Reproducibility of left ventricular mass measurements by two-dimensional and M-mode echocardiography

Both two-dimensional and M-mode echocardiography provide accurate estimates of left ventricular mass. However, their reproducibility in serial studies has not been compared, although this issue is critical to evaluation of regression of hypertrophy. To determine which technique provides more reproducible estimates of left ventricular mass, three serial studies were performed prospectively in each of eight normal adults over 5 months. Both two-dimensional and M-mode echocardiograms were obtained at each of these 24 studies. Measurements were performed by two independent observers who did not know patient identity. For the two-dimensional method, left ventricular mass was determined with use of a computer light-pen system and the truncated ellipsoid formula. For the M-mode method, mass was calculated from Penn convention measurements with use of the cube formula. At study 1 the group mean left ventricular mass by two-dimensional echocardiography (115 +/- 20 g) did not differ from that by M-mode study (127 +/- 37 g, p = NS). However, serial estimates of left ventricular mass were more reproducible by two-dimensional echocardiography. The mean difference among the three serial two-dimensional studies in each individual was 4.8 +/- 4 g (4.2 +/- 3%) by the two-dimensional method, but was 18.5 +/- 13 g (14.9 +/- 10%) by the M-mode method (p = 0.01). Interobserver results for left ventricular mass by two-dimensional echocardiography correlated closely (r = 0.95, n = 24, p less than 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)