- Email: [email protected]
Ethical considerations in newer reproductive technologies
Ethical Considerations in Newer Reproductive Technologies Frank A. Chervenak*, Laurence B. McCullough†, and Zev Rosenwaks*
Ethics is an essential dimension of newer reproductive technologies. In this rapidly evolving field, there is a need for an ethical framework to guide both clinical practice and research. In this article, we provide such a framework that incorporates ethical principles, professional virtues, and the concept of the fetus as a patient. We then apply this framework to an important current clinical issue: the number of embryos to be transferred during in vitro fertilization; in addition to an emerging research issue: gene transfer research on in vitro embryos. © 2003 Elsevier Inc. All rights reserved. thics, as it has been practiced for centuries in all global cultures, is the disciplined study of morality. Medical ethics has also existed for centuries. The central question of medical ethics is, What are the doctor’s moral responsibilities to the patient, to health care organizations, and to society?1 Throughout its history, medical ethics has successfully addressed many clinical controversies. It has accomplished this by using the secular philosophical tools of ethical principles and professional virtues to create clinically applicable ethical frameworks.2,3 The ethical framework that we use in this article derives from the history of medical ethics.
E
Ethical Principles The ethical principle of beneficence has played a central role in the history of Western medical ethics since the time of Hippocrates. Beneficence identifies the physician’s responsibilities to protect and to promote the health-related interests of the patient, as these are understood from a rigorous clinical perspective. The principle of beneficence includes the principle of nonmaleficence, sometimes known as primum non nocere. The principle of respect for autonomy entered the history of medical ethics in the 19th and 20th centuries. This ethical principle identifies the physician’s responsibility to protect and to promote the patient’s interest, as these are understood from the patient’s perspective. The principle of justice, developed since the 18th century, identifies the physician’s responsibilities to protect and to promote the interests of health care organizations and society, particu-
larly their economic interests in the allocation of scarce resources.1,2
Professional Virtues The professional virtues became prominent in the 18th century in the English-language literature of medical ethics by 2 physician-philosophers, Dr. John Gregory and Thomas Percival.4-6 Their goal in writing and teaching medical ethics was to reform the market-driven, entrepreneurial, self-interested practice of medicine that was the dominant practice model of their day. Gregory and Percival called for a professional life of service to patients grounded in the professional virtues of the physician as moral fiduciary of the patient. This concept requires that the physician be scientifically and clinically competent and use his or her knowledge and skills primarily to benefit the patient and not primarily to advance the physician’s self-interest. Professional virtues create moral responsibilities for physicians by blunting self-interest and making the patients’ interests the primary consideration in the physician’s practice and research. Two such virtues are especially relevant.1 Self-sacrifice creates the responsibility to make reasonable From the *Department of Obstetrics and Gynecology, Weill Medical College of Cornell University, New York, NY; and †Center for Medical Ethics and Health Policy, Baylor College of Medicine, Houston, TX. Address reprint requests to Frank A. Chervenak, MD, The New York-Presbyterian Hospital, Weill Medical College of Cornell University, 525 East 68th St, M-7, New York, NY 10021; e-mail: [email protected]. © 2003 Elsevier Inc. All rights reserved. 0146-0005/03/2706-0002$30.00/0 doi:10.1053/j.semperi.2003.10.002
Seminars in Perinatology, Vol 27, No 6 (December), 2003: pp 427-434
427
428
Chervenak, McCullough, and Rosenwaks
sacrifices of self-interest, eg, income and prestige, as required to protect the health-related interests of the patients. Integrity creates the responsibility to practice medicine consistently according to standards of intellectual and moral excellence. Practicing medicine according to the best evidence, as opposed to idiosyncratic bias, is required by the professional virtue of integrity. An ethical framework for clinical decisionmaking, practice, and research integrates these principles and virtues into a coherent approach that guides and evaluates clinical judgment, practice, and research. This framework treats each principle and virtue as prima facie, ie, as binding unless it is shown by argument to be overridden by another principle or virtue.1,2 This approach avoids the intellectual gridlock and clinical disutility of absolute principles and virtues, ie, treating them as never allowing exceptions. A framework of prima facie principles and virtues makes possible more nuanced and therefore more clinically applicable judgments about the physician’s moral responsibility in matters of clinical ands research controversy.1
The Fetus as a Patient Ethical issues in reproductive technologies are a subset of obstetric ethics in which the concept of the fetus as a patient plays a central role.1 This concept therefore needs to be part of our ethical framework. The concept of the fetus as a patient should not be understood on the basis of the independent moral status of the fetus, ie, some feature(s) of the fetus that, independently of other entities, including the pregnant woman, physician, and the state, generates obligations of others to it, because all attempts to establish such status are doomed to failure. This failure results from the irreconcilable differences among philosophical and theological methods that have been deployed over the centuries of debate about the independent moral status of the fetus. Instead, the moral status of the fetus as a patient depends on 1) whether it is reliably expected later to achieve the relatively unambiguous moral status of becoming a child and, still later, the more unambiguous moral status of being a person; and 2) whether the fetus is presented for care. The fetus is a patient when reliable links exist between it and its later achiev-
ing moral status as a child and then person, and when it is presented to a physician.1 The first link between a fetus and its later achieving moral status as a child and then person is being presented to a physician and viability is the ability of the fetus to exist ex utero with the levels of technological support required to supplant immature or impaired anatomy and physiology through the neonatal period and into the second year of life, times at which no one disputes that childhood and then personhood exist. Viability is not an intrinsic characteristic of the fetus, but a function of both biology and technology. In developed countries, fetal viability occurs at approximately the 24th week of gestational age, as determined by reliable ultrasound dating.7 The second link between a fetus and its later achieving moral status as a child and then person is being presented to a physician and the decision of the pregnant woman to continue a previable pregnancy to viability and thus to term. That is, the only link between a previable fetus and its later achieving moral status as a child and then person is the pregnant woman’s autonomy, exercised in the decision not to terminate her pregnancy and to present the fetus (and herself) to the physician, because technological factors do not exist that can sustain the fetus ex utero. There are 4 important implications of the concept of the fetus as a patient. First, the concept does not rely on the concept of fetal rights or fetal personhood, but, in particular, the right to life. Second, for the previable fetus, including in vitro embryos, the woman’s autonomous decision to confer the status of being a patient is a necessary condition. Therefore, not all previable fetuses or in vitro embryos are patients. Patienthood status is therefore not dependent on location (in vitro versus in vivo), but rather is dependent on the woman’s establishing a linkage between the previable fetus or embryo and the child that the fetus or embryo can later become. Third, a woman is free to withdraw the status of being a patient from the previable embryo even when she has previously conferred such status. Therefore, within beneficence-based clinical judgment, abortion or fetal reductions are ethically justifiable options prior to fetal viability, because these do not involve killing a patient,
Ethics in Reproductive Technologies
which is (usually) prohibited in medical ethics.1,8
