Today, most causes of infertility are successfully treated. Yet there is still a subgroup of female infertility affecting around 4%, which so far is untreatable because of an absolute uterine factor. To acquire motherhood, these women are today referred to either adoption or surrogacy. Research in the field of uterus transplantation has been evaluated in different animal models for decades and has presently reached a human clinical application as a possible treatment for absolute uterine factor infertility. Organ transplantation is no longer reserved to those with a life-threatening disease and neither is organ transplantation together with concurrent immunosuppression prohibiting pregnancy. Uterus transplantation involves four parties – recipient, donor, partner of recipient and future child – and is a subject of ethical controversies.
Introduction
Becoming a parent is one of the most common and wanted expectations in adulthood and most adults include having children in their life plan/dream. For some, this dream does not come true so easily and the feeling of loss can have deep negative consequences on the affected couple. Infertility is a worldwide health issue with similar prevalence between countries . Many infertile couples overcome their situation, after successful treatment or through adoption or surrogacy, but a large amount of couples are left with no possible option to achieve parenthood. Women with absolute uterine factor infertility (AUFI) represent the last major group of untreatable female infertility. Transplantation of organs/tissues is today no longer limited to those with life-threatening disease, as also quality-of-life enhancing transplants, such as face, limbs and larynx, have become a medical reality. Uterus transplantation (UTx) could be a treatment for AUFI and it may well become the first type of quality-of-life-enhancing transplantation that is also life propagating. A clinical introduction of UTx raises ethical questions, in many ways different from those of other organ transplantations. Still classified as experimental, UTx, although tested in animal studies and attempted in eleven human cases, stands to prove a sufficiently developed procedure before further human cases are conducted.
Causes of infertility
Uterine factor infertility
The prevalence of current infertility in women of fertile age has been estimated to be about 9% . Within couples, female factors correspond to approximately 35% , whereof infertility due to a uterus factor makes up approximately 3–5% . Uterine factor-related infertility (UFI) has many causes, is either congenital or acquired and can be absolute or relative. AUFI affects around one in every 500 women of fertile age , corresponding to around 200,000 women in Europe and 85,000 in the US. Women with UFI may benefit from UTx; however, before UTx is considered in females with relative UFI, all other possible treatments, such as corrective surgery, should naturally have been ruled out.
Congenital uterine malformations ( Fig. 1 )
Congenital uterine malformations, not always associated with infertility, have a prevalence of around 7% . The majority of these malformations present themselves as septate within the uterus, a result of a failed absorbance of the partition between the two fused Müllerian ducts, or as a bicornuation of the uterus. Although these two conditions often can be corrected by surgery, other malformations such as unicornuation or didelphic uterus (accounting for 20% of the uterine malformations) cannot . Even more severe is the absence of a uterus as in the Mayer – Rokitansky – Küster – Hauser (MRKH) syndrome. This rare (1:4000) condition is characterized by Müllerian duct aplasia, resulting in a lack of the uterus and the upper two-thirds of the vagina but the presence of normal ovaries . These females are usually identified in their teens when presenting with primary amenorrhoea and generally undergo surgical creation of a neovagina.
Acquired uterine malformations ( Fig. 1 )
Myomas do not always affect the fertility, yet they represent the most common cause of acquired UFI. Around 10% of the population between 33 and 40 years have myomas , and the incidence increases with age. Corrective surgery is the primary option for a woman of fertile age but effectiveness on infertility is dependent on the position, number and size of the myomas. An estimated 1% of females with myoma in the age group 30–34 years will need a hysterectomy and 2.5% in the age group 35–39 .
Intrauterine adhesions can cause both infertility and recurrent miscarriage. Adhesions can occur secondary to intrauterine infections, surgical abortions or genital tuberculosis, with the latter cause being more common in third-world countries . Even in spite of adhesiolysis, around 50 % of women with adhesions remain infertile .
Cervical cancer is the second most common female form of cancer worldwide , with around 50% of the affected women being below 40 years , and with a non-negligible group of patients below 30 . Fertility-sparing surgery (trachelectomy) is restricted to tumours with an invasion <10 mm and a diameter not exceeding 20 mm . Half of the group with cervical cancer <40 years are suitable for this type of intervention . The remaining group of women is treated with radical hysterectomy with preservation of the ovaries. In the US, the annual prevalence of hysterectomy, due to benign or malign causes, is estimated to be 0.2/1000, in women <24 years, or in absolute numbers about 5,000 . Although efforts are made not to perform hysterectomies in fertile women with a child wish, it may still be indicated as an emergency peripartum intervention in the case of severe haemorrhage .
