We welcome the opportunity to respond to Finsterer and Stöllberger’s comment and thank these investigators for their interest in our report.

Finsterer and Stöllberger continue to use the term “left ventricular hypertrabeculation” instead of “left ventricular noncompaction” (LVNC), which is confusing. Most researchers and clinicians adhere to the statements of international societies and use the term “noncompaction.” Noncompaction represents a failure of compaction of the noncompacted myocardium during embryogenesis and thus leads to its typical morphologic appearance. “Hypertrabeculation” implies an increased number of normally formed trabeculations. We agree with Finsterer and Stöllberger that hypertrabeculation, defined as an increased number of normally formed trabeculations, can occur in patients without genetic backgrounds (e.g., in hypertensive heart disease, systemic right ventricle, and valvular heart disease). This is comparable to the difference between left ventricular hypertrophy in the presence of hypertension and hypertrophic cardiomyopathy. Nobody doubts a genetic cause of hypertrophic cardiomyopathy only because hypertrophy of the left ventricle can also occur in other disease states; besides, histology is different in hypertrophic cardiomyopathy and left ventricular hypertrophy due to other causes. Comparably, there is an important difference between “hypertrabeculation” and LVNC: the histologic appearance of the myocardium in patients affected by LVNC has been shown to be abnormal. We emphasize a statement from Anderson et al, who also doubted that the term “hypertrabeculation” should replace the much more popular and more appropriate alternative “LVNC.”

In our study, we examined the cardiovascular outcomes of patients with isolated LVNC. This was the reason we did not include patients with LVNC associated with other congenital heart lesions or myopathies. Congenital heart disease and inherited myopathies themselves are known to be important determinants of cardiovascular complications.

We agree with Finsterer and Stöllberger that using a distinct ratio of noncompacted to compacted myocardium in isolation may not be an optimal diagnostic criterion. We thus use a number of diagnostic characteristics, as outlined in our report and previous publications by our group. It is to emphasize that we validated the echocardiographic criteria with pathoanatomic heart preparations. Although there may be some noncompacted myocardium in patients with valvular or hypertensive heart disease or dilated cardiomyopathy, applying all our criteria helps differentiate these entities from isolated LVNC. As discussed in our report, we believe that a ratio of noncompacted to compacted myocardium of 2:1 used as 1 of the diagnostic criteria allows the identification of a high-risk group of symptomatic patients with LVNC. From a clinical standpoint, it may thus be reasonable to use this cutoff until better and more specific diagnostic criteria become available. Roberts et al recently emphasized that cuts of the ventricles parallel to the posterior atrioventricular sulcus after fixation or short-axis views provide the best opportunity to demonstrate the 2-layered structure of the myocardial wall, consisting of a compacted epicardial layer and a markedly thickened noncompacted endocardial layer. These cuts of the anatomic specimens mirror our approach by assessing the myocardial morphology in the short-axis views.

Interestingly, Finsterer and Stöllberger have modified their diagnostic criteria for LVNC in recent years, added new criteria (e.g., a ratio of noncompacted to noncompacted segment >2.0 at end-diastole), and shifted away from their original description. This reflects the diagnostic difficulty. Presently, echocardiographic uncertainties and pitfalls regarding the diagnosis of LVNC cannot completely be extinguished. Echocardiographers must be cautious not to overdiagnose LVNC. The morphologic spectrum of trabeculations, from normal variants to pathologic trabeculations with the morphologic features of noncompaction must be carefully distinguished.

Regarding the association of LVNC and congenital heart disease, we disagree with Finsterer and Stöllberger that this always looks similar to isolated LVNC; in Ebstein’s anomaly and in muscular ventricular septal defects, the noncompacted myocardium seems to be morphologically slightly different in many patients compared to isolated LVNC.

We agree that the risk for systemic embolization may have been overestimated in the past when affected cohorts consisted mainly of highly symptomatic patients with often poor systolic left ventricular ejection fractions. Our report spans the clinical experience of almost 30 years. Given our initial experiences, we were relatively liberal starting patients on anticoagulation in the past. More recently, on the basis of our recent clinical experiences and the data of our current analysis, we reserve anticoagulation for patients with moderately to severely impaired left ventricular ejection fractions or for patients with definite indications for anticoagulation, such as atrial fibrillation.