INTRODUCTION

Polycystic ovary syndrome (PCOS) is the most common (5-8%) multisystemic endocrine and metabolic disorder in women of reproductive age. Its incidence may be up to 13% according to the population studied (1,2). The presence of at least two of the criteria for oligoovulation and/or anovulation, biochemical or clinical evidence of hyperandrogenism, and polycystic ovarian morphology is sufficient to diagnose PCOS (3). Although most women with PCOS experience oligomenorrhea or amenorrhoea due to anovulatory cycles, both heavy and irregular bleeding can sometimes occur. Although oligo-ovulation is accepted as the main cause of infertility in patients with PCOS, even assisted reproductive techniques do not satisfactorily increase the decreased pregnancy outcomes in patients with PCOS (4). These data suggest that anovulation alone is not responsible for subfertility in PCOS. In connection with the last sentence, increased early pregnancy loss in women with PCOS suggests that parameters such as endometrial dysfunction, obesity, and hyperandrogenemia also affect the fertility status of patients (5).

The PCOS endometrium differs from the healthy endometrium in terms of histomorphology, steroid receptors and coactivators, and receptivity modulators. Therefore, the receptive status of the endometrium is often not suitable for implantation, and endometrial dysfunction in addition to oligoanovulation also contributes to subfertility. In PCOS, progesterone-dependent changes in the endometrium do not occur because of oligo-ovulation or anovulation. Increased androgen-estrogen conversion in peripheral tissue and increased free estradiol and testosterone levels due to hyperinsulinemia may cause unpredictable bleeding in the absence of progesterone. The endometrium of women with PCOS exhibits increased expression of ERa and AR compared with healthy controls (6,7). Differences in receptor expression in PCOS patients may be related to unopposed estrogen elevation. Both circulating portions of estrogen and progesterone and differences in receptor expression prevent the opening of the implantation window in PCOS. On the other hand, the PCOS endometrium fails to produce a physiological response to progesterone, which is an important evidence of progesterone resistance (8). In summary, decreased progesterone levels and receptor expression defects in patients with PCOS lead to the presence of unopposed estrogen, leading to disruption of the normal menstrual cycle and subfertility. Most PCOS patients present with oligo- or amenorrhoea although there is rarely heavy bleeding due to increased estrogen levels.

In the absence of ovulation and progesterone, the endometrium does not undergo secretory transformation and is constantly exposed to estradiol. However, although this rarely leads to excessive bleeding and hyperplasia, in most PCOS cases, menstrual cycle intervals are prolonged and become oligomenorrheic. Although making the cycles ovulatory with medical agents restarts the menstrual cycle, the picture of amenorrhoea reappears when the treatment is stopped. Platelet-rich plasma (PRP), because of its regenerative potential, is more widely used in reproductive medicine, as it is in every field of medicine (9,10). Although there are studies showing improvement in fertility outcome after the use of intraovarain or intraendometrial PRP in reproductive medicine, most of them are studies with low level of evidence due to both design, PRP standardization, and inhomogeneous participant population (9,10). A thin endometrium is defined as an endometrial thickness (EMT) of 7 mm or less on the day of hCG or embryo transfer (11). The number of studies investigating the effects of PRP application on EMT and fertility outcome in an infertile patient population is quite limited. Although EMT varies in different phenotypes of PCOS, the incidence of a thin endometrium is higher in women with PCOS where hyperandrogenemia is dominant (12). In patients who have ovulation induction with clomiphene, the EMT is further thinned. The primary aim of this study was to investigate the effects of autologous intrauterine PRP (IU-PRP) application during ovulation induction with clomiphene citrate (CC) on EMT and fertility outcomes in patients with PCOS with thin endometrium (<7 mm) resistant to adjuvant treatments. The secondary outcome was to detect possible transformations in oligomenorrheic cycles after PRP.

