Welcome to Fertility and Sterility On Air, the podcast where you can stay current on the latest global research in the field of reproductive medicine. This podcast brings you an overview of this month's journal, in-depth discussions with authors and other special features. FNS On Air is brought to you by the Fertility and Sterility family of journals, in conjunction with the American Society for Reproductive Medicine, and is hosted by Dr. Kurt Barnhart, Editor-in-Chief, Dr. Eve Feinberg, Editorial Editor, Dr. Micah Hill, Media Editor, Dr. Pietro Bortoletto, Interactive Associate-in-Chief, and Associate Editor, Dr. Kate Devine.
Good morning and welcome to another edition of Fertility and Sterility On Air. We are in October 2026 and you wouldn't know what that means. That means it is ASRM season and we are excited to see all of you in Baltimore.
Come swing by the ASRM booth where we'll be recording Fertility and Sterility On Air. We are volume 126, number 4. I am joined by my partners in crime, Pietro, good morning sir, how are you? Hi Micah, I love October when it comes to podcast time, lots of content coming out in the month of October, post-meeting, some of the best on-the-fly interviews, finding people who give killer talks that you otherwise wouldn't know about unless you've attended, lots of stuff to look forward to on the podcast coming up. Can't wait.
Kate ma'am, how are you today? I'm great Micah, thanks for asking. I love October too. Halloween is sneakily my favorite holiday and then also sneakily this issue of FNS for October is absolutely killer.
I think we're only going to be able to cover the tip of the iceberg today but everybody should dive in. It's one of the best issues I can remember. Good stuff and last but never least, our intrepid editor-in-chief, Kurt, how are you sir? I am well.
I also have a very much an autumn person, like the crisp air, helps you read FNS, then you go about your work and it's crisp and nice again in the evening and you read FNS again. And it is also football season. Unfortunately, we have a lot of Philadelphia Eagles here on the talk today but we'll put up with that as we get to the science.
Go Berners, they pulled it out again last night. Barely over the Titans, yes. All right, we're gonna dive right in.
Here we go with the views and reviews by editorial editor Catherine Rakowski and some experts on uterine transplantation including Branstrom and many others. Really good views and reviews on balancing innovation risk and access. I think, you know, this is a nice unbiased sort of look at the nuances that are involved as we balance innovation risk and access to a very complex surgical process such as uterine transplantation from the worldwide experts.
Very well worth the read. We have a reflection from editorial editor Anuja Dokras and experts Teed and others on reflecting the journey of renaming PCOS to PMOS. I still can't even say it quite right.
I still fumble over it with patience. I don't know if you remember this, Kate, but when you and I were fellows way back in the day at the NIH, they had a meeting at the NIH on PCOS and the thing that came out of that meeting is we need to rename it. And Anuja was there and I don't even remember that because I was a fellow and I didn't know Anuja, but it literally took that long.
It took a decade and a half for this to have expert consensus to rename it. We can debate whether the change in the name matters or not. We're all old enough to remember like cervical incompetency being renamed cervical insufficiency.
We're old enough to remember premature ovarian failure being primary ovarian insufficiency. Has that actually moved the needle on patient outcomes on research dollars? To me, that would be the big critique. We can all debate the name.
We'll see if this moves the needle or not. But the reflection is really interesting on a 15-year labor of love from Dr. Dokris and others on renaming this. I got to say, Micah, just in the last month or two, I've had patients wink at me when I say PCOS.
They're like, they're just looking at me waiting for me like, oh, and of course, it's now PMOS because they know about it. It's on social media. It's on their TikTok.
It's on their feed. It's been really, really cool to see something that came out in our literature make it to the lay press so quickly and so easily. And if someone, if I never hear the words again, I was told I have lots of cysts on my ovaries.
It will all be worth that 15-year wait. It'll be worth just that, yeah. The big difference is this one really has reached the patients.
And it's almost made it more empowering for our patients that they feel that they can do something about it. So I'm really pleased that this one has garnered so much attention. So read Anuja's article.
It was a long time coming. I think it was necessary. Certainly not going to hurt, but it'll make you in the know.
So make sure you, from now on, say PMOS. That's fascinating. I think that the difference in terms of the quick dissemination to our patient population is completely attributable to social media.
I mean, it's amazing. That's a positive difference, I think. Yeah.
Compared to the examples I gave, this may just be the new times. Back in the examples I gave, you would have had to have access to journals that were making these changes, and most patients didn't. So the age of social media.
So please read that. A long labor of love from Robert Norman, Anuja Dokris, Teed, a lot of other very smart people and leaders in our field. And last but not least, ASRM Pages has diagnosis and treatment of luteal phase deficiency that is out.
I read through that. There are no major changes from the last one. I think the big take-home point is similar to before.
You do not need to measure a luteal progesterone to give patients a diagnosis of luteal phase deficiency. We still get patients that come into our office. I do, probably monthly, and ask for this.
Remember, progesterone is highly pulsatile because LH is highly pulsatile with a 20 to 40 minute half-life. So measuring a progesterone, if you don't like it, check it in an hour. You'll like that one better.
I guarantee you it's going to bounce and regress back to the mean. It is a clinical diagnosis. So a luteal phase less than 10 days is a clinical diagnosis and could be something to look into.
To me, the biggest change, and this is the second practice bulletin in a row that has done it. If you remember, for those of us who are old, three practice bulletins ago, you have to go back about a decade and a half, said the clinical irrelevance of luteal phase insufficiency and being on those practice committee discussions, we felt like maybe that title sort of squelched the impetus to do research into that, and so we don't want to do that. And this practice bulletin is saying we should do research into this.
We shouldn't ignore research into it. We just don't have clinical evidence that measuring a single progesterone level is what's going to give you that diagnosis. So that's the high-level take-home point, but for all the fellows out there, please, every month, read every practice committee front to back.
Pietro? I had an anecdote on this. I told a patient who was asking me the exact same question that I'm sure all of us have been asked before. It's like, oh, can we measure my progesterone level? I think my progesterone levels are low in the luteal phase.
