An Assessment of Protocol Utilization and Efficacy when Transitioning from Infusion to Subcutaneous Insulin at an Academic Medical Center

Background: The transition from intravenous to subcutaneous insulin is a complicated process often associated with poor glycemic control. Our academic medical center has protocols for guiding providers through this transition. Our protocols have previously been shown to improve glycemic control. We sought to assess protocol utilization and whether it continues to produce successful insulin transitions. Methods: Data was collected through retrospective chart review of all transitions from intravenous to subcutaneous insulin completed at our medical center from April 2015 through June 2015. A successful transition was defined as one in which the insulin drip was not restarted, mean blood glucose was less than 180 mg/dL for the first 48 hours after transition, and there was no hypoglycemia (blood glucose less than 70 mg/dL). Results: In total, 267 transitions from intravenous to subcutaneous insulin occurred at our medical center over the study period, and 211 were included in the analysis. For the group of patients with diabetes and transitioned to long acting subcutaneous insulin, 25 out of 1(23.1%) were transitioned using either our paper or computer protocol, and 82 out of 1(75.9%) were not transitioned via protocol. Of the 25 patients transitioned via protocol, 15 (60%) were successfully transitioned from intravenous to subcutaneous insulin. Of the 82 patients transitioned without using the protocol, 39 (47.6%) were successful. Among patients without a history of diabetes, 62 out of 65 patients (95.4%) were successfully transitioned from intravenous to subcutaneous insulin. The majority of these patients maintained good glycemic control with correctional insulin only. Conclusions: Our protocol is effective in producing successful transitions from intravenous to subcutaneous insulin, however it is underutilized in our academic medical center. Disclosure L. Vincent: None. K. Box: None. K. Kulasa: None.