Living donor uterus transplant in the UK: A case report

Uterine transplantation (UTx) is the only intervention that restores reproductive anatomy and functionality in women with absolute uterine factor infertility (AUFI). UTx involves transplantation of the uterus, cervix, surrounding ligamentous tissues, associated blood vessels and a vaginal cuff to offer women with AUFI an alternative to adoption and surrogacy as a route to motherhood.1 UTx provides women with AUFI with the unique prospect to gestate and give birth to genetically related offspring, enabling them to acquire biological, social, and legal parenthood, thereby circumventing many of the challenges associated with adoption and surrogacy. Following more than 90 cases, and the achievement of 49 livebirths,2 there is no question regarding the feasibility of the procedure as it now transitions from research concept to clinical practice.3 Given that the goal is to achieve the livebirth of healthy offspring, it is important to note that there does not appear to be an increased risk of fetal growth restriction in cases so far,4 and age-appropriate neonatal and developmental outcomes have been observed for the first 2 years of life.5 However, it remains a procedure associated with significant risk. Detailed analysis of the first 45 cases revealed that more than a quarter of grafts needed to be removed prematurely because of complications, and in living donor cases, one in ten donors required further surgery as a consequence of complications.6 We describe herein, the first UTx case performed in the UK using a living donor. Both the donor and recipient have given written consent for the case to be reported. The recipient was a 34-year-old woman with Type I Mayer–Rokitansky–Küster–Hauser syndrome without any other medical or psychological history and had not undergone any previous surgery. She had received dilator therapy to acquire an appropriate calibre vagina, which was 8 cm in length. She was a non-smoker, drank minimal alcohol, did not take any medications, and had a body mass index of 21 kg/m2. She was married, and had a supportive family. Her blood group was B positive. She was cytomegalovirus (CMV) IgG negative. All other serum virologies were unremarkable. She was negative for human papillomavirus, chlamydia and gonorrhoea. She was comprehensively counselled regarding her other options to acquire motherhood. She specifically desired a biological relation to her offspring, and expressed the desire to gestate and give birth, instead of using surrogacy. She underwent in vitro fertilisation, including two cycles of controlled ovarian stimulation using an antagonist protocol. The oocytes were fertilised by intracytoplasmic sperm injection and cultured to day 5 blastocysts. These were subjected to pre-genetic testing for aneuploidy and then cryopreserved by vitrification. Eight day 5 blastocysts were found to be euploid, but two were discarded because of aneuploidy. The donor was the recipient's sister, who was 40 years old at the time of donation. She did not have any significant medical or psychological history. Her BMI was 22 kg/m2. She was multiparous, having had two previous normal vaginal deliveries, both of which were at term, with normal birthweights and following uncomplicated pregnancies. She had completed her family. Her blood group was B positive. She was CMV IgG positive and all other serum virology was unremarkable. She was negative for human papillomavirus, chlamydia and gonorrhoea. Ultrasound, computed tomography and magnetic resonance imaging demonstrated normal pelvic anatomy and vasculature and revealed a duplex right kidney with two proximal ureters, and a normal left kidney and ureter. High-resolution donor and recipient human leucocyte antigen (HLA) typing were performed by next-generation sequencing (AllType™; OneLambda). HLA-DRB1 expression was confirmed in both individuals by complement-dependent cytotoxicity monoclonal typing (LMT First HLA Class II; OneLambda). Pre- and post-transplant antibody profiling was performed using Luminex® single antigen bead testing (LABScreen Single Antigen HLA Class I and II; OneLambda). Pre-transplant autologous and allogeneic T-lymphocyte and B-lymphocyte flow cytometric and complement-dependent cytotoxicity crossmatches were performed using in-house methods. The HLA mismatch was 1-1-1 and the calculated reaction frequency was 78%. All crossmatching was negative despite low-level HLA class II DSA, including a self-reactive antibody against one product of the shared haplotype (a putative HLA-DR4, -7, -9 epitope) and the purely allogeneic HLA-DR52. Both donor and recipient were extensively counselled preoperatively and were reviewed by gynaecologists, transplant surgeons, obstetricians, reproductive and transplant psychologists, anaesthetists and pharmacists. Psychological evaluation and counselling were based on the experience from the Swedish and Dallas UTx programmes, using the suggested framework that was subsequently produced.7 They were both assessed by a Human Tissue Authority Independent Assessor to ensure that they had capacity, were fully aware of the associated risks of the retrieval and grafting procedures, and that both parties were entering into the surgery of their own free will. The case was reviewed by a panel at the Human Tissue Authority and permission was granted to proceed with the transplant. The programme was approved by the Technology Advisory Group at Oxford University Hospitals NHS Trust and