Avoiding multiple pregnancies in ART

Multiple pregnancies: a test case for the moral quality of medically assisted reproduction

Guido Pennings

Department of Philosophy, Free University Brussels, Pleinlaan 2, Lok. 5 C 442, B-1050 Brussels, Belgium.E-mail: gpenning{at}vub.ac.be


    Abstract
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
Although most professional societies have issued guidelines to diminish the number of embryos to be transferred during assisted reproduction techniques, the incidence of multiple pregnancies remains unacceptably high. The negative psychological, social and medical consequences for the patients and their offspring easily outweigh the benefits in terms of increased success rates. Multiple pregnancies would never be tolerated if the `best interest of the child' standard was applied as strictly to these consequences, as it is to controversial family forms. The persistence of high multiple pregnancy rates is largely due to the pressure brought to bear on the physicians to increase the overall success rate. The fertility specialist should inform the patients about the risks and benefits of a multiple transfer but ultimately the specialist should decide how many embryos to transfer. Multifetal reduction is an ethically acceptable solution if, and only if, the physician has taken all reasonable steps to prevent the occurrence of a multiple pregnancy. Finally, an additional strategy to decrease the incidence of multiple pregnancies is proposed, i.e. to extend the professional responsibility of the fertility specialist to all steps of procreation including pregnancy, birth and neonatal care.

Key words: child welfare/embryo reduction/ethics/multiple pregnancy/success rate


    Introduction
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
`The establishment of the pregnancy was successful. Regrettably, the children are handicapped and the mother suffers from depression'. This paraphrase of the famous criticism of medicine `The operation was a success. Regrettably, the patient died' expresses the practice of IVF when several embryos are replaced routinely. Somehow the establishment of a pregnancy has become separated from the general goal of the well-being of the patient and her future children. This is clearly demonstrated by the fact that multiple pregnancies are counted as successes in the general standard used to evaluate fertility centres. The consequences for the patient have faded out of the picture.

The accumulation of information about the possible consequences of multiple pregnancies presents a very disturbing picture. The `blessing' of twins and triplets has turned out to be a `poisoned gift'. Multiple pregnancies are the cause of a strong increase in obstetric complications, perinatal morbidity, congenital malformations, maternal and fetal mortality and long-term social, psychological and economic difficulties (Hazekamp et al., 2000Go; Olivennes, 2000Go). The call to do something about the very high incidence of multiple pregnancies among the pregnancies established by assisted reproductive treatment started about a decade ago and grows ever louder. Although most major professional societies issued guidelines or recommendations about the number of embryos to be transferred, the effect in the field is limited (Templeton, 2000Go). This finding is a strong argument against the feasibility of professional self-regulation. It should not come as a surprise that, when the recommendations are not spontaneously accepted, rules will be imposed externally (Cohen, 1998Go).


    A new standard
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
The introduction of the `birth per embryo transferred' as a new standard to evaluate the efficiency of the programmes and to rank fertility centres can promote a structural change in practice (Hazekamp et al., 2000Go). For too long, standards have been used which encouraged the generation of multiple pregnancies. The ideal treatment procedure should combine quantity (rate of pregnancy) with quality (healthy singletons). The challenge is to design a procedure that `walks the thin line between low pregnancy rates and multiple pregnancies' (Yaron et al., 1997Go). The pressure on the physicians to increase success rates is very strong. The low success rate is still a major argument against most applications of assisted reproduction. In some countries, there have been threats by the Ministry of Health to withdraw the licence of centres that do not reach a minimum level of success. This focus on the success rate is meant to protect the patient against clinics where the rates are very poor (Faber, 1997Go). This is obviously a laudable goal. However, when the intake of patients is determined by the success rate, statistics become publicity. The new standard might prevent possible (or probable) misinterpretations by patients of the tables of success rates published in countries such as the UK and The Netherlands. Centres with a large number of multiple pregnancies should end up at the end of the list.

The number of new patients attracted by the success rate has a strong influence on the realization of two major self-serving goals of infertility practitioners, namely professional status and profit. Several authors have pointed at the commercialization of IVF as the cause of the current practice of multiple embryo transfer (Faber, 1997Go; Templeton, 2000Go). This development raises urgent moral questions when it interferes with the treatment decisions of the practitioners. It is generally agreed that the first obligation of a doctor is to act for the best interests of the patient. The self-interest of the physician is not intrinsically opposed to the best interests of the patient. On the contrary, most medical acts are part of a practice that is mutually beneficial for both patient and physician. However, when physicians start taking risks in order to realize a self-serving goal, they fail in their first obligation as physicians. The physician is guilty of recklessness when he helps to create a multiple pregnancy in order to boost his overall success rate. In that case, he uses the patient as a means to his end.