Numbers of Embryos to be Transferred During In Vitro Fertilization How many embryos to transfer during in vitro fertilization (IVF) cycles is a current controversy in the practice of the new reproductive technologies.9-22 Four ethical factors shape this controversy. First, the issue is addressed by the medical outcomes of pregnancy rate, live birth rate, and the rate of multiple pregnancies stratified by high-risk groups and techniques. This first factor dominates the literature. However, 3 other factors have been discussed: the role of the patient’s preferences, costs, and market forces of providers.23 The ethical framework set out above creates the basis for integrating all 4 factors into a coherent, clinically applicable account of the doctor’s responsibility regarding the number of embryos to be transferred in IVF.22 In IVF, the physician has beneficence-based obligations to the woman who wants to become pregnant and to fetal patients and future possible children who will result from the process. Beneficence-based obligations to the woman balance pregnancy success rates against the healthrelated risks of pregnancy with multiple gestations. The replacement of more embryos (provided that the replacement of more embryos results in higher pregnancy and live birth rates) favors success rates, while the replacement of fewer embryos (provided that the replacement of fewer embryos results in fewer multiple gestations) reduces risks of multiple pregnancy. Considered by itself, the principle of beneficence, applied to the pregnant woman’s healthrelated interests, does not automatically favor either more or fewer embryos being replaced. There are 2 important subgroups to be considered. If fetal reduction is not an option (ie, the woman would not withdraw the status of being a patient from any fetus), and the IVF procedure results in possibly more than 2 gestations and definitely more than 3 gestations, the health-related interests of the fetal patients are adversely affected by increased morbidity and mortality from prematurity risks.24 In this situation when fetal reduction is not an ethically acceptable option for the woman, based on her moral commitments and judgments, there is a
429
strong beneficence-based obligation to replace fewer embryos. In the situation in which fetal reduction is an ethically acceptable option of the pregnant woman, this strong beneficence-based obligation to replace fewer embryos is mitigated, but not entirely. This is because fetal reduction has an intrinsic risk of loss of the entire pregnancy.8 The principle of respect for autonomy is put into clinical practice through the informed consent process. This process has 3 basic elements: 1) the physician’s obligation to disclose an adequate amount of information about the clinical benefits and risks of IVF in terms of the number embryos to be replaced, and high risk groups and techniques; 2) the patient’s understanding of this information; and 3) the patient’s process of deciding, including making a voluntary decision.1,2,25 Beneficence-based clinical judgments and recommendations depend on the competence of physicians to evaluate patients’ health-related interests as required by the concept of being a moral fiduciary.5 Although pregnancy and childbirth have health-related aspects, the decisions to become pregnant, how aggressively to pursue treatment of infertility, and how many children should result from infertility treatment appeal predominantly to non-health–related values such as personal preference, family traditions and expectations, economics and social resources, and religious beliefs. Neither medical education nor clinical expertise bestows authoritative judgment. These matters appropriately fall under respect for the woman’s autonomy. Thus, the woman’s preference concerning how many children she would like to result from IVF treatment must be taken into clinical account in determining the number of embryos transferred.1 Decisions about IVF for the woman are cognitively demanding, as are many other clinical decisions made by patients. When patients struggle with such cognitively demanding decisions, the principle of respect for autonomy creates a responsibility for physicians to structure the consent process in ways that empower the patient’s decision-making capacities. Fisk and Trew9 have recently suggested that “even with counseling, the risks of triplets are generally underestimated by parents.” They conclude that the number of embryos to replace should not be left to the
430
Chervenak, McCullough, and Rosenwaks
woman’s choice. We do not agree. Instead, women should be assisted as necessary to estimate such risks reliably and balance them against the benefits of IVF. Fisk and Trew9 provide no evidence that women are incapable of benefiting from such assistance or performing this cognitively demanding task. They do suggest that assisted reproduction technology (ART) specialists are distanced from clinical practice. To the contrary, we believe that most ART practitioners are especially cognizant of such risks. Some patients may find it useful to consult with a maternal-fetal specialist who has experience with multiple gestations and can explain risks and benefits and assist her in her decision-making process. These decisions especially apply to the older patient when the transfer of 3 embryos appears to result in higher pregnancy rates.21 Some may be skeptical about the pressures on voluntary decision-making by women seeking IVF, thereby weakening autonomy-based obligations. Again, we disagree, because autonomybased obligations remain intact and take the form of efforts to shield women from undue pressure from family members, and take the time necessary to elicit her values and preferences from her, and assure her that the physician will support her in whatever decision she ultimately makes. Another crucial responsibility regarding respect for autonomy is for the physician to take steps to ensure that the pregnant woman’s decision about conferring or withdrawing patient status on the previable fetus(es) is ultimately hers. She is free to take account of the preferences of her partner, husband, or other family members and should be assured that they do not control this decision. In particular, while the husband and father’s interests are surely at stake, there is obvious moral asymmetry in that the burdens of pregnancy, fetal reduction, and childbirth fall on the woman and not the husband.1 In fulfilling the disclosure of obligation of the consent process, the physician is obligated to present and explain only medically reasonable alternatives, ie, those that are expected on balance to protect the patient’s health-related interests.1,25 Therefore, ineffective or harmful alternatives should not be offered. The patient exercises autonomy to select from among the medically reasonable alternatives, not to define
what those are. When adding embryos is not reliably expected to increase pregnancy or live birth rates, then that number adds no benefit, but introduces unnecessary incremental risk of multiple gestations and therefore is medically unreasonable. When added embryos decrease pregnancy or live birth rates, they are intrinsically harmful and also add to the risk of multiple gestations, making them even more medically unreasonable. It has long been known that good ethics should be based on good facts. Currently, the facts appear to support that pregnancy and live birth rates increase when more than 2 embryos are placed, especially for certain high-risk groups.22 There are major justice-based issues about IVF and about assisted-reproduction technologies generally. Chief among these is whether these services should be covered by insurance and national health plans, and to what degree. This is a very large issue because it concerns rights or entitlements to health care and health care policy, and whether infertility is or is not a disease, goes far beyond the scope of this articles. The authors believe that infertility is a serious condition worthy of medical attention. Faber19 has recently called attention to one justice-based consideration that is directly relevant to the subject at hand. He points out that in the United States most insurance plans do not cover ART procedures, but do cover pregnancy. Thus, patients take economic risks in starting a pregnancy, but not in continuing it. Thus, from the perspective of payers, a perverse economic incentive is introduced that gives patients little reason to take into account the overall subsequent costs of managing pregnancies initiated by IVF and ART.19 This is obviously an incoherent policy that, like all incoherent health care policy, strains against the demands of justice. As in many other medical specialties, when patient services in IVF produce a significant revenue stream, and when patient services bring prestige to individual physicians and their sponsoring institutions, self-interest can become a predominant consideration for the physician. This threatens the physician’s professional virtues. Professional integrity requires vigorous adherence to evidence-based standards and to the responsibilities of the informed consent process. In particular, the woman seeking IVF services
Ethics in Reproductive Technologies
should be provided with institution-specific information about that institution’s outcomes of pregnancy rates, live birth rates, and rates of multiple gestation per number of embryos transferred, stratified by high-risk group, embryo quality, and techniques used, as well as national information on the same parameters, so that the patient can make an informed comparative choice. However, there are limitations and perhaps even bias related to patient selection and practice in registries that can affect outcomes. Individual physicians and centers must be willing to accept the consequences of the virtue of self-sacrifice when patients, on the basis of this information, elect to go to centers with better outcomes. After all, IVF services in the United States function in a marketplace model and the ethics of the market strictly require informed consumer choice. Thus, failure to provide such information violates the virtues of integrity and self-sacrifice and the ethics of the market. Professional integrity also requires that the informed consent process be as free as possible from the bias introduced by self-interest, eg, a physician who starts the informed consent process by emphasizing the increased pregnancy rates from added embryos while under-emphasizing the risks of multiple gestation, to gain market advantage from high-pregnancy rates. This is obviously inappropriate, and self-sacrifice as well as integrity prohibit this morally corrupt practice.
