The aim of this study was to assess the impact of vancomycin (VAN) versus linezolid (LZD) on renal function in patients with renal failure (RF) admitted to intensive care units. This was a multicenter, retrospective, comparative cohort study. Renal failure patients were treated with VAN or LZD for proven or suspected infections by multiresistant Gram-positive cocci. Changes in plasma creatinine levels and creatinine clearance at the start and end of treatment were used as endpoints. A total of 147 patients were treated with VAN (group A, n = 68) or LZD (group B, n = 79). Group B included more patients with diabetes mellitus [9 (13.2%) vs. 25 (31.6%); p = 0.007], septic shock [39 (57.4%) vs. 60 (75.9%); p = 0.013] and greater RF (mean ClCr 42.24 ml/min vs. 37.57 ml/min; p = 0.04). Renal function improved in patients from both groups who did not require renal replacement therapy. A greater improvement was seen in group B [percent decrease in Cr (27.94 vs. 9.48; p = 0.02) and percent increase in ClCr (95.96 vs. 55.06; p = 0.05)]. In group A, nine patients (13.2%) experienced an antibiotic-related increase in RF, and antibiotic was discontinued in five patients due to adverse effects. It is reasonable to avoid use of VAN in critically ill patients with acute renal failure.