Despite these advances in paediatric cardiology, the stratification of CHD severity using a simple and reproducible classification has not been established, as can be the NYHA functional class in adult heart failure. Various CHD classifications have been used in CHD, focusing on anatomical lesions, complexity of care, or physiological status, but their prognostic value has not been determined.

We aimed to compare the accuracy of the main existing CHD classifications (Uzark, Stout and Bethesda classifications), in the prediction of functional status in children with CHD, as determined by cardiopulmonary fitness.

Longitudinal cohort study.

The CHD population having had 2 CPET included 296 subjects (n=129 female). The time between the first (T1) and second CPET (T2) assessments was 4.1±1.6years. The performance of classifications according to VO₂max at T1 was better for Uzark classification. The VO₂max Z-score decreased significantly according to the severity group (groups 1 and 2>group 3>group 4) and group 4 had a significant VO₂max decrease of −6.68 [−10.69; −2.67] mL/kg/min compared to group 1. The prediction of classifications at T2 according to VO₂max was better for Uzark classification with AUC values of 0.62 [0.55–0.69], compared to 0.59 [0.51–0.66] for Stout and 0.55 [0.48–0.62] for Bethesda (Fig. 1).

Among the existing cardiovascular risk classifications for CHD, the Uzark classification appeared to be the most reliable for discriminating the severity of CHD according to exercise capacity and for predicting the VO₂max impairment than the other classifications tested. This longitudinal study also showed the continued decline in exercise capacity, whatever the CHD, and recalls the interest of regular monitoring to offer care adapted to these patients (such as rehabilitation programs) for primary prevention of the added cardiovascular risks of adulthood.