Articles

Frozen versus fresh single blastocyst transfer in ovulatory

women: a multicentre, randomised controlled trial
Daimin Wei*, Jia-Yin Liu*, Yun Sun*, Yuhua Shi*, Bo Zhang*, Jian-Qiao Liu, Jichun Tan, Xiaoyan Liang, Yunxia Cao, Ze Wang, Yingying Qin,
Han Zhao, Yi Zhou, Haiqin Ren, Guimin Hao, Xiufeng Ling, Junzhao Zhao, Yunshan Zhang, Xiujuan Qi, Lin Zhang, Xiaohui Deng, Xiaoli Chen,
Yimin Zhu, Xiaohong Wang, Li-Feng Tian, Qun Lv, Xiang Ma, Heping Zhang, Richard S Legro, Zi-Jiang Chen

Summary
Background Elective single embryo transfer (eSET) has been increasingly advocated, but concerns about the lower Published Online
pregnancy rate after reducing the number of embryos transferred have encouraged transfer of multiple embryos. February 28, 2019
http://dx.doi.org/10.1016/
Extended embryo culture combined with electively freezing all embryos and undertaking a deferred frozen embryo S0140-6736(18)32843-5
transfer might increase pregnancy rate after eSET. We aimed to establish whether elective frozen single blastocyst
See Online/Comment
transfer improved singleton livebirth rate compared with fresh single blastocyst transfer. http://dx.doi.org/10.1016/
S0140-6736(19)30426-X
Methods This multicentre, non-blinded, randomised controlled trial was undertaken in 21 academic fertility centres *These authors contributed
in China. 1650 women with regular menstrual cycles undergoing their first cycle of in-vitro fertilisation were enrolled equally to this Article
from Aug 1, 2016, to June 3, 2017. Eligible women were randomly assigned to either fresh or frozen single blastocyst Center for Reproductive
transfer. The randomisation sequence was computer generated, with block sizes of two, four, or six, stratified by study Medicine, Cheeloo College of
Medicine, Shandong
site. For those assigned to frozen blastocyst transfer, all blastocysts were cryopreserved and a delayed frozen-thawed University, Jinan, China
single blastocyst transfer was done. The primary outcome was singleton livebirth rate. Analysis was by intention to (D Wei MD, Prof Y Shi MD,
treat. This trial is registered at the Chinese Clinical Trial Registry, number ChiCTR-IOR-14005405. Z Wang MS, Prof Y Qin MD,
Prof H Zhao MD,
Prof Z-J Chen MD); The Key
Findings 825 women were assigned to each group and included in analyses. Frozen single blastocyst transfer resulted Laboratory of Reproductive
in higher rates of singleton livebirth than did fresh single blastocyst transfer (416 [50%] vs 329 [40%]; relative risk Endocrinology of Ministry of
[RR] 1·26, 95% CI 1·14–1·41, p<0·0001). The risks of moderate or severe ovarian hyperstimulation syndrome (four of Education, Jinan, China (D Wei,
Prof Y Shi, Z Wang, Prof Y Qin,
825 [0·5%] in frozen single blastocyst transfer vs nine of 825 [1·1%] in fresh single blastocyst transfer; p=0·16),
Prof H Zhao, Prof Z-J Chen);
pregnancy loss (134 of 583 [23·0%] vs 124 of 481 [25·8%]; p=0·29), other obstetric complications, and neonatal National Research Center for
morbidity were similar between the two groups. Frozen single blastocyst transfer was associated with a higher risk of Assisted Reproductive
pre-eclampsia (16 of 512 [3·1%] vs four of 401 [1·0%]; RR 3·13, 95% CI 1·06–9·30, p=0·029). Technology and Reproductive
Genetics, Jinan, China (D Wei,
Prof Y Shi, Z Wang, Prof Y Qin,
Interpretation Frozen single blastocyst transfer resulted in a higher singleton livebirth rate than did fresh single Prof H Zhao, Prof Z-J Chen);
blastocyst transfer in ovulatory women with good prognosis. The increased risk of pre-eclampsia after frozen Department of Reproductive
blastocyst transfer warrants further studies. Medicine, First Affiliated
Hospital of Nanjing Medical
University/Jiangsu Province
Funding The National Key Research and Development Program of China. Hospital, Nanjing, China
(X Ma MD, Prof J-Y Liu MD);
Copyright © 2019 Elsevier Ltd. All rights reserved. State Key Laboratory of
Reproductive Medicine,
Nanjing, China (X Ma,
Introduction embryo development increases in women seeking a Prof J-Y Liu); Center for
Studies have shown that elective single embryo transfer blastocyst-stage embryo transfer compared with women Reproductive Medicine, Ren Ji
(eSET) is the most efficient approach to reduce the risk of with a cleavage-stage embryo transfer,6 especially in those Hospital, Shanghai, China
(Prof Y Sun MD, Prof Z-J Chen);
multiple gestations and their associated risks to mothers with few day-3 cleavage-stage embryos and those with School of Medicine, Shanghai
and children after in-vitro fertilisation (IVF).1 Despite poor prognosis. Some cleavage-stage embryos that do Jiao Tong University, Shanghai,
strong advocacy for its universal adoption,2 its widespread not survive prolonged in-vitro culture, however, may China (Prof Y Sun, Prof Z-J Chen);
uptake is slow because of concerns about the lower continue to develop into viable pregnancies if they are Shanghai Key Laboratory of
Assisted Reproduction and
pregnancy rate after reducing the number of embryos transferred into the uterus on day 3,7 because in-vitro Reproductive Genetics,
transferred.3 Many efforts had been made to improve the culture condition differs from the in-vivo environment. Shanghai, China
selection of a single embryo that is likely to implant and As a result, single blastocyst transfer strategy is recom­ (Prof Y Sun, Prof Z-J Chen);
thus to increase pregnancy rate after eSET. mended primarily to women with a good prognosis,8 who Center for Reproductive
Medicine, Maternal and Child
Extending embryo culture to blastocyst from cleavage will probably have more cleavage-stage embryos available Health Hospital in Guangxi,
stage allows for better evaluation of the implantation for extended culture. Guangxi, China
potential of the embryo.4 The implantation rate is higher In addition to embryo selection, improving the peri- (Prof B Zhang MD); Department
after blastocyst transfer than after cleavage-stage embryo implantation uterine environment could also contribute of Reproductive Medicine, Key
Laboratory for Major Obstetric
transfer during fresh embryo transfer cycles.5 However, to better success rates after single embryo transfer. The Diseases of Guangdong
there are drawbacks to blastocyst culture. The failure to supra-physiological level of administered gonadotropins Province, and Key Laboratory
reach an embryo transfer because of poor or arrested or resultant increase in steroid hormones after ovarian for Reproduction and Genetics

