Do we still need OCBs in MS diagnosis and how many?

BackgroundA standard cut-off of ≥2 CSF-specific OCBs is used in multiple sclerosis but some studies suggest ≥3 CSF OCBs maybe more specific.AimTo determine if a cut-off of ≥2 or ≥3 CSF OCBs should be used in MSMaterials and MethodsOligoclonal bands were tested using standardised methods incorporating iso- electric focussing. Samples sent between July 2018-June 2019 were included. CSF OCB number (1, ≥2,≥3, or unclear) was assessed by two blinded raters and case records were reviewed.ResultsCSF OCBs were detected in 196/746 (26%) samples (CSF OCB=1; 14/196 (7%), ≥2=182/196 (93%), ≥3=169/196 (86%). Clinical data were available in 157 patients (MS=101, other inflammatory=37, unclear significance=19). Excluding CSF OCB results, 68/101 (67%) cases fulfilled MS diagnostic criteria. Of patients with 1, ≥2 and ≥3 CSF OCBs, 1/8 (12%), 100/182 (55%) and 98/169 (58%) had MS respectively. The positive predictive value of 1, ≥2, and ≥3 OCBs was 0.35, 0.89 and 0.92 respectively in CNS inflammatory disease.ConclusionDetection of CSF OCBs was necessary to fulfil MS diagnostic criteria in approximately 30% of cases. Detection of ≥3 OCBs marginally improved the likelihood of an MS diagnosis. Overall measures of sensitivity, specificity and receiver operator curves will be presented.giovanna.campagna@thewaltoncentre.nhs.uk