300 RBMO VOLUME 41 ISSUE 2 2020

ARTICLE

Maternal and perinatal outcomes in

programmed versus natural vitrified–warmed
blastocyst transfer cycles
BIOGRAPHY
Dr Reeva Makhijani obtained her medical degree from NYU School of Medicine in 2014.
She completed her Obstetrics and Gynecology residency in 2018 at Women and Infants
Hospital in Rhode Island. She is currently a fellow in Reproductive Endocrinology and
Infertility at the University of Connecticut.

Reeva Makhijani, Chantal Bartels, Prachi Godiwala, Alison Bartolucci,

John Nulsen, Daniel Grow, Claudio Benadiva, Lawrence Engmann*

KEY MESSAGE
Programmed frozen embryo transfer cycles are associated with more obstetric complications, particularly
hypertensive disorders of pregnancy, compared with natural cycles. Therefore, natural frozen embryo transfer
protocols should be recommended for endometrial preparation in eligible patients.

ABSTRACT
Research question: Do maternal and perinatal outcomes differ between natural and programmed frozen embryo
transfer (FET) cycles?
Design: Retrospective cohort study at a university-affiliated fertility centre including 775 patients who underwent
programmed or natural FET cycles resulting in a singleton live birth using blastocysts vitrified between 2013 and 2018.
Results: A total of 384 natural and 391 programmed FET singleton pregnancies were analysed. Programmed FET
resulted in higher overall maternal complications (32.2% [126/391] versus 18.8% [72/384]; P < 0.01), including higher
probability of hypertensive disorders of pregnancy (HDP) (15.3% [60/391] versus 6.3% [24/384]; P < 0.01), preterm
premature rupture of membranes (2.6% [10/391] versus 0.3% [1/384]; P = 0.02) and caesarean delivery (53.2%
[206/387] versus 42.8% [163/381]; P = 0.03) compared with natural FET. After controlling for potential confounders,
including age, body mass index, parity, smoking status, history of diabetes or chronic hypertension, infertility
diagnosis, number of embryos transferred and use of preimplantation genetic testing, the adjusted odds ratio for
HDP was 2.39 (95% CI 1.37 to 4.17) and for overall maternal complications was 2.21 (95% CI 1.51 to 3.22) comparing
programmed with natural FET groups. The groups did not significantly differ for any perinatal outcomes analysed,
including birth weight (3357.9 ± 671.6 g versus 3318.4 ± 616.2 g; P = 0.40) or rate of birth defects (1.5% [6/391]
versus 2.1% [8/384]; P = 0.57), respectively.
Conclusion: Vitrified–warmed blastocyst transfer in a programmed cycle resulted in a twofold higher probability of
HDP compared with transfer in a natural cycle. Natural FET cycle should, therefore, be recommended as first line for
all eligible patients undergoing FET to reduce the risk of HDP.

Center for Advanced Reproductive Services, Division of Reproductive Endocrinology and Infertility, University of KEYWORDS
Connecticut School of Medicine, 2 Batterson Park Road, Farmington CT, USA
Hypertensive disorders in pregnancy
In vitro fertilization
© 2020 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved.
Obstetric outcome
*Corresponding author. E-mail address: [email protected] (L Engmann). https://doi.org/10.1016/j.rbmo.2020.03.009
1472-6483/© 2020 Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. Perinatal outcome
Declaration: The authors report no financial or commercial conflicts of interest. Vitrified–warmed embryo transfer
 RBMO VOLUME 41 ISSUE 2 2020 301

