Glycemic Status and Glycemic Variability Related to Mortality and Morbidity in Critically III Diabetic and Non-Diabetic Patients: A Prospective Observational Study

Background: Glycemic dysregulation is frequent in critically ill patients and may influence prognosis regardless of pre-existing diabetes status. This study aimed to evaluate the impact of glycemic status and glycemic variability on mortality and morbidity in diabetic and non-diabetic patients admitted to the intensive care unit (ICU). Methods: This prospective observational study included 244 critically ill patients. Demographic, clinical, biological, and glycemic data were collected during ICU hospitalization. Glycemic parameters included admission glucose, mean glucose, glucose standard deviation, coefficient of glycemic variability, hyperglycemia > 140 mg/dL, hyperglycemia > 180 mg/dL, hypoglycemia < 70 mg/dL, and intravenous insulin requirement. The primary outcome was in-hospital mortality. Results: In-hospital mortality was 58.6%. Non-survivors had significantly higher admission glucose, mean glucose, glycemic standard deviation, and coefficient of glycemic variability compared with survivors. Hyperglycemia > 140 mg/dL, hyperglycemia > 180 mg/dL, poor glycemic control, and intravenous insulin therapy were significantly associated with mortality. Patients with high glycemic variability, defined as coefficient of variation (CV) > 36%, had higher mortality than those with lower variability. In multivariable analysis, mean glucose remained the most robust glycemic predictor associated with mortality, while glycemic variability retained prognostic relevance in alternative models. Conclusions: Glycemic dysregulation was common and clinically relevant in critically ill diabetic and non-diabetic patients. Dynamic glucose monitoring, including mean glucose and glycemic variability, may improve risk stratification and support structured ICU glycemic management.