We read with interest the report by Roberts et al regarding 3 explanted hearts that fulfilled anatomically the echocardiographic definition of left ventricular noncompaction/hypertrabeculation (LVHT). Photographs of 18 cases were reviewed, but only 7 were assessed as representing clear examples of LVHT. It was assumed that LVHT is overdiagnosed in the other cases, at least morphologically. We have several concerns and questions: What is the specificity of the anatomic definition of noncompacted portion of left ventricular free walls ≥2 times as thick as the compacted portions of these walls? Magnetic resonance imaging shows that areas of noncompaction are common in healthy volunteers, athletes, patients with hypertrophic or dilated cardiomyopathy, and patients with aortic stenosis. Additionally, the ratio is dependent on the point at which it is measured. Where should the ratio be measured, or is it recommend to take the highest value?

We agree with the investigators that after formalin fixation, LVHT is better visible than by looking at fresh hearts at autopsy. However, we cannot follow their arguments that cuts according to the flow of blood will result in flattened cardiac chambers. Although it was stated that 7 published cases present clear examples of LVHT, only 6 references are given. However, in these 6 publications, we found photographs of 10 hearts, and we do not know which of them were “included.” In 2 of the “included” reports, the hearts were opened along the longitudinal axis, and no cuts of the ventricles were made parallel to the posterior atrioventricular sulcus as recommended. When looking at the photographs of the “excluded” reports, and comparing them with the “included” cases, we cannot find substantial differences. Thus, the reasons why cases were included or excluded should be specified. It should be also mentioned to which figures in the cited reports the investigators refer.

Because the 3 patients had histories of heart failure, most probably they underwent various echocardiographic examinations. Did LVHT change over time? It is stated that none of them had neuromuscular disorders. Were they investigated by neurologists? If they were seen by neurologists, which investigations were carried out to confirm or rule out the diagnosis of a neuromuscular disorder? Were mild clinical manifestations regarded as indicators for muscle disease? Were family histories taken in these patients, or were family members investigated for familial LVHT? Did any of the family members develop neuromuscular disorder?

Using only short-axis views, the differentiation among LVHT, false tendons, prominent papillary muscles, and muscle bands is impossible, because the course of structures within the left ventricular cavity along the longitudinal axis cannot be assessed. This is substantiated by the figures in the report: echocardiographically, long-axis as well as short-axis views are depicted, whereas anatomic pictures are given only from short-axis cuts.

We are impressed by the discrepancy between the fine trabecular meshwork seen anatomically and the coarse trabeculations visible on echocardiography. Whereas counting the number of trabeculations within 1 imaging plane, as recommended by us, is feasible echocardiographically, this does not seem to be practical in anatomic studies. Is it useful in anatomic studies to consider the thickness of the trabeculations and to establish a cutoff level of 2 mm in diameter, as suggested by Boyd et al?

In conclusion, LVHT is neither anatomically nor echocardiographically well defined. Because of improvements in image quality and resolution capacity of echocardiography and other imaging technologies, structures within the left ventricle are better visible nowadays. The distinction between normal and pathologic findings can be achieved only by cooperation, comparison of findings, and discussion among clinicians, anatomists, and pathologists.