INTRODUCTION cycles in women with PCOS (Chen MATERIALS AND METHODS

A
et al., 2016). This difference in results
ssisted reproductive may be attributed to the method of Study design
technologies (ART) have endometrial preparation used by the This was a retrospective cohort study
been associated with an disparate patient cohorts in the two conducted at a single university-affiliated
increased risk of obstetric studies. This is further supported by fertility centre comparing the difference
and perinatal complications, including other studies that have shown that the between maternal and perinatal
hypertensive disorders of pregnancy risk of HDP does seem to be increased outcomes after programmed versus
(HDP), antepartum haemorrhage, in programmed FET cycles compared natural FET carried out between March
preterm delivery, low birthweight and with natural or modified natural cycles 2013 and October 2018. Only cycles in
congenital anomalies compared with (Ernstad et al., 2019; Saito et al., which previously vitrified blastocyst(s)
spontaneous conception (Helmerhorst 2019; Versen-höynck et al., 2019a). derived from autologous oocytes were
et al., 2004; Jackson et al., 2004; Additionally, the corpus luteum, which transferred and resulted in a singleton
Pandey et al., 2012; Hansen et al., is present in natural and absent in live birth were included. Only one cycle
2013; Tandberg et al., 2015; Sunderam programmed cycles, has been identified per patient was included in the analyses.
et al., 2019). Hypertensive disorders as possibly protective against the Exclusion criteria included donor oocyte
of pregnancy refer to a spectrum of development of HDP (Arthur et al., cycles, FET cycles resulting in multiple
hypertensive disorders in pregnancy 1996; Strauch et al., 2018; Versen- births, cycles in which cleavage stage
ranging from gestational hypertension höynck et al., 2019b). embryos were transferred or cycles in
to eclampsia. The higher incidence which the embryos that were transferred
of multiple pregnancy after ART and The nature of IVF treatment is rapidly had been cryopreserved using slow
proportion of women of advanced changing, and it is challenging to obtain freezing. In addition, modified natural
maternal age undergoing IVF may partly data that reflect new techniques and FET cycles involving ovulation induction
explain the increased occurrence of are generalizable to one’s patient with letrozole, clomiphene citrate
these complications (Gardner et al., population. The two largest studies or gonadotrophins, as well as those
1995; Tandberg et al., 2015; Qin et al., that have compared the difference requiring HCG triggering to induce
2017; Sunderam et al., 2019). Laboratory in obstetric and perinatal outcomes ovulation, were excluded. This study was
procedures, such as cryopreservation between programmed and natural FET approved by our university’s institutional
methods, extended blastocyst culture cycles were conducted internationally, review board on 15 February 2019.
and use of preimplantation genetic limiting their generalizability to the US
testing (PGT), however, may also population (Ernstad et al., 2019; Saito Treatment protocol
contribute to such adverse obstetric et al., 2019). Furthermore, these studies IVF stimulation protocols have been
and perinatal outcomes (Palomba et al., included both cleavage- and blastocyst- previously described and were decided
2016). In fact, compared with fresh stage embryo transfer cycles and one based on patient factors and physician
embryo transfers, pregnancies after included both vitrification and slow preference (Diluigi et al., 2011; Johnston-
frozen embryo transfer (FET) cycles are freezing protocols for cryopreservation, MacAnanny, et al., 2011). Gonadotrophin
at higher risk of HDP, placenta accreta, although multicellular embryo transfer dosing was adjusted according to
infants who are large for gestational age and slow freezing are no longer widely patient response. Trigger for final oocyte
or have birth defects (Roque et al., 2013; used (Ernstad et al., 2019; Saito et al., maturation was administered when
Wennerholm et al., 2013; Ishihara et al., 2019). The only study using a US cohort three or more follicles reached 18 mm
2014; Centers for Disease Control and included mainly blastocyst transfer and in diameter or above. Either HCG
Prevention, 2016; Maheshwari et al., a few cleavage stage embryo transfer; 3300–10000 IU SC (Pregynl, Merck,
2016; Luke, 2017; Robles et al., 2017). however, it was limited by sample size Kenilworth, NJ, USA; Novarel, Ferring
Given the growing utilization of FET and the cryopreservation technique used Pharmaceuticals, Parsippany, NJ, USA),
cycles, which account for more than was unclear (Versen-höynck et al, 2019a; gonadotrophin releasing hormone
one-half of all the IVF cycles carried 2019b). (GnRH) agonist 1 mg (leuprolide acetate,
out in the USA, it is vital to identify the Abbott Laboratories, Abbott Park, IL,
cause underlying the association between Therefore, it is important to examine USA) or a combination was used for
FET cycles and higher rates of such the effect that method of endometrial trigger based on physician preference,
complications to improve patient safety preparation has on obstetric and protocol type and risk of ovarian
and implement preventative measures. perinatal outcomes in a cohort of hyperstimulation syndrome. Transvaginal
patients who underwent FET cycles ultrasound-guided oocyte retrieval was
Previous studies have suggested that using only previously vitrified blastocysts carried out 35 h after trigger injection.
particular aspects of FET cycle protocols resulting in singleton birth to both Oocytes were fertilized with either
may affect obstetric and perinatal reduce the effect of other cofounding intracytoplasmic sperm injection or
outcomes. For example, a recent study variables and reflect current practice conventional insemination.
demonstrated no difference in HDP in US fertility centres. Therefore, we
between fresh transfer and natural sought to determine whether maternal Only good-quality blastocysts according
FET cycles in ovulatory women (Shi and perinatal outcomes differed between to Gardner criteria (3BB or higher)
et al., 2018) whereas the same research programmed and natural FET cycles (Gardner and Schoolcraft, 1999)
group previously showed a higher involving transfer of previously vitrified were vitrified. If PGT was being used,
prevalence of HDP after programmed blastocysts that resulted in a singleton trophectoderm biopsy was carried
FET cycles compared with fresh IVF live birth. out, embryos vitrified and either array
302 RBMO VOLUME 41 ISSUE 2 2020

