Abstract

Background. Ceftolozane/tazobactam (C/T) efficacy and safety in ventilator-associated pneumonia (VAP) is being evaluated at a double dose by several trials. This dosing is based on a pharmacokinetic (PK) model that demonstrated that 3 g q8h achieved >= 90% probability of target attainment (50% fT > minimal inhibitory concentration [MIC]) in plasma and epithelial lining fluid against C/T-susceptible P. aeruginosa. The aim of this study was to evaluate the efficacy of different C/T doses in patients with lower respiratory infection (LRI) due to MDR- or XDR-P. aeruginosa considering the C/T MIC. Methods. This was a multicenter retrospective study of 90 patients with LRI caused by resistant P. aeruginosa who received a standard or high dose (HDo) of C/T. Univariable and multivariable analyses were performed to identify independent predictors of 30-day mortality. Results. The median age (interquartile range) was 65 (51-74) years. Sixty-three (70%) patients had pneumonia, and 27 (30%) had tracheobronchitis. Thirty-three (36.7%) were ventilator-associated respiratory infections. The median C/T MIC (range) was 2 (0.5-4) mg/L. Fifty-four (60%) patients received HDo. Thirty-day mortality was 27.8% (25/90). Mortality was significantly lower in patients with P. aeruginosa strains with MIC <= 2 mg/L and receiving HDo compared with the groups with the same or higher MIC and dosage (16.2% vs 35.8%; P = .041). Multivariate analysis identified septic shock (P < .001), C/T MIC >2 mg/L (P = .045), and increasing Charlson Comorbidity Index (P = .019) as independent predictors of mortality. Conclusions. The effectiveness of C/T in P. aeruginosa LRI was associated with an MIC <= 2 mg/L, and the lowest mortality was observed when HDo was administered for strains with C/T MIC <= 2 mg/L. HDo was not statistically associated with a better outcome.