And I was like, listen, for me to get an honest sense of what your progesterone levels look like, I need to measure progesterone level every hour for 24 hours and do an area under the curve. And she's like, okay, when can we set that up? I wasn't recommending it. I was trying to dissuade her from the single progesterone measurement, but boy, was she motivated to find that out.
That'll teach her. Yeah, never again. Oh, that's a good anecdote.
Kate and I actually have done that at the NIH. We have admitted people for Q1 hour blood draws for 24 hours, and you can get that area under the curve, but that's a lot to do from a clinical diagnostic standpoint that has a low yield. I'll fax over an outside monitoring order to Shady Grove.
Oh, man, I shouldn't have opened my mouth. Thanks, Pietro. All right, so let's move on to the research.
Pietro, you are up first. We're going to go straight into the world of andrology. We're going to talk about male factor infertility and male age and what that means with outcomes in IVF.
Thanks, Mike. It's not always that I get a male factor paper. My wheelhouse is usually the uterus, the endo, and the AI, but here we are.
So this article is entitled Male Factor Infertility Diagnosis, but Not Male Age Significantly Affects the Yield of Usable Embryos in Donor Oocyte Cycles. This is by first author Anat Chemerinsky and friends of the podcast, Sarah Morelli and David Shin from New Jersey, both Hackensack and Rutgers. So donor oocytes from an investigational perspective are some of the cleanest natural experiments we have, particularly when we're trying to isolate certain things.
In this case, we're trying to isolate the sperm's contribution because egg quality presumably is held roughly constant in the donor oocyte population. The title says male factor matters and male age doesn't, but the data potentially tells a bit more of a nuanced story. Let me dig in a little bit more with you.
So embryo aneuploidy is overwhelmingly a maternal age story. The sperm contribution is small, and we've talked about this in previous podcasts, likely in the low single digit range. But fathers are getting older, and there's still no agreed upon definition of what advanced paternal age is.
You could draw that line in the sand at 40, 45, 50, 50 plus. There's been lots of prior work on this question of paternal age and embryo ploidy, and it's conflicting. Most of it, unfortunately, is small, single center, uses autologous eggs.
So you have a kind of an entanglement of male and female age here in the data. The male factor literature has the same problem. Plus, it also has a lot of older technology.
If we look at what the male infertility world has contributed, it's stuff from the era of cleavage stage biopsy, fish, array CGH. So still very, very tough to draw meaningful conclusions from. So there's been this big gap where we needed something that was large, modern, donor egg only looking.
So this is exactly what the authors decided to do. We have the wonderful benefit of having the SARTCORS data available to us in the United States for answering some of these big questions. They looked at 2017 to 2022 data.
They specifically look at non-identified donor oocyte IVF cycles with PGTA, donors 33 or younger, blastocyst stage only, and the excluded shared donor cycles. This gave them about 10,000 cycles, 70% of which were fresh oocytes, 30% of which were frozen oocytes. And it's a study that was well-powered for older men, about 5,000 cycles in men between the ages of 40 to 49, 1,700 cycles in men 50 and over, only 150 cycles in men under 30.
There are two outcomes where embryos biopsy per cycle and embryos, quote unquote, suitable for transfer per cycle. This is an important caveat. SART does not capture euploidy rates.
So suitable for transfer is defined by each clinic and is a presumably a proxy for euploidy, although it's not the same thing. So what do they find? With regard to male age, there was no difference across age groups in embryos biopsy. The median was five in every age group.
And there was also no difference in embryos suitable for transfer. The median was three to four across all of the groups. So no statistical difference across the different age thresholds.
Male factor, the same number were biopsied across the different age groups, or excuse me, in the presence of male factor and non-male factor, and the median was five. But one fewer suitable embryo was found in the median, three versus four. In the adjusted models, any male factor diagnosis was associated with about 4% fewer embryos biopsied and about 4% fewer suitable embryos.
Age, male age, showed a very small adjusted effect on suitable embryos, roughly 2% per decade. So while the authors call it clinically significant, it is not clinically meaningful, 2% per decade. One important thing that we should talk about here is that the yield versus the ploidy question in this study is really important.
So male factor reduced embryos biopsied and embryos suitable by the same amount. That means that the proportion of BLAST deemed suitable did not change. This looks like a BLAST yield effect, not a ploidy effect.
So the conclusions language about the impact on embryo ploidy goes a bit further than the data, because again, SART does not capture ploidy rates. And then the other thing that's really important to know is that there are some big missing variables. So fresh versus frozen, oocytes was the strongest determinant of embryo number in the data set, and it was not adjusted for.
Neither were oocytes retrieved, neither was fertilization method, or the use of ICSI. So what does this mean for, I think, our patients? So if you're reading this and you're wondering, what do I tell my patients who are coming through my door and interested about this question about older paternal age, at least in older male partners in a donor cycle, it seems to be reassuring. Any age effect on usable embryo appears to be trivial, and no change in approach is warranted.
You do not need to replace a older partner for a younger partner. Male factor, I think it's reasonable to counsel that there's a modestly lower blast yield, or probably not a different euploidy rate per blast. Again, this is a dark spot in the paper that they can't definitively tell us by nature of SART not capturing ploidy data.
Now, one thing that I think we've talked a lot about in this podcast is the ongoing use of PGTA and donor oocyte cycles. It still remains very common in donor cycles, despite the explicit guidance against it, and recent registered data showing no benefit from it, and potentially even lower live birth rates in fresh donor cycles. Micah, Kate, Kurt, why do we continue to use PGTA and donor egg cycles? Is there something that we're missing that patients want to know? Is there something that we're not counseling patients sufficiently on? I think it's the latter.
So many folks are not counseling patients sufficiently. The only scenario I see where it's rational is maybe, maybe, maybe in patients that are pursuing a surrogacy journey, just because there's a supply-demand problem with qualified gestational carriers, and they also may be not counseled as well as we'd like, and see that as an advantage in terms of a family that they'd like to work with. So that's just, I think, an idiosyncrasy that I hope that we will overcome as a field.
But otherwise, I really, it's extremely frustrating to me. But boy, is that an exceptionally small group, though. Donor egg, gestational carrier, I can think of a couple dozen patients that I've seen over the course of many years.
