Can the physical act of sexual intercourse before frozen embryo transfer improve IVF treatment outcome? A Randomized Clinical Trial

Background: Exposure of the female reproductive tract to either seminal plasma or uid component of the ejaculate is benecial to achieving successful embryo implantation and normal embryo development. But whether the "physical" component of of sexual intercourse during the peri-transfer period have any inuence on FET pregnancy outcome is not clear. Methods: We conducted a randomized trial that included 223 patients undergoing IVF treatment at a University-aliated reproductive center from 19 June to 20 August 2018. Enrolled patients undergoing IVF treatment were randomized either to engage sexual intercourse using the barrier contraception (n=116) or to abstain (n=107) one night before FET. The follow-up parameters included biochemical pregnancy, clinical pregnancy, early abortion and implantation rate. Results: Patients having intercourse had higher clinical pregnancy rate (51.72% vs. 37.07%, P=0.045) and implantation rate (38.31% vs. 24.77%, P=0.005) compared to those did not engage intercourse. However, there was no signicant difference of the spontaneous abortion rate between two groups (11.67% 33 vs. 14.63%, P=0.662). Conclusions: Sexual intercourse before the day of embryo transfer may improve the clinical pregnancy and implantation rates during the frozen- thawed embryo transfer (FET) cycle. However, it should be noted that patients choose only one time for sexual intercourse, that is, the night before embryo transfer. study


Background
In the process of natural sexual reproduction, the act of intercourse and presence of seminal uid are essential factors for pregnancy. With respect to the latter, studies have shown that the soluble and exosome-born signaling agents present in the semen interact with the female reproductive tract and induce an immune response that may bene t the pregnancy outcome [1]. However, whether physical act of sexual intercourse has any bene cial effect on the chance of pregnancy during the cycle of ART is still controversial. Studies have shown that sexual intercourse may cause subclinical upper reproductive tract infection that is associated with decreased implantation rates [2] as well as ascending uterine infection during third-trimester gestation [3]. Indeed, the uterus is very vulnerable to pathogens introduced by sexual intercourse because the cervical mucus barrier to ascending infection is disrupted by mechanical passage of the catheter during the embryo transfer procedure of IVF treatment. Moreover, the act of intercourse may induce the event of orgasm that increases the activity of uterine myometrium [4]. Therefore, the intercourse-induced increases in levels of uterine activity are highly correlated with uterine contractions that may be bene cial for embryo implantation and affect IVF outcome [5].
On the other hand, sexual intercourse has bene cial effects on assisted implantation. Studies performed using animal models revealed that exposure of the female reproductive tract to either seminal plasma or uid component of the ejaculate is bene cial to achieving successful embryo implantation and normal embryo development [6][7][8]. For instance, rodent studies performed using arti cial insemination or embryo transfer with being exposed to seminal plasma have substantially higher pregnancy rates than those without being exposed [6,7]. Similarly, rat embryos inseminated with spermatozoa before blastocyst transfer have a higher implantation rate than those without spermatozoa exposure [9]. Additionally, increasing evidence suggests that women who have tissue contact with husband's semen during ART treatment improve their pregnancy outcomes [10][11][12]). A meta-analysis study performed using seven randomized controlled trials that involved 2,204 patients showed that women with either sexual intercourse or passive exposure to semen around the time of IVF treatment had an increase in clinical pregnancy rate (approximately 23 %) [11]. Collectively, these ndings indicate that components in the semen may induce an immune response that may positively affect the endometrial receptivity and pregnancy outcomes.
Interestingly, sexual arousal has been shown to improve the motility of spermatozoa, although the detailed molecular mechanism is not clear [13]. Furthermore, studies revealed that patients with sexual intercourse without contraception between the day of ovulation and embryo transfer had a higher incidence of multiple pregnancies and ectopic pregnancy during their frozen-thawed embryo transfer (FET) cycles [14]. At present, it remains unclear whether the bene cial effects of sexual intercourse on the endometrial receptivity and embryo implantation are due to the transportation of semen into the vagina or the sexual behavior itself. Therefore, we aimed to investigate whether the act of sexual intercourse itself (using condoms as contraception) has any in uence on clinical pregnancy rates in patients with IVF assisted treatment.

Study design
This is a unblinded randomized study conducted at the Center for Reproduction and Genetics of the A liated Hospital of Shandong University of Traditional Chinese Medicine. This study was approved by the University Research Ethics Committee and registered at the China Clinical Trial Registration Center (registration number ChiCTR1800017209). All participants signed an informed consent form prior to the study. The sample size calculation was done based on previous studies, in which the e cacy of sexual intercourse during the peri-transfer period was 24% [10], α = 0.05 and β= 0.2. According to the ratio of 1:1, 10% of the lost visit rate, there were expected to be 104 patients in the the barrier contraception group and 104 patients in the abstain group.