    Decisional authority
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
The question raised is: who should decide how many embryos should be replaced? Should the balancing of success rate and risk of multiple pregnancies be done by the candidate parents or by the physician? The answer depends on three elements: the (mainly psychological) capacity and competence of the would-be parents, the welfare of the children and the model of the patient–physician relationship. Part of the reason why physicians would refuse to replace more than two or three embryos is paternalistic in the strict sense of the term: they want to protect the best interests of the couple or mother. However, the estimation of the capacity to raise twins or triplets looks like a highly personal decision best left to the prospective parents (Gleicher et al., 1995Go). If this were the sole relevant dimension of the decision, the reproach of medical paternalism might apply. Still, the autonomous and informed character of this decision can be questioned. Two conditions need to be fulfilled to enable a person to think rationally about a decision: (i) the absence of overwhelming emotions; and (ii) the availability of objective and sufficient information. There are strong indications that infertile people's decision-making is mainly guided by their desire to become parents. Their wish for a child makes them underestimate the difficulties of raising a child (let alone more than one child), among which there may be a child with special needs. The fact that women with children are significantly more negative about having multiples than childless women corroborates this hypothesis (Leiblum et al., 1990Go). It would be an interesting test for this claim to question parents about the burden and benefit of having a twin or triplet before the treatment and again a year after the birth of multiples. If the perception by the parents differs significantly before and after, this proves that the original preference and consent for multiple embryo transfer was ill-informed. The bias caused by the strength of the desire for a child can be reinforced when financial considerations limit the number of IVF cycles for which the patients can pay (Murdoch, 1998Go). Undoubtedly, some fertility specialists, like a large number of their patients, are convinced that two surely is better than none. After the accumulation and publication of the consequences of multiple pregnancies, they can no longer believe that even twins are a happy or acceptable outcome of an IVF cycle. Ignorance, which is one of the most general categories of excuse, is no longer available to the fertility practitioner. As professionals with the capacity to handle the statistics and to judge the impact of certain outcomes on people's lives, the physician should do more than simply inform the patient of the risk and benefits.

The second aspect of the decision focuses on the well-being of the future offspring. The physician should decide whether the children that are likely to come into existence will have a life that fulfils the `reasonable welfare' criterion (Pennings, 1999Go). The provision of medical assistance in procreation is acceptable when the child born as a result of the treatment will have a reasonably happy life. The probability that this is not the case for twins and triplets is unacceptably high, considering that this outcome can in many cases be avoided. The seriousness of the negative consequences combined with the moral principle of doing no harm, strongly favours erring on the safe side at the expense of the pregnancy rate. Since the focus is on the mental and physical health of the future children, the refusal by the physician to accede to patients' requests is not paternalistic. He is not trying to impose his conception of the patients' good on his patients; he is shouldering his responsibility towards the future child. The physician's position on this issue should fit the attitude he takes when considering offering infertility treatment to patients with an increased risk of conceiving a child with a serious genetic disease.

The third moral element in the decision is the perception of the physician–patient relationship. The content of this relationship determines the goals of the interaction, the physician's obligations, the role of patient values and the concept of patient autonomy (Emanuel and Emanuel, 1992Go). Two positions can be adopted: firstly, the physician is convinced that he is morally obliged as a physician to respect the patient's wishes. However, in that case he wrongfully neglects the other obligations and responsibilities that are part of a physician's role. Secondly, the physician sees himself as the `technical agent' whose job it is to try to realize the patients' wishes. He is only responsible for the skill, proficiency, and technique with which his interventions are performed but not for the value choices that are made (Blustein, 1993Go). The weakness of this position can be shown by looking more closely at the answers of the infertile women. One author (Leiblum et al., 1990Go) found that 83% of the IVF women would elect to have quadruplets rather than forfeit the opportunity to have biological children. Does this make it all right for physicians to collaborate in the creation of quadruplets? They ask for it, they will get it. If not, how can this restriction of the parents' wishes be justified by the defenders of the patients' autonomy? Although leaving the decision to the patient looks like a means of diminishing the physician's personal responsibility, there is no way in which the practitioner can deny his responsibility. Even when he sees himself as only helping others, he remains accountable for his collaboration and is an accessory to a negative outcome.


