Spare Parts or Saviour Sibling? The Birth of an Ethical Dilemma

Artist Credit: Editorial Board

I        Introduction

Zain Hashimi suffered from beta thalassaemia major, an inherited blood disorder requiring extensive medicating and regular blood transfusions to treat (Laurance, 2018). By the time Zain was two and a half years old, his symptoms had become severe, his treatments were painfully invasive, and his life expectancy was uncertain (Deech 2007, 66). Whilst he could have been cured with a stem cell transplant, none of Zain’s siblings were a match and no other donor had been found (Deech 2007, 66). By age 6, Zain’s parents were desperate to alleviate his suffering. The couple underwent six unsuccessful attempts at in vitro fertilisation (IVF), each ending in miscarriage or a failure to produce a tissue match (Dyer 2005, 330). They then sought permission from the United Kingdom’s Human Fertilisation and Embryology Authority to create a saviour sibling for Zain (Zúñiga-Fajuri 2018, 229).

II       What are Saviour Siblings?

“Saviour siblings” are children created to act as donors for an existing sick child (henceforth referred to as an “affected child”) in the family (Zúñiga-Fajuri 2018, 229). Saviour siblings may be conceived naturally; however, since there is only a one in four chance of naturally conceiving a match (Robertson 2002, 35), sometimes assisted reproductive technology is used to conceive a donor child that is both free from the condition and a tissue match to the affected child (Whelan 2021, 81). Three technological techniques are used: first, IVF combines eggs and sperm to create fertilised embryos outside of the body; second, pre-implantation genetic diagnosis (PGD) is used to screen for specific genetic conditions or chromosomal abnormalities by taking a biopsy of a few cells and analysing them; and finally, human leukocyte antigen (HLA) tissue typing is used to select embryos compatible with the affected child (James 2019). They are then implanted in the mother’s uterus and, upon birth, the donor child’s stem cells are collected and transplanted to the affected child (James 2019).

Hematopoietic stem cells (HSCs) are found in umbilical cord blood, peripheral blood, and bone marrow (Devolder 2005, 582). HSC transplants are currently either the best or only treatment for certain inherited conditions such as Huntington’s disease, thalassemia, cystic fibrosis, muscular dystrophy, and non-inherited conditions including leukaemia, Hodgkin’s lymphoma, and Fanconi anaemia (Robertson 2002, 35). Medical interventions for these conditions include blood transfusions, bone marrow transplants, and organ transplants (Zúñiga-Fajuri 2018, 230). Approximately 75% of bone marrow donations for children are from their siblings (Packman 2011, 701).¹ The transplant’s success depends on the compatibility of the donor’s and recipient’s tissue; thus siblings have a much higher success rate than alternative donors (Devolder 2005, 582).

Creating saviour siblings has clear benefits. Using tissue typing PGD allows parents to save the life of their seriously ill child and welcome another child into the family. However, while many experts agree that the creation of saviour siblings is an ethically permissible practice (Zúñiga-Fajuri 2018, 229), the subject has attracted considerable debate due to the complex medical, legal, and ethical issues that arise (Robertson 2002, 35).

III      New Zealand Law

A         Legal Framework

In New Zealand, assisted reproductive technology (ART) is regulated by the Human Assisted Reproductive Technology Act 2004 (the HART Act). The Act divides ART into three categories: prohibited procedures (Human Assisted Reproductive Technology Act 2004); established procedures, which can be performed without approval; ² and assisted reproductive procedures, which cannot (Human Assisted Reproductive Technology Act 2004). The Human Assisted Reproductive Technology Order 2005, made under s 6 of the HART Act, lists the established procedures, including routine PGD (Human Assisted Reproductive Technology Act 2004). However, when carried out in conjunction with HLA tissue typing, it is an assisted reproductive procedure requiring ethical approval (Advisory Committee on Assisted Reproductive Technology 2008). The Act also sets up two statutory bodies. The Advisory Committee on Assisted Reproductive Technology (ACART) creates, recommends, and issues guidelines on ART procedures and research, and advises the Minister of Health. The Ethics Committee on Assisted Reproductive Technology (ECART) assesses individual applications for approval against ACART’s guidelines on a case-by-case basis (Nguyen 2015, 1790).

B         Previous Legal Position

The original 2005 guidelines restricted tissue typing PGD to situations where the affected child has a genetic disorder, and the donor child was at a 25–50 percent risk of that same disorder (Advisory Committee on Assisted Reproductive Technology 2008). The rationale for this restriction was that the donor child would also benefit from the procedure because they would be born free from the genetic disorder (Advisory Committee on Assisted Reproductive Technology 2008). Several other requirements also had to be met: there could be no other possibilities for treatment of the affected child, the planned treatment would utilise only cord blood, and the donor child would be the affected child’s sibling (Henaghan 2007). 

