Singleton term pregnancies resulting from frozen-thawed embryo transfer in hormone replacement cycles increase the risk of aberrant placentation, including velamentous umbilical cord insertion.

Pregnancies achieved through frozen–thawed embryo transfer (FET) in the hormone replacement cycle have a higher risk of developing abnormal placenta, including velamentous umbilical cord insertion and placenta accreta spectrum, than do pregnancies achieved in the ovulatory cycle. Therefore, the advantages and disadvantages of FET in the alternative endometrial preparation regimen of hormone replacement or the ovulatory cycle should be re-evaluated to ensure safety during the perinatal period.

This observational retrospective cohort study was conducted over 8 years from January 2016 to July 2024, in accordance with Japanese laws and STROBE guidelines. Obtaining informed consent from the patients was waived because this study had an observational retrospective cohort design. The study protocols conformed to the provisions of the Declaration of Helsinki (revised in Tokyo, 2004). The study was approved by the Kubonoya Women’s Hospital Ethics Committee and Review Board, which approved the use of an electronic medical record database for this clinical study (approval no. 2024–1). A total of 11,354 consecutive singleton labor and delivery cases managed at Kubonoya Women’s Hospital in Chiba, Japan, were analyzed from the hospital’s obstetric database. From these, only cases conceived via FET ( n  = 1225) were extracted. All FET cases were referred from other fertility treatment clinics, where the endometrial preparation method for each FET cycle was arbitrarily determined. Detailed ART information, including indications of ART treatment, fertilization method (conventional in vitro fertilization [cIVF] or intracytoplasmic sperm injection [ICSI]), developmental embryo stage at transfer (cleavage stage or blastocyst), and endometrial preparation methods, was obtained from a referral letter or an interview survey from each patient. Vitrification was applied to all embryos for cryopreservation. High-risk pregnancies (e.g., higher-order multiple births, preterm labor < 35 weeks of gestation, severe HDP, placenta previa, vasa previa, and pregnancies with uncontrolled medical disease) are not managed in our hospital per the hospital’s policy for patient safety and management. These cases were transferred to a vicinal specialized perinatal center beforehand. We could not obtain sufficient information on referred cases afterward; therefore, we had to exclude higher risk cases from our study population. The inclusion criteria were women with singleton term pregnancy achieved via FET using an autologous oocyte who delivered a live baby at our facility ( n  = 1201). The exclusion criteria were cases with missing or incomplete core data on endometrial preparation methods ( n  = 22), patients who had undertaken preimplantation genetic testing for aneuploidy ( n  = 5), patients with a uterine anomaly deforming the shape of the uterine cavity ( n  = 6), and singleton cases showing two gestational sacs at the early gestation stage because vanishing twins can increase the risk of VCI ( n  = 7) [ 22 ]. A total of 1161 patients who conceived through FET were finally recruited as eligible cases. The patients were allocated to two study groups according to the endometrial preparation methods: NC-FET ( n  = 315, 27.1%) and HRC-FET ( n  = 846, 72.9%). FET with no medication during the cycle and FET using ovulation induction agents followed by human chorionic gonadotropin (hCG) trigger, luteal support, or both were classified into the NC-FET group. FET using estradiol (E2) and progesterone (P4) preparations with or without prior downregulation using GnRH agonists/antagonists was allocated to the HRC-FET group. Information on assisted hatching for FET embryos was unavailable because it was not included in the referral letter or the patients were not informed about it. Perinatal data, including maternal age, gravidity, parity, gestational weeks at birth, pregestational body mass index (BMI), smoking habits during pregnancy, delivery mode, complications during labor and delivery, blood loss volume, blood transfusion, neonatal weight at