Noncompacted myocardium (NCM) is a cardiac abnormality involving the myocardial wall, characterized by numerous, excessively prominent trabeculations and deep intertrabecular recesses penetrating into the midmyocardium ( Fig. 64.1 ). Noncompaction of the left ventricle can occur as an isolated cardiac feature (left ventricular noncompaction [LVNC]) or in association with other primary cardiomyopathies as idiopathic dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), or even restrictive cardiomyopathy (RCM). In addition, NCM can occur in metabolic and genetic syndromes, in which it can be associated with HCM or DCM (eg, Barth syndrome) or with congenital heart defects (nonisolated LVNC), and in neuromuscular disorders. It was first described in 1926 by Grant as spongy myocardium, but later the definition of isolated LVNC was first used by Chin et al. in 1990 to describe eight pediatric cases with the cardiac phenotype in the absence of congenital heart defects. Although much attention has been given to this entity in recent years, there is currently no consensus on cause, pathogenesis, diagnosis, or management of LVNC. In the 1995 World Health Organization (WHO)/International Society and Federation of Cardiology (ISFC) classification of cardiomyopathies, LVNC was not included ; it was included in the North American revision and listed among the genetically determined cardiomyopathies, but not all experts agree that it represents a cardiomyopathy or a separate disease.
The diagnosis of noncompaction is challenging and its nosology is still debated. However, when a definite diagnosis of noncompaction is made, the diagnostic process should orient toward a genetic disease with a relatively high probability of sarcomeric mutations. The clinical presentation is with a sporadic or familial heart muscle disorder with presumed persistence of the embryonic pattern of trabeculation of the left ventricle (spongy myocardium). The disruption of the normal trabeculation and compaction processes that occurs between days 30 and 70 of embryogenesis is thought to predispose to systolic and diastolic dysfunction, cavity dilatation, hypertrophy, and arrhythmia.
Many terms were introduced in the literature over the years to identify this heterogeneous entity, showing the need for an internationally recognized definition ( Table 64.1 ).
|Angelini et al.||Spongy myocardium|
|Oechslin et al. , Lofiego et al.||Isolated left ventricular noncompaction|
|Dellegrottaglie et al. , Biagini et al.||Isolated ventricular noncompaction|
|Sasse-Klaassen||Isolated noncompaction of the left ventricular myocardium|
|Stanton et al.||Isolated left ventricular noncompaction syndrome|
|Towbin et al. , Zhang||Left ventricular noncompaction cardiomyopathy|
|Yin||Ventricular noncompact syndrome|
|Ikeda et al.||Isolated left ventricular noncompaction cardiomyopathy|
|Bleyl et al.||Noncompaction of the ventricular myocardium|
|Stöllberger et al.||Left ventricle hypertrabeculation/noncompaction|
|Freedom et al. , McNally and Dellefave||Ventricular noncompaction|
|Arbustini et al. , Captur and Nihoyannopoulos , Choudhary et al.||Left ventricular noncompaction|
|Patrianakos et al.||Noncompaction myocardium|
|André et al.||Noncompacted myocardium|
Although this condition is thought to be rare, its incidence and prevalence are not yet well defined. However, because of increasing awareness and improvements in the imaging techniques, it is considered the third most common among the cardiomyopathies. In an epidemiological study in Australian children, it has been reported that 9.2% of the population was affected by isolated ventricular noncompaction, ranking third after dilated and hypertrophic cardiomyopathies, similar to what has been reported by Pignatelli et al. According to different echocardiographic studies, the reported prevalence of isolated noncompaction in adults varied from 0.014% to 0.05%, 0.14%, and 0.26%, whereas its prevalence increased to 3.7% among adults with decreased (<45%) ejection fraction (EF) and to 3%, and 4% in heart failure patients. It is important to underline that half of the LVNC patients are children. In most cases the disease occurs congenitally; however, in some it can manifest later in life. In affected families, at least 25% of asymptomatic relatives present with echocardiographic features.
Pathology and Pathogenesis
Noncompacted myocardium is characterized by a primary involvement of apical and midventricular segments of the left ventricular trabeculated component with two myocardial layers, a thick noncompacted endothelial layer, and a thin compact epicardial layer in the transmural wall (see Fig. 64.1 ). The ratio between noncompacted and compacted layers is of importance for clinical and pathologic diagnoses, and although imaging criteria cannot be translated directly to morphologic evaluation, it is accepted that pathologic criteria are a ratio of greater than 2 between the noncompacted and compacted layers or the presence of intertrabecular recesses extending to half the thickness of the myocardium. The endocardium contains deep intertrabecular recesses that communicate with the ventricular cavity, without evidence of communication with the epicardial coronary arteries and prominent hypertrabeculation (see Fig. 64.1 ), which result from an arrest of compaction of myocardial fibers during embryogenesis, as the most popular pathogenetic hypothesis recognizes. This arrest usually occurs at 5 to 8 weeks of gestation when the myocardium is gradually compacted, the intertrabecular spaces are transformed into capillaries, and the coronary circulation develops. The process occurs from the epicardium toward the endocardium and from the base toward the apex. The right ventricle can also be affected in isolation or with the left ventricle, namely biventricular involvement, in less than 50% of patients, although the evaluation of the right ventricle could be somewhat more difficult. The interventricular septum is spared. Absence of well-formed papillary muscle in the left ventricle is considered an important diagnostic marker.
