Re: 23-weekers -- I suspect this varies quite a lot by hospital. 23 weekers are right at the stage where the lungs are differentiating from a solid mass of tissue to something with air sacs, so some 23 weekers have nothing to inflate with oxygen yet. Those who do would require specialized equipment (particularly very very small equipment) so to some extent it would come down to whether or not you were at a very high-level NICU and then it would depend on the developmental stage the lungs were at. As I understand it steroids cannot act on the undifferentiated tissue, because it hasn't yet developed the proper receptors.
Re: tests -- Standard of care will take a while -- in particular, they'll need to see if administering the test improves outcomes! It seems to me that it is likely to do so, if it allows time for steroids, but it's always possible that babies will be delivered earlier in response to a positive predictive test and that this will lead to poorer outcomes than not administering the test. So yeah, further study.
Here's the abstract
for the journal article I find the most interesting. :) It says 1) The PlGF/sEng ratio and its delta and slope had an excellent predictive performance for the prediction of early-onset preeclampsia, with very high likelihood ratios for a positive test result and very low likelihood ratios for a negative test result; and 2) Although the positive likelihood ratios are high and the positive predictive values low, the number of patients needed to be closely followed is 4:1 for the PlGF/sEng ratio and 3:1 for the slope of PlGF/sEng.
This is of course math talk. For those who don't speak math: "ratio" just means the difference between the two. One part PlGF to two parts sEng might be normal, but one part PlGF to fifteen parts sEng might not be. "Slope" means "rate of change between two points" -- in other words they will compare the levels from your 12 week bloodwork and your 24 week bloodwork and if they are changing faster than some cutoff it will meet diagnostic criteria. So for example -- in completely imaginary numbers -- if sFlt levels doubled, that wouldn't be diagnostic, but if they quadrupled, they would. And "delta" means that if the rate is accelerating, curving upwards on a graph, at some particular rate, that you'll be diagnosed. For that they would need three points, to make a curve, so I'm assuming this means MOAR BLOODWORKS.
It's also possible to use these tests just once, as a "spot test", with an absolute cutoff. Say 99.5% of people with levels above 6 widgets are severely preeclamptic -- so they'd check your bloodwork to see if your levels were above 6, and if they were they'd treat you as if you were a severe preeclamptic. This would be a lot faster than waiting for a 24-hour urine.
So as I understand things, the idea is to use sFlt and sEng to develop a test to predict preeclampsia onset before 34 weeks gestation, also usable as a confirmatory diagnostic test in the hospital. But it won't be usable at 24 weeks to predict PE at 37 weeks. Still, at that gestational age you'd just deliver; what we really want to know is who's going to get sick early, so we can watch them closely and get steroids on board before things go truly pear-shaped.
Also, the reason I keep saying "sFlt" and the abstract says "PlGF" is that soluble Flt binds to Placental Growth Factor, so measuring one just is measuring the other. IIRC it was easier to measure PlGF because they already had an assay for it.