Clinical Significance of Perioperative Hyperglycemia in Radical Gastrectomy Among Non- diabetic Patients: Evidence for Better Perioperative Glucose Management in Different Anastomotic Methods

Background: Uncontrolled perioperative hyperglycemia (HG) in patients undergoing surgery was determined as a risk factor for postoperative complications. Few studies have focused on the effects of HG in gastric cancer patients undergoing different anastomotic methods after radical gastrectomy. Methods: : We performed a double-institutional dataset study involving 811 patients who had undergone radical gastrectomy between 2014 and 2017. Patients with diabetes mellitus were excluded. Propensity-score-matching (PSM) analysis was performed to strictly balance the significant variables. The association between any elevated perioperative glucose value (HG≥7 mmol/L) and postoperative complications in patients treated with different anastomotic methods was assessed. Results: : Among the 742 non-diabetic patients with gastric cancer, 100 (13.48%) and 148 (19.95%) experienced preoperative and postoperative HG, respectively. Perioperative HG was not significantly associated with postoperative complications. On comparing the different anastomotic methods, differences in postoperative complication incidence were exclusively identified between postoperative hypoglycemia and HG (20.32% vs. 34.62%, P=0.025) among patients underwent Billroth-I anastomosis. Further logistic regressive analysis found HG to be independently associated with postoperative complications before (odds ratio [OR]:1.989, 95% confidence interval [CI]:1.031–3.837, P=0.040) and after (OR:3.341, 95%CI: 1.153–9.685, P=0.026) PSM. Finally, preoperative HG remained a significant predictor of postoperative HG (OR:3.718, 95%CI: 1.673–8.260, P=0.001). Conclusions: : Postoperative HG, rather than preoperative HG, was significantly associated with worse postoperative outcomes in non-diabetic patients who underwent Billroth-I anastomosis after radical gastrectomy. However, preoperative HG was associated with postoperative HG, suggesting that improved preoperative glycemic management may help reduce postoperative hyperglycemic events.