Why CSF is still relevant in MS


Int J Mol Sci. 2017 Sep 27;18(10). pii: E2061. doi: 10.3390/ijms18102061.

Clinically Isolated Syndrome According to McDonald 2010: Intrathecal IgG Synthesis Still Predictive for Conversion to Multiple Sclerosis.

Schwenkenbecher P, Sarikidi A, Bönig L, Wurster U, Bronzlik P, Sühs KW, Pul R, Stangel M, Skripuletz T.


While the revised McDonald criteria of 2010 allow for the diagnosis of multiple sclerosis (MS) in an earlier stage, there is still a need to identify the risk factors for conversion to MS in patients with clinically isolated syndrome (CIS). Since the latest McDonald criteria were established, the prognostic role of cerebrospinal fluid (CSF) and visual evoked potentials (VEP) in CIS patients is still poorly defined. We conducted a monocentric investigation including patients with CIS in the time from 2010 to 2015. Follow-ups of 120 patients revealed that 42% converted to MS. CIS patients with positive oligoclonal bands (OCB) were more than twice as likely to convert to MS as OCB negative patients (hazard ratio = 2.6). The probability to develop MS was even higher when a quantitative intrathecal IgG synthesis was detected (hazard ratio = 3.8). In patients with OCB, VEP did not add further information concerning the conversion rate to MS. In patients with optic neuritis and negative OCB, a significantly higher rate converted to MS when VEP were delayed. In conclusion, the detection of an intrathecal IgG synthesis increases the conversion probability to MS. Pathological VEP can help to predict the conversion rate to MS in patients with optic neuritis without an intrathecal IgG synthesis.

I’ve often asked myself the question if Ian McDonald (of the McDonald criteria) hadn’t been who he was (undoubtedly a leader in the field of MS) or an MRI aficionado, where would the diagnostic criteria for MS be now? This is not in any way a criticism of a great mind and visionary – I met Ian McDonald in Queen Square during my PhD, and he had an aura that simply shimmered.

Prominent leaders inspire great change; often for the betterment of society as a whole, but how many of these decisions are directly or indirectly based on intrinsic satisfaction? 

One of the earlier derivations of the MS diagnostic criteria was the Poser criteria (see above), introduced in 1983 and had 94% sensitivity for MS versus post-mortem analysis. The criteria took into consideration the oligoclonal band (OCB) test and other paraclinical measures, which were subsequently dropped in the McDonald 2010 criteria. This decision to move forward with MRI as the sole deciding factor in MS diagnosis also has had some unforeseen knock on effects on MS research; for instance, research on MS histology and the CSF largely stagnated, as a result.

Even in the presence of many lesions (> 10 or > 8), magnetic resonance imaging could not accurately rule multiple sclerosis in (likelihood ratio of a positive test result 3.0 and 2.0, respectively)” – Whiting P

Fast-forward more than a decade, and researchers such as Schwenkenbecher are re-evaluating this stance. They found that in those presenting with the initial demyelinating event (CIS), OCB and VEP both contributed separately to increasing the likelihood of recurrent demyelinating events. The usefulness of OCB test is demonstrated pictographically below. Conversion to MS was higher in those who were OCB+ (55%) than those who were OCB- (21%), and OCB+ individuals were twice as likely to convert to MS. Similarly, IgG synthesis (i.e. presence of antibodies) in the CSF was found in 66% of those who went on to develop MS. Now, if you look at the presence of immune cell activity in the CSF, it was present in 59% who later developed MS. Moreover, if you combined both baseline MRI and OCB findings together the sensitivity of the pick up for the future development of  MS increased to 65%. When it comes to abnormal findings on VEP following optic neuritis, in the absence of OCB positivity there was still an additional predictive value of VEP as far as conversion to MS at a future date was concerned.

