Background and purpose: The accuracy of prevalent risk models recommended by American Heart Association/American College of Cardiology(AHA/ACC) and European Society of Cardiology(ESC) for hypertrophic cardiomyopathy(HCM) patients is suboptimal. We aim to assess the utility of cardiac magnetic resonance(CMR) -3D left ventricular mechanical dispersion(LVMD) in the prognostic stratification of HCM, and explore the correlation with myocardial fibrosis heterogeneity. Methods and results: 159 HCM patients were retrospectivly enrolled, 31 experienced primary endpoint events, 49 experienced secondary endpoint events. 3D LVMD were computed as the standard deviations(SD) of the time from the peak of the R wave and the negative strain peak across 16 segments longitudinally, circumferentially, and radially on cine images. The heterogeneity parameter entropy was calculated utilizing the generic Python algorithm. Patients experienced endpoint events exhibited significantly higher radial LVMD(LVMD-R), circumferential LVMD(LVMD-C), and longitudinal LVMD(LVMD-L) (p<0.001). Kaplan-Meier analyses demonstrated that HCM patients with elevated LVMD had a higher risk of primary and secondary endpoint events(p<0.05). In multivariable Cox analysis incorporated the 2024 AHA/ACC and 2022 ESC guidelines risk classification, both LVMD-C and LVMD-L emerged as significant predictors for both endpoint events(p<0.05). We developed a model incorporated the 2022 ESC risk stratification in combination with LVMD-C and LVMD-L. ROC analyses revealed that this combined model(area under the curve[AUC] for primary endpoint events: 0.854, AUC for secondary endpoint events: 0.802) significantly outperformed the guidelines alone(Delong test, p<0.001). Additionally, entropy demonstrated linear correlations with LVMD. Conclusion: The inclusion of LVMD evaluation in HCM patients adds further value to risk stratification beyond existing guidelines.