Ectopic pregnancy is associated with increased risk of displaced implantation window: a retrospective study.

A total of 934 patients were included in the study. The patients underwent an average of 4.04 ART cycles and 2.86 ET cycles. Among the 934 patients, ERT showed that 562 patients were in the receptive phase (60.17%), 368 patients were in the pre-receptive phase (39.40%), and four patients were in the post-receptive phase (0.43%) (Fig.  1 ). The overall WOI displacement rate is 39.83%. Most WOI displacements are pre-receptive (98.9%) compared to post-receptive (1.1%). Table  1 shows the characteristics of the WOI and displaced WOI groups. Age, infertile type, BMI, ectopic pregnancy history, and number of ART cycles were not statistically different between the two groups. The age in the displaced WOI group is slightly older than the WOI group ( p -value = 0.06). Though age, infertile type, BMI, and ectopic pregnancy history were not significantly different between the two groups, multivariate regression analyses were performed to analyze the relationship between these factors and WOI displacement. On the one hand, some of the variables are correlated with each other. For example, ectopic pregnancy is correlated with the type of infertility. Patients with ectopic pregnancy are diagnosed as secondary infertility but not primary infertility. On the other hand, it is unknown whether there is a threshold effect for the continuous variable such as age or BMI on WOI displacement. Fig. 1 Pie plot showing the composition ratio of ERT results Pie plot showing the composition ratio of ERT results Table 1 Summary descriptives table by groups of endometrial receptivity test (ERT) result WOI Displaced WOI P -value N  = 562 N  = 372 Female age (year) 33.1(± 4.40) 33.7(± 4.93) 0.060 Infertile type : 0.793  Primary infertility 299(53.2%) 202(54.3%)  Secondary infertility 263(46.8%) 170(45.7%) BMI (kg/m 2 ) 21.8(± 2.77) 22.0(± 2.75) 0.218 Ectopic pregnancy history : 0.254  No 496(88.3%) 318(85.5%)  Yes 66(11.7%) 54(14.5%) No of ART cycle 3.90(± 1.86) 4.08(± 1.84) 0.166 Note: ERT: endometrial receptivity testing; WOI: window of implantation; BMI: body mass index; ART: assisted reproductive technology Summary descriptives table by groups of endometrial receptivity test (ERT) result Note: ERT: endometrial receptivity testing; WOI: window of implantation; BMI: body mass index; ART: assisted reproductive technology The generalized estimation models were used to test the factors associated with WOI displacement. Variables included in the model included ectopic pregnancy (EP), female age, BMI, and infertile type. Model 1 shows that patients with EP history had a 58% higher risk of WOI displacement than those without EP (aOR 1.58, 95% CI 1.01–2.46, p -value 0.046); The risk of WOI displacement increased by 3% within every 1 year increasing of age (aOR 1.03, 95%CI 1.00-1.06, p -value 0.068) without statistical significance; Secondary infertility showed a 25% lower risk of WOI displacement compared to primary infertility (aOR 0.75, 95% CI 0.54–1.03, p -value 0.078) without statistical significance; The risk of WOI displacement increased 3% within every 1 kg/m 2 increasing of BMI (aOR 1.03, 95%CI 0.98–1.08, p -value 0.3) without statistical significance. The results of the GEE model1 are shown in Table  2 . Table 2 Multivariate regression showing the associations between the window of implantation (WOI) displacement with ectopic pregnancy (EP), female age, infertile type, and body mass index (BMI) using generalized estimation equation (GEE) models GEE model 1 § GEE model 2 † Variables aOR 95%CI P -value aOR 95%CI P -value Ectopic pregnancy  No ref ref  Yes 1.58 1.01–2.46 0.046 1.62 1.03–2.53 0.035 Female age (Year) 1.03 1.00-1.06 0.068  < 35 ref  ≥ 35 1.50 1.12-2.00 0.007 Infertile type  Primary infertility ref ref  Secondary infertility 0.75 0.54–1.03 0.078 0.74 0.54–1.02 0.062 BMI (Kg/M 2 ) 1.03 0.98–1.08 0.3  < 22 ref  ≥ 22 1.25 0.94–1.67 0.12 Note: Variables included in the models were ectopic pregnancy, female age, infertile type, and BMI. §: Female age and BMI were treated as continuous variables in GEE model 1; †: Female age and BMI were transformed into categorical variables in GEE model 2. WOI: window of implantation; GEE: generalized estimation equation; aOR: adjusted odds ratio; CI: confidence interval; ref: reference; BMI: body mass index Multivariate regression showing the associations between the window of implantation (WOI) displacement with ectopic pregnancy (EP), female age, infertile type, and body mass index (BMI) using generalized estimation equation (GEE) models Note: Variables included in the models were ectopic pregnancy, female age, infertile type, and BMI. §: Female age and BMI were treated as continuous variables in GEE model 1; †: Female age and BMI were transformed into categorical variables in GEE model 2. WOI: window of implantation; GEE: generalized estimation equation; aOR: adjusted odds ratio; CI: confidence interval; ref: reference; BMI: body mass index Furthermore, the generalized additive model (GAM) was used to explore the non-linear relationship between continuous variables (age and BMI) with WOI displacement. The GAM analysis adjusted with confounders showed that age has a curved positive relationship with the risk of WOI displacement. The threshold detected by segmented regression was 35 (Fig.  2 ); the risk of WOI displacement increases after 35 years old. GAM analysis showed that BMI was positively related to the risk of WOI displacement; the threshold detected by segmented regression was 22 (Fig.  2 ). Fig. 2 Adjusted generalized additive model showing the relationship of female age or BMI with WOI displacement Adjusted generalized additive model showing the relationship of female age or BMI with WOI displacement Therefore, in the GEE model2, we transformed the age and BMI from a continuous variable to a categorical variable (Female age: <35 years or ≥ 35 years; BMI: <22 kg/m 2 or ≥ 22 kg/m 2 ). Results of the GEE model2 showed that patients with an EP history had a 62% higher risk of WOI displacement than those without EP (aOR 1.62, 95%CI 1.03–2.53, p -value 0.035); The WOI displacement risk was 50% higher in patients over 35 years old than those under 34 (aOR 1.50, 95% CI 1.12-2.00, p -value 0.007). Secondary infertility showed a 26% lower risk of WOI displacement than primary infertility (aOR 0.74, 95% CI 0.54–1.02, p -value 0.062) without statistical significance. The WOI displacement risk was 25% higher in patients with BMI ≥ 22 kg/m 2 than those with BMI < 22 kg/m2 (aOR 1.25, 95% CI 0.94–1.67, p -value 0.12) without statistical significance. The above analyses showed that ectopic pregnancy and advanced age are risk factors for WOI displacement risk. We further compare the WOI displacement rate in patients grouped by ectopic pregnancy or female age. As Fig.  3 ; Table  3 show, the overall WOI displacement rate is 39.83% (372/934). The WOI displacement rate in patients with EP is 46.67% (56/120), and the WOI displacement rate in patients without EP is 38.82% (316/814), considering the correlation of EP with infertile type, the WOI displacement rate was further calculated in the secondary infertile patients, the WOI displacement rate in secondary infertile patients without EP is 36.51% (115/315). 13 patients conceived ectopic pregnancy following embryo transfer, the WOI displacement rate of the 13 patients were 69.23% (9/13). The WOI displacement rate in patients who conceived EP following embryo transfer was significantly higher than that of patients without EP irrespective of infertile type ( p -value = 0.04). The WOI displacement rate in patients with EP history was higher than that of patients without EP history but without statistical significance ( p -value = 0.1), however, when considering the correlated effect of infertile type, the WOI displacement rate was higher in patients with EP history than secondary infertile patients without EP history ( p -value = 0.05). Fig. 3 Bar plot showing the WOI displacement rate of different populations Bar plot showing the WOI displacement rate of different populations Table 3 Rates of window of implantation (WOI) displacement in different populations All patients All non-EP Patients with EP Non-EP with secondary infertility EP following ET Age < 35 Age ≥ 35 Number of patients 934 814 120 315 13 594 340 Number of WOI displacement 372 316 56 115 9 224 148 Rates of WOI displacement 39.83% 38.82% 46.67% 36.51% 69.23% 37.71% 43.53% Note: WOI: window of implantation; EP: ectopic pregnancy; non-EP: patients without history of ectopic pregnancy; ET: embryo transfer Rates of window of implantation (WOI) displacement in different populations Note: WOI: window of implantation; EP: ectopic pregnancy; non-EP: patients without history of ectopic pregnancy; ET: embryo transfer The WOI displacement rate in women aged < 35 was 37.71% (224/594), while the WOI displacement rate in women aged ≥ 35 was 43.53% (148/340). The WOI displacement rate was higher in patients ≥ 35 years than in patients < 35 years but without statistical significance ( p -value = 0.08).

