
patients (65%) met screening criterion 1, with sensitivity 100%, specificity 36%, PPV 3.6% and
NPV 100%. Criterion 2 was met by 76 (44.7%) patients, with sensitivity 75%, specificity 56%,
PPV 3.9% and NPV 98.9%. Forty-five patients (26.5%) had conotruncal heart defects and fulfilled criterion 3, with sensitivity 50%, specificity 74%, PPV 4.4% and NPV 98.4%. The ROC
curves are shown in Figure 1. All the areas under the ROC curves were less than 0.5 [criterion 1:
0.247 (P = 0.083); criterion 2: 0.295 (P = 0.161); and criterion 3: 0.374 (P = 0.388)].
Sao Paulo Med J. 2014; 132(2):125-6
125
COCHRANE HIGHLIGHTS
DOI: DOI: 10.1590/1516-3180.20141322T1
Oral treatments for fungal
infections of the skin
of the foot
Sally E. M. Bell-Syer, Sameena M. Khan, David J. Torgerson
The independent commentary was written by Leontina da
Conceição Margarido
ABSTRACT
BACKGROUND: About 15% of the world population have fungal infections of the feet (tinea pedis or athlete’s foot). There are many clinical
presentations of tinea pedis, and most commonly, tinea pedis is seen
between the toes (interdigital) and on the soles, heels, and sides of the
foot (plantar). Plantar tinea pedis is known as moccasin foot. Once acquired, the infection can spread to other sites including the nails, which
can be a source of re-infection. Oral therapy is usually used for chronic
conditions or when topical treatment has failed.
OBJECTIVE: To assess the effects of oral treatments for fungal infections
of the skin of the foot (tinea pedis).
METHODS:
Search methods: For this update we searched the following databases
to July 2012: the Cochrane Skin Group Specialized Register, CENTRAL in
The Cochrane Library, MEDLINE (from 1946), EMBASE (from 1974), and
CINAHL (from 1981). We checked the bibliographies of retrieved trials for
further references to relevant trials, and we searched online trials registers.
Selection criteria: Randomized controlled trials of oral treatments in participants who have a clinically diagnosed tinea pedis, confirmed by microscopy and growth of dermatophytes (fungi) in culture.
Data collection and analysis: Two review authors independently undertook study selection, “Risk of bias” assessment, and data extraction.
Main results: We included 15 trials, involving 1,438 participants.
The 2 trials (71 participants) comparing terbinafine and griseofulvin
produced a pooled risk ratio (RR) of 2.26 (95% confidence interval (CI)
1.49 to 3.44) in favors of terbinafine’s ability to cure infection. No significant difference was detected between terbinafine and itraconazole, fluconazole and itraconazole, fluconazole and ketoconazole, or between
griseofulvin and ketoconazole, although the trials were generally small.
Two trials showed that terbinafine and itraconazole were effective compared with placebo: terbinafine (31 participants, RR 24.54, 95%CI 1.57 to
384.32) and itraconazole (72 participants, RR 6.67, 95%CI 2.17 to 20.48).
All drugs reported adverse effects, with gastrointestinal effects most
commonly reported. Ten of the trials were published over 15 years ago,
and this is reflected by the poor reporting of information from which to
make a clear “Risk of bias” assessment. Only one trial was at low risk of
bias overall. The majority of the remaining trials were judged as “unclear”
risk of bias because of the lack of clear statements with respect to methods of generating the randomization sequence and allocation concealment. More trials achieved blinding of participants and personnel than
blinding of the outcome assessors, which was again poorly reported.
AUTHORS’ CONCLUSIONS: The evidence suggests that terbinafine
is more effective than griseofulvin; and terbinafine and itraconazole
are more effective than no treatment. In order to produce more reliable data, a rigorous evaluation of different drug therapies needs to be
undertaken with larger sample sizes to ensure they are large enough
to show any real difference when two treatments are being compared.
It is also important to continue to follow up and collect data, preferably for six months after the end of the intervention period, to establish
whether or not the infection recurred.
This is the abstract of a Cochrane Review published in the Cochrane
Database of Systematic Reviews (CDSR) 2012, issue 10, Art. No.
CD003584. DOI: 10.1002/14651858.CD003584.pub2 (http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003584.pub2/full). For full
citation and authors details see reference 1.
The full text is available from: http://onlinelibrary.wiley.com/
store/10.1002/14651858.CD003584.pub2/asset/CD003584.pdf?v=1&t=
hralbcjp&s=8a122343e91a325aeb541587596a057767c41698
The abstract is also available in the Portuguese, Spanish and Chinese
languages from: http://summaries.cochrane.org/CD003584/oral-antifungal-drugs-for-treating-athletes-foot-tinea-pedis#sthash.i0KoYuHh.dpuf
REFERENCE
1. Bell-Syer SE, Khan SM, Torgerson DJ. Oral treatments for fungal
infections of the skin of the foot. Cochrane Database Syst Rev.
2012;10: CD003584.
COMMENTS
“Tinea pedis”, which includes interdigital mycosis (i.e. between the toes)
and onychomycosis (nail mycosis), is a very important and frequent
problem and needs to be well treated. Nail mycosis can be a source of
reinfection. Itraconazole is the most effective agent since it eliminates
dermatophytes, yeasts and also filamentous fungi, while terbinafine is
effective against dermatophytes.
