Therapeutic options for vancomycin-resistant enterococcal (VRE) bloodstream infections are extremely limited. Chloramphenicol is effective when VRE isolates are susceptible to this agent. However, longitudinal trends in chloramphenicol-resistant VRE (CR-VRE) are unknown. The possible association between CR-VRE and antibiotic use has not been studied. We analyzed the antimicrobial susceptibility profiles of all VRE blood isolates from 1991-2000 at our institution. We performed a correlational study to examine the relationship between annual hospital-wide use of specific antibiotics and antibiotic classes and CR-VRE prevalence. During the 10-year study period, the prevalence of CR-VRE increased from 0 to 11% ( P< 0.001, trend). CR-VRE prevalence was correlated only with chloramphenicol use (P=0.05 ) and quinolone use (P= 0.01 ). If these trends continue, dependence on newer, more expensive agents will increase. The correlation between both chloramphenicol use and quinolone use and the prevalence of CR-VRE suggests that efforts to preserve the utility of chloramphenicol in VRE infections may depend on optimizing the use of these agents.