FUNDAMENTO: Para discriminar risco coronariano elevado, indicadores de obesidade central são melhores do que o Índice de Massa Corpórea (IMC), que é ainda o índice antropométrico (IA) mais utilizado para seguimento após intervenção coronariana percutânea (ICP). OBJETIVO: Reconhecer, entre os índices antropométricos (IA), os que melhor se correlacionam com ocorrência de desfechos após intervenção coronariana percutânea (ICP). MÉTODOS: Foram considerados 308 pacientes (p), idade média de 61,92±11,06 anos, 60,7% do sexo masculino, submetidos a ICP com stent. Após seis meses, pesquisaram-se os desfechos: óbito, reintervenção por ICP ou cirurgia cardíaca, exame não-invasivo alterado por isquemia ou sintomas anginosos. Os p foram divididos em: Grupo 1 (com desfechos, n=91; 29,5%) e Grupo 2 (sem desfechos, n=217; 70,45%). No sexo masculino e feminino, os IA estudados e seus respectivos pontos de corte foram: circunferência abdominal (CA) > 90/80 cm, relação cintura-quadril (RCQ) > 0,90/0,80cm, índice de conicidade (IC) >1,25/1,18 e índice de massa corpórea (IMC) >30 para ambos os sexos. RESULTADOS: Os grupos diferiram quanto à maior ocorrência de histórico familiar e infarto prévio no Grupo 2. No sexo masculino, CA > 90 cm (p=0,0498) foi, em análise multivariada, preditor independente de desfechos. IMC não foi preditor de eventos. No Grupo 1, a probabilidade de ocorrência de IMC alterada é significativamente menor do que a ocorrência dos outros IA estudados (p<0,0001). CONCLUSÃO: CA anormal comportou-se como preditor independente de ocorrência de desfechos no sexo masculino dessa população pós-ICP. IMC elevado não foi preditor de desfechos e foi o índice antropométrico menos prevalente em pacientes com eventos.
ARTIGO ORIGINAL
Hospital Universitário Evangélico de Curitiba, Curitiba, PR - Brasil
Desde 1983, quando foram publicados os resultados do estudo de Framingham relacionados à obesidade, observa-se forte correlação desse fator de risco com doença arterial coronariana1. No Brasil, a prevalência de obesidade é de cerca de 8% para os homens, e de 12,4% para as mulheres2. O somatório de sobrepeso e obesidade atinge cifras em torno de 38,5% e 39%, para cada sexo. Nos Estados Unidos, esse índice é de aproximadamente 30,5% e 64,5%, ou seja, mais da metade da população norte-americana apresenta sobrepeso ou obesidade3.
Utilizando como marcador de obesidade o IMC, importantes estudos epidemiológicos mostraram que obesidade está associada com morbidade e mortalidade cardiovascular1,4. A associação da obesidade com fatores de risco convencionais, como hipertensão arterial e diabete melito, e com disfunção endotelial, resistência insulínica e inflamação pode contribuir para o aumento do risco de desfechos desfavoráveis após intervenção coronária percutânea (ICP) em obesos5.
Apesar das evidências de risco conferidas pelo IMC elevado, há relatos na literatura da existência de proteção paradoxal da obesidade em pacientes submetidos a ICP6-9. Assim, nessas publicações, pacientes com IMC elevado apresentaram menor incidência de eventos cardíacos em 1 ano6,7, menor risco de eventos intra-hospitalares8, e menor mortalidade após o procedimento9.
Embora o IMC seja uma medida simples, conveniente e até agora válida para estudo da obesidade, medidas de obesidade central (principalmente circunferência abdominal e relação cintura-quadril alteradas)10 têm se mostrado mais relacionadas tanto com risco coronariano elevado11,12 quanto com infarto agudo do miocárdio13,14. Apesar disso, IMC é o índice antropométrico até agora mais utilizado na cardiologia intervencionista para seguimento clínico após intervenção coronariana.
O objetivo deste estudo é reconhecer, entre os índices antropométricos de obesidade, os que melhor se correlacionam com ocorrência de desfechos após ICP.
