Understanding Noncompaction Cardiomyopathy: A Brief Comprehensive Review of A Controversial Entity

Noncompaction cardiomyopathy is a heterogeneous and complex entity characterized by hypertrabeculation, typically of the left ventricle. Uncertainties regarding pathogenesis, classification as primary genetic or unclassified cardiomyopathy, diagnostic criteria, and risk stratification have contributed to fuel the discussion surrounding this disorder. Meanwhile, noncompaction phenotype is thought to be the morphological expression of different underlying pathophysiological mechanisms, genetics, and pathologies. Recent studies suggest that distinguishing genetic from nongenetic causes allows risk stratification and may support clinical management and counselling of patients and their relatives. Additionally, advanced cardiac imaging techniques have demonstrated a complementary role in outcome prediction. The purpose of this review is to provide a brief comprehensive review of this controversial entity.


Introduction
Noncompaction cardiomyopathy (NCCM), first described by Grant in 1926 as "left ventricular noncompaction" 1 , is a heterogeneous myocardial disorder characterized by prominent trabeculae, intratrabecular recesses and a bilayered myocardium composed of a compacted and noncompacted layer 2, 3 . It has been the subject of several studies and publications, but numerous controversies remain 3,4 . While NCCM is classified as a distinct primary genetic cardiomyopathy by the American Heart Association 5 , it is regarded as unclassified familial cardiomyopathy by the European Society of Cardiology 6 . Indeed, despite being traditionally a familial presentation entity it may occur sporadically, isolated or associated with other congenital defects, affecting only the left ventricle (LV) or both 4,7,8 . Moreover, noncompaction may also present as a congenital or acquired morphological trait shared by many distinct cardiomyopathies (e.g. hypertrophic (HCM), dilated (DCM)) 6,9 and may become evident in some pressure/volume overload conditions 10,11 . In the current paper, we aim to provide a brief and comprehensive review of this controversial disorder.

Pathophysiology
It is postulated that NCCM is the result of abnormal intrauterine heart compaction 3,4,13 . Apical trabeculations appear around day 26 of gestation and, by the seventh week, both ventricles are similarly and primarily filled with a trabeculation network, with at least twice the thickness of the compact ventricular layer 13 . This is believed to enhance oxygen and nutrients exchange, before coronary artery development, and to compartmentalize the blood before heart septation. Further on, ventricular thickening and remodelling occurs as the compact layer develops through different processes of proliferation, hypertrophy, and layering, to form the specialized contractile structure 14 . The molecular and cellular mechanisms involved are not completely understood and it is not clear how the mesh of ventricular trabeculations contributes to the compact layer development 13,14 . However, several signalling pathways (e.g. NOTCH) involving many genes suspected to contribute to this process have been described, specifically BMP10 (bone morphogenic protein-10) and Nkx2.5 that, when knocked out in mice, are associated with hypertrabeculation [15][16][17] . Moreover, it was speculated that hypertrabeculation may result from a compensation mechanism of impaired attempt of myocardial growth and hypertrophy: abnormal cardiomyocyte adhesion due to gap junction malfunction or altered endocardium signalling leading to persisting sinusoids penetrating the ventricular wall 18 .
Nonetheless, some findings oppose this "embryogenic hypothesis" 15,18 . In fact, hypertrabeculation may also develop during adulthood in a normally compacted LV as an adaptive response to abnormal loading conditions (e.g. athletes, pregnancy, anaemia): a more trabeculated ventricle allows higher stroke volumes with lower wall stress 7,19,20 . Gap junction malfunction due to intraventricular pressure and microinfarcts secondary to vascularization mismatch in a compensating hypertrophic ventricle are mechanisms that may contribute to acquired hypertrabeculation 19,21 .

