Résumé : Ventilator-associated pneumonia (VAP) occurs in nearly one-third of mechanically ventilated patients in the Intensive Care Unit. Piperacillin/tazobactam (TZP) is currently recommended in the empirical treatment of VAP, but intermittent dosing may result in inadequate serum concentrations. The efficacy and costs of continuous infusion (CI) of TZP, using therapeutic drug monitoring for real-time dose adjustment, was assessed in a prospective pilot study of 16 patients with VAP. TZP was given as a loading dose of 2.0/0.25 g followed by a CI of 10.0/1.25g daily. Rapid antimicrobial susceptibility testing was used to determine the minimum inhibitory concentration (MIC) of the pathogens. TZP concentrations were determined by high-pressure liquid chromatography before and at 1, 6, 12, 24, 48, 72 and 96 h after the onset of administration. Dosages were adjusted to maintain piperacillin concentrations four-fold above the MIC (T>4 × MIC) of the pathogen, with a maximum dose of 16.0/2.0 g. The cost of the total TZP administered was compared with the cost of a standard TZP regimen (16.0/2.0 g) if given over the same period of time. The median MIC for TZP was 1 μg/mL (range 0.025-32 μg/mL). TZP concentrations were adequate for 71% of pathogens on the first day of therapy. Clinical cure was achieved in 9/10 patients who had adequate drug concentrations and in 3/6 patients with insufficient levels. The daily dose of TZP received by CI was 37.5% less than that of a standard regimen, which corresponds to a saving of €15 on daily therapy costs compared with the standard regimen. In conclusion, CI of TZP achieved optimal drug concentrations in most patients with VAP, with a favourable impact on costs. Adequate drug concentrations were achieved for MIC ≤ 4 μg/mL, but higher dosages should be considered for the treatment of pathogens with low susceptibility thresholds.