Articles

Livebirth after uterus transplantation from a deceased donor

in a recipient with uterine infertility
Dani Ejzenberg*, Wellington Andraus*, Luana Regina Baratelli Carelli Mendes, Liliana Ducatti, Alice Song, Ryan Tanigawa, Vinicius Rocha-Santos,
Rubens Macedo Arantes, José Maria Soares Jr, Paulo Cesar Serafini, Luciana Bertocco de Paiva Haddad, Rossana Pulcinelli Francisco,
Luiz Augusto Carneiro D’Albuquerque, Edmund Chada Baracat

Summary
Background Uterus transplantation from live donors became a reality to treat infertility following a successful Swedish Lancet 2018; 392: 2697–704
2014 series, inspiring uterus transplantation centres and programmes worldwide. However, no case of livebirth via Published Online
deceased donor uterus has, to our knowledge, been successfully achieved, raising doubts about its feasibility and December 4, 2018
http://dx.doi.org/10.1016/
viability, including whether the womb remains viable after prolonged ischaemia.
S0140-6736(18)31766-5
See Comment page 2657
Methods In September, 2016, a 32-year-old woman with congenital uterine absence (Mayer-Rokitansky-Küster-
*These authors contributed
Hauser [MRKH] syndrome) underwent uterine transplantation in Hospital das Clínicas, University of São Paulo, equally to the manuscript
Brazil, from a donor who died of subarachnoid haemorrhage. The donor was 45 years old and had three previous Division of Gynaecology
vaginal deliveries. The recipient had one in-vitro fertilisation cycle 4 months before transplant, which yielded (D Ejzenberg PhD,
eight cryopreserved blastocysts. J M Soares Jr PhD,
P C Serafini PhD,
Prof E Chada Baracat PhD) and
Findings The recipient showed satisfactory postoperative recovery and was discharged after 8 days’ observation Division of Obstetrics
in hospital. Immunosuppression was induced with prednisolone and thymoglobulin and continued via (R Pulcinelli Francisco PhD),
tacrolimus and mycophenalate mofetil (MMF), until 5 months post-transplantation, at which time azathioprine Department of Obstetrics and
replaced MMF. First menstruation occurred 37 days post-transplantation, and regularly (every 26–32 days) Gynaecology, Hospital das
Clínicas, University of
thereafter. Pregnancy occurred after the first single embryo transfer 7 months post-transplantation. No blood São Paulo School of Medicine,
flow velocity waveform abnormalities were detected by Doppler ultrasound of uterine arteries, fetal umbilical, or São Paulo, Brazil; and Digestive
middle cerebral arteries, nor any fetal growth impairments during pregnancy. No rejection episodes occurred Organs Transplant Division,
after transplantation or during gestation. Caesarean delivery occurred on Dec 15, 2017, near gestational week 36. Gastroenterology Department,
University of São Paulo School
The female baby weighed 2550 g at birth, appropriate for gestational age, with Apgar scores of 9 at 1 min, 10 at of Medicine FM-USP, São Paulo,
5 min, and 10 at 10 min, and along with the mother remains healthy and developing normally 7 months post Brazil (W Andraus PhD,
partum. The uterus was removed in the same surgical procedure as the livebirth and immunosuppressive L R Baratelli Carelli Mendes PhD,
therapy was suspended. L Ducatti MD, A Song PhD,
R Tanigawa MD,
V Rocha-Santos PhD,
Interpretation We describe, to our knowledge, the first case worldwide of livebirth following uterine transplantation R Macedo Arantes MD,
from a deceased donor in a patient with MRKH syndrome. The results establish proof-of-concept for treating uterine L Betocco de Paiva Haddad PhD,
Prof L A Carneiro
infertility by transplantation from a deceased donor, opening a path to healthy pregnancy for all women with uterine
D’Albuquerque PhD)
factor infertility, without need of living donors or live donor surgery.
Correspondence to:
Dr Dani Ejzenberg, Disciplina de
Funding Fundação de Amparo à Pesquisa do Estado de São Paulo and Hospital das Clínicas, University of São Paulo, Ginecologia, Departamento de
Brazil. Obstetrícia e Ginecologia,
Hospital das Clínicas, Faculdade
de Medicina da Universidade de
Copyright © 2018 Elsevier Ltd. All rights reserved. São Paulo, São Paulo 05403‑000,
Brazil
Introduction uteri.3,4 All previous successful livebirths were limited daejz@hotmail.com
Infertility affects 10–15% of couples of reproductive age. to patients with live donors. The use of dead donors
In this population, one in 500 women has absolute uterine would greatly broaden access to this treatment, but
infertility, with uterine agenesis (Mayer-Rokitansky- uterine transplantation in 2011 from a deceased donor
Küster-Hauser [MRKH] syndrome) or unexpected in Turkey led to an unsuccessful pregnancy and
hysterectomy, malformation, or sequelae of infection or miscarriage 2 years later5,6 despite apparent health of
surgery.1 Hysterectomy is the most common cause of the graft, casting uncertainty on feasibility of the use of
uterine infertility, while MRKH syndrome affects one in dead donors for patients with infertility.
4500 women.2 For these women, the only available options In our manuscript, we describe the first successful
to have a child were adoption or surrogacy, until the first case of livebirth after a uterine transplantation from a
childbirths following uterine transplantation from living deceased donor, by a collaborating team of transplantation
donors in Gothenburg, Sweden, in September, 2013.3 and endocrine reproductive specialists, on Sept 20, 2016,
Only one Swedish and one US centre have previously at the Hospital das Clínicas, University of São Paulo
published findings of livebirths from transplanted School of Medicine, Brazil.

