Geometry of the pulmonary arteries before the Fontan operation: can we influence it during the Norwood procedure?

Abstract

OBJECTIVES The right ventricle-to-pulmonary artery (RV-PA) shunt provides stable haemodynamics after the Norwood procedure but can influence development of the central pulmonary arteries (PAs). The goal of this study was to analyse the geometry of the central PAs in children with hypoplastic left heart syndrome before the Fontan operation with respect to the RV-PA shunt site and the type of the second-stage operation. METHODS A total of 161 children with hypoplastic left heart syndrome, median age 2.7 (range 1.3–9.8) years and median weight 12.7 (range 7.6–26.1) kg, underwent the Fontan operation after having had the Norwood procedure with an RV-PA shunt. The patients were divided into 2 groups: left-sided RV-PA (L-RV-PA) (n = 129) with the shunt on the left and right-sided RV-PA (n = 32) with the shunt on the right side of the neoaorta. Angiographic data obtained before the Fontan and all cardiac catheterization interventions were analysed retrospectively. RESULTS Between the second and third stages, as well as directly before the Fontan operation, the L-RV-PA group required more PA catheter interventions (P = 0.001 and P = 0.03). In this group, the minimal left PA diameter was smaller than that in the R-RV-PA group (P = 0.021). Leaving the shunt open until the Fontan operation increased the rate of PA interventions in the L-RV-PA group (P = 0.001), but there is no evidence of the impact on the development of the left PAs (P = 0.075). There is also no evidence that the type of the second-stage procedure influences the intervention rate before the Fontan procedure (P = 0.14). CONCLUSIONS Children who have the L-RV-PA shunt require more PA catheter interventions. The right-sided RV-PA shunt and the subsequent Glenn anastomosis in the place of the shunt are associated with distortion-free and more symmetrical development of the central PAs.