LAUSR

Question In patients with type 2 diabetes mellitus, what are the efficacy and safety of rapid-acting insulin analogues compared with regular human insulin (RHI)? Review scope Included studies compared rapid-acting insulin analogues (insulin lispro, insulin aspart, insulin glulisine, or biosimilars) with RHI in adults 18 years of age who had type 2 diabetes. Studies were excluded if insulin was not administered subcutaneously via syringe, pen, or pump; treatment duration was <24 weeks; or pregnant women were included. Outcomes included mortality, macrovascular and microvascular complications, severe hypoglycemic episodes, glycemic control (hemoglobin [Hb] A1c), and nonsevere hypoglycemic episodes. Review methods EMBASE/Excerpta Medica (Oct 2017); MEDLINE, Cochrane Central Register of Controlled Trials, ClinicalTrials.gov, and WHO International Clinical Trials Registry Platform (Oct 2018); and reference lists, abstracts from major diabetology meetings, and pharmaceutical companies' trial registers were searched for randomized controlled trials (RCTs). The 3 main pharmaceutical companies producing rapid-acting insulin analogues, experts, and approval agencies were consulted. 10 RCTs (n =2751, mean age 57 y, 55% men when reported), ranging in size from 12 to 892 patients, met the selection criteria. Trial duration ranged from {24}* to 104 weeks (mean 41 wk). 5 trials had adequate random sequence generation, and 5 adequately reported allocation concealment. All trials were open-label, with no blinding of patients or investigators. 8 trials were at least partially commercially funded, and 2 did not report funding. Main results The results of meta-analyses for mortality, glycemic control, and nonsevere hypoglycemic episodes are in the Table. Heterogeneity of reporting precluded meta-analysis of severe hypoglycemic episode results; none of the 6 trials that reported this outcome showed a difference between treatment groups. No trials reported on macrovascular and microvascular complications. Conclusion In patients with type 2 diabetes, rapid-acting insulin analogues do not differ from regular human insulin for mortality or glycemic control. Rapid-acting insulin analogues vs regular human insulin (RHI) in patients with type 2 diabetes Outcomes Number of trials (n) Weighted event rates At 24 wk Insulin analogues RHI RRI (95% CI) Mortality 6 (2519) 0.4% 0.2% 66% (53 to 555) Mean difference (CI) Glycemic control (HbA1c) 9 (2608) 0.03% (0.16 to 0.09) Nonsevere hypoglycemic episodes (mean episodes /patient/mo) 7 (2667) 0.08 (0.00 to 0.16) Hb = hemoglobin; other abbreviations defined in Glossary. RRI and CI calculated from RHI event rate and odds ratio in article. Commentary The recent Cochrane review by Fullerton and colleagues found that rapid-acting analogue insulins did not reduce HbA1c more than RHI. The quality of some of the included trials may be questionable, but the results are similar to those of a 2014 report that found that 7 of 9 RCTs of patients with type 2 diabetes showed no difference in HbA1c between analogue insulins and RHI (1). These results are in contrast to the repeated recommendations from specialty organizations and specialists to use analogue insulins (2). Why? Perhaps it is based on the small pharmacokinetic and pharmacodynamic differences between analogue insulins and RHIs. However, with 20% to 30% intraindividual day-to-day variability in response to insulin, these minor pharmacokineticpharmacodynamic differences are clinically irrelevant. For instance, most of us were taught that RHI should be injected 20 to 30 minutes before a meal. However, a crossover study revealed no difference in glucose levels for 6 weeks after regular insulin was given either 20 minutes or just before meals (3). The cost of insulin tripled from 2001 to 2010 and doubled yet again between 2012 and 2016, with analogue insulins costing 2.5 times as much as RHI in the USA. Because of these increased costs, 25% of patients who require insulin are not taking the full amount of their prescribed doses (4). The results of the review by Fullerton and colleagues show that there is no reason not to use RHIs, which are less costly than analogue insulins.