In treatment of Pseudomonas aeruginosa with piperacillin tazobactam, addition of low dose gentamicin markedly reduces emergence of resistance and increases antibacterial effect.

Objective. To define pharmacokinetic/pharmacodynamic targets for combination therapy of piperacillin/tazobactam (P/T) plus gentamicin against Pseudomonas aeruginosa in vitro, and assess dosing strategies to minimize emergence of resistance (EoR).

Method. Kill curve methodology over 48 hours was used to assess the antibacterial effect (ABE) of combinations of P/T and gentamicin against a wild type clinical isolate of P. aeruginosa. EoR to both antimicrobials was assessed at 24 and 48 hours.

An in vitro dynamic pharmacokinetic model was used to assess ABE and EoR over 72 hours. The percentage of time above the MIC over 24 hours (%T>MIC) required to obtain a static and a three log(10) kill ABE was determined for P/T alone, and for P/T in the presence of gentamicin. The pharmacokinetic parameters modeled were: P/T: 8hrly dosing, t1/2=1hour, and T>MIC ranging from 10% to 100%; gentamicin (a) static gentamicin at 0.5mg/l, and (b) 24hrly dosing, Cmax=12mg/l; t1/2=1hour.
ABE was determined by serial quantitive culture on antibiotic free media. The minimum level of detection was 10(2) CFU/ml. EoR was determined by quantitive culture on agar containing 4 and 16 times the MIC of gentamicin or P/T, at 24, 48 and 72 hours.

Results. Time kill demonstrated that where P/T (196mg/l) was used in combination with gentamicin (0.5mg/l), ABE at 24hours was increased by >2 log(10) compared to the most potent agent alone. High level resistance to P/T (16 x MIC) emerged by 48hours in the absence, but not in the presence of gentamicin.
In the dynamic models the addition of gentamicin markedly decreases the P/T %T>MIC required to achieve static effect and 3 log(10) kill. The EoR seen with P/T alone, followed the classic bell shaped pattern, with maximal evolution of resistance at moderate %T>MIC, and less at very low and very high %T>MIC. By 72 hours only where %T>MIC = 10%, or %T>MIC = 100%, was resistance undetectable. In contrast, in the models with the addition of gentamicin, no isolates resistant to P/T at 4xMIC or 16xMIC were detected.


Conclusion. In treatment of P. aeruginosa clinically, it can be difficult to achieve the T>MIC targets for B-lactam antibiotics, particularly where the isolate has a raised MIC. We show that in vitro, addition of gentamicin markedly reduces the target T>MIC for P/T against a clinical isolate, and that gentamicin protects against emergence of resistance against P/T regardless of the P/T T>MIC achieved. These effects are robust to changes in the gentamicin dosing regimen modeled and occur at levels of gentamicin exposure below that usually considered to be nephrotoxic in humans.