Previous Abstract | Next Abstract
Printable Version
October 17, 2012
8:00:00 AM - 9:30:00 AM
Room 101
Impact of an Intraoperative Diabetes Notification System on Perioperative Outcomes
Jesse M. Ehrenfeld, M.D.,M.P.H., Kelsey L. McCartney, M.B.A., Jason Denton, B.S., Brian Rothman, M.D., Robert Peterfreund, M.D.,Ph.D.
Vanderbilt University, Nashville, Tennessee, United States

Although there is evidence in the literature suggesting that hyperglycemia and diabetes mellitus are both independent predictors of elevated perioperative cardiovascular morbidity and mortality, no studies to date have allowed us to understand the most appropriate management strategy for diabetic patients undergoing surgery. However, most would agree that monitoring glucose intraoperatively is appropriate in diabetic patients - particularly those receiving insulin. We therefore undertook a multi-center prospective trial at two large academic medical centers designed to evaluate the impact on perioperative outcomes of a system in which providers were electronically reminded to monitor glucose during surgery.


After obtaining IRB approval, we launched a prospective diabetes quality improvement initiative in which we provided notifications to anesthesia providers to remind them to perform intraoperative glucose testing. A matched historical control group from the prior year was used as a comparison. The “Intraoperative Glucose Monitoring Alert” recommended to providers that they measure glucose every hour if insulin was administered during surgery, and every 120 minutes if a diabetic patient was undergoing surgery, but not receiving intraoperative insulin. A sample notification is shown in Figure 1. We then measured whether the implementation of the point-of-care suggestions improved (1) adherence to glucose monitoring and (2) a series of patient outcomes.


Preliminary data from one center is now available and shown in Tables 1 and 2. Data from the second center will be forthcoming. A total of 8,243 patients were enrolled (4447 control; 3,796 intervention). More intraoperative glucose tests were performed in the intervention group (p<0.0001). A statistically significant reduction in PACU length of stay (p=0.0004) and 14 day readmission (p=0.0467) were observed in the intervention group. A non-statistically significant trend towards reduction in surgical site infection rates was also observed.


We report the preliminary results from one center in a two center prospective trial of an intraoperative glucose monitoring notification. We observed a reduction in PACU length of stay which is likely due to fewer patients having glucose markedly out of control in the recovery room. We also observed a reduction in 14 day readmission rate, which requires further investigation to fully explain. Ultimately, we believe increased perioperative glucose monitoring will help define practice guidelines and allow us to further optimize perioperative care of diabetic patients.

Figure 1
Figure 2

Copyright © 2012 American Society of Anesthesiologists