Prognostic value of mean glycemia and glycemic variability in medical, surgical, and cardiovascular intensive care units at a Lebanese tertiary care center.

Background and objectives: Stress-induced hyperglycemia is common in intensive care units (ICUs) and has been linked to adverse outcomes. Although mean glycemia (MG) has been extensively studied, the benefits of strict glycemic control remain controversial, and the impact of glycemic variability (GV) is less clearly defined. No consensus currently exists regarding GV thresholds, and limited evidence is available across different ICU settings, with data from the Middle East region particularly lacking. This study aimed to assess the relationship between MG and GV with key clinical outcomes, including hospital and ICU length of stay (LOS), renal function, and in-hospital mortality, among patients admitted to three ICUs at a Lebanese tertiary care center. Methods: We retrospectively reviewed the medical records of patients admitted during July and August 2024 to the surgical, medical, and cardiovascular ICUs at the Hôtel-Dieu de France Hospital. Baseline characteristics, MG, GV, total hospital/ICU LOS, in-hospital mortality, and glomerular filtration rates (GFR) were analyzed. Results: GV was significantly associated with prolonged total and ICU LOS, reduced GFR, and increased in-hospital mortality. Patients with GV >30% had a markedly higher risk of death. In contrast, no significant association was found between MG and said outcomes. MG differed across ICU subunits, reaching its highest levels in the cardiovascular ICU, while GV did not vary significantly between units. Conclusion: GV, rather than MG, emerged as a key predictor of adverse outcomes in ICU patients, being associated with longer hospital and ICU LOS, renal impairment, and increased mortality. These findings highlight GV as an important therapeutic target in the management of critically ill patients.