Evidence-Based Guidelines for Management of Poor Ovarian Response in Assisted Reproduction

INTRODUCTION

Controlled ovarian stimulation (COS) is indeed a critical determinant of success in assisted reproductive technology (ART), with the number of retrieved mature oocytes serving as a key prognostic marker for live birth rates (LBRs) in in vitro fertilisation (IVF)/intra cytoplasmic sperm injection (ICSI) cycles.^[1,2] In a subset of women, poor ovarian response (POR) occurs when fewer than 3-5 follicles/oocytes are retrieved, leading to higher cancellation rates and reduced chances of conception, compromised embryo availability, and overall diminished ART success^.[2] The prevalence of POR is reported to range between 9-24% among women undergoing ART, highlighting its significance in reproductive medicine.^[3] Reports from the American Society for Reproductive Medicine (ASRM) and the Society for ART (SART) further highlight that POR contributes substantially to cycle cancellations, accounting for more than half of such cases. These statistics emphasise the burden POR places on patients and clinicians alike, both in terms of psychological distress and resource utilisation.^[4.5]

Efforts to optimise outcomes in women with POR have led to the exploration of diversified clinical approaches. These include modifying ovarian stimulation protocols through individualised gonadotropin dosing, varying pituitary suppression regimens, and employing alternative approaches such as mild or dual stimulation cycles.^[6,7] Adjunctive therapies, such as androgen supplementation, growth hormone administration, and antioxidant use, have also been investigated for their potential to enhance ovarian response or improve oocyte competence.^[6] Preimplantation genetic testing for aneuploidy (PGT-A) is increasingly used in IVF to select embryos with the correct number of chromosomes (euploid embryos) for transfer, aiming to increase implantation rates and reduce miscarriages. While it offers significant benefits, particularly for women of advanced maternal age, its integration into IVF treatment adds substantial complexity, cost, and procedural, legal, and ethical challenges.^[7]

Despite such advances, the management of POR remains challenging due to the intrinsic heterogeneity of the affected population. One of the fundamental issues has been the lack of uniform diagnostic guidelines. While the Bologna criteria provided a standardised definition, their clinical applicability has been limited by variability in patient characteristics and response profiles.^[8] In recognition of this limitation, the POSEIDON (patient-oriented strategies encompassing individualised oocyte number) classification was launched in 2016.^[9-12] This framework stratifies patients into subgroups based on age, ovarian reserve markers, and prior response to stimulation, thereby enabling more precise prognostication and the tailoring of individualised therapeutic strategies. Nevertheless, a universally accepted definition and management algorithm for POR continues to be lacking, leading to heterogeneity in clinical practice worldwide.^[13] Although the availability of multiple stimulation protocols and adjunctive therapeutic strategies, none has been conclusively established as the universally optimal approach. Consequently, clinicians are often required to individualise treatment decisions based on patient characteristics and clinical judgment.^[13,14] This emphasises the inherent complexity of managing POR and highlights the necessity for continued research, systematic evaluation, and consensus development to refine evidence-based practices in ART. Given these challenges, there is a pressing need for evidence-based, context-specific guidance to support clinicians in optimising treatment for women with POR. Recognising this gap, the Indian Fertility Society (IFS) initiated the development of structured, evidence-based guidelines. The primary aim of these guidelines is to synthesise current evidence and expert consensus into clear, pragmatic recommendations that can be applied in clinical practice. These guidelines address key aspects of POR, including diagnostic criteria, stimulation strategies, adjunctive therapies, and patient-centred outcome measures, while ensuring safety and feasibility. By providing a structured framework, the IFS guidelines are intended to harmonise practice patterns, enhance clinical decision-making, and ultimately improve reproductive outcomes for this challenging patient group.

MATERIAL AND METHODS

The IFS systematically developed evidence-based guidelines for the management of POR in ART. A panel of experts initiated the process by outlining the scope, key clinical questions, outcomes, and objectives. A multidisciplinary Guideline Development Group (GDG) was formed [Figure 1], and questions were framed using the PICO (population, intervention, comparator, outcome) approach. The key outcomes incorporated included measures of efficacy (e.g., cumulative LBR, miscarriage rate), safety (e.g., ovarian hyperstimulation syndrome [OHSS], multiple pregnancy), and patient-related outcomes (e.g., cycle cancellation rates, patient preferences).

Close

Figure 1: Guideline Development Group

Export to PPT

Close

Figure 2: PRISMA Flowchart

Export to PPT

A comprehensive literature search was performed across various databases such as PUBMED/MEDLINE, Cochrane Library, EMBASE, Web of Sciences, and Scopus databases, incorporating studies published up to October 2023. The search strategies were in accordance with the PICO framework and executed independently by two reviewers to ensure consistency using keywords. Eligible studies included meta-analyses, randomised controlled trials (RCTs), and observational studies, while case reports, non-English publications, and non-relevant study designs were excluded. Titles and abstracts underwent initial screening, followed by full-text reviews to assess relevance and evidence quality.

