What is the normal value of the neutrophil-to-lymphocyte ratio? (2017) Forget, De Kock, et al.

Background
Neutrophil-to-lymphocyte ratio (NLR) has proven its prognostic value in cardiovascular diseases, infections, inflammatory diseases and in several types of cancers. However, no cut-off has been proposed on the basis of reference values coming from healthy population.

Methods
Routine blood samples were obtained (n = 413) from workers (age: median 38, range: 21–66 years) involved in a health care prevention program, to determine means, standard deviations (SDs), 95% confidence intervals (95% CI), percentiles P2.5 and P97.5. A second independent sample of healthy volunteers is compared (n = 29).

Results
The mean NLR is 1.65 [±1.96 SD: 0.78–3.53] (95% CI [0.75–0.81] and [3.40–3.66]). In the second cohort (healthy control), the NLR values are in the same range, whichever the used analyzer. No NLR assessed in the validation series is out of the proposed limits.

Conclusions
We have identified that the normal NLR values in an adult, non-geriatric, population in good health are between 0.78 and 3.53. These data will help to define the normal values of the NLR.

Electronic supplementary material
The online version of this article (doi:10.1186/s13104-016-2335-5) contains supplementary material, which is available to authorized users.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5217256/

 

I haven't heard of this method of checking for inflammation previously.

 

Nor have I. The authors of this paper have no idea what they are doing - yet again. Normal ranges with percentiles are of no particular value in establishing levels of clinical interest. The level of rate at which you get gout is well within the 'normal range'. But it matters because it is the level at which crystal deposit and cause arthritis.

 

How is it meant to be worked out what value of some finding has clinical interest?

 

It entirely depends on what the clinical interest is. So for blood pressure it is a matter of demonstrating risk of stroke or heart attack, for sugar it is a matter of showing risk of diabetic complications, for neutrophils the risk of infection and so on. And of course there may be several different risks to consider.

So centile-based 'normal ranges' are not 'healthy ranges'. They just give a statistical reference for the population.

 

But what do they do with a marker like CRP, which could be associated to a variety of different sorts of conditions?

 

When you say 'they do' I guess you are referring to the lab that reports the result? 'They' will give a range based on centiles for a reference healthy population. But the physician who is familiar with using the test ignores that and makes a decision on clinical interest based on the context. That will depend on what question the clinical is asking. It might be whether the CRP suggests any inflammatory process that is unlikely to be just due to the normal daily insults that put the CRP up and down in the reference range - in which case the reference range at least gives the rough probabilities. Or it might be a much more specific question where a lower or a higher figure was relevant - relating to possible TB or the activity of someone's RA.

One of the problems with CRP is that people are calibrated very differently. Levels of 1-10 in one person equate to 3-30 in another. When following patients long term you can make use of that,but not for one off measurements.

 

That's what I suspected. Thanks for explaining.

This case actually illustrates that. The paper in the OP establishes a reference value that incorporates as "normal" many of the values flagged in the papers cited as being warnings that a procedure may not go well, or something else adverse might happen in various specific populations.

The reason I picked this specific paper to post is that it referenced many different populations being studied, so one could see that the test might have broad applicability (although it might be hard to guess what it might predict in a new population--and the prediction value is poor--but in an area where there are no useful tests it seems to be better than nothing at showing "there's something going on weirdly with the immune system, and something could possibly go wrong").

Thus it might be interesting to add to ME studies, knowing that it would be a very general test more like CRP than like anti-ds-DNA.

 

I don't follow why this 'test' which is a ratio of two meaningful tests with different implications, should be of any extra value.