Endometrial scratching in frozen embryo transfer: A stratified cohort study

Abstract

Background:

Endometrial scratching (ES) is widely used in assisted reproduction. However, randomized trials have shown no benefits in unselected populations, and evidence specific to frozen embryo transfer (FET) is limited. Therefore, we estimated the association between ES and the clinical pregnancy rate (CPR) separately within strata defined by the embryo transfer (ET) history.

Methods:

In this retrospective cohort study at a tertiary center in Taiwan (January 2017-January 2021), nulliparous women aged <40 years who underwent FET were eligible. ES was performed with a Novak curette in the cycle preceding FET, and analyses were stratified by the ET history (none, 1-2, or ≥3). The primary outcome was the CPR at 6 to 8 weeks of gestation. Multivariate logistic regression within each stratum was adjusted for prespecified covariates. Differences between strata were not formally tested.

Results:

Of the 822 women who were enrolled, 176 (21.4%) underwent ES. The overall CPR did not differ between the groups (46.6% vs 53.6%; p = 0.100). In women undergoing first FETs (n = 457), ES was associated with a lower crude CPR (39.4% vs 58.1%; odds ratio: 0.47, 95% CI, 0.27-0.80; p = 0.005) but not after adjustment (adjusted odds ratio [aOR]: 0.59, 95% CI, 0.31-1.15; p = 0.121). The CPR was numerically higher with ES in women with 1 to 2 previous ETs (n = 295; 50.0% vs 46.9%; aOR: 1.29, 95% CI, 0.62-2.66; p = 0.499) and in those with ≥3 previous ETs (n = 70; 54.2% vs 45.7%; aOR: 1.54, 95% CI, 0.19-12.40; p = 0.687); all adjusted associations were nonsignificant. Early miscarriage and ectopic pregnancy rates did not differ across the strata.

Conclusion:

ES was not associated with improved clinical pregnancy in this FET cohort, and its routine use in treatment-naïve patients with good prognosis was not supported. In women with previous ET failures, the estimates favored ES but were imprecise, nonsignificant, and exploratory. Therefore, adequately powered randomized controlled trials are required.

Lay Summary: Endometrial scratching is a minor procedure deliberately disturbing the lining of the uterus before an embryo is transferred, to promote embryo implantation. Whether this works is unclear, and little is known about its use in frozen embryo transfer. We reviewed the records of 822 women aged <40 years who had not previously given birth. Because the procedure can be offered both before the first transfer and after failed transfers, we examined three groups separately: women approaching their first transfer, those with one or two previous failures, and those with three or more previous failures. Pregnancy rates did not improve with scratching in any group, and the procedure did not increase the risk of miscarriage or ectopic pregnancy. Therefore, routine scratching is not recommended for women approaching their first transfer with a good prognosis. Whether this helps women with previously failed transfers remains unknown and requires large, randomized trials.

1. INTRODUCTION

Endometrial scratching (ES), defined as the deliberate mechanical disruption of the endometrial lining using a Pipelle catheter or curette, has been proposed as a simple and cost-effective adjunct to assisted reproductive technology (ART). Early observational data have suggested improvements in implantation rates,^1–6 leading to rapid and widespread clinical uptake.^7,8 The proposed mechanisms include the modulation of endometrial receptivity-related gene expression, activation of localized inflammatory responses conducive to implantation, and decidual transformation after tissue injury.^5,9–11

However, subsequent high-quality randomized controlled trials (RCTs) have not supported these initial observations. The Pipelle for Pregnancy (PIP) trial (n = 1364), the largest such RCT to date, revealed no difference in live birth rates (LBR) between the ES and control groups (adjusted odds ratio [aOR]: 1.00; 95% CI, 0.78-1.27).¹² Similarly, in a large RCT involving women undergoing their first in vitro fertilization (IVF) cycle (n = 1048), Metwally et al¹³ found no improvement in the LBR after mid-luteal ES (38.6% vs 37.1%; p = 0.621). This finding was corroborated by a subsequent systematic review and meta-analysis by the same authors.¹⁴

Despite these negative results, uncertainties remain. The PIP trial has been criticized for its methodological heterogeneity, including the use of both fresh and frozen cycles (26% frozen embryo transfer [FET]), patients with 0 to 2 previous failed cycles, and a broad ES window, which may have diluted subgroup-specific effects. Most RCTs have focused on fresh embryo transfer cycles, whereas FET, which eliminates the influence of ovarian hyperstimulation, remains underexplored. Emerging evidence from meta-analyses suggests that the effects of ES may not be uniform, with potential benefits confined to women experiencing recurrent implantation failure (RIF) rather than to those undergoing their first treatment cycle.^6,15,16

