Do diabetes and poor control of acute stress-related hyperglycemia increase the risk of ICU-acquired infections? A retrospective assessment in patients with septic shock

Annals of Intensive Care volume 15, Article number: 168 (2025) Published: 22 October 2025

Background

Patients with septic shock who survive the early resuscitation phase are prone to ICU-acquired infections. Although hyperglycemia harbors potent immunomodulatory properties, the impact of preexisting diabetes and the control of acute stress-induced hyperglycemia on the risk of further infections remains unclear.

Materials and methods

We conducted a retrospective (2008–2023) single-center study in patients with septic shock who remained alive in the ICU after 72 h. Glycemic control was assessed during the first 72 h. Mild and severe hyperglycemia were defined by blood glucose levels > 8 mmol/L and > 10 mmol/L, respectively. Poor glycemic control was defined when blood glucose levels were above 8 mmol/L for more than 20% of time. The primary outcome was ICU-acquired infections.

Results

The study involved 901 patients, with preexisting diabetes present in 22% of them. Most patients (71%) experienced hyperglycemic episodes > 8 mmol/L, prompting fast-acting insulin treatment. ICU-acquired infections developed in 243 patients (26.9%), with median time from ICU admission to diagnosis of 9 days, interquartile range [6–13]. There was no association between preexisting diabetes and ICU-acquired infections. Patients with further ICU-acquired infections displayed poorer control of stress-induced hyperglycemia, with longer exposure to hyperglycemia (78% with mild or severe hyperglycemia for more than 20% of time compared to 68% of patients without subsequent infections (p = 0.005)). Poor glycemic control was independently associated with the development of ICU-acquired infections.

Conclusion

72-hour poor glycemic control, but not preexisting diabetes, was independently associated with an increased risk of ICU-acquired infections in septic shock patients and may therefore contribute to the post-aggressive immunosuppressive response. This argues for effective glycemic management to improve outcomes in this setting.