Technical Refinements in Brain-dead Donor Uterus Transplantation: Lessons From the 3 First Cases in France

Abstract

Background.

Uterus procurement from brain-dead donors during multiorgan procurement for uterus transplantation (UTx) presents a technical challenge because the uterus is not a vital organ and also because of the need to harvest uterine and ovarian veins for transplantation. The aim of this article is to describe the technical refinements in uterus procurement from brain-dead donors during multiorgan procurement and evaluate their impact on graft viability in UTx.

Methods.

According to a prospective interventional study within a national multiorgan procurement network, we described uterus procurement during multiorgan retrieval with protocol modifications between the first and second procedures, including bilateral femoral artery cannulation for pelvic perfusion and in situ identification of uterine and pelvic vessels before cold flushing. The main outcomes were the feasibility of procurement, graft viability, surgical parameters (ischemia times, operative duration), and short-term graft outcomes.

Results.

Among 8 evaluated recipients, 3 UTx procedures from brain-dead donors were performed. The first transplantation resulted in early graft loss because of bilateral arterial thrombosis requiring graft explantation on postoperative day 1. The second and third transplantations demonstrated sustained graft viability with normal perfusion on Doppler imaging up to 2 and 7 mo, resumption of menstruation, and only 1 episode of mild rejection responsive to corticosteroids. Technical refinements reduced back-table complexity and operative time (5 h 40 min versus 4 h 15 min) and enabled improved venous outflow through additional anastomoses. Bilateral femoral artery cannulation allowed effective pelvic perfusion without interfering with standard multiorgan procurement.

Conclusions.

UTx from brain-dead donors is feasible within a multiorgan procurement setting. Bilateral femoral cannulation combined with in situ vascular identification improves graft preparation and may enhance reproducibility without compromising retrieval of vital organs. These refinements represent important steps toward standardization of deceased donor UTx.

INTRODUCTION

Uterus transplantation (UTx) represents a major breakthrough both in the transplantation field and in reproductive medicine. Since the proof of concept established by Mats Brännström in 2014, which led to the first live birth of a healthy child following UTx, 141 UTxs have been performed worldwide.^1-3 The majority have involved living donors, most often the mothers of recipients born without a uterus, while approximately 25% have been performed using grafts retrieved from deceased donation after brain death (DBD).⁴

To date, only 6 countries worldwide have reported UTxs using grafts from DBD: Brazil, the Czech Republic, Turkey, the United States, Italy, and Great Britain.⁴ Brazil was the first to demonstrate the feasibility of achieving a live birth with a uterus graft retrieved from a DBD donor.⁵ However, early experiences with DBD UTx were sometimes unsuccessful.⁶ In Brazil, only the second DBD UTx performed by the team led to the birth of a healthy child.⁵ In the Czech Republic, the first case required graft removal on postoperative day 7 because of graft thrombosis.^7-9 In the United States, the consortium involving 4 centers (Cleveland, OH; Baylor; University of Pennsylvania [Philadelphia, PA]; and University of Alabama [Birmingham, AL]) reported explantation of the uterus graft 12 d after the first DBD transplantation.^4,10 In Italy, 3 UTx were performed, and 1 graft was lost 3 mo after UTx.¹¹ Nevertheless, today UTx from DBD is expanding worldwide, and the graft survival rate is about the same whether using a living or deceased donor at the beginning of the learning curve: recent reports show rates between 79% and 84%, respectively.^1,3

Uterus graft retrieval from a DBD during a multiorgan procurement (MOP) is a major determinant of UTx success and poses challenges distinct from those encountered in living donor retrieval. In the setting of MOP, uterus procurement must be performed without compromising the retrieval of other vital organs, in a context of unplanned surgery, frequently at night, involving multiple surgical teams that may not be familiar with each other. Uterus graft removal requires some adaptations during the MOP procedure, including specific cannulation to allow pelvic washing. Currently, the technical descriptions of uterus graft procurement from DBD donors during multiorgan retrieval vary considerably between countries, potentially leading to inconsistent reproducibility and outcomes.^5,10,12,13

The aim of the present article is to describe the technical modifications implemented between the first 2 UTxs from DBD donors performed in France, a step toward MOP with uterus graft retrieval standardization.

