Background. This study was undertaken to test the hypothesis that administration of superoxide dismutase (SOD) restores the contractile and metabolic dysfunction in coronary microembolization and that these beneficial effects of SOD are attributable to the restoration of 5'-nucleotidase activity and subsequent augmentation of adenosine release.Methods and Results. In 78 dogs before and after an injection of microspheres (15 pm in diameter) into the left anterior descending coronary artery, regional coronary blood flow (CBF), fractional shortening (FS), and lactate extraction ratio (LER) were measured with and without administration of recombinant human SOD (50 ,g/kg/min i.c.). In the untreated dogs (n=6), both FS and LER decreased after coronary microembolization (2.0x10' microspheres per ml CBF [mLIminJ). FS and LER decreased from 24.2±13% to 5.1±1.2% and from 23.0+1.1% to -10.5+2.9%o, respectively. These ischemic changes were associated with coronary hyperemic flow (141±8 versus 92±1 mL/100 g/min) and adenosine release (5.8±0.5 versus 0.4±0.1 nmol/100 g/min). Pretreatment with SOD augmented the hyperemic flow to 164±4 mVJ100 g/min and enhanced the release of adenosine (9.6±0.6 nmol/100 g/min) associated with improvement of functional and metabolic dysfunction (FS, 14.8+2.3%; LER, 15.1±3.1%). Administration of SOD at 10 minutes (n=5) and 30 minutes (n=5) after coronary embolization restored the contractile function and lactate metabolism (at 10 minutes: FS, 16.7±2.2% and LER, 16.7±3.9%o; at 30 minutes: FS, 11.1±1.3% and LER, 7.2±3.1%). However, administration of SOD 60 minutes after coronary embolization (n=6) did not restore the contractile and metabolic dysfunction. The restoration of the contractile and metabolic dysfunction by SOD treatment was blunted by adenosine receptor blockade with 8-phenyltheophylline (n=5). Myocardial 5'-nucleotidase activity at 2 hours after embolization was restored with SOD treatment at 10 minutes (n=5) and 30 minutes (n=5) after embolization. However, SOD treatment at 60 minutes after embolization (n=6) did not restore 5'-nucleotidase activity compared with the SOD pretreatment group. Furthermore, coronary submaximal vasodilation induced by papaverine (n=5) and adenosine (n=5) abolished the beneficial effects of SOD.Conclusions. We conclude that 1) in sustained myocardial ischemia, SOD treatment attenuates ischemic injury caused by coronary microembolization by restoration of 5'-nucleotidase activity and augmentation of adenosine release; 2) this beneficial effect of SOD is observed even after coronary microembolization; and 3) the beneficial effects of SOD are attributable to coronary vasodilation produced by augmented adenosine release. (Circulation 1993;87:982-995 To test the hypothesis that SOD restores contractile and metabolic dysfunction after the onset of coronary microembolization, we administered SOD before and 10, 30, 60, and 120 minutes after embolization of microspheres. We also determined whether the beneficial effects of SOD are attributable to coronary vasodilati...