Association of Follicular Distribution Patterns in Polycystic Ovaries with Clinical Outcomes in Assisted Reproductive Technology Cycles: A Prospective Cohort Study.

Polycystic ovary syndrome (PCOS) is an endocrine and metabolic disorder that affects most women of reproductive age, and the alarming prevalence rate among teenagers (97.83 per 100,000) is also concerning ( 1 ). This syndrome is frequently associated with conditions such as diabetes, cardiovascular diseases, and metabolic disorders ( 2 ). Clinical manifestations of this disease are characterized by menstrual disorders (oligo/amenorrhea), hyperandrogenism, infertility, and hormones-related disorders. Also, ultrasound imaging, a non-time-consuming and inexpensive method, is effective in diagnosing this syndrome ( 3 ). Its diagnostic criteria on the basis of the Rotterdam criteria include the presence of ≥12 follicles with a size of 2 to 9 mm in at least one ovary volume is more than 10 cm3, in at least one ovary ( 4 ). Recently, an important role of ovarian morphology in the diagnostic criteria of the syndrome has been emphasized more ( 5 ). In women with the ovaries which have PCO morphology, various folliculogenesis physiological processes are disturbed that may lead to different patterns of follicular distribution ( 3 , 6 ). There is still much debate about how best to determine the ovaries' appearance in PCOS ( 7 ). While the number of follicles is used to define this syndrome, the follicular distribution pattern (FDP) in the ovarian stroma is less discussed ( 3 ). The two classifications based on the distribution of antral follicles are defined, the classic pattern of a string of pearls is called as peripheral cystic pattern (PCP) and the second category is known as a general cystic pattern (GCP) in which the follicles are existed in the entire parenchyma of the ovary ( 3 ). First of all, Takahashi et al. ( 8 ), reported that serum androstenedione levels and the luteinizing hormone (LH)/ follicle stimulating hormone (FSH) ratio in women with PCP were remarkably higher than those with GCP ovaries. This finding indicated an endocrine disorder in patients with ovaries or a peripheral scattering pattern ( 3 ). Furthermore, a positive connection between the number of follicles, ovarian volume, stromal characteristics, and insulin resistance markers has been reported ( 6 ). Recent ly, Mills et al. ( 3 ) concluded that menstrual irregularities and specific serum steroid levels are directly associated with ovarian morphology. It remains uncertain whether ovarian morphology, testosterone levels, or LH levels directly cause menstrual disorders or if these factors are coregulated by an underlying mechanism that contributes to both hormonal imbalances and menstrual irregularities. A limited number of studies ( 3 , 6 , 8 , 9 ) have investigated the different pattern of follicular distribution in the ovaries of PCOS patients and its relationship with the prevalence of infertility and the level of sex hormones and metabolic disorders, therefore this subject has attracted a great deal of attention. Given the high prevalence of infertility and the use of assisted reproductive technology (ART) in women with PCOS, examining the different follicular distribution pat terns in the ovaries of patients undergoing ART cycles could be valuable. It may help predict the optimal ovarian response to treatment and enable personalized counseling for these patients. Therefore, this study was designed to evaluate the relationship of the FDP in the ovaries and the ovarian response in different treatment plans, including in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI), as well as the level of sex hormones and clinical manifestations in patients with PCOS diagnosis.

