PRIMARY ROSS OPERATION MAY BE PREFERABLE TO ATTEMPTS AT AORTIC VALVE REPAIR IN OLDER CHILDREN WITH PRIMARY AORTIC INSUFFICIENCY

Abstract

BACKGROUND: In older children with severe aortic insufficiency (AI), a strategy of aortic valve REPAIR is often pursued in order to delay ROSS procedure or aortic valve replacement with a mechanical/bioprosthetic device (AVR). We aimed to test this strategy of valve REPAIR by evaluating its durability. METHODS: From 2001-2012, 90 children with severe AI underwent: REPAIR (N1⁄446, 51%), ROSS (N1⁄421, 23%) or AVR (N1⁄423, 26%). Repeated measures (N1⁄41081 echos) mixed model analysis were used to evaluate haemodynamic outcomes. RESULTS: Mean age at operation was identical for REPAIR and ROSS (11 years), but slightly older for AVR (13 years). Annular dimensions were largest for REPAIR versus either AVR or ROSS (P1⁄4.01, figure). Cardiopulmonary bypass times were significantly shorter for REPAIRS versus either ROSS or AVR (figure, P<.01). REPAIRS were initially attempted in 52; 6 switched strategy during surgery to ROSS (5) or AVR (1). REPAIRS were associated with more multiple bypass runs than primary strategies of ROSS or AVR. Need for multiple runs did not influence survival or need for re-operation. After REPAIR, peak left ventricular outflow tract (LVOT) gradients were highly variable (figure). Some offered reasonable durability with peak LVOT w50 mmHg at 5-8 years, however recurrent stenosis was often evident within 2-3 years (figure). Overall, time-related LVOT gradients were significantly higher after REPAIR versus either AVR or ROSS (P1⁄4.04). ROSS offered the best long-term freedom from residual LVOT stenosis (figure). Long-term freedom from AI was comparable between REPAIR and ROSS: approximately 80% had mild+ or better AI at 8 years and 20% had moderate or worse. AVR offered the best freedom from AI. There have been no deaths. Estimated freedom from surgical re-operation is 74%, 100% and 63% at 5 years for REPAIRS, ROSS and AVR respectively (P1⁄4.05). 4 of 5 reoperations after AVR were for failed non-mechanical devices. CONCLUSION: Durability after aortic valve REPAIR for AI in children is unpredictable and this strategy may offer little benefit in off-setting need for ROSS or AVR. Small annular sizes as a driver for a REPAIR strategy (therefore allowing for continued annular-growth) seems incongruous, as mean annular dimensions were already favourable in the REPAIR group (mean 2.2 cm). Primary ROSS or AVR (in warfarintolerant teenagers) may be favourable over attempt at REPAIR. 233 CLINICAL OUTCOME SCORE PREDICTS ADVERSE NEURODEVELOPMENTAL OUTCOME AFTER INFANT HEART SURGERY