Summary: The purpose of this study was to determine whether the selective type IV cAMP-phosphodiesterase inhibitor rolipram could reduce the reperfusion injury that occurs during myocardial infarction in the anesthetized dog. This question was tested in pentobarbital-anesthetized dogs subject to 90 min of regional myocardial ischemia and 5 h of reperfusion. Dogs were treated with 1 mg/kg of rolipram (i.v., 15 min before reperfusion) followed by a 1 mg/kg/h infusion over the duration of the 5 h of reperfusion. Rolipram was tested in vitro for efficacy in inhibition of isolated human neutrophil superoxide generation. Rolipram produced significant inhibition of superoxide production over the concentration range of 0.1–100 μM- rolipram when neutrophils were stimulated with a 10 -7M concentration of the chemotactic peptide f-Met-Leu-Phe. Rolipram significantly inhibited superoxide generation from human and canine granulocytes in whole blood stimulated by zymosan. Therapeutic concentrations of rolipram in the blood of dogs were achieved during the course of the experiments with a plasma concentration of 0.761 ± 0.095 μg/ml (2.76 ± 0.34 μ- M) at the time of reperfusion, and 0.574 ± 0.098 μg/ml (2.08 ± 0.36 μ M) at the end of the reperfusion period. The relative severity of myocardial ischemia between the two treatment groups was similar as assessed with radiolabeled microsphere measurement of myocardial blood flow. Transmural myocardial blood flows were not significantly different between the two groups after coronary occlusion (control, 0.05 ± 0.01 ml/min/g, n = 6, vs. rolipram, 0.18 ± 0.07 ml/min/g, n = 6; p = 0.48). There was not a significant reduction in the accumulation of neutrophils in the myocardium as assessed by the marker enzyme myeloperoxidase (control, 2.33 ± 1.07 U/g, n = 5, vs. rolipram, 0.54 ± 0.28 U/g, n = 6; p = 0.0548). Myocardial infarct size determined postmortem was not limited by rolipram treatment (control, 31.1 ± 2.6% infarct/ area at risk, n = 6, vs. rolipram, 23.3 ± 7.8% infarct/area at risk, n = 6; p > 0.05). Even though rolipram did not result in statistically significant protection as measured by infarct size, there was a significant attenuation of the “no-reflow” phenomenon with rolipram treatment. Circumflex coronary blood flow declines that were evident in the vehicle-treated group did not occur in the rolipram treatment group. These data indicate that cAMP-phosphodiesterase inhibition with rolipram is not sufficient to provide protection against myocardial reperfusion injury.