Background and Purpose—Seasonal variation in stroke incidence was examined using 14-year stroke registration data in a Japanese population. We also examined if this variation was modified by conventional stroke risk factors hypertension, diabetes mellitus, drinking, and smoking.

Methods—Data were obtained from the Takashima Stroke Registry, which covers a stable population of 55 000 in Takashima County in central Japan. There were 1665 (men, 893; women, 772) registered first-ever stroke cases during 1988 to 2001. The average age of stroke onset for men and women patients was 69.4 and 74.2 years, respectively. Incidence rates (per 100 000 person-years) and 95% CI were calculated by gender, age, and stroke subtype for winter, spring, summer, and autumn. After stratifying patients by their risk factor history, the OR (with 95% CI) of having a stroke in autumn, winter, and spring were calculated, with summer serving as a reference.

Results—Among the seasons, stroke incidence per 100 000 person-years was highest in the spring (231.3; 95% CI, 211.1 to 251.5). Spring incidence was highest in both men (240.8; 95% CI, 211.5 to 270.2) and women (222.1; 95% CI, 194.4 to 249.9), and in subjects younger than 65 years (72.6; 95% CI, 60.0 to 85.3) and 65 years or older (875.9; 95% CI, 787.5 to 964.3). Among stroke subtypes, spring incidence was highest for cerebral infarction (154.7; 95% CI, 138.2 to 171.2) and cerebral hemorrhage (53.7; 95% CI, 44.0 to 63.4). The spring excess in stroke incidence was observed regardless of the presence or absence of the risk factor histories.

Conclusions—Stroke incidence appears to be highest in the spring among a Japanese population regardless of conventional risk factor history. Factors that explain this excess need further investigation.