Gene Transfer Research In the not-so-distant future, gene transfer technology will be combined with IVF and introduced into the clinical setting as human subjects research, which will raise significant ethical issues.26-28 Recent studies on the experience of subjects of research have shown that patients are not always aware when they are subjects of research.29 It has been suggested that the language used during the consent process may contribute to this disturbing lack of understanding.29 In our judgment, the use of the word “therapy” should be avoided to prevent this lack of understanding. Phrases such as “innovative therapy,” “gene therapy,” and “experimental therapy” therefore should never be used in consent forms or in discussions with women about their participation in gene transfer research. Instead, the con-
431
sent process and form should be explicit about the fact that the clinical application of gene therapy to the fetus at this time is research or experimentation. It is never obligatory for an individual who can consent for himself or herself to consent to become a subject for research. Nor is it obligatory for a surrogate, such as a parent, to consent for a patient not capable of participating in the consent process, eg, an infant or very young child, to become a subject of research.25 Therefore, it follows that no woman has an obligation to consent to gene therapy research on her in vitro embryo, even when she gives the embryo the status of being a patient. The consent process should make this moral fact very clear to the woman and to others who might be involved with her in the consent process. Institutional review boards should scrutinize consent forms and procedures to require efforts on the part of investigators to prevent coercion of a woman’s decision by internal factors such as unreasoning desperation, and external factors such as partners and family members. These recommendations parallel the ethically justified practice of protecting women from subtle coercion in decisions about using assisted-reproductive technologies generally. Gene therapy research will be new and the informed-consent process should be structured with this fact in mind. Therefore, we recommend that the consent process should begin with the research team inviting the woman to state what she understands about the embryo’s diagnosis, available alternatives for managing that diagnosis, including nontransfer, and the benefits and risks of those alternatives. She should be asked what she understands the prognosis to be if gene transfer is successful. In particular, women will need to understand the distinction between gene transfer for uniformly lethal conditions and gene transfer for conditions that result in serious morbidity. The research team should be attentive to factual errors and incompleteness in the woman’s fund of knowledge. Educating her about the protocol should begin by making sure her initial fund of knowledge is accurate, thus laying a solid informational foundation for the rest of the consent process. That process should continue with an explanation of the embryo’s genetic condition and
432
Chervenak, McCullough, and Rosenwaks
how the gene transfer research is designed to address that condition. She should be given information about the results of animal studies, especially about documented benefits and risks identified in such studies. She should also be informed about the unknown risks that transferred genes could function in unpredictable ways (the law of unintended consequences). She should then be assisted to identify her relevant values and beliefs. This can be accomplished in a nondirective fashion by asking what is important to her about becoming pregnant, about having children, and about having children with health problems that can range in their severity. She can then be asked to assess the offered gene therapy research on the basis of her values and beliefs, thus enhancing her autonomy in the consent process. Throughout the consent process and in the consent form, the options of abortion and nonintervention during pregnancy (assuming transfer is successful) should be presented as entirely acceptable to the research team.30 We make this recommendation to reinforce the nondirective character of the informed-consent process for gene therapy research. Current US federal regulations continue to require paternal consent.31 On the account we have given of the fetus as a patient, the father of the fetus does not determine whether the fetus is a patient. There is an obvious moral asymmetry between the father and the pregnant woman during pregnancy such that his role in decisions about interventions on the fetus should be a function of the pregnant woman’s autonomy. Even though the legal regulation is therefore ethically without support, civic responsibility requires that the regulation be followed (until such time as it is changed). Gene transfer research will be used especially to try to reduce the mortality of uniformly lethal conditions, such as alpha thalassemia, as the ability to diagnose them in vitro is created. The traditional logic of beneficence that drives such research has been that every reduction of mortality from such conditions is worth whatever morbidity that might result for survivors. In the clinical setting, especially in critical care, the traditional logic of beneficence has been appropriately challenged when morbidities eliminate or greatly impair developmental capacity of survivors. As McCormick32 put it over 30 years ago,
when critical care results in all of the patient’s energies being used in an irreversible struggle to survive, critical care intervention can be stopped. The moral lessons for gene transfer research are 2-fold. First, if animal studies reduce mortality but survivors are left with devastating morbidity, then human trials should not be started until animal outcomes improve. Second, human trials should include, as a stopping rule, high rates of occurrence of devastating fetal morbidity. Recall that the in vitro embryo is a patient solely as a function of the autonomy of the woman into whom the embryo will be transferred. For gene transfer on embryos, the exercise of such autonomy is greatly restricted in the absence of prenatal diagnosis to determine the effectiveness of the gene transfer. In particular, some women may want to terminate a pregnancy before viability, when there is no laboratory evidence of successful transfer. In our view, therefore, offering prenatal diagnosis should be required by Institutional Review Boards for gene therapy interventions with previable fetuses. The consent process should include careful explanation about the potential for false-negative and false-positive results. It is an accepted feature of study design in general that clinical trials should be conducted in such a way as to control for the idiosyncratic effects of patients’ preferences on results. This, for example, justifies a double-blind study design. For gene transfer research this general rule of study design raises significant ethical issues. On the one hand, to get the cleanest results one would not want any pregnancies in which gene transfer occurred to result in elective abortions. On the other, it would be desirable to prevent adverse outcomes of gene transfer through abortion in a study population of women who would accept this option. To address the first problem, one would exclude women who indicated any willingness to consider elective abortion. To address the second problem, one would exclude women who were opposed to abortion. Both solutions share a common and disabling ethical problem: They decide for the woman whether the previable fetus is a patient, thus unjustifiably overriding her autonomy in favor of research considerations, a paternalistic abuse of research subjects.
