Progestin-primed milder stimulation with clomiphene citrate yields fewer oocytes and suboptimal pregnancy outcomes compared with the standard progestin-primed ovarian stimulation in infertile women with polycystic ovarian syndrome

A retrospective cohort trial was conducted at the department of assisted reproduction of the Ninth People’s Hospital of Shanghai Jiaotong University School of Medicine. This study was approval by the Ethics Committee (Institutional Review Board) of Shanghai Ninth People’s Hospital. The PCOS diagnosis criteria followed the Rotterdam consensus. PCOS was diagnosed by the presence of menstrual disturbance combined with either hyperandrogenism (hirsutism or hyperandrogenaemia) or polycystic ovary on ultrasonography (defined as an ovary that contained ≥12 antral follicles) and excluded other causes of hyperandrogenism (congenital adrenal hyperplasia, Cushing’s syndrome, androgen-producing tumours) and ovulation dysfunction (hyperprolactinaemia and thyroid dysfunction). In addition, this study only included women no more than 40 years of age and with baseline serum FSH no more than 10 mIU/ml. Women with functional cysts on the ovaries or medical conditions that contraindicated assisted reproductive technology and/or pregnancy were excluded. A total of 220 infertile women with PCOS from April 2014 to November 2015 were included and classified into the study group (HMG + MPA + CC) and the control group (HMG + MPA). A flowchart of the study is shown in Fig. 1.

In the study group, a low dose of HMG (150 IU daily), CC 50 mg and MPA 10 mg daily were started from cycle day 3 or after an episode of withdrawal bleeding. MPA was used to prevent premature ovulation during the ovarian stimulation. Follicle monitoring by transvaginal ultrasound and serum hormone measurements (FSH, LH, E₂ and progesterone) were performed 5 days later. HMG doses were then adjusted according to the ovarian response (range 150–300 IU daily). Oocyte maturation was triggered by triptorelin 0.1 mg (Decapeptyl, Ferring Pharmaceuticals, Germany) and urinary human chorionic gonadotropin (hCG 1000 IU, Lizhu Pharmaceutical Trading Co., China) when at least three follicles reached diameters of 18 mm or more. Cumulus oocyte complexes were collected 36 h later. All follicles larger than 10 mm in diameter were aspirated.

In the control group, HMG 225 IU and MPA 10 mg daily were initiated from cycle day 3 or after an episode of withdrawal bleeding. Follicle monitor and hormone assay were performed 5 days later. HMG dose was then adjusted according to the ovarian response, and MPA dose was consistent up to the trigger day. The criteria of mature follicle and the trigger methods were the same as above.

Fertilization was carried out in vitro after oocyte retrieval depending on the semen parameters and previous fertilization situation. Embryos were examined for the number or regularity of blastomeres and the degree of fragmentation. All top-quality cleavage-stage embryos (grade 1 and grade 2, 6-cell embryos and above) were frozen within three days after oocyte retrieval. The non-top-quality embryos were placed in further extended culture, and good-morphology blastocysts were frozen. Cleavage-stage embryos and blastocysts were frozen by vitrification as described previously [17].

Endometrium preparation for FET was arranged on the second cycle after oocyte retrieval. The first choice was using a letrozole-induced- ovulation cycle. Letrozole 5 mg was administered for 5 days, and then, follicle growth was monitored beginning on day 10. At times, a low dose of HMG (75 IU/day) was used to stimulate follicle growth and endometrial lining. The timing of FET was performed 4 or 5 days later, after a spontaneous or hCG-induced LH surge. Hormone replacement treatment was recommended for patients with thin endometrium and patients in whom letrozole failed. Oral ethinyl oestradiol 75 mcg/day was administered from cycle day 3 onwards. Once the endometrial lining was > 8 mm thick, femoston (Solvay Pharmaceuticals B.V.) 8 mg/day was started. The time of thawing and transfer was determined on the third day after femoston administration [17]. Each patient received no more than two embryos at one time. Once pregnancy was achieved, the luteal support was continued until 10 weeks of gestation.

Serum FSH, LH, E₂, and progesterone were measured on menstrual cycle day 3, day 8–11 (after 5–7 days of stimulation), the trigger day and the day after trigger. Hormone levels were determined with chemiluminescence (Abbott Biological B.V. Netherlands). The lower limits of sensitivity were as follow: FSH 0.06 mIU/ml, LH 0.09 mIU/ml, E₂ 10 pg/ml and progesterone 0.1 ng/ml. The upper limit of E₂ measurement was 5000 pg/ml. If serum E₂ on the trigger day or the day after was higher than the upper limit, it was recorded as 5000 pg/ml.

