Skip to Main Content

By the time Sarah Chamberlin’s fertility doctor declared her genes “incompatible” with her husband’s and said the clash might be preventing her from having children, she’d had five rounds of artificial insemination and two cycles of in vitro fertilization — and precisely zero pregnancies. So when the physician suggested she try a drug that is ordinarily prescribed to cancer patients, to get her immune system to stop attacking her embryos, she didn’t hesitate.

“I was 41 by then,” said Chamberlain, 47, who lives on New York’s Long Island and at the time was a consultant at her husband’s restaurant. “When they say they have one more thing you can try, it gives you hope.”

advertisement

She had already spent tens of thousands of dollars in an effort to conceive — her insurance didn’t cover assisted reproduction — and had grown inured to the fertility-related shots that turned her into a human pin cushion. With all she had been through, it seemed foolish to hesitate over six additional daily injections and another $3,000 for the drug, called Neupogen; the idea that she might finally have a baby was irresistible.

In the 40 years since the world’s first “test tube baby,” fertility clinics have cooked up nearly three dozen such “add-ons,” or supplementary procedures. Like immune therapy for supposed genetic incompatibility, they’re not essential to IVF. Instead, clinics offer procedures such as “assisted hatching” and “embryo glue” and “uterine artery vasodilation” as purportedly science-based options that increase the chance of having a baby. Except there is little to no evidence that the vast majority of IVF add-ons do any such thing, conclude four papers published on Tuesday in Fertility and Sterility, the journal of the American Society for Reproductive Medicine.

Together, the studies offer the most comprehensive look at current practices in a little-regulated industry for which there is, “at best, extremely weak or contradictory evidence of benefit,” said biostatistician Jack Wilkinson of England’s University of Manchester, who led one of the analyses. “At worst, there is good evidence that some of the add-ons lower the chance” of having a baby through IVF.

advertisement

Wilkinson and his colleagues describe the widespread use of add-ons as a “free-for-all driven largely by commercial pressures.” Add-ons, he said in an interview, “are introduced into routine practice before they have been shown to improve the live birth rate.”

In the United Kingdom, 74% of IVF patients reported using at least one add-on. There are no U.S. data on these supplementary procedures, said Dr. Alan Penzias, a Massachusetts-based fertility doctor and chair of the ASRM practice committee. But they are widely used, he said, especially when, as in Chamberlin’s case, the core procedures haven’t worked.

“Patients and providers have the same interests,” Penzias said. “A pregnancy with a healthy singleton baby as soon as possible. We’re all on the same side.”

Assisted Hatching
Hyacinth Empinado/STAT

The critics say they want patients to understand that the add-ons’ scientific basis is as squishy as quicksand. “Patients are given the impression that the procedures have been studied and shown to be effective,” said Pamela Mahoney Tsigdinos, Wilkinson’s co-author and an IVF patient in the early 2000s (she remains childless) who advocates greater transparency. “But in most cases they haven’t been. You’re on your own at a time when you’re in no position to be objective.”

Add-ons typically enter clinical practice in either of two ways.

Researchers or doctors might simply have a brainstorm. For instance, the idea that observing how fertilized eggs develop over the course of five days in an incubator, rather than examining how they look only at the end, will reveal the best ones led to the adoption of time-lapse systems that watch embryonic development 24/7.

Alternatively, physicians might notice that IVF patients who underwent an unusual procedure had more success. “Endometrial scratching,” for instance, arose about 20 years ago when doctors in Israel observed that women who had undergone a biopsy of the uterine lining seemed to have higher rates of pregnancy. That led to the idea that using a pipette to scratch the endometrium might trigger a hormonal response that makes this tissue more receptive to embryo implantation.

Case reports, in which fertility doctors reported that a few selected patients who underwent the $500 procedure seemed to give birth at higher rates than other patients, seemed to validate this notion.

Artificial egg activation
Hyacinth Empinado/STAT

Unfortunately, case series are a weak form of proof: There is no way to know whether the women would have conceived without the procedure or whether they differed in important ways from women who did not have it. A large (1,364 women) randomized controlled study — the strongest form of evidence — published in the New England Journal of Medicine this year found that endometrial scratching did not raise the rate of live births.

Yet the procedure was used in 27% of IVF cycles in the U.K. in 2018; over 80% of IVF clinicians surveyed in New Zealand, Australia, and the U.K. said they recommended it, found Sarah Lensen of the University of Auckland and her colleagues in their Fertility and Sterility paper.

The reason such procedures are so easily added to what former IVF patient Katy Lindemann calls “a Chinese menu of choices” is that IVF clinics are regulated lightly, if at all. Although the U.S. Food and Drug Administration requires a rigorous assessment of safety and efficacy of procedures that manipulate human cells “more than minimally,” no fertility procedure has been deemed to do so. The FDA does not require proof that a procedure (unlike a drug) benefits patients. In the United Kingdom, regulators can prohibit an IVF procedure only on the grounds of safety; they are not allowed to even require evidence of effectiveness. In neither the U.S. nor the U.K. are clinics required to tell patients whether research shows that an add-on will improve their chances.