birth, neonatal sex, condition of the neonate at birth, placental weight, placental shape (e.g., succenturiate placenta), and umbilical cord insertion site, were collected from the electric obstetric database. The attending doctor performed a morphological examination of the delivered placenta at birth and determined the umbilical cord insertion site. VCI is defined by the direct insertion of umbilical vessels into the chorio-amniotic membranes outside the placental margin, running within the membranes to the placental parenchyma. A succenturiate placenta is a placental morphological variation with a smaller accessory lobe separated from the main placental disc. HDP is defined as hypertension (blood pressure ≥ 140/90 mmHg) in pregnancy, which is classified according to a revised Japan Society for the Study of Hypertension in Pregnancy (JSSHP) statement of 2005 [ 23 ]. HDP in this study included gestational hypertension and preeclampsia but excluded chronic hypertension in pregnancy. PAS is diagnosed by pathological examination on the basis of the FIGO classification for clinical diagnosis of PAS and by the condition in which the retained placenta requires manual removal, interventional radiology, or surgical treatment [ 24 ]. SGA and LGA infants are defined as infants with a birth weight below the 10th percentile and above the 90th percentile of the Japanese reference curve, respectively. The primary outcomes of this study were the incidence rate of VCI, HDP, and PAS, which results from abnormal placentation. The secondary outcomes were obstetric (delivery mode, placental abruption, abnormal blood loss during labor, and blood transfusion) and neonatal outcomes (weight at birth, sex, conditions at birth, and congenital anomalies). The maternal characteristics and perinatal outcomes were compared between the NC-FET and HRC-FET groups. All statistical data analyses were performed using EZR software (version 1.65) [ 25 ] and a Microsoft Excel (2018) spreadsheet. Continuous variables are expressed as the mean ± standard deviation, and the Student’s t test or Mann–Whitney U test was used for comparisons between the two groups. Categorical variables are described as frequencies and percentages, which were compared between groups using Pearson’s Chi-squared or Fisher’s exact test. Univariate and multivariate logistic regression analyses were conducted to evaluate the effect of HRC-FET on the incidence of perinatal outcomes, including VCI, HDP, and PAS, compared to that of NC-FET. The confounding factors arising from the inherent differences between the two applied comparative groups were adjusted. Confounding factors included maternal age, parity, maternal BMI, smoking habits, fetal sex, fertilization method, and developmental stage of the embryo at transfer. The results were described as the crude odds ratio (OR), adjusted OR (aOR), 95% confidence interval (CI), and a two-sided P value of < 0.05 was considered statistically significant.

Maternal baseline characteristics, including maternal age at conception, gravidity, parity, maternal BMI at conception, smoking habits during pregnancy, causes of infertility, history of cesarean delivery, fertilization method for embryo production, and developmental stage of the embryo at transfer, are summarized in Table  1 . The average maternal age of the NC-FET group was slightly higher than that of the HRC-FET group. Other demographic parameters were comparable between groups. Table 1 Maternal demographics of patients conceiving through HRC-FET and NC-FET Characteristic HRC-FET group ( n  = 846) NC-FET group ( n  = 315) P value Age (years) 35.7 ± 3.8 36.2 ± 3.6 0.038  (Range) (24–47) (24–44) Gravidity Nulligravida 388 (45.9) 140 (44.5) 0.691 Multigravida (range) 458 (2–7) 175 (2–6) Number of pregnancies 2 266 (31.4) 105 (33.3) 3 128 (15.1) 41 (13.0) 4 41 (4.8) 15 (4.8) 5 15 (1.8) 12 (3.8) 6 5 (0.6) 2 (0.6) 7 3 (0.4) 0 (0) Parity Nullipara 552 (65.2) 206 (65.4) 1.000 Multipara (range) 294 (1–3) 109 (1–2) Number of births 1 279 (33.0) 101 (32.1) 2 12 (1.4) 8 (2.5) 3 3 (0.4) 0 (0) Causes of infertility a 0.119 Male factor 189 (22.3) 82 (26.0) Ovulation disorder 67 (7.9) 12 (3.9) Tubal factor 23 (2.7) 11 (0.3) Endometriosis 82 (9.7) 35 (11.1) Habitual abortion 5 (0.6) 2 (0.6) Unexplained infertility 501 (59.2) 184 (58.4) BMI (kg/m 2 ) 21.6 ± 3.1 20.8 ± 2.7  < 0.001 Smoking habit 5 (0.6) 0 (0.0) 0.332 Previous cesarean delivery 74 (8.7) 21 (6.7) 0.280 Fertilization method cIVF/ICSI 359/487 117/198 0.180 Developmental stage of embryo at transfer Cleavage stage/blastocyst 110/736 24/291 0.001 The results are presented as follows: categorical data are expressed as frequencies (n) and percentages (%), and continuous variables are expressed as the mean ± standard deviation P  < 0.05 was considered statistically significant HRC-FET frozen–thawed embryo transfer in a programmed hormone replacement cycle, NC-FET frozen–thawed embryo transfer in an ovulatory cycle, BMI body mass index, cIVF conventional in vitro fertilization, ICSI intracytoplasmic sperm injection a Double counting allowed Maternal demographics of patients conceiving through HRC-FET and NC-FET The results are presented as follows: categorical data are expressed as frequencies (n) and percentages (%), and continuous variables are expressed as the mean ± standard deviation P  < 0.05 was considered statistically significant HRC-FET frozen–thawed embryo transfer in a programmed hormone replacement cycle, NC-FET frozen–thawed embryo transfer in an ovulatory cycle, BMI body mass index, cIVF conventional in vitro fertilization, ICSI intracytoplasmic sperm injection a Double counting allowed Pregnancy outcomes of the two study groups are shown in Table  2 . Gestational days at birth, HDP, placental abruption, placental weight, succenturiate placenta, distribution of neonatal sex, SGA and LGA infants, neonatal conditions at birth, congenital anomalies, and neonatal intensive care unit admission were comparable between groups. Patients with VCI, PAS, abnormal blood loss during labor and delivery, blood transfusion, emergency cesarean delivery, and instrumental delivery were observed more in the HRC-FET group than in the NC-FET group. Table 2 Pregnancy outcomes in patients who underwent HRC-FET and NC-FET Characteristic HRC-FET ( n  = 846) NC-FET ( n  = 315) P value Gestational age (days) 277.0 ± 7.9 276.0 ± 7.2 0.064 Emergency cesarean delivery 204 (24.1) 52 (16.5) 0.005 Instrumental delivery 212 (25.1) 53(16.8) 0.003 Hypertensive disorders of pregnancy (HDP) 77 (9.1) 21 (6.7) 0.194 Abnormal blood loss during labor/delivery a 212 (25.1) 39 (12.4)  < 0.001 Blood transfusion 40 (4.7) 4 (1.3) 0.005 Placenta accreta spectrum (PAS) 43 (5.1) 3 (1.0)  < 0.001 Placental abruption 13 (1.5) 1 (0.3) 0.129 Placental weight (g) 636.7 ± 122.6 608.9 ± 110.0  < 0.001 Velamentous umbilical cord insertion (VCI) 59 (7.0) 8 (2.5) 0.003 Succenturiate placenta 6 (0.7) 1 (0.3) 0.681 Neonatal sex  Male/Female 436/410 157/158 0.644 Neonatal weight at birth (g) 3147.2 ± 365.2 3074.1 ± 356.1 0.002 Small for gestational age infant b 35 (4.1) 14 (4.4) 0.870 Large for gestational age infant b 147 (17.3) 36 (11.4) 0.014 Neonatal asphyxia c 27 (3.2) 13 (4.1) 0.470 Major congenital anomaly d 12 (1.4) 2 (0.6) 0.347 NICU admission 31 (3.7) 7 (2.2) 0.268 The results are presented as follows: categorical data are expressed as frequencies (n) and percentages (%), and continuous variables are expressed as the mean ± standard deviation P  < 0.05 was considered statistically significant HRC-FET frozen–thawed embryo transfer in a programmed hormone replacement cycle, NC-FET frozen-thawed embryo transfer in an ovulatory cycle, BMI body mass index, cIVF conventional in vitro fertilization, ICSI intracytoplasmic sperm injection, NICU neonatal intensive care unit a Abnormal blood loss during labor was defined by blood loss of ≥ 800 mL and ≥ 1400 mL within 2 h after vaginal and cesarean delivery, respectively b Small and large for gestational age infants were defined as infants with a birth weight below the 10th percentile and above the 90th percentile of the Japanese reference curve, respectively c Neonatal asphyxia was the condition of neonates with low umbilical artery pH value (< 7.100) and/or low Apgar score (< 7 at 5 min after birth), necessitating resuscitation d Major congenital anomaly was defined as the structural or functional abnormality recognized at birth, requiring therapeutic interventions Pregnancy outcomes in patients who underwent HRC-FET and NC-FET The