In each of the next five chapters—beginning with this one—we will discuss in detail a separate, discrete area of our larger domain of inquiry. Each of these chapters will be structured as follows. First, the chapter will review the relevant techniques and practices; next, it will address the ethical considerations; and finally, it will consider the existing regulatory activities.

For reasons discussed above, we will take the practice of assisted reproduction as our fundamental point of departure. Although readers are no doubt familiar with the main features of assisted reproduction techniques and practices, we will give a detailed account of them in order to clarify which aspects might give rise to a need for monitoring, oversight, or regulation.

I. Techniques and Practices

Most methods of assisted reproduction involve five discrete phases: (1) collection and preparation of gametes; (2) fertilization; (3) transfer of an embryo or multiple embryos to a woman’s uterus; (4) pregnancy; and (5) delivery and birth. We will discuss each phase separately. Additional issues connected with recruitment, intake, and possible payment of gamete donors will be discussed extensively in Chapter 6 (on commerce).

The precursors of human life are the gametes: sperm and ova. Parents seeking to conceive through assisted reproduction usually provide their own gametes. In the United States in the year 2001, 75.2 percent of the ART cycles undertaken used never-frozen, nondonor ova or embryos and another 13.7 percent used frozen nondonor ova or embryos. Of the remaining 11.1 percent of cycles using donor embryos, the breakdown is as follows: 3.2 percent of the embryos were previously cryopreserved, and 8 percent were not.^{i 1}

Sperm are typically acquired directly from the male prospective parent. The minority of men who cannot ejaculate, or who have a blocked reproductive tube, may undergo assisted sperm retrieval (ASR). Alternatively, sperm precursor cells obtained by testicular biopsy may be used for purposes of insemination (though this yields a lower pregnancy rate).

Acquiring ova for use in artificial reproduction is significantly more onerous, painful, and risky than acquiring sperm (though its risks are still low in absolute terms). In the normal course of ovulation, one mature oocyte is produced per menstrual cycle. However in assisted reproduction—to increase the probability of success—many more ova are typically retrieved and fertilized. Thus, the ova source (who is usually also the gestational mother) undergoes a drug-induced process intended to stimulate her ovaries to produce many mature oocytes in a single cycle. This procedure, commonly referred to as “superovulation,” requires the daily injection of a synthetic gonadatropin analog, accompanied by frequent monitoring using blood tests and ultrasound examinations. This treatment begins midway through the previous menstrual cycle and continues until just before ova retrieval. The synthetic gonadatropin analogs give the clinician greater control over ovarian stimulation and prevent premature release of the ova.

A very small percentage of women using assisted reproduction (in 2001, fewer than 1 percent of assisted reproduction patients) opted not to undergo ovarian stimulation prior to ova retrieval.² In such “unstimulated” procedures, the clinician monitors the development of an ovarian follicle (via ultrasound) and uses daily blood sampling to predict the moment of ovulation. Only one follicle develops and the timing of maturation and release is not controlled. Because there are fewer embryos for transfer, this process yields a lower success rate than does in vitro fertilization (IVF) following ovarian stimulation.

When blood testing and ultrasound monitoring suggest that the ova are sufficiently mature, the clinician attempts to harvest them. This is typically achieved by ultrasound-guided transvaginal aspiration. In this procedure, a needle guided by ultrasound is inserted through the vaginal wall and into the mature ovarian follicles. An ovum is withdrawn (along with some fluid) from each follicle. This is an outpatient procedure. Risks and complications are low, but may include accidental puncture of nearby organs such as the bowel, ureter, bladder, or blood vessels, as well as the typical risks accompanying outpatient surgery (for example, risks related to administration of anesthesia, infection, etc.).

Once sperm and ova have been collected, they are cultured and treated to maximize the probability of success. Ova are transferred into a culture medium containing the intended mother’s blood serum. The seminal fluid is removed from sperm and replaced with an artificial medium. For infertile men, the clinician removes excess material and concentrates the motile sperm.ⁱⁱ

B. Fertilization

Once the ova and sperm have been properly prepared, the clinician attempts to induce fertilization—the union of sperm and ovum culminating in the fusion of their separate pronuclei and the initiation of a new, integrated, self-directing organism. It is common practice to attempt to fertilize all available ova.ⁱⁱⁱ Fertilization can be achieved through a number of means including (1) “classical” IVF, (2) gamete intrafallopian transfer (GIFT)^iv, (3) intracytoplasmic sperm injection (ICSI), and (4) various other methods of zona pellucida manipulation.^v

IVF is the most common method of artificial fertilization. In 2001, it was used by 99 percent of ART patients.³ As noted previously, both sperm and ovum are cultured to maximize the probability of fertilization. The ova are examined and rated for maturity in an effort to calculate the optimal time for fertilization. They are usually placed in a tissue culture medium and left undisturbed for two to twenty-four hours. The sperm are prepared as described above. Once the gametes are adequately prepared, thousands of tiny droplets of sperm are placed in the culture medium containing a single ovum. After 24 hours, each of the oocytes is examined to determine whether fertilization has occurred.

