Validity of the CDC 2-tiered Test for Late Lyme Borreliosis

Medical topics with questions, information and discussion related to Lyme disease and other tick-borne diseases.
Post Reply
TicksSuck
Posts: 100
Joined: Thu 31 May 2012 20:25

Validity of the CDC 2-tiered Test for Late Lyme Borreliosis

Post by TicksSuck » Tue 19 Mar 2013 17:50

I have asked a seemingly knowledgeable person before in other topics, but never got a specific answer... Maybe there was too much noise from shillers and the question got lost, or (what I suspect), the person just don't have an answer. I will therefore ask more generally in a new noise-free topic.

Can somebody please provide me with a study that shows that the 2-tired test advocated by the CDC for Lyme Disease (http://www.cdc.gov/lyme/healthcare/clin ... otier.html) is sensitive in cases of late stage (>30 days) disease, but doesn't suffer from selection bias?

Henry
Posts: 1108
Joined: Thu 10 Nov 2011 18:49

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by Henry » Tue 19 Mar 2013 23:17

Try this reference for starters: http://www.ncbi.nlm.nih.gov/pubmed/19947857

TicksSuck
Posts: 100
Joined: Thu 31 May 2012 20:25

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by TicksSuck » Wed 20 Mar 2013 0:22

Henry wrote:Try this reference for starters: http://www.ncbi.nlm.nih.gov/pubmed/19947857
Thank you Henry!

I will read this more thoroughly, but at first glance, it would appear to suffer from the usual selection bias:
Patients and Methods

Study subjects
The Human Investigations Committees at Massachusetts General Hospital (MGH), New York Medical College and CDC approved the study. All patients categorized as having LD met the CDC surveillance criteria for the diagnosis [16].


...and guess what this CDC surveillance criteria requires for diagnosis of late cases of Lyme disease?

Lorima
Posts: 914
Joined: Mon 29 Oct 2007 20:47

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by Lorima » Wed 20 Mar 2013 1:11

Here are two more papers that use a positive two-tier test as a selection criterion, and then claim to show that the test is 100% sensitive in late disease:

Steere et al 2008 
http://cid.oxfordjournals.org/content/47/2/188.full.pdf 

Branda et al 2011 
http://cid.oxfordjournals.org/content/53/6/541.full.pdf 

We should compile a list.
"I have to understand the world, you see."
Richard Feynman

Henry
Posts: 1108
Joined: Thu 10 Nov 2011 18:49

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by Henry » Wed 20 Mar 2013 14:08

Well, then tell me WHAT you would use for starters -- as an independent comparative standard? What you fail to understand is that the specimens used were derived from patients that have been well-characterized for their Lyme disease so that their specimens are truly representative. There is a well-defined clinical history associated with each specimen used. That is the value of a study such as this. These people are not stupid. However, I must confess that I expected such a response from both of you. It would not matter to you in the least that other newly developed test , e.g., the C6 ELISA give results that are consistent with those of the two-tiered test. There are many who post on this website who praise IGenex's IgG Western blot; however, except for 31 and 34 kDa which are not acceptable for reasons that I noted earlier, it does not detect any new or different markers than the 10 associated with the CDC's criteria for IgG Western blot. It requires the presence of FEWER bands to be positive (3 of 5 possible bands, including 31 and 34 kDa. How is that an advance? It surely results in more false positive tests. But, believe what you may; it is your body and your health.

Also, keep in mind that much of the data generated for 2-tiered testing comes from the many State Laboratories of Public Health that actually do these tests; part of the reporting procedure includes not only the results of serological tests, but also clinical data from the patient's medical history. All that is submitted to the CDC independently. That's where the CDC gets much of its reporting data, although they do their own research as well. However, there is much to much distrust to really have a fruitful discussion about these and other issues.

Lorima
Posts: 914
Joined: Mon 29 Oct 2007 20:47

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by Lorima » Wed 20 Mar 2013 15:35

Henry, why do you get so angry when discussing these matters? 

It gives the impression that you are so upset that intelligent people might question the official disease model and its trappings, that you can't execute the usual sequence of careful, impartial analysis, upon which science depends. 

I grant that it is very upsetting to think about the possibility that scientific and medical knowledge is this fragile, and this easily manipulated. I couldn't speak about it for years, after I caught on, without raising my voice. It felt like a betrayal. 