As such, this retrospective cohort study focused on nulliparous women aged <40 years who underwent FET. Treatment-naïve patients and patients with repeated unsuccessful transfers constitute distinct clinical decision-making contexts in which the indication for ES is evaluated independently; therefore, separate effect estimates are required. Patients were stratified according to the number of previous embryo transfer (ET) attempts, and the association between ES and the clinical pregnancy rate (CPR) was estimated within each stratum. We hypothesized that ES would not improve CPR in women undergoing their first FET but might be associated with a higher CPR in women who meet our operational definition of RIF (ie, three or more previous failed ET cycles).

2. METHODS

2.1. Study design and setting

This retrospective cohort study was conducted at Taipei Veterans General Hospital, a tertiary referral center in Taiwan. Medical records from January 2017 to January 2021 were reviewed. The study protocol was approved by the Institutional Review Board (IRB) of Taipei Veterans General Hospital (IRB no. 2026-03-005AC).

2.2. Participants

Eligible participants were nulliparous women aged <40 years who underwent FET. Women with documented intrauterine abnormalities, including endometrial polyps, submucosal fibroids, endometrial hyperplasia, or endometrial malignancy identified on baseline hysteroscopy or transvaginal ultrasound within three months before embryo implantation were excluded, leaving 822 patients for analysis (Fig. 1).

Fig. 1.

Flow diagram of the study design and process. ET = embryo transfer; FET = frozen embryo transfer.

Participants were stratified into three groups based on their cumulative ET history: no previous ET (n = 457); 1 to 2 previous ETs (n = 295); and ≥3 previous ETs (n = 70). We used three or more previously failed ET cycles as a working clinical definition for RIF, in accordance with Polanski et al.¹⁷ This operational definition does not meet the current diagnostic criteria¹⁸ but reflects the clinical norms of the study period. ES was not assigned randomly and was more frequently applied to patients with prior implantation failure, at the physician’s discretion.

2.3. ES procedure

To minimize procedural heterogeneity and obtain a consistent local injury, the ES procedure was protocolized using a conventional Novak curette, adapted from a previously described method.¹⁹ Following speculum placement and cervical antisepsis, the curette was advanced into the uterine cavity under real-time transabdominal ultrasound guidance to confirm the instrument position and uterine axis. Systematic scraping was performed from the fundus to the internal os along the anterior and posterior uterine walls, followed immediately by diagnostic hysteroscopy for direct visual inspection of the treated cavity. This approach was intended to reduce the variability inherent in blind, limited endometrial biopsies. All procedures were performed by trained reproductive endocrinologists in the cycle preceding the planned FET (within 90 days), and the timing of ES was verified by histological dating.

2.4. Endometrial preparation and ET protocol

The endometrium was prepared using either a hormone replacement therapy (HRT) protocol or the natural cycle. Most patients received HRT (91.4% [751/822]), consisting of oral estradiol valerate (2-6 mg daily), initiated on cycle days 2 to 3. Progesterone supplementation (vaginal progesterone gel [Crinone; Merck KGaA, Darmstadt, Germany] three times daily plus micronized progesterone 200 mg once every night) was started once the endometrial thickness reached ≥7 mm. The remainder (8.6% [71/822]) underwent natural-cycle FET, and ovulation was confirmed using serial ultrasonography and urinary luteinizing hormone testing.

Only fair- to good-quality embryos—defined as G1-G2 for cleavage-stage embryos and ≥BB for blastocysts, according to the 2011 Istanbul consensus on embryo assessment²⁰—were transferred, so that the baseline embryo quality was clinically controlled. The ET was performed under transabdominal ultrasound guidance using a soft catheter.

2.5. Outcome measures

The primary outcome was the CPR, defined as the presence of ≥1 intrauterine gestational sac or a fetal heartbeat noted on transvaginal ultrasound at 6 to 8 weeks of gestation. Secondary outcomes included live birth (viable birth after 24 weeks of gestation), biochemical pregnancy loss (positive serum β-human chorionic gonadotropin ≥10 IU/L at 10 days post-ET without subsequent ultrasound confirmation of an intrauterine gestational sac), early miscarriage (spontaneous miscarriage occurring before 12 weeks of gestation), and ectopic pregnancy. Multiple pregnancies were defined as the presence of more than one intrauterine gestational sac on transvaginal ultrasound.

