The neurotrophic protein S100B has been promoted as a neuromarker for decades, and to reflect the severity of brain injury. On the other hand, S100B is a tumor marker. The interpretation of its serum levels may be altered by a contribution from extracerebral sources and its renal elimination.

In the present study we investigate the relevance of S100B as a prognostic factor, as well as the correlation with different CT classifications in a large cohort of patients with and without brain injury. Furthermore, we examine whether S100B is elevated in brain tumors.

Patients of the Department of Neurosurgery, University of Erlangen-Nuremberg, undergoing surgery for sellar lesions were prospectively included. The study protocol was approved by the local Ethical Committee. Informed written consent was given by the participants or the next-of-kin in each case. Exclusion criteria comprised of those below 18 years of age, pregnancy and a drug intolerance.

Patients baseline information included age, gender, clinical presentation, and preexisting medical conditions. In four different cohorts of patients, S100B was measured:

1. ICU TBI, patients treated on the ICU with brain injury
2. ICU tumor, patients treated on the ICU because of an intracranial tumor
3. ICU surgery, patients treated on the ICU following surgery without brain injury
4. ICU control, patients treated on the ICU without TBI, tumor or surgery

The performed diagnostic tests included blood, cerebrospinal fluid and urine samples. In all subjects, blood (4 mL), cerebrospinal fluid (4mL) and urine (4mL) samples were collected daily as part of the clinical routine at 6:00 AM.

Samples were immediately centrifuged for 10min at 1.300xG and 4°C and stored at a temperature of -80°C until the assays were performed. Analysis was performed with commercially available kits on automated immunoanalyzers (LIAISON® Sangtec®100 by chemiluminescence immunoassay, Diasorin). The sensitivity of the assay was 0.02ng/ml.

In all cohorts the clinical status was documented with Glasgow Coma Score (GCS), and the outcome was assessed applying the Glasgow Outcome Score (GOS) or the Karnofsky Status Scale. The radiological work-up included a computed tomography (CT) or a magnetic resonance imaging (MRI). In brain injured patients, the CT was classifiied by the Marshall and Rotterdam Score. Extracranial injuries or surgical procedures were documented

Statistical analysis was performed with SPSS, and p<0.05 was accepted as significant.

Inclusion Criteria:

• patients of the Department of Neurosurgery, University of Erlangen Nürnberg, patients with traumatic brain injury,
• patients with intracranial tumor,
• intensive care patients,
• informed written consent was given by the Patient or the next of kin in each case.

Exclusion Criteria:

• age under 18
• pregnancy

Patients of the Department of Neurosurgery, University of Erlangen-Nuremberg,