Causes of infertility
Uterine factor infertility
The prevalence of current infertility in women of fertile age has been estimated to be about 9% . Within couples, female factors correspond to approximately 35% , whereof infertility due to a uterus factor makes up approximately 3–5% . Uterine factor-related infertility (UFI) has many causes, is either congenital or acquired and can be absolute or relative. AUFI affects around one in every 500 women of fertile age , corresponding to around 200,000 women in Europe and 85,000 in the US. Women with UFI may benefit from UTx; however, before UTx is considered in females with relative UFI, all other possible treatments, such as corrective surgery, should naturally have been ruled out.
Congenital uterine malformations ( Fig. 1 )
Congenital uterine malformations, not always associated with infertility, have a prevalence of around 7% . The majority of these malformations present themselves as septate within the uterus, a result of a failed absorbance of the partition between the two fused Müllerian ducts, or as a bicornuation of the uterus. Although these two conditions often can be corrected by surgery, other malformations such as unicornuation or didelphic uterus (accounting for 20% of the uterine malformations) cannot . Even more severe is the absence of a uterus as in the Mayer – Rokitansky – Küster – Hauser (MRKH) syndrome. This rare (1:4000) condition is characterized by Müllerian duct aplasia, resulting in a lack of the uterus and the upper two-thirds of the vagina but the presence of normal ovaries . These females are usually identified in their teens when presenting with primary amenorrhoea and generally undergo surgical creation of a neovagina.
Acquired uterine malformations ( Fig. 1 )
Myomas do not always affect the fertility, yet they represent the most common cause of acquired UFI. Around 10% of the population between 33 and 40 years have myomas , and the incidence increases with age. Corrective surgery is the primary option for a woman of fertile age but effectiveness on infertility is dependent on the position, number and size of the myomas. An estimated 1% of females with myoma in the age group 30–34 years will need a hysterectomy and 2.5% in the age group 35–39 .
Intrauterine adhesions can cause both infertility and recurrent miscarriage. Adhesions can occur secondary to intrauterine infections, surgical abortions or genital tuberculosis, with the latter cause being more common in third-world countries . Even in spite of adhesiolysis, around 50 % of women with adhesions remain infertile .
Cervical cancer is the second most common female form of cancer worldwide , with around 50% of the affected women being below 40 years , and with a non-negligible group of patients below 30 . Fertility-sparing surgery (trachelectomy) is restricted to tumours with an invasion <10 mm and a diameter not exceeding 20 mm . Half of the group with cervical cancer <40 years are suitable for this type of intervention . The remaining group of women is treated with radical hysterectomy with preservation of the ovaries. In the US, the annual prevalence of hysterectomy, due to benign or malign causes, is estimated to be 0.2/1000, in women <24 years, or in absolute numbers about 5,000 . Although efforts are made not to perform hysterectomies in fertile women with a child wish, it may still be indicated as an emergency peripartum intervention in the case of severe haemorrhage .
Uterus transplantation
Research in organ transplantation
In the development of novel surgical techniques, different phases require diverse models. In organ transplantation, non-rejecting models such as autologous or syngeneic models enable the evaluation of surgical techniques without interference of immunosuppression (IS). In the autologous model, the organ/tissue is retrieved and retransplanted into the same individual, whereas in the syngeneic model individuals with an identical genome are used. The rejecting (allogeneic) model, where the organ/tissue is retrieved from one individual and transplanted to another, requires the use of IS and facilitates the possibility of studying the disposition of rejection of the organ/tissue.
History of UTx
Research in UTx has been conducted in several different animal models. Excluding the early experiments, when dogs were a common animal model , initial studies have usually been done in rodents . Rodents are commonly used in medical research and although far from a humanlike setting, the knowledge is good concerning immunology, reproductive physiology and genetics making rodents good as an experimental model at a low cost. Findings from these experiments could be transferred to small domestic species such as the rabbit . To achieve a more humanlike setting, subsequent UTx experiments were performed in larger animals like sheep and pig . Three species of non-human primates have been subjected to UTx research, cynomolgus macaque , rhesus macaque and baboon , and this is being considered as the last step before clinical introduction to make the transition to human UTx as safe as possible. No species of animals has such a resemblance to the human when it comes to anatomy and physiology of the reproductive organs as the non-human primates; however, the administration of IS is technically challenging .
