Pii: S1010-7940(02)00295-6

Objective: The optimal time and approach of repair of tetralogy of Fallot (TOF) remain controversial. The purpose of this study was to evaluate the outcome following repair of TOF in infants with particular regard to the surgical approach used. Patients: One hundred and sixty infants (mean age 195 ^ 89 days, range 11–364 days) undergoing repair of a simple TOF were studied. Between 1974 and 2000, a transventricular approach (RV) was used in 91 and between 1988 and 2000, a transatrial (RA) approach in 69 infants. Ten of these infants (6.2%) had a previous palliative shunt (four in the RV versus six in the RA group). A transannular patch (TAP) was inserted in 96 (60%) infants (76 versus 20). Follow-up was complete (mean 14.5 ^ 5.2 versus 6 ^ 1 years). Results: There were three operative deaths (1.9%), (two in RV versus one in RA group). A re-operation for right ventricular outflow tract obstruction (RVOTO) was performed in 19 patients (3 versus 16). Ten-year freedom from re-operation for RVOTO (^standard error of the mean) was 88 ^ 4% (98 ^ 2 versus 72 ^ 6%, P , 0:0001). Within the RA group, 5-year freedom from re-operation for RVOTO for those who had a TAP was 79 ^ 9% and it was 75 ^ 4% for those having a simple repair. Six patients in the RV group required pulmonary valve replacement (PVR). Ten-year freedom from PVR was 98 ^ 1% (97 ^ 2 versus 100%, P 1⁄4 0:3). There were two late deaths, one in each group. Ten-year survival was 97 ^ 1%. One patient in the RV group developed late recurrent ventricular tachycardia requiring the implantation of a defibrillator. At most recent echocardiography, all but the patient who had the defibrillator had good right and left ventricular function. Conclusions: Transventricular and transatrial repair of TOF in infancy, are associated with an acceptable operative risk, low incidence of late arrythmias, good biventricular function and excellent survival. In our experience, however, transatrial repair has a disturbing incidence of early and midterm residual or recurrent RVOTO, even when a TAP has been used. q 2002 Elsevier Science B.V. All rights reserved.