Comparison of various basal insulin dose adjustments for inpatients while unable to eat

Background The American Diabetes Association recommends basal insulin or basal plus correctional insulin regimen for non-critically ill patients with type 2 diabetes mellitus unable to eat. There is limited evidence available examining ideal basal insulin dose reductions in this patient population. Aim This study aimed to determine the percent reduction of maintenance basal insulin that would provide the least hypoglycemic incidence in patients with type 2 diabetes mellitus in the non-intensive care unit setting. Methods This retrospective cohort study evaluated adult patients with type 2 diabetes mellitus prescribed outpatient basal insulin with a minimum unable to eat status of two hours. Patients were divided into four groups; <25%, 25-50%, 51-75%, > 75% of basal insulin administered compared to home dose. The primary endpoint was the incidence of hypoglycemia while unable to eat. Secondary endpoints included incidence of hyperglycemia, severe hypoglycemia, median daily blood glucose and hospital length of stay. Results A total of 173 patients were included. The primary outcome of hypoglycemia (5.9% vs. 8.8% vs. 14.3% vs. 12.3%; P = 0.578) was similar in all groups. There were no differences in hyperglycemia (P = 0.0701), severe hypoglycemia (P = 0.578) and median daily blood glucose (P = 0.428). Patients receiving 25-50% of home basal insulin had the longest unable to eat duration (11.5 h; P = 0.026); however, this was not statistically significant when adjusted using the Bonferroni correction for multiple tests. Conclusions No differences were observed in hypoglycemic events for patients unable to eat receiving various basal insulin dose reductions.