AIM: One of the objectives of this study was to assess the emergency medical dispatchers (EMDs) ability for the identification and prioritisation of cardiac arrest (CA) cases, and offering and achievements of dispatcher-assisted bystander cardiopulmonary resuscitation (CPR). The other objective was to give an account of the frequency of agonal respiration in cardiac arrest calls and the caller's descriptions of breathing. METHODS: Prospective study evaluating 100 tape recordings of the EMD calls of emergency medical service (EMS)-provided advanced life support- (ALS) cases, of out-of-hospital cardiac arrest. RESULTS: The quality of EMD-performed interviews was highly commended in 63% of cases, but insufficient or unapproved in the remaining 37%. The caller's state of mind was not a major problem for co-operation. Among the 100 cases, 24 were suspected to be unconscious and in respiratory arrest. A further 38 cases were presented as unconscious with abnormal breathing. In only 14 cases dispatcher-assisted bystander CPR was offered by the EMD, and in 11 of these it was attempted, and completed in eight. Only four of the cases were unconscious patients with abnormal breathing. The incidence of suspected agonal breathing was estimated to be approximately 30% and the descriptions were; difficulty, poorly, gasping, wheezing, impaired, occasional breathing. CONCLUSIONS: Among suspected cardiac arrest cases, EMDs offer CPR instruction to only a small fraction of callers. A major obstacle was the presentation of agonal breathing. Patients with a combination of unconsciousness and agonal breathing should be offered dispatcher-assisted CPR instruction. This might improve survival in out-of hospital cardiac arrest.