W. Weber*, L. Harnisch, M. Zimmer, C. Rokitta, P. Mendes
Clinical Research, Hoechst AG, Frankfurt, Germany
Introduction
For a successful therapy with beta-lactam antibiotics, it is mandatory to maintain the drug plasma concentrations above the MIC (minimal inhibition concentration) of the respective pathogen for a maximum period of time within a given dosing interval. Due to the uncertainty regarding the causative pathogen(s) it is essential that the drug concentrations in general and the trough levels in particular remain as high as possible in order to obtain a sufficient antibacterial effect. On the other hand it is necessary to consider the dosing regimen in order to minimise the risk of adverse events.
Prevailing problems
In a twice daily dosing regimen, the trough levels at 12 h post dose are not readily detectable, since the plasma concentrations are above the present LOQ (limit of quantification) of the assay for only up to six hours after the end of infusion.
Former solution to the problem
The analysis comprised the extrapolation of the biexponential plasma pharmacokinetics during the first six hours after administration to 12 hours. This approach predicts very low trough levels. Unlike the plasma concentrations however, the urinary drug concentrations are detectable in the individual urine collection periods over the entire dosing interval. With the available plasma and urinary concentration data, it is now possible to re-evaluation the biexponential model using the population approach.
Contradictions in the biexponential model
Normal extrapolation of the biexponential pharmacokinetics predicts a lower renal drug excretion than is actually measured in urine. This contradiction suggests a triexponential disposition, in which, due to the sparse individual data, the terminal phase can not be adequately described with conventional pharmacokinetic methods.
New solution to the problem
Integration of urinary excretion data into the NONMEM-analysis allows a re-evaluation of the hypothesis regarding triexponential pharmacokinetics.
Experimental evaluation
The pharmacokinetics of cefotaxime was studied in 12 healthy subjects who received a single 2 g iv. dose administered as a short infusion. The cefotaxime and desacetyl-cefotaxime plasma concentrations were below the LOQ six and eight hours, respectively after the infusion. Concentrations of both the parent compound and the metabolite were detectable in urine up to the 8-16 collection period.
Results of the NONMEM-analysis
The urinary excretion of cefotaxime in the terminal phase is significantly better described in a triexponential than in a biexponential model. The Likelihood-Ratio test was used to determine the differences between the two models.
Therapeutic consequences
The trough levels of cefotaxime following twice daily dosing are about 30-fold higher than those anticipated until now and as a result are above the MIC of many therapeutically relevant pathogens. Additionally, the metabolite, desacetyl-cefotaxime, also contributes to the antibiotic effect. Accordingly, a twice daily dosing regimen for cefotaxime is adequate and this is supported by data generated in several clinical studies [Shah et al, Drugs 1988; (35 Suppl.2); 190-194; Wittmann & Frommelt, Infection 1985, 13 (Suppl. 1); 37-42].