Oral dydrogesterone along with vaginal micronized progesterone supplementation for luteal phase support in IVF patients, and its impact on pregnancy and live birth rates: a prospective randomized trial.

In a prospective, open, randomized study, a total of 700 infertile patients aged 27–41 years, undergoing in vitro fertilization treatment in a private ART Unit (One Day Medical Center, Rome, Italy), were enrolled from January 2017 to December 2022. The patients had infertility attributable to tubal factors, male factors, moderate endometriosis, combined male and female factors, or unexplained infertility. They exhibit a serum hormonal profile within the normal range (Anti Mullerian Hormone AMH  ≥  1 ng/mL), regular menstrual cycles; the presence of a normal uterine cavity; and a body mass index (BMI) of 20–26 kg/m2. Eighty patients were excluded from the study due to: gynecologic abnormality or disease, previous poor response to gonadotropins stimulation, history of severe ovarian hyperstimulation syndrome (OHSS), history of severe endometriosis, male partner had azoospermia and genetic disorder. The remaining 620 patients who satisfied the inclusion criteria underwent IVF/ICSI treatment. Frozen embryo transfer was not included in this study (Fig.  1 ). Fig. 1 Flow chart of the study population Flow chart of the study population The primary endpoints were the clinical pregnancy rate, live birth rate, implantation rate, and miscarriage rate. The secondary endpoints were the mean duration of stimulation, total dose of Follicle Stimulating Hormone (FSH), serum oestradiol level and endometrial thickness on the day of Human Chorionic Gonadotropin (hCG) administration, number of retrieved oocytes, and number of embryos transferred. All endpoints were analyzed statistically. The study has been retrospectively registered with the Clinical Trials registry reference number ISRCTN52148405 ( http://isrctn.org/ ) on the 15th of August 2023. The participating patients had a mild stimulation protocol with a starting dose of rFSH modulated on the base of AMH level, ranging between 75 and 150 IU, and 0.25 mg/day GnRH antagonist administered when the leading follicle was 14 mm. The FSH dose was adjusted when necessary according to the follicular size and oestradiol level. Final oocyte maturation was triggered by the administration of 10,000 IU of hCG. Retrieved oocytes were assessed for their maturity and inseminated by standard IVF or ICSI. Embryos were transferred on day 3 after oocyte insemination, no more than three embryos per patient were transferred. On the day of oocyte recovery, the patients were randomized into two groups: group A ( n  = 310) had luteal phase support by vaginal administration of capsules of micronized progesterone (800 mg/day), that is routine supplementation in our centre; Group B ( n  = 310) had luteal phase support by a combination of vaginal administration of a gel of micronized progesterone (90 mg/day) and oral dydrogesterone 10 mg twice a day. In the case of confirmed pregnancy, luteal support in Group A was continued until the 10th week of pregnancy, in Group B vaginal micronized gel was suspended at the 5th week of pregnancy, whereas dydrogesterone administration (20 mg/day) continued until the 10th week of pregnancy. Clinical pregnancy was confirmed 6 weeks after embryo transfer by ultrasonography. Randomization was performed using a computer-generated random assignment schedule for each patient. In order to conceal the treatment allocation a sealed and numbered envelope was given to the doctor deputy to follow each patient in order to prescribe the indicated therapy for LPS. Statistical analysis was performed using JMP software (version 17; SAS, Inc., Cary, NC, USA). The statistical power was calculated based on an α level of 0.05 (two-tailed t-test) with 80% power to detect a 10% difference between treated groups. The sample size was calculated using the two-tailed test, 95% confidence intervals (CI), and anticipated effective size (Cohen’s D) of 0.5 (medium size). Based on the two-tailed hypothesis, the minimum total sample size required was 620 patients – at least 310 valuable patients per group. The data were analysed using the two-tailed Student’s t-test for independent data, Fisher’s exact test, and a two-by-two table between groups where appropriate. P  ≤  0.05 was considered statistically significant.

