Background: blockade of the RAAS lowered the incidence of Diabetes Mellitus in randomized clinical trials, but whether the de-activation of the RAAS can elicit the same beneficial effect in patients with renovascular hypertension (RVH) is unknown. Aim: to verify if endovascular treatment (EVT) could improve glycemic control in patients with atherosclerotic RVH. Methods: in the METRAS study (http://clinicaltrials.gov/show/NCT01208714, a randomized clinical trial comparing the effect of EVT on top of optimal medical therapy versus medical therapy alone (OMT) on GFR of the ischemic and contralateral kidney in patients with atherosclerotic RVH), glycemic control, as assessed by HbA1c at baseline and repeatedly during follow-up, was a pre-specified secondary endpoint of the study. Results: between June 2010 and March 2014, 16 patients were randomly assigned to EVT plus optimal medical therapy (n = 9) or OMT alone (n = 7). At baseline the 2 groups showed no significant difference of age, clinical and demographical features. All the patients, except two in the EVT group, were on a RAAS blocker. At 2 years follow-up HbA1c increased in patients on OMT (59±12 mmol/mol) from baseline values (45±16 mmol/mol, p<0.001). By contrast, in the patients assigned to EVT HbA1c remained stable (42±7 mmol/mol at baseline vs 41±5 mmol/mol at follow-up; p = NS; p<0.001 vs OMT at follow-up). When analyzed after multivariate adjustment for age, HbA1c at enrollment, and presence/absence of DM, the difference between the arms remained highly significant (p<0.001). Conclusions: in patients with atherosclerotic RVH and chronic activation of the RAAS, deactivation of the RAAS by means of EVT was associated with no increase in plasma levels of HbA1c over long term follow-up. At variance, similar patients assigned to optimized medical therapy alone showed worsened glycemic control over time. These findings support the contention of a causative role of RAAS activation in the incidence/progression of DM.