Uterus Transplant in Women With Absolute Uterine-Factor Infertility

Key Points

Question

Is uterus transplant safe and feasible, resulting in subsequent live births?

Findings

In this series of 20 participants, uterus allograft was successful in 70% of recipients and was associated with at least 1 live birth for all recipients with a successful allograft. Adverse events were common, with medical and surgical risks affecting recipients as well as donors.

Meaning

Uterus transplant was feasible and was associated with at least 1 subsequent live birth following a successful allograft.

Abstract

Importance

Uterus transplant in women with absolute uterine-factor infertility offers the possibility of carrying their own pregnancy.

Objective

To determine whether uterus transplant is feasible and safe and results in births of healthy infants.

Design, Setting, and Participants

A case series including 20 participants with uterine-factor infertility and at least 1 functioning ovary who underwent uterus transplant in a large US tertiary care center between September 14, 2016, and August 23, 2019.

Intervention

The uterus transplant (from 18 living donors and 2 deceased donors) was surgically placed in an orthotopic position with vascular anastomoses to the external iliac vessels. Participants received immunosuppression until the transplanted uterus was removed following 1 or 2 live births or after graft failure.

Main Outcomes and Measures

Uterus graft survival and subsequent live births.

Results

Of 20 participants (median age, 30 years [range, 20-36]; 2 Asian, 1 Black, and 16 White), 14 (70%) had a successful uterus allograft; all 14 recipients gave birth to at least 1 live-born infant. Eleven of 20 recipients had at least 1 complication. Maternal and/or obstetrical complications occurred in 50% of the successful pregnancies, with the most common being gestational hypertension (2 [14%]), cervical insufficiency (2 [14%]), and preterm labor (2 [14%]). Among the 16 live-born infants, there were no congenital malformations. Four of 18 living donors had grade 3 complications.

Conclusions and Relevance

Uterus transplant was technically feasible and was associated with a high live birth rate following successful graft survival. Adverse events were common, with medical and surgical risks affecting recipients as well as donors. Congenital abnormalities and developmental delays have not occurred to date in the live-born children.

Trial Registration

ClinicalTrials.gov Identifier: NCT02656550

This study assesses the long-term outcomes of uterus transplant to determine whether uterus transplant is feasible, safe, and results in births of healthy infants.

Introduction

Assisted reproductive technologies have provided millions of women with fertility options. However, absolute uterine infertility, affecting 1 in 500 women, remains a barrier in reproductive medicine. Absolute uterine infertility can be acquired or congenital and due to either a dysfunctional or an absent uterus. Affected women who want to have children traditionally choose between adoption or surrogacy where legally available. The development of uterus transplant offers an option for women with absolute uterine infertility to experience pregnancy and childbirth. Since the first successful uterus transplant in 2011,¹ more than 100 uterus transplant procedures have been performed worldwide, using grafts from living or deceased donors. Although no registry exists that precisely reports the number of live births resulting from these procedures, it is estimated that there are more than 70 live births worldwide.^2,3 As of May 1, 2024, there have been 48 uterus transplants and 33 live births in the US.⁴

The Dallas Uterus Transplant Study (DUETS) was designed to assess the long-term outcomes of uterus transplant following an institutional review board (IRB)–approved protocol. The study was considered closed once all enrolled uterus transplant recipients had either a graft hysterectomy or at least 1 live birth. This study reports the outcome data of the completed study; an interim analysis was previously reported.^4,5

Methods

Study Setting and Population

DUETS was approved by the Baylor University Medical Center IRB (No. 015-158) and registered (NCT02656550). The study began enrollment in 2016 using either living or deceased donors (Table 1 and Figure 1). Approval of an amended protocol expanded the number of potential participants from 10 to 20 in 2018 (Supplement 1). All eligible enrolled recipients and living donors provided verbal and written consent for research, using the IRB-approved consent form. Information regarding race and ethnicity was self-reported and collected to help the care team communicate effectively with patients, as well as understand a patient’s culture, which may affect their health. Study transplants were performed between September 2016 and August 2019. Changes to surgical and medical aspects of the protocol over time were inherent in this study, based on clinical judgment and evolution of treatment options. The study followed the Reporting of Studies Conducted Using Observational Routinely Collected Health Data (RECORD) statement (an extension of the Strengthening the Reporting of Observational Studies in Epidemiology [STROBE] reporting guideline), provided by the EQUATOR Network.⁶

