Continuous antibiotic prophylaxis and cerebral spinal fluid infection in patients with intracranial pressure monitors.

INTRODUCTION

Inconsistencies in the recommendation of prophylactic antibiotics for patients with intracranial pressure monitors compelled us to assess the effect of our standard regimen of continuous antibiotic prophylaxis on cerebrospinal fluid infection. We examined the rate, possible risk factors, causative organisms, and characteristics of infection.

METHODS

Three hundred eleven patients admitted between September 1998 and February 2001 with an intracranial pressure monitoring device in place were included. Two hundred eleven patients received a ventriculostomy, 95 an intraparenchymal fiber optic intracranial pressure monitor (ICPM), and 5 both an ICPM and a ventriculostomy.

RESULTS

The overall infection rate was 5.5% (17/311). No patient with an ICPM developed CSF infection. The infection rate among ventriculostomy patients was 8.1% (17/211). The majority of infections (82%) were caused by Gram-positive species. Younger age (OR=1.04 for each year, 95% CI=1.01-1.08, p=0.03) and increasing duration of ventriculostomy insertion (OR=1.2 for each day of catheter insertion, 95% CI=1.1-1.3, p<0.001) were risk factors for CSF infection in multivariate analysis. Infected patients experienced longer lengths of stay in the NICU (p<0.001) and hospital (p<0.001); however, infection did not impact clinical outcome, as measured by mortality and discharge GCS.

CONCLUSION

ICP monitors have a low overall infection rate. When infection occurs, gram positive organisms predominate. For patients with ventriculostomy, duration of catheter insertion strongly predicts infection, but did not alter in-hospital mortality.