    Moral responsibility
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
In the ethical analysis of an act, a distinction is made between intended and foreseeable consequences. The Practice Committee of the American Society for Reproductive Medicine opens its report on the number of embryos transferred with the statement `Multiple gestation is an unintended result of the assisted reproductive technologies (ART)' (American Society for Reproductive Medicine, 1998Go). No physician directly intends to create a twin, a triplet or a higher order pregnancy. His immediate reason for replacing three or more embryos is to increase the chances of establishing a pregnancy. However, the physician knows that the multiple pregnancy rate increases with the number of embryos that are transferred. When he knows or is able to foresee that in a number of cases his actions will lead to adverse results, he is morally responsible and accountable to those who suffer the harm (Dworkin, 1987Go). In itself, the causal contribution of the physician to the multiple pregnancy is not sufficient to make him morally responsible. Whether the causal responsibility leads to the attribution of moral responsibility depends on the justification the physician can offer for replacing more than one embryo. He can justify this decision by pointing out that he has taken all the necessary steps and all the precautions to avoid a multiple pregnancy while simultaneously preserving an acceptable chance of success. Unless the physician has indications that a specific woman has a very low chance of pregnancy when only one embryo is replaced (due to age, embryo quality or previous treatment history), it is highly unlikely that the difference in success rate between replacing one embryo and two or three embryos outweighs the difference in expected well-being between a singleton and a child born to a higher order pregnancy.


    Inconsistent application of the `best interest of the child' standard
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
The well-being of the future child is generally considered as a crucial element in the decision about whether or not to proceed with the treatment. However, the criterion is very inconsistently applied. Post-menopausal pregnancies have been rejected by large numbers of fertility centres (as demonstrated by the adoption of an age limit) because of the adverse consequences for the child (being raised by an elderly mother) and because of the obstetric and health risks for the mother (Braverman et al., 1993Go). Lesbian women are denied access to infertility treatment because their children will not be brought up in optimal circumstances. A debate is going on about the acceptability of intracytoplasmic sperm injection (ICSI) when there is a possible transfer of infertility to the next generation. This is no plea in favour of dropping this criterion. However, as Templeton (2000) pointed out concerning the debate about the acceptability of ICSI, `any putative risks in this respect pale into insignificance when compared with the morbidity in children born of higher order multiple pregnancy'. The comparison of the negative impact of multiple pregnancies with the possible effect of other controversial applications of assisted reproduction shows that the `welfare of the child' is used as an instrument to forbid or deny what one does not like (Pennings, 1999Go). Apparently practitioners tolerate far more harm to mother and child when this is part of an attempt to increase the success rate than when they deliberate on morally controversial requests by patients.


    Multifetal reduction
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
The key word is prevention (Nisand and Shenfield, 1997Go). If multiple pregnancies can be prevented by reducing the number of embryos replaced, then this should be preferred to multifetal reduction. Looked at from this position, the woman's predicament is iatrogenic. A number of people will find it very difficult to accept that doctors first create a problem that they then solve by another medical intervention. However, this is not unusual. A physician often prescribes drugs to help a patient and afterwards prescribes other drugs to manage the side-effects of the first drug. Multifetal reduction can be seen in the same light. The scheme implies that the replacement of multiple embryos was necessary to increase the chances of pregnancy of a woman significantly and that the side-effect (the multiple pregnancy) can be treated by selective abortion.

Multifetal reduction is ethically acceptable if, and only if, the physician has done what he should have done to prevent the occurrence of the multiple pregnancy. The benefits of reducing a multiple pregnancy largely exceed the disadvantages of carrying the pregnancy to term or risking miscarriage (Bergh, 1998Go). However, this presupposes that there already is a multiple pregnancy. When the medical risks associated with the procedure and the high ethical and psychological stress for the couple are weighed against the disadvantages of not being pregnant in this cycle, then the balance tips in favour of trying again later. This balance also points at a frequent misrepresentation of the options; the choice is not between having no children at all and having multiples but between having no child now and having a singleton later. A large majority of the women who do become pregnant with multiples will become pregnant with a singleton if the number of cycles is increased.

One of the remaining problems is the moral significance of an abortion for the patient and for the doctor. When the patients feel very uncomfortable or unable to cope with the decision of a possible reduction, they should at the very least have the right to choose fewer embryos for transfer than the number proposed by the physician (Kanhai et al., 1994Go). The ultimate decision about the abortion stays with the woman. Her decision not to have a multifetal reduction even when she carries a higher order pregnancy should be respected (although this is clearly an irrational and self-defeating decision). However, I think that the active contribution of the physician in the creation of the risk pregnancy and his ability to foresee this consequence overrides his right of conscience not to perform an abortion for ethical reasons. If he made a woman pregnant with quadruplets by replacing four or more embryos, he should accept the obligation to terminate the lives of some of these embryos if the woman requests this (Howe, 1994Go).


    Fifth strategy
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
Finally, I would like to suggest a fifth way, in addition to the strategies mentioned by Hazekamp et al. (2000), to decrease the number of multiple pregnancies. This can be done by extending the professional responsibility of the fertility specialist to all steps of procreation including pregnancy, birth and neonatal care. The specialist who transfers the embryos should also be responsible for the performance of the multifetal reduction and for counselling the patients about this decision (Faber, 1997Go). It might also help to call in infertility specialists at regular intervals to the labour ward or to the neonatal unit where they can observe at first hand the final results of their own decisions (Templeton, 2000Go). These proposals are rather unrealistic (medical specialization will not be reversed) but they emphasize the importance of feedback from, and interaction with, the other medical and psychological carers involved in the treatment of mother and child. In fact, this feedback should remind the fertility specialists that their job does not end with a successful conception.