Table 1. Demographic Characteristics at Baseline for Recipients of Uterus Transplant and Uterus Donors Included in the Dallas Uterus Transplant Study and Reproductive Outcome Data for the 14 Recipients Who Achieved Live Birth.

Characteristic	No. (%)
	Recipients	Donors
Female sex	20 (100)	20 (100)
Age, median (range), y	30 (20-36)	37 (30-56)
Body mass indexa	25 (19-34)	25 (19-28)
Ethnicityb
Hispanic or Latino	1 (5)	1 (5)
Non-Hispanic or Latino	19 (95)	19 (95)
Raceb
American Indian or Alaska Native	0	0
Asian	2 (10)	0
Black or African American	1 (5)	0
Native Hawaiian or Other Pacific Islander	0	0
White	16 (80)	20 (100)
Other	1 (5)	0
Parity, median (range)	0 (0-0)	2 (1-7)
Previous cesarean delivery	0	5 (25)
Indication for uterus transplant
Mayer-Rokitansky-Küster-Hauser syndrome	18 (90)	NA
Previous hysterectomy	2 (10)	NA
Donor type
Living nondirected	NA	17 (85)
Living directed	NA	1 (5)
Deceased	NA	2 (10)
Transplant/donation surgery
Laparotomy	20 (100)	15 (75)
Robotically assisted	NA	5 (25)
Operative time, median (range), min
Laparotomy	310 (241-463)	394 (313-454)
Robotically assisted	NA	648 (565-730)
Hospital stay, median (range), d
Laparotomy	6 (5-16)	6 (4-8)
Robotically assisted	NA	4 (3-6)
Surgery at graft hysterectomy (19 grafts removed)
Cesarean hysterectomy	6	NA
Laparotomy	10c	NA
Robotically assisted	3	NA
No. of embryo transfers per recipient to first live birth, median (range)	1 (1-9)	NA
Clinical pregnancy rate per embryo transfer	0.52	NA
Embryo transfer resulting in miscarriage	6 (13)	NA
Live birth rate per embryo transfer	0.35	NA
Time to first live birth, median (range), d	434 (334-1638)	NA
Time to second live birth, median (range), d	1032 (894-1170)	NA
Total graft time (from transplant to hysterectomy), median (range)
Without live birth, d	6 (0-14)	NA
With live birth, mo	22 (12-53)	NA

Abbreviation: NA, not applicable.

^a Calculated as weight in kilograms divided by height in meters squared.

^b Self-identified race and ethnicity from fixed categories. For the “other” category, patients did not have to specify their group identification.

^c Six uteri removed prior to live birth.

Figure 1. Potential Uterus Transplant Recipients and Donors Considered, Evaluated, and Enrolled in the Dallas Uterus Transplant Study.

Exclusion causes for evaluated potential recipients and living donors are shown in the respective categories. HSV indicates herpes simplex virus; IVF, in vitro fertilization; and STI, sexually transmitted infection.