www.thelancet.com Vol 392 December 22/29, 2018 2697

 Articles
Research in context
Evidence before this study
We did a systematic literature analysis using PubMed, LILACS,
and the Cochrane Database, limiting the search to
English-language publications on human patients, from the
year 1956 to May, 2018. In PubMed, our search for the terms
(“uterus” [Mesh]) AND (“transplantation” [Mesh]) yielded
2377 studies, including 1412 in patients. In LILACS, we found
18 articles using a search for terms “uterus AND transplantation
AND human”, but we classified them as irrelevant. We found
16 articles in the Cochrane Database via a search for the terms
“uterus AND transplantation AND human”, but we also found
these articles irrelevant. Despite advances in uterine
transplantation research in several countries, we found only
reports of births following uterine transplantation in the
nine-case series in Gothenburg (2014) and one in the USA,
in Dallas (2017). These births resulted from uterine
transplantations from live donors. A transplantation in Turkey
from a deceased donor (2011) yielded pregnancy, but gestation
ended in miscarriage.
Methods
Ethical details and participants
The procedure is part of a research protocol approved by
Ethics Committee of the School of Medicine (CAPPesq;
online register number 13122), by the National Ethics
Committee (CONEP; CAAE: 45068815.3.0000.0065), and
the Brazilian National Transplantation System (SNT;
1140/2016). The patient agreed to participate by informed
consent. The donor’s family consented to donation of
her heart, liver, and kidneys, and separately, to uterus
donation.
The recipient and compatible deceased donor were
matched solely by blood type (ABO system). Inclusion
criteria for recipient selection were: age 21–38 years;
body-mass index of less than 30; primary uterine
infertility; in a stable relationship for at least 2 years;
agreement with Terms of Informed Consent. Exclusion
criteria were: any conditions contraindicating pregnancy
(eg, serious cardiopathy, pulmonary hypertension,
decompensated autoimmune disease, or decompensated
type 1 diabetes); diminished ovarian reserves (follicle
stimulation hormone >10 IU during early follicular
phase or antral follicle count <7); presence of HIV,
hepatitis B or C, or HTLV I or II infection; or absence of
viable spermatozoa in the recipient’s partner assessed by
semen analysis.
The patient
A 32-year-old woman, married for 5 years and diagnosed
with MRKH syndrome. The recipient presented with
uterine agenesis, but no cardiac, renal, or bone structure
dys­
functions. Examinations listed below, including
antibodies and serology (table), served to ensure that she
was a suitable available uterus recipient.
2698 www.thelancet.com Vol 392 December 22/29, 2018
Added value of this study
To our knowledge, this is the first report of a livebirth after
uterus transplantation from a deceased donor. We also
observed viability of the explanted uterus after about 8 h
ischaemia, more than double that previously described for
those of live donors, and found that reducing postsurgical time
to embryo transfer by 5 months versus that reported in
previous live-donor cases yielded no rejection, while also
reducing costs and patient risk. We also suggest a modification
in the number of venous anastomoses that can be used in the
future to improve venous drainage.
Implications of all the available evidence
The results provide proof-of-concept for a new treatment