METHODS

This study was conducted on 35 anovulatory PCOS patients, aged between 22 and 29 years, who applied to our gynecology and obstetrics clinic for infertility treatment. The patients had a history of a thin endometrium in their previous treatment history. EMT 7 mm was considered thin endometrium. The diagnosis of PCOS was made according to the revised Rotterdam criteria. Those who met at least two of the criteria for amenorrhoea/oligomenorrhea with chronic anovulation, clinical and/or biochemical evidence of hyperandrogenism, and ultrasonographic appearance of PCOS were accepted as having PCOS. The diagnosis of anovulation was confirmed by the patient’s history and hormonal measurements. From the anamnesis of the patients, it was noted that adjuvant treatments, such as oral or patch-form estrogen supplementation, low-dose aspirin, and vaginal sildenafil, were applied to increase the EMT, but they did not benefit.

It was explained to the patients that PRP treatment is not a routine practice for the treatment of infertility and for increasing the thickness of the endometrium. It was emphasized that there were studies that reported positive results in EMT and fertility outcome after PRP, and studies that reported no benefit. The study was initiated after obtaining permission from the Clinical Research Ethics Committee of Zeynep Kamil Gynecology and Pediatrics Training and Research Hospital (decision no: 14, date: 11.01.2023). In addition, an informed volunteer consent form was obtained from all patients regarding the procedure to be performed. Twenty patients who accepted PRP and 15 patients who did not were divided into two groups and included in the CC treatment protocol. CC (Serophene, Serono, Roma, Italy) was administered to the patients in both groups for 5 days, starting from the 3^rd day of progesterone-induced withdrawal bleeding at a dose of 100 mg/day. The patients were called for control on the sixth and eighth days of treatment. Follicular development and EMT were recorded using transvaginal ultrasonography (TV-USG). IU-PRP infusion was planned for cases with EMT <7 mm on the eighth day of follow-up. Endometrial PRP infusion was applied to the patients in the PRP group, four days after the 5-day CC treatment, i.e., on the ninth day of the cycle. In the CC alone group, PRP was not applied. In the TV-USG examination after PRP, both EMT and follicle diameter were measured. Timed intercourse was recommended for cases with a follicle with a mean diameter of at least 16-18 mm. The biochemical and clinical pregnancy rates of both groups were recorded. Clinical pregnancy was defined by the presence of an intrauterine gestational sac confirmed by TV-USG.

Autologous PRP Preparation and Endometrial Infusion

Autologous PRP was prepared from each patient following a two-stage centrifugation using autologous blood. On the 9^th day of ovulation stimulation with CC, 15 mL of venous blood was drawn into a syringe containing acid citrate A anticoagulant solution and immediately centrifuged at 300 rpm for 5 min. In the first procedure, red blood cells were removed from the medium. Subsequently, the plasma was centrifuged again at 700 rpm for 17 min and autologous PRP was obtained. Platelet activation was achieved by adding CaCl₂ and human thrombin. Activated PRP was finally centrifuged at 3000 rpm for 20 min, and the supernatant was collected. The collected pellets contained erythrocyst- and leukocyte-free PRP. Approximately 0.5-1 mL of PRP was infused with the IUI catheter into the endometrial cavity of the patient in the lithotomy position, accompanied by ultrasonography. No anesthesia or sedation was used during PRP infusion (13). EMT was measured and recorded again 3 and 6 days after PRP infusion.

Statistical Analysis

The data were analyzed using the Statistical Package for Social Sciences software 18.0 for Windows package software (SPSS). Normality of data was examined using the Kolmogorov-Smirnov test. Mean values were calculated for data showing abnormal distribution and compared with the non-parametric Mann-Whitney U test.  Categorical data are described either by number of cases or percentages. Categorical variables were presented as percentage values and compared using the chi-square/Fisher’s Exact test. Continuous variables were analyzed using the Mann-Whitney U test. The results are presented as mean ± standard deviation. For all tests, p-value <0.05 was considered statistically significant.

