Influence of broad-spectrum antibiotic prophylaxis on intracranial pressure monitor infections and subsequent infectious complications in head-injured patients.

BACKGROUND: The benefit of antibiotic prophylaxis for intracranial pressure (ICP) monitors remains controversial, and clinical practice varies widely. Whether any antibiotic coverage, particularly broad-spectrum coverage, reduces monitor-related infections remains unproved, and exposure to antibiotics may affect the susceptibility patterns of pathogens producing subsequent infectious complications. Despite the lack of data supporting its use, our level I trauma center had a long-standing ICP monitor prophylaxis protocol that provided broad-spectrum coverage that included ceftriaxone. In April 2002, a protocol change was instituted that substituted cefazolin for ceftriaxone as single-agent prophylaxis for ICP monitors. HYPOTHESIS: Broader-spectrum antibiotic prophylaxis does not reduce ICP monitor-related infections but is associated with acquisition of more drug-resistant infections than narrow-spectrum prophylaxis.
METHODS: To evaluate the influence of broad- versus narrow-spectrum prophylaxis, a three year period encompassing each practice was selected. All injured patients with ICP monitors placed between January 1, 2001, and December 31, 2003 (n = 279), were identified using the Vanderbilt trauma database. Antibiotic prophylaxis for ICP monitors was determined using the hospital financial database to identify all antibiotics given to individual patients and subsequent chart review to identify those antibiotics given solely for ICP prophylaxis. A total of 119 patients received narrow-spectrum (either cefazolin or vancomycin; n = 100) or no (n = 19) prophylaxis, whereas 160 received broad-spectrum prophylaxis (ceftriaxone or ciprofloxacin). The two groups did not differ with respect to baseline demographics, type of ICP monitor, or duration of monitor placement. Infectious complications were determined by continuous infection surveillance utilizing standard U.S. Centers for Disease Control and Prevention National Nosocomial Infection Surveillance System (CDC-NNIS) definitions and maintained in a contemporary database. The influence of broad-spectrum antibiotic prophylaxis on both ICP monitor infections and subsequent infections outside the central nervous system (CNS) was determined.
RESULTS: Nine patients (3.2%) developed CNS infections; two of 119 patients (1.7%) who received narrow-spectrum or no prophylaxis versus seven of 160 patients (4.4%) who received broad-spectrum prophylaxis (p = NS). Only the duration of monitor placement and Injury Severity Score were associated with the infection rate. In the total population, 185 infections occurred in 93 patients (33%). Infection rates did not differ between patients who received narrow-spectrum or no prophylaxis (32%) and those who received broad-spectrum prophylaxis (34%). However, patients who received broad-spectrum prophylaxis acquired gram-negative infections with significantly greater antibiotic resistance.
CONCLUSIONS: Broad-spectrum antibiotic prophylaxis of ICP monitors does not reduce CNS infections, but is associated with a shift to resistant gram-negative pathogens in subsequent infectious complications. Thus, broad-spectrum antibiotic prophylaxis of ICP monitors should be eliminated or minimized unless data from randomized trials prove its utility.