Normal Cardiac Magnetic Resonance Measurements and Interobserver Discrepancies in Volumes and Mass Using the Papillary Muscle Inclusion Method

Background: The potential of cardiac magnetic resonance (CMR) imaging to provide precise and reproducible measurements might not be realized currently. Methods: We acquired CMR images on 30 healthy adults and compared readings of 2 experienced physicians to determine sources of measurement variability. Results: Simultaneous review of tracings revealed 3 major sources of interobserver variability: 1) choice of end- diastolic/end-systolic frames; 2) choice of the most basal left and right ventricular short-axis slices; and, 3) approach to endocardial edge selection. Study Plan. These objectives were to be achieved through a prospective study in an adequate number of healthy subjects. Qualifying subjects were to undergo a stan- dardized functional CMR evaluation. Predefined cardiovas- cular measurements were made using workstation-specific methods. Key volume and mass measurements were to be performed independently by 2 observers to improve accu- racy, to determine reproducibility, and to develop consensus methods. Results were to be used for clinical and research