Associations of heat and cold with hospitalizations and post-discharge deaths due to acute myocardial infarction: what is the role of pre-existing diabetes?

BACKGROUND The existing evidence suggests that pre-existing diabetes may modify the association between heat and hospitalizations for acute myocardial infarction (AMI). METHODS This study included patients who were hospitalized for AMI from 1 January 2005 to 31 December 2013 in Brisbane, Australia, and also included those who died within 2 months after discharge. A time-stratified case-crossover design with conditional logistic regression was used to quantify the associations of heat and cold with hospitalizations and post-discharge deaths due to AMI in patients with and without pre-existing diabetes. Stratified analyses were conducted to explore whether age, sex and suburb-level green space and suburb-level socio-economic status modified the temperature-AMI relationship. Heat and cold were defined as the temperature above/below which the odds of hospitalizations/deaths due to AMI started to increase significantly. RESULTS There were 14 991 hospitalizations for AMI and 1811 died from AMI within 2 months after discharge during the study period. Significant association between heat and hospitalizations for AMI was observed only in those with pre-existing diabetes (odds ratio: 1.19, 95% confidence interval: 1.00-1.41) [heat (26.3°C) vs minimum morbidity temperature (22.2°C)]. Cold was associated with increased odds of hospitalizations for AMI in both diabetes and non-diabetes groups. Significant association between cold and post-discharge deaths from AMI was observed in both diabetes and non-diabetes groups. CONCLUSIONS Individuals with diabetes are more susceptible to hospitalizations due to AMI caused by heat and cold.