GIFT was introduced in 1984 as an alternative to standard IVF. Today, attempts at fertilization via GIFT are rare. In 2001, they accounted for less than 1 percent of all attempts at fertilization used by ART patients.⁴ As the name suggests, fertilization using GIFT occurs within the woman’s body. Ovarian stimulation and retrieval are performed in the same manner as in IVF. In a single procedure, ova are retrieved, combined with the sperm outside the body, and then transferred back into the fallopian tube where it is hoped that fertilization itself will occur. Typically, two or more ova are retrieved and transferred. GIFT requires only one functional fallopian tube to succeed. Because fertilization takes place inside the woman’s body, substantially less lab work is required and there is no need for embryo culturing. For the same reason, however, if several ova are transferred, GIFT exposes the patient to a higher-than-normal risk of multiple gestations. Moreover, when GIFT does not succeed practitioners frequently cannot determine why it failed, for example, whether the ovum was not fertilized or the embryo did not implant.

A new and increasingly popular technique for fertilization is intracytoplasmic sperm injection. As the name implies, with ICSI, ovum-sperm fusion is accomplished not by chance, but by injecting a single sperm directly into an oocyte. The oocyte is treated with an enzyme that removes certain cells that surround it (“nurse cells”). The sperm are placed in a viscous solution that greatly slows their motility. A single sperm is selected and drawn into a thin pipette from which it is injected into the cytoplasm of the ovum cell.

ICSI is indicated in cases of severe male-factor infertility, in which male patients have either malformed sperm or an abnormally low sperm count. ICSI is also ideal for patients whose sperm would not otherwise penetrate the exterior of  an oocyte.^vi ICSI was used in 49.2 percent of all ART cycles in 2001.⁵ However, 42.2 percent of those ICSI cycles were undertaken by couples without male-factor infertility.⁶ The growing popularity of this technique most likely has to do with the wish to increase the control over, and success rates for, fertilization: ICSI, unlike standard IVF, guarantees the entrance of a single sperm directly into a single egg.^vii

Clinicians can also attempt to induce fertilization artificially through manipulation of the zona pellucida, the thick extra-cellular covering that surrounds the ovum. To assist the sperm’s penetration of the ovum, clinicians perforate the zona pellucida using an acidic solution (“zona drilling”) or a needle or pipette (“partial zona dissection”). Alternatively, clinicians inject sperm underneath the zona pellucida, but not directly into the ovum’s cytoplasm (“subzonal insemination”). Zona drilling results in few pregnancies and has been linked to inhibition of early embryo growth, perhaps due to the acidic solution entering the ovum itself.⁷ Few embryos conceived through partial zona dissection have a normal appearance, but it is not definitively known why this is so or whether the difference is significant in any way to the health of the developing child. Subzonal insemination can be effective in the hands of a skilled practitioner, but frequently results in unfertilized oocytes or fertilization by multiple sperm, rendering the embryo unusable.⁸ The safety risks associated with these procedures are discussed below.

A recently developed adjunct to IVF is ooplasm transfer. This procedure has been used for women whose fertilized ova do not develop normally owing to a deficiency in their mitochondria. To remedy this problem at the time of fertilization, the oocyte is injected with donor cytoplasm that contains healthy mitochondria. Because the new cytoplasm contains the donor’s mitochondrial DNA, the resulting child will have inherited DNA from three individuals: the father, the mother, and mitochondrial DNA from the ooplasm donor. Moreover, the donor mitochondria could be passed on to future generations through the resulting child. To date, there have been thirty children born worldwide as a result of this procedure.⁹ However, for reasons discussed elsewhere in this document, this technique is not currently approved for use in clinical practice in the United States.^viii

Once fertilization has occurred, the new embryos remain in the culture medium. Nutrients are added to the medium. Some commercially produced preparations exist but, typically, ART clinics make their own on-site. Some clinics co-culture developing embryos: that is, they culture the embryos in a medium containing other cells that enhance the growth of the embryos and remove toxins. Various types of cells have been used for such co-culture, including cells extracted from the uterus or fallopian tubes of patients or donors, rat liver cells, monkey kidney cells, cow uterine cells, and human ovarian cancer cells. The embryos remain in culture and are warmed in an incubator until they are either transferred into the recipient’s uterus or cryopreserved.

Because in many cases not all embryos are transferred in each cycle, cryopreservation of embryos has become an integral part of ART.^ix The American Society for Reproductive Medicine (ASRM) has deemed cryopreservation “essential” to the practice of assisted reproduction and provides extensive guidance to technicians as to the maintenance of cryopreservation facilities. Cryopreservation is a complicated process that requires embryo preparation, sophisticated freezing technology, reliable storage, and meticulous record keeping. To guard against the formation of ice crystals that could destroy the embryo, the clinician introduces a cryoprotectant solution into the early-stage embryo’s interior. The prepared embryos are then placed in a straw-like structure that is gradually frozen. Once frozen, these structures are stored in canisters at very low temperature (typically around minus 196 degrees centigrade). Some researchers suggest that it may be possible to cryopreserve embryos safely for fifty years or longer.¹⁰ A recently reported study by the Society for Assisted Reproductive Technology and RAND estimates that 400,000 embryos are in cryostorage in the United States.¹¹

Most ART patients do not receive cryopreserved embryos. In 2001, only 14 percent of all ART cycles involved transfer of frozen embryos.¹² The rate of live births for cycles using cryopreserved embryos is significantly lower than it is for never-frozen embryos (23.4 percent versus 33.4 percent).¹³Experts estimate that only 65 percent of frozen embryos survive the thawing process.¹⁴ There are, however, incentives for couples to use cryopreserved embryos; doing so eliminates the cost and effort of further oocyte retrieval. This can decrease the cost of a future cycle by roughly $6,000.¹⁵ Transfer of cryopreserved embryos might be preferable also for recipients who are suffering from ovarian hyperstimulation syndrome (discussed below). Because pregnancy aggravates this disorder, delayed transfer can be helpful, and cryopreservation allows such delay. The additional control over the timing of transfer conferred by cryopreservation is also helpful to women whose uterine lining is not fully prepared to receive an embryo at the time of its creation. Cryopreservation also reduces pressure to implant all embryos at once, thus reducing the risk of high-order multiple pregnancies.

