Higher Programmatic Volume in Neonatal Heart Surgery Is Associated With Lower Early Mortality

Abstract

The early results of congenital heart surgery in neonates remain a challenge. We sought to determine the nature of the association between annual center volume of neonatal cardiac surgery and operative mortality using a multicenter cohort. The dataset consists of 27,556 neonatal procedures performed between 1999 and 2015 in 90 centers participating in the European Congenital Heart Surgeons Association database. Centers with mean annual volume load of six or more that submitted data for at least 3 consecutive years were included. World Bank annual gross national index per capita was utilized as an indicator of temporal national affluence. Multilevel logistic regression was used to create a model including the significant risk factors and to calculate odds ratios for operative mortality. Iterative modeling of the dataset incrementally excluding centers with lower annual caseload was used to identify the relationship between annual volume and mortality. In the model thus calculated including The Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery (STAT) mortality score, operative weight and age, noncardiac genetic anomalies, and annual volume of operations were independent risk factors for operative mortality in the analysis of the entire cohort. In the model containing these variables, annual gross national index and year of surgery were not significantly associated with mortality. In the iterative process, annual volume ceased to be a risk factor when units operating on fewer than 60 neonates annually were excluded. In neonatal congenital heart surgery, the risk of operative death decreased with the increase of volume load. The cutoff point in this cohort was a mean annual volume of 60 neonatal operations per year.