Quality of evidence was assessed by GDG by using ‘A Measurement Tool to Assess Systematic Reviews’ (AMSTAR)-2, and in RCTs and observational studies, a structured checklist was employed to evaluate biases (selection, performance, detection, and attrition). Evidence certainty was graded by using the GRADE (grading of recommendations, assessment, development, and evaluation) framework. The high (⊛⊛⊛⊛), moderate (⊛⊛⊛⊘), low (⊛⊛⊘⊘), or very low (⊛⊘⊘⊘)was based on study design, bias risk, effect size, and confounding factors. The evidence summaries for every question included the study design, population, comparators, interventions, outcomes, conclusions, quality assessment, and grading.

The evidence-based guidelines were formulated by the GDG using the GRADE methodology, which systematically assessed the strength and certainty of the available evidence. The process incorporated evaluation of the balance between benefits and potential harms, feasibility of implementation, acceptability to stakeholders, and resource considerations. Recommendations within the guidelines were categorised as either strong or conditional, reflecting both the quality of evidence and the consensus reached among GDG members and the stakeholders. Every recommendation had a detailed rationale that accounted for the trade-off between the desirable and undesirable effects, implications for health equity, and anticipated impact on resource utilisation.

The draft was presented to the IFS executive committee and then circulated for stakeholder review from February to May 2024, via the online platforms and also by direct invitations. Feedback was collated through the IFS website and by email, reviewed by GDG, and then incorporated into the final document, with a detailed summary recorded in the Stakeholder Review Report.

RESULTS

This developed guideline comprehensively addresses critical domains in the management of POR, with an emphasis on optimising ovarian stimulation, improving the clinical pregnancy rates (CPRs) and the live birth rates (LBRs), while also maintaining patient safety, adherence, and individualised treatment strategies. A total of 45 recommendations were developed in response to 37 predefined key PICO questions. Among these, 44 were evidence-based recommendations, out of which 32 were graded as strong recommendations and 12 as conditional recommendations, complemented by 1 good practice point (GPP).

With respect to the quality of evidence, 1 recommendation was supported by high-quality evidence, 6 were derived from moderate-quality evidence, 25 were based on low-quality evidence, 8 were informed by very low-quality evidence, and 4 lacked evidence with recommendations for further research in 4. This distribution suggests the existing evidence gaps in POR management and highlights the necessity for further high-quality studies in this field. The complete set of recommendations has been presented in [Table 1 and 2]. Additional supporting material, including detailed evidence profiles, rationale for recommendations, and the detailed version of these IFS guidelines, is accessible at: https://indianfertilitysociety.org/wp-content/uploads/2024/07/POR-guideline-print-version-2july24.pdf.

DISCUSSION

All recommendations presented in these guidelines are primarily formed by evidence from systematic reviews, meta-analyses, randomised controlled trials (RCTs), and well-conducted cohort studies. Among the various hormonal biomarkers evaluated, anti-Müllerian hormone (AMH) has consistently demonstrated the highest reliability for predicting POR. AMH measurement offers the advantage of cycle independence and has been shown to accurately predict both low (<4 oocytes) and high (>15 oocytes) ovarian responses.^[15-17] Similarly, antral follicle count (AFC) is recognised as a robust predictor of ovarian response in IVF cycles.^[18] Both AMH and AFC exhibit comparable predictive accuracy and clinical utility; however, AFC provides a more precise assessment of ovarian reserve than all other ultrasound-based parameters, such as ovarian volume.^[19-21]

Comparisons between the gonadotropin-releasing hormone (GnRH) agonist protocols and the antagonist protocols have been evaluated in patients with POR. Evidence from multiple meta-analyses indicates no statistically significant differences between the two approaches with respect to CPRs, ongoing pregnancy rates (OPRs), or the number of oocytes retrieved.^[22,23] While few studies suggest that long and short GnRH agonist regimens may yield equivalent or occasionally better outcomes compared with the antagonist protocols, the evidence remains inconclusive, necessitating further high-quality trials to validate these observations.^[24,25]

The impact of stimulation intensity and protocol modification has also been examined in POR patients. Available studies demonstrate no statistically significant difference in fresh LBRs when comparing the low-dose gonadotropin regimens, with/ without antagonist use, against the conventional stimulation protocols employing GnRH agonists or antagonists.^[26] Evidence of moderate certainty from RCTs and meta-analyses indicates that oral ovarian stimulation agents, such as Clomiphene citrate or Letrozole, when combined with the low-dose gonadotropins, achieve comparable CPRs, OPRs, cumulative LBRs, and fresh LBRs to standard stimulation regimens. This combination may therefore be considered a viable alternative in selected patients, although it is associated with fewer oocytes retrieved and higher cancellation rates in some studies.^[26,27]

At present, evidence remains insufficient to support the use of letrozole monotherapy as a substitute for conventional stimulation protocols in POR. Moreover, escalation of gonadotropin doses beyond conventional thresholds has not been shown to improve LBRs in expected poor responders, as demonstrated consistently across several RCTs.^[28-32]

