Intraoperative detection of segmental wall motion abnormalities with transesophageal echocardiography

Purpose: To compare two methods of analysis of regional wall-motion (RWM) u sing transesophageal echocardiography (TEE). Methods: Thirty patients undergoing coronary artery bypass surgery were stu died. The transgastric short axis view at the mid-papillary level was recor ded before and after cardiopulmonary bypass. All images were reviewed by an anesthesiologist trained in TEE and an echocardiographer, Regional wall mo tion was graded: I normal, 2 hypokinetic, 3 akinetic, and 4 dyskinetic. The left ventricle was evaluated according to the guidelines of the American S ociety of Echocardiography using 6-segment, and 4-segment models. Agreement between observers (interobservers), and for one observer at two different moments (intraobservers), for grading each segment was defined as RWM abnor mality scores within I grade. A wall-motion score index (WMSI), which is th e sum of individual scores divided by the number of segments visualized, wa s calculated. A Bland Altman analysis was used to assess interobserver vari ability. Results: Agreement between observers occurred in 96% and 94% of the examine d segments, using 4- and 6segment models respectively. Intraobserver agreem ent was 99% and 97% for the 4- and 6-segment models. The mean differences ( bias) of the interobserver variability in grading the segments were 0.04 +/ -. 0.79 and 0 +/- 0.72 using a 4- or 6-segment model. The mean difference o f the interobserver variability in WMSI were -0.05 +/- 0.42 and 0.05 +/- 0. 37 using a 4- or a 6-segment model. Conclusion: Both methods, using either a 4- or a 6-segment model, result in a high intraobserver and interobserver agreement, and a low interobserver variability.