Ethics in Reproductive Technologies
To avoid this unacceptable ethical problem there should be no exclusion criteria for fetal gene transfer based on willingness to countenance elective abortion. Study designs would therefore have to include elective abortion and birth of adversely affected infants as hard endpoints. Germ-Line Risks
5.
6.
7.
Fletcher and Richter have raised the important ethical concern for germ-line harm that could result from the unknown harms of gene transfer. They propose that somatic cell gene transfer research “ought not be approved unless investigation in animal studies shows that the vector does not convey copies of exogenous genetic material into sex cells of fetuses.”27 This, in our view, is a prudent recommendation designed to prevent unnecessary harm to future generations. As fetal gene transfer research matures, this position may need to be reconsidered, especially when it is reliably thought that germ-line benefits convincingly outweigh germ-line harms. Any attempt to address this question will be controversial. 27
8.
9.
10.
11.
12.
13.
Conclusion 14.
Clinical practice and research in newer reproductive technologies is best guided and evaluated by an appropriate ethical framework. We have presented such a framework and have shown it enhances both clinical insight and management, and the design and conduct of research. We believe that clinical practice and research of newer reproductive technologies should incorporate ethics as an essential dimension.
15. 16.
17.
18.
References 1. McCullough LB, Chervenak FA: Ethics in Obstetrics and Gynecology, New York, NY, Oxford University Press, 1994 2. Beauchamp TL, Childress JF: Principles of Biomedical Ethics, ed 4. New York, NY, Oxford University Press, 1994 3. Gillon R: Principles of Health Care Ethics, Chichester, NY, John Wiley and Sons, 1994 4. Gregory J: Lectures on the duties and qualifications of a physician, in McCullough LB (ed): John Gregory’s Writings and Medical Ethics and Philosophy of Medicine
19.
20.
21.
433
Dordrecht, The Netherlands, Kluwer Academic Publishers, 1998, pp 161-245 McCullough LB: John Gregory and the Invention of Professional Medical Ethics and the Profession of Medicine, Dordrecht, The Netherlands, Kluwer Academic Publishers, 1998 Percival T: Medical Ethics, or a Code of Institutes and Precepts, Adapted to the Professional Conduct of Physicians and Surgeons, London, Johnson and Bickerstaff, 1803 Chervenak FA, McCullough LB: The limits of viability. J Perinatal Med 25:418-420, 1997 Chervenak FA, McCullough LB, Wapner R: Three ethically justified indications for selective termination in multifetal pregnancy: A practical and comprehensive management strategy. J Assist Reprod Genet 12:531-536, 1995 Fisk NM, Trew G: Commentary: Two’s company, three’s a crowd for embryo transfer. The Lancet 354:1572-1573, 1999 Bergh T, Ericson A, Hillensjo¨ T, et al: Deliveries and children born after in-vitro fertilisation in Sweden 198295: A retrospective cohort study. The Lancet 354:15791585, 1999 Schieve LA, Peterson HB, Meikle SF, et al: Live-birth rates and multiple-birth risk using in vitro fertilization. JAMA 282:1832-1838, 1999 Templeton A, Morris JK: Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. New Engl J Med 339:573-577, 1998 Elsner CW, Tucker MJ, Sweitzer CL, et al: Transactions of the fifty-ninth annual meeting of the South Atlantic Association of Obstetricians and Gynecologists. Am J Ob Gyn 177:350-357, 1997 The Practice Committee Report. Guidelines on Number of Embryos Transferred. Amer Soc Reprod Med Nov 1999; 1-2. Hecht BR, Magoon MW: Can the epidemic of iatrogenic multiples be conquered? Clin Ob Gyn 41:127-137, 1998 Roest J, Mous HVH, van Heusden AM, et al: A triplet pregnancy after in vitro fertilization is a procedure-related complication that should be prevented by replacement of two embryos only. Fert Steril 67:290-295, 1997 Matson PL, Browne J, Deakin R, et al: Clinical assisted reproduction: the transfer of two embryos instead of three to reduce the risk of multiple pregnancy: A retrospective analysis. J Assist Reprod Genetics 16:1-5, 1999 Vauthier-Brouzes D, Gonzales J, Lefebvre G, et al: How many embryos should be transferred in in vitro fertilization? A prospective randomized study. Fert Steril 62:339342, 1994 Faber K: IVF in the US: Multiple gestation, economic competition, and the necessity of excess. Human Reprod 12:1614-1616, 1997 Damario MA, Davis OK, Rosenwaks Z: The role of maternal age in the assisted reproductive technologies. Reprod Med Rev 7:141-160, 1999 Veeck LL, Davis OK, Rosenwaks Z: How to avoid multiple pregnancy during artificial reproductive treatment in iatrogenic multiple pregnancy, in Blickstein I, Keith LG (eds): Iatrogenic Multiple Pregnancy: Clinical Implications London, Parthenon, 2001, pp 45-64
434
Chervenak, McCullough, and Rosenwaks
22. Chervenak FA, McCullough LB, Rosenwaks Z: Ethical dimensions of the number of embryos to be transferred in in vitro fertilization. J Assist Reprod Genet 185:583587, 2001 23. Nelson LJ: Preimplantation diagnosis. Clin Perinatol 30: 67-80, 2003 24. Blickstein I, Weissman A: Estimating the risk of cerebral palsy after assisted conceptions. Lancet 341:1313-1314, 1999 25. Faden RR, Beauchamp TL: A History and Theory of Informed Consent, New York, NY, Oxford University Press, 1986 26. Chervenak FA, McCullough LB: Ethics on the frontier of fetal research, in Carrera JM, Kurjak A, Chervenak FA, (eds): Controversies in Perinatal Medicine. New York, NY, Parthenon Publishing, 2003, pp 296-302
27. Fletcher JC, Richter G: Human fetal gene therapy: moral and ethical questions. Human Gene Therapy 7:16051614, 1996 28. Anderson WF, Fletcher JC: Gene therapy in human beings: When is it ethical to begin? N Eng J Med 303:12931297, 1980 29. Sugarman J, Kass NE, Goodman SN, et al: What patients say about medical research. IRB 20:1-7, 1998 30. Chervenak FA, McCullough LB: A comprehensive ethical framework for fetal research and its application to fetal surgery and spina bifida. Am J Obstetf Gynecol 187:10-14, 2002 31. Brody BA: The Ethics of Research: An International Perspective. New York, NY, Oxford University Press, 1998 32. McCormick RJ: To save or let die: the dilemma of modern medicine. JAMA 229:172-176, 1974
Ethics is an essential dimension of newer reproductive technologies. In this rapidly evolving field, there is a need for an ethical framework to guide both clinical practice and research. In this article, we provide such a framework that incorporates ethical principles, professional virtues, and the concept of the fetus as a patient. We then apply this framework to an important current clinical issue: the number of embryos to be transferred during in vitro fertilization; in addition to an emerging research issue: gene transfer research on in vitro embryos. © 2003 Elsevier Inc. All rights reserved. thics, as it has been practiced for centuries in all global cultures, is the disciplined study of morality. Medical ethics has also existed for centuries. The central question of medical ethics is, What are the doctor’s moral responsibilities to the patient, to health care organizations, and to society?1 Throughout its history, medical ethics has successfully addressed many clinical controversies. It has accomplished this by using the secular philosophical tools of ethical principles and professional virtues to create clinically applicable ethical frameworks.2,3 The ethical framework that we use in this article derives from the history of medical ethics.
