The obstetrical and infant outcomes of trophectoderm biopsy on preimplantation genetic testing embryos after frozen single embryo transfer cycles

Abstract

Purpose

To investigate if trophectoderm (TE) biopsy on preimplantation genetic testing (PGT) increases the risk of obstetrical and infant outcomes after frozen single blastocyst transfer cycles.

Methods

This retrospective cohort study included all frozen single blastocyst transfer cycles conducted between January 2012 and December 2017 at Peking University Third Hospital. A total of 1492 patients in the PGT group and 600 patients in the non-PGT group were analyzed. The primary outcomes were clinical pregnancy rate, live birth rate, and early motor development milestones. Secondary outcomes included miscarriage, obstetric complications, and infant outcomes.

Results

The PGT group had younger parental ages and higher antral follicle counts, whereas the non-PGT group had lower sperm quality. The clinical pregnancy rate (52.7 vs. 45.2%, p = 0.002) and live birth rate (44.0 vs. 32.8%, p < 0.001) were significantly higher in the PGT group, while the miscarriage rate (13.6 vs. 22.5%, p < 0.001) was lower. In terms of early motor development, infants in the PGT group achieved milestones such as independent sitting, hand and knee crawling, and standing earlier (p < 0.001), although the timing of independent walking did not significantly differ between groups (p = 0.304). No significant differences were found between the two groups regarding pregnancy complications, infant outcomes, or early motor development. These findings were further confirmed through adjusted analyses.

Conclusion

PGT was associated with higher live birth rate and lower miscarriage rate in first single blastocyst transfer cycles. Blastocyst biopsy may not increase the risk of pregnancy complications or negatively impact early motor development in infants.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-025-03512-7.

Introduction

Preimplantation genetic testing (PGT) has advanced rapidly in recent years, becoming a crucial intervention for preventing heritable genetic disorders, identifying chromosomal abnormalities, and thereby reducing the risk of recurrent miscarriages and implantation failure (RIF) [1, 2]. In its early days, blastomere biopsy was used on six to eight cell embryos [3]. However, some studies suggest that blastomere biopsy may negatively impact embryonic development and reduce implantation success, leading to a shift toward alternative methods [4, 5].

The development of embryo culture and vitrification techniques has made trophectoderm biopsy the preferred approach. This method targets cells destined to form the placenta, while leaving the inner cell mass—the future fetus—untouched, and has gained widespread acceptance due to its presumed safety for embryonic development [6]. Despite its growing popularity, PGT imposes certain burdens on patients, including financial costs, emotional stress, and the invasive nature of the procedure. Moreover, the uncertainty of results in cases of mosaicism, segmental abnormalities, or rare findings such as haploidy and triploidy highlights the need to carefully evaluate whether patients truly benefit from in vitro fertilization (IVF) treatment.

Most previous studies have focused on short-term pregnancy outcomes such as clinical pregnancy rates, miscarriage rates, and live birth rates. However, limited research has explored the relatively long-term effects of PGT on neonatal and early childhood development, particularly regarding motor outcomes in neonates [7]. The potential impact of PGT on early motor development remains a novel area of investigation, with few studies having provided substantial data on this aspect.

This study aims to address this gap by examining early neonatal motor outcomes following PGT, offering evidence to inform clinical practice. Additionally, this retrospective cohort study provides a comprehensive assessment of obstetrical and infant outcomes, including clinical pregnancy, live birth rates, miscarriage rates, and pregnancy complications. A particular focus is placed on whether trophectoderm biopsy contributes to adverse outcomes in frozen single blastocyst transfer cycles.

Materials and methods

Study design and population

This retrospective cohort study was approved by the Ethics Committee of Peking University Third Hospital (No. IRB00006761-M2021612) and was conducted using data from the assisted reproductive technology (ART) medical record database of Peking University Third Hospital from January 2012 to December 2017. Infant follow-up was conducted through December 2021.

The study included patients who did not receive fresh cycle embryo transfer and underwent their first vitrified frozen single blastocyst transfer cycles. A total of 1492 patients in the PGT group and 600 patients in the ICSI (unbiopsied) group were eligible for inclusion. Cycles were excluded if any of the following criteria were met: (i) the oocyte/spermatozoa were donated; (ii) the oocyte had undergone in-vitro maturation (IVM) or assisted oocyte activation (AOA); (iii) two blastocysts were transferred; (iv) the patient was lost to follow-up. Data on patient characteristics, treatment procedures, obstetrical outcomes, and infant outcomes for each cycle were collected from the clinical database.

