Background: At present, a considerable number of patients survive who underwent an atrial switch operation for correction of complete transposition. Our study aimed to assess their long-term exercise performance and the serial evolution of cardiac function. Methods: We studied 22 patients 5 to 17 years after an atrial switch operation, and followed them serially for 3.5 ± 2 years after the first evaluation. Cardiorespiratory exercise function was assessed by analysis of gas exhange and by determination of the ventilatory anaerobic threshold. Echocardiography was performed on all evaluations. Results: All patients were in Class I of the classification of the New York Heart Association at all assessments. Ventilatory anaerobic threshold, however, was significantly lower than normal. It averaged 77.9% ± 13.7 of the normal mean value at the initial evaluation, and remained stable when re-evaluated later (76.2 ± 13.7%). At the initial study, the increase in oxygen uptake during graded exercise was below the 95% confidence limit in 6 of the patients, and was below this level in 10 patients at re-assessment. The subnormal values for oxygen uptake during submaximal exercise were associated with moderate to severe haemodynamic dysfunction. At echocardiography, 15 of 17 patients studied twice had mild to moderate right ventricular dilation and tricuspidregurgiation, which remained virtually the same at reasssesment. A stable sinus rhythm was initially present in 17 patients, and persisted in 15 patients during follow-up. Conclusion: At medium term follow-up, cardiorespiratory exercise performance remains stable in patients after atrial switch repair. Serial exercise testing appears useful, because in individual patients in the present study, a decreasing exercise tolerance correlated with development of haemodynamic sequels.

Atrial correction for complete transposition involves a potential risk of later development of right ventricular dysfunction. To determine the reasons contributing to such systemic ventricular failure, radioisotope first-pass and SPECT perfusion studies at rest after one dose of Technetium-99m MIBI were performed in 26 asymptomatic patients, 5.9±2.7 years after a Mustard or Senning repair. Mean age at operation was 4.3±4.3 years. All patients were followed with 24-hour Holter monitoring and echocardiography. Normal perfusion was observed in 11 patients with a mean age at operation of 2.25±1.8 years and mean right ventricular ejection fraction of 49.0±8.7%. Perfusion defects were found in 15 patients with a mean age at operation 5.8±4.8 years (p<0.05) and mean right ventricular ejection fraction of 43.2±9.3% (NS). In the group of older patients, 10 had extensive perfusion defects with significantly lower right ventricular ejection fraction (39.7±9.3%, p<0.05) and even older mean age at the surgery (7.1±5.6 years p<0.02). Mild or moderate tricuspid regurgitation was found significantly more frequently in those patients with perfusion defects (73%) than in the group of younger patients (36.3%), p<0.05. There was no difference in the frequency and type of arrhythmias, nor in the extent of follow-up, between the two groups. We conclude that, in patients undergoing surgery at an older age, it is more common to find perfusion defects and, often, lower right ventricular ejection fraction. Perfusion abnormalities found in such asymptomatic patients may be a result of several factors: impaired right ventricular function as a consequence of pressure and volume overload, enhanced myocardial fibrosis, or long-lasting preoperative hypoxemia and tricuspid valvar dysfunction secondary to ventricular dilation. We suggest that the perfusion defects observed in this study may be a better indicator of systemic ventricular impairment.

Between December 1982 and April 1991, balloon atrial septostomy was performed in the intensive care unit under echocardiographic control in 60 neonates. Of the patients, 58 had complete transposition. Two patients had double outlet right ventricle with a sub-pulmonary ventricular septal defect. Associated lesions included a patent arterial duct in 19 patients, ventricular septal defect in nine, obstruction of the left ventricular outflow tract in six, aortic coarctation in two and tricuspid atresia in one. The mean age at septostomy was four days (range 4 hours - 25 days) and the mean weight 3.19 kg (range 1.17–4.25 kg). In 39 (65%) patients, an infusion of prostaglandin was in progress prior to the septostomy and 22 (37%) were being ventilated. Standard subcostal four-chamber echocardiographic views were used to show the atrial septum and to guide the catheter used for septostomy. Venous access was obtained via the femoral vein in 43 (by percutaneous puncture in 40 and by cutdown in three) and the umbilical vein in 17. Transient atrial arrhythmias were common during the septostomy but no acute hemodynamic disturbances or deaths occurred during the procedure. The size of the atrial septal defect as measured by echocardiography after the septostomy ranged from three to 12 mm in diameter. In only one patient was this inadequate. Three (5%) patients died between two and 10 days after the septostomy, two due to necrotizing enterocolitis and one from persistent hypoxemia. One patient had a cerebral thrombosis and convulsions immediately after the septostomy but made a good neurological recovery. Corrective surgery was performed in 52 (86.6%), two (3.3%) had palliative surgery and two were considered unsuitable for total correction, of whom one has died. One patient died whilst awaiting correction. We conclude that balloon atrial septostomy using echocardiographic guidance can be safely and effectively performed in the intensive care unit.