This study was designed to use real-time imaging to test the hypothesis that delayed cardiac protection induced by volatile anesthetics inhibits apoptosis. Rats were divided into two groups. One group was exposed to 120 min of 33 % O 2 [control group (CON group)] and the other group was exposed to 2.5 % sevoflurane in 33 % O 2 for 120 min [sevoflurane group (SEVO group)]. Both groups were allowed to return to their cages for 24 h. After 24 h recovery, all rats underwent 30 min myocardial ischemia by occluding coronary artery followed by 2 h of reperfusion. After reperfusion, technetium-99m-labeled annexin-V was administered intravenously to identify apoptosis. Left ventricular samples were obtained to measure infarct size and radionuclide imaging and caspase-3. Radionuclide imaging indicated that apoptosis was reduced in SEVO group (0.78 % ± 0.82) when compared with the CON group (1.15 % ± 0.61), and the infarct size was also decreased in the SEVO group (40 % ± 7). The transferase dUTP nick end labeling (TUNEL)-positive cardiomyocytes in the SEVO group (16 % ± 6) were significantly decreased in the peri-infarct zone when compared with the CON group (28 % ± 4). After reperfusion, caspase-3 expression was significantly blunted in the SEVO group than in CON group (50 % ± 11 vs. 68 % ± 10, p \ 0.05). This study used technetium-99m-labeled annexin-V of real-time imaging to detect cardiomyocyte apoptosis and the results were confirmed by the TUNEL assay and caspase-3 expression. We concluded that delayed volatile anesthetic preconditioning (APC) protects against I/R in vivo. The method of technetium-99m-labeled annexin-V of real-time imaging can be used to detect cardiomyocyte apoptosis in delayed APC during ischemia/reperfusion.