Given the short-comings of randomized controlled trials, such as low numbers of patients and events, we aimed to perform a network meta-analysis of observational trials on the effects of non-insulin glucose-lowering drugs on all-cause and cardiovascular mortality in type 2 diabetes.

We searched MEDLINE and EMBASE through August 2017. From the 38 eligible studies, we extracted patient characteristics, exposure, outcome, confounder-adjusted effect measures, and risk of bias. All-cause and cardiovascular mortality risk were assessed with random-effects network meta-analysis.

In 32 studies reporting on all-cause mortality (1,661,945 persons) and in 11 studies on cardiovascular mortality (358,784 persons), 119,591 (7.2%) and 9,221 (2.6%) persons, respectively, died during follow-up. Compared to metformin, sulfonylurea drugs (SU) increased all-cause mortality risk (HR: 1.42 (95% CI: 1.29-1.56)). Dipeptidyl peptidase-4 inhibitors (DDP4i) added to metformin significantly decreased all-cause mortality risk compared to metformin monotherapy (HR: 0.76 (0.63-0.93)), any other drug class as monotherapy, and SU + metformin dual therapy (HR: 0.69 (0.60-0.79)). DPP-4i + metformin dual therapy was associated with even larger reductions in risk of cardiovascular mortality; HR: 0.40 (0.21-0.79) when compared to metformin monotherapy, and HR: 0.41 (0.23-0.74) when compared to SU + metformin dual therapy. Glucagon-like-peptide-1 receptor agonists + metformin dual therapy tended to, but did not significantly, decrease mortality risk when compared to metformin monotherapy (all-cause mortality, HR: 0.88 (0.53-1.47); cardiovascular mortality, HR: 0.74 (0.30-1.82)).

In conclusion, this large network meta-analysis evidently confirmed that SU, compared to metformin, increased all-cause mortality risk. DPP-4i + metformin dual therapy, compared to other treatments, resulted in a consistent risk reduction of both all-cause and cardiovascular mortality.


B. Bongaerts: None. L.M. Lindner: None. A. Hoyer: None. O. Kuss: None. C. Herder: Other Relationship; Self; Sanofi, Eli Lilly and Company. H. Al-Hasani: None. K. Müssig: None. M. Roden: Speaker's Bureau; Self; Boehringer Ingelheim GmbH. Research Support; Self; Boehringer Ingelheim GmbH. Consultant; Self; Poxel SA. Research Support; Self; Danone Nutricia Early Life Nutrition, GlaxoSmithKline plc., Nutricia Advanced Medical Nutrition, Sanofi. W. Rathmann: Advisory Panel; Self; AstraZeneca. Research Support; Self; Novo Nordisk A/S.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at