The short- and long-term hemodynamic effects of vasodilators in patients with primary pulmonary hypertension have been studied, but whether they affect survival is unknown. We measured the short-term response to nifedipine and hydralazine in 23 patients with primary pulmonary hypertension and followed their clinical course over 2 years. A favorable drug response, defined as a fall in the pulmonary vascular resistance of 20% or greater, occurred in 18 patients (78%). Half of the patients who exhibited a favorable short-term response were treated with long-term vasodilator therapy. Their clinical course was compared with that of responders who were not treated and with that of the nonresponders. Of the responders who were treated, two improved, four had no change, and three died; of the responders who were not treated, one improved, three had no change, and five died. Using stepwise Cox regression, we evaluated age, sex, functional class on entry, pulmonary arterial pressure, pulmonary vascular resistance, and short-term drug response as predictors of survival and found only functional class and a favorable short-term drug response to be significant predictors (p less than .01); however, there was no difference in survival between the responders who were treated and those who were not. We conclude that the ability to respond to short-term nifedipine or hydralazine therapy predicts longer survival for patients with primary pulmonary hypertension, but placing patients with a favorable short-term response on long-term vasodilator therapy does not affect the overall outcome.