Selection of resistance during sequential use of preferential antibiotic classes

OBJECTIVE: To determine the effect of antibiotic class pressure on the susceptibility of bacteria during sequential periods of antibiotic homogeneity. DESIGN AND SETTING: Prospective study in a mixed ICU with three separated subunits of eight, eight, and ten beds. PATIENTS AND PARTICIPANTS: The study examined the 1,721 patients with a length of stay longer than 2 days. INTERVENTIONS: Three different antibiotic regimens were used sequentially over 2 years as first-choice empirical treatment: cephalosporins, fluoroquinolone, or a penicillin-beta-lactamase inhibitor combination. Each regimen was applied for 8 months in each subunits of the ICU, using "latin square" design. RESULTS: We treated 731 infections in 546 patients (32% of patients staying more than 48 h). There were 25.5 ICU-acquired infections per 1,000 patient-days. Infecting pathogens and colonizing bacteria were found in 2,739 samples from 1,666 patients (96.8%). No significant change in global antibiotic susceptibility was observed over time. However, a decrease in the susceptibility of several species was observed for antibiotics used as the first-line therapy in the unit. Selection pressure of antibiotics and occurrence of resistance during treatment was documented within an 8-month rotation period. CONCLUSIONS: Antibiotic use for periods of several months induces bacterial resistance in common pathogens