In 1969, aiming to describe rejection patterns of allogeneic transplanted uteri, female ( n = 9) and male ( n = 5) dogs received a uterus after UTx . The internal iliac vessels of the graft were anastomosed to the internal iliac vessels of the recipient and azathioprine was given post-operatively. Of eight long-term survivors, five showed a viable uterus at termination after 45 days.
During the late 1970s to the 1990s, little attention was paid to UTx with the successful introduction of in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI), as infertility treatments, being possible causes. Nevertheless, in the early 2000s, the research field was reopened, evolving rapidly at the same time as being highly debated.
Rodents and small animals
In 2000s, a study describing successful UTx after syngeneic transplantations in mouse were published. One uterus horn, including vascular pedicels up to the aorta/vena cava, was anastomosed to the aorta/vena cava. The transplanted cervix was left unattached in the abdomen and the native uterus left in situ. In 90% of the surviving animals, long-term graft function was demonstrated, and concurrent embryo transfer (ET) resulted in pregnancy. In a follow-up study, the cervix was placed as a cervical-cutaneous stoma . Pregnancy rates were similar following when the native and transplanted uteri were compared. The birth weight, postnatal development and fertility of the pups were normal. The same syngeneic UTx model was used to evaluate the influence of ischaemia on the transplanted uterus . The graft was prior to transplantation exposed to cold ischaemia for 24 or 48 h, and the uteri preserved for 24 h showed normal morphology and blood flow after 2 weeks while grafts with longer ischaemia showed decreased blood flow and necrosis. In 5/6 animals transplanted with a uterus preserved for 24 h, ET resulted in pregnancy and normally developed offspring.
Initial experiments in rats used similar methods as in mice and the single uterus horn included a vascular pedicle up to aorta/vena cava or the common iliacs . The first rat UTx model was syngeneic and the uterus was anastomosed to the native aorta/vena cava, whilst the native uterus was kept in situ . Although an animal survival of 95% was observed, the study struggled with thrombosis that caused loss of one-third of the transplants. A rat model, allowing spontaneous mating, where the common iliac vessels were used for vascular anastomoses and the cervix was attached to the vagina was developed . Pregnancy rates of 50% occurred in both control and transplanted animals with no difference between the groups regarding the number of pups and postnatal development.
The first pregnancy after allogeneic UTx was reported in rats also showing similar rates of pregnancies between control and transplanted animals . In follow-up studies, the pups developed normally into adulthood (C. Díaz-García, personal communication).
The feasibility of UTx has been evaluated in rabbits where uteri including aorta and vena cava were recovered from deceased donors, and after 1 h transplanted . Surgical survival was good and a short-term follow-up presented viable uteri, but several post-operative complications like paraplegia, haemorrhage and pulmonary embolism occurred.
Large animals
Two autologous UTx studies in pig have been published . The pig model is, due to the size of the uterus in correlation to the size of the vessels, considered difficult and the time of vascular anastomosis, where the graft is warmed, long. The results accordingly were poor, showing thrombi in the anastomosis sites or ischaemia-reperfusion damage .
Allogeneic UTx was performed in miniature swine, and the graft included the aorta/vena cava . The uterus was placed retroperitoneally and the cervix exteriorized as a cutaneous stoma. The native uterus was kept in situ. IS was given, and after 1 year 50 % of the animals was alive and episodes of rejection had been successfully treated.
Autologous UTx in sheep with anastomosis of the uterine vessels, including part of the internal iliac vessels, to the external iliac vessels showed blood reperfusion in 5/7 animals and after three hours visible blood flow in the tissue and spontaneous uterus contractility . The technique was modified to include one uterus horn and the associated oviduct and ovary to allow a test of fertility . The graft was replanted after 8 h of surgery. Animal survival was 50%, and in 60% of the mated ewes pregnancy occurred . The lambs were comparable in size to the lambs from control ewes.