Interested potential recipients and living donors independently contacted our institution. There was no advertising or recruiting for the clinical trial aside from information on the hospital website and listing on ClinicalTrials.gov (NCT02656550).⁷ Evaluation of potential recipients and donors are described in eAppendixes 1 and 2 in Supplement 2 and was modeled on existing evaluation processes for kidney and liver transplants. When the trial was initiated, there were no specific guidelines or regulations related to uterus transplant. All aspects of the evaluation process, perioperative management, follow-up, and data recording in the trial were directly derived from the existing framework used in the institution’s solid organ transplant program. During the course of the trial, the United Network for Organ Sharing (https://unos.org/) established rules for uterus transplant in accordance with the existing rules for other solid organ and vascular composite allograft transplants. A multidisciplinary committee, including transplant physicians, obstetricians and gynecologists, psychologists, donor advocates, and nurse transplant coordinators, approved recipients and donors. Eligibility criteria for recipients are described in detail in Supplement 1 and included being individuals of childbearing age, 20 to 40 years, with absolute uterine-factor infertility and at least 1 functioning ovary. Furthermore, the recipients had to meet the medical and psychological protocol criteria and be willing to undergo in vitro fertilization (IVF). Vaginal length was observed carefully, and all recipients were instructed to self-dilate at home with the help of a dilator or intercourse to reach or maintain a vaginal length of more than 5 cm. Recipients were required to have obtained a minimum of 2 euploid good-quality blastocysts prior to the uterus transplant (this minimum was increased to 4 for recipients 11-20).

Eligibility criteria for donors are presented in detail in Supplement 1 and included being between 25 and 65 years of age with at least 1 prior term live birth with no relevant medical or psychological comorbidities.

Procurement of Donor Organ

The donor operative procedures are described in detail in eAppendix 3 in Supplement 2 (sections 3.1 and 3.2). The previously described techniques for living donor hysterectomy included open laparotomy (first 13 living donors) and a minimally invasive robotic-assisted approach, with transvaginal allograft extraction for the subsequent 5 living donors.^8,9 Techniques similar to the open procedure were used for allograft recovery from both deceased donor allografts.¹⁰

Uterine Transplant

eAppendix 3 in Supplement 2 (section 3.3) describes the technical details of the uterus transplant, including use of an infraumbilical midline laparotomy.^5,11 Vascular anastomoses are performed to the external iliac vessels bilaterally with the uterine veins to create 2 to 4 venous anastomoses. Following graft reperfusion, the graft’s vaginal rim is anastomosed to an orifice created on the recipient’s vaginal vault.

Graft Surveillance

Duplex ultrasound is performed on postoperative days 1 and 5. To detect early rejection of the transplanted uterus, biopsies of the cervix were performed using a Tischler biopsy punch (MedGyn) at day 5, week 2, and then monthly until the start of pregnancy. During pregnancy, cervical biopsies were performed twice (12-14 and 24-26 weeks of gestation). Biopsies were assessed and graded for acute cellular rejection (none, borderline, mild, moderate, severe).¹²

Immunosuppression Protocol and Rejection Surveillance

eAppendix 4 in Supplement 2 lists the details of the previously described uterus transplant immunosuppression protocol.¹³ The immunosuppressive protocol for the uterus transplant recipients in this trial was based on the existing protocol used in recipients of kidney transplants. All recipients had normal kidney function prior to the initiation of calcineurin inhibitor immunosuppression. Antithymocyte globulin (4.5 mg/kg in 3 divided doses) and methylprednisolone (1000 mg) were used for induction followed by tacrolimus combined with an antimetabolite for maintenance. Initially (cases 1-4), a 1-month steroid taper protocol was used. For recipients 4 to 8, the antimetabolite mycophenolate mofetil was used for the first 6 months. To avoid delay in embryo transfer, after recipient 9, azathioprine replaced mycophenolate mofetil in the maintenance immunosuppression regimen. The tacrolimus dosing aimed for a trough level of 8 to 10 ng/mL within the first 3 months, decreasing to 3 to 5 ng/mL by the first year post transplant. Trough levels were not altered during the pregnancy.¹² Moderate and severe rejections were treated with a 3-day bolus of intravenous methylprednisolone (day 1, 1000 mg; days 2-3, 500 mg). Successful treatment was verified with a follow-up cervical biopsy 1 to 2 weeks post treatment. Steroid-resistant rejection was treated with antithymocyte globulin.

