RESUMO
BACKGROUND: Recently, a new risk model was developed, namely hypertrophic cardiomyopathy (HCM) risk for cerebrovascular accident, for estimating the risk of thromboembolism (TE) in patients with HCM. There is no study about the external validation of this model. METHODS: We evaluated the performance of the model for predicting TE in 417 patients with HCM recruited between 2008 and 2016, from a tertiary referral centre. The primary end point was 5-year TE, and the risk was calculated using the model formula. RESULTS: During a median follow-up of 3.5 years, 25 (6.0%) patients reached the TE end point, and 22 (5.3%) patients within the first 5 years. Within a 5-year time frame, the model showed a possibly helpful discrimination for TE (C-index for the whole cohort: 0.67, C-index for the subgroup without atrial fibrillation: 0.67) relative to its original C-index of 0.75. However, the calibration was not perfect, which suggested that there was an underestimation of 5-year TE risk in the whole cohort and different risk groups. CONCLUSIONS: HCM risk for cerebrovascular accident demonstrated a possibly helpful discrimination for TE when applied in a new set of patients with HCM. However, the accurate estimation of absolute risk should be explored in future studies.
