Shunt Reintervention and Time-Related Events After Norwood Operation: Impact of Shunt Strategy

Background The use of right ventricle-to-pulmonary artery shunt (RV-PA) is thought to improve postoperative hemodynamics in hypoplastic left heart syndrome (HLHS). Proximal shunt obstruction often prompts early reintervention. Description of our technique and analysis of outcomes after Norwood operation using RV-PA shunt (NRVPA) was undertaken. Methods Between January 2005 and June 2009, 100 consecutive neonates with HLHS underwent stage-1 palliation (S1P). Sixty-five had NRVPA. From 56 survivors, 52 (86. 2%) reached pre-Glenn cardiac catheterization. The RV-PA shunt was constructed using nonringed polytetrafluoroethylene with reversed beveling and hood augmentation of its anterior wall. The PA size, Nakata index, pulmonary to systemic blood flow ratio, and any pre-Glenn intervention were studied. The RV function and NRVPA diastolic reversal flow were assessed. Results Mean age and weight at S1P were 6. 6 ± 3. 5 days and 3. 21 ± 0. 57 kg, respectively. No shunt reversal flow to RV was noted the first 48 hours after S1P by echocardiogram. No patient had shunt revision after hospital discharge from S1P. Mean age and weight at stage-2 palliation were 14. 9 ± 5. 1weeks and 5. 4 ± 2. 1 kg, respectively. No patient had proximal shunt stenosis. Six patients with distal shunt stenosis and associated central PA hypoplasia (32% ± 17% PA diameter reduction), 3 with left PA and 2 with right PA) stenoses had balloon angioplasty. The pulmonary to systemic blood flow ratio between patients requiring pre–stage-2 palliation intervention and patients without intervention was 1. 05 ± 0. 3 versus 1. 13 ± 0. 37 ( p = 0. 09). Right PA and left PA growth was balanced (right PA/left PA 0. 99 ± 0. 3; right PA 26. 8 ± 11. 5 mm 2 versus left PA 27. 2 ± 12. 3 mm 2 ; p = 0. 1). Freedom from significant RV dysfunction for S1P survivors at 3, 6, 12, 36, and 48 months was 94.4% ± 3.1%, 88.7% ± 4.4%, 86.7% ± 4.7%, 83.9% ± 5.3%, and 79.7 ± 6.7%, respectively. Conclusions Norwood RV-PA promotes balanced branch PA growth. The RV-to-PA shunt with proximal valvelike hood modification significantly reduces proximal anastomotic stenosis and need for early shunt revision. Limited early shunt diastolic reversal flow into the RV and subsequent early unloading of RV during BDG might preserve ventricular performance.