Now I know I was naive. Science only works, if everyone involved is very competent, very careful, and very resistant to being lured into short-cuts by expedience. Like all prestigious and lucrative human endeavors, it is susceptible to error and distortion at many points, and for many reasons. The culture of intellectual independence that is necessary for it to work, is in constant danger of deteriorating, and has to be actively maintained by all those involved. Call it entropy, or whatever you want. Maybe it's time to jettison your unquestioning faith that the LD medical dogma is "evidence-based", and look at the actual evidence itself. Yes, it can be tedious, and it won't please your teammates. But if you choose not to look, please don't act like we're the ones who are failing to examine the evidence. 

Now, can you find the paper that claims to have not used serology to select the great majority of the late-stage LD patient samples? I seem to recall that I found one, once. But there were so many others that made the tautological error we've been discussing, that the evidence seemed overwhelming to me that the 2-tier test as formulated by Steere, has no real scientific support for being used as a diagnostic requirement; and now I've forgotten where I found that one paper. 

Looking forward to hearing from you, 
Lorima
"I have to understand the world, you see."
Richard Feynman

Henry
Posts: 1108
Joined: Thu 10 Nov 2011 18:49

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by Henry » Wed 20 Mar 2013 16:25

Lorima: Why do you characterize my response as being angry? Let me re-phrase what I consider to be the real issue. What objective standard/test would you use to diagnose a patient as having Lyme disease who is seronegative by currently used tests, all of which are based on an analysis of specimens that have a well-defined clinical history? Please tell me -- I'm waiting to hear your answer. What do you have to offer that is superior? Both physicians and patients are waiting to hear your answer.

Granted, all diagnostic tests can be improved so that they are are more sensitive and specific. But, if the existing diagnostic tests are so deficient and biased as you claim, what would you recommend in their place? How would you justify treating a patient with antibiotics -- or some other therapeutic-- when they are seronegative and no Borrelia can be detected by culture or other means? Why would you be so certain that the patient that stands before you has Lyme disease and not some another medical condition with similar symptoms? How can you be so sure that it is Lyme disease and not some other medical condition that deserves appropriate treatment and care? One does not have to be angry to ask such questions that I think are entirely reasonable. You say that patients who are seronegative have Lyme disease. I'm simply asking how do you know and upon what objective evidence do you base such a claim? Could it be that YOU are angry because you can not answer my question?

Please keep in mind that I by no means believe that these patients are not suffering from something that deserves to be treated appropriately, once diagnosed. I know many physicians who treat patients with Lyme disease. Some have had patients referred to them after being treated for many months without success for what was misdiagnosed by someone else as Lyme disease; some of these patients were later found to have a vitamin B12 deficiency, hyothyroidism, leukemia, MS, ALS, Parkinson's disease or other non-infectious conditions. Also, I know how medical research and science works; it is based on the best evidence available at any given point in time. Changes/modification of the current working model or paradigm do occur and are based on the results of new evidence-based research, not intuition or "gut feelings". I am not insensitive to the needs and suffering of patients who are seronegative for Lyme disease; in the absence of objective evidence supporting a diagnosis of Lyme disease, I think it is reasonable to explore other possible causes for their symptoms -- if one really and truly wishes to help them. Now, what's wrong with such a view?

Lorima
Posts: 914
Joined: Mon 29 Oct 2007 20:47

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by Lorima » Wed 20 Mar 2013 18:56

Henry, I'll just quote myself. This time, I will highlight my proposal; maybe that will make it more memorable. I'll start collecting examples in the literature, where other respected, mainstream researchers, have made similar proposals. 

http://www.lymeneteurope.org/forum/view ... 33&p=35000
Lorima wrote:It has to do with the trade-off between sensitivity and specificity. In Craven et al, 1996, referenced in my post above, one can see that three of the academic experts (presumably Steere, Sigal, and Dattwyler) successfully argued for the test to be biased toward specificity, at the expense of sensitivity. It also reveals an overreliance on arthritis patients in sampling late-stage cases, which you've already admitted are known to exhibit an unusually robust, varied, and prolonged antibody response. The five-band criterion is what is absurd, in my scientific judgment. Apparently there was a lot of doubt about the test in the "insiders" community too, as late as 1998, looking at the meeting agenda I also quoted, above. 

snip

I suspect that a requirement for two Bb specific bands, would only allow a rare patient with some kind of unusual cross-reactivity, or some other borreliosis, to be falsely positive; and it would surely keep many, many patients from going undiagnosed, and even worse, falsely assured that whatever they have, it is not Lyme disease, merely because they happen not to be making the high and diverse IgG response required by the Dressler criteria, at the time they are tested.