MATERIALS AND METHODS

Preparation for UTx in the Rennes Teaching Hospital (France) included cooperation with the Gynecologic department in Limoges Teaching Hospital, France (to ensure the best geographical conditions for some patient follow-up). The UTx study was permitted by the Ministry of Health of France in June 2023 and approved by the Ethics Committee. The study was officially begun in September 2023 (NCT05726305). A formal collaboration was implemented with 10 hospitals authorized for graft procurement from DBD (Brest, Vannes, Saint-Brieuc, Rennes, Avranches, Nantes, Angers, Tours, Caen, and Grenoble Hospitals).

Waiting List Recipient

For the purpose of this study, the recipients were evaluated with regard to the number of fertilized oocytes and gathered frozen embryos before UTx, the UTx procedure, surgical time, estimated blood loss, complications in the postoperative course, inpatient stay length, and cervical biopsies for organ rejection monitoring. All evaluated recipients were women with absolute uterus factor infertility, defined by the absence of a uterus (congenital or acquired), with no previous major surgery, and with six frozen blastocyst-stage embryos (obtained by IVF). For women with congenital uterus aplasia (Mayer-Rokitansky-Küster-Hauser syndrome), a vaginal length >7 cm was required for inclusion. When women fulfilled all the inclusion criteria, they were included in the French national transplantation waiting list (named CRISTAL).

Deceased Donor Uterus Retrieval—Surgery Description

The study was initiated only after completing a first protocol for uterus procurement for scientific purposes (ie, without transplantation afterward) during MOP procedures performed for heart, liver, and/or kidney transplantations. Ethical approval (number 21.37) and agreement (PFS21-010) from the French Biomedicine Agency were obtained. Organ donation network coordinators informed families about organ donation and decided whether to include information about consent to uterus retrieval. After oral consent to both organ donation and uterus donation was obtained, the procurement surgery was organized. The surgical team performed 10 uterus procurements (without transplantation) following the surgical approach described by Gauthier et al¹⁴ (see below). These 10 uterus procurement procedures showed no impact on the surgical removal of other grafts.

Eligible uterus donors for transplantation were all deceased donors with established brain death according to valid criteria set by French transplantation law, who also fulfilled the following study criteria: female sex, up to 42 y old (a compromise between various age limits available in the literature^8,12), without a history of cancer, endometriosis, or myoma.

Procedure for Uterus Procurement for the First Uterus Transplantation of the Study

For the first UTx, we used the uterus procurement technique described by Gauthier et al.¹⁴