gained Senior Management Team approval before commencement. NHS England were also consulted and authorised the procedure. The donor retrieval was undertaken via a Maylard incision. Ureteric stents were inserted bilaterally to facilitate the identification of the ureters. A modified radical hysterectomy, with substantial retrieval of the round and uterosacral ligaments, as well as an extensive area of bladder peritoneum, with long vascular pedicles, was undertaken. The arterial pedicles included the uterine arteries to the level of the anterior division of the internal iliac on the left, and the lateral uterine arteries on the right, which in this case were duplicated. It was not possible to retrieve a cuff of the anterior division of the internal iliac on this side, because the superior vesical artery was between the two uterine arteries and there were concerns of devascularisation of the bladder and other pelvic structures. The venous pedicles included the utero-ovarian veins bilaterally as well as short segments of uterine veins. A vaginal cuff of 10–15 mm was included. Both ovaries were preserved. Bilateral salpingectomy was undertaken. The blood loss was 900 mL and the retrieval operative time was 8 hours 12 minutes. Following removal of the graft, it was flushed with cold Custodiol (histidine–tryptophan–ketoglutarate; Köhler Chemie GmbH) solution. As it was not possible to retrieve the cuff of the anterior division of the internal iliac on the right, the two uterine arteries were pantalooned to form a common stem using 7/0 prolene, with a resultant internal diameter of 3 mm. The left-sided artery was spatulated, with an internal diameter of 2 mm. The recipient surgery was commenced 1 hour before the anticipated uterine explant. It was performed through an infraumbilical midline incision. The external iliac arteries and veins were skeletonised and prepared for anastomosis. The uterine remnant horns and fallopian tubes were subsequently excised, leaving the vagina closed. The graft was then brought into the pelvis in an orthotopic position and the vascular anastomoses were undertaken. All vascular anastomoses were undertaken with 7/0 polypropylene, using a continuous suture technique. The arteries were anastomosed end to side with the recipient's external iliac arteries. Two right-sided end-to-side venous anastomoses were performed, between the utero-ovarian vein of the graft to the recipient's external iliac vein, and a further utero-ovarian vein to the recipient's ovarian vein. The left-sided venous drainage was established between the utero-ovarian vein and the inferior epigastric vein after an unsuccessful attempt was made between the utero-ovarian and the external iliac, which was complicated by tearing of the vessel during the anastomosis. The vagina was subsequently opened, and the vaginal anastomosis was undertaken using interrupted sutures. The round and uterosacral ligaments were subsequently re-attached, and the bladder peritoneum from the graft was fixed above the recipient's bladder, to enhance the structural support of the graft. Following reperfusion, the uterus changed from a dusky pale colour to a pink and well-perfused appearance, as demonstrated in Figure 1. A 20-MHz Cook–Schwartz Doppler probe (CookMedical) was placed around the left uterine artery, which was exteriorised through the midline incision, to allow continuous monitoring of blood flow postoperatively. The total blood loss was 2000 mL, but 1000 mL were reinfused using cell-salvage homologous transfusion (Sorin Electa and XTRA Autotransfusion systems, LivaNova Plc). A further two units of packed red blood cells were transfused postoperatively. The implantation operative time was 9 hours 20 minutes. Cold ischaemic time was 3 hours 20 minutes and warm ischaemic time was 2 hours 1 minute. After induction, the recipient was given 5000 units of subcutaneous unfractionated heparin. Before organ perfusion, she was given 500 mg of intravenous methylprednisolone (Solu-Medrol; Pfizer) and 30 mg of alemtuzumab (Campath-1H; Berlex). Maintenance immunosuppression was commenced with tacrolimus (Adoport; Sandoz Limited) 0.05 mg/kg twice daily, with target trough levels between 8 and 12 ng/mL. Mycophenolate mofetil 750 mg twice daily was commenced – to be switched to azathioprine 3 months postoperatively. Immunosuppression doses were modified in response to trough levels and bone marrow response. She was given intravenous antimicrobial and antifungal prophylaxis with piperacillin/tazobactam 4.5 g three times daily and caspofungin 50 mg once daily for 5 days. Prophylaxis for CMV and pneumocystis pneumonia was commenced with valganciclovir 900 mg once daily and co-trimoxazole 480 mg once daily, both of which will be continued for 6 months. Inpatient thromboprophylaxis with 7500 units of unfractionated heparin twice daily was used and adjusted based upon thromboelastography results. This was switched to dalteparin 5000 units daily on discharge, which was continued for 6 weeks. Aspirin 75 mg once daily was commenced and will be continued while the graft remains in situ. The donor was discharged on postoperative day 5, and had an uneventful recovery. The recipient was discharged on day 10 postoperatively. Her prolonged admission was the result of excessive lymph output, initially from the drain, and then through the incision, which resolved with conservative management. Following discharge, she was closely followed up, twice weekly for the first postoperative month and