C. Transfer

Following the creation of a human embryo by IVF, the next discrete phase in the assisted reproduction process is transfer of the embryo into the uterus of the mother (or gestational carrier^x).

Typically, the embryos are transferred on the second or third day after fertilization, at the four- to eight-cell stage. To maximize the probability of implantation, some clinicians cultivate embryos until the blastocyst stage (five days after fertilization) before transferring them to the uterus.¹⁶ Prior to transfer, the clinician evaluates the embryos’ shape and appearance. There is believed to be some correlation between the external appearance of an embryo and its likelihood of implantation and successful development, but appearances can also be misleading. Some unhealthy-looking embryos implant and develop into healthy fetuses and children, and some healthy-looking embryos fail to implant or experience developmental problems.¹⁷ Other methods of evaluation include analysis of chemicals produced by the embryos in culture and pre-evaluation of the quality of sperm and ovum.

A more recently developed method of embryo analysis is preimplantation genetic diagnosis. In PGD, one or more cells are extracted from the eight- to sixteen-cell embryo by means of biopsy. The clinician tests the sample cell(s) for chromosomal or genetic characteristics, including the sex of the embryo, with special attention to any genetic disorder for which the relevant mutation has been identified in the parents or an earlier child. (PGD will be discussed further in Chapter 3.)

Prior to transfer, some clinicians attempt to facilitate implantation by means of a process called assisted hatching. Several days after fertilization, an embryo must break out of the zona pellucida so that it can implant into the uterine wall. In some instances, the zona pellucida proves to be too hard to break and implantation fails as a result. To aid in hatching, clinicians use chemicals, lasers, or mechanical manipulation of the zona pellucida.¹⁸

Once the embryos have been selected and prepared, they are transferred into the uterus. The total number of embryos transferred per cycle varies, usually according to the age of the recipient. For women under 35, the average number of never-frozen embryos transplanted per transfer procedure was 2.8. For women 35 to 37, 38 to 40, and 41 to 42, the average numbers of never-frozen embryos transplanted per transfer procedure were, respectively, 3.1, 3.4, and 3.7.¹⁹ The Centers for Disease Control (CDC) report notes that in 32 percent of ART cycles using never-frozen, nondonor ova or embryos in 2001, 4 or more embryos were transferred.²⁰

Typically embryos are transferred into the uterus using a catheter. The catheter is inserted through the woman’s cervix and the embryos are injected into her uterus (along with some amount of the culture fluid). This procedure does not require anesthesia. Following injection, the patient must lie still for at least one hour. While the transfer procedure is regarded as simple, different practitioners tend to achieve different outcomes.²¹

An alternative method of embryo transfer is zygote intrafallopian transfer (ZIFT). In ZIFT, the embryo is placed (via laparoscopy) directly into the fallopian tube, rather than into the uterus. In this way, it is similar to the transfer of gametes in GIFT. Some individuals opt for ZIFT on the theory that it enhances the likelihood of implantation, given that the embryo matures en route to the uterus, presumably as it would in natural conception and implantation. Additionally, many patients prefer ZIFT to GIFT because the process of fertilization and early development of the embryo may be monitored.²² However, ZIFT remains a rare choice, accounting for 0.8 percent of all ART cycles in 2001.²³

D. Pregnancy

Successful implantation of an embryo in the uterine lining marks the beginning of pregnancy. In 2001, 32.8 percent of the ART cycles undertaken resulted in clinical pregnancy.^{xi 24} This number varied according to patient age.²⁵ After the inception of pregnancy, patients are carefully monitored and treated by an obstetrician. Pregnancies resulting from assisted reproduction are sometimes treated as high risk.²⁶ Clinicians recommend prenatal diagnosis and testing for many pregnancies resulting from assisted reproduction.

There are a number of medications and procedures that may be indicated during a pregnancy facilitated by assisted reproduction. It is typical for a patient to receive progesterone injections to support key functions necessary to pregnancy.

Multiple gestations are common among pregnancies facilitated by assisted reproductive technologies. The rate of multiple-fetus pregnancies from ART cycles using never-frozen, nondonor ova or embryos in 2001 was 36.7 percent.^xii  For the same time period, the multiple infant birth rate in the United States was 3 percent. The extraordinarily high rate of multiple pregnancies resulting from assisted reproduction is almost entirely attributable to the transfer of multiple embryos per cycle.^xiii

In an effort to reduce the risks of multiple pregnancy, practitioners sometimes employ a procedure termed “fetal reduction,” the reduction in the number of fetuses in utero by selective abortion. Fetuses are selected for destruction based on size, position, and viability (in the clinician’s judgment).²⁷ The clinician, using ultrasound for guidance, inserts a needle through the mother’s abdomen (transabdominal multifetal reduction) through the uterine wall. The clinician then administers a lethal injection to the heart of the selected fetus—typically potassium chloride. The dead fetus’s body decomposes and is resorbed. To be effective, transabdominal multifetal reduction must be performed at ten to twelve weeks’ gestation. In an alternative procedure, transvaginal multifetal reduction, a needle is inserted through the vagina. Transvaginal multifetal reduction must be performed between six and eight weeks gestation (eight weeks is recommended).

