Abstract 10353: Independent Validation of the Ecg Risk Score for Sudden Death Risk Stratification in Pediatric Hypertrophic Cardiomyopathy

Abstract

Background: Hypertrophic cardiomyopathy (HCM) is the leading cause of sudden cardiac death (SCD) in children and young adults. An ECG risk score tested in a Swedish cohort of 155 children with HCM had 97% sensitivity, 80% specificity, 45% positive predictive value (PPV), and 99% negative predictive value (NPV) for SCD risk prediction. Our goal was to validate the ECG risk score in an independent HCM cohort. Methods: Phenotype-positive pediatric HCM patients seen at a single center between 1987-2019 with at least 5 years follow-up from diagnosis, or a SCD event within 5 years of diagnosis, were retrospectively identified. First-evaluation ECGs were analyzed by a single operator blinded to patient outcomes. ECG risk score was calculated using QRS electrical axis, amplitude, duration, ST/T-wave abnormalities and QTc duration. The primary outcome was a composite of SCD, aborted sudden cardiac arrest, and appropriate shock from a prophylactic ICD. Results: Of 144 eligible patients with a median follow-up 14.6 yrs, 22 experienced SCD events within five years from diagnosis at a median age 12.1 yrs. The ECG risk score in patients experiencing a SCD event was 2-fold higher i.e. 8 (IQR 7-10) compared to the remaining cohort i.e. 4 (IQR 2-8) (p<0.001). An ECG risk score >5 had a sensitivity of 95% (CI 77%-100%), specificity 56% (46%-65%), PPV 28% (24%-33%), NPV 99% (91%-100%), and a receiver operating characteristic area under the curve of 0.76 for its ability to discriminate between patients with and without SCD events ( Figure 1 ). Compared to the discovery cohort, the score had strong sensitivity and NPV, but lower specificity and PPV. Conclusions: The ECG risk score for SCD risk stratification performed well in an independent Canadian validation cohort but overestimated SCD risk. Futher analysis will determine if inclusion of the ECG risk score can enhance the performance of currently available SCD risk prediction models for pediatric HCM.