The Efficacy of an Artificial Pancreas Device for Achieving Tight Perioperative Glycemic Control in Living Donor Liver Transplantation

Abstract

Intraoperative hyperglycemia during liver transplantation can induce infectious bacterial complications after surgery. The aim of this study was to evaluate the efficacy of the artificial endocrine pancreas in achieving perioperative blood glucose control and preventing infection in patients undergoing living donor liver transplantation. An artificial endocrine pancreas device was used for intraoperative glycemic control in 14 patients who underwent transplantation. In this study, we aimed to control the perioperative glucose levels (target range, 100-110 mg/dl) consecutively for 24 hours from the induction of anesthesia. We compared the 14 patients with an artificial endocrine pancreas device to 14 patients who underwent glycemic control using the sliding scale method with respect to perioperative blood glucose level and postoperative infection. The average blood glucose level in the artificial pancreas group was significantly lower than that in the sliding scale group (118 vs. 141 mg/dl, p< 0.05). In the sliding scale group, 11 of the 14 patients had postoperative infections within one month after LDLT. Extreme hypoglycemia (<40 mg/dl) did not occur in either group. The postoperative infections included bacteremia (n=3), pneumonia (n=7), urinary tract infection (n=1), bacterial peritonitis (n=4), and cholangitis (n=1). In contrast, the postoperative infections of the artificial pancreas group included bacteremia (n=1), pneumonia (n=1), and bacterial peritonitis (n=2). The postoperative infection rate of the artificial pancreas was significantly lower than that of the sliding scale group within one month after LDLT (35.7% vs. 78.6%, P<0.05). In conclusion, the artificial endocrine pancreas enabled the perioperative glucose level to be stably controlled without hyperglycemia or hypoglycemia. Blood glucose control is considered to be one of the factors that can reduce the incidence of postoperative infection after LDLT.