Ventriculostomy-Related Infections: Reduced Incidence by Terminating Cerebrospinal Fluid Screening

Abstract

The external ventricular drain (EVD) is an important instrument in managing patients in neurointensive care (NICU). A frequent and sometimes severe complication is ventriculostomy-related infection (VRI). This study aimed to evaluate the effects of biweekly VRI screening by CSF sampling on the clinical course and rate of VRI. Patients with implanted EVDs were retrospectively identified and a cohort screened twice per week was compared with a cohort sampled only on clinical indication. VRI was defined either as a suspected case or a culture-confirmed case. Length of stay in the NICU and CSF shunt dependency were used as outcome parameters. A total of 562 patients were included. The overall proportion of patients treated for VRI was 22% and was not affected by screening (odds ratio [OR], 0.90; 95% confidence interval [CI], 0.57-1.54). Screening was associated with a higher proportion of patients with culture-confirmed VRI (OR, 2.86; 95% CI, 1.36-6.73). The main risk factor for VRI was the number of days with EVD. Positive bacterial culture was associated with both a longer time in the NICU (β= 3.6; P < 0.01) and higher risk for shunt surgery (OR, 1.95; 95% CI, 1.04-3.64). Screening was associated with a higher number of culture-confirmed VRI cases. In addition, screening did not detect VRI at an earlier stage and was not associated with a reduction in the rate of permanent hydrocephalus. No clinical benefit of screening was found. Frequent CSF sampling might contribute to infection. Based on these findings, CSF screening for VRI cannot be recommended.