Cracks, scaling or maceration in the toe-web spaces (interdigital mycosis) may be the sites where beta-hemolytic streptococci enter to
cause cellulitis of the legs. However, these sites are often little heeded.
In 85% of a group of patients with cellulitis and tinea pedis, cultures
yielded beta-hemolytic streptococci (Lancefield group A in 4, group B
in 3, group C in 1 and group G in 9). In contrast, in a control group of
30 patients with tinea pedis but without cellulitis, beta-hemolytic streptococci were not isolated from the interdigital spaces. The growth of
Gram-negative bacilli and Staphylococcus aureus was similar in the two
patient populations.1 The bacteria may cause cellulitis by entering the
skin at these sites or by spreading to contiguous cutaneous surfaces
and invading through a disrupted epidermis or through areas of impaired local defenses.
This correlation aims to emphasize the importance of seeking and treating tinea pedis, so as to prevent recurrent episodes of cellulitis, which
are common in patients who have had a previous attack.
Leontina da Conceição Margarido, MD, PhD. Professor at the Biological
and Health Sciences Center, Mackenzie Presbyterian University; Chairman of the Department of Dermatology, Associação Paulista de Medicina (APM); Counselor of the Brazilian Society of Dermatology.
REFERENCE
1. Semel JD, Goldin H. Association of athlete’s foot with cellulitis of the
lower extremities: diagnostic value of bacterial cultures of ipsilateral
interdigital space samples. Clin Infect Dis. 1996;23(5):1162-4.
Sao Paulo Med J. 2014; 132(2):127 127
COCHRANE HIGHLIGHTS
Interventions for preventing
obesity in children
Elizabeth Waters, Andrea de Silva-Sanigorski, Belinda J.
Burford, Tamara Brown, Karen J. Campbell, Yang Gao,
Rebecca Armstrong, Lauren Prosser, Carolyn D. Summerbell
The independent commentary was written by Angela
Maria Spinola e Castro
ABSTRACT
BACKGROUND: Prevention of childhood obesity is an international
public health priority given the significant impact of obesity on acute
and chronic diseases, general health, development and well-being.
The international evidence base for strategies that governments, communities and families can implement to prevent obesity, and promote
health, has been accumulating but remains unclear.
OBJECTIVE: This review primarily aims to update the previous Cochrane
review of childhood obesity prevention research and determine the effectiveness of evaluated interventions intended to prevent obesity in
children, assessed by change in Body Mass Index (BMI). Secondary aims
were to examine the characteristics of the programs and strategies to
answer the questions “What works for whom, why and for what cost?”
METHODS:
Search methods: The searches were re-run in CENTRAL, MEDLINE, EMBASE, PsychINFO and CINAHL in March 2010 and searched relevant
websites. Non-English language papers were included and experts
were contacted.
Selection criteria: The review includes data from childhood obesity prevention studies that used a controlled study design (with or without
randomisation). Studies were included if they evaluated interventions,
policies or programs in place for twelve weeks or more. If studies were
randomized at a cluster level, six clusters were required.
Data collection and analysis: Two review authors independently extracted data and assessed the risk of bias of included studies. Data
was extracted on intervention implementation, cost, equity and outcomes. Outcome measures were grouped according to whether they
measured adiposity, physical activity (PA)-related behaviours or dietrelated behaviours. Adverse outcomes were recorded. A meta-analysis
was conducted using available BMI or standardized BMI (zBMI) score
data with subgroup analysis by age group (0-5, 6-12, 13-18 years, corresponding to stages of developmental and childhood settings).
MAIN RESULTS: This review includes 55 studies (an additional 36 studies found for this update). The majority of studies targeted children aged
6-12 years. The meta-analysis included 37 studies of 27,946 children and
demonstrated that programmes were effective at reducing adiposity, although not all individual interventions were effective, and there was a high
level of observed heterogeneity (I2 = 82%). Overall, children in the intervention group had a standardised mean difference in adiposity (measured as
BMI or zBMI) of -0.15kg/m2 (95% confidence interval (CI): -0.21 to -0.09). Intervention effects by age subgroups were -0.26kg/m2 (95% CI -0.53 to 0.00)
(0- 5 years), - 0.15 kg/m2 (95% CI -0.23 to -0.08) (6-12 years), and -0.09 kg/m2
(95% CI -0.20 to 0.03) (13-18 years). Heterogeneity was apparent in all three
age groups and could not explained by randomisation status or the type,
duration or setting of the intervention. Only eight studies reported on adverse effects and no evidence of adverse outcomes such as unhealthy dieting practices, increased prevalence of underweight or body image sensitivities was found. Interventions did not appear to increase health inequalities
although this was examined in fewer studies.
