Low body temperature and mortality in critically ill patients with coronary heart disease: a retrospective analysis from MIMIC-IV database

Abstract

Abstract Background This study was aimed to investigate the correlation between low body temperature and outcomes in critically ill patients with coronary heart disease (CHD). Methods Participants from the Medical Information Mart for Intensive Care (MIMIC)-IV were divided into three groups (≤ 36.5 ℃, 36.6–37.4 ℃, ≥ 37.5 ℃) in accordance with body temperature measured orally in ICU. In-hospital, 28-day and 90-day mortality were the major outcomes. Multivariable Cox regression, decision curve analysis (DCA), restricted cubic splines (RCS), Kaplan–Meier curves (with or without propensity score matching), and subgroup analyses were used to investigate the association between body temperature and outcomes. Results A total of 8577 patients (65% men) were included. The in-hospital, 28-day, 90-day, and 1-year overall mortality rate were 10.9%, 16.7%, 21.5%, and 30.4%, respectively. Multivariable Cox proportional hazards regression analyses indicated that patients with hypothermia compared to the patients with normothermia were at higher risk of in-hospital [adjusted hazard ratios (HR) 1.23, 95% confidence interval (CI) 1.01–1.49], 28-day (1.38, 1.19–1.61), and 90-day (1.36, 1.19–1.56) overall mortality. For every 1 ℃ decrease in body temperature, adjusted survival rates were likely to eliminate 14.6% during the 1-year follow-up. The DCA suggested the applicability of the model 3 in clinical practice and the RCS revealed a consistent higher mortality in hypothermia group. Conclusions Low body temperature was associated with increased mortality in critically ill patients with coronary heart disease.