The outcome in 135 patients with pulmonary atresia with intact ventricular septum was reviewed in terms of actuarial survival and "suitability" for eventual definitive repair (defined as having a tricuspid valve diameter larger than 2.4 standard deviations below the mean normal beyond the neonatal period). Of patients who underwent an initial closed valvotomy, 50% were dead, 22% were alive and suitable, and the remaining 28% were alive but unsuitable for definitive repair at 5 years. None of 66 patients who underwent primary shunt alone achieved suitability and 52% were dead at 5 years. No mode of neonatal closed valvotomy was consistently satisfactory: in only 10 of 26 survivors with serial measurements did the neonatal valvotomy alone achieve decompression to subsystemic pressures. Only half of the decompressed ventricles achieved growth of the tricuspid valve disproportionate to somatic growth as a result of the neonatal procedure. Strategies for the neonatal treatment of pulmonary atresia with intact ventricular septum should not rely on closed operations on the pulmonary valve to optimized right ventricular growth.