results are presented as follows: categorical data are expressed as frequencies (n) and percentages (%), and continuous variables are expressed as the mean ± standard deviation P  < 0.05 was considered statistically significant HRC-FET frozen–thawed embryo transfer in a programmed hormone replacement cycle, NC-FET frozen-thawed embryo transfer in an ovulatory cycle, BMI body mass index, cIVF conventional in vitro fertilization, ICSI intracytoplasmic sperm injection, NICU neonatal intensive care unit a Abnormal blood loss during labor was defined by blood loss of ≥ 800 mL and ≥ 1400 mL within 2 h after vaginal and cesarean delivery, respectively b Small and large for gestational age infants were defined as infants with a birth weight below the 10th percentile and above the 90th percentile of the Japanese reference curve, respectively c Neonatal asphyxia was the condition of neonates with low umbilical artery pH value (< 7.100) and/or low Apgar score (< 7 at 5 min after birth), necessitating resuscitation d Major congenital anomaly was defined as the structural or functional abnormality recognized at birth, requiring therapeutic interventions Table 3 shows the incidence, univariate, and multivariate analysis of VCI classified by the three ART-related variables, including endometrial preparation method for FET, fertilization method to produce an embryo, and developmental stage of the embryo at transfer in FET cycles. The incidence of VCI in HRC-FET was significantly higher than that in NC-FET. The VCI incidence rate in blastocyst transfer was also higher than that in cleavage-stage embryo transfer. HRC-FET (aOR: 3.07, 95% CI: 1.44–6.54, P  = 0.004) and blastocyst transfer in the FET cycle (aOR: 4.69, 95% CI: 1.13–19.50, P  = 0.034) were two independent predictive factors for the placenta developing VCI. However, the fertilization method (cIVF or ICSI) did not show any statistical difference in terms of the VCI incidence rate (aOR: 0.74, 95% CI: 0.45–1.22, P  = 0.242). Table 3 Incidence, univariate, and multivariate analyses for VCI classified by ART-related predictor variables in FET cycles Predictor variable Incidence rate of VCI (number (%) of participants) Crude OR (95% CI, P value) Adjusted OR (95% CI, P value) a Endometrial preparation method for FET  NC-FET ( n  = 315) 8 (2.5) Reference Reference  HRC-FET ( n  = 846) 59 (7.0) 2.87 (1.35–7.05, 0.003) 3.07 (1.44–6.54, 0.004) Fertilization method  cIVF ( n  = 476) 33 (6.9) Reference Reference  ICSI ( n  = 685) 34 (5.0) 0.70 (0.41–1.19, 0.161) 0.74 (0.45–1.22, 0.242) Developmental stage of embryo at FET  Cleavage stage ( n  = 134) 2 (1.5) Reference Reference  Blastocyst ( n  = 1027) 65 (6.3) 4.46 (1.16–38.04, 0.018) 4.69 (1.13–19.50, 0.034) P  < 0.05 was considered statistically significant VCI velamentous umbilical cord insertion, ART assisted reproductive technology, OR odds ratio, CI confidence interval, FET frozen–thawed embryo transfer, NC-FET frozen–thawed embryo transfer in an ovulatory cycle, HRC-FET frozen–thawed embryo transfer in a programmed hormone replacement cycle, cIVF conventional in vitro fertilization, ICSI intracytoplasmic sperm injection a Adjusted for maternal age, parity, maternal body mass index, smoking habit, and neonatal sex Incidence, univariate, and multivariate analyses for VCI classified by ART-related predictor variables in FET cycles P  < 0.05 was considered statistically significant VCI velamentous umbilical cord insertion, ART assisted reproductive technology, OR odds ratio, CI confidence interval, FET frozen–thawed embryo transfer, NC-FET frozen–thawed embryo transfer in an ovulatory cycle, HRC-FET frozen–thawed embryo transfer in a programmed hormone replacement cycle, cIVF conventional in vitro fertilization, ICSI intracytoplasmic sperm injection a Adjusted for maternal age, parity, maternal body mass index, smoking habit, and neonatal sex Table 4 indicates univariate and multivariate logistic regression analyses of pregnancy outcomes classified according to endometrial preparation methods for FET. HRC-FET was an independent risk factor for PAS (aOR: 5.73, 95% CI: 1.76–18.70, P  = 0.004), abnormal blood loss during labor and delivery (aOR: 2.35: 95% CI: 1.61–3.41, P  < 0.001), blood transfusion (aOR: 3.93, 95% CI: 