www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5 1

 Articles
of Guangdong Higher
Education Institutes, the Third
Affiliated Hospital of
Guangzhou Medical University,
Guangzhou, China
(Prof J-Q Liu MD); Reproductive
Medicine Center, Department
of Obstetrics and Gynecology,
Shengjing Hospital, China
Medical University, Shenyang,
China (Prof J Tan PhD);
Reproductive Medicine
Research Centre, the
6th Affiliated Hospital of
Sun Yat-sen University,
Guangzhou, China
(Prof X Liang MD); Department
of Obstetrics and Gynecology,
Reproductive Medicine Center,
The First Affiliated Hospital,
Anhui Medical University,
Hefei, China (Prof Y Cao MD);
Center for Reproductive
Medicine, Qingdao Women’s
and Children’s Hospital,
Qingdao University, Qingdao,
China (Y Zhou MD);
Department of Reproductive
Medicine, Shenyang Dongfang
Jinghua Hospital, Shenyang,
China (H Ren MD); Department
of Reproductive Medicine,
the Second Hospital of Hebei
Medical University,
Shijiazhuang, China
(Prof G Hao MD); Department
of Reproductive Medicine, the
affiliated Obstetrics and
Gynecology Hospital with
Nanjing Medical University,
Nanjing Maternity and Child
Health Care Hospital, Nanjing,
China (Prof X Ling MD);
Reproductive Medical Center,
the Second Affiliated Hospital
of Wenzhou Medical College
and Yuying Children’s hospital,
Wenzhou, China
(Prof J Zhao MD); Center for
Reproductive Medicine, Tianjin
Central Hospital of Obstetrics
and Gynecology, Tianjin, China
(Prof Y Zhang MD); Center for
Reproductive Medicine,
the Affiliated Hospital of
Qingdao University, Qingdao,
China (Prof X Qi MD);
Department of Occupational
Hygiene, School of Public
Health and Management,
Weifang Medical University,
Weifang, China (L Zhang PhD);
Center for Reproductive
Medicine, Qilu Hospital of
Shandong University, Jinan,
China (Prof X Deng MD);
Division of Reproductive
Medicine, Department of
Obstetrics and Gynecology,
Sun Yat-Sen Memorial
Hospital, Sun Yat-Sen
2 www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5
Research in context
Evidence before this study
We searched Pubmed and Cochrane Library from database
inception to May 1, 2018, with the keywords “frozen embryo”
OR “frozen-thawed cycle” OR “cryopreservation” OR
“vitrification” OR “freeze all” AND “fresh embryo”.
We identified one Cochrane systematic review published in
2017, and five additional randomised trials that found
conflicting results. The Cochrane review reported four
randomised trials comparing fresh embryo transfer versus
elective frozen embryo transfer. The authors concluded that
frozen embryo transfer resulted in lower rates of miscarriage
and ovarian hyperstimulation syndrome (OHSS), but a higher
rate of pregnancy complications. No difference in the
cumulative livebirth rate (based on subsequent embryo
transfers of embryos cryopreserved from the study cycle of
ovarian stimulation) was found. There was great heterogeneity
among the trials included in the Cochrane review in terms of
study populations, developmental stages of the transferred
embryos, freezing methods, and the number of embryos
transferred. The result was dominated by the trial undertaken in
women with polycystic ovary syndrome (PCOS). Subsequently,
two large randomised trials were undertaken in ovulatory
women with cleavage-stage embryo transfer with consistent
results showing that elective frozen embryo transfer led to
similar rates of pregnancy, pregnancy loss, and livebirth
compared with fresh embryo transfer. Another randomised trial
compared frozen versus fresh euploid blastocyst transfer after
preimplantation genetic screening and found higher rates of
pregnancy and livebirth after frozen embryo transfer. The risk
of obstetric complications was not reported. There were two
other trials that were respectively undertaken in women with
gonadotropin releasing hormone (GnRH) antagonist regimen
and GnRH agonist trigger for ovarian stimulation and in
women with elevated progesterone on the day of triggering;
results showed no significant difference in the rates of
stimulation might adversely affect the endometrial de­
velop­ment.9 The endometrium after ovarian stimulation
has been shown to exhibit histological advancement,
alteration in gene expression, and structural abnor­
malities.10 With the development in cryopreservation
technology, embryos could be more safely frozen and
preserved for later use.11 Elective frozen embryo transfer
avoids the exposure of the endometrium to the adverse
sequelae of ovarian stimulation and has been shown to
result in a higher rate of livebirth than has fresh embryo
transfer in women with polycystic ovary syndrome
(PCOS).12 However, in ovulatory women, frozen embryo
transfer seemed to be as efficient and as safe as fresh
embryo transfer to achieve a livebirth.13,14 In previous
trials, embryo transfer was done at cleavage stage and up
to two embryos were transferred; the proportion of
multiple pregnancies was high at approximately 30%,
leading to increased maternal and fetal morbidity.13,14 In
pregnancy and livebirth. In all these trials, up to two embryos
were transferred in both the fresh and frozen embryo transfer
groups, leading to higher rates of multiple pregnancies and
their associated perinatal morbidity. Whether frozen single
blastocyst transfer could improve singleton livebirth rate
compared with fresh single blastocyst transfer remained to
be determined.
Added value of this study
In this multicentre randomised trial, 1650 ovulatory women
with good prognosis from 21 fertility centres in China were
randomly assigned to undergo either a frozen single blastocyst
transfer or a fresh single blastocyst transfer. Frozen single
blastocyst transfer resulted in a higher rate of singleton
livebirth attributed to a higher rate of implantation than did
fresh single blastocyst transfer. Frozen single blastocyst transfer
also led to a higher singleton birthweight, which was
accompanied by a higher risk of pre-eclampsia. The risks of
OHSS, pregnancy loss, and other obstetric complications
including preterm delivery and congenital anomalies were
similar after frozen and fresh single blastocyst transfer.
Implications of all the available evidence
A strategy to transfer a single frozen blastocyst versus
two cleavage-stage embryos results in a marked decrease in
twin livebirth rates with a comparable overall livebirth rate.
The available evidence on so-called freeze-all strategy
suggested the risk–benefit ratio of elective frozen embryo
transfer was influenced by several factors, including the patient
diagnosis and the stage of embryo transferred. Elective frozen
embryo transfer seems a better choice to achieve livebirth for
women with PCOS, women with a higher risk of OHSS, and
women with good prognosis who are planning to undergo
single blastocyst transfer. However, its potential for increased
maternal pre-eclampisa, as well as the long-term effects on
offspring, warrant further studies.
the past 5 years, with the refinement of techniques for
blastocyst culture and in compliance with the guidelines
for reducing the risk of multiple pregnancies,15 the
application of single blastocyst transfer has become
increasingly popular. Frozen single blastocyst transfer
could optimise pregnancy rates and maintain perinatal
safety compared with fresh single blastocyst transfer.
In this randomised trial, we compared pregnancy
outcomes and obstetric and perinatal complications
after frozen versus fresh single blastocyst transfer.
Methods
Study design and participants
This study was a non-blinded, multicentre, randomised
controlled trial undertaken in 21 academic fertility centres
in China. The trial was approved by the ethics committees
of all study sites. All the couples including female and
male partners gave written informed consent. A data and