comparative genomic hybridization or complications, was self-reported and sent entry method, and all the predictor
next-generation sequencing was used to the Society of Assisted Reproductive variables were tested in one block to
for aneuploidy testing. Vitrification was Technology. These data were collected by assess their predictive ability while
by rapid exposure to a cryoprotectant nurses who telephoned each patient after controlling for other predictors in the
solution consisting of 15% ethylene delivery to obtain delivery details and model. Both crude and adjusted odds
glycol, 15% dimethylsulphoxide, 20% information on any complications. ratios with 95% confidence intervals
dextran and 0.5 mmol/l sucrose in a were calculated. In addition, subgroup
Cryolock® device (Irvine Scientific, Santa The primary outcomes were overall rate analyses were conducted stratifying
Ana, CA, USA). Before vitrification, a of maternal complications, particularly groups by age (<35 and ≥35 years), BMI
laser pulse of 300 µs (constant 0.9 J) was HDP, which included the diagnoses of (≤30 and >30 kg/m2) and use of PGT.
applied to collapse the blastocoel. Before pre-eclampsia, gestational hypertension, A two-sided P-value of less than 0.05
embryo transfer, all blastocysts were the syndrome of haemolysis, elevated was considered statistically significant.
rapidly warmed in solutions of 0.5 and liver enzymes and low platelet count or Bonferroni correction, however, was
0.2 mmol/l sucrose and rinsed through eclampsia. Secondary outcomes included used for multiple comparisons. For
HEPES-buffered human tubal fluid with maternal delivery characteristics and rate the 10 maternal outcomes assessed, P
12 mg/ml human serum albumin. To allow of other obstetric complications, e.g. <0.005 would be considered statistically
time for re-expansion, embryos were mode of delivery, gestational diabetes, significant. For the 15 perinatal outcomes
warmed and equilibrated in culture 1–2 h postpartum haemorrhage, preterm assessed, P < 0.003 would be considered
before transfer. premature rupture of membranes statistically significant.
(PPROM), chorioamnionitis, placenta
As previously described (Kaye et al., accreta, placenta previa, postpartum RESULTS
2018), FET was carried out in either haemorrhage and placental abruption,
a natural or programmed cycle. The as well as perinatal outcomes (rate of Baseline characteristics
decision to use a programmed or birth defects, birth weight, gestational The baseline cycle and demographic
natural FET cycle was based on the age at birth, newborn intensive care unit information for each FET group of
patient’s ovulatory status and physician admission and fetal loss.) Birth defects patients is presented in TABLE 1. A total of
preference. Programmed cycles were classified as major and minor as 775 FET cycles comprising 384 natural
consisted of downregulation with a described in previous studies (Bonduelle and 391 programmed FET cycles were
GnRH agonist in the luteal phase of the et al., 2002). A major congenital anomaly included for analysis. Women who
preceding cycle followed by increasing was defined as a condition that reduces underwent natural FET cycles were
doses of oral or transdermal oestradiol the viability or compromises the quality significantly older (35.0 ± 3.7 versus 33.9
after menses. Intramuscular progesterone of life and requires medical treatment ± 3.9 years; P < 0.01) and had lower
was started when endometrial thickness or a condition that causes functional BMI (25.9 ± 5.7 versus 27.1 ± 6.2kg/m2;
measured about 8 mm. On the sixth impairment or requires surgical correction. P < 0.01) than those who underwent
day of intramuscular progesterone, programmed FET, respectively. Women
FET was carried out. In a natural FET Statistical analysis who underwent natural FET cycles were
cycle, patients were monitored with IBM SPSS© Statistics version 26.0 was also more likely to have used PGT (P
daily bloodwork beginning on cycle used for statistical analysis. Student’s = 0.01) and transferred fewer embryos
day 10. A transvaginal ultrasound was t-test or Mann–Whitney U was used to (P < 0.01). Infertility diagnosis was
also carried out at this time to ensure analyse continuous variables depending significantly different between the two
a follicle had been recruited and that on whether the variable was normally groups (TABLE 1). To determine which
the endometrial stripe measured about distributed. Chi-squared or Fischer’s groups were significantly different,
8 mm. Six days after detecting the LH exact test was used for categorical a post-hoc analysis was conducted
surge (LH ≥20 IU/l), FET was carried out variables and, where indicated, a involving pairwise comparisons using
and the luteal phase supplemented with post-hoc analysis involving pairwise the z-test of two proportions with a
vaginal progesterone (Crinone, Merck, comparisons was conducted if there Bonferonni correction. This showed
Kenilworth, NJ, USA; and Endometrin, were three or more independent groups, that a significantly higher proportion of
Ferring Pharmaceuticals, Parsippany, NJ, using z-test of two proportions with women with the diagnosis of anovulation
USA), commencing 2 days after the LH a Bonferroni correction. Continuous or polycystic ovary syndrome used a
surge (Bartels et al., 2019). variables were presented as mean ± SD, programmed FET protocol compared
and categorical variables presented as with a natural cycle. In contrast, more
Outcome measures percentage and count. A binary logistic women with unexplained or male factor
Demographic and baseline cycle regression was conducted to control infertility underwent a natural FET
information were collected from the for potential covariates that may be protocol. A comparable number of
electronic medical records at our fertility associated with an increased risk of women in both groups were smokers
centre, including age, body mass index HDP, including age, BMI, and number and primigravida at time of conception.
(BMI), smoking status, parity, history of embryos transferred as continuous No differences were observed between
of pre-existing diabetes or chronic variables and smoking status, parity, groups in the proportion of women with
hypertension (HTN), primary infertility history of diabetes mellitus or HTN, history of diabetes mellitus and HTN.
diagnosis, mean number of embryos primary infertility diagnosis and use of
transferred and use of PGT. Outcome PGT as categorical variables (Vidaeff Maternal outcomes
data, including mode of delivery and et al., 2019; Zhang et al., 2019). All the Maternal delivery characteristics and
occurrence of maternal or perinatal variables were entered using a forced outcomes are presented in TABLE 2
 RBMO VOLUME 41 ISSUE 2 2020 303