Of the 157 patients affected by PCOS who evaluated in our hospital during the study period, 73 patients were classified in the PCP ovaries group and 84 patients in the GCP ovaries group. The baseline characteristics of patients were compared between two groups ( Table 1 ). There was no significant difference in age and body mass index between these two groups. A significant relationship was observed between the FDP and the baseline serum levels of LH, AMH, total testosterone, and 17-OHP. The mean levels of LH, AMH, total testosterone, and 17-OHP were significantly higher in the PCP group compared to the GCP group (P=0.022, P=0.023, PP<0.001, and P=0.028, respectively). No statistically significant difference was found between the two groups regarding the serum level of DHEAS. Baseline characteristics of PCOS patients stratified by the ovarian morphology pattern Values are presented as the mean ± standard deviation (SD) and number (%). *; Obtained by independent t test and chi square test. Significant level was considered at P<0.05. LH; Luteinizing hormone, PCOS; Polycystic ovary syndrome, PCP; Peripheral cystic pattern, GCP; General cystic pattern, AMH; Müllerian hormone, DHEA-S; Dehydroepiandroster one sulphate, and 17-OHP; 17-Hydroxyprogesterone. No statistically remarkable difference was observed be tween the two groups in terms of clinical manifestations such as oligomenorrhea diagnosis as well as the presence of acne and alopecia. However, a significant association was found between the presence of hirsutism and follicular distribution pattern. So that the frequency of cases with hirsutism in the PCP group was significantly higher than the GCP group (P=0.011, Table 1 ). The ovarian stimulation outcomes were compared be tween groups in Table 2. Statistical analysis showed that the number of retrieved oocytes and the percentage of OHSS at risk cases in the PCP group were significant ly higher than those of in the GCP group (P=0.036 and P=0.013, respectively). However, no statistically significant difference was observed regarding the oocyte quality (total number of MII oocytes) between groups. In addition, the number of germinal vesicles and MI oocytes was similar between groups. Ovarian stimulation cycle outcomes stratified by the ovarian morphology pattern *; Values are presented as the mean ± standard deviation (SD) and number (%). Obtained by independent t test and chi square test. Significant level was considered at P<0.05. MII; Metaphase II, GV; Germinal vesicle, OHSS; Ovarian hyperstimulation syndrome, PCP; Peripheral cystic pattern, and GCP; General cystic pattern.

The main objective of the present study was to as sess whether the FDP in PCOM ovaries of infertile women diagnosed with PCOS can be useful in predicting the total number of retrieved, MII oocytes, as well as the risk of OHSS. From a clinical point of view, it is important to understand how different PCOM morphologies are related to the severity of the disease itself. The results of the present study in dicated that some signs of disease severity such as high serum levels of LH, AMH, total testosterone and 17-OHP were significantly related to PCP follicular distribution pattern. In the follow-up of patients after COS, it was found that the PCP follicular pattern was associated with an increase in the number of retrieved oocytes and the risk of OHSS; however, the quality of the obtained oocytes was not related to the type of follicular distribution pattern. In similar way, Mills et al. ( 3 ) in a retrospective study reported that the PCP follicular distribution was more strongly associated with menstrual irregulari ties and high levels of serum LH and total and free testosterone. These findings approved the results of the previous studies demonstrating a relationship be tween FDP and hyperandrogenism, supporting the hypothesis that PCP and GCP ovarian morphologies may exhibit differences in their endocrine and pathophysiological mechanisms. The occurrence of PCP ovarian morphologies is thought to result from a stromal core characterized by heightened density and increased vascular blood flow, which may influence the peripheral development of ovarian follicles. Prior research has indicated that stromal density and vascular flow are predictive of the severity of PCOS, as they correlate with levels of ovarian hyperandrogenism ( 3 , 8 , 14 ). Conversely, GCP ovaries do not demonstrate elevated the stromal density and are likely to induce a lesser degree of androgenic disturbance in affected individuals ( 3 ). The current study demonstrated no significant relation ship between menstrual irregularity and either unilateral or bilateral PCP or GCP. This finding contrasts with Mills and colleagues’ research, which indicated that women with unilateral PCP follicular development patterns were more prone to experience menstrual irregularities compared to those with bilateral GCP ( 3 ). The discrepancy may be due to the use of oral contraceptive pills prior to the onset of the menstrual cycle. Furthermore, a more