Genetics
Genetics plays an important role in NCCM since in up to 50% of patients there is evidence of a genetic cause, whether because a cardiomyopathy mutation is found and/or because at least one family member presents nonischemic cardiomyopathy 9,22,23 .
The genetic causes are heterogeneous but share a final common pathway, similar to other types of cardiomyopathy with multiple causes and several mechanisms at molecular and cellular levels 3,8,16 . In the familial form, autosomal dominant transmission (with reduced penetrance) is the most common mode of transmission 7,22 , followed by X-linked, mitochondrial inheritance and chromosomal abnormalities 8,24,25 . In a systematic review of 541 patients, van Waning et al. 25 showed that autosomal dominant inheritance (mostly missense mutations) was the most frequent pattern of transmission (83%), with more than half of the genetic defects being reported in sarcomere genes (MYH7, MYBPC3, ACTC1, and TTN). These mutations were usually present in adults and associated with less severe outcomes. On the other hand, in 7% of the patients an X-linked inherited genetic defect was reported (mostly in TAZ, causing Barth's syndrome), a mitochondrial defect was found in 7%, while 6% presented a chromosome defect (frequently 1p36 deletion). These abnormalities were mostly observed in children and were associated with a higher risk for severe outcomes, congenital heart defects (CHD), and neuromuscular symptoms 25 .
Meanwhile, as in sporadic forms, a pathological genetic defect may be also absent in a considerable number of patients with familial NCCM, indicating that many genetic causes are still unknown 4,9 . However, a recent retrospective analysis showed that the risk of left ventricular systolic dysfunction (LVSD) was lower in these two scenarios compared to patients with a mutation and that LVSD was associated with cardiac events in mutation carriers but not in sporadic cases 7 .

Epidemiology
The real incidence and prevalence of NCCM remain unknown 3 . In adults, the prevalence may vary from 0.14% in the general population up to 3.7% in patients with LVSD 26,27 . In children, it represents 4.8% to 9% of all cardiomyopathies, the third after DCM and HCM 12,28 .

Diagnosis
The morphological hallmark of NCCM consists of prominent trabeculae, intratrabecular recesses and a bilayered myocardium composed of a compacted and noncompacted layers 2,3 . Despite lacking gold standard criteria, the diagnosis mostly relies on non-invasive imaging techniques, usually transthoracic echocardiography (TTE) and cardiac magnetic resonance (CMR) 3,4,29 . Additionally, there is evidence that these features may represent a remodelling process secondary to abnormal volume/ pressure loading conditions, whether physiological and transient (pregnancy, athleticism) or pathological (hypertension, sickle cell disease) 7,10,11,29 , and that a phenotypic overlapping with other cardiomyopathies may also be seen 3,7,9,29 . Considering the hypothesis that noncompaction may represent an epiphenomenon under certain conditions, we agree with Negri et al. 29 that the fulfilment of imaging criteria should be complemented with clinical (including family history), electrocardiographic, and genetic findings to establish the diagnosis. Accordingly, we propose the use of a multifactorial algorithm to establish the diagnosis of NCCM (Figure 1). This complete evaluation also allows the establishment of eight phenotypical subtypes of NCCM, as initially proposed by Towbin et al. 3 , each one being associated with different outcomes. A summarized description of the NCCM subtypes, based on simple cardiac imaging measurements (e.g. LV wall size and thickness), co-existing findings such as CHD or arrhythmias, and corresponding prognosis, is described in Table 1.

Electrocardiogram
Electrocardiographic abnormalities can be seen in more than 80% of cases, most frequently including ventricular hypertrophy voltage criteria, T-wave inversion, ST-segment strain, prolonged QT, fragmented QRS, right/ left bundle branch block, and pre-excitation 3,22,26 . Regular Holter monitoring should be used in symptomatic patients, especially in case of LSVD, to identify clinically relevant arrhythmias 24,26 .

