Evaluation of Ovarian Stimulation Initiated From the Late Follicular Phase Using Human Menopausal Gonadotropin Alone in Normal-Ovulatory Women for Treatment of Infertility: A Retrospective Cohort Study

Early versus late follicular phase ovarian stimulation: a randomized controlled trial

RESEARCH QUESTION

Is late follicular phase stimulation as efficient as early follicular phase stimulation in a gonadotrophin-releasing hormone (GnRH) antagonist protocol in oocyte donors in terms of the number of oocytes.

DESIGN

In this open label, phase 3, non-inferiority, randomized controlled trial using a two-arm design with a 1:1 allocation ratio, 84 oocyte donors were allocated to the early follicular start group (control group, n = 41) or the late follicular start group (study group, n = 43). In the control group, women followed a fixed GnRH antagonist protocol with recombinant FSH (r-FSH) 225 IU. In the study group, r-FSH 225 IU was initiated in the late follicular phase. The primary outcome was the number of oocytes. The secondary outcomes were the number of mature oocytes, consumption of gonadotrophins and GnRH antagonist, and cost of medication.

RESULTS

The number of oocytes did not differ between the control group and the study group (intent-to-treat analysis 15.5 ± 11.0 versus 14.0 ± 10.7, P = 0.52; per-protocol analysis 18.2 ± 9.7 versus 18.8 ± 7.8, P = 0.62). In addition, the number of mature oocytes did not differ between the groups (14.1 ± 8.1 versus 12.7 ± 8.5, P = 0.48). The duration of stimulation was shorter in the control group (10.0 ± 1.4 versus 10.9 ± 1.5 days, P = 0.01). The total amount of r-FSH used was lower in the control group (2240.7 ± 313.9 IU versus 2453.9 ± 330.1 IU, P = 0.008). A GnRH antagonist was used for approximately 6 days in the control group, while a GnRH antagonist was only prescribed for one woman in the study group (6.0 ± 1.4 days versus 0.13±0.7 days, P < 0.001). There was a significant difference in the cost of medication per cycle between the groups (1147.9 ± 182.8€ in control group versus 979.9 ± 129.0€ in study group, P < 0.001).

CONCLUSIONS

Late follicular phase stimulation is as efficient as early follicular phase stimulation in terms of the number of oocytes.

The dark side of random-start ovarian stimulation: ovarian hyperstimulation syndrome due to inadvertent pregnancy

RESEARCH QUESTION

Can inadvertent pregnancies go unnoticed when initiating random-start ovarian stimulation (RSOS) despite monitoring?

DESIGN

Case series at a university-based tertiary care fertility clinic.

RESULTS

Between June 2022 and December 2023, two cases of undetected early pregnancy at the onset of RSOS were identified, both leading to severe ovarian hyperstimulation syndrome (OHSS) with hospitalization.

CONCLUSION

RSOS protocols add flexibility in fertility clinics when there is no intention of a fresh embryo transfer, but may be associated with insidious risk of OHSS. The authors advocate for comprehensive consultation and serial monitoring of human chorionic gonadotrophin during ovarian stimulation, while cautioning against over-reliance on baseline hormone concentrations when initiating RSOS. If the benefits of RSOS seem limited, healthcare providers should consider delaying ovarian stimulation to avert health, but also medicolegal and financial, complications.

Free your patients and yourself from day 2-3: start ovarian stimulation any time in freeze-all cycles

Ovarian stimulation for assisted reproductive technology is traditionally started in the early follicular phase. The essential rationale is to allow timely follicle growth and oocyte retrieval to ensure synchronization of the in-vitro cultured embryos with the receptive period of the endometrium in a fresh transfer cycle. In addition, conventional thought suggested that follicle recruitment happened only once, around menstruation. A deeper understanding of folliculogenesis, advances in cryobiology and an increasing proportion of freeze-all cycles provide a unique opportunity here. Experience from oncofertility patients as well as infertile women and oocyte donors who underwent ovarian stimulation in different phases of the menstrual cycle, dubbed 'random start' cycles, suggests that the number of oocytes collected and their reproductive potential do not depend on the time of starting ovarian stimulation, although the duration of stimulation and gonadotrophin consumption can vary slightly. It may be time to free both patients and clinics from the obsession with starting ovarian stimulation in the early follicular phase in planned freeze-all cycles. The flexibility provided by random start cycles is one aspect of individualizing treatment to patients' needs.

Clinical Value of Three-Dimensional Transvaginal Ultrasound in Diagnosis of Endometrial Receptivity and Ovarian Function in Patients with Infertility

Objective

A case-control study was conducted to explore the clinical value of three-dimensional transvaginal ultrasound in the diagnosis of endometrial receptivity (ER) and ovarian function in patients with infertility.

