RESUMO
A hospital-based case-control study was undertaken at Prasat Neurological Hospital and Institute, Bangkok, in 1990 to determine the association between blood lipid variables (i.e., total cholesterol and triglycerides) and cerebral infarction. To alleviate the unwanted effects of sample selection a number of inclusion and exclusion criteria were also made. A total of 193 newly diagnosed cerebral infarction patients together with an equal number of age-sex-matched controls, male and female 45 to 84 years of age, were studied. The association was assessed by logistic regression analysis. Other potential risk factors for cerebral infarction were entered in the logistic models as covariates to be adjusted. The results showed that, total cholesterol and triglycerides were statistically significant related to cerebral infarction (p < 0.0001) in univariated analysis. After adjusting for the covariates through multivariate logistic regression analyses, total cholesterol remained significantly related to cerebral infarction (p = 0.0430) but triglycerides was not significant (p = 0.3953). Other factors, namely, hypertension, cardiac impairment, diabetes mellitus and hemoglobin concentration were strongly related to cerebral infarction in both the univariate and multivariate analyses. The interaction between blood lipids variables and potential risk factors was also assessed. The finding showed significant multiplicative interaction between triglyceride levels and diabetes mellitus status (p = 0.0026). Other interaction terms did not approach statistical significance.