Amostra composta por 308 pacientes consecutivos, idade média de 61,92±11,06 anos (variando de 34 a 88 anos), sendo 60,7% do sexo masculino, submetidos, no período de maio de 2005 a setembro de 2006, a ICP com
As ICP foram realizadas via artéria femoral com a técnica padrão baseada em consenso atual15. Os pacientes foram pré-tratados com dupla inibição plaquetária, tendo recebido AAS 100 mg e ticlopidina 250 mg, duas vezes ao dia, de início 48 horas antes, ou clopidogrel 75 mg (pelo menos iniciado 24 horas antes do procedimento, uma vez ao dia ou em dose de ataque de 300 mg, seis horas antes do procedimento), conduta mantida por 30 dias após a intervenção. Nenhum paciente recebeu inibidores da glicoproteína IIb IIIa. O sucesso de procedimento foi definido como a obtenção de estenose residual < 30%, sem ocorrência de eventos clínicos maiores (óbito, infarto agudo do miocárdio [IAM] ou necessidade de cirurgia de emergência), durante a fase hospitalar.
Após período de seis meses, os pacientes foram contatados, na busca dos seguintes desfechos: óbito, reintervenção por ICP ou cirurgia cardíaca vaso-relacionada, exame não-invasivo alterado por isquemia, infarto agudo do miocárdio ou recorrência de sintomas anginosos16,17. Considerou-se como desfecho o evento mais grave relatado. Correlacionou-se a ocorrência de desfechos desfavoráveis com os índices antropométricos obtidos. No sexo masculino e feminino, os índices antropométricos estudados e seus respectivos pontos de corte foram: circunferência abdominal > 90/80 cm, relação cintura-quadril >0,90/0,80 cm18, índice de conicidade >1,25/1,1811 e índice de massa corpórea (IMC)
Quanto à análise estatística, os grupos foram comparados entre si utilizando-se, para variáveis dicotômicas, teste exato de Fisher, e para variáveis contínuas, teste
Um total de 308 pacientes que realizaram implante de
Observa-se diferença estatisticamente significativa entre os grupos nos ítens história familiar de coronariopatia e infarto prévio, mais prevalentes no Grupo 2 (p=0,02 e 0,03, respectivamente). As demais características clínicas e angiográficas foram semelhantes entre os grupos. A taxa de sucesso imediato, estenose residual média final e o índice de complicações foram similares entre os grupos.
A análise estatística univariada e multivariada (nesta última foram incluídos todos os parâmetros antropométricos alterados disponíveis, com exceção de relação cintura-quadril) da ocorrência de índices antropométricos alterados e desfechos, na população geral e nas mulheres, não demonstrou influência independentemente das variáveis estudadas sobre a ocorrência de eventos. Já nos homens, circunferência abdominal >90 em análise multivariada (p=0,0498) esteve independentemente relacionada com ocorrência de desfechos. IMC não foi preditor de eventos em nenhum dos sexos (
Apesar de a amostra ser composta de pacientes, na sua maioria, com índices antropométricos anormais, verificou-se a ordem de ocorrência de cada medida antropométrica alterada no grupo com desfechos. Dessa forma, no Grupo 1, a probabilidade da ocorrência de circunferência abdominal (61,36%), relação cintura-quadril (94,48%) e índice de conicidade (64,29%) anormais é significativamente diferente da probabilidade de anormalidade de IMC (26,30%), com p<0,0001 para todas as combinações (
Neste estudo de pacientes submetidos a ICP com
Vários trabalhos publicados evidenciam a existência de um efeito protetor da obesidade (de acordo com IMC), nos pacientes submetidos a ICP. Assim, Gruberg e cols.6 relataram menor quantidade de eventos cardíacos em obesos, proteção que permaneceu por até um ano após ICP. Ellis e cols.20, antes do advento dos
Há, no entanto, controvérsia em relação ao assunto. Alguns estudos não evidenciaram efeito protetor da obesidade após intervenção coronariana: dados a partir do estudo TAXUS-IV, com controle angiográfico após ICP com implante de
Muito embora índices antropométricos de obesidade central tenham melhor correlação com eventos coronarianos14, risco coronariano elevado11 e IAM13, não há estudos que os correlacionem com evolução após ICP. Na verdade, apesar de o IMC ser uma medida simples e conveniente para o diagnóstico de obesidade, é cada vez mais evidente a importância do efeito deletério da obesidade abdominal sobre doença arterial coronariana10. Comparados ao IMC, a circunferência abdominal, o índice de conicidade e a relação cintura-quadril têm se mostrado superiores para identificar adiposidade visceral e, por conseguinte, desordens metabólicas e risco cardiovascular12. Num estudo de fatores de risco na cidade de São Paulo e no INTERHEART, relação cintura-quadril elevada (tercil intermediário
Todos esses recentes achados sobre a importância da obesidade central (circunferência abdominal, relação cintura-quadril e índice de conicidade) em detrimento das medidas de obesidade generalizada (IMC), somados com os resultados deste estudo, nos levam a questionar a real existência do paradoxo da obesidade, achado que se baseou apenas no cálculo do IMC. Pesquisas adicionais – sobretudo com número maior de indivíduos alocados – acerca da influência dos índices antropométricos no cenário da cardiologia intervencionista devem ser realizadas para conclusões definitivas sobre o tema.
