Noncompaction cardiomyopathy

Article written and reviewed by Dr Luca Moretti

Noncompaction cardiomyopathy describes a condition in which a deeply trabeculated ventricular surface is not merely an anatomic variant but is part of true myocardial disease. The left ventricle typically shows prominent trabeculae, recesses communicating with the cavity and a wall apparently organized into an inner trabeculated component and an outer more compact component. For this appearance to acquire pathologic significance, however, findings such as dysfunction, dilation, fibrosis, arrhythmias, familial aggregation or a consistent genetic cause must coexist. Isolated morphology is not sufficient for diagnosis.

This clarification is essential because modern techniques identify marked trabeculation in a non-negligible proportion of people without heart disease. Athletes, women during pregnancy and individuals undergoing magnetic resonance for other reasons may exceed criteria historically called diagnostic without developing events. Left ventricular hypertrabeculation is therefore an anatomic descriptor, whereas cardiomyopathy requires demonstration of a pathologic myocardial substrate.

Nomenclature remains controversial. The 2006 AHA classification included noncompaction among primary genetic cardiomyopathies, whereas the 2008 ESC position placed it among unclassified forms. The 2023 ESC guidelines prefer to describe hypertrabeculation as a phenotypic trait, potentially isolated or associated with dilated, hypertrophic, arrhythmogenic, congenital or syndromic cardiomyopathies. The term noncompaction cardiomyopathy retains clinical usefulness only if it is not converted into an automatic imaging diagnosis.

The practical consequence is a change in perspective. The question should not simply be whether the ratio between trabeculated and compacted layers exceeds a threshold, but whether the patient has a disease capable of explaining symptoms, functional abnormalities and evolution. Modern diagnosis integrates ventricular phenotype, tissue characterization, electrophysiology, family history and genetics. This multiparametric integration simultaneously reduces the risk of overdiagnosis in healthy individuals and underestimation in patients with true cardiomyopathy.

Clinical definition and nosologic boundaries

There is no universally accepted histologic, genetic or imaging gold standard. Echocardiographic and CMR criteria were developed in small cohorts, often selected for symptoms or dysfunction, and use different cardiac phases, planes and denominators. None establishes by itself that trabeculae are the cause of cardiomyopathy. A clinically robust definition therefore requires a convincing trabecular phenotype together with at least one independent piece of evidence of heart muscle disease.

Independent evidence includes reduced systolic function or regional abnormalities not explained by coronary artery disease, dilation not attributable to loading conditions, nonischemic fibrosis, significant ventricular arrhythmias, conduction disease, a cardiomyopathic phenotype in relatives, or a pathogenic variant with consistent segregation. Congenital heart disease, chromosomal syndromes and neuromuscular disorders may also confer significance on the finding. The etiologic context is more informative than simple apical extent.

The expression isolated form must be used rigorously. It may mean absence of congenital malformations, absence of another cardiomyopathic morphology, or complete normality of function, rhythm and family history, three very different concepts. A patient with trabeculae and dysfunction but no structural defects is not equivalent to an asymptomatic person with a normal heart. The operational definition must be stated in every report and study.

Right ventricular involvement is difficult to classify because right ventricular trabeculation is physiologically more pronounced. Criteria derived from the left ventricle do not have sufficient specificity when transferred to the right. In the presence of dilation, dysfunction or arrhythmias, congenital heart disease, pulmonary hypertension and arrhythmogenic cardiomyopathy should be investigated, describing biventricular involvement without creating right ventricular noncompaction based on geometry alone.

Cardiac development and pathologic mechanisms

The traditional model attributed the disease to arrested embryonic compaction. According to this hypothesis, the initially spongy myocardium should progressively transform into a compact wall from base to apex and from epicardium to endocardium. Persistence of an inner trabeculated layer would represent the residue of an incomplete process. Contemporary developmental anatomy studies, however, do not support such a linear conversion and make it inappropriate to infer embryonic arrest from an adult image.

