Reproducibility of assessment of left-ventricular function using intraoperative transesophageal echocardiography.

Reproducibility of results is an important point in assessing the utility of intraoperative transesophageal echocardiography for evaluating changes in left-ventricular function. The purpose of the present study was to define the intra- and interobserver reproducibility of the qualitative assessment of left-ventricular regional wall motion and the quantitative assessment of global left-ventricular function. In addition, the interstudy reproducibility of two examinations was tested when the probe was displaced and replaced in the esophagus. A transesophageal short-axis view at the level of the papillary muscles was obtained in 86 patients undergoing cardiac surgery. In the 80 patients with adequate images, regional wall motion was visually graded and area ejection fraction was calculated by two observers and assessment was repeated by the same observer one day later. The same observer graded wall motion differently in only 5% (24/480) of segments. Grading by two observers differed in 9% (43/480) of segments. Assessment differed by one grade at the most and in not more than 2 out of 6 segments per patient. Repeated measurements of area ejection fraction (AEF) by the same observer correlated well (r = 0.97 before and r = 0.97 after cardiopulmonary bypass) with a mean percent difference of 6%. A similarly close correlation was found for measurements of two observers (r = 0.90 and r = 0.93, respectively) with a mean percent difference of 10% for area ejection fraction. The correlation for the first and second examination in the same patient by one observer was acceptable (r = 0.78 and r = 0.80, respectively) with a mean percent difference of 15% for area ejection fraction.(ABSTRACT TRUNCATED AT 250 WORDS)