• -Step 1: Vessel preparations for pelvic, abdominal, and thoracic levels. The donor was in a supine position. The first step of the procedure was the dissection of the bilateral femoral vessels through 2 inguinal incisions in preparation for later catheter placement, allowing flushing of the pelvic organs (ie, uterus). One femoral vein was prepared for later catheter placement, allowing venous discharge (Figure 1). This procedure, including cannulation of both femoral arteries and 1 femoral vein, was performed in all donors. To secure and not modify abdominal organ procurements (ie, standard multiorgan retrieval procedure) involving several varying teams, an abdominal aorta catheter was also used for abdominal organ retrieval, such as the kidney, liver, or pancreas. Thus, after a midline incision was made from the pubis to the sternum, the aorta and inferior vena cava were prepared for later catheter placement. Lungs, heart, pancreas, liver, and kidneys dissection fields were prepared before flushing.
• -Step 2: Macroscopic assessment of uterus quality.
• -Step 3: Cannulation of all the vessels. Catheters were placed inside the aorta, inferior vena cava, femoral arteries, and 1 femoral vein; the contralateral femoral vein was ligated to avoid an unnecessary loss of preservation solution.
• -Step 4: “Flushing.” During thoracic aorta cross-clamping and celiac aorta clamping, flushing with IGL-1 solution (Institut Georges Lopez, Lyon, France) at 4 °C through aorta and femoral artery catheters was continued in all the abdominal and pelvic areas, and the venous discharge catheters were opened. The thoracic, abdominal, and pelvic cavities were filled with sterile ice. Flushing was stopped once the discharge fluid was clear. Then, all the organs selected for procurement were successively retrieved: heart, lung, liver, kidneys, and pancreas. The uterus was removed last because of its non–life-saving nature.
• -Step 5: Uterus procurement. The aim was to preserve the uterus vasculature. The round ligaments were cut close to the pelvic side wall. The ovaries and fallopian tubes were separated from the uterus and left in place. The paravesical and pararectal spaces were dissected to individualize the parametria. The vesicocervical and rectovaginal spaces were dissected well below the vaginal fornix. No attempts were made to identify the uterine vessels. The ureter was cut just below its crossing with the iliac vessels. It was not dissected in its parametrial part to avoid tearing the uterine vessels. Laterally, the internal iliac vessels were cut above their anterior division. The anterior internal iliac trunks were gently dissected from the pelvic side wall to preserve the cervix and the uterine arteries and veins. The goal was to secure the bilateral vascular internal iliac pedicles. The umbilical, bladder, and vaginal vessels were cut below the parametria. At the end of the procedure, the explanted specimen consisted of the uterus and proximal vagina with the parametria up to the anterior internal iliac vessels. The specimen was packed in a cold preservation solution (IGL-1).
• -Step 6: Travel from the organ procurement hospital to the UTx hospital (Rennes Teaching Hospital, France).
• -Step 7: Uterus graft preparation. The uterus graft was prepared on the back table and placed on sterile gauze over crushed ice to maintain a temperature of approximately 4 °C throughout the procedure. Preparation began with hysteroscopic evaluation of the uterine cavity to confirm its integrity and exclude any intracavitary abnormalities. The ureter was divided immediately before its entrance into the parametrium. The intraparametrial segment of the ureter was intentionally preserved within the graft. The vaginal vascular pedicles were secured with vascular clips to prevent leakage during subsequent graft reperfusion. Both internal iliac arteries were then catheterized and flushed with cold IGL-1 preservation solution to assess vascular patency and identify potential leakage sites. Any leaking branches were repaired using 6-0 polypropylene (Prolene) sutures. The collateral branches arising from the internal iliac arteries were progressively ligated, leaving only the uterine arteries patent for transplantation. The venous system was subsequently evaluated: both internal iliac veins were flushed with IGL-1 preservation solution to verify vascular integrity and detect any leakage points, which were repaired as necessary. After completion of all vascular checks and reconstructions, a final graft perfusion was performed through both internal iliac arteries using 1 liter of cold IGL-1 preservation solution.

FIGURE 1.

Femoral cannulation for pelvic flushing in uterus retrieval procedures for transplantation from brain-dead donors. A, Preparation of the femoral vessels on both sides, with cannulas inserted during the abdominal preparation phase. B, View of the inguinal side: both arterial (*) and venous (▲) vessels are cannulated. This procedure, including cannulation of both femoral arteries and 1 femoral vein, was performed in all donors.

Modified Procedure for Uterus Procurement for the Second and Other Uterus Transplantations in the Study

After the first UTx from DBD in the study, we proposed some amendments to the protocol to optimize uterus procurement and graft preparation. Thus, we described the new protocol for uterus procurement from MOP surgery below.

• -Step 1 and step 2 were similar to the above description.
• -A new step was added after step 2. During the thoracic incision, heart and lung preparation fields, and before catheters were placed in the aorta and inferior vena cava, femoral arteries and veins, uterus, and pelvic vessels were prepared. During this new step, named “uterus and pelvic vessels preparation,” the round ligaments were cut between ligatures close to the pelvic side wall, the peritoneum above the urinary bladder was preserved with the uterus graft, and vesico-uterine and vesico-vaginal spaces were developed. The paravesical and pararectal spaces were dissected to individualize the parametria. The rectovaginal space was dissected well below the vaginal fornix. The utero-ovarian veins were identified. The ovaries and tubes were removed after ligatures. Lastly, pelvic vessels were prepared: internal iliac arteries and veins were identified, and their branches were gently dissected from the pelvic sidewall, and some of them were ligated at the level of pelvic entry. Umbilical arteries were marked with 2 clips and sectioned. All the vessels of the graft were marked with rubber slings to optimize graft preparation on the back table.
• -The next steps were similar to the ones previously described, starting from step 3. The final step (previous step 5: “uterus procurement”) was quicker because the vessels and the grafts were prepared before flushing. The uterus was removed last, because of its non–life-saving nature. Vessels identified between rubber slings were cut, and ureters were identified with 3 clips. Parameters were sectioned against the pelvic wall to preserve uterine vessels, and the vagina was finally sectioned.
• -For step 8, the bilateral utero-ovarian veins were also catheterized and flushed with IGL-1 preservation solution to assess their patency and integrity. Any vascular defects or leakage sites identified during this assessment were repaired using interrupted 8-0 polypropylene (Prolene) sutures when required. To facilitate rapid identification of the vascular pedicles during implantation, each internal iliac artery was marked with a long suture. Similarly, the 4 venous outflow tracts (bilateral internal iliac veins and bilateral utero-ovarian veins) were individually tagged with long sutures. This standardized labeling allowed immediate identification of all vascular structures upon transfer of the graft to the recipient operative field and minimized warm ischemia time during vascular anastomosis.

