Glycemic Variability With Insulin Glargine Versus Detemir in Hospitalized Patients With Diabetes

Abstract

Background: Prior research has demonstrated increased mortality with increasing glycemic variability (GV) in hospitalized patients with diabetes. Objective: We aimed to compare glycemic variability (GV) of insulin glargine to detemir in the inpatient setting. Methods: This single-center, retrospective, cohort study evaluated noncritically ill patients with diabetes on long-acting insulin at a large academic medical institution between 2010 and 2017. This study was reviewed and approved by the Institutional Review Board. The formulary transitioned from insulin glargine to detemir in December 2013; therefore, patients were compared before and after transition. The primary endpoint was to compare coefficient of variation (CV), a measure of GV, between detemir and glargine. Secondary endpoints included GV measured by standard deviation (SD), CV within 72 hours of long-acting insulin initiation, length-of-stay (LOS), in-hospital mortality, and comparison between subgroups. Results: 2334 patients were included in the study, and there were 1167 in each group. CV was significantly less variable with detemir compared to glargine (33.7% versus 34.8%, difference = 1.09, p = 0.02) and remained significant after controlling for confounders. Similarly, SD was significantly less with detemir (p = 0.048). CV within 72 hours, LOS, and in-hospital mortality were not statistically different. Lastly, GV was higher in medical patients compared to surgical. Conclusion: Insulin detemir exhibited less GV than insulin glargine, although the small difference is unlikely to be clinically significant. Application of this data will aid in formulary decisions and support the use of either agent within the hospital setting.

Biosimilar insulin concepts

“…There is evidence that the glucose CV remains high in persons with diabetes treated with insulin glargine. An observational study of 1167 hospitalized persons with T1D and T2D treated with insulin glargine showed a glucose CV of 34.8%, 20 an outpatient continuous glucose monitoring study of 17 persons with T2D treated with insulin glargine 0.2 U/kg/day with 24‐h mean glucose 9.5 mmol/L showed glucose CV 26.8%, 21 and a study based on SMBG measured by 88 persons with T1D using insulin glargine as basal insulin showed the CV of the fasting and predinner glucose was 41.1% and 40.1%. 22 In considering the applicability to insulin prescription of the three levels of similarity defined by the FDA for biologics, then, one should realize that a given dose of insulin has highly variable expected action, reminding us of the dictum of Elliot Joslin in 1923, just 2 years after the discovery of insulin: “Insulin is a remedy primarily for the wise and not for the foolish, whether they be patients or doctors.” 23…”