Relationship of blood pressure to coronary and stroke morbidity and mortality in clinical trials and epidemiological studies.

We compared pooled mortality and morbidity results from nine randomized trials of antihypertensive drug treatment (total n = 43,139), in which diastolic blood pressure (DBP) was reduced by a mean of 5.8 mmHg, with mortality and morbidity differences associated with the same DBP difference in two large prospective, observational studies [the follow-up study of the Multiple Risk Factor Intervention Trial (MRFIT) screenees and the Framingham Study]. These results indicate that the effect on stroke of BP reduction in the trials (-36%, 95% confidence interval +/- 10%) was about 80% of that estimated from the epidemiological studies, suggesting that the effects of long-term BP elevation on the cerebral vasculature are mostly reversible over 5-6 years of BP reduction. For coronary heart disease (CHD), the effect of BP reduction in the trials (-9 +/- 10%) was about 36% of that estimated from the epidemiological studies. This difference may reflect chronic processes involved in the pathophysiological association of BP with CHD; however, selective treatment of control patients at high risk of CHD and an adverse effect of diuretics on serum cholesterol may have influenced the effect of treatment in the trials.