E
Ethical Principles The ethical principle of beneficence has played a central role in the history of Western medical ethics since the time of Hippocrates. Beneficence identifies the physician’s responsibilities to protect and to promote the health-related interests of the patient, as these are understood from a rigorous clinical perspective. The principle of beneficence includes the principle of nonmaleficence, sometimes known as primum non nocere. The principle of respect for autonomy entered the history of medical ethics in the 19th and 20th centuries. This ethical principle identifies the physician’s responsibility to protect and to promote the patient’s interest, as these are understood from the patient’s perspective. The principle of justice, developed since the 18th century, identifies the physician’s responsibilities to protect and to promote the interests of health care organizations and society, particu-
larly their economic interests in the allocation of scarce resources.1,2
Professional Virtues The professional virtues became prominent in the 18th century in the English-language literature of medical ethics by 2 physician-philosophers, Dr. John Gregory and Thomas Percival.4-6 Their goal in writing and teaching medical ethics was to reform the market-driven, entrepreneurial, self-interested practice of medicine that was the dominant practice model of their day. Gregory and Percival called for a professional life of service to patients grounded in the professional virtues of the physician as moral fiduciary of the patient. This concept requires that the physician be scientifically and clinically competent and use his or her knowledge and skills primarily to benefit the patient and not primarily to advance the physician’s self-interest. Professional virtues create moral responsibilities for physicians by blunting self-interest and making the patients’ interests the primary consideration in the physician’s practice and research. Two such virtues are especially relevant.1 Self-sacrifice creates the responsibility to make reasonable From the *Department of Obstetrics and Gynecology, Weill Medical College of Cornell University, New York, NY; and †Center for Medical Ethics and Health Policy, Baylor College of Medicine, Houston, TX. Address reprint requests to Frank A. Chervenak, MD, The New York-Presbyterian Hospital, Weill Medical College of Cornell University, 525 East 68th St, M-7, New York, NY 10021; e-mail: [email protected]. © 2003 Elsevier Inc. All rights reserved. 0146-0005/03/2706-0002$30.00/0 doi:10.1053/j.semperi.2003.10.002
Seminars in Perinatology, Vol 27, No 6 (December), 2003: pp 427-434
427
428
Chervenak, McCullough, and Rosenwaks
sacrifices of self-interest, eg, income and prestige, as required to protect the health-related interests of the patients. Integrity creates the responsibility to practice medicine consistently according to standards of intellectual and moral excellence. Practicing medicine according to the best evidence, as opposed to idiosyncratic bias, is required by the professional virtue of integrity. An ethical framework for clinical decisionmaking, practice, and research integrates these principles and virtues into a coherent approach that guides and evaluates clinical judgment, practice, and research. This framework treats each principle and virtue as prima facie, ie, as binding unless it is shown by argument to be overridden by another principle or virtue.1,2 This approach avoids the intellectual gridlock and clinical disutility of absolute principles and virtues, ie, treating them as never allowing exceptions. A framework of prima facie principles and virtues makes possible more nuanced and therefore more clinically applicable judgments about the physician’s moral responsibility in matters of clinical ands research controversy.1
The Fetus as a Patient Ethical issues in reproductive technologies are a subset of obstetric ethics in which the concept of the fetus as a patient plays a central role.1 This concept therefore needs to be part of our ethical framework. The concept of the fetus as a patient should not be understood on the basis of the independent moral status of the fetus, ie, some feature(s) of the fetus that, independently of other entities, including the pregnant woman, physician, and the state, generates obligations of others to it, because all attempts to establish such status are doomed to failure. This failure results from the irreconcilable differences among philosophical and theological methods that have been deployed over the centuries of debate about the independent moral status of the fetus. Instead, the moral status of the fetus as a patient depends on 1) whether it is reliably expected later to achieve the relatively unambiguous moral status of becoming a child and, still later, the more unambiguous moral status of being a person; and 2) whether the fetus is presented for care. The fetus is a patient when reliable links exist between it and its later achiev-
ing moral status as a child and then person, and when it is presented to a physician.1 The first link between a fetus and its later achieving moral status as a child and then person is being presented to a physician and viability is the ability of the fetus to exist ex utero with the levels of technological support required to supplant immature or impaired anatomy and physiology through the neonatal period and into the second year of life, times at which no one disputes that childhood and then personhood exist. Viability is not an intrinsic characteristic of the fetus, but a function of both biology and technology. In developed countries, fetal viability occurs at approximately the 24th week of gestational age, as determined by reliable ultrasound dating.7 The second link between a fetus and its later achieving moral status as a child and then person is being presented to a physician and the decision of the pregnant woman to continue a previable pregnancy to viability and thus to term. That is, the only link between a previable fetus and its later achieving moral status as a child and then person is the pregnant woman’s autonomy, exercised in the decision not to terminate her pregnancy and to present the fetus (and herself) to the physician, because technological factors do not exist that can sustain the fetus ex utero. There are 4 important implications of the concept of the fetus as a patient. First, the concept does not rely on the concept of fetal rights or fetal personhood, but, in particular, the right to life. Second, for the previable fetus, including in vitro embryos, the woman’s autonomous decision to confer the status of being a patient is a necessary condition. Therefore, not all previable fetuses or in vitro embryos are patients. Patienthood status is therefore not dependent on location (in vitro versus in vivo), but rather is dependent on the woman’s establishing a linkage between the previable fetus or embryo and the child that the fetus or embryo can later become. Third, a woman is free to withdraw the status of being a patient from the previable embryo even when she has previously conferred such status. Therefore, within beneficence-based clinical judgment, abortion or fetal reductions are ethically justifiable options prior to fetal viability, because these do not involve killing a patient,
Ethics in Reproductive Technologies
which is (usually) prohibited in medical ethics.1,8
Numbers of Embryos to be Transferred During In Vitro Fertilization How many embryos to transfer during in vitro fertilization (IVF) cycles is a current controversy in the practice of the new reproductive technologies.9-22 Four ethical factors shape this controversy. First, the issue is addressed by the medical outcomes of pregnancy rate, live birth rate, and the rate of multiple pregnancies stratified by high-risk groups and techniques. This first factor dominates the literature. However, 3 other factors have been discussed: the role of the patient’s preferences, costs, and market forces of providers.23 The ethical framework set out above creates the basis for integrating all 4 factors into a coherent, clinically applicable account of the doctor’s responsibility regarding the number of embryos to be transferred in IVF.22 In IVF, the physician has beneficence-based obligations to the woman who wants to become pregnant and to fetal patients and future possible children who will result from the process. Beneficence-based obligations to the woman balance pregnancy success rates against the healthrelated risks of pregnancy with multiple gestations. The replacement of more embryos (provided that the replacement of more embryos results in higher pregnancy and live birth rates) favors success rates, while the replacement of fewer embryos (provided that the replacement of fewer embryos results in fewer multiple gestations) reduces risks of multiple pregnancy. Considered by itself, the principle of beneficence, applied to the pregnant woman’s healthrelated interests, does not automatically favor either more or fewer embryos being replaced. There are 2 important subgroups to be considered. If fetal reduction is not an option (ie, the woman would not withdraw the status of being a patient from any fetus), and the IVF procedure results in possibly more than 2 gestations and definitely more than 3 gestations, the health-related interests of the fetal patients are adversely affected by increased morbidity and mortality from prematurity risks.24 In this situation when fetal reduction is not an ethically acceptable option for the woman, based on her moral commitments and judgments, there is a
