Our aim was to compare Sodium-glucose co-transporter 2 inhibitors (SGLT-2i) to Dipeptidyl peptidase-4 inhibitors (DPP-4i) as add-on therapy to metformin.

We searched for randomized trials comparing SGLT-2i to DPP-4i as add-on therapy to metformin in Type 2 diabetes.We pooled trials reporting outcomes between 12 and 26 weeks together while trials reporting results ≥52 weeks were pooled together. The primary outcomes were the change in haemoglobin A1c (A1c) at ≤26 and ≥52 weeks. Sensitivity analyses were performed according to the dose of SGLT-2i and according to baseline A1c for the primary outcomes.

Seven trials met our inclusion criteria. There was a statistically significant reduction in A1c at ≥52 weeks favouring SGLT-2i compared to DPP-4i (MD [95% CI]=−0.11% [−0.20, −0.03]) but no significant difference at ≤26 weeks (MD [95% CI]=−0.05% [−0.16, 0.05]). SGLT-2i caused significantly more weight loss compared to DPP-4i at ≤26 weeks and ≥52 weeks (MD [95% CI]=−2.31kg [−2.66, −1.96] and −2.45kg [−2.83, −2.07], respectively). SGLT-2i treated patients had a significantly more genital infection compared to DPP-4i. On restricting the analysis according to the SGLT-2i FDA-approved dose, only higher doses at ≥52 weeks showed a statistically significant reduction in A1c compared to DPP-4i. On restricting the analysis according to baseline A1c, results favoured DPP-4i if baseline A1c was<8.5%, but favoured SGLT-2i if ≥8.5%.

While both SGLT-2i and DPP-4i can reduce A1c, SGLT-2i causes a more robust A1c reduction and more weight loss but with more genital infections. Higher doses of SGLT-2i showed more efficacy when compared to DPP-4i; however, this data should be interpreted cautiously given the limited number of trials.