Effect of Papillary Muscles and Trabeculae on Left Ventricular Measurement Using Cardiovascular Magnetic Resonance Imaging in Patients with Hypertrophic Cardiomyopathy

Objective: To evaluate the influence of papillary muscles and trabeculae on left ventricular (LV) cardiovascular magnetic resonance (CMR) analysis using three methods of cavity delineation (classic or modified inclusion methods, and the exclusion method) in patients with hypertrophic cardiomyopathy (HCM). Materials and Methods: This retrospective study included 20 consecutive HCM patients who underwent 1.5-T CMR imaging with short-axis cine stacks of the entire LV. LV measurements were performed using three different methods of manual cavity delineation of the endocardial and epicardial contours: method A, presumed endocardial boundary as seen on shortaxis cine images; method B, including solely the cavity and closely adjacent trabeculae; or method C, excluding papillary muscles and trabeculae. Ascending aorta forward flow was measured as reference for LV-stroke volume (SV). Interobserver reproducibility was assessed using intraclass correlation coefficients. Results: Method A showed larger end-diastole and end-systole volumes (largest percentage differences of 25% and 68%, respectively, p < 0.05), compared with method C. The ejection fraction was $55.7{\pm}6.9%$ for method A, $68.6{\pm}8.4%$ for B, and $71.7{\pm}7.0%$ for C (p < 0.001). Mean mass was also significantly different: $164.6{\pm}47.4g$ for A, $176.5{\pm}50.5g$ for B, and $199.6{\pm}53.2g$ for C (p < 0.001). LV-SV error was largest with method B (p < 0.001). No difference in interobserver agreement was observed (p > 0.05). Conclusion: In HCM patients, LV measurements are strikingly different dependent on whether papillary muscles and trabeculae are included or excluded. Therefore, a consistent method of LV cavity delineation may be crucial during longitudinal follow-up to avoid misinterpretation and erroneous clinical decision-making.