TABLE 1 BASELINE CHARACTERISTICS FOR NATURAL AND PROGRAMMED FROZEN EMBRYO TRANSFER GROUPS

Baseline characteristics Natural FET (n = 384) Programmed FET (n = 391) P-value

Age, years, mean ± SD 35.0 ± 3.7 33.9 ± 3.9 <0.01

Age group, % (n) <0.01
<35 years 49.7 (191/384) 62.9 (246/391)
≥35 years 50.3 (193/384) 37.1 (145/391)
2
BMI (kg/m ), mean ± SD) 25.9 ± 5.7 27.1 ± 6.2 <0.01
BMI group, % (n) 0.03
≤30 kg/m2 79.2 (304/384) 72.6 (284/391)
>30 kg/m2 20.8 (80/384) 27.4 (107/391)
Smoker, % (n) 3.1 (12/384) 3.3 (13/389)a 0.87
Parity, % (n) 0.43
Parous 32.6 (125/384) 29.9 (117/391)
Nulliparous 67.4 (259/384) 70.1 (274/391)
History of pre-existing diabetes, % (n) 0.25
Yes 0.0 (0/384) 0.8 (3/391)
No 100.0 (384/384) 99.2 (388/391)
History of chronic HTN, % (n) 0.22
Yes 0.3 (1/384) 1.3 (5/391)
No 99.7 (383/384) 98.7 (386/391)
Diagnosis, % (n) <0.01
Unexplained 26.6 (102/384) 12.0 (47/391)
Anovulation/PCOS 3.9 (15/384) 37.3 (146/391)
Male factor 31.0 (119/384) 21.2 (83/391)
DOR 4.7 (18/384) 2.8 (11/391)
Endometriosis 6.5 (25/384) 7.2 (28/391)
Fibroids 3.9 (15/384) 2.3 (9/391)
RPL 8.9 (34/384) 7.2 (28/391)
Tubal factor 8.3 (32/384) 6.1 (24/391)
Other 6.3 (24/384) 3.8 (15/391)
Cycle characteristics
Mean number of embryos transferred (n, mean ± SD) 1.2 ± 0.4 1.3 ± 0.5 <0.01
PGT, % (n) 35.2 (135/384) 26.3 (103/391) 0.01
a The denominator is less than total number of patients in the group because of missing data for that variable.
BMI, body mass index; DOR, diminished ovarian reserve; HTN, chronic hypertension; PCOS, polycystic ovary syndrome; PGT, preimplantation genetic testing; RPL, recur-
rent pregnancy loss.

and observations specific to HDP are remained significantly different after even after groups were stratified by age
shown in FIGURE 1. Programmed FET was the Bonferroni correction was applied. (15.9% [39/246] versus 6.8% [13/191],
associated with more overall maternal No significant difference was observed P < 0.01; 14.5% [21/145] versus 5.7%
complications (32.2% [126/391] versus between groups in the rate of gestational [11/193], P < 0.01, P < 35 and P ≥ 35
18.8% [72/384]; P < 0.001) than natural diabetes mellitus or placenta accreta, years, respectively) (FIGURE 1A), BMI
FET, respectively. More specifically, although a few cases of placenta accreta (12.0% [34/284] versus 3.9% [12/304],
women in the programmed FET group were reported. P < 0.01; 24.3% [26/107] versus 15.0%
had a higher probability of HDP than the [12/80], P = 0.12), BMI ≤30 and >30 kg/
natural FET group (15.3% [60/391] versus In view of the differences in baseline m2, respectively) (FIGURE 1B) and the use
6.3% [24/384]); P = 0.001), respectively. characteristics as well as known risk of PGT (13.9% [40/288] versus 5.2%
Moreover, women in the programmed factors for HDP, subgroup analyses were [13/249], P < 0.01; 19.4% [20/103] versus
FET group more often experienced conducted. These subgroup analyses 8.1% [11/135], P = 0.01, without PGT
PPROM and were delivered via caesarean showed that a higher probability of HDP versus with PGT, respectively) (FIGURE 1C).
section. Only the rate of overall maternal in the programmed compared with Although HDP was significantly higher
complications and rate of HDP, however, natural FET group generally persisted in programmed compared with natural
304 RBMO VOLUME 41 ISSUE 2 2020