Histologic examination confirms that the spongy appearance is due to the deep intertrabecular recesses, lined by endothelium, which spread close to the epicardial surface. This feature strongly resembles the spongy myocardium pattern of nonmammalian vertebrates such as fish, amphibians, and reptiles. Increased subendocardial fibrosis is a common feature (see Fig. 64.1 ). In a series of 14 cases, predominantly consisting of autopsied hearts, endocardial fibroelastosis was another characteristic histologic finding. A subendocardial scar suggesting ischemic damage is often found.
NCM is frequently associated with other cardiac defects, especially when causing sudden death in infants and children. Associated cardiac anomalies may include valvular abnormalities, ventricular septal defects, persistent left superior vena cava, histiocytoid cardiomyopathy, partial anomalous pulmonary venous return, and coronary ostial stenosis and conotruncal defects.
No differences in gross or histologic patterns of the noncompacted regions between the isolated and nonisolated NCM have been reported. A wide range of heterogeneity in patterns of trabeculation and recesses can be identified: broad anastomosing trabeculae, coarse trabeculae resembling multiple papillary muscles and interlacing smaller papillary muscles, or a relatively smooth endocardial surface with compressed recesses. At macroscopic evaluation, endocardial fibroelastosis can frequently be detected involving the entire cavity or focal areas. Mural thrombi can be entrapped in the recesses, providing a substrate for systemic embolization. The diagnosis of NCM can be made, even in the setting of other cardiomyopathies, as dilated, hypertrophic, or restrictive. Although the noncompaction hypothesis, as arrest in compaction of the embryonic hypertrabeculated myocardium, is recognized as the most likely pathogenetic hypothesis, other hypotheses have been proposed to explain this entity. The compensatory hypothesis, also potentially responsible for the primary form, recognizes a genetic defect that should produce an adaptive reaction to a poorly contracting myocardium. Secondary NCM forms recognize a hemodynamic hypothesis according to which ischemia or microinfarcts result in myocardial hypoxia or even an unlikely myocarditis hypothesis. Diversity of mutated genes and frequent familial occurrence suggest a complex interplay between structural, contractile, and metabolic factors producing arrest of the compaction process or induction of noncompaction.
Current Diagnostic Criteria
The diagnosis of LVNC is difficult. Current diagnostic criteria relying on standard transthoracic echocardiography, contrast echocardiography, and cardiovascular magnetic resonance imaging have been developed but are still under discussion regarding different measuring modalities, (short- or long-axis view, end-systole vs. end-diastole, linear vs. mass measurements) ( Table 64.2 ). One of the key features is a two-layered myocardium, a thin and compacted layer adjacent to the epicardium and a thick noncompacted layer near the endocardium, with prominent trabeculations separated by recesses. Because the morphopathologic substrates are not always easily detected with the imaging modalities, global or regional myocardial dysfunction and electromechanical features should be considered to improve sensibility and specificity of diagnosis. The echocardiographic criteria are the most discussed and agreed upon ( Fig. 64.2 ). The first proposed criteria by Chin et al. require an X/Y ratio less than or equal to 0.5 at end-diastole. Because the differentiation of the two layers in diastole is difficult, the subsequent set of criteria, suggested by Jenni et al. on parasternal short-axis view, focus on a noncompacted (N) to compacted (C) ratio measured at end-systole, and require an N/C ratio greater than 2 in adults and greater than 1.4 in children. In addition, Jenni et al. proposed additional diagnostic requirements: the absence of coexisting cardiac structural abnormalities; the demonstration of numerous, excessively prominent trabeculations and deep intratrabecular recesses supplied by intraventricular blood on color flow Doppler; the prevalent localization of noncompact regions to the lateral, inferior, or apical left ventricular segments. It is relevant to mention that the distribution of these abnormalities is segmental in LVNC, whereas it is often diffuse in non-LVNC hearts. The most recently proposed criteria by Stöllberger et al. focus on more than three trabeculations protruding from the left ventricular wall, apically to the papillary muscles, visible in a single image plane and on the finding of intertrabecular spaces perfused from the ventricular cavity on color Doppler imaging. These authors justified their approach as a way to overcome the technical difficulties in the differentiation between papillary muscles, aberrant bands, false tendons, and hypertrabeculations in the short-axis views used by Jenni et al. It should be noted that most recently Finsterer and Stollberger have proposed that if both the Stollberger and Jenni criteria are satisfied, the diagnosis should be considered as “definite” LVNC, whereas if only one set is satisfied, the diagnosis should be “probable” LVNC. Patients with less than four trabeculations or an N/C ratio less than 2 would be considered as “possible” LVNC.
