Cardiac MRI-derived left atrial function and parameters for predicting new onset atrial fibrillation in patients with hypertrophic cardiomyopathy
P Bunditboondee, S Kongrat, V Prakongwong, T Lueangklanlayanakhun, W Prasertkulchai, T TangcharoenAbstract
Background
Atrial fibrillation (AF) is a common comorbidity in patients with hypertrophic cardiomyopathy (HCM), 4-6-fold more prevalent than in age-matched patients without HCM. Incident AF is clinically important because anticoagulation treatment is needed to prevent stroke. The cardiac substrate for AF can be structural and electrical changes of the left atrium (LA). Cardiac MRI (CMR) can provide comprehensive atrial structural and functional information, including left atrial strain and left atrial enhancement, which may be associated with the progression of atrial fibrillation.
Purpose
This study aims to evaluate whether comprehensive left atrial structural and functional parameters from cardiac MRI are predictors of atrial fibrillation in patients with hypertrophic cardiomyopathy. We also aim to compare CMR parameters with the conventional HCM-AF score for predicting new-onset AF.
Methods
Single-center, retrospective cohort study, collecting data from identified hypertrophic cardiomyopathy patients by CMR without documented atrial fibrillation at baseline during January 2011-June 2025
Results
138 enrolled patients with new-onset of AF in 32 patients (23.19%) following CMR. The incidence of AF is 5.84%/year with a median follow-up time of 3.97 years. By univariate analysis, CMR parameters associated with incidence of AF are maximum LA volume index (HR 1.02, P 0.005), minimum LA volume index (HR 1.02, HR 0.007), LA reservoir strain (HR 0.95, P 0.048), LA conduit strain (HR 0.89, P 0.007), left ventricular stroke volume (LVSV) index (HR 1.05, P 0.006) and mitral regurgitation (MR) volume (HR 1.05, P 0.03). For multivariate models, maximum LA volume index (HR 1.02, P 0.035), LVSV index, and MR volume are independent factors. While LA conduit strain is lower in developed AF patients (12.37±4.83 VS 9.96±4.38, HR 0.89, p = 0.007). Nevertheless, LA hyperenhancement and LA pumping strain do not differ between AF and non-AF patients (p = 0.424 and p = 0.632, respectively). The AUC of the conventional HCM-AF score for incident AF diagnosis in the high-risk (score ≥ 22) and low-to-intermediate-risk (score < 22) scores are 0.625 and 0.374, respectively. Incremental value analysis by adding CMR parameter (maximum LA volume index and LVSV index) could increase the diagnostic performance, especially in low/intermediate risk patients (AUC from 0.374 to 0.695)
Conclusions
In HCM patients, maximum LA volume index, LVSV index, and MR volume by CMR are independent predictors of new-onset AF. While the higher LA conduit strain is associated with lower AF incidence. In addition, adding the maximum LA volume index and LVSV index to the conventional HCM-AF score enhances the performance of the conventional score, especially in the low/intermediate risk group.table univariate and multivariate analysisgraph area under ROC curve in HCM-AF score