Independent external evaluation of pediatric HCM Risk Scores in predicting severe ventricular arrhythmias

Introduction

Sudden cardiac death (SCD) is the most common cause of death in childhood hypertrophic cardiomyopathy (HCM). Recently, two risk scores have been developed to estimate the 5-year risk of SCD.

Objective

We aimed to assess their respective performances in an independent cohort of primary prevention children with HCM.

Methods

All patients with HCM < 18-year-old from a single-center were retrospectively included between 2003 and 2023. Secondary and syndromic causes of HCM were excluded as well as children with inaugural sustained ventricular arrythmias. HCM Risk-Kids and PRIMaCY risk scores were calculated at diagnosis and during follow-up. The primary composite outcome included sustained ventricular arrhythmia, appropriate ICD therapy, aborted cardiac arrest, or SCD.

Results

Hundred primary prevention children were included (mean age 7.1 ± 5.6 years, 59.0% males), with a mean follow-up of 8.6 ± 5.5 years.13 (13.0%) patients experienced the primary composite outcome. When only considering events during the 5 first years, Harrel's C index was 0.52 (95% CI: 0.27–0.77) for HCM Risk-Kids (≥ 6%) and 0.70 (95% CI: 0.59–0.80) for PRIMaCY (> 8.3%), with 1 patient potentially treated by ICD for every 25 ICDs implanted for HCM Risk Kids and 1 for every 14 ICDs implanted for PRIMaCY. When risk scores were repeated and all primary outcomes during follow-up considered, all events except one (93.2%) were correctly identified using both risk scores, with 1 patient potentially treated by ICD for every 5.6 ICDs implanted for HCM Risk Kids and 1 for every 5.3 ICDs implanted for PRIMaCY. Among 44 (44.0%) patients implanted with an ICD, all primary prevention patients who had ≥ one appropriate ICD therapy during follow-up had HCM Risk-Kids ≥ 6% and PRIMaCY > 8.3% at implantation.

Conclusion

Our findings suggest imperfect discrimination between low and high-risk HCM patients using these two risk scores. The performance or risk scores was substantially improved by periodic re-assessment during follow-up.