How predictive is PK/PD for antibacterial agents? Niels Frimodt-Møller * Microbiological R&D, Statens Serum Institut, 5 Artillerivej, DK-2300 Copenhagen S, Denmark Abstract The pharmacodynamic (PD) parameters most often used in studies of antibiotic effect include the following relationships between the antibiotic concentration curve in serum as a surrogate marker for the antibiotic concentration at the infection site, the peak/ minimal inhibitory concentration (MIC) ratio, the area under the curve (AUC)/MIC ratio and the duration of time the concentration exceeds the MIC (T MIC ). The MIC plays an important role also as a PD marker, and its precision in this respect is discussed. The predictive role of T MIC is important for drugs showing minimal concentration dependent effect such as the b-lactam antibiotics, the macrolides and others. The time can be calculated as the chronological time measured or as the (cumulative) per cent of the dosing interval covered by the dose. Several clinical studies have confirmed this relationship. It can be deduced from experimental as well as clinical studies that there is a minimal effective time (MET), which needs to be covered by the antibiotic concentration at the site of infection in order to achieve cure. Dosing according to this MET will result in the least antibiotic needed for the shortest duration. In several cases a single dose will suffice to cover the MET. If this is not possible the antibiotic should be dosed in a way, that each dose will surpass the MIC for at least 40  /50% of the dosing interval. For antibiotics with a clear concentration-dependent bacterial killing effect the most important pharmacokinetic/pharmacodynamic (PK/PD) index is the peak/ MIC ratio (or the AUC/MIC ratio). This is the case for aminoglycosides and fluoroquinolones, and for both classes a peak/MIC ratio of at least 10 within the first 24 h of treatment has been shown to result in around 90% bacteriological as well as clinical cure. One consequence of clinical dosing has been the once-a-day (OD) dosing for aminoglycosides, which is the standard mode of therapy in many countries. Clinical studies in the field of antibacterial PD are still relatively scarce, and much information is needed to enable relevant dosing strategies for all types of antibiotics against all common infections and micro-organisms. # 2002 Published by Elsevier Science B.V. and International Society of Chemotherapy. Keywords: PK/PD; Antibacterial agents; b-lactam antibiotics; MIC; Aminoglycosides; Fluoroquinolines; Pharmacodynamics; Pharmacokinetics 1. Introduction Extensive efforts have been invested during the last 20 years in studying the role of PK parameters in order to achieve optimal effects of antibiotics in vivo. Although soon after the appearance of penicillin, Eagle and co- workers had determined its PD properties by using elegant experimental studies [1,2], (i.e. the importance of the time the penicillin concentration remained above the minimal inhibitory concentration (MIC) of the infecting pathogen), this knowledge was never really implemented for dosing strategies. Only during the 1980 and 1990s has renewed interest in this field clarified the importance of PD for dosing of antibiotics, not only for optimizing the effect but also for minimizing the risk of resistance developing during treatment. In the following review I will try to evaluate the predictive value of pharmacokinetic/pharmacodynamic (PK/PD) parameters and indices for their effect in vivo, drawing mainly on experience from clinical studies rather than experimental animal studies. 2. The importance of the MIC The basic PK/PD indices which have been used in most studies to correlate with efficacy are those shown in Fig. 1: the ratio of the peak antibiotic concentration and MIC, the area under the concentration  /time curve and MIC (AUC/MIC) ratio and the time the antibiotic concentration remains /MIC (T MIC ). Although other parameters could be equally relevant, these are used most often because they are factors which can be * Tel.:  /45-32-68-3646; fax:  /45-32-68-3887. E-mail address: nfm@ssi.dk (N. Frimodt-Møller). International Journal of Antimicrobial Agents 19 (2002) 333  /339 www.isochem.org 0924-8579/02/$ - $20 # 2002 Published by Elsevier Science B.V. and International Society of Chemotherapy. PII:S0924-8579(02)00029-8