• Continuous infusion;
  • GISA;
  • MRSA;
  • serum concentration;
  • vancomycin


  1. Top of page
  2. Abstract
  3. References

Vancomycin serum concentrations were determined for 1737 patients treated with either 2 × 1 g of vancomycin or 4 × 500 mg daily (780 patients), according to current nomograms, or by continuous infusion (957 patients) with a loading dose (1 g) and a total of 2–6 g daily. Trough serum concentrations were determined after 36–48 h. Adequate serum levels for the treatment of a normal methicillin-resistant Staphylococcus aureus (MRSA) and a glycopeptide-intermediate S. aureus (GISA) were observed in 81% and 20.9% of patients, respectively. The data support theoretical arguments that higher and more sustained serum levels of vancomycin, obtained by continuous infusion, may enhance clinical efficacy.

More than 45 years after the introduction of vancomycin for the treatment of severe antibiotic-resistant staphylococcal infections, there are still controversies regarding the determination and interpretation of vancomycin serum concentrations [1–7]. Older studies focused on toxicity without considering clinical efficacy [8–11]. When performed and reported, serum assays always involved very few patients and the interpretation of the results was debatable. The aims of the present study were to determine the trough vancomycin serum concentrations in a large number of patients, and to consider the rationale for modifying the currently recommended ‘acceptable therapeutic range’, taking into account the pharmacokinetics, toxicity and antibacterial activity of vancomycin in combination with the clinical results.

Serum assays are performed routinely in the Hospital Saint Joseph, Paris, for all patients receiving vancomycin therapy. In total, 1737 patients treated during 1998–2004 were enrolled in the study. The patients received either 1 g of vancomycin every 12 h, or 500 mg every 6 h (according to current nomograms), or were treated by continuous infusion with a loading dose (1 g) and a total daily dose of 2–6 g. The daily dose was then adapted after further monitoring. The first trough serum assays were performed at steady state after treatment for 36–48 h; 5–10 mL of blood was drawn from the controlateral vein, and assays were performed within 1 h by the fluorescence polarisation immunoassay (TDx; Abbott Diagnostics, Rungis, France). The results are presented in Table 1.

Table 1.  Trough vancomycin serum levels (first assay only) obtained in 1737 patients after either 2–4 separate doses (n = 780) or administration by continuous infusion (n = 957)
Serum level (mg/L)No. patients with the indicated serum levela
2–4 separate dosesContinuous infusion
  • a

    Dialysed patients excluded.

<5148 (19.0%) 15
5–10204 60 (7.9%)
20–25 70 (68.8%)176
25–30 35121 (71.2%)
30–35 22 74
35–40 12 59
>40 26 (12.2%) 67 (20.9%)

When the ‘Mississippi mud’ was released during the 1950s, it contained impurities responsible for renal and ototoxicity or allergic reactions [8–11]. Cases of toxicity occurred during the first 10 years of vancomycin use, often in patients receiving other nephrotoxic compounds, or who had pre-existing renal dysfunctions. In the present study, <1% of the 1737 patients treated during the 7-year period of the study with trough concentrations ≤40 mg/L suffered from vancomycin-related toxicity (N. Desplaces and A. Ben Ali, personal communications). Vancomycin has a narrow spectrum of activity, restricted to most Gram-positive bacteria, and is the drug of choice for the treatment of methicillin-resistant Staphylococcus aureus (MRSA). The vancomycin MIC for MRSA is 1–2 mg/L for fully vancomycin-susceptible strains. All strains with MICs ≥ 4 mg/L can be considered to be (hetero) glycopeptide-intermediate S. aureus (GISA) on the basis of population analysis profiles [12,13].

Vancomycin inhibits peptidoglycan synthesis by binding to the D-Ala-D-Ala terminus of the nascent murein monomer, resulting in the inhibition of cell-wall synthesis. Only 50% of the vancomycin arriving at the surface of a staphylococcus will reach the target site. GISA are characterised by a thicker cell-wall with increased amounts of peptidoglycan, and the increased quantities of unprocessed D-Ala-D-Ala cause increased ‘trapping’ and ‘clogging’, resulting in higher vancomycin MICs and the increased inoculum effect observed with GISA in comparison with fully vancomycin-susceptible strains [12–15]. This can explain the therapeutic failures observed in the treatment of GISA infections, where a concentration of 8–16 mg/L is required at the site of infection in order to totally inhibit large numbers of bacteria (108 CFU/g), in contrast to 3–4 mg/L for a fully vancomycin-susceptible strain [12–14]. Continuous infusion increases diffusion of vancomycin in body fluids and tissues, with the result that a more sustained concentration is achieved more quickly.