Another study in allogeneic UTx in sheep treated with IS showed a viable uterus and vascular patency in 60% of the animals after 6 months . In a follow-up study, 12 ewes were transplanted with allogeneic uteri . IS was increased compared to the previous experiment, and after 4 months five ewes were subjected to ET . Three pregnancies occurred resulting in one live birth by caesarean section. Allogeneic UTx was also performed in ewes with anastomosis of the aorta/cava unilateral to the external iliac vessels . IS was maintained by cyclosporine and mycophenolate mofetil (MMF) and rejections monitored by vaginoscopy, MRI and second-look laparotomy. All transplants showed thrombosis and necrosis after 10 weeks with poor fixation of the graft, rejection and insufficiently achieved vessel lengths speculated to be the causes.
Non-human primates
Prior to the first published human UTx case, an autologous baboon model for UTx-like experiments was used . Sixteen baboons were included and the first eight UTx were performed with vascular anastomosis of the uterine vessels end to end resulting in thrombosis in a majority of cases. It was not mentioned whether also uterine veins were reconnected or if the ovarian veins were preserved. The technique was altered and the sites of anastomoses instead being the uterine vessels to the internal iliac vessels which improved the vascular patency to 90% after 6–12 weeks. However, exact data are not provided.
In our initial study on autologous UTx in the baboon, the uterus, the ovaries and Fallopian tubes were transplanted . Bilateral uterine arteries with the anterior divisions of the internal iliac arteries and the ovarian veins were included. The ovaries permitted a non-invasive way of graft assessment, since hormonal cyclicity could be monitored as cyclic perineal changes. In addition, menstruation provided data on uterine functionality. During back-table preparation, the vascular pedicles were anastomosed side to side to create one artery and one vein, which, at transplantation, could be anastomosed unilaterally to the external iliac vessels. Despite the long and complicated surgery, 9/10 animals showed long-term survival. Five animals presented ovarian cyclicity, with menstruation seen in two animals.
In a follow-up study on autologous UTx in the baboon , we initially repeated the surgical procedure of our first baboon study . Among the six auto-transplanted animals, four survived the post-operative period. They all resumed ovarian cyclicity but not menstruation. Autopsies revealed normal ovaries but small fibrotic uteri. As the same artery supplied both the uterus and ovaries, the results indicate that the ovary has a better capacity to survive ischaemia or/and that the ovary’s capacity to induce neovascularization from adjacent tissues is greater. Several modifications were tested to improve the results . The perfusion of the graft during ischaemia was altered and the main trunks of the internal iliac arteries were included to achieve a larger arterial end. At back-table preparation, one iliac artery was coupled end to end to the contralateral posterior branch of the internal iliac. This construct was then anastomosed end to end to the internal iliac artery. On the venous side, there were extended dissections of the ovarian veins including patches of the vena cava and the left kidney vein, to acquire veins with walls of adequate thickness for easier anastomosis to the external iliac veins. The results improved considerably with animal survival of 100%, cyclic hormonal patterns in 80% and menstruation in 60% . Yet pregnancy following natural mating failed to occur. Post-mortem analysis revealed tubal blockage as a likely cause.
In 2011, a Japanese team performed autologous UTx in two cynomolgus macaques . During 6–8 h, the uterus and uterine vessels were recovered. This size of the cynomolgus macaque, approximately half the size of the baboon, is most likely the cause of the time-consuming recovery. Vascular anastomosis was done bilaterally of the uterine vessels end to side to the external iliac vessels. One animal died of assumed acute renal failure. The surviving animal resumed menstruation after 4–5 months. In a follow-up study, four cynomolgus macaques underwent autologous UTx and a unilateral anastomosis of one uterine artery, and one uterine vein was proposed to provide the graft with sufficient blood supply . The total surgery duration was 12.5–17.5 h and only one animal showed long-term uterine survival. Using the same autologous model as in the previous studies , the same group reported the first pregnancy ever after any type of UTx in a non-human primate species , Natural mating resulted in a pregnancy, developing normally until placental abruption occurred near term. A live offspring was delivered but it showed signs of foetal distress.
The first study of allogeneic UTx in a non-human primate included the recovery of a uterus with extensive vascular pedicles in the rhesus macaque . The anastomosis site of the recipient followed a preferred hierarchy of internal iliac vessels, external iliac vessels, common iliac vessels and aorta/vena cava. Monotherapy with cyclosporine was used as IS. Unfortunately, neither IS doses or blood levels nor survivals of the grafts or animals were reported.