option for absolute uterine factor infertility, enabling
pregnancy even for those women without living uterine
donors—making uterine transplantation a therapeutic option
for those women who until now had only adoption or
surrogacy as options.
The angio-CT scan showed normal vessel anatomy
and prominent ovaries, with a follicle cyst in the right
ovary measuring 1·4 cm and elongated (3·5 cm) pelvic
structures, each with extremities neighbouring the
respective ovaries, corresponding to fibrosis in the
uterus site. Gynaecological examination showed a 5 cm
vagina with fibrosis in the third distal portion. In
April, 2016, the patient had in-vitro fertilisation in the
Human Reproduction Center of the Hospital das
Clínicas, yielding 16 eggs obtained from a single cycle,
and cryopreservation of eight good-quality blastocysts.
Before inclusion of the patient for the transplantation
procedure, social and psychological conditions were
assessed. Throughout the process, the couple received
monthly psychological counselling from professionals
specialised in transplants and fertility.
Donor selection and operation
Donor inclusion criteria were: age of at least 45 years, at
least one previous livebirth. The uterus was retrieved
from a 45-year-old multiple-organ (heart, liver, and
kidneys) donor of blood type O Rhesus positive,
diagnosed with Fischer 4 subarachnoid haemorrhage
that caused brain death. Based on her three vaginal
deliveries and no reported sexual disease, the donor was
accepted as a candidate for donation. She had received
high doses of vasoactive drugs (norepinephrine
0·39 µg/kg per min and vasopressin 10 mL/h) for 4 h.
Human leucocyte antigen matching was not performed.
All antibody assessments during the post-transplantation
period and pregnancy were negative.
Donor surgery commenced by abdominal–thoracic
midline incision. Heart, liver, kidneys, and uterus were
retrieved in that order. Uterus dissections before (warm)
 Articles

and after (cold) aortic clamping occurred last. The uterus

Result
was dissected along with its ligaments to suspend it after
implantation. We dissected and isolated ovarian veins, Treponema pallidum Negative

ureters, and the uterine arteries and veins, which did not HTLV 1/2 Negative
require separation from the distal ureters as needed in HIV 1/2 Negative
living donors. The vagina was sectioned at sufficient Herpes 1/2 Negative
length (2 cm) for anastomosis in the recipient. We Chagas Negative
cannulated the proximal common iliac arteries bilaterally Cytomegalovirus IgM negative and IgG positive
and closed the femoral arteries, for pelvic retention of Hepatitis A IgM negative and IgG positive
histidine-tryptophan-ketoglutarate preservation solution Hepatitis B Negative (anti-HBs and anti-HBc)
(HTK, Custodiol). We began perfusion before the other Hepatitis C Negative
transplantation teams, as ischaemia commenced with Blood type O Rh factor positive
distal aorta ligation. We used 1 L of preservation solution Table: Antibodies and serology of the recipient before uterine
through each uterine artery. Each uterine vessel branch transplant
was cut, retaining an internal iliac patch for anastomosis,
and the uterus was then immediately perfused with 1 L
of preserving liquid. The uterus weighed 225 g. Back-
table preparation of the uterus lasted 1·5 h (figure 1),
and great care was taken to dissect, recognise and test
small vessels, and complete bilateral salpingectomy and
oophorectomy.

