Diminished ovarian reserve (DOR) poses a significant challenge in the field of reproductive medicine [1]. Characterized by a reduction in both the quantity and quality of ovarian follicles, DOR leads to decreased fertility potential and is a major cause of infertility in women [2, 3]. Epidemiological data indicate a rising prevalence of DOR, reflecting a concerning trend not only in aging populations but also among younger women of reproductive age [4]. Although current evidence implicates multifactorial etiologies, including genetic predisposition, environmental exposures, and lifestyle influences, the precise molecular mechanisms underlying DOR pathogenesis remain elusive, thereby hindering the development of mechanism-based therapeutic interventions [5, 6].

With the advancement of assisted reproductive technologies (ART), an increasing number of DOR patients have achieved successful pregnancies through in vitro fertilization and embryo transfer (IVF-ET) [7, 8]. Nevertheless, optimizing ovarian response in patients with DOR undergoing controlled ovarian hyperstimulation (COS) remains a significant clinical challenge, often resulting in suboptimal follicular recruitment, cycle cancellations, and reduced live birth rates [9, 10]. Premature ovulation, poor ovarian response to stimulation, low oocyte yield, and a higher incidence of aneuploid embryos are common challenges that can significantly impact the success rates of IVF-ET. These challenges adversely impact not only clinical success rates but also impose profound psychological distress and significant economic burdens on affected individuals and their families. Given the limited efficacy of conventional therapeutic approaches, the development of novel, cost-effective, and patient-tailored ovulation induction strategies for DOR management represents a critical unmet need in reproductive medicine.

Recent advances in COS have led to the development of innovative protocols specifically designed to enhance treatment outcomes for patients with DOR [11]. Notably, progestin-primed ovarian stimulation (PPOS) and luteal-phase ovarian stimulation (LPOS) have emerged as promising alternatives to conventional COS approaches [12, 13]. These novel protocols were designed to overcome the inherent limitations of traditional stimulation methods by providing better regulation of the ovarian environment and effective prevention of premature luteinizing hormone (LH) surges. The PPOS protocol employs progestins to suppress the endogenous LH surge, thereby enabling more controlled follicular development [14], whereas LPOS leverages on the physiological characteristics of the luteal phase to initiate ovarian stimulation [15]. Although both protocols offer theoretical benefits, comparative data on their relative efficacy, cost-effectiveness, and clinical outcomes in patients with DOR remain limited. To address this critical knowledge gap, the present study employs a self-controlled design to conduct a comprehensive evaluation of PPOS versus LPOS, aiming to provide evidence-based recommendations for optimizing treatment strategies in this clinically challenging population.

Management of DOR poses substantial challenges in reproductive medicine, largely due to a reduced ovarian follicular pool and compromised oocyte quality, which collectively adversely affect ART outcomes [19, 20, 21]. In this comparative study, we systematically evaluated the efficacy of two ovarian stimulation protocols—PPOS and LPOS—in patients with DOR undergoing IVF-ET. Our findings yield clinically relevant insights into the comparative performance of these protocols, elucidating their respective advantages and limitations in this challenging patient population.

Notably, our data reveal that the PPOS protocol is associated with a significantly shorter stimulation duration and reduced total Gn requirements compared with the LPOS approach. This differential carries important clinical implications, directly affecting both treatment costs and patient tolerability. The reduced Gn consumption observed with PPOS may be explained by effective suppression of premature LH surges through MPA administration, thereby enabling more controlled follicular recruitment [14, 22, 23]. In contrast, the LPOS protocol’s higher Gn requirements likely reflect luteal-phase initiation, where elevated endogenous progesterone levels may induce relative gonadotropin resistance.

Interestingly, although the LPOS protocol showed a trend toward a higher day-3 high-quality embryo rate, the difference did not reach statistical significance. This suggests that both protocols can generate embryos with comparable developmental potential, likely through different mechanisms of follicular recruitment and maturation. Clinically, both protocols yielded similar pregnancy outcomes, including implantation, clinical pregnancy, and miscarriage rates. This similarity in clinical outcomes is noteworthy, given the distinct differences in Gn dosage and stimulation duration between the two protocols. The comparable pregnancy rates indicate that embryos derived from both PPOS and LPOS protocols possess adequate developmental competence to achieve implantation and clinical pregnancy in DOR patients. Nevertheless, the higher miscarriage rate observed in the LPOS group, although not statistically significant, warrants further investigation. Although the LPOS group demonstrated a higher miscarriage rate numerically, the difference did not reach statistical significance. This suggests that the observed variation may reflect random error or insufficient sample size rather than a genuine difference in protocol safety. Miscarriage in ART can be influenced by multiple factors, including embryo quality, endometrial receptivity, and patient-specific characteristics [24, 25, 26]. The trend toward a higher miscarriage rate in the LPOS group may reflect underlying differences in these factors that were not fully controlled for in this study.

Our findings indicate that both PPOS and LPOS are effective for women with DOR undergoing IVF, with PPOS associated with improved oocyte maturation and fertilization rates. These results are consistent with a recent meta-analysis reporting that PPOS improves clinical pregnancy rates and embryo quality while reducing gonadotropin consumption compared with other protocols. Nevertheless, while both studies suggest benefits of PPOS, large-scale, well-designed trials are still required to validate these findings and evaluate long-term outcomes [27].

