First Do No Harm? Altering Fetal Development

By Mark Sloan | December 4, 2012

Congenital adrenal hyperplasia (CAH) is a genetic disorder of steroid production that can lead to malformation of the developing female genital tract. A woman at risk of having a child with CAH may be offered prenatal dexamethasone, a synthetic steroid, in an attempt to ensure normal genital development.  The intervention is described as both highly effective and “safe for mother and child” by its leading proponent, Dr. Maria New, a renowned pediatric endocrinologist at Mount Sinai Hospital in New York.

Yet researchers at Swedens Karolinska Institutet have recently reached very different safety conclusions.  Concerned with a number of “severe adverse events” detected in an ongoing prospective study of prenatal dexamethasone and CAH–such as poor postnatal growth and developmental delays, including mental retardation–they have chosen not to enroll new patients into their study. “We do not consider it ethical to initiate further treatment,” writes Dr. Svetlana Lajic of the Karolinska team. “We find it unacceptable that, globally, fetuses at risk of CAH are still treated prenatally without followup.”


In addition to safety concerns, the exposure of potentially affected fetuses to dexamethasone raises complex ethical questions about attempts to alter fetal development, the nature of cultural gender expectations, and even human experimentation.

The adrenal glands of CAH-affected individuals produce inadequate amounts of cortisol and other essential steroid hormones. The severe, “classical” form occurs in about 1 in 15,000 births, while milder variants can occur as often as 1 in 1,000 births. Because CAH is an autosomal recessive genetic disorder, the children of parents who are both carriers of a CAH gene have a one in four chance of inheriting the condition. In its classical form CAH can be fatal without daily, lifelong replacement of the inadequately produced steroids.

The female genital malformation that can accompany CAH is the result of a futile attempt by the adrenal glands to compensate for the defect in cortisol production. Other adrenal steroids are overproduced, particularly androgens, the class of hormones responsible for the development of the external male genitalia, among other functions. In a female fetus, high levels of androgens can lead to masculinization, or virilization, of the external genital tract. The internal genitalia–the uterus, ovaries and fallopian tubes–are not affected by CAH.

Virilization in a female with CAH can range from mild enlargement of the clitoris to the formation of a urogenital sinus, a condition in which the vagina and urethra are joined together in a common channel. Though milder cases may pass unnoticed by parents and physicians and require no treatment, surgeons may recommend correction of more significant anomalies in early childhood.

The primary purpose of administering prenatal dexamethasone is to prevent virilization of the female genitalia. Dexamethasone doesn’t “cure” CAH–with or without prenatal dexamethasone, many affected children will still require lifelong hormone  therapy.

Since the virilizing effects of CAH can begin as early as seven weeks post-conception, dexamethasone is started as soon as an at-risk woman learns she is pregnant. Fetal gender cant reliably be determined until chorionic villus sampling (CVS) is performed at 10-12 weeks of gestation, however, which means that 50 percent of the exposed fetuses–the males, whose genital development is unaffected by CAH–will receive no benefit at all while still being exposed to potentially serious side effects.

Furthermore, as only one in four of the females will actually have CAH, overall only 1 in 8 exposed fetuses (12.5 percent) are at risk for genital virilization and thus stand to benefit from the intervention.  In other words, nearly 90 percent of fetuses at risk for CAH will be subjected to an intervention that cannot possibly benefit them.

Though dexamethasone is stopped for male fetuses and for CAH-unaffected females after CVS results are known, they are nonetheless exposed for several weeks. For fetuses believed to be CAH-affected females, daily dexamethasone, which bathes the developing brain in 60-100 times the normal level of glucocorticoids (the steroid family that includes both cortisol and dexamethasone), continues until birth.

Exposing so many who cannot benefit wouldn’t a problem in the case of a drug intervention that is completely “safe for mother and baby,” of course. But is prenatal dexamethasone really harmless? Shockingly, after three decades of clinical use, the question of safety is far from scientifically settled.

* * *

Dr. Maria New has, by far, the most experience with prenatal dexamethasone administration for CAH of any physician in the United States, and perhaps in the world. Though exact numbers are difficult to ascertain, since 1986 she and her team have intervened in the pregnancies of at least several hundred women.

It was that lack of precise data that raised a red flag in 2009 for Dr. Alice Dreger, Dr. Ellen Feder, and Anne Tamar-Mattis, a trio of researchers who describe themselves as “history, philosophy, and legal scholars interested in the medical treatment of children with atypical sex.” Dreger and colleagues were concerned to find that prenatal dexamethasone administration–an intervention consistently labeled by expert panels as experimental, and one without clearly established risks–was routinely offered to at-risk women without truly informed consent or rigorous follow-up. What became of the treated children later in life, they discovered, was anyone’s guess.

As described in a 2012 review, Dreger et. al. note that follow-up studies of children who have been exposed to prenatal dexamethasone for CAH are of very poor quality. A systematic review and meta-analysis by Fernandez-Balsells, et. al., (2010) examined 1,083 follow-up studies and discovered that only four provided any meaningful data. The authors found that the studies provided “no data of long-term follow-up of physical and metabolic outcomes in children exposed to dexamethasone.”

