To assess the relative efficacy and safety of basal insulin regimens in adults with type 1 diabetes mellitus (T1DM).
A systematic review and Bayesian network meta-analysis (NMA) of randomized controlled trials comparing two or more basal insulin regimens were conducted. The following basal insulin regimens were included: Neutral Protamine Hagedorn (iNPH) (once [od], twice [bid], and four times daily [qid]), insulin detemir (iDet) (od and bid), insulin glargine 100 IU (iGlarg) (od), and insulin degludec (iDegl) (od). We searched the following databases: MEDLINE via OVID, Embase via OVID, and the Cochrane Library (Wiley). Study quality was appraised using Cochrane risk-of-bias checklist for randomized controlled trials. Two outcomes (change in hemoglobin A [HbA ] and rate of severe/major hypoglycemia [SH]) were analyzed. Network inconsistency was assessed using Bucher and chi-square tests.
Thirty studies met the eligibility criteria. Twenty-five were included in the HbA network and 16 in the SH network. All studies were of moderate quality. No network inconsistency was evident in the HbA network. Of the seven regimens of interest, iDet (bid) had the highest probability of being best (mean change in HbA −0.48; 95% credible interval −0.69 to −0.29). In contrast, the SH network demonstrated both considerable uncertainty and significant network inconsistency (χ test, P = 0.003).
Of the specified frequency regimens, iDet (bid) had the highest probability of being the best basal insulin regimen in terms of reduction in HbA . Ranking of the regimens in terms of the SH rate was highly uncertain and no clear conclusion could be made.