In Vitro Fertilization, Embryo Transfer, and Pregnancy

Prior to uterus transplant, all recipients underwent IVF. eAppendixes 5 and 6 in Supplement 2 contain detailed descriptions of the previously described IVF (eAppendix 5) and embryo transfer (eAppendix 6) procedures.^14,15 All embryos were created with the recipients’ own oocytes. For the first 8 recipients, single embryo transfer was attempted after 6 months; for subsequent recipients, single embryo transfer was attempted after 3 months. Conception was detected 9 days after embryo transfer by serum trace levels of human chorionic gonadotropin.

Obstetrical care was provided by specialists in maternal-fetal medicine. Prenatal visits occurred every 2 to 3 weeks from gestational week 20, and included fetal growth, biophysical profiles per the treating physician’s discretion, transvaginal sonographic assessment of the cervix, and umbilical arterial Doppler. The cervical length was determined by measuring a straight line between the external os and internal os. Delivery by cesarean was planned at 37 weeks or more of gestation based on expected completed neurological, lung, and digestive system development. The delivery date relied on the known embryo transfer date and early ultrasound dating.

Follow-Up

Uterine graft hysterectomy followed the first or second delivery. We previously published our decision-making process regarding the timing of graft removal.¹⁶ Potential second pregnancies were planned according to the recipient’s desire and lack of clinical complications, including complications of immunosuppression and rejection, maternal complications, or obstetrical complications.¹⁶ Per-protocol uterine graft hysterectomy was to be performed at the time of cesarean delivery, during the postpartum period (<8 weeks), or after the postpartum period (eAppendix 3 in Supplement 2, section 3.4). In cases of postpartum graft removal, the hysterectomy was either performed with an open or minimally invasive surgical technique.

Follow-up for all children occurred via protocolized pediatrician visits. Recipients were consented to allow for data collection for all children through their pediatrician visits.

Primary and Secondary End Points

The primary outcome for the study was procedure efficacy, defined as graft survival with at least 1 live birth following uterus transplant. Secondary outcomes were procedure safety, including patient survival and complications occurring in all participants (recipients, living donors, children).

Analysis

Data were recorded prospectively in the Simmons Transplant All-Organ Research System database. Quantitative and categorical data are presented in medians and ranges.

Results

Recipient Demographics

Twenty recipients, with a median age of 30 years (range, 20-36), were enrolled from 701 self-reported women assessed for eligibility (Figure 1). Eighteen had congenital absence of the uterus (Mayer-Rokitansky-Küster-Hauser syndrome) without surgical reconstruction of the vagina and 2 had a previous hysterectomy for benign disease. Living donors were selected from a pool of 424 donors (Figure 1).⁷ The 18 living donors had a median age of 37 years (range, 30-56) and a median parity of 2 (range, 1-7). Both deceased donors, aged 30 and 43 years, had a parity of 2. Table 1 displays the baseline demographic data at baseline for uterus transplant recipients and uterus donors.

Procurement of Donor Organ and Associated Complications

The median operative time for donor hysterectomy was 394 minutes (range, 313-454) for laparotomy and 648 minutes (range, 565-730) for the robot-assisted minimally invasive approach. Table 2 summarizes the 4 Clavien-Dindo¹⁷ grade 3 complications. Two donors experienced complications following laparotomy (vaginal cuff dehiscence, repaired through a transvaginal approach, and a fecal impaction requiring fecal disimpaction under anesthesia). Two donors experienced complications following robotic-assisted hysterectomy. One donor experienced a ureteral obstruction due to a clot in the ureter that immediately resolved following ureteroscopy and extraction. The other donor had a thermal injury to the ureters bilaterally resulting in a urinary leak and required stenting and unilateral subsequent robot-assisted reimplantation of the ureter. None of these 4 patients had any clinical sequelae at a minimum follow-up of 4 years. The median follow-up for the living donors after donation was 5 years 3 months (range, 4 years 5 months to 7 years 4 months).

Table 2. Adverse Events for the 18 Living Uterus Donors.

