Evaluation of Basal Insulin Dose Reductions in Hospitalized Patients With Diabetes While Unable to Eat

Background: The American Diabetes Association guidelines recommend a basal plus correction or basal insulin regimen for patients with type 2 diabetes mellitus (T2DM) receiving nothing by mouth (NPO; nil per os) in the non-intensive care unit setting. In the perioperative setting, 60% to 80% of long-acting insulin or half-dose morning insulin NPH is recommended. Objective: The goal of this study was to determine the impact of basal insulin dose reduction for hospitalized patients with insulin-dependent T2DM while NPO. Methods: This retrospective, single-center study evaluated patients admitted to the non-intensive care unit setting. Administration of >50% of home basal insulin was compared with administration of ≤50% of home basal insulin. The primary outcome was the difference in hypoglycemic events (blood glucose [BG] < 70 mg/dL). Secondary outcomes included comparing severe hypoglycemic events (BG < 40 mg/dL), hyperglycemic events (BG > 180 mg/dL), and hospital length of stay (LOS). Results: Two hundred fifty-eight patient encounters were included, of which 85 and 173 patients received ≤50% and >50% of their home basal insulin dose, respectively. There were no significant differences in hypoglycemia (21.2% vs 21.4%; P = .97), severe hypoglycemia (1.2% vs 2.9%; P = .67), and hospital LOS (3 [IQR 2.13-6.74] days vs 4.66 [IQR 2.94-8.17] days; P = .74). Hyperglycemia occurred at a higher rate in patients receiving ≤50% of their home basal insulin dose (97.6% vs 89%; P = .02). Conclusions: No differences were observed in hypoglycemic events between those patients receiving ≤50% and >50% of their home basal insulin.