K.J.B. Lamers

The Committee of the European Concerted Action for Multiple Sclerosis (Charcot Foundation) organised five workshops to discuss CSF analytical standards in the diagnosis of multiple sclerosis. This consensus report from 12 European countries summarises the results of those workshops. It is hoped that neurologists will confer with their colleagues in clinical chemistry to arrange the best possible local practice. The most sensitive method for the detection of oligoclonal immunoglobulin bands is isoelectric focusing. The same amounts of IgG in parallel CSF and serum samples are used and oligoclonal bands are revealed with IgG specific antibody staining. All laboratories performing isoelectric focusing should check their technique at least annually using "blind" standards for the five different CSF and serum patterns. Quantitative measurements of IgG production in the CNS are less sensitive than isoelectric focusing. The preferred method for detection of blood-CSF barrier dysfunction is the albumin quotient. The CSF albumin or total protein concentrations are less satisfactory. These results must be interpreted with reference to the age of the patient and the local method of determination. Cells should be counted. The normal value is no more than 4 cells/microliters. Among evolving optional tests, measurement of the combined local synthesis of antibodies against measles, rubella, and/or varicella zoster could represent a significant advance if it offers higher specificity (not sensitivity) for identifying chronic rather than acute inflammation. Other tests that may have useful correlations with clinical indices include those for oligoclonal free light chains, IgM, IgA, or myelin basic protein concentrations.

OBJECTIVE: To study the ability of glial (glial fibrillary acidic protein [GFAP] and S100b) and neuronal (neuron specific enolase [NSE]) protein levels in peripheral blood to predict outcome after severe traumatic brain injury. METHODS: Eighty-five patients with severe traumatic brain injury (admission Glasgow Coma Score [GCS] < or = 8) were included. Blood samples taken at the time of hospital admission were analyzed for S100b, GFAP, and NSE. Data collected included demographic and clinical variables. Outcome was assessed using the Glasgow Outcome Scale (GOS) at 6 months post injury. RESULTS: The median serum levels of S100b, GFAP, and NSE were raised 18.3 fold (S100b), 4.6 fold (GFAP), and twofold (NSE) compared to normal reference values. S100b, GFAP, and NSE serum levels correlated significantly with the injury severity score and CT findings but not with age, sex, or GCS. S100b, GFAP, and NSE levels were significantly higher in patients who died or had a poor outcome 6 months post injury than in those who were alive or had good outcome. S100b level >1.13 microg/L was the strongest predictor of death with 100% discrimination, but GFAP (>1.5 microg/L) and NSE (>21.7 microg/L) levels also strongly predicted death (adjusted odds ratios 5.82 [for GFAP] and 3.91 [for NSE]). S100b, GFAP, and NSE all strongly predicted poor outcome (adjusted odds ratios 5.12 [S100b], 8.82 [GFAP], and 3.95 [NSE]). CONCLUSIONS: These results suggest that determination of serum levels of glial and neuronal proteins may add to the clinical assessment of the primary damage and prediction of outcome after severe traumatic brain injury.

BACKGROUND: Biomarker levels in blood after traumatic brain injury (TBI) may offer diagnostic and prognostic tools in addition to clinical indices. This study aims to validate glial fibrillary acidic protein (GFAP) and S100B concentrations in blood as outcome predictors of TBI using cutoff levels of 1.5 μg/L for GFAP and 1.13 μg/L for S100B from a previous study. METHODS: In 79 patients with TBI (Glasgow Coma Scale score [GCS] ≤12), serum, taken at hospital admission, was analyzed for GFAP and S100B. Data collected included injury mechanism, age, gender, mass lesion on CT, GCS, pupillary reactions, Injury Severity Score (ISS), presence of hypoxia, and hypotension. Outcome was assessed, using the Glasgow Outcome Scale Extended (dichotomized in death vs alive and unfavorable vs favorable), 6 months post injury. RESULTS: In patients who died compared to alive patients, median serum levels were increased: GFAP 33.4-fold and S100B 2.1-fold. In unfavorable compared to favorable outcome, GFAP was increased 19.8-fold and S100B 2.1-fold. Univariate logistic regression analysis revealed that mass lesion, GFAP, absent pupils, age, and ISS, but not GCS, hypotension, or hypoxia, predicted death and unfavorable outcome. Multivariable analysis showed that models containing mass lesion, pupils, GFAP, and S100B were the strongest in predicting death and unfavorable outcome. S100B was the strongest single predictor of unfavorable outcome with 100% discrimination. CONCLUSION: This study confirms that GFAP and S100B levels in serum are adjuncts to the assessment of brain damage after TBI and may enhance prognostication when combined with clinical variables.

In this study levels of neuron-specific enolase (NSE), S-100 protein (S-100) and myelin basic protein (MBP) in cerebrospinal fluid (CSF) of children and adults with distinct neurological disorders were examined. A previous study from our department demonstrated age related reference values for these brain-specific proteins in CSF. The median concentration level of the 3 proteins in 17 different neurological disease groups versus the reference group was compared. Significantly higher MBP values were observed in patients with multiple sclerosis (MS), cerebrovascular accident (CVA), metabolic disorder and infection. Furthermore, significantly higher values were demonstrated for S-100 in CVA and for NSE in metabolic diseases. In CVA, the NSE and S-100 values were significantly related with MBP values, whereas in MS the NSE and S-100 were not related with MBP values.