Patients may accept significant risks in their pursuit of a family.1 Assisted reproductive techniques (ART) have safely eliminated many obstacles in this pursuit, but risks and complications can still occur while attempting to conceive.2
Transplantation medicine and surgery have advanced sufficiently since the first kidney transplant to allow nonvital organ allografts.3 Reproductive organ transplantation has been reported,4,5 and uterine transplantation has recently been proposed to treat infertility in a select group of patients.6,7 We report our initial experience with a local organ donor network for the retrieval of the human uterus from multi-organ, heart-beating, brain-dead donors.
MATERIALS AND METHODS
Research approval for the human uterus transplantation project was received from the New York Downtown Hospital Institutional Review Board and the New York Organ Donor Network before initiation at all participating organ donor network hospitals. Currently, consent for uterus retrieval is obtained separately from other organ donation consents by organ donor network coordinators. Organ procurement proceeded through the usual mechanisms, ie, participation with the United Network for Organ Sharing as stipulated by the National Organ Transplant Act of 1984.
Mandatory donor preoperative testing included all donor network requirements, eg, testing for the presence of human immunodeficiency virus (HIV), hepatitis, and terminal blood gases. Additional requirements for an acceptable uterus retrieval included recent normal Pap test, human papillomavirus (HPV) negative status, and pelvic sonogram or magnetic resonance imaging. Additional criteria for uterus donation included donor ages between 16 and 45 years, regular menses, and no known infertility before the incident illness. Donor exclusion criteria included leiomyomata larger than 1 cm on imaging or gross inspection, history of: pelvic inflammatory disease, cervical dysplasia, endometrial or ovarian cancer, endometriosis or adenomyosis, and any other anatomical abnormalities of the uterus.
Procurement surgery was initiated after review of the donor's history, confirmation of consent, and documentation of brain death. ABO blood type, review of appropriate serology tests, and preoperative hemodynamic and respiratory data were obtained per organ donor network policy.
Donor surgery was in the supine position, with 100% oxygen throughout the recovery, and used a midline incision from pubis to sternotomy. The only additional step needed for uterus retrieval as part of a multi-organ retrieval protocol is placement of bilateral femoral artery catheters before laparotomy for the uterus flush as described later.
After the multi-organ dissection was complete and otherwise ready for flushing, attention was turned to the uterus for inclusion in the multi-organ recovery. Based on our techniques reported in different animal models (including rat, rabbit, pig, and nonhuman primates), two surgical approaches were used in these seven human cases (Del Priore G, Zhang JJ, Diflo T, Silber S, Smith JR. Ovary and uterine transplant: a feasible rat model [abstract]. Fertil Steril 2005;84:s58; Smith JR, Boyle D, Ungar L, Corless DJ, Del Priore G. Uterine transplant: a successful porcine model [abstract]. Fertil Steril 2001;76:s106).8,9
In kidney-only procurement donors, the vena cava and aorta was dissected from the level of the inferior mesenteric artery to the bifurcation of the common iliac vessels, continued inferiorly and medially until the uterine vessels were identified in the retroperitoneum by following the internal iliac vessels' anterior division. In pancreas and liver transplant donors, this techniques was followed, but the iliac system was eventually used as a part of the other organ transplants, ie, recreation of pancreatic and hepatic anastomoses. Heart or lung retrieval or both did not interact with the uterus procurement at all.
For adequate exposure in the pelvis, the round ligament was divided close to the pelvic sidewall. The paravesical and pararectal spaces were then developed with blunt dissection. This allowed the cardinal ligament and parametrium to be directly accessed by placing one's fingers in the newly opened spaces. The dissection of the external iliac vessels continued caudally until the circumflex iliac vessels were encountered proximal to the inguinal ligament. The inguinal ligaments were transected and the vessels dissected to the previously placed femoral artery catheters.
Sharp dissection was used to create vesicocervical and rectovaginal spaces by freeing the bladder and rectum from the uterus down to the level well below the vaginal fornix. The ureters were dissected bilaterally as they passed under the uterine vessels and retracted laterally to allow careful preservation of the uterine vessels. Ureter length retrieved is not reduced by uterus retrieval by retracting the cardinal ligament medially and inferiorly, thus allowing the ureter to be transected at the bladder and retracted cephalad with the kidney.
