Background:Combination schemes are commonly used for the treatment of infections due to carbapenemase-producing Klebsiella pneumoniae (CP-Kp). We therefore investigated the in vitro effectiveness of double and triple combinations of meropenem, colistin and tigecycline against CP-Kp isolates with different resistance mechanisms in a static broth microdilution model and a pharmacokinetic-pharmacodynamic model. Methods:One WT isolate and seven CP-Kp isolates with different carbapenem resistance mechanisms and increasing MICs of meropenem (4-512 mg/L), colistin (0.5-32 mg/L) and tigecycline (0.25-4 mg/L) were tested with a 3D chequerboard microdilution method. Combinations were then assessed in an in vitro pharmacokinetic-pharmacodynamic model simulating 50 and 100 mg of tigecycline q12h as a 1 h infusion, 4.5 million units of colistin q12h as a 1 h infusion and 1 g of meropenem q8h as 1 and 0.5 h infusions for 2 days. Results:In the chequerboard assay, interactions within the triple combination were mainly additive with a median (range) fractional inhibitory index of 0.66 (0.22-1.26). In the dynamic model, meropenem alone was bactericidal against isolates with MICs up to 4 mg/L, whereas bactericidal activity was found with the double combination meropenem + colistin and the triple combination meropenem + colistin + tigecycline against CP-Kp isolates with meropenem MICs of 16 and 256 mg/L, respectively. A high dose (100 mg) of tigecycline and a prolonged infusion (1 h) of meropenem increased the efficacy of the triple combination. Conclusions:Despite the merely additive interactions in the chequerboard assay, the triple combination of meropenem, tigecycline and colistin was bactericidal in the dynamic model against highly resistant CP-Kp isolates. This effect was more pronounced if prolonged infusion of meropenem and high tigecycline dosing were used.