Real-time CGM in Perioperative Glycemic Management

Diabetes and stress hyperglycemia are common among perioperative surgical patients and are associated with higher risks of postoperative infection, complications and prolonged hospital stay. Real-time continuous glucose monitoring (rtCGM) can capture continuous interstitial glucose data including asymptomatic hypo- and hyperglycemia, which may optimize perioperative glycemic control. This single-center, prospective, open-label, randomized parallel-controlled trial aims to compare perioperative glycemic management guided by real-time continuous glucose monitoring versus conventional point-of-care capillary glucose testing (POCT) among adults with abnormal glucose receiving non-emergency elective or semi-elective general or orthopedic surgery. One-hundred-twenty eligible subjects will be randomized 1:1 into experimental rtCGM group or POCT control group. Experimental group uses visible rtCGM (Sinocare iCan h6) as main monitoring tool, supplemented by POCT when clinically indicated. Control group wears blinded rtCGM sensor (data and alarms disabled for clinical use) and is managed exclusively by routine POCT finger-stick glucose. Both groups target perioperative glucose 5.6-10.0 mmol/L, and clinical treatment including anti-hyperglycemic drugs and surgical procedures follows standard hospital practice without protocol-mandated modifications. Primary endpoint is hierarchical composite of postoperative length of hospital stay (first priority) and CGM Time-In-Range (TIR, second priority), evaluated with win-ratio analysis. Secondary endpoints cover multiple glycemic variability indices, rtCGM accuracy metrics (MARD, Clarke error grid, Bland-Altman), postoperative complication rates, hypoglycemia and severe metabolic event rates, 30-day unplanned readmission / emergency visit rates, total hospitalization costs, patient and healthcare-worker satisfaction. Safety assessment collects all adverse events, serious adverse events and device-related adverse reactions up to 30 days post-surgery. Study includes screening/baseline period, perioperative in-hospital intervention period, discharge visit, and post-operative 30-day ±3-day telephone follow-up. Statistical analysis will adopt full analysis set (FAS) as primary population; per-protocol set (PPS) and safety set (SS) will also be analyzed.