RESUMEN
OBJECTIVE: Our objective was to evaluate whether microalbuminuria predicts myocardial infarction (MI) in a Mexican population. METHODS: The study was a prospective, population-based cohort. Baseline examination was carried out in 1989; the first follow-up in 1993 and the second in 1997. All men and non-pregnant women between 35 and 64 years of age at the start of the study were considered eligible. Clinical, anthropometric, and laboratory characteristics were evaluated. All patients with macroalbuminuria at baseline were excluded from the present analyses, as were all prevalent cases with MI. Remaining patients were classified as with or without microalbuminuria. Incident cases of MI were identified during follow-up phases using an electrocardiogram (according to the Minnesota Code) or the death certificate (in which underlying cause of death was listed as MI, Causes of Death codes 410.0-410.9). Results. From 2196 individuals, 1586 satisfied the inclusion criteria. Two hundred fifteen (13.6%) had microalbuminuria, and 1371 (86.4%) did not. During follow-up, 10 patients with microalbuminuria and 31 patients without microalbuminuria developed an MI. Using robust logistic regression, the probability of developing MI, adjusting by Framingham score, was estimated to be 1.90 (95% CI,.97-3.72) times higher in patients with microalbuminuria as compared with patients without microalbuminuria. CONCLUSION: We found that in a Mexican population the relationship between microalbuminuria and incidence of MI was borderline statistically significant after adjusting for other cardiovascular risk factors.