This is not the choice that Steere, and his followers and collaborators, made. But there is nothing irrational about it.


Going back over that conversation, I see that I let myself get distracted by a side-issue that Martian introduced. I will now take the time to write a response in which I specifically address each of the points you raised, in your posts adjacent to mine that I quoted above. This may take a while, as I intend to cite literature supporting each point I make. It would be nice if you would take the time to cite the literature that supports your points; otherwise I have to do the literature research, for both sides of the argument.

To summarize, I think two Bb-specific bands should be sufficient for provisionally diagnosing, and treating, a case with symptoms suspicious for LD. It certainly should be a clear sign not to rule LD out, in the differential diagnosis.

Thanks for writing, I do appreciate your comments.
Lorima
"I have to understand the world, you see."
Richard Feynman

Henry
Posts: 1108
Joined: Thu 10 Nov 2011 18:49

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by Henry » Wed 20 Mar 2013 19:44

Lorima: You state:

"I suspect that a requirement for two Bb specific bands, would only allow a rare patient with some kind of unusual cross-reactivity, or some other borreliosis, to be falsely positive; and it would surely keep many, many patients from going undiagnosed, and even worse, falsely assured that whatever they have, it is not Lyme disease, merely because they happen not to be making the high and diverse IgG response required by the Dressler criteria, at the time they are tested."

What two bands are you referring to and where is the evidence to justify that the presence of only these two bands would be sufficient to make a diagnosis? If they are bands for 31 and 34 kDA, you would be wrong since these have no diagnostic value. These are proteins that are expressed by Borrelia mainly when grown on artificial laboratory media and in the midgut of ticks, not during mammalian infection -- except when there is longstanding Lyme induced arthritis with the production of large amounts of inflammatory cytokines that alters (actually heightens) the immune response significantly.

In closing, recent comparative studies show that the results obtained with the C6 ELISA test are in complete agreement with those of the two-tiered test. Although this suggest that the C6 ELISA test might be a appropriate single-step assay for the diagnosis of Lyme disease, thereby replacing the two-tiered test, the FDA would like to have additional supportive data before making a final decision in that regard. The main point here is, could the same results obtained by two independent tests be all that wrong?

The 10 protein bands that comprise the CDC criteria for the IgG Western blot were derived from extensive studies with well-characterized clinical specimens on the frequency of bands that appear during late Lyme disease. Although much of this was developed from the work of Ma et al, (J. Clin. Microbiol. 30: 370-376, 1992), It has been confirmed by several more recent studies, mostly conducted by Aguero-Rosenfeld who is an expert on the diagnosis of Lyme disease.

Ma et al wrote that, when antibodies are detected in specimens from patients with late Lyme disease (when the antibody response is mostly -- if not exclusively--IgG), " antibodies against the 31 and 34 kDa proteins were rarely detected and, consequently, became less significant when compared to other protein bands in this study." When bands at 31 and 34 kDa are observed in IgG Western blots, they are virtually always in the context of a robust IgG response to a large number of the protein bands than comprise the CDC criteria for IgG positive Western blots. So, from what is known about the patients who have late Lyme disease and are known to be seropositive, it would be highly unusual -- if not almost beyond belief-- that there be antibody vs the 31 and 34 kDa bands and NOT any of the 10 bands that comprise the CDC criteria. But, since you didn't say, perhaps you are thinking of two different bands. If so, which ones are they?

duncan
Posts: 1370
Joined: Wed 5 Sep 2012 18:48

Re: Validity of the CDC 2-tiered Test for Late Lyme Borrelio

Post by duncan » Wed 20 Mar 2013 20:15

I feel compelled to interject. To me the point of this thread is to call into question the validity of the CDC 2-tiered test. It is not necessarily to propose an alternative.

To defend a test simply on the merits that it is currently the best available - which is what it appears to me Henry is attempting - is not the same as saying it is adequate. It is not the same as stating unequivocally that misdiagnosis are not rooted in the current testing methodology in itself. It is not the same as declaring those tested via the present method can be given a guarantee they are being diagnosed correctly, or with a five 9's confidence factor, or anything even approaching that. Or merely that the criteria by definition are not excluding individuals the test is meant to catch in the first place, jettisoned by the test itself.

If the current two-tiered test is flawed, it is flawed not at the expense of researchers who promote it, or physicians who appeal to it, but rather the patients who founder without recourse due to its misdirection.

Post Reply