The current limitations in the diagnostic criteria have been recently highlighted by Kohli et al. who compared the three sets of LVNC criteria in an adult population referred to a heart failure clinic and found limited concordance and lack of specificity, especially in Afro-Carribbeans. These authors found an unexpected high proportion of heart failure patients (23.6%) and controls (8%) fulfilling one or more of the echocardiographic definitions and suggested that current criteria may be too sensitive, particularly in black subjects. Belanger et al. proposed a modification of the criteria requiring the presence of prominent apical trabeculations in any view and the blood flowing through the area of noncompaction: a maximal systolic N/C ratio in apical four-chamber view with a grading score of mild, moderate, and severe. They proposed a planimetric evaluation of the noncompacted area. Further work is needed to define the range of normal and LVNC morphology in different racial groups. Contrast echocardiography can improve the identification of noncompacted and compacted layers and detection of ventricular dysfunction.
New echocardiographic techniques, such as tissue Doppler imaging, strain rate imaging, and speckle tracking, have been used to provide a more objective and quantitative assessment of LVNC. However, these techniques are still very much user dependent in performing and interpreting the imaging findings.
More recently, cardiac magnetic resonance (CMR) has been adopted as a more sensitive tool in the diagnosis of noncompaction, because it can offer reliable measurements of morphology, structure, and function, thanks to high spatial resolution and high contrast between tissue and blood, without the acoustic window restriction and although no definitive diagnostic criteria have been accepted so far. Two different diagnostic criteria have been applied to CMR that identify the two layers of ventricular parietal walls (see Fig. 64.2 ): The first criteria was proposed by Petersen and involves linear perpendicular measurement of noncompacted versus compacted myocardium at the most trabeculated segments in long-axis steady-state free-precession (SSFP) cines at end-diastole with a ratio greater than 2.3; sensitivity and specificity are 86% and 93.7% respectively. The second criteria, proposed by Jacquier, includes mass calculation of the percentage of the noncompacted layer versus the total trabecular mass in short-axis SSFP cines at end-diastole; a value greater than 20% is positive. The trabecular mass should include blood volume between the trabeculae and the papillary muscle in the compacted layer. Sensitivity and specificity are 93.7% and 93.7%, respectively.
Clinical Presentation (Cardiac and Noncardiac Findings)
Clinical features in LVNC in major published series are summarized in Table 64.3 . Noncardiac features in isolated LVNC include dysmorphic facial appearance, prominent forehead, bilateral strabismus, low-set ears, micrognathia, high-arched palate, and contractures of the left elbow. Facial dysmorphism and motor delay are more common in children than in adults. In its extreme pediatric form, LVNC may have severe presentation as fetal hydrops, neonatal heart failure, or ventricular fibrillation, which, if occurring in the absence of prodromes, may manifest as sudden infant death syndrome. However, most pediatric cases are asymptomatic, and diagnosis is made at family screening, or following an abnormal EKG or chest-X ray. LVNC seems to be more common in children than in adults and may account for up to 10% of the pediatric cardiomyopathies in children younger than 10 years. In symptomatic pediatric cases, tachypnea is the most common presentation, whereas arrhythmia and sudden death seem to be less common. Wolff-Parkinson-White syndrome is found in up to 17% of pediatric cases, but is rare in adults. Heart failure in the pediatric forms has been reported with a frequency ranging from 30% to 89% (see Table 64.3 ). Death in children with LVNC ranged from 7% to 38%. Heart failure symptoms with left ventricular dilatation and reduced systolic function by echocardiography are also common in adults. Thromboembolic events and tachyarrhythmia, with or without associated Wolff-Parkinson-White syndrome, are known features (see Table 64.3 ). EKG abnormalities in LVNC, described in up to 91% of cases, may include paroxysmal supraventricular tachycardia, atrial fibrillation, Wolff-Parkinson-White syndrome, complete heart block, abnormal Q waves, ST-segment abnormalities, T wave changes, left ventricular hypertrophy, bigeminal ventricular ectopic beats, left bundle branch block, ventricular tachycardia, and fibrillation.
|Features||Pediatric Patients||Adult Patients|
|Captur and Nihoyannopoulos||Pignatelli et al.||Oechslin et al.||Ritter et al.||Pascal et al.||Bleyl et al.||Aras et al.||Zaragoza et al.||Sandhu et al.|
|Number of patients||8||27||36||66||34||32||45||62||65|
|Median age at diagnosis (years)||7||5||0.3||4||40||49||37 (mean)||50 (mean)||42|
|Ratio of i-LVNC: ni-LVNC (number of patients)||8:0||27:0||31:5||25:41||34:0||29:3||45:0||62:0||65:0|
|Follow-up (years)||Up to 5||Up to 17||Up to 12||Up to 4||Up to 11||—||—||—||Up to 16|
|Bundle branch block (%)||25||15||0||—||56||—||29||23||32|
|Ventricular tachycardia (%)||38||0||2.7||—||41||—||20||18||6|
|Heart failure (%)||63||30||89||68||68||62.5||62||73||34|
|Systemic emboli (%)||38||0||0||—||21||—||4||—||5|
|Pulmonary emboli (%)||0||7||0||—||9||—||—||—||—|
|Ventricular thrombi (%)||25||0||2.7||—||9||6||—||—||—|
|Neuromuscular disorders (%)||—||—||14||—||—||—||—||82||9|