OCB, IgG Index, and the McDonald Criteria Predictive Values - Sensitivity and Specificity

[COBOB]

Just for grins and giggles I figures I'd do some research to see what the predicative values of IgG Index, OCBs and the McDonald Criteria were in MS/CIS.  The McDonald Criteria is for CIS, since it no longer applies after diagnosis.

OCBs (by IEF)   Sensitivity: 90%   Specificity 94%
CSF IgG Index   Sensitivity  70%   Specificity 96%

So here is where the "10% of MS patients don't have bands" number comes from.  90% of clinically positive MS patients have OCBs.  6% of clinically negative patients also have OCBs.  Only 70% of clinically positive MS patients have an elevated IgG index.  4% of clinically negative MS patients have an elevated IgG index.

based on data from "Cerebrospinal Fluid Oligoclonal Bands in the Diagnosis of Multiple Sclerosis" (2003)
http://ajcp.ascpjournals.org/content/120/5/672.full.pdf

-------------------------
"The newer McDonald criteria demonstrated sensitivity of 74%, specificity of 86%, and accuracy of 80% in predicting conversion to clinically definite MS."  (misses 26% of clinically positive cases reports positive on 14% of clinically negative cases.)

So this breaks down to:  If you Dr. is using the McDonald Criteria and he says he's 80% sure, that is  statistically the best answer you are going to get.

based on data from "DIAGNOSING MS IN CLINICALLY ISOLATED SYNDROME—THE ROLE OF MRI" (2005)
http://www.neurologyreviews.com/sep05/MRIandMS.html  

Bob

[lovemy6monkeys]

Oh goodness! Part of me feels badly that my "80% sure it's MS" starting this...but a bigger part of me is thrilled that my comment caused all this wonderful info to be posted!

I think I should clear up a few things about my case, because I do think I'm one of the lucky ones who did not spend a lot of time in Limboland.

My neurological symptoms started in November of last year. The exception is that I experienced overwhelming fatigue after the birth of my youngest in April of last year. I saw my PCP about a month after the neurological symptoms started, at that point MS was my greatest concern. He referred me to a neurologist and sent me for an MRI of the head and c-spine. I saw the neuro in January. She said MRI was normal did an exam and said "Do you think you are just anxious."

I left and refused to go back to see her. I was given the recommendation of my current neuro. He has a reputation for aggressively looking for a diagnosis...and aggressively treating. I saw him in February and he sent me for another MRI (head, c-spine, full spine). This was done in early March. That MRI was considered normal. My neuro wants to see you when any issue starts so over the following weeks I was in his office at least 3 times a month.

At the end of April I had vertigo that my neuro considered to be from the brainstem. He also noticed nystagmus for the first time during the exam. I was started on a course of steroids. Within a few days the vertigo went away, except for slight off balance feelings when tired. From that point on I showed Romburg's sign at every exam.

At the end of May I had buzzing/vibrating feeling that started in my right foot and traveled up my leg. The weather was warming up, and I was showing signs of Uthoff's and L'Hermittes. Again I was given a course of steroids, and again it cleared it up.

In August I had a return of the vertigo, though not as severe as the first time. They also redid a few eye tests that they had done back in March, VEP and the other one that is like an MRI of the eye. My left optic nerve still showed slight thinning but the same as last time and still within normal limits. The VEP in March showed an 8 millisecond delay (not enough to point to MS, but enough to warrant further MS testing) and in August the delay had increased to 10 milliseconds.  This is when my neuro made the diagnosis and offered DMD. I guess I fit the criteria for dissemination in time and space.

My neuro is aggressive in treating MS. He left the decision of starting DMD up to me, but did say he believes I'm still "early" enough in the disease process that DMDs would give me the best chance of slowing the progression of disability.

I do know I'm very lucky to have found the neuro I did. And I'm even luckier to have found this forum with such smart and caring people! :-)
~Jess

[supermum_ms]

From my perspective this "Basically, Dr.s like signs, not symptoms.   Symptoms are hard to demonstrate in court at a malpractice trial.  Signs are easy.  Sometimes I think this is what drives much of the diagnostic process. " assumes defensive medicine as the norm, not the exception, pitty.