then weekly thereafter. Each appointment included an assessment of the recipient's clinical signs, a clinical examination including a speculum, high vaginal swabs and urine for microscopy, culture and sensitivities, blood tests and cervical biopsies. Cervical biopsies were taken on day 5 postoperatively, then after 2 weeks and 1 month, and then monthly thereafter. All the biopsies were categorised as normal, with no evidence of rejection, as per the histopathological grading system used in the Dallas UtErus Transplant Study (DUETS),8 which has evolved from the initial grading system proposed from the first Swedish series.9 Transabdominal ultrasound scans were undertaken that demonstrated stable uterine size and echogenicity, evolving endometrial thickness throughout the menstrual cycle, and normal uterine arterial and intramyometrial Doppler waveforms. She had her first period 2 weeks postoperatively. At the time of reporting, 10 weeks postoperatively, she has had her third menstrual period postoperatively and remains well. Our team have undertaken preclinical research into UTx for more than 25 years, including performing the procedure in several animal species, including rabbits, pigs and sheep.10-12 Our animal model research ceased following the first successful live birth reported in Sweden in 2014.13 Our work on uterine vasculature, an essential part of UTx, has led to the development of multiple other surgical procedures, including the development of the abdominal radical trachelectomy,14 which has enabled fertility preservation of many thousands of women with early-stage cervical cancer, and the modified Strassman procedure,15 which has been used to preserve the reproductive potential of women with placental site trophoblastic tumour, and other pathologies. We have also undertaken a series of qualitative research studies to assess the perceptions and motivations of potential recipients and donors towards the concept.16, 17 As a result of the prolonged surgical retrievals in the initial operations using living donors, and the significant associated surgical and thromboembolic associated risks, our initial intention was to establish a deceased donor programme.18 However, subsequent evolution in the surgical technique, with modification to the venous drainage from the uterine veins to the utero-ovarian or ovarian veins,6 greatly reduced the operative time and surgical risk involved in living donor UTx. As a result, we developed a living donor programme, the first case of which is presented herein, which was performed with the support of the Baylor uterus transplant team in Dallas. Following various institutional delays in establishing both our living and deceased donor programmes, and after a 3-year postponement during the COVID-19 pandemic, both programmes are now live. The process of UTx is associated with significant risk, including multiple major surgeries, the use of immunosuppression while the graft is in situ, and the need to undergo in vitro fertilisation to achieve pregnancy. Our implantation was complicated by a prolonged implantation surgical time, and a much higher than expected blood loss, although the consequences of this were negated by the proactive use of cell-salvage homologous transfusion. The excessive blood loss resulted predominantly from bleeding from the vaginal cuff following reperfusion. Whereas in routine hysterectomy, energy devices would be used during the colpotomy to cauterise the vaginal edge, we proactively sought to preserve the tissue integrity to facilitate subsequent vaginal anastomosis healing in the recipient. However, in future cases, greater attention to achieving haemostasis in individual vessels in the vaginal cuff during the retrieval will be essential, to reduce subsequent donor blood loss. Our case has a number of novel aspects that add to the previously published literature. First, this is the first successful case where vascular reconstruction was undertaken. Necessitated by atypical vasculature, where there were two right-sided uterine arteries, which is a rare variant,19 they were pantalooned to form a common stem, which was anastomosed to the recipient's external iliac artery. There have been two UTx cases undertaken so far where vascular reconstruction was attempted. During the first UTx case, which was performed in Saudi Arabia, the uterine vessels were elongated using saphenous vein extensions.20 Another case in Dallas created a conduit with a branch of the internal iliac artery, which also resulted in thrombosis.21 In other solid organ transplants, ex vivo vessel reconstruction is associated with a greater risk of arterial thrombosis,22 so it should only be considered in UTx if there is no alternative. Another novelty from this case is the venous drainage used; this is the first case that has used the recipient's inferior epigastric vein for venous drainage. The typical venous drainage employed in UTx cases so far includes the uterine veins or the utero-ovarian veins, or a combination of both.6 Due to the complex uterine venous dissection required to acquire long lengths of the uterine vein, with the associated increased surgical time and risk, we purposely acquired shorter lengths of uterine veins, with an intention to focus on prioritising the drainage from the utero-ovarian vessels. However, as we experienced technical difficulties with the left-sided venous anastomosis, which resulted in the anastomosis being torn, we mobilised the left larger epigastric venae comitantae, before