Short-Term Exposure to Low and High Temperatures and Mortality Among Patients With Heart Failure in Sweden

Importance Patients with heart failure may be particularly susceptible to nonoptimal temperature exposure, but the associations between short-term low and high temperature exposure and mortality in this population remain unclear, especially in Sweden—a high-latitude country where no nationwide study has been conducted. Objective To investigate the associations between short-term exposure to low and high ambient temperatures and all-cause and cardiovascular mortality among Swedish patients with heart failure. Design, Setting, and Participants This nationwide, time-stratified case-crossover study was conducted in Sweden among 250 640 patients with heart failure who died from any cause from 2006 to 2021, identified from the Swedish National Patient Register and the Cause of Death Register. Exposure Daily mean ambient temperature was assessed at 1 × 1-km spatial resolution. To account for regional adaptation, temperature exposures were defined using municipality-specific percentiles, with low and high temperatures corresponding to the 2.5th and 97.5th percentiles, respectively. Main Outcomes and Measures The primary outcome was all-cause and cardiovascular mortality among patients with heart failure. Results The mean (SD) age at death among patients with heart failure was 84.3 (9.4) years, with 121 061 female patients (48.3%). Short-term exposure to ambient temperature demonstrated a U-shaped association with both all-cause and cardiovascular mortality. For all-cause mortality, odds ratios (ORs) were 1.130 (95% CI, 1.074-1.189) for low temperatures and 1.054 (95% CI, 1.017-1.093) for high temperatures over the entire study period. For cardiovascular mortality, low temperatures were associated with an OR of 1.160 (95% CI, 1.083-1.242) over the entire study period, and high temperatures with an OR of 1.084 (95% CI, 1.014-1.159) during 2014-2021. The mortality risk associated with high temperatures was more pronounced during the 2014-2021 period compared to 2006-2013. Male patients, those with comorbid diabetes, and diuretic users were more susceptible to low temperatures, whereas high temperature was more strongly associated with mortality in patients with comorbid atrial fibrillation or flutter and those exposed to elevated ozone levels. Conclusions and Relevance This nationwide Swedish time-stratified case-crossover study indicates that short-term exposure to both low and high temperatures was associated with increased risk of all-cause and cardiovascular mortality in patients with heart failure. The mortality risk associated with high temperatures appears to be increasing over time, emphasizing the need for adaptation, even in high-latitude regions.