Ｏｂｊｅｃｔｉｖｅ　To study the efficiency of ventilationin in out-of-hospital cardiopulmonary resuscitation(CPR) by performing compression-only CPR or conventional CPR in rats of different causes of cardiac arrest(CA).　Methods　Forty five SD rats were randomly divided into 5 groups: asphyxia-conventional group(group A, n=10), asphyxia-compression-only group(group B, n=10), ventricular fibrillation-conventional group(group C, n=10), ventricular fibrillation-compression-only group(group D, n=10), group sham(n=5). The model of asphyxial CA was made by clamping endotracheal tube at expiration in group A and group B. The model of ventricular fibrillation CA was made by transoesophageal cardiac pacing in group C and group D. Bystander CPR was initiated 4 min later. Conventional CPR(compression∶ventilation=30∶2) was performed in group A and group C, compression-only CPR was performed in group B and group D. For animals that did not attain return of spontaneous circulation(ROSC) during the 8 min bystander CPR period, advanced life support(ALS) was provided. The animals which did not attain ROSC after 10 min ALS should be given up. The pathological specimens of brain, ventricular wall and right lower lobe of the lung were collected after 24 h, and the samples were collected in the death of 24 h. Post-resuscitation treatment had been performed in all successfully resuscitated animals for 1 h. At 24 h after ROSC, rats were sacrificed for histological analysis. The vilai signs was monitored for 1 h in goup sham, rats were sacrificed for histological analysis after survival 24 h. The major outcome measure was rates of ROSC, and the secondary was the survival rates of 24 h. Blood samples were drawn for blood gas analysis during resuscitation in the pathologic findings.　Results　Rates of ROSC were higher in group A and group C compared with group B and group D, but the difference was not significant. Survival rates of 24 h were compared, group A and group C were significantly higher than group B and group D(P<0.05), and the tissue damage in group A and group C were milder than group B and group D.　Conclusions　Conventional bystander CPR was associated with a higher survival of 24 h than compression-only CPR. Conventional bystander CPR might be the preferable approach of resuscitation for adult patients with witnessed out-of-hospital CA, whether cardiogenic or non-cardiogenic CA.