BACKGROUND: A large proportion of cardiac arrests outside hospital are caused by ventricular fibrillation. Although it is frequently used, the exact role of treatment with adrenaline in these patients remains to be determined. AIM: To describe the proportion of patients with witnessed out-of-hospital cardiac arrest found in ventricular fibrillation who survived and were discharged from hospital in relation to whether they were treated with adrenaline prior to hospital admission. PATIENTS AND TREATMENT: All the patients with out-of-hospital cardiac arrest found in ventricular fibrillation in Göteborg between 1981 and 1992 in whom cardiopulmonary resuscitation (CPR) was initiated by our emergency medical service (EMS). During the observation period, some of the EMS staff were authorized to give medication and some were not. RESULTS: In all, 1360 patients were found in ventricular fibrillation and detailed information was available in 1203 cases (88%). Adrenaline was given in 417 cases (35%). Among patients with sustained ventricular fibrillation, those who received adrenaline experienced the return of spontaneous circulation more frequently (P < 0.001) and were hospitalized alive more frequently (P < 0.01). However, the rate of discharge from hospital did not differ significantly between the 2 groups. Among patients who converted to asystole or electromechanical dissociation, those who received adrenaline experienced the return of spontaneous circulation more frequently (P < 0.001) and were hospitalised alive more frequently (P < 0.001). However, the rate of discharge from hospital did not differ significantly between the 2 groups. CONCLUSIONS: On the basis of 2 treatment regimens during a 12-year survey, we explored the usefulness of adrenaline in out-of-hospital ventricular fibrillation. Both patients with sustained ventricular fibrillation and those who converted to asystole or electromechanical dissociation had an initially more favourable outcome if treated with adrenaline. However, the final outcome was not significantly affected. This study does not confirm the hypothesis that adrenaline increases survival among patients with out-of-hospital cardiac arrest who are found in ventricular fibrillation.