We can sum up the new uterus procurement protocol from DBD as follows:

• -Step 1: Vessel preparations for the pelvic, abdominal, and thoracic levels.
• -Step 2: Macroscopic assessment of uterus quality.
• -Step 3: Uterus and pelvic vessel preparation.
• -Step 4: Cannulation of all vessels.
• -Step 5: “Flushing.”
• -Step 6: Uterus procurement.
• -Step 7: Travel to the transplantation center.
• -Step 8: Uterus graft preparation.

The main technical modifications introduced after the first UTx are summarized in Figure 2 and Table 1. These amendments aimed to shift a substantial part of the vascular dissection from the after-clampage procedure and back table to the before-clampage donor procedure.

FIGURE 2.

Surgical times of the first 2 organ procurement procedures of the study. Surgical times between the first MOP with uterus removal and subsequent procedures. In the first case, uterus preparation was performed after aortic clamping and the removal of all vital organs. In the second and third cases, uterus preparation was performed while the heart was still beating. Refining this technique did not prolong the duration of cold ischemia in other organs, which actually decreased from 446 min in the first case (for the last vital organ) to 370 min in the third case. This protocol also reduced the duration of cold ischemia of the uterus during MOP, from 519 min in the first case to 390 min in the third case. MOP, multiorgan procurement.

TABLE 1.

Technical modifications introduced after the first uterus transplantation

Surgical phase	First uterus transplantation	Modified protocol (second and subsequent transplantations)	Expected benefit
Internal iliac vessels	Internal iliac vessels identified only during uterus procurement.	Internal iliac arteries and veins dissected and partially prepared before flushing.	Easier identification and safer preservation of vascular pedicles.
Utero-ovarian veins	Not specifically identified during procurement.	Bilateral utero-ovarian veins systematically identified and preserved before flushing.	Additional venous outflow options and improved graft venous assessment.
Management of vascular branches	Branches managed mainly during back-table preparation.	Several pelvic vessel branches ligated during donor surgery before flushing.	Reduced back-table preparation time.
Vessel marking during procurement	No systematic marking of graft vessels.	All vascular pedicles marked with rubber slings before flushing.	Rapid identification of vessels during procurement.
Uterus procurement after flushing	Extensive dissection still required after organ retrieval.	Most dissection already completed; procurement consisted mainly of vessel division and graft extraction.	Shorter procurement time and reduced cold ischemia.
Identification of vessels for implantation	No systematic labeling after graft preparation.	Internal iliac arteries and all 4 venous outflow tracts individually tagged with long sutures.	Immediate identification during implantation and reduced warm ischemia time.
Overall strategy	Procurement-focused approach with most vascular preparation performed on the back table.	Anticipatory procurement approach with extensive vascular preparation performed before flushing.	Shorter graft preparation time, easier implantation, and optimization of ischemia times.

Uterus Transplantation—Surgery Description

The surgery was performed via a lower midline laparotomy. The vaginal vault was prepared for the anastomosis. External iliac veins and arteries were dissected and successively clamped for anastomosis after an intravenous heparin bolus. The graft was placed in an orthotopic position. The uterus implantation began with the graft internal iliac vein anastomosis in an end-to-side fashion with the recipient external iliac vein (using continuous sutures with polypropylene 6-0, Prolene). The internal iliac artery anastomosis was also performed in an end-to-side fashion with the recipient external iliac artery. Once anastomoses on both sides were completed, the clamps were removed, and the graft was reperfused.

The last part of the transplantation involved the completion of the vaginal anastomosis (using individual stitches with Vicryl 2-0).

The round ligaments and the peritoneal flap of the uterus graft were sutured to the rudimentary round ligaments and the bladder peritoneum of the recipient, respectively.

