Are wall thickening measurements reproducible?

An early transesophageal echocardiography (TEE) study reported that there was high inter- and intraobserver variability in measurements of wall thickening (WT). This study reevaluated TEE for the measurement of WT in the transverse plane and, for the first time, evaluated it in the longitudinal plane. Ten patients were studied with a biplane TEE probe inserted and positioned to obtain a transverse short-axis (SA) view of the left ventricle at the midpapillary muscle level. A longitudinal transgastric long-axis (LA) view of the left ventricle was then obtained. This measurement sequence was performed four times in each patient. Off-line analysis was performed independently by two observers. In each tomographic cut of the ventricle, four locations for measurement were defined. WT was measured in three consecutive beats, and the results averaged (WT₁). The observers repeated the measurements (WT₂) 2 wk later, and the intra- and interobserver differences (mm) were calculated. Average differences for the intraobserver comparisons of wall thickening were small (0.08-0.24 mm), but the high standard deviations (1.17-2.44 mm) suggest significant variability between individual measurements. Thus it appears that measurements of wall thickening with small mean intraobserver bias may be possible, but to offset the potential for variability, multiple measurements are necessary. Also, careful standardization of location and edge definition is important to limit interobserver variability.