C         Recommendations

In 2007, a Human Genome Research Project report commissioned by the New Zealand Law Commission recommended that tissue typing PGD regulations should be expanded to allow treatment for any serious or life-threatening condition caused by spontaneous mutation, for which stem cell transplantation can treat or cure (Dunedin 2007, 3). They also recommended that using tissue, blood, and bone marrow should be permitted and that the donor child is assigned a qualified third-party advocate (New Zealand Law Commission 2007).

Following the release of this report, ACART first consulted with the public in 2008 and issued draft guidelines to the Minister of Health (Duff 2014). After considering public submissions, ethical literature, and national and international policy developments, ACART recommended that tissue-matching to existing individuals with non-genetic disorders should be permitted and that the procedure should be allowed to benefit close relatives other than siblings (Advisory Committee on Assisted Reproductive Technology 2014). ACART also considered that the clinical team and parents must agree that the condition of the affected person is “sufficiently severe” to justify the procedure (Advisory Committee on Assisted Reproductive Technology 2014). Objections based on the potential psychological damage to the donor child, the adverse impact on family dynamics, and the commodification of children were rejected Advisory Committee on Assisted Reproductive Technology 2014).

D         Current Legal Position

In 2014, ACART published new guidelines adopting several of the above recommendations, which remain the law in New Zealand today. Notably, ACART extended the procedure’s permitted use to benefit children affected by non-genetic disorders Advisory Committee on Assisted Reproductive Technology, 2014), and added bone marrow transplantation, in addition to cord blood, as a treatment option (Advisory Committee on Assisted Reproductive Technology 2014). However, the guidelines still require that the resulting donor child will be a sibling of the affected child Advisory Committee on Assisted Reproductive Technology 2014). The guidelines also did not adopt the recommendation that the donor child be assigned an advocate.

IV      The Legal Debate

A         Therapeutic Objective

When examining both the previous and current legal positions on whether limiting PGD’s use to inheritable genetic conditions is justified, a key difference emerges. The previous position was that PGD must have a therapeutic objective; it must benefit the donor child in some way (Zúñiga-Fajuri 2018, 230). The problem with tissue typing PGD is that it subjects the embryo to the potential risks of undergoing the procedure solely for the benefit of another, ³ while producing no therapeutic benefit to that embryo (Sheldon 2005, 73). Therefore, tissue typing PGD to ensure compatibility with an affected child could previously only be undertaken when the embryo was at risk of developing the same genetic condition. 

The main difficulty with this justification is that no embryos subjected to PGD receive any therapeutic benefit. PGD screens for abnormalities; it does not prevent, cure, or treat a disorder that the embryo may have (Sheldon 2005, 525). Rather, the embryo is selected because it is already healthy. The only arguable benefit an embryo gains is its selection over other embryos for implantation (Deech 2007, 73). Therefore, if there is no benefit to the embryo in either case, restrictions for non-inheritable conditions cannot be justified on that basis. This reasoning supports ACART’s expansion to allow tissue typing PGD for the benefit of children with serious non-inheritable conditions.

B         Informed Consent and “Best Interests”

“Informed consent” means consent given voluntarily, with enough information about what is being consented to and its associated risks to make an informed decision (Health and Disability Commissioner 2021). However, informed consent creates difficulties concerning saviour siblings. Children, depending on their developmental stage, have a limited capacity to understand the implications of proposed procedures (Lufti 2015, 24). As the affected child’s condition progresses, parents cannot wait for the donor child to reach an age where they can competently consent to donation. Generally, parents can consent on their child’s behalf to undergo necessary medical procedures (World Health Organization 2010). However, parents in this context may have a conflict of interest, being responsible for the welfare of both the donor and the affected child (World Health Organization 2021). Objections from the donor child will likely supersede parents’ wishes, ⁴ particularly if they are 16 or older (Human Tissue Act 2008). Nonetheless, donor children may find it nearly impossible to refuse, especially if they were created for that purpose (Assche 2016, 266).

Due to the complexities of obtaining informed consent, the best interests of the child should be the primary consideration (United Nations 1989). The moral problem of non-therapeutic medical interventions is that they fail to meet the generally accepted minimum conditions required to justify medical interventions for people unable to consent (Zúñiga-Fajuri 2018, 229). This is because the donor child receives no medical benefit. However, in the United Kingdom, the child’s “best interests” are not confined to medical benefits; they can include potential psychological effects and the interests of their entire family, including their sibling (Zúñiga-Fajuri 2018, 231). The child certainly benefits from growing up in an intact family, if the affected sibling survives (Pennings 2002, 536). They have an interest in the well-being of the recipient (Pennings 2002, 536); therefore, it could be said that donation is in their best interests. 

