OBJECTIVEThe measurement of neuromarker/neuroproteins in the cerebrospinal fluid (CSF) is gaining increased popularity. However, insufficient information is available on the rostrocaudal distribution of neuroproteins in the CSF to guarantee an appropriate interpretation of ventricular versus lumbar concentrations.METHODSIn 10 patients treated with both an external ventricular and a lumbar CSF drain, we collected concomitant CSF samples. We measured CSF concentrations of the glial S100B protein, the neuron-specific enolase (Cobas e411®; Roche Diagnostics), the leptomeningeal β-trace protein (BN Pro Spec®; Dade Behring/Siemens), and the blood-derived albumin (Immage; Beckman Coulter). Statistical analysis was performed with a paired Wilcoxon signed ranks test.RESULTSIn patients with a free CSF circulation without any recent neurosurgical procedure, S100B and neuron-specific enolase concentrations did not differ between the ventricular and lumbar CSF while β-trace and albumin levels were significantly higher in the lumbar than in the ventricular CSF (p=0.008 and p=0.005). Following posterior fossa tumor surgery, all proteins accumulate in the lumbar CSF.CONCLUSIONFor brain-derived proteins, we could not confirm a rostrocaudal CSF gradient while lepto-meningeal and blood-derived proteins accumulate in the lumbar CSF. We conclude that for the interpretation of protein CSF concentrations, the source of the sample is of crucial importance.