Influence of systolic and diastolic blood pressure on stroke risk: a prospective observational study.

The purpose of this study was to estimate the influence of systolic (SBP) and diastolic blood pressure (DBP) on stroke risk. The Copenhagen City Heart Study is a prospective survey of 19,698 women and men who were invited to two cardiovascular examinations at 5-year intervals. Blood pressure was measured in participants once at each examination, together with other variables. Initial cases of stroke and transient ischemic attack were recorded from hospital records and death certificates from 1976 through 1988. When entered separately in the Cox regression model, both SBP and DBP had significant effects on stroke risk. In the lower 60% of the blood pressure distribution in the population, the relative risk of stroke was nearly constant, followed by a gradual increase in the upper 40% of blood pressure distribution. However, when SBP and DBP were entered simultaneously in the model, the effect of DBP vanished, while the pattern of the association between SBP and stroke risk remained unchanged. Persons on antihypertensive treatment had higher risk for stroke than non-treated persons with the same blood pressure, relative risk = 1.6 (95% confidence interval (CI) 1.2-2.2). The relative risk for the highest SBP levels, shared by nearly 3% of the population, was 4.0 (95% CI 2.2-7.3). The attributable risk of SBP in the upper 40% of SBP distribution, i.e., above the mean for each age and sex group, was 22%. Our results indicate that: 1) the association between blood pressure and stroke risk was not log-linear, and 2) SBP was a stronger stroke predictor than DBP.