Emergence of highly antibiotic-resistant Pseudomonas aeruginosa in relation to duration of empirical antipseudomonal antibiotic treatment.

Objective: This study examines antibiotic resistance in Pseudomonas aeruginosa in hospitalized patients in relation to prior empirical antibiotic therapy.

Design: Two retrospective case analyses comparing patients who manifested P aeruginosa with differing patterns of antibiotic resistance.

Setting and participants: Patients acquiring P aeruginosa in a community hospital.

Measures: Patients were compared on duration of hospitalization and days and doses of antibiotics prior to recovery of P aeruginosa. Patients were grouped, based on susceptibility patterns of their P aeruginosa isolates classified as follows: (1) fully susceptible (susceptible to all classes of antipseudomonal antibiotics [SPA]), (2) multidrug-resistant (resistant to two classes of antipseudomonal antibiotics [MDRPA]), or (3) highly drug-resistant (resistant to > or = 6 classes of antipseudomonal antibiotics [HRPA]). To control for duration of hospitalization, antibiotic treatments of HRPA and SPA patients were compared during the first 21 days of care.

Results: Prior to recovery of HRPA, six HRPA patients received greater amounts of antibiotics, both antipseudomonal and non-antipseudomonal, than did six SPA patients prior to recovery of SPA. For 14 patients with hospital-acquired SPA who later manifested MDRPA, duration and dosage of antipseudomonal antibiotics, but not all antibiotics, were significantly higher for the SPA-to-MDRPA interval than for the preceding admission-to-SPA interval. The median duration of antipseudomonal antibiotic treatment prior to the recovery of P aeruginosa was 0 days for SPA, 11 days for MDRPA, and 24 days for HRPA.

Conclusion: Duration of empirical antipseudomonal antibiotic treatment influences selection of resistant strains of P aeruginosa; the longer the duration, the broader the pattern of resistance.