A total of 700 infertile patients, aged between 27 and 41, were enrolled in the study, of which 80 patients were excluded and 620 patients underwent IVF/ICS treatment. The patients ( n  = 620) were randomized on the day of egg retrieval into two groups based on the luteal phase support therapy (Fig.  1 ). As shown in Table  1 , the two groups were comparable regarding demographic characteristics: mean age, BMI, and aetiology of infertility. Table 1 Demographic characteristics Group A (vaginal) Group B (combined) P value Patients 310 310 Mean age ± SD 36.1 ± 2.4 36.4 ± 2.7 0.91* Mean BMI ± SD 23.7 ± 1.4 24.1 ± 1.9 0.88* Tubal factor % 26.1 (81/310) 23.2 (72/310) 0.40** Male factor % 21.9 (68/310) 22.3 (69/310) 0.92** Moderate endometriosis % 8.1 (25/310) 9.3 (29/310) 0.56** Combined male and female factor % 39 (93/310) 32.5 (101/310) 0.48** Unexplained infertility % 13.9 (43/310) 12.6 (39/310) 0.63** * Two tailed student’s t-test (statistical significance when p  ≤  0.05) ** Chi square test (statistical significance when p  ≤  0.05) No significant differences between the two groups BMI = body mass index Demographic characteristics * Two tailed student’s t-test (statistical significance when p  ≤  0.05) ** Chi square test (statistical significance when p  ≤  0.05) No significant differences between the two groups BMI = body mass index There were no significant differences between the two groups regarding stimulation outcomes: days of stimulation, total dose of FSH, estradiol level and endometrial thickness on hCG day. No statistically significant differences were observed between the two groups in terms of the mean number of retrieved oocytes and the mean number of transferred embryos ( P  < 0.32 and P  < 0.15 respectively) (Table  2 ). Table 2 Stimulation and embryological outcomes Group A (vaginal) Group B (combined) P value Patients 310 310 Mean duration of stimulation (days) ± SD 12 ± 2.3 11 ± 2.6 0.84* Total rFSH dose IU 1300 ± 480 1250 ± 470 0.86* Estradiol level on hCG day (pg/ml) 1270 ± 435 1195 ± 420 0.78* Endometrial thickness on hCG day (mm) 10.7 ± 1.9 11 ± 1.7 0.83* Mean number of retrieved oocytes ± SD 5.2  ±  2.3 5 + 2.7 0.32* Mean number of transferred embryos ± SD 1.9 + 0.84 2 + 0.9 0.15* * Two tailed student’s t-test (statistical significance when p  ≤  0.05) No significant differences between the two groups rFSH = recombinant follicle-stimulating hormone hCG = human chorionic gonadotropin Stimulation and embryological outcomes Mean duration of stimulation (days) ± SD * Two tailed student’s t-test (statistical significance when p  ≤  0.05) No significant differences between the two groups rFSH = recombinant follicle-stimulating hormone hCG = human chorionic gonadotropin As depicted in Table  3 , sixteen patients had no viable embryos for replacement due fertilization failure, embryo cleavage failure or embryo developmental arrest: 7 in group A and 9 in Group B. The remaining 303 patients in group A and 301 patients in Group B had viable embryos for replacement. Table 3 Clinical outcomes Group A (vaginal) Group B (combined) P value RR (CI 95%) Patients at oocyte recovery 310 310 Patients with failure of fertilization or failure of embryo cleavage % 0.022 (7/310) 0.029 (9/310) 0.61* 1.00 (0.9812–1.0328) Clinical pregnancy per patient (%) 34.9 (108/310) 35.7 (111/310) 0.80* 1,02 (0.7863–1.2040) Clinical pregnancy per embryo transfer (%) 35,6 (108/303) 36.8 (111/301) 0.75* 0.96 (0.7823 to 1.1942) Live birth per patient (%) 30.6 95/310 29.2 (91/310) 0.72* 0,98 (0.8207–1.3279) Live birth per embryo transfer (%) 31.3 (95/303) 30.2 (91/301) 0.76* 1.03 (0.8164 to 1.3173) Miscarriage rate (%) 12 (13/108) 18 (20/111) 0.21* 0.66 (0.3501 to 1.2747) Implantation rate (%) 18.6 (114/612) 17.1 (109/636) 0.49* 0.98 (0.9324 to 1.0343) * Chi square (statistical significance when p  ≤  0.05) No significant differences between the two groups RR = relative risk Clinical outcomes 0.022 (7/310) 0.029 (9/310) 1.00 (0.9812–1.0328) 34.9 (108/310) 35.7 (111/310) 1,02 (0.7863–1.2040) 35,6 (108/303) 36.8 (111/301) 0.96 (0.7823 to 1.1942) 30.6 95/310 29.2 (91/310) 0,98 (0.8207–1.3279) 31.3 (95/303) 30.2 (91/301) 1.03 (0.8164 to 1.3173) 12 (13/108) 18 (20/111) 0.66 (0.3501 to 1.2747) 18.6 (114/612) 17.1 (109/636) 0.98 (0.9324 to 1.0343) * Chi square (statistical significance when p  ≤  0.05) No significant differences between the two groups RR = relative risk No statistically significant differences were observed between the two groups in terms of clinical pregnancy rate (34.9% in group A vs. 35.7% in group B, P  < 0.96) or per embryo transfer (35.6% in group A vs. 336.8% in group B, P  < 0.96), live birth rate per treated patient (30.6% in group A vs. 29.2% in group B, P  < 1.0) or per embryo transfer (31.3% in Group A vs. 30.2% in Group B, P  < 1.03), and implantation rate (18.6% in group A vs. 17.1% in group B, p  < 0.97). A higher, though not statistically significant, miscarriage rate was observed in group B compared to group A (18% vs. 12% respectively, P  < 0.7) (Table  3 ). No relevant side effects were observed in both treated groups.

A normal luteal phase function is an essential factor for maintaining pregnancy; luteal phase deficiency decreases embryo implantation and pregnancy rate and increases the early miscarriage rate. Accordingly, for stimulated in vitro fertilization-embryo transfer (IVF-ET) patients, luteal phase support becomes an important step of the procedure. The mechanism of luteal insufficiency after ovarian stimulation for IVF treatment remains an open door for discussion. The abnormal luteal phase function in stimulated cycles could be due to a variety of reasons: the continuous down-regulation with gonadotropin-releasing hormone (GnRH) agonist may negatively affect the pituitary recovery [ 1 , 2 ]; multiple follicle development itself could influence the duration of the luteal phase [ 3 , 4 ]; the elevated level of steroids due to the higher number of corpora lutea during the early luteal phase may inhibit luteinizing hormone (LH) release via negative feedback actions [ 5 , 6 ]; the removal of large quantities of cumulus cell mass during egg retrieval could alter the luteal phase function by diminishing the most important source of progesterone synthesis by corpus luteum [ 7 , 8 ]. It is widely established that the use of GnRH-agonists and recently antagonists for pituitary down-regulation are useful adjuncts to in vitro fertilization treatment cycles, as they decrease the incidence of premature LH surge. Nevertheless their beneficial effect during ovarian stimulation, these drugs may lead to significant alterations in the hormonal milieu of the luteal phase [ 9 , 10 ]. Therefore, luteal phase support with exogenous supplementation therapy became an essential component of IVF treatment as it can significantly improve pregnancy outcomes [ 11 , 12 ]. The use of progesterone as a supplementation therapy for luteal phase support has demonstrated a significant improvement in pregnancy maintenance and clinical outcomes [ 13 ]. Accordingly, progesterone supplementation became mandatory for luteal phase support in stimulated IVF/ICSI (in vitro fertilization/intracytoplasmic sperm injection) cycles, as stated recently by the European Society for Human Reproduction and Embryology (ESHRE) Guideline group on ovarian stimulation 2020 [ 14 ]. Progesterone administration has been implemented in different routes either intramuscularly (IM), subcutaneously, vaginally, rectally, or orally, with controversial results. Intramuscular administration of natural progesterone has been considered the best choice for luteal phase support in IVF-ET cycles. Recently the vaginal administration of micronized natural progesterone has gained popularity in assisted reproductive technology (ART) as it alleviates the side effects and low compliance of intramuscular progesterone. A recent meta-analysis revealed no differences between vaginally and intramuscular routes of progesterone administration in terms of efficacy and pregnancy outcome [ 15 ]. However, other studies reported that the incidence of side effects associated with intramuscular progesterone seems to be higher than that with vaginal progesterone [ 16 , 17 ]. Another natural progesterone formulation which can be administered subcutaneously has recently become available for luteal phase support, and the results obtained have shown no statistical differences when compared with the intramuscular route in terms of live birth rate [ 18 ]. Recently, another interesting concept was focused on the oral route of progesterone. administration in ART, given the fact that dydrogesterone (DG) has long been used to treat other conditions associated with progesterone deficiency [ 19 ]. Dydrogesterone is a 6-dihydro-retroprogesterone characterized by high oral bioavailability, good tolerability, easy absorption, and higher selectivity for progesterone receptors [ 20 ]. Its oral route of administration is considered to be a more patient-friendly protocol that can improve treatment compliance. Up to date, only a few studies have considered the use of oral dydrogesterone for luteal phase support in fresh IVF cycles, with variable results summarized in a recent review [ 21 ]. Oral dydrogesterone has been compared with vaginal micronized natural progesterone and the reported results, in terms of live birth and ongoing pregnancy rates, are still controversial [ 22 , 23 ]. It has also been proposed that the association of oral dydrogesterone with vaginal micronized progesterone for luteal phase support may improve clinical outcomes. As far as we know, only one recent retrospective study reported that the use of oral dydrogesterone combined with vaginal micronized progesterone for luteal support significantly improves clinical and ongoing pregnancy rates when compared to the use of vaginal micronized progesterone alone in patients undergoing fresh embryo transfer [ 24 ]. Although several luteal phase supplementation protocols have been implemented, no accepted regimen has been established yet for optimal luteal phase support. This study aimed to compare the administration of a combined protocol of progesterone vaginal gel and oral dydrogesterone, with the administration of micronized vaginal progesterone alone for luteal phase support in patients undergoing GnRH antagonist-mild stimulation regimen.

Our study shows no statistical differences observed between the combined route group and single route group in terms of clinical outcomes, and the combination of oral dydrogesterone and vaginal micronized progesterone seems to be non-superior to vaginal progesterone alone, at least in mild stimulation setting. Further randomized studies, taking into account also the different protocols of ovarian stimulation, are needed to understand whether vaginal micronized progesterone alone has the same or more efficacy than the association of vaginal micronized progesterone with oral dydrogesterone for the luteal phase support. Indeed, searching for optimal luteal phase support still needs further investigation.

A growing body of evidence demonstrates that progesterone supplementation for luteal phase support highly improves the IVF outcome. Therefore, luteal phase supplementation with progesterone became a mandatory step in ART cycles [ 14 ]. Different routes of natural progesterone administration are available for luteal phase support in ART cycles: intramuscular, vaginal, subcutaneous, and oral. While the oral administration of natural progesterone has shown to be less efficacious and can cause systemic adverse effects, all the other routes of natural progesterone administration have shown no statistical differences in terms of pregnancy rates and live birth rates [ 14 ]. Vaginal administration of micronized progesterone, which is available either as a vaginal gel (90 mg) or vaginal capsules (200 mg), has gained popularity and has largely replaced the intramuscular route because of its fewer side effects and comparable efficacy [ 15 – 17 ]. Indeed, each route of progesterone administration is burdened by different side effects that reduce the patient’s compliance with the treatment [ 25 ]. Recently, Dydrogesterone has been introduced as an oral progesterone administration route for luteal phase support in IVF patients with variable success. Barbosa et al.2016 [ 26 ] in a meta-analysis, comparing vaginal progesterone and oral dydrogesterone found no differences between the two routes of administration in terms of pregnancy rate or live birth rate. However, in another meta-analysis, Griesinger et al. 2020 [ 23 ] reported a higher pregnancy rate and live birth rate when oral dydrogesterone is compared with micronized vaginal progesterone for luteal phase supplementation. The recent ESHRE COS guidelines stated that “dydrogesterone can be recommended for luteal phase support, although the level of evidence was moderate [ 14 ]. Eventually a recent meta-analysis has shown the absence of additional risk of congenital anomalies after dydrogesterone use [ 27 ]. On the other hand, in daily practice, different products containing progesterone with different routes of administration are often used in a combined regimen. In a recent survey on the luteal phase support practice up to 17,7% of clinicians declared preferred combined route to support the luteal phase in the ART cycle, whereas the oral route of administration alone was preferred by a minority of clinicians (2,3%) [ 28 ]. In our study, we compared the use of dydrogesterone (20 mg daily) associated with 90 mg daily of vaginal gel of progesterone with the use of 800 mg/die of vaginal micronized progesterone alone for the luteal phase supplementation in IVF patients. The primary endpoints evaluated were pregnancy, implantation, live birth and miscarriage rates. The use of an association of different progesterone products with different routes of administration relies on a few preliminary reports that seem to suggest an improvement in IVF outcomes. Patki et al. (2007) have compared a combined regimen of micronized progesterone (600 mg/day) and oral dydrogesterone (20 mg/day) with 600 mg of vaginal micronized progesterone alone for the luteal phase support in two groups of stimulated patients, in a GnRH agonist long protocol. They observed a higher clinical pregnancy rate in the combined medication group than in the single treatment group [ 29 ]. Other studies have compared the association of those different routes of administration of natural progesterone. Tomic et al. (2011) have compared the results following the administration of 90 mg of progesterone vaginal gel applied daily in combination with 100 mg of oral micronized progesterone administered three times a day with the administration of 90 mg of progesterone vaginal gel applied daily, in a GnRH agonist long stimulation protocol. The combined group showed a higher ongoing pregnancy rate and lower miscarriage rate compared to the single-treatment group [ 30 ]. Similar results were obtained by Devine et al. (2018) when comparing single administration of vaginal micronized progesterone (400 mg/die) with the combined administration of intramuscular (50 mg/day) and vaginal progesterone (400 mg/day) in frozen embryo transfer cycles [ 31 ]. Recently, in a retrospective study of a large group of patients undergoing fresh embryo transfer, in a GnRH antagonist protocol, De M et al. (2023) evaluated the IVF outcome in a group of patients who received vaginal micronized progesterone (90 mg/day) associated with oral dydrogesterone (20 mg/day) compared to the group treated only with vaginal micronized progesterone (90 mg/die). After a complex statistical evaluation, they found a statistically higher pregnancy rate in the combined group [ 24 ]. In our study, no statistical differences were observed between the combined route group and single route group in terms of pregnancy rate, live birth rate, miscarriage and implantation rate. As far as we know, this is the first prospective randomized study evaluating the efficacy of the combination of oral dydrogesterone (20 mg/day) and vaginal micronized progesterone (90 mg/day) in comparison to the vaginal administration of 800 mg/day of micronized progesterone alone in patients undergoing ART in a fresh transfer after an antagonist stimulation protocol. The combination of the two different formulations of progesterone (vaginal and oral) for luteal phase support does not show any improvement in final results, either clinical pregnancy, implantation, miscarriage or live birth rates when compared to the vaginal route of progesterone administration alone. Interestingly, a difference between our study and other similar available studies seems to rely on the ovarian stimulation protocol used, which may impact the luteal phase function. In fact, in our study, the patients underwent a mild stimulation protocol to recover a mean number of 4–5 oocytes per patient. Only two other similar studies, where the number of retrieved oocytes was reported are published by Tomic et al. (2011) [ 30 ] and De et al. (2023) [ 24 ]. They reported a higher number of recovered oocytes with a mean number of 7 and 9 oocytes per patient respectively. Accordingly, the higher level of circulating oestradiol due to high ovarian stimulation could be responsible for a major impact on luteal phase adequacy (32). Consequently, this may necessitate a more enhanced luteal phase supplementation. Furthermore, in Tomic et al. study [ 30 ], a short agonist protocol was used which can lead to a more compromised luteal phase due to not only the higher level of circulating oestradiol but also to a decreased pituitary secretion in the luteal phase [ 1 ], This peculiarity could in part explain the higher results obtained in term of pregnancy rates and LBR rate in combined supplementation groups in Tomic (2011) [ 30 ] and De (2023) [ 24 ]. Differently, the results of our study where two supplementation protocols (combined vs. vaginal alone) were used in a mild stimulation protocol and lower oestradiol levels show no statistical differences in terms of clinical outcomes. The efficiency of oral dydrogesterone alone compared with vaginal micronized progesterone in luteal phase support is still under evaluation. Barbosa et al. (2018) [ 22 ] in a meta-analysis of 7 papers concluded that oral dydrogesterone shows comparable clinical outcomes to vaginal micronized progesterone. The studies evaluated in this meta-analysis took into consideration different stimulation protocols (long GnRH agonist protocols and GnRH antagonist protocols) and different clinical settings (frozen embryo transfers and fresh embryo transfers). Contrary, in a more recent meta-analysis available Griesinger et al. (2020) [ 23 ] concluded that higher results in terms of pregnancy rate and live birth rate may be observed in women receiving oral dydrogesterone compared to vaginal micronized progesterone for luteal phase support. Also, this meta-analysis takes into account the papers evaluating different protocols of ovarian stimulation, different daily doses of oral dydrogesterone, and different starting days of progesterone administration. Given these results, the data reported appear contradictory and remain controversial.