The artificiality of this argument can be striking for some. The parents clearly do not intend to benefit the donor child when they select an embryo based on tissue typing at the outset – their concern is entirely for the affected child (Pennings 2002, 536). Perhaps it is more accurate to concede that a donation is not in the donor child’s best interests, their moderate sacrifice is justified by the significant benefits to the affected child.

V       The Ethical Debate

A         Welfare of Donor Child

Ethical concerns about saviour siblings naturally focus on the physical and psychological welfare of the donor child. The UN Convention on Children’s Rights requires states to protect children against all forms of exploitation prejudicial to any aspect of the child’s welfare (United Nations 1989). Moreover, several statutory enactments aim to protect the welfare of New Zealand children (Children’s Commissioner Act 2003; Care of Children Act 2004; Education and Training Act 2020; New Zealand Bill of Rights Act 1990).

1          Physical Welfare

While the risks of embryo biopsies were previously uncertain, there is no conclusive evidence that pre-implantation genetic testing has any short- or long-term adverse effects on health outcomes for resulting children (Alteri 2023, 291). Further, because stem cells are usually collected from umbilical cord blood, the donor child is usually not subjected to direct physical intervention (Pennings 2002, 537). However, if cord blood is inadequate to treat the affected child, bone marrow may be used. Bone marrow collection requires putting the donor under general anaesthesia, and some donors experience nausea, fatigue, headaches, bleeding, difficulty walking, sleep problems, bruising, pain, or discomfort (New Zealand Bone Marrow Donor Registry, 2024). Nonetheless, bone marrow donation is generally considered unproblematic due to the tissue’s ability to regenerate and the low risk of complications despite requiring physical intrusion (Pennings 2002, 537).

2          Psychological Welfare

Critics have also raised concerns about the donor child’s emotional and psychological welfare (Ram 2006, 280). Sources of potential harm include emotionally distant family relationships, awareness of the circumstances surrounding their conception, and feelings of inadequacy or fault if the treatment fails (Sheldon 2005, 536) However, such concerns remain speculative. In most cases, the donor is too young to comprehend the situation. Once older, they will value their sibling relationship and thus likely agree with their parents’ decision (Pennings 2002, 538). Knowing they were conceived to help their sibling could give them a greater sense of self-esteem and purpose than those conceived accidentally or without conscious thought (Pennings 2002, 537). Equally, it could cause them to feel subservient to their sibling, diminishing their self-worth. The impact on the donor child depends entirely on their specific experience.

Several studies can provide insight into the psychological impacts on adolescent sibling donors, although not saviour siblings specifically. In 1999, William Packman conducted a study on the psychosocial impact of paediatric bone marrow transplantation on siblings. Participants expressed feeling they had “no choice” but to donate, and experienced loneliness following donation (Packman 1999, 701-706). A 2003 study supported the finding that sibling donors perceived a lack of choice, although only one-third felt this was due to external pressure (MacLeod 2003, 226). Nearly all donors whose transplantations were successful believed their participation positively impacted most aspects of their lives; however, unsuccessful donors expressed feeling responsible for the death of their sibling, supporting the idea that treatment outcomes influence the donor’s psychological impacts (MacLeod 2003, 227-228). It is difficult to determine whether a child’s purposeful conception for donation would exacerbate these feelings. However, if a child already in the family feels they have no say in the decision to donate, a child conceived solely for that purpose would almost certainly feel this way (Whelan 2021, 253).

3          A Possible Justification

One possible justification for allowing some harm to the donor child’s welfare is that they are inherently benefitted by the saviour sibling process due to the simple fact that, without it, they would not exist (Zúñiga-Fajuri 2018, 230). This has been coined the “non-identity problem,” which posits that life is always better than non-life and that, given the option, everyone would choose to be born (Whelan 2021, 253). Thus, any potential physical or psychological harm to the donor child’s welfare must be so great that one can conclude it would have been in the child’s best interests that they were never born (Whelan 2021, 524).

The difficulty with the non-identity problem is that it would justify the use of assisted reproductive technologies that create harm to children conceived as a result, provided they do not have a “life not worth living” (Taylor-Sands 2015, 54). This argument ignores the fact that a life of substantial suffering is still harmful, and incorrectly assumes that a person has an interest in existing prior to conception (Taylor-Sands 2015, 53).

B         Instrumentalisation

The instrumentalisation of children is another concern raised in the saviour sibling context. Lord Winston, a doctor who developed the technology required for PGD, expressed that the process risks turning babies into “commodities” (Collins 2025). Bulletin of Medical Ethics editor Richard Nicholson described saviour siblings as “slave siblings” and “a source of spare parts for an existing child (Devolder 2005, 584). The most common objection is that it is wrong to bring children into existence “conditionally” (Boyle 2001). This objection is founded on the Kantian imperative that one should always treat people as an end rather than a means (Kant 1998,1964).