    Conclusions
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
The effort made by fertility practitioners to reduce the multiple pregnancy rate expresses the degree to which they have the patients' best interests in mind. It would be an oversimplification to put the blame for the high rate of multiple pregnancies solely on the shoulders of the physicians. Competition between centres, cost-effectiveness and the increasing commercialization of assisted reproduction all contribute to the attempt to maximize the success rate. Add to these the profit motive and the gain in professional status and you end up with a strong cocktail of forces which all deflect the practice from the real goal, i.e. the birth of a healthy child. It is up to the practitioners in the field to show that they can resist the temptations.


    Notes
 
This debate was previously published on Webtrack, September 19, 2000


    References
 Top
 Abstract
 Introduction
 A new standard
 Decisional authority
 Moral responsibility
 Inconsistent application of the...
 Multifetal reduction
 Fifth strategy
 Conclusions
 References
 
American Society for Reproductive Medicine (1998) Practice Committee Report: Guidelines on number of embryos transferred. American Society for Reproductive Medicine, USA.

Bergh, C. (1998) Pregnancy outcome and psychological follow-up of families after embryo reduction. Assist. Reprod., 9, 104–111.

Blustein, J. (1993) Doing what the patient orders: maintaining integrity in the doctor–patient relationship. Bioethics, 7, 289–314.[ISI]

Braverman, A.M. and Ovum Task Force of the Psychological Special Interests Group of The American Fertility Society (1993) Survey results on the current practice of ovum donation. Fertil. Steril., 59, 1216–1220.[ISI][Medline]

Cohen, J. (1998) How to avoid multiple pregnancies in assisted reproduction. Hum. Reprod., 13 (Suppl. 3), 197–218.[Medline]

Dworkin, G. (1987) Intention, foreseeability, and responsibility. In Schoeman, F. (ed.), Responsibility, Character, and the Emotions: New Essays in Moral Psychology. Cambridge University Press, Cambridge, UK, pp. 338–354.

Emanuel, E.J. and Emanuel, L.L. (1992) Four models of the physician-patient relationship. J. Am. Med. Assoc., 267, 2221–2226.[ISI][Medline]

Faber, K. (1997) IVF in the US: multiple gestation, economic competition, and the necessity of excess. Hum. Reprod., 12, 1614–1616.[ISI][Medline]

Gleicher, N., Campbell, D.P., Chan, C.L. et al. (1995) The desire for multiple births in couples with infertility problems contradicts present practice patterns. Hum. Reprod., 10, 1079–1084.[Abstract]

Hazekamp, J., Bergh, C., Wennerholm, U.-B. et al. (2000) Avoiding multiple pregnancies: considerations of new strategies. Hum. Reprod., 15, 1217–1219.[Abstract/Free Full Text]

Howe, E.G. (1994) Clinical dilemmas when patients want assistance in dying. J. Clin. Ethics, 5, 3–9.[ISI][Medline]

Kanhai, H.H.H., de Haan, M. and van Zanten, L.A. (1994) Follow-up of pregnancies, infants, and families after multifetal pregnancy reduction. Fertil. Steril., 62, 955–959.[ISI][Medline]

Leiblum, S.R., Kemmann, E. and Taska, L. (1990) Attitudes toward multiple births and pregnancy concerns in infertile and non-infertile women. J. Psychosom. Obstet. Gynaecol., 11, 197–210.[ISI]

Murdoch, A.P. (1998) How many embryos should be transferred? Hum. Reprod., 13, 2666–2670.[Free Full Text]

Nisand, I. and Shenfield, F. (1997) Multiple pregnancies and embryo reduction: ethical and legal issues. In Shenfield, F. and Sureau, C. (eds.), Ethical Dilemmas in Assisted Reproduction. Parthenon Press, New York, London, pp. 67–76.

Olivennes, F. (2000) Double trouble: yes a twin pregnancy is an adverse outcome. Hum. Reprod., 15, 1663–1665.[Free Full Text]

Pennings, G. (1999) Measuring the welfare of the child: in search of the appropriate evaluation principle. Hum. Reprod., 14, 1146–1150.[Free Full Text]

Templeton, A. (2000) Avoiding multiple pregnancies in ART: replace as many embryos as you like – one at a time. Hum. Reprod., 15, 1662.[Free Full Text]

Yaron, Y., Amit, A., Kogosowski, A. et al. (1997) The optimal number of embryos to be transferred in shared oocyte donation: walking the thin line between low pregnancy rates and multiple pregnancies. Hum. Reprod., 12, 62, 699–702.[Abstract]