Por se tratar de uma amostra populacional do mundo real, composta de pacientes consecutivos submetidos a ICP, a maioria das mulheres apresentou obesidade central, o que pode explicar a não-significância estatística nesse grupo. Por ter ocorrido elevado porcentual de anormalidade (94,48%) da relação cintura-quadril na população geral, com ausência de casos no grupo das mulheres de relação cintura-quadril
Circunferência abdominal anormal, índice antropométrico que reflete obesidade central, comportou-se como preditor independente de ocorrência de desfechos no sexo masculino dessa população pós ICP. IMC elevado não foi preditor de desfechos em nenhum dos sexos e foi o índice antropométrico menos prevalente em pacientes com desfechos, apesar de ainda ser o mais utilizado na literatura mundial.
Declaro não haver conflito de interesses pertinentes.
O presente estudo não teve fontes de financiamento externas.
Não há vinculação deste estudo a programas de pós-graduação.
Correspondência: José Carlos Estival Tarastchuk Rua Padre Anchieta 1721/191, Bigorrilho 80730-000, Curitiba, PR - Brasil E-mail:
Artigo recebido em 03/05/2007; revisado recebido em 08/01/2008; aceito em 15/01/2008.
BACKGROUND: Central anthropometric indexes are better than the body mass index to discriminate elevated coronary risk. However, the Body Mass Index (BMI) is still the most frequently studied anthropometric index on outcomes of patients undergoing percutaneous coronary angioplasty (PCI). OBJECTIVE: To recognize, among several anthropometric indexes of obesity, which one best discriminates MACE (Major Adverse Cardiac Events) after PCI. METHODS: Subjects were 308 patients (mean age 61.92±11.06 years, 60.7% of them men) who had undergone successful coronary angioplasties. Six months after the procedure, patients were contacted for clinical follow-up. Major Adverse Cardiac Events included death, acute myocardial infarction, cardiac surgery, reintervention, angina, or evidence of myocardial ischemia on a non-invasive test. Patients were divided into 2 groups: Group 1 (with MACE, n=91, 29.5%), Group 2 (with no MACE, n= 217; 70.45%). For men and women, the anthropometric indexes studied and their respective cut-off points were waist circumference >90/80 cm, Waist-Hip Ratio > 0.90/0.80cm, Conicity Index > 1.25/1.18, and Body Mass Index > 30. RESULTS: There were more cases of familial history and previous infarct in Group 2. For men, waist circumference >90cm (p=0.0498) in multivariate analyses was an independent predictor of MACE. BMI was not related to MACE. In Group 1, the prevalence of an elevated BMI was significantly different compared to the other anthropometric indexes studied (p<0.0001). CONCLUSION: Waist circumference was an independent predictor of MACE in men. Body Mass Index was not related to MACE and was the least frequent anthropometric index in the MACE group.
Hospital Universitário Evangélico de Curitiba, Curitiba, PR - Brazil
Mailing address
Since 1983 when the results of the Framingham Study related to obesity were published, a strong correlation between this risk factor and coronary artery disease has been observed1. In Brazil, the prevalence of obesity is approximately 8% for men and 12.4% for women2. The joint effect of overweight and obesity reaches figures of around 38.5% and 39% for men and women, respectively. In the United States, this rate is approximately 30.5% and 64.5%, for men and women, respectively i.e., more than half of the North American population is overweight or obese3.
Using the BMI as a marker for obesity, significant epidemiological studies have shown that obesity is associated with cardiovascular morbidity and mortality1,4. The association of obesity with conventional risk factors, such as arterial hypertension and diabetes mellitus, and with endothelial dysfunction, insulin resistance, and inflammation may contribute to the increase in untoward outcomes after Percutaneous Coronary Interventions (PCI) in obese individuals5.
Despite evidence of risk conferred by an elevated BMI, there are reports in literature of a paradoxal protection afforded by obesity in patients submitted to PCI6-9. Thus, in these publications, patients with high BMIs showed lower rates of cardiac events in one year6,7, lower risks of intrahospital events8, and lower mortality rates after the procedure9.
