# Comprehensive risk stratification model for sudden death in hypertrophic cardiomyopathy: integration of cardiac magnetic resonance and clinical parameters

## TL;DR

A new model combining MRI scans and clinical data better predicts sudden death risk in heart patients than existing methods.

## Contribution

A novel risk prediction model integrating CMR parameters with clinical variables improves sudden cardiac death risk stratification in HCM.

## Key findings

- The comprehensive model achieved a C-index of 0.607, outperforming traditional and clinical-only models.
- CMR ejection fraction and left ventricular diastolic pressure were identified as independent predictors of sudden death.
- The model effectively stratified patients into low, intermediate, and high-risk groups with significant discrimination.

## Abstract

Current risk stratification for sudden cardiac death (SCD) in hypertrophic cardiomyopathy (HCM) relies primarily on the HCM Risk-SCD score, but its predictive accuracy remains suboptimal.

This study aimed to develop and validate a comprehensive risk prediction model integrating cardiac magnetic resonance (CMR) parameters with clinical and biomarker variables.

We analyzed 576 consecutive HCM patients from a tertiary referral center. The primary endpoint was sudden death or appropriate implantable cardioverter-defibrillator therapy. We developed four prediction models: (1) Traditional (HCM Risk-SCD alone), (2) Clinical (traditional + clinical variables), (3) CMR (traditional + CMR parameters), and (4) Comprehensive (integrating all variables). Model performance was assessed using C-index and time-dependent receiver operating characteristic (ROC) analysis.

During median follow-up of 3 years, 58 patients (10.1%) experienced the primary endpoint. The comprehensive model demonstrated superior performance (C-index 0.607) compared to traditional (0.565), clinical (0.598), and CMR (0.607) models. In multivariable analysis, CMR ejection fraction (HR: 0.94, 95% CI: 0.91–0.97, P < 0.001) and left ventricular diastolic pressure (HR: 0.94, 95% CI: 0.89–0.98, P = 0.010) were independent predictors. Time-dependent ROC analysis showed maintained predictive accuracy over 3 years (AUC 0.78–0.85). Risk stratification using the comprehensive model effectively discriminated low, intermediate, and high-risk groups (log-rank P < 0.001).

Integration of CMR parameters with clinical variables significantly improves SCD risk prediction in HCM compared to traditional risk stratification. The comprehensive model provides enhanced accuracy for identifying high-risk patients who may benefit from primary prevention implantable cardioverter-defibrillator therapy.