precise classification of each ovary regarding follicular development patterns (PCP or GCP) could have facilitated a better understanding of the relationship between cycle irregularities and ovarian morphology. In contrast to the present findings, Christ et al. ( 6 ) conducted a cross-sectional observational study involving 49 women with PCOS. They concluded that there was no significant relationship between the FDP and any reproduction or metabolic markers associated with PCOS. Several factors may explain this discrepancy. Firstly, the study population in Christ et al.'s ( 6 ) study differs from that in the current study. In our study, the participants were infertile PCOS women undergoing in vitro maturation (IVM) or IVF/ICSI cycles, whereas Christ et al. ( 6 ) selected patients based on the National Institutes of Health (NIH) diagnostic criteria. Additionally, evidence of hyperandrogenism was used as an inclusion criterion in their study. Secondly, the sample size in Christ et al.'s ( 6 ) study was considerably smaller than in the present study. In a study by Alviggi et al. ( 9 ), the ovarian ultrasono graphic characteristics of women with PCOS and insulin resistance were compared to those with a hyperandro genic profile. The study identified two distinct ovarian patterns: group A, characterized by follicles primarily measuring between 5 and 9 mm with a prominent "neck lace" sign, and group B, where the follicles were smaller (2-4 mm) and more evenly distributed, with no "neck lace" sign. The researchers suggested that insulin resist ance might be linked to specific ovarian morphology patterns, especially the pattern seen in group A, where insulin resistance was associated with lower BMI, waist to-hip ratio, and HOMA compared to those in group B ( 9 ). However, our findings differ from Alviggi et al.'s ( 9 ) study. In our study, PCOS patients with the classic "necklace" sign in their ovaries, resembling the morphology of group A, were associated with more severe PCOS symptoms, including menstrual disorders and abnormal hormone levels. As Alviggi et al. ( 9 ) proposed, insulin resistance may alter ovarian morphology, but we did not investigate insulin resistance markers in this future research. This would provide further insight into how insulin resistance and ovarian morphology may in teract in PCOS, especially considering the differences observed between group A and group B of Alviggi et al. ( 9 ) study. The strengths of the present study include its prospective design, with all sonographic evaluations and laboratory tests conducted by a dedicated team of specialists. However, a limitation of our study was the absence of a control group of PCOS patients with normal ovarian morphology. Additionally, we did not investigate insulin resistance markers or lipid metabolism, which could have provided valuable insights into the metabolic as pects of PCOS and helped explain the findings related to ovarian stimulation outcomes. The varying sizes of the observed follicles were consistent with the number of recruitable follicles, reflecting the degree of metabolic dysfunction in PCOS and the potential risk of OHSS. Future research should consider categorizing antral fol licle counts into distinct groups to further refine our understanding of these relationships.

The current study found that PCP follicular distribu tion was associated with clinical manifestations, including hyperandrogenism, as well as elevated levels of LH and AMH in women with PCOS. Additionally, in COS cycles, both the number of retrieved oocytes and the risk of OHSS were significantly higher in patients with this specific ovarian morphology. These findings suggest that conducting more detailed studies to identify distinguishing factors among individuals with PCOS could aid in the development of personalized treatment strategies and provide more accurate predictions of treatment response.

This prospective cohort study was carried out at the infertility center of Arash Women’s Hospital (a university affiliated infertility center in Tehran, Iran) from March 2022 to September 2023. The scientific board and the ethics committees of the Tehran University of Medical Sciences was approved the study protocol (IR.TUMS.MEDI CINE.REC.1402.209). Eligible patients provided written informed consent to participate in the study. All women with PCOS diagnosis who were referred for IVF and/or ICSI treatment cycles due to several ovulation induction failures and/or intrauterine insemination (IUI) cycles were screened. The exclusion criteria were as follows: history of clinical or biochemical signs of thy roid disorders, hyperprolactinemia, Müllerian anomalies, history of hypothalamic pituitary dysfunction or ovarian failure, ovarian and adrenal androgen-secreting tumors and congenital adrenal hyperplasia, presence of function al cyst >3 cm in initial ultrasound assessment of pelvic, endometriosis diagnosis, history of unilateral oophorectomy, abnormal karyotype of women or her husband, recurrent implantation failures and repeated pregnancy