Our team performed allogeneic UTx in a baboon model with transplantation from living donors . The duration of the donor surgery was 3 h, and the post-operative survival was 100%. Various IS protocols were tested and it was found that induction therapy with anti-thymocyte globulin followed by triple IS with tacrolimus, MMF and corticosteroids was compatible with long-term graft survival. The short-term survival rate of the recipients was 100%. Hormonal cyclicity reappeared in some animals but menstruation did not.
Grafts from deceased donors have been used in baboon UTx. Anastomoses were performed with the aorta/vena cava to the recipient’s aorta/vena cava (our unpublished results). Initially, the IS protocol was identical to that of our live donor UTx study , and after some months monotherapy with tacrolimus could be introduced. Episodes of graft rejection were successfully treated and 1-year graft survival proved.
Human UTx
Human trials
In 2000, the first human case of UTx was performed in Saudi Arabia . A 26-year-old woman, who previously had undergone a peripartum hysterectomy, received a uterus including oviducts from an unrelated 46-year-old living donor going through elective surgery because of bilateral benign ovarian cysts. The uterine vessels obtained were short and had to be elongated by segments of the saphenous veins. The extended pedicles were anastomosed bilaterally to the external iliac vessels. Triple IS was given and one episode of acute rejection was successfully resolved. The uterus was responsive to oestrogen and progesterone treatment with endometrial proliferation and bleeding. After 3 months, necrosis and thrombosed vessels were observed and the uterus was removed. The exact cause of these events are unclear but the authors suggest that prolapse of the uterus, as a result of inadequate structural support, with secondary thrombosis of the vessels led to uterus necrosis.
In 2007, an attempt to retrieve the uterus including complete internal iliac vessels bilaterally was made as part of a multi-organ donation . The sufficient length of the procured vessels was only achieved in 2/7 cases.
We performed a study aiming to evaluate the feasibility of living donor UTx . During radical hysterectomies to treat cervical cancer, parts of a uterus recovery were simulated, by adding dissection of the uterine veins. This extra dissection did not affect the morbidity. The procured uterine arteries measured 65–70 mm and the uterine veins 50–55 mm. It was concluded that these lengths of the vessels would enable a bilateral end-to-side anastomosis to the external iliac vessels in a potential living donor UTx.
In 2011, the second human UTx attempt was performed in Turkey . The recipient was a young female with MRKH and she was transplanted with a uterus from a multi-organ donor with the same age as herself. The graft included the uterus and bilateral uterine vessels including the common iliac vessels. The procedure took around 5.5 h and the common iliac vessels were anastomosed to the external iliac vessels bilaterally. The graft was fixed by the structure of the round ligaments and the uterovesical peritoneum to the inguinal ligaments and the bladder fundus. Immunosuppressive therapy was with induction with thymoglobulin for 10 days and maintained by prednisolone, MMF and tacrolimus. Menstruation was reported after 20 days and the graft has so far survived with regular menstrual pattern without reported rejection episodes. Two early miscarriages after the first and second ET have been reported .
The first controlled research-based clinical trial of a series of UTx took place in 2012/2013 . After extensive investigations, nine women were included, due to either congenital uterine absence ( n = 8) or hysterectomy ( n = 1). The patients were transplanted with grafts from living donors, the majority of the donors being recipients’ mothers ( n = 5). Induction therapy was with thymoglobulin and maintained by triple IS, with tapered doses, the intention being removal of MMF after 6–9 months, well before ET, and introduction of monotherapy with tacrolimus. Patients were monitored with clinical examinations, blood chemistry, ultrasound and cervical biopsies. The durations of donor and recipient surgery ranged from 10 to 13 h and 4 to 6 h, respectively. The transplanted uterus was removed in two cases: due to bilateral thrombotic uterine artery occlusions and due to persistent intrauterine infection. The remaining seven transplanted uteri remain viable 6 months post-operatively, all exhibiting cyclical menstruation. Mild rejection episodes occurred in some patients and were effectively reversed by corticosteroid boluses. In vitro fertilization was performed prior to transplantation and ET is planned to start at least 1 year post-operatively.
Uterus recipient
The uterus recipient has to be thoroughly investigated prior to the transplantation regarding both the physical and psychological ability to handle the transplantation itself and the after coming treatment and pregnancy ( Table 1 ).
