The association between glucose variability (GV) and adverse perioperative outcomes in type 2 diabetes mellitus (T2DM) patients undergoing orthopedic surgery was investigated. A retrospective cohort study was performed by analyzing data on T2DM patients receiving continuous blood glucose (BG) monitoring and continuous subcutaneous insulin infusion treatment due to poorly controlled preoperative BG prior to orthopedic surgery. GV was assessed with coefficient of variation (CV). Postoperative and perioperative CV, hypoglycemia cases, and other perioperative outcomes (diabetes preparation time [DPT], length of stay [LOS], and perioperative and infective complication cases) were analyzed. Results showed that a total of 168 patients were grouped into preoperative CV tertiles: 1st (n=56): 0-0.2921, 2nd (n=58): 0.2922-0.3779, and 3rd (n=54): 0.3780-0.5750. Fasting blood glucose (FBG), perioperative CV, rate of hypoglycemia cases (OR: 5.53, 95%CI: 2.43-12.59) (all P<0.001) and DPT (P=0.024) were higher in the 3rd than in the 1st tertile. After adjustments of covariates, regression analysis indicated that the 3rd tertile was associated with increased perioperative CV (adjusted coefficient=0.515, P<0.001), DPT (adjusted coefficient =0.169, P=0.073), rate of hypoglycemia cases (OR: 6.72, 95%CI: 2.69-16.82, P<0.001) and perioperative complication cases (OR: 2.50, 95%CI: 0.90-7.01, P=0.080). In conclusion, preoperative GV is associated with increased perioperative GV and adverse perioperative outcomes including longer DPT and higher rates of hypoglycemia and perioperative complications.
Keywords: glycemic variability; orthopedics; perioperative period; type 2 diabetes mellitus.