Abstract
BACKGROUND: Sepsis patients hospitalized in the intensive care unit (ICU) often have comorbid diabetes mellitus (DM). However, the clinical impact of DM on the clinical outcomes of critically ill sepsis patients has yet to be determined. Therefore, the current study aimed to analyze the association of comorbid DM with the prognosis of sepsis patients in the ICU. METHODS: Data of patients with sepsis and comorbid DM were obtained from a large-scale intensive care database. The primary outcome was 28-day mortality after ICU admission. Associations of comorbid DM with the primary outcome were assessed using a multivariable Cox regression model. Different adjusted models, such as the propensity score method, were used to determine the prognosis of the patients. RESULTS: Overall, 12,321 sepsis patients were enrolled, including 3,509 (28.48%) with comorbid DM. After adjusting and matching, we found that comorbid DM was not an independent risk factor for 28-day mortality in critically ill sepsis patients and was even associated with lower mortality. Propensity score matching showed a dramatically lower 28-day mortality for sepsis patients with comorbid DM in comparison to patients without comorbid DM [hazard ratio (HR): 0.86, 95% confidence interval (CI): 0.77-0.97, P=0.0167]. The relationship of comorbid DM with 28-day mortality was broadly consistent for all subgroup variables. In the stratified analysis, a significant interaction was observed only for glucose concentration (P<0.0001). Patients with comorbid DM and a glucose level of 140-200 mg/dL (7.8-11.1 mmol/L) or ≥200 mg/dL (11.1 mmol/L) had a significantly lower 28-day mortality rate (HR 0.83, 95% CI: 0.71-0.98, P=0.0250 and HR: 0.49, 95% CI: 0.38-0.64, P<0.0001, respectively). CONCLUSIONS: Critically ill patients with sepsis and comorbid DM were not found to have increased 28-day mortality compared to those without comorbid DM, and may even have a lower risk of mortality. Notably, this association remained in the setting of hyperglycemia.