Complication	No. (%)
	Laparotomy (n = 13)	Robotically assisted surgery (n = 5)
Any complication	8 (62)	3 (60)
Major complication (Clavien-Dindoa grade ≥3)	2 (15)	2 (40)
Intraoperative complication (Clavien-Dindoa grade ≥3)	1 (8)b	0
Postoperative complication (Clavien-Dindoa grade ≥3)	2 (16)	2 (40)

^a The Clavien-Dindo system¹⁷ is a standardized system for grading adverse events resulting from surgical procedures and divides complications into 5 categories based on the therapy needed to treat the complication.

^b One donor in the series had short-term demand gluteal ischemia when climbing stairs.

Uterus Transplant and Procedural Complications

The median operative time for the recipients was 310 minutes (range, 241-463 minutes), with a median hospital length of stay of 6 days (range, 5-16 days). Table 3 lists the complications. Six recipients experienced graft loss, a Clavien-Dindo grade 3 complication. One of these recipients had an external iliac dissection requiring an interposition graft repair. That repair ultimately stenosed, necessitating a femoral-femoral arterial bypass, and 5 years after the bypass the patient had no sequelae. Another recipient with graft loss experienced hemorrhagic shock after an arterial bleed within hours after the transplant, prompting the removal of the graft. None of the 14 recipients who had a successful uterus allograft transplant had a severe complication.

Table 3. Adverse Events for the 20 Recipients of Uterus Transplant.

Event	No. (%)
Graft failure	6 (30)
Infectiona	9 (45)
Vaginal stricture	10 (71)
Rejection (n = 14)
Nonpregnant, treated with steroid cycling or thymoglobulin	9 (64)
Pregnant, treated with steroid cycling	2 (14)
Maternal complications (n = 14)
Gestational hypertension	2 (14)
Gestational diabetes	1 (7)
Preeclampsia	1 (7)
Obstetrical complications (n = 14)
Insufficient cervix	2 (14)
Preterm labor	2 (14)
Placenta previa	1 (7)

^a Including urinary tract infection, cytomegalovirus, yeast infection, incision infection, herpes outbreak, pelvic abscess, and sinus infection.

Graft Survival and Rejection

Six uterus transplant grafts were unsuccessful (1/2 deceased donor and 5/18 living donor).⁵ All graft failures occurred within 2 weeks of the transplant. Five of the failures were attributed to either problems with the vascular anastomosis or graft/donor selection, either as thrombosis of the arterial inflow or the venous outflow with evidence of parenchymal necrosis, or as existing atherosclerotic changes in the microvasculature of the uterus. One was attributed to hemorrhagic shock after an arterial bleed within hours after the transplant.

In the 14 patients without graft failure, 9 had cervical biopsy evidence of an acute cellular rejection; 3 had more than 1 episode. The median time to the first rejection episode was 3.6 months (Figure 2). Rejection was detected in 1 recipient during pregnancy; the episode resolved after steroid treatment. Most episodes (91.7%) were treated with either an increase in the calcineurin inhibitor immunosuppression dose or with steroid treatment. Rejection episodes that were not treated were determined to be borderline or mild in character and were followed with repeat biopsy within 2 weeks to confirm resolution or persistence of the borderline inflammation, in which case no action was taken.

Figure 2. Timeline of Events After Uterus Transplant.

The bars are medians with box ends representing first and third quartiles with whiskers extending to upper and lower boundaries and dots representing more extreme data.