safety monitoring board was established to oversee the
study. We have previously published the protocol.16
This trial included women with regular menses who
were undergoing the first cycle of IVF with or without
intra­cytoplasmic sperm injection with an indication of
tubal, male, or unexplained infertility. Eligible women
were aged 20–35 years, and had a menstrual cycle length
of 21–35 days indicative of regular ovulation. Women who
were planning cycles of preimplantation genetic testing
were excluded from this study, as were those with a
diagnosis of a congenital or acquired uterine abnormality
(such as a uterine malformation, adenomyosis, sub­
mucous myoma, or intrauterine adhesion). We also
excluded women with medical conditions that are
contraindications to IVF procedures or pregnancy, such as
uncontrolled hypertension, known symptomatic heart
disease, poorly controlled type 1 or type 2 diabetes,
undiagnosed liver disease or dysfunction, renal disease,
severe anaemia, history of deep venous thrombosis,
history of pulmonary embolus, previous cerebrovascular
accident, or history of cervical cancer, endometrial cancer,
or breast cancer.
Randomisation and masking
The randomisation sequence was computer generated by
statisticians in the data coordinating centre in Shandong
University. Blocked randomisation was done with dy­
namic block sizes of two, four, or six and was stratified by
study site. This sequence was entered into the central
online database, which was secured by the username and
password login.
Randomisation was done on the third day after oocyte
retrieval—ie, day 3 of embryo culture. We chose this point
in the IVF cycle for randomisation to ensure comparable
ovarian stimulation between groups in this non-blinded
trial and to minimise exclusions or crossovers after
randomisation due to a low number of embryos or poor
embryo development. Women who had already planned
to undergo frozen embryo transfer before the day of
randomisation at the discretion of local physicians
because of hydrosalpinx, premature elevation of proges­
terone, or a high risk of ovarian hyperstimulation
syndrome (OHSS) were excluded. Women were randomly
assigned to either fresh single blastocyst transfer group
or frozen single blastocyst transfer group by 1:1 ratio.
Only women who had four or more high-grade embryos
on day 3 of embryo culture were randomly assigned.
Procedures
All participants were given gonadotropin releasing
hormone (GnRH) antagonist regimen for ovarian stimu­
lation. Recombinant follicle-stimulating hormone (rFSH,
PUREGON; MSD Organon, Oss, Netherlands) was
started on day 1–3 of menstrual cycle. The dose adjustment
of gonadotropin and the initiation of GnRH antagonist
(ganirelix, MSD Organon, Oss, Netherlands) were done as
in our previous report.12,13 When at least two follicles were
www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5 3
Articles
18 mm or greater in mean diameter, human chorionic University, Guangzhou, China
gonadotropin (hCG) at a dose of 4000–10 000 IU was (X Chen MD); Department of
Reproductive Endocrinology,
administered to induce the final maturation of oocytes. Women’s Hospital, School of
Oocyte retrieval was done 34–36 h after hCG injection by Medicine, Zhejiang University,
experienced physicians. Luteal phase support was started Hangzhou, China
from the day of oocyte retrieval with vaginal progesterone (Prof Y Zhu MD); Reproductive
Medical Center, Tangdu
gel (Crinone, Merck Serono, Watford, UK) 90 mg per day Hospital, the Fourth Military
and oral dydrogesterone (Duphaston, Abbott, OLST, Medical University, Xi’an, China
Netherlands) 10 mg twice daily. (Prof X Wang MD); Reproductive
On day 3 of embryo culture, embryos were graded by Medical Center, Jiangxi
Provincial Maternal and Child
morphological criteria on the basis of the number and Health Hospital, Nanchang,
size of blastomere and the percentage of fragmentation.17 China (L-F Tian MS); Center for
Women who had four or more high-grade embryos with Reproductive Medicine,
scores of three or four were randomly assigned to the Sichuan Provincial People’s
Hospital, Chengdu, China
fresh or frozen blastocyst transfer group.17 Blastocyst (Prof Q Lv MS); Department of
culture was done with sequential media in all centres. Biostatistics, Yale University
On day 3, embryos were removed from cleavage media School of Public Health,
New Haven, CT, USA
and replaced in blastocyst media. The embryo score
(Prof H Zhang PhD); and
on day 5 was assessed according to Gardner morpho­ Department of Obstetrics and
logical criteria,18 on the basis of the degree of expansion Gynecology, Penn State College
and the development of the inner cell mass and of Medicine, Hershey, PA, USA
(Prof R S Legro MD)
trophectoderm.
For the fresh blastocyst transfer group, a single blasto­ Correspondence to:
Prof Zi-Jiang Chen, Center for
cyst was selected and transferred on day 5 of embryo Reproductive Medicine, Cheeloo
culture (details of embryo transfer procedure are provided College of Medicine, Shandong
in the appendix). The selection of the single blastocyst University, Jinan 250021, China
gave priority to the score of the inner cell mass, and the chenzijiang@hotmail.com
score of trophectoderm was also considered—ie, the rank See Online for appendix
of blastocyst grade from top to good was AA, AB, BA, BB,
AC, and BC (details of blastocyst score are shown in
the appendix). If two or more blastocysts were of equal
grade, their early scores at cleavage stage were referred
for the selection of the single blastocyst. Supernumerary
embryos were frozen on day 5 or 6 according to embryo
development. If pregnancy was achieved after fresh single
blastocyst transfer, luteal phase support was continued
until 10 weeks’ gestation.
For the frozen blastocyst transfer group, all blastocysts
were vitrified on day 5 or day 6 according to embryo
development. Luteal phase support was stopped after
randomisation. At least 4 weeks later, the endometrium
was prepared either with a natural cycle regimen or a
programmed cycle regimen, at the discretion of local
investigators. For the natural ovulatory cycle regimen,
ovulation was determined by ultrasound monitoring.
Oral dydrogesterone (Duphaston, Abbott, OLST,
Netherlands) 10 mg three times daily was administered
for luteal phase support after ovulation. A single frozen-
thawed blastocyst was transferred on the 5th day after
ovulation. If pregnancy was achieved after frozen
blastocyst transfer, luteal phase support was continued
until 10 weeks’ gestation. For the programmed cycle
regimen, oral oestradiol valerate (Progynova, Delpharm
Lille, Lys-Lez-Lannoy, France) at a dose of 4–8 mg daily
was started on day 1–3 of menstrual cycle. Vaginal
progesterone gel (Crinone, Merck Serono, Watford, UK)
90 mg per day and oral dydrogesterone 10 mg twice daily
 Articles

2708 patients assessed for eligibility

1058 ineligible
734 did not meet inclusion criteria
672 less than four embryos with top or good score on day 3
32 cancelled cycles due to poor ovarian response
11 with irregular menstrual cycle
10 previous IVF cycles
7 age ≥35 years or <20 years
2 IVM cycle
324 met exclusion criteria
6 with uterine abnormality
9 abnormal karyotype in patients or partners
5 oocyte cryopreservation
3 did not start cycle because of natural pregnancy
82 at high risk of OHSS
93 with elevated progesterone or thin endometrium
54 unable to comply with the study protocol
72 withdrew consent

