This study was designed to evaluate the pharmacodynamics of doripenem, imipenem and meropenem as a predictor of clinical success, mortality, 28 day recurrence and development of resistance in patients treated for Pseudomonas aeruginosa ventilator-associated pneumonia (VAP).Patients and methods
Previously published demographic and outcome data derived from patients treated for P. aeruginosa VAP with doripenem, imipenem or meropenem were utilized. Patient-specific data were used in conjunction with published population pharmacokinetic models to construct concentration–time profiles for each patient. Etest MICs were used to determine pharmacodynamic profiles. Classification and regression tree (CART) analysis was utilized to partition pharmacodynamics based on outcomes with P values of 0.05.Results
Eighty-six patients were included in the analysis. Initial carbapenem MICs ranged from 0.03 to 32 mg/L. VAP recurred in 28 patients; of these, 17 patients were initially infected with susceptible organisms, and 14 of them developed resistance. CART fT>MIC partitions identified for clinical success and survival were 19.2% (P = 0.016) and 47.9% (P < 0.001), respectively. No statistically significant partitions for fT>MIC were identified for recurrence or resistance development.Conclusions
We identified fT>MIC cut-offs for positive clinical outcomes in patients with P. aeruginosa VAP that were similar to those observed in animal models of infection for stasis (∼20%) and 1 log decreases in cfu (∼40%). Although in vitro studies have suggested a link between drug exposure and development of resistance, we were unable to identify such a relationship clinically.