1.38–11.10, P  = 0.010), emergency cesarean section (aOR: 1.63, 95% CI: 1.14–2.33, P  = 0.007), and instrumental delivery (aOR: 1.70, 95% CI: 1.21–2.40, P  = 0.003). HRC-FET was not a risk factor for HDP (aOR: 1.24, 95% CI: 0.74–2.07, P  = 0.421), placental abruption (aOR: 5.03, 95% CI: 0.65–38.90, P  = 0.122), and succenturiate placenta (aOR: 2.16, 95% CI: 0.25–18.50, P  = 0.481). HRC-FET was a risk factor for LGA; however, this finding was not reconfirmed after adjusting for confounding factors (aOR: 1.39, 95% CI: 0.94–2.08, P  = 0.103). Table 4 Univariate and multivariate analyses of pregnancy outcomes classified by endometrial preparation methods for FET Pregnancy outcome Crude OR (95% CI, P value) Adjusted OR (95% CI, P value) a Hypertensive disorders of pregnancy (HDP)  NC-FET Reference Reference  HRC-FET 1.40 (0.84–2.44, 0.194) 1.24 (0.74–2.07, 0.421) Placenta accreta spectrum (PAS)  NC-FET Reference Reference  HRC-FET 5.57 (1.82–17.05, 0.001) 5.73 (1.76–18.70, 0.004) Placental abruption  NC-FET Reference Reference  HRC-FET 4.90 (0.73 – 208.74, 0.129) 5.03 (0.65–38.90, 0.122) Succenturiate placenta  NC-FET Reference Reference  HRC-FET 2.24 (0.35–14.23, 0.681) 2.16 (0.25–18.50, 0.481) Abnormal blood loss during labor/delivery b  NC-FET Reference Reference  HRC-FET 2.37 (1.62–3.52, < 0.001) 2.35 (1.61–3.41, < 0.001) Blood transfusion  NC-FET Reference Reference  HRC-FET 3.86 (1.38–14.95, 0.005) 3.93 (1.38–11.10, 0.010) Instrumental delivery  NC-FET Reference Reference  HRC-FET 1.65 (1.17–2.36, 0.003) 1.70 (1.21–2.40, 0.003) Emergency cesarean delivery  NC-FET Reference Reference  HRC-FET 1.61 (1.14–2.30, 0.005) 1.63 (1.14–2.33, 0.007) Small for gestational age infant c  NC-FET Reference Reference  HRC-FET 0.93 (0.48–1.89, 0.870) 1.01 (0.53–1.91, 0.982) Large for gestational age infant c  NC-FET Reference Reference  HRC-FET 1.63 (1.09–2.48, 0.014) 1.39 (0.94–2.08, 0.103) P  < 0.05 was considered statistically significant FET frozen-thawed embryo transfer, OR odds ratio, CI confidence interval, NC-FET frozen-thawed embryo transfer in an ovulatory cycle, HRC-FET frozen-thawed embryo transfer in a programmed hormone replacement cycle a Adjusted for maternal age, parity, maternal body mass index, smoking habit, and infant sex b Abnormal blood loss during labor/delivery was defined by blood loss of ≥ 800 mL and ≥ 1400 mL within 2 h after vaginal and cesarean delivery, respectively c Small and large for gestational age infants were defined as infants with a birth weight below the 10th percentile and above the 90th percentile of the Japanese reference curve, respectively Univariate and multivariate analyses of pregnancy outcomes classified by endometrial preparation methods for FET P  < 0.05 was considered statistically significant FET frozen-thawed embryo transfer, OR odds ratio, CI confidence interval, NC-FET frozen-thawed embryo transfer in an ovulatory cycle, HRC-FET frozen-thawed embryo transfer in a programmed hormone replacement cycle a Adjusted for maternal age, parity, maternal body mass index, smoking habit, and infant sex b Abnormal blood loss during labor/delivery was defined by blood loss of ≥ 800 mL and ≥ 1400 mL within 2 h after vaginal and cesarean delivery, respectively c Small and large for gestational age infants were defined as infants with a birth weight below the 10th percentile and above the 90th percentile of the Japanese reference curve, respectively

To the best of our knowledge, this study is the first to investigate the effect of the two different endometrial preparation methods in FET cycles on the prevalence rate of VCI among women achieving pregnancy through FET. Three-fourths of the FET-conceived pregnancies managed at our facility used hormone replacement cycles. This trend is the same at other obstetric facilities in Japan and in the U.S. [ 26 , 27 ]. However, in this study, pregnancies achieved via HRC-FET led to a higher incidence of developing VCI and PAS than those achieved by NC-FET, resulting in adverse prenatal outcomes such as increased cases of emergency cesarean/instrumental delivery, abnormal blood loss during labor, and blood transfusion. In the present study, HRC-FET and blastocyst transfer were independent predictive factors for developing the placenta