Further alarming Dreger and colleagues was New’s secondary treatment goal–the prevention of “behavioral masculinization” that can accompany CAH. In a 2001 lecture to a CAH parent support group, New declared her intention to “restore this baby to the normal female appearance–with her eventually becoming somebody’s wife, and having normal sexual development, and becoming a mother.”

Rather than a legitimate treatment goal, Dreger, Feder, and Tamar-Mattis instead saw this as a problematic attempt to medically “prevent” lesbianism. And indeed, while not overtly declaring lesbianism as a target, Dr. New does describe prenatal dexamethasone treatment of CAH-induced behavioral masculinization as a “paradigm for prenatal diagnosis and treatment.”

Whether or not prenatal dexamethasone for CAH could someday lead to an “anti-lesbian drug,” as has been speculated in the popular press–a claim that seems doubtful at best–it is true that exposure to increased androgens in utero can alter psychosexual development.  Though most adult women with CAH identify themselves as heterosexual, studies have generally found a higher prevalence of homosexuality or bisexuality (3-31 percent, compared with approximately five percent in the general population) as well as an increased tendency to be employed in more traditionally male fields and to pursue rougher sports and recreational activities.

Women with CAH tend to score lower than unaffected peers on quality of life surveys, too, most significantly in the subset of women with severe CAH. Complicating these assessments, though, are factors such as parental acceptance of a childs condition, the child’s feelings of isolation from more culturally “female” peers and activities, and the fact that many of the women studied were subjected to aggressive early surgical repair, a trend that has lessened in recent years.

A recurring theme throughout these assessments, as in those of people with a variety of chronic conditions, is that women with good social support and coping skills tend to report a better quality of life than those who lack them–a finding that speaks to the need for comprehensive, multidisciplinary support for girls and women with CAH.

A more pressing concern for parents, physicians, and medical ethicists is the lack of long-term safety data. In addition to the concerns recently voiced by the  Karolinska Institutet team (cited above), experts in the field have expressed strong reservations about prenatal dexamethasone for more than a decade. Some examples:

  • In 2001 the Section on Endocrinology and Committee on Genetics of the American Academy of Pediatrics criticized the practice, pointing to animal and human studies that found high-dose steroid therapy to be harmful to the developing brain. The Section report concluded that prenatal treatment should be “confined to centers doing controlled prospective, long-term studies”– the type of study the Karolinska Institutet has undertaken, and that Dr. New and colleagues, to date, have not.
  • Dr. New’s reply to the AAP report, in which she defended prenatal dexamethasone as “effective” and “safe for both fetuses and mothers,” triggered a strong reply from the report’s authors: “We believe that just as it is inappropriate to stigmatize prenatal therapy before all the data are in [as New charged], it is equally inappropriate to assume it is safe until proven harmful.” The authors then unanimously restated their conclusion: “[P]renatal glucocorticoid therapy for CAH should be confined to centers doing controlled prospective, long-term studies. The memory of diethylstilbestrol [DES] and thalidomide demands no less.”
  • In 2008 Dr. Walter Miller, chief of endocrinology at the University of California, San Francisco, went even further, declaring that “this experimental treatment is not warranted and should not be pursued, even in prospective clinical trials.”
  • A 2010 Endocrine Society Task Force statement declared that prenatal dexamethasone treatment “should continue to be regarded as experimental” and be pursued only through Institutional Review Board-approved protocols.
  • In a 2012 Clinical Opinion in the American Journal of Obstetrics & Gynecology, Dr. Walter Miller of UCSF and Dr. Selma Witchel of the Division of Pediatric Endocrinology, Childrens Hospital of Pittsburgh, concluded: “[I]t is the clinical opinion of the authors that the benefits do not warrant the risks, and hence that first-trimester dexamethasone should not be used to ameliorate genital virilization.”

Despite these admonishments, Dr. New and her team at Mount Sinai continue to promote prenatal dexamethasone to at-risk women without rigorous long-term follow-up of their patients. And despite a troubling lack of scientific evidence to support its claims, the web site of the Maria New Children’s Hormone Foundation continues to state:

“Dr. New maintains contact with all children treated prenatally, and has found not [sic] permanent adverse effects of treatment on mother or fetus. Thus, with nearly 20 years experience, the treatment has been found safe for mother and child.”

* * *

In the end, of course, whether to accept prenatal dexamethasone or not is a parent’s decision to make. It can be a heart-wrenching choice. Virtually all women who are offered treatment already have CAH-affected daughters, some of whom have undergone genital surgery. (It is very rare for both parents to know they are CAH carriers before a first affected pregnancy.)

For parents facing the prospect of another affected daughter, the potential surgery-sparing benefits of prenatal treatment can be very compelling.  For some families, too, as Dreger and colleagues worry, the prevention of “masculinized” behavior may also play a role.

But truly informed consent requires that parents understand the very significant downside of the intervention as well. After all, nearly 90 percent of exposed babies will receive no benefit, while 100 percent will be exposed to risks that are potentially serious and life-long–and that may not be clearly defined for years.

What will the future bring? Will prenatal dexamethasone for CAH ultimately be abandoned as unethically risky? Will it become the source of another DES horror story? Or will it be vindicated as the safe and effective treatment its advocates claim today?

Here’s the sadly frustrating part of the story: Had three decades of exposed mothers and babies been followed more closely, we might already know.