Abstract

Despite balloon atrial septostomy within the first days of life, some patients with complete transposition of the great arteries die before reaching elective definitive surgery in the second six months of life. To discover why, we analysed the fate of 144 patients who had balloon atrial septostomy after 1966, using a modified logrank survival test with multivariate capability. Patients were withdrawn "alive" on reaching definitive surgery. The following largely independent factors were associated with a statistically significant excess mortality: pulmonary hypertension, the presence and size of a ventricular septal defect of persistent ductus arteriosus, relative anaemia, absence of left ventricular outflow tract obstruction, low arterial oxygen saturation, aortic stenosis and coarctation, and balloon atrial septostomy between 1 week and 1 month of life. Those of the above factors which can be determined at balloon atrial septostomy or at routine cardiac catheterisation at 3 months of age were then introduced into discriminant function analysis on survival to 6 months. Hence the probability of any individual patient dying in the first six months was calculated, allowing for these factors. This prediction was correct in 76 per cent of the patients studied. By offering earlier definitive correction to patients thus identified as being at high risk of premature death, it should prove possible to reduce overall mortality in transposition of the great arteries.