Historically, indications for ventricular septal defect closure have included congestive heart failure, pulmonary hypertension, aortic insufficiency with or without aortic valve prolapse, and prior bacterial endocarditis. However, controversy exists as to how the lifetime risk of an isolated, nonoperated restrictive ventricular septal defect compares with the risk of surgical closure in an asymptomatic child. Between 1980 and 1991, cardiac catheterization and elective ventricular septal defect closure (age > 1 year, pulmonary to systemic flow ratio < 2.0) were performed in 141 patients aged 1 to 23 years (mean age, 6.1 +/- 4.7 years). Mean systolic pulmonary artery pressure was 26.9 +/- 13.0 mm Hg, and mean pulmonary to systemic flow ratio was 1.6 +/- 0.3. Aortic valve prolapse was present in 63 patients (45%), aortic insufficiency was present in 25 (18%), and 5 (3.5%) had prior bacterial endocarditis. There were no early or late deaths or major morbidity. No patient required a ventriculotomy to accomplish ventricular septal defect closure. Mean postoperative intensive care unit stay was 1.3 +/- 0.9 days, and mean hospital stay was 5.5 +/- 1.9 days. There were no instances of permanent complete atrioventricular dissociation, reoperations for bleeding, postoperative wound infections, or reoperations for residual or recurrent ventricular septal defect. These improved results justify a reevaluation of historic indications for ventricular septal defect closure.(ABSTRACT TRUNCATED AT 250 WORDS)