Study participants
A total of 300 infertile patients undergoing IVF frozen-thawed cycle embryo transfer (FET) were recruited from the Center for Reproduction and Genetics of the A liated Hospital of Shandong University of Traditional Chinese Medicine, from 19 June 2018 to 20 August 2018, during this period, patients undergoing embryo transfer will be included after signing informed consent. Inclusion criteria were: 1) at least one cycle of IVF/ICSI performed in our reproductive center; 2) female age ≤ 35 years; 3) a body mass index (BMI) between 18.5 and 25; 4) regular menstrual cyclicity (24-35 days); 5) the attending physician and embryo transfer being the same person for all patients; 6) the tubal factor is the cause of female infertility( bilateral tubal obstruction, or hydrosalpinx treated using laparoscopic surgery); 7) 4 or more embryos available for transfer with at least one viable day 3 embryo with a score of 6 or above according to the Istanbul Consensus [20] and without receiving preimplantation genetic screening/preimplantation genetic diagnosis (PGS/PGD). Exclusion criteria were 1) patients with known congenital uterine malformation, uterine broids, or endometriosis; 2) chromosomal abnormalities in any of infertile couples; 3) female with immune diseases. Three hundred patients (all due to tubal factors) requiring IVF treatment were randomly divided into two groups, with 150 patients in each group. Group A (n = 150) was subjected to have intercourse using a condom at the night before embryo transfer and, while group B (n = 150) was subjected to abstain for the entire IVF cycle. Patients engaged in intercourse one night before embryo transfer using a barrier contraception (condoms). All patients participating in the study underwent vaginal scrubbing prior to embryo transfer.

Randomization
The randomization sequence was generated using a computer program by a statistician, who was not clinically involved in the trial. Signed written informed consents were obtained from all participants before the study. On the day the patient determines the date of FET, a total of 300 patients were divided into 2 groups according to the principle of random control with 150 cases in each group. Information about whether patients engage sexual intercourse using the barrier contraception or to abstain one night before FET is sealed in an opaque envelope with only one number on it. In order to prevent patients from not following the doctor's orders, we will contact the patients and their spouses again according to the results after unblinding to con rm that the content provided is true.

Interventions
In the present study, arti cial endometrium preparation by the exogenous administration of estrogen and progesterone was used for the thawed embryo transfer cycle. To prepare the endometrium, all patients self-administered oral estradiol valerate 8 mg per day started at the third to fourth day of the menstrual period for 5 days, and then 12 mg per day for the remaining days of the cycle, which was adjusted according to the clinical situation during this period. Estrogen administration was continued until the endometrium reaches a thickness of 8 mm (determined using an ultrasonographic examination). and then progesterone was combined to initiate the secretory phase in preparation for FET. Luteal support was routinely provided after FET for 14 days irrespective of pregnancy. For patients con rmed with a clinical pregnancy,estradiol valerate and progesterone were continued until 10 weeks of gestation, which was gradually reduced until detecting the fetal heart beats.
Participant follow-up and data collection Serum β-hCG levels were examined using electrochemiluminescence analysis (Cobas e411 System Product, Germany) on day 14 following embryo transfer in all patients. Implantation was de ned as a serum quantitative β-hCG ≥10 u/L. All pregnancies were followed until 12 weeks after FET. A biochemical pregnancy is de ned as a transient but signi cant increase in serum β-hCG ≥10 u/L between days 12 and 20 after embryo transfer without detecting a gestational sac by an ultrasonography. A clinical pregnancy is de ned as the detection of fetal heart beats using ultrasonographic examinations 35 days after FET. An early abortion is de ned as the detection of an empty gestational sac using ultrasonographic examinations before 12 weeks of gestation. Implantation rate is de ned as the ration of the number of embryos implanted and the number of transferred embryos.

Statistical analysis
The statistical software package, IBM SPSS Statistics for Windows, version 22.0 (IBM Corporation, Armonk, NY, USA), was used for all data analyses. The data are expressed as mean ± standard deviation (x ± s). Among those were normally distributed and homogeneity of variance (homoscedasticity) was satis ed, a t test was performed; otherwise, Mann-Whitney U test was performed. The treatment outcomes for each group were counted and expressed as a frequency. When the total sample size was greater than 40 and the minimal theoretical frequency was greater than 5, we applied the Chi-square test.
When the total sample size was less than 40 or the minimal theoretical frequency was less than 5, we applied the Fisher exact-probability method. P 0.05 was considered statistically signi cant.

