We read with interest the report by Greutmann et al on the predictors of adverse outcomes in adult patients with noncompaction, also known as left ventricular hypertrabeculation (LVHT). We have the following concerns.

The investigators characterize LVHT as “cardiomyopathy with genetic heterogeneity.” False. There is considerable doubt regarding the designation of LVHT as hereditary and even as a cardiomyopathy. LVHT occurs in a number of patients without genetic backgrounds. LVHT may even be acquired. Furthermore, in patients with genetic backgrounds, it has not been proved thus far that LVHT results directly from 1 of the mutations described in association with LVHT. A further argument against a primary hereditary basis for LVHT is that many different mutated genes and a large number of chromosomal disorders have been suggested to be involved in the pathogenesis of LVHT, which makes it hardly traceable why so many different mutated genes should cause the same morphology. More likely than a distinct cardiomyopathy, LVHT appears to represent an adaptation to impaired cardiac function or triggered remodeling than a distinct phenotypic expression.

It is also not comprehensible why neuromuscular disorders (NMDs) are an exclusion criterion when diagnosing LVHT. What is the difference between LVHT in patients with and without NMDs? We found LVHT in a large number of patients with various different NMDs, but the morphology of LVHT was never different between these patients and those without NMDs, not on echocardiography, cardiac magnetic resonance imaging, on autopsy, or histology. Furthermore, the definition applied by Greutmann et al allows the inclusion of LVHT associated with chromosomal disorders but not of NMDs, a discrepancy that is not comprehensible. According to our own clinical and autopsy data, LVHT is morphologically the same in patients with NMDs, chromosomal disorders, and nonhereditary disease. It is also not conceivable to differentiate between isolated and nonisolated LVHT and to include congenital heart disease as an exclusion criterion, because LVHT in these patients is again not different compared to patients with LVHT without congenital heart disease. It is also irrational to exclude congenital LVHT with the explanation that congenital LVHT is “often accompanied by syndromic associations.” False.

The application of a distinct figure for the ratio between the noncompacted and compacted layers is arbitrary and not based on pathoanatomic or autopsy studies. The use of a distinct ratio is questionable, because it is not specified at which point of the myocardium or image plane the ratio should be measured, making the measurement highly variable and thus unreliable. Because there is also no consensus as to whether the highest or lowest value should be used, some investigators have proposed abstaining from using a ratio and instead including exclusively the 2-layered structure as a diagnostic criterion. A disadvantage of many studies of LVHT is that there is no consensus on the definition and diagnostic criteria of LVHT and that many different definitions have been proposed. There are also hardly any studies available comparing the sensitivity or specificity of different diagnostic criteria in the same cohort of patients.

Which was the indication for oral anticoagulation in 48 of the patients? Did these patients all present with atrial fibrillation, or was anticoagulation given simply because of the presence of LVHT? Because it is strongly under debate if the stroke rate is truly increased in patients with LVHT, there is actually no general need to apply anticoagulation in patients with LVHT in the absence of atrial fibrillation or severe heart failure. Because the authors mention only systemic embolism as a complication of LVHT but do not specify whether cerebral events were included, it would be helpful to list the rates of ischemic stroke, transitory attack, and syncope separately. Compared to our own data, it is surprising that only 1 in 132 patients developed stroke, whereas a study of 144 patients with LVHT found a stroke rate of 15%.

Given these objections, the diagnoses, management, and outcomes of patients with LVHT remain matters of concern. The lack of a consensus on cause, pathogenesis and diagnostic criteria, however, does not mean that predictors of outcomes such as arrhythmia, cardioembolism, and systolic dysfunction differ between studies.