TABLE 2 MATERNAL OUTCOMES FOR NATURAL AND PROGRAMMED FROZEN EMBRYO TRANSFER GROUPS

Maternal complications Natural FET Programmed FET Crude OR (95% CI) Adjusted OR (95% CI)a P- valuea
Overall, %, n 18.8 (72/384) 32.2 (126/391) 2.06 (95% CI 1.48 to 2.87) 2.21 (95% CI 1.51 to 3.22) <0.01
Gestational diabetes, %, n 7.3 (28/384) 10.7 (42/391) 1.53 (95% CI 0.93 to 2.52) 1.69 (95% CI 0.95 to 2.99) 0.07
Hypertensive disorders of 6.3 (24/384) 15.3 (60/391) 2.56 (95% CI 1.56 to 4.25) 2.39 (95% CI 1.37 to 4.17) <0.01
pregnancy, %, n
Preterm premature rupture of 0.3 (1/384) 2.6 (10/391) 10.05 (95% CI 1.28 to 78.91) 11.71 (95% CI 1.44 to 95.55) 0.02
membranes, %, n
Chorioamnionitis, % (n) 0.5 (2/384) 0.5 (2/391) 0.98 (95% CI 0.25 to 8.89) 0.76 (95% CI 0.08 to 7.17) 0.81
Placental abruption, % (n) 0.5 (2/384) 0.8 (3/391) 1.48 (95% CI 0.25 to 8.94) 1.11 (95% CI 0.12 to 10.15) 0.93
Placenta accrete, % (n) 0.3 (1/384) 0.5 (2/391) 1.97 (95% CI 0.18 to 21.81) 2.98 (95% CI 0.25 to 34.95) 0.38
Placenta previa, (%) n 2.3 (9/384) 2.6 (10/391) 1.09 (95% CI 0.44 to 2.72) 1.50 (95% CI 0.57 to 3.89) 0.42
Postpartum haemorrhage, % (n) 1.0 (4/384) 1.3 (5/391) 1.23 (95% CI 0.33 to 4.62) 2.02 (95% CI 0.49 to 8.30) 0.33
Caesarean delivery, % (n) 42.8 (163/381)b 53.2 (206/387)b 1.52 (95% CI 1.14 to 2.02) 1.44 (1.04 to 2.01) 0.03
a Binary logistic regression controlling for covariates, including age, body mass index, smoking status, parity, history of diabetes mellitus or chronic hypertension, primary
infertility diagnosis, number of embryos transferred and use of preimplantation genetic testing.
b The denominator is less than total number of patients in the group because of missing data for that outcome variable.
FET, frozen embryo transfer.

FET cycles in women with BMI 30 kg/m2 Perinatal outcomes outcomes between the two methods of
or above, statistical significance was not No significant difference was observed endometrial preparation.
reached for HDP rates in programmed between groups in any of the perinatal
FET cycles in women with BMI above 30 outcomes, including the mean birth These results have significant implications
kg/m2, mainly because of the relatively weight and the probability of having for current practice patterns in US
smaller sample size in the BMI above an infant with birth defects (TABLE 3). fertility centres. Previous studies
30 kg/m2 group. After adjusting for Furthermore, no statistically significant comparing methods for endometrial
covariates of age, BMI, diagnosis, smoking difference was found in mean birthweight preparation in FET protocols
status, history of diabetes mellitus between groups, including the proportion demonstrated equivalent pregnancy
or HTN, primary infertility diagnosis, of macrosomic infants (≥4000 g) in the outcomes, including clinical pregnancy,
number of embryos transferred and use programmed compared with the natural ongoing pregnancy and live birth rates
of PGT, the probability of developing FET groups (14.0% [54/386] versus (Groenewoud et al., 2013; Peeraer
HDP was twofold higher in women who (9.4% [36/381]; P = 0.07)). In addition, et al., 2015; Yarali et al., 2016). Given
used programmed compared with natural no minor birth defects were reported in this information, decisions about which
FET cycles (adjusted OR 2.39; 95% CI either group and no significant difference protocol to choose for FET cycles have
1.37 to 4.17; P < 0.01) (TABLE 2). was found between groups in reported largely been left to physician preference
major birth defects (2.1% [8/384] versus if a particular patient’s characteristics
In addition, a secondary analysis was 1.5% [6/391]; P = 0.57). allowed for both protocol types. In fact,
conducted in which patients with programmed FET is often selected simply
anovulatory infertility were excluded DISCUSSION because it offers the most flexibility in
because they may have a higher terms of scheduling owing to the ability
likelihood of maternal complications, Our study evaluated the effect of to avoid weekend embryo transfers. Our
which could confound results. When endometrial preparation on obstetric results argue against this approach and
patients with anovulatory infertility and perinatal outcomes in a large US provide further justification to strongly
were excluded from the analysis, the cohort, which included only singleton recommend natural over programmed
rate of HDP remained significantly deliveries after transfer of blastocysts FET cycles in all eligible patients.
higher in the programmed FET group that were previously vitrified. We showed
compared with the natural FET group that the probability of developing HDP Our results are consistent with the
(15.1% [37/245] versus 6.5% [24/369]; was twofold higher in women who findings of previous studies, which most
P = 0.001), respectively. Similarly, the conceived after programmed compared notably demonstrate a significantly higher
rate of overall maternal complications with natural FET cycles (adjusted OR risk of HDP in pregnancies conceived
also remained significantly higher in 2.39, 95% CI 1.37 to 4.17; P < 0.01). after programmed FET cycles compared
the programmed FET group (34.7% This result persisted after controlling for with natural FET cycles (Ernstad et al.,
[85/245] versus 19.0% [70/369]; P < important covariates known to be risk 2019; Saito et al., 2019; Versen-höynck
0.001) compared with the natural FET factors for developing HDP, including age, et al., 2019b). In the present study,
group, respectively. In this subgroup BMI, diagnosis, smoking status, history this held true even after stratification
analysis, the odds ratio for HDP was of diabetes mellitus or HTN, primary and controlling for covariates that
2.56 (95% CI 1.49 to 4.40) and for infertility diagnosis, number of embryos are potential risk factors for the
overall maternal complications was 2.27 transferred and use of PGT. We did development of HDP. Our results, more
(95% CI 1.57 to 3.29). not detect any differences in perinatal importantly, confirm an increased risk
 RBMO VOLUME 41 ISSUE 2 2020 305

