Abstract Body (Do not enter title and authors here): Background
Myocardial fibrosis could be detected by cardiac MRI (CMR) with late gadolinium enhancement (LGE) which provides important prognostic information of risk stratification for adult hypertrophic cardiomyopathy (HCM). However, in childhood HCM, it remains uncertain the prevalence, pattern of LGE, and whether it is associated with adverse outcomes.
Aims
This study aims to explore the characteristics of LGE in childhood HCM and to evaluate the prognostic value of LGE for adverse outcomes.
Methods
In this retrospective study, a total of 231 consecutive childhood patients with primary HCM who were ≤18 years of age at diagnosis and underwent contrast-enhanced CMR from January 2011 to December 2019 were enrolled. The extent of LGE was quantified by measuring areas with the increased signal intensity of ≥6 standard deviations above the mean of normal myocardium. The primary outcome included a composite of sudden cardiac death (SCD) or equivalent events (resuscitated sudden cardiac arrest or aborted SCD), heart failure-related events (heart transplantation, death from heart failure, rehospitalization for heart failure).
Results
Patients were 15±3 years of age at baseline and 65% were male. During a median follow-up of 61.7 months (IQR:39.2-84.5), 26(11.3%) patients with HCM reached the primary end points, 13 of whom experienced SCD events. LGE was present in 195(84.4%) childhood HCM patients with a median LGE extent 4.77%(IQR:2.01-9.18) and higher in the mid regions. After univariable analysis, multivariable Cox analysis adjusting clinical and cardiac functional factors, LGE extent was an independent predictor for the primary endpoints (HR, 1.11; P<0.01) and SCD event (HR, 1.09; P<0.01). Time-dependent receiver operating characteristic (ROC) analysis showed that LGE extent performed a superior efficacy of the five-year risk prediction of primary endpoints (AUC=0.80) and SCD event (AUC=0.82). Kaplan-Meier analysis showed a significantly increased risk of primary endpoints (log-rank P<0.01) and SCD event (log-rank P<0.01) in high LGE group (LGE extent>mean). LGE extent could improve the predictive model, such as HCM Risk-Kids model (C statistic 0.65 vs 0.79, P=0.011) and PRIMaCY SCD risk prediction model (C statistic 0.61 vs 0.80, P=0.014).
Conclusion
LGE assessed by CMR is a common characteristic and an independent predictor of adverse outcomes among childhood HCM, providing incremental value to improve risk stratification in current clinical risk models.