Methods

A total of 308 infertile women treated in our hospital from March 2020 to June 2021 were enrolled as the observation group, and another 300 women of childbearing age who underwent physical examination in the same period were enrolled as the control group. The clinical value of three-dimensional transvaginal ultrasound in ER in patients with infertility was analyzed by comparing the classification of endometrial and subendometrial blood perfusion, endometrial AUC value and Pi value, and subendometrial AUC value and Pi value. According to the number of oocytes obtained, the patients were assigned into the normal response group (182 cases, ≥5 oocytes) and the low response group (126 cases, <5 oocytes). The levels of some hormones, such as follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), and FSH/LH, were measured. Transvaginal ultrasonography was performed to detect ovarian volume (OV), antral follicle count (AFC), and peak flow rate of the ovarian interstitial artery (PSV). The peak of end-diastolic flow velocity (EDV) and other indexes were analyzed. The correlation between FSH level, FSH/LH, and ultrasound indexes was analyzed, and the ROC curve was established to analyze the value of transvaginal Doppler ultrasound in evaluating ovarian reserve function and predicting ovulation.

Results

There were significant differences in late proliferation type I and type III, ovulatory type II and type III (P < 0.05). There exhibited no significant difference in late proliferation type II, ovulation stage type I, and implantation window stage type I, type II, and type III (P > 0.05). Regarding the endometrial AUC and Pi values, the endometrial AUC and Pi values in the observation group were lower compared to the control group during late proliferation and ovulation (P < 0.05). There exhibited no significant difference in AUC and Pi (P > 0.05). Regarding the subintimal AUC and Pi values, the subintimal AUC and Pi values in the observation group were higher compared to the control group during late proliferation and ovulation (P < 0.05). There exhibited no significant difference in AUC and Pi during the implantation window (P > 0.05). There exhibited no significant difference in menarche age, age, body mass index, and menstrual cycle between the normal response group and the low response group (P > 0.05). The levels of EDV, OV, AFC, and PSV in the normal response group were higher compared to the low response group (P < 0.01). Compared with the low response group, the levels of FSH and FSH/LH in the normal response group were lower, but the levels of LH and E2 in the normal response group were higher (P < 0.05). The results of correlation analysis of FSH, FSH/LH, and ultrasound parameters between the normal response group and the low response group indicated that FSH was negatively correlated with E2, EDV, OV, AFC, and PSV in 308 infertile women (r = −0.817, -0.846, -0.707, -0.845, -0.911, P < 0.01), but it was positively correlated with FSH/LH (r = 0.714, P < 0.01). The ultrasound parameters of ovarian reserve function in the normal response group and the low response group were compared with the indexes that predicted ovulation. The results of ROC curve analysis indicated that the cutoff values of EDV, OV, AFC, and PSV were 4.141, 3.726, 4.106, and 13.944, respectively, the specificity of each index was higher than 90.00%, and the sensitivity was higher than 80.00% except PSV.

Conclusion

Transvaginal ultrasound can not only accurately evaluate the ER of infertile women but also directly observe follicular development and monitor ovulation, which is of high value in evaluating ovarian reserve function and predicting ovulation.

The euploid blastocysts obtained after luteal phase stimulation show the same clinical, obstetric and perinatal outcomes as follicular phase stimulation-derived ones: a multicenter study

STUDY QUESTION

Are the reproductive outcomes (clinical, obstetric and perinatal) different between follicular phase stimulation (FPS)- and luteal phase stimulation (LPS)-derived euploid blastocysts?

SUMMARY ANSWER

No difference was observed between FPS- and LPS-derived euploid blastocysts after vitrified-warmed single embryo transfer (SET).

WHAT IS KNOWN ALREADY

Technical improvements in IVF allow the implementation non-conventional controlled ovarian stimulation (COS) protocols for oncologic and poor prognosis patients. One of these protocols begins LPS 5 days after FPS is ended (DuoStim). Although, several studies have reported similar embryological outcomes (e.g. fertilization, blastulation, euploidy) between FPS- and LPS-derived cohort of oocytes, information on the reproductive (clinical, obstetric and perinatal) outcomes of LPS-derived blastocysts is limited to small and retrospective studies.