The ‘therapeutic range’ for vancomycin was established in the early 1980s in four healthy subjects who each received 500 mg of vancomycin, yielding peak concentrations of 20–40 mg/L and trough concentrations of 5–10 mg/L, obviously without any clinical correlation [13]. Vancomycin is 50% protein-bound in serum, which explains the 2–4-fold MIC increase in vivo and the occurrence of breakthrough positive blood-cultures, particularly in patients infected with GISA strains [12–15]. The tissue concentration of vancomycin does not exceed 30–40% of the serum concentration [13]. Other important factors to be considered include the decreased in-vivo bactericidal activity, the inoculum effect, the attachment of staphylococci to foreign bodies, and the production of biofilm. A trough vancomycin serum level of 15–25 mg/L is mandatory for the treatment of a fully glycopeptide-susceptible MRSA, with 30–40 mg/L (administered by continuous infusion) being required for treatment of a GISA infection [12,16]. In the present study of 1737 patients who were treated either conventionally or with continuous infusion, 19% and 7.9%, respectively, of those infected with a fully vancomycin-susceptible S. aureus, and 87.8% and 79.1%, respectively, of those infected with a GISA isolate failed to receive an adequate dose. Similarly, a study of paediatric patients reported that <20% had adequate therapeutic serum levels [6]. Theoretical aspects, often unfounded and misinterpreted, such as nomograms based solely on creatinine clearance, are likely to be inaccurate for some patients [7]. Other practical considerations, including cost, should not counterbalance the risk to the patient and the consequences of inadequate treatment [1].

Several previous reports have claimed that clinical cure does not always correlate with vancomycin serum concentrations [2,3], and no clinical cure attributed to vancomycin can be expected if the serum concentration remains below the MIC, particularly in the case of GISA. Sabath [14] reported that up to 6 g vancomycin/day were required to eliminate strains that, from the increase in MICs observed with a heavy inoculum (from 2 to 32–64 mg/L), were probably GISA. Nomograms are necessary, but monitoring is essential for dose adjustment and follow-up. The data presented above strengthen the theoretical arguments that higher and more sustained serum levels, obtained by continuous infusion, may contribute to a better clinical efficacy.

Some investigators have suggested that determination of vancomycin serum concentrations is unnecessary until the relationship between serum concentration and clinical outcome has been demonstrated [2]. However, after waiting for such results for more than 30 years, it seems appropriate to continue monitoring vancomycin serum levels in order to ensure effective therapeutic concentrations until the results of well-designed clinical studies become available.


  1. Top of page
  2. Abstract
  3. References
  • 1
    Darko W, Medicis JJ, Smith A, Guharoy R. Lehmann DE. Mississippi mud no more: cost-effectiveness of pharmacokinetic dosage adjustment of vancomycin to prevent nephrotoxicity. Pharmacother 2003; 23: 643650.
  • 2
    Edwards DJ, Pancorbo S. Routine monitoring of serum concentrations: waiting for proof of its value. Clin Pharm 1987; 6: 652654.
  • 3
    Moellering RC. Monitoring serum vancomycin levels: climbing the mountain because it is there? Clin Infect Dis 1994; 18: 544546.
  • 4
    Roghmann MC, Perdue BE, Polish L. Vancomycin use in a hospital with vancomycin restriction. Infect Control Hosp Epidemiol 1999; 20: 6063.
  • 5
    Tobin CM, Darville JM, Thomson AH et al. Vancomycin therapeutic drug monitoring: is there a consensus view? The results of a UK national external quality assessment scheme (UK NEQAS) for antibiotic assays questionnaire. J Antimicrob Chemother 2002; 50: 713718.
  • 6
    Nandi-Lozano E, Ramirez-Lopez E, Avila-Figueroa C. Pharmacologic clinical monitoring of serum vancomycin levels in pediatric patients. Rev Invest Clin 2003; 55: 276280.
  • 7
    Rodvold KA, Zokufa H, Rotschafer JC. Routine monitoring of serum vancomycin concentrations: Can waiting be justified ? Clin Pharm 1987; 6: 655658.
  • 8
    Bailie GR, Neal D. Vancomycin ototoxicity and nephrotoxicity. A review. Med Toxicol Adverse Drug Exper 1988; 3: 376386.
  • 9
    Geraci JE, Heilman FR, Nichols DR, Wellman WE, Ross GT. Some laboratory and clinical experience with a new antibiotic, vancomycin. In: Antibiotics annual 1956–57. New York: Medical Encyclopedia Inc., 1957; 90106.
  • 10
    Rybak MJ, Albrecht LM, Boike SC, Chandrasekar PH. Nephrotoxicity of vancomycin alone and with an aminoglycoside. J Antimicrob Chemother 1990; 25: 679687.
  • 11
    Farber BF, Moellering RC. Retrospective study of the toxicity of preparations of vancomycin from 1974 to 1981. Antimicrob Agents Chemother 1983; 23: 138141.
  • 12
    Goldstein FW, Kitzis MD. Vancomycin-resistant Staphylococcus aureus: no apocalypse now. Clin Microb Infect 2003; 9: 761765.
  • 13
    Hiramatsu K. Vancomycin-resistant Staphylococcus aureus: a new model of antibiotic resistance. Lancet Infect Dis 2001; 1: 147155.
  • 14
    Sabath LD. Vancomycin dosing in life-threatening staphylococcal infections: need for measuring and achieving high peak serum levels for inoculum-related resistance (‘Cryptic vancomycin resistance’) [abstract 1774]. In: Program and abstracts of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington, DC: American Society for Microbiology, 1999: 43.
  • 15
    Laplante KL, Rybak MJ. Impact of high-inoculum Staphylococcus aureus on the activities of nafcillin, vancomycin, linezolid, and daptomycin, alone and in combination with gentamicin, in an in vitro pharmacodynamic model. Antimicrob Agents Chemother 2004; 48: 46654672.
  • 16
    Wysocki M, Delatour F, Faurisson F et al. Continuous versus intermittent infusion of vancomycin in some staphylococcal infections: prospective multicenter randomized study. Antimicrob Agents Chemother 2001; 45: 24602467.