Objective: The purpose of this study was to determine the impact of scheduled changes of antibiotic classes, used for the empirical treatment of suspected or documented Gram-negative bacterial infections, on the occurrence of inadequate antimicrobial treatment of nosocomial infections.
Design: Prospective observational study.
Setting: Medical (19-bed) and surgical (18-bed) intensive care units in an urban teaching hospital.
Patients: A total of 3,668 patients requiring intensive care unit admission were prospectively evaluated during three consecutive time periods.
Interventions: During each time period, one antibiotic class was selected for the empirical treatment of Gram-negative bacterial infections as follows: time period 1 (baseline period) (1,323 patients), ceftazidime; time period 2 (1,243 patients), ciprofloxacin; and time period 3 (1,102 patients), cefepime.
Measurements and main results: The overall administration of inadequate antimicrobial treatment for nosocomial infections decreased during the course of the study (6.1%, 4.7%, and 4.5%; p = .15). This was primarily because of a statistically significant decrease in the administration of inadequate antibiotic treatment for Gram-negative bacterial infections (4.4%, 2.1%, and 1.6%; p < .001). There were no statistically significant differences in the overall hospital mortality rate among the three time periods (15.6%, 16.4%, and 16.2%; p = .828) despite a significant increase in severity of illness as measured with Acute Physiology and Chronic Health Evaluation (APACHE) II scores (15.3 +/- 7.6, 15.7 +/- 8.0, and 20.7 +/- 8.6; p < .001). The hospital mortality rate decreased significantly during time period 3 (20.6%) compared with time period 1 (28.4%; p < .001) and time period 2 (29.5%; p < .001) for patients with an APACHE II score > or = 15.
Conclusions: These data suggest that scheduled changes of antibiotic classes for the empirical treatment of Gram-negative bacterial infections can reduce the occurrence of inadequate antibiotic treatment for nosocomial infections. Reducing inadequate antibiotic administration may improve the outcomes of critically ill patients with APACHE II scores > or = 15.