OBJECTIVES: We conducted a systematic review and meta-analysis comparing the incidence of contrast-induced nephropathy (CIN) in patients at risk with chronic kidney disease (CKD) undergoing computed tomography (CT). METHODS: Randomized controlled clinical trials designated to evaluate the nephrotoxicity related to iso-osmolar contrast media (IOCM) compared to low-osmolar contrast media (LOCM) were searched in the following electronic databases: PubMed/MEDLINE, EMBASE, ISI Web of Knowledge and Virtual Health Library (BVS-BIREME), as well as abstracts presented at related scientific societies meetings. Prior to data extraction, definitions of nephrotoxicity were established. We applied a random effects model of DerSimonian and Laird, with heterogeneity (Q statistic), publication bias (Egger and Begg test) and sensitivity analyses. RESULTS: Five studies were included with 716 randomized patients. When CIN was defined as increased serum creatinine (SCr) ≥ 25%, the relative risk (RR) was 0.71 (CI95% 0.40 to 1.26) – in favor of IOCM – and when it was defined as SCr ≥ 0.5 mg/dL it showed a RR 1.48 (CI95% 0.37 to 5.87) – favoring LOCM – in the four studies used this criterion. CONCLUSIONS: In patients with CKD undergoing CT there is a similar risk of CIN with the administration of any contrast media studied. CIN incidence depends on the chosen criteria and is lower with the definition of SCr ≥ 0.5mg/dL at 24-72h.