E. Delivery

In 2001, for never-frozen nondonor ova or embryos, the overall rate of live births per cycle^xiv was 27 percent (33.4 percent live births per transfer).^{xv 28} Among these pregnancies, 82.2 percent resulted in live births.²⁹ Of these resulting 21,813 live births, 35.8 percent were multiple infant births (32 percent twins and 3.8 percent triplets or more).^{xvi 30} One 1993 Canadian study showed that nearly 25 percent of all births facilitated by ART are premature, and 30 percent of the resulting infants had low birthweight.^{xvii 31} While this low birthweight may be attributable to the high rate of multiple pregnancies, one 1987-89 French study reported that even for singleton births facilitated by ART, the rate of prematurity and low birthweight was twice that of children conceived by natural means.³² Another study suggests that women using ART are more likely to induce labor and undergo elective caesarian section delivery.³³

F. Disposition of Unused Embryos

As mentioned above, in many cases of ART there are in vitro embryos that remain untransferred following a successful cycle. There are five possible outcomes for such an embryo: (1) it may remain in cryostorage until transferred into the mother’s uterus in a future ART cycle; (2) it may be donated to another person or couple seeking to initiate a pregnancy; (3) it may be donated for purposes of research; (4) it may remain in cryostorage indefinitely; or (5) it may be thawed and destroyed.

G. Projected Techniques/Recent Experiments

There is a range of research in the reproductive technology area that is now experimental and in some cases speculative, but still worth noting. One such area of research is “nuclear transfer,” which involves transplanting the nucleus from a fertilized human egg into an enucleated fertilized human egg.^xviii The process is similar to somatic cell nuclear transfer (or human cloning), except that the nucleus inserted into the egg comes from another fertilized egg rather than from a somatic cell of a living child or adult. The resulting child could conceivably carry genetic material from three (perhaps four) people: the male and female progenitors of the original fertilized human egg and at least the mitochondrial DNA from the donor of the egg into which the embryo’s nucleus is inserted. In experiments in China in 2003, researchers reported achieving a triplet pregnancy with such embryos, though none of the fetuses survived to birth (a result they attribute to substandard obstetrical care).³⁴ Researchers have also begun investigating whether ovarian tissues from aborted fetuses may be developed in the lab in hopes of one day providing mature eggs suitable for IVF.^xix In July 2003, researchers announced that they obtained ovarian follicles from aborted fetuses aged between twenty-two and thirty-three weeks gestation, and were able to develop the follicles in culture to a secondary stage. The researchers are working to improve the culture media and prolong the culture period to completely develop the follicles as a source for human eggs.³⁵

In their quest for alternative sources of gametes, researchers are working to develop human eggs and sperm from embryonic stem cells. There has already been some success coaxing embryonic stem cells from mice to develop into sperm and eggs, and some researchers project that this technology will succeed with human embryonic stem cells in “about ten years.”^xx This would make possible the novel prospects of producing male-derived eggs or female-derived sperm, and of producing children whose biological progenitors were embryos that were disaggregated for their stem cells. There has also been an experiment that fused blastomeres from two separate embryos to produce a single (in this case, hybrid male-female) embryo.³⁶

Most speculative is research aimed at engineering uterine lining tissue outside the body, for use as a diagnostic tool to study implantation. Researchers have transferred human embryos to an artificial endometrium, to which these embryos attached and began to develop. The implanted, developing embryos were grown for six days, but researchers did not attempt to cultivate them further.³⁷ It is not possible now to predict just how much further in vitro human embryos may someday be developed with such “uterine-like” substitutes. Another area of highly speculative research involves uterus transplants, contemplated as a means to enable women with damaged or absent uteri to bear children.³⁸ There has also been speculation about the prospect of implanting human embryos into specially prepared non-human animal uteruses in order to study their development, but there are as yet no reports of such experiments having taken place with any noteworthy results.

II. Ethical Considerations

The development and practice of assisted reproductive technologies have yielded great goods. They have relieved the suffering of many who are afflicted with infertility, helping them to conceive biologically related children. Yet these activities also raise a variety of ethical issues. Some concern the well-being of the participants in assisted reproduction: gamete donors, prospective parents, and their resulting children. Other issues arise from the expansion of control over reproduction, including current and projected possibilities for altering the biological relationships central to human procreation. Still other issues concern the use and disposition of human embryos that are incident to these new capacities and techniques.

The intersection of two key factors—patient vulnerability and novel (in some cases untested) technology—defines much of the arena of concern. First, assisted reproduction is generally practiced on patients who are experiencing great emotional strain. When it succeeds it can be a source of great joy—as it has been for tens of thousands of parents each year. But success is far from universal, especially for older patients; and even when it happens, the process and the circumstances surrounding it can be difficult to bear. Those suffering from infertility often come to practitioners of assisted reproduction after prolonged periods of failure and dismay. This vulnerability may lead some individuals to take undue risks (such as to insist on transferring an unduly large number of embryos). The occasional irresponsible clinician may even pressure patients to take such risks, for the sake of improving his reportable success rates.

