Abstract
BACKGROUND: Atrial arrhythmia is prevalent in patients with hypertrophic cardiomyopathy (HCM) and is prognostically deleterious. Catheter ablation (CA) has poor efficacy in HCM, making predictors of outcome valuable. Left atrial size is associated with morbidity and mortality, and E/e' has been proposed. The current study evaluates whether these predict non-paroxysmal atrial arrhythmia, interventions for atrial arrhythmia and all-cause mortality. METHODS: 233 patients from the echocardiography database were included; patients with any mitral annular calcification were excluded. Medical records were reviewed for sample characteristics. Survival analysis was performed for left atrial diameter (LAD), left atrial volume index (LAVi, by area-length method), and E/e' with respect to endpoints, namely: non-paroxysmal atrial arrhythmia; atrial intervention (composite of CA and pace-and-ablate); and all-cause mortality. If LAD or LAVi were significant in multivariate analysis, ROC analysis and DeLong's test were performed to evaluate and compare their discriminative power. RESULTS: Over a median follow-up of 8 years (median age 53, 73% male), the overall prevalence of non-paroxysmal atrial arrhythmia was 31%. In multivariate models, LAVi was predictive of non-paroxysmal atrial arrhythmia and atrial intervention; LAD was only predictive of arrhythmia. LAVi demonstrated greater discriminative power for predicting atrial intervention. E/e' was not predictive of any outcome. No echocardiographic variable predicted death in multivariate analysis. CONCLUSION: LAVi may have greater clinical utility compared to two-dimensional measurements like LAD. Future work should clarify which measures of left atrial size are most appropriate in clinical practice.