Although the BMI is a simple and convenient measurement considered valid up until now for the study of obesity, measurements of central obesity (primarily Waist Circumference and altered Waist-Hip Ratios)10 have proved to be more closely related both to elevated coronary risks11,12 and to acute myocardial infarcts13,14. Even so, to date, BMI is the anthropometric index most used in interventional cardiology for clinical follow-up after a coronary intervention.
The objective of this study is to recognize, among several anthropometric indices of obesity, those that best correlate with the occurrence of post-PCI outcomes.
The sample is composed of 308 consecutive patients, mean age 61.92±11.06 years (varying from 34 to 88 years), 60.7% of them male, successfully submitted to PCI with a conventional stent during the period from May 2005 to September 2006. For all study subjects, a routine medical history was taken with the collection of information on risk factors for coronary disease. Patients were submitted to physical examinations, with weight and height measurements. The BMI was calculated by the weight in kilograms divided by the height in square meters (kg/m2). Obesity was considered when BMI
CPIs were carried out via femoral artery using a technique chosen by current consensus15. Patients were pretreated with double platelet inhibition, receiving aspirin 100mg and ticlopidine 250mg, twice a day, initiated 48 hours before, or clopidogrel 75mg (initiated at least 24 hours before the procedure, once a day, or with a loading dose of 300mg 6 hours before the procedure), and this approach was maintained for 30 days after the intervention. No patient received glycoprotein IIb/IIIa inhibitors. Success in the procedure was defined as the achievement of residual stenosis < 30%, with no occurrence of a significant clinical event (death, acute myocardial infarct (AMI), or need for emergency surgery), during the hospital phase.
After a 6-month period, patients were contacted in search of the following outcomes: death, reintervention with PCI or vessel-related heart surgery, non-invasive test altered by ischemia, acute myocardial infarct, or recurrence of anginous symptoms16,17. Outcomes were considered the most serious events reported. The occurrence of unfavorable outcomes was correlated with the anthropometric indices obtained. In men and women, the anthropometric indices studied and their respective cut-off points for both genders were Waist Circumference > 90/80cm, Waist-Hip Ratio >0.90/0.80cm18, Conicity Index >1.25/1.1811, and Body Mass Index (BMI)
As to statistical analyses, the groups were compared with the use of dichotomic variables, Fisher's Exact Test, and, for continuous variables, Student's t Test (independent samples), except for the "initial lesion and residual lesion" variables which were analyzed with the Mann-Whitney non-parametric test. To study dichotomized anthropometric variables, Fisher's Exact Test and Wald's Test were used. In order to verify the prevalence of abnormality of each anthropometric index studied, the Binomial Test was used. All anthropometric variables studied were submitted to univariate and multivariate analyses. P values less than 5% (p<0.05) were considered significant.
Three hundred and eight patients who successfully received metallic stent implants between May 2005 and September 2006 were studied. According to the outcomes, they were divided into Group 1 (with outcomes, n=91; 29.5%) and Group 2 (no outcomes, n=217; 70.45%).
A statistically significant difference was observed between the groups for the items familial history of coronary disease and prior infarct, more prevalent in Group 2 (p=0.02 and 0.03, respectively). The other clinical and angiographic characteristics were similar between the groups. The rates of immediate success, mean final stenosis, and complications were similar between the two groups.
No independent influence was shown of the variables studied on the occurrence of events in the univariate and multivariate statistical analyses (the latter included all altered anthropometric parameters available, except the Waist-Hip Ratio) of the occurrence of altered anthropometric indices and outcomes in the general population and in women. In men, on the other hand, the Waist Circumference >90cm in the multivariate analysis (p=0.0498) was independently related to the occurrence of outcomes. The BMI was not a predictor of events in either gender (
Despite the fact that most of the patients in the sample had abnormal anthropometric indices, we noted the order of occurrence of each altered anthropometric measurement in the group with outcomes. In this way, in Group 1, the probability of the occurrence of abnormal Waist Circumferences (61.36%), Waist-Hip Ratios (94.48%), and Conicity Indices (64.29%) is significantly different from the probability of BMI abnormalities (26.30%), with p<0.0001 for all combinations (
In this study of patients submitted to PCI with a conventional stent, it was noted that the altered waist circumference stands out as an independent predictor for the occurrence of late outcomes in the male sub-group. The BMI was not associated with a better or worse clinical progress after PCI.
