Impact of intracranial pressure monitor prophylaxis on central nervous system infections and bacterial multi-drug resistance.

BACKGROUND

Routine intracranial pressure monitor (ICP) prophylaxis is not practiced at our institution. Nevertheless, some patients receive de facto prophylaxis as a result of the use of antibiotics for injuries such as open or facial fractures. We tested the hypothesis that prophylactic antibiotics do not reduce the incidence of central nervous system (CNS) infections but instead are associated with the acquisition of multi-drug resistant (MDR) bacterial infections.

METHODS

Patients admitted to the trauma intensive care unit (TICU) from January, 2001 through December, 2004 with blunt, non-operative traumatic brain injury who were managed solely with an ICP monitor were identified from our trauma registry and divided into two groups: (1) Those receiving no antibiotics prior to or during ICP monitoring (NONE; n = 71); and (2) those already receiving antibiotics at the time of ICP monitor insertion (PRO; n = 84). Groups were stratified on the basis of age, Injury Severity Score (ISS), Glasgow Coma Scale (GCS) Score, base excess (BE), ICP days, transfusions in 24 h, ICU days, ventilator days, head Abbreviated Injury Score (AIS), and chest AIS. The study groups did not differ with respect to age, ISS, GCS, BE, ICP days, 24-h transfusions, ICU days, ventilator days, head AIS, or length of stay. In all, 183 patients were identified, of whom 28 died within seven days and were excluded from the analysis. All patients were followed until discharge for both CNS infections and subsequent infectious complications.

RESULTS

Only two patients, both in the PRO group, developed CNS infection. Both infectious complications (0.7 vs 1.4 per patient; p < 0.05) and infections secondary to MDR pathogens (0.03 vs. 0.33 per patient; p < 0.01) were significantly more common in the PRO group. Twenty-nine percent of the ventilator-associated pneumonias and 33% of the blood stream infections in the PRO group were MDR, whereas only two blood stream infections in the NONE group (4% of the total infections) were MDR.

CONCLUSIONS

The routine use of prophylactic antibiotics for ICP monitor insertion is not warranted. This practice does not reduce the CNS infection rate and is associated with more MDR pathogens in any subsequent infectious complications.