The superior vesicle and all collateral vessels of the internal iliac system were ligated except for the uterine branches. No attempt was made to identify a single uterine vein. Instead, a generous amount of parametria surrounding the more obvious uterine artery was preserved with its attachment to the internal iliac vein. The surrounding fat and lymphoid tissue was dissected from the vessels after removing the uterus from the donor. The ovarian vessels, ovary, and adjacent fallopian tubes were retained with the uterus for use as an additional backup vascular pedicle in the event that the pelvic dissection might fail to secure usable vessels.
In general, the goal was to secure bilateral vascular pedicles of the greatest length possible by dissecting laterally from the uterus to at least the anterior division of the internal iliac vein. Attempts were made to secure the entire internal iliac vessels, if possible, just distal of their origins from the common iliac vessels. This would allow donor vessels to be anastomosed to the recipient's bilateral internal iliac vessels.
When all retrieval teams (thoracic, abdominal, and pelvic) were ready with all organs satisfactorily dissected, a wedge section of the uterine fundus including endometrium was removed for future studies. A heparin bolus of 30,000 units was given in concordance with the other organ teams. Because the abdominal aorta is ligated distal to the inferior mesenteric artery to concentrate the flush in the upper abdominal target organs, the femoral artery catheters were used for the uterus flush. At the time of thoracic aortic cross-clamping, done according to standard multi-organ procurement procedures, the femoral arteries were also clamped distal to the previously placed bilateral femoral artery catheters. Flush with University of Wisconsin solution at 4°C was continued at all sites until the incised vena cava was clear. The visceral cavities including pelvis were filled with sterile ice slush. Immediately after flushing, another sample was obtained from the uterine fundus including endometrium. More uterine tissue samples were obtained from the specimens at 15- to 30-minute intervals for 12 hours and stored for future analysis. After the retrieval of the other donor organs to be used in transplant recipients, the last step of the uterine recovery began by entering the vagina in a circumferential fashion, thus freeing the uterus entirely. Warm ischemic time for the uterus is the same as that of the other organs, ie, essentially none.
Additional dissection to complete the preparation of the removed uterus for transplantation took place on an extracorporeal ice bath. These additional steps included removing the ovaries while preserving the utero-ovarian ligaments and confirming venous return through the uterus from the cannulated arteries.
Since both approvals (approximately 6 months), there have been nearly 150 standard multi-organ procurements from approximately 1,800 eligible donors. Of these only nine families specifically consented to allow the uterus retrieval as part of the multi-organ donation. The nine donors were representative of the general organ donor pool. Causes of death included cerebral vascular accidents (2), cardiac arrest during electrophysiologic testing (1), and traumatic brain injury (6). All were parous females, ages 30–45 years, with one to three deliveries of healthy children. One consented donor did not have retrieval surgery because of rapidly deteriorating clinical status before a full retrieval team could be gathered.
Flow rates and pressure through the femoral arteries remained constant throughout the uterine flush, suggesting intact arterial and venous systems when femoral arteries were used for flushing. Extracorporeal perfusion with mechanical perfusion pump demonstrated free flow pressures of 25–40 mmHg. Without the addition of the femoral artery catheters, the only flush to the uterus is through the ovarian arteries proximal to the ligated abdominal aorta. Extracorporeal flush pressures and flow rates indicated that this is inadequate. Placement of an additional aortic flushing catheter distal to the upper abdominal cannulation site was possible, but cumbersome, because of the additional line in the operating field.
Additional operating time ranged from 30 to 10 minutes. There was little-to-no warm ischemia time. By the fourth patient, the entire uterine dissection added approximately 15 minutes to the multi-organ retrieval. Additional blood loss was 10–20 mL, exclusive of one case which had a 250-mL loss due to a lacerated internal iliac vein.
Although attempts were made in each case to obtain the greatest length of the internal iliac vessels possible, only 2 specimens included the entire length of both artery and vein to the common iliac origin (fourth and seventh donors) (Fig. 1). The remaining five specimens contained ample lengths of the anterior division of the iliac vessels. However, two of the specimens (first and fifth donors) had unilateral loss of uterine vessels. In these two specimens, the ovarian vessels could be cannulated, and venous return flow appeared normal.
Shown in Figure 2 is the biopsy after 12 hours of cold ischemia from a single donor. There was no evidence of morphologic changes in the myometrium or endometrium of any section.
Experimental reproductive organ transplantation had been performed for decades in animals before current ART obviated the need to continue these early experiments.4 With the improvement of ART, many causes of infertility have been overcome. However, success in these areas has now again raised the question of what can be done for patients with no functional uterus.6
However, nonvital organ transplantation raises ethical issues. Hand and facial allografts have recently been reported,3 but the only uterine transplant carried out in the human was controversial and unsuccessful7
The safety of both the mother and future child must be considered in balancing the risk of uterus transplantation. The uterine transplant recipient mother must assume the risk of short-term immunosuppression.
