The goal of the proposed research is to develop a 3-dimensional scanning biplane transesophageal echo (3D TEE) probe to measure left ventricular (LV) function during surgery, to validate the accuracy of quantitative analysis of regional and global function in 3-dimensions using this probe, and to investigate the suitability of such derived measures in outcome investigations in coronary artery or related surgery. We have demonstrated the feasibility of 3D multiplanar TEE interrogation of the LV during surgery in both animal and man. Global end diastolic (EDV) and end systolic (ESV) volumes, stroke volume, and ejection fraction (EF) measurements were shown in animals to be sufficiently accurate to supplant current clinical techniques. We developed a method for measuring regional EF from 3D reconstructions where computations are portrayed as histograms and/or color-coded 3D reconstructions. Cardiac surgery patients were studied, of whom 15/23 had regional akinesis and 8/23 normal function. The 3D TEE histograms agreed with contrast ventriculography 82% of the time. Of the 18% discrepancy, 82% of it was due to improper imaging of the basal or apical areas. Due in part to uncertainty in aligning 3D reconstructions to ventriculogram orientations, regional comparisons of the EF color-coded reconstructions gave lower agreement except in the apical regions for long axis scans (agreement - 83%). Although preliminary, the potential for quantitative 3D analysis is evident. However, the need for a probe that will scan the entire LV and a means of better orientation for comparative studies is also clear. Consequently, we propose to develop a new biplane probe which overcomes these current limitations. We will add to our interrogation and analysis the study of the epicardium and develop a method for correcting for heart translocation. An animal coronary artery occlusion model will be used to test the accuracy and diagnostic sensitivity of global and regional LV function measurement. The normal mean and standard deviation for regional LV assessment by our method will be defined from the data of patients with a low probability of heart disease who are undergoing orthopedic or gynecological surgery. Repeatability, accuracy and precision of measures will be investigated during cardiac surgery and postoperative recovery in the intensive care unit. Finally we will investigate whether we can detect changes in myocardial performance, before and after surgery, in regions receiving grafts. Of particular interest will be to learn if we can identify regions of the myocardium that may be stunned or hibernating.
Showing the most recent 10 out of 23 publications
