I’ve now heard from multiple people that their doctors or endocrinologists have told them laboratory tests for oestrogen levels are either inaccurate or cannot detect exogenous oestrogen, so there’s either no reason to test or no reason to take action based on test results. This always seems to be used to deny increase in hormone doses, but for decreases the blood levels are always trusted. Interesting that…

What are we measuring anyway?

Oestrogen isn’t one substance, but the oestrogens we’re interested in are oestrone (E1) and oestradiol (E2). In general E2 is the predominant oestrogen in adults and the most common form for HRT. In New Zealand we only have tests for oestradiol available, which lines up nicely with what we want to measure.

What is HRT oestrogen?

All HRT oestrogen is oestradiol. There are two common forms of oestradiol in New Zealand, oestradiol valerate - the ester of valeric acid¹ and oestrdiol - and oestradiol hemihydrate which is pure oestradiol with extra water in the crystalline structure (yes I’m vastly simplifying this) ².

Oestradiol valerate is a prodrug - it is not biologically active, but it is metabolised in to oestradiol and valeric acid in the body ³. There’s no need to measure oestradiol valerate as it’s not active, only the oestradiol it’s metabolised in to.

Oestradiol hemihydrate doesn’t need any metabolism, it’s already oestradiol. The extra water is broken away as the oestradiol hemihydrate dissolves in to your bloodstream.

In both cases the biologically active component is identical to oestradiol produced in your body, so would be detected by any system that detects oestradiol.

How do we measure it?

For the full details you’ll have to read Challenges to the Measurement of Estradiol: An Endocrine Society Position Statement ⁴, but here’s a TLDR. The best method is a complicated system of isotope dilution/gas chromatography coupled with mass spectrometry but that’s far too slow for practical use. Most modern systems use immunoassays.

What’s an immunoassay?

Chances are you’ve used one! The general concept is the same as the COVID-19 self-test I’m sure many of us have used, except with far tighter tolerances so actual numbers can be derived. The basic principal is an antibody is developed that attaches to the desired molecule - oestradiol in this case - and the number of matches is counted. This is done by introducing a second competitive molecule that attaches to the same antibodies but fluoresces, then you can measure the light given off to determine the amount of antibodies that weren’t bound to the oestradiol. If you know how many antibodies you included in the first place and the amount of light each molecule will emit you can calculate the total oestradiol there must have been! Simple!

Simple?

Not really. There’s a bunch of consideration here. The affinity of the antibodies for oestradiol, the sensitivity of the photosensor, the variability of fluorescence, how consistent the reagents are in antibody count, the potential for things other then oestradiol to also bind to the antibody, and a whole bunch of other things I don’t understand.

Fortunately modern laboratories can quantify these problems and produce error ranges for their assays. However there are limits to the detection, and different assays are often not comparable - that is if you use a single assay type you can compare the numbers, but if your lab changes their system then the numbers can’t be compared.

What assays does NZ use?

It’s hard to tell, labs don’t always disclose this which is quite annoying.

From what I’ve been able to find New Zealand’s labs have standardised on Roche assays ⁵. I found the Roche technical documentation on their oestradiol tests, and while I can’t be certain this is what the labs use it lines up with their stated performance and machines.

This assay can detect down to 18.4pmol/L, but can only accurately quantify levels over 91.8pmol/L. Even then their data tables show the measurements are only accurate to ±7% at 100pmol/L, but this improve to ±3% at 200pmol/L and ±2% by 500pmol/L. So a constant level may fluctuate slightly over multiple readings, but it’s good enough to establish trends over multiple tests. A single test with an unexpected level should be repeated though as there is a remote possibility for a bad read (However depending on the records the lab has about your gender they may have already done this). Additionally the amount of oestradiol in your blood will vary based on activity, time of day, diet, all sorts, even on a fairly constant dose system like patches.

So can we measure HRT oestradiol, and accurately?

Yes, we can detect HRT oestradiol. It’s identical to regular human oestradiol.

Yes, the tests used in New Zealand are accurate for levels over 100pmol/L. There will be variance because the tests aren’t exact, but they are accurate enough to establish trends - if your oestadiol levels are going up the results will trend upwards, or vice versa. Or be around the same values.