RESULTS

Demographic, hormonal, and reproductive parameters of the CC plus PRP and CC alone groups are presented in Table 1. Both groups were similar in terms of participant age and body mass index. All participants in the CC plus PRP group were successfully infused with autologous PRP on the ninth day of the cycle. No PRP application was applied to the CC alone group. Serum estradiol, testosterone, luteinizing hormone (LH), follicle-stimulating hormone (FSH), and LH/FSH ratios of both groups were similar. CC doses and administration times were similar in both groups. Biochemical pregnancy and clinical pregnancy rates of the CC plus PRP group were significantly higher than those of the CC alone group (p<0.03 and p<0.02, respectively). Although clinical pregnancy was detected in 5 individuals in the PRP group (25%), clinical pregnancy was recorded in 2 individuals in the CC alone group (13.3%). No significant change was observed in the oligo/anovulatory cycle patterns of patients with and without PRP.

A detailed presentation of the change in the EMT values of the PRP and CC alone groups is shown in Table 2. EMT values on the sixth (4.96±2.11 mm vs. 4.68±2.47 mm, p<0.37) and eighth days were similar between the two groups (5.11±3.10 mm vs. 5.29±3.01 mm, p<0.51). EMT was not evaluated because PRP was performed on the ninth day. Compared with the CC alone group, the EMT values measured both at day 12 (6.34±1.09 mm vs. 5.47±3.90 mm, p<0.02) and at day 15 (7.44±2.60 mm vs. 6.23±2.70 mm, p<0.01) in the PRP group were found to be significantly higher.

DISCUSSION

Implantation in mammals is a very complex process that requires coordinated cross-talk between the blastocyst and endometrium through molecular pathways. Before implantation, the endometrium must be exposed to adequate levels of estrogen and progesterone to make it susceptible to blastocyst invasion. Ovarain sex steroids exert their effects on the endometrial epithelium and decidualized stromal cells through mediator molecules and immune cells. Peripheral blood-derived immune cells, leukocytes, and platelets mediate the effects of estrogen and progesterone. These effects are mediated by paracrine mechanisms and prepare the endometrium for implantation by regulating the release of growth factors and cytokines. This multi-molecular pathway results in endometrial decidualization, proliferation in epithelial and stromal cells, increased synthesis of receptivity modulators, and provision of an end-point receptive endometrium (14,15). However, pre-implantation reactions at the histomorphological, molecular, and genomic levels in the endometrium can keep the endometrium in the receptive phase within a period of approximately 6 days. Although the endometrium exhibits these cyclic changes regularly in most women of reproductive age, implantation failure remains the most frequent and serious problem in assisted reproductive techniques (16). Therefore, it is critical to determine whether the endometrium is in the receptive phase before implantation and to choose a treatment approach accordingly. Endometrial receptivity tests developed for this purpose in the last decade have not been routinely used because of their invasive nature and low sensitivity and specificity. EMT measurement with ultrasonography, an old but effective marker, remains the only non-invasive and reproducible method available to evaluate the receptive endometrium.

In the ultrasonographic evaluation of the endometrium, a prediction can be made about endometrial receptivity by evaluating echogenicity and thickness. In addition to the hypoechogenic appearance in sonographic evaluations performed on the day of ovulation or hCG, the detection of an EMT of 7 mm or more strongly supports the presence of a receptive endometrium (17). However, pregnancy has also been reported in hyperechogenic endometriums with a thickness of approximately 4 mm (18). Although the presence of high EMT is not always compatible with receptivity status, the chance of pregnancy is lower in women with thin endometrium. A value of <7 mm, measured on the day of ovulation or hCG, is considered as thin endometrium and means that the success rate for implantation is low (11). PCOS is an endocrine cause of subfertility with ovulatory dysfunction, receptivity defect, and thin endometrial development. Although EMT is normal in some PCOS phenotypes, a thin endometrium is more common, especially in PCOS phenotypes with hyperandrogenemia (12). In particular, in cycles using CC for ovulation induction, EMT is seen to be more thinned (19). Although EMT has been increased with methods such as sildenafil citrate, aspirin, and mechanical endometrial injury, the results are heterogeneous (20). There is no study design using PRP to increase EMT in patients with PCOS using CC for ovulation stimulation. This is the first clinical study investigating the effects of PRP on EMT and fertility outcome in patients with PCOS who could not conceive due to thin endometrium in previous cycles. EMT values of both groups were similar before PRP application. We found that EMT measurements performed 3 and 6 days after IU-PRP application increased significantly. Similarly, both biochemical and clinical pregnancy rates in the PRP group increased significantly compared with the control group. Because the oligo-anovulatory cycle patterns of both groups did not change, we can argue that PRP has a phase-specific effect and has no long-term and clear effect on ovulatory functions.