Second, some assisted reproductive technologies have been used in clinical practice without prior rigorous testing in primates or studies of long-term outcomes. IVF itself was performed on at least 1,200 women before it was reported to have been performed on chimps, although it had been extensively investigated in rabbits, hamsters, and mice.³⁹ The same is true for ICSI. The reproductive use of ICSI was first introduced by Belgian researchers in 1992.⁴⁰ Two years later, relying on a two-study review of safety and efficacy, ASRM declared ICSI to be a “clinical” rather than “experimental” procedure. Yet the first non-human primate conceived by ICSI was born only in 1997 and the first successful ICSI procedure in mice was reported in 1995. ⁴¹ Absent long-term studies of the children conceived using ICSI or other novel procedures, it is unclear to what extent these alterations in the ART process affect the health and development of the children so conceived.⁴²

Below, we survey the ethical concerns raised by ART in four specific areas: (1) the well-being of children born with the aid of ART; (2) the well-being of women in the ART process; (3) the meaning of enhanced control over procreation; and (4) the use and destruction of embryonic human life. As we proceed, two points are worth noting. First, we raise these areas of concern solely to enable us to diagnose whether the current regulatory system is adequately protecting the human goods at stake. In no way have we lost sight of the human goods made possible by ART—most notably, the treatment of infertility and the creation of biologically related children for couples who desire and could not otherwise have them. Second, we shall be raising three different kinds of questions: First, questions of fact, such as whether a certain assisted reproduction technique is safe. Second, questions of principle, such as the moral significance of embryo destruction incident to fertility treatment or the significance of using fetal gametes for reproductive purposes. Third, questions of judgment, such as what degree of risk to the carrying mother or child conceived with assisted reproduction is justified in cases where bearing such risks is the only way for individuals or couples to have a biologically related child. Connected to this last question is the issue of who should make such judgments—individuals, doctors, or society as a whole acting through public institutions. For each of these questions—questions of fact, questions of principle, and questions of judgment—both better data and more public discussion are crucial.

A. Well-Being of the Child

The central figure in the process of assisted reproduction, directly affected by every action taken but incapable of consenting to such actions, is the child born with the aid of ART. Each intervention or stage in the ART process might affect this child’s health and well-being: gamete retrieval and preparation, fertilization, embryo culture, embryo transfer, pregnancy, and of course birth.⁴³

The health of the child born through ART may be affected by actions taken as early as gamete retrieval and preparation. Some studies show that superovulation decreases embryo and fetal viability (compared with those in unstimulated cycles).⁴⁴ One study of embryos created during stimulated cycles revealed a high level of “developmental arrest, embryonic aneu- ploidy, mosaicism, apoptosis and failure of cytokinesis.”⁴⁵ It is possible that lesser abnormalities, compatible with birth, make their way into the children born alive.

There have been very few comprehensive or long-term studies of the health and well-being of children born using ART, although more than 170,000 such children have been born in the United States.⁴⁶ The fact that no major investigation or public study has yet been called for in this area might suggest that there is no discernible health crisis in assisted reproduction, as does the fact that demand for ART has grown substantially and continuously since its inception. At the same time, however, our ability to know this with certainty is limited, both because of the absence of major longitudinal studies of the well-being of children born using different assisted reproduction techniques, and because the oldest person conceived through ART is only in her mid-twenties.

Some recent studies have associated various birth defects and developmental difficulties with the uses of various technologies and practices of assisted reproduction. None of these studies provide a causal link between ART and the dysfunctions observed, and some commentators have taken issue with some of the methodologies used. Nevertheless, these findings have raised some concerns. One such study concluded that children conceived by assisted reproduction are twice as likely to suffer major birth defects as children conceived without such assistance.^{xxii 47} Other recent studies have reached similar conclusions.⁴⁸ Additional studies have associated the use of assisted reproduction technologies with a higher incidence of diseases and malformations, including Beckwith-Wiedemann syndrome (BWS),^xxii rare urological defects, retinoblastoma,⁴⁹ neural tube defects,⁵⁰ and Angelman syndrome.⁵¹

While many are concerned about the increased risk to children suggested by these studies, the overall incidence of such harms is low enough that infertile couples have not been deterred in their efforts to conceive using IVF or ICSI. Indeed, ART clinicians (and in some cases the authors of these studies)⁵² advise their patients that such data should not dissuade them from pursuing infertility treatment.

ICSI has raised concerns among some observers largely for the very reasons that it has proven so successful as a means of fertilization: ICSI circumvents the ovum’s natural barrier against sperm otherwise incapable of insemination. Some suspect that removing this barrier may permit a damaged sperm (for example, aneuploid or with damaged DNA) to fertilize an ovum, resulting in spontaneous abortion or harm to the resulting child. Some male ART patients have a gene mutation or a chromosomal deletion that renders them infertile. Yet, if a sperm can be retrieved from these patients, they may be able to conceive a child via ICSI, possibly passing along the genetic abnormality to the resulting child. For example, two-thirds of men with congenital bilateral absence of the vas deferens (rendering them unable to ejaculate) carry certain cystic fibrosis mutations.⁵³ ICSI may permit these men to overcome their infertility, but the resulting child will (in 50 percent of the cases) bear this genetic mutation. Similarly, another form of male factor infertility characterized by a very low sperm count is associated with a particular Y-chromosome deletion. The use of ICSI in such cases risks transferring this chromosome deletion to the resulting child, rendering any male child infertile, and, according to some studies, at risk for sex-chromosome aneuploidy.⁵⁴ Additional studies have associated the use of ICSI with an increased incidence in novel chromosomal abnormalities and mental developmental delays.⁵⁵

It is a matter of concern that there have been few longitudinal studies analyzing the long-term effects of ICSI on the children born with its aid. The Belgian group that pioneered ICSI has collected a database that details neonatal outcome and congenital malformations in children conceived through ICSI.⁵⁶ But there do not seem to be any ongoing or published studies of this kind investigating the long-term effects of ICSI beyond the neonatal stage.