During development, both the compact portion and trabeculae increase in volume at different rates and with different geometries. Thickening of the outer wall may occur through growth, proliferation and remodeling without trabecular tissue having to merge completely with it. Endocardial-myocardial signaling pathways, including the NOTCH network, regulate wall maturation and growth, but mechanisms observed in animal models do not justify a single explanation for all human cases. Differential growth replaces the simplistic concept of obligatory compaction.

In clinical cardiomyopathy, the phenotype may arise from different mechanisms. Sarcomeric variants alter force and mechanosensitivity; cytoskeletal and nuclear defects alter stress transmission; mitochondrial diseases reduce energy availability; syndromic conditions interfere with development and cell survival. Dilation may also separate pre-existing trabeculae and make them more visible. Pathogenic heterogeneity explains why the same morphology accompanies opposite functional profiles and prognoses.

When dysfunction and fibrosis are present, pathophysiologic consequences resemble those of other cardiomyopathies. Reduced contractility increases end-systolic volume and wall stress, while neurohormonal activation and remodeling worsen heart failure. Disarray, scar and conduction heterogeneity may sustain arrhythmias. Deep recesses have been implicated in stasis, but thrombotic risk depends primarily on function, rhythm and documented thrombus. The functional substrate, not the surface alone, generates complications.

Genetics, syndromes and familial phenotypes

Genetics confirms that noncompaction cardiomyopathy is not a uniform monogenic entity. In selected cohorts, diagnostic yield varies with age, family history, dysfunction and panel breadth; a substantial proportion of patients remain without a molecular cause. Many identified variants belong to the same genes implicated in dilated and hypertrophic cardiomyopathies. The genetic overlap supports classification based on the dominant phenotype rather than on a gene exclusive to trabeculation.

Sarcomeric genes are frequent, with particular representation of MYH7, followed in several series by MYBPC3, TTN, ACTC1 and other contractile genes. Not all variant classes carry the same significance: a missense variant in a functional region and a truncating variant require specific evidence, population frequency, segregation and functional data. Attributing causality simply because a gene appears on a panel creates false positives. Gene-disease validity precedes interpretation of the individual result.

Some associations provide recognizable phenotypic clues. Variants in HCN4 may accompany sinus bradycardia, ascending aortic dilation and trabeculation; variants in MIB1 link the NOTCH pathway to families with cardiomyopathy; alterations in TAZ cause Barth syndrome with cardiomyopathy, neutropenia and 3-methylglutaconic aciduria. Mitochondrial diseases, RASopathies and 1p36 deletion further broaden the spectrum. The syndromic signature guides testing and counseling better than an indiscriminate panel.

In children, the likelihood of a genetic or multisystem cause is greater, especially with fetal or neonatal onset, biventricular dysfunction, dysmorphic features, hypotonia, growth delay or metabolic abnormalities. Examination should include development, muscle strength, CK and targeted investigations according to clinical suspicion, without applying identical metabolic batteries to everyone. The extracardiac phenotype may be the key to distinguishing a treatable disease or a syndrome with specific risk.

Genetic testing is most strongly indicated when there is a defined cardiomyopathy, early onset, family history, congenital heart disease, conduction disorder or a syndromic presentation. A pathogenic or likely pathogenic variant may guide screening, prognosis and sometimes gene-specific management. A variant of uncertain significance does not confirm the diagnosis and should not be used for predictive testing in relatives. Genetic counseling should precede and follow the analysis.

Clinical presentation and age spectrum

The presentation ranges from an incidental finding to advanced heart failure. Dyspnea, exercise intolerance, orthopnea, edema and fatigue reflect systolic or diastolic dysfunction and are not specific to trabeculation. Palpitations may result from ectopy, supraventricular tachycardias, atrial fibrillation or ventricular arrhythmias. Syncope during exertion or without prodromes requires prompt evaluation because it may reflect electrical instability or severe hemodynamic compromise.

The electrocardiographic picture is heterogeneous. Repolarization abnormalities, hypertrophy, bundle branch block, atrioventricular conduction disorders, pre-excitation and arrhythmias may occur, but no pattern is diagnostic. A normal ECG reduces the likelihood of a clinically relevant phenotype when function and family history are also normal, without absolutely excluding an early form. The electrical profile acquires value only when integrated with imaging and history.

