What is the range of % of active B12 in a serum B12 test

[Ernest12]

I seem to remember that normally about 20% of the total serum B12 in blood test is active, but I also think that I saw a range quoted of between 7% and 20%.

Anyone have a medical document ref on the subject that I could use to convince my doctor?

[fbirder]

Never seen anything accurate.

It would be difficult, because the serum test can give a range of more than 20% when testing the same sample. And Active is about +/- 5%

So, if the true values were: total 100, Active 20.

Your ratio could range from total 120, Active 19 (ratio 6.3:1 or 16% Active) to total 80, Active 21 (ratio 3.8:1 or 26% Active).

[Ernest12 in reply to fbirder]

Thanks for the info.

I found where the 7% to 20% active figures came from : a paper called "B12 Deficiency in the Body", link on this site somewhere.

With your accuracy figures of ‎+ or - 20% for measured total serum value, then if total serum B12 measured value was 500, say then Active B12 could be somewhere between 600 x .2 = 120 and 400 x.07 = 28. Quite a big range!

Medichecks say that a "safe" level for Active B12 is 37.5.

So a safe value for a total serum B12 test might be somewhere between

37.5 x (100/7) x (120/100) = 643!

and 37.5 x (100/20) x (80/100) = 150!

and note 150 would only be safe if it was -20% accuracy against the real total serum value and 20% of the total serum value was active.

No wonder a lot of statements are made about the inaccuracy of total serum B12 tests in determining B12 deficiency ... and the importance of the reported symptoms of the patient.

Of course this just fiddling with figures to get an idea of the implications. A proper statistical treatment should be made, i.e. does your +or- 20% on the total serum value represent a standard deviation about a mean value derived from a normal distribution of test results, and is the 7% to 20% range of active B12 a similar SD about a mean value of a normal distribution etc. Basically the probabilities of each level of inaccuracy needs to be involved in the calculations.

I'll ask my son about it in the next couple of weeks. He has a PhD in Bayesian statistics... I expect he'll say that there are so many assumptions that any conclusion would be fairly meaningless. But what the hell, it's only people's wellbeing.

Perhaps the only conclusion to be drawn is that using 150, as I think I have seen quoted, as the safe lower limit is very dangerous.

[fbirder in reply to Ernest12]

If your son said that then he would be correct.

There is little point counting the number of B12 molecules that can dance on the head of a pin. The problem isn’t the accuracy (or inaccuracy) of the actual tests. The problem is that there’s a big overlap between the range where people might be deficient and the range where people aren’t deficient. See my post here - healthunlocked.com/pasoc/po....

That’s why people who know about B12 say that the only gold standard for a B12 deficiency is - does the patient respond well to B12 treatment?

[FlipperTD]

There are multiple problems here!

'Active B12' measures B12 bound to Transcobalamin; Serum B12 measures total B12 after chemical release of any B12 in the serum, so that includes 'bound to Transcobalamin', [active] and 'bound to Haptocorrin'. As far as I'm aware there's no reliable data on the ratio of Transcobalamin to Haptocorrin in the serum, so I would expect it may vary quite a bit. On top of that, any test we have has a degree of imprecision, and some assays are more precise than others. Given that we have two different analytical processes here, each with its inherent imprecision, then you'll end up multiplying entities, and coming up with even more imprecision. As fbirder says, does B12 replacement help, and that's probably more important than comparing different assay methods. That's a sensible place to start, but it's hard for me to admit, having been responsible for hundreds of thousands of Serum B12 assays by various methods!

Good luck.

[tomdickharry]

Can't say for certain as different labs get different results. I recently had an active B12 of 180, then, the next week and a different lab, total B12 was 800. I've seen percent active quoted as anything from 20% to 35%. The NHS only tests total via GP but will do paid tests of active. Doesn't make sense to me.