STUDY DESIGN, SIZE, DURATION

Multicenter study conducted between October 2015 and March 2019 including all vitrified-warmed euploid single blastocyst transfers after DuoStim. Only first transfers of good quality blastocysts (≥BB according to Gardner and Schoolcraft's classification) were included. If euploid blastocysts obtained after both FPS and LPS were available the embryo to transfer was chosen blindly. The primary outcome was the live birth rate (LBR) per vitrified-warmed single euploid blastocyst transfer in the two groups. To achieve 80% power (α = 0.05) to rule-out a 15% difference in the LBR, a total of 366 first transfers were required. Every other clinical, as well as obstetric and perinatal outcomes, were recorded.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Throughout the study period, 827 patients concluded a DuoStim cycle and among them, 339 did not identify any transferable blastocyst, 145 had an euploid blastocyst after FPS, 186 after LPS and 157 after both FPS and LPS. Fifty transfers of poor quality euploid blastocysts were excluded and 49 patients did not undergo an embryo transfer during the study period. Thus, 389 patients had a vitrified-warmed SET of a good quality euploid blastocyst (182 after FPS and 207 after LPS). For 126 cases (32%) where both FPS- and LPS-derived good quality blastocysts were available, the embryo transferred was chosen blindly with a 'True Random Number Generator' function where '0' stood for FPS-derived euploid blastocysts and '1' for LPS-derived ones (n = 70 and 56, respectively) on the website random.org. All embryos were obtained with the same ovarian stimulation protocol in FPS and LPS (GnRH antagonist protocol with fixed dose of rec-FSH plus rec-LH and GnRH-agonist trigger), culture conditions (continuous culture in a humidified atmosphere with 37°C, 6% CO2 and 5% O2) and laboratory protocols (ICSI, trophectoderm biopsy in Day 5-7 without assisted hatching in Day 3, vitrification and comprehensive chromosome testing). The women whose embryos were included had similar age (FPS: 38.5 ± 3.1 and LPS: 38.5 ± 3.2 years), prevalence of male factor, antral follicle count, basal hormonal characteristics, main cause of infertility and previous reproductive history (i.e. previous live births, miscarriages and implantation failures) whether the embryo came from FPS or LPS. All transfers were conducted after warming in an artificial cycle. The blastocysts transferred after FPS and LPS were similar in terms of day of full-development and morphological quality.

MAIN RESULTS AND THE ROLE OF CHANCE

The positive pregnancy test rates for FPS- and LPS-derived euploid blastocysts were 57% and 62%, biochemical pregnancy loss rates were 10% and 8%, miscarriage rates were 15% and 14% and LBRs were 44% (n = 80/182, 95% CI 37-51%) and 49% (n = 102/207, 95% CI 42-56%; P = 0.3), respectively. The overall odds ratio for live birth (LPS vs FPS (reference)) adjusted for day of blastocyst development and quality, was 1.3, 95% CI 0.8-2.0, P = 0.2. Among patients with euploid blastocysts obtained following both FPS and LPS, the LBRs were also similar (53% (n = 37/70, 95% CI 41-65%) and 48% (n = 27/56, 95% CI 35-62%) respectively; P = 0.7). Gestational issues were experienced by 7.5% of pregnant women after FPS- and 10% of women following LPS-derived euploid single blastocyst transfer. Perinatal issues were reported in 5% and 0% of the FPS- and LPS-derived newborns, respectively. The gestational weeks and birthweight were similar in the two groups. A 5% pre-term delivery rate was reported in both groups. A low birthweight was registered in 2.5% and 5% of the newborns, while 4% and 7% showed high birthweight, in FPS- and LPS-derived euploid blastocyst, respectively. Encompassing the 81 FPS-derived newborns, a total of 9% were small and 11% large for gestational age. Among the 102 LPS-derived newborns, 8% were small and 6% large for gestational age. No significant difference was reported for all these comparisons.

LIMITATIONS, REASONS FOR CAUTION

The LPS-derived blastocysts were all obtained after FPS in a DuoStim protocol. Therefore, studies are required with LPS-only, late-FPS and random start approaches. The study is powered to assess differences in the LBR per embryo transfer, therefore obstetric and perinatal outcomes should be considered observational. Although prospective, the study was not registered.

WIDER IMPLICATIONS OF THE FINDINGS

This study represents a further backing of the safety of non-conventional COS protocols. Therefore, LPS after FPS (DuoStim protocol) is confirmed a feasible and efficient approach also from clinical, obstetric and perinatal perspectives, targeted at patients who need to reach the transfer of an euploid blastocyst in the shortest timeframe possible due to reasons such as cancer, advanced maternal age and/or reduced ovarian reserve and poor ovarian response.

STUDY FUNDING/COMPETING INTEREST(S)

None.

TRIAL REGISTRATION NUMBER

N/A.

