Excessive Trabeculations and Prognosis: The Plot Thickens.

Abstract

In this issue of Circulation: Cardiovascular Imagin g, Ivanov et al1 provide further evidence that the presence of excessive trabeculations may be benign. In this prospective cohort study of 700 patients (all comers) who were clinically referred for a cardiovascular magnetic resonance (CMR) scan at a single center with a low-volume CMR service, the extent of trabeculations by 4 different imaging criteria—Petersen,2 Stacey,3 Jacquier,4 and Captur5 methods—was assessed. Over a median follow-up of ≈7 years, 209 patients (30%) had adverse events (primary combined end point of all-cause mortality, ischemic stroke, ventricular tachycardia/ventricular fibrillation, or hospitalization for heart failure). None of the 4 CMR criteria for the so-called left ventricular noncompaction (LVNC) was associated with this primary composite outcome in multivariate analysis. However, limited conclusions could be made regarding the Captur criterion because only 3% (23/700) exceeded the cutoff, and thus, too few events occurred during the follow-up for sufficient power to exclude a false-negative finding. See Article by Ivanov et al Diagnosing a heart condition is typically done taking into account the clinical context. According to Bayes principles, making a diagnosis is based on the posttest probability, which in turn is determined by the pretest probability and the result and diagnostic accuracy of a test, such as CMR.6 Here, the authors use the term left ventricular noncompaction synonymously to positive imaging criteria of excessive trabeculations without considering the pretest probability and refer to the subgroup of patients as high risk LVNC who fulfill at least 1 of the 4 CMR criteria and have a high pretest probability of having a clinical diagnosis of LVNC. High pretest probability was defined as the presence of at least 1 of the following characteristics: (1) positive family history of LVNC; (2) symptoms suggestive of a diagnosis of LVNC; …