The added effects of cold spells on stroke admissions: Differential effects on ischemic and hemorrhagic stroke.

Abstract

Epidemiological evidence suggests an association between low ambient temperature and stroke risk, but available data are limited particularly on associations with different stroke subtypes. The aim of this study is to estimate the relationship between cold spells and stroke admissions, including the effect of cold spells on different stroke subtypes (ischemic stroke and intracerebral hemorrhage (ICH)). A total of 144,405 stroke admissions from the Tianjin Centre for Health and Meteorology Multidisciplinary Innovation in China, covering the period from January 2016 to December 2020, were studied, as well as meteorological and air pollutant data. A generalized additive model with a distributed lag nonlinear model was employed to assess the relationship, considering 12 different definitions of a cold spell based on various temperature thresholds and durations. The analysis controlled for lagged and nonlinear effects of temperature. Analyses were performed on all strokes as well as ischemic stroke and ICH. There was a significant increase in stroke admissions during cold spells. Generally, the increased risk during cold spells increased as the temperature threshold decreased, but was not significantly affected by the duration. The optimal model was obtained using the cold-spell definition based on an average daily temperature below the 10th percentile (0.11°C) for 2 or more consecutive days. According to this model, the effect of cold spells on ischemic stroke admissions had a significant lag effect and was long-lasting, with a single-day effect occurring on lag 7d, peaking on lag 13d (relative risk (RR) = 1.05; 95% confidence interval (CI) = 1.02 to 1.09), and lasting until lag 20d. In contrast, the effect on ICH was immediate and short-lived, with the most significant single-day effect occurring on the current day (RR = 1.17; 95% CI = 1.06 to 1.29) and limited within 3 days. 14.15% of stroke cases could be attributed to cold spells, with ICH exhibiting a higher burden than ischemic stroke except for strict temperature threshold definitions. Cold spells are associated with an increased stroke risk. Different patterns of association were seen for different stroke subtypes. The effect on ischemic stroke had a lag effect and a longer duration, whereas the effect on ICH had an immediate effect and a shorter duration. These findings support the development and improvement of stroke cold-spell early warning systems and highlight the importance of public health interventions to mitigate the adverse health impacts of cold spells.