1650 randomly assigned

825 assigned to fresh embryo transfer 825 assigned to frozen embryo transfer

703 adhered to protocol
685 had D5-blastocyst transfer
18 had D6-blastocyst transfer
122 had protocol deviation
13 did not undergo embryo transfer
98 transferred frozen embryo
5 with D3 (one embryo [n=1],
two embryos [n=4])
73 with D5 (one embryo [n=67],
two embryos [n=6])
20 with one D6 embryo
11 transferred two fresh embryos
8 with D3 embryos
3 with D5 embryos
724 adhered to protocol 101 had protocol deviation
668 had D5-blastocyst transfer 29 did not undergo embryo transfer
56 had D6-blastocyst transfer 37 transferred fresh embryo
7 with D3 (one embryo [n=2],
two embryos [n=5]
29 with D5 (one embryo [n=27],
two embryos [n=2])
1 with one D6 embryo
35 transferred two frozen embryos
9 with D3 embryos
20 with D5 embryos
6 with D6 embryos

3 lost to follow-up 2 lost to follow-up

276 delivered live infants 65 delivered live infants 403 delivered live infants 36 delivered live infants

Figure: Trial profile

IVF=in-vitro fertilisation. IVM=in-vitro maturation. OHSS=ovarian hyperstimulation syndrome.

were added when the endometrial thickness reached Outcomes

7 mm or more. A single frozen-thawed blastocyst was
transferred on the 5th day after progesterone initiation.
If pregnancy was achieved, oral oestradiol valerate was
continued until 8 weeks’ gestation, and vaginal
progesterone gel and oral dydrogesterone were
continued until 10 weeks’ gestation. The selection of the
frozen blastocyst for thawing was based on the blastocyst
grade before freezing with the same rule as for the fresh
blastocyst transfer group. If the first thawed blastocyst
did not survive, a second blastocyst was thawed and
transferred.
The reproductive outcomes of the subsequent frozen
embryo transfer within 12 months after the unsuccessful
first transfer were also followed up.
The primary outcome was singleton livebirth. Secondary
outcomes were rates of conception, clinical pregnancy,
ongoing pregnancy, pregnancy loss, livebirth, moderate
and severe OHSS, ectopic pregnancy, pregnancy and
perinatal complications, neonatal complication and other
adverse events, and birthweight (detailed definitions are
provided in the appendix). We did a post-hoc analysis for
the outcomes of small for gestational age (SGA), large
for gestational age (LGA), the rate of cumulative live­
birth within 12 months after the first embryo transfer,
the number of embryos remaining, and time to live­
birth. The determination of SGA and LGA was based on
the birthweight reference for Chinese populations
adjusted for sex and gestational age.19 SGA was defined

4 www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5

 Articles

Frozen embryo Fresh embryo Frozen embryo Fresh embryo p value

transfer (n=825) transfer (n=825) transfer (n=825) transfer (n=825)
Age (years) 28·8 (3·0) 28·8 (3·0) Days of ovarian stimulation 9·3 (1·5) 9·3 (1·5) 0·33
Body-mass index (kg/m²) 22·4 (3·2) 22·5 (3·1) Total gonadotropin dose (IU) 1604·4 (501·1) 1598·6 (480·0) 0·81
Duration of infertility (years) 3·2 (2·1) 3·3 (2·3) Oestradiol level on HCG trigger day (pmol/L)* 12 419 (6254) 12 341 (6467) 0·81
Previous conception 363 (44·0%) 338 (41·0%) Progesterone level on HCG trigger day (nmol/L)† 3·7 (1·7) 3·7 (1·7) 0·77
Indications for IVF Endometrial thickness on HCG trigger day (mm) 10·6 (2·0) 10·6 (2·0) 0·63
Tubal factor 460 (55·8%) 443 (53·7%) Number of oocytes retrieved 14·0 (5·6) 13·8 (5·7) 0·50
Male factor 173 (21·0%) 191 (23·2%) Number of high-score embryos on day 3‡ 7·1 (3·3) 6·8 (2·9) 0·08
Unexplained infertility* 43 (5·2%) 33 (4·0%) Regimen of endometrial preparation for frozen embryo transfer
Combined factors 142 (17·2%) 155 (18·8%) Natural cycles 474/759 (62·5%) ·· ··
Others 7 (0·8%) 3 (0·4%) Programmed cycles 276/759 (36·4%) ·· ··
Antral follicle count in both 16·4 (5·5) 16·0 (5·1) Minimal ovarian stimulation cycles 9/759 (1·2%) ·· ··
ovaries† Stage of embryo transferred 0·54
Baseline sex hormone Blastocyst transfer 780/796 (98·0%) 799/812 (98·4%) ··
FSH (IU/L) 6·4 (1·5) 6·4 (1·5) Cleavage-stage embryo transfer 16/796 (2·0%) 13/812 (1·6%) ··
LH (IU/L) 4·8 (2·0) 4·8 (2·2) Mean number of embryos transferred 1·05 (0·22) 1·03 (0·16) 0·0056
Oestradiol (pmol/L) 138·6 (56·4) 139·7 (61·3) One embryo transferred 754/796 (94·7%) 791/812 (97·4%) 0·0054
Total testosterone (nmol/L)‡ 0·9 (0·5) 0·9 (0·5) Two embryos transferred 42/796 (5·3%) 21/812 (2·6%) ··
Data are mean (SD) or n (%). FSH=follicle-stimulating hormone. LH=luteinising Number of patients who did not undergo 29/825 (3·5%) 13/825 (1·6%) 0·012
hormone. IVF=in-vitro fertilisation. *Unexplained infertility is rarely used as a embryo transfer
diagnosis indicator for IVF in China because it is a controversial diagnosis. †Antral No embryo obtained 10 (1·2%) 9 (1·1%) ··
follicle count was missing for ten patients in the frozen embryo transfer group Personal issue 10 (1·2%) 3 (0·4%) ··
and for 17 patients in the fresh embryo transfer group. ‡Total testosterone was
missing for 28 patients in the frozen embryo transfer group and for 23 patients in Natural conception after oocyte retrieval 9 (1·1%) 1 (0·1%) ··
the fresh embryo transfer group.
Data are mean (SD), n/N (%), or n (%). HCG= human chorionic gonadotropin. *Number of observations was 725 in the
Table 1: Baseline characteristics of the intention-to-treat population frozen embryo transfer group and 735 in the fresh embryo transfer group. †Number of observations was 823 in the frozen
embryo transfer group and 820 in the fresh embryo transfer group ‡Number of observations was 811 in the frozen
embryo transfer group and 811 in the fresh embryo transfer group.

