Background: Heterogeneous phenotypic expression in hypertrophic cardiomyopathy (HCM) results in challenging risk stratification. Risk stratification is limited due to lack of hypertrophy visualized by transthoracic echocardiography (TTE) particularly of the lateral wall and apical hypertrophy. Cardiac magnetic resonance (CMR) provides better visualization and more accurate assessment of LV wall thickness and mass. In addition, the role of LV mass in risk stratification of HCM has not been clearly understood.
Objective: There were two main objectives of the study. Firstly, to investigate the effect of maximum LV wall thickness (LVWTmax) measured by MRI and TTE on the 5-year ESC risk of sudden cardiac death. Secondly, to investigate the relationship of indexed LVmass on CMR with the incidence of insertion of an implantable cardioverter defibrillator (ICD).
Methods: HCM subjects were retrospectively analysed for a mean duration of 5.7 ± 3.3 years in a single tertiary centre. LVmass and LVWTmax on CMR and TTE were measured and ESC risk scores were calculated.
Results: 250 HCM patients (males =178, females = 72) were followed up for 5.7 ± 3.3 years. Age of diagnosis was 55.8 ± 17.1 years.
CMR versus TTE based ESC risk score increased the likelihood of ICD implantation (36% versus 12.8%, p=<0.01) due to higher measured LVWTmax secondary to a better resolution. Higher risk group (ESC risk 34%) on CMR had higher LVmass (92.9 ± 38.4 g/m2 versus 78.5 ± 5.2 g/m2, p = 0.01) when compared to the low-risk group.
LVmass was also higher (88.3 ± 32.4 g/m2 versus 71.7 ± 38 g/m2, p<0.01) in the subjects with ventricular arrhythmias (n=69) with no differences seen for LVWT.
Forty-three patients had ICD implanted during follow up (primary = 36, secondary = 7). LVmass was higher for patients who received an ICD (92.6 ± 31.1 g/m2 versus 79.4 ± 29 g/m2, p = 0.01).
ROC analysis of the variables shows that MRI based ESC risk score, MRI LVWTmax, and MRI LVmass are superior to Echo based ESC risk score and Echo LVWTmax as tests for identifying possible high-risk patients.
Using multivariate analysis, MRI LVWTmax (Odds ratio = 1.34, 95% CI: 1.18–1.53, p=<0.001) was an independent predictor for ICD insertion in patients with HCM.
Conclusion: LVWT derived from MRI is more likely to provide an accurate assessment of risk calculated by the ESC score due to improved myocardial resolution. LVmass is more likely to identify subjects with a higher risk of ventricular arrhythmias as opposed to LVWT on TTE or MRI. Further prospective randomised studies need to be performed to establish if LVmass is a predictor of ventricular arrhythmias.