The lion's share of patients are using it autologously in their own uterus, but still really leaning into PGTA as a tool to help them conceive. Kurt? PGTA has such a good publicist, that the story is so intuitive, and that the layman's description of how it improves your pregnancy rate and gives you a healthy pregnancy is just so easily grasped that patients want it. And there's not a lot of incentive to talk them out of it from an REI practice.
So I think that's where we stand, is that we let them say it's a really good thing, and we say, well, okay, if you want it, we'll do it. Rather than taking the other approach, which is, this really isn't helping you, it's not cost effective, and for all we know, it might actually be less effective. So it's a difficult counseling tool.
We actually have a program, and we're digressing a little bit from the main thrust of this paper, but where we have a money-back guarantee, a refund for patients based on live birth from donor eggs, or the egg bank that we work with does this. And they actually have PGTA as an exclusion for the refund program. So one small step towards evidence-based counseling.
Although I saw that there's also a paper in press in FNS that suggested that doing PGTA on cycles using donor eggs does not decrease, or sorry, does not increase the number of transfers needed. So maybe we'll review that one in a future podcast. But there's obviously- Correct, correct.
But that's not the title of the paper. Right. All right.
Well, good stuff. Thank you, Pietro. Pietro, we're sticking with you, and we're moving on into epidemiology.
So you also, you not only got male factor, you got epidemiology, even though Curt is on. So we're throwing you some curveballs here. Listen, this is epidemiology, and we're talking about diet.
We are not talking about GLP-1. So stay tuned. I know that's all the rage, and hopefully those papers will come.
This is a paper entitled Preconception Dietary Patterns in Association with Live Birth and Pregnancy Loss, a Couples-Based Approach. This is an article from PhD Kyle Busse from UPenn, and our friends of the podcast, Drs. Enrique Shisterman, Jenny Ryan, Erica Johnston, James Hotailing, and Sunny Mumford.
Micah, just like the question that we get about, do I have cysts from patients? Should I use PGTA and donor eggs? The other question I get a lot is, is there anything I should be taking or anything I should be eating? That comes up at nearly every patient visit, and there's a whole industry of fertility diet books and programs that have been ready to answer that question on our behalf. Every new patient consult, they pretty much ask this question to us. Totally.
So this study puts two purpose-built fertility diets, and I'm using air quotes here for our listeners, up against two ordinary healthy eating patterns. And it does a really interesting thing where it looks at both partners, because we know it takes two to tango, or at least it can take two to tango. The short version is that the boring advice of eat healthy, do your best holds up at least as well as the specialized dietary advices.
But notably, the effect sizes are small and nothing diet-wise really changed miscarriage rates. That's the punchline if you stop listening at this point, but I hope you continue to stick with us. So plenty of individual foods and nutrients have been linked to fertility outcomes, but no overall dietary pattern has enough evidence behind it to be a slam dunk recommendation.
There are two fertility-specific patterns that have been proposed. One is this fertility diet, which came out of the Nurses Health Study 2, and was built around ovulatory patients with infertility. And then the pro-fertility diet came out of an IVF cohort and was built around ART outcomes.
Prior studies of dietary patterns have mostly been small, single center, IVF only, female only. And of course, as you can expect, the results have been mixed. Male pattern diets have mostly been against semen parameters, and they've rarely studied it against live birth outcomes.
So there's always been this natural gap of a large couples-based cohort across the full range of fertility treatments with, again, the most important outcome, live birth, central to it. This podcast loves the secondary analysis of other data. We have so many wonderful papers that have looked at rich data sets to try to come up with some more important conclusions for our field.
So this paper is actually a secondary analysis of the FAST study, which is the NICHD Folic Acid and Zinc Trial in Men, which our listeners may recall was null, plus IDEAL, its companion cohort of female partners. So this was nearly 2,300 couples across four U.S. centers in Utah, Iowa, Chicago, and Minneapolis that were enrolled between 2013 and 2017 that were planning any kind of fertility treatment. And of that cohort, about 20% went on to IVF, 40% went on to IUI, 17% had ovulation induction, and a quarter had no treatment.
Each partner at the time of enrollment answered 125-item food frequency questionnaire covering the prior three months before they presented for treatment. And they gave patients four scores per partner. The HEI 2015, in alignment with the U.S. Dietary Guidelines, the Mediterranean Fertility Diet, and the Pro Fertility Diet.
And the couple's scores was the sum of both partners. Importantly, no one was assigned to a diet. So adherence here means how closely reported intake happened to line up with each dietary pattern.
And their outcome, like I told you, was live birth and pregnancy loss before 20 weeks. So what did they find? 44% of the couples conceived, 35% of them had a live birth, and 287 had a pregnancy loss. And when we look at the healthy patterns, in each standard deviation that you increased adherence to the Mediterranean or this HEI 2015, there was an association with about a 5% per 6% higher relative likelihood of live birth.
The estimate was the same whether you looked at her diet, his diet, or the couple's diet. So in absolute terms, this was roughly a live birth rate of 35% going to 37%. The fertility diet showed the same size association in women, but nothing really in men or in the couples that went analyzed together.
The Pro Fertility Diet, there was no association in any individual group. And like I told you at the beginning with the punchline, there was no association in any dietary pattern with pregnancy loss. When you did the subgroup analysis and you looked where these signals might have been strongest, it's kind of exactly where I thought it would end up.
The signal was strongest in patients with BMI 30 or higher, in women with PMOS or anovulation, and it was present in patients undergoing OI and IUI, but absent in patients with IVF. Now, one important thing that if you're paying good statistical attention to this paper that I want to draw the reader to once they land on the PDF is that each confidence interval touches or crosses one. So the abstract mentioning that there were associations with is a bit more confident than the number suggests.
These are small signals, not firm findings. There is also this healthy user effect and access to care. These are patients who are generally leaner, more educated, had higher income, less likely to smoke, and patients more likely to get IVF treatment than their counterparts.
So it's always one of the natural confoundings when you're doing these dietary adherence studies. And finally, the people in this cohort were pretty good prognosis. So the mean female age was 31.