The pregnancy rate of in vitro fertilization (IVF) and embryo transfer (ET) is influenced by various factors, with endometrial receptivity being a significant factor [ 1 ]. The embryo can implant into the endometrium only during the window of implantation (WOI). WOI is a specific time frame when the endometrium becomes receptive to the blastocyst, it usually occurs between days 7 to 10 post-ovulation during the secretory phase [ 2 ]. It is reported that endometrial receptivity may last between 12 h to 2 days with individual variability. WOI may be advanced or retarded thus preventing the embryo from successful implantation [ 3 ]. Previous genomic signatures of the endometrium show the incidence of a WOI displacement is up to 25% [ 4 ], which may lead to poorer pregnancy outcomes following embryo transfer. Nowadays, endometrial genomic signatures-based technologies such as endometrial receptivity array (ERA) or endometrial receptivity testing (ERT) are used to evaluate the receptivity of the endometrium to optimize ET timing [ 5 ]. However, the effect of ERA-guided or ERT-guided personalized embryo transfer (pET) on pregnancy outcomes remains controversial. Some research revealed a positive role of ERA-guided pET in improving pregnancy outcomes in specific patient cohorts, especially in patients with recurrent implantation failure [ 6 – 8 ]. However, other studies found that ERA-guided pET does not improve pregnancy outcomes [ 9 – 14 ]. Routine ERA in the first embryo transfer cycles does not improve the live birth rate [ 15 ]. ERA-guided pET during either autologous or donor cycles after a failed transfer attempt did not improve reproductive outcomes [ 16 ]. A recent meta-analysis fails to reveal a significant improvement in pregnancy rate following ERA-guided pET [ 14 ]. What is worth noting is that the most considerable heterogeneity between these studies is the different study populations, which may contribute to explaining the inconsistent results. Therefore, identifying the population that may benefit from ERA is urgently needed. Recent studies focusing on individual variances in endometrial receptivity across different patient groups reveal that patients with a history of implantation failure due to endometriosis exhibit a significantly altered WOI [ 17 , 18 ]. Patients experiencing recurrent implantation failure have an increased likelihood of an abnormal implantation window, with pregnancy rates improving after ERA-guided pET [ 19 , 20 ]. Additionally, studies indicate that endometritis and obesity may impact endometrial receptivity [ 21 , 22 ]. In summary, current evidence did not support a significant improvement in reproductive outcomes following routine ERA-guided pET in the overall population [ 9 – 16 ]. ERA-guided pET tends to improve reproductive outcomes in particular populations such as RIF patients who experience a higher risk of WOI displacement [ 6 – 8 ]. Therefore, identifying the specific population who experience a higher risk of WOI displacement may help to find potential patients who may benefit from ERA. However, the risk factors associated with a higher risk of WOI displacement are not well understood. Whether an association exists between ectopic pregnancy or maternal age with WOI displacement has not been evaluated. This study explored the risk factors associated with WOI displacement based on ERT and confirmed that ectopic pregnancy and advanced maternal age are independent risk factors for WOI displacement. This study helps to target the potential population that may benefit from ERA. This study suggests that future EAR/ERT investigations are better conducted in selected populations with one or more risk factors of WOI displacement, such as patients with advanced age and/or with a history of ectopic pregnancy.

