Experience has shown that better quality CPR leads to a greater chance of a patient surviving a cardiac arrest. Simple CPR techniques, such as using only chest compressions, lead to better skill retention and greater willingness to attempt resuscitation on strangers. However, it is not clear from clinical or experimental studies whether such techniques offer any physiological benefit over more usual 5:1 or 15:2 compression:ventilation ratios. Computer simulations of blood flow and gas exchange during CPR showed that continuous chest compressions produced much greater blood flow (1.39 l/min) than 5:1 (0.73 l/min), 15:2 (0.86 l/min) or 50:5 (0.94 l/min) ratios. However, the ratio of 5:1 produced the highest arterial oxygen levels, with continuous chest compressions the lowest. The most appropriate measure of CPR efficiency appears to be the amount of oxygen delivered to the body during CPR. The ratios of 15:2 and 50:5 produced significantly greater oxygen delivery to the body than 5:1, the greater blood flow with these techniques offsetting the slightly lower arterial oxygen levels. The best oxygen delivery was provided by continuous chest compression in the early stages of CPR. After 3-4 min however, hypoxia meant that continuous compressions became worse than the other techniques. Spontaneous gasping by the patient during CPR was able to extend the effectiveness of chest compression only CPR for much longer.