Polycystic Ovary Syndrome (PCOS) is the most common endocrine pathology in women of childbearing age. Its overall prevalence in the world is between 6% and 21% [1] [2] . It is characterized to varying degrees by the association of clinical and/or biological hyperandrogenism (HA), ovulation disorders (OA) and the presence of polycystic ovaries on ultrasound (PCO). We thus distinguish four phenotypes. PCOS is the major cause of anovulatory infertility [3] [4] [5] [6] ; In Vitro Fertilization (IVF) constitutes the final stage of care for this group of patients and according to several authors, the outcome in IVF/ICSI (In Vitro Fertilization/Intracytoplasmic Sperm Injection) would depend on the PCOS phenotype [3] [6] [7] [8] . Thus, given the recurrence of cases of PCOS admitted for ovarian stimulation for IVF/ICSI at the Hospital Center for Research and Application in Endoscopic Surgery and Human Reproduction (CHRACERH) and the fact that no study on this subject has been carried out in our context which we proposed to carry out this study.

This was a cohort study with historical-prospective data collection which took place at CHRACERH-Yaoundé-Cameroon. It took place over a period ranging from 1^st of January 2016 to the 30^th of May 2023, i.e., a period of 7 years and 4 months. After obtaining approval from the ethics committee of Faculty of Medicine and Biomedical Science of the University of Yaounde I (Ref N 00273/12^th, May 2023), and authorizations from the competent authorities, we recruited patients admitted to IVF/ICSI with ovarian stimulation at CHRACERH during the period. The data collected was reported in the previously established technical sheet.

The “exposed” group consisted of PCOS patients meeting the Rotterdam criteria who had a usable file or who had agreed to participate in the study. The “control” group constituted the files of patients who did not have signs of PCOS. The PCOS group was THEN subdivided into four subgroups corresponding to the following PCOS phenotypes: Phenotype A (HA, hyperandrogenism + ovulation disorders, OA+ polycystic ovaries, PCO on ultrasound), Phenotype B (HA + OA), Phenotype C (HA+PCO) and Phenotype D (OA + PCO) (See Figure 1).

The variables of interest were socio-demographic and clinical, assessment of ovarian reserve, and IVF outcome.

Stimulation protocol: patients were placed either on an antagonist protocol or on an agonist protocol depending on the availability of drugs. The dose of gonadotropin required depended on the ovarian reserve and varied from 150 to

300 IU. The doses must be revised downwards depending on the evolution. Thus, the monitoring of IVF cycles consisted of regular monitoring by endovaginal ultrasound carried out by a gynecologist or a radiologist with an ACUSON X150 type ultrasound device (Figure 2).

Aimed at counting antral follicles and measuring the thickness of the endometrium and by measuring the level of estradiol, progesterone from the 5^th-6^th day of the cycle depending on the protocol. Dosage of serum estradiol (E2) and progesterone (P4) carried out by a biologist on a Cobass e 411 HITACHI type device (Figure 3).

Then every 2 - 3 days depending on progress, until at least 2 follicles of 17 mm are obtained. Triggering was based on an intramuscular injection of 10,000 IU of Human Chorionic Gonadotropin (HCG) for the agonist protocols and triptorelin (decpeptyl) 0.1 mg or HCG in the case of the antagonist protocol. The oocyte

retrieval was carried out 36 hours later. The luteal phase was supported by daily intake of micronized progesterone 600 mg/day vaginally from the day of oocyte retrieval until the pregnancy test (12 days later). The embryo transfer was most often carried out on D2 or D3 after the puncture under ultrasound control. It could be carried out in the same cycle or postponed to a later cycle due either to the high progesterone level on the day of initiation or risk of hyperstimulation. Pregnancy was initially diagnosed by a positive plasma HCG level on day 12 after embryo transfer. The data were entered and coded in CS Pro version 6.2 software; then imported and analyzed in IBM SPSS version 23.0 software for statistical analysis. Comparisons between PCOS phenotypes were made by ANOVA tests (LSD post-hoc Tukey test) for continuous variables and the Chi Square test (or the Fisher’s Exact test) for categorical variables. Results were expressed as mean ± standard deviation for continuous variables and as frequencies for categorical variables. P values less than 0.05 were considered statistically significant.

Age was comparable in the different phenotypes. These results were close to

those of the study by Wang et al. carried out in China in 2022 which found an average age of between 28.87 ± 3.02 and 29.49 ± 3.46 (p = 0.095) [3] . PCOS phenotypes with hyperandrogenism (B > A > C) had a higher BMI than the D phenotype without hyperandrogenism. These results were consistent with those found in Belgium by Mackens et al. with a significantly higher BMI in phenotypes with hyperandrogenism (p < 0.001) [8] ; this can be justified by the pathophysiology of PCOS involving hyperandrogenism in the occurrence of obesity (metabolic syndrome) by mechanism of insulin resistance (and reactive hyperinsulinemia with adipogenesis) and obesity itself increasing hyperandrogenism. Concerning the different phenotypes, the phenotypes presenting the ultrasound criterion for PCOS (A, C and D) had a significantly higher ovarian reserve than phenotype B without ultrasound criterion. These data are in strong agreement with those found in Iran by Ramezanali et al. who found an AMH varying between 5.8 ± 2.4 and 6.8 ± 2.8 ng/ml and a CFA (from 33.2 ± 3.9 to 35.4 ± 4.8) [7] , in Italy by Cela et al. (AMH: 7.78 ± 2.63 to 10.57 ± 0.79 ng/ml and a CFA: 25.89 ± 8.19 to 36.43 ± 8.18) [5] and in China by Wang et al. (AMH: 8.86 ± 4.37 to 12.35 ± 6.03 ng/ml and CFA ranging from 27.74 ± 8.62 to 33.73 ± 11.64) [3] ; The duration of stimulation was comparable between the different groups. These results confirmed those of Wang et al. with stimulation durations of between 10.04 ± 2.40 and 10.55 ± 2.52 days [3] . Studies by Cela in Italy and Ramezanali in Iran also found similar results but with shorter average durations varying between 9.34 ± 1.87 and 11.75 ± 2.06 days and between 10.1 ± 2.3 and 10.4 ± 2.1 days respectively [6] [7] .