Ectopic pregnancy and advanced age (≥ 35) are significantly associated with increased risk of WOI displacement. Primary infertility and higher BMI (≥ 22 kg/m 2 ) tend to increase the risk of WOI displacement without statistical significance. ERT can be applied to particular populations with one or more risk factors including ectopic pregnancy, advanced age, higher BMI, or primary infertility.

The current study finding ectopic pregnancy and advanced age (≥ 35) are independent risk factors for Window of implantation (WOI) displacement. Primary infertility and higher BMI (≥ 22 kg/m 2 ) tend to increase the risk of WOI displacement but without statistical significance. WOI denotes a specific phase in the menstrual cycle, usually occurring between days 19 and 23, during which the endometrium is primed to facilitate embryo implantation. The methods of detecting endometrial receptivity are mainly based on the following techniques: histopathology examination, ultrastructure observation, and molecular detection. In 1975, Noyes et al. [ 28 ] proposed the criteria for endometrial morphological dating, which divided the endometrium into early-proliferative, middle-proliferative, late-proliferative, early-secretive, middle-secretive, and late-secretive phases. WOI usually opens in the middle-secretive phase. Noyes’ endometrial dating is a crude histomorphological method to assess endometrial receptivity in early times. However, morphological observation is relatively subjective. There will be inconsistent results between two observers [ 29 ]. Moreover, some fine-structure changes could not be observed through endometrial dating. Subsequently, pinopodes have been recognized as an ultrastructural indicator for WOI [ 30 ]. However, using pinopodes as a maker for endometrial receptivity remains controversial. Some studies have reported that pinopodes are present throughout the luteal phase rather than specific to a particular period during WOI. Nowadays, molecular methods such as ERA or ERT have been applied to accurately define the receptive phase. However, inconsistencies are reported for the effect of ERA-guided pET on reproductive outcomes. Recent studies fail to support the routine use of ERA in the overall population [ 12 ]. The biggest heterogeneity between the studies is the different study populations. Therefore, identifying the specific population that may benefit from ERA is urgently needed. Detecting the factors associated with the increased risk of WOI displacement may help to find potential patients who may benefit from ERA. Our study has found that ectopic pregnancy is significantly associated with the increased risk of WOI displacement, which has not been reported before. Previous studies found that patients with EP may encounter hormonal imbalances and inflammatory responses, potentially affecting endometrial receptivity. The tubal factor is a known risk factor for EP. Tubal factors can indirectly influence endometrial receptivity by impacting the expression of HOXA10 [ 31 ], implying the possible association between EP and endometrial receptivity. Our study shows a significant relationship between EP and a higher risk of WOI displacement. However, the causal relationship between EP and WOI displacement is unclear. Whether EP-associated pathophysiologies lead to WOI displacement or WOI displacement leads to the occurrence of ectopic pregnancy need to be further studied. It is well-known that advanced age has a detrimental effect on pregnancy outcomes probably due to decreased oocyte quality and endometrial receptivity [ 32 , 33 ]. Our study has identified that advanced age (≥ 35) is a significant independent risk factor for WOI displacement (Table  2 ). WOI displacement risk increases after 35 years old (Fig.  2 ). Previous studies found that female age begins to affect endometrial gene expression at 35 years, with the differential genes significantly enriched to cilia motility and ciliogenesis [ 34 ], indicating diminished endometrial function in older women. The reported endometrial functional dysregulations