RESEARCH ALERT: Surprisingly high gestational carrier (GC) and neonatal morbidity in GC cycles
By Norbert Gleicher, MD, Medical Director and Chief Scientist at The Center for Human Reproduction in New York City. He can be contacted though The Reproductive Times or directly at either ngleicher@thechr.com or ngleicher@rockefeller.edu.
BRIEFING: A recent study in a prestigious general medical journal reported surprisingly high adverse maternal as well as neonatal complications in pregnancies of gestational carriers (GCs). Dr. Gleicher in this article builds on this recent publication to address the rapidly increasing phenomenon in the U.S. of GC pregnancies.
Despite rapidly increasing costs, the use of gestational carriers (GCs) – by some also erroneously called “surrogates” (a surrogate also contributes her own egg) – is steadily gaining in popularity and, indeed, has become its own “industry,” with new agencies popping up almost daily. The underlying assumption for most GC cycles is that the genetic mother because of medical reasons cannot or should not carry a pregnancy and, therefore, chooses a GC to carry a pregnancy in her place. A second and rapidly increasing indication for the utilization of GCs is in offering fatherhood to single men and/or male same-sex couples.
Especially in more recent years, the utilization of GCs has, however, expanded beyond just medical and biological indications. Increasingly – and often even widely publicized by celebrities – the indications are social, and we have also noticed an uptick in GC utilization among female business executives.
Whatever the indication for the use of a GC may be, an underlying assumption in choosing a GC, of course, will always be that a GC is fertile (to offer the best chance of pregnancy with transfer of embryos) and healthy (to offer the best chance for a normal pregnancy outcome). GCs, therefore, must be very carefully selected and not only have a clean past medical history and family history but, ideally, should have given birth before, should neither be too young or too old, and be socially responsible individuals. Assuming a proper GC selection process, one should therefore also be able to expect that GCs not only should conceive quickly but almost universally should also experience uncomplicated pregnancies.
But is this really true? And if not, why not?
Interestingly, this had never before been systematically investigated. A very recent paper by Canadian colleagues on the subject appeared, out of all places, in the Annals of Internal Medicine (1). And this paper likely made it into this prestigious general medical journal because it surprised!
From among all eligible singleton births the Canadian colleagues investigated, 846,124 (97.6%) occurred through unassisted conception, 16,087 (1.8%) through IVF, and only 806 (0.1%) using a GC. The risk for severe maternal morbidity was 2.3% with unassisted conception, 4.3% with IVF, and 7.8% in GC cycles. Weighted relative risks generated through modified Poisson regression models were 3.30 (95% CI, 2.59 to 4.20) comparing GC cycles with unassisted conception and 1.86 (CI, 1.36 to 2.55) comparing GC cycles with IVF. Respective risks for severe neonatal morbidity were 5.9%, 8.9%, and 6.6%, generating weighted relative risks of 1.20 (CI, 0.92 to 1.55) for GC pregnancies versus unassisted conception and 0.81 (CI, 0.61 to 1.08) for GC pregnancies versus IVF. GCs also demonstrated more hypertensive disorders, postpartum hemorrhages, and preterm births at less than 37 weeks in comparison to both comparison groups. Though GCs demonstrated more preterm births, the evidence of severe neonatal morbidity was less than the evidence for severe GC morbidity and adverse pregnancy outcomes.
The authors concluded that potential mechanisms for higher maternal morbidity among GCs require elucidation and that GC cycles require development of “special care plans.”
They are to be congratulated on having published a very important study which will likely have a great impact on who uses GCs and how GCs are selected. Women who for medical reasons have no choice and men who need a woman to carry their child, of course, have also no choice. Both groups will, therefore, continue using GCs. But we can see social indications for GC utilization somewhat declining. And while the authors, of course, are correct in that underlying causes for poorer outcome of GC cycles should be further investigated, we find this argument a little bit too correct politically.
There, indeed, is no big mystery hiding behind the findings pointed out by the paper. In principle, we know of two reasons. One is poor GC selection, a subject we will return to, and the other is the GC’s immune system. And here is why: When a woman spontaneously conceives, the pregnancy is a so-called 50% allograft. This means that half of the genetic background of the pregnancy is maternal, and the other half is paternal and, therefore, allogeneic. In a GC pregnancy the situation differs: For the GC, the pregnancy is 100% allogeneic because the GC has zero genetic contribution to the pregnancy. And 100% allogeneic pregnancies (other examples are pregnancies established with donor eggs and the father’s semen or with donor eggs and donor sperm) are well known to experience increased pregnancy complications for the person carrying the pregnancy and the offspring of such pregnancies, as recent paper well reviewed (2).
One, of course, can also not preclude that women who for medical reasons may need a GC may be producing “poorer” eggs than most other women, but women undergoing IVF should have represented an appropriate control for such a possibility in the recent paper. Moreover, if egg quality was the reason, the difference, likely, would have been primarily seen in neonatal morbidity and not GC morbidity, as the study suggested.
In a sense, this paper, however, did not surprise. Especially since COVID, the quality of GCs appears to have significantly declined. The CHR therefore recommends to all patients that they allow for an interview of their GC with a CHR physician before making any downpayments to an agency. Though we noticed a rise even before COVID, rejecting GC candidates was then a relatively rare event; but after COVID, rejections have almost become the norm. For example, very recently we discovered that a proposed GC was a diabetic.
With patients in a large majority of cases finding their GCs through GC agencies, one must hold the GC industry at least partially responsible. Young, fertile, healthy, and socially responsible GCs should, themselves, have lower than average complications in a pregnancy and should have lower—not higher—complication rates in pregnancies than unselected control populations. Though as noted above, carrying a 100% allogeneic pregnancy undoubtedly increases risks to GC and offspring, the observed rate in increase still appears excessive and, therefore, very likely also includes a GC selection contribution.
Does this mean—as the authors suggested—that special care plans must be developed for GC cycles?
To a degree probably yes, but one plan that fits all, of course, makes little sense, considering what all can be wrong with a GC. What definitely has to change, however, are four aspects of current GC practice: (i) As already noted, GC selection must be tightened; (ii) The informed consent given to patients using GCs must include the new recognition that GC cycles do not necessarily produce low-risk pregnancies; (iii) GCs must be informed that GC pregnancies likely carry higher risk than the pregnancies they already experienced with use of their own eggs; and (iv) obstetricians must also be made aware of the fact that GC pregnancies must be expected to be more complicated than prior pregnancies of the GC with her own eggs.
And, yes, the rapidly growing GC industry must acknowledge all of these new observations and must recognize that growing demand for GCs cannot be satisfied by lowering the criteria for GC selection. To the contrary, with growing demand, the responsibility of GC agencies in selecting only the best qualified GCs actually increases. Their efforts at GC selection, therefore, must be tightened.
References
1. Velez et al., Ann Intern Med 2024; doi: 10.7326/M24-0417. Online ahead of print.
2. McCoy et al., Early Hum Development 2024;196:106072