with VCI in FET cycles. The underlying cause of the higher incidence of VCI in pregnancies resulting from HRC-FET and blastocyst transfer compared to that resulting from NC-FET remains unclear. Macroscopic abnormal placentation, typified by VCI, increases in singleton pregnancies resulting from ART (approximately 6–14%) compared to that in natural singleton pregnancies (approximately 0.23–1.6%) [ 28 – 30 ]. Furthermore, the incidence of VCI in twin pregnancies (approximately 1.6–40%) is higher than that in singleton pregnancies [ 30 ]. Notably, ART treatment is a significant protective factor against VCI in twin pregnancies achieved through ART, compared to ones achieved by natural conception (OR: 0.45, 95% CI: 0.21–0.95) [ 31 ]. HRC-FET does not affect the incidence rate of morphological placental abnormalities compared to NC-FET or natural conception [ 17 ]. Some studies have investigated the influence of various ART procedures on the development of VCI. A recent meta-analysis showed that ART was a risk factor for pregnancies with VCI compared to those achieved through natural conception (OR: 2.14, 95% CI: 1.64–2.79) and ART types (fresh embryo vs. frozen–thawed and cleavage-stage embryo vs. blastocyst) did not affect the VCI incidence rate [ 16 ]. ICSI can lead to a higher VCI prevalence than cIVF [ 28 ]. Moreover, blastocyst transfer is associated with a lower VCI incidence rate than cleavage-stage embryo transfer in fresh embryo transfer cycles (OR: 0.5, 95% CI: 0.3–0.9) [ 32 ]. In contrast, in a previous study, blastocyst transfer increased the risk of VCI compared to that with cleavage-stage embryo transfer (OR: 4.33, 95% CI: 1.86–12.68). However, cIVF and ICSI did not affect the VCI incidence rate in this study (OR: 1.16, 95% CI: 0.65–2.12) [ 15 ]. These controversial links between VCI and ART could be explained by the in vitro embryo culture environment, micromanipulation procedures, and blastocyst transfer [ 33 , 34 ]. Blastocyst orientation during implantation can be impeded by the artificial displacement of cultured embryos, which could result in the development of VCI [ 34 , 35 ]. Furthermore, the altered hormone milieu during implantation could change endometrial gene expression patterns, resulting in abnormal implantation and placentation [ 36 ]. Epigenomic modifications, which occur more frequently during blastocyst culture than during cleavage-stage embryo culture, can alter normal placentation [ 37 ]. Endometrial thickness in the HRC-FET cycle is a crucial index for abnormal placentation during embryo implantation [ 38 ]. Therefore, the recent increase in blastocyst transfers in the HRC-FET cycle is concerning because it creates unfavorable opportunities for forming the placenta with VCI, which is directly linked to perinatal outcomes [ 30 , 39 ]. To ensure the safety of both the mother and fetus, we recommend proactively identifying the umbilical cord insertion site during routine prenatal ultrasound examinations. This recommendation is particularly relevant, as our findings demonstrate an increased risk of complications even in a low-risk population as well as in a high-risk population, highlighting the need for vigilance across all risk categories. However, further research is required to answer the following two important clinical questions: (1) Does earlier and accurate detection of VCI directly improve perinatal outcomes? (2) How should current prenatal management strategies be adapted following the identification of VCI? In this study, the risks of PAS and post-partum massive hemorrhage resulting in blood transfusion were higher in pregnancies achieved via HRC-FET than in those resulting from NC-FET, similar to the findings of other studies [ 10 , 40 – 42 ]. The HRC-FET cycle requires external medication, which may create a less favorable physiological endometrial environment for blastocyst implantation [ 43 ]. P4 induces endometrial stromal cell decidualization and regulates extravillous trophoblast (EVT) invasion and maternal vascular reconstruction [ 44 ]. Insufficient decidualization and excessive EVT invasion into the myometrium, often caused by lower P4 levels in the HRC-FET cycle, can lead to PAS disorders [ 14 ]. Furthermore, previous uterine surgery, including cesarean