CSF inflammatory markers differ in gram-positive versus gram-negative shunt infections

Gwenn L. SkarEmail author View ORCID ID profile,
David Synhorst,
Matthew Beaver and
Jessica N. Snowden

Journal of Neuroinflammation201916:7

https://doi.org/10.1186/s12974-019-1395-6

Received: 31 August 2018
Accepted: 2 January 2019
Published: 9 January 2019

Abstract

Background

Cerebrospinal fluid (CSF) shunt placement is frequently complicated by bacterial infection. Shunt infection diagnosis relies on bacterial culture of CSF which can often produce false-negative results. Negative cultures present a conundrum for physicians as they are left to rely on other CSF indices, which can be unremarkable. New methods are needed to swiftly and accurately diagnose shunt infections. CSF chemokines and cytokines may prove useful as diagnostic biomarkers. The objective of this study was to evaluate the potential of systemic and CSF biomarkers for identification of CSF shunt infection.

Methods

We conducted a retrospective chart review of children with culture-confirmed CSF shunt infection at Children’s Hospital and Medical Center from July 2013 to December 2015. CSF cytokine analysis was performed for those patients with CSF in frozen storage from the same sample that was used for diagnostic culture.

Results

A total of 12 infections were included in this study. Patients with shunt infection had a median C-reactive protein (CRP) of 18.25 mg/dL. Median peripheral white blood cell count was 15.53 × 103 cells/mL. Those with shunt infection had a median CSF WBC of 332 cells/mL, median CSF protein level of 406 mg/dL, and median CSF glucose of 35.5 mg/dL. An interesting trend was observed with gram-positive infections having higher levels of the anti-inflammatory cytokine interleukin (IL)-10 as well as IL-17A and vascular endothelial growth factor (VEGF) compared to gram-negative infections, although these differences did not reach statistical significance. Conversely, gram-negative infections displayed higher levels of the pro-inflammatory cytokines IL-1β, fractalkine (CX3CL1), chemokine ligand 2 (CCL2), and chemokine ligand 3 (CCL3), although again these were not significantly different. CSF from gram-positive and gram-negative shunt infections had similar levels of interferon gamma (INF-γ), tumor necrosis factor alpha (TNF-α), IL-6, and IL-8.

Conclusions

This pilot study is the first to characterize the CSF cytokine profile in patients with CSF shunt infection and supports the distinction of chemokine and cytokine profiles between gram-negative and gram-positive infections. Additionally, it demonstrates the potential of CSF chemokines and cytokines as biomarkers for the diagnosis of shunt infection.