This is topic Know Lyme test differences. ELISA vs Western Blot (why ELISA is a junk test) in forum Medical Questions at LymeNet Flash.

To visit this topic, use this URL:

Posted by bottle (Member # 31863) on :
I've spent some time learning about the ELISA tests and the Western Blot test. To explain the concept to a friend, I made the graphics below which I think would be useful to others too. Just skip ahead to them and you can come back to the written explanation if they don't make sense.

An ELISA test is a much weaker test than a Western Blot. ELISA looks at a single antigen (bacterial protein) to say if a patient has Lyme or not. In October 2010 a University of Medicine and Dentistry of New Jersey (UMDNJ) published a study saying that there are 13 strains of Borrelia (Lyme) Here is the story in Discover Magazine( ). The current ELISA tests are based on just one strain, B-31. The ELISA test uses a single antigen from this strain as an attachment point for a patient's antibodies. Clearly this recent discovery of many more strains could mean that the single antigen of the ELISA test is not shared by all the strains. This might explain why the ELISA test often misses cases of Lyme. Some companies even use a synthetic version of the antigen.

ELISA is usually done because it is a cheaper test at about $75, while a Western Blot can run around $200 to check for early & late antibodies (IgM (~$100) and IgG (~$100)).

The Western Blot test runs a patients sample through an gel with the use of an electric current. The patients antibodies are complexed (bound) to broken up bacterial parts in the sample. The electric current drags these bound pairs through the gel. The different weighted pairs end up in different places in the gel with the lighter pieces moving the farthest. These different places are represented as bands in the gel. The bands create a pattern and there are patterns that indicate if antibodies have been bounded to Lyme bacterial parts. This can indicate if Lyme is present. The advantage of the Western Blot over the ELISA test is that far more indicators are examined than the single antigen in ELISA. Perhaps the 13 strains will differ in a couple of particular antigens, but when you start looking at many antibody complexes it will become more difficult to miss Lyme's presence.

Think about comparing cars, with different models representing different strains. If you check for only one thing like GPS navigation to define the vehicle as a car, you will miss a lot of models that aren't equipped with that feature. This one feature test for determining if a thing is a car would be like using the single feature ELISA test. But suppose you start comparing a bunch of features like GPS navigation, steering wheel, 4 doors, fuel injection, foglights, and Air conditioning as defining a thing as a car. With a greater selection of features you will properly identify more things as cars. This is like using the Western Blot test which is looking at a broader scope in indentifying Lyme's presence through antibody activity.

Pictures are below.



Western Blot Test process:


(The band information I got from a number of websites. The International Lyme and Associated Diseases Society (ILADS) may use only a subset of those bands. However the idea of the band usage to identify Lyme's presence is the same. This is one of the webpages: )

Antibodies Complexed with bacterial part examples


Hope this helps everyone in understanding the test differences.


[ 06-14-2011, 05:21 AM: Message edited by: bottle ]
Posted by Tincup (Member # 5829) on :
Bottle. Wow! You are amazing. This is amazing!

One correction though. You said you spent "some" time on this. From what I see you spent much more than "some" time.

You spent a LOT of time! And it is appreciated!

I like your car parts as the example too. Good way to bring the concept to life.

Do you by chance have this on a pdf or website that we can refer folks to? It would be quite useful. You can do graphs that most can't do and it gives a great visual for folks.

Again, thanks for sharing this and doing all this work to help others. Very impressive!

[Big Grin]
Posted by Lymetoo (Member # 743) on :
Great info!! I'll post it in our Popular Topics and under the Western Blot link.

Thanks!! [Smile]
Posted by bottle (Member # 31863) on :
Thanks Tincup & Lymetoo. The differences for the tests have to get out out there so more people know to ask for the Western Blot. A visual was needed to show why ELISA is a weak teast.

Thanks Lymetoo for cross-posting. I am very new and don't know this site yet.

Tincup, let me see if I can make a PDF. I should be able to if I put these into a Word file and convert. I don't have a website. Do I just post the PDF in a new topic? Or where does it go on this site?

Feel free to spread the images around. There are some slight variations of the ELISA and Western Blot test, but this explanation captures the core spirit of them.

And 'yes', you are right, it was more than a little time. :-)
Posted by philly78 (Member # 31069) on :
Wow indeed! Very useful information.
Posted by lululymemom (Member # 26405) on :
Great Post! Valuable informaton..
Posted by bottle (Member # 31863) on :
Thanks lululymemom.

Tincup, I made a pdf. Please tell me where you would like it.
Posted by Lymetoo (Member # 743) on :
Posted by bottle (Member # 31863) on :

The concepts of the Western Blot Test described above are sound.

You have included the specific chemical steps which are not needed in the conceptual explanation. The actual blotting from one membrane over to the other is not needed either and just adds confusion. The final output of the Western Blot and how it analyzes data points on a broader scale than the ELISA is what is important, and that is what is concentrated on in the above diagrams.

When a patient goes to a physician the above diagrams and their information is what is relevant and poignant. The patient does not need to describe in agonizing detail the percentage of acrylamide gel that is used (which can be variable by the way), the membrane was of nitrocellulose nature due to nitrocellulose's affinity in binding of proteins, the pH , etc.

If you have more recent antibody-antigen-bounded weight information, please feel free to post it in this thread instead of keeping it private or referencing some other site.
Posted by ChuckG (Member # 19093) on :
Do you believe that the lyme antigen and antibody actually move together on the gel under the impetus of the applied electric field?
Posted by bottle (Member # 31863) on :
If you have some useful knowledge to contribute, then feel free to contribute.

If you have better diagrams, then post them.

If you have better sources, then cite them.

Most people from Berkeley would do those things.

What's stopping you?

If you know things so well, where are your drawings? Where is your explanation?

