RESUMO
BACKGROUND: Risk factors for cardiovascular mortality have barely been investigated in very elderly persons and there may be differences compared with younger individuals. METHODS: This is a cohort study of all inhabitants over 80 years of age in the city of Veranópolis, Brazil. The association of demographic, anthropometric, physical, and medical characteristics with mortality by any cause and by cardiovascular disease (CVD) was investigated by means of Cox regression models. RESULTS: The mean age of the participants was 83.6 ± 3.3 years. Vital status and cause of death was ascertained in 96.9% of the participants after a mean follow-up of 8.7 ± 3.8 years. Systolic and diastolic blood pressure showed a U-shape relationship with cardiovascular and total mortality. Blood pressure lower than 140/90 mmHg was associated with a higher risk for cardiovascular mortality (HR 4.76, 95% CI 1.56-14.28, p = 0.006). Duration of sleep was inversely associated with the risk of cardiovascular death (HR 0.83, 95% CI 0.73-0.95, p = 0.007), while apoA-I was inversely associated only with the risk of all-cause mortality (HR 0.99, 95% CI 0.98-1.00, p = 0.041). Anthropometric indexes, smoking, cholesterol, LDL-cholesterol, HDL-cholesterol, and other traditional risk factors were not associated with cardiovascular mortality. CONCLUSION: Many traditional risk factors are not associated with cardiovascular mortality in the very elderly. Longer sleep duration is associated with lower cardiovascular mortality of very elderly individuals, while low blood pressure identifies very elderly individuals at higher risk of dying from cardiovascular causes.