delivery, myomectomy, and uterine curettage, is also an independent maternal risk factor for PAS in HRC-FET [ 45 ]. HDP occurs more in pregnancies conceived via HRC-FET than in those resulting from NC-FET [ 10 , 40 – 42 , 46 ]. Abnormal maternal vascular remodeling and adaptation during the early gestation period can cause HDP because of the absence of the corpus luteum (CL) [ 47 ]. Vasoactive substances produced from the CL, such as relaxin and vascular endothelial growth factor, are essential for initiating normal placentation [ 48 – 51 ]. However, HRC-FET cycles occur without the CL; therefore, pregnancies achieved via the HRC-FET cycle could hamper healthy placentation in early pregnancy, causing HDP in the later stages of pregnancy [ 10 , 12 , 51 , 52 ]. In contrast to these findings, this study did not show an increased risk of HDP in HRC-FET cycles. However, considering the restricted inclusion criteria, many HDP cases may have been overlooked because the study only included term cases and excluded high-risk pregnancies, such as those with severe HDP, chronic hypertension, and cases with uncontrollable medical conditions that are prone to complications from HDP. The strength of this study is that the study population originated from complete consecutive perinatal cohort records of singleton term labor and delivery cases treated under uniform management policies at a single obstetric facility. Furthermore, the information collected under stringent inclusion and exclusion criteria from a single hospital database was correct and valid. Therefore, to the best of our knowledge, this study is the first to elucidate the direct relationship between HRC-FET and VCI. However, this study also has some limitations. First, the difference in baseline characteristics between the comparative groups could not be excluded because of the intrinsic nature of the observational retrospective cohort study. Only low-risk pregnancies (low-risk in appearance) were included in our study population because of our hospital policy of patient management. In particular, cases with placenta previa and vasa previa, which are highly associated with pregnancies achieved by ART, were excluded. There was also a large difference in the number of cases between the HRC-FET and NC-FET groups because of the recent trend of increased preference for HRC-FET treatment over NC-FET in Japan. Furthermore, cases with chronic anovulation, oligomenorrhea, polycystic ovary syndrome (PCOS), and premature ovarian insufficiency (POI) were usually allocated to the HRC-FET group [ 19 ]. PCOS cases are especially complicated by gestational diabetes mellitus more often than non-PCOS cases, which may have some adverse effects on perinatal outcomes. Therefore, our study population has an original selection bias, making it challenging to standardize the findings to different populations, and the results need to be interpreted with caution. Second, heterogeneity involving various protocols and medications used for NC-FET and HRC-FET was observed. The NC-FET group included cycles with no medication and cycles treated with or without ovulation induction agents, hCG triggers, and P4 administration. For the HRC-FET group, the route and dosage of supplemented E2/P4 differed among the cases. Furthermore, the reason for ART clinics selecting a specific protocol was not described in the referral letter, and this information could not be acquired easily from the patients. In Japan, an increasing number of patients are undergoing the HRC-FET cycle as a viable method. However, the lack of a consistent policy across ART clinics makes conducting randomized controlled studies challenging. Third, this study spanned 8 years, during which improvements and changes in the FET procedure could have influenced outcomes. Therefore, an extended research period is necessary to analyze the effects of relatively uncommon events such as VCI and PAS, ensuring a sufficient number of relevant cases for analysis. Fourth, misclassification may occur because patients using NC-FET who received P4 and E2 in the luteal phase can be classified as HRC-FET [ 35 ]. In addition, an interview survey sometimes conveys incorrect information owing to misunderstanding or lapse in memory, thereby causing