Perioperative Real Time Glucose Assessment as a Predictive Tool for Complications After Pancreatic Resection in Non-diabetic Patients- a Prospective Single Center Pilot Study

Background Postoperative hyperglycaemia has been described as an early marker of complications after pancreatic resection. However, evidence is based on retrospective assessment of arbitrary serum glucose measurements. In contrast, continuous glucose monitoring (CGM) systems allow real-time monitoring of glucose fluctuations. Aims The aim of this study is to investigate continuous perioperative glucose changes after pancreatic resection and the impact on postoperative complications. Methods Twenty (n=20) consecutive patients undergoing pancreatic resection were prospectively enrolled. In addition, n=10 patients undergoing other major abdominal surgery served as control group. Dexcom G6 CGM system was used. Time in euglycemic range (TIR) and peak glucose levels were analyzed. Routine serum glucose measurements and daily C-reactive protein (CRP) levels were also assessed. Comprehensive Complication Index (CCI) was used to quantify postoperative complications. Results No adverse events related to CGM devices were observed. Glucose levels increased significantly from a median of 7mmol/l (IQR 6-8mmol/l) to 9 mmol/l (IQR 8-1mmol/l, P=0.026) after pancreatic resection. Correspondingly, the TIR decreased from 86.5% (IQR 85-96%) to 78.1% (IQR 34-89%, p=0.042). Perioperative glucose levels (p=0.623) and TIR (p=0.408) remained unchanged in the control group. Linear regression showed a significant correlation between peak glucose levels on day 1, measured by CGM (R=0.738, p=0.004, Figure 1A), and CRP levels on day 2 (R=0.528, p=0.034, Figure 1B) with CCI. In contrast, routine serum glucose levels did not predict complications. Conclusion In this pilot study, peak glucose levels on day 1 after pancreatic resection were associated with adverse events. CGM may be a valuable tool to identify patients at risk of complications.Figure 1.CGM Correlation of glucose measurement and complications. Linear Regression revealed a linear correlation of CGM peak levels on day 1 (A), p=0.004) and CRP on day 2 (B, p=0.034) with complications, displayed as CCI.