[Seasonal variation in strokes incidence and the influence of the meteorological conditions].

The aim of this paper was to study the seasonal and Monthly distribution of different subtypes of strokes, and also the influence of the most usual meteorological factors on their incidence. The study, based on the population-based data of the Dijon register of stroke (France), involved 3287 patients with a cerebrovascular event during the Years 1985-1998. The seasonal distribution pointed out a summer decrease for all stroke subtypes. The cerebrovascular risk was the highest in autumn for small artery atheroma infarcts and for cerebral infarcts and, among these later, for cardioembolic infarcts and large artery atheroma infarcts, as well as for subarachnoid hemorrhages and for the total number of strokes. It was followed by spring (total number of strokes, small artery atheroma infarcts, cerebral infarcts) and/or winter (subarachnoid hemorrhages, cerebral infarcts, cardioembolic infarcts and large artery atheroma infarcts). For intracerebral hemorrhages, the risk peaked in spring and secondly in autumn. But the difference from one season to another was significant only for total number of strokes, cerebral infarcts and, among the later, for cardioembolic infarcts and large artery atheroma infarcts. As regards the Monthly distribution, the difference between Months was significant only for the total number of strokes and cerebral infarcts, with a minimum from July to September and a maximum in October for the later. Correlations with meteorological data were found for the total number of strokes, cerebral infarcts, cardioembolic infarcts, large artery atheroma infarcts and small artery atheroma infarcts. They showed an influence of temperature and relative humidity of the day of stroke or of the one to five days before. Correlations with wind speed, duration of sunshine or snow, even if less frequent, could also be found. Such results sometimes differ from those of some earlier studies, for which the climate and the risk factors were not the same, showing that regional epidemiological studies are necessary in order to determine the relations between seasons, meteorological factors and strokes.