Objective To investigate the association between ambient average temperature and emergency hospital visits (EHVs) for actue cerebrovascular disease in Changsha, China.

Methods The daily EHVs for acute cerebrovascular disease were gathered from the third Xiangya hospital during 2008–2009 and the corresponding meteorological, air pollution data were collected. The seasonal-stratified case-crossover design was used for the data analysis.

Results After adjusting the influence of relative humidity, wind speed and atmospheric pressure, a 1°C increase of daily average was associated with cerebral haemorrhage (OR = 1.278.95%CI: 1.045–1.563), cerebral infarction (OR = 1.223.95%CI: 1.056–1.416), TIA (OR = 1.318.95%CI: 1.099–1.581 ) and total acute cerebrovascular disease (OR = 1.239.95%CI: 1.085–1.414) (P < 0.05) in spring. It was also associated with cerebral infarction (OR = 1.201.95% CI: 1.006–1.434) and total acute cerebrovascular disease (OR = 1.179.95%CI: 1.002–1.387) (P < 0.05) in summer. In autumn, it was associated with cerebral haemorrhage (OR = 0.803. 95%CI: 0. 683–0.944), cerebral infarction (OR = 0.765.95%CI: 0.664–0.882), TIA (OR = 0.794.95%CI: 0.671–0.940) and total acute cerebrovascular disease (OR = 0.792.95%CI: 0.699–0.898) (P < 0.05).

Conclusions These data on outcomes suggested that the elevated level of daily average temperature would increase the daily EHVs for cerebral haemorrhage, cerebral infarction, TIA and total acute cerebrovascular disease in spring while it would decrease the diseases above in autumn. In summer, it would increase the daily EHVs for cerebral infarction and total acute cerebrovascular disease.