HCG with Dr. Vivienne Meljen
/Today we welcome Dr. Vivienne Meljen, senior resident at Duke OB/GYN, to talk to us about one of her favorite topics — bHCG, the “pregnancy hormone.” There’s a lot to unpack — so these notes are broken down as our Q&A conversation!
What is hCG?
Human chorionic gonadotropin or hCG is part of a family of glycoprotein hormones including TSH, FSH, and LH.
These are hormones produced by the pituitary gland. hCG included!
They are each heterodimers and have an alpha and a beta subunit. The alpha subunits are identical and beta subunits are all a bit different
In fact, luteinizing hormone (LH) and hCG are super similar and come from shared genes, so when LH is being made in large amounts, hCG is also made and may be elevated.
HCG is metabolized by the liver and kidneys.
Where does hCG come from?
Traditionally we think of hCG as being produced by “a pregnancy”. The part of a pregnancy that makes hCG is the trophoblast - what will eventually become the placenta.
It also comes from the pituitary gland as we mentioned earlier. And some cancers can make hCG
What does hCG do?
To understand what hCG does, we need to back track a bit and remember how a pregnancy starts.
To get to hCG production, there must be fertilization and then implantation of the pregnancy. This process of reaching implantation takes about 7 – 10 days as the embryo floats through the tube to its final implantation site.
Once it has implanted, the trophoblast makes hCG which then stimulates the corpus luteum to keep making progesterone, which helps continue to support gestation. At about 10-12 weeks, hCG levels usually peak and thereafter, the placenta takes over hCG production.
This is apparent in patients’ thyroid testing. In normal pregnancy with an asymptomatic patient, TFTs could be slightly abnormal at this point due to the cross-reactivity and homology between TSH and hCG.
So someone can be pregnant, but not necessarily have hCG? Is that right?
Yes, while that blastocyst is making its way to implantation, the corpus luteum is making its own progesterone and there is no hCG yet so a pregnancy test will be negative. In order to minimize variation in ability to detect a pregnancy, we should time pregnancy testing to be done at 15 days after LH surge, instead of testing at the time of expected menses.
This is why in contraception management if we cannot reasonably exclude pregnancy, we recheck in ~ 2 weeks because we would “capture” most pregnancies at that point.
I know there is hCG, but I always hear about “betas” or “checking a beta”. What’s that about?
Like I mentioned, hCG is a heterodimer with an alpha and a beta subunit. Most people say “check a beta” meaning “check a quant”. The reference to betas is actually due to a misunderstanding back when the hCG immunoassays were first being created, and a bottle of antibodies was labeled as “beta subunit” and it was interpreted as the test only testing for beta subunits. In fact, most hCG tests are detecting all different types of hCG when you get a quant.
What can we use hCG testing for?
Detecting pregnancy
Tumor marker monitoring in molar pregnancy and GTN
Screening for trisomies in pregnancy with quad screen for example
What can we use hCG itself for?
One of the evidence-based and legitimate ways to use it is in the REI world for ovarian hyperstimulation for ART. HCG is the “trigger shot” to help mature eggs because it simulates an LH surge due to its homology.
Note- by the time you’d want to test for pregnancy in a woman undergoing ART, this should have cleared so it shouldn’t affect testing.
What’s the normal rate of rise for hCG in a normal pregnancy?
We use serial hCG concentration measurements as a tool to help us differentiate normal from abnormal pregnancies
Back in the day, the rule of thumb was that hCG would rise ~51% every 48 hours.
Now we know that hCG rate of rise depends on the initial value with the % rise being greater at lower values. For an initial hCG level of < 1,500 the expected rate of rise is > 49% versus starting over 3,000 the expected rate of rise is in the ~33% range. Most normal pregnancies will rise faster than this.
What about hCG levels going down?
While we have some clear guidelines about a 15+% decrease in hcg levels between days 4 and 7 following mtx administration for suspected ectopic pregnancy, it is less clear in other situations.
There is no specific rule, but per ACOG, following an SAB hCG levels should normalize within 2-6 weeks. This likely depends on the initial value.
What can cause a positive pregnancy test?
Pregnancy (normal, abnormal, ectopic)
HCG doping (people use it for weight loss and for sports)
A slew of possible causes of false positives we will get into later
What’s the deal with urine and serum testing for hCG?
Urine tests are QUALITATIVE – positive or negative
Most Serum tests are QUANTITATIVE – give you a number amount usually in IU/L
Urine tests will typically turn positive at the same point at which a serum test is > 20-25. However, you can have a false negative urine study if the urine is very dilute and a woman’s hCG level is low.
Most of these test employs antibodies that sandwich hCG when present and then are converted into a test result.
Because they use antibodies though, some other things can cause antibodies to link and create a false positive or negative result that you need to be aware of.
On a typical urine test or a “UPT” seeing 1 line is a sign that the control and test works, 2 lines means hCG is present.
You can get a false negative test with a rare situation called the “hook effect” when you have SUCH a high level of hCG (think GTN levels) that the sandwich antibodies are saturated on both ends and can’t link together to give a signal.
False positive tests seem to be fairly common and weird. Can you take us through those?
Heterophile antibodies are often referred to as “phantom hCG”. These are nonspecific, low affinity antibodies that can crosslink the antibodies in the test and make it seem positive.
Where do these antibodies come from? - Some people at greater risk of these heterophile antibodies are people who have worked on a farm or in a veterinary facility, women with rheumatologic conditions, patients who have rec’d recombinant antibodies for medical treatment, some who have received plasma exchange from an unknowingly pregnant donor. You can filter out for these by cross checking with urine that should be negative, trying a different assay, or having your lab perform serial dilutions or use a heterophile-blocking agent.
Pituitary hCG is often seen in perimenopause or post-chemotherapy. This makes sense because of the homology to LH. So when LH is high, hCG rises with it. You can suppress this by giving OCPs to see if it normalizes in a few weeks.
“Chemical pregnancy” is another potential “false” but not really false positive. This is an implanted conception that produces hCG but results in a SAB by the time of expected menses.
Familial hCG is a bizarre and rare cause where there is a hereditary cause of hCG production that can be present throughout ones entire life at low levels.
Kidney disease is one of the more common causes I’ve been called for and this is a tricky one. The exact mechanism for why patients with ESRD have false positive hCG is not known but we think its related to impaired renal clearance and increased gonadotropin levels. The tough part is that these patients often do NOT produce urine so they get serum testing which if positive, we can’t cross check with urine to see easily if it is a heterophile. Often, these patients are tested in the setting of prepping for transplant and sometimes, in the final stretch before going to the OR for a new kidney; so a positive test can really complicate things. To make matters worse, they often have complicated menstrual histories so it is very hard to cross check timing of a possible pregnancy. It’s estimated that approximately 1.5% of dialysis receiving reproductive-aged women conceive over ~2 years so it’s possible. Usually, using a good history one can work this out and often these patients have a series of low positive quants over the course of years to help support that it is a false positive test. Some authors suggest using a good old fashioned progesterone level to help clarify the situation, though it isn’t fool proof.