Produce something with your knowledge.
Posted by ChuckG (Member # 19093) on :
Originally posted by bottle:

The concepts of the Western Blot Test described above are sound.

If you have more recent antibody-antigen-bounded weight information , please feel free to post it in this thread instead of keeping it private or referencing some other site.

antibody-antigen-bounded weight information is what I object to. It does not exist to the best of my knowledge in lyme Western Blotting.

"An aliquot equivalent to 50 ug of pooled B31 and 297 B. burgdorferi sonicated cell lysate (l00ug/ml) was fractionated by SDS-PAGE on 11.3% acrylamide gel..."

"fractionated". Bb proteins are sorted by molecular mass.

" ...and transferred to Protran BA nitrocellulose membrane (Schleicher & Schuell, Keene, NH) in Tris-glycine-methanol transfer buffer pH 8.3 [25 mM Tris, 192 mM glycine, 20% (vol/vol) methanol]."

"transferred to Protran BA nitrocellulose membrane". Bb proteins sorted by molecular mass are "transferred" to a membrane. The Bb proteins are removed from the gel and adhere to the membrane in the same relative location as they had in the gel preserving their spatial relationship.

The job of the gel is done. Electrophoresis is done. Antibodies now enter the picture.
Posted by ChuckG (Member # 19093) on :

Viramed Biotech AG Borrelia B31 ViraBlot

page 3.
Summary and explanation of the test

...A Western blot can be considered a modified solid-phase enzyme linked immunosorbent assay. Electrophoretically isolated antigens are bound to a solid phase nitrocellulose support membrane called an immunoblot strip. For each test to be performed, the immunoblot strip and diluted test sera is added to a immunoblot strip well designed to containing the immunoblot strip and sample. If specific antibodies that recognize an antigen are present, they will bind to the specific antigens on the strip.

Posted by bottle (Member # 31863) on :
Thank you ChuckG. This kind of datasheet is just what I needed. Ok, I see some changes should be made. I appreciate your input & the links.

I will try to update the pictures as soon as possible and re-upload them.
Posted by bottle (Member # 31863) on :
Ok I modified the pictures. They are visible above since I re-uploaded them on the hosting site.

They are more accurate now with the test process. Thank you for the references ChuckG.
Posted by ChuckG (Member # 19093) on :
Probably you should write a very short description of scoring. I hate writing! So edit away.

The degree of ab binding to antigen determines the strength of the visualized band. And it is the strength (grayscale) of the band that determines how it is scored. Relative to a "standard" band.

Clongen had a pdf, which I downloaded fortunately but is no longer available, that displayed the bands and the machine scores.

Setting up a machine to read the bands is a non-trivial task.

Some lyme ELISAs used multiple antigens.

This links to Dr. C's explanation. It dates to about 1990. The descriptions for 83 and 93 are absurd as I have pointed out more than once.

dates to around 2000. And it has major errors too. There is no more recent and up-to-date list on the net that I am aware of. I suppose I should generate one.
Posted by Lymetoo (Member # 743) on :
It should be noted that you can take a POSITIVE ELISA TO THE BANK. There are extremely few "false positives" when it come to the ELISA.

What we have to worry about is a falsely NEGATIVE ELISA.
Posted by bottle (Member # 31863) on :
ChuckG thanks for the idea for scoring. I don't mind writing. I just need the time and that is short right now and will be for awhile. If it is ok with you, I will contact you later about where to go for the best technical scoring papers.

Thanks for the multiple antigen testing on the ELISA. The ones I read just said one. The name of one of those companies was I- m - m - u - n - e - t - i - c - s (spaced it out so a search engine doesn't find this and then they give us crap about it). And what they used was a synthetic one too.

Thanks for the relatively newer band info too.
Posted by bottle (Member # 31863) on :
Thank you Lymetoo for that note on the strength of a positive ELISA. I appreciate your input.
Posted by lorima (Member # 11925) on :
Regarding the specificity of various bands, my favorite paper has always been Ma et al., 1992.

Here's a link to the pdf - look at the bar graph on journal page 373 (page 4 of the pdf). It summarizes results from 186 patients with Lyme, compared with 320 normal controls.

I keep a printout of this page in my Western blot folder for quick reference.
Posted by lorima (Member # 11925) on :
About the cost, HJ in her book:

says that in 2005, the "list price" for a Western blot (a CDC one, not a good one) was $115.88, but the price the insurance company paid was $12.41.

That's right: in a reasonable world, one could get both IgM and IgG Western blots for a mere $25. There would be no extra work involved in reporting the full set of bands rather than the limited set that are now reported, either. All IDSA has to do is say the word.

It would save money for the insurance company, to treat people with Lyme disease with oral antibiotics, which are pretty cheap even long term. Rather than to run all the diagnostic tests to rule out all the other possible diseases, and then to pay for expensive meds for fibromyalgia, chronic fatigue syndrome, MS, and all the other stuff Lyme patients get misdiagnosed with.

But it might crash the New England economy if people knew what a plague zone it is. If you had a choice between sending your kid to college or summer camp in the Northeast, or somewhere less afflicted, would you chose, say, Yale, knowing what we know? Or to a summer camp along the Hudson?

Maybe they're waiting for the whole country to be endemic, and then they'll let it be diagnosed, so they can sell their next vaccine to a big market. But that assumes they know what they're doing, and I doubt that. My opinion of MD brain power is now extremely low. A few of them make stuff up, and the rest of them repeat what they're told.
Posted by bottle (Member # 31863) on :
Thanks for the information Lorima. That is great stuff.

I promise to incorporate it into the slides. However that day maybe 2 months away. I'm in a time crunch for many things.

Thank you though, and feel free to keep adding. I will eventually be freed up.

Powered by UBB.classic™ 6.7.3