Natural FET Programmed FET Natural FET Programmed FET

100.0 100.0
90.0 90.0
80.0 80.0
70.0 70.0
Rate of HDP

Rate of HDP
60.0 60.0
50.0 50.0
40.0 40.0 P = 0.01 P < 0.01
P < 0.01 P < 0.01
30.0 30.0
19.4%
20.0 15.9% 14.5% 20.0 13.9%
6.8% 5.7% 8.1%
10.0 10.0 5.2%

0.0 0.0
<35 ≥35 With PGT Without PGT
Age groups (years) Groups divided by use of PGT

Natural FET Programmed FET

100.0
90.0
80.0
70.0
Rate of HDP

60.0
50.0
P = 0.07
40.0
P < 0.01
30.0 24.0%
20.0 12.9%
12.5%
10.0 4.8%
0.0
≤30 >30
BMI groups (kg/m2)

FIGURE 1 Rate of hypertensive disorders in pregnancy stratified by (A) age; (B) body mass index; and (C) use of preimplantation genetic testing for
aneuploidies. BMI, body mass index; FET, frozen embryo transfer; HDP, hypertensive disorders of pregnancy; PGT, preimplantation genetic testing.

of HDP in current IVF practices, which of cleavage-stage embryos (>50% of involved blastocyst transfers (>95%),
include blastocyst culture, vitrification the embryos in the Swedish study and their results were limited by sample size,
cryopreservation protocol and use of 20–30% of the embryos in the Japanese and it was unclear what proportion of
modern PGT techniques. study.) Furthermore, in the Swedish embryos were cryopreserved using slow
study, embryos were transferred that freezing. Furthermore, it was also unclear
To the best of our knowledge, this is had been cryopreserved using slow whether multiple cycles per patient were
the largest US cohort of patients that freezing. Although the investigators included, which would affect their results
addresses differences in maternal state that they controlled for freezing (Versen-höynck et al., 2019b).
and perinatal outcomes after method, they did not specify what
programmed and natural FET cycles. proportion of cycles used slow freezing. It has been proposed that this
Two previous studies with large cohorts Nevertheless, given that their registry heightened risk of HDP is because
addressing this question have been included cycles spanning back as far as programmed FET protocols do not
conducted internationally, limiting 2005, one can assume that a significant mimic physiologic conditions, resulting
their generalizability to our population number of cycles were carried out in abnormal placentation (Conrad
(Ernstad et al., 2019; Saito et al., 2019). before the widespread adoption of et al., 2017). It is accepted that HDP
Both of these studies, one conducted vitrification. One prospective study disorders develop from uteroplacental
in a Swedish cohort and the other in has been conducted in a US cohort. insufficiency (Mol et al., 2019); however,
a Japanese cohort, demonstrated a The investigators found a significantly more recently it has been proposed that
higher rate of HDP in programmed higher rate of HDP in programmed imbalances in angiogenic factors play a
FET cycles compared with natural FET FET cycles. Although their analysis role in the pathogenesis of HDP (Vidaeff
cycles. They both, however, included a covered only procedures carried out et al., 2019). Versen-Höynck et al. (2019a;
significant number of cycles with transfer in recent years (2011–2017) and mostly 2019b ) were the first to show that the
306 RBMO VOLUME 41 ISSUE 2 2020

TABLE 3 PERINATAL OUTCOMES FOR NATURAL AND PROGRAMMED FROZEN EMBRYO TRANSFER GROUPS

Perinatal birth characteristics and complications Natural FET Programmed FET P-value
Mean gestational age, weeks, mean ± SD 38.7 ± 2.3 38.7 ± 3.2 0.85
Gestational age group, % (n) 0.59
≤28 weeks 1.0 (4/384) 1.5 (6/391)
>28 to <34 weeks 1.8 (7/384) 3.1 (12/391)
>34 to <37 weeks 9.6 (37/384) 8.4 (33/391)
≥37 weeks 87.5 (336/384) 87.0 (340/391)
Weight, g, mean ± SD 3318.4 ± 616.2 3357.9 ± 671.6 0.40
Weight group, % (n) 0.14
a a
≤2500 g 7.1 (27/381) 8.0 (31/386)
>2500 to <4000 g 83.5 (318/381) 78.0 (301/386)
≥4000 g 9.4% (36/381) 14.0 (54/386)
Overall complications, % (n) 16.1 (62/384) 18.7 (73/391) 0.35
NICU admission, % (n) 4.7 (18/394) 6.4 (25/391) 0.30
Fetal loss <24 weeks, % (n) 0.3 (1/384) 0.3 (1/391) 1.0
Fetal loss ≥24 weeks, % (n) 0.0 (0/384) 0.0 (0/391) 1.0
Major birth defects, % (n) 2.1 (8/384) 1.5 (6/391) 0.57
Minor birth defects, % (n) 0.0 (0/384) 0.0 (0/391) 1.0
a The denominator is less than total number of patients in the group because of missing data for that outcome variable.
FET, frozen embryo transfer; NICU, newborn intensive care unit.