128
Sao Paulo Med J. 2014; 132(2):128-9
DOI: 10.1590/1516-3180.20141322T2
AUTHORS’ CONCLUSIONS: We found strong evidence to support beneficial effects of child obesity prevention programmes on BMI, particularly for programmes targeted to children aged six to 12 years. However,
given the unexplained heterogeneity and the likelihood of small study
bias, these findings must be interpreted cautiously. A broad range of
programme components were used in these studies and whilst it is not
possible to distinguish which of these components contributed most to
the beneficial effects observed, our synthesis indicates the following
to be promising policies and strategies:
• school curriculum that includes healthy eating, physical activity and
body image;
• increased sessions for physical activity and the development of fundamental movement skills throughout the school week;
• improvements in nutritional quality of the food supply in schools;
• environments and cultural practices that support children eating
healthier foods and being active throughout each day;
• support for teachers and other staff to implement health promotion
strategies and activities (e.g. professional development, capacity
building activities);
• parent support and home activities that encourage children to be
more active, eat more nutritious foods and spend less time in screen
based activities. However, study and evaluation designs need to be
strengthened, and reporting extended to capture process and implementation factors, outcomes in relation to measures of equity, longer
term outcomes, potential harms and costs. Childhood obesity prevention research must now move towards identifying how effective
intervention components can be embedded within health, education and care systems and achieve long term sustainable impacts.
This is the abstract of a Cochrane Review published in the Cochrane
Database of Systematic Reviews (CDSR) 2011, issue 12, Art. No.
CD001871. DOI: 10.1002/14651858.CD001871.pub3 (http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD001871.pub3/abstract). For
full citation and authors details see reference 1.
The full text is available from: http://cochrane.bvsalud.org/doc.php?d
b=reviews&id=CD001871&lib=COC
REFERENCE
1. Waters E, de Silva-Sanigorski A, Burford BJ, et al. Interventions
for preventing obesity in children. Cochrane Database Syst Rev.
2011;(12):CD001871.
COMMENTS
Obesity during childhood and adolescence has become an issue of major concern over the last 10 years, around the world, regardless of race.
This event has been named an epidemic and has much to do with profound
changes not only in economic issues through higher per capita income, but
also especially in dietary habits and lifestyle, in parallel with decreased physical activity due to a variety of reasons. Despite all the concerns about overweight/obesity and the advent of numerous diets for weight reduction, it
appears that most of them have been ineffective in reducing weight. Thus,
this review is very timely, particularly given the importance of evaluating the
effectiveness of various intervention programs for obese children.
Worldwide experience in this regard has been very extensive, involving
governments, communities and families. Nonetheless, at the present
time, there is no good evidence about the best strategy for health promotion relating to weight loss.
COCHRANE HIGHLIGHTS
The first objective of this review was to determine the effectiveness of
intervention programs for preventing obesity in children, as assessed
by the body mass index (BMI). This form of evaluation of programs has
questionable sensitivity among children, whose BMI is not a parameter
that reflects weight loss fairly, since only the influence of growth itself can
interfere with BMI, even if there is only a slight weight reduction. This becomes more important during puberty. Likewise, children and especially
teenagers who are committed towards impact sports that have important influences on body composition will also present divergent results.
The second objective can be summarized as a review of the characteristics of the programs and strategies that fit these patients, as well as
their costs and benefits.
In accordance with the review criteria, all studies with control groups
were included. However, not all studies reviewed were randomized,
which may have facilitated entry in the group of children and or families
who were already motivated to lose weight, which may have interfered
with the results obtained.
The authors included 37 studies in the review, corresponding to 27,496
children aged 6 to 12 years, and they concluded that the programs were
effective for reducing adiposity, although not all interventions have
reached good results. Furthermore, there was great heterogeneity in the
results found, which could not be adequately explained by the review
authors, and they suggested that caution is needed in interpreting the results. Some of the issues that were pointed out here, along with the study
population itself, and the possibility of bias would probably explain this
heterogeneity, including the need for better assessment of the behavior
of randomized trials, in comparison with non-randomized trials. Evaluation of the percentage of the children who did not lose weight but had
stabilized, or those who may have gained weight, would perhaps have
been helpful for our understanding of the results obtained. Moreover, it
also needs to be borne in mind that the clinical approach towards obese
patients should always be individualized, especially in relation to children
who are exposed to different kinds of influences, from families, peers,
school, the environment and so on.
This review was also unable to identify which aspects of the programs
have in fact contributed to the slight weight loss. Nevertheless, in the
discussion, which is very well written, the authors stressed some very
important issues: the need to improve environmental conditions and
cultural practices so as to emphasize healthy food intake; the need for
a curriculum, including notions of healthy eating, physical activity and
body image; and educational support for teachers in relation to health
promotion activities. However, it is important to stress that, without adequate participation and awareness among families, it will be difficult
for any program to succeed. Given that these authors also suggested
that increased physical activity during the school week is important, it
must also be said that in certain areas and countries, it is very difficult to
implement these activities due to the lack of public policies.