429
strong beneficence-based obligation to replace fewer embryos. In the situation in which fetal reduction is an ethically acceptable option of the pregnant woman, this strong beneficence-based obligation to replace fewer embryos is mitigated, but not entirely. This is because fetal reduction has an intrinsic risk of loss of the entire pregnancy.8 The principle of respect for autonomy is put into clinical practice through the informed consent process. This process has 3 basic elements: 1) the physician’s obligation to disclose an adequate amount of information about the clinical benefits and risks of IVF in terms of the number embryos to be replaced, and high risk groups and techniques; 2) the patient’s understanding of this information; and 3) the patient’s process of deciding, including making a voluntary decision.1,2,25 Beneficence-based clinical judgments and recommendations depend on the competence of physicians to evaluate patients’ health-related interests as required by the concept of being a moral fiduciary.5 Although pregnancy and childbirth have health-related aspects, the decisions to become pregnant, how aggressively to pursue treatment of infertility, and how many children should result from infertility treatment appeal predominantly to non-health–related values such as personal preference, family traditions and expectations, economics and social resources, and religious beliefs. Neither medical education nor clinical expertise bestows authoritative judgment. These matters appropriately fall under respect for the woman’s autonomy. Thus, the woman’s preference concerning how many children she would like to result from IVF treatment must be taken into clinical account in determining the number of embryos transferred.1 Decisions about IVF for the woman are cognitively demanding, as are many other clinical decisions made by patients. When patients struggle with such cognitively demanding decisions, the principle of respect for autonomy creates a responsibility for physicians to structure the consent process in ways that empower the patient’s decision-making capacities. Fisk and Trew9 have recently suggested that “even with counseling, the risks of triplets are generally underestimated by parents.” They conclude that the number of embryos to replace should not be left to the
430
Chervenak, McCullough, and Rosenwaks
woman’s choice. We do not agree. Instead, women should be assisted as necessary to estimate such risks reliably and balance them against the benefits of IVF. Fisk and Trew9 provide no evidence that women are incapable of benefiting from such assistance or performing this cognitively demanding task. They do suggest that assisted reproduction technology (ART) specialists are distanced from clinical practice. To the contrary, we believe that most ART practitioners are especially cognizant of such risks. Some patients may find it useful to consult with a maternal-fetal specialist who has experience with multiple gestations and can explain risks and benefits and assist her in her decision-making process. These decisions especially apply to the older patient when the transfer of 3 embryos appears to result in higher pregnancy rates.21 Some may be skeptical about the pressures on voluntary decision-making by women seeking IVF, thereby weakening autonomy-based obligations. Again, we disagree, because autonomybased obligations remain intact and take the form of efforts to shield women from undue pressure from family members, and take the time necessary to elicit her values and preferences from her, and assure her that the physician will support her in whatever decision she ultimately makes. Another crucial responsibility regarding respect for autonomy is for the physician to take steps to ensure that the pregnant woman’s decision about conferring or withdrawing patient status on the previable fetus(es) is ultimately hers. She is free to take account of the preferences of her partner, husband, or other family members and should be assured that they do not control this decision. In particular, while the husband and father’s interests are surely at stake, there is obvious moral asymmetry in that the burdens of pregnancy, fetal reduction, and childbirth fall on the woman and not the husband.1 In fulfilling the disclosure of obligation of the consent process, the physician is obligated to present and explain only medically reasonable alternatives, ie, those that are expected on balance to protect the patient’s health-related interests.1,25 Therefore, ineffective or harmful alternatives should not be offered. The patient exercises autonomy to select from among the medically reasonable alternatives, not to define
what those are. When adding embryos is not reliably expected to increase pregnancy or live birth rates, then that number adds no benefit, but introduces unnecessary incremental risk of multiple gestations and therefore is medically unreasonable. When added embryos decrease pregnancy or live birth rates, they are intrinsically harmful and also add to the risk of multiple gestations, making them even more medically unreasonable. It has long been known that good ethics should be based on good facts. Currently, the facts appear to support that pregnancy and live birth rates increase when more than 2 embryos are placed, especially for certain high-risk groups.22 There are major justice-based issues about IVF and about assisted-reproduction technologies generally. Chief among these is whether these services should be covered by insurance and national health plans, and to what degree. This is a very large issue because it concerns rights or entitlements to health care and health care policy, and whether infertility is or is not a disease, goes far beyond the scope of this articles. The authors believe that infertility is a serious condition worthy of medical attention. Faber19 has recently called attention to one justice-based consideration that is directly relevant to the subject at hand. He points out that in the United States most insurance plans do not cover ART procedures, but do cover pregnancy. Thus, patients take economic risks in starting a pregnancy, but not in continuing it. Thus, from the perspective of payers, a perverse economic incentive is introduced that gives patients little reason to take into account the overall subsequent costs of managing pregnancies initiated by IVF and ART.19 This is obviously an incoherent policy that, like all incoherent health care policy, strains against the demands of justice. As in many other medical specialties, when patient services in IVF produce a significant revenue stream, and when patient services bring prestige to individual physicians and their sponsoring institutions, self-interest can become a predominant consideration for the physician. This threatens the physician’s professional virtues. Professional integrity requires vigorous adherence to evidence-based standards and to the responsibilities of the informed consent process. In particular, the woman seeking IVF services
Ethics in Reproductive Technologies
should be provided with institution-specific information about that institution’s outcomes of pregnancy rates, live birth rates, and rates of multiple gestation per number of embryos transferred, stratified by high-risk group, embryo quality, and techniques used, as well as national information on the same parameters, so that the patient can make an informed comparative choice. However, there are limitations and perhaps even bias related to patient selection and practice in registries that can affect outcomes. Individual physicians and centers must be willing to accept the consequences of the virtue of self-sacrifice when patients, on the basis of this information, elect to go to centers with better outcomes. After all, IVF services in the United States function in a marketplace model and the ethics of the market strictly require informed consumer choice. Thus, failure to provide such information violates the virtues of integrity and self-sacrifice and the ethics of the market. Professional integrity also requires that the informed consent process be as free as possible from the bias introduced by self-interest, eg, a physician who starts the informed consent process by emphasizing the increased pregnancy rates from added embryos while under-emphasizing the risks of multiple gestation, to gain market advantage from high-pregnancy rates. This is obviously inappropriate, and self-sacrifice as well as integrity prohibit this morally corrupt practice.