Recipient operation and graft preparation

The recipient received an infraumbilical laparotomy.
Surgical time from iliac and pelvic vessel preparation to
uterus implantation was 2 h. Dissection of iliac vessels was
straightforward, but synchronisation between recipient
and donor surgeries was not ideal, causing a prolonged
time between retrieval of donor tissue and transplantation
into the recipient. Anastomosis of left and right uterine
veins and arteries lasted a total of 56 min (figure 2), and
vessels were revascularised con­ currently. Ovarian veins
were clamped during the procedure, and reflow was
permitted after the revascularisation of the uterus and
verification of good blood flow. After confirming their
substantial flow, we anastomised the respective external
iliac veins. After revascularisation of the uterus, careful
haemostasis of several small vessels was needed, which
took longer than 1 h to avoid large sutures and we used
bipolar cauterisers cautiously. The rudimentary uterus was
then opened to access the vaginal canal, and its vaginal
Figure 1: Uterus retrieved from 45-year-old dead donor after backtable preparation
stump (2 cm) sutured to the recipient’s vagina, guided by a
plastic tube, and the transplanted uterus set with rounded
and uterine–sacral ligaments. 1·5 mg/kg of thymoglobulin as induction therapy
Duration of cold ischaemia (from clamping of the aorta (figure 4). Immunosuppression therapy remained the
in the recipient until placement of the organ into the same until the last days of hospital stay, with weekly
recipient) was 6 h 20 min and warm ischaemia (from adjustment of tacrolimus trough concentrations to
placement of the uterus in the recipient cavity until 8–10 ng/mL, and mycophenolate mofetil (MMF; 720 mg
vascular anastomoses) was 1 h 30 min. The recipient every 12 h) until 5 months after uterus transplantation,
received 600 mL of Ringer lactate solution, produced when we replaced MMF with azathioprine 100 mg
800 mL of urine and lost 1200 mL of blood, mostly from per day and prednisone 10 mg per day until the end of
the transplanted organ. Total recipient operating time pregnancy.
was 10 h 30 min (figure 3). Antimicrobials administered included piperacillin-
tazobactam 4∙5 g every 6 hours and fluconazole 200 mg
Postoperative regimen twice a day, which were both given intraoperatively and
Immunosuppression followed the Swedish protocol,3,7 for the next 7 days; and ivermectin 200 µg/kg once a day
including 1 g intraoperative methylprednisolone and on postoperative days one and two, then 15 days later

