A total of 466 neonates with transposition of the great arteries, all less than 15 days of age, have been entered into a 20-institution study between Jan. 1, 1985, and June 1, 1987. Seventy-three percent were less than 48 hours old when entered into the study. Two hundred twelve of these underwent an arterial switch repair, and the 1-week, 1-year, and 2 1/2-year survival rates were 82%, 79%, and 78%, respectively. The usual coronary anatomy was present in 67% of the patients. The arterial switch repair was performed in 16 of the institutions. Six among these were shown to be "low-risk" institutions, with the prevalence of the demographic and surgical variables seeming to be the same in these as in the other institutions. The 1-week, 1-year, and 2 1/2-year survival rates after the arterial switch repair in low-risk institutions for simple transposition were 96%, 91%, and 90%, respectively; for transposition with ventricular septal defect they were 84%, 83%, and 83%. Only older age at operation (over 14 days of age), in the case of simple transposition, and transposition with ventricular septal defect were risk factors for death in these six institutions. Among the patients as a whole, freedom from reoperation for pulmonary outflow obstruction at 1 week and 1 year was 99% and 89%, respectively. A previous pulmonary artery banding and, possibly, one institution were identified as risk factors for reoperation. Inferences: In low-risk institutions, the arterial switch repair can be accomplished with good early results, which suggests the possibility that the late results will be better than after the atrial switch repair. Since young age was not a risk factor for the arterial switch repair of transposition and ventricular septal defect, this type of repair for this anomaly, as well as for simple transposition, may be more advantageously performed early in life than at 2 to 3 months of age.