Temperature-Defined Heat-Alert-Threshold Days and Acute Myocardial Infarction Admissions and Mortality: A Nationwide Hungarian Registry-Based Cohort Study.

backgroundAlthough extreme heat is associated with adverse cardiovascular outcomes, the relationship between heat-alert-threshold days, acute myocardial infarction (AMI) admissions, and post-AMI mortality remains uncertain. We aimed to evaluate the association between temperature-defined heat-alert-threshold days and (i) daily AMI admissions and (ii) cumulative all-cause mortality after hospitalized AMI in Hungary.

methodsWe conducted a nationwide registry-based study using data from the Hungarian Myocardial Infarction Registry (HMR). All AMI admissions between 1 January 2018 and 31 December 2019 were eligible, with mortality follow-up through 16 June 2021. Heat-alert-threshold days were defined using the temperature criterion applied in the Hungarian national heat-health action plan (daily mean temperature ≥25 °C). AMI admissions were analyzed using adjusted quasi-Poisson regression models and cumulative mortality using stratified Cox proportional hazards models. This operational temperature threshold was used as the exposure definition and does not represent linkage to administrative heat-alert declarations.

resultsThe cohort included 30,883 AMI events from 29,596 unique patients (mean age, 67.2 [SD 12.8] years; 60.3% male). Patients admitted on heat-alert-threshold days had baseline clinical characteristics similar to those admitted on non-alert days (all standardized mean differences <0.10). Heat-alert-threshold days were associated with fewer recorded AMI admissions during summer (adjusted incidence rate ratio [aIRR] 0.93, 95% CI 0.90-0.97). In contrast, no association was observed between heat-alert-threshold exposure and subsequent all-cause mortality after AMI hospitalization (adjusted hazard ratio [aHR] 0.96, 95% CI 0.87-1.05). These findings remained consistent across sensitivity analyses.

conclusionsIn this nationwide Hungarian registry-based study, temperature-defined heat-alert-threshold days were associated with fewer recorded AMI admissions during summer but not with differences in post-AMI mortality among hospitalized patients. The observed inverse association should not be interpreted as evidence of a protective effect of heat exposure. Alternative explanations include behavioral adaptation, delayed care-seeking, exposure misclassification, residual confounding, and the possibility of unmeasured out-of-hospital cardiovascular events, which were not captured by the registry. Because administrative heat-alert declarations were not linked to the analytic dataset, the findings should be interpreted as associations with temperature-defined heat-alert-threshold days rather than evaluations of the effectiveness of the Hungarian heat-alert system.