25-Year outcomes following the regionalization of three competing pediatric cardiac surgery programs.

Abstract

The concept of regionalizing pediatric cardiac surgery centers to elevate care, increase surgeon volume, and improve outcomes has been discussed for several decades. We examined the outcomes after regionalizing 3 competing pediatric cardiac surgery programs. The procedures from children (<18 years of age) requiring pediatric cardiac surgery between 1991 and 2024 were divided into 2 groups: (1) preregionalization (1991-1998) when surgery was performed by 3 independent programs (Buffalo, Syracuse, Rochester) and (2) regionalization (2000-2024) when surgery was performed by 1 program at 2 sites (Syracuse, Rochester). Geographic information system software estimated the travel distance between the patient's residence and the surgical center. Logistic regression evaluated the impact of regionalization on hospital mortality. From 7234 procedures, 2326 were performed during preregionalization and 4908 after regionalization. Surgeon volume (72.5 [interquartile range, 67.4-85.8 cases/year] vs 193.0 [interquartile range, 88.0-216.5 cases/year]; P < .001), the percentage of neonatal procedures (17.3% vs 24.0%; P < .001), and risk-adjusted congenital heart surgery 5 and 6 cases (2.3% vs 4.8%; P < .001) increased significantly after regionalization. Travel distance was longer (20.0 [interquartile range, 19.0-78.0 miles] vs 78.0 [interquartile range, 32.0-88.0 miles]; P < .001), but the rates of unplanned reoperation (5.4% vs 1.5%; P < .0001) and mortality (6.7% vs 4.2%; P < .001) were lower after regionalization. Multivariable analysis demonstrated that regionalization independently lowered mortality (odds ratio, 0.483, 95% CI, 0.306-0.761; P = .002). After regionalization, travel distance and surgeon volume increased, and mortality decreased. Our model suggests that regionalizing pediatric cardiac surgical programs can increase surgeon volume and improve outcomes.