Abstract 16697: Trends in Incidence, Prevalence and Outcomes of Hypertrophic Cardiomyopathy in a Real-World Community Based Cohort

Abstract

Introduction: Hypertrophic cardiomyopathy (HCM) is a common cause of sudden cardiac death (SCD) in young individuals. However, data on incidence of SCD has largely come from large tertiary hospital referral populations, which may capture a higher proportion of sicker patients. Accordingly, we performed a community-based study of SCD in HCM. Methods: We collected data from all persons within the eREP (extended Rochester Epidemiology Project), a community-based population cohort between 2010-2022 with an ICD code diagnosis of HCM followed by manual chart review to confirm HCM by the AHA diagnostic criteria. Incidence and prevalence of HCM and SCD per 100,000 person years, and rate of all-cause and cause specific mortality were reported. Cox proportional hazards model was used to compare the performance of SCD prediction scores. Results: 805 patients (46.3% women and median [IQR] age 62 [51-72] yrs) met the HCM diagnostic criteria with an average incidence of 8.7 (95% CI: 7.4-10) per 100,000 person years and a prevalence of 65.8 (95% CI: 57.3-74.4) per 100,000. At a median (IQR) follow-up of 8.9 (4.4-15.1) yrs, there were 226 (28%) deaths with 14 (6.2%) from SCD with a biphasic distribution in incidence of HCM and consistent decline in the SCD rate ( Fig 1a-b ). Distribution of cause of death was similar among women and men ( Fig 1c-d ). Probability/risk of non-SCD mortality was higher throughout the follow-up period ( Fig 2a-b ). HCM related CV mortality constituted 45% of the deaths with HF being the leading cause in both women and men ( Fig 2c ). None of the risk prediction scores were significantly able to predict SCD ( Fig 2d ). Conclusion: The contemporary incidence and prevalence of HCM in the community is much lower than previously reported rates, as is the rate of HCM related SCD regardless of gender. Patients with HCM are most likely to die of non-HCM related conditions or of HF rather than SCD. Further, existing SCD prediction tools are ineffective in the real world setting. These results may help guide/revisit resource allocation among patients with HCM and guide further investigation based on unmet needs.