could contribute to diminished embryo implantation with aging. Our study further underscores the significance of female age on endometrial receptivity, suggesting the consideration of advanced age as one of the indications of ERA. Subsequent studies are needed to demonstrate if advanced age-related WOI displacement underlies some cases of implantation failure and whether ERA-guided pET is efficient in improving pregnancy outcomes in these cases. Besides, our study indicates that higher BMI (≥ 22 kg/m 2 ) and primary infertility tend to increase the risk of WOI displacement without statistical significance. Research demonstrated that higher BMI is associated with adverse pregnancy and perinatal outcomes [ 35 , 36 ]. Obesity is associated with significant endometrial transcriptome changes compared to non-obese patients and may impact endometrial receptivity [ 22 , 37 , 38 ]. Using the generalized additive model, Zheng et al. [ 35 ] found that the cumulative live birth rate (CLBR) declined with increasing BMI, with 24 kg/m 2 as a threshold. However, our adjusted GAM identified 22 kg/m 2 as a threshold for WOI displacement. However, the result is non-statistically significant ( p -value = 0.12) partly due to the sample size. Therefore, the results should be explained with caution. Further studies are needed to investigate whether BMI is associated with WOI displacement and detect the threshold. Our study showed that primary infertility is associated with a higher risk of WOI displacement without statistical significance ( p -value = 0.062). Though the result failed to show a statistical significance, the tendency could not be ignored, for the non-statistically significant result may be attributed to insufficient sample size. Lessey et al. found patients with primary infertility have defective endometrial receptivity compared to fertile women or secondary infertile women [ 39 ]. The overall WOI displaced rate of our study is higher than previous study [ 4 ]. The explanations for our higher WOI displacement rate include: (1), the population included in the study are patients who experienced at least one failure cycle; (2) the patients who performed ERT in our study are under HRT protocol, the superphysiological hormone levels may impact WOI; (3) the effect of ethnicity could not be ruled out. The following limitations of this study should be underlined: First, the patients included in the current study experienced at least one implantation failure cycle, all interpretations of the results are based on this population, and whether the results can be extrapolated to other populations is unknown. Second, though the potential factors are adjusted, confounding factors that were not included in the database were not able to be analyzed. Third, the patients performed only one ERT, the effect of WOI variation in the same patient could not be assessed. Because inconsistencies are reported for the ERA in the same patient, indicating some month-to-month variation. Finally, the retrospective nature of this study is another limitation. Nevertheless, this study has found ectopic pregnancy to be a novel significant risk factor for WOI displacement. Besides, we have detected a threshold effect of female age on WOI displacement, WOI displacement risk increases after 35 years old. Moreover, multivariate regression analyses by GEE and GAM models are used to control the confounding factors. In summary, our study provides information about the risk factors of WOI displacement, finding that ectopic pregnancy and advanced age (≥ 35) are significantly associated with increased risk of WOI displacement, primary infertility and higher BMI (≥ 22 kg/m 2 ) tend to increase the risk of WOI displacement though without statistical significance. Our study helps to find the target population that may benefit from ERT. However, the effect of ERT-guided pET on patients with EP and advanced age is not investigated. Future prospective studies are needed to explore other possible risk factors and validate the value of ERT in the selected populations.

