Since the first baby was born after a uterus transplantation in Sweden in 2014, uterus transplantation has been rapidly transitioning toward clinical reality.1 Several teams in the United States and multiple teams worldwide have performed the procedure, with the total number of worldwide surgeries performed nearing 100.
Dr. Rebecca Flyckt
Uterus transplantation is the first and only true treatment for women with absolute uterine factor infertility – estimated to affect 1 in 500 women – and is filling an unmet need for this population of women. Women who have sought participation in uterus transplantation research have had complex and meaningful reasons and motivations for doing so.2 Combined with an accumulation of successful pregnancies, this makes continued research and technical improvement a worthy endeavor.
Most of the births thus far have occurred through the living-donor model; the initial Swedish trial involved nine women, seven of whom completed the procedure with viable transplants from living donors, and gave birth to eight healthy children. (Two required hysterectomy prior to attempted embryo transfer.3)
The Cleveland Clinic opted to build its first – and still ongoing – trial focusing on deceased-donor uterus transplants on the premise that such an approach obviates any risk to the donor and presents the fewest ethical challenges at the current time. Of eight uterus transplants performed thus far at the Cleveland Clinic, there have been three live births and two graft failures. As of early 2021, there was one ongoing pregnancy and two patients in preparation for embryo transfer.
Dr. Elliot G. Richards
Thus far, neither the living- nor deceased-donor model of uterus transplantation has been demonstrated to be superior. However, as data accrues from deceased donor studies, we will be able to more directly compare outcomes.
In the meantime, alongside a rapid ascent of clinical landmarks – the first live birth in the United States from living-donor uterus transplantation in 2017 at Baylor University Medical Center in Houston,4 for instance, and the first live birth in the United States from deceased-donor uterus transplantation in 2019 at the Cleveland Clinic – there have been significant improvements in surgical retrieval of the uterus and in the optimization of graft performance.5
Most notably, the utero-ovarian vein has been used successfully in living donors to achieve venous drainage of the graft. This has lessened the risks of deep pelvic dissection in the living donor and made the transition to laparoscopic and robotic approaches in the living donor much easier.
Donor procurement, venous drainage
Adequate circulatory inflow and outflow for the transplanted uterus are essential both for the prevention of ischemia and thrombosis, which have been major causes of graft failure, and for meeting the increased demands of blood flow during pregnancy. Of the two, the outflow is the more challenging component.
Courtesy Cleveland Clinic
Deceased-donor uterine graft
Venous drainage traditionally has been accomplished through the use of the uterine veins, which drain into the internal iliac veins; often the vascular graft will include a portion of the internal iliac vessel which can be connected via anastomoses to the external iliac vein classically in deceased donors. Typically, the gynecologic surgeon on the team performs the vaginal anastomosis and suspension of the uterus, while the transplant surgeons perform the venous and arterial anastomoses.
In the living-donor model, procurement and dissection of these often unpredictable and tortuous complexes in the deep pelvis – particularly the branching uterine veins that lie in close proximity to the ureter, bladder, other blood vessels, and rectum – can be risky. The anatomic variants in the uterine vein are numerous, and even in one patient, a comprehensive dissection on one side cannot be expected to be mirrored on the contralateral side.
Courtesy Cleveland Clinic
Figure B. Recipient upper vagina prepared for anastomosis with mucosal layer tagged.
In addition to the risk of injury to the donor, the anastomosis may be unsuccessful as the veins are thinly walled and challenging to suture. As such, multiple modifications have been developed, often adapted to the donor’s anatomy and the caliber and accessibility of vessels. Preoperative vascular imaging with CT and/or MRI may help to identify suitable candidates and also may facilitate presurgical planning of which vessels may be selected for use.
Recently, surgeons performing living-donor transplantations have successfully used the more accessible and less risky ovarian and/or utero-ovarian veins for venous anastomosis. In 2019, for instance, a team in Pune, India, reported laparoscopically dissecting the donor ovarian veins and a portion of the internal iliac artery, and completing anastomosis with bilateral donor internal iliac arteries to recipient internal iliac arteries, and bilateral donor ovarian veins to recipient external iliac veins.6 It is significant that these smaller-caliber vessels were found to able to support the uterus through pregnancy.
Courtesy Cleveland Clinic
Figure C: Deceased donor uterus immediately prior to implantation into the recipient. Note long vascular pedicles.
We must be cautious, however, to avoid removing donors’ ovaries. Oophorectomy for women in their 40s can result in significant long-term medical sequelae. Surgeons at Baylor have achieved at least one live birth after harvesting the donor’s utero-ovarian veins while conserving the ovaries – a significant advancement for the living-donor model.4
There is tremendous interest in developing minimally invasive approaches to further reduce living-donor risk. The Swedish team has completed a series of eight robotic hysterectomies in living-donor uterus transplantations as part of a second trial. Addressing the reality of a learning curve, their study was designed around a step-wise approach, mastering initial steps first – e.g., dissections of the uterovaginal fossa, arteries, and ureters – and ultimately converting to laparotomy.7 In the United States, Baylor University has now completed at least five completely robotic living-donor hysterectomies with complete vaginal extraction.
Published data on robotic surgery suggests that surgical access and perioperative visualization of the vessels may be improved. And as minimally invasive approaches are adopted and improved, the length of donor surgery – 10-13 hours of operating room time in the original Swedish series – should diminish, as should the morbidity associated with laparotomy.
Courtesy Cleveland Clinic
Figure D: Transplanted uterus immediately following graft reperfusion.
Surgical acquisition of a uterine graft from a deceased donor diminishes concerns for injury to nearby structures. Therefore, although it is a technically similar procedure, a deceased-donor model allows more flexibility with the length, caliber, and number of vessels that can be used for anastomosis. The internal iliac vessels and even portions of the external iliac vessels and ovarian vessels can be used to allow maximum flexibility.8