The primary measure analysed was the cumulative ongoing pregnancy rate, which was defined as the proportion of patients with ongoing pregnancy after the gestation age of 12 weeks. The secondary measures included the stimulation duration, gonadotropin consumption, incidence of premature LH surge and OHSS, the number of oocytes retrieved, the number of viable embryos, the proportion of mature oocytes and the live birth rate. The implantation rate was calculated as the number of gestational sacs visualized on transvaginal ultrasound divided by the number of transferred embryos. Clinical pregnancy was defined as the presence of foetal cardiac activity confirmed by transvaginal ultrasound. The cumulative live birth rate was defined as the total number of live births divided by all participants.

The data were evaluated by Student’s t-test for continuous variables of normal distribution, the Mann-Whitney U-test for continuous variables of non-normal distribution, the x²-test or Fisher’s exact for categorical variables, as appropriate. All tests were two-sided, and P < .05 was considered statistically significant. All data were analysed using the Statistical Package for the Social Sciences for Windows (SPSS, version 19).

The basal demographical and hormonal characteristics are shown in Table 1. A total of 220 patients completed this trial. There were no significant between-group differences in age, body mass index (BMI), previous IVF failures, infertility duration, menstrual cycle, indication for IVF or basal hormonal profile. All women completed one oocyte retrieval cycle, 208 patients had 1–15 viable embryos harvested, and 12 cases were cancelled before transfer due to either non-fertilization or no transferrable embryos. A total of 205 women completed 287 FET cycles in the following two years.

Clinical and cycle characteristics of ovarian stimulation in both groups are shown in Table 2. The study group (HMG + MPA + CC protocol) had a similar stimulation duration (9.2 ± 1.3 days vs 9.1 ± 1.2 days, P > 0.05) and consumed less HMG (1470.0 ± 360.1 IU vs 1943.8 ± 372.0 IU, P < 0.05). The numbers of oocytes retrieved, MII oocytes, and fertilized oocytes in the study group were significantly lower than those in the control group (P < 0.05). Consequently, the number of viable embryos in the study group was significantly lower than in the control group (4.8 ± 3.5 vs. 6.2 ± 3.7, P < 0.05). No between-group differences were found in the oocyte retrieval rate, but the maturation rate and the proportion of viable embryos per oocyte retrieved were better in the study group (respectively, 87.4% vs 80.3 and 39.5% vs 34.0%, both P < 0.05). The cycle cancellation due to zero viable embryos was significantly higher in the study group (9.1% vs 1.8%, P < 0.05). One patient experienced moderate or severe OHSS in each group (P > 0.05).

The serum concentrations of FSH, LH, E₂ and P in the two groups are presented in Fig. 2. FSH in the study group was slightly lower than in the control group during the mid-follicular phase (P < 0.05). LH gradually decreased during ovarian stimulation in the control group; in contrast, LH in the study group showed a slight rise initially, followed by a downward trend, and the mean LH value on the trigger day was significantly higher than in the control group (4.49 ± 2.49 mIU/ml vs 2.52 ± 2.09 mIU/ml, P < 0.05). No endogenous LH surge occurred in either group (P > 0.05). The LH value on the post-trigger day showed a dramatic increase in the two groups, with no between-group difference (P > 0.05).

E₂ increased gradually, accompanied by with multiple growing follicles during the ovarian stimulation, and no difference was found between the two groups (P > 0.05). The measured E₂ values were underestimated in 114 cases due to the upper limit of 5000 pg/ml, so the comparison of E₂ between the two groups was compromised. Serum P showed a gradual increase during ovarian stimulation and increased significantly after trigger in both groups.

The pregnancy outcomes from FET are shown in Table 3. A total of 287 FET cycles were completed in the two groups, including 66 women who finished at least two transfers. The control group yielded more embryos, which were able to finish more FET cycles in the following two years. The mean transfer cycles per patient were 1.5 in the control group and 1.1 in the study group in the following two years. A total of 560 embryos were thawed and the survival rate was 99.3% (556/560). The remnant embryos were, respectively, 350 and 303 in the control and study group, which were the suplus embryos in pregnant cases except for three cases in study group without their transfer.

The synchronization methods of endometrium and embryo were similar between the two groups. The ongoing pregnancy rate per transfer and the implantation rate were significantly higher in the study group (respectively, 66.4% vs 53.6%; 52.2% vs 42.7%, P < 0.05) but the live birth rate per transfer was comparable between two groups (59.7% vs 49.4%, P > 0.05). Sixty-four women experienced twin pregnancies, including 6 women with vanishing syndrome. One triplet pregnancy occurred in each group, and both resulted in live births after operation of multifetal reduction. The proportions of multiple pregnancies, miscarriage and ectopic pregnancy were similar between groups (P > 0.05). The cumulative ongoing pregnancy and live birth rate per patient were lower in the study group but did not reach the significant difference (respectively 71.8% vs 81.8%; 64.5% vs. 75.5%; P > 0.05).

All newborns were examined with no congenital malformation except that oesophageal atresia was found in one baby of the control group and ventricular septal defect in one of the twin babies of the study group.