OBJECTIVE--To describe the demographic and haemodynamic variables of children presenting with primary pulmonary hypertension or pulmonary hypertension appearing or persisting after surgical correction of congenital heart defects and to assess the acute effect of vasodilator drugs on their pulmonary vascular bed. DESIGN--Retrospective review. SETTING--Paediatric cardiology department and intensive care unit of a large tertiary centre. PATIENTS--Fourteen consecutive patients presenting with primary pulmonary hypertension (group 1) or pulmonary hypertension persisting or appearing late after complete surgical repair (group 2). INTERVENTION--Baseline haemodynamic measurements were taken in room air at rest and repeated in 100% oxygen. With constant monitoring of heart rate and pulmonary and systemic arterial pressures, patients were given serial intravenous, sublingual, or oral incremental doses of vasodilators (mean 4.1 trials per patient). The maximum effect of the drug was charted. MAIN OUTCOME MEASURES--A positive response to acute vasodilator tests was defined as a decrease in mean pulmonary or mean systemic arterial pressure > 15% with the mean pulmonary artery pressure not reaching the systemic level and either no change or an increase in mean systemic arterial pressure. Haemodynamic variables between groups (1 v 2, responders v non-responders, patients experiencing or not experiencing adverse effects to vasodilators) were compared by a two tailed unpaired Student's t test, and their survival curves were compared by the log rank statistic. RESULTS--Groups are small and definitive conclusions are difficult to draw, but the baseline haemodynamic assessments were not significantly different between group 1 and 2 or between responders and non-responders to vasodilators. Patients experiencing adverse effects had a higher pulmonary vascular resistance (p = 0.04) and wedge pressure (p = 0.02) than those without adverse effects. Of the vasodilators used, tolazoline, hydralazine, salbutamol, phentolamine, and phenoxybenzamine were ineffective. A positive response was seen in five of 13 patients given oxygen, in one of eight given prostacyclin, four of 12 given nifedipine, four of eight given diltiazem, one of six given captopril, and two of seven given glyceryl trinitrate. Estimates of survival of the population with primary pulmonary hypertension were 37% at one year and 12% at 2.5 years. Survival was significantly shorter in the non-responders than in the responders (p = 0.005). CONCLUSION--Children with primary pulmonary hypertension present to the cardiologist at a young age (five of eight were younger than 7 years) but with advanced pulmonary vascular disease and have a poor prognosis. 64% of group 1 and group 2 patients had a positive response to acute treatment with at least one vasodilator. Calcium channel blockers were the most effective agents. There was a positive response to drugs despite a negative response to acute treatment with oxygen. The survival of non-responders was shorter than that of the responders.