Genetics, Clinical Features, and Long-Term Outcome of Noncompaction Cardiomyopathy

Background

Noncompaction cardiomyopathy (NCCM), which is also known as left ventricular noncompaction (LVNC), is a cardiomyopathy with excessive trabeculations of the LV, with a >2-fold thickening of the endocardial noncompacted layer compared with the epicardial compacted layer of the myocardium (NC/C >2). The clinical outcomes of noncompaction cardiomyopathy (NCCM) range from asymptomatic to heart failure, arrhythmias, and sudden cardiac death (Elliott 2008).. Genetics play an important role in NCCM because up to 50% of patients have a family member with cardiomyopathy, This study investigated the correlations among genetics, clinical features, and outcomes in adults and children diagnosed with NCCM.a

Methods

A retrospective multicenter study from 4 cardiogenetic centers in the Netherlands. Diagnosis of NCCM was based on consensus of re-evaluated echocardiography and magnetic resonance imaging (MRI) according to the Jenni and Petersen criteria (Jenny 2001).

DNA was isolated from peripheral blood of the patients to analyze coding regions of 45 core cardiomyopathy genes (ABCC9, ACTC1, ACTN2, ANKRD1, BAG3, CALR3, CRYAB, CSRP3, DES, DMD, DSC2, DSG2, DSP, EMD, GLA, JPH2, JUP, LAMA4, LAMP2, LMNA, MYBPC3, MYH6, MYH7, MYL2, MYL3, MYPN, MYOZ1, MYOZ2, PKP2, PLN, PRKAG2, RBM20, RYR2, SCN5A, SGCD, TAZ, TCAP, TMEM43, TNNC1, TNNI3, TNNT2, TPM1, TTN, VCL and LDB3).

Patients were classified as genetic if they had a likely pathogenic mutation; as probably genetic if they had a family history of cardiomyopathy and if DNA testing did not identify a mutation; and as sporadic if had no mutation or family history were present. Patients were classified as familial, if at least 1 first-degree or 2 second-degree relatives were reported to have cardiomyopathy. In families with multiple cases of NCCM, only the index case (i.e. the proband), was included in the study.
Clinical data and Information on the occurrence of clinical events at follow-up was collected from the medical records. The occurrence of cardiac death, implantation of a LV assistance device, heart transplantation, (aborted) sudden cardiac death, appropriate implantable cardioverter-defibrillator shock, or ischemic stroke were classified as MACE.

Clinical features and major adverse cardiac events (MACE) during follow-up were compared across the children and adults.

Results

A total of 327 unrelated NCCM patients were classified into 3 categories: 1) genetic, 32% (81 adults; 23 children) of patients; 2) probably genetic in 16% (45 adults; 8 children) of patients; or 3) sporadic, in 52% (149 adults; 21 children) of patients.

The median age at diagnosis of NCCM was 41 years (range: 0 to 79 years. Only 16 (30% of the children) were diagnosed at younger than 1 year of age.

Of the 104 genetic cases, i.e. pathogenic variant found on gene testing. In 82% the mutation involved a sarcomere gene: MYH7, MYBPC3, and TTN mutations were the most common mutations (71%) found in genetic NCCM.

Mutations (45 vs. 30%, p = 0.04) as well as complex genotypes were more frequent in children and were associated with MACE (p = 0.025). Adults were more likely to have sporadic NCCM (54% vs. 40% in children). Mutations occurred more frequently in female patients irrespective of age at diagnosis (males: 27% vs. females: 38%; p = 0.039). Of the 120 familial cases, 56% (n = 67) had a pathogenic mutation as compared to only 18% of the 207 patients without a family history of cardiomyopathy (p < 0.001).

The predominant phenotypes in family members were DCM and HCM. Sudden death of a relative (younger than 50 years) was reported in 7% (n = 23) of the families.

Congenital heart defect (CHD) was observed in 9% of the patients (n = 28), more often in children (p = 0.027) who had atrial septal defects (ASD) and ventricular septal defects (VSD). Two children and 2 adults had Ebstein anomaly and an MYH7 mutation.
The prevailing majority (~85%) of the patients studied were symptomatic on presentation. Heart failure (27%) and arrhythmias (26%) were the most common presentations in children and adults, 4% presented with sudden death and 3% with thromboembolism. LV systolic dysfunction occurred in 44% of the children and 58% of the adult patients. The risk of having reduced (LV) systolic dysfunction was higher for genetic patients compared with the probably genetic and sporadic cases (p = 0.024). The highest risk was detected in patients with multiple mutations and TTN mutations.
RV dysfunction was less common. Hypertension and LBBB were more common in adults with sporadic NCCM.