An RCT conducted in women aged ≥35 years who underwent IVF demonstrated that human menopausal gonadotropin (hMG) and recombinant follicle-stimulating hormone (rFSH) achieved comparable LBRs per initiated cycle, despite a higher oocyte yield in the rFSH group.^[33] These findings were corroborated by a large retrospective cohort study in poor ovarian responders, which reported similar oocyte recovery rates, CPRs, LBRs, and cycle cancellation rates between the two treatments.^[34] Importantly, no cases of OHSS were reported in either group, underscoring the safety profile of both regimens. Collectively, these results support the use of either rFSH or hMG, enabling clinicians to adopt individualised stimulation strategies without compromising treatment efficacy or patient safety.

In women demonstrating suboptimal ovarian response to rFSH during long GnRH agonist cycles, an RCT reported that mid-cycle supplementation with hMG improved oocyte recovery and LBRs more effectively than either maintaining the current rFSH dose or increasing the rFSH dose.^[35] However, two additional RCTs yielded inconclusive results.^[36,37] Similarly, low-quality evidence from three observational studies in antagonist cycles produced inconsistent findings regarding the benefit of adding hMG compared with continuation of rFSH alone.^[38-40] Based on the available evidence, the mid-cycle addition of hMG in long agonist protocols may be considered for patients who are hyporesponsive to rFSH, although further confirmatory studies are required.

Evidence from a systematic review and meta-analysis demonstrated that co-treatment with recombinant luteinizing hormone (rLH) and rFSH significantly improved CPRs compared with rFSH monotherapy.^[41] Enhanced implantation rates and higher oocyte yields were also observed with combined therapy. Subsequent systematic reviews and individual studies consistently supported the benefit of rLH supplementation, particularly among poor ovarian responders, with improvements noted in CPRs and overall reproductive outcomes.^[42-44] Furthermore, an RCT reported a lower incidence of complete pregnancy failure when r-hFSH was combined with r-hLH, compared to r-hFSH alone, highlighting the superiority of combined gonadotropin therapy in optimising clinical outcomes for this challenging patient population.^[45]

RCTs have demonstrated that the initiation of rLH at either the early or mid-follicular phase yields similar outcomes in women with POR. In a study involving 202 patients, no significant difference was found in the number of oocytes obtained when rLH supplementation was commenced early versus mid-cycle in combination with rFSH.^[46] Similarly, another RCT found no variation in oocyte recovery between early and late initiation of rLH in POR patients undergoing IVF.^[47] Collectively, these findings suggest that there is no significant difference in oocyte retrieval rate with respect to the timing of rLH addition, supporting the flexibility of either strategy in clinical practice.

Clinical evidence also indicates that Corifollitropin alfa (CFA) achieves outcomes comparable to daily rFSH or hMG in poor responders and women aged 35-45 years undergoing IVF.^[48-54] Although no significant differences were observed in LBRs or pregnancy rates between CFA and conventional stimulation regimens, CFA offers the advantage of reduced injection frequency, thereby enhancing patient convenience and compliance.^[48,55] Additionally, several studies have evaluated dual triggering (GnRH agonist plus hCG) compared with conventional hCG triggering, demonstrating higher numbers of mature oocytes, improved fertilisation and implantation rates, and increased CPRs with dual triggers.^[56-60] However, other trials reported similar outcomes between the two strategies.^{[56,59,61,62]} Given the heterogeneity of available evidence, both dual trigger and hCG-only trigger are conditionally recommended for POR patients undergoing GnRH antagonist cycles.

The GDG also identified several interventions that should not be routinely implemented in clinical practice due to insufficient or low-quality evidence. Different interventions like genetic polymorphism testing, luteal-phase estrogen pretreatment, luteal-phase stimulation, modified natural cycles, adjuvant therapies with growth hormone, DHEA, testosterone, or glucocorticoids, complementary or alternative medicine interventions, follicular flushing, PGT-A, elective freeze-all strategies, intraovarian platelet-rich plasma (PRP), intraovarian stem-cell therapy, and in vitro activation of ovarian tissue showed inconsistent results. The lack of supportive evidence should not be interpreted as proof of ineffectiveness but rather as an indication of insufficient data. Well-designed, RCTs with good sample sizes, complemented by periodic evidence updates, remain essential to guide future recommendations in this evolving domain.

CONCLUSION

The management of POR in IVF remains challenging due to patient heterogeneity and the absence of universally standardised protocols. Serum AMH and AFC are validated as the most reliable predictors for identifying women at increased risk of POR. While multiple stimulation protocols are available, no single regimen demonstrates consistent superiority across all patient populations, reinforcing the need for individualised treatment strategies.

Emerging approaches such as using dual-trigger, combined rLH/rFSH regimens, and flexible rLH administration demonstrate potential benefits, whereas CFA offers a patient-friendly option with reduced injection burden and comparable efficacy to conventional gonadotropins