Reproducibility of M-mode echocardiographic assessment of left ventricular function. Significance of the temporal range of measurements.

The aim of this study was to evaluate the reproducibility of M-mode echocardiography in sequential studies on left ventricular (LV) function. To that end, 3 to 8 serial recordings were performed on 4 groups of healthy subjects (total number, 49) within dissimilar time periods: (1) within 3 h, (2) within 24 h, (3) within 1 to 2 weeks and (4) within 14 to 18 months. The examination technique was carefully standardized and the recordings were analysed by digitization and computer assistance; the data were averaged from 5 beats on each recording. The random variation of LV transverse dimensions and performance was roughly equal in the first 3 of these studies; the mean coefficients of variation ranged from 1.2 to 1.8% for end-diastolic dimension, from 1.7 to 2.3% for end-systolic dimension, and from 2.5 to 3.6% for ejection fraction. In the long-term study the respective mean coefficients were larger: 3.2%, 3.7% and 5.5%. The coefficients of variation for estimates of stroke volume, cardiac output, and peripheral vascular resistance ranged from 4.1 to 8.4% in studies completed within 1 day or inside 1 or 2 weeks but increased to 10-15.5% in the long-term study. The beat-to-beat variability of LV dimensions and performance, as assessed in a subgroup of 11 subjects, was roughly equal in quantity to the variability observed in the long-term study.(ABSTRACT TRUNCATED AT 250 WORDS)