Results
Owing to non-compliance with medication and personal concerns resulting in cycle cancelation, several patients lost to follow up in this study. Ultimately, 223 patients were included in the analysis, with 116 cases in group A and 107 cases in group B (Fig. 1). The characteristics of trial participants were similar, con rming the adequacy of the randomization process. No signi cant difference in major parameters between the two groups, including maternal age (

Discussion
Assisted reproductive technology (ART) has evolved rapidly over the past 30 years, which is as frequent for conceiving a child as any other minor medical procedure. Despite the recent advances in IVF, the successful rates (clinical pregnancy and subsequent live birth rates) remains disappointingly low because of the multiple causes of infertility, individual patient differences, and intolerance to embryo transfer. Among these multiple factors, the implantation rate is limited despite the transfer of excellentquality embryos. Indeed, repeated failures of implantation due to inadequate uterine receptivity has become a compelling challenge in reproductive medicine. A meta-analysis revealed a statistically signi cant improvement in clinical pregnancy rate when women are exposed to seminal plasma (unprotected sexual intercourse, intra-vaginal, intracervical, or intrauterine application) around the time of oocyte pick-up or embryo transfer [11]. Studies performed using animal models and human samples suggest an immune-regulatory role for seminal plasma in embryo implantation [11]. But at present, no trial study has investigated whether the "mechanical" act of sexual intercourse before the time of embryo transfer has any in uence on implantation and clinical pregnancy.
The results obtained from this randomized study are the rst to provide evidence that the act of sexual intercourse (excluding the effect of semen plasma) during the time of embryo transfer is not detrimental to implantation or pregnancy outcome. Furthermore, female patients having sexual intercourse exhibited substantial improvement in both implantation and pregnancy rates. All patients underwent arti cial cycle FET. Therefore, natural conception of the enrolled patients in the sexual intercourse group was unlikely to occur.
The observed ndings that sexual intercourse had a bene cial effect on pregnancy outcome may be explained by several factors. The physical act of sexual intercourse arousal has been shown to increase the blood ow of female genital tract [15], which could enhance the blood supply of the endometrium, resulting in improvement of endometrial receptivity. Indeed, increasing evidence suggests that adequate uterine perfusion contributes to the uterine receptivity, and an increased impedance in uterine ow is associated with infertility [16,17]. Future studies are required to investigate the distribution of endometrial blood ow and the impedance changes in uterine artery blood ow before and after sexual intercourse. Previous studies also showed that sexual intercourse can activate various female genital tract re exes, which may further induce several bene cial effects, such as increased production of sex hormones and maintenance of general tract functionality [13]. The other potential bene cial impact of sexual intercourse on the pregnancy outcome is that coitus is considered as a type of behavioral treatment approach that might be e cacious for emotional aspects of infertility. A study performed including 54 infertile patients showed that behavior treatment signi cantly decreased degree of anxiety, depression, and fatigue, but increased vigor, of which 34% of these women became pregnant later on [18]. Therefore, if the act of sexual intercourse can make women less stressful, it may, to some extent, reduce the anxiety of patients perceived prior to embryo transfer. Moreover, encouraging couples to engage in sexual intercourse during their IVF treatment gains an additional psychological advantage, as they might feel more natural and comfortable during the process of conception. Certainly, all these proposed mechanisms are issues worth exploring further.
Of course, there are some limitations to this study. For example, it is important to note that our selected time of coitus for patients was only once, the night before the embryo transfer. Our interpretation of the results obtained in the present study has been too brief an interval to show bene cial effects. It is di cult to determine which day is the best time point for sexual intercourse between oocyte pick-up and embryo transfer. However, a study performed using animal models showed that the pregnancy rates after arti cial insemination in ewes were much lower than those after natural mating, although the mechanism remains unclear [19]. Therefore, the results obtained from animal studies suggest that the act of natural mating does not appear to increase uterine activity and subsequent pregnancy rate. Future studies aimed at addressing these issues by conducting a study design with various time points of sexual intercourse to investigate the effects on subsequent pregnancy outcomes. Besides, limitations of our study also include the non-blindness and small sample size which will reduce the persuasiveness of our experiments. It also cannot be ignored the fact that the number of good quality embryos transferred in group A has a borderline signi cance which will also have an impact on the experimental results.
In summary, clinical data obtained from this study demonstrate that the "physical" component of sexual intercourse before the day of embryo transfer may improve the clinical pregnancy and implantation rates during FET cycles. However, there is much work to be done in investigating the functional relationships and the underlying mechanisms to support the theoretical basis. A large-sample, multicenter, randomized controlled trial study is therefore needed to con rm the effects of sexual intercourse on IVF outcomes. Indeed, it is practically not feasible to control the process and duration of sexual intercourse, leading to uncertain whether these factors might have an impact on IVF outcomes.

Abbreviations
FET frozen-thawed embryo transfer IVF in vitro fertilization Declarations Ethics approval and consent to participate This study was approved by the University Research Ethics Committee. All participants signed an informed consent form prior to the study. Consent for publication: Publication consent was obtained from all authors.
Availability of data and material: Data and material are available with the corresponding author.
Competing interests: Competing interests: All authors declare that they have no competing interests.
Funding: This study was funded by the National Natural Science Foundation (81674018). All authors have no con ict of interest.
Authors' contributions: Li-Hua Yuan and Xian-Ling Cao responsible for data collection and thesis writing, Zhen-Gao Sun responsible for the revision of articles and enrollment of patient.

Item
Group A Group B P Group B: patients were engaged to abstain BMI, body mass index; bFSH, basal follicle-stimulating hormone (on day 2); data are expressed as means ± variance (s).  Figure 1 Flow chart of case selection.