This review has not added any new facts and, in effect, does not indicate
what type of intervention promotes better outcomes. Its conclusions
are well known in approaches towards obese children that have already
become part of routine care. Perhaps the time available for action in
the studies included was very short, such as the minimum of 12 weeks,
which may have been insufficient to promote changes in behavior. On
the other hand, this underlines the need to establish public policies
that allow teachers and educators to teach about nutrition, increased
physical activity in schools and creation of community spaces for practicing exercises, with full participation by families. Perhaps, rather than
focusing only on weight, it is more important to work on adherence to
healthier eating habits and exercise, not only towards weight loss but
also towards health promotion. Assessment of body composition is also
becoming more important than BMI seen in isolation. Future evalua-
tions need to focus on following up the evolution of weight after the
intervention and on maintaining the knowledge acquired over the long
term, because only through education will it be possible to prevent
obesity and its comorbidities, as a lifetime program.
Angela Maria Spinola e Castro. MD, PhD. Associate Professor of the Department of Pediatrics, Escola Paulista de Medicina, Universidade Federal de São Paulo (EPM-Unifesp); Head of Pediatric Endocrinology and
Chairman of the Department of Endocrinology, Associação Paulista de
Medicina (APM), São Paulo, Brazil.
Sao Paulo Med J. 2014; 132(2):128-9 129
ERRATUM
In the article “Recurrence of cervical intraepithelial neoplasia
in human immunodeficiency virus-infected women treated
by means of electrosurgical excision of the transformation
zone (LLETZ) in Rio de Janeiro, Brazil”, published in the São
Paulo Medical Journal, volume 131, issue number 6, 2013, the
correct name of the fourth author is Beatriz Grinsztejn and
not Beatriz Gilda Jegerhorn Grinstejn. The article should be
130
Sao Paulo Med J. 2014; 132(2):130
cited thus: Russomano F, Paz BR, Camargo MJ, Grinsztejn B,
Friedman RK, Tristao MAP, Oliveira CA. Recurrence of cervical intraepithelial neoplasia in human immunodeficiency
virus-infected women treated by means of electrosurgical
excision of the transformation zone (LLETZ) in Rio de Janeiro,
Brazil. Sao Paulo Med J. 2013; 131(6):405-10. PubMed PMID:
24346780. DOI: 10.1590/1516-3180.2013.1316578.
SÃO PAULO MEDICAL JOURNAL/EVIDENCE FOR HEALTH CARE
Indexing and scope
The São Paulo Medical Journal/Evidence for Health Care was
founded in 1932. Its articles are indexed in Medline, Lilacs, SciELO,
Science Citation Index Expanded, Journal Citation Reports/Science
Edition (ISI) and EBSCO Publishing.
Published bimonthly by the Associação Paulista de Medicina, the
journal accepts articles in the fields of clinical health science (internal
medicine, gynecology and obstetrics, mental health, surgery, pediatrics and public health). Articles will be accepted in the form of original articles (clinical trials, cohort, case-control, prevalence, incidence,
accuracy and cost-effectiveness studies and systematic reviews with
or without meta-analysis), narrative reviews of the literature, case
reports, short communications and letters to the editor. Papers with a
commercial objective will not be accepted.
The Journal’s policy and procedures
After receipt of the article by the Scientific Publications Sector, the
authors will be provided with a protocol number. This number serves
to maintain good understanding between the authors and the Scientific Publications Sector. Following this, the article will be read by the
Editor, who will verify whether it is consonant with the journal’s policy and interests, i.e. whether the research or review is within the fields
of health or public health.
Next, the Scientific Publications Sector will verify whether the
text complies with the journal’s Instructions for Authors. If the text
is incomplete or if it is not organized as required, the authors will be
asked to resubmit their text after resolving such problems. When its
format is acceptable, the Scientific Publications Sector will submit the
manuscript to closed peer review, in which the reviewers will not sign
their verdict and will not know the names of the authors. Each paper
will be reviewed by at least three reviewers: one expert in the field, one
associate editor (who will evaluate the article from the reader’s perspective) and one ad hoc editorial advisor (who will assess methodological aspects of the study).
The authors will then receive the reviewers’ evaluation and will
be asked to resolve all the problems that have been pointed out. Once
the Scientific Publications Sector receives the manuscript again, the
text will be sent to the scientific editor and the proofreader, who will
point out problems with sentence construction, spelling, grammar,
bibliographical references and other matters. The authors should then
provide all further information and corrections requested and should
mark in the text all the points at which modifications have been made,
using different colors or electronic text marking systems, so that these
modifications are easy to see.
When the text is considered acceptable for publication, and only
then, it will enter the queue for publication and the author will receive
a letter of acceptance of the article. The Scientific Publications Sector
will provide a proof, including any tables and figures, for the authors
to approve. No article is published without this last procedure.
Instructions for authors
General guidelines: for all types of articles
Texts must be submitted exclusively through the Internet,
using the electronic submission system, which is available at
http://www.spmj.hemeroteca.com.br. Submissions sent by e-mail
or through the post will not be accepted.
The manuscript must be submitted in English. Nonetheless, it
must also include a summary and five key words both in Portuguese
and in English. The key words must be selected from the DeCS and
MeSH lists only, as explained in detail below (no other key words will
be accepted).
Papers submitted must be original and therefore all the authors
need to declare that the text has not been and will not be submitted
for publication in any other journal. Papers involving human beings
(individually or collectively, directly or indirectly, totally or partially,
including the management of information and materials) must be
accompanied by a copy of the authorization from the Research Ethics
Committee of the institution in which the experiment was performed.