Cardiac imaging
Transthoracic echocardiography remains the most common and initial diagnostic strategy, mainly due to its widespread availability, ease of interpretability, and low cost 3,36 . It allows a simple and reliable evaluation of ventricular size and function, wall thickness, and eventually associated CHD. Among the echocardiographic criteria, those proposed by Chin et al. 2 and by Jenni et al. 37 are the most widely used although the latter is usually preferred given its higher specificity and necropsy validation 4,29 . Jenni et al. 37 consider for the diagnosis of NCCM the presence of a bilayered myocardial structure consisting of a compacted thin epicardial band (C) and a much thicker non-compacted endocardial layer (NC), with a maximal NC/C ratio >2 at end-systole on short-axis parasternal views, in association with deep perfused intertrabecular recesses on colour Doppler. In this study, abnormalities were typically located in the mid-lateral, mid-inferior, and apical LV areas, and patients with CHD were excluded. As stated, the right ventricle may also be involved 4,26 , although no recognized criteria are available 3 . Contrast-enhanced TTE may be useful in cases of poor acoustic window and in differentiating NCCM from HCM 36 . Recently, advanced techniques (strain, strain rate, torsion) are being used 3,4 and the presence of base-apex longitudinal strain gradient may assist in differentiating between NCCM and DCM 38 .
Given the higher quality and sensitivity in the detection of myocardial trabeculations, CMR has an important role in the confirmation of NCCM and also allows the identification/exclusion of thrombi and other coexisting abnormalities 4,36 . The same controversies remain regarding diagnostic criteria, however, the most used are those of Peterson et al. 39 , consisting of a NC/C myocardium ratio >2.3 at end-diastole, and Jacquier et al. 40 which requires a NC ventricular mass >20% of the global LV mass for diagnosis. Aside from ventricular strain 36 , CMR also allows tissue characterization using late gadolinium enhancement (LGE) and T1-mapping to identify myocardial fibrosis, which could behave as a prognostic marker 4,30 . In a multicentre study by Andreini et al. 41 , LGE (replacement fibrosis), LVSD and LV dilation were independently associated with cardiac events, while the degree of trabeculation was not. Additionally, T1-mapping demonstrates an expansion of the extracellular space by diffuse myocardial fibrosis, normally not identifiable by LGE (less sensitive), which is correlated with myocardial dysfunction and VA 42 .
Finally, cardiac computed tomography is useful for the anatomical and functional assessment of both ventricles and allows the exclusion of coronary artery disease 3,4 , which is typically absent 3,16 .

Genetic testing
Although the diagnostic yield is usually low 7,34,43 , genetic screening should be offered to (index) patients with an established clinical diagnosis of NCCM 44,45 . Additionally, recent evidence emphasizes the importance of routine genetic screening given its role in risk stratification and prognosis 7,25 .

Treatment
There are no specific guidelines for the therapeutic management of NCCM, so it is largely dictated by clinical findings 3 . In case of HF, patients with LSVD should be managed based on existing recommendationstypically including oral angiotensin-converting enzyme inhibitors, β blockers, and aldosterone antagonist 46 , while treatment of those presenting preserved ejection fraction is more challenging and targeted on symptom relief and comorbidities 47 . Loop diuretics should be used in case of pulmonary or peripheral congestion or volume overload 46,47 . Cardiac resynchronisation may be also considered and seems to provide beneficial clinical effects and LV function improvement 48 . Ventricular assist devices and cardiac transplantation are possibilities for end-stage disease 3 .
As NCCM carries an increased risk of VA and SCD 7,26,30,34 , implantable cardioverter defibrillator is recommended in case of ventricular tachyarrhythmias (VT) associated with syncope or SCD and should be implanted in the presence of non-sustained VT and LVSD 45 .
Despite the association with increased thromboembolic risk 23,30 , routine use of anticoagulants is still a matter of debate 4 . While oral anticoagulation is recommended in patients with atrial fibrillation and/or previous embolic event 45 , it should be used in case of intracardiac thrombi 49 and may be reasonable in individuals with LVSD 30,45,49 . Vitamin-K antagonists have been advocated as a preferred choice due to some concerns regarding non-vitamin-K oral anticoagulants 49 , although these drugs proved to be safer and equally effective in adults with non-valvular atrial fibrillation 50 .

Family Screening
Given that up to 50% of cases may be familial 9,22,23 , mutation-specific testing is recommended for family members and appropriate relatives following the identification of a NCCM-causative mutation in the index case 44,45 . Regular cardiac follow-up (including TTE and electrocardiogram) is recommended for family members presenting the same mutation and for first-degree relatives of an index case with no mutation or not tested, since childhood up to 60 years-old 43 .

Conclusions
Almost a century has passed since its first description and NCCM remains a chameleonic entity: from aetiopathogenesis to clinical presentation and prognosis. Recently, the emphasis has been given to genetics and genotype-phenotype relation, with many studies suggesting that noncompaction may be found in a considerable proportion of patients without causative mutation or familial history (sporadic) as a consequence of ventricular remodelling due to certain physiological/pathologic overload conditions. These patients usually present a good prognosis, contrary to those with an identified genetic abnormality. Additionally, advanced imaging techniques (LGE and T1-mapping) have demonstrated an increasingly important role in risk stratification. However, these and many other aspects are still not fully understood and multicentre registries/studies are required to assess the utility of diagnostic criteria, provide risk stratification tools, and determine the appropriate management/treatment.