Immunosuppression Issues

The protocol was designed to minimize recipients’ time receiving immunosuppression through early embryo transfer. The median total immunosuppression duration for the 14 successful uterus recipients was 21 months (range, 12 months to 4 years 2 months). The white blood cell count at the initiation of antimetabolites was a median of 6.6 × 10⁹/L for the 20 recipients. This increased to a median of 10.8 × 10⁹/L at the time of graft hysterectomy and to a median of 5.8 × 10⁹/L 3 months after graft hysterectomy. The degree of leukopenia in our recipients was never severe enough to require the administration of colony-stimulating factor. The maximum serum creatinine values in nonpregnant and pregnant recipients were 1.0 mg/dL (median) with respective ranges of 0.75 to 1.89 mg/dL and 0.66 to 1.47 mg/dL. At 6 months post graft hysterectomy, the median creatinine level was 0.86 mg/dL (range, 0.49-0.96). One recipient had calcineurin inhibitor–induced kidney toxicity.

Conception

The details of IVF are described in eAppendix 5 in Supplement 2. The median time interval from uterus transplant to the first embryo transfer was 4.1 months (range, 2.1-7.7 months). Seven recipients (50%) experienced a live birth after their first embryo transfer, 2 recipients after 2 embryo transfers, and 5 recipients after multiple embryo transfers. The overall live birth rate per embryo transfer was 43%.

Pregnancy and Pregnancy Complications

Umbilical artery Doppler flow was normal throughout the pregnancy for all patients. Sixteen of the 24 clinical pregnancies resulted in live births. Early pregnancy losses occurred in 4 recipients; 2 recipients had 2 losses. Three second trimester pregnancy losses occurred at gestational weeks 17, 17, and 19 due to cervical insufficiency (n = 2) or chorioamnionitis (n = 1). Table 3 lists the maternal and/or obstetrical complications experienced in 50% (8/16) of the pregnancies. Gestational hypertension was the most common maternal complication (14.3%), and preterm labor and cervical insufficiency were the most common obstetrical complication (14.3% each). Three recipients experienced vaginal bleeding during their pregnancy (cases 4, 9, and 18). Two recipients had abnormal placental location and received blood transfusions. One recipient (case 4) had a subchorionic hematoma diagnosed in gestational week 13. Five live births occurred earlier than planned at a median gestational age of 33 weeks 6 days (range, 30 weeks 6 days to 36 weeks 6 days). Two patients had preterm labor (35 weeks 5 days and 30 weeks 6 days), 1 had preeclampsia (36 weeks 6 days), 1 had vaginal bleeding spanning 2 trimesters and preterm contractions (32 weeks 4 days), and 1 had calcineurin inhibitor–induced kidney toxicity (33 weeks 1 day). Two recipients had cervical insufficiency and underwent cerclage placement. Supplement 2 contains additional details on recipient pregnancy (eAppendix 7) and delivery (eAppendix 8).

Live Births

All pregnancies concluded with cesarean delivery per protocol. Delivery was planned at a target gestational age of 37 weeks or more. Fourteen recipients had a least 1 live birth and 2 had a second live birth. Birth occurred at a median of 14.2 months (range, 11-53.9 months) following uterus transplant for the first delivery and 34 months (range, 29.3-38.5 months) for the second delivery (Figure 2). Six recipients had their uterus removed at the time of their first or second delivery.

Children

eAppendix 9 in Supplement 2 describes the health of the children. The median gestational age at live birth was 36 weeks 1 day (range, 30 weeks 6 days to 38 weeks 1 day). The sexes of the children were equally distributed, with 7 males and 9 females. Detailed outcomes of the neonates and follow-up to 2 years of age have been published elsewhere.^18,19 There were no congenital malformations in any of the children. The median follow-up for the children was 3 years 8 months after birth (range, 8 months to 6 years 2 months). One child (child 5) did not meet his communicative milestones at the 12-month examination. This child was referred and followed by an audiologist and diagnosed with autism at 24 months. Of interest, his sister born after him did not show any signs of developmental delays. At the 18-month examination, 2 additional children showed transient cognitive deviations (children 1 and 10). Both children showed significant improvement at subsequent follow-ups.

