Optimal Shunt Type for Norwood Procedure: Predicted Adverse Impact of Discordant Surgical Approach.

Background: The Single Ventricle Reconstruction (SVR) trial compared survival after Norwood procedure with either modified Blalock Taussig shunt (MBTS) or right ventricle pulmonary artery shunt (RVPAS).

Methods: Data from all 549 participants in the SVR trial were used to develop the MBTS TFSA algorithms, which predict the transplantation-free survival advantage (TFSA) after MBTS vs RVPAS at 1 and 6 years after Norwood procedure. Linear regression analysis of the MBTS TFSA values was performed to identify factors related to more optimal outcomes with MBTS at each timepoint. The impact of discordant management (ie, predicted shunt type did not equal the one actually received) on outcomes and the extent of inconsistencies between predictions were evaluated.

Results: The MBTS TFSA algorithm favored MBTS over RVPAS for only 6.2% of participants at 1 year and for 27.0% at 6 years. In terms of both 1- and 6-year outcomes, MBTS was favored with younger age at Norwood procedure and pre-Norwood intubation, while RVPAS was favored with younger gestational age and metrics indicating larger right ventricle size in the parasternal echocardiographic views. Other predictors were timepoint-specific. MBTS TFSA based allocation could have led to an absolute risk reduction in heart transplantation and mortality of 8.0% at 1 year and 16.8% at 6 years, mostly by preventing discordant MBTS management. Notably, separate predictions from the 1-year and 6-year algorithms produced discordant predictions for 136 participants (24.8%).

Conclusions: The incorporation of data-derived patient-specific factors for selection of shunt type for the Norwood procedure may produce more optimal transplantation free survival. These precision medicine algorithms require prospective validation.