Reducing empirical use of fluoroquinolones for Pseudomonas aeruginosa infections improves outcome

Abstract

We previously reported ciprofloxacin resistance (CR) and empirical use of fluoroquinolones as predictors of mortality in patients infected with Pseudomonas aeruginosa in a case-control study. Here, we assessed the clinical impact of reducing empirical fluoroquinolone use for P. aeruginosa infections in hospitalized patients by performing a follow-up study in 2005-06 [period 2 (P2)] and comparing this with prior data from 2001-02 [period 1 (P1)]. Medical charts of infected patients who received at least 72 h of antibiotic therapy were reviewed. Patients were subgrouped based on the susceptibility of infected strains into the CR or ciprofloxacin-susceptible group. Antibiograms, patient and treatment variables and outcome measures were compared between groups and between study periods. Study patients were elderly (median age, 76 years), had a median of three co-morbidities and a median APACHE II score of 13. Most (75%) had pneumonia or urosepsis. Empirical use of fluoroquinolones was reduced by 30% in P2 versus P1, with a corresponding 39% increase in piperacillin/tazobactam use. The resultant positive impact observed in the CR group during P2 includes shortened delay to receipt of effective therapy (1 versus 3.5 days, P < 0.0001), reduced length of stay (13 versus 16 days, P = 0.03) and 2-fold lower mortality (9% versus 22%, P = 0.05). Susceptibility of P. aeruginosa improved by 10% to all antipseudomonal agents tested. In settings where high rates of fluoroquinolone resistance exist, use of non-fluoroquinolone-based empirical regimens for P. aeruginosa infections improves patient outcomes and organism susceptibility over time.