Methods: Two populations with a diagnosis of HCM undergoing CMR were identified for retrospective validation (Derivation cohort n=109) and an additional prospective evaluation (Validation cohort n=109). CMR predictors of AF were evaluated with logistic regression analysis. Receiver operating characteristic (ROC) curves were used to determine the diagnostic accuracy of the identified CMR predictors of AF, and cut-off values with optimal sensitivity and specificity were calculated. A goodness-of-fit was applied in the Derivation cohort to evaluate how well the models were calibrated.
Results:Median age was 60±13 years, 51% of the patients were male. 15% of the population had a documented episode of AF (n=33). Risk of AF was associated with higher left atrial volume index (LAVi) (OR=1.09, CI=1.05-1.14; p<0.001) and a reduced mitral annular plane systolic excursion(MAPSE) (OR=0.41, CI=0.27-0.63; p<0.001, Figure A). Moreover, the amount of LGE in the myocardial basal segments was correlated with AF risk (OR=1.018, CI=1.004-1.033; p=0.014). The area under the ROC curve for LAVi (cut-off 50 ml/m2) was 0.776 (p<0.001; sensitivity=88% and specificity=67%) while for MAPSE (cut-off 10 mm) was 0.809 (p<0.001; sensitivity=88 % and specificity=74%). The area under the ROC curve for the LAVi/MAPSE ratio (cut-off=5.0) was 0.820 (p<0.001; sensitivity=93% and specificity=71%, Figure B). These results were confirmed in the Validation cohort with a goodness-of-fit test that showed a good calibration in AF prediction between cohorts.
Conclusion: CMR assessment of MAPSE holds promise in refining risk assessment of AF in HCM patients. Combining structural and functional information in a simple ratio (LAVi/MAPSE) may represent a novel risk stratification tool with high predictive accuracy to detect patients with increased risk of AF (Figure C).