Effect of dexamethasone administration for postoperative nausea and vomiting prophylaxis on glucose levels in adults with diabetes undergoing elective surgery: a systematic review with meta-analysis

OBJECTIVE: The objective of this review was to evaluate the effect of intravenous dexamethasone given intraoperatively for postoperative nausea and vomiting prophylaxis on maximal blood glucose level within the initial 24 hours following elective surgery for patients with diabetes. INTRODUCTION: Postoperative nausea and vomiting is a prevalent adverse effect of anesthesia that leads to morbidity, increased health care costs, and unanticipated hospital admissions. Dexamethasone is an effective prophylactic agent that confers secondary analgesic and anti-inflammatory benefits. However, its use in patients with diabetes remains controversial due to the potential for increased postoperative blood glucose levels. INCLUSION CRITERIA: This review considered studies with participants 18 years of age or older with type 1 or 2 diabetes undergoing an elective surgical procedure. Eligible studies reported postoperative blood glucose levels in adults with diabetes after receiving a single 4-10 mg prophylactic dose of intravenous dexamethasone intraoperatively for postoperative nausea and vomiting. The primary outcome was maximum blood glucose level in the first 24 hours after surgery. All study designs were eligible for inclusion. Studies were excluded if they lacked a control group with diabetes or if they did not report maximum blood glucose values in both groups. METHODS: A search of MEDLINE, CINAHL Complete, Embase, Web of Science, TRIP database, and the Cochrane Database of Systematic Reviews was completed in October 2021. Gray literature resources were also searched. No date or language restrictions were applied. Methodological quality was assessed using JBI appraisal tools for randomized controlled trials, cohort studies, and case-control studies. A meta-analysis of maximal postoperative blood glucose level within 24 hours of surgery was performed, as well as subgroup analyses by dexamethasone dose, insulin treatment, and study design type. RESULTS: Eleven studies (4 randomized controlled trials, 6 cohort studies, and 1 case-control study) were included in this review, with 1 study excluded from meta-analysis and results reported narratively. The total sample size of studies included in meta-analysis was 2567. The administration of dexamethasone significantly increased maximal blood glucose levels in the 24 hours immediately following surgery compared with control groups with diabetes, as demonstrated by randomized controlled trials (mean difference [MD] 39.56 mg/dL; 95% CI 16.18 to 62.94; P < 0.001; I2 = 87%) and observational studies (MD 26.31 mg/dL; 95% CI 7.10 to 45.52; P = 0.007; I2 = 92%). This increase in blood glucose was significant for all doses of dexamethasone: 4 mg (MD 40.81 mg/dL; 95% CI 2.42 to 79.19; P = 0.001; I2 = 91%), 8 mg (randomized controlled trials only; MD 39.45 mg/dL; 95% CI 15.32 to 63.58; P = 0.001; I2 = 86%), and mixed 4-10 mg dose (MD 30.82 mg/dL; 95% CI 6.75 to 54.88; P < 0.012; I2 = 93%). Postoperative hyperglycemia persisted in studies using insulin treatment as well as those not using insulin protocols. The overall certainty of the findings ranged from very low for outcomes that included cohort studies to moderate when outcomes from randomized controlled trials were analyzed separately. However, the quantitative findings of the experimental and observational studies were clinically similar. Risk of bias presented minimal concerns in all included studies. CONCLUSIONS: Dexamethasone leads to transient postoperative hyperglycemia in patients with diabetes undergoing elective surgery when given as a single 4-10 mg intravenous dose for postoperative nausea and vomiting prophylaxis. The clinical relevance of hyperglycemia is debatable given its small magnitude and transient nature. Without more tightly controlled data, methodological consistency, and baseline blood glucose values, it is impossible to test causal links between hyperglycemia and pre-existing patient factors (eg, hemoglobin A1C levels) or postoperative complications. REVIEW REGISTRATION: PROSPERO CRD42020185607.