For the fetus, transplant safety data are reassuring. Pregnancies in organ transplant recipients, including liver, lung, and kidney, have been reported almost as long as there have been transplants. When controlled for other factors, pregnancy outcomes appear acceptable.10,11 Fortunately, long-term safety data are available because generations have now become pregnant after organ transplants.11
Our transplant techniques have developed from our previous animal studies and radical cancer surgery experience (Del Priore et al [abstract], 2005; Smith et al [abstract], 2001).6,12 The surgical techniques needed for transplantation retrieval are direct extensions of newly developed gynecologic oncology surgeries, such as the abdominal radical trachelectomy and the laterally extended endopelvic resection.12,13 Our experience with actual human multi-organ retrieval has shown us that separate flushing is necessary for the uterus. The femoral arteries are preferred for flush cannulation because they do not interfere at all with existing retrieval protocols.
Since 1995, we and others have been building on fertility preservation in cancer patients to make uterine transplantation a theoretical possibility.14,15 The actual retrieval of a usable human uterus is an additional step in this direction. Our hope is to eventually restore reproductive function through transplantation of a human uterus.
1. Chalian R, Licciardi F, Rebarber A, Del Priore G. Successful infertility treatment in a cancer patient with a significant personal and family history of cancer. J Womens Health 2004;13:235–7.
2. Del Priore G, Ungar L, Smith JR. Complications after fertility-preserving radical trachelectomy. Fertil Steril 2006;85:227.
3. Devauchelle B, Badet L, Lengele B, Morelon E, Testelin S, Michallet M, et al. First human face allograft: early report. Lancet 2006;368:203–9.
4. Scott JR, Hendrickson M, Lash S, Shelby J. Pregnancy after tubo-ovarian transplantation. Obstet Gynecol 1987;70:229–34.
5. Silber SJ, Lenahan KM, Levine DJ, Pineda JA, Gorman KS, Friez MJ, et al. Ovarian transplantation between monozygotic twins discordant for premature ovarian failure. N Engl J Med 2005;353:58–63.
6. Del Priore G, Smith JR, Boyle DC, Corless DJ, Zacharia FB, Noakes DA, et al. Uterine transplantation, abdominal trachelectomy, and other reproductive options for cancer patients. Ann NY Acad Sci 2001;943:287–95.
7. Fageeh W, Raffa H, Jabbad H, Marzouki A. Transplantation of the human uterus. Int J Gynaecol Obstet 2002;76:245–51.
8. Sieunarine K, Moxey P, Boyle DC, Ungar L, Lindsay I, Del Priore G, et al. Selective vessel ligation in the pelvis: an invaluable tool in certain surgical procedures. Int J Gynecol Cancer 2005;15:967–73.
9. Sieunarine K, Hakim NS, Corless D, Ungar L, Del Priore G, Smith JR, et al. Is it feasible to use a large vessel patch with a uterine allograft en bloc for uterine transplantation? Int Surg 2005;90:257–61.
10. Scott JR, Wagoner LE, Olsen SL, Taylor DO, Renlund DG. Pregnancy in heart transplant recipients: management and outcome. Obstet Gynecol 1993;82:324–7.
11. Armenti VT, Radomski JS, Moritz MJ, Gaughan WJ, McGrory CH, Coscia LA. Report from the National Transplantation Pregnancy Registry (NTPR): outcomes of pregnancy after transplantation. Clin Transpl 2003:131–41.
12. Ungar L, Palfalvi L, Hogg R, Siklos P, Boyle DC, Del Priore G, et al. Abdominal radical trachelectomy: a fertility-preserving option for women with early cervical cancer. BJOG 2005;112:366–9.
13. Hockel M. Laterally extended endopelvic resection. Novel surgical treatment of locally recurrent cervical carcinoma involving the pelvic side wall. Gynecol Oncol 2003;91:369–77.
14. Wranning CA, Molne J, El-Akouri RR, Kurlberg G, Brannstrom M. Short-term ischaemic storage of human uterine myometrium–basic studies towards uterine transplantation. Hum Reprod 2005;20:2736–44.
15. Racho El-Akouri R, Wranning CA, Molne J, Kurlberg G, Brannstrom M. Pregnancy in transplanted mouse uterus after long-term cold ischaemic preservation. Hum Reprod 2003;18:2024–30.