The surgical technique has evolved in response to initial setbacks met during the first UTx. The 3 major changes after the first UTx were:

• -to suture the posterior wall of the recipient’s vagina with the posterior vaginal section of the graft before vascular anastomosis, allowing for correct vascular anastomosis placement and orientation, thus avoiding kinking.
• -to add bilateral anastomosis between the external iliac vein of the recipient and the utero-ovarian veins of the graft to increase the blood outflow of the graft.

Postoperative Immunosuppressive Treatment

Immunosuppression induction was performed with a bolus of methylprednisolone intraoperatively and a Simulect injection. Maintenance immunosuppression was based on oral prednisolone, tacrolimus, and mycophenolate mofetil. The target blood levels of tacrolimus were 8–10 ng/mL in the first 4 wk posttransplantation, and 6–8 ng/mL until embryo transfer. Mycophenolate mofetil (CellCept Genentech) was given at an oral dose of 1000 mg twice a day.

RESULTS

Recipients

Between September 2023 and November 2025, 8 patients were evaluated and included in the study. Each patient had 6 or more blastocyst-stage embryos. All the patients were Mayer-Rokitansky-Küster-Hauser patients with no history of vaginoplasty, but rather conservative approaches, that is, vaginal dilatations (Frank methods) or sexual activity. The mean age of the 8 patients was 31.5 y old (±3.2 y old). The first patient was included in the French national waiting list “CRISTAL” in December 2024, 4 patients in January 2025, and 3 patients in September 2025.

Death Brain Donors

Between December 2024 and December 2025, among the 10 organ procurement hospital centers included in the study network, 10 potential donors were identified by the organ donor network coordinators (among around 250 organ brain death donors identified during the same period). One ideal eligible uterus donor was not selected because of family refusal. Three proposals were declined because of blood group incompatibility (ie, no AB or B blood group in the 8 recipients). Two donors were also declined because of gynecologic diseases: 1 donor had a history of bowel endometriosis, and another donor had an undetermined ovarian mass detected by the body scan. One donor was also refused because the uterus transplantation team was not available. Three donors were selected for uterus donation. The characteristics of the donors are summarized in Table 2. Surgery times of all organ retrieval procedures are shown in Figure 2.

TABLE 2.

Characteristics of the donors

UTx number	Age (y)	BMI (kg/m2)	Cause of death	Parity	Smoking habit (packs per year)	Other organ removed during the procedure
No. 1	35	22.8	Subarachnoid hemorrhage	0	15	Heart, lung, liver, kidneys, pancreas
No. 2	30	35.3	Subarachnoid hemorrhage	0	10	Heart, lung, liver, kidneys, pancreas (islets)
No. 3	24	22.4	Road traffic accident	0	0	Heart, lung, liver, kidneys

BMI, body mass index.

Uterus Transplantation Surgery

For the first UTx, we used the uterus procurement technique described by Gauthier et al¹⁴ and above. The back-table duration was 135 min. A lot of time was lost for the identification of the vessels, especially the veins. Surgery duration for the UTx was 5 h 40 min, with 2 h of warm ischemia during vessel anastomoses (2 arterial anastomoses and 2 vein anastomoses). The total duration of ischemia between vessel clamping in the donor and unclamping in the recipient was 11 h.

For the following UTx, we used the modified procedure (8 steps) described above (in the Materials and Methods section; Figure 2). For the second UTx, the back-table duration was 120 min. Surgery duration for this second UTx was 4 h 15 min, with 100 min of warm ischemia during vessel anastomoses (with 2 arterial anastomoses and 4 venous anastomoses [2 uterine veins and 2 utero-ovarian veins]). The total duration of ischemia between clamping in the donor and unclamping in the recipient was 10 h.

For the third UTx, the back table was 90 min. Surgery duration for this second UTx was 4 h 10 min, with 120 min of warm ischemia during vessel anastomoses (2 arterial anastomoses and 4 vein anastomoses). The total duration of ischemia between clamping in the donor and unclamping in the recipient was 7 h.

