Restricting orders for insulin glargine may lower hypoglycemia rates, study finds

Restricting inpatient ordering of high-dose insulin glargine significantly lowered hypoglycemia rates at 1 hospital, according to a recent retrospective cohort study.

As part of an effort to reduce insulin-related hypoglycemia at Boston Medical Center, non-endocrinologist clinicians who electronically order glargine at a dose of 0.5 units/kg or greater incur a warning screen, which prevents order completion and advises either lowering the dose or contacting the diabetes consult service for approval. Results of the intervention were published online on Aug. 19 by Endocrine Practice.

Study participants were nonpregnant adults outside the ICU who received glargine during the first 24 hours of admission: 692 patients in the preintervention group and 651 patients in the postintervention group. Researchers assessed hypoglycemia (blood glucose level ≤70 mg/dL [≤3.89 mmol/L]), hyperglycemia (blood glucose level ≥200 mg/dL [≥11.11 mmol/L]), and glycemic control (mean blood glucose levels) during the first 48 hours of admission.

In the preintervention group, high-dose glargine (≥0.5 units/kg) was administered on admission in 45 patient admissions (5.2%), whereas only 3 patient admissions (0.3%) in the postintervention group received high-dose glargine on admission (P<0.001).

The postintervention group had significantly lower incidences of hypoglycemia per admission than the preintervention group (17.8% vs. 20.1%; P<0.001), and the decrease remained statistically significant after controlling for multiple variables. Variables predicting a higher incidence of hypoglycemia per admission were a lower body mass index, a lower blood glucose level, and a higher dose of glargine.

Mean (± SD) blood glucose levels were not significantly different between groups at 197.6 ± 10 mg/dL (10.98 ± 0.55 mmol/L) pre-intervention versus 199.7 ± 11 mg/dL (11.09 ± 0.61 mmol/L) postintervention (P=0.80). Rates of hyperglycemia were also not significantly different between groups.

These results suggest that “the dose restriction was effective in reducing hypoglycemia without impacting overall glycemic control,” the study authors wrote. They noted limitations to their study, such as its retrospective design and the omission of potentially important confounders because of limitations in the data (e.g., serum hematocrit). They were also unable to account for the potential effects of concurrent short- or intermediate-acting insulins.