as birth­weight lower than the 10th percentile of
referential birthweight. LGA was defined as birthweight
higher than the 90th percentile of referential birthweight.
Statistical analysis
The livebirth rate after single fresh blastocyst transfer in
women younger than 35 years was about 50% in our
retrospective clinical database. We assumed that an
absolute difference of 10% in livebirth rate was of clinical
significance and thus aimed to test a difference of 10% of
livebirth rate between treatment groups at a significance
level of 0·01 with statistical power of 90%. The minimal
sample size calculated was 735 for each group. In con­
sideration of a dropout rate of 10%, we planned to enrol
817 women in each group.
The primary outcome was analysed according to the
intention-to-treat principle. The difference in the prima­ry
outcome—ie, singleton livebirth rate—between the
two treat­ment groups was analysed by the Pearson χ² test.
The relative risk and 95% CIs were calculated. The
between-group differences in secondary outcomes were
compared with the Pearson χ² test. The mean birthweight
was compared by the Student’s t test. Secondary per-
protocol and per-treatment analyses were done among
those who adhered to the protocols and according to the
actual treatment that participants received, respectively.
Continuous data were expressed as mean (SD), and
between-group differences were tested by the Wilcoxon
Table 2: Outcomes of ovarian stimulation and embryo culture and transfer
rank sum test. Categorical data were represented as
frequency and percentage; differences in these variables
between the treatment groups were assessed by χ² ana­
lysis, with Fisher’s exact test for expected frequencies less
than five.
We performed post-hoc subgroup analyses based on
the concentrations of oestradiol and progesterone on the
day of hCG administration and according to the cycle
regimens of endometrial preparation for frozen embryo
transfer. All analyses were done with SAS software
(version 9.4).
This trial is registered at the Chinese Clinical Trial
Registry, number ChiCTR-IOR-14005405.
Role of the funding source
The sponsors of the study had no role in study
design, data collection, data analysis, data interpretation,
or writing of the report. The corresponding author
had full access to all the data in the study and had
final responsibility for the decision to submit for
publication.
Results
Recruitment was done between Aug 1, 2016, and
June 3, 2017. 1650 women were included and randomly

www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5 5

 Articles

796 women in the frozen embryo trans­fer group and 21

Frozen embryo Fresh embryo Relative risk in p value
transfer group transfer group frozen embryo (3%) of 812 in the fresh embryo transfer group had a
(n=825) (n=825) group (95% CI) transfer of two embryos (p=0·0054; table 2). More than
Livebirth 98% of women in both groups had the embryo transfer at
Singleton livebirth per woman 416 (50·4%) 329 (39·9%) 1·26 (1·14–1·41) <0·0001 blastocyst stage according to the protocol, and the other
Twin livebirth per woman 23 (2·8%) 12 (1·5%) 1·92 (0·96–3·83) 0·0602
2% of women in each group had embryo transfer at
Total livebirth per woman 439 (53·2%) 341 (41·3%) 1·29 (1·16–1·43) <0·0001
the cleavage stage (table 2). The overall proportion of
Birthweight*
deviations from the protocol was similar between the
frozen (101 of 825 [12%]) and fresh (122 of 825 [15%])
Singleton (g) 3407·9 (476·2)† 3293·1 (513·5) ·· 0·0018
embryo transfer groups (p=0·13; figure). Four participants
Twin (g) 2544·8 (468·9) 2523·8 (472·7) ·· 0·86
assigned to the frozen single blastocyst transfer group and
Gestational weeks (week) 38·9 (1·7) 38·8 (1·9) ·· 0·41
two assigned to the fresh single blastocyst transfer group
Pregnancy
but who received a frozen single blastocyst did not have a
Conception per woman‡ 583 (70·7%) 481 (58·3%) 1·21 (1·13–1·30) <0·0001
viable embryo after thawing of the first blastocyst and had
Clinical pregnancy per 512 (62·1%) 401 (48·6%) 1·28 (1·17–1·39) <0·0001
woman§
a second blastocyst thawed and transferred.
Singleton pregnancy 491 (59·5%) 395 (47·9%) 1·24 (1·14–1·36) <0·0001
416 (50·4%) of 825 women in the frozen single blastocyst
Twin pregnancies¶ 21 (2·5%) 6 (0·7%) 3·50 (1·42–8·63) 0·0036
transfer group had a singleton livebirth, which was higher
Monozygotic twin 19 (2·3%) 14 (1·7%) 1·36 (0·69–2·69) 0·38
than in the fresh singleton blastocyst trans­ fer group
pregnancies|| (329 of 825 [39·9%]; relative risk 1·26, 95% CI 1·14–1·41;
Implantation per embryo** 524/838 (62·5%) 406/833 (48·7%) 1·28 (1·18–1·40) <0·0001 table 3). The total livebirth rate including singleton and
Ongoing pregnancy per 458 (55·5%) 355 (43·0%) 1·29 (1·17–1·43) <0·0001 twin was also higher in the frozen single blastocyst
woman†† transfer group (table 3). Frozen single blastocyst transfer
Pregnancy loss was associated with higher rates of implantation, clinical
Total pregnancy loss among 134/583 (23·0%) 124/481 (25·8%) 0·89 (0·72–1·10) 0·29 pregnancy, and ongoing pregnancy than was fresh single
conception blastocyst transfer (table 3). The rate of twin pregnancies
Biochemical miscarriage 65/583 (11·1%) 68/481 (14·1%) 0·79 (0·57–1·08) 0·14 was also higher in the frozen embryo transfer group
Clinical pregnancy loss 69/512 (13·5%) 56/401 (14·0%) 0·97 (0·70–1·34) 0·83 (table 3). However, the rate of pregnancy loss was similar
First trimester pregnancy 54/512 (10·5%) 46/401 (11·5%) 0·92 (0·63–1·33) 0·66 between the two groups (table 3).
loss
Singleton birthweight after frozen single blastocyst
Second trimester 15/512 (2·9%) 10/401 (2·5%) 1·17 (0·53–2·59) 0·69 transfer was higher than that after fresh single blastocyst
pregnancy loss
transfer (table 3).
Data are n (%), mean (SD), or n/N (%). *Absolute difference in birthweight was 114·8 g (95% CI 43·0–186·6) for Risk of moderate or severe OHSS did not differ
singleton livebirth and 21·0 g (–215·2 to 257·3) for twin livebirth. †Birthweight of ten newborn babies was missing. significantly between the frozen and fresh single blasto­
‡Conception: serum human chorionic gonadotropin ≥10 mIU/mL. §Clinical pregnancy: detection of a gestational sac in
the uterine cavity. ¶Twin pregnancies were defined as detection of two gestational sacs by ultrasound scan 5 weeks cyst transfer groups (table 4). Frozen single blastocyst
after embryo transfer. ||Monozygotic twin pregnancies were defined as detection of two fetal heart beats 5 weeks after transfer was associated with a higher risk of pre-eclampsia
single embryo transfer. **Implantation rate: number of gestational sac divided by number of embryos that were (table 4). The rates of other obstetrical complications
transferred. ††Ongoing pregnancy: detection of a viable fetus with fetal heartbeat at 11–12 weeks’ gestation.
including preterm delivery (detailed in the appendix)
Table 3: Livebirth, birthweight, pregnancy, and pregnancy loss were similar between the two groups (table 4). The risks
of neonatal morbidities including congenital anomalies
were also similar between the two groups (appendix).
assigned to either fresh or frozen single blastocyst The results of our per-protocol analyses and per-
transfer groups (figure). The fresh and frozen groups treatment analyses were consistent with the primary
were comparable in baseline demographics and clinical intention-to-treat analysis for the rates of singleton live­
characteristics (table 1) and the outcomes of ovarian birth, livebirth, pregnancy, and implantation (appendix).
stimulation (table 2). In these secondary analyses, frozen single blastocyst
98 (12%) women who were assigned to the fresh embryo transfer was associated with lower risks of ectopic
transfer group actually underwent a frozen embryo pregnancy and SGA, and higher risks of gestational
transfer, while 37 (5%) women assigned to the frozen diabetes, pre-eclampsia, and LGA.
embryo transfer group actually had a fresh embryo
transfer (p<0·0001; figure). The main reasons for patients Discussion
converting to the frozen embryo transfer group were the In ovulatory women with a good prognosis, a frozen
risk of OHSS (51 of 98) and lack of blastocyst formation at single blastocyst transfer resulted in a higher rate of
day 5 of embryo culture (14 of 98), while converting to the singleton livebirth, which was mainly mediated by a
fresh embryo transfer group was because of patients’ higher rate of implantation, than did fresh single blasto­
request (32 of 37) and poor embryo quality (five of 37). cyst transfer. The rate of pregnancy loss after frozen and
Additionally, although most women complied with the fresh single blastocyst transfer was similar. Frozen
protocol and underwent single embryo transfer, 42 (5%) of blastocyst transfer led to a higher singleton birthweight

