Comparisons of diabetic retinopathy events associated with glucose‐lowering drugs in patients with type 2 diabetes mellitus: A network meta‐analysis

Aim To assess the comparative effects of glucose‐lowering drugs (GLDs) on the risk of diabetic retinopathy (DR) in patients with type 2 diabetes mellitus (T2DM). Methods We systematically searched Cochrane Central Register of Controlled Trials, PUBMED and EMBASE from inception to January 17, 2017 to identify randomized controlled trials (RCTs) that reported DR events among T2DM patients receiving any GLD. Random‐effects pairwise and network meta‐analyses were performed to calculate odds ratios (ORs) with 95% confidence intervals (CIs). Results A total of 37 independent RCTs with 1806 DR events among 100 928 patients with T2DM were included. The mean duration of diabetes was 8.7 years and mean baseline HbA1c was 8.2% (SD, 0.5%). Our network meta‐analysis found that DPP‐4i (OR, 1.20; 95% CI, 0.87‐1.65), GLP‐1RA (OR, 1.19; 95% CI, 0.94‐1.52) and SGLT2 inhibitors (OR, 0.79; 95% CI, 0.49‐1.28) were not associated with a higher risk of DR than placebo; however, a significantly increased risk of DR was associated with DPP‐4i in the pairwise meta‐analysis (OR, 1.27; 95% CI, 1.05‐1.53). Sulfonylureas, on the other hand, were associated with a significantly increased risk of DR compared to placebo (OR, 1.67; 95% CI, 1.01‐2.76). Conclusions Current evidence indicates that the association between DPP‐4i, GLP‐1RA or SGLT2 inhibitors and risk of DR remains uncertain in patients with T2DM. Some evidence suggests that sulfonylureas may be associated with increased risk of DR. However, given that DR events were not systematically assessed, these effects should be explored further in large‐scale, well‐designed studies.