Surveillance and correlation of antimicrobial usage and resistance of Pseudomonas aeruginosa: a hospital population-based study.

This retrospective study evaluated trends and association between resistance of Pseudomonas aeruginosa isolated from patients with hospital-acquired infections (HAIs) and hospital antimicrobial usage from 2003 through 2011 in a tertiary care hospital in northeast China. HAI was defined as occurrence of infection after hospital admission, without evidence that infection was present or incubating (≦48 h) on admission. In vitro susceptibilities were determined by disk diffusion test and susceptibility profiles were determined using zone diameter interpretive criteria, as recommended by Clinical and Laboratory Standards Institute (CLSI). Data on usage of various antimicrobial agents, expressed as defined daily dose (DDD) per 1,000 patients-days developed by WHO Anatomical Therapeutical Chemical (ATC)/DDD index 2011, were collected from hospital pharmacy computer database. Most of 747 strains of P. aeruginosa were collected from respiratory samples (201 isolates, 26.9%), blood (179, 24.0%), secretions and pus (145, 19.4%) over the years. Time series analysis demonstrated a significant increase in resistance rates of P. aeruginosa to ticarcillin/clavulanic acid, piperacillin/tazobactam, cefoperazone/sulbactam, piperacillin, imipenem, meropenem, ceftazidime, cefepime, ciprofloxacin, and levofloxacin except aminoglycosides over time in the hospital (P<0.001). The rates of carbapenem-resistant P. aeruginosa (CRPA) isolated from patients with HAIs were 14.3%, 17.1%, 21.1%, 24.6%, 37.0%, 48.8%, 56.4%, 51.2%, and 54.1% over time. A significant increase in usage of anti-pseudomonal carbapenems (P<0.001) was seen. ARIMA models demonstrated that anti-pseudomonal carbapenems usage was strongly correlated with the prevalence of imipenem and meropenem-resistant P. aeruginosa (P<0.001). Increasing of quarterly CRPA was strongly correlated at one time lag with quarterly use of anti-pseudomonal carbapenems (P<0.001). Our data demonstrated positive correlation between anti-pseudomonal antimicrobial usage and P. aeruginosa resistance to several classes of antibiotics, but not all antimicrobial agents in the hospital.