I also have the experience of having definitive signs, and although they are there, they are being dismissed because the presumption is I shouldn't have any with the dx of 'chronic ischemic white matter disease' which is stated on the radiology report. Look for lesions with MRI to explain the signs, find them but call them something else so therefore logically (in neuro world of logic) the signs dont 'really' exist.lol

Examples:
Romburg's sign: swaying of the body or falling when the eyes are closed.
Clonus: a continuous rhythmic reflex tremor initiated by the spinal cord below an area of spinal cord injury, set in motion by reflex testing.
Nastagmus unilateral: acquired horizontal jerk nystagmus there is a slow and fast phase

I could give a few more but you get the picture, most i'm sure are 'unfakeable' signs, all my sx are consistent with a dx of MS, I meet the McDonald criteria even if the subjective sx (eg Uhthoffs etc) are not included, though I do not have a dx and still in the gathering evidence stage lol!

Cheers.......JJ

[jumpinjiminy]

I mean no MG, no testing for it. But the MS doc said he saw a possible lesion in my cervical cord, and insisted I need an MRI there. And I can't get my neuro to justify doing it.

[jumpinjiminy]

No. I don't know that I've shared the whole story in one post. I think it's been spread out over several. Maybe my first post. I'll go look and bump it up.

Then I can write it up in my journal if you like. It keeps growing. :-) There are new things that have come to light since I first posted here. Very significant new things. And they don't indicate MS, but lupus instead. One of them happened, as if on cue, right after getting home from seeing the MS doc on Tuesday.

[Quixotic1]

JIJ- I'm going tol look for the post where you tell us your whole story.  I don't remember everything anymore.

Would you look for it, too, or write up the whole enchilada in your journal?  

On the way there, tho, I have a question.  Have you been checked for Myasthenia Gravis?

Quix

[jumpinjiminy]

I think I need to move on to other docs. The neuro I'm seeing out here has no idea what to do with me. He didn't think a rheumatologist would be able to find anything either because of the negative ANAs, CRPs and RFs.

The speciality neuro at THE place to go for MS in the big city didn't offer any other possibilities either. He gave me the names of the resident neuro and her attending neuro that saw me one night at the big university ER and said they would be interested in following up.

I'm not sure what to do. I think I will hit up my GP once more. He's always good about running tests. After that, maybe schedule with a rheumy at the university in Chicago. I'm torn between seeing the attending neuro or a rheumatologist.

I don't like how they won't even consider MS without a brain lesion. It may never show up. Then what? This is hitting my diaphragm, throat, vocal cords, muscles under my tongue, and other really scary places. I don't like it, and I want answers this year.

[Quixotic1]

JIJ - Your neuro is full of prunes.  At least 5% of people with definite MS have lesions ONLY in the spinal cord.  I guess he is comfortable missing 1 in 20.  Shame on him.  He had his back up and was trying to take the wind out of your sails.  That's it!  Let's remove hope from this person that they will ever get a diagnosis.

Since he is SO sure of what you don't have, any offer or suggestion of what it "might" be?

How useless!  You are looking for a proper doctor that is less involved with himself and more interested in the patient, aren't you?

Quix

[jumpinjiminy]

So why the hard line in the sand when the MS doc said even if my story said MS, and if in six months I ended up with a brain lesion he would say it's MS?  He said even if I had CSF suggestive of MS, AND spinal cord lesions, they wouldn't call it MS or treat it until a brain lesion showed up.

Does patchy loss of sensation on hands and feet correlate with spinal cord lesions? Or is that a no lesion category?

I do not have the cognitive skills to go so in depth will all of the arguments above right now. Maybe with some treatment I would get that ability back. But so far, it's not likely to happen until the damage is done.

[Quixotic1]

Bob - Silly Rabbit!!  I sure didn't mean to lecture you.  I was hoping to add fire to your very accurate assertion that an 80% confidence level was VERY strong and that DMDs should probably already have been started.  I think because 73% of labs do the O-Band testing wrong, the Sensitivity and Specificity quoted are invalid for use with the public outside a rigorous study.

