Title: Evaluation of Safety Outcomes in Patients with Diabetes Receiving Perioperative Dexamethasone
Investigators: Christian Garner, Devon Johnson, Spencer Livengood
Practice Site: ECU Health Medical Center
Contact: [email protected]Background: Postoperative nausea and vomiting (PONV) occurs in ~30% of surgical patients and is more common in high-risk procedures. Intravenous dexamethasone is often utilized for PONV prophylaxis but may increase the risk of hyperglycemia in patients with diabetes. At ECU Health Medical Center (ECUHMC), cases of diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state (HHS) have been observed following perioperative dexamethasone use. However, the incidence of these complications remains poorly described in current literature. This research project aimed to evaluate the incidence of DKA and HHS in patients with diabetes receiving perioperative dexamethasone and identify associated risk factors for postoperative insulin drip initiation.
Methods: This single-center, retrospective observational review evaluated adult patients with diabetes who underwent surgery and received perioperative dexamethasone at ECUHMC from July 2024 to July 2025. Patients were excluded for conditions likely to confound glycemic outcomes, including chronic steroid use, perioperative insulin infusion, or administration of additional steroids. The primary outcome was the incidence of postoperative DKA or HHS as defined by the 2026 American Diabetes Association. Secondary outcomes included hospital length of stay and mean 72-hour postoperative blood glucose levels. Descriptive statistics were used to summarize primary and secondary outcomes. A multivariate analysis was used to identify an association between independent variables and insulin infusion initiation.
Results: Of 1,450 patients screened, 381 met inclusion criteria. Two patients (0.5%) developed confirmed DKA, and no cases of HHS were identified. Median postoperative length of stay was 92.2 hours and mean 72-hour postoperative blood glucose was 172 mg/dL, with 65.9% having a 24-hour postoperative blood glucose >200 mg/dL. Seventeen patients (4.5%) required initiation of an insulin infusion within 72 hours of dexamethasone administration but were missing necessary information to confirm DKA or HHS diagnosis. Results from the multivariate analysis confirmed higher admission blood glucose was independently associated with insulin infusion initiation (p=0.0435).
Conclusion: Although perioperative dexamethasone use was associated with a low incidence of confirmed DKA and HHS, a notable proportion of patients required insulin infusion, indicating a higher burden of postoperative hyperglycemia and potential under-recognition of hyperglycemic crises. Elevated preoperative blood glucose levels were associated with increased risk of insulin infusion initiation, supporting closer perioperative glucose monitoring in patients with diabetes.
