The ideal valve substitute for surgical intervention of congenital aortic valve disease in children remains unclear. Data on outcomes beyond 10-15 years after valve replacement are limited but important for evaluating substitute longevity. We aimed to describe up to 25-year death/cardiac transplant by type of valve substitute and assess the potential impact of treatment centre. Our hypothesis was that patients with pulmonic valve autograft would have better survival than mechanical prosthetic. This is a retrospective cohort study from the Pediatric Cardiac Care Consortium, a multi-institutional US-based registry of paediatric cardiac interventions, linked with the National Death Index and United Network for Organ Sharing through 2019. Children (0-20 years old) receiving aortic valve replacement (AVR) from 1982 to 2003 were identified. Kaplan-Meier transplant-free survival was calculated, and Cox proportional hazard models estimated hazard ratios for mechanical AVR (M-AVR) versus pulmonic valve autograft. Among 911 children, the median age at AVR was 13.4 years (IQR=8.4-16.5) and 73% were male. There were 10 cardiac transplants and 153 deaths, 5 after transplant. The 25-year transplant-free survival post AVR was 87.1% for autograft vs 76.2% for M-AVR and 72.0% for tissue (bioprosthetic or homograft). After adjustment, M-AVR remained related to increased mortality/transplant versus autograft (HR=1.9, 95% CI=1.1 to 3.4). Surprisingly, survival for patients with M-AVR, but not autograft, was lower for those treated in centres with higher in-hospital mortality. Pulmonic valve autograft provides the best long-term outcomes for children with aortic valve disease, but AVR results may depend on a centre's experience or patient selection. © Author(s) (or their employer(s)) 2022. No commercial re-use. See rights and permissions. Published by BMJ.