The in vivo activity of tigecycline was evaluated in an experimental pneumonia model (C57BL/6 mice) by Acinetobacter baumannii. Two clinical strains were used: minimum inhibitory concentrations (MICs) of imipenem and tigecycline 1 and 2 microg/mL (imipenem-susceptible, IPM-S), and 8 and 2 microg/mL (imipenem-intermediate, IPM-I), respectively. For imipenem (30 mg/Kg), T/CMI (h) were 1.04 and 0.51 for IPM-S and IPM-I, respectively. For tigecycline (5 mg/Kg), the area under the concentration-time curve (AUC)/MIC(0-24 h) (serum and lung) were 9.24 and 4.37 (for the two strains), respectively. In the efficacy experiments with the IPM-S, imipenem (log CFU/g 3.59 +/- 0.78, p = 0.006) and tigecycline (2.82 +/- 1.2, p = 0.054) decreased the bacterial counts in lungs with respect to its controls; with the IPM-I, both imipenem (1.21 +/- 0.52, p = 0.002) and tigecycline (3.21 +/- 0.28, p = 0.035) decreased the bacterial counts with respect to the controls. In the survival experiments, with the IPM-S, the mortality was the same in the control (67%) and in the tigecycline (77%) groups, and imipenem reduced it (21%, p = 0.025); with the IPM-I, the mortality was the same in the control (87%) and in the tigecycline (85%) groups, and imipenem (0%) reduced it (p < 0.001). In summary, the present study shows that tigecycline is less efficacious than imipenem in the treatment of experimental A. baumannii pneumonia caused by IPM-S and IPM-I strains.