Automatic real-time detection of myocardial ischemia by epicardial accelerometer

Objective: Myocardial ischemia may be detected with epicardial accelerometers. We developed and tested automated algorithms for real-time detection of myocardial ischemia by accelerometer measurements in both experimental and clinical settings. Methods: In 10 pigs, an accelerometer was fixed to the epicardium in the area perfused by left anterior descending coronary artery. Acceleration and electrocardiogram were simultaneously recorded, and the QRS complex was automatically detected for exact timing of systole. Peak circumferential velocity and displacement were automatically calculated from epicardial acceleration signal within 150 milliseconds after peak R on electrocardiography. Global myocardial function was reduced by esmolol infusion, and regional function was altered by temporary left anterior descending occlusion. Automated ischemia detection analyses were tested in 7 patients during off-pump coronary artery bypass grafting. Left anterior descending coronary artery was occluded for 3 minutes before grafting. In both models, echocardiographic myocardial circumferential strain was used to confirm ischemia. Results: Systolic displacement changed most during left anterior descending occlusion. Negative displacement during ischemia was found in pigs (11.5 +/- 2.3 to -1.2 +/- 2.8 mm, P < .01); regional hypokinesia was found in clinical study (12.8 +/- 8.1 to 3.5 +/- 4.4 mm, P < .01). Ischemia was confirmed by echocardiography in both settings. Esmolol infusion induced smaller changes in automated accelerometer measurements than did left anterior descending occlusion (P < .01). Conclusions: Automatic real-time detection of myocardial ischemia with epicardial accelerometer was feasible. Automated ischemia detection analysis may be used for continuous monitoring of myocardial ischemia during cardiac surgery. (J Thorac Cardiovasc Surg 2010;139:1026-32)