2.6. Statistical analysis

Continuous variables are expressed as mean ± SD and were compared using the t-test, and categorical variables are expressed as counts with percentages and were compared using the chi-squared or Fisher’s exact test. Analyses were stratified by ET history. ES is more often applied to patients with prior treatment failure; therefore, a history of ETs is both the clinical context in which the decision arises and the principal source of the measured imbalance between the exposure groups. Stratifying addressed this imbalance through design.

Multivariate logistic regression was performed within each of the previous ET strata, and the aORs with 95% CIs were calculated. Covariates were selected a priori based on clinical relevance and prior literature, including age, anti-Müllerian hormone (AMH) level, body mass index (BMI), blastocyst transfer (yes vs no), infertility type (primary vs secondary), and number of embryos transferred.^13,21–23 Propensity score matching was not performed because the baseline characteristics were generally comparable between the two groups (Table 1), and the primary potential confounder was addressed through stratification. Matching would have reduced the sample size without addressing further confounding.²⁴ No substantial collinearity was identified; all variance inflation factors were <10. The events-per-variable (EPV) ratio met the conventional requirement of at least 10 in the two larger strata but not in the ≥3 ET stratum (n = 70), indicating a risk of overfitting. Therefore, estimates from this stratum were exploratory. Because the objective was to estimate the association within each clinical context rather than to compare contexts, interaction models were not fitted, and no between-stratum comparisons were performed. Exploratory secondary analyses examined ES cycles by ES phase and ES–FET interval (Supplementary Tables S1 and S2, https://links.lww.com/JCMA/A434). They compared the primary models with models omitting the number of embryos transferred (Supplementary Table S3, https://links.lww.com/JCMA/A434). All tests were two-sided, and p values <0.05 indicated statistical significance. All statistical analyses were performed using SPSS (version 26.0; IBM Corp., Armonk, NY).

Table 1.

Baseline characteristics of the study cohort

Characteristic	Scratch	Non-scratch	p
	n = 176	n = 646
Age, y	34.91 ± 3.13	34.63 ± 3.26	0.308
BMI, kg/m2	21.86 ± 3.27	22.32 ± 3.61	0.131
AMH, ng/mL	4.32 ± 4.62	4.32 ± 3.75	0.998
Cause of subfertility			0.296
Ovarian	35 (19.9%)	134 (20.7%)
Male	61 (34.7%)	235 (36.4%)
Tubal	29 (16.5%)	72 (11.1%)
Other	51 (29.0%)	205 (31.7%)
Type of subfertility (n, %)			0.56
Primary infertility	85 (48.3%)	328 (50.8%)
Secondary infertility	91 (51.7%)	318 (49.2%)
Previous ET attempts			<0.001*
0	66 (37.5%)	391 (60.5%)
1-2	86 (48.9%)	209 (32.4%)
≥3	24 (13.6%)	46 (7.1%)
FET protocol			0.250
Artificial cycle	157 (89.2%)	594 (92.0%)
Ovulatory cycle	19 (10.8%)	52 (8.0%)
Embryo transfer day
EM(ET) thickness	10.37 ± 2.55	10.54 ± 4.26	0.607
PGTR(ET) level	27.95 ± 18.68	26.62 ± 17.12	0.369
ET embryo number			0.012*
≤2	41 (23.3%)	221 (34.2%)
3	45 (25.6%)	164 (25.4%)
4	90 (51.1%)	261 (40.4%)
Days of ET embryo
Blastocyst transfer	22 (12.5%)	113 (17.5%)	0.113

Data are presented as number (%) or mean ± SD.

EM_(ET) thickness = endometrial thickness on ET day; PGTR_(ET) = progesterone level on ET day.

^*p < 0.05; bold values indicate statistical significance.

3. RESULTS

3.1. Study population and baseline characteristics

Of the 822 women, 176 (21.4%) underwent ES in the preceding cycle, and 646 (78.6%) served as controls. The baseline characteristics of the cohort are summarized in Table 1. Approximately 60% of the participants were aged ≥35 years. The two groups were comparable in age, BMI, AMH level, cause and type of infertility, FET protocol, endometrial thickness, and progesterone level on the day of ET but differed in ET history and the number of embryos transferred. The ES group had significantly more previous ET attempts compared with controls (p < 0.001): 37.5% vs 60.5% vs 48.9% had no previous ET; 32.4% had 1 to 2 previous ETs; and 13.6% vs 7.1% had ≥3 previous ETs.

