Predictive value of D‐dimer in patients with unexplained recurrent implantation failure during freeze‐thaw embryo transfer cycles

This study aimed to evaluate whether D‐dimer can predict the clinical outcomes of patients with unexplained recurrent implantation failure (URIF) during freeze‐thaw embryo transfer (FET) cycles.


BACKGROUND
Even after years of development, implantation failure in in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI)-embryo transfer (ET) remains an impenetrable barrier to increasing successful pregnancy outcomes. In many cases, women who undergo repeated IVF/ICSI-ET attempts with good-quality embryos ultimately fail to become pregnant. As early as 1983, the term recurrent implantation failure (RIF) was used to describe embryo implantation failure following IVF/ICSI. To the best of our knowledge, there is still a lack of a specific standard that clearly defines RIF, which include the number of failed cycles, the type of transfer cycle (fresh embryo transfer cycle or freeze-thaw embryo transfer cycle) and the total number of transferred embryos that have not been successfully implanted. [1][2] In previous research, Polanski proposed two conditions that can be used to determine the occurrence of RIF: at least two fresh or freeze-thaw embryo transfer implantation failures and either at least four highquality cleavage-stage embryos transferred or at least two blastocysts transferred. 3 Implantation is a complex physiological process involving the embryo and the endometrium that involves adhesion, nidation and invasion of the trophoblast within the endometrial tissues. Current studies have found that many factors can affect implantation failure.
However, for most of these factors, the corresponding pathogenesis has not been established. It has been proven that natural prothrombotic events are caused by physiological changes during pregnancy. 4 The hemostatic balance in placental vessels may be altered during pregnancy. These events may cause hypercoagulable states, which may in turn lead to inadequate fetomaternal circulation due to reduced placental perfusion. 5 Previous studies have suggested that some cases of recurrent pregnancy loss are a consequence of an exaggerated hemostatic response. [6][7][8] Although the association between thrombophilia and recurrent pregnancy loss is well known, there is limited research on the association between thrombophilia and RIF. Similar to recurrent miscarriage, thrombophilia plays an important role in RIF through various mechanisms, which has also drawn the attention of several researchers. During implantation, upregulated expression of tissue factors such as thrombin generation can promote human endometrial stromal cell hemostasis by protecting against bleeding due to endometrial capillaries being invaded by implanting cytotrophoblasts. 9 It has been hypothesized that the invasion of maternal vessels by syncytiotrophoblasts can be affected by localized thrombosis at the implantation site, leading to IVF failure. 10 Foad Azem et al. claimed that inherited thrombophilia had some impact on the etiology of RIF. 11 In their case-control study, they found that hereditary thrombophilia was more prevalent in patients with a history of RIF than in healthy women.
In recent years, increasing evidence has implied that correlations exist between RIF and thrombophilia (both inherited and acquired). [12][13] However, the exact effect of thrombophilia on URIF is still unclear.
Early detection of the coagulation cascade's progressive activation in thromboembolic diseases can enable management of the problem. D-dimer is currently considered a useful biochemical marker in thromboembolism. 14 It is a product of increases in fibrin degradation during acute thrombotic processes due to secondary activation of the fibrinolytic system. The level of plasma D-dimer may reflect the coagulation status in URIF patients.
In this study, we determined the association between plasma Ddimer levels and clinical outcomes in patients with URIF during FET cycles. Moreover, we aimed to evaluate whether plasma D-dimer levels can be used to predict the pregnancy outcomes of URIF patients in clinical practice.

Clinical management
The FET cycle protocol was based on patient characteristics. All patients underwent either a natural cycle or a substituted cycle.
After spontaneous menses, a vaginal ultrasound examination was performed on the 10th−12th day of the cycle to detect the leading follicle.
Based on the ultrasound examination, patients had their plasma levels of luteinizing hormone (LH), estradiol (E 2 ) and P checked. ET was scheduled between 3 and 6 days after ovulation upon ultrasound examination. Luteal phase support was commenced on the day of ET using dydrogesterone (Abbott Healthcare Products B.V., Netherlands) at a dose of 10 mg twice daily for 14 days. If the pregnancy test was positive, luteal phase support was continued until the 10th gestational week.
(2) This protocol was used in patients with irregular menses (>37 days Otherwise, it was stopped.

Laboratory tests
The plasma D-dimer and other coagulation indictors of the patients on the day of embryo transferred were monitored. The levels of D-dimer and other coagulation indictors were detected using an ACL-Advance automatic coagulation analyzer (ACL-TOP-700, Werfen Inc, Spain).
Total platelet count in blood samples was measured using a Sysmex XN-9100 Hematology Analyzer (Sysmex, Kobe, Japan).