loss, severe male infertility factor (azoospermia). The PCOS diagnosis was defined according to the Rotterdam criteria ( 4 ). Definitions of menstrual disorders and hyperandrogenism, along with the Ferriman-Gallwey scoring system for hirsutism, are outlined in previous publications ( 10 , 11 ). The presence of 12 or more ovarian cysts with 2-9 mm diameter per ovary and/or ovarian volume ≥10 cm 3 was defined as polycystic ovary morphology (PCOM) ( 12 ). Menstrual regularity was detected by asking the patient regarding the duration of most menstrual cycles at the first visit in the fertility clinic. Oligomenorrhea was defined as infrequent menstrual cycles occurring more than 35 days apart or <8 cycles per year ( 12 ). Vaginal ultrasound was done on day 2-5 of a spontaneous or progestin induced menstrual cycle. Follicle count and distribution was evaluated by two expert gynecologists using a Philips Affiniti 70w with a trans-vaginal 3.5-10 MHz probe (USA) and images were classified into two groups based on ovarian morphology ( 3 ), including: i. PCP: 12 or more follicles distributed around a dense stromal core peripherally for at least 50% of the ovarian diameter, ii. GCP: 12 or more follicles located throughout the ovary and not more than 49% in a peripheral distribution. Each specialist was blinded to the diagnoses made by the others. If there was a discrepancy between the two specialists’ diagnoses, a third radiologist was consulted to make the final decision. The final diagnosis was based on the agreement between two out of the three special ists, ensuring consensus. No other guidelines or criteria, beyond the specialists’ diagnoses, were used to determine the final decision. In the current investigation, the study sample consisted of patients with identical morphology in both ovaries. This implies that both ovaries exhibited either PCP or GCP morphology. Baseline serum levels of follicle-stimulating hormone (FSH), LH, anti-Mullerian hormone (AMH), total testos terone, dehydroepiandrosterone sulfate (DHEAS), and 17-Hydroxyprogesterone (17-OHP) were determined from venous blood samples collected from each participant on the second or third day of their menstrual cycle. The concentrations of FSH, LH, total testosterone, DHE AS, and 17-OHP were quantified using Electrochemilumi nescence Immunoassay (ECLIA) with a fully automated Immulite 20000 analyzer (SIEMENS, Germany) and the following commercial kits: FSH Kit (103249, SIEMENS, Germany); LH Kit (103248, SIEMENS, Germany); To tal Testosterone Kit (103244, SIEMENS, Germany); DHEAS Kit (103253, SIEMENS, Germany) and 17-OHP Kit (103252, SIEMENS, Germany). The measurement of AMH concentration was performed using a commercial enzyme-linked immunosorbent assay (ELISA) kit (Di aZist, Iran), which demonstrated inter- and intra-assay coefficient of variations (CVs) of less than 5%. All the studied population of our participants received the same controlled ovarian stimulation (COS) ( 13 ). Ovar ian quiescence was defined by the observation of ovarian suppression. Serum estradiol (E2) levels were measured and found to be below 50 pg/mL. This was determined through baseline ultrasounds and hormonal assessments. These assessments were conducted on the second or third day of the menstrual cycle. Following this evaluation, a starting dose of 150 IU of recombinant human FSH (Cinnal-f, Cinagen, Iran) was administered daily for a duration of five days. The serial vaginal ultrasound assessments were carried out for follicles growing monitoring and, the dosage of gonadotropins was adjusted on the basis of the ovarian response in each patient ( 13 ). The administration of the gonadotropin-releasing hormone (GnRH) antagonist (Cetronax ®, Ronak Pharmaceutical Company, Iran) at a dose of 0.25 mg per day via subcutaneous injection was initiated upon the the observation of follicle(s) with an average diameter of 13 mm or greater. This treatment continued until the day of the final oocyte triggering. The final maturation of the oocytes was induced using a GnRH agonist (0.2 mg decapeptyl®, Ferring Pharmaceuticals, Australia) once at least two follicles measuring 18 mm or more in diameter were identified. The risk of ovarian hyperstimulation syndrome (OHSS) was defined as a serum E2 level> 5000 pg/ml and/or the presence of more than 20 follicles on ultrasound evaluation on the day of oocyte triggering. In high-risk patients, the strategy of freezing all embryos was considered. Transvaginal ultrasound-guided oocyte retrieval was performed 35 to 36 hours after the final oocyte triggering. The IVF/ICSI procedure was conducted using ejaculated sperm on mature or metaphase II (MII) oocytes, following standard protocols ( 13 ). The statistical analysis was performed using SPSS software (version 22; IBM Corp., Chicago, IL, USA). Data were presented as mean ± Standard deviation (SD) or number (percentage). The two-tailed Student's t test and Chi-square test were used to compare groups, as appropriate. A P<0.05 was considered to be statistically significant.