Additionally, a higher proportion of women in the ES group received 4 embryos per transfer (51.1% vs 40.4%; p = 0.012). However, within individual previous ET strata, these distributions did not differ significantly between the groups (p = 0.200, 0.112, and 0.644, respectively; Supplementary Table S4, https://links.lww.com/JCMA/A434), indicating that the overall difference reflected the composition of the strata rather than divergent transfer practices within them. Single-embryo transfer (SET) was uncommon (15/822, 1.8%) and occurred exclusively in the non-ES group (12 cleavage-stage and 3 blastocyst transfers). Blastocyst transfers were less frequent in the ES group (12.5% vs 17.5%; p = 0.113).

3.2. Overall pregnancy outcomes

Pregnancy outcomes in the overall cohort are shown in Table 2. No significant differences were observed between the ES and control groups in biochemical pregnancy rate (57.4% vs 61.1%; p = 0.374), CPR (46.6% vs 53.6%; p = 0.100), ongoing pregnancy rate (36.9% vs 42.3%; p = 0.200), or LBR (35.8% vs 41.6%; p = 0.165).

Table 2.

Pregnancy outcomes in nulliparous women who underwent FET

Outcome	Scratch	Non-scratch	p
	(n = 176)	(n = 646)
Biochemical pregnancy rate, %	57.4%	61.1%	0.374
Clinical pregnancy rate, %	46.6%	53.6%	0.100
Ongoing pregnancy rate, %	36.9%	42.3%	0.200
Live birth rate, %	35.8%	41.6%	0.165

FET = frozen embryo transfer.

The reproductive outcomes, according to the number of embryos transferred, are presented in Supplementary Table S5, https://links.lww.com/JCMA/A434. The transfer of four embryos was not associated with a higher CPR or LBR, whereas multiple pregnancy was numerically but not significantly more frequent (22.2% of clinical pregnancies vs 19.0% and 16.5% after three and ≤2 embryos, respectively; p = 0.434).

3.3. Stratified analysis according to previous ET attempts

As previous ET attempts were significantly associated with the ES assignment (p < 0.001), stratification was used to reduce confounding. The subgroup analyses are presented in Fig. 2 and Table 3. Among women undergoing their first FET cycle (n = 457), ES was associated with a significantly lower CPR (39.4% vs 58.1%; crude OR: 0.47, 95% CI, 0.27-0.80; p = 0.005). However, this association was no longer statistically significant after adjustments (aOR: 0.59, 95% CI, 0.31-1.15; p = 0.121).

Fig. 2.

Clinical pregnancy rate according to endometrial scratching, stratified by previous ET attempts. Bars show unadjusted clinical pregnancy rates with the number of clinical pregnancies over the number of cycles. *p = 0.005 for the unadjusted comparison with the ET = 0 stratum (Pearson’s chi-squared test); this association was no longer significant after multivariate adjustments (adjusted odds ratio: 0.59, 95% CI, 0.31-1.15; p = 0.121). Adjusted analyses were nonsignificant across all three strata (Table 3), and differences between strata were not formally tested. ES, endometrial scratching; ET, embryo transfer (ET = 0: no previous ET attempt; ET = 1-2: 1 or 2 previous ET attempts; ET≥3: 3 or more previous ET attempts).

Table 3.

Association between ES and CPR, stratified by previous ET attempts

ET attempt	Group	CPR (%)	OR (95% CI)	p	aOR# (95% CI)	p
0	ES	39.4%	0.47* (0.27, 0.80)	0.005 *	0.59 (0.31, 1.15)	0.121
	Non-ES	58.1%	Reference		Reference
1-2	ES	50.0%	1.13 (0.69, 1.87)	0.627	1.29 (0.62, 2.66)	0.499
	Non-ES	46.9%	Reference		Reference
≥3	ES	54.2%	1.41 (0.52, 3.79)	0.499	1.54 (0.19, 12.40)	0.687
	Non-ES	45.7%	Reference		Reference

#: Adjusted for age, AMH level, BMI, blastocyst transfer, infertility type, and number of embryos transferred (≤2, 3, or 4).ES, endometrial scratching; CPR, clinical pregnancy rate; ET, embryo transfer; OR, crude odds ratio; aOR, adjusted odds ratio; AMH, anti-Müllerian hormone; BMI, body mass index; CI, confidence interval.

^*p < 0.05; bold values indicates statistical significance.