Only a quarter of women carried an infertility diagnosis and a quarter of couples were actually never treated and were able to conceive. So essentially, the cohort that this data is derived from is an early in workup, good prognosis group. Now, what do I do with this data? Next patient shows up and asks me, what should I eat? I think it's reasonable to tell him off the findings in the study that a general eating pattern for both partners is reasonable.
It's low risk. It's good for pregnancy and long term health. The fertility benefit is real, but it is small.
Imagine a 2% increased live birth rate is an important number to counsel. But there's no support here for a restrictive, expensive, specialized fertility diet above and beyond just general, ordinary, healthy eating. And the best diet is the one that you can stick to.
I think we can reasonably tell patients that healthy eating, said broadly with air quotes for the listeners, is a reasonable approach and they don't need to spin their wheels on the expensive dietary changes that some find very taxing in addition to the stress of infertility. Micah, Kurt, Kate, what do you guys tell patients when they ask you, what should I be eating? Are you a eat healthy, all good things in moderation counselor, or are you specifically pulling out a diet book or pointing them towards a resource? First thing that you said, and I, you know, I hate it when I like studies that reaffirm my biases or my practice patterns, but this one did. That's exactly what I tell patients, Pietro.
I mean, you said it very well. Be healthy, do what you can do to exercise well, eat well. But beyond that, I don't think a specific fertility diet has strong evidence and the systematic reviews on this, the meta-analyses on this pretty much confirm that.
And so what the patient, like you said, the best diet is the one that you can stick to. So eat healthy, be healthy, live healthy. When we start doing our fertility treatments, we start manipulating physiology sort of to the extremes of biology that we can, and from a physiologic mechanism, it's unlikely that a specific one diet or another is going to change that when we start manipulating things through IVF or IUIs or the other things that we do.
So I like it when things challenge my beliefs, but in this case, it reaffirmed what I think. Kurt, what were your thoughts? So I agree. The big picture is, I don't think the diet really is that important, but I think when you're talking to an individual patient, a lot of the art of medicine comes through.
There are a lot of people that come in and they think they're eating healthy and they think their diet is the problem, so they want to change. And sometimes, you know, telling them that this is the diet to use or having them see a nutritionist, which has far more credibility than I do, actually is what they're looking for. So in the back of my mind, I may know that a thousand pages that I see, it really doesn't matter, but it doesn't take away from the individualization that some people really want expert advice and really want to follow a pattern.
So it's never going to hurt, I guess, and people can certainly drink grapefruit juice for too long or something like that that can hurt them. But, you know, it's a balance is what I'm saying. Yeah.
You're saying it's a potential impetus for change for a healthier diet for them that they're going to use that moment to, yeah, that's great. And it's an opportunity for control, which has a huge psychological benefit. I continue to be shocked by how important that is for folks.
I mean, even with the new recommendations about check DFI for patients who have RPL, I get frustrated with that because like, what are we going to do with that? And I spoke with, you know, my friend Ruth Lothy and she's like, oh, it's because it's an opportunity to counsel on lifestyle. I mentioned that to a friend and I said, isn't that so or to a patient rather? And I said, isn't that so frustrating? She said, no, it's great. You know, I love to have something that I can control.
So, you know, it's an important thing not to underestimate as much as, as a from a scientific perspective, it can be somewhat frustrating. It's an interesting psychological comment from Kate and Kurt there. I love that.
Yeah, but Micah, you're right, though. It's reassuring to know that if I forgot to talk to people about diet or didn't get no specific diet, I'm not harming them anyway. And ultimately they're still OK.
So it's just it's just a question of how you're looking at this. I love your point about involving a nutritionist or a dietician in their care. I think with the rise of the telemedicine platforms, insurance coverage, at least in Massachusetts, has been excellent for involving some of these folks.
And you're right. My bandwidth is small and my expertise is narrow when it comes to dietary and behavior counseling. So involving people with the expertise is a great way to connect the dots for them, hopefully in an evidence based way.
But I think the big takeaway, if you're listening to this and want to do the quick and dirty counseling is any of these diets seems to be reasonable. And the diet that you can stick to is the diet that's likely to move the needle for you the most. But if resources are needed above and beyond that, then, like Kurt said, involving some of these other allied health professionals can be a really powerful tool for restoring a semblance of control and purpose for patients who are on their fertility journey.
It was a fantastic summary of a fantastic study from some of the top experts in our field in the world of EPI and a trial from FAST. So really good stuff. Kate, we're sort of moving from the macro of sort of diet, high level view into the micro of science and something that's actually really important to us when we talk about PGTM.
And we're going to talk about allele dropout and the study that we have looking at that for monogenetic disorders. So tell us what we learned from this study this month. Yeah, so I personally learned a ton.
And I just want to say at the at the front of this discussion that I love this paper. And I think it's a really great paper for every fellow and every practitioner and REI that practices and uses PGTM for their patients to help understand it well, which is so important, especially at a time in our field where things are likely to be changing rapidly. And so this paper is Systemic Assessment of Allele Dropout in Preimplantation Genetic Testing for Monogenic Disorders, Incidence Detection and Clinical Testing Strategies with first author Zhang and colleagues from Shanghai.
And this is also an esteemed author group with, you know, multiple other high quality publications on this topic. So I'm so happy to see this paper in Fertility and Sterility. So just kind of high level, I wanted to step back and remind folks, because sometimes, you know, we get our results.
We're not thinking about everything on a micro level, as this paper calls for. So what is allele dropout? So allele dropout is where only one allele is amplified from the two parental haplotypes. And this is particularly important in the setting of PGTM, where we are trying to assess a small amount of DNA from embryos to determine whether the pathogenic allele or alleles are present.
So if we're only seeing one allele, and particularly if we're only seeing the wild type or quote unquote normal allele for a particular disease state, then this would put that patient at risk for a false negative in their biopsy sample and their PGTM results, potentially exposing that family to an affected child. And so any new technology that we evaluate for PGTM, allele dropout really needs to be front and center as a phenomenon that we are hoping to avoid as an artifact and limitation of the testing. And so what these authors did was to endeavor to assess two different methodologies for PGTM in order to assess the risk for allele dropout in each methodology.
