Late gadolinium enhancement distribution patterns in non-ischaemic dilated cardiomyopathy: genotype-phenotype correlation

Abstract

Aims: Late gadolinium enhancement (LGE) is frequently found in patients with dilated cardiomyopathy (DCM); there is little information about its frequency and distribution pattern according to the underlying genetic substrate. We sought to describe LGE patterns according to genotypes and to analyse the risk of major ventricular arrhythmias (MVA) according to patterns.

Methods and results: Cardiac magnetic resonance findings and LGE distribution according to genetics were performed in a cohort of 600 DCM patients followed at 20 Spanish centres. After exclusion of individuals with multiple causative gene variants or with variants in infrequent DCM-causing genes, 577 patients (34% females, mean age 53.5 years, left ventricular ejection fraction 36.9 ± 13.9%) conformed to the final cohort. A causative genetic variant was identified in 219 (38%) patients, and 147 (25.5%) had LGE. Significant differences were found comparing LGE patterns between genes (P < 0.001). LGE was absent or rare in patients with variants in TNNT2, RBM20, and MYH7 (0, 5, and 20%, respectively). Patients with variants in DMD, DSP, and FLNC showed a predominance of LGE subepicardial patterns (50, 41, and 18%, respectively), whereas patients with variants in TTN, BAG3, LMNA, and MYBPC3 showed unspecific LGE patterns. The genetic yield differed according to LGE patterns. Patients with subepicardial, lineal midwall, transmural, and right ventricular insertion points or with combinations of LGE patterns showed an increased risk of MVA compared with patients without LGE.

Conclusion: LGE patterns in DCM have a specific distribution according to the affected gene. Certain LGE patterns are associated with an increased risk of MVA and with an increased yield of genetic testing.

Keywords: cardiac magnetic resonance; dilated cardiomyopathy; genetics; late gadolinium enhancement; sudden cardiac death.

Graphical Abstract

A cohort of 577 individuals with DCM phenotyped with genetic testing and CMR was analysed. A causative genetic variant was identified in 219 (38%) patients, and 147 (25.5%) had LGE. LGE patterns in genetic DCM have a specific distribution. Patients with subepicardial, lineal midwall, transmural, and right ventricular insertion points or with combinations of LGE patterns showed an increased risk of MVA compared with patients without LGE. Abbreviations: CMR, cardiovascular magnetic resonance; DCM, dilated cardiomyopathy; HR, hazard ratio; IQR, interquartile range; LGE, late gadolinium enhancement; L.Midwall, lineal midwall; LVEF, left ventricular ejection fraction; MVA, major ventricular arrhythmias; RV, right ventricle.

Figure 1

Examples of LGE patterns extracted from the cohort. Abbreviations: LGE, late gadolinium enhancement; RV, right ventricle.

Figure 2

LGE patterns according to the affected gene in DCM. Simplified LGE pattern distribution based on the causative underlying gene shows a cluster in three categories: subepicardial, unspecific, and absent/rare. Abbreviations: DCM, dilated cardiomyopathy; LGE, late gadolinium enhancement.

Figure 3

LGE presence according to LVEF. The presence of LGE was not significantly associated with lower LVEF. Abbreviations: LGE, late gadolinium enhancement; LVEF, left ventricular ejection fraction.

Figure 4

Kaplan–Meier curve of MVA according to LGE patterns. Certain LGE patterns are associated with an increased risk of MVA. Abbreviations: CI, confidence interval; HR, hazard ratio; LGE, late gadolinium enhancement; L.Midwall, lineal midwall; MVA, major ventricular arrhythmias; RV, right ventricle.