Objective To investigate the effect of diabetes on coronary collateral development. Methods A dog model of diabetes was induced by intravenous injecting alloxan and streptozotocin. Dogs were subjected to brief (2 min), repetitive coronary artery occlusions (8 times per day for 21days) in the absence (ischemia n=8) or presence of diabetes (diabetes+ischemia n=6). A sham group (n=8) was instrumented identically but received no occlusion. Myocardial blood flow was determined in the myocardium perfused by the left anterior descending coronary artery (LAD) and the left circumflex coronary artery (LCCA). Peak reactive hyperemia response (PRH) was recorded on experimental days 1, 7, 14 and 21. Colla teral blood flow was expressed as the percentage of normal zone blood flow. Results Fasting blood glucose in the diabetes+ischemia group (17.5 mmol·L -1±1.1 mmol·L -1, 18.6 mmol·L -1±1.1 mmol·L -1, 21.6 mmol·L -1±1.5 mmol·L -1 and 21.9 mmol·L -1±1.5 mmol·L -1 on days 1, 7, 14 and 21, respectively) were significantly higher than those in the ischemia group (3.9 mmol·L -1±1.0 mmol·L -1, P<0.01). Coronary collateral blood flow was significantly increased in ischemia dogs (71%±7%) after 21 days but remained unchanged in sham dogs. The collateral blood flow in the diabetes+ischemia group after 21 days was only 17%±6%, which was significantly lower than that in the ischemia group (P<0.01). PRH in the ischemia group was 7.6 Hz·102±2.9 Hz·102 on days 21, which was significantly lower than that on day 1 (16.4 Hz·102±2.9 Hz·102, P<0.05). PRH remained unchanged in the diabetes+ischemia and sham group. PRH in diabetes+ischemia dogs (15.6 Hz·102±2.5 Hz·102) on days 21 was significantly higher than that in ischemia dogs (P<0.05). Conclusions Diatetes inhibits the development of coronary collateral blood flow. Objective To investigate the effect of diabetes on coronary collateral development. Methods A dog model of diabetes was induced by intravenous injecting alloxan and streptozotocin. Dogs were subjected to brief (2 min), repetitive coronary artery occlusions (8 times per day for 21 days) in The absence (ischemia n = 8) or presence of diabetes (diabetes + ischemia n = 6). A sham group (n = 8) was instrumented identically but received no occlusion. Myocardial blood flow was determined in the myocardium perfused by the left anterior Peak reactive hyperemia response (PRH) was recorded on experimental days 1, 7, 14 and 21. Colla teral blood flow was expressed as the percentage of normal zone blood flow (LCC) Results Fasting blood glucose in the diabetes + ischemia group (17.5 mmol·L -1 ± 1.1 mmol·L -1, 18.6 mmol·L -1 ± 1.1 mmol·L -1, 21.6 mmol·L -1 ± 1.5 mmol · L -1 and 21.9 mmol·L -1 ± 1.5 mmol·L -1 on days 1, 7, 14 and 21, r espectively) were significantly higher than those in the ischemia group (3.9 mmol·L -1 ± 1.0 mmol·L -1, P <0.01). Coronary collateral blood flow was significantly increased in ischemic dogs (71% ± 7%) after 21 The collateral blood flow in the diabetes + ischemia group after 21 days was only 17% ± 6%, which was significantly lower than that in the ischemia group (P <0.01). PRH in the ischemia group was 7.6 Hz · 102 ± 2.9 Hz · 102 on days 21, which was significantly lower than that on day 1 (16.4 Hz · 102 ± 2.9 Hz · 102, P <0.05). PRH remained unchanged in the diabetes + ischemia and sham group . PRH in diabetes + ischemia dogs (15.6 Hz · 102 ± 2.5 Hz · 102) on days 21 was significantly higher than that in ischemic dogs (P <0.05). Conclusions Diatetes inhibits the development of coronary collateral blood flow.