dividing the inferior epigastric vein and performing an end-to-end anastomosis with the donor utero-ovarian vein. Although prompted by adversity in this case, it demonstrates an alternative option for venous drainage when the typical venous anastomoses are not possible. Consideration of the CMV mismatch between donor and recipient is important. CMV infection is the most common viral infection experienced in transplant recipients,23 where it can have severe consequences. Moreover, infection during pregnancy is associated with adverse outcomes, including miscarriage and fetal growth restriction, and it can result in sensorineural hearing loss, visual impairment and cerebral palsy in offspring.24 Given that the ultimate goal of UTx is to achieve a live birth of a healthy infant, and considering the treatment of CMV is teratogenic,25 it is of paramount importance to minimise the risk of CMV infection during pregnancy. Recipients who are seronegative in receipt of a uterus from a seropositive donor, such as this case, have the highest risk of acquiring primary CMV infection post-transplantation.23 As such, we will complete 6 months of anti-viral prophylaxis before a 3-month period of CMV polymerase chain reaction surveillance, to ensure that she remains infection free, before performing embryo transfer. This is the first case to use alemtuzumab as an induction immunosuppression agent. The vast majority of cases so far have used the polyclonal antibody anti-thymocyte globulin (ATG).6 However, alemtuzumab is commonly used in kidney and pancreas transplantation and is standard induction therapy for most solid organ transplants in our unit. It is associated with less biopsy-proven rejection when compared with ATG,26 and has been demonstrated to reduce maintenance immunosuppression and steroid exposure post-transplantation.27 Although it is not licensed for use in solid organ transplantation, given the theoretical advantages over ATG, it should be considered in UTx. When clinical UTx programmes are established, it is essential to be in the context of appropriate expertise and regulation. As described herein, our team gained experience performing the procedure in a variety of animal models and has established other surgical procedures closely related to the UTx surgical technique. Moreover, we collaborated closely with the Dallas UTx team, who have provided insightful experience and expertise. In the future, whether within the remit of research or as an innovative surgical procedure, it is essential that outcomes are accurately monitored, published and reported in the International Society of Uterus Transplantation (ISUTx) international registry, to enable performance and safety monitoring, including appropriate follow up of donors, recipients and offspring. Moreover, membership with ISUTx is advised, and attendance at their annual meetings is recommended, to refine and collaborate on research, as well as to reflect and learn from experience elsewhere. In summary, we present herein the first case of UTx performed in the UK. Following three regular menstrual periods, with ultrasonographic evidence of cyclical endometrial proliferation, the uterus is functional, and there has been no evidence of rejection so far. BPJ conceived the article, helped organise and assist with the surgery, reviewed the patient postoperatively and wrote the manuscript. SV helped organise and assist with the surgery, reviewed the patient postoperatively, collected data and revised the final draft. SS assisted in performing the surgery and helped revise the final draft. AD provided pharmacological input and revised the final draft. TBM, MYT and JN performed the fertility treatment, provided expertise and revised the final draft. BK and CR undertook the histological analysis and revised the final draft. VS and NB provided imaging expertise and revised the final draft. AO assisted in performing the surgery and revised the final draft. MB provided tissue typing expertise and revised the final draft. PD, AK and RK provided anaeshetic input and support and revised the final draft. LJ provided guidance and expertise throughout the entire process and revised the final draft. CDG and VU assisted and performed the surgery and revised the final draft. PF provided transplantation expertise and revised the final draft. IQ and JRS led the surgical teams, performed the surgery, reviewed the patient postoperatively and revised the final draft. We would like to thank Ms Barbara Antoniak-Jakubowska, Mrs Jeanette Ayres, Mr Antonio Barbosa, Professor Jayanta Chatterjee, Dr Mian Chen, Ms Gabriela Chesaru, Miss Rebecca Cullen, Professor Sadaf Ghaem-Maghami, Professor Peter Friend, Ms Nicki Hayward-Priest, Dr Maria Jalmbrant, Dr Katie Jeffrey, Ms Mirka Kazmierczwk, Mrs Yassmin Khater, Ms Lian Lee, Miss Bryony Lennon, Ms Angela Losekann, Ms Samantha Madzikanda, Ms Josefina Marticio, Ms Joanna Nawrocka, Mr Roman Nebres, Ms Alicja Opalczewska, Mr Ahmad Sayasneh, Mr Sanjay Sinha, Ms Clare Snelgrove, Dr Giuliano Testa, Ms Anu Thomas, Mr Joseph Yazbek and all the staff on the Transplant ward and Transplant Immunology Laboratory team at the Churchill Hospital in Oxford. The study was funded by the registered charity Womb Transplant UK (1138559). JRS is Chairman of the registered charity Womb Transplant UK (1138559), which funded the case. All other authors: none declared. Not applicable. The data that support the findings of this study are available on request from the corresponding author.