The median follow-up was 60 months (IQR: 18 to 113 months). MACE occurred in 27% children, and 21% adults, during a median follow-up of 25 months (IQR: 4 to 58 months). An increased risk of MACE was observed in children with genetic NCCM (p = 0.025), children diagnosed at age younger than 1 year, and patients with LV systolic dysfunction. To the contrary, the risk for adverse events in children with sporadic NCCM was low. In adults the differences in outcome appeared to be less pronounced among the NCCM subgroups, when comparing to children, Interestingly, all patients with MYH7 mutations had low risk for MACE (p = 0.03).

Cardiac arrest and ischemic stroke occurred significantly more in female patients than in male patients (11% vs. 2%; p = 0.003, and 9% vs. 3%; p = 0.045, respectively).

Stroke was also associated with congenital heart disease (with CHD: 56% vs. without CHD: 5%; p = 0.001).

Discussion

The authors conclude that NCCM is a heterogeneous condition, and that genetic stratification has a role in clinical care: in prediction of risk for MACE, to guide management and follow-up of patients and their relatives.

MACE were quite common occurring at ~1% per month. Since this study enrolled mostly symptomatic patients and included only the index cases in families, it potentially overestimated the disease severity and its outcomes. Sporadic asymptomatic cases, such as those incidentally diagnosed during screening, should have a better prognosis but were obviously underrepresented in these series.

In approximately one-half (48%) of the NCCM cohort genetics played a role either through a definite gene mutation or a familial disease without a mutation. Sporadic NCCM was more prevalent in adults than in children, suggesting that apart from genetic causes for NCCM, acquired (nongenetic) causes for hypertrabeculation might play a role (Anderson 2017).

Because acquired causes for NCCM are identified in a large proportion of adult patients, the prevailing assumption that NCCM is an embryonic developmental disorder, might not be true. Alternatively, late-onset NCCM might be explained by an enhancement of a latent asymptomatic congenital defect through disruptions of cardiac homeostasis later in life. Recent studies endorsed the hypothesis that acquired causes such as increased loading conditions might lead to characteristic hypertrabeculation. Genetic variants such as TTN mutations may constitute a genetic background predisposing for late-onset expression of NCCM after an additional stress.

Finally, the proportion of genetic patients may be more than 48% reported in this study. Some of the patients without a mutation were Sanger-tested for a small number of cardiomyopathy genes, while others (41%) had VUS that could be reclassified in the future as likely pathogenic. Novel cardiomyopathy genes, to be identified in future, might play a role in the familial cases without a mutation, as well as in some of the sporadic cases.

An accompanying editorial (Oechslin and Jenny 2018) discuss the added value of this elaborate and so far the largest study on NCCM. The phenotype of noncompaction may constitute a physiologic, reversible remodeling of the LV myocardium with prominent trabeculations in athletes or pregnant women as opposed to a pathologic remodeling in patients with a cardiomyopathy of any cause including pathological loading conditions.

The association between a genetic mutation and LV dysfunction as a risk factor for outcome, with LVNC as the myocardial phenotype, strengthens the hypothesis that LVNC is a genetically determined cardiomyopathy. The results of this study emphasize the importance of routine genetic testing for diagnosis, risk stratification and to establish a genotype–phenotype in LVNC.

On the other hand, even if hypertrabeculations result in the morphological appearance of LVNC, the patients should not unnecessarily labeled as NCCM (Gati 2003). This is in analogy of hypertrophy of the myocardium which does not automatically imply hypertrophic cardiomyopathy. The challenge is to differentiate physiologic remodeling from pathologic remodeling of the myocardium (Oechslin 2017). Multimodality imaging, gene testing and deconditioning, as well as family screen, might be useful to resolve this diagnostic problem.

Another issue not yet resolved is why a given mutation results in different phenotypic expression among family members with cardiomyopathy. Genetic modifiers, not yet identified, may be those leading to morphological appearance of LVNC in patients with genetic cardiomyopathy as well as under loading conditions in apparently normal individuals.