The outcome of 143 male hypertensive patients, investigated in the period 1972-1982, was ascertained in 1989 to determine if brachial artery pressure measured during a progressive graded exercise test on the bicycle ergometer is a better predictor of mortality and cardiovascular events than pressure at rest. During the total follow-up time of 1,573 patient years, 27 patients suffered at least one fatal or nonfatal cardiovascular event and 13 patients died. Using the Cox regression model, the age-adjusted relative hazard rates of systolic pressure at supine rest, at 50 W, at peak work load, and at 50% of peak exercise capacity were significant for total mortality (p less than or equal to 0.01) and for cardiovascular events (p less than or equal to 0.03). Pressure during exercise, however, did not significantly (p = 0.11-0.97) predict the outcome of the patients when age and pressure at rest were taken into account. The results were similar for diastolic pressure. In conclusion, intra-arterial pressures at rest and during submaximal and peak exercise significantly predict mortality and the incidence of cardiovascular events in hypertensive men, independent of age. However, there is no additional prognostic precision of the exercise pressures when age and the rest pressure are taken into account.