Use of the t > MIC to choose between different dosing regimens of beta-lactam antibiotics

Sir, The major pharmacodynamic index correlating with in vitro and in vivo efficacy has been shown to be the time the concentration of an antibiotic remains above the MIC, or in short, the t > MIC.1 This relationship has been shown both in in vitro pharmacokinetic models and numerous animal experiments, and the analysis of clinical trials in humans seems to confirm this.2 This relationship between t > MIC and efficacy is beginning to be used to define or compare dosage regimens in humans. However, there are three major problems with this approach. The first is that, although there is a good relationship between t > MIC and effect, there is no universal agreement on the duration of this time. Should it be 40%, 70% or 100% of a dosing interval? The second problem is the MIC itself. If various dosage regimens of one drug, or even different drugs, are compared with each other with respect to the t > MIC, at which concentration should this be done? Finally, the issue on which we will focus in more detail is the calculation of the t > MIC for different dosage regimens. If a proper comparison is to be made between different regimens, the t > MIC should be calculated. In some recent papers, this approach has been used to compare dosage regimens of the same drug,3,4 or of different drugs.3,5 However, the disadvantage in these publications is that comparisons are made at one concentration only, or at two at the most, which does not furnish a full view of the subject. A more proper approach, we propose, would be to compare the t > MIC for the various dosing regimens over the whole range of MICs. This would establish a general pic-ture, and enable anyone to draw their own conclusions regarding the comparability of different dosing regimens. Differences and absolute values of t > MIC would be readily visible. An example is shown in the Figure for various dosing regimens of co-amoxiclav. Recently, a new formulation was introduced, containing 875 mg amoxycillin instead of 500 mg. One of the advantages of this would be that doses of the drug could be administered less frequently. Various claims have been made of equal or almost equal t > MICs at various concentrations (MICs),6 but an overall picture is generally unavailable. The major disadvantage of this approach, and perhaps one of the reasons that an illustration such as that shown in the Figure has not been used widely, is that computation of the t > MIC is difficult as soon as calculations involve more than one constant, as is the case in two-compartment models (one for distribution and one for elimination), or drugs given orally (one for absorption and one for elimination). The absorption case is complicated further by two components of the concentration–time curve, one ascending and one descending. These derivations are explicit, which implies that either the various dosing regimens have to be simulated and the t > MIC calculated by hand from the resulting concentration–time relationship, which is a tedious undertaking, or that the use of a numerical approach to solve the equations is necessary. To obtain the t > MIC relationships shown in the Figure, we used an Excel program that was developed for this specific purpose and readily shows the t > MIC–MIC relationship for various dosing regimens. We conclude that to compare various dosing regimens of one or more drugs, a figure showing the t > MIC–MIC relationship would be most helpful, and that the tools to reveal such relationships are now readily available. t > MIC–MIC relationship for four different dosing regimens of amoxycillin. Data for smulation were taken from Woodnutt & Berry (1999).4 The t > MIC was calculated as a percentage of the dosing interval at steady state, assuming a bioavailability of 80%. (•), 500 mg q6h; (▪), 500 mg q8h; (♦), 875 mg q8h; (□), 875 mg q12h.