Neonates and infants may present with tachypnea, feeding difficulty, poor growth, hepatomegaly, arrhythmias or shock. In pediatric series, early onset, hemodynamic instability, dilation and severe dysfunction identify the most unfavorable courses. The associated dilated, hypertrophic or restrictive phenotype changes the risk of death and transplantation more than the general noncompaction label. Pediatric stratification should therefore be prompt and repeated during growth.

In adults, the finding may emerge during evaluation for heart failure, arrhythmia, stroke, family screening or imaging performed for other reasons. Tertiary-center cohorts overestimate severity because they collect symptomatic patients, whereas population studies include many morphologies without consequences. An identical label can therefore encompass opposite pretest probabilities. Cohort selection explains much of the discordant epidemiologic estimates.

Echocardiography and morphologic criteria

Transthoracic echocardiography is the initial examination because it defines dimensions, function, valves, pressures and flow within the recesses. Apical views must avoid foreshortening, which can truncate the apex and create a false appearance of wall thickening. Assessment should include ejection fraction, volumes, right ventricular function, strain, any congenital defects, and echocardiographic contrast when the endocardium is not delineated. Acquisition quality is more important than the apparent precision of a ratio.

In Chin's original pediatric series, an end-diastolic ratio was calculated between the epicardium-to-recess-floor distance and the epicardium-to-trabecular-peak distance, with a value no greater than 0.5 considered positive. Jenni instead used end systole: a two-layered wall, with the inner component more than twice as thick as the outer component, typical distribution and recesses perfused from the cavity; the original definition also required absence of coexisting cardiac abnormalities. Because they use different phases and geometric constructions, the two echocardiographic criteria do not constitute independent biological confirmations.

The Stöllberger criteria describe more than three prominent trabeculae distal to the papillary muscles, visible in the same imaging plane, with synchronous movement and intertrabecular spaces perfused from the cavity. They are relatively simple but depend on distinguishing trabeculae from false tendons and papillary muscles. No criterion has an external biological reference standard and agreement among operators and systems is imperfect. The plurality of criteria should not be mistaken for convergent validation.

Global longitudinal strain may identify subclinical dysfunction and show an abnormal regional distribution, but it does not specifically distinguish this cardiomyopathy from others. A reduced value may depend on loading conditions, scar, associated heart disease or image quality. Three-dimensional echocardiography improves visualization of architecture and volumes but retains limited apical resolution. Myocardial function should be interpreted as prognostic information, not as etiologic proof of trabeculation.

The report should specify the criterion, cardiac phase, segment, image quality and associated abnormalities, avoiding the isolated phrase compatible with noncompaction. It is preferable to describe excessive trabeculation and conclude whether signs of cardiomyopathy are present. Comparison with prior studies may reveal progressive dilation or, conversely, regression after changes in loading conditions. A graded conclusion makes imaging clinically actionable without converting a threshold into certainty.

Cardiac magnetic resonance and tissue characterization

Cardiac magnetic resonance visualizes the entire ventricle, measures volumes and mass with high reproducibility, and characterizes tissue. It is particularly useful when the echocardiographic apex is inadequate, function appears discordant, thrombus is suspected, or overlapping phenotypes need to be distinguished. Greater anatomic sensitivity, however, also increases incidental findings. Multiparametric CMR is valuable because it adds function and fibrosis, not because it makes trabecular measurement infallible.

Petersen measures on long-axis images at end diastole, excluding the apical tip, the ratio between noncompacted and compacted thickness: the proposed threshold is greater than 2.3. Jacquier shifts the analysis from wall thickness to mass and considers a trabeculated component greater than 20% of total ventricular mass positive. In the Grothoff method, the noncompacted component exceeds 25% and is greater than 15 g/m² after indexing; a noncompacted-to-compacted ratio of at least 3 in at least one of segments 1–3 or 7–16 and at least 2 in segments 4–6 completes the combined system. The CMR discordance reflects different methods and derivation populations.

How software assigns trabeculae and papillary muscles to the cavity or myocardial mass simultaneously changes volume, mass and ejection fraction. Cardiac phase, minimum compacted-wall thickness and selection of the segment with the maximum value also influence the result. Fractal analysis and automated segmentation quantify endocardial complexity on a continuous scale but require standardization and demonstration of outcome utility. Reproducible metrology is a necessary, not sufficient, condition for a biological diagnosis.