All articles submitted must comply with the editorial standards
established in the Vancouver Convention (Uniform Requirements for
Manuscripts Submitted to Biomedical Journals)1 and the specific quality guidelines for papers reporting on clinical trials (CONSORT),2 systematic reviews and meta-analyses (PRISMA),3,4 observational studies (STROBE)5,6 and accuracy studies on diagnostic tests (STARD).7,8
The style known as the “Vancouver Style” is to be used not only for
the format of the references, but also for the whole text. The Editors
recommend that authors should familiarize themselves with this style
by accessing http://www.icmje.org.
Abbreviations must not be used, even those in common use.
Drugs or medications must be referred to using their generic names,
avoiding unnecessary mention of commercial or brand names, and
should be followed by the dosage and posology. Any product cited
in the Methods section, such as diagnostic or therapeutic equipment,
tests, reagents, instruments, utensils, prostheses, orthoses and intraoperative devices must be described together with the manufacturer’s name and place (city and country) of manufacture in parentheses.
Grants, bursaries and any other financial support for studies must
be mentioned separately after the references, in a section named
“Acknowledgements”, along with any other acknowledgements to
individuals or professionals who have helped in producing the study
but whose contribution does not constitute authorship (we recommend that the item “Authorship” at http://www.icmje.org should be
read to obtain clarifications regarding the criteria for authorship).
For any type of study, all statements in the text that are not results
from the study presented for publication in the São Paulo Medical Journal/Evidence for Health Care, but are data from other studies already published elsewhere must be accompanied by citations
of the pertinent literature. Thus, statements about the incidence or
Sao Paulo Med J. 2014; 132(2):I-V
I
These instructions are updated periodically. We recommend that they are consulted online at: www.scielo.br/spmj
prevalence of diseases, costs, frequency of use of certain therapies
References
and epidemiological data in general should be followed by the refer-
The list of references (in the “Vancouver style”, as indicated by the
ences for the surveys that generated this information, even if the data
International Committee of Medical Journal Editors, ICMJE) should be
come from government institutions or databases, given that these are
laid out in the final part of the article, after the conclusions and before the
data from other studies.
tables and figures. In the text, the references must be numbered accord-
Format
ing to the order of citation. The citation numbers must be inserted after
First page (cover page)
periods/full stops or commas in sentences (see examples in the preced-
The first page must contain:
ing section), and must be in superscript form (without using parentheses
1) the type of paper (original article, review or updating article,
or square brackets). References cited in the legends of tables and figures
short communication or letter to the editor);
2) the title of the paper in English and Portuguese, which must
be short but informative;
must maintain sequence with the references cited in the text.
In the list of references, all the authors must be listed if there are
up to and including five authors; if there are six or more, the first three
3) the full name of each author (the editorial policy of the São
should be cited, followed by the expression “et al.” For books, the city
Paulo Medical Journal is that abbreviations for authors’ names
of publication and the name of the publishing house are mandatory.
must not be used; thus, names should either be sent complete
For texts published on the internet, the complete uniform resource
or with middle names omitted, for example: an author whose
locator (URL) or address is necessary (not only the main home page
full name is John Richard Smith can be presented as John
of a website or link), so that by copying the complete address into their
Smith or John Richard Smith, but not as John R. Smith; like-
computer internet browsers, the journal’s readers will be taken to the
wise, use Christopher Smith and not Chris Smith, or William
exact document cited, and not to a general website. The following are
Smith and not Bill Smith, and so on)), his/her academic titles
some examples of the most common types of references:
(abbreviated in English), in the order obtained (for example:
Article in journal
MD for medical doctor, MSc for holders of a master’s title,
- Hurt AC, Hardie K, Wilson NJ, et al. Community transmis-
PhD for holders of a doctorate or BSc for bachelor of sci-
sion of oseltamivir-resistant A(H1N1)pdm09 influenza.
ence, such as in biology), and the positions currently held
N Engl J Med. 2011;365(26):2541-2.
(for example, Doctoral Student, Attending Physician, Adjunct
Chapter of book
Professor, Associate Professor, Head of Department, etc.), in
- Miller WI, Achernabb JC, Fluck CE. The adrenal cortex and
the department and institution where he/she works, and the
its disorder. In: Sperling M. Pediatric endocrinology. 3rd ed.
city and country;
Elsevier Health Sciences; 2008. p. 444-511.
4) the place where the work was developed;
Text on the internet
5) the complete address (name of street or avenue, building
- Centers for Disease Control and Prevention. Children’s food
number, city) of the corresponding author, telephone and
environment State Indicator Report, 2011. Available from:
e-mail that can be published together with the article.
http://www.cdc.gov/obesity/downloads/ChildrensFoodEnvi-
Second page: abstract (English and Portuguese) and key words
The second page must include the title and an abstract (Eng-
ronment.pdf. Accessed in 2012 (Mar 7).
