Aggressive glycemic control using insulin infusion has been associated with decreased mortality in subsets of surgical and medical ICU patients. There has been little published on safety or efficacy for insulin infusion use on general medical/surgical floors, where most patients are located in the hospital.
Baseline data came from 249 patients on a general medical/surgical floor who had an insulin infusion from October 2004 to October 2005. The baseline patients were managed using a protocol with a goal glucose range of 120–180 mg/dL. A new insulin infusion protocol was developed with a goal glucose of 100–150 mg/dL. The intervention data are from 260 patients from November 2006 to October 2007. Five glucometrics were followed: (1) patient‐stay, the mean glucose (MG) for the patient during the entire hospital admission; (2) infusion episode, the MG for each insulin infusion episode; (3) patient‐day, the MG for the patient in one 24‐hour period; (4) interval, the MG for each 6‐hour interval; and (5) all capillary blood glucose (CBG) values. A validated case severity index score, length of stay (LOS), and surgical site infections were recorded. Surgical site infections were obtained from the National Surgical Quality Improvement Program (NSQIP) data. Comparisons between baseline and intervention groups were made using independent‐sample f tests, Wilcoxon rank‐sum tests, chi‐square tests, and Fisher's exact test where appropriate. A model for generalized estimating equations for correlated data was used to assess the association between surgical site infection and an intervention effect, while controlling for LOS and case severity index score and accounting for the correlation of patients with multiple hospital stays.
The intervention insulin infusion protocol was statistically significantly associated with an improvement in ail 5 glucometrics, and did not increase hypoglycemia (Table 1). In addition, at the patient‐stay level, the case severity index score did not differ between the groups [median 2, interquartile range (IQR) 1–3; P = 0.39], nor did LOS [median 7 days, IQR 4–12; P = 0.91]. However, NSQIP surgical site infections (n = 228 patient‐stays) decreased from 7 of 84 (8.3%) to 2 of 144 (1.4%), P = 0.014. After adjusting for LOS and case severity index score, patients in the intervention group were less likely to have experienced a surgical site infection than were patients in the baseline group (OR: 0.18,95% Cl: 0.04–0.84, P = 0.03).
The use of insulin infusion on a general medical/ surgical floor improved glucometrics without an increase in hypoglycemia, and decreased the odds of surgical site infections by 82%. The current glycemic control guideline for patients on general medical/surgical floors recommends a target of 110–180 mg/dL. Our insulin infusion protocol produced results closer to 180 mg/dL but was associated with an improvement in an important clinicaf outcome.
T. W. Barrett, none; C. Locke, none; L Snodgrass, none; C. Tillotson, none; M. Mori, none; J. McEuen, none; B. Lesselroth, none; R. Felder, none.
To cite this abstract:Barrett T, Locke C, Snodgrass L, Tillotson C, Mori M, McEuen J, Lesselroth B, Felder R. Use of an Insulin Infusion to Improve Glycemic Control and Decrease Surgical Site Infections on a General Medical/Surgical Floor. Abstract published at Hospital Medicine 2009, May 14-17, Chicago, Ill. Abstract 11. Journal of Hospital Medicine. 2009; 4 (suppl 1). https://www.shmabstracts.com/abstract/use-of-an-insulin-infusion-to-improve-glycemic-control-and-decrease-surgical-site-infections-on-a-general-medicalsurgical-floor/. Accessed May 26, 2019.