The total dose of gonadotropins used during ovarian stimulation was lower in phenotype D. These results agree with those of Cela et al. who found a lower total dose of gonadotropins used (1377.92 ± 618.0 IU) in phenotype D. Hyperandrogenism may alter folliculogenesis by disrupting the meiotic cell cycle of the oocyte, leading to premature cessation of oocyte development. This mechanism compromises oocyte maturation and the complete acquisition of its skills. The exact mechanisms of this maturation defect in PCOS remain poorly understood compared to other phenotypes [6] ; but do not corroborate those of Wang et al. in whom the total doses were comparable in the different phenotypes [3] .

The estradiol level on the day of ovulation triggering was higher in phenotypes with ultrasound criterion (A, C and D). The studies by Cela and Wang also went in the same direction, but found lower doses respectively ranging from 1479.77 ± 828.05 to 2034.25 ± 1732.5 ng/ml and from 4168.88 ± 2488.18 to 4882.83 ± 2918.41 ng/ml. This discrepancy is explained by the use of the short agonist protocol “microflare” in a large proportion of PCOS patients due to the unavailability of antagonists at the given times.

The numbers of follicles collected after puncture and mature oocytes obtained after decoronization appeared to be lower in the phenotypes with hyperandrogenism A, B, C without significant difference. These results follow the same trend as those of Cela et al. with lower numbers of follicles ranging from 7.1 ± 2.56 to 8.54 ± 3.59 follicles and 3.58 ± 1.50 to 5.11 ± 1.45 M2 in hyperandrogenic phenotypes (A, B, C) versus 9.36 ± 4.36 follicles and 6.84 ± 3.15 M2 in phenotype D without hyperandrogenism [6] . However, Wang et al. reported comparable results between the different phenotypes.

The results did not show any differences in terms of immature oocytes in the different phenotypes. But the number of immature oocytes were significantly higher in the group of PCOS patients (2.22 ± 3.60 VG and 1.70 ± 3.10 M1) compared to the control group (0.64 ± 1.32 VG and 0.80 ± 1.03 M1) (p < 0.001). These results can be justified by data from the literature which states that hyperandrogenism alters folliculogenesis and compromises oocyte maturation [9] - [15] .

Ovarian hyperstimulation syndrome occurred mainly in phenotype D without hyperandrogenism. These data are similar to those found in the literature reporting that the occurrence of OHSS varies between 3% - 8% [8] ; but a study conducted in France by Isnard et al. on ovarian stimulation with gonadotropins in patients with female polycystic ovarian syndrome reported two cases of OHSS with a lower prevalence of 1.1% which could be explained by the exclusive use of the antagonist protocol.

The fertilization rate was lower in phenotypes A, B, C with hyperandrogenism without significant difference (p = 0.461). This result could be explained by the fact that hyperandrogenism alters the quality of oocytes [11] [12] [13] [14] [15] . But once the embryos were obtained, the average number of top quality embryos was comparable between the different phenotypes (p = 0.207). The same results were found by Wang et al., Ramezanali et al. and by Cela et al. Biological pregnancy rates were comparable in the different groups. These results were similar to those of Ramezanali et al. which found biological pregnancy rates of 32.5%, 26.4%, 36.8%, 53.3% and 45.1% in phenotypes A, B, C, D.

Phenotypes D and A were respectively the most represented. Age was comparable between the different groups. BMI was higher in phenotypes with hyperandrogenism. Ovarian reserve was significantly higher in phenotypes with ultrasound criterion (A, C, D). Ovarian hyperstimulation syndrome occurred mainly in phenotype D. Phenotype D seemed to have a higher pregnancy rate. This raises the question of systematic treatment of hyperandrogenism before or during ovarian simulation in these PCOS phenotypes.

The authors declare no conflicts of interest regarding the publication of this paper.

Vanina, N.A., Minala, N., Etienne, B., Voundi, V., Pascale, M., Serges, N., Kasia, O.Y., Joselyne, C., Florence, K., Abega, A. and Marie, K.J. (2024) Association between the Different Phenotypes of Polycystic Ovary Syndrome and the Outcome in in Vitro Fertilization at Human Reproductive Center Paul et Chantal Biya-Yaoundé. Open Journal of Obstetrics and Gynecology, 14, 18-28. https://doi.org/10.4236/ojog.2024.141003