Last page
lish and Portuguese, maximum of 250 words each),9 structured in
The last page must contain:
five items:
1) the date and place of the event at which the paper was presented, if
1) context and objective;
2) design (type of study) and setting (place where the study was
developed);
3) methods (described in detail);
4) results; and
applicable, such as congresses or dissertation or thesis presentations;
2) sources of support in the forms of finance for the project, study
bursaries or funding for purchasing equipment or drugs. The
protocol number for the funding must be presented;
3) description of any conflicts of interest held by the authors. We
5)conclusions.
recommend that the item “Conflicts of interest” at http://www.
The abstract (both in English and in Portuguese) should contain
icmje.org should be read to obtain clarifications regarding what
five key words. The English terms must be chosen from the Medical
Subject Headings (MeSH) list of Index Medicus, which are available on
may or may not be considered to be a conflict of interest;
Figures and tables
the internet (http://www.ncbi.nlm.nih.gov/sites/entrez?db=mesh).10
Images must have good resolution (minimum of 300 DPI) and be
The Portuguese terms must be chosen from the Descritores em Ciên-
recorded in “.jpg” or “.tif ” format. Do not attach images inside Micro-
cias da Saúde (DeCS), developed by Bireme, which are available on
soft PowerPoint documents. If photographs are inserted in a Micro-
the internet (http://decs.bvs.br/).11
soft Word file, the images should also be sent separately. Graphs must
II
Sao Paulo Med J. 2014; 132(2):I-V
These instructions are updated periodically. We recommend that they are consulted online at: www.scielo.br/spmj
be prepared in Microsoft Excel (do not send them in image formats)
and must be accompanied by the tables of data from which they have
been generated. The number of illustrations must not exceed the total
number of pages minus one.
All figures and tables must contain legends or titles that precisely
describe their content and the context or sample from which the information was obtained (i.e. what the results presented are and what the
kind of sample or setting was). The legend or title sentence should be
short but comprehensible without depending on reading the article.
All the figures and tables should be cited in the text.
São Paulo Medical Journal/Evidence for Health Care is for now
published in black-and-white in its printed version. Photographs,
photomicrographs, bar and line graphs and any image to be published
must be prepared considering that there will be no color differentiation (any color information will be discarded). Shades of gray and
printing patterns (dots, stripes and others) should be used instead,
with good contrast.
Original articles
Clinical trials, cohort, case-control, prevalence, incidence, accuracy and cost-effectiveness studies, and systematic reviews with or
without meta-analysis, are considered to be original articles.
The São Paulo Medical Journal/Evidence for Health Care supports the clinical trial registration policies of the World Health
Organization (WHO) and the International Committee of Medical Journal Editors (ICMJE) and recognizes the importance of
these initiatives for registration and international dissemination
of information on randomized clinical trials, with open access.
Thus, from 2008 onwards, manuscripts on clinical trials have been
accepted for publication only if they have received an identification number from one of the clinical trial registers that have been
validated in accordance with the criteria established by WHO and
ICMJE. Authors of randomized clinical trials must thus register
their studies before submitting them for publication in the São
Paulo Medical Journal/Evidence for Health Care. The addresses
for these registers are available from the ICMJE website (http://
www.icmje.org). The identification number should be declared at
the end of the abstract.
Authors will be required to comply with the guidelines for writing
each type of original article, as follows:
1. Observational articles: STROBE Statement;5,6
2. Clinical trials: CONSORT Statement;2
3. Accuracy studies on diagnostic tests: STARD Statement;7,8
4. Systematic reviews of the literature and meta-analyses:
PRISMA4
The São Paulo Medical Journal takes the view that these guidelines not only aid in writing and organizing the content of articles in a
standardized manner, thereby improving their quality and facilitating
reading and assessment, but also these guidelines help to avoid situations in which important information on the methodology of studies
remains outside of the manuscript.
As a partner institution of the Cochrane Collaboration and the
Brazilian Cochrane Center, the Associação Paulista de Medicina considers that production of articles in accordance with these guidelines
also aids in future production of systematic reviews of the literature
and meta-analyses. Thus, articles submitted for publication that are
not in accordance with these norms may be returned to their authors
for adjustment before the peer review process begins.
Original articles must be structured so as to contain the following parts: Introduction, Objective, Methods, Results, Discussion and
Conclusion. The text must not exceed 5,000 words (excluding tables,
figures and references), from the introduction to the end of the conclusion, and must include a structured abstract with a maximum of
250 words.9 “Structured abstract” means that the abstract must contain the following items: Context and objective, Design and setting,
Method, Results and Conclusion.
The structure of the document should follow the format laid out below:
1) Title and abstract: the study design and/or the way participants were allocated to interventions, for example “randomized” or “retrospective” study, should be mentioned in the title
and in the abstract. The abstract should provide a summary of
what was done and what was found.
2)Introduction: specify the reasons for carrying out the study,
describing the present state of knowledge of the topic.
Describe the scientific background and “the state of the art”.
Do not include here any results or conclusions from the study.
Use the last paragraph to specify the principal question of the
study, and the principal hypothesis tested, if there is one. Do
not include discussions about the literature in the introduction; the introduction section should be short.
3)Objective: describe briefly what the main objective or question of the study was. Clearly describe the pre-specified
hypotheses.
