P5283The prognostic value of contouring papillary muscles vs. simplified contouring in CMR volumetry of the left ventricle

Abstract

Abstract Background Cardiovascular magnetic resonance (CMR) has evolved as a gold standard for the quantitative assessment of functional parameters. While only inclusion of trabecles and papillary muscles (TPM) accurately reflects anatomy, a simplified approach is frequently used, that “cuts off” this tissue when contouring. Current recommendations favour their inclusion yet also allow for a simplified method. No data exist on any impact of the selection of methods on the prognostic value of the CMR findings. Purpose We aimed to assess the prognostic value of left ventricular (LV) volumetry in non-ischemic dilated cardiomyopathy (NIDCM) assessed by CMR with (1) the exclusion method (accounting TPM to LV cavity volume) or (2) the inclusion method (accounting TPM to LV mass). Furthermore, the predictive value of measuring myocardial mass in end-systole compared with end-diastole was evaluated. Methods Three-hundred forty-six patients with NIDCM who had undergone CMR examination between 2005 and 2013 were enrolled for retrospective analysis. Left ventricular endo- and epicardial contours were acquired by semi-automatic threshold detection in end-diastole and in end-systole, both with the inclusion and the exclusion method. The combined endpoint was cardiac death, heart transplantation, cardiac decompensation, ventricular tachycardia or ventricular fibrillation. Cox regression analysis was performed to evaluate prognostic impact. Results Primary endpoint occurred in 76 patients during a median follow-up of 4.5 years. A significant absolute difference between the two measurement methods was shown for ejection fraction (EF), left ventricular end-systolic volume (LVESV), left ventricular end-diastolic volume (LVEDV) and myocardial mass both in end-systole and end-diastole (p<0.05). Correlation analyses showed significant direct correlation between the two measurement methods. However, in multivariate Cox regression, there was no significant difference in prediction of the combined endpoint between using the inclusion or the exclusion method for EF, LVESV, LVEDV and myocardial mass. Similarly, there was no significant difference in analysis of myocardial mass in end-diastole compared with end-systole. Conclusion In patients with non-ischemic dilated cardiomyopathy, there is no impact of the contouring method (inclusion vs. exclusion of trabecular tissue an papillary muscles from myocardial tissue) on the prognostic value of a CMR volumetry. Although there is a significant absolute difference in measurements, the prediction of adverse events is not influenced.