BACKGROUND: Attenuation of myocardial stunning by several angiotensin-converting enzyme (ACE) inhibitors has been demonstrated. However, the signal cascade mediating such protective effect has not been analyzed in detail so far. METHODS AND RESULTS: In a first protocol, we addressed the role of bradykinin and analyzed the effect of the ACE inhibitor ramiprilat without and with added bradykinin B2 receptor antagonist HOE 140 on regional myocardial blood flow (colored microspheres) and function (sonomicrometry). Thirty-two enflurane/N2O-anesthetized open-chest dogs were subjected to 15 minutes of occlusion of the left circumflex coronary artery (LCx) and 4 hours of subsequent reperfusion. Eight dogs served as placebo controls (group 1), and 8 dogs received ramiprilat (20 micrograms/kg IV) before LCx occlusion (group 2). Eight dogs received a continuous intracoronary infusion of HOE 140 [0.5 ng/(mL.min) IC] during ischemia and reperfusion (group 3), and in 8 dogs HOE 140 was infused continuously during ischemia and reperfusion, starting 45 minutes before the administration of ramiprilat (group 4). Mean aortic pressure was kept constant with an intra-aortic balloon, and heart rate did not change throughout the experimental protocols. Under control conditions and during myocardial ischemia, posterior transmural blood flow (BF) and systolic wall thickening (WT) were not different in the four groups of dogs. However, at 4 hours of reperfusion, WT was still depressed in groups 1 (-10 +/- 20% of control [mean +/- SD]), 3 (-18 +/- 12% of control), and 4 (-12 +/- 21% of control), whereas WT in group 2 had recovered to 55 +/- 20% of control (P < .05 versus group 1). BF at 4 hours of reperfusion was not different in the four groups of dogs. Thus, the beneficial effect of ramiprilat on the functional recovery of stunned myocardium was obviously mediated by bradykinin. Since bradykinin stimulates the formation of both prostaglandins and nitric oxide, we tested in a second protocol which of these mediators was further involved in the beneficial effects of ramiprilat. Twenty-four additional dogs were subjected to 15 minutes of LCx occlusion and 4 hours of reperfusion. Six dogs received the cyclooxygenase inhibitor indomethacin (10 mg/kg IV) (group 5) and 6 dogs a combination of indomethacin with ramiprilat (group 6) before LCx occlusion. Six dogs received the nitric oxide synthase inhibitor NG-nitro-L-arginine methyl ester (L-NAME) (20 mg/kg IV) (group 7) and 6 dogs a combination of L-NAME with ramiprilat (group 8) before LCx occlusion. BF and WT before and during myocardial ischemia were not different in groups 5 and 6 and groups 7 and 8. However, at 4 hours of reperfusion, WT was still depressed in groups 5 (-10 +/- 38% of control), 6 (-7 +/- 18% of control), and 7 (-12 +/- 14% of control), whereas WT in group 8 had recovered to 47 +/- 28% of control (P < .05 versus group 7). BF at 4 hours of reperfusion was not different in the four groups of dogs. CONCLUSIONS: In summary, the attenuation of stunning by the ACE inhibitor ramiprilat involves a signal cascade of bradykinin and prostaglandins but not nitric oxide.