So they looked at 516 families, and this is really kind of, I think, a benefit of their patient population that they had really amazing, an amazing population to work with and amazing adherence, and we'll get to that in a little bit. They had 2,949 embryos, and some of those embryos, they were looking for multiple variants, so 3,942 variants, a lot more variants than embryos even, and 188 genes at 425 loci. So they took a rigorous approach.
They obtained peripheral blood from both parents, as well as from at least one additional relative. They used a whole genome amplification methodology that I had to educate myself on, which is called multiple displacement amplification. It's an amplification methodology that is often used specifically for PGTM, but a little bit different than what's typically used for PGTA.
And what they assessed was which of these two methodologies was more likely to be subject to the pitfall of allele dropout. And so what they found was that for 11% of all loci in their entire population, allele dropout was identified, and that NGS was better than Sanger sequencing. So basically, when they used NGS methodology for sequencing of the pathogenic variant, they only found allele dropout to be present in 4% of the loci versus 7.4% with Sanger sequencing.
So more likely to have allele dropout with Sanger sequencing. And this is a clinically relevant question. Different laboratories are evaluating which technology to use here.
So I mean, if you're running a PGT lab, this is even more relevant for you, but for us SREIs prescribing these technologies, it's important to understand as well. There were only 28 of the 425 loci for which they observed allele dropout that both technologies had allele dropout. So in most cases, one or the other was able to get a comprehensive result.
One of the things I really liked a lot about this paper and a way in which it's a real privilege for these investigators to be doing this science in such a compliant population is that they asked them all to get amniocentesis, and they all got amniocentesis. So 322 amniocentesis that were 100% consistent with the PGTM results. So 100% specificity there, which is really kind of nice to see.
They did transfer two embryos in this population where they had observed allele dropout. So with presumably appropriate counseling of the patients that might not be a comprehensive result for them. But neither of those two embryos resulted in pregnancy.
So these were all embryos where they were able to give a clear result. They also did a nice analysis as to what were the characteristics of some of the genes that were more prone to allele dropout. So they found that the two most common genes were polycystic kidney disease and GJB2.
And what they noted there was they went on to, again, assess the characteristics and they found that a higher GC, so guanine cytosine content of the genes, as well as poorer whole genome amplification quality, so that part's not surprising, were associated for a higher risk of allele dropout. Fun fact that I also had to remind myself of, GC has three hydrogen bonds versus AT has two. So the reason for that is that there's more hairpins and loops in GC-enriched content, which results in a higher energy of activation for PCR and then subsequently worse quality amplification.
So interesting to note that some of those cases could be mitigated actually with specific optimization of the case. Really interesting discussion then of things like the allele frequency in the algorithms that they were using, as well as sequencing depth and how those things might be able to be optimized in challenging cases. And I learned a lot, but the take-home message here for labs is for the overwhelming majority of PGTM cases, NGS is a better technology in terms of accuracy and completeness of the results from an allele dropout standpoint.
Sanger sequencing does have its place in some specific challenging cases and really just a good read all around for updating your understanding of PGT technology. I really think that we all need to understand this so much better than most of us do. And I think that the reason it's even more important than ever is that there are so many emerging technologies that tout the ability to do accurate PGTM and accurate detection of lots of other types of variants by direct detection.
And that is the holy grail, right? If our patients don't have to have these multiplex probes set up, you know, obviously it's a much smoother, faster process. We can get a lot more information a lot more efficiently, but we got to make sure that all of these technologies are being validated appropriately as we consider them for our patients. So we're entering into a new era.
I was so again, maybe slightly out of my wheelhouse. Pietro, we're all kind of did like musical chairs on the areas of expertise in this one. But I also learned so much and would highly recommend this paper to our readership.
What did you guys take away from this one? This is a killer fellow paper. If you're really trying to understand the what is happening behind the tests that we are ordering and the pitfalls of it. Boy, is this a good one to read? I don't think this is ever going to come up on your board.
So it's not that kind of utility of test. But man, does it make you a stronger clinician to understand some of the pitfalls for these kinds of things when we're doing PGTM? Killer, killer paper. And like you said, Kate, it's really the only kind of thing that could be done in China.
100% amniocentesis rate. I really like this paper for a couple of reasons. And one was, I think as our eyes, we really do need to know the genetics of what we're doing.
We can't just know this kind of hand waving that we send cells off to a genetic labs. They give us a result and then we act on the result. And what struck me was how different this this PGTM is from PGTA.
I mean, it's just incredibly different. It's got different pitfalls. It's got different methodologies.
It's got it's got really different technical limitations. And when you think about really the goal of avoiding a disease in somebody with PGTM, I mean, this this kind of detail is necessary. And I'm glad it got worked out.
I do think it's a must read for everybody in our field. While we don't do it every day, knowing the difference between what we do more every day, like PGTM and the difference of techniques makes you a better REI and a better geneticist. And we should know this stuff.
So I'm glad it's an FNS. I recently had a patient that, you know, we're running a study right now looking at some I would still consider experimental PGTA technologies. And it was a patient who has a Wnt10a mutation, had previously done PGTM with more standard methodologies.
And and this lab felt very strongly that they were going to be able to detect this mutation with direct detection, you know, without essentially a probe setup. And I ultimately think they were correct. But, you know, I felt really good about being able to have a rigorous conversation with this lab about what was the risk of allele dropout for this patient.
And they they talked me through how they were able to, you know, satisfy themselves and the patient that they were going to have a high degree of accuracy. But these things are going to come up. They're going to come up.
You know, the future is now. So read the paper so that you can understand it. The one thing I struggled with is should I feel reassured by this paper or not reassured because they say only 6.5% had allele dropout concordant with both tests.
But to me, I would want it less than 1% when we're talking about PGTM. But then when they did the fetal amnios, that wasn't any different. So clinically, I guess I'm reading that as there wasn't the manifestation of bad clinical outcomes because of that allele dropout.
