Time-kill curves as a tool for targeting ceftazidime serum concentration during continuous infusion

Sir, The time above the minimum inhibitory concentration (t > MIC) is considered to best predict the in vivo activity of β-lactam antibacterial agents, with administration by continuous infusion (CI) allowing optimization of the t > MIC over a 24 h period.1 Preclinical and clinical evidence indicates that the drug concentration in serum should reach 4–5 times the MIC value to exert the maximum bactericidal effect during CI.1,2 However, such a value may be hard to reach in infections caused by difficult to treat Gram-negative bacteria, like Pseudomonas aeruginosa. Ceftazidime is one of the more extensively studied drugs that are currently given by CI administration and its clinical efficacy has been confirmed in different patient populations.3–5 The in vitro study we report here was undertaken to assess the bactericidal activity of different ceftazidime concentrations on either susceptible or resistant P. aeruginosa isolates. All of the study strains were obtained from respiratory samples from patients hospitalized in intensive care units. Six of the P. aeruginosa strains were susceptible (CTZ-S) with MIC values ≤ 8 mg/L; four strains were intermediate (MIC = 16 mg/L; CTZ-I) and three strains were resistant (CTZ-R) with MIC values ≥ 32 mg/L. MIC values were determined for each of the study strains by the NCCLS methodology.6 Bactericidal activity was defined as a ≥3 log10 reduction in a 105 cfu/mL inoculum. Bacterial suspensions were prepared from an overnight agar plate culture, then transferred into Mueller–Hinton broth supplemented with calcium (25 mg/L) and magnesium (12.5 mg/L) and incubated for 2 h at 37°C to reach logarithmic phase. Ceftazidime concentrations of 2 ×, 4 × and 8 × MIC as exact multiples of the MIC value were added to the medium and bacterial counts carried out after 0, 4, 6, 8 and 24 h. In order to avoid ceftazidime carry-over, 0.1 mL aliquots were sampled with a detection limit of 10 cfu/mL. All the experiments were repeated two times. Ceftazidime exerted maximal bactericidal activity within 6–8 h in all the experiments (Figure 1). Eleven out of 13 strains and 12 out of 13 strains, exposed to 4 × and 8 × MIC, respectively, showed a decrease of more than 3 log10 cfu/mL at 6 h after antibiotic exposure. In the experiments where the strains were exposed to 2 × MIC, a decrease of 2.4–4.8 log10 cfu/mL was observed after 6–8 h of antibiotic exposure, even though four (MIC range 4–32 mg/L) out of 13 strains showed re-growth after 24 h. No variations in the bactericidal activity of ceftazidime could be correlated with the MIC value for the different strains. Results when the strains were exposed to a concentration of 4 × MIC were comparable to those at 8 × MIC, suggesting that no further bactericidal effect can be achieved by increasing the ceftazidime concentration above 4 × MIC (no concentration effect). At 2 × MIC, bactericidal activity was seen against all isolates; in general, this activity was comparable to that observed with 4 × and 8 × MIC, even on CTZ-R strains, but in one CTZ-R, one CTZ-I and two CTZ-S strains, re-growth at 24 h was observed. CI has been reported to be at least as effective as intermittent administration in severe infections.3–5 β-Lactam clinical efficacy may be optimized using administration by CI providing that serum drug concentrations are in excess of the MIC value of the infecting bacteria. From our results, the use of ceftazidime at a concentration of 4 × MIC gave excellent bactericidal activity within 6–8 h of antibiotic exposure and increasing the drug concentration to 8 × MIC did not increase the bacterial killing. Ceftazidime concentrations of 2 × MIC still demonstrated in vitro bactericidal activity against P. aeruginosa, but at 6–8 h after drug exposure, thus supporting the hypothesis that the 2 × MIC value may be sufficient to achieve bacterial killing over a 24 h period. Should such an assumption be confirmed, a reduction in the ceftazidime daily dosage, or treatment of more resistant bacteria, could be introduced. Before any conclusions can be reached as to clinical validity of our findings, the results of the present time–kill experiments should be integrated with pharmacokinetic/pharmacodynamic findings and confirmed in clinical trials using different patient sub-populations. Corresponding author. Tel: +39-045-9219573; Fax: +39-045-9218043; E-mail: fem75838@gsk.com Figure 1. Bactericidal activity of ceftazidime at concentrations of 2 ×, 4 × and 8 × MIC, against susceptible (a), intermediate (b) and resistant (c) P. aeruginosa strains. Values are mean ± s.d. of each strain result. ΔLog10 cfu/mL is the difference in cfu/mL at the time after antibiotic exposure starting from t = 0. 1Medicines Research Centre, GlaxoSmithKline; 2Anti-bacterials—Medical Department, GlaxoSmithKline S.p.A., Verona, Italy