Gene Transfer Research In the not-so-distant future, gene transfer technology will be combined with IVF and introduced into the clinical setting as human subjects research, which will raise significant ethical issues.26-28 Recent studies on the experience of subjects of research have shown that patients are not always aware when they are subjects of research.29 It has been suggested that the language used during the consent process may contribute to this disturbing lack of understanding.29 In our judgment, the use of the word “therapy” should be avoided to prevent this lack of understanding. Phrases such as “innovative therapy,” “gene therapy,” and “experimental therapy” therefore should never be used in consent forms or in discussions with women about their participation in gene transfer research. Instead, the con-
431
sent process and form should be explicit about the fact that the clinical application of gene therapy to the fetus at this time is research or experimentation. It is never obligatory for an individual who can consent for himself or herself to consent to become a subject for research. Nor is it obligatory for a surrogate, such as a parent, to consent for a patient not capable of participating in the consent process, eg, an infant or very young child, to become a subject of research.25 Therefore, it follows that no woman has an obligation to consent to gene therapy research on her in vitro embryo, even when she gives the embryo the status of being a patient. The consent process should make this moral fact very clear to the woman and to others who might be involved with her in the consent process. Institutional review boards should scrutinize consent forms and procedures to require efforts on the part of investigators to prevent coercion of a woman’s decision by internal factors such as unreasoning desperation, and external factors such as partners and family members. These recommendations parallel the ethically justified practice of protecting women from subtle coercion in decisions about using assisted-reproductive technologies generally. Gene therapy research will be new and the informed-consent process should be structured with this fact in mind. Therefore, we recommend that the consent process should begin with the research team inviting the woman to state what she understands about the embryo’s diagnosis, available alternatives for managing that diagnosis, including nontransfer, and the benefits and risks of those alternatives. She should be asked what she understands the prognosis to be if gene transfer is successful. In particular, women will need to understand the distinction between gene transfer for uniformly lethal conditions and gene transfer for conditions that result in serious morbidity. The research team should be attentive to factual errors and incompleteness in the woman’s fund of knowledge. Educating her about the protocol should begin by making sure her initial fund of knowledge is accurate, thus laying a solid informational foundation for the rest of the consent process. That process should continue with an explanation of the embryo’s genetic condition and
432
Chervenak, McCullough, and Rosenwaks
how the gene transfer research is designed to address that condition. She should be given information about the results of animal studies, especially about documented benefits and risks identified in such studies. She should also be informed about the unknown risks that transferred genes could function in unpredictable ways (the law of unintended consequences). She should then be assisted to identify her relevant values and beliefs. This can be accomplished in a nondirective fashion by asking what is important to her about becoming pregnant, about having children, and about having children with health problems that can range in their severity. She can then be asked to assess the offered gene therapy research on the basis of her values and beliefs, thus enhancing her autonomy in the consent process. Throughout the consent process and in the consent form, the options of abortion and nonintervention during pregnancy (assuming transfer is successful) should be presented as entirely acceptable to the research team.30 We make this recommendation to reinforce the nondirective character of the informed-consent process for gene therapy research. Current US federal regulations continue to require paternal consent.31 On the account we have given of the fetus as a patient, the father of the fetus does not determine whether the fetus is a patient. There is an obvious moral asymmetry between the father and the pregnant woman during pregnancy such that his role in decisions about interventions on the fetus should be a function of the pregnant woman’s autonomy. Even though the legal regulation is therefore ethically without support, civic responsibility requires that the regulation be followed (until such time as it is changed). Gene transfer research will be used especially to try to reduce the mortality of uniformly lethal conditions, such as alpha thalassemia, as the ability to diagnose them in vitro is created. The traditional logic of beneficence that drives such research has been that every reduction of mortality from such conditions is worth whatever morbidity that might result for survivors. In the clinical setting, especially in critical care, the traditional logic of beneficence has been appropriately challenged when morbidities eliminate or greatly impair developmental capacity of survivors. As McCormick32 put it over 30 years ago,
when critical care results in all of the patient’s energies being used in an irreversible struggle to survive, critical care intervention can be stopped. The moral lessons for gene transfer research are 2-fold. First, if animal studies reduce mortality but survivors are left with devastating morbidity, then human trials should not be started until animal outcomes improve. Second, human trials should include, as a stopping rule, high rates of occurrence of devastating fetal morbidity. Recall that the in vitro embryo is a patient solely as a function of the autonomy of the woman into whom the embryo will be transferred. For gene transfer on embryos, the exercise of such autonomy is greatly restricted in the absence of prenatal diagnosis to determine the effectiveness of the gene transfer. In particular, some women may want to terminate a pregnancy before viability, when there is no laboratory evidence of successful transfer. In our view, therefore, offering prenatal diagnosis should be required by Institutional Review Boards for gene therapy interventions with previable fetuses. The consent process should include careful explanation about the potential for false-negative and false-positive results. It is an accepted feature of study design in general that clinical trials should be conducted in such a way as to control for the idiosyncratic effects of patients’ preferences on results. This, for example, justifies a double-blind study design. For gene transfer research this general rule of study design raises significant ethical issues. On the one hand, to get the cleanest results one would not want any pregnancies in which gene transfer occurred to result in elective abortions. On the other, it would be desirable to prevent adverse outcomes of gene transfer through abortion in a study population of women who would accept this option. To address the first problem, one would exclude women who indicated any willingness to consider elective abortion. To address the second problem, one would exclude women who were opposed to abortion. Both solutions share a common and disabling ethical problem: They decide for the woman whether the previable fetus is a patient, thus unjustifiably overriding her autonomy in favor of research considerations, a paternalistic abuse of research subjects.
