Background and Aims: several studies have explored the impact of nonalcoholic fatty liver disease (NAFLD) on risk of incident type 2 diabetes mellitus (T2DM). However, the extent to which NAFLD may confer risk of incident T2DM remains uncertain. We performed a meta-analysis of relevant studies to quantify the magnitude of the association between NAFLD and risk of incident T2DM. Methods: we searched PubMed, Scopus and Web of Science (up to July 31, 2017) using pre-defined key-words to identify large observational cohort studies with a follow-up duration >1 year, in which NAFLD was diagnosed by imaging methods (mainly ultrasonography). No studies with biopsy-proven NAFLD were available for the analysis. Data from selected studies were extracted, and meta-analysed using random-effects modelling. Results: a total of 19 observational cohort studies with 296,439 individuals (30.1% with NAFLD) and nearly 16000 cases of incident T2DM over a median follow-up of 5 years were included in the final analysis. Patients with NAFLD had a higher risk of incident T2DM than those without NAFLD (random-effects hazard ratio 2.22, 95%CI 1.8-2.6; I2=79.2%). Notably, since we always used the fully adjusted hazard ratios for each eligible study, this random-effects hazard ratio was independent of multiple common risk factors for T2DM. Patients with more “severe” NAFLD (according to either ultrasonography or non-invasive fibrosis markers) were also more likely to develop incident T2DM (n=4 studies; random-effects hazard ratio 2.63, 95%CI 1.6-3.7; I2=82.4%). This risk was even greater among NAFLD patients with high NAFLD fibrosis score (n=1 study; random-effects hazard ratio 4.74, 95%CI 3.5-5.9). Sensitivity analyses did not alter these findings. Funnel plot and Egger’s test did not reveal significant publication bias. Conclusions: this comprehensive meta-analysis shows that imaging-diagnosed NAFLD is associated with an approximate doubling of the long-term risk of incident T2DM. However, the observational design of the eligible studies does not allow for proving causality. These findings pave the way for future large, prospective, histologically-based studies.