Uterus Transplantation Outcomes at 6 mo

For the first UTx, an angioscan was performed on day 1 according to the study protocol. Blood perfusion was bilaterally halted in the graft 2 centimeters after the arterial anastomoses. A new surgery under general anesthesia was therefore performed: the graft was pale, no pulse was found when examining its arteries, and the graft showed no bleeding with a scalpel injury. These observations led us to explant the graft on day 1 after transplantation. Pathological analysis showed bilateral arterial thrombi next to 2 atherosclerotic plaques within the uterine arteries, reducing the arterial lumen by 40% on the right side and 30% on the left side (Figure 3). The patient quickly recovered, with hospital discharge occurring on day 7 after uterus explantation. The postoperative follow-up showed no identified sequelae, with a good aspect of the vaginal cuff and no sexual disturbance 4 mo later.

FIGURE 3.

Pathological analysis of explanted uterus graft. A, Right artery showing a narrowing of the lumen because of recent thrombosis (★) next to focal atherosclerosis (▲) (HES ×20). B, Left artery showing a narrowing of the lumen because of recent thrombosis (★) next to focal atherosclerosis (▲) with no calcification (HES ×40). C, Normal myometrium and endometrium with no edema or necrosis (HES ×10). D, Vessels in myometrium without thrombosis or vasculitis (HES ×40). HES, haematoxylin, eosin, and saffron.

For the second UTx, postoperative Dopplers showed good graft perfusion on day 1, 3, and 7 and 6 mo later. Menstruation started 5 wk after the transplantation and was regular to this day. Hospital discharge occurred 16 d after UTx, and the patient returned to work 4 mo after the UTx. During follow-up, we observed 1 episode of grade 1 rejection, according to the 2017 classification,¹⁵ 2 mo after the transplantation, solved within15 d after increasing the daily dosage of betamethasone. Speculum evaluation of the cervix showed a normal aspect 2 mo after UTx (Figure 4).

FIGURE 4.

Aspect of the cervix in the second recipient and uterine artery Dopplers. A, Cervix aspect of graft at day 31 after uterus transplantation. B, Doppler of right artery at day 18 after uterus transplantation. C, Doppler of left artery at day 18 after uterus transplantation. D, Sonographic aspect of uterus graft at day 38 after uterus transplantation. During follow-up of the second recipient, speculum examination showed a normal appearance of the grafted cervix on day 31 (A) and a normal Doppler spectrum of the uterine arteries on day 18 (B and C). Uterine ultrasound performed on day 38 showed a normal appearance of the graft with a 3-layered endometrium in sagittal view (D).

For the third UTx, postoperative Dopplers showed good graft perfusion on day 1, 3, and 7 and 2 mo later. At day 2 posttransplantation, a hemoperitoneum was identified, and a new surgery was performed. The peritoneal cavity was cleaned of blood hemorrhage, but no active bleeding was identified during this new surgery. A grade 1 rejection occurred in the first few weeks and was managed by solumedrol bolus.

DISCUSSION

This study demonstrates the feasibility of UTx from brain-dead donors without compromising the procurement of other vital organs. Given the limited number of procedures performed worldwide (35 UTx; Table 3), standardization of surgical techniques remains a critical challenge.^1,2 We acknowledge that graft loss in the first transplantation was likely multifactorial and primarily related to donor-related atherosclerotic vascular pathology, rather than a technical issue alone. Accordingly, the procurement modifications should be interpreted as improvements in workflow and graft preparation efficiency rather than interventions expected to prevent donor-intrinsic vascular complications such as thrombosis on preexisting arterial disease.

TABLE 3.

Published data about uterus transplantation from brain-dead donors

Country	No. of DBD UTx	No. of live births	Graft failure
Turkey	2	2	0
United States (4 centers)	19	7	3
Brazil	2	1	1
Czech Republic	5	1	2
Italy	3	1	1
United Kingdom	1	1	0
France	3	0	1
Total	35	13	23% (8/35)

DBD, donation after brain death; UTx, uterus transplantation.

A key finding of our work is the importance of the procurement sequence, particularly the early identification of the uterine vascular structures. After cold perfusion, anatomical landmarks become difficult to identify on the back table, potentially affecting graft preparation and outcomes, as previously described.¹² This technical limitation supports the need for optimized in situ identification before flushing. In addition, graft thrombosis remains the main perioperative complication, justifying immediate postoperative anticoagulation.⁵ Following the first UTx in the present trial, several modifications were introduced to the procurement protocol to optimize graft retrieval and back-table preparation. These changes were driven by the observation that a substantial part of vascular identification was performed after organ retrieval, leading to prolonged back-table preparation. The revised strategy therefore shifted key steps of vascular dissection and identification to the donor procedure, with the aim of improving anatomical clarity and procedural standardization (Table 1). In subsequent transplantations, these modifications resulted in a more streamlined workflow, with faster graft preparation and easier identification of vascular structures at implantation, thereby reducing both cold and warm ischemia times.

