Friday, June 5, 2009

The murky world of Western Blots

Lyme Western Blots can be quite different.
Some labs make their own kits while other labs buy commercially prepared kits.
The "mill labs" Labcorp and Quest simply report whether or not bands are present or not. Other labs grade Blot/band responses.

Specialty labs typically grade bands on a scale of 1 to 4 and may also use the indeterminate designation. These results are manually read by the person performing the test.

The dark line or blot seen on the patient strip (blot) are visually compared to the control strip. The ratings are subjective and may very from one observer to the next.
The human eye unfortunately is not all that reliable.

I recently looked at some blot results with Dr. K. He has used both manual blot and computerized blot technologies.

He offers state of the art computerized WB results.
Here the computer digitizes the the images seen on the Western Blot strip. The computer counts the pixels and compares the intensity of the patient test reaction with that of the control. To my eye, patient bands may appear equal to the control reaction but may be read negative by the computer.

He uses the "Mayo Clinic" methods. Bands are reported as positive only if the pixelated blot/band is 90% of the control. The report indicates a percent reading as it relates to images present on the control strip. For example, an IgG 41 band might be reported as 300%, this means it has 3X as many pixels as the control band as read by the computer software.

This technology which should be more accurate in fact tends to show less positive bands than manual WB results. This is because the human eye tends to see bands as being more intense than perhaps they actually are. We are biased to see positive bands. However-- since WB have always been performed manually one could question the validity of the computer generated reports.

A strip for one of my patient's recently showed a 34 band with a reading of 85%. Although this is 5% less than the Mayo Clinic criteria, it is pretty clear to me that this is a positive band.

When I look at strips I frequently see weak bands at key locations which are reported as negative. This is based on the set-point of the software. Truly "negative" bands(in my opinion) show no reactivity at all- the strip is blank. Any bands which are visible, even if they are a low percent of the control need to be considered, especially when they are present at critical locations. When physicians are presented with the actual strips it provides more information. Information is always a good thing.

IgeneX makes it own kits and interprets the results manually.
Clongen uses commercial kits, which are CDC approved( I don't know if this is good or bad) and offers both manual and computerized interpretations if requested.

The mill labs only report if bands are present. They do not indicate their cut-off points or methodology.

When WBs are sent to three labs at the same time different results are obtained quite frequently.

As you can see the world of Lyme Western Blots is murky like everything else Lyme; so--what else is new?


bobcat said...

What about Stony Brook? (I am not affiliated with them in any way) Are they pretty reliable compared to specialty labs like Igenex with their Western Blots?

I realize they don't report band intensity, but I have still gotten good results with them in terms of showing many bands. But I don't hear much about Stony Brook from Lyme patients.

Note: I think with Stony Brook you have to request that they report the bands that were non-CDC significant bands (or at least this used to be true?). Don't quote me on that.

michele said...

Yes,you need to request non-CDC significant bands when dealing with Stonybrook lab. My daughter's labwork was sent to 3 different labs - Stonybrook being one of them- Take care Michele