Cold Spells and the Onset of Acute Myocardial Infarction: A Nationwide Case-Crossover Study in 323 Chinese Cities

BACKGROUND: Few studies have explored the relationships between cold spells and acute myocardial infarction (AMI) using the information of symptom onset. OBJECTIVES: We assessed the impact of cold spells on AMI onset and the potential effect modifiers. METHODS: We conducted a time-stratified case-crossover study among 456,051 eligible patients with AMI from 2,054 hospitals in 323 Chinese cities between January 2015 and June 2021 during cold seasons (November to March). Nine definitions of cold spells were used by combining three relative temperature thresholds (i.e., lower than the 7.5th, 5th, and 2.5th percentiles) and three durations of at least 2–4 consecutive d. Conditional logistic regressions with distributed lag models were applied to evaluate the cumulated effects of cold spells on AMI onset over lags 0–6 d, after adjusting for daily mean temperature. RESULTS: The associations generally appeared on lag 1 d, peaked on lag 3 d, and became nonsignificant approximately on lag 5 d. Cold spells defined by more stringent thresholds of temperature were associated with higher risks of AMI onset. For cold spell days defined by a daily mean temperature of [Formula: see text] percentile and durations of [Formula: see text] , [Formula: see text] , and [Formula: see text] , the percentage changes in AMI risk were 4.24% [95% confidence interval (CI): 2.31%, 6.20%], 3.48% (95% CI: 1.62%, 5.38%), and 2.82% (95% CI: 0.98%, 4.70%), respectively. Significant AMI risks associated with cold spells were observed among cases from regions without centralized heating, whereas null or much weaker risks were found among those from regions with centralized heating. Patients [Formula: see text] years of age were more susceptible to cold spells. DISCUSSION: This national case-crossover study presents compelling evidence that cold spells could significantly increase the risk of AMI onset. https://doi.org/10.1289/EHP11841