Intraoperative continuous glucose monitor accuracy compared with point of care glucose testing and assessment of preoperative medication hold parameters on glycemic control during orthopedic procedures: a observational study

Perioperative glycemic control includes a multitude of variables which are patient specific and may not distill to a one size fits all patient approach. Continuous glucose monitoring (CGM) systems present the opportunity to improve care by constantly monitoring patient glucose control perioperatively. The primary objectives of this observational study were to evaluate the performance of a CGM during the perioperative period while also assessing the glycemic control results from following preoperative medication hold times recommended by the American Diabetes Association Standards of Care (ADA SOC). CGM systems were applied prior to admission in this single center study using recommendations from the ADA SOC for medication hold times. Point of care (POC) glucose measurements were taken every 30 min intraoperatively for comparison to CGM glucose values. Phase I focused solely on CGM performance, phase II added a calibration step preoperatively to determine if performance could be further improved. Accuracy was assessed with mean absolute relative difference (MARD) between POC and CGM glucose values and error grid analysis (EGA). Glycemic control was evaluated via preoperative POC and CGM glucose values and intraoperative time below range (TBR), time in range (TIR), and time above range (TAR). Sixty patients completed the study with combined preoperative glucose values of mean (± SD) 131.1 (± 26.1) mg/dL for POC glucose and 150.7 (± 35.2) mg/dL for CGM. Intraoperative MARD in phase I was 18.1% (± 17.2) with 64.7% of CGM-POC pairs in zone A, 33.6% in zone B, and 1.7% in zone C on EGA. Phase II intraoperative MARD was 9.8% (± 6.9) with 93.6% of CGM-POC pairs in zone A, 6.4% in zone B. Combined phase I/II intraoperative TBR was 0.7%, TIR 91.5%, TAR 7.8%. One intraoperative hypoglycemic event (69 mg/dL) occurred in phase II. Postoperative (12 h) MARD for phase I was 11.4% (± 9.1), phase II 7.1 (± 5.4). CGM performance was within acceptable limits of agreement with POC glucose testing during the intraoperative period which improved following the addition of a preoperative calibration step. Glycemic control values obtained both pre and intraoperatively were within target range providing support for use of current recommended hold times.