Medical Examiner

How Do IVF Babies Turn Out?

Fertility specialists confront disturbing evidence.

“Would you like a free sample?” said the sales rep from Cryogenic Laboratories, handing out a white plastic pen, unremarkable except for the fact that it was shaped like a sperm. It was a freebie that could only be considered normal here, at the annual convention of fertility doctors, held last month in Philadelphia by the American Society for Reproductive Medicine. In another booth the fertility clinic Genetics and IVF Institute was advertising its patented sex-selection technology by serving pink-and-blue M&Ms in paper urine-specimen cups, while an enterprise called was peddling its new line of where-you-came-from-IVF-children’s-books by giving out test tubes filled with mints. Urologists and reproductive endocrinologists wandered from booth to booth, snacking on egg-shaped chocolates and eying the latest in hormone-delivery systems, embryo-transfer catheters, and terrifyingly long uterine surgery scissors.

It was an impressive display of 21st-century baby-making technology and 21st-century baby-making commerce—and a far cry from 1981, when a doctor named Howard Jones stood in a small IVF clinic in Norfolk, Va., anticipating the birth of the first U.S. child conceived through in vitro fertilization. As recounted by Robin Marantz Henig in her book Pandora’s Baby, Jones had two different commentaries prepared that day, one to read if the child emerged tragically deformed, the other ready if (as turned out to be the case) she emerged looking OK. At that time, nobody involved in this nascent science had any idea, really, how IVF babies were going to turn out. The striking news from this year’s conference is that, gleaming displays notwithstanding, they still basically don’t.

“This is the most important question that we need to answer; I can’t believe that there aren’t hundreds more of us here in this room,” said one doctor, speaking at a workshop exploring whether IVF babies are more likely to develop certain chromosomal and genetic abnormalities and, if so, whether doctors should warn patients. The workshop was one of at least a dozen presentations that explored possible adverse outcomes for IVF babies, an issue that seems to become more urgent, yet more difficult to answer, with each passing year. In the quarter century since the first child was conceived outside the body, IVF and its variants have been pursued by an ever-widening circle of would-be parents; in 2001 as many as 40,000 (1 percent of the babies born in the United States) were conceived through IVF, according to the most recent figures from the Centers for Disease Control. Created as a way of enabling women with blocked fallopian tubes to bear children, IVF is now sought by women suffering from everything from polycystic ovaries to aging egg reserves, not to mention an equally large cohort of infertile men, as well as cancer patients, single parents, and gay and lesbian partners.

The issue of IVF babies’ health burst into the open two years ago, when a group of researchers published a controversial study in the New England Journal of Medicine. The study found that IVF children were more than twice as likely as naturally conceived children to have been diagnosed with a major birth defect by 1-year-old. They were also more likely to be delivered by C-section, to have low birth weight, and to be born before term. Subsequent studies, including two published this year in Obstetrics & Gynecology, have confirmed the findings and added to the list of possible complications, suggesting a greater propensity among IVF babies for certain cancers, such as retinoblastoma, as well as urogenital problems. Meanwhile, pediatricians working with a group of children suffering from Beckwith-Wiedemann syndrome, a condition in which they are prone to disproportionate growth and with it cancerous tumors, found—entirely by happenstance—a higher percentage of IVF children than in the ordinary population.

Many doctors and scientists argue—rightly—that the vast majority of IVF children are normal, their parents grateful, and their development sound. Yet nagging concerns have prompted a flurry of new studies, a number of which were presented at the ASRM conference. Most are disturbing in some way; none seem conclusive; a few are contradictory. In what will probably emerge as one of the more significant, Mary Croughan of the University of California, San Francisco, has received that rare thing in this field—federal funding—to track the development of IVF children and compare it to a cohort group of naturally conceived children. At work for more than a year now, Croughan presented preliminary research that showed an increased risk of certain birth defects, cognitive delays, and behavioral problems among IVF children; to date, her only cases of ADD and autism are in the IVF kids.