4)Methods
4.1)Type of study: describe the design of the study and specify,
if appropriate, the type of randomization (the way in which
draws were conducted), the blinding (how this was ensured),
the diagnostic test standards (gold standard or range of normal values) and the time direction (retrospective or prospective). For example: “randomized clinical trial”, “double-blind
placebo-controlled clinical trial”, “cross-sectional accuracy
study”, “retrospective cohort study”, “cross-sectional prevalence study” or “systematic review of clinical trials”.
4.2)Sample, participants or patients: describe the eligibility criteria for participants (inclusion and exclusion criteria) and
the sources and procedures for selection or recruitment. In
case-control studies, describe the rationale for distributing
the subjects as cases and controls, and the matching criteria. The numbers of patients at the beginning and end of
the study (after exclusions) must be made clear. A flow diagram showing the initial recruitment, the exclusions and the
Sao Paulo Med J. 2014; 132(2):I-V
III
These instructions are updated periodically. We recommend that they are consulted online at: www.scielo.br/spmj
final sample of patients included should be produced and
inserted in the article.
4.3)Setting: indicate the place where the study was carried out,
including the type of healthcare provided (i.e. whether primary or tertiary; and whether in a private or in a public hospital). Avoid stating the name of the institution where the study
was developed (for blinding purposes in the peer review).
Only the type of institution should be made clear, for example: “public university hospital” or “private clinic”.
4.4)Procedures (intervention, diagnostic test or exposure):
describe the principal characteristics of any intervention,
including the method, the timing and the duration of its
administration or of data collection. Describe the differences
in interventions administered to each group (if the study is
controlled). Detail the procedures in such a way that other
researchers will be able to repeat them in other localities.
4.5)Main measurements, variables and outcome: state what the
primary and secondary outcomes analyzed in the study are.
Describe the method of measuring the primary result, in
the way in which it was planned before data collection. For
each variable of interest, detail the assessment methods. If
the hypothesis of the study was formulated during or after
data collection (and not before), this needs to be declared.
Describe the methods used to enhance the quality of measurements (for example, multiple observers, training, etc.) and
to avoid bias. Explain how quantitative variables were handled
in the analyses.
4.6)Sample size and statistical analysis: describe the sample size
calculation method, or the study period in the event that
patients were consecutively admitted over a period. Readers need to understand why a given number of patients was
used. The planned statistical analysis, the statistical tests used
and their significance levels, along with any post hoc analyses, should be presented in this section. Describe the methods used to control for confounding factors and variables, and
explain how missing data and cases lost from the follow-up
were dealt with.
4.7)Randomization: describe the method used to implement the
random allocation sequence (for example, sealed envelopes
containing random sequences of numbers or software for
generating random numbers). If appropriate, report that the
study used “quasi-randomization”.12 In addition, describe who
generated the random sequence, who assigned the participants to each group (in the case of controlled trials) and who
recruited the participants.
5)Results: describe the main findings. If possible, these should be
accompanied by their 95% confidence intervals and the exact
level of statistical significance (it is not enough to write “P <
0.05”: the exact P value should be supplied). For comparative
IV
Sao Paulo Med J. 2014; 132(2):I-V
studies, the confidence interval must be stated for the differences between the groups.
5.1)Participant flow diagram: describe the flow of participants
through each stage of the study (inclusions and exclusions)
and the follow-up period, and the number of participants
completing the study (or lost from the follow-up). Use a flow
diagram to demonstrate the numbers of patients, from the
initial recruitment to the end of the study, and the reasons
for exclusions. If there was any “intention-to-treat” analysis,
describe it.
5.2)Deviations: if there was any deviation from the protocol, away
from what was initially planned, describe it and the reasons
for it.
5.3)Adverse events: describe any side effect, adverse event or
complication.
6)Discussion: provide an interpretation of the results, taking
into account the study hypotheses and conclusions. Emphasize the new and important factors encountered in the study,
which will form part of the conclusion. Do not repeat data
presented in the introduction or results in detail. Mention any
limitations of the findings that should be noted and any possible implications for future research. Describe any potential bias. Report any relevant findings from other studies: it is
important to review the recent literature to seek new evidence
that may have been published, which needs to be discussed.
State whether the findings can be generalized to populations
(i.e. whether the findings have external validity). It is recommended that the last two paragraphs should contain implications for practice and for further research.
7)Conclusions: specify only the conclusions that can be sustained by the results, together with their clinical significance
(avoiding excessive generalization). Draw conclusions based
on the objectives and hypotheses of the study. The same
emphasis should be placed on studies with positive and negative results.
Systematic reviews with or without meta-analyses should comply
with the same publication norms established for original articles, and
be produced in accordance with PRISMA4 and the Cochrane Collaboration’s systematic review Handbook.13 The text should not exceed
5,000 words (excluding tables, figures and references)
Short communications, case reports or case series
Short communications and case reports must be limited to 3,000
words (from the introduction to the end of the conclusion). Short
communications are reports on the results from ongoing studies or
studies that have recently been concluded for which urgent publication is important. They should be structured thus: Introduction,
Objective, Methods, Results, Discussion and Conclusion, like in original articles. Individual case reports should contain: Introduction, Case
Report, Discussion and Conclusion. Reports on case series constitute
These instructions are updated periodically. We recommend that they are consulted online at: www.scielo.br/spmj
observational studies and these should be structured in accordance
with the norms of the STROBE Statement.5
Both short communications and case reports must be submitted with
abstracts and key words. The abstracts in short communications should
be structured with: Context and objective, Design and setting, Methods, Results and Conclusion, like in original articles. The abstracts in
case reports and case series should contain: Context, Case Report (with a
description of the case and a pertinent discussion) and Conclusion.