absence of corpus luteum may affect correction was applied, however, difference in gestational diabetes mellitus
vascular health during early pregnancy this result was no longer significant. as described by Saito et al. (2019).
by altering maternal circulation. In an Furthermore, there does not seem to
initial study, the investigators analysed be a physiological explanation for the Although multiple studies have
a small cohort of patients undergoing increased rate of PPROM. Moreover, we demonstrated a higher incidence
IVF based on number of corpus luteum did not find an increased risk of preterm of macrosomic infants born after
present, and found that, in the absence delivery, so it is unlikely that this finding programmed FET (Ishii et al., 2018;
of a corpus luteum, mean arterial is clinically significant. Additionally, Ernstad et al., 2019; Saito et al., 2019),
pressure did not decline as expected in unlike previous reports (Ishihara et al., we did not detect a significant difference
pregnancy and also that the measures 2014; Saito et al., 2019), we did not in mean birth weight according to FET
of endothelial cell function and arterial find a higher rate of placenta accreta protocol nor when birth weight was
stiffness were significantly altered. in programmed FET pregnancies. It is stratified into groups. In general, it is
In addition, circulating angiogenic reasonable to assume that, if placental difficult to reconcile how programmed
progenitor cells were lower and the development is compromised in FET cycles result in both high rates of
concentration of relaxin, a potent programmed FET cycles, there should HDP, which is often associated with fetal
vasodilator secreted by corpus luteum be a higher risk of disorders of abnormal growth restriction, and fetal macrosomia.
was lower; this contributes to maternal placentation, including placenta accreta Potential explanations may include
vascular adaptations in pregnancy as and placenta previa. Placenta accreta, distinctive parental traits, such as more
shown in animal models (Marshall et al., however, is a rare event, ranging gestational diabetes mellitus in women
2016). In a subsequent study, they further from 1 in 272 to 1 in 533 pregnancies who give birth to macrosomic infants.
showed that vascular compliance was (American College of Obstetricians We also did not detect a significant
blunted in the absence of a corpus and Gynecologists and Society for the difference in major or minor birth
luteum (Arthur et al., 1996; Strauch Maternal-Fetal Medicine, 2018). It is, defects, but as birth defects are so rare, a
et al., 2018; Versen-höynck et al., 2019b; therefore, notable that, in the study by much larger data set may be required to
2019a). On the basis of these findings, Saito et al. (2019), the prevalence of discern such a difference.
one might argue that natural or modified placenta accreta was less than 1% in
natural FET protocols, in which one or both the programmed and natural cycles Our study is unique because we included
more corpora lutea are present, should groups (0.1% versus 0.9%; P < 0.01), only singleton deliveries that resulted
be recommended as the FET protocols which is consistent with the probability from transfer of vitrified–warmed
of choice for eligible patients to reduce noted in our study (0.3% versus 0.5%, autologous blastocysts. One may argue
their risk of HDP. P = 1.0). They found a statistically that the findings of this study are limited
significant difference, however, because by the fact that all cycles were performed
We observed a higher rate of PPROM of their large sample size and the clinical at a single fertility centre, but it is also
in the programmed compared with the significance of this finding is debatable. a strength as each group was treated
natural FET group. When the Bonferroni Similarly, we did not detect a significant uniformly in the medications that were
 RBMO VOLUME 41 ISSUE 2 2020 307

used and how they were monitored. In important to highlight that HDP is one REFERENCES
contrast, in the Swedish cohort study of the leading causes of maternal and
of 10,000 singleton births discussed perinatal death around the world (Vidaeff American College of Obstetricians and
above (Ernstad et al., 2019), the type of et al., 2019). Currently, no accurate Gynecologists and Society for the Maternal-
Fetal Medicine Obstetrics Care Consensus:
FET cycle carried out and medications screening methods are available for
Placenta Accreta Spectrum. Obstet. Gynecol.
used were assumed based on what the early detection of women at risk of 2018; 132: e259–75
drugs had been prescribed for each developing HDP. Low-dose aspirin may Almog, B., Levin, I., Wagman, I., Kapustiansky,
patient. Therefore, it was possible that modestly reduce the risk of developing R., Lessing, J.B., Amit, A., Azem, F. Adverse
centres across the country were using HDP, but no interventions can completely obstetric outcome for the vanishing twin
syndrome. Reprod. Biomed. Online. 2010; 20:
vastly different protocols and that the eliminate it (ACOG, 2018). Therefore,
256–260
medications prescribed were not the the most effective strategy to reduce a Arthur, I.D., Anthony, F.W., Thomas, E.J. Serum
ones necessarily used by the patient. patient’s risk for HDP may be to evaluate relaxin and the major endometrial secretory
Moreover, our study was conducted for modifiable risk factors, one of which proteins in in-vitro fertilization and down-
retrospectively, so there is a risk of is selecting the type of FET protocol regulated hormone-supported and natural
cycle frozen embryo transfer. Hum. Reprod.
confounding and bias. Logistic regression type used by women during IVF. As
1996; 11: 88–91
was conducted to control for potential such, natural FET protocols should be Bartels, C.B., Ditrio, L., Grow, D.R., O’Sullivan,
significant confounders, including age, recommended as first-line for all eligible D.M., Benadiva, C.A., Engmann, L., Nulsen,
BMI, smoking status, parity, history of patients undergoing IVF to reduce the J.C. The window is wide: flexible timing for
pre-existing diabetes or HTN, primary risk of HDP. vitrified-warmed embryo transfer in natural
infertility diagnosis, number of embryos cycles. Reprod. Biomed. Online. 2019; 39:
241–248
transferred and use of PGT, and still Bonduelle, M., Liebaers, I., Deketelaere, V., Derde,
found a significantly higher risk of ACKNOWLEDGEMENTS M.P., Camus, M., Devroey, P., Van Steirteghem,
HDP in the programmed FET group. A. Neonatal data on a cohort of 2889 infants
Furthermore, as significantly more We would like to thank the physicians born after ICSI (1991-1999) and of 2995
women had anovulatory infertility in the and nursing staff at CARS for their infants born after IVF (1983-1999). Hum.
Reprod. 2002; 17: 671–694
programmed FET group, we carried support. Centers for Disease Control and Prevention, 2016.
out a secondary analysis and excluded 2016 Assisted Reproductive Technology Fertility
these patients as they may have a higher Clinic Success Rates Report.
probability of maternal complications. Chen, Z., Shi, Y., Sun, Y., Zhang, B., Liang, X., Cao,
We still found that rate of HDP and Y., Yang, J., Liu, J., Wei, D., Weng, N., Tian, L.,
Hao, C., Yang, D., Zhou, F., Shi, J., Xu, Y., Li, J.,
overall maternal complications was
Yan, J., Qin, Y., Zhao, H., Zhang, H., Legro, R.S.
significantly higher in the programmed Fresh versus Frozen Embryos for Infertility
FET group. in the Polycystic Ovary Syndrome. N. Eng. J.
Med. 2016; 375: 523–533
Nevertheless, we recognize that other Conrad, K.P., Belen, M., Post, E.D. Emerging role
for dysregulated decidualization in the genesis
possible confounders may exist that
of preeclampsia. Placenta. 2017; 60: 119–129
that we could not account for, such as Diluigi, A.J., Maier, D., Engmann, L., Schmidt,
history of other concurrent medical D.W., Benadiva, C., Nulsen, J.N. A randomized
conditions that may increase the risk of trial of microdose leuprolide acetate protocol
HDP (renal disease, autoimmune disease) versus luteal phase ganirelix protocol in
predicted poor responders. Fertil. Steril. 2011;
or prior history of HDP. In addition, we
95: 2531–2533
did not have information on vanishing Ernstad, E.G., Wennerholm, U., Khatibi, A.,
twins or fetal demise of a twin. Research Petzold, M., Bergh, C. Neonatal and maternal
suggests that pregnancies complicated outcome after frozen embryo transfer:
by vanishing twin syndrome may be Increased risks in programmed cycles. Am. J.
Obstet. Gynecol. 2019; 221: 126–e1
associated with increased perinatal
Evron, E., Sheiner, E., Friger, M., Sergienko, R.,
complications, including low birthweight Harlev, A. Vanishing twin syndrome: Is it
and preterm delivery (Almog et al., 2010; associated with adverse perinatal outcome?
Evron et al., 2015; Timur et al., 2018). To Fertil. Steril. 2015; 103: 1209–1214
our knowledge, however, vanishing twin Gardner, D.K., Schoolcraft, W.B. Culture and
transfer of human blastocysts. Curr. Opin.
syndrome has not been associated with
Obstet. Gynecol. 1999; 11: 307–311
increased risk of HDP. Nevertheless, it Gardner, M.O., Goldenberg, R.L., Cliver, S.P.,
may represent a potential confounder Tucker, J.M., Nelson, K.G., Copper, R.L.
for outcomes. As such, we hope that our The origin and outcome of preterm twin
results prompt further investigation. A pregnancies. Obstet. Gynecol. 1995; 85:
prospective randomized controlled trial 553–557
Groenewoud, E.R., Cantineau, A.E.P., Kollen,
would be the optimal method to validate B.J., Macklon, N.S., Cohlen, B.J. What is the
our findings. optimal means of preparing the endometrium
in frozen – thawed embryo transfer cycles?
In conclusion, blastocyst transfer in a A systematic review and meta-analysis. Hum.
programmed FET cycles is associated Reprod. Update. 2013; 19: 458–470
Hansen, M., Kurinczuk, J.J., Milne, E., de
with a higher risk of overall maternal Klerk, N., Bower, C. Assisted reproductive
complications, particularly HDP. It is technology and birth defects: A systematic
308 RBMO VOLUME 41 ISSUE 2 2020

