Circadian, circaseptan and circannual periodicity of cardiac arrest.

In this issue, Arntz and colleagues published an interesting report on the daily (circadian), weekly (circaseptan) and monthly (circannual) variation in the occurrence of sudden death in West Berlin during the 5-year period from 1987 to 1991. The city of West Berlin, with about 2 million inhabitants, offered a unique opportunity of studying a population that was clearly defined within ‘the Wall’. Moreover, this population remained nearly constant on a day-to-day basis because of the almost complete absence of ‘noise’ produced by commuters to and from the study area. During this 5 year period, demographic data were collected prospectively on 24 061 consecutive cases of cardiac arrest of presumed cardiac origin attended by the West Berlin emergency medical services. The authors estimated that the figure of approximately 5000 events per year represents about 50% of all sudden deaths of presumed cardiac origin. This amounts to an annual incidence of 10 000 sudden deaths of presumed cardiac origin, corresponding to an incidence of five per 1000 inhabitants per year. According to the authors, this surprisingly high figure is probably related to the unusually elderly age profile of the population. The figure contrasts with the data from MONICA, in which three German collaborative centres had an annual incidence of heart attacks ranging from 276 to 376 per 100 000 for males and from 49 to 78 per 100 000 for females. Certainly, there are differences between the MONICA and the Berlin study populations: in MONICA there is an age limit for inclusion in the database of 65 years, whereas in Berlin there is no age limit. Assuming an observed case fatality rate of 50%, with 60% of deaths occurring outside the hospital, the figure of sudden death of presumed cardiac origin obtained in Berlin remains high. The chronobiological data of this unusual study population are supported, however, by extensive literature from other parts in the world. No