Many adjuncts to the fertilization and transfer process raise concerns for the health and well-being of the children born as a result.^xxiii Some have speculated that factors such as culture conditions and length of time an embryo spends in culture may affect the development of the children later born.⁵⁷ Some authorities claim that differences in salt or amino acid concentrations in the culture media can affect gene expression.⁵⁸ Additionally, one researcher notes that the process of extended culture in mice (for example, permitting extended embryo development prior to transfer) can cause imprinting problems leading to abnormal development.⁵⁹

Still other adjuncts to fertilization and transfer may not be risk-free. Cryopreservation might affect gene expression or lead to other molecular effects such as “telomere shortening and replicative senescence, damage to plasma and nuclear membranes, and inappropriate chromatin condensation.”⁶⁰  Similarly, ooplasm transfer has been linked to an unusually high rate of Turner syndrome.⁶¹ Finally, assisted hatching (or any technique that results in manipulation of the zona pellucida) has been associated with a higher incidence of monozygotic twinning and an increased risk of twins carried in the same amniotic sac, which can lead to malformation, disparities in growth, and pregnancy complications.⁶²

Multiple gestations, far more common in the context of assisted reproduction than in natural conception,^{xxiv 63} have a higher incidence of adverse impacts on the health of the children born.⁶⁴ Such pregnancies greatly increase the risk of prenatal death.⁶⁵ Multiple pregnancies are also more likely to lead to premature birth; and prematurity is associated with myriad health problems including serious infection, respiratory distress syndrome, and heart defects.⁶⁶ One in ten children born following high-order pregnancies dies before one year of age.⁶⁷ Children born following a multiple pregnancy are at greater risk for such disabilities as blindness, respiratory dysfunction, and brain damage.⁶⁸ Moreover, infants born following such a pregnancy tend to have an extremely low birthweight, which is itself associated with a number of health problems, including some that manifest themselves only later in life, such as hypertension, cardiac disease, stroke, and osteoporosis in middle age.⁶⁹ Interestingly, the higher incidence of low birth-weight may not be limited to infants born from multiple pregnancies. According to recent studies, singletons born with the aid of ART tend to have an abnormally high incidence of prematurity and low birthweight.⁷⁰ 

So-called “fetal reduction” aims to reduce the problems associated with multiple pregnancy. But fetal reduction is itself potentially associated with a number of adverse effects on the children who remain following the procedure. One study shows that following transabdominal multifetal reduction there is a miscarriage rate of 16.2 percent, and 16.5 percent of the remaining pregnancies end in premature birth.⁷¹ The alternative method, transvaginal multifetal reduction, carries a higher risk of infection and has been associated with a higher risk of infant mortality than its counterpart.⁷² It has been observed that children born following fetal reduction (by either method) tend to be premature, thus exposing them to the complications described above.⁷³ One study has suggested that children born following fetal reduction are more vulnerable to periventricular leukomalacia, which is characterized by brain dysfunction and developmental difficulties.⁷⁴

Taken together, the significance of these various studies is uncertain. They raise a broad range of concerns, but the scale of the research has been limited. In many cases, there are observed correlations between ART and a higher incidence of certain health problems in the resulting children. But in most studies, there is no demonstrable causal relationship between a particular facet of ART and the undesirable health effect. Infertile individuals seeking assisted reproduction may be disproportionately afflicted with heritable disorders, and these may in part account for the higher incidence of birth and developmental abnormalities in ART children compared to those conceived in vivo. The results are therefore still preliminary.  The need seems clear for more data to determine what risks, if any, different assisted reproduction techniques present to the well-being of the future child. Moreover, in cases where ART is the only available means for individuals or couples to conceive a biologically related child, it is an important ethical and social question what level of increased risk can be privately justified by patients and doctors, and what level of increased risk should be publicly justified by society as a whole, especially should the society bear the costs of caring for any resulting health problems.

Another concern is for the well-being of the women who participate directly in the process of assisted reproduction.

Aside from the discomforts and burdens of ovarian stimulation and monitoring, there are also some risks attached to hormonal stimulation. One such risk is “ovarian hyperstimulation syndrome,” characterized by dramatic enlargement of the ovaries and fluid imbalances that can be (in extreme cases) life threatening.^xxv Complications can include rupture of the ovaries, cysts, and cancers. The reported incidence of severe ovarian hyperstimulation syndrome is between 0.5 and 5.0 percent.⁷⁵ Additionally, adverse side effects of the hormones administered during superovulation have included memory loss, nerological dysfunction, cardiac disorders, and even sudden death.⁷⁶ There do not appear to be any studies on the incidence of such side effects.⁷⁷

Some women who become pregnant with the aid of assisted reproduction are treated as “high-risk” patients and experience a higher incidence of complications than do women with natural pregnancies. Some commentators have suggested that this is due to the age of the patients (who tend to be older than most childbearing women) and the high rate of multiple pregnancies.⁷⁸

Multiple pregnancies are far more common following ART, owing especially to the practice of transferring multiple embryos but also to the higher incidence of spontaneous twinning with any single embryo. Multiple pregnancies pose greater risks to mothers than do singleton pregnancies. A woman carrying multiple fetuses has a greater chance of suffering from high blood pressure, anemia, or pre-eclampsia.⁷⁹ Because multiple-gestation pregnancies are generally more taxing on the mother’s body, they are likelier to aggravate pre-existing medical conditions.⁸⁰ Moreover, such pregnancies expose the woman to higher risks of uterine rupture, placenta previa, or abruption.⁸¹ One commentator noted in 1995 that the added expense growing out of complications from multiple-gestation pregnancies is one of the primary reasons private health insurance generally does not cover assisted reproduction.⁸² Both professional societies and advocates for infertile patients argue that mandating insurance coverage could reduce multiple- gestation pregnancies because it would reduce financial pressure to succeed in the first attempt.^xxvi

