Although atrial switch is rarely performed today except as part of a "double switch" operation, there continues to be interest in the long-term outcome of the procedure because of the many Mustard and Senning survivors who are in follow-up. In contrast to most other reported series on atrial switch, this study by the Congenital Heart Surgeons Society (CHSS) is a prospective multiinstitutional study of patients encountered in a relatively short time frame during the late 1980s. Between 1985 to 1989, 20 centers with surgeons belonging to the CHSS prospectively registered patients <15 days of age presenting with simple or complex transposition of the great arteries (TGA). The patients were assigned to protocol groups based on intent to treat (arterial switch, Mustard, or Senning). Data were abstracted in the Data and Analysis Center, which in most cases, conducted the annual follow-up. Among the 281 patients who had an atrial switch, there were 108 Mustard and 173 Senning procedures. For the combined atrial switch population, the survival at 1 month, 5 years and 10 years was 90%, 85%, and 84%, respectively. Results for the Mustard were better than for the Senning with survival at 1 month, 5 years, and 10 years being 96% versus 86%, 95% versus 80%, 93% versus 78%; (P <.001) for Mustard versus Senning. While the major mortality risk occurred in the first postoperative month for both groups, thereafter, the late rate of death from 1 to 10 years after operation was 0.78%/year Senning versus 0.23%/year Mustard (P <.05). TGA/ventricular septal defect (VSD), lower weight, younger age, cardiac positional anomalies, and procedures on the left ventricle (LV) outflow all correlated significantly with poor outcome. There were 19 reoperations including 2 for right ventricle (RV) failure, 12 for pathway obstruction, and 5 for baffle leak with a 36% overall mortality. Freedom from any pathway obstruction at 10 years was 95%. A permanent pacemaker was required in 21 patients with the significant risk factors including TGA/VSD and a Senning, and previous atrial septectomy. Freedom from a pacemaker insertion at 10 years was 91%. For both Mustard and Senning, 60% of patients were in New York Heart Association (NYHA) functional Class I with the remaining in Class II. This study of atrial switch outcomes by the CHSS suggests that the subgroup undergoing a Mustard procedure had better early and late survival than those undergoing a Senning operation. The overall incidence of baffle pathway obstruction was low though reoperation for this complication carried high risk. Serious arrhythmia requiring medication and the need for a permanent pacemaker were both relatively uncommon. The incidence of severe right heart failure was very low, although only 60% of patients are unrestricted in their activity. Copyright 2000 by W.B. Saunders Company