I recently had a number of medical tests, one of which was erythrocyte sedimentation rate (or ESR, or send rate). My test result fell in what would be considered the normal range using the “modified Westergren method” but there was a note under the results stating that it was done using “the Alifax ESR instrument” which is not based on the Westergren method, and that sensitivity and specificity might be different from Westergren. It said to “please refer to new established reference ranges.”
I’ve been trying to understand this, and so far what it looks like to me is that the Alifax results are pretty prone to inaccuracy*, but maybe it’s good enough? But if I’m understanding correctly, my “normal” range number under the Westergren scale has a plus or minus range that could take it from almost the minimum number to 5 or 10 points into the high range.
My doctor told me the result was normal, but I don’t know if there even is a reference range to determine that. I can understand the statements in the papers I’m reading that say you can’t use these results interchangeably, but much of the rest is over my head. Anyone want to take a shot at explaining it, and or finding the applicable reference range?
Here’s an example of the kind of info I’ve been finding:
My understanding of the sed rate (however it’s performed) is that it’s a relatively insensitive and nonspecific test which is only useful in making a diagnosis when combined with other tests as well as clinical evaluation and physical exam. It can also be used to monitor the course of some inflammatory conditions.
I suggest discussing your concerns with your physician. He/she may tell you that a normal or even slightly elevated value is not a significant concern, nor does the sed rate alone establish or rule out a diagnosis.
My understanding is that it’s very non-specific, but it’s a decent measure of overall health. An abnormal sed rate means something is wrong. The something still needs to be diagnosed.
The issue that I’m asking about, though, is that for the method used, there doesn’t seem to be any established normal range, so the number is completely meaningless.
I have and will continue to talk to my doctor.
It just seems to me that given the method that was used, my doctor may not have been able to say that my result was “normal” or not.
I’m curious too about how labs can adopt a new way of measuring something like this without having the data to establish a new reference range or close correlation with the established range.
And it likely then listed a reference range of normal for test results on that instrument, as standardized on that instrument. Which may be a different range of normal than the range of normal on a different instrument.
IOW it is a heads up to the clinician that a result of say 30 on this instrument is not necessarily the same as the same number on another one. 30 may be considered elevated on one of them and not considered elevated on another. The result of “within normal range” or outside of it means the same, but the exact numbers may not track. This is important when the ESR is being used to monitor an identified condition over time.
It is separate issue than the crudeness of ESR (often mocked as “estimated sick rate”) and the relative advantages and disadvantages of other markers of general inflammation as screening tools for certain sorts of conditions. Speaking as a clinician the ESR as a screening tool is only useful if it is very high. A little bit high for the specific instrument doesn’t mean much.
To more directly address the reference range question, this study may be helpful. It looked at an (the?) Alifax automated system in a varied patient population, and lists reference ranges (“normal” values). They cite 0-20 mm/h as normal for men and 0-30 mm/h as normal for women.
As you can see from the data, “normal” values can be considerably influenced by lifestyle and age. For instance, an elderly, sedentary, obese smoker who doesn’t drink will likely have a higher ESR than someone of similar age of normal weight who exercises regularly and drinks in moderation. Throw in other potentially important confounding factors and you can see why establishing “normality” for a given patient involves clinical judgment as well as lab precision. As suggested previously, an ESR value that’s a little high is probably, in and of itself, not a cause for worry.
Interestingly, laboratorians (a term I loathe, but whatever) still consider the manual Westergren method of determining the sed rate as the gold standard, despite its venerable status, potential variation in performance/interpretation between labs, and findings that automated systems may be more accurate and reproducible.
I think the issue is that Alifax is measuring something different, which allows it to essentially estimate what the sed rate would be using Westergren, which turns out to not correlate super well. And it correlates the least well at high levels, where it’s used to monitor disease and treatment. It’s being adopted because Westergren is a manual test that is read after an hour, and Alifax is an automated test that takes 20 seconds. But sed rate is based on the amount of red blood cells that have reached the bottom of the container in an hour. (Roughly described). So the 20 second test may be more convenient, but it seems like it is providing way less information because there’s a very big plus or minus variation from the “actual” sed rate using Westergren.
Got it. The reference range is listed, and now that I have more info, I know that it’s different from the Westergren range. So instead of a reading of less than or equal to 30 (I think) being normal, it’s less than or equal to 39.
In my case, it was testing for something specific – GCA – where it would be expected to be very elevated.
I strongly think GCA is the issue, but my sed rate and CRP came back normal and barely elevated, respectively. It looks like my sed rate is low enough that the difference between tests wouldn’t matter much.
From what I’ve read, GCA can still be present with normal labs, but I don’t know if I should be pushing for more testing. I had a period of 3 weeks with very painful and concerning symptoms, but they stopped a few days before I saw my doctor. I don’t know if I should wait for them to come back before asking for more testing – I don’t want to risk blindness, but I also don’t want to have more negative test results if they are based on bad timing.