Kant’s dictum demands that a child be wanted for their own sake, and not simply for the sake of others (Ram 2006, 279). Thus, conceiving a child to harvest their stem cells is unethical because it treats the child as a means to the end of saving the affected sibling’s life. However, parents seeking to create saviour siblings are unlikely to have such simplistic motives. Kant’s dictum was to never use people solely as a means. Thus, there is nothing objectionable about treating a child as a means, provided they are also viewed and treated as a person in their own right (Sheldon 2005 534). Families who have applied to use tissue typing PGD have stated a pre-existing desire to have more children (Ram 2006, 279). Moreover, parents have children for all kinds of instrumental reasons: for example, to benefit the couple’s marriage, give an existing child a companion, or continue the family name (Trifiolis 2014, 432). What matters most is how they treat their child, not their motives for having them (Devolder 2005, 584). Provided parents sincerely love their child, problems with that child benefitting others are minimal. Indeed, it is reasonable to believe that parents willing to go to such extraordinary lengths for their sick child would care deeply for each of their children in their own right (Ram 2006, 279).

Saviour sibling opposers also fear that once the donor child is born, it will be “at lifelong risk of exploitation,” and subjected to repeated testing and harvesting procedures (Ram 2006, 279). However, the law does not create loopholes for the parents of saviour siblings; they enjoy the same legal protection from exploitation as any other child. Therefore, just as no naturally conceived child could function as a “spare parts factory,” neither could a child born for the purpose of donation (Ram 2006, 280).

C         Designer Babies

The final major ethical concern is that tissue typing PGD will set society on a slippery slope towards the selection of embryos based on desirable non-medical characteristics; in other words, it will encourage the creation of “designer babies” (James 2006, 10). This is because the purpose of PGD has shifted from disease prevention towards the positive selection of embryos based on desired characteristics (i.e., tissue type) (Smith 2010, 16). Some argue that the slippery slope has already begun; since the embryo itself does not benefit from tissue typing, this ought to be regarded as a social purpose, rather than a medical one (Pennings 2002, 535).

There are two types of arguments here. The first is that if we allow something acceptable or only slightly bad to occur, it will cause something else to happen that is very bad or clearly wrong (Sheldon 2005, 534). Applied in this context, the idea is that allowing the creation of saviour siblings (slightly bad) will lead to the creation of designer babies (very bad). The second argument relates to consistency: allowing saviour siblings and designer babies are not morally different, therefore, if we ban one, we should ban the other (Sheldon 2005, 534).

Three objections to “slippery slope” arguments can be made. Firstly, one can reject the premise that allowing people to choose embryos for their characteristics is wrong (Sheldon 2005, 534). Indeed, in some cases, choosing embryos for their characteristics may be medically and ethically desirable (Damewood 2001, 3144). Secondly, permitting saviour siblings need not cause society to become permissive about designer babies (Sheldon 2005, 534). Finally, and relatedly, saviour siblings and designer babies are morally different, therefore one can oppose the latter and not the former without any inconsistency (Sheldon 2005, 534). While the reasons for choosing designer characteristics – such as eye colour, hair colour, sex, and so on – are generally trivial, saviour siblings are chosen to save the life of another child (Sheldon 2005, 534). New Zealand’s current policy merely permits a shift in focus from the health of the embryo to the health of another. Therefore, tissue typing is an intermediate procedure; while not strictly part of the medical model, it still serves a medical purpose (de Wert 2005, 3262). Both routine PGD and tissue typing PGD are aimed at the same end: to procure healthy children.

VI      Conclusion

Saviour siblings will forever attract medical, legal, and ethical debate. A regulatory framework that enables technological innovation whilst balancing legitimate moral concerns is notoriously difficult to achieve (James 2006, 2). ACART’s expansion of tissue typing PGD’s uses appears to be based on well-founded justifications. However, there is still scope for disagreement. Among the main arguments opposed to saviour sibling creation, concerns regarding the welfare of the donor child hold the most merit. Therefore, the donor child’s welfare should remain the paramount consideration for lawmakers and regulatory bodies.

¹William Packman, “Psychological impact of pediatric BMT on siblings,” Bone Marrow Transplantation 24, no. 7 (1999): 701. See also Katherine Lissienko, “Blood & Bone Marrow Transplant” KidsHealth NZ, September 13, 2011. 

²Human Assisted Reproductive Technology (HART) Order, 2005 (SR 2005/181), sec. 4(b); HART Act, sec. 3(e). 

³N. R. Ram, “Britain’s new preimplantation tissue typing policy: An ethical defence,” Journal of Medical Ethics 32, no. 5 (2006): 279; Guido de Wert, “Preimplantation genetic diagnosis: The ethics of intermediate cases,” Human Reproduction 20, no. 12 (2005): 3263. 

⁴United Nations Convention on the Rights of the Child (entered into force 1989), art. 12(1); World Health Organization, 5.