We therefore stand a decade on, expectantly for those who develop the MS diagnostic criteria to take note of this and similar findings. In reality, I’m yet to see this happen. In the US, lumbar punctures are hardly preformed and yet we face a new era of B cell therapies – ocrelizumab, in particular, selected cases based on OCB positivity. The irony of this is that those who designed the study understood the value of having a positive OCB result.

About the author

Neuro Doc Gnanapavan


  • My diagnosis for MS was really only closed when I did the lumbar puncture to check the OCBs, which were positive.
    Because if it were only for the MRI would not close, I only have a spinal (cervical) lesion and a brain in the left parietal…

    I think that if it were not for OCBs research, I would still be without a diagnosis, and I was able to have the diagnosis closed within 02 months as soon as I had the onset of the releapse at the time.

    • So this is not something people consider, but the speed at which a diagnosis is made when OCB result is available is important.

    • If the MacDonald criteria 2010 is diagnosis by MRI only, how can this be attributed to Professor W.Ian MacDonald who sadly passed away in 2006? Surely an important decision like this would have been questioned, by neurologists after his death.

    • Their is a previous 2001, 2005 McDonald criteria which was revised in 2010. The 2005 criteria still had CSF.
      The argument was that the 2010 criteria versus the 2005 resulted in faster diagnosis, which is true. But also interestingly the likelihood of including OCB negative cases as MS also increased.
      There is a need for strong voices on diagnostic guideline panels which question the utility of what is constructed. What is included often is dependent on what is published and sadly the research into OCBs has reduced substantially, where there is constant research around MRI CIS accuracy etc . To date I have focused my efforts into introducing neurofilament testing as a prognosticator and part of NEDA criteria in clinical practice. What is translated is pretty much related to exposure.

  • Dear Doc Gnanapavan, I read back old posts trying to understand. What is the average percentage of OCB- that is actually MS? I read in one post that it is around 35% but in another one I read that with newer machines its 5%? Also, has the theory that OCB- has better progression and even be another form of MS or not MS at all any value? The difference with the old – new machines and the collected data confused my conclusions.
    What do you think?
    Thank you.

    • The current sensitivity of isoelectric focusing technique stands at 5% for OCB negativity in those with a clinical diagnosis of MS – some argue that these cases are not intact the typical MS and may have other disease processes at play which are predominant than B cell/plasma cell activity.
      OCB positivity is more likely to progress, in particular IgM OCB versus the typical oligoclonal pattern which are used for diagnosis are considered a bad prognostic marker correlating with disability progression.
      Unfortunately isoelectric focusing as a methodology is very user dependent are requires training, now there are international standards which are distributed around Europe as part of laboratory quality standards to ensure that laboritories are interpreting these accurately.

  • In 2011 a large lesion was seen on my spinal cord. NMO was suspected but I was neagative for that. Then I had a LP and that was negative. I had another LP in 2012 which was still negative. then in 2016 had new mri and LP, the LP was positive for OCBs but the large lesion on the spinal cord was no longer present. MRI was repeated on 3T machine and the cord appeared normal. still have the foot drop from that cervical lesion. I got the MS dx but do not as yet qualify for a DMD. I often wonder was that last LP a false positive.

    • There is a false positive rate to any test. Depending on where you had the rest performed sometimes the labs store the CSF sample and it can be re-run for confirmation. If it is there then it's there. I must point out that the sensitivity of spinal cord MRI is not great, if you remyelinate the lesion it may not be visible. Remyelination of a lesion doesn't automatically mean that the neuronal connections have re-established so you'll still be left with the clinical deficit.

    • Thankyou for the detailed reply. I feel it must be a goo sign for the lesion to remyelinate. All the sensory symptoms gradually disappeared and so have bladder issues completely. Just the spasticity remains and I suppose it is there to stay.

    • Thanks for asking
      Eyesterday I got my wisdom tooth remove and it hurt like cr@#+()//z&&&zy. Compare to the spinal tap today i Will do it 2 in a row no prob …Only thing, is ,you have to stand still for 2 days if not you gonna get dizzy (a lot)




Recent Posts

Recent Comments