6 www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5

 Articles

but also a higher risk of LGA than did fresh single

Frozen embryo Fresh embryo Relative risk in p value
blastocyst transfer. The incidence of pre-eclampsia was transfer group transfer group frozen embryo
higher after frozen single blastocyst transfer than after group (95% CI)
fresh single blastocyst transfer. The risk of moderate or Maternal complications
severe OHSS was similar in both groups. Moderate or severe OHSS* 4/825 (0·5%) 9/825 (1·1%) 0·44 (0·14–1·44) 0·16
The higher implantation rate after frozen single Ectopic pregnancy† 6/583 (1·0%) 12/481 (2·5%) 0·41 (0·16–1·09) 0·065
blastocyst transfer might involve three factors and their Gestational diabetes‡ 52/512 (10·2%) 32/401 (8·0%) 1·27 (0·84–1·94) 0·26
interaction: an embryo with implantation competency, Gestational hypertension‡ 13/512 (2·5%) 8/401 (2·0%) 1·27 (0·53–3·04) 0·59
a more receptive endometrium, and improved develop­
Pre-eclampsia‡ 16/512 (3·1%) 4/401 (1·0%) 3·13 (1·06–9·30) 0·029
mental synchrony between the embryo and the endo­
Placenta previa‡ 8/512 (1·6%) 5/401 (1·2%) 1·25 (0·41–3·80) 0·69
metrium.20 Ovarian stimulation and possibly the resultant
Preterm rupture of membrane‡ 49/512 (9·6%) 44/401 (11·0%) 0·87 (0·59–1·28) 0·49
supra-physiological oestrogen concentrations might have
Preterm delivery‡ 32/512 (6·3%) 26/401 (6·5%) 0·96 (0·58–1·59) 0·89
a detrimental effect on endometrial develop­ ment
Post-partum haemorrhage§ 8/441 (1·8%) 1/342 (0·3%) 6·20 (0·78–49·37) 0·09
compared with the endometrium in natural cycles.10,21
Neonatal complications
Frozen embryo transfer allows for the removal of
Small for gestational age¶ 29/452 (6·4%) 33/353 (9·3%) 0·69 (0·43–1·11) 0·12
iatrogenically administered gonadotropins and recovery
Large for gestational age¶ 84/452 (18·6%) 41/353 (11·6%) 1·60 (1·13–2·26) 0·0067
of the stimulated ovaries. The subsequent shedding of the
Neonatal hospitalisation >3 days|| 50/443 (11·3%) 30/347 (8·6%) 1·31 (0·85–2·01) 0·22
exposed endometrium after ovarian stimu­ lation and a
Neonatal jaundice among live 79/443 (17·8%) 58/347 (16·7%) 1·07 (0·78–1·45) 0·68
fresh start and regrowth under alternative less intensive
newborns||
endometrial preparation regi­mens could provide a more
Neonatal infection among live 15/443 (3·4%) 10/347 (2·9%) 1·17 (0·53–2·58) 0·69
favourable uterine environ­ment for embryo implantation newborns||
with frozen embryo transfer than with fresh transfer. Congenital anomalies** 12/464 (2·6%) 11/355 (3·1%) 0·83 (0·37–1·87) 0·66
Transfer of a blastocyst into this comparatively more
favourable endometrial environment during a frozen Data are n/N (%). OHSS=ovarian hyperstimulation syndrome. *The denominator was number of women randomly
assigned to each group. †The denominator was number of conception in each group. ‡The denominator was number of
embryo transfer cycle more closely mimics the process of clinical pregnancy in each group. §The denominator was number of delivery including livebirths and stillbirths.
natural implantation compared with a frozen cleavage- ¶The denominator was number of newborn babies in each group. Birthweight of ten newborn babies in the frozen
stage embryo transfer cycle or a fresh embryo transfer embryo transfer group was missing. ||The denominator was number of newborn babies in each group. A total of
19 newborn babies in the frozen embryo transfer group and six in the fresh embryo transfer group were lost to follow-up.
cycle. The implantation rate after frozen or fresh blastocyst **The denominator was number of live newborns plus number of stillborn babies.
transfer in this trial was higher than that after either
frozen or fresh cleavage-stage embryo transfer (41% and Table 4: Maternal and neonatal complications
42%, respectively) in our previous trial in ovulatory
women.13 The better embryo selection by extended culture increased risk of pre-eclampsia and LGA was unclear;
to blastocyst could also have contributed to the increased however, embryo cryopreservation was suggested to alter
implantation rate compared with cleavage-stage embryo the epigenetics of embryo in in-vitro experiment22 and
transfer.5 Furthermore, the livebirth rate after frozen animal studies,23 and epigenetic dysregulation in turn
single blastocyst transfer is very similar to that after the was associated with abnormal placentation and fetal
transfer of two cleavage-stage embryos (49% and 50% after growth.24 Further study could also show that the decrease
two frozen or fresh cleavage-stage embryos transfer, in SGA with frozen blastocyst transfer that we noted in
respectively);13 however, the rate of twin livebirth (17% and our secondary analysis could counterbalance concerns
16%, respectively) and the risks of preterm birth (16% and about the increased prevalence of LGA babies. The
13%, respectively) were significantly reduced in the increase in birthweight predominantly in male babies in
ovulatory women in this study. the frozen blastocyst transfer group might actually favour
The risk of pre-eclampsia was higher after frozen survival of preterm males; however, this post-hoc finding
single blastocyst transfer than after fresh single blastocyst requires replication and a sound biological explanation.
transfer, which was consistent with our previous findings The risk of moderate or severe OHSS was similar
from the randomised trial in women with PCOS.12 In between the frozen and fresh single blastocyst transfer
ovulatory women with cleavage-stage embryo transfer, groups, which contrasted with our previous findings that
incidence of pre-eclampsia did not differ significantly frozen embryo transfer had a lower risk OHSS.12,13 The
after frozen embryo transfer (4% vs 3%, p=0·28).13 Frozen absolute incidence of OHSS in both groups was lower in
blastocyst transfer was also associated with a higher this study than in our previous two trials.12,13 The
incidence of LGA than was fresh blastocyst transfer. randomisation was done 3 days after oocyte retrieval
According to the definition of LGA, the incidence of LGA in this study whereas in the previous studies it was done
is 10% in the reference population that was mainly on the day of oocyte retrieval. The longer period of
composed of natural conception.19 Thus, frozen single observation allowed for the exclusion of patients who
blastocyst transfer seems to increase the risk of LGA developed or were at risk for developing OHSS before
compared with natural conception and fresh single randomisation. The loss of protection against OHSS in
blastocyst transfer. The underlying mechanism for the the frozen single blastocyst transfer group compared