www.thelancet.com Vol 392 December 22/29, 2018 2699

 Articles
External iliac vein
Uterine
artery
Vaginal anastomosis
Figure 2: Schematic drawing of uterine vessels anastomosis in external iliac veins and donor vaginal stump
sutured to the recipient’s vagina
repeated for 2 days, due to unavailability in Brazil of
Strongyloides serology. We administered 5 mg/kg per day
of prophylactic ganciclovir throughout the hospital
stay, and on discharge from hospital replaced it
with valganciclovir 900 mg once a day for 3 months.
Prophylactic sulfamethoxazole (400 mg) and trimeth­
oprim (80 mg) was given once a day from postsurgical
day one for 6 months.
The patient received 40 mg enoxaparin twice a day
subcutaneously and 100 mg aspirin (AAS) orally on
postsurgical day one. On postoperative day two she
received 60 mg enoxaparin once a day and 100 mg of
AAS, and thereafter enoxaparin 40 mg a day and 100 mg
AAS orally. Aspirin was suspended at 34 weeks after
gestation and heparin suspended 36 h before birth.
The patient stayed in an intensive care unit for 2 days,
then 6 days on a specialilsed transplant ward. She
was discharged after drug concentrations stabilised and
drugs could be administered orally.
Biopsies of the uterine cervix to detect rejection were
taken 1 week after uterus transplantation and weekly
during the first month, twice a month during the second
month, and monthly thereafter until gestation week 20.
Biopsies were resumed at gestation week 30, and
performed again at week 34. No alterations suggesting
rejection7 were found. Immunosuppressive therapy was
suspended after birth.
2700 www.thelancet.com Vol 392 December 22/29, 2018
In-vitro fertilisation
Before transplantation, the patient underwent one in-
vitro fertilisation cycle, with gonadotropin-releasing
External iliac artery
hormone agonist (Lupron-Abbott) administered after
confirming ovulation by ultrasonography, maintained for
14 days before initiating ovarian stimulation, and a total
of 1800 IU urinary follicle stimulating hormone and
luteal hormone (FSH+LH; MENOPUR-Ferring) for
8 days, yielding 16 matured oocytes by vaginal aspiration.
Fifteen oocytes were fertilised, which produced eight
good-quality blastocysts that were frozen by vitrification.8
Immunosuppression therapy was changed 5 months
after uterine transplantation to prepare for embryo
transfer. MMF was replaced by azathioprine 100 mg per
day to avoid teratogenic effects.
Ovarian vein
Embryo transfer was done with ultrasonographic
monitoring and hormone doses in the spontaneous
menstrual cycle (day 14 of menstrual cycle LH 17·1 IU/L,
Uterine FSH 4·4 IU/L, oestradiol 236·9 pg/mL, progesterone
vein 0·7 ng/mL) and endometrial thickness 8·6 mm. 7 months
after trans­plantation and 19 days after menstruation, one
embryo B3AB was transferred in a natural cycle
supplemented with natural micronised progesterone
400 mg vaginally starting 48 h after ovulation.8
Role of the funding source
Those funding the study had no role in study design,
data collection, analysis, or interpretation, or writing the
report. All authors had access to primary data and accept
responsibility for accuracy and completeness of data. The
corresponding author had access to all data and final
responsibility for the decision to submit for publication.
Results
The recipient’s first menstruation occurred 37 days after
transplantation, and her second occurred 26 days later.
After transplantation, she had a viral infection of the
upper respiratory tract (postoperative day 45), and a
vaginal secretion treated with secnidazole orally and
metronidazole vaginally on day 57. 5 months after
transplantation, the organ remained in good condition,
confirmed by Doppler ultrasound, regular menses, and
absence of rejection signs.
We intended to transfer an embryo to the uterus
6 months after transplantation, but after 14 days of
preparation the endometrial thickness was not above the
threshold of 7 mm required for embryo transfer, so we
postponed transfer for one cycle.
The first quantitative blood pregnancy test was
performed 10 days after embryo transfer (203·60 mIU/mL)
and repeated after 48 h (556·90 mIU/mL). 4 weeks after
embryo transfer, the gestational sac was identified in
sonographic examination with cardiac beats of 113 bpm
and crown–rump length of 0·33 cm, compatible with a
gestational age of 6 weeks. Non-invasive prenatal testing
was done at 10 weeks, showing a normal female fetus.
Ultrasound scans at 12 and 20 weeks revealed no
 Articles

Implanted uterus

Figure 3: Implanted uterus in recipient after reperfusion

Induction therapy Maintenance therapy

Intraop Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8
Thymoglobulin 1·5 mg/kg 60 mg
Methylprednisolone 1g 200 mg 160 mg 120 mg 80 mg 40 mg
Tacrolimus 3 mg every 12 h 4 mg every 12 h 5 mg every 12 h 6 mg every 12 h
Mycophenolate mofetil 720 mg every 12 h
Prednisone 20 mg 20 mg 20 mg