information bias. In conclusion, this study demonstrated that conceptions achieved through the HRC-FET cycle have a higher risk of being affected by VCI as well as abnormal placentation, including PAS. As these hazardous complications are observed even in low-risk (initially low-risk in appearance) singleton term pregnancies resulting from HRC-FET, it must be assumed that their prevalence and severity would be even higher in high-risk populations. Therefore, these pregnancies should be managed as high-risk from the outset of conception. Practitioners engaged in ART treatment must seriously consider these results for patient safety during the perinatal period. HRC-FET is more convenient for patients and practitioners than NC-FET; however, whether HRC-FET is the better choice for patients must be evaluated. Considering the diverse background of each patient (e.g., age, cause of infertility, history of ART treatment, and patient’s desire), we should prioritize selecting optimal patients for ART treatment with NC-FET more proactively rather than relying on HRC-FET. Standardizing the treatment policy will directly minimize the sequela accompanied by HRC-FET cycles. However, a large-scale randomized controlled study between HRC-FET and NC-FET, ideally by selecting only FET cases with regular ovulatory cycles, is required to verify the novel findings of this study more comprehensively. Further investigation is also needed to confirm the close relationship between abnormal placentation and the development of the placenta with VCI.

The transition from slow-freezing techniques to vitrification for embryo and oocyte preservation has marked the advent of a new era in assisted reproductive technology (ART). The primary benefits of vitrification are improved gamete and embryo survival rates, cumulative pregnancy rates, and ART safety through single embryo transfer to avoid multiple gestation [ 1 ]. Recently, many completed oocyte retrieval cycles used a “freeze-all” strategy using vitrification to eliminate ovarian hyperstimulation syndrome. These cycles were followed by frozen–thawed embryo transfer (FET). In Japan, over 90% of newborns conceived through ART in 2021 were born following FET [ 2 ], and approximately 80% of all egg retrieval cycles started in 2022 in the U.S. used embryo cryopreservation for subsequent FET [ 3 ]. Therefore, reappraising the value of FET as a central component in contemporary ART treatment is crucial. FET is a beneficial technique for most patients treated with ART [ 4 ]. Pregnancies resulting from FET have advantages over fresh embryo transfer, such as lower risks of preterm birth and small for gestational age (SGA) infants, lower incidence of placenta previa and abruption, and lower perinatal mortality [ 5 – 8 ]. However, FET can also lead to a higher incidence of large for gestational age (LGA) infants, increased risks of hypertensive disorders of pregnancy (HDP), and placenta accreta spectrum (PAS) [ 5 – 11 ]. The pathophysiology of HDP and PAS is examined from the perspective of microscopic abnormal placentation resulting from FET [ 12 – 14 ]. Pregnancies achieved by ART have more macroscopic (morphological) abnormal placentation than unassisted natural pregnancies, typified by a placenta with velamentous umbilical cord insertion (VCI). Additionally, FET is not directly associated with the increased incidence of VCI compared to fresh embryo transfer [ 15 ]. However, the relationship between FET and macroscopic abnormal placentation is unclear [ 16 , 17 ]. Two endometrial preparation protocols exist for the FET cycle: FET in the ovulatory cycle (NC-FET) and the programmed hormone replacement cycle (HRC-FET). Although NC-FET has a higher chance of clinical pregnancy and live birth than HRC-FET [ 18 , 19 ], HRC-FET is more widely and readily applied than NC-FET because of its convenience for patients and practitioners [ 20 , 21 ]. Thus, we hypothesized that the difference in endometrial preparation methods in the FET cycle would affect the etiopathogenesis of abnormal placentation such as VCI, HDP, and PAS. This study aimed to determine whether NC-FET and HRC-FET affect the incidence of VCI, HDP, and PAS and to assess the effects of each protocol on prenatal/neonatal outcomes.