So as a clinician, sort of what should my take home be? Should I be concerned? Should I feel reassured when I'm ordering these tests? What was your as you struggled through this paper? What was your thoughts on that? My strong suspicion? Well, they said that they didn't transfer any embryos that had had allele dropout and they were able to assess whether there was allele dropout or sorry, they transferred to and they did not obtain a pregnancy. And so in our practice modality in the United States, if we have when we refer the case to the PGT lab, if they don't have that greater than 98% accuracy in most cases, they're not taking the case. So here they just kind of went ahead and did all of them.
It seems like or that part might not have been reported. But the good news is most of the labs that we're working with are able to identify a priori whether they're going to have high accuracy or whether the particular allele is prone to allele dropout. And maybe you need to look for another lab that uses a different methodology.
But there are still some PGTM variants that we can't detect as reliably as we'd like. And I agree with you. I'm not reassured in that case to tell the patient that they should be doing IVF just for this reason.
That 30,000 foot view is helpful. Kurt, we are moving on to you. We are coming to infertility and peripartum hysterectomy.
An interesting combination. Probably need a really large sample size to look at a question like this. And we're kind of in your wheelhouse because this I think with this kind of sample size is an epi question.
So what did we learn from this study? Thanks, Micah. And this is one of my to coin a phrase, let's let this data flow over you because there's lots of this. You're not going to take this paper and memorize the number.
So this is the title of this paper is infertility treatment is associated with peripartum hysterectomy, a population based cross-sectional study. The senior epidemiologist is Kande Anath, who's done a lot of good work in this space. And it's really one of these things where you can kind of it's what is our field doing or what is the infertility treatment doing to the population more so than what you're going to tell your patients specifically what the risks are.
So I think we all know the story. You know, the children conceived with IVF have different outcomes than children that are conceived without assistance. Now, those perinatal outcomes have gotten and have been improved dramatically over the last decades, most likely due to the fact that we have single embryo transfers and we have eliminated twins.
But there's still this question about, is there a difference in the physiology when you use assisted reproduction versus unassisted? And this paper is taking it beyond just that. It's basically saying we should also be considering the maternal health, not just the child's health. And there's been a lot of studies on maternal health as well.
And this one is specifically looking at a very large database, about 80 percent of the country in terms of hospitalizations in the country, and using peripartum hysterectomy as a surrogate for maternal health. In other words, is there something about the conception that is affecting the mother and ultimately in the hysterectomy, even though it's a relatively rare outcome? So we're going beyond just intrauterine growth restriction, preeclampsia, placenta accreta, although some of these things are intertwined and obstetrically. So this study got a hold of a database, which is, again, most hospitalizations in the country, 16.7 million hospitalizations.
It's looking at data from 2000 to 2020, of which 85,000 plus were coded with infertility treatment. Now, I'm going to get to this because it's an important part of the paper and a teaching tool. That's only 0.5 percent of deliveries, and I think we would argue that that's probably an underestimate.
But just hold that number for a second. And then they basically looked at the peripartum hysterectomy as the primary outcome. And they noticed a difference.
There is roughly 400 per 100,000 deliveries for infertility treatment versus around 150, actually 180 per 100,000 deliveries for unassisted conceptions. Now, that's a raw number. There's lots of confounding.
I had a really good time just reading table one of this paper, which just shows you the differences between the populations of who use and conceive with ART versus who don't. It really is everything we've learned for the last 30 years. There's difference in whether you're urban or rural.
There's difference in age. There's difference in race. I mean, it's just it's just very obvious that the people who use ART are very different than the rest of the country.
We'll just leave it at that. So you have to control for that. And they did a nice job.
And I won't bore you with all the specifics of how they controlled work. But after adjustment from maternal age and race and ethnicity and multiple gestation, socioeconomic factors, hospital characteristics, the infertility treatment, they found a significant increase in hysterectomy for women who conceived with infertility treatments at around 33 percent. When they took only ART, IVF took out the infertility treatment themselves.
The risk is around 40 percent higher. Again, statistically significant. So this corresponds to use Micah's favorite statistic, that the number needed to treat or the number needed to harm in this case.
So one additional hysterectomy for every 413 already associated births. That's not insignificant. Our labor for might not see that, you know, that much, but that's a real number to pay attention to.
And again, they're just taking this as a surrogate for effect on the mother and maternal health, not that we're worried about necessarily that relatively rare outcome. So the increase was was they did a nice job trying to figure this out. And through some sensitivity analysis, the increase was seen across all racial and age groups, all that, although, as we would expect, the absolute risk was a lot higher for African-American because their baseline risk for hysterectomy was so much higher.
But the relative risk was the same in all in all racial groups. You could not explain it by just what are called obstetrical complications. In other words, they said, well, maybe it's because you have a credo or maybe it's because you have postpartum hemorrhage that that's the reason you got your hysterectomy.
And when you when you look at the placental complications as a whole, like they could also find you had a placental complication, the risk was still 40 percent higher. When you looked at patients that had no associated obstetrical complication like a credo or postpartum hemorrhage, it was still 28 percent higher. So this suggests suggested again that the placental abnormalities and hemorrhage didn't fully account for the observed differences.
Then what's the intermediate? That's a biological guess. We're not sure. The point was it's not just a quote unquote causal pathway.
You know, ART, abnormal placenta, postpartum hemorrhage. There seems to be more than just that. They also found a weaker association with infertility compared to ART.
And that's interesting, too. And again, it's just really a thought provoking experiment, which is why it's air. So you can look at it two ways, that there's some risk even in a patient who has infertility, even if they don't have IVF, or you could argue it that IVF confers even greater risk on top of infertility because of the consequences of IVF.
Now, we've had that debate before. That's nothing new. So let me give you a little more of the complications of the study.
Again, this is not like any epidemiologic study. You can't take this as gospel. There's always limitations.
There are some limitations in the database and that we can't look at things like fresh versus frozen. We really couldn't get a deep dive into diagnosis for what was the reason for the infertility. There was some limit in BMI.
But the biggest limitation I do want to just talk about, because I found it a kind of neat analysis, and I learned something from this analysis. Remember, I said that the incidence of infertility in the database was only a half a percent, and that seems low. So they understood that it's misclassification.
