Prolonged Glucose Lowering Therapy In Post-Cardiac Surgery-Induced Stress Hyperglycemia

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SUN 807-838-Diabetes - Diagnosis, Complications & Outcomes
Clinical
Sunday, June 16, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SUN-822
I-Tsyr Shaw*1, Ryan Deiter2 and Kathleen M Dungan3
1Ohio State University Medical Center, Columbus, OH, 2Ohio State University Medical Center, 3The Ohio State University, Columbus, OH
Introduction: Post cardiac surgery, tight glucose control is standard of care. A subset of non-diabetic patients with stress hyperglycemia (SH) still requires treatment at discharge. Objective: To evaluate predictors and outcomes of prolonged glucose lowering treatment in non-diabetic patients 1 year after the implementation of an electronic medical record (EMR). Methods: Patients with SH (A1c <6.5%, not on medication) who were discharged on any antidiabetic agent following cardiac surgery, were included. Outcomes of interest were antidiabetic medication at discharge and at follow up and clarity of discharge instruction. Results: A total of 598 patients without diabetes received insulin during the study period. AM glucose was >180 mg/dL at discharge in 51 patients who were not discharged on glucose lowering therapy, 47 of whom did not have a diabetes consult. Another 37 patients were discharged on glucose lowering therapy and served as the study sample of interest. Cardiac bypass (CB) was performed in 49% and 40% had valve surgery (VS). Discharge regimen included oral agents only (27%), insulin only (41%), or both (32%). At 6 weeks, 56% of patients were still prescribed at least one antidiabetic agent. VS was associated with less insulin the day prior to discharge (p=0.01) and less oral agent use at follow up (p=0.02) compared to non-VS. CABG had no effect on treatment compared to non-CABG. A diabetes consult was obtained in 68%. Those with a diabetes consult were more likely to be discharged on oral agents than those without (80% vs.17%, p=0.0007). Discharge instructions were unclear for 20 patients. Issues included inconsistencies between the instructions and prescriptions, incomplete information (lack of or unclear frequency of insulin use, no detail regarding sliding scale). There was a trend for decreased clarity of discharge instructions in patients without a diabetes consult (44% vs 75%, p=0.09). Discharge instructions were more likely to lack clarity in those who were discharged on insulin (p=0.002). Conclusion: We found that treatment requirements for SH may be prolonged in 16% of patients undergoing cardiac surgery, but differ by procedure. Although the EMR is meant to enhance discharge procedures through features such as medication reconciliation, the interface for insulin dosing is sub-optimal. Discharges may be improved with the assistance of a diabetes consult in this group of patients new to antidiabetic agents.

Nothing to Disclose: ITS, RD, KMD

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm