The comparison of two different protocols ultra-long versus medroxyprogesterone acetate in women with ovarian endometriosis: a prospective randomized controlled trial.

BACKGROUND: This study aimed to investigate the medroxyprogesterone acetate (MPA) + HMG protocol vs ultra-long gonadotrophin releasing hormone (GnRH) agonist protocol in patients with advanced ovarian endometriosis who received in vitro fertilization (IVF).
METHODS: Three hundred patients with advanced ovary endometriosis who underwent IVF were included, and embryological and clinical outcomes were assessed between March 2017 and September 2017. Patients were divided into MPA + HMG group and 1-month ultra-long GnRHa protocol group.
RESULTS: Lower hMG dose and shorter medication time were found in the MPA + HMG group than in the GnRHa group (P < 0.05). Follicle to-Oocyte Index was significantly different between MPA + HMG group and GnRHa group (P < 0.001). No differences were found in the ovary response and numbers of mature oocytes, fertilized oocytes and viable embryos. The clinical pregnancy and live birth outcomes were similar between MPA + HMG group and GnRHa group, and these outcomes were independent of fresh or frozen embryo transfer in the GnRHa protocol group. There were no significant differences in the time to embryo transfer, medical cost and adverse effects.
CONCLUSION: The number of oocytes retrieved and pregnancy outcomes after MPA + HMG protocol are similar to those after ultra-long GnRHa protocol in women with ovarian endometriosis. MPA + HMG protocol may be an alternative to ultra-long GnRHa protocol for IVF in ovary endometriosis patients. Trial registration The trial was registered in the Chinese Clinical Trial Registry (ChiCTR-INR-17010924) In conclusion, the administration of MPA in COH showed similar number of oocytes retrieved, no premature LH surge, and similar pregnancy and live birth outcomes in patients with advanced ovarian endometriosis undergoing IVF/ICSI as compared to the one-month long protocol. The use of MPA in COH appears to be promising although many questions remain to be elucidated, including the dose and time of progestin priming as well as its possible influence on the oocyte development potential and microenvironment. Given their good tolerability, few metabolic influence, and low cost, progestogens provide a novel alternative to the conventional protocol for patients with endometriosis.

Background

Endometriosis refers to the presence of endometrial-like tissues outside the uterine cavity. It is a chronic, estrogen-dependent inflammatory status, and affects approximately 10% of women of reproductive age and 20%–50% of infertile women [1, 2]. Women with endometriosis usually have a low pregnancy rate, and thus assisted reproductive techniques are often employed to improve the pregnancy rate. In vitro fertilization (IVF) is an important strategy for pregnancy, especially for infertile women non-responsive to surgical treatment. However, even mild endometriosis may adversely affect the fertility via influencing the oocyte development, embryogenesis and implantation [3, 4]. Nevertheless, the exact pathophysiological mechanism by which the endometriosis adversely affects the fertility remains unknown [5-7].

In recent years, some studies have been conducted to explore the optimal protocol for IVF in women suffering from endometriosis. There is evidence showing that gonadotropin-releasing hormone (GnRH) agonist used before IVF/ intracytoplasmic sperm injection (ICSI) is able to improve the pregnancy rate. The Cochrane guidelines in 2006 recommend the use of GnRH analogs for more than 2 months before IVF to “improve the inflammatory status”, thereby increasing the pregnancy rate [8]. Progestins have been used in endometriosis therapy for many years. It has been reported that progestin can improve the endometriosis-associated pelvic pain via suppressing CCL5/RANTES (Regulated upon Activation, Normal T cell Expressed and presumably Secreted) production and inhibiting inflammation in the pelvis [9]. Fechner et al. [10] found that progestins could regulate local E2 biosynthesis and inhibit the growth of ectopic endometrium in the endometriosis women. In addition, progestins can reduce E2 level and have good tolerability, few metabolic effects and low cost. Medroxyprogesterone acetate (MPA) has been used in patients undergoing controlled ovarian hyperstimulation (COH) for IVF, and may serve as an effective oral alternative for women with advanced endometriosis [11, 12]. However, the time of pituitary down-regulation is still controversial in the endometriosis patients. Moreover, down-regulation is based on the consequence of fresh embryo transplantation in endometriosis. Most recent studies have revealed that long-term pituitary down-regulation has limitations, especially in severe endometriosis patients receiving frozen-thawed embryo transfer (FET) [13]. Therefore, it is necessary to investigate measures that can ensure a good pregnancy rate while reducing the side effects during the treatment. Health economics (such as time to embryo transfer, medical cost and adverse effects of drugs used) have become a focus of COH for patients with endometriosis. Whether MPA treatment is effective to improve the oocyte quality, embryo quality and pregnancy outcome has not been studied. In the present study, women with ovarian endometriosis were included. To shed more light on this debated and intriguing issue, the included patients had normal ovarian reserve function. Patients receiving 1-month down-regulation served as controls. This study aimed to explore the efficiency and safety of MPA with hMG in advanced endometriosis patients with normal ovarian reserve function during IVF as compared to one-month ultra-long protocol.

Materials and method

Study design

This was a prospective non-inferiority randomized controlled study that was carried out at the Department of Assisted Reproduction of the Ninth People's Hospital, School of Medicine, Shanghai Jiaotong University between March 2017 and September 2017. The study was approved by the Institutional Review Board of the Ninth People's Hospital of Shanghai. The trial was registered in the Chinese Clinical Trial Registry (ChiCTR-INR-17010924). All procedures were performed in accordance with the ethical standards of the responsible committee on human trials and with the Helsinki Declaration of 1964 and its later amendments. Informed consent was obtained from all patients for being included in the study.

Patients

Patients undergoing the first IVF/ICSI cycle were recruited into present study. The inclusion criteria were as follows: (1) Laparoscopy or laparotomy before IVF showed severe ovarian endometriomas and recurrence of ovarian cyst after the operation, and patents were diagnosed with stage III-IV endometriosis according to the revised American Fertility Society (AFS) classification [14]; (2) the ovarian endometriomas was identified as “chocolate” cysts (> 3 cm) and multiple cysts were confirmed by ultrasonography repeatedly; (3) patients were aged ≤ 40 years; (4) the menstrual cycle was regular (25–35 day per cycle) in the prior 3 months; (5) the antral follicle count (AFC) was more than 4 and less than 20 on menstrual cycle day 2–3; and (6) the basal serum follicle-stimulating hormone (FSH) concentration was ≤ 10 IU/L.

The exclusion criteria were as follows: (1) there was documented ovarian failure, including basal FSH > 10 IU/L or no antral follicles on ultrasound examination; (2) patients were diagnosed with polycystic ovarian syndrome; (3) there was hydrosalpinx; (4) there was adenomyosis on the laparoscopy or laparotomy, ultrasonography displayed disordered myometrial echo, or magnetic resonance imaging showed mild adenomyosis; (5) patients received hormone treatment within the prior 3 months; (6) patients had mild endometriosis or peritoneal endometriosis; (7) there was moderate to severe intrauterine adhesion.

Sample size

This was a prospective non-inferiority trial. The sample size was estimated according to previously reported [15]. The primary end-point was the number of oocytes retrieved. The assumed mean number of oocytes retrieved in the ultra-long GnRHa group was 10.0, and the number of oocytes for non-infertility margin was 2.0. The assumed number of normal ovulation obtained from women with endometriosis would be similar to that in patients with normal ovulation. Thus, the required sample size would be 130 for each group to achieve an α = 0.05 and the power of 0.8 (PS power and sample size calculations, version 2.1.30). Given the possibility of 10% dropout, 150 women were included in each group. Secondary end-points were the number of metaphase II (MII) oocytes, pregnancy rate and live birth rate (Figs. 1, 2).

Fig. 1

Stimulation protocols in ovarian endometriosis during IVF

Fig. 2

Flowchart for patient recruitment

Randomization

Patients were recruited in a 1:1 ratio to receive treatment with either MPA + HMG protocol or ultra-long GnRH agonist protocol via the use of a random number table based on a computer-generated drawing of numbers. Relevant embryologists were blind to the grouping in the trial. The physicians and participants were not blind to the grouping. The randomization was prepared by an independent statistician, and performed after confirmation of the inclusion/exclusion criteria and signing the informed consent. All of the participants provided informed consents after they consulted for infertility treatments and routine IVF procedures.

Ovarian stimulation and embryo culture

Women undergoing IVF followed MPA + hMG protocols according to previously reported [11, 12]. The final stage of oocyte maturation was triggered with decapeptyl (0.1 mg) (Ferring International Center SA, Germany) and hCG (2000 IU; Lizhu Pharmaceutical Trading Co, China) [11]. Viable embryos were cryopreserved for later transfer. The ultra-long GnRHa protocol was used in the control group. Patients were administered with ultra-long triptorelin (3.75 mg) on the second day of menstrual cycle, and then hMG (225 IU) was administered 5–6 weeks later. In the ultra-long GnRHa protocol, the final stage of oocyte maturation was triggered using hCG (5000 IU, Lizhu Pharmaceutical Trading Co., Zhuhai, China). Fresh embryo transfer was the first choice in this group except for the patients who had to undergo frozen embryo transfer due to some reasons. The embryo quality (number/uniformity of blastomeres and degree of fragmentation) was assessed [16]. The embryo morphology was scored according to the criteria of Cummins [16]. All follicles with a diameter larger than 10 mm were retrieved. Oocytes are fertilized conventionally or by intracytoplasmic sperm injection. Embryos were graded in the same way on the third day. One or two embryos with good quality were transferred, and the procedures for embryo freezing and thawing were carried out the same as in the MPA + hMG group. If the fresh embryo transfer (ET) was not prepared due to high risk of ovarian hyperstimulation syndrome (OHSS), elevated progesterone (P) on the triggering day, unqualified endometrium, or other personal reasons, all good-quality embryos were frozen, and frozen-thawed embryo transfer (FET) was done later.

Laboratory protocols, transfer of cryopreserved-thawed embryos and vitrification

Details on ovarian stimulation, oocyte retrieval and IVF/ICSI procedures have been previously described. Briefly, conventional IVF or ICSI was carried out based on semen parameters and previous fertilization history. Fertilization was assessed 16–18 h post insemination/injection. Zygotes were subsequently transferred into a dish containing preequilibrated culture medium. All embryos were cultured under mineral oil in an incubator with 5% O2 and 6% CO2 at 37 °C. Embryo development was evaluated on Day 3, and only high-quality cleavage-stage embryos (at least six blastomeres with ≤ 20% fragmentation) were selected for cryopreservation. Except for the change in culture medium, all other IVF protocols and laboratory conditions remained constant throughout the study period [17]. Embryo morphology assessment was evaluated on day 3, day 5, and day 6. Cleavage-stage embryos with at least 6 blastomeres and fragmentation < 20% were regarded as high-quality embryos [16]. Blastocysts were scored according to the Gardner and Schoolcraft grading system [18] and recorded as high-quality ones if they reached at least an expansion stage 4 with C or C for inner cell mass and trophectoderm. Expansion was categorized based on the following degrees: 1, an early blastocyst with its blastocoele filling < 50% of the embryo; 2, an early blastocyst with a blastocoele filling > 50% of the embryo; 3, a full blastocyst with a blastocoele filling the entire blastocyst; 4, an expanded blastocyst with a blastocoele larger than its size and a thin zona pellucida; 5, a hatching blastocyst out of the zona pellucida; and 6, a hatched blastocyst that has completely escaped from the zona pellucida. The inner cell mass (ICM) was graded as follows: A, numerous tightly packed cells; B, a few loosely grouped cells; and C, very few cells, The TE was evaluated including following grades: A, many cells many cells organized in the epithelium; B, several cells organized in a loose epithelium; and C, very few cells pushed to the side.

In all FET cycle, no more than 2 embryos can be transferred. The endometrial preparation in FET cycles was performed in natural cycles, mild stimulation cycles, or hormone therapy cycles [19]. The vitrification and thawing procedure were previously described in our previous studies [20]. Briefly, embryo vitrification was performed via Cryotop carrier system, in conjunction with dimethylsulfoxide-ethylene glycol–sucrose as cryoprotectants. For thawing, embryos were transferred into dilution solution in a sequential manner (1 mol/L to 0.5 mol/L to 0 mol/L sucrose).

Data collection

The primary outcome of this study was the number of oocytes retrieved. The secondary outcomes included the duration and dosage of hMG, number of mature oocytes (MII oocyte) retrieved, the ratio between the total number of oocytes collected at the end of ovary stimulation (Follicleto-Oocyte Index [FOI]) [21], health economic indicators, cycle cancellation rate, fertilization rate, implantation rate, clinical pregnancy rate, and live birth rate. Clinical pregnancy and ongoing pregnancy were considered in the presence of a gestational sac with fetal heart activity, as assessed by ultrasonography at 7 and 12 weeks of gestation, respectively. The implantation rate was defined as the number of gestational sacs divided by the number of embryos transferred. The early miscarriage rate was defined as the proportion of patients with spontaneous pregnancy termination before the gestational age of 12 weeks [19].

Hormone measurement

The AFC was monitored by vaginal ultrasonography, and anti-mullerian hormone (AMH) was measured by chemiluminescence assay. The serum levels of FSH, luteinizing hormone (LH), estradiol (E2) and progesterone were measured on day 3 of the stimulation cycle, the triggering day, and the day after triggering (12 h later after the injection of GnRHa and hCG). Hormone levels were measured by chemiluminescence assay (Abbott Biologicals B.V, Netherlands).

Statistical analysis

Continuous data were tested for normal distribution using the Kolmogorov–Smirnov test. Continuous data with normal distribution are presented as mean ± standard deviation (SD), and those with non-normal distribution as median (range or interquartile range). Categorical variables are presented as frequency with proportion. The homogeneity of variances was examined with Levene test. T-tests or analysis of variance (ANOVA) test were used to analyse continuous variables and the Pearson Chi-square test to analyse categorical variables. A value of P < 0.05 was considered statistically significant. All data were analysed using the SPSS version 16.0 for Windows (IBM, Armonk, NY, USA).

Results

Patients’ characteristics

A total of 300 patients were included in the present study (MPA + HMG group: n = 150; ultra-long regimen group: n = 150). Among them, 300 patients completed the oocytes retrieval cycles. In the MPA + HMG group, 150 completed the oocytes retrieval cycles, there were 129 patients with viable embryos after oocytes retrieval cycles, 113 patients had 157 FET cycles, and there were 69 neonates. In the ultra-long protocol group, 150 completed the oocytes retrieval cycles, 137 patients had viable embryos after oocytes retrieval cycles. One hundred patients had 100 fresh embryo transfer cycles and subsequent 48 FET cycles. In the ultra-long protocol group, 49 babies were born after fresh embryo transfer and 27 after FET.

Baseline characteristics of patients in two groups are summarized in Table 1. No differences were found in the age, body mass index (BMI), duration of infertility, and primary infertility rate between two groups. The number of antral follicles (MPA + HMG group: 10.35 ± 3.92; ultra-long protocol group: 10.57 ± 4.66), and the level of AMH (MPA + HMG group: 3.26 ± 1.43 ng/mL; ultra-long protocol group: 3.05 ± 1.16 ng/mL) were also comparable between two groups (Table 1).

Table 1

Characteristics of women in two groups

Characteristics	MPA + HMG Group(n = 150)	Ultra-long Group(n = 150)	P
Number of cycles	150	150
Age, years	32.80 ± 3.43	32.46 ± 3.48	0.391
BMI (kg/m2)	21.61 ± 2.15	21.31 ± 2.68	0.231
Duration of infertility, y	3.15 ± 2.33	3.27 ± 2.57	0.672
Number of ET failures	0.32 ± 0.79	0.35 ± 1.09	0.761
Primary infertility, %	66 (99/150)	70.67 (106/150)	0.385
Antral follicle count, n	10.35 ± 3.92	10.57 ± 4.66	0.675
AMH (ng/ml)	3.26 ± 1.43	3.05 ± 1.16	0.562
Cyst surgery, %	68 (102/150)	64 (96/150)	0.465
Day 3 measures
FSH, IU/L	6.11 ± 1.46	5.74 ± 1.93	0.089
LH, IU/L	3.41 ± 1.39	3.28 ± 1.54	0.061
E2, pg/mL	40.10 ± 16.65	38.90 ± 17.29	0.588
P, ng/mL	0.29 ± 0.12	0.29 ± 0.14	0.595

P < 0.05: MPA + HMG group vs. Ultra-long group

BMI, body mass index; FSH, follicle stimulating hormone; LH, luteinizing hormone; E2, estrogen; P, progesterone; AMH, anti-mullerian hormone

Changes in hormones

The blood levels of FSH, LH, E2 and P in the MPA + HMG group are shown in Fig. 3. FSH level increased significantly after hMG administration and remained stable during ovarian stimulation. However, in the ultra-long protocol group, the FSH level was stable during the COH, and there was significant difference in the time to first hospital visit (MPA + HMG group: 16.62 ± 4.79 IU/L; ultra-long protocol group: 5.94 ± 6.25 IU/L; P < 0.001). No difference was found in the FSH level on the triggering day between two groups (MPA + HMG group:16.90 ± 4.45 IU/L; ultra-long protocol group: 16.00 ± 4.40 IU/L; P = 0.085). The FSH level on the day after triggering was markedly higher in the MPA + HMG group than in the ultra-long protocol group (MPA + HMG group: 26.15 ± 7.73 IU/L; ultra-long protocol group: 12.09 ± 4.02 IU/L; P < 0.001). The basal LH level had no significant difference between two groups; however, during the COH, the LH level reduced in both groups. The LH level further reduced significantly in the MPA + HMG group than in the ultra-long protocol group on the first visit (3.25 ± 5.77 vs 1.04 ± 2.02) and the triggering day (1.92 ± 1.38 vs 0.75 ± 0.59). LH level gradually decreased during ovarian stimulation, and the mean LH level on the triggering day was significantly lower than the basal LH level in two groups (MPA + HMG group: 1.92 ± 1.38 vs 3.41 ± 1.39 IU/L; ultra-long protocol group: 0.75 ± 0.59 vs 3.28 ± 1.54 IU/L) and then it increased significantly (MPA + HMG group: 59.17 ± 27.32 IU/L vs ultra-long protocol group: 1.65 ± 5.97 IU/L; P < 0.001) at 10 h after triggering. The LH level remained at a low level during COH, and there was no premature LH peak. No patient had premature ovulation. Serum E2 level gradually increased accompanied by the increase in the number of follicles during ovarian stimulation after triggering, but E2 level remained unchanged on the triggering day (2706.83 ± 1424.27 pg/ml vs 2612.68 ± 1447.95 pg/ml) and the day after triggering (2944.71 ± 1437.54 pg/ml vs 2971.19 ± 1507.85 pg/ml). The E2 level was markedly higher in the MPA + HMG group than in the ultra-long protocol group on the first visit (2971.19 ± 1507.85 pg/ml vs 1701.65 ± 1106.47 pg/ml; P < 0.05). The P level remained at a low level during ovarian stimulation (0–1 ng/ml) (Fig. 3).

Fig. 3

Hormone profiles of the MPA + hMG group and the ultra-long GnRHa group with triggering by GnRH-a or hCG. The mean ± SD show the temporal associations among blood levels of follicle stimulating hormone (FSH), luteinizing hormone (LH), estrogen (E2), and progesterone (P). Green line: MPA + HMG group, red line: ultro-long GnRHa group. *P < 0.05 MPA + HMG group vs ultro-long protocol group at the same time point

Ovarian stimulation, follicle development and oocyte performance

The AFC at 2–5 days of menstrual cycle was 10.35 ± 3.92 in the MPA + HMG group and 10.57 ± 4.66 in the ultra-long protocol group, showing no significant difference (P > 0.05). The number of follicles with the diameter > 10 mm (MPA + HMG group: 10.70 ± 5.27; ultra-long protocol group: 12.28 ± 6.60, P < 0.05) or 14 mm (MPA + HMG group: 7.09 ± 3.82; ultra-long protocol group: 8.79 ± 5.11, P < 0.05) was significantly different between two groups. The MPA + HMG protocol group was characterized by a lower dose of hMG (MPA + HMG group: 1882.50 ± 388.77 IU; ultra-long protocol group: 2456.50 ± 560.17 IU; P < 0.001), and the duration of stimulation was also markedly different between two groups (MPA + HMG group: 8.72 ± 1.48 days; ultra-long protocol group: 11.31 ± 2.26 days; P < 0.001). The FOI was significantly different between MPA + hMG group and GnRHa group (89.83% vs 97.02%, P < 0.001).

No significant differences were noted in the number of oocytes retrieved (MPA + HMG group: 9.30 ± 5.73; ultra-long protocol group: 9.33 ± 5.36), maturation (MPA + HMG group: 7.77 ± 5.23; ultra-long protocol group: 8.23 ± 4.93), number of high-quality embryos (MPA + HMG group: 2.69 ± 2.43; ultra-long protocol group: 2.43 ± 2.44;) and number of viable embryos (MPA + HMG group: 3.09 ± 2.46; ultra-long protocol group: 3.07 ± 2.38). There were no significant differences in the rate of high-quality embryos (D3 6-cell grade I and II and above) (MPA + HMG group: 28.89%, ultra-long protocol group: 26.07%) and the viable embryo rate per oocyte retrieved (MPA + HMG group: 33.26%, ultra-long protocol group: 32.93%) (P > 0.05).

The rate of maturation was significantly lower in the MPA + HMG group than in the ultra-long protocol group (83.58% vs 88.21%; P < 0.001), but the number of MII was similar between two groups. The fertilization rate for ICSI was higher in the MPA + HMG group than in the ultra-long protocol group (90.13% vs 84.40%; P = 0.023). The cleavage rate was similar between two groups (97.03% vs 97.81%). The cycle cancellation rate due to absence of viable embryos was not different between two groups (14% vs 8.67%; P > 0.05). The rates of hyper-response (18.67% vs 10.67%) and hypo-response (0.67% vs 2.67%) were also comparable between two groups (P > 0.05). No patients experienced moderate or severe OHSS in the whole study (Table 2).

Table 2

Cycle characteristics and embryological outcomes of women in two groups

Characteristics	MPA + HMG Group(n = 150)	Ultra-long Group(n = 150)	P
hMG dose (IU)	1882.50 ± 388.77	2456.50 ± 560.17*	< 0.001
hMG duration, days	8.72 ± 1.48	11.31 ± 2.26*	< 0.001
hMG dose per follicle, IU	225.25 ± 130.47	282.18 ± 267.05*	0.02
The level of E2 per mature follicle (ng/ml)	272.47 ± 91.68	232.55 ± 92.32*	< 0.001
FOI	89.83 (1395/1553)	97.02 (1400/1443)	< 0.001
Cancelled cycle	0	0	/
No. of > 10 mm follicles on the triggering day	10.70 ± 5.27	12.28 ± 6.60*	0.023
No. of > 14 mm follicles on the triggering day	7.09 ± 3.82	8.79 ± 5.11*	0.001
No. of > 10 mm follicles on oocytes pick up day	12.41 ± 5.47	14.21 ± 6.96*	0.013
No. of oocytes punctured	12.68 ± 7.11	13.55 ± 7.49	0.301
Oocytes retrieved (n)	9.30 ± 5.73	9.33 ± 5.36	0.959
MII oocytes (n)	7.77 ± 5.23	8.23 ± 4.93	0.434
Fertilized oocytes (n)	6.73 ± 5.00	6.92 ± 4.35	0.722
Top-quality embryos (n)	2.69 ± 2.43	2.43 ± 2.44	0.368
Viable embryos (n)	3.09 ± 2.46	3.07 ± 2.38	0.943
Oocyte retrieval rate, %	73.34 (1395/1902)	68.86 (1400/2033)	0.38
Mature oocyte rate, %	83.58 (1166/1395)	88.21 (1235/1400)*	< 0.001
Fertilization rate, %	65.23 (910/1395)	65.35 (915/1400)	0.945
ICSI rate, %	31.33 (47/150)	40.6 7(61/150)	0.092
Fertilization rate for ICSI, %	90.13 (283/314)	84.40 (357/423)*	0.023
Cleavage rate, %	97.03 (883/910)	97.81 (995/915)	0.292
High quality embryo rate, %	28.89 (403/1395)	26.07 (365/1400)	0.095
Viable embryo rate, %	33.26 (464/1395)	32.93 (461/1400)	0.852
Cancellation rate, %	14 (21/150)	8.67 (13/150)	0.380
Premature LH rate, %	0	0	/
Incidence of OHSS, %	0	0	/

*P < 0.05 MPA + HMG group vs. Ultra-long group

MPA, medroxyprogesterone acetate; HMG, human menopausal gonadotropin

Pregnancy outcomes in FET cycles

The FET pregnancy outcomes between two groups are presented in Table 3. A total of 262 women (MPA + HMG group: n = 129; ultra-long protocol group: n = 133; fresh-ET: n = 100. 15 patients who were non-pregnant after fresh ET received subsequent frozen ET. In the ultra-long protocol group, 33 patients did not receive fresh ET. Thus, 48 patients received FET) in two groups completed a total of 325 FET cycles (MPA + HMG group: n = 157; ultra-long protocol group: n = 168; fresh-ET: n = 100, frozen-ET: n = 68). The implantation rate was comparable between two groups (MPA + HMG group: 34.27%; ultra-long protocol group, fresh-ET: 33.85%, frozen-ET: 39.67%) (P > 0.05). The clinical pregnancy rate (MPA + HMG group: 50.31%; ultra-long protocol group, fresh-ET: 55%, frozen- ET: 48.53%), miscarriage rate, multiple pregnancy rate, ongoing pregnancy rate and cumulative pregnancy rate were similar between two groups (all P > 0.05). No significant difference was observed in the live birth outcome between two groups. The live birth rate was also similar (MPA + HMG group: 43.59%; ultra-long protocol group, fresh-ET: 49%, frozen- ET: 39.71%). There were no significant differences in the complications during pregnancy and postpartum and in the birth malformation rate between two groups (all P > 0.05).

Table 3

Pregnancy outcomes from fresh/frozen-thawed embryo transfers in two groups

Characteristics	MPA + HMG Group-FET	Ultra-long Group-ET	Ultra-long Group-FET	P
Patients (n)	129	100	33
Patients of ET (n)	0	100
Patients of FET	113		48
FET cycles (n)	157		68
Embryo before thawed	286	/	115
Viable embryo after thawed	286	/	115
Transferred embryos per cycle (n)	1.9 ± 0.4	1.8 ± 0.4	1.8 ± 0.5	0.236
Endometrial thickness (mm)	11.3 ± 1.6	11.7 ± 2.9	11.5 ± 2.3	0.351
Clinical pregnancy rate per (%)	50.31 (79/157)	55(55/100)	48.53 (33/68)	0.665
Implantation rate (%)	34.27 (98/286)	33.85 (65/192)	39.67 (48/121)	0.517
Ectopic pregnancy rate (%)	1.27 (1/79)	1.81 (1/55)	0 (0/33)	0.747
Multiple pregnancy rate (%)	30.38 (24/79)	21.81 (12/55)	36.36(12/33)	0.312
Miscarriage rate (%)	11.39 (9/79)	9.09 (5/55)	21.21 (7/33)	0.229
Ongoing pregnant rate per transfer (%)	43.95 (69/157)	49 (49/100)	38.23 (26/68)	0.384
Cumulative pregnancy rate per woman (%)	61.01 (69/113)	50 (50/100)	56.25 (28/48)	0.479
Gestation, w	37.87 ± 2.07	38.33 ± 2.10	37.00 ± 2.22	0.879
Birth length, cm	49.35 ± 1.35	49.56 ± 1.23	49.16 ± 1.22	0.568
Birth weight, g	3030.6 ± 642.8*	3087.5 ± 564.01#	2633.9 ± 634.9*#	0.007
Child’s sex-male, no. (%)	55.43 (51/92)	52.72 (39/55)	51.72 (15/29)	0.484
Live birth, no. (%)	43.95 (69/157)	49 (49/100)	39.71 (27/68)	0.48
Complications during pregnancy and postpartum (%)	8.85 (10/113)	5 (5/100)	8.51 (4/47)	0.532
Birth malformation rate (%)	0.98 (1/102)	0 (0/100)	0 (0/43)	0.602

*P < 0.0167 MPA + HMG group vs Ultra-long Group-ET group; #P < 0.0167 Ultra-long Group-ET vs Ultra-long Group-FET group

FET, frozen-thawed embryo transfer

In our study, the medical cost in the ultra-long protocol group and MPA + HMG group was 4.61 ± 0.23 thousand vs. 4.52 ± 0.24 thousand $, respectively (excluding the cost for diet, accommodation and transportation), showing no significant difference (P = 0.241); the number of hospital visit was 10.1 ± 1.27 vs 10.42 ± 1.29 times (time from first visit to transplantation), respectively, showing no marked difference (P = 0.325); the average time spent was significantly different between two groups (66.97 ± 4.18 days vs 61.06 ± 2.63 days; P < 0.001) (from first visit to transplantation) (Fig. 4).

Fig. 4

Health economic indicators in the two groups. 1: ultra-long GnRHa group; 2: MPA + HMG group

Discussion

This study aimed to investigate the efficiency of MPA + hMG in patients with ovarian endometriosis undergoing COH compared with conventional one-month ultra-long protocol and to explore the clinical outcomes in fresh ET as well as subsequent frozen ET. Although the numbers of oocytes retrieved and MII oocytes were similar between two groups, the fertilization rate for ICSI was significantly higher in the MPA group than in the ultra-long protocol group. In the MPA group, with mild pituitary suppression, normal FOI > 50% and fewer side-effects, the dose of HMG significantly decreased. Meanwhile, the average duration of stimulation was shorter in the MPA + hMG group than in the ultra-long protocol group. The number of cleaved embryos, ovary response, cancellation, pregnancy and live birth outcomes after frozen ET were similar between two groups.

In the conventional ultra-long protocol, more than two doses of long-acting triptorelin was administered prior to COH, aiming to create a low estrogen status, which facilitates the atrophy of ectopic endometrium, and then gonadotropin (Gn) is administered in the IVF [13]. Although this regimen for COH can improve the pregnancy rate to a certain extent in patients with ovarian endometriosis, the treatment cycle is long, the dose of Gn used is high, and the side effects (such as hectic fever and night sweating) are significant. Therefore, how to ensure an acceptable pregnancy rate while reducing the side effects during the IVF has become a focus in recent studies. In the study of Benschop et al., results showed no significant differences in the pregnancy outcomes between GnRH agonists and GnRH antagonists during the in vitro fertilization—embryo transfer (IVF-ET) (OR = 0.81, 95% CIs = 0.26–2.54) [22]. Hughes et al. showed similar benefits for the oral contraceptives vs no treatment [23]. Recent studies have indicated that long-term pituitary down-regulation has limited effect on the FET in severe endometriosis patients [13]. However, down-regulation is based on the results of fresh embryo transplantation. At present, there is still controversy on the time of down-regulation in clinical practice. The conventional down-regulation may not improve the reduced quality of oocytes and embryos after ovulation in the endometriosis patients. In addition, in the ultra-long protocol, the prolongation of down-regulation may increase the hospital visit and thereafter increase the duration of medication, which increases the medical cost. Moreover, the long-lasting down-regulation may compromise the compliance of patients to the treatment, leading to the reduced clinical outcomes. Thus, in the present study, one-month protocol was employed for down-regulation.

It has been confirmed that endometriosis may adversely affect ovaries. Toxic contents (such as free iron, reactive oxygen species, proteolytic enzymes and inflammatory factors) released in the endometrioma may lead to unfavorable events such as increased oxidative stress, thereafter reducing follicular maturation and impaired fertilization [24, 25]. There is evidence showing that a 3-month ultra-long GnRHa may create a more favorable environment for oocyte maturation, with better oocyte and embryo quality as well as fertilization rate. Whether this environment can also be improved by MPA treatment in the IVF/ICSI is still controversial. MPA is clinically used to alleviate pain and improve overall comforts in most women with endometriosis [26, 27]. The mechanism underlying the effects of MPA and other progestins remains to be investigated. It has been shown that their efficacy stems from pituitary inhibition and atrophy of endometriotic lesions [28]. Another possible explanation is that MPA alleviates inflammation at the site of endometriotic implants. Prolonged treatment (8 days) with MPA decreases the luciferase activity by 36% and reduces CCL5/RANTES protein production by 50%; however, shorter treatment (2 or 4 days) with MPA has no significant effect on the CCL5/RANTES protein production. We also found that 8-d MPA treatment increased progesterone receptor (PR) expression [9]. In the present study, there was no significant difference in the clinical outcomes of patients with fresh -ET or FET after using two protocols. This may be explained as that the clinical outcomes are influenced by many factors. However, the fertilization rate in the MPA + HMG group increased. Whether it is related to the reduced oxidative stress and inflammation as well as improved microenvironment for follicular development in case of endometriosis remains to be further verified.

There were several novelties in our study. First, this was a randomized controlled trial, and patients with ovarian endometriosis and normal ovarian function were recruited. Indeed, patients with poor ovarian reserve function were excluded to avoid the influence on the embryo and pregnancy outcomes. Second, MPA possesses significant anti-inflammatory and anti-oxidative activities, which can improve the microenvironment for follicular development. Third, fewer side effects were noted in the MPA + HMG protocol with lower dose and shorter duration. This implicates that the ovarian sensitivity can be reduced in the ultro-long group. In the study, the socioeconomics indicators (time to embryo transfer, medical cost and side effects) were comparable between two groups. If the time of down-regulation is longer than one cycle, the times of hospital visit and the duration of medication in the ultra-long protocol group will increase significantly. Therefore, the MPA + HMG protocol is a choice for patients with endometriosis under the premise of FET.

In this study, the MPA + HMG protocol was compared with the one-month pituitary down-regulation protocol, not the more than 2–3 month-long regimen, in patients with endometriosis. Whether the duration of down-regulation would affect the oocyte and pregnancy outcomes is still unclear and more studies are needed to investigate this issue. Moreover, there is no consensus about the time and dose of progestin priming. In the present study, MPA was administered at 10 mg/day for more than consecutive 10 days in the MPA protocol. Whether the dose and time of MPA affect the endocrine characteristics and the level of inflammatory factors are needed to be verified. Although the patients included in this study had ovarian endometriomas and recurrence of ovarian cyst after surgery, the ovarian function (AFC and AMH) was normal. Whether the conclusion can be expanded to patients with decreased ovarian reserve function should be further investigated. Another probable bias is that some patients did not receive laparoscopy or open surgery, and ultrasonography was employed for the diagnosis. The types of lesions can be reliably identified by transvaginal ultrasonography [29, 30], chocolate cyst was punctured during oocyte retrieval, but this requires experience and skills. Severe endometrioma may be accompanied by peritoneal and deep endometriosis. This study was a prospective randomized controlled study. One major limitation of this study was the small sample size. Furthermore, insufficient FET cycles and the limited data on neonatal outcomes also reduced the power of this study. Due to the side effects of ultra-long downregulation and the fact that patients are unwilling to accept ultra-long downregulation due to the long treatment, the study may be terminated early. In addition, ovarian endometriosis may have concomitant peritoneal and/or deep endometriosis, and the heterogeneity of the population may also bias our findings. Although blind randomization was done, the treatment was hard to be blind to patients due to the different ways by which the protocols were applied, which also compromised the power of the study. Besides, the person who managed the entire protocol was not blind to the hMG dosage, the trigging time and the endometrial preparation method in this trial. The randomization in allocation did not thoroughly control all the confounding factors between two groups. The lack of a double-blind approach decreased the power of the evidence, although this trial was strictly conducted according to good clinical practice guidelines. A double-blind randomized controlled trial with large sample size is needed to confirm our findings.

Conclusion

In conclusion, the administration of MPA in COH showed similar number of oocytes retrieved, no premature LH surge, and similar pregnancy and live birth outcomes in patients with advanced ovarian endometriosis undergoing IVF/ICSI as compared to the one-month long protocol. The use of MPA in COH appears to be promising although many questions remain to be elucidated, including the dose and time of progestin priming as well as its possible influence on the oocyte development and microenvironment. Given their good tolerability, few metabolic influence, and low cost, progestogens provide a novel alternative to the conventional protocol for patients with endometriosis.