Multivariate analysis confirmed that the LPOS protocol was independently associated with a higher MII oocyte rate, after adjustment for age, BMI, and ovarian reserve indicators. This supports the inference that LPOS may enhance both cytoplasmic and nuclear maturation. However, no statistically significant associations were found between protocol type and clinical pregnancy or live birth after adjustment, indicating that other patient-level or embryo-endometrial factors may play a predominant role in determining final reproductive outcomes. Nevertheless, the limited sample size constrains the precision of these estimates, particularly for low-incidence outcomes (e.g., live birth). Future prospective studies with larger cohorts and stratified randomization are needed to further validate these associations.

Although the LPOS protocol showed numerically higher implantation and clinical pregnancy rates, these differences were not statistically significant. Nonetheless, the direction and magnitude of these trends may indicate potential clinical relevance. Our post hoc power analysis revealed that the current sample size (n = 130 per group) was insufficient to detect an absolute 8–10% difference in pregnancy-related outcomes with adequate statistical power. Specifically, detecting a 10% absolute difference in clinical pregnancy rates with 80% power would require approximately 370 patients per group. Therefore, the lack of statistical significance in the present study may reflect an underpowered design rather than the absence of a true effect. Future large-scale, multicenter trials with adequate sample sizes are needed to validate these findings and to more definitively compare the reproductive outcomes of PPOS and LPOS protocols.

Although the self-controlled design employed in this study effectively minimized inter-subject variability, a critical methodological limitation lies in the non-randomized sequence of protocol administration. Patients received PPOS and LPOS in a non-random order determined by clinical decisions or logistical considerations, which may introduce first-protocol effect bias. For example, patients undergoing PPOS first may have exhibited a stronger ovarian response simply due to being earlier in their treatment course, whereas those receiving LPOS as a subsequent cycle could have experienced physiological fatigue, diminished follicular recruitment, or psychological stress stemming from prior suboptimal outcomes.

The lack of randomization in protocol order compromises the internal validity of intra-individual comparisons, particularly for cycle-dependent outcomes such as oocyte yield, embryo quality, and implantation potential. Additionally, cumulative effects from prior stimulation, such as residual suppression or hormonal milieu carryover, may confound the observed differences in response between PPOS and LPOS. To mitigate this source of bias, future studies should adopt a randomized crossover design, with patients prospectively assigned to receive either PPOS or LPOS first, followed by crossover to the alternative regimen. Such a design would enable balanced sequencing and appropriate washout intervals, thereby enhancing the robustness of within-subject comparisons and strengthening causal inferences regarding protocol efficacy.

The self-controlled study design employed here offers distinct methodological advantages, particularly by minimizing inter-subject variability and enabling direct comparison of treatment efficacy within the same patient cohort. However, this approach entails several challenges. The non-randomized sequence of protocol administration may influence outcomes, as ovarian response could be affected by treatment order due to factors including cumulative hormonal exposure or psychological effects from prior cycle results. Furthermore, attributing clinical outcomes (e.g., implantation and pregnancy rates) to specific protocols becomes inherently complex when both interventions are applied sequentially within the same patient.

Future research should prioritize larger, multicenter trials to confirm these findings and explore the mechanisms underlying driving the differences between PPOS and LPOS protocols. Additionally, studies incorporating patient-reported outcomes, including treatment satisfaction and quality of life, will provide a more comprehensive understanding of the clinical utility and patient acceptability of these protocols.

In this study comparing the PPOS and LPOS protocols in patients with DOR, both regimens achieved comparable clinical outcomes in terms of implantation rate, clinical pregnancy rate, and live birth rate. Additionally, the PPOS protocol required a shorter duration and lower total gonadotropin dose, suggesting a more economical and patient-friendly stimulation strategy.

Importantly, despite the higher oocyte maturation rate in the LPOS group, live birth rates were nearly identical between the two protocols, underscoring that enhanced laboratory metrics do not necessarily translate into improved clinical outcomes. These findings highlight the need for individualized stimulation strategies that consider both laboratory efficiency and patient tolerance. Larger randomized controlled trials are warranted to validate these results and optimize protocol selection in this challenging patient population.

The datasets generated or analyzed during the current study are not publicly available due to patient privacy and institutional data protection policies, but are available from the corresponding author upon reasonable request. Requests for access to anonymized data will be reviewed and granted for non-commercial academic purposes in compliance with ethical and legal guidelines.

ML and QCL designed the study and performed literature research. ML conducted the experimental studies, clinical studies, and manuscript preparation. YLZ designed the research study. QCL defined the intellectual content. YJG conducted clinical studies. HLZ acquired data. JLW performed data analysis and statistical analysis. ML wrote the main manuscript text and reviewed the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Approval for the study was obtained from the Ethics Committee of The First Affiliated Hospital of Fujian Medical University [Date.2022.12; No.2022(657)]. All participants provided written informed consent prior to inclusion in the study, including authorization for clinical data collection and analysis.

Not applicable.

This project was funded by Fujian Medical University Sailing Foundation (NO.2022QH1095).

The authors declare no conflict of interest.

During the preparation of this work the authors used ChatGpt-3.5 in order to check spell and grammar. After using this tool, the authors reviewed and edited the content as needed and takes full responsibility for the content of the publication.