Ambient cold exposure and myocardial infarction risk in Tianjin, China: a distributed lag non-linear analysis

Abstract

Abstract Background Ambient temperature is an important environmental determinant of cardiovascular health. Although cold exposure has been associated with acute myocardial infarction (MI), evidence on delayed effects and subtype-specific differences remains limited in northern China. We examined the short-term association between temperature and MI incidence in Tianjin. Methods We conducted a time-series study integrating daily MI counts from the Tianjin Chest Pain Center with meteorological data from 2020 to 2023. Distributed lag non-linear models with quasi-Poisson regression were used to estimate non-linear and lagged effects of temperature on MI, adjusting for long-term trends, seasonality, relative humidity, day of the week, and public holidays. Extreme cold was defined as the 5th percentile of daily mean temperature (− 3.4 °C). The minimum morbidity temperature (MMT) was used as the reference temperature. Results During the 1,461-day study period, 34,940 MI events were recorded, including 21,814 (62.4%) STEMI cases. Compared with the median temperature (15.4 °C), extreme cold was associated with a cumulative 7-day RR of 1.53 (95% CI: 1.25–1.87), with peak effects observed at lag 2 days. The cumulative risk appeared higher for STEMI (RR = 1.62, 95% CI: 1.31–2.01) than NSTEMI (RR = 1.38, 95% CI: 1.05–1.82), but this difference was not statistically significant ( P heterogeneity = 0.367). Referenced to the MMT, the temperature–MI association showed a non-linear, approximately U-shaped pattern, with extreme cold associated with increased risk (RR = 1.66, 95% CI: 1.35–2.04). The bootstrap-derived MMT was 24.1 °C (95% CI: 22.6–26.2 °C), consistent with the primary estimate. Conclusions Ambient cold exposure is associated with a short-term increase in MI risk in Tianjin, peaking approximately 2 days after exposure. Although STEMI showed higher descriptive risk than NSTEMI, no statistically significant difference was observed. Heat-related associations were not statistically significant and should be interpreted cautiously due to limited exposure frequency and statistical power. These findings provide evidence for cold-related cardiovascular risk in temperate regions. Clinical trial registration Not applicable.