Various studies published have shown the existence of a protective effect of obesity (according to the BMI) in patients submitted to PCI. Gruberg et al6 reported a lower number of cardiac events in obese individuals, a protection that remained for up to one year after the PCI. Ellis et al20, before the advent of stents, observed poorer intrahospital clinical developments only in patients with extreme BMIs (equal to or lower than 25 and greater than 35), and this finding is a predictor of death after PCI20. There are reports of greater risks of vascular complications in patients with extreme BMIs when compared to moderate obesity, which supports the existence of a paradox of obesity21. In studying patients submitted to the primary PCI in the Cadillac study, Nikolski et al9 observed a correlation between elevated BMIs and lower mortality9. In evaluating the relationship between the BMI and one-year clinical progress after PCI, Kelly et al7 observed that the BMI was associated with greater efficacy and less bleeding after PCI. In this sample, there was an extra advantage for patients with high BMIs randomized to receive clopidogrel7. The so-called "obesity paradox" has also been observed in patients after heart surgery22 and in those with congestive heart failure23. The BARI study reported a better short-term clinical progress in obese individuals in the PCI arm, and the BMI showed no association with the 5-year mortality rate8. Presently, there is no plausible explanation for the paradoxal protection of obesity. A greater coronary diameter in the obese, with a smaller chance for restenosis8, and the influence of age, which may be lower in the obese patients studied, may have influenced the superior clinical course24; even excessive anticoagulation in patients with higher BMIs6 is a possible mechanism to justify these findings in medical literature. More recently, Rubinstein et al25 studied the severity of coronary artery disease in obese subjects submitted to PCI and concluded that the obese showed a smaller prevalence of serious coronary lesions, which might also explain the better clinical progress in this group of patients.
There is controversy, however, regarding this issue. Some studies have not shown the protective effect of obesity after coronary intervention: data from the TAXUS-IV study, with angiographic control post-PCI with metallic stent implantation, showed a worse clinical progress in the obese when compared to patients with normal BMIs26. A sub-analysis of the ARTS study27, in observing outcomes three years after PCI or cardiac surgery, failed to show any association between the BMI and significant cardiac events. Rana et al24 did not note any relationship between the metabolic syndrome or any of its components (among them, an elevated BMI) and a smaller occurrence of post-PCI outcomes. Poston et al28, in evaluating the impact of obesity after PCI (quality of life or health status after 12 months in one large cohort of patients with 1631 individuals), perceived that there was no long-term difference in health status, quality of life, need for repeated procedures, or survival in the different classifications of BMI28. This author, when studying 903 patients after angioplasty with conventional stents, found no protection afforded by obesity against outcomes 6 months after the procedure. In the normal BMI, overweight, and obese groups, the occurrence of untoward events did not differ significantly29.
Although anthropometric indices of central obesity bear a better correlation with coronary events14, high coronary risk11, and AMI13, there are no studies correlating them with the clinical progress after PCI. In fact, despite the fact that BMI is a simple and convenient measurement for the diagnosis of obesity, the importance of the deleterious effect of abdominal obesity on coronary artery disease is increasingly more evident10. Compared to the BMI, the Waist Circumference, Conicity Index, and Waist-Hip Ratio have proved to be superior for identifying visceral adiposity and, consequently, metabolic disorders and cardiovascular risk12. In a study on risk factors in the city of São Paulo and in INTERHEART, a high Waist-Hip Ratio (intermediate tercile versus the inferior tercile) was independently associated with AMI, which did not occur with an altered BMI13,14.
All these recent findings on the importance of central obesity (Waist Circumference, Waist-Hip Ratio, and Conicity Index) in detriment of generalized obesity measurements (BMI), added to the results of this study, lead us to question the true existence of the paradox of obesity, a finding based merely on the calculation of the BMI. Additional research - especially with a greater number of individuals allocated - on the influence of anthropometric indices on the interventional cardiology scene should be carried out for definitive conclusions on the theme.
Abnormal Waist Circumference, an anthropometric index that reflects central obesity, behaved as an independent predictor of the occurrence of outcomes in the male individuals of this population post-PCI. An elevated BMI was not a predictor of outcomes in either gender, and was the least prevalent anthropometric index in patients with outcomes, despite being the most frequently used in worldwide literature.
No potential conflict of interest relevant to this article was reported.
There were no external funding sources for this study.
This study is not associated with any graduation program.