Late gadolinium enhancement identifies focal fibrosis and has greater prognostic value than the amount of trabeculation in several cohorts. It may appear midmyocardial, subepicardial or at insertion points, without an exclusive pattern. T1 mapping and extracellular volume can document diffuse abnormalities, while elevated T2 suggests edema and calls for an inflammatory differential diagnosis. Tissue characterization helps identify the true cardiomyopathic substrate and competing causes.

Cardiac CT may provide an anatomic alternative when echocardiography and CMR are not feasible, especially when the coronary arteries also need to be studied. It exposes the patient to radiation and iodinated contrast and offers inferior tissue characterization. Ventriculography and endomyocardial biopsy are not routine diagnostic tests: biopsy is reserved for specific inflammatory, infiltrative or metabolic suspicions in which the result may change therapy. The clinical question should guide every modality.

Integrated diagnosis and differential diagnosis

The pathway begins with technical confirmation of the finding and proceeds to identification of the pathologic phenotype. History, examination, ECG, echocardiography and CMR are integrated with rhythm monitoring, exercise or cardiopulmonary testing and targeted laboratory studies. In a pedigree spanning at least three generations, heart failure, transplantation or ICD implantation, sudden deaths, early cerebrovascular events, congenital heart disease and neuromuscular disorders should be sought. The diagnostic sequence prevents a suggestive image from retrospectively conditioning every other datum.

Dilated cardiomyopathy is the most frequent overlap. A dilated cavity may accentuate normal trabeculae, while a sarcomeric or TTN variant may simultaneously produce dysfunction and a trabecular phenotype. Whether the condition is called DCM with excessive trabeculation or noncompaction cardiomyopathy is less important than identifying cause, function, scar and risk. Phenotypic dominance should guide treatment and screening.

An international cohort published in 2026 studied 1,160 patients with dilated cardiomyopathy who underwent CMR, most of whom also had genetic testing. Hypertrabeculation, present in about 30% by fractal analysis or the Petersen criterion, did not add independent risk of embolism, major ventricular arrhythmias or advanced heart failure. The result supports DCM-guided management rather than trabeculation-specific therapy, but cannot automatically be extended to children, syndromes or nondilated cardiomyopathic forms. Validity within DCM should be respected without generalization.

Myocarditis, sarcoidosis and inflammatory cardiomyopathies may cause dysfunction, arrhythmias and LGE with incidental trabeculation. Coronary artery disease produces regional abnormalities and subendocardial or transmural scar in a vascular territory. Apical hypertrophic cardiomyopathy, endomyocardial fibrosis and thrombus may alter the apical contour; bifid papillary muscles and false tendons may mimic trabeculae. The distribution of injury and contrast-enhanced sequences distinguish structures and tissue.

Adaptations to preload require particular caution. During pregnancy, previously unapparent trabeculae may develop and regress after delivery; in athletes, physiologic dilation makes the endocardium more evident; anemia and high-flow states produce similar effects. These findings are not automatically benign when symptoms, reduced function, LGE, arrhythmias or family history coexist. Reversibility with loading conditions supports adaptation, whereas isolated persistence does not prove disease.

A credible conclusion classifies the case as physiologic morphology, indeterminate phenotype or cardiomyopathy, documenting the evidence supporting the category. In indeterminate cases, follow-up is used to observe function, rhythm and evolution, not to repeatedly measure ratios indefinitely. The diagnosis may be revised if a family event, a reclassified variant or new dysfunction emerges. Proportionate certainty is preferable to unsupported absolute labels.

Prognosis and risk stratification

Early series reported very high mortality, transplantation, embolic and arrhythmic rates because they included symptomatic adults with severe dysfunction. Later studies showed that the risk associated with incidental trabeculation is much lower. Meta-analyses indicate that, among patients selected as having cardiomyopathy, overall prognosis is similar to that of comparable dilated cardiomyopathy. Ventricular function remains the most consistent determinant of heart failure and mortality.