www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5 7

 Articles
with fresh single blastocyst transfer could be due to the
selection bias towards women with a low risk of OHSS.
The rate of pregnancy loss was similar after frozen and
fresh single blastocyst transfer in this trial. In this
environment, the maintenance of pregnancy was not
adversely affected by previous ovarian stimulation. This
lack of association with pregnancy loss was consistent
with the result of our previous trial in ovulatory women
with cleavage-stage embryo transfer.13 However, it con­
trasts with the results in women with PCOS, in whom
frozen embryo transfer was associated with a lower rate
of pregnancy loss than was fresh embryo transfer.12
Although the underlying mechanism is still unclear, the
effect of supra-physiological oestrogen concentrations on
pregnancy loss might vary between ovulatory women and
women with PCOS.25
A Cochrane meta-analysis including four randomised
trials (1892 women) showed that the rates of clinical
pregnancy and ongoing pregnancy were similar after
frozen versus fresh embryo transfer, whereas frozen
embryo transfer was associated with a lower rate of mis­
carriage and an increased risk of pregnancy complications
after the first transfer.26 However, the result of this
Cochrane review was dominated by our previous trial, the
largest trial in women with PCOS (1508 women), who
could be more susceptible to an increased rate of
pregnancy loss after fresh embryo transfer than ovulatory
women.25 Since the publication of the Cochrane review,
five new randomised trials comparing frozen with fresh
embryo transfer in different populations have been
published.13,14,27–29 The two large trials in ovulatory women
with cleavage-stage embryo transfer showed no difference
in the rates of implantation, pregnancy, pregnancy loss,
or livebirth.13,14 The trial comparing frozen versus fresh
euploid blastocyst transfer showed higher rates of
implantation and livebirth in the frozen-thawed cycle
than in the fresh cycle.27 The mechanism underlying the
discrepant result between blastocyst-stage embryo
transfer and cleavage-stage embryo transfer is unclear.
However, since ovarian stimulation advances the window
of implantation, it could decrease endometrial receptivity
during a fresh cycle. This notion was supported by a
study in which the rates of implantation and pregnancy
after fresh single blastocyst transfer were significantly
reduced when the normally developing embryo was
electively transferred on day 6 compared with on day 5.30
Alternatively, transfer of a cleavage-stage embryo into the
uterus might promote the synchronised development
between embryo and endometrium at the time of
implantation, while extended in-vitro culture might
perturb the kinetics of embryonic development and
disrupt the synchrony with endometrial development.
Furthermore, two embryos were usually transferred at a
time when cleavage-stage embryo trans­ fer was done
while only one blastocyst was typically transferred in the
blastocyst-stage embryo transfer cycle. Two embryos
within the uterine cavity may compete with each other for
8 www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5
implantation and consequently result in a decreased
implantation rate of each embryo. Future studies are
needed to elucidate the physiological differ­ ence in
implantation after blastocyst transfer and cleavage-stage
embryo transfer.
The strengths of this study include the large sample size
that allows for an accurate estimate of the primary outcome
(singleton livebirth), and the multicentre setting and
pragmatic design that improves extrapolation of our
results. However, there are limitations in this study. First,
we included only young women with a good prognosis;
more than 25% of the screened women were excluded
because of fewer than four high-grade embryos on day 3,
poor ovarian response, or a previous failed IVF cycle. We
should be cautious to generalise the results to women with
an unfavourable or even less favourable prognosis. Second,
this study was a pragmatic trial and a reflection of clinical
practice. About 5% of women in the frozen embryo
transfer group had a transfer of two embryos, which was
higher than the proportion in the fresh embryo transfer
group. These deviations were mostly because of patients
insisting on two embryos being transferred during the
wait for a frozen embryo transfer. The rate of twin
pregnancies was higher in the frozen blastocyst transfer
group probably because of this iatrogenic tendency to
transfer more than one embryo in this group. This
performance bias probably led to a higher rate of livebirth
in the frozen blastocyst transfer group, but alone does not
account for the difference in livebirth rate between groups.
Results of the per-protocol analysis in women who
underwent single blastocyst transfer supported those of
the intention-to-treat analysis. Furthermore, because this
was a non-blinded study, the question is not whether there
was treatment bias, because this influenced treatment
crossovers and multiple embryos transfer, but whether we
are underestimating its effects. Finally, elements of the
pragmatic design such as type of embryo media (we used
sequential media, not single-step) or choice of frozen cycle
regimen could have affected results. Although this study
had a relatively large sample, it was neither designed nor
powered to show differences in obstetric and neonatal
complications. Future meta-analysis pooling all trials
might obtain a consolidated conclusion.
Nonetheless there are practice changing implications to
our findings. Our results suggest that frozen single
blastocyst transfer is better to achieve singleton livebirth
than fresh single blastocyst transfer in women with
good prognosis. Compared with our previous studies
that allowed multiple cleavage-stage embryo transfers,12,13
the practice of single frozen blastocyst transfer reduces
multiple pregnancy rates and associated morbidities,
while maintaining livebirth rate. Its potential for increased
maternal pre-eclampsia and its long-term effects on
offspring warrant further studies.
Contributors
DW, YSh, Z-JC, HeZ, and RSL designed the trial. YSh and Z-JC were in
charge of the trial conduct. J-YL, YSu, BZ, YHS, J-QL, JT, XLia, YC, ZW,

