To determine correlates of acute vasodilator responsiveness in primary pulmonary hypertension, we retrospectively studied 25 patients, comparing 41 resting echo/Doppler and nine resting catheterization variables with the maximal reduction in pulmonary vascular resistance achieved during vasodilator trials. Twelve vasodilators were tested (mean, 5.6 drugs per patient; range, three to eight). Eight patients were vasodilator responsive, as defined by a reduction in pulmonary vascular resistance greater than or equal to 30 percent in response to at least one agent. Univariate and multivariate analyses revealed only Doppler pulmonic peak flow velocity to be an independent correlate of responsiveness (p less than 0.05). Responders differed from nonresponders in having a higher Doppler pulmonic peak flow velocity (PV) (SD 81 +/- 24 vs 64 +/- 15 cm/s; p = 0.05), lower mean right atrial pressure (RAP) (6 +/- 4 vs 13 +/- 7 mm Hg; p = 0.04), and longer median survival (37 vs 5 months; p = 0.03). Seven of eight responders had RAP less than or equal to 10 mm Hg, and all responders had PV greater than 60 cm/s. Seven of ten patients with both RAP less than or equal to 10 and PV greater than 60 and one of the 15 remaining patients were vasodilator responsive (p less than 0.001). Thus, echo/Doppler and invasive hemodynamic parameters correlate with acute vasodilator responsiveness in primary pulmonary hypertension. Patients with low PV and high RAP values were almost never vasodilator responsive. Doppler pulmonic peak velocity and mean RAP may be useful in identifying patients most likely to respond to acute vasodilator trials and those in whom testing is unlikely to yield positive results.