Genetic Architecture of Acute Myocarditis and the Overlap With Inherited Cardiomyopathy

Abstract

Background: Acute myocarditis is an inflammatory condition that may herald the onset of dilated cardiomyopathy (DCM) or arrhythmogenic cardiomyopathy (ACM). We investigated the frequency and clinical consequences of DCM and ACM genetic variants in a population-based cohort of patients with acute myocarditis.

Methods: This was a population-based cohort of 336 consecutive patients with acute myocarditis enrolled in London and Maastricht. All participants underwent targeted DNA sequencing for well-characterized cardiomyopathy-associated genes with comparison to healthy controls (n=1053) sequenced on the same platform. Case ascertainment in England was assessed against national hospital admission data. The primary outcome was all-cause mortality.

Results: Variants that would be considered pathogenic if found in a patient with DCM or ACM were identified in 8% of myocarditis cases compared with <1% of healthy controls (P=0.0097). In the London cohort (n=230; median age, 33 years; 84% men), patients were representative of national myocarditis admissions (median age, 32 years; 71% men; 66% case ascertainment), and there was enrichment of rare truncating variants (tv) in ACM-associated genes (3.1% of cases versus 0.4% of controls; odds ratio, 8.2; P=0.001). This was driven predominantly by DSP-tv in patients with normal LV ejection fraction and ventricular arrhythmia. In Maastricht (n=106; median age, 54 years; 61% men), there was enrichment of rare truncating variants in DCM-associated genes, particularly TTN-tv, found in 7% (all with left ventricular ejection fraction <50%) compared with 1% in controls (odds ratio, 3.6; P=0.0116). Across both cohorts over a median of 5.0 years (interquartile range, 3.9-7.8 years), all-cause mortality was 5.4%. Two-thirds of deaths were cardiovascular, attributable to worsening heart failure (92%) or sudden cardiac death (8%). The 5-year mortality risk was 3.3% in genotype-negative patients versus 11.1% for genotype-positive patients (Padjusted=0.08).

Conclusions: We identified DCM- or ACM-associated genetic variants in 8% of patients with acute myocarditis. This was dominated by the identification of DSP-tv in those with normal left ventricular ejection fraction and TTN-tv in those with reduced left ventricular ejection fraction. Despite differences between cohorts, these variants have clinical implications for treatment, risk stratification, and family screening. Genetic counseling and testing should be considered in patients with acute myocarditis to help reassure the majority while improving the management of those with an underlying genetic variant.

Keywords: arrhythmogenic right ventricular dysplasia; cardiomyopathy, dilated; connectin; death, sudden, cardiac; desmoplakins; heart failure; myocarditis.

Figures

Figure 1.

Bar chart showing the age and sex distributions of patients with a primary or secondary diagnosis of acute myocarditis Across all hospitals in National Health Service England from 2016 to 2018. Age indicated in years.

Figure 2.

Detailed review of the genotype and phenotype of all cases with truncating variants of known pathogenicity in DSP. Cardiac magnetic resonance short-axis images (T2-STIR sequence for myocardial oedema, top; late gadolinium enhancement, bottom) are presented for the 3 cases from London and histopathology for the 1 case from Maastricht. Images are taken at 100 µm and demonstrate staining for (A) hematoxylin-eosin, (B) CD45, (C) CD68, and (D) Sirius red. ACM indicates arrhythmogenic cardiomyopathy; FHx, family history; gnomAD, Genome Aggregation Database; Het, heterogeneous; LVEF, left ventricular ejection fraction; P, patient; SCD, sudden cardiac death; and Trop, troponin.

Figure 3.

Distribution of LVEF assessed by cardiac magnetic resonance in patients with acute myocarditis recruited in London (cohort 1; n=230) and Maastricht (cohort 4; n=106) stratified by presence of likely pathogenic variants in ACM- and DCM-associated genes. Black lines indicate median; blue shading shows interquartile range. Dots refer to individual patients. Note that genes with a single patient affected have left ventricular ejection fraction (LVEF) shown as an absolute value (applies to DSG2, JUP, and TNNC1). ACM indicates arrhythmogenic cardiomyopathy; and DCM, dilated cardiomyopathy.

Figure 4.

Cumulative incidence curves for the study end points over follow-up by the presence or absence of a protein-altering variant across both cohorts. Major arrhythmia composite includes hemodynamically unstable ventricular tachycardia, aborted sudden cardiac death, implantable cardioverter defibrillator, or heart block (second or third degree). Major heart failure composite includes heart transplantation, left ventricular assist device implantation, or heart failure hospitalization.