Ovarian stimulation and oocyte retrieval

All women underwent ovarian stimulation using either gonadotrophin-releasing hormone (GnRH) agonist or GnRH antagonist protocols [8]. Final oocyte maturation was triggered by injecting 250 μg human chorionic gonadotropin (hCG) (Choriogonadotropin alfa, Merck Serono, Germany) when at least two dominant follicles measuring over 18 mm in diameter were observed by ultrasound. For dual trigger cycles, a combination of 250 μg/125 μg hCG and GnRH agonist (GnRHa) 0.2 mg was used. For a high-risk population of ovarian hyperstimulation syndrome (OHSS) in GnRH antagonist cycles, 0.2 mg GnRHa was used as the final trigger. Oocyte retrieval was carried out 36–38 h after the trigger.

ICSI and embryo culture

All oocytes were fertilized through ICSI. The cumulus-oocyte complexes (COC) were digested in the culture medium containing hyaluronidase, and granulosa cells were removed by blowing with the capillary glass tube. Then the mature oocytes of the MII stage were obtained. Normal fertilization was assessed by the presence of two pronuclei 16–18 h after sperm injection. All embryos were cultured in 20 Ll G-1 PLUS™ medium (Vitrolife, Sweden) covered with mineral oil (Vitrolife, Sweden) at 37 ℃, 6% CO₂, and 100% humidity environment inside the embryo incubator (Thermo Fisher, USA) [9].

Blastocyst biopsy, vitrification, and warming

Embryos were cultured to days 5–7, the fully developed blastocysts were transferred into biopsy buffer medium (G-MOPSTM PLUS, Vitrolife, Sweden); 5–10 cells were obtained from trophectoderm biopsy. The quality of embryos was evaluated by three aspects: the size of blastocoel, inner cell mass, and trophoblasts condition [10]. All blastocysts were vitrified, and the biopsied trophectoderm cells were washed with phosphate-buffered saline (PBS), stored in lysis buffer in 0.2-mL polymerase chain reaction (PCR) tube, and then subjected to genetic analysis [11].

Embryo transfer and follow-up

Our primary outcomes were clinical pregnancy rate, live birth rate, and early motor development milestones. Secondary outcomes included miscarriage, preterm delivery, ectopic pregnancy, monozygotic twins, cesarean section, newborn weight, newborn length, hypertensive disorder complicating pregnancy, and gestational diabetes mellitus. Frozen blastocysts were transferred in natural or programmed cycles. The serum hCG level was evaluated 12–14 days after embryo transfer, with levels above 30 mIU/mL considered positive for pregnancy tests. Clinical pregnancy was confirmed by ultrasound detection of a gestational sac in the uterus 4 weeks after frozen embryo transfer (FET). Monozygotic twin pregnancies were also evaluated by ultrasound during the first trimester. In accordance with the World Health Organization’s definition of the lower limit of fetal viability, we defined live birth as the delivery of a live infant after 22 weeks of gestation. Miscarriage was defined as the spontaneous loss of an intrauterine pregnancy prior to 22 completed weeks of gestation, with early miscarriage specifically referring to fetal loss occurring before 12 weeks of gestation or with a birth weight > 500 g [12]. Preterm delivery is defined as deliveries between 28 and 36 + 6 gestational weeks. Hypertensive disorder complicating pregnancy (HDCP) and gestational diabetes mellitus (GDM) were diagnosed according to standard obstetric criteria. The diagnosis of HDCP was based on the criteria established by the American College of Obstetricians and Gynecologists (ACOG), encompassing gestational hypertension, preeclampsia, and eclampsia [13]. GDM was diagnosed using the one-step approach with a 75-g oral glucose tolerance test. GDM was confirmed if any of the following thresholds were met or exceeded: fasting ≥ 5.1 mmol/L, 1 h ≥ 10.0 mmol/L, or 2 h ≥ 8.5 mmol/L [14].

According to the routine process at our center, all clinical pregnancies were followed up by trained nurses through telephone calls within 2 months after the expected date of childbirth. The follow-up contained obstetric complications, date of birth, newborn weight/lengths, sex of babies, and any birth defects. For the early motor development, additional phone calls were made by our researchers in the study period from 2019 to 2021. Patients were asked about their children’s early motor development milestones and their answers were recorded. All the results were compiled into a comprehensive database, which served as the foundation for our study. We focused on early motor milestones, as they are objective indicators of neuromuscular development and are typically achieved within the first 24 months of life, according to the World Health Organization (WHO) [15].

