Comparison of transesophageal echocardiographic and scintigraphic estimates of left ventricular end-diastolic volume index and ejection fraction in patients following coronary artery bypass grafting

Abstract

Transesophageal echocardiography (TEE) has become a commonly used monitor of left ventricular (LV) function and filling during cardiac surgery. Its use is based on the assumption that changes in LV short-axis ID reflect changes in LV volume. To study the ability of TEE to estimate LV volume and ejection immediately following CABG, 10 patients were studied using blood pool scintigraphy, TEE, and thermodilution cardiac output (CO). A single TEE short-axis cross-sectional image of the LV at the midpapillary muscle level was used for area analysis. Between 1 and 5 h postoperatively, simultaneous data sets (scintigraphy, TEE, and CO) were obtained three to five times in each patient. End-diastolic (ED(a)) and end-systolic (ES(a)) areas were measured by light pen. Ejection fraction area (EF(a)) was calculated (EF(a) = (ED(a) - ES(a))/ED(a)). When EF(a) was compared with EF by scintigraphy, correlation was good (r = 0.82 SEE = 0.07). ED(a) was taken as an indicator of LV volume and compared with LVEDVI which was derived from EF by scintigraphy and CO. Correlation between ED(a) and LVEDVI was fair (r = 0.74 SEE = 3.75). The authors conclude that immediately following CABG, a single cross-sectional TEE image provides a reasonable estimate of EF but not LVEDVI.