Factors associated with marked reduction in mortality for Fontan operations in patients with single ventricle.

The Fontan operation was originally employed for patients with tricuspid atresia, but its application has been extended to those with a variety of complex defects in which there is functionally only a single ventricular chamber. The outcome of 225 modified Fontan operations carried out between 1984 and 1990 at the Children's Hospital, Boston, for patients with defects other than tricuspid atresia was reviewed. Overall 30 patients (13.3%) had failure of this operation (death or takedown). Results improved significantly during the period of the study, with failure rates of 6.5% (2 of 31) and 3.4% (2 of 59) in the last 2 years of the review. Multivariate analysis showed that pulmonary artery distortion, pulmonary artery pressure, age less than 3 years, use of a nonoxygenated glucose K+ cardioplegic solution, and cardiopulmonary bypass time greater than 180 minutes were associated with worse outcome, while "favorable" atrioventricular valve anatomy (non-left atrioventricular valve stenosis/atresia or common atrioventricular valve) and age greater than 9 years were associated with improved outcome. Excluding cardiopulmonary bypass time from the multivariate analysis, the technique of atrial partitioning for patients with left atrioventricular valve atresia/stenosis became the most important variable, followed by the others noted in the initial multivariate analysis except for age greater than 9 years. In the last 2 years of the study 31.1% of patients were less than 3 years of age. During the period of the study there was no significant decline in preoperative risk factors. These results show that modified Fontan operations can be carried out with a high likelihood of success in properly selected patients with complex defects in whom there is functionally a single ventricle.