LGE adds information even when ejection fraction is not severely reduced. In a meta-analysis, the presence of LGE was associated with events and cardiac death, whereas no hard events were observed in the subgroup with preserved function and no LGE, albeit with limited numbers. This does not create an individual guarantee, but shows that myocardial fibrosis discriminates risk better than the noncompacted-to-compacted ratio.

A multicenter cohort of 585 patients confirmed ejection fraction as the variable most strongly associated with events and the incremental value of LGE. Age, sex, ECG abnormalities, cardiovascular factors and familial aggregation contributed to the model; patients with a normal ECG, LVEF of at least 50%, no LGE and negative family screening had no events during cohort follow-up. This low-risk combination should not be replaced by a single anatomic measurement.

For arrhythmic risk, previous cardiac arrest or sustained tachycardia, suspected arrhythmic syncope, reduced function, scar, ectopic burden or nonsustained ventricular tachycardia, and specific genotypes are important. The maximum trabecular ratio may correlate in some series but does not independently justify an ICD. Right ventricular involvement, dilation and advanced heart failure worsen the overall profile. Arrhythmic stratification follows the principles used for cardiomyopathies and ventricular arrhythmias.

In children, prognosis depends strongly on age and the associated phenotype. The pediatric registry showed differences in death or transplantation among dilated, hypertrophic, restrictive and isolated forms, underscoring the limitation of a single category. Prenatal onset, severe dysfunction, arrhythmias and systemic syndromes require intensive surveillance, whereas a stable child with normal function needs a proportionate plan. The growth trajectory can modify expression and risk.

Heart failure, arrhythmias and devices

No drug can selectively normalize trabeculae. Heart failure with reduced ejection fraction is treated according to guidelines with renin-angiotensin system inhibition or ARNI, beta-blockers, mineralocorticoid antagonists and SGLT2 inhibitors, adapting sequence and doses to blood pressure, renal function, potassium, age and tolerance. Diuretics control congestion without directly modifying the disease course. The foundational therapy is that of the functional phenotype.

Mildly reduced or preserved ejection fraction requires evaluation of causes and comorbidities, control of congestion and application of the relevant heart-failure recommendations. In children, doses and evidence of efficacy differ and management belongs in expert centers. The response may include reverse remodeling, but persistence of trabeculation does not equal treatment failure. Therapeutic success is measured by symptoms, function, hospitalizations and survival.

Atrial fibrillation, supraventricular tachycardias and ectopy are treated according to symptoms, function and thromboembolic risk. Prolonged monitoring is appropriate with syncope, palpitations, known arrhythmias, scar or reduced function; frequent Holter monitoring is not required for morphology alone in a healthy adult. Ablation may be indicated for specific arrhythmias without expecting it to modify the anatomic phenotype. Rhythm management should address a documented arrhythmia.

An ICD is indicated for secondary prevention after cardiac arrest due to VT/VF or sustained ventricular tachycardia with hemodynamic compromise, in the absence of a reversible cause. For primary prevention, the decision integrates ejection fraction after optimized therapy, LGE, nonsustained arrhythmias, syncope, family history and genotype, following guidelines for cardiomyopathies and ventricular arrhythmias. There is no specific indication based on the Petersen or Jenni criterion. The defibrillator decision requires absolute risk and informed preferences.

Cardiac resynchronization follows QRS duration and morphology, function and symptoms as in other forms of heart failure. In refractory cases, hemodynamic assessment, ventricular assist support and transplantation should be considered early, especially in children or rapidly progressive syndromes. Neuromuscular and multisystem diseases require individualized assessment of candidacy and extracardiac outcomes. Advanced therapy should not be delayed by nominal uncertainty about the cardiomyopathy.

Thromboembolism and anticoagulation

The historical association with stroke and embolism encouraged broad anticoagulation, but studies do not show that isolated trabeculation by itself increases risk sufficiently. Events occur mainly with atrial fibrillation, severe dysfunction, dilation, visible thrombus or previous embolism. Recesses may contribute to stasis in a hypokinetic ventricle, but this plausibility does not replace a clinical indication. Morphology without additional risk factors does not justify permanent anticoagulation.

