Transitioning from intravenous to subcutaneous insulin in the medical intensive care unit.

Background: There is a paucity of studies on transitions from IV insulin infusion (IVII) to subcutaneous (SC) insulin in the medical ICU (MICU).

Methods: We conducted a retrospective study of patients admitted to the Cleveland Clinic MICU from June 2013 to January 2014 who received IVII. We compared blood glucose (BG) control between 3 cohorts based on timing of basal insulin dose: (1) NB (no basal), (2) IB (incorrect basal), (3) CB (correct basal) at 5 time points post-IVII discontinuation (1, 4, 8, 12, and 24h). Insulin doses used for transitioning were compared with 80% of estimated 24h IVII total. Analysis was done using chi-square, ANOVA and t-tests.

Results: There were 269 patients (NB 166, IB 45, CB 58), 55% male with a mean age 58±16years. 103 patients (38%) had a transition attempted (IB 21%, CB 17%). The NB cohort had better BG than the IB cohort at all time points (p<0.001) but also lower HbA1c, prior DM diagnosis and home insulin use (p<0.001). IB and CB did not have significantly different BG with mean BG>180mg/dL at 4/5 time intervals. However, the dose of basal insulin used was less than 80% of estimated 24h IVII total (IB 21.4 vs 49.6U, CB 25vs 57.1U). Despite this, 15% of patients in the IB cohort and 24% of patients in the CB had hypoglycemic events.

Conclusion: The low rates of IV to SC insulin transitions raises the question of challenges to transitions.