Impact of Perioperative Glycemic Variability on Surgical Outcomes in Patients With and Without Diabetes: A Multicenter Prospective Observational Study

Background: Perioperative hyperglycemia is associated with postoperative morbidity, but a single glucose measurement may not capture the physiological instability represented by repeated fluctuations.Glycemic variability may therefore provide prognostic information beyond diabetes status and mean glucose concentration.This multicenter prospective observational study evaluated the association between perioperative glycemic variability and 30-day surgical outcomes in patients with and without diabetes.Methods: We analyzed 360 adults undergoing major non-cardiac surgery at Gujranwala Teaching Hospital and Mayo Hospital.Diabetes was defined as a documented diagnosis or glycated hemoglobin ≥6.5%.A minimum of six glucose values obtained from the preoperative period through 72 hours after surgery were required to calculate glycemic variability as the coefficient of variation (CV = standard deviation/mean glucose × 100).High variability was defined as the upper cohort tertile (CV ≥17.5%).The primary outcome was a composite of any 30-day postoperative adverse event: surgical-site infection, wound complication, pulmonary complication, acute kidney injury, sepsis, cardiac event, venous thromboembolism, ileus, anastomotic leak, reoperation, or death.Secondary outcomes were Clavien-Dindo grade III or higher complications, intensive care unit admission, readmission, and length of stay.Multivariable logistic regression adjusted for age, American Society of Anesthesiologists (ASA) class, diabetes status, surgical urgency, operative duration, and intraoperative transfusion.Results: Of 360 patients, 109 (30.3%) had diabetes and 120 (33.3%) had high glycemic variability.At least one adverse outcome occurred in 138 patients (38.3%).The composite outcome occurred in 62.5% of patients with high variability compared with 26.3% of those with lower variability (p < 0.001).Among patients without diabetes, event rates were 50.9% with high variability and 23.7% with lower variability; corresponding rates among patients with diabetes were 73.0% and 37.0%, respectively.High glycemic variability remained independently associated with the composite outcome (adjusted odds ratio (aOR): 2.44; 95% CI: 1.42-4.19;p = 0.001).This association persisted in a sensitivity analysis additionally adjusting for mean perioperative glucose ≥180 mg/dL and intraoperative hypotension (aOR: 2.33; p = 0.003).Diabetes (aOR: 2.49; 95% CI: 1.44-4.31),emergency surgery (aOR: 2.68; 95% CI: 1.56-4.58),ASA class III or higher (aOR: 2.00; 95% CI: 1.07-3.75),operative duration >180 minutes (aOR: 1.66; 95% CI: 1.01-2.72),and intraoperative transfusion (aOR: 1.84; 95% CI: 1.05-3.24)were also independently associated with the composite outcome.The diabetes-by-variability interaction was not statistically significant (p = 0.703).Conclusions: Perioperative glycemic variability was independently associated with postoperative morbidity in patients with and without diabetes, with the greatest absolute risk observed among patients with both diabetes and high glycemic variability.However, reverse causation cannot be excluded, as complications arising during the 72-hour glucose-monitoring window may themselves increase glycemic variability.Prospective studies incorporating time-resolved analyses are needed to clarify the temporal relationship and determine whether improving glucose stability reduces postoperative complications.