In the western world, ∼20% of all deaths among people between 20 and 80 years old are sudden and unexpected and this has not changed much during recent decades, whereas mortality from cardiac disease has diminished substantially.1 Sudden death is most often caused by the onset of ventricular fibrillation (VF) or ventricular tachycardia, interfering with the heart's function to pump blood into the circulation and thus causing a cardiac (and circulatory) arrest. We know that immediate termination of the life-threatening arrhythmia by means of defibrillation, thereby restoring spontaneous circulation, provides many cardiac arrest victims several ‘saved’ life-years of satisfactory quality.2 Over the years numerous articles have addressed two crucial questions: ‘Can we identify a person with increased risk of cardiac arrest before the event?’ and, ‘How to optimize the resuscitation attempt and the management of the victim after return of a stable cardiac rhythm?’ Regarding the first question, we know that our current methods of risk stratification identify only a small fraction of the victims before the event.3–6 Also, no new methods have been published that provide cardiac arrest predictors with sufficient sensitivity and specificity to both identify a substantial portion of all arrest victims in advance and to be specific enough to make preventive action (defibrillator implantation) practical and cost-effective. This somewhat disappointing conclusion forces us to concentrate on the second question. Several recent articles discussed our current insights into incidence of cardiac arrest, profile of the victim, rhythm at arrest, and results of the resuscitation attempt.7–10 They show that outcomes of witnessed cardiac arrests have improved in recent years due to greater emphasis on resuscitation …