Improved Sensitivity of CSF Oligoclonal Bands in the Diagnosis of Multiple Sclerosis

2004, Volume 15, Number 1


David F. Keren, M.D.

The initial diagnosis of Multiple Sclerosis (MS) is often delayed because of its variable presentation. Clinical suspicion of MS begins when patients develop recurrent neurological symptoms that indicate the presence of anatomic lesions at two or more locations of the central nervous system (CNS). Some of the symptoms are specific such as diplopia or trigeminal neuralgia, whereas others are vague such as weakness and fatigue.

The anatomic lesions that cause these symptoms are areas of demyelination which contain myelin-reactive T cells and clones of plasma cells that produce autoantibodies against myelin, myelin oligodendrocyte glycoprotein and other, as yet unidentified, neural antigens. These oligoclonal bands found in the CSF, but not in the serum, of patients with MS have been used since the 1960s to support the diagnosis of MS.

Unfortunately, detection of O-bands in the CSF is not specific for MS, even when these bands are absent from the serum. Inflammation, infections, cerebrovascular accidents and structural CNS lesions have also been associated with the presence of CSF O-bands.

Nonetheless, the presence of O-bands in the CSF, that are not in the serum, remains the best single laboratory test to support the clinical diagnosis of MS. Because the methods originally used to detect the O-bands had limited sensitivity, most articles in the medical literature required only two or more bands to be present in the CSF, that were not present in the serum, to confirm the diagnosis of MS. In 1994, the use of isoelectric focusing (IEF) combined with IgG immuno-identification of the bands was shown to be the most sensitive method for the detection of O-bands (Andersson et al.). In the past year, semiautomated methods for this technique have become available.

However, with the improved sensitivity of the method, in order to maintain high specificity, the number of bands required must increase from 2 to 4, as reported by Fortini et al. They found that the new semiautomated IEF-IgG method had 90% sensitivity vs. 60% for their older high-resolution agarose method. The enhanced sensitivity for detecting O-bands gave a significant improvement in the reproducibility of the IEF-IgG method. With agarose, the average positive case contained only 2.2 O-bands (with a requirement of 2 or more bands for a positive result), whereas the IgG-IEF technique had an average of 8.0 bands in a positive case (with a requirement of 4 or more bands for a positive test result). Therefore, the new method provides improvement in the confidence of a positive result. Further, in their study, 17 of the 20 cases of definite MS demonstrated 8 or more bands by the IEF-IgG method.

Our own in-house studies confirm the improved sensitivity and reproducibility of the IEF-IgG method and we have adopted this new method and the requirement of 4 or more O-bands for positive supporting evidence for the diagnosis of MS. Samples with 1-3 O-bands are considered only equivocal supporting evidence.

Although no laboratory test is specific for the diagnosis of MS, the new IEF-IgG method will provide more reliable and sensitive information than previous methods.


  1. Andersson M, et al. Cerebrospinal fluid in the diagnosis of multiple sclerosis: a consensus report. J Neurol Neurosurg Psychiatry. 1994;57:897-902.
  2. Fortini AS et al. Cerebrospinal fluid oligoclonal bands in the diagnosis of multiple sclerosis: isoelectric focusing with IgG immunoblotting compared with high-resolution agarose gel electrophoresis, and cerebrospinal fluid IgG index. Am J Clin Pathol 2003;120:672-675.
  3. Keren, DF. Optimizing detection of oligoclonal bands in cerebrospinal fluid by use of isoelectric focusing with IgG Immunoblotting. Am J Clin Pathol 2003;120:649-651.