The preoperative dose response to inhaled nitric oxide (NO) was compared with the need for and response to NO after cardiac surgery in patients with congenital heart defect and secondary pulmonary hypertension. In a preoperative vasodilator test with inhaled NO 20, 40 and 80 ppm and oxygen, mean pulmonary artery pressure (PAP) was at least 40 mmHg and/or the pulmonary vascular resistance index (PVRI) 4 Wood units. Preoperatively, NO 40 ppm and FiO2 0.9 reduced systolic pulmonary/systemic arterial pressure (PAPs/SAPs) from 0.89 (SD 0.10) to 0.80 (0.18) and pulmonary/systemic vascular resistance (PVR/SVR) from 0.26 (0.13) to 0.13 (0.08). Haemodynamic assessment was repeated in 11 patients postoperatively. NO treatment was started if PAPs/SAPs rose to 0.8 or the pulmonary oximetry fell below 40%. Postoperatively, eight of 11 patients, including 6 patients with Down's syndrome, needed NO. PAPs/SAPs decreased more than preoperatively: 48.5% vs 11.2, p = 0.0045. Pulmonary oximetry increased by 15.7%, p = 0.02. The degree of preoperative response to NO did not differ between the patients with postoperative pulmonary hypertension and the other children. Patients with early pulmonary hypertensive crisis (first 24 h; n = 6) had a higher PVRI (7.6 vs 4.4 Um2; p = 0.003) and PVR/SVR (0.34 VS 0.17; p = 0.02) preoperatively. Two patients died in pulmonary hypertensive crisis.