Cold Spells, Heat Waves, and Nonoptimal Air Temperature and Risk of Heart Failure Hospitalization in Sweden

Extreme temperature events are projected to increase with climate change; yet, associations of short-term heat and cold exposure with heart failure (HF) exacerbations remain incompletely characterized, especially in high-latitude settings.

The purpose of this study was to investigate the associations of short-term exposure to cold spells, heat waves, and nonoptimal ambient temperature with HF hospitalization in Sweden.

We conducted a nationwide time-stratified case-crossover study including 482,000 HF hospitalizations recorded in the Swedish National Patient Register (2006-2021). Cold spells and heat waves were defined as ≥2 consecutive days with daily mean temperature ≤5th percentile (October to March) or ≥95th percentile (April to September), respectively, using municipality-specific temperature distributions. We applied conditional logistic regression combined with distributed lag nonlinear models to estimate lag-specific (lag 0-6 days) and cumulative associations with the odds of HF hospitalization.

During the cold season, cold spells and lower ambient temperature exposures were associated with a higher risk of HF hospitalization at lag 2 to 6 days (significant at lag 3-5 days). Cumulative associations over lag 2 to 6 days showed ORs of 1.085 (95% CI: 1.042-1.129) for cold spells and 1.008 (95% CI: 1.005-1.011) per 10-percentile-point decrease in temperature. During the warm season, higher temperature exposure was associated with higher odds at lag 1 to 3 days with attenuation at longer lags, with a cumulative association over lag 0 to 3 days (OR: 1.009; 95% CI: 1.006-1.013, per 10-percentile-point increase in temperature). Heat waves showed nonsignificant positive associations with HF hospitalizations that were more likely to capture acute exacerbations, whereas no associations were observed using the broader HF definition that included chronic conditions.

In this Swedish nationwide study, short-term exposure to cold spells, lower temperatures, and higher temperatures was associated with a higher risk of HF hospitalization, with delayed cold effects and more immediate heat effects.