To date, experience with brain-dead donor UTx remains limited, with approximately 35 procedures reported worldwide and variable outcomes, including a substantial graft loss rate of 23%.^4,5,9,10 Surgical approaches differ significantly between teams. Some groups have proposed procurement before multiorgan retrieval under beating-heart conditions to facilitate vascular identification.¹² However, this strategy may interfere with standard MOP and raises concerns. Indeed, in the French national system, as in most international organ procurement frameworks, nonvital organs such as the uterus cannot be retrieved before life-saving organs. Consequently, strategies proposing full uterus procurement before multiorgan retrieval under beating-heart conditions are not widely applicable and are not reproducible in many centers worldwide.

Furthermore, in contrast to approaches relying on selective iliac artery cannulation for pelvic perfusion,^5,8 our strategy is based on bilateral femoral artery cannulation combined with a unilateral venous drainage line, thereby maintaining standard aortic and cava cannulation used for MOP. While iliac cannulation may provide direct pelvic access, it can also introduce technical complexity in the abdominal phase of procurement, particularly because of spatial conflict between upward-oriented aortic and downward-oriented iliac arterial catheters. This may increase procedural clutter within the operative field and potentially complicate coordination between surgical teams during multiorgan retrieval. By contrast, femoral cannulation offers a more linear and anatomically neutral access route that avoids interference with standard upper abdominal dissection and preserves the established sequence of life-saving organ procurement. In addition, this configuration maintains full compatibility with conventional multiorgan retrieval protocols, which is essential within current national and international frameworks where procurement of life-saving organs must not be compromised by adjunct research procedures. In the longer term, this approach may represent a step toward an integrated procurement strategy combining uterus and abdominal organ retrieval through a simplified and standardized cannulation scheme, potentially based on bilateral femoral arterial inflow and venous drainage, which could flush pelvic and abdominal organs. However, at the present stage, such an evolution cannot be implemented in routine practice, as MOP protocols must prioritize life-saving organs. Our current strategy therefore represents a research-adapted compromise designed to optimize uterus graft quality while preserving the integrity of established multiorgan retrieval procedures.

This study has limitations. First, no pregnancies have yet been achieved, as embryo transfer is planned in a subsequent phase of the protocol. Second, we acknowledge that it is not possible to attribute the initial graft loss solely to donor-related atherosclerotic disease, nor to conclude that the outcomes of the second and third transplantations are exclusively the result of the technical refinements introduced in our procurement protocol. Indeed, graft failure is likely multifactorial, involving both donor-intrinsic vascular pathology and procedural factors, and the relative contribution of each cannot be definitively established based on the present series. Nevertheless, the technical modifications described here may contribute to improving reproducibility and safety in deceased donor uterus procurement, which are essential for broader clinical adoption.

CONCLUSIONS

Bilateral femoral artery cannulation enables effective pelvic perfusion and allows the procurement of a uterus suitable for transplantation without disrupting standard multiorgan retrieval. These technical refinements may improve reproducibility and support the wider development of UTx programs.

ACKNOWLEDGMENTS

The authors thank Maela Le Lous (Obstetrics), Sylvie Jaillard (Genetics), Auriane Cospain (Genetics), Arthur Legal (Intensive care), Maria Lahlou (Anesthesia), Axelle Maurice (Anesthesia), Thierry Frouget (Nephrologist), Cecile Vigneau (Nephrologist), Leonard Golbin (Nephrologist), Marie Pascale Morin (Nephrologist), Matthieu Revest (Infectiologist), and Marion Baldeyrou (Infectiologist), who are involved in the present uterus transplantation trial. The authors thank Céline Parc and Gildas Leborgne for their support in the regulatory procedures. The authors also thank the CHPOTs (Brest, Vannes, Saint-Brieuc, Rennes, Avranches, Nantes, Angers, Tours, Caen, and Grenoble hospitals) for their participation in the TULIP project.