Figure 4: Immunosuppression strategy during the patient’s hospital stay

abnormalities in the fetal brain, face, spine, heart,
stomach, bowel, kidneys, or limbs.
Blood pressure remained normal throughout gestation.
Prenatal weight gain of the mother was 15 kg. Glucose
tolerance test was normal at gestation week 25. Mean
serum tacrolimus concentration measured from the
mother’s blood was 8–10 ng/mL during the first half and
6 ng/mL during the second half of gestation, until birth.
No graft rejection episodes were detected during the
prenatal period, confirmed by cervical biopsies.7 Cervix
length was 40 mm at 12 weeks, 39 mm at 19 weeks, and
30 mm at 34 weeks. During gestation week 32 the patient
presented signs of pyelonephritis and was admitted to
hospital for 48 h and treated with ceftriaxone for 10 days.
Caesarean delivery occurred at 35 weeks and 3 days,3
which is in line with the protocol used by Brännström
and colleagues.1 The patient was admitted to hospital
12 h before delivery. The newborn baby girl weighed
2550 g and measured 45 cm in length (figure 5). Apgar
score was 9 at 1 min, 10 at 5 min, and 10 at 10 min. The
infant required no glucose or ventilation support. She
stayed with her mother in the same room, and both were
discharged 3 days after birth. The transplanted uterus
was removed during caesarean section, and some
adhesions dissected. Implanted vessels were tied,
leaving a small patch in the recipient vessels to avoid
stenosis. The graft’s vaginal patch was removed and the
vaginal orifice closed. The uterine allograft weighed
915 g and showed usual gestational changes, with­
out evidence of rejection. The uterine arteries were
thickened by intimal fibrous hyperplasia, without lumen
obliteration (figure 6). The veins were un­remarkable.

www.thelancet.com Vol 392 December 22/29, 2018 2701

 Articles
Figure 5: Picture of neonate immediately after birth
1000 µm
Figure 6: Uterine artery with mild intimal fibrous hyperplasia, without medial calcification or inflammation
Haematoxylin and eosin (H&E) stain of the uterine artery.
The placenta weighed 405 g and showed no
abnormalities.
The mother and child had uneventful early follow-up.
The mother’s wound healed well. At the time of writing,
at age 7 months 20 days, the baby continued to breastfeed
and had normal growth parameters (7·2 kg in weight
and 62 cm in height).
Discussion
To our knowledge, this is the first livebirth reported
worldwide from a uterus transplanted from a deceased
2702 www.thelancet.com Vol 392 December 22/29, 2018
donor, and the first uterine transplantation in Latin
America.9 The success expands the prospects for
childbirth among women with infertility attributable to
uterine factors, by establishing proof-of-principle and
thus feasibility. Previous successful childbirths from
transplanted uteri were restricted to those from uteri of
living donors.10
The success of pregnancies resulting from such
transplantations in a Swedish series published in 20143
constituted a medical milestone in assisted reproduction
and transplantation, creating the possibility of childbirth
for many infertile women with access to suitable donors
and the needed medical facilities. Nevertheless, the
requirement of a live uterine donor is a major limitation,
as such donors are in short supply and often unavailable,
and the practice typically relies on availability of willing
and eligible family members or close friends. Therefore,
despite its complexities, our demonstration of successful
pregnancy following uterine transplantation from a
deceased donor enables use of a much wider potential
donor population, as the numbers of people willing and
committed to donate organs upon their own deaths are
far larger than those of potential live donors. Further
incidental but substantial benefits of the use of deceased
donors include lower costs and avoidance of live donors’
surgical risks.
Uterine transplantation is a recent advance, still
considered experimental in many countries.10 Our success
was enabled by important steps introduced by the Swedish
team to diminish complication risks, including stan­
dardisation of surgical technique, immuno­­sup­pression
protocol, and method for monitoring rejection episodes.
Nonetheless, due to unique features of the deceased donor
modality, we developed it as a new procedure that will
inevitably evolve with further refinements on a case-by-
case basis.
Management of total ischaemic time to the donor tissue
was a crucial issue in our case and is a key target for future
improvements. Ischaemia time was longer than planned,
because of the time needed for the preceding organ
retrieval (heart, liver, and kidney) that were necessarily
prioritised in the retrieval sequence, and coordination
issues between our transplantation team and the other
transplantation teams. Also, we performed six small-
vessel anastomoses, prolonging warm ischaemia time.
For young multiple-organ donors, in future the additional
cold ischaemia time needed to retrieve such prioritised
multiple organs can be anticipated, and the coordination
of teams managed accordingly. Other centres have
suggested measures to reduce cold ischaemia times such
as retrieving the uterus before other organs or referring
donors to the hospital where the uterine transplantation
will occur. These measures may help to optimise success
in cases using deceased donors.11,12
Our findings reveal that at least some uteri can
sur­vive almost 8 h ischaemia without definitive endo­
metrial damage, allowing successful transplantation and