Premature ovarian insufficiency patients with viable embryos derived from autologous oocytes through repeated oocyte retrievals could obtain reasonable cumulative pregnancy outcomes following frozen-embryo transfer

Background

Women with premature ovarian insufficiency (POI) are often discouraged from using autologous oocytes; however, some patients have a strong desire to be genetically linked to their offspring. In the present study, we aimed to estimate cumulative pregnancy outcomes following frozen-embryo transfer (FET) in POI patients who could obtain viable embryos with their eggs during in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) treatments.

Methods

In this matched-retrospective cohort study, only patients undergoing IVF/ICSI treatments with a freeze-all strategy were screened, and 103 POI patients were matched with 515 normal controls in terms of the same number of viable embryos obtained at the same age. The primary outcome was the cumulative clinical pregnancy rate (CCPR) following FET per patient.

Results

Patients with POI and normal ovarian reserve had comparable CCPRs of 62.14% (64/103) and 65.24% (336/515), respectively (P=0.547), and no statistical difference was found in the cumulative live-birth rate (CLBR) between the study group (43.69%) and the control group (53.01%). Based on binary logistic regression, the CCPR and CLBR showed no association with the type of ovarian function (POI or normal ovarian reserve). The number of embryos per transfer and the sum of all viable embryos per patient were positively associated with the CCPR and CLBR. The clinical pregnancy rate (CPR) per FET cycle was 38.17% for the study group and 52.1% for the control group, while the CPRs per oocyte retrieval cycle in the 2 groups were 11.25% and 69.9%, respectively, and both were statistically different (P<0.05). Moreover, POI patients had a lower implantation rate (27.8% vs. 37.94%) and a higher early miscarriage rate per transfer (26.76% vs. 15%) than patients in the control group (P<0.05).

Conclusions

Cumulative pregnancy outcomes following FET were reasonable for POI patients using viable embryos derived from autologous oocytes through repeated oocyte retrievals.

Mathematical Modeling and Simulation Provides Evidence for New Strategies of Ovarian Stimulation

New approaches to ovarian stimulation protocols, such as luteal start, random start or double stimulation, allow for flexibility in ovarian stimulation at different phases of the menstrual cycle. It has been proposed that the success of these methods is based on the continuous growth of multiple cohorts ("waves") of follicles throughout the menstrual cycle which leads to the availability of ovarian follicles for ovarian controlled stimulation at several time points. Though several preliminary studies have been published, their scientific evidence has not been considered as being strong enough to integrate these results into routine clinical practice. This work aims at adding further scientific evidence about the efficiency of variable-start protocols and underpinning the theory of follicular waves by using mathematical modeling and numerical simulations. For this purpose, we have modified and coupled two previously published models, one describing the time course of hormones and one describing competitive follicular growth in a normal menstrual cycle. The coupled model is used to test ovarian stimulation protocols in silico. Simulation results show the occurrence of follicles in a wave-like manner during a normal menstrual cycle and qualitatively predict the outcome of ovarian stimulation initiated at different time points of the menstrual cycle.

Late Follicular Phase Ovarian Stimulation Without Exogenous Pituitary Modulators

Introduction: A gonadotropin-releasing hormone antagonist is the most common modulator used to prevent the premature luteinizing hormone (LH) surge when ovarian stimulation was initiated in the late follicular phase. We aimed in this study to evaluate the feasibility of performing ovarian stimulation in the late follicular phase without the use of exogenous pituitary modulators. Methods: Data were retrospectively collected from 404 normo-ovulatory patients who underwent their first in vitro fertilization (IVF)/intracytoplasmic sperm injection (ICSI) treatment in our department. One hundred sixteen subjects in the study group received ovarian stimulation when a dominant follicular diameter of ≥ 10 mm was confirmed by transvaginal ultrasonography after menstrual cycle day 6, which entailed a daily injection of gonadotropin until the trigger day, while 288 subjects in the control group received ovarian stimulation in the early follicular phase under a progesterone protocol. The primary outcome was the number of mature oocytes. Results: There was no statistical difference in the number of mature oocytes between the two groups (9.67 ± 5.33 in the study group vs. 9.38 ± 5.15 in the control group, P = 0.693). No secondary LH surges in the study group and no premature LH surges in the control group were found during ovarian stimulation. The good-quality embryo rate per oocyte retrieved showed no significant difference between the two groups (35.22 vs. 35.91%, P = 0.665). The clinical pregnancy rate per transfer was 54.55% in the study group and 56.48% in the control group (P = 0.718), and the implantation rate was similar between the two groups (36.94 vs. 37.77%, P = 0.829). Conclusions: Our study revealed that late follicular phase ovarian stimulation could be performed without an exogenous pituitary modulator.