The study is a retrospective study that investigates the risk factors associated with WOI displacement and confirms whether ectopic pregnancy is associated with WOI displacement based on endometrial receptivity testing (ERT). The retrospective study consists of 934 patients who underwent ERT at the Reproductive Medicine Center of Xiangya Hospital from January 2020 to April 2024. This study was approved by the Ethics committee of Xiangya Hospital (Approval number: 2023006). Each patient experienced at least one implantation failure cycle. The generalized estimation equation (GEE) models were used to test the factors associated with WOI displacement, which were adjusted with confounding factors. The association between ectopic pregnancy and WOI displacement was confirmed by the GEE model adjusted with female age, body mass index (BMI), and infertile type. We detected the relationships between continuous variables (such as female age and BMI) with the risk of WOI displacement by generalized additive models (GAM) adjusted with confounding factors. If a non-linear relationship exists between the continuous variables with the risk of WOI displacement, segmented regression was performed to detect the thresholds. In this study, ectopic pregnancy is diagnosed with sonography evidence or surgical pathology evidence combined with hCG testing. Ectopic pregnancy can be diagnosed when all the following three characteristics are met: (1) positive hCG testing; (2) no gestational sac sonography in the uterine cavity, or uterine curettage pathology showed no embryo tissue; (3) gestational sac is seen outside the uterine cavity by sonography, or surgery is performed and confirmed the embryo tissues outside of the uterine cavity by pathologic examination. Treatments of ectopic pregnancy include surgical removal of pregnancy tissues, salpingectomy, or methotrexate. The time interval from ectopic pregnancy to assisted reproduction treatment is not limited in this study. The inclusion criteria in this study are patients who experienced one or more implantation failures and underwent endometrial receptivity testing. The exclusion criteria are patients who have a history of suspected ectopic pregnancy but can not be confirmed by sonography or pathology evidence. PGT patients are not excluded from the study because PGT is not likely to impact the result of ERT. The patient underwent endometrial receptivity testing (ERT) following a hormone replacement therapy (HRT) cycle. Estradiol valerate was initiated at a dosage of 4 mg, which was then increased to 6 mg or higher until achieving an appropriate endometrial thickness (> 7 mm). Progesterone supplementation commenced after at least 12 days of estrogen administration if the endometrium was > 7 mm. The initiation day of progesterone supplementation was designated as P  + 0 and endometrial tissues were obtained on P  + 5. The cervix was cleansed with saline before endometrial sampling. An endometrial sampler was placed into the uterine fundus with 5 ~ 10mm 3 of endometrial tissues aspirated into the sampler. The aspirated endometrial tissues were immediately placed into a 1.5 ml RNAlater buffer (AM7020; Thermo Fisher Scientific, Waltham, MA, USA) for tissue preservation. RNA sequencing was carried out within 7 days after endometrial sampling. RNA extraction, library preparation, and sequencing were carried out on endometrial biopsy specimens following the protocol reported in a prior study [ 23 ]. Our established predictive model for WOI [ 23 ], known as rsERT, was applied to estimate the most receptive timing of each endometrial sample. The receptivity status of the endometrium was categorized into three phases according to the previously established machine learning model: pre-receptive phase, receptive phase, and post-receptive phase. The pre-receptive or post-receptive phases were defined as WOI displacement in this study. The categorical variables were presented as percentages and frequencies, the continuous variables with normal distribution were presented as means and standard deviations (SDs), and the continuous variables with non-normal distribution were presented as medians and 25th-75th percentiles. A Shapiro-Wilks test was used to decide the normal or non-normal distribution. We compared categorical variables between groups using the χ 2 test, the Fisher’s exact test was taken when the expected frequency of one or more cells is less than 5. The continuous variables with normal distribution were compared using the t test, and the continuous variables with non-normal distribution were compared using the Mann-Whitney-U test. Descriptive statistics and comparisons between the study groups were performed using the “compareGroups” R package (version 4.8.0) [ 24 ]. The relationship between ectopic pregnancy or other factors with the risk of WOI displacement is analyzed by multivariate regression using the generalized estimation models (GEE) adjusted with confounding factors, the GEE analysis was performed using the “geepack” R package (version 1.3.10) [ 25 ]. The relationship between continuous variables (such as female age and BMI) with WOI displacement was detected by generalized additive models adjusted with confounding factors using the “mgcv” R package (version 1.9-1) [ 26 ]. The segmented regression was performed to detect the cut-off values of the continuous variables using the “segmented” R package (version 2.1-0) [ 27 ]. All statistical analyses were 2-sided, and a p -value of less than 0.05 was considered statistically significant. The statistical analyses were performed by the R software (version 4.3.2, www.R-project.org ).