Ethics in Reproductive Technologies
To avoid this unacceptable ethical problem there should be no exclusion criteria for fetal gene transfer based on willingness to countenance elective abortion. Study designs would therefore have to include elective abortion and birth of adversely affected infants as hard endpoints. Germ-Line Risks
5.
6.
7.
Fletcher and Richter have raised the important ethical concern for germ-line harm that could result from the unknown harms of gene transfer. They propose that somatic cell gene transfer research “ought not be approved unless investigation in animal studies shows that the vector does not convey copies of exogenous genetic material into sex cells of fetuses.”27 This, in our view, is a prudent recommendation designed to prevent unnecessary harm to future generations. As fetal gene transfer research matures, this position may need to be reconsidered, especially when it is reliably thought that germ-line benefits convincingly outweigh germ-line harms. Any attempt to address this question will be controversial. 27
8.
9.
10.
11.
12.
13.
Conclusion 14.
Clinical practice and research in newer reproductive technologies is best guided and evaluated by an appropriate ethical framework. We have presented such a framework and have shown it enhances both clinical insight and management, and the design and conduct of research. We believe that clinical practice and research of newer reproductive technologies should incorporate ethics as an essential dimension.
15. 16.
17.
18.
References 1. McCullough LB, Chervenak FA: Ethics in Obstetrics and Gynecology, New York, NY, Oxford University Press, 1994 2. Beauchamp TL, Childress JF: Principles of Biomedical Ethics, ed 4. New York, NY, Oxford University Press, 1994 3. Gillon R: Principles of Health Care Ethics, Chichester, NY, John Wiley and Sons, 1994 4. Gregory J: Lectures on the duties and qualifications of a physician, in McCullough LB (ed): John Gregory’s Writings and Medical Ethics and Philosophy of Medicine
19.
20.
21.
433
Dordrecht, The Netherlands, Kluwer Academic Publishers, 1998, pp 161-245 McCullough LB: John Gregory and the Invention of Professional Medical Ethics and the Profession of Medicine, Dordrecht, The Netherlands, Kluwer Academic Publishers, 1998 Percival T: Medical Ethics, or a Code of Institutes and Precepts, Adapted to the Professional Conduct of Physicians and Surgeons, London, Johnson and Bickerstaff, 1803 Chervenak FA, McCullough LB: The limits of viability. J Perinatal Med 25:418-420, 1997 Chervenak FA, McCullough LB, Wapner R: Three ethically justified indications for selective termination in multifetal pregnancy: A practical and comprehensive management strategy. J Assist Reprod Genet 12:531-536, 1995 Fisk NM, Trew G: Commentary: Two’s company, three’s a crowd for embryo transfer. The Lancet 354:1572-1573, 1999 Bergh T, Ericson A, Hillensjo¨ T, et al: Deliveries and children born after in-vitro fertilisation in Sweden 198295: A retrospective cohort study. The Lancet 354:15791585, 1999 Schieve LA, Peterson HB, Meikle SF, et al: Live-birth rates and multiple-birth risk using in vitro fertilization. JAMA 282:1832-1838, 1999 Templeton A, Morris JK: Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. New Engl J Med 339:573-577, 1998 Elsner CW, Tucker MJ, Sweitzer CL, et al: Transactions of the fifty-ninth annual meeting of the South Atlantic Association of Obstetricians and Gynecologists. Am J Ob Gyn 177:350-357, 1997 The Practice Committee Report. Guidelines on Number of Embryos Transferred. Amer Soc Reprod Med Nov 1999; 1-2. Hecht BR, Magoon MW: Can the epidemic of iatrogenic multiples be conquered? Clin Ob Gyn 41:127-137, 1998 Roest J, Mous HVH, van Heusden AM, et al: A triplet pregnancy after in vitro fertilization is a procedure-related complication that should be prevented by replacement of two embryos only. Fert Steril 67:290-295, 1997 Matson PL, Browne J, Deakin R, et al: Clinical assisted reproduction: the transfer of two embryos instead of three to reduce the risk of multiple pregnancy: A retrospective analysis. J Assist Reprod Genetics 16:1-5, 1999 Vauthier-Brouzes D, Gonzales J, Lefebvre G, et al: How many embryos should be transferred in in vitro fertilization? A prospective randomized study. Fert Steril 62:339342, 1994 Faber K: IVF in the US: Multiple gestation, economic competition, and the necessity of excess. Human Reprod 12:1614-1616, 1997 Damario MA, Davis OK, Rosenwaks Z: The role of maternal age in the assisted reproductive technologies. Reprod Med Rev 7:141-160, 1999 Veeck LL, Davis OK, Rosenwaks Z: How to avoid multiple pregnancy during artificial reproductive treatment in iatrogenic multiple pregnancy, in Blickstein I, Keith LG (eds): Iatrogenic Multiple Pregnancy: Clinical Implications London, Parthenon, 2001, pp 45-64
434
Chervenak, McCullough, and Rosenwaks
22. Chervenak FA, McCullough LB, Rosenwaks Z: Ethical dimensions of the number of embryos to be transferred in in vitro fertilization. J Assist Reprod Genet 185:583587, 2001 23. Nelson LJ: Preimplantation diagnosis. Clin Perinatol 30: 67-80, 2003 24. Blickstein I, Weissman A: Estimating the risk of cerebral palsy after assisted conceptions. Lancet 341:1313-1314, 1999 25. Faden RR, Beauchamp TL: A History and Theory of Informed Consent, New York, NY, Oxford University Press, 1986 26. Chervenak FA, McCullough LB: Ethics on the frontier of fetal research, in Carrera JM, Kurjak A, Chervenak FA, (eds): Controversies in Perinatal Medicine. New York, NY, Parthenon Publishing, 2003, pp 296-302
27. Fletcher JC, Richter G: Human fetal gene therapy: moral and ethical questions. Human Gene Therapy 7:16051614, 1996 28. Anderson WF, Fletcher JC: Gene therapy in human beings: When is it ethical to begin? N Eng J Med 303:12931297, 1980 29. Sugarman J, Kass NE, Goodman SN, et al: What patients say about medical research. IRB 20:1-7, 1998 30. Chervenak FA, McCullough LB: A comprehensive ethical framework for fetal research and its application to fetal surgery and spina bifida. Am J Obstetf Gynecol 187:10-14, 2002 31. Brody BA: The Ethics of Research: An International Perspective. New York, NY, Oxford University Press, 1998 32. McCormick RJ: To save or let die: the dilemma of modern medicine. JAMA 229:172-176, 1974