Results of long-term vasodilator therapy in patients with refractory congestive heart failure.

The long-term effects of vasodilator therapy with oral hydralazine and long-acting nitrates were studied in 34 patients with refractory heart failure. Seven patients who had marginal hemodynamic improvement despite optimal hydralazine therapy were not maintained on vasodilators, and eight who had a favorable hemodynamic response subsequently discontinued hydralazine therapy because of side effects. Of these 15 patients, four (27%) died and 11 remained in New York Heart Association functional class II or IV at a mean follow-up of 10 +/- 2 months (SEM). The 19 patients who received chronic therapy for 8 +/- 2 months were divided into nine late responders (47%), who improved to functional class I or II, and 10 late nonresponders (53%), who remained in functional class III or IV. Only one of the nine late responders (11%) died, compared with seven of the 10 late nonresponders (70%) (p less than 0.01). The actuarially determined survival at 1 year was 100% for late responders and 13 +/- 12% for late nonresponders (p less than 0.01). No clinical variable could distinguish late responders from late nonresponders. Hemodynamic variables measured before vasodilator therapy showed that late responders had a lower mean right atrial pressure (8 +/- 1 vs 17 +/- 3 mm Hg, p less than 0.01) and lower mean pulmonary artery wedge pressure (20 +/- 2 vs 30 +/- 2 mm Hg, p less than 0.005), higher stroke, volume index (27 +/- 2 vs 20 +/- 1 ml/m2, p less than 0.005) and higher stroke work index (32 +/- 4 vs 19 +/- 2 g-m/m2, p less than 0.01) than late nonresponders. There were no significant differences in the acute response to vasodilators between the two groups. We conclude that (1) a substantial portion of patients with refractory congestive heart failure either do not have a beneficial response to vasodilator therapy or discontinue it because of side effects; (2) about half of the patients who are maintained on chronic vasodilator therapy (or about one-fourth of the patients in whom therapy is initiated) had sustained clinical benefit; and (3) the initial hemodynamics, but not the clinical variables, are predictive of late mortality and late clinical response. Patients with evidence of more severe left ventricular dysfunction have an unfavorable course.