Acid-base changes occurring during cardiac arrest and subsequent CPR are related to a complex low-perfusion state characterized clinically by venous and tissue hypercarbic and metabolic (lactic) acidosis. This low-flow state is a dynamic process dependent on the time intervals between onset of arrest, initiation of CPR, and restoration of adequate spontaneous circulation. Increased release of CO2 from ischemic tissues and reduced CO2 transport from the tissues to the lungs result in profound tissue acidosis. However, recent experimental data suggest that even very low pH is compatible with neurologically intact survival. Thus, the clinical use of buffer agents, and especially of sodium bicarbonate, is currently controversial. Because results of controlled clinical studies are not available, a careful review of well-performed experimental studies is necessary. So far, the use of either CO2-generating or CO2-consuming buffers has not been proved conclusively to increase neurologically intact long-term survival after CPR. More importantly, adequate ventilation and effective chest compressions must be quickly established after cardiac arrest. This will counterbalance the hypercarbic and metabolic acidemia of cardiac arrest by creating concurrent hypocarbic arterial alkalemia during at least the early phase of CPR. Thus, the treatment of the complex acid-base changes associated with CPR is based primarily on the classical maneuvers of A and B (airway and breathing = adequate oxygenation and ventilation), C (chest compressions), and D (early defibrillation for rapid restoration of spontaneous circulation). In cases of prolonged cardiac arrest or preexisting metabolic acidemia, buffer therapy may be indicated.