“People are not being given complete information about add-ons,” said Lindemann, a London-based digital product strategist. Her four IVF cycles starting in 2015 resulted in two pregnancies but no live births. She received an add-on that involved assessing an embryo’s quality by analyzing a single one of its scores of cells. “By not hearing what evidence does or does not exist about the efficacy of these procedures, people are being misled,” she said.

When the U.K.’s Human Fertilisation and Embryology Authority gathered the scientific evidence for the 11 most common IVF add-ons, it has found that none had been shown in large, rigorous studies to improve a woman’s chance of having a baby.

Embryo glue
Hyacinth Empinado/STAT

That includes popular procedures such as “assisted hatching,” in which embryologists use acid, lasers, or other tools to poke a hole in or simply thin the zona pellucida that covers ova, in an effort to increase the chance of fertilization. This procedure might improve pregnancy rates (by 16%, a 2016 analysis of 36 individual studies found) but not the rate of live births, due to higher rates of miscarriage.

Nor is there evidence for the effectiveness of having fertilized eggs spend their first few hours in the womb in an “intrauterine culture device” rather than a lab incubator.

There is slightly better evidence for a handful of add-ons. In “artificial egg activation,” clinicians douse fertilized eggs with chemicals called calcium ionophores in order to induce the process that kicks off embryo development. Some studies suggest it might improve fertilization rates when sperm are injected into an egg, called intracytoplasmic sperm injection (ICSI), while other research finds no benefit.

The jury is also out for adding “embryo glue” (a chemical called hyaluronan) to the lab dish where embryos develop to increase the chance of implantation, but overall the research suggests it improves the chances of a live birth by about 10%. In the U.S., that translates to 39% of IVF cycles resulting in a live birth, compared to 35% otherwise.

Sarah Chamberlin and her husband were told they had “incompatible” genes. Alice Proujansky for STAT

Despite the weak evidence for most IVF add-ons, it is not unusual for women to receive one after another as they go through ever more rounds of IVF.

Chamberlin had five rounds of IVF in only 13 months, ending in 2014, with a total of 24 embryos transferred. Each time, she received $200 “intralipid infusions” (concoctions of soy oil, glycerine, and egg yolks, given intravenously), which many fertility clinics describe as normalizing levels of the immune system’s natural killer cells so they do not attack the early embryo. Starting in round two, she received injections of human growth hormone, at hundreds of dollar a pop, which her physician said could improve her egg quality and increase her chance of successful IVF (but which left her so dizzy she fell off a treadmill).

With each failure, doctors offered a new explanation, including that she and her husband were “genetically incompatible.” If that were so, Chamberlin wondered, why wasn’t she told until after two IVF failures, in what seemed to be an attempt to keep her trying — and paying? After the five IVF cycles, Chamberlin and her husband decided not to go forward with any additional procedures.

“Fertility clinics have gotten bolder and bolder about using procedures that have little to no scientific validation,” Manchester’s Wilkinson said. “It lets them tell patients there’s one more thing they can try.”

Endometrial scratching
Hyacinth Empinado/STAT

Patients desperate for help feel they’re in no position to question add-ons. “Because patients want to do all they can to succeed, it’s hard to say no,” said Tsigdinos, who paid $500 for assisted hatching. “Many patients believe that clinics offer procedures only if they have validated science behind them, but that’s not the case. I remember what it feels like to be on the receiving end of such sales pitches. They know they’ve got $15,000 from you, and from there it’s all upselling.”

Fertility clinics do not claim that add-ons will guarantee a live birth. “But there was an implicit message that it would improve my chances,” Chamberlin said, referring to the immune therapy she received. “It didn’t really sink in, when the doctor described this as cutting-edge, that that meant experimental.”

In the worst cases, add-ons can do the opposite. Despite the seemingly rational case for time-lapse systems to monitor IVF embryos — for which clinics charge about $1,000 — a 2019 analysis found that the systems had a miscarriage rate of 4% to 14%, compared to 4% with conventional incubation.

But because time-lapse has a live birth rate per IVF cycle of 27% to 40%, compared to 35% with standard incubation, Penzias believes “the jury is still out” on its value. “I do not recommend using it routinely,” he said. “But some clinics are buying the machines and charging patients for it.”

Immune suppression
Hyacinth Empinado/STAT

One popular add-on indisputably hurt women’s chances of giving birth. Called preimplantation genetic testing for aneuploidy (PGT-A), a chromosomal abnormality, it samples one cell from a 3-to-5-day-old embryo. Embryos with an abnormality were not implanted, and often discarded. But later research, including a large randomized controlled trial published last year, showed that a single abnormal cell does not doom an embryo and that PGT-A “makes no difference to live birth rates,” concluded the European Society of Human Reproduction and Embryology.

Patients who abandoned some embryos based on PGT-A, which typically costs $5,000, therefore lost potentially viable ones. “They paid more and worsened their chances of a live birth,” Wilkinson said.

If so, Penzias said, it was well-intentioned. “Sometimes, when you’re down to your last hurrah” with a patient who’s running out of money, time, or the emotional and physical strength that IVF requires, “you want to throw everything you’ve got at it.”

STAT encourages you to share your voice. We welcome your commentary, criticism, and expertise on our subscriber-only platform, STAT+ Connect

To submit a correction request, please visit our Contact Us page.