What Can We Learn from Studies Comparing Linezolid with Vancomycin in Neutropenic Patients When Vancomycin Dosages Are Not Optimized?

To the Editor—Jaksic et al. [1] compared the safety and efficacy of linezolid (600 mg every 12 h), and vancomycin (1 g every 12 h) in febrile neutropenic patients. Although the authors specified that coinvestigators were allowed to monitor vancomycin levels in serum, in accordance with local practice, the article does not indicate whether adjustment of the vancomycin dosage could be performed throughout the study. The usual recommended intravenous dose of vancomycin is 30 mg/kg/day in adults; use of this dosage achieves serum concentrations of 25 mg/L (peak level) and 2 mg/L (trough level) in healthy volunteers [2], but pharmacokinetic studies of vancomycin therapy in neutropenic patients have shown a 3-fold increase of the initial volume of distribution and a shortened (3-fold) half-life, compared with values in healthy subjects [3, 4]. Although there have never been any definitive data correlating vancomycin efficacy and concentration, most experts will aim to achieve trough serum concentrations of ⩾15 mg/L in patients with potentially severe infections [5], as recommended by the American Thoracic Society—Infectious Diseases Society of America guidelines for the treatment of methicillin-resistant Staphylococcus aureus pneumonia [6]. The pharmacodynamic parameters that best express vancomycin bactericidal activity are the time during which serum concentration is greater than the MIC for the organism [2] and the area under the curve divided by the MIC [7]. In neutropenic patients, pharmacodynamic principles suggest that an optimal regimen of a time-dependent killing agent with little postantibiotic effect, such as vancomycin, should achieve serum concentrations greater than the MIC 100% of the time. This may be easily obtained with therapeutic drug monitoring or when vancomycin dosages are adapted to the patient's weight (i.e., 30 mg/kg/day) [8]. On the other hand, with the fixed dosages of vancomycin used in the study by Jaksic et al. [1] (1 g every 12 h), many patients may not achieve this pharmacodynamic parameter. Could the authors detail the range of vancomycin trough serum concentrations obtained in the health care centers where vancomycin serum levels were monitored? In addition, given that the duration of neutropenia is an acknowledged risk factor for infectious complications [9], the delayed absolute neutrophil count recovery in patients receiving linezolid is of concern, despite the authors' assumption that “it may be attributable to physiological processes during recovery from acute bacterial infection” [1, p. 605]. Indeed, these physiological processes would not explain why neutrophil count recovery was significantly delayed in patients receiving linezolid, compared with those receiving vancomycin. Reports of linezolid myelotoxicity could be a serious limitation to the use of this agent in neutropenic patients [10]. Potential conflicts of interest. P.T. and C.C.: no conflicts.