Atrial fibrillation is managed using current thromboembolic algorithms, whereas ventricular thrombus requires anticoagulation and follow-up imaging according to location, morphology and bleeding risk. Contrast echocardiography or CMR clarifies suspicious masses when the apex is not visible. After systemic embolism, alternative sources, selected thrombophilias and vascular disease should also be investigated. Embolic causality should not be automatically attributed to trabeculae.

There are no dedicated randomized trials establishing the optimal drug and duration in noncompaction cardiomyopathy. The choice between a vitamin K antagonist and a direct oral anticoagulant depends on the specific indication, available evidence for atrial fibrillation or ventricular thrombus, renal function and interactions. Specific constraints apply in children and pregnancy. Bleeding risk assessment must always accompany estimation of benefit.

Pregnancy, sports and special situations

Preconception assessment considers function, functional class, previous heart failure, arrhythmias, medications, genotype and familial transmission. Maternal risk depends much more on myocardial reserve than on the trabecular ratio. A woman with normal function and no other marker may tolerate pregnancy well, whereas dysfunction, a previous event or arrhythmia requires a pregnancy heart team. Preconception planning also allows fetotoxic medications to be replaced.

During pregnancy, increased plasma volume may make trabeculae and recesses more evident. A newly positive morphologic threshold does not establish cardiomyopathy and should be reassessed after delivery. Symptoms, biomarkers, function and temporal evolution help distinguish adaptation, peripartum cardiomyopathy and pre-existing disease. Obstetric timing prevents confusion between reversible remodeling and a pathologic process.

Exercise prescription starts from symptoms, function, blood-pressure response, arrhythmias, LGE and genotype. A person with isolated excessive trabeculation, normal function and reassuring electrical evaluation should not be automatically excluded from sports. In established cardiomyopathy, moderate recreational activity is often possible, whereas intense or competitive exercise requires shared decision-making and periodic reassessment. Sports restriction should be proportionate to actual risk.

In athletes, balanced dilation, normal diastolic function, an appropriate increase in contractility with exercise, absence of LGE and no arrhythmia favor physiologic adaptation. Reduced contractile reserve, scar, complex arrhythmias or a positive family history instead favor pathology. Deconditioning may help in selected cases but is not a universal diagnostic test. The athletic gray zone requires combined expertise in imaging and sports cardiology.

Family members, follow-up and future perspectives

When the proband has a credible cardiomyopathy, first-degree relatives should undergo history, ECG and echocardiography, with CMR in doubtful cases or when required by the familial phenotype. Age at initiation and screening intervals depend on the gene, age of onset in the family and guidelines, because penetrance is often incomplete and age-dependent. Clinical family screening can detect dysfunction or arrhythmias before symptoms.

If a pathogenic or likely pathogenic variant is identified, cascade testing distinguishes relatives who require surveillance from those who did not inherit the familial cause. A VUS should not drive predictive decisions. In families without an identified variant, clinical screening remains necessary while familial risk remains plausible. Phenotype segregation may also contribute to future molecular reclassification.

Follow-up of the patient with cardiomyopathy assesses symptoms, functional capacity, ECG, function and rhythm; frequency and tools are adapted to severity and trajectory. Repeat CMR is useful if a change could alter therapy or risk, not at rigid intervals simply to remeasure trabeculae. New syncope, sustained palpitations, functional deterioration, pregnancy or a family event prompts earlier reassessment. Dynamic surveillance follows the pathologic process rather than the label.

Diagnostic communication should acknowledge uncertainty without minimizing findings. Explaining that a morphology may be common but becomes important when independent markers coexist reduces anxiety and improves adherence. The record should clearly distinguish trabecular phenotype, dominant cardiomyopathy, genetic cause and level of risk. Terminologic precision prevents a historical diagnosis from being carried forward for years without reassessment.

Future research must move beyond cohorts defined solely by imaging ratios. Prospective registries with appropriate controls, standardized segmentation, rigorous genotyping, tissue characterization and independent outcomes are needed. Automated models may quantify geometry and strain, but must demonstrate incremental value beyond function, LGE, ECG and family history. Outcome validation will determine whether biologically distinct subtypes truly specific to noncompaction exist.