Abstract

Incidence of both primary and secondary central nervous system (CNS) malignancies are increasing, yet detection and diagnosis of which remains challenging. Neurological examination, CSF cytology and neuroimaging each offer diagnostic value, however, individually, each method has relatively low sensitivity. CSF cytology is often routinely requested as part of the diagnostic work-up. Therefore, we reviewed the value of CSF cytology in predicting CNS neoplastic process.

We systematically analysed 296 patients who had had CSF cytology. All consecutive patients over the last 2 years were included. The 12 patients positive for CSF cytology were also found to have concurrent; clinical findings (92%), radiological findings (92%), or both (83%). In no patient CNS malignancy was identified by CSF cytology alone. Additionally, only 7 of the 12 CSF positive cases were collected by means of lumbar puncture, the rest were carried out during neurosurgical procedures.

We conclude that if there is neither clinical findings nor abnormal neuroimaging, assessing CSF cytology does not contribute towards diagnosis of CNS malignancy. Yield of CSF cytology is high when both clinical and radiological evidence are present. Where neuroimaging is normal but suspicion remains, the low sensitivity of the CSF-cytology would make it a weak corroborator to exclude malignancy.