Among women with 1 to 2 previous ETs (n = 295), the CPR was higher in the ES group (50.0% vs 46.9%) but was not statistically significant (aOR: 1.29, 95% CI, 0.62-2.66; p = 0.499). Among patients with ≥3 previous ETs (n = 70), the CPR was 54.2% in the ES group and 45.7% in the control group. The point estimate favored ES but was not significant and carried a very wide CI (aOR: 1.54, 95% CI, 0.19-12.40; p = 0.687). As between-stratum comparisons were not performed, these estimates were not compared. In the sensitivity analysis that omitted the number of embryos transferred, the estimates were essentially unchanged within each stratum (Supplementary Table S3, https://links.lww.com/JCMA/A434).

3.4. Exploratory subgroup analyses by ES timing and interval

Most procedures were performed during the luteal phase (77.3% [136/176]); the remainder were performed during the follicular phase (22.7% [40/176]). Exploratory subgroup analyses by ES timing revealed no significant differences in CPR (45.0% vs 47.1%, p = 0.819), LBR (35.0% vs 36.0%, p = 0.905), or early miscarriage rate (10.0% vs 11.0%, p = 0.853). Similarly, no consistent pattern was observed across the ES–FET intervals (<30, 31-60, and 61-90 days) for either the CPR or LBR. The exploratory results are presented in Supplementary Tables S1 and S2, https://links.lww.com/JCMA/A434.

3.5. Pregnancy risk and safety concerns

The rates of early miscarriage and ectopic pregnancy did not differ between the subgroups (Table 4).

Table 4.

Pregnancy risk-subgroup analysis according to previous ET failure

	Group	Biochemical loss		Early miscarriage		Ectopic pregnancy
		%	p	%	p	%	p
ET = 0 (n = 457)	ES	12.1	0.117	7.6	0.193	0	>0.99
	Non-ES	6.6		13.3		0.8
ET = 1-2 (n = 295)	ES	10.5	0.714	14.0	0.333	0	>0.99
	Non-ES	9.1		10.0		0.5
ET ≥3 (n = 70)	ES	8.3	>0.99	8.3	>0.99	0	>0.99
	Non-ES	8.7		8.7		0

Early miscarriage: pregnancy loss before 12 wks.

4. DISCUSSION

In this retrospective cohort, the association between ES and the CPR in FET cycles was estimated separately within the strata defined by a history of ET. ES conferred no significant benefits to the overall cohort. Among women undergoing their first FET, a lower crude CPR did not remain significant after adjustment. In women with prior ET failures, the point estimates favored ES but were imprecise and not significant. The safety outcomes were comparable across all strata.

In the unselected population, our findings were consistent with randomized evidence. The PIP trial found no difference in the LBR,¹² and Eskew et al.²⁵ reported a numerically lower CPR with ES (47.2% vs 59.6%, OR: 0.61; p = 0.22) in a trial that was halted early for predefined futility. In treatment-naïve patients, the RCT by Metwally et al.¹³ (n = 1048) demonstrated comparable LBRs between the ES and non-ES groups in first-cycle fresh IVF cycles (38.6% vs 37.1%; p = 0.621), and their subsequent meta-analysis identified no significant effect on either live births (OR: 1.17, 95% CI, 0.76-1.79) or the CPR (OR: 1.18, 95% CI, 0.82-1.72).²⁶ This study extends these results to FET. Because the crude reduction in the CPR did not persist after adjustments, we did not consider it as evidence of harm. These data do not support routine ES in patients undergoing first-cycle FET with a favorable prognosis.

Among women with 1 to 2 previous failures, a modest absolute gain of 3.1% in the CPR was recorded in the ES group, consistent with the results of the SCRaTCH trial, which reported a numerically higher LBR with ES in women with ≥1 previous failed ETs (23.7% vs 19.1%; RR: 1.24, 95% CI, 0.96-1.59).¹⁶ Among women with ≥3 previous ETs, the point estimate likewise favored ES, as did the corresponding subgroup of the meta-analysis by Iakovidou et al.²⁷ in which trials enrolling patients with at least three previous failed cycles reported a higher LBR with ES (RR: 1.70, 95% CI, 1.14-2.54; 3 trials, 547 participants). However, these comparisons were made across studies rather than across the strata within our study. Although the subgroup-specific point estimates differed numerically, all adjusted analyses remained nonsignificant with wide CIs, and between-stratum comparisons were not performed. These findings should be interpreted as exploratory and should warrant prospective evaluations in adequately powered trials.

