OBJECTIVE: Management of pulmonary atresia with intact ventricular septum (PAIVS) can be directed to either biventricular repair or univentricular palliation. The optimal management strategy has yet to be defined. METHODS: All patients operated at the Royal Children's Hospital, Melbourne for PAIVS between 1990 and 2006 (n = 81) were reviewed. Patients were retrospectively stratified into a simple three-tiered classification based on right ventricle (RV) size. Multivariate logistic regression analysis was performed to identify risk factors of mortality. RESULTS: The distribution of RV sizes was normal in 11 (14%), moderate hypoplasia in 45 (56%), and severe hypoplasia in 25 (31%) patients. RV-to-coronary-artery connections were present in 33 (41%) and RV coronary dependence in six patients (7%). Sixteen patients died (20%). The end-status of the remaining patients was biventricular repair in 31/81 (38%), 1½-ventricle repair in 10/81 (12%), Fontan circulation in 14/81 (17%), transplantation in 1/81 (1%), and still awaiting repair in 9/81 (11%). Ten-year survival was 80% (95% confidence interval (CI): 71-87%). Independent predictors of mortality were lower tricuspid valve (TV) annulus size Z-score and the presence of RV-to-coronary-artery connections. CONCLUSIONS: A simple three-tiered classification based on RV size may allow initial stratification into biventricular or univentricular repair for patients with normal RV size and severe RV hypoplasia. In patients with moderate RV hypoplasia, the presence of RV-to-coronary-artery connections or a TV Z-score<-2 should caution one against attempting biventricular repair. Crown
OBJECTIVE: Management of pulmonary atresia with intact ventricular septum (PAIVS) can be directed to either biventricular repair or univentricular palliation. The optimal management strategy has yet to be defined. METHODS: All patients operated at the Royal Children's Hospital, Melbourne for PAIVS between 1990 and 2006 (n = 81) were reviewed. Patients were retrospectively stratified into a simple three-tiered classification based on right ventricle (RV) size. Multivariate logistic regression analysis was performed to identify risk factors of mortality. RESULTS: The distribution of RV sizes was normal in 11 (14%), moderate hypoplasia in 45 (56%), and severe hypoplasia in 25 (31%) patients. RV-to-coronary-artery connections were present in 33 (41%) and RV coronary dependence in six patients (7%). Sixteen patients died (20%). The end-status of the remaining patients was biventricular repair in 31/81 (38%), 1½-ventricle repair in 10/81 (12%), Fontan circulation in 14/81 (17%), transplantation in 1/81 (1%), and still awaiting repair in 9/81 (11%). Ten-year survival was 80% (95% confidence interval (CI): 71-87%). Independent predictors of mortality were lower tricuspid valve (TV) annulus size Z-score and the presence of RV-to-coronary-artery connections. CONCLUSIONS: A simple three-tiered classification based on RV size may allow initial stratification into biventricular or univentricular repair for patients with normal RV size and severe RV hypoplasia. In patients with moderate RV hypoplasia, the presence of RV-to-coronary-artery connections or a TV Z-score<-2 should caution one against attempting biventricular repair. Crown