Intermittent bolus dosing of ceftazidime in critically ill patients.

Ceftazidime is frequently used in critically ill patients, particularly for the treatment of Pseudomonas aeruginosa infections. The recommended dosing regimen is based on pharmacokinetic data obtained in healthy volunteers and may not be appropriate in the critically ill. We administered ceftazidime in the maximum recommended dose (2 g i.v. every 8 h) to ten critically ill patients with normal plasma creatinine. Eighteen arterial blood samples were taken from each patient over the first 8 h for measurement of ceftazidime concentrations and subsequent compartmental pharmacokinetic analysis. An additional trough sample was taken from each patient on day 3. Although mean pharmacokinetic variables did not differ from previously reported data in normal volunteers there was wide variability in plasma drug concentrations. Three of our patients had plasma ceftazidime concentrations less than the MIC for P. aeruginosa (8 mg/L) and nine had concentrations less than 5 x MIC, which has been recommended to ensure efficacy. On day 3 trough ceftazidime concentrations were less than the MIC in four out of the seven patients in whom measurements were made and less than 5 x MIC in the remaining three. There was no clinical predictor of which patients would have low plasma concentrations. Our results show that plasma concentrations of ceftazidime are very variable when the recommended intermittent bolus dosing regimen is used and may result in inadequate plasma concentrations of drug in critical infections. This may result in treatment failure and the emergence of antibiotic resistance. A loading dose followed by continuous infusion should overcome these problems but this awaits in-vivo evaluation.