Abstract
Cardiopulmonary Exercise Testing (CPET) is considered the gold standard assessment of peak oxygen uptake ( V˙O2 peak), and consequently, cardiorespiratory fitness (CRF). Common practice in children with congenital heart diseases (CHD) is to compare individual test results to reference values in healthy children. This approach does not account for the diversity of heart defects and may also be demotivating for children with CHD. This retrospective multicentre study aimed to establish individualized CRF reference models specific to children with CHD, to facilitate a better understanding and management of their condition. A total of 1475 CPETs until exhaustion were performed in 943 children and adolescents at two hospitals in Norway. The children and adolescents were categorized into three groups: simple defects (VSD, ASD, CoA, LVOTO; n = 497, 38% female), moderate defects (TGA, Fallot; n = 299, 41% female), and univentricular defects with a Fontan circulation (n = 147, 42% female). Multivariable mixed-effects models with individuals as random intercepts were used to develop group-specific reference models for CRF. Estimated predictive models including covariance matrices for each group and outcome are provided, allowing for the generation of reference values with confidence intervals for clinical and research applications. Conclusion: Treadmill-based reference models for maximal and submaximal CRF variables in the most prevalent types of CHD are presented. Using individualized reference values has the potential to improve clinical decision-making for children with CHD and to enhance their sense of mastery. A web-based calculator is available for quick and easy clinical use.