Comparative pharmacodynamics of azithromycin and roxithromycin with S-pyogenes and S-pneumoniae in a model that simulates in vitro pharmacokinetics in human tonsils

Most macrolides penetrate and persist in peripheral tissues, irrespective of plasma concentrations. For this reason, comparative pharmacodynamics of macrolides might be better based on tissue rather than plasma pharmacokinetics. The present study compares the antimicrobial effects of azithromycin and roxithromycin on Streptococcus pyogenes and Streptococcus pneumoniae using in vitro simulations of steady-state pharmacokinetics in human tonsils expected after a third 500 mg dose of azithromycin administered once a day and after a sixth 150 mg dose of roxithromycin administered twice a day. Clinical isolates of S. pyogenes and S. pneumoniae (MICs 0.12 and 0.47 mg/L of azithromycin, and 0.15 and 0.60 mg/L of roxithromycin, respectively) were used. More pronounced antistreptococcal effects were observed with azithromycin than with roxithromycin. Despite similar rates of initial killing of S. pyogenes and S. pneumoniae, the respective 12 h areas between the control growth curve and the time-kill curve of antibiotic-exposed bacteria (ABBCs) were 22% and 36% greater with azithromycin than roxithromycin. Moreover, with azithromycin, viable bacterial counts reached the theoretically achievable limit of detection (10 cfu/mL) 8-10 h after drug administration, with no regrowth within 48 h. In contrast to azithromycin, S. pyogenes and S. pneumoniae exposed to roxithromycin regrew 26 and 6 h, respectively, after initial reduction of the starting inoculum. Further in vitro simulations of tissue pharmacokinetics might be useful for pharmacodynamic comparisons among other macrolides.