YaQ, HaZ, YZho, HR, GH, XLin, JZ, YZha, XQ, XD, XC, YZhu, XW,
L-FT, QL, and XM enrolled participants. DW, ZW, and LZ did the
statistical analyses and prepared the tables with oversight by HeZ.
DW, HeZ, RSL, and Z-JC drafted the manuscript. Z-JC had a primary
responsibility for final content. All authors were involved in data
collection, interpreted the data, provided critical input to the manuscript,
and approved the final manuscript.
Declaration of interests
HeZ has received grants from the National Institute of Health (NIH)
and National Science Foundation during the conduct of the study.
RSL reports grants from NIH and Guerbet; grants and consultant’s fees
from Ferring; and consultant’s fees from Bayer, Abbvie, Fractyl, and
Ogeda, outside the submitted work. All other authors declare no
competing interests.
Data sharing
The study protocol and statistical analysis plan will be available online
with publication. Data collected for the study, including specified dataset
and a data dictionary defining each field in the set, will be made available
to others with publication. Investigators can request data sharing by
emailing the corresponding author. Our publication committee
established for this trial will review and approve the request.
An agreement on how to collaborate will be reached based on the
overlaps and conflicts between the proposal and our ongoing efforts.
Acknowledgments
This trial was funded by the National Key Research and Development
Program of China (2017YFC1001000), the State Key Program of
National Natural Science Foundation of China (81430029), National
Natural Science Foundation of China (81701515), Natural Science
Foundation of Shandong (ZR2017BH011), and Thousand Talents
Program (RSL and HeZ). We thank all participants in this study, all
research staff in study sites, staff of Resman central randomisation
platform, and members of the data safety monitoring board for their
oversight for this trial.
References
1 Practice Committee of Society for Assisted Reproductive Technology,
Practice Committee of American Society for Reproductive Medicine.
Elective single-embryo transfer. Fertil Steril 2012; 97: 835–42.
2 Harbottle S, Hughes C, Cutting R, Roberts S, Brison D.
Elective single embryo transfer: an update to UK Best Practice
Guidelines. Hum Fertil (Camb) 2015; 18: 16583.
3 Bhattacharya S, Kamath MS. Reducing multiple births in assisted
reproduction technology. Best Pract Res Clin Obstet Gynaecol 2014;
28: 191–99.
4 Glujovsky D, Farquhar C. Cleavage-stage or blastocyst transfer:
what are the benefits and harms? Fertil Steril 2016; 106: 244–50.
5 Papanikolaou EG, Camus M, Kolibianakis EM, Van Landuyt L,
Van Steirteghem A, Devroey P. In vitro fertilization with single
blastocyst-stage versus single cleavage-stage embryos. N Engl J Med
2006; 354: 1139–46.
6 Glujovsky D, Farquhar C, Quinteiro Retamar AM, Alvarez Sedo CR,
Blake D. Cleavage stage versus blastocyst stage embryo transfer in
assisted reproductive technology. Cochrane Database Syst Rev 2016;
6: CD002118.
7 Gleicher N, Kushnir VA, Barad DH. Is it time for a paradigm shift
in understanding embryo selection? Reprod Biol Endocrinol 2015;
13: 3.
8 Practice Committee of Society for Assisted Reproductive Technology,
Practice Committee of American Society for Reproductive Medicine.
Blastocyst culture and transfer in clinical-assisted reproduction:
a committee opinion. Fertil Steril 2013; 99: 667–72.
9 Bourgain C, Devroey P. The endometrium in stimulated cycles for
IVF. Hum Reprod Update 2003; 9: 515–22.
10 Mirkin S, Nikas G, Hsiu JG, Diaz J, Oehninger S. Gene expression
profiles and structural/functional features of the peri-implantation
endometrium in natural and gonadotropin-stimulated cycles.
J Clin Endocrinol Metab 2004; 89: 5742–52.
www.thelancet.com Published online February 28, 2019 http://dx.doi.org/10.1016/S0140-6736(18)32843-5 9
Articles
11 Rienzi L, Gracia C, Maggiulli R, et al. Oocyte, embryo and blastocyst
cryopreservation in ART: systematic review and meta-analysis
comparing slow-freezing versus vitrification to produce evidence for
the development of global guidance. Hum Reprod Update 2017;
23: 139–55.
12 Chen Z-J, Shi Y, Sun Y, et al. Fresh versus frozen embryos for
infertility in the polycystic ovary syndrome. N Engl J Med 2016;
375: 523–33.
13 Shi Y, Sun Y, Hao C, et al. Transfer of fresh versus frozen embryos
in ovulatory women. N Engl J Med 2018; 378: 126–36.
14 Vuong LN, Dang VQ, Ho TM, et al. IVF transfer of fresh or frozen
embryos in women without polycystic ovaries. N Engl J Med 2018;
378: 137–47.
15 Practice Committee of Society for Assisted Reproductive Technology,
Practice Committee of American Society for Reproductive Medicine.
Guidance on the limits to the number of embryos to transfer:
a committee opinion. Fertil Steril 2017; 107: 901–03.
16 Wei D, Sun Y, Liu J, et al. Live birth after fresh versus frozen single
blastocyst transfer (Frefro-blastocyst): study protocol for a randomized
controlled trial. Trials 2017; 18: 253.
17 Puissant F, Van Rysselberge M, Barlow P, Deweze J, Leroy F.
Embryo scoring as a prognostic tool in IVF treatment. Hum Reprod
1987; 2: 705–08.
18 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.
19 Dai L, Deng C, Li Y, et al. Birth weight reference percentiles for
Chinese. PLoS One 2014; 9: e104779.
20 Teh WT, McBain J, Rogers P. What is the contribution of
embryo-endometrial asynchrony to implantation failure?
J Assist Reprod Genet 2016; 33: 1419–30.
21 Basir GS, O WS, Ng EH, Ho PC. Morphometric analysis of
peri-implantation endometrium in patients having excessively high
oestradiol concentrations after ovarian stimulation. Hum Reprod
2001; 16: 435–40.
22 Hiura H, Hattori H, Kobayashi N, et al. Genome-wide microRNA
expression profiling in placentae from frozen-thawed blastocyst
transfer. Clin Epigenetics 2017; 9: 79.
23 Wang Z, Xu L, He F. Embryo vitrification affects the methylation of
the H19/Igf2 differentially methylated domain and the expression
of H19 and Igf2. Fertil Steril 2010; 93: 2729–33.
24 Januar V, Desoye G, Novakovic B, Cvitic S, Saffery R. Epigenetic
regulation of human placental function and pregnancy outcome:
considerations for causal inference. Am J Obstet Gynecol 2015;
213: S182–96.
25 Wei D, Yu Y, Sun M, et al. The effect of supraphysiological estradiol
on pregnancy outcomes differs between women with PCOS and
ovulatory women. J Clin Endocrinol Metab 2018; 103: 2735–42.
26 Wong KM, van Wely M, Mol F, Repping S, Mastenbroek S.
Fresh versus frozen embryo transfers in assisted reproduction.
Cochrane Database Syst Rev 2017; 3: CD011184.
27 Coates A, Kung A, Mounts E, et al. Optimal euploid embryo transfer
strategy, fresh versus frozen, after preimplantation genetic screening
with next generation sequencing: a randomized controlled trial.
Fertil Steril 2017; 107: 723–30.e3.
28 Aflatoonian A, Mansoori-Torshizi M, Farid Mojtahedi M, et al.
Fresh versus frozen embryo transfer after gonadotropin-releasing
hormone agonist trigger in gonadotropin-releasing hormone
antagonist cycles among high responder women: a randomized,
multi-center study. Int J Reprod Biomed (Yazd) 2018; 16: 9–18.
29 Aghahosseini M, Aleyasin A, Sarfjoo FS, Mahdavi A, Yaraghi M,
Saeedabadi H. In vitro fertilization outcome in frozen versus fresh
embryo transfer in women with elevated progesterone level on the
day of HCG injection: an RCT. Int J Reprod Biomed (Yazd) 2017;
15: 757–62.
30 Poulsen V, Ingerslev HJ, Kirkegaard K. Elective embryo transfers on
Day 6 reduce implantation compared with transfers on Day 5.
Hum Reprod 2017; 32: 1238–43.