Statistical analysis

SPSS 26.0 (IBM, Armonk, NY, USA) was used for descriptive statistics. For continuous variables, t-tests or Mann–Whitney U tests were used based on data distribution, while categorical variables were analyzed using chi-square or Fisher’s exact tests as appropriate. A non-conditional multiple logistic regression model was used for multivariate analysis. P < 0.05 was considered statistically significant.

Results

Basic characteristics of patients in two groups

A total of 2092 FET cycles with single vitrified-warmed blastocyst transfer were studied. There were 1492 patients in the PGT group, and 600 patients in the ICSI (unbiopsied) group. Women and their partners in the PGT group were slightly younger. The BMI of couples and baseline female hormone levels were similar between the two groups, although AFC was higher in the non-PGT group. The average endometrial thickness in FET was also greater in the non-PGT group. The proportions of programmed cycles in both groups are similar. Additionally, the non-PGT group exhibited a higher rate of primary infertility (69.7 vs. 32.6%) and poorer semen quality (Table 1).

Table 1.

Basic characteristics of patients in the two groups

	PGT group (n = 1492)	Non-PGT group (n = 600)	P-value
Female age	31.0 ± 4.4	31.8 ± 4.8	0.001
Male age	32.4 ± 5.2	33.4 ± 6.4	0.002
Female BMI (kg/m2)	22.5 ± 3.6	22.2 ± 3.5	0.128
Male BMI (kg/m2)	25.5 ± 3.7	25.7 ± 4.0	0.216
Type of infertility
Primary infertility rate (%)	32.6 (487/1492)	69.7 (418/600)	< 0.001
Secondary infertility rate (%)	67.4 (1005/1492)	30.3 (182/600)
FSH (mIU/mL)	5.7 ± 2.6	5.9 ± 3.0	0.213
E2 (pmol/L)	149.7 ± 81.1	157.2 ± 68.0	0.058
AFC	13.1 ± 5.0	14.4 ± 6.1	< 0.001
Endometrial thickness (mm)	10.0 ± 1.8	10.4 ± 1.8	< 0.001
FET protocol
Natural cycles (%)	54.5 (813/1492)	53.0 (318/600)	0.536
Programmed cycles (%)	45.5 (679/1492)	47.0 (282/600)
Sperm concentration (106/mL)	42.9 ± 46.0	24.8 ± 34.6	< 0.001
Spermatozoa motilities (%)	27.9 ± 19.4	17.8 ± 16.0	< 0.001
Sperm progressive motility (a + b, %)	24.3 ± 17.5	15.4 ± 14.5	< 0.001

BMI body mass index, FSH follicle stimulating hormone, E2 estradiol, AFC antral follicle count

Pregnancy outcomes of patients in two groups

The clinical pregnancy rate for the first transfer in the PGT group was 52.7%, which was significantly higher than that in the non-PGT group (45.2%, p = 0.002). Similarly, the live birth rate in the PGT group was 44.0%, which was also significantly higher than in the non-PGT group (32.8%, p < 0.001). The total miscarriage rate (13.6 vs. 22.5%, p < 0.001) and early miscarriage rate (12.5 vs. 19.6%, p = 0.005) were lower in the PGT group. The incidence of monozygotic twins was significantly lower in PGT group (0.9 vs. 3.3%, p = 0.011). No significant differences were observed between the two groups in terms of ectopic pregnancy rate, preterm delivery rate, or Caesarean section rate. When considering pregnancy complications, the incidences of HDCP and GDM were also similar between groups. Newborn weight and length were also similar between the two groups (Table 2).

Table 2.

Pregnancy outcomes in the two groups

	PGT group (n = 1492)	Non-PGT group (n = 600)	P-value
Live birth rate (%)	44.0 (656/1492)	32.8 (197/600)	< 0.001
Clinical pregnancy rate (%)	52.7 (787/1492)	45.2 (271/600)	0.002
Ectopic pregnancy rate (%)	0.6 (5/787)	1.1 (3/271)	0.714
Miscarriage rate (%)	13.6 (107/787)	22.5 (61/271)	< 0.001
Early miscarriage (%)	12.5 (98/787)	19.6 (53/271)	0.005
Monozygotic twins (%)	0.9 (7/787)	3.3 (9/271)	0.011
Preterm delivery (%)	6.9 (45/656)	7.1 (14/197)	1
Cesarean section rate (%)	65.5 (430/656)	72.6 (143/197)	0.079
Newborn weight (g)	3418.5 ± 524.1	3431.2 ± 641.0	0.793
Newborn length (cm)	50.2 ± 2.4	50.5 ± 3.9	0.331
HDCP (%)	4.6 (36/787)	3.0 (8/271)	0.328
GDM (%)	7.4 (58/787)	8.1 (22/271)	0.788