Now, it's possible the misclassification was differential, meaning if you had a hysterectomy, you're more likely to code infertility. That would bias you towards a higher relative risk, but that doesn't seem right. But if it's a non-differential misclassification, meaning you're just missing in most people, then it's going to clearly bias you to the null and a relatively small relative risk.
There are other studies that have looked at this in much smaller databases that found relative risks of five and six versus hysterectomy. So they kind of have a study that is still significant, but under what other people have performed. So they did something called a probabilistic bias analysis, which has got a great name.
It's actually not all that difficult. They assumed that bias, the misclassification was non-differential, and they basically did kind of Monte Carlo simulations by changing the sensitivity, how often that you listed IVF or infertility on your birth certificate, higher and lower, and specificity, how often you correctly said there was no infertility treatment. And they changed those assumptions kind of randomly within ranges and ran, you know, 500,000 simulations to get an idea.
If this wasn't it, and we assumed it was random, we did it a whole bunch of times, half a million times, what really would the answer be? And the answer came up quite a bit to around 2.8, which is more in line with what some of the others said. So they recognize that they have a great database, huge database from our country, but they realize there's limitations with it. And what I think is wonderful to see in these kind of papers is not just reporting the crude answer and trying to interpret to others.
They actually really did try to, torture is the wrong word, be very rigorous in their analysis to look at it in many, many different ways to say, is this answer as robust as we can get it? So the other thing you should read is Tom Molinaro wrote a wonderful inklings. Now, Tom Molinaro, my former fellow was a real, he did the master's degree with us. He was a great epidemiologist and he's taken all those skills and runs now a multinational IVF practice.
You can see how it translates directly. So, but I can see from writing this perspective, he does still have a little epidemiology in his heart. And he recognizes that we shouldn't just be looking at IVF success rates.
We shouldn't just be looking at perinatal success rates. We should deem success as optimal maternal health, not just neonatal, avoidance of severe maternal mortality, preservation of future fertility, and a full-term singleton birth without major complications. So he's really ringing true to us that while we're in the weeds trying to help every single patient and keep our success rates, the global health picture is IVF really helping the population should still be our concern to some degree.
So the bottom line for this paper is that pregnancies following infertility treatments, particularly ART, were associated with a modest but clinically increased risk of peripartum hysterectomy. Why is that important? It's a marker of maternal health in general. So there is something going on about maternal health after they conceive with IVF that we should be paying attention to.
The absolute risk remains rare, but it does support the findings of, again, the population health, attention to maternal outcomes, risk assessment, and it's still, in my mind, got probably to do with it starting as early as placentation in the first trimester, all the way through the third trimester, and then something about delivery that is just not working. It's very close to normal, but it's still just not normal. So that's the flow over to you now, that remember we're physicians in the globe, in the country, not just like what happens to the patient in front of us, and this is another well-done study that shows us there's still a lot for us to learn.
What did you guys think of this? You're going to tell your patients now you have a 30% increased risk of hysterectomy? I'm not going to tell the patient that, Kurt, but I am going to highlight the importance of having a nice handoff, a discharge from our practice to the obstetricians. I think we are the ones who understand where the risks are. We understand that tough retained products that we had to excise.
We understand that cornea that's obliterated. We understand that they conceived on a thin lining. We understand that it was a frozen ember that was biopsied.
We have bits and pieces of this increased relative risk that we have at our disposal that if we're not doing a good job communicating to our obstetricians who are taking care of these patients, it's potentially a missed opportunity for risk reduction, for studying, for increasing people's whiskers and making sure that they're paying attention to that sticky placenta or that excess bleeding at the time of delivery. And my bias, of course, is my wife is an MFM, so I hear about it all the time from her, but this is not so much a conversation that we will have with patients, but I'm hoping it's a conversation we will have with our colleagues who are inheriting these pregnancies we're conceiving. Yeah, to put it back in perspective, Pietro, you're right.
I just reviewed an editorial in the New England Journal about the role of IVF in population, and I don't know the number in the United States. It's probably three to five percent now, but there's some countries where they're approaching 10 percent of the population is with ART. So this is no longer a niche.
This is not clearly just women that can't get pregnant. This is a large percentage of the population. So it's going to spill over to population health, not just how good is your IVF lab.
So that's what I think the purpose of this paper is. I just want to give kudos to these authors for elevating us as REIs in the U.S. I mean, they're making use of this huge U.S.-based dataset, right? And we see so many of these come out of Scandinavian countries. Our data may not be as clean, but that doesn't make it any less important.
And they were able to look across races and ethnicity. I think this paper is a huge contribution. And while, as you so eloquently explained, Curt, we cannot infer causality even from this well-performed analysis, we need to stay hard on ourselves that IVF might be causing some of these adverse outcomes that ART might be.
My mind went first to, OK, how do we improve frozen embryo transfer protocols to reduce some of these complexities? But I mean, this study goes back from 2000 to 2020. Most of this is not FET. So, you know, we really need to be thinking hard.
There's so much more that we don't understand. And it's these big papers that help us recognize that we still have a lot to learn. Great.
Thank you, Curt, for breaking that down. You, sir, disarmed all of the comments that I had with your comments beforehand. So I have nothing else to say about it.
That was a great presentation. I enjoyed it. You knew what the things that I were going to say, and you addressed all of them.
So I really enjoyed that. I let the science wash over me, as you well advise me to do wisely sometimes. All right.
We have one more paper. We're moving on to the world of epidemiology. I love the title.
We're deconstructing the triglyceride to glucose body mass index in normal weight women with PMOS. This is from Daniel Domicic, senior author single out of UCLA. So to remind those of us who haven't looked at this in a while, I hadn't looked at this ratio for several years.
This is triglyceride times glucose. So both fasting divided by two times the BMI. And so it's sort of a way if you're thinking about what the heck does that mean? Micah, it's a way of looking at insulin resistance without doing something like the HOMA-IR or a euglycemic clamp.
The HOMA-IR being fasting insulin times glucose divided by 405. The 405 is just to normalize it to one. Big picture though, all this math, all that means is we're trying to get an idea of how much insulin resistance is there in a patient.
