Objective: Previous studies suggest that epinephrine may alter the correlation of perfusion with measures of Pco(2) during cardiopulmonary resuscitation. This study investigated the effects of epinephrine on Pace, and mixed venous Pco, in a high-flow reperfusion model of cardiac arrest. Design: Prospective, block randomized, blinded, placebo controlled, laboratory study. Subjects: Thirty mixed breed canines. Interventions: After a 12-min ventricular fibrillation cardiac arrest, 30 mixed breed ca. nines were reperfused with standardized (3200 revolutions/min) cardiopulmonary bypass and were given placebo (n = 10), standard dose epinephrine (0.02 mg/kg; n = 10), or high-dose epinephrine (0.2 mg/kg; n = 10). Arterial and mixed venous blood gases, coronary perfusion pressure, pump flow and peripheral vascular resistance were compared between groups during the early reperfusion period using analysis of variance with a post hoc Tukey's multiple comparison test. Measurements and Main Results: Baseline variables were similar between groups. With reperfusion, the high-dose epinephrine group had higher coronary perfusion pressures (p < .002), lower systemic pump flow (p < .01), and higher peripheral vascular resistance (p < .001). In the high-dose epinephrine group, both Paco(2) (high-dose epinephrine, 40 +/- 6 torr [5.3 +/- 0.8 kPa]; standard dose epinephrine, 45 +/- 7 torr [6.0 +/- 0.9 kPa]; placebo, 54 +/- 4 torr [7.2 +/- 0.5 kPa];p < .01) and mixed venous Pco(2) (high-dose epinephrine, 55 +/- 10 torr [7.3 +/- 1.3 kPa]; standard dose epinephrine, 57 +/- 9 torr [7.6 +/- 1.2 kPa]; placebo, 67 +/- 4 torr [8.9 +/- 0.5 kPa]; p < .05) were significantly decreased, and arterial pH, Pao(2), and mixed venous Po-2 were significantly increased compared with the placebo group. Conclusion: In this model, when ventilation and CO2 production are constant, the decrease in Paco(2) with epinephrine is due to decreased pulmonary blood flow (flow to membrane oxygenator) and peripheral shunting.