Assessment of left ventricular volume and mass by cine magnetic resonance imaging in patients with anterior myocardial infarction intra-observer and inter-observer variability on contour detection.

Abstract

Remodeling of the left ventricle after myocardial infarction can be documented by calculation of left ventricular volume and mass, using endocardial and epicardial tracings of multilevel multiphase short-axis cine magnetic resonance (MR) imaging series. We assessed left ventricular volume and mass from 8 slices and during 12 phases of the cardiac cycle in seven patients with an anterior wall myocardial infarction; one patient was studied twice, leaving eight MR examinations to be evaluated. Purpose of this study was to assess the intra-observer and inter-observer variability of epicardial volume, endocardial volume, and left ventricular mass from contours manually traced by two independent observers. For the eight MR examinations, epicardial volume was found to be 292 +/- 51 ml (mean +/- SD) at end-diastole, which decreased to 237 +/- 55 ml at end-systole. Endocardial volume was 141 +/- 31 ml at end-diastole, which decreased to 79 +/- 27 ml at end-systole. Left ventricular ejection fraction was 45 +/- 8%. Mean left ventricular mass, when averaged over all patient studies and all phases, was 159 +/- 30 g. Intra-observer and inter-observer variability were found to be 3.5% and 5.2% for endocardial volume, 2.0% and 2.5% for epicardial volume, and 3.6% and 3.6% for left ventricular mass, respectively. The contour analysis showed a statistically significant phase effect in the endocardial contour in the midventricular slices, which was resolved after establishing a more precise definition for the tracing of the endocardial border. In conclusion, left ventricular volume and mass in patients with an anterior wall myocardial infarction can be assessed with high reproducibility and reliability from manual contour tracings. A precise protocol for the definition of endocardial and epicardial contours is required to obtain reproducible and reliable results.