Background: A large proportion of patients who suffer out-of-hospital cardiac arrest have asystole as the initial recorded arrhythmia. Since they have a poor prognosis, less attention has been paid to this group of patients. Aim: To describe a consecutive population of patients with out-of-hospital cardiac arrest with asystole as the first recorded arrhythmia and to try to define indicators for an increased chance of survival in this population. Setting: The community of Gothenburg. Patients: All patients who suffered out-of-hospital cardiac arrest during 1981 to 1992 and were reached by our emergency medical service (EMS) system and where cardiopulmonary resuscitation (CPR) was attempted. Results: In all there were 3434 cardiac arrests of which 1222 (35%) showed asystole as the first recorded arrhythmia. They differed from patients with ventricular fibrillation by being younger, including more women and having a longer interval between collapse and arrival of the first ambulance. In all 90 patients (7%) were hospitalized alive and 20 (2%) could be discharged from hospital. Independent predictors for an increased chance of survival were: (a) a short interval between the collapse and arrival of the first ambulance (P < 0.001) and the time the collapse occurred (P < 0.05). Initial treatment given in some cases with adrenaline, atropine and tribonate were not associated with an increased survival. Conclusions: Of all the patients with out-of-hospital cardiac arrest, 35% were found in asystole. Of these, 7% were hospitalized alive and 2% could be discharged from hospital. Efforts should be made to improve still further the interval between collapse and arrival of the first ambulance.