One explanation for this difference was the variation in the underlying endometrial state rather than a true treatment effect. Young nulliparous women without previous ETs likely possess well-preserved endometrial receptivity under standardized hormonal preparations, in which mechanical disruption may transiently destabilize endometrial integrity or perturb the local immunological milieu. In contrast, women with RIF may have subclinical endometrial dysfunction that is undetectable on conventional assessments. In such cases, ES may induce a regenerative response, alter cytokine signaling, or promote angiogenesis and tissue remodeling.^10,28–30 The present data could not test these hypotheses, and further mechanistic studies are needed to confirm them.

This study had three strengths. First, restricting the analysis to nulliparous women aged <40 years reduced heterogeneity related to reproductive aging and parity-associated uterine changes, thereby improving internal validity. Second, stratification yielded an estimate for each clinically defined group rather than a single pooled figure, which would obscure the differences among them. Finally, ES followed a uniform protocol under real-time ultrasound guidance with immediate hysteroscopy. This limited the spatial variability inherent in blind endometrial biopsy. Furthermore, ultrasound guidance during intrauterine instrumentation is an established technique reported to reduce procedural complications.^31,32

However, this study had several limitations. First, ES was not randomly assigned but was more commonly offered to patients with previous ET failure. Although stratification and within-stratum adjustments addressed the measured imbalance, the influence of unmeasured factors could not be excluded.

Second, the transfer practices reflected the standards of the study period rather than current recommendations. Four embryos were transferred in 40.4% to 51.1% of the cycles, predominantly in older women (58.9% of those aged ≥35 years vs 19.5% of younger women; p < 0.001). During the study period, PGT-A was not routinely available, and higher-order transfers of untested cleavage-stage embryos remained an accepted practice for patients with unfavorable prognoses, according to the ASRM committee opinion at that time.³³ The transfer of four embryos did not yield a higher CPR or LBR, whereas multiple pregnancy was numerically more frequent, but the difference was not statistically significant. However, these comparisons were confounded by indication, as women receiving four embryos were systematically older. Our center now prioritizes elective SET.

Third, embryo-level morphological and euploidy data were unavailable for adjustments because PGT-A was not routinely performed during the study period. Adjustments for cycle-level variables and restrictions to fair- to good-quality embryos could not replace detailed embryo-level adjustments. Therefore, the observed associations should be regarded as hypothesis-generating rather than causal.

Fourth, reviews contemporaneous with the study period noted the absence of a universally accepted definition of RIF.²¹ We adopted three or more previous failed ET cycles as a working definition.¹⁷ However, morphological and euploidy data for these historical transfers were unavailable, and our study population with RIF may not align with contemporary RIF populations defined by the failure of high-quality or euploid embryos.¹⁸ Therefore, our RIF subgroup results should be interpreted with caution.

Fifth, the subgroup with ≥3 previous ET failures was small (n = 70), limiting statistical power and resulting in wide CIs. The post-hoc power was not calculated for this nonsignificant result, as it provided no additional information beyond the effect estimate and CI. A future prospective study targeting this subgroup would require approximately 542 to 680 participants per arm (two-sided α = 0.05, 80% power) to detect an effect with the magnitude observed here.

Finally, the external validity was limited. The cohort came from a single center and was confined to nulliparous women aged <40 years, most of whom underwent HRT-FET with cleavage-stage transfer. These findings may not extend to older or parous women, natural-cycle FET, or programs that routinely perform elective single-blastocyst transfer or PGT-A.

In conclusion, the association between ES and clinical pregnancy in FET cycles was estimated separately, stratified by ET history. The routine use of ES in treatment-naïve FET patients with a good prognosis was not supported by our findings and should not be recommended as standard practice. In individuals with previous ET failures, the present estimates were imprecise, nonsignificant, and exploratory. Adequately powered RCTs are required before any clinically meaningful benefits can be established for this population.

ACKNOWLEDGMENTS

This study was supported by grants from Taipei Veterans General Hospital (grant nos. V114EA-017 and V115A-006).

We thank Ms. Hsin-Yi Huang from the Department of Information Management and Big Data Center, Taipei Veterans General Hospital, and Professor Mei-Hsuan Lee at the Institute of Clinical Medicine, National Yang Ming Chiao Tung University, for their statistical advice and review. We thank Taipei Veterans General Hospital for their help with sample preparation.

APPENDIX A. SUPPLEMENTARY DATA

Supplementary data related to this article can be found at https://links.lww.com/JCMA/A434.

Supplementary Material

ca9-89-750-s001.pdf ^{(651.4KB, pdf)}