The effect of anesthetics on ischemic myocardium to which blood was supplied by a stenotic coronary artery was investigated in dogs. The ischemia was assessed by regional wall motion (ultrasonic dimension technique) using fractional shortening (FS) [(EDL - ESL)/EDL x 100] and end-systolic pressure-segment length relationships (ESPLR). The latter is considered to be a more load-independent measure of regional myocardial function. Isoflurane and fentanyl were chosen as anesthetics of current interest. On reducing the left circumflex coronary artery (LCX) flow to approximately 50% of its resting value, a decrease in FS and a rightward shift in ESPLR were observed in myocardium perfused by the LCX. Simultaneously, increases in FS were observed in the nonischemic area perfused by the left anterior descending coronary artery (LAD), which was most likely due to the intraventricular unloading effect. No significant changes of ESPLR were observed in the area supplied by LAD. Isoflurane induced a dose-dependent decrease in FS and a rightward shift in ESPLR in the ischemic myocardial segment, whereas fentanyl caused an increase in FS and tended to shift ESPLR leftward in the same area. The results suggest that isoflurane may have deleterious effects on preexisting myocardial ischemia, whereas fentanyl may not when loading conditions are taken into consideration. Fractional shortening and ESPLR seem to provide similar information about regional myocardial function.