Study period and groups
The study was divided into two parts. First, enrolled patients were collected from January 2017 to January 2020 and divided into two groups according to clinical pregnancy outcomes. Patients with at least one live birth (LB) were classified as LB (+), while the others were classified as LB (−). We compared clinical data, especially plasma D-dimer, between these groups. Additionally, the simultaneous relationship of multiple prognostic factors for pregnancy outcome was assessed using binary logistic regression analysis. Moreover, we adopted receiver operating characteristic (ROC) curve analysis to determine the cutoff value of plasma D-dimer for the live birth rate (LBR).
In the second part, we prospectively included patients from January 2020 to December 2020. We used the cutoff value of plasma D-dimer from the retrospective study to classify the included patients into two groups, that is, a low D-dimer group and a high D-dimer group, and compared their clinical outcomes.

Definition of outcomes
The LBR and clinical pregnancy rate (CPR) were the major outcomes of this study. A clinical pregnancy is defined as a pregnancy diagnosed by ultrasonographic visualization of one or more gestational sacs. The CPR is defined as the percentage of clinical pregnancies occurring in every 100 ET cycles. The LBR is defined as the percentage of deliveries that resulted in at least one live birth, expressed per 100 ET cycles. The following clinical outcome was also included: implantation rate, which was defined as the number of gestational sacs detected divided by the number of embryos transferred.

Statistical analysis
All analyses were performed using SPSS version 11.0 (SPSS Inc., Chicago, IL). The significance of differences in proportions was tested using chi-square and Fisher's exact statistics. Differences in continuous variables were analyzed using Student's t-test and the Mann-Whitney U test. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.
Binary logistic regression analysis was used to assess whether the pregnancy outcomes were affected by maternal age, BMI (kg/m 2 ), infertility type (primary infertility or secondary infertility), endometrial thickness, the number of embryos transferred, and the type of embryos transferred (cleavage stage embryo or blastocyst). An ROC curve was drawn to obtain the plasma D-dimer threshold, and the area under the curve (AUC) was calculated. Values of P < .05 were considered statistically significant unless otherwise noted.  All data was recorded as mean ± standard, median (1st quartile-3rd quartile), or coefficient of variation (%). The significance of differences in proportions was tested using chi-square statistics. Differences in continuous variables were analyzed using Student's t-test and the Mann-Whitney U test.   Table 2.

Outcomes of patients in the retrospective study
To determine the normal plasma D-dimer concentration threshold to apply in our trial, we used ROC curve analysis to analyze all included patients, as is shown in Figure 2 The

Outcomes of patients in the prospective study
About 113 patients underwent freeze-thaw cycles during the prospective study period. Among these patients, no one was lost to follow up. We analyzed the characteristics and pregnancy outcomes of these patients and categorized them into low and high D-dimer groups (>0.22 mg/L and ≤0.22 mg/L, respectively) based on the ROC curve results from the retrospective study. Table 3    All data was recorded as mean ± standard, median (1st quartile-3rd quartile), or coefficient of variation (%). The significance of differences in proportions was tested using chi-square and Fisher's exact statistics. Differences in continuous variables were analyzed using Student's t-test and the Mann-Whitney U test. Abbreviaton: BMI, body mass index.
to implantation failure. In clinical work, many couples transfer goodquality embryos cycle by cycle but fail to achieve clinical pregnancy. The majority of these couples do not find any positive result in the routine examination, a condition called URIF. URIF not only causes physical and mental harm to patients but is also considered an important cause of infertility.
Interest in thrombophilia in RIF patients is largely influenced by the clinical association between alteration of hemostasis and a trend toward a hypercoagulable state and recurrent pregnancy loss. In recent years, several researchers have found that thrombophilic single nucleotide polymorphisms and acquired thrombophilia were more prevalent in the IVF failure group. [12][13]15 A possible mechanism for URIF is thrombosis of the maternal vessels, especially the decidual or chorionic vessels, which can reduce perfusion of the intervillous space, preventing embryo implantation. 16 The prethrombotic state is a pathological state involving hemo- In research on recurrent pregnancy loss, Wang T et al. found that Ddimer is a useful predictor of unexplained recurrent pregnancy loss. 7 Wang P et al. also found that the concentration of D-dimer can not only distinguish patients with recurrent pregnancy loss from those with normal fertility but also more accurately indicates recurrent spontaneous abortion prognosis. 18 Interest in thrombophilia in RIF patients is based on the hypothesis that microvascular thrombosis at the decidua can reduce perfusion of the intervillous space, leading to placental failure. [19][20] However, only a few studies on female per-implantation blood coagulation status have elucidated the relationship between coagulation status and URIF. [11][12][13]15 Physiologically In the prospective part of the present study, we verified this result. Our data showed that the basal plasma D-dimer level was a strong predictor of pregnancy results in URIF patients receiving FET.
The implantation rate, CPR and LBR were higher in URIF patients Our studies were limited by the small sample size and the retrospective nature of the first part of the study. In addition, our results did not take consider uterine receptivity 32 and other clinical parameters, such as thromboelastographic parameters, into account. A larger prospective study is needed to validate the results.

CONCLUSIONS
In conclusion, our results suggested that URIF patients with plasma D-dimer levels lower than 0.22 mg/L had a higher CPR and LBR in FET cycles. Plasma D-dimer levels may be used as predictors of clinical outcomes in URIF patients undergoing FET cycles and may guide appropriate anticoagulant treatment.