The ability to initiate fertilization artificially may also profoundly affect the character of human reproduction and our attitudes toward it, as well as the relationships between parents and children and across generations. Three potential hazards or concerns seem especially worthy of note. First, ART raises novel possibilities for altering the biological relationships that are central to normal sexual reproduction, and thus for confounding the human relationships that follow from it. Through ART, it is now possible for a surrogate (or an adoptive parent) to carry and give birth to another couple’s biological child; it is possible for a woman to become pregnant with an anonymous donor’s sperm; it is possible for a deceased male to become a biological father after death; and it is possible to produce a child with genetic material from three progenitors. Moreover, current research might one day make it possible to use gametes from aborted fetuses, and thus make such fetuses into biological parents, and to produce eggs from male-derived embryonic stem cells or sperm from female-derived embryonic stem cells, which would in theory allow for the creation of a child with two male or two female embryonic progenitors. Second, ART raises the possibility of moving human procreation in the direction of manufacture, by introducing technical approaches or attitudes into the activity of human reproduction. And finally, ART might affect our general understanding of or attitudes about parenthood and childhood, by making sexual reproduction simply one option among many, with no special significance for how we understand the coming-to-be of the next generation.

Particular techniques raise certain specific concerns in this regard. Cryopreservation, ooplasm transfer, and the possible use of fetal oocytes directly raise concerns about the definition and identity of “father” and “mother.” Cryopreservation of sperm and embryos makes posthumous parentage possible. For instance, some American soldiers have been reported to store up sperm on the eve of shipping out to a battle zone. And instances have been reported in which women have requested that their newly deceased husband’s sperm be harvested via assisted sperm retrieval from the corpse and used for artificial insemination. If techniques for cryopreservation of ova are ever perfected, or if ova can be derived from adult stem cells, new opportunities for posthumous conception involving deceased women will also arise.

Ooplasm transfer raises a slightly different issue. Because donated ooplasm contains mitochondrial DNA from the donor, the resulting child receives a small genetic contribution from a third person. Moreover, because mitochondrial DNA is maternally inherited, if the resulting child is female, she will pass on to her own offspring the genetic contribution of both her mother and the female ooplasm donor.

A projected technique that raises new ethical concerns is the harvesting and use of fetal oocytes. Some researchers have posited that oocytes (or their precursors) might be harvested from aborted fetuses and used as donated ova (once they have matured in vitro) for patients who have impaired ovarian function.^xxvii The aborted fetuses would be the genetic mothers of any resulting children. If recent studies in which mouse oocytes have been derived from mouse embryonic stem cells⁸³ can be replicated in humans, a five-day-old embryo (the age of the mouse embryo when cells were retrieved) could also become the biological progenitor of new children.⁸⁴

These procedures, and others like them, raise the possibility that children conceived through ART might be connected to their biological parents in fundamentally different ways than children conceived and born without artificial intervention. In some cases, children conceived with these technologies might be denied the biparental origins that human beings have always taken for granted and that have always been the foundation of familial relations and generational connections. ART techniques do not have to disrupt such relations, but they might be used in ways that confound parentage, involve more or fewer than two biological parents, or otherwise depart from the biologically grounded parent-child relation.

Fetal reduction raises its own distinct set of concerns. In this procedure, parents effectively choose to have some developing fetuses (each of which was conceived in the hope that it would be developed to term) live and some not, and they use surgical procedures to reduce the number of living fetuses in utero.

Assisted reproduction usually entails the loss of human embryos, especially when superovulation is used and many ova are fertilized at once. Large numbers of embryos die at all stages of assisted reproduction (in vitro and in vivo).^xxviii An unknown number of additional embryos are discarded when it is determined that they are no longer needed or desired. Still others are donated to researchers, who use them in biomedical or scientific experiments that involve or lead to their destruction. Thousands of embryos are cryopreserved for indefinite periods of time. As previously noted, an estimated 400,000 embryos were in cryostorage in the United States as of April 2002.

Actions that result in the end of embryonic life are morally significant and require careful consideration and attention. We consider the ethical significance and current regulation of human embryo research in Chapter 5.

III. Current Regulation

The following discussion provides an overview of the current state of regulation of the biotechnologies and practices discussed above. The discussion will be broadly divided into sections treating the governmental (federal and state) and nongovernmental regulation of assisted reproduction, both direct and indirect. Each source of regulation will be described in terms of its aims, animating values, jurisdictional scope and requirements, mechanisms of regulation, and efficacy.

1. Federal Oversight.

a. Consumer protection and embryo laboratory standards. There is only one federal statute that aims at the regulation of assisted reproduction: the Fertility Clinic Success Rate and Certification Act of 1992 (“the Act”).⁸⁵ The purposes of the statute and its related regulations are twofold: (1) to provide consumers with reliable and useful information about the efficacy of ART services offered by fertility clinics, and (2) to provide states with a model certification process for embryo laboratories.

