The ductal flow velocities in 37 newborns (group 1: persistent pulmonary hypertension [n = 16], transient tachypnea [n = 3], other [n = 2]; group 2: respiratory distress syndrome [n = 16]) were prospectively evaluated by Doppler ultrasound for the purpose of deriving systolic pulmonary artery pressures. Maximal tricuspid regurgitant Doppler velocity in 21 of these patients was used to validate the pulmonary artery pressures derived from ductal flow velocities. There was a significant linear correlation between tricuspid regurgitant Doppler velocity and pulmonary artery systolic pressure derived from ductal Doppler velocities in patients with unidirectional (pure left to right or pure right to left) ductal shunting (p less than 0.001, r = 0.95, SEE 8) and in those with bidirectional shunting (p less than 0.001, r = 0.95, SEE 4.5). Systolic pulmonary artery pressure in group 1 (67 +/- 13 mm Hg) was significantly higher than that in group 2 (39 +/- 10 mm Hg) (p less than 0.001). In those with bidirectional shunting, duration of right to left shunting less than 60% of systole was found when pulmonary artery pressure was systemic or less, whereas duration greater than or equal to 60% was associated with suprasystemic pulmonary artery pressures. Serial changes in pulmonary artery systolic pressure, reflected by changes in ductal Doppler velocities, correlated with clinical status in persistent pulmonary hypertension of the newborn. Persistently suprasystemic pulmonary artery pressure was associated with death in five group 1 patients. It is concluded that ductal Doppler velocities can be reliably utilized to monitor the course of pulmonary artery systolic pressures in newborns.