HDCP hypertensive disorder complicating pregnancy, GDM gestational diabetes mellitus

Early motor development of infants

Infants in the PGT group achieved milestones such as independent sitting, hand-knee crawling, and independent standing earlier than those in the non-PGT group (p < 0.001). However, there was no significant difference in the age at which solitary walking was achieved between the two groups (p = 0.937) (Table 3).

Table 3.

Analysis of developmental milestones in children

	Independent sitting (month)	Hand-knee crawling (month)	Independent standing (month)	Solitary walking (month)
PGT group (n = 588)	5.4 ± 0.9	7.2 ± 1.2	9.7 ± 1.7	12.5 ± 1.0
Non-PGT group (n = 170)	5.8 ± 0.8	7.7 ± 1.1	10.3 ± 1.5	12.5 ± 1.1
P-value	< 0.001	< 0.001	< 0.001	0.937

Correlation analysis of pregnancy outcomes in different groups

Multiple logistic regression analysis was performed to evaluate the association between biopsy and clinical pregnancy rate, miscarriage rate, live birth rate, ectopic pregnancy rate, monozygotic twin rate, preterm birth rate, and the incidence of HDCP and GDM (Table 4). After adjusting for factors of female age, male age, AFC, sperm concentration, sperm motility, sperm progressive motility (a + b), endometrial thickness, and type of infertility, the PGT group demonstrated a significantly higher live birth rate for the first transfer compared to the non-PGT group (OR 1.570, 95%CI 1.156–2.140). Additionally, the miscarriage rate was lower in the PGT group than in the non-PGT group (OR 0.533 95%CI 0.316–0.906). The incidence of monozygotic twins was significantly higher in the non-PGT group compared to the PGT group (OR 0.047, 95%CI 0.002–0.310). No significant differences were observed between groups for clinical pregnancy rate, preterm delivery rate, or the incidence of HDCP and GDM.

Table 4.

Associations between PGT and clinical outcomes: logistic regression analysis

	P-value	OR	95% CI
Clinical pregnancy rate	0.500	0.880	(0.606, 1.271)
Miscarriage	0.019	0.533	(0.316, 0.906)
Live birth rate	0.004	1.570	(1.156, 2.140)
Ectopic pregnancy	0.917	1.147	(0.107, 28.109)
Monozygotic twins	0.007	0.047	(0.002, 0.310)
Preterm delivery	0.440	1.516	(0.527, 4.361)
HDCP	0.298	2.308	(0.574, 15.632)
GDM	0.490	0.752	(0.341, 1.741)

Female age, male age, AFC, sperm concentration, sperm motility, sperm progressive motility (a+b), endometrial thickness, and infertility type were adjusted

Discussion

The present study examined maternal and infant outcomes in frozen single blastocyst transfer cycles, comparing cases with PGT involving trophectoderm biopsy to those with ICSI alone (without biopsy). The findings indicate that trophectoderm biopsy does not appear to increase the risk of pregnancy complications or negatively impact early motor development in infants.

The baseline characteristics of patients in the two groups showed differences in semen quality, likely due to the asymmetry in sample size and varying indications for IVF. To better evaluate the impact of biopsy on maternal and infant outcomes, the non-PGT group was limited to the first frozen blastocyst transfer cycle, resulting in a smaller sample size for the control group. Additionally, ICSI is typically used to address male factor infertility, including conditions such as low sperm count, poor motility, and abnormal morphology. This explains the lower semen quality observed in the non-PGT group [16].

In our study, the live birth and pregnancy rates for the first transfer were higher in the PGT group compared to the non-PGT group. However, after adjusting for confounding factors such as age, AFC, sperm quality, endometrial thickness, and infertility type, the pregnancy rates between the two groups showed no significant difference, while the live birth rate in the PGT group remained significantly higher. The role of PGT in improving live birth rates remains a controversial area. For example, it is acknowledged that chromosomal aneuploidy in embryos is a major cause of pregnancy loss and recurrent implantation failure [17, 18]. To address this challenge, preimplantation genetic testing for aneuploidies (PGT-A) has been proposed as a tool for selecting chromosomally normal embryos for transfer. Bhatt et al. demonstrated that PGT-A improves both clinical pregnancy and live birth rates in patients with recurrent pregnancy loss [19], while Vinals et al. reported that single blastocyst transfer following PGT-A significantly increased clinical pregnancy and live birth rates in older women, with a concomitant reduction in miscarriage rates [20]. However, other studies argue that PGT-A has not consistently been shown to improve live birth rates [21–24]. It suggests that the observed benefits may reflect patient selection and the use of elective single embryo transfer, rather than the genetic testing itself. This ongoing debate highlights the need for further investigation into the long-term impact of PGT-A on overall reproductive outcomes.

