Addressing resistance evolution in Pseudomonas aeruginosa using pharmacodynamic modelling: application to meropenem dosage and combination therapy.

Isolates of Pseudomonas aeruginosa (n = 208) were collected from an 810-bed hospital in Connecticut, USA. A model employing the pharmacokinetic properties of meropenem, susceptibility results and Monte Carlo simulation was used to analyse four different dosing regimens of meropenem at pharmacodynamic endpoints. Cumulative fraction of response (CFR) was assessed at bacteriostatic and bactericidal endpoints for the entire population of isolates, as well as for isolates from principal anatomical sites. CFR was also evaluated at endpoints shown to suppress emergence of resistance in 'susceptible'P. aeruginosa with either monotherapy or combination therapy. The bacteriostatic/bactericidal CFR of meropenem 1 g every 8 h (q8h), 2 g q8h, 1 g q8h infused over 3 h (3-h INF), and 2 g q8h 3-h INF were 76%/73%, 80%/76%, 77%/75% and 79%/78%, respectively. At the monotherapeutic suppressive endpoint, CFRs against susceptible isolates were 21%, 35%, 32% and 50%, respectively. When combination therapy with an aminoglycoside was simulated, the CFRs for the same regimens were 50%, 64%, 65% and 79%, respectively. Bactericidal CFRs for all regimens against wound isolates were significantly higher (p <0.03 for each regimen) than CFRs for the entire population. Meropenem 2 g q8h with a 3-h infusion in combination with an aminoglycoside provides the greatest likelihood of P. aeruginosa coverage, and may help to prevent development of resistance, although local MIC data are essential to inform therapeutic decisions.