Abstract 2350: The Adverse Implications Of Early Re-intervention After Biventricular Repair For Neonates With Critical Avs.
The 5-year incidence of unplanned re-intervention in 362 prospectively enrolled neonates with critical aortic stenosis was less than 5% following univentricular repair (UVR). The corresponding rate following biventricular repair (BVR) was instead 51% (p<.01). We therefore sought to determine the impact of re-intervention on survival following BVR. Multiphase parametric modeling and competing risks methodology were used to examine transition to mutually exclusive endstates. Multivariate analysis was then used to identify incremental risks associated with each outcome. Biventricular repair was pursued in 139 (71% 5year survival). Index procedure was balloon valvotomy (105), surgical valvotomy (27) or complex repair (Ross/Yasui, 7). Reintervention was required in 64 (46%) and was not influenced by the type of index procedure. Earlier need for re-intervention was an independent predictor of death (p<.01). Incremental predictors of earlier re-intervention included left ventricular dysfunction (p<.01), fewer aortic cusps (p<.01) and sub-aortic obstruction (p<.01). Re-intervention required within 2 years following index procedure, predicted 5-year survival of 67%. Re-intervention within 30 days predicted 5-year survival of 54% and re-intervention within 2 days predicted 42% 5year survival. UVR offers better predicted survival for those infants requiring re-intervention within 30 days of index BVR (figure⇓).
Conclusions Early re-intervention after intended BVR is associated with poor survival. Patient-specific characteristics identify those BVR patients at risk of requiring early re-intervention. Pursuit of UVR may instead be preferable for such patients.