pregnancy to term. The maximum cold ischaemia
time for uteri of dead donors that will allow successful
transplantation is unknown, but the uterus in our case
had an ischaemia time of 7 h 50 min. The previous
published maximum time with live donors is 3 h
25 min.1 Experimental findings in animals have
suggested that uteri may remain stable, and thus
potentially trans­ plantable, for up to 24 h,11 but for
humans no such data are yet available.
With only limited guidance available from few
relatively similar previous cases for this novel procedure,
by way of caution we used broad-ranging immuno­
suppression therapy, adapting a treatment scheme
similar to those we and others use in transplanting
moderately immunogenic organs such as the pancreas
and kidneys. Future experience might enable further
reduction of immuno­suppression.
One factor mitigating risks in patients undergoing
uterine transplantation, whether from dead or living
donors, is their generally healthy state, which diminishes
risk of perioperative complications compared with that
for recipients of other transplanted organs, who
invariably have severe underlying diseases.
To maximise venous flow during the procedure we
performed bilateral venous anastomoses, which gave
apparently robust blood outflow from the transplanted
organ. Our decision to do this procedure aimed to
minimise thrombosis risk, and we suggest considering
this procedure in cases when the ovarian veins show
abundant flow after reperfusion.
Substantial blood loss occurred following reperfusion
in this case. Such bleeding may perhaps have been
reduced had more small vessels around the internal iliac
vessels been tied, but doing so would have increased
warm dissection time, perhaps disrupting coordination
among the other solid organ teams. The bleeding seen
was high but manageable, and the uterus exhibited good
perfusion. Therefore, the need for multiple small sutures
resulted from our strategy to minimise ischaemia time.
On pathological examination of the uterus after
livebirth, the allograft showed mild and asymmetrically
distributed intimal fibrous hyperplasia of the uterine
arteries, probably reflecting the donor’s age. Other age-
related degenerative changes that occur in uterine
arteries include intimal proliferation, fibrosis, and
medial calcification (Mönckeberg sclerosis), but these
were absent. It is unlikely that allograft vasculopathy, a
form of chronic rejection and major cause of late graft
dysfunction that limits survival of cardiac and probably
renal and lung allografts, occurred in our case,
considering the absence of acute rejection episodes and
donor-specific antibodies.13
After each uterine transplantation from a live donor,
Brännström and colleagues waited a year before embryo
transfer. Their decision was empirical based on the
time needed for healing and to stabilise the immuno­
suppression.3 We aimed to shorten the time between
www.thelancet.com Vol 392 December 22/29, 2018 2703
Articles
the uterine transplant and embryo transfer, planning
to do so at 6 months, but then transferring at 7 months,
to permit reaching adequate endometrial thickness.
Advantages of the shorter total time of immuno­
suppressive therapy we used versus the longer times
used in previous participants include the lower total cost
and lower risk of adverse side-effects to the patient.
We performed the caesarean 4 days before 36 weeks
gestation, following Brännström and colleagues’ recom­
mendation to deliver at 35–36 weeks to avoid potential
risks to fetal vitality beyond that time. An ultrasound
scan the day before delivery showed an estimated fetal
weight of 2500 g (40th percentile on Hadlock curve) and
normal vitality.
Like most countries, Brazil has a high demand but a
short supply of solid organ donors. Brazil’s donor rate is
16 per million population per year and is increasing
annually. In its population of 220 million, this estimate
yields about 3500 donors per year. Therefore, the number
of potential uterus donors is high, but implementing wide
