Brain natriuretic peptide (BNP) blunts hypoxic pulmonary hypertension in animal models, but its acute hemodynamic effects in patients with pulmonary arterial hypertension (PAH) are not known. The aim of this study was to determine if human B-type natriuretic peptide is a safe and efficacious pulmonary vasodilator in patients with PAH and if the pulmonary hemodynamic effects are potentiated by phosphodiesterase inhibition. Open-label study. Medical ICUs of three tertiary care hospitals in New England. Thirteen consecutive adult patients undergoing right-heart catheterization and a pulmonary vasodilator trial for the initial evaluation of PAH. Patients were administered inhaled nitric oxide (iNO), i.v. epoprostenol, and a 3-h infusion of BNP alone and 1 h after an oral dose of the phosphodiesterase-5 inhibitor sildenafil. iNO and sildenafil alone decreased mean pulmonary artery pressure (mPAP) without a significant fall in pulmonary vascular resistance (PVR). Epoprostenol decreased both mPAP and PVR. BNP alone had no significant effect on pulmonary hemodynamics, but the combination of sildenafil plus BNP decreased mPAP and PVR for up to 6 h after stopping BNP. The decrease in mPAP with sildenafil plus BNP (+/- SE) was greater than after 1 h of sildenafil alone (44.6 +/- 3.8 to 40.6 +/- 3.9 mm Hg, p = 0.027). An acute vasodilator response, defined as a decrease in mPAP > 10 mm Hg and end mPAP < 40 mm Hg, was seen in 0 of 8 patients with iNO, 1 of 13 patients with epoprostenol, 0 of 13 patients with BNP, and 4 of 12 patients with sildenafil plus BNP. BNP decreased mean systemic arterial pressure (5.6 +/- 2.8 mm Hg) but had no effect on cardiac output or systemic vascular resistance. A 3-h BNP infusion does not significantly improve pulmonary hemodynamics in most patients with PAH but is well tolerated and augments the acute pulmonary vasodilator effect of sildenafil.