gms | German Medical Science

Objective: The astrocyte-derived protein S100B has been proposed as a biomarker predicting the severity of brain injury after different types of neurotrauma. Beside S100B measurements in blood, a positive correlation between S100B levels in cerebrospinal fluid (CSF) and worse outcome following traumatic brain injury (TBI), intracerebral hemorrhage or stroke has been reported. However, while CSF levels may reflect the S100B release into the brain extracellular fluid more accurately than blood levels, little information is available on the distribution of S100B in CSF – although a concentration gradient of 3.5:1 from ventricular to lumbar has been proposed in the literature.

Methods: We analyzed concomitantly ventricular and lumbar S100B CSF concentrations (Cobas e411^® electrochemiluminescence assay, Roche Diagnostics) in 10 neurosurgical patients who were simultaneously treated with an external ventricular and a lumbar CSF drainage and calculated the ventricular/lumbar ratio (V/L-ratio). Seven patients suffered from a communicating hydrocephalus following aneurysmal subarachnoid hemorrhage (SAH, n=3), TBI (n=2) or meningitis (n=2), and 3 patients suffered from an occlusive hydrocephalus following posterior fossa surgery.

Results: The V/L-ratios of S100B ranged from 0.97 to 1.28 following meningitis, from 0.80 to 1.22 in SAH and from 0.95 to 1.53 in TBI. Following posterior fossa tumor surgery, the S100B V/L-ratios were significantly lower (0.45, 0.02 and 0.08, p<0.001).

Conclusions: The ventricular to lumbar CSF concentration gradient of S100B depends on the underlying pathology. As long as the CSF distribution is free, we could not confirm any V/L concentration gradient. Infratentorial lesions result in higher lumbar S100B concentrations in comparison with ventricular levels and consequently to a decreased V/L-ratio. We conclude that for the interpretation of S100B CSF levels, both the knowledge of the underlying pathology and the source of CSF sample are of crucial importance.