Type 2 diabetes mellitus (T2DM) is associated with an increased risk of stroke and an unfavourable outcome following stroke. Apart from pioglitazone, glucose-lowering modalities have not been shown to protect against stroke. Nevertheless, there is evidence from experimental studies of potential neuroprotective effects with dipeptidyl peptidase (DPP)-4 inhibitors, especially if treatment starts before stroke.

To perform a meta-analysis of available evidence regarding the risk of stroke in individuals taking DPP-4 inhibitors.

All available data from prospective randomized placebo-controlled trials involving DPP-4 inhibitors in T2DM patients published up to December 2015 were considered. The included trials reported data on the incidence of stroke with a recruitment rate of at least 100 diabetes patients and a follow-up of at least 12 weeks.

A total of 19 small randomized clinical trials (RCTs) evaluating the efficacy and safety of gliptins (n=9278), along with three multicentre prospective double-blind placebo-controlled RCTs assessing cardiovascular outcomes as the primary endpoint and involving 36,395 T2DM patients, were included in the analysis. Pooled analysis of the small RCTs showed a non-significant trend towards benefit with DPP-4 inhibitors against stroke [odds ratio (OR): 0.639, 95% confidence interval (CI): 0.336–1.212; P=0.170]. In contrast, in the analysis of RCTs reporting on cardiovascular safety, there was no difference in the risk of stroke with gliptin treatment compared with a placebo (OR: 0.996, 95% CI: 0.850–1.166; P=0.958).

The promising data from experimental studies regarding cardioprotective gliptin-associated effects against stroke were not supported by available data from trials specifically looking at cardiovascular safety.