I wrote 90% of that epic post for the benefit of the forum.  It's what I do - use all the words I have access to for a given day.  

OH!  Here's a word I had left over

IDIOSYNCRACY

Sorry, Bob, I would not ever mean to try to scratch your thick hide.  Your presence on the forum is too special and too welcome.  I rarely give the scientific side of things. Many of our members have advanced degrees and that is the kind of data they are looking for.  You are very much what this forum needs.

Be sure to know you are welcome here.

Quix, the Sheepish

[COBOB]

Oh, by the way.  I have to fix the chicken coup roof, get the cow's water tank ready for he first freeze, cut the lawn for the last time this year (We only have 6 acres,) and rake to poop out of the dog run.  Any bets on how many naps will be required.  I hate to think about winter and having to pitch hay bales over the fence for five months.  Don't think I have any spare time this weekend :-P

[Quixotic1]

Here is the link for the HP on the History of the Diagnosis of MS

http://www.medhelp.org/health_pages/Multiple-Sclerosis/History-of-the-Diagnosis-of-MS/show/158?cid=36

Read this one before the one on the McDonald Criteria.

solly

[COBOB]

I think I realize most of what you said.  There is a huge variability in lab reproducibility.   Most of this comment was a response to a post that when a Dr said they were 80% sure it was MS, that it wasn't good enough to start DMD.  My argument is that given the current testing technology and how some Dr.s are evaluating the clinical symptomatology and the exam signs, 80% may be the best we can get. Given he stories some of us have, if the Dr told me 50%/50% I request that he start DMDs.  Given the conversion rates over time, that might me the safest course.

Basically, Dr.s like signs, not symptoms.   Symptoms are hard to demonstrate in court at a malpractice trial.  Signs are easy.  Sometimes I think this is what drives much of the diagnostic process.  

And I realize some of the issues have been raised before, lots of new folks...me and others that have not heard all the "war stories," yet.

Bob (thick skin not even slightly scratched)  

[Quixotic1]

THE REAL PURPOSE OF THE MCDONALD CRITERIA

Hi, all.  the purpose of the McDonald Criteria - or the reason the international group of experts got together to develop the MC - was to provide diagnosing doctors with a way to supplement the clinical data (history and phsyical) to make a diagnosis earlier.  This meant that a person who had only had one attack (a Clinically Isolated Syndrome or CIS) might still be able to get a firm diagnosis if the MRI provided certain types and quantities of lesion information.

They also provided info that could be used if the person has had two attacks (so NOT a CIS) but only had evidence on exam of one clinical lesion.  The MRI could be used to provide evidence of that needed 2nd lesion.

Definition of "Clinical Lesion" An abnormality on exam that shows without a doubt that a lesion (or area of damage) exists in the Central Nervous System.  Example, a person has hyperactive reflexes in a leg with clonus.  That can only be caused by a lesion in the spinal cord - whether or not such a hyperintensity is seen on MRI!  Those pathologic reflexes are a "clinical lesion".

As always, a person could be diagnosed without reliance on the McDonald Criteria if there was evidence of 2 or more attacks, 2 or more clinical lesions, and other reasonable explanations of the patient's problems had been ruled out.

It seems from my viewpoint that the majority of non-MS neurologists believe that the McD Criteria are the template that must be followed to make a diagnosis.  Nothing could be farther from the truth.  It was put forth as an adjunct when there weren't enough attacks or clinical lesions yet, so the patient didn't have to wait months or years for the next attack or lesion.

THE DATA ON O-BANDS CANNOT BE COMPARED TO WHAT HAPPENS IN REAL LIFE

Bob - you joined us fairly recently and so have missed some of our discussions about the REAL sensitivity/specificity of O-Bands.  The studies aren't flawed, but reality IS quite flawed.  For one that test requires that the tech be well-trained both in running the test and in interpreting the test.  It requires that usual method of testing for bands be aumented by using a technique called IsoElectric Focusing.  It also requires that the testing of the CSF be partnered with a simultaneous testing of the serum for comparison.

When they do studies they follow all procedures rigorously and conscientiously.  The problem is that in real life the testing is often done poorly, inaccurately or imcompletely.

