Left ventricular measurements with cine and spin-echo MR imaging: a study of reproducibility with variance component analysis

The authors studied the reproducibility of repeated measurements of left ventricular volumes, weight, and wall stress as determined with cine and spin-echo magnetic resonance (MR) imaging. Two observers analyzed 40 serial MR imaging examinations twice, yielding a total of 160 repeated measurements. Estimates of 95% ranges for change were ejection fraction, 10% and 12%; wall mass, 16% and 21%; and wall stress, 22% and 20%, for measurements derived from cine MR imaging and spin-echo MR imaging, respectively. Reproducibility was not significantly different between cine and spin-echo MR imaging. Intraobserver, interobserver, and interexamination errors were quantitated with variance component analysis. Interexamination variability was the single most important contributor to total variance. Reproducibility of left ventricular chamber volume measurements with MR imaging is superior to that of other imaging modalities. In addition, MR imaging can provide reliable estimates of wall mass and wall stress. Efforts to improve reproducibility should be aimed at lowering interexamination variability.