The São Paulo Medical Journal/Evidence for Health Care is interested in publishing rare or instructive case reports, accompanied by
a systematic search of the literature, in which relevant studies found
(based on their level of evidence) are presented and discussed.14 The
results from the systematic search of the main databases — Medline
(via PubMed), Embase, Lilacs and Cochrane Library — should be
presented in a table with the search strategy for each database and the
number of articles obtained.
Narrative reviews
Narrative reviews may be accepted by the São Paulo Medical Journal/Evidence for Health Care and should be structured with: Introduction, Objectives, Methods, Results, Discussion and Conclusions.
The abstract must be structured with: Context and objective, Design
and setting, Methods, Results and Conclusions, like in original articles. The manuscript must comply with the norms of the Vancouver
style1 and must include a systematic search in the main databases:
Medline, Embase, Lilacs and Cochrane Library. The search strategy for each database and the number of articles obtained from each
database should be presented in a table. The access route to the electronic databases used should be stated (for example, PubMed, OVID,
Elsevier or Bireme). For the search strategies, MeSH terms must be
use for Medline, LILACS and Cochrane Library. DeCS terms must be
used for LILACS. EMTREE terms must be used for Embase. Also, for
LILACS, search strategy must be performed, at the same time, with
English (MeSH), Spanish (DeCS) and Portuguese (DeCS) terms. The
search strategies must be presented exactly as they were used during the search, including parentheses, quotation marks and Boolean
operators (AND, OR, AND NOT).
Letters to the editor
Letters to the editor may address articles published in the São
Paulo Medical Journal/Evidence for Health Care publication or may
deal with health issues of interest. Case reports must not be submitted
as letters. In the category of letters to the editor, the text has a free format, but must not exceed 500 words and five references.
Documents cited
1. Internal Committee of Medical Journal Editors. Uniform
requirements for manuscripts submitted to biomedical journals,
writing and editing for biomedical publications. Available from:
http://www.icmje.org. Accessed in 2012 (Aug 6).
2. The CONSORT Statement. Available from: http://www.consort-statement.org/consort-statement/. Accessed in 2012 (Aug 6).
3. Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomised controlled trials: the
QUOROM statement. Lancet. 1999;354(9193):1896-900. Available
from:
http://www.thelancet.com/journals/lancet/article/PIIS01406736(99)04149-5/abstract. Accessed in 2012 (Aug 6).
4. PRISMA. Transparent Reporting of Systematic Reviews and
Meta-Analyses. Available from: http://www.prisma-statement.org/
index.htm. Accessed in 2012 (Aug 6).
5. STROBE Statement. Strengthening the reporting of observational studies in epidemiology. What is strobe? Available from: http://
www.strobe-statement.org/. Accessed in 2012 (Aug 6).
6. von Elm E, Altman DG, Egger M, et al. The Strengthening the
Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-9.
7. STARD Statement. STAndards for the Reporting of Diagnostic accuracy studies. Available from: http://www.stard-statement.org/.
Accessed in 2012 (Aug 6).
8. Rennie D. Improving reports of studies of diagnostic tests: the
STARD initiative. JAMA. 2003;289(1):89-90.
9. Haynes RB, Mulrow CD, Huth EJ, Altman DG,
Gardner MJ. More informative abstracts revisited. Ann Intern Med.
1990;113(1):69-76.
10. National Library of Medicine. Medical Subject Headings:
annotated alphabetic list. Bethesda: NLM; 1998. Available from: http://
www.ncbi.nlm.nih.gov:80/entrez/query.fcgi?db=mesh. Accessed in
2012 (Aug 6).
11. BVS Biblioteca Virtual em Saúde. Descritores em Ciências
da Saúde. Available from: http://decs.bvs.br/. Accessed in 2012
(Aug 6).
12. Reeves BC, Deeks JJ, Higgins JPT, Wells GA. Including nonrandomized studies. In: Cochrane Non-Randomised Studies Methods Group. The Cochrane Book Series. England: John Wiley & Sons;
2008. Available from: http://hiv.cochrane.org/sites/hiv.cochrane.org/
files/uploads/Ch13_NRS.pdf. Accessed in 2012 (Aug 6).
13. The Cochrane Collaboration. Cochrane Handbook for Systematic Reviews of Interventions. Available from: http://www.cochrane.
org/training/cochrane-handbook/. Accessed in 2012 (Aug 6).
14. Phillips B, Ball C, Sackett D, et al. Oxford Centre for Evidencebased Medicine Levels of Evidence (May 2001). Available from: http://
www.cebm.net/index.aspx?o=1047. Accessed in 2012 (Aug 6).
Sao Paulo Med J. 2014; 132(2):I-V
V