review and meta-analysis. Hum. Reprod. and perinatal outcomes after high technology Technology Surveillance - United States, 2016.
Update 2013; 19: 330–353 infertility treatment: A comprehensive MMWR. Surveill. Summ. 2019; 68: 1–23
Helmerhorst, F.M., Perquin, D.A.M., Donker, systematic review. Reprod. Biol. Endocrinol. Tandberg, A., Klungsøyr, K., Romundstad, L.B.,
D., Keirse, M.J.N.C. Perinatal outcome of 2016; 14: 1–25 Skjærven, R. Pre-eclampsia and assisted
singletons and twins after assisted conception: Pandey, S., Shetty, A., Hamilton, M., Bhattacharya, reproductive technologies: Consequences of
A systematic review of controlled studies. Br. S., Maheshwari, A. Obstetric and perinatal advanced maternal age, interbirth intervals,
Med. J. 2004; 328: 261–264 outcomes in singleton pregnancies resulting new partner and smoking habits. BJOG An
Ishihara, O., Ph, D., Araki, R., Ph, D., Kuwahara, from ivf/icsi: A systematic review and meta- Int. J. Obstet. Gynaecol. 2015; 122: 915–922
A., Ph, D. Impact of frozen-thawed single- analysis. Hum. Reprod. Update. 2012; 18: Timur, H., Aksoy, R.T., Tokmak, A., Timur, B.,
blastocyst transfer on maternal and neonatal 485–503 Coskun, B., Uygur, D., Danisman, N. Maternal
outcome : an transfer cycles from 2008 to Peeraer, K., Couck, I., Debrock, S., De Neubourg, and perinatal outcomes of dichorionic
2010 in Japan. Fertil. Steril. 2014; 101: 128–133 D., De Loecker, P., Tomassetti, C., Laenen, A., diamniotic twin pregnancies diagnosed with
Ishii, R., Shoda, A., Kubo, M., Okazaki, S., Suzuki, Welkenhuysen, M., Meeuwis, L., Pelckmans, vanishing twin syndrome: A retrospective
M., Okawa, R., Enomoto, M., Shitanaka, M., S., Meuleman, C., Hooghe, T.D. Frozen- analysis from a single clinical center. Ginekol.
Fujita, Y., Nakao, K., Shimada, N., Horikawa, thawed embryo transfer in a natural or mildly Pol. 2018; 89: 30–34
M., Negishi, H., Taketani, Y. Identifying a hormonally stimulated cycle in women with American College of Obstetricians and
possible factor for the increased newborn regular ovulatory cycles: a RCT. Hum. Reprod. Gynecologists ACOG Committee Opinion
size in singleton pregnancies after assisted 2015; 30: 2552–2562 No. 743. Low dose Aspirin Use in Pregnancy.
reproductive technology using cryopreserved Qin, J.-B., Sheng, X.-Q., Wu, D., Gao, S.-Y., Obstet. Gynecol. 2018; 132: 44–52
embryos, in comparison with fresh embryos. You, Y.-P., Yang, T.-B., Wang, H. Worldwide Von Versen-höynck, F., Narasimhan, P., Seda, E.,
Reproductive Medicine and Biology. 2018; 17: prevalence of adverse pregnancy outcomes Tierney, S., Martinez, N., Conrad, K.P., Baker,
307–314 among singleton pregnancies after in vitro V.L., Winn, V.D. Absent or Excessive Corpus
Jackson, R.A., Gibson, K.A., Wu, Y.W., Croughan, fertilization/intracytoplasmic sperm injection: Luteum Number Is Associated With Altered
M.S. Perinatal outcomes in singletons a systematic review and meta-analysis. Arch. Maternal Vascular Health in Early Pregnancy.
following in vitro fertilization: A meta-analysis. Gynecol. Obstet. 2017; 295: 285–301 Hypertension. 2019; 73: 680–690
Obstet. Gynecol. 2004; 103: 551–563 Robles, A., Vidal, M., Checa, M.A., Prat, M. Von Versen-höynck, F., Schaub, A.M., Chi, Y.,
Johnston-MacAnanny, E.B., DiLuigi., A.J., Perinatal outcomes in children born after Chiu, K., Liu, J., Lingis, M., Williams, R.S.,
Engmann, L., Maier, D., Benadiva, C., Nulsen, fresh or frozen embryo transfer : a Catalan Rhoton-vlasak, A., Nichols, W.W., Fleischmann,
J. Selection of first in vitro fertilizztion (IVF) cohort study based on 107. Fertil. Steril. 2017; R.R., Zhang, W., Winn, V.D., Segal, M.S.,
cycle stimulation protocol for good prognosis 107: 940–947 Conrad, K.P., Baker, V.L. Preeclampsia
patients: GnRH antagonist vs. agonist Roque, M., Lattes, K., Serra, S., Solà, I., Geber, Increased Preeclampsia Risk and Reduced
protocols. J. Reprod. Med. 2011: 12–16 S., Carreras, R., Checa, M.A. Fresh embryo Aortic Compliance With In Vitro Fertilization
Kaye, L., Marsidi, A., Rai, P., Thorne, J., Nulsen, J., transfer versus frozen embryo transfer in in Cycles in the Absence of a Corpus Luteum.
Engmann, L., Benadiva, C. Frozen blastocyst vitro fertilization cycles: A systematic review Hypertension. 2019; 73: 640–649
transfer outcomes in immediate versus and meta-analysis. Fertil. Steril. 2013; 99: Vidaeff, A., Pettker, C.M., Simhan, H. Gestational
delayed subsequent cycles following GnRH 156–162 Hypertension and Preeclampsia ACOG
agonist or hCG triggers. J. Assist. Reprod. Saito, K., Kuwahara, A., Ishikawa, T., Morisaki, PRACTICE BULLETIN. Clinical Management
Genet. 2018; 35: 669–675 N., Miyado, M., Miyado, K., Fukami, M., Guidelines for Obstetrician-Gynecologists. Am.
Luke, B. Pregnancy and birth outcomes in Miyasaka, N., Ishihara, O., Irahara, M., Saito, Coll. Obstet. Gynecol. 2019; 133: 1–25
couples with infertility with and without H. Endometrial preparation methods for Wennerholm, U., Henningsen, A.A., Romundstad,
assisted reproductive technology: with an frozen-thawed embryo transfer are associated L.B., Bergh, C., Pinborg, A., Skjaerven, R.,
emphasis on US population-based studies. with altered risks of hypertensive disorders of Forman, J., Gissler, M., Go, K., Tiitinen, A.
Am. J. Obstet. Gynecol. 2017; 217: 270–281 pregnancy, placenta accreta, and gestational Perinatal outcomes of children born after
Maheshwari, A., Raja, A., Ph, D., Bhattacharya, diabetes mellitus. Hum. Reprod. 2019: 1–9 frozen-thawed embryo transfer : a Nordic
S., D, M. Obstetric and perinatal outcomes Shi, Y., Sun, Y., Hao, C., Zhang, H., Wei, D., Zhang, cohort study from the CoNARTaS group.
after either fresh or thawed frozen embryo Y., Zhu, Y., Deng, X., Qi, X., Li, H., Ma, X., Ren, Hum. Reprod. 2013; 28: 2545–2553
transfer : an analysis of recorded in the H., Wang, Y., Zhang, D., Wang, B., Liu, F., Wu, Yarali, H., Polat, M., Mumusoglu, S., Yarali, I.,
Human Fertilisation and Embryology Q., Wang, Z., Bai, H., Li, Y., Zhou, Y., Sun, M., Bozdag, G. Preparation of endometrium
Authority anonymized dataset. Fertil. Steril. Liu, H., Li, J., Zhang, L., Chen, X., Zhang, S., for frozen embryo replacement cycles : a
2016; 106: 1703–1708 Sun, X., Legro, R.S., Chen, Z.J. Transfer of systematic review and meta-analysis. J. Assist.
Marshall, S.A., Leo, C.H., Senadheera, S.N., fresh versus frozen embryos in ovulatory Reprod. Genet. 2016: 1287–1304
Girling, J.E., Tare, M., Parry, L.J. Relaxin women. N. Engl. J. Med. 2018; 378: 126–136 Zhang, W.Y., von Versen-Höynck, F., Kapphahn,
deficiency attenuates pregnancy-induced Strauch, N.K., Von Versen-ho, F., Liu, J., Chi, Y., K.I., Fleischmann, R.R., Zhao, Q., Baker, V.L.
adaptation of the mesenteric artery to Keller-woods, M., Conrad, K.P., Baker, V.L. Maternal and neonatal outcomes associated
angiotensin II in mice. Am. J. Physiol. Regul. Effect of Mode of Conception on Maternal with trophectoderm biopsy. Fertil. Steril. 2019;
Integr. Comp. Physiol. 2016; 310: R847–57 Serum Relaxin, Creatinine, and Sodium 112: 283–290
Mol, B.W., Ph, D., Barnhart, K.T. Scratching the Concentrations in an Infertile Population.
Endometrium in In Vitro Fertilization — Time Reproductive Sciences. 2018; 26: 412–419 Received 16 December 2019; received in revised
to Stop. N. Engl. J. Med. 2019: 391–392 Sunderam, S., Kissin, D.M., Zhang, Y., Folger, form 7 March 2020; accepted 16 March 2020.
Palomba, S., Homburg, R., Santagni, S., La Sala, S.G., Boulet, S.L., Warner, L., Callaghan,
G.B., Orvieto, R. Risk of adverse pregnancy W.M., Barfield, W.D. Assisted Reproductive