One of the possible reasons for the significant increase in EMT and clinical pregnancy rates in PCOS patients administered PRP compared with control PCOS patients is the regenerative effects of PRP on the endometrium. PRP is a platelet product that is free from leukocytes and erythrocytes. In addition to many growth factors in PRP, anti-inflammatory and pro-inflammatory molecules create a suitable microenvironment for implantation by regulating decidualization and redox reactions (13,21). In animal studies, treatment of mating-induced endometrial inflammation with PRP infusion supports the immunomodulatory and anti-inflammatory role of PRP (22). The decrease in endometrial development and blood flow due to CC may have been caused by the intense cytokine and growth factor content of PRP. Increased expression of receptivity genes in PRP-treated animals suggests that a similar effect may occur in patients with PCOS treated with PRP (19,23). The presence of intense interleukin, tumour necrosis factor-a, and interferon-a in PRP may regulate clomiphene-induced endometrial developmental defects and decreased blood flow. The fact that an increase in both EMT and endometrial vascularization was reported in Power Doppler analysis studies performed after PRP is an important evidence of the net effect of PRP on vascular demodulation (24).

In addition to studies reporting improvement in EMT and fertility outcome after IU-autologous PRP infusion (10), there are also studies reporting that PRP has no effect on EMT and implantation (21,25). Although the regenerative effect of PRP has been demonstrated in a murine model of endometrial damage, no human studies have investigated the effect of PRP on receptivity modulators (26). Our study is clinically important in terms of presenting the first clinical data reporting that clomiphene-induced thin endometrial growth is improved with PRP. However, the low number of cases and the application of PRP once is an important handicap. Multiple infusions of autologous PRP may reveal the relationship between EMT and clinical pregnancy rates more clearly. The lack of standardization of the PRP preparation methods may be the main reason for the difference in the results. Applying the buffy coat obtained using only two-stage centrifugation to the endometrial cavity is not a real PRP. Because leukocytes are not removed from the environment in such infusions, we cannot determine whether the effect on the endometrium is due to leukocytes or cytokines. Similarly, the administration of PRP without activation limits its effectiveness. Activated PRP infusion should also be performed quickly and without release of platelet contents. In our study, activated PRP was infused as soon as it was prepared. The fact that the participants were PCOS patients using CC and serial EMT measurements is critical in determining PRP efficacy.

CONCLUSION

In conclusion, autologous IU-PRP increases both EMT and clinical pregnancy rates in patients with PCOS with thin endometrium due to CC. However, the application of IU-PRP did not cause a significant change in the oligo/anovulatory cycle patterns of the patients. We can make a clearer judgment about the effectiveness of PRP in PCOS using case-controlled studies with multiple PRP applications.

Acknowledgment: We thank the participants, anonymous reviewers, and section editor for their excellent criticism of the article.

ETHICS

Ethics Committee Approval: The study was initiated after obtaining permission from the Clinical Research Ethics Committee of Zeynep Kamil Gynecology and Pediatrics Training and Research Hospital (decision no: 14, date: 11.01.2023).

Informed Consent: In addition, an informed volunteer consent form was obtained from all patients regarding the procedure to be performed.

Authorship Contributions

Surgical and Medical Practices: R.Ö., E.B., Concept: R.Ö., E.B., M.B.Y., Design: R.Ö., E.B., Data Collection or Processing: R.Ö., E.B., M.B.Y., Analysis or Interpretation: R.Ö., E.B., Literature Search: R.Ö., E.B., M.B.Y., Writing: R.Ö., E.B.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: The authors declare that this study received no financial support.