This study was undertaken to compare long-term results of various types of surgical repairs for either congenitally corrected transposition of the great arteries or double outlet right ventricle with discordant atrioventricular connections, and to analyze the risk factors that affect early and late mortality and reintervention.
Between January 1972 and September 2005, a total of 189 patients (median age 8.3 years, range 2 months to 47 years old) with congenitally corrected transposition of the great arteries or double outlet right ventricle with discordant atrioventricular connections underwent definitive repairs. The definitive repairs comprised a conventional repair (atrial septal defect, or ventricular septal defect closure with or without pulmonary stenosis release, or isolated tricuspid valve surgery) in 36 patients (group I), conventional Rastelli in 31 patients (group II), double-switch operation (atrial switch plus arterial switch) in 15 patients (group III), atrial switch plus intraventricular rerouting (with or without extracardiac conduits) in 69 patients (group IV), and a Fontan-type repair in 38 patients (group V). The mean follow-up period was 10.1 years. Hospitalization and late mortality and reoperation were indicated as events. Risk factors for these events were analyzed by logistic regression for hospital death and a Cox proportional hazards model for late events.
The Kaplan-Meier survival including hospital and late mortality was 62.4% at 32 years in group I, 78.5% at 27 years in group II, 74.5% at 15 years in group III, 80% at 16 years in group IV, and 79.3% at 22 years in group V. The reoperation-free ratio was 64.2% in group I, 76.6% in group II, 84.4% in group III, 89.6% in group IV, and 91.3% in group V. Risk analyses showed that the risk for hospital death was preoperative in patients with more than moderate tricuspid regurgitation and a cardiopulmonary bypass time of more than 240 minutes. A risk for late mortality was the presence of tricuspid regurgitation. Risks for reoperation were preoperative cardiomegaly, preoperative tricuspid regurgitation of more than grade II, ventricular septal defect enlargement, and body weight less than 10 kg. Risks for pacemaker implantation, as indicated by multivariate analysis, were ventricular septal defect enlargement during operation and age less than 3 years.
There were no statistical differences between long-term survival rates of patients who underwent conventional surgical repair versus those of patients who underwent anatomic surgical repair. Results of conventional repair were satisfactory except in patients with significant tricuspid regurgitation. Results of anatomic repair were also satisfactory even for patients with significant tricuspid regurgitation, and therefore, anatomic repair should be the procedure of choice for those patients.