Background. - Methicillin-resistant Staphylococcus aureus (MRSA) accounts for 10-40% of hospital-acquired pneumonia, and even more in intensive care units. The current guidelines for the treatment of MRSA nosocomial pneumonia include vancomycin and linezolid. The authors of 2 prospective randomized trials comparing vancomycin and linezolid in nosocomial pneumonia had concluded to the non-inferiority of linezolid. A slight superiority of linezolid was observed in the MRSA pneumonia subgroup, in terms of clinical success and survival, but no definite conclusion could be drawn. Methods. - A prospective randomized study was made to compare a fixed linezolid dose to dose-optimized vancomycin for the treatment of bacteriologically proven MRSA nosocomial pneumonia (ZEPHyR Study). Results. - Among the 165 patients treated by linezolid (57.6%) in the PP population, 95 were clinically cured at the end of the study, compared to 81 of the 174 patients treated by vancomycin (46.6%) (IC 95% of the difference 0.5%-21.6%, P = 0.042). Nephrotoxicity in the mITT population reached 8.4% in the linezolid group compared to 18.2% in the vancomycin group. Conclusion. - LNZ was superior to vancomycin for the treatment of MRSA nosocomial pneumonia. (C) 2013 Elsevier Masson SAS. All rights reserved.