And in this case, we're normalizing it to BMI. And so they specifically looked at PMOS women who had a diagnosis of PMOS, but were normal BMI. So we're taking obesity out of the equation here.
So that's our patient population. And what are we looking at? Well, we're looking at their outcomes with IVF. So do they do the same in IVF as women who are similar in their BMI, who do not have this diagnosis of PMOS? So they match the patients to this.
My one critique, and I kind of wish the reviewers had pushed them a little more, is they told us they matched to age and BMI, but they weren't specific in that. So any of us who have done matching studies know you have to make a decision and how you're going to match. Is it within one age? Is it the first patient? Is it within two years, three years, five years? And there actually ended up being some differences because the ages ended up being off by about two years.
BMI was 0.5. So that got pretty close. And then they ended up with 21 patients in the study group and 19 in the control in the matches. So if you're matching one to one, it should be 21 to 21, but you end up with 21 to 19.
So that's kind of in the minutia. It's a minor critique. I don't think it affects how I think about the study.
I just wish the reviewers had pushed them a little more to tell us, hey, how exactly did you do the match? Why did we end up with unequal numbers? But from a high-level perspective, essentially, again, what they're looking at is lean women with PMOS and do they have insulin resistance based upon this marker that's a little bit less invasive to measure? And does that affect their IVF outcomes? There's some things I like that they did that I thought were really smart. They only matched for age and BMI. The reason they did that is they only had 21 and 19 patients and they only adjusted for those two things in their models.
And the reason being they just don't have a lot of patience and sometimes I think, you know, back 10, 15 years ago, we saw people in studies, maybe not doing enough regression analysis, not adjusting for enough confounders. But now what you sometimes see is people just throwing the kitchen sink at it and they put in 20 things into their regression model. Well, when you have 21 and 19 patients and you're going to adjust for a lot of things, sometimes that actually makes your model less accurate, not more accurate.
So sticking to the things that they think will actually matter for this study, they decided age and BMI a priori. I think that was really smart. They also didn't adjust for things that could be collinear because they had a lot of data based upon measurements that they did in this study that could cause collinearity.
And when you have collinearity, if you're thinking about your DAG analysis and how you get from your effect through your causal pathway to your outcome, if you're measuring things that are collinear that are essentially affecting themselves, that can also mess with your model. So from a statistical standpoint, while I made a couple critiques at the start, I actually think this was a really smart design based upon the data that they had. So ultimately, what did they find? As you would expect, there were some differences in women with PMOS versus the controls.
So these are age matched within a couple of years and their BMI matched within 0.5. Their LH was higher with PMOS. Their total testosterone, free testosterone was higher. Their androstenedione was higher.
Their free fatty acids were higher. Ultimately, what's really interesting is they also had higher levels of insulin resistance when measured with actual markers of true insulin resistance. And so the take home from that is this ratio, and this is consistent, I guess, with other literature, is that this formula of fasting triglycerides times fasting glucose divided by two times their BMI is actually predictive of insulin resistance in women with PMOS and women without PMOS who are undergoing IVF.
But at the high level, the big picture is that when you look at their IVF outcomes, there were no differences. Now, you could reasonably say, Micah, we had 21 patients in one group, 19 patients in another group. Can we actually detect IVF differences and outcomes? And I would say, no, we couldn't.
And the authors did not try to hide that. I actually really like this line from their discussion. Accordingly, the multivariable model was intentionally parsimonious and domain guided, so its results would be interpreted as exploratory and hypothesis generating rather than definitive, where smaller effect sizes cannot be excluded.
So they conclude ultimately that this ratio predicts insulin resistance, but it does not predict IVF outcomes. But larger studies might find that. And I just highlighted and specifically read that comment because I think a lot of times there's this misinterpretation in fellows and people trying to publish within FNS that you have to force a positive outcome, something that is going to drive the field forward.
And I think that's actually the opposite of what you will see in how FNS reviews papers. We want you to be critical and thoughtful about how you review your data. A negative finding, such as this one is, is actually helpful.
And acknowledging that you were underpowered, but you did this deep analysis and do something that's endocrinological, that we don't have a lot of other data on, actually helps move the field forward and help generate hypotheses for future studies. If these authors had taken the opposite approach and tried to force a positive finding and spin it in that direction, I think it probably wouldn't have been published in FNS. So the reason I read that sentence that they wrote is I thought it was a very elegant sentence that sort of framed and contextualizes how they approached this study.
So I'm interested if you guys have any thoughts on that. It's kind of a niche study, but one that I think definitely belonged in FNS, and I was glad to read it and see it published. Is this the first PMOS titled paper in FNS after the name change? Or have we had one before? I think we've had one other one since at least on the podcast that I've noticed.
So they're starting to come. People are adapting already. An early signal.
All right. Well, as always, we highlighted just some of the articles that in this month, it was Eve Feinberg who selected our studies. And I just want to say congrats to Eve.
This is my last podcast as the media editor. Eve will be replacing me in that role. So congratulations to her.
After five years of me replacing Kurt in this role, it's time for me to move on to something else with the journal. And I'm excited to do that. But we are very excited to have Eve get into this role and lead these going forward.
As always, we have a lot of other articles in the journal. We have some really interesting letters to the editors, some fantastic video articles. I don't recommend that you use this podcast to supplement reading the journal.
So dive into the journal this month. As Kate said, it is a fantastic month that Kurt and all of the editors have curated from the amazing authors around the world. Keep submitting to us.
We encourage you to do that. Kurt, any final word from you as we close out our podcast? I'm very, very impressed on how loyal people are to listening to this podcast. I think it's a great way to digest the material.
I'm glad you're listening. Keep it up. Share it with a friend.
But continue reading the journal. This concludes our episode of Fertility and Sterility On Air, brought to you by Fertility and Sterility in conjunction with the American Society for Reproductive Medicine. This podcast is produced by Dr. Molly Kornfield, Dr. Adriana Wong, Dr. Elena HogenEsch, Dr. Selena Park, Dr. Carissa Pekny, and Dr. Nicholas Raja.