Graft Hysterectomy

The median time from uterus transplant to graft hysterectomy was 22 months (range, 12-53) (Figure 2). Of the 13 hysterectomies performed, 7 followed live births as the advance choice of the recipient and 6 were due to concerns of pregnancy- or immunosuppression-related complications. In 6 recipients, the uterus was removed at the time of cesarean delivery of either their first delivery (n = 5) or their second delivery (n = 1). In the remaining 7 recipients, the graft hysterectomy was done post partum, either via open midline incision (n = 3) or a robotically assisted approach (n = 4). The recipients’ median follow-up was 4 years after graft hysterectomy (range, 8 months to 7 years and 4 months). Only 1 recipient still had the transplanted uterus in situ for a second pregnancy.

Discussion

This case series of 20 women in a single institution demonstrated acceptable graft survival following a multidisciplinary approach to uterus transplant, conception, and subsequent live birth. During the study period, the technical success of graft survival improved with time and experience. All 14 participants with successful uterus transplant had a subsequent live birth. National US data for infertility treatment in women younger than 35 years using their own ova describe a live birth rate of 68.9% following any number of cycles.²⁰ The live birth success rate in this study suggests that a successfully transplanted uterus is capable of functioning at least on par with a native, in situ uterus. A prior study of 9 patients reported a 67% live birth rate following uterus transplant.²¹ Two of these recipients had graft failure prior to embryo transfer and 1 had graft removal after multiple unsuccessful pregnancy attempts.

Safety concerns included risks for donors. Because of the lack of data specific to hysterectomy for living donation, the reference data discussed with the donors at the time of consent were derived from the experience with living donation for liver and kidneys, in terms of general risk such as infection, bleeding, and thrombosis events, as well as known complications of radical hysterectomy, in terms of injury to surrounding organs and structures, such as ureters, bladder, and rectum. Ureter injury during hysterectomy is well reported in the literature for both open laparotomy and minimally invasive approaches, with rates varying from 0.02% to 0.78%.^22,23,24 The reported incidence, though, is higher for abdominal radical hysterectomy, to which living donor graft hysterectomy should be compared, where as many as 1 in 10 women may have a ureteral injury.²⁵ The reported incidence in living donor hysterectomy varies from 0% to 20% for open hysterectomies^5,26,27 to 40% in the initial experience with robotic-assisted living donor hysterectomy.⁵

Eleven recipients experienced at least 1 complication. Risks for the uterus transplant recipient included surgical risks for the initial transplant surgery, as well as subsequent surgical procedures (1-2 cesarean deliveries for delivery and graft hysterectomy), graft failure, and exposure to immunosuppression with the risk of nephrotoxicity. The long-term impact of immunosuppression in these otherwise healthy women remains unknown. Preliminary data seemed to indicate that kidney function completely or mostly recovered at a follow-up of 12 months from immunotherapy cessation.²⁸ Inherent in the desire to conceive and bear a child are the risks of assisted reproduction, pregnancy, and delivery.

There was no fetal growth restriction in any study pregnancy.^18,29 This contrasts with registry data of mothers with other solid organ transplants reporting a 16% frequency of children small for gestational age at birth.³⁰ Initial data on children born after uterus transplant also indicate that no increased incidence of abnormal physical or mental development has been detected.^18,19,31

Limitations

The study had several limitations. First, the sample size was small. Second, the results reflect the experience of a single center. Third, generalizability is limited due to the currently prohibitive cost of uterus transplant.

Conclusions

Uterus transplant was technically feasible and was associated with a high live birth rate following successful graft survival. Adverse events were common, with medical and surgical risks affecting recipients as well as donors. Congenital abnormalities and development delays have not occurred to date in the live-born children.

Supplement 1.

Trial Protocol

Supplement 2.

eAppendix 1. Graft Acquisition

eAppendix 2. Donor Evaluation

eAppendix 3. Surgical Procedures

eAppendix 4. Immunosuppression

eAppendix 5. In Vitro Fertilization

eAppendix 6. Embryo Transfer

eAppendix 7. Pregnancy

eAppendix 8. Delivery

eAppendix 9. Children

eReferences

Supplement 3.

Data Sharing Statement