The clinical, hemodynamic, and angiographic data on 92 patients with severe isolated aortic stenosis were reviewed to determine the incidence and mechanism of pulmonary hypertension. The status of each of these patients was determined 1 to 8 years after diagnosis by cardiac catheterization. Patients were divided into three groups on the basis of the pulmonary artery systolic pressure: group 1 (less than or equal to 30 mm Hg), 46 patients; Group 2 (31 to 50 mm Hg), 31 patients; and Group 3 (greater than 50 mm Hg), 15 patients. The prevalence of pulmonary hypertension was 50% (46/92) and that of severe pulmonary hypertension, 16% (15/92). There was no significant difference in age, aortic valve gradient, or valve area among the three groups. There was a significant positive correlation in left ventricular end-diastolic pressure (group 1, 15.5 +/- 7.2 mm Hg; group 2, 23.3 +/- 8.1 mm Hg; and group 3, 29.5 +/- 5.8 mm Hg; R = 0.56, p less than 0.01). There was also a significant negative correlation in left ventricular ejection fraction (group 1, 67.5% +/- 14%; group 2, 62.3% +/- 13.8%; and group 3 49.9% +/- 18.3%; R = 0.43, p less than 0.01). Of the 92 patients, 85 had aortic valve replacement with four (4.7%) hospital deaths. Follow-up showed excellent symptomatic relief in all three groups. Thirteen of the 15 patients in group 3, with severe pulmonary hypertension, had aortic valve replacement. There were no hospital deaths and only one noncardiac death at follow-up in Group 3 patients, and 11 of the 12 surviving patients were in New York Heart Association functional class I. We conclude that pulmonary hypertension is common in isolated aortic stenosis and is related to an elevated left ventricular end-diastolic pressure, frequently with preserved systolic function. Surgical results are excellent.