ObjectiveTo evaluate the blood flow and perfusion pressure differences observed during open- vs. closed-chest cardiopulmonary resuscitation (CPR), including the effects of epinephrine and sodium bicarbonate administration.DesignProspective, randomized, controlled trial.SettingExperimental animal laboratory in a university hospital.SubjectsA total of 35 anesthetized piglets.InterventionsAfter tracheostomy and insertion of arterial, right atrial, and pulmonary arterial catheters, thoracotomy was performed with placement of a pulmonary arterial flow probe and left atrial catheter. Ventricular fibrillation was induced and followed by 15 mins of either open-chest (n = 14) or closed-chest (n = 21) CPR. A 4-min infusion of 50 mmol of sodium bicarbonate or saline was added at the start of CPR. After 8 mins of CPR, 0.5 mg of epinephrine was given intravenously, and after 15 mins, direct current (DC) shocks were used to revert the heart to sinus rhythm.Measurements and Main ResultsBlood flow was studied using transit-time ultrasound flowmetry. In an extended group, intrathoracic pressure was measured for calculation of transmural pressure. Before epinephrine administration, mean pulmonary arterial flow (cardiac output) was reduced: a) during closed-chest CPR relatively more than pulmonary perfusion pressure but in proportion to systemic perfusion pressure; b) during open-chest CPR relatively less than pulmonary perfusion pressure but still in proportion to systemic perfusion pressure. Epinephrine administration temporarily increased systemic perfusion pressure during both closed- and open-chest CPR but temporarily decreased pulmonary perfusion pressure only during closed-chest CPR. After epinephrine administration, cardiac output temporarily decreased during both closed- and open-chest CPR.ConclusionsOpen-chest CPR resulted in better cardiac output and systemic perfusion pressure than closed-chest CPR. However, cardiac output values obtained with both methods were much lower than previously reported. After epinephrine administration, cardiac output became extremely low with both methods.(Crit Care Med 1995; 23:715-725)