We have previously described the activity of low-dose clindamycin in extended-interval dosing regimens by determination of bactericidal titer in serum. In this study, we used a one-compartment in vitro dynamic infection model to compare the pharmacodynamics of clindamycin in three intravenous-dosing regimens (600 mg every 8 h [q8h], 300 mg q8h, and 300 mg q12h) against three clinical isolates of Staphylococcus aureus and two clinical isolates of Streptococcus pneumoniae. Test organisms were added to the central compartment of the model to yield a starting inoculum of 10⁶ CFU/ml. Clindamycin was injected as a bolus into the central compartment at appropriate times over 48 h to simulate the q8h or q12h dosing regimens. Drug-free culture medium was then pumped through the system to mimic a half-life of 2.4 h. At predetermined time points during the experiment, samples were removed from the central compartments for colony count determination and drug concentration analysis. The rates of killing did not significantly differ among the three clindamycin dosing regimens against either S. aureus or S. pneumoniae (P = 0.88 or 0.998, respectively). Likewise, no significant differences in activities were detected among the three regimens against staphylococci (P = 0.677 and 0.667) or pneumococci (P = 0.88 and 0.99). Against an S. aureus isolate exhibiting inducible macrolide-lincosamide-streptogramin B resistance, none of the three clindamycin regimens prevented regrowth of the resistance phenotype in the model. In this model, clindamycin administered at a low dose in an extended-interval regimen (300 mg q12h) exhibited antibacterial activity equivalent to that of the 300- or 600-mg-q8h regimen.