Last spring, I think, I found a study that survey a random collection of about 130 labs - private, hospital, medical center, academic center.  Their questions included asking whether these labs used BOTH Isoelectric Focusing AND tested the serum simultaneously.  ONLY 27% of all the labs surveyed did the test correctly.  That is a whole lot of missed O-Bands.  I had wondered out loud here on the forum for the last three years about this very topic and predicted that most/many labs did the testing wrong.

UNDERSTAND THE BASICS OF HOW THE DIAGNOSIS IS MADE

To all:  I know they are long and terribly wordy, but there are two Health Pages that tell about the information and process that had been is and now is used to diagnose MS.  It makes it clear that a diagnosis with a good amount of certainty can be made without doing an MRI.  Even the McDonald Criteria say this.  However, the vast majority of MS neurologists DO want to see "some" kind of abnormalitiy on the MRI as confirmatory evidence.

Now, ever since the revised McDonald Criteria came out in 2005 they have been trying to simply them and there are some good candidates that are being validated.  I expect a revision soon.

Please take the time and check out these sites:

For this first site, go to this address and scroll a little more than halfway down the page to the section called, "McDonald Criteria for the Diagnosis of MS"  Click on the "Pocket Card".  YOu will see a table that shows you what information is needed.  Read from left to right.  The first option is 2 attacks, 2 clinical lesions = no further evidence "needed", but MRI evidence desirable.  This means some lesion that is "consistent with MS"  It does NOT mean that the evidence must be classic or even characteristic.

http://www.nationalmssociety.org/for-professionals/healthcare-professionals/resources-for-clinicians/index.aspx

This next Health Page is a history of how MS has been diagnosed since the 1960's when the first guidelines were written.  They really help to understand the discussion on the McDonald Criteria.

http://www.nationalmssociety.org/for-professionals/healthcare-professionals/resources-for-clinicians/index.aspx

Finally the last link is my attempt to describe exactly how the McDonald Criteria SHOULD be used as per the group's intention that formulated them.  You will see that the first information that is sought is the number of attacks and the number of clinical lesions.  All the rest is an attempt to "fill in" a missing attack or a missing 2nd lesion.

http://www.medhelp.org/health_pages/Multiple-Sclerosis/Diagnosing-MS---The-McDonald-Criteria-revised-2005/show/370?cid=36

I hope that all of these together will give you a better idea of what the thought processes behind the Criteria were and how they were to be used.  I do not understand why the neuros do not understand these.  They take a tiny bit of thinking, but I got it and I was a pediatrician!!!! (Neuros look down their noses generally at pedicatricians, lol)

I really think that if you take the time this will make more sense.

FINALLY!

There is a difference between diagnosing a CIS (precursor to full-blown MS) and diagnosing definite MS.  A person with CIS must wait for the 2nd attack or lesion (whichever hasn't happened yet) or look to the McDonald Criteria to fulfill the evidence that takes a CIS to a Definite Diagnosis.

There is also a difference between making the diagnosis of MS and having a high enough suspicion of MS to begin a DMD medication.  Again, too many neuros are unsure of themselves and seem to want a 100% assurance.  As Bob says, "It ain't going to happen!"  It's a matter of the accumulation of evidence and the most important evidence is from the patients's history and from the neuro exam.

Quix - wheww!!!!!!!!

[sllowe]

After you visit Audrey's, my house needs some attending to toooooo

I'm letting the predictability info sink in here....

Since you mention CIS, I think you'd be interested in the 06 recommendations which detailed the syndrome in depth, or I should say much moreso than in 03 and 09. In the comparative HP there is a link to all three, 06 in particular is good reading on CIS (it's in the lower bottom corner of the page).

Bob, would you mind clarifying your mention of the McD criteria being for CIS? I think that could leave many confused. It is for dxing MS and inclusive of CIS (more in recent yrs than past, etc) Thanks! And, thanks to your work too for being slow! Leaving you here w/us to post!
ttys,
shell

[Audrey_M]

Thanks for the info, and yes, you did have time on your hands today!

Want to come to my house and wash the windows, power wash the siding, use my rider mower to do the lawn, and on and on?

Giggles to you!

Audrey