Pregnancy complications such as HDCP, GDM, preterm delivery, and monozygotic twins are noteworthy indicators to measure the safety of PGT, as they pose significant risks to both mother and fetus. Among these, HDCP has a profound impact on maternal and child health, with its progression often leading to vasospasms and target organ damage in the liver, kidney, brain, and blood system. Insufficient trophoblastic invasion and impaired vascularization, which hindered placenta development, are believed to be key factors in the pathogenesis of pregnancy-related hypertension resulting from poor placenta formation. As blastocyst biopsy involves the removal of cells destined to form the placenta, concerns arise regarding whether it affects proper placental function, causes placental disorder, or increases the risk of hypertensive disorders [25, 26]. Jing et al. compared the incidence of pregnancy-induced hypertension among patients undergoing blastocyst biopsy with frozen transfer, cleavage-stage biopsy with frozen transfer, and cleavage-stage biopsy with fresh transfer. They found that the incidence rates were 9.0%, 9.1%, and 2.3%, respectively, attributing the increased incidence in the blastocyst biopsy with frozen transfer group partly to vitrification technology [27]. However, a multicenter study using multivariate logistic and linear regression models and accounting for confounders showed that embryo biopsy did not increase adverse obstetric or neonatal outcomes, particularly preeclampsia [28]. Consistent with this, our data didn’t show biopsy increases the incidence of HDCP. In our analysis, the OR for HDCP was 2.308. Although this result was not statistically significant, the wide confidence interval may be explained by the small number of HDCP cases in the non-PGT group (only 8 cases), which may have led to instability in the regression estimate. While the elevated OR might suggest a possible tendency toward increased HDCP risk in the PGT group, this interpretation should be approached with caution given the limited sample size and lack of statistical significance. Future research with larger, multicenter cohorts and more comprehensive control of confounding variables is warranted to validate this potential association.

Cozzolino’s study reported no significant difference in the incidence of GDM between the PGT and non-PGT group, a finding consistent with ours [28]. Preterm delivery, on the other hand, is more dependent on obstetric factors during pregnancy, such as intrauterine or systemic infections, uterine tension, and cervical conditions. Similarly, our study found no significant difference in the incidence of preterm delivery between the PGT and non-PGT groups (6.9 vs. 7.1%, p = 1), which aligns with the conclusions of most current research [29, 30].

Twin pregnancies significantly increase maternal blood volume and cardiac workload, thereby increasing the risk of complications such as hypertension, preeclampsia, and other cardiovascular issues. They also pose significant fetal risks, including growth restriction, abnormal blood flow, miscarriage, and preterm birth. In our study, the incidence of monozygotic twin pregnancies was 3.3% in the non-PGT group, significantly higher than the 0.7% observed in the PGT group. These findings indicate that blastocyst trophoblast biopsy does not appear to increase the risk of monozygotic twin pregnancies. One potential confounding factor that could influence the observed differences between the two groups is maternal age. In our study, the maternal age in the PGT group was significantly lower than in the non-PGT group, and advanced maternal age may be associated with an increased risk of monozygotic twin pregnancies [31]. Although we adjusted for age in the statistical analysis, caution is still needed when interpreting these results. In addition, factors such as blastocyst grading, extended embryo culture, and transfer procedures may affect the research results [32–34]. Given the small sample size (only 16 cases of monozygotic twins in the two groups), further large-scale prospective studies on a more diverse patient population are needed to more accurately assess the impact of these variables on the incidence of monozygotic twins. Studies have shown that biopsy does not negatively affect fetal development during pregnancy [35, 36]. In our study, there were no significant differences between the groups in terms of newborn weights (3418.5 ± 524.1 vs. 3431.2 ± 641.0 g, p = 0.793) or length (50.2 ± 2.4 vs. 50.5 ± 3.9 cm, p = 0.331). Additionally, the cesarean section rates were comparable between the two groups.