procurement and transplantation will require national
standardisation and establishing some new supporting
infrastructure—challenges that all nations aiming to
implement similar programmes must overcome.
In Brazil, surrogacy is permitted, yet it is difficult for
most women to find such a volunteer. The option of
uterine transplantation can thus expand fulfilment of
women’s desire to carry their own pregnancies. Given
the existence of well-established national systems in
many countries for regulating and distributing grafts for
transplantation, we predict that widespread adoption of
the deceased donor uterus transplant modality will prove
quite feasible and successful so that not only women
with willing donors in their families, but all women with
uterine factor infertility, can have a real option of a
healthy pregnancy.
Contributors
DE and WA both initiated the study, performed surgery, followed up the
patient, and wrote the report. DE did the in-vitro fertilisation.
LRBCM assisted patient selection and follow-up. LD performed surgery
and follow-up. AS followed up after surgery and during pregnancy.
RT did histological analyses and wrote the report. VR-S and RMA
performed surgery. JMS Jr, PCS, and LBdPH wrote the report.
RPF followed up during pregnancy and wrote the report. LACD’A and
ECB initiated the study and wrote the report.
Declaration of interests
We declare no competing interests.
Acknowledgments
The authors would like to thank the surgeons of the gastroenterology
transplantation group, the residents’ medical group, the human
reproduction centre team, and the prenatal and obstetrics team, and also
the Departments of Gynecology and of Gastroenterology for their
support. Financial support was provided by Fundação de Amparo à
Pesquisa do Estado de São Paulo (process number 2016/01223-1) and the
Hospital das Clínicas of University of São Paulo School of Medicine
(FM-USP).
References
1 Brännström M, Johannesson L, Dahm-Kähler P, et al. First clinical
uterus transplantation trial: a six-month report. Fertil Steril 2014;
101: 1228–36.
 Articles
2 Bagnoli VR, Fonseca AM, Fassolas G, Arie MHA, Arie WMY,
Baracat EC. Conduta frente às malformacões genitais uterinas:
revisão baseada em evidências. Femina 2010; 38: 217–28.
3 Brännström M, Johannesson L, Bokström H, et al. Livebirth after
uterus transplantation. Lancet 2015; 385: 607–16.
4 Testa G, McKenna GJ, Gunby RT Jr, et al. First live birth after uterus
transplantation in the United States. Am J Transplant 2018; 18: 1270–74.
5 Erman Akar M, Ozkan O, Aydinuraz B, et al. Clinical pregnancy
after uterus transplantation. Fertil Steril 2013; 100: 1358–63.
6 Flyckt R, Kotlyar A, Arian S, Eghtesad B, Falcone T, Tzakis A.
Deceased donor uterine transplantation. Fertil Steril 2017; 107: e13.
7 Mölne J, Broecker V, Ekberg J, Nilsson O, Dahm-Kähler P,
Brännström M. Monitoring of human uterus transplantation with
cervical biopsies: a provisional scoring system for rejection.
Am J Transplant 2017; 17: 1628–36.
8 Gardner DK, Lane M, Stevens J, Schlenker T, Schoolcraft WB.
Blastocyst score affects implantation and pregnancy outcome:
towards a single blastocyst transfer. Fertil Steril 2000; 73: 1155–58.
2704 www.thelancet.com Vol 392 December 22/29, 2018
9 Soares JM Jr, Ejzenberg D, Andraus W, D’Albuquerque LA,
Baracat EC. First Latin uterine transplantation: we can do it!
Clinics 2016; 71: 627–28.
10 Ejzenberg D, Mendes LR, Haddad LB, Baracat EC,
D’Albuquerque LA, Andraus W. Uterine transplantation:
a systematic review. Clinics 2016; 71: 679–83.
11 Tricard J, Ponsonnard S, Tholance Y, et al. Uterus tolerance to
extended cold ischemic storage after auto-transplantation in ewes.
Eur J Obstet Gynecol Reprod Biol 2017; 214: 162–67.
12 Testa G, Anthony T, McKenna GJ, et al. Deceased donor uterus
retrieval: a novel technique and workflow. Am J Transplant 2018;
18: 679–83.
13 Merola J, Jane-Wit DD, Pober JS. Recent advances in allograft
vasculopathy. Curr Opin Organ Transplant 2017; 22: 1–7.