ObjectiveTo determine if the administration of epinephrine changes the partial pressure of end-tidal CO2during cardiac arrest, as previously reported. Such a change could diminish the demonstrated ability of end-tidal CO2measurements to predict resuscitation from cardiac arrest.DesignThe partial pressures of end-tidal CO2of adult cardiac arrest patients who received iv epinephrine in doses from 1 to 15 mg were monitored throughout arrest.SettingEmergency department of a university hospital.PatientsAdults (n = 64) in cardiac arrest with a mean age of 70 ± 12 yrs, of whom 35 were males and 15 had a mean time of return of spontaneous circulation of 6.5 ± 11 hrs.InterventionsEnd-tidal CO2(in torr) was analyzed on arrival, before the first dose of epinephrine, and 4 mins after epinephrine was administered in varying doses chosen by the supervising physician.Measurements and ResultsThe end-tidal CO2decreased an average of 0.3 torr (0.04 kPa) after epinephrine was administered. Patients with a return of pulse had a decrease of −2 torr (-0.3 kPa) vs. an increase of 0.3 torr (0.04 kPa) for those patients with no return of pulse (p =.07). In 33% of patients, there was no change; in 28%, the partial pressure of end-tidal CO2increased, and in 39%, it decreased. There was no correlation between the change in end-tidal CO2after epinephrine and whether or not patients regained a pulse (r2=.08, p =.07), although a decrease in end-tidal CO2was most often associated with return of pulse. At a threshold of 10 torr (1.3 kPa), the first end-tidal CO2had a positive predictive value for return of pulse of 50% and a negative predictive value of 82%. Just before epinephrine administration, the positive predictive value was 71% and the negative predictive value was 83%; 4 mins after epinephrine administration, the positive predictive value was 64% and the negative predictive value was 86%. A decrease in end-tidal CO2after epinephrine had a positive predictive value of 53% and a negative predictive value of 92%. End-tidal CO2readings predicted resuscitation most accurately when taken after initial stabilization and before administration of epinephrine.ConclusionsAlthough epinephrine administration may decrease end-tidal CO2tensions in cardiac arrest, it does so unpredictably in individual patients, and it does not eliminate the predictive value of this measurement.