But it was another study that was the only one to be granted a press conference at the ASRM gathering. Conducted by the Genetics & Public Policy Center at Johns Hopkins University, it was not a new cohort study, exactly, but rather a meta-analysis that attempted to sort through existing studies and locate common findings. The Johns Hopkins group found no greater risk of birth defects or developmental disorders, a conclusion that the lead scientist, Kathy Hudson, said should be reassuring. Even so, a patients’ advocacy group is now launching its own study.

Yet as every researcher acknowledges, infertility outcome studies themselves suffer from unusual limitations. For one thing, it’s impossible to get a perfect control group. When you compare children of infertile women with children of fertile ones, you cannot know whether any problems in the IVF children are due to the procedure itself, the drugs the women take, or the underlying condition—including, simply, age—that created the infertility in the first place. As one doctor pointed out, women undergoing IVF “are not the normal population; they may not have a normal uterus or vascular supply.” Moreover, Hudson’s group agreed that existing studies suggest that IVF may interfere with the delicate process of genetic expression that takes place during the early days of embryonic development.

She also confirmed what has now become accepted truth: that, puzzlingly, even singleton IVF babies are born at a lower birth weight than naturally conceived ones and are more likely to be premature and to have perinatal complications. One explanation currently in vogue is the “vanishing twin” hypothesis: Many IVF singletons actually start as twins, but at some point one the growth of one ceases and it disappears, leaving a sibling who is smaller than normal.

Because of course—as every last person there was aware—the main risk to IVF babies is that they will be born as part of a set. In a very real sense IVF children are different from other children, or more than half of them are different from other children, simply because they are born as twins or triplets or even quadruplets. As one doctor pointed out, it’s absurd to worry about infinitesimally higher risks of birth defects and to ignore what he referred to as the “uncle with the stinky cigar,” the presence in the room that everyone has gotten used to: The fact that doctors routinely transfer three or more embryos to ensure that even one will take. More than one-third of IVF births involve multiples; more than half of IVF children are multiples. At which point, the procedure itself hardly matters. Simply by dint of being a twin or triplet, a child is much likelier to be stillborn, to be small, to suffer from developmental delay, to have cerebral palsy. * Because of this, a number of European countries have mandated a one-embryo-transfer policy. But in this country the issue is much more difficult to resolve. There are no hard and fast rules here; doctors want the best outcome numbers; and many IVF patients themselves pray for twins. Because American parents so often must pay for the procedure themselves—and because they have waited so long—they often hope that a single try will yield more than a single child. They are, literally, buying in bulk.

What’s also interesting is how each new scientific advance creates new potential dangers. Many doctors believe it’s possible to avoid multiples through a technique called blastocyst culture. By growing embryos in the Petri dish for five days instead of three, they believe that the healthiest embryos will emerge, and the best can then be chosen and only one—at most two—transferred. Other doctors believe, however, and some scientific evidence exists, that prolonged exposure to culture medium can itself interfere with embryonic development.

And that is basically where the science is: inconclusive and rapidly moving ahead. Except there is one clear-cut yet little-discussed issue that has surfaced: Whatever problems may be due to infertile women and their treatment, one should not underestimate the problems created by infertile men and their treatment. Thanks to a relatively new but now common technique called ICSI, in which a single sperm is injected directly into an egg, virtually any man can now become a biological father, no matter how weak or slow or scarce or chromosomally problematic his sperm. This breakthrough is, as one doctor put it, “anti-Darwinian.” It allows the propagation of a sperm that would never on its own have been able to penetrate an egg. Increasingly, doctors suspect that ICSI may interfere with imprinting—the switching on and off of genes in the early days of embryonic development—possibly leading to genetic syndromes like Beckwith-Wiedemann.

If nothing else, ICSI clearly is enabling infertile men to produce infertile sons. “We’re creating a whole new generation of patients,” acknowledged one doctor at the conference, laughing at the irony. Well, sort of laughing.

Correction, Nov. 4, 2004: The original version of this article incorrectly stated that children born as multiples are at a greater risk of cystic fibrosis. They are at a greater risk of cerebral palsy. ( Return to the corrected sentence.)