Key points are not available for this paper at this time.
Cerebrospinal fluid and serum immunoglobulin G from 1007 patients with suspected neurological disease were analysed by 2 methods: isoelectric focusing for the detection of oligoclonal banding, and quantitative measurement of IgG and albumin for the formulation of a Log IgG-Index. A comparison of the 2 methods in the detection of local synthesis of IgG showed that isoelectric focusing not only gave a much higher yield overall, with 282 patients showing local synthesis versus 225 for the Log IgG-Index, but also a higher specificity, with a false positive rate of 0% versus 3.5%. In addition, of the 282 patients positive by isoelectric focusing only 163 (58%) were positive by the Log IgG-Index. Of the 1007 patients studied, 206 had multiple sclerosis (MS), and isoelectric focusing showed local synthesis in 95% of clinically definite cases, with a 90% detection rate overall. The Log IgG-Index was positive in only 67% of clinically definite cases, with an overall 59% detection rate. Thus with the exceptions noted above, local synthesis of IgG as defined by isoelectric focusing is confined to demyelinating, inflammatory, infectious and postinfectious disorders. Our results compare very favourably with the published sensitivities of magnetic resonance imaging in the detection of abnormalities in multiple sclerosis, and better than those for evoked potentials. Where both these investigations are readily available isoelectric focusing provides a useful adjunct. For the majority of physicians and neurologists who do not have ready access to magnetic resonance imaging, isoelectric focusing is an excellent alternative. We would also recommend that it become the standard for the measurement of IgG abnormalities in the cerebrospinal fluid and that the use of quantitative data be abandoned for routine purposes.
McLean et al. (Mon,) studied this question.