Unfortunately, the effect of serum protein binding on antimicrobial activity still attracts considerable controversy and confusion among many clinicians and researchers. The proportional reduction of antimicrobial activity in the presence of serum or binding proteins has been thoroughly demonstrated for several anti-infectives in susceptibility testing.
Although of high interest, there are too few examples of well-controlled experiments that demonstrate the importance of serum protein binding on efficacy in vivo. The difficulty in showing the importance of protein binding in animal models largely lies in the fact that the class of anti-infectives with the greatest variability in serum protein binding are the P-lactam antibiotics. Since the in vivo efficacy of these agents is dependent upon T > MIC and they have relatively short half-lives in small animals, large differences in serum protein binding are required to produce significant differences in free-drug T> MIC. Merriken et al.  demonstrated the importance of serum protein binding on the efficacy of several structurally related analogs of penicillin in a mouse model of sepsis due to Staphylococcus aureus. All of the agents had similar in vitro potency against the test organism (MIC between 0.25 and 0.5 mg/L) and pharmacokinetic properties, but the percent bound to serum proteins ranged between 36% and 98%. Although the differences in pharmacokinetic properties of total drug were small (2.5-fold range), there was a 70fold difference among agents in dose required for survival in 50% of animals (ED50 ranged from 0.7 to 49.7). In a neutropenic mouse thigh model, we also compared the efficacy of three cephalosporin analogs with varying MICs to 2 strains of methi-cillin-resistant Staphylococcus aureus, pharmacokinetics, and protein binding. As shown in Fig. 7, bacterial killing was best described by the number of hours that free-drug concentrations exceeded the MIC.
Was this article helpful?