2. State Oversight.

There are a variety of state laws that bear directly on the clinical practice of assisted reproduction. The vast majority of state statutes directly concerned with assisted reproduction, however, are concerned mostly with the question of access to such services. These states have legislative directives as to whether and to what extent assisted reproduction services will be covered as insurance benefits. Other state statutes regarding assisted reproduction aim to prevent the malfeasance of rogue practitioners (for example, California criminalizes unauthorized use of sperm, ova, and embryos). Still others focus on the regulation of gamete and embryo donation (for example, California sets forth screening requirements for donated sperm). There are a host of states whose laws dictate parental rights and obligations in the context of assisted reproduction.⁹⁵ A few jurisdictions (such as New Hampshire and Pennsylvania) have statutes that provide for fairly comprehensive regulation of the practitioners and participants in ART. Many jurisdictions have statutes that bear generally on the treatment and disposition of embryos, but only a subset of these jurisdictions explicitly speaks to the treatment of embryos in the context of assisted reproduction (including Louisiana, New Mexico, and South Dakota).

New Hampshire has an “In Vitro Fertilization and Pre-embryo Transfer” statutory scheme that provides that “IVF will be performed in accordance with the rules adopted by the [state] department of Health and Human Services.”⁹⁶ The state additionally specifies who may receive IVF treatment, namely, a woman who is at least twenty-one years of age, who has been medically evaluated for her “acceptability” to undergo the treatment (it is unclear what this means), and who has undergone requisite counseling.⁹⁷ New Hampshire likewise extends the medical and counseling requirement to the woman’s husband.⁹⁸

Pennsylvania also regulates ART as such, but focuses its efforts on record keeping and standards for maintenance of clinical facilities.⁹⁹ All IVF practitioners are required to submit reports and be available for inspection. The reports must include the names of the practitioners, their locations, the number of ova fertilized, the number of embryos destroyed or discarded, and the number of women “implanted with a fertilized egg.”

Louisiana, New Mexico, and South Dakota, as noted, have embryo experimentation statutes that directly speak to assisted reproduction.¹⁰⁰ The New Mexico statute prohibits any “clinical research activit[ies] involving fetuses, live-born infants or pregnant women.”¹⁰¹ Clinical research “includes research involving human in vitro fertilization, but . . . shall not include human in vitro fertilization performed to treat infertility; provided that this procedure shall include provisions to insure that each living fertilized ovum, zygote or embryo is implanted in a human female recipient . . .”¹⁰²There have been no court opinions interpreting this language, but some commentators suggest that this effectively proscribes the practice of IVF except in cases in which all embryos are transferred to the mother.¹⁰³

South Dakota, like New Mexico, prohibits “non-therapeutic research” on embryos. In contrast to New Mexico, however, it explicitly exempts from this definition “IVF and transfer, or diagnostic tests which may assist in the future care of a child subjected to this test.” Again, there are no cases interpreting this language, but it seems that this statute would not require the transfer to a uterus of all embryos created in the process of IVF.

Louisiana’s regulation of ART provides the highest level of protection to human embryos of any U.S. jurisdiction. It defines the in vitro embryo as a “juridical person” with nearly all of the attendant rights and protections of infants.^xxxi It stipulates that the use of an in vitro embryo must be solely for “the support and contribution of the complete development of human in utero implantation.” The production, culture, or use of human embryos for any other purpose is proscribed. An in vitro embryo is not considered the property of the clinician or the gamete donors. If the ART patients identify themselves as the embryo’s progenitors, they are deemed parents according to the Louisiana Civil Code. If the ART patients do not identify themselves, the “physician shall be deemed to be the temporary guardian . . . until adoptive implantation can occur.” The physician who creates the embryo through IVF is directly responsible for its safekeeping. The gamete donors owe the embryo “a high duty of care and prudent administration.” They may, however, renounce their parental rights through a formal proceeding, after which the embryo shall be available for adoptive implantation. Donors may convey their parental rights to another married couple, but only if “the other couple is willing and able to receive” the embryo. Under Louisiana law, the judicial standard governing any disputes involving the embryo is “the best interests of the embryo.” Thus, there can be no intentional destruction of a viable embryo.

Louisiana has also set standards for who may perform IVF and where IVF may be performed: It may be practiced only by a licensed physician in medical facilities that meet “the standards of [ASRM] and the American College of Obstetricians and Gynecologists.”

Some states have enacted statutes that preclude “experimentation” on human embryos. Given the experimental nature of certain ART procedures (such as preimplantation genetic diagnosis), these statutes might be construed broadly to reach such practices. Some individuals have challenged such statutes on constitutional grounds, arguing that the operative terms are so vague as to violate the constitutional guarantee of due process.^xxxii Practitioners have argued that they have not been adequately informed about which procedures could expose them to criminal liability. Courts in three jurisdictions have invalidated such statutes on these grounds.¹⁰⁴ One court among these three struck down the statute on the additional ground that it impermissibly infringed the plaintiff’s right to choose a particular means of reproduction, noting: “It takes no great leap of logic to see that within the cluster of constitutionally protected choices that includes access to contraceptives, there must be included within that cluster the right to submit to a medical procedure that may bring about, rather than prevent, pregnancy.”¹⁰⁵

In short, there are very few state laws that bear directly on assisted reproduction. Most of these laws relate to the provision of insurance coverage for infertility treatment. A few state laws directly relating to ART focus on health and safety concerns; a handful of states provide modest consumer protections. Some state laws regulating embryo research may indirectly affect the practice of assisted reproduction, though the decisional law in this area is unsettled. In the main, however, assisted reproduction is regulated at the state level by the same mechanisms that apply to the practice of medicine more generally, namely, through the licensure and certification of practitioners.

There are a number of state and federal governmental authorities that do not explicitly aim at the regulation of ART, but indirectly and incidentally provide some measure of oversight and direction.

2. State Oversight.