Some researchers have conducted long-term follow-up studies on offspring from cleavage-stage biopsy, using tools such as the Wechsler Preschool and Primary Scale of Intelligence and other professional assessments in areas like psychology, motor skills, and overall development. These studies found that offspring from cleavage-stage biopsy generally met standard developmental indices, although they exhibited slightly poorer performance in certain specific tasks [37–39]. However, because blastocyst biopsy is a relatively recent development, there are still limited studies on the long-term outcomes of offspring resulting from this procedure. We referred to the motor development milestones published by WHO, including the age (in months) at which children achieve independent siting (3.8–9.2), crawling on hands and knees (5.2–13.5), standing independently (6.9–16.9), and walking independently (8.2–17.6). All motor developmental milestones in both groups were achieved within the normative time frames defined by the WHO. Although our findings showed that children in the PGT group achieved certain motor milestones—such as independent sitting, crawling, and standing—slightly earlier than those in the non-PGT group, the clinical significance of these differences remains uncertain. One possible explanation is differences in patient populations, as parents in the PGT group may have higher socioeconomic status and place more emphasis on early developmental milestones, potentially influencing motor outcomes. Nevertheless, due to the reliance on parental recall via telephone follow-up, the possibility of recall bias cannot be excluded. Future studies employing standardized developmental assessments and long-term follow-up are warranted to further clarify these findings.

Currently, there is limited literature comparing the impact of different types of PGT on clinical outcomes. In our study, we analyzed all patients according to the indications for PGT and compared live birth rates, clinical pregnancy rates, miscarriage rates, and preterm birth rates. The results showed no significant differences in reproductive outcomes among the three groups (Supplemental Table 1). Some studies suggest that preimplantation genetic testing for monogenic (PGT-M) or preimplantation genetic testing for chromosomal structural rearrangements (PGT-SR) typically do not have a background of infertility and may lack the intrinsic factors that contribute to infertility. Therefore, the outcomes in these patients may require further research to better assess the effects of embryo manipulation, potentially avoiding major confounding factors associated with infertility [40].

Our study has several notable strengths. First, we analyzed a comprehensive cohort of 2092 FET cycles, providing a robust sample size that enhances the reliability of our findings. Second, we conducted relatively long-term follow-up on early motor development in infants through 2021, offering well-substantiated data on early growth and developmental outcomes. These strengths further support the potential practical application of PGT.

However, there are several limitations of our study that should be acknowledged. First, our analysis focused solely on the outcomes of patients’ first frozen-thawed blastocyst transfer, without evaluating the cumulative live birth rate for the entire oocyte retrieval cycle. Second, the long-term effects of blastocyst biopsy on health and development—particularly in accurately assessing cognitive, psychological, or behavioral outcomes from infancy through early childhood—were not evaluated. Further follow-up and long-term studies are needed to fully evaluate the safety and prognosis of trophoblast biopsy in PGT.

Conclusion

TE biopsy in PGT was not associated with a significant increase in adverse obstetrical or infant outcomes, including HDCP, GDM, and preterm delivery, when compared with ICSI alone. PGT was associated with higher live birth rate and lower miscarriage rate in first single blastocyst transfer cycles. Although these findings suggest that trophectoderm biopsy is unlikely to have a negative impact, the potential risks associated with the procedure cannot be entirely ruled out. Therefore, it should be applied with caution and only when there are clear clinical indications.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

RY, JX.W, and YW conceived and designed the study; RY, JX.W, TT, and LX.C conducted the analysis and designed the figures and tables; YW, CG, JL, RL, PL, and JQ reviewed the literature; LX.C helped to collect the data; RY, JX.W, and YW wrote the first draft and revised the manuscript. All authors participated in the discussion of the analysis and interpretation of data in this article.

Funding

National Key Research and Development Program (2022YFC2702905, 2024YFC2706702, and 2021YFC2700605); National Natural Science Foundation of China (82171632); Capital’s Funds for Health Improvement and Research of Beijing (2024–2-40911).

Data availability

The corresponding authors can be contacted on reasonable data request.

Declarations

Ethics approval

All methods in this study were carried out in accordance with the guidelines and regulations in the Declaration of Helsinki. The Ethics Committee of Peking University Third Hospital approved the study (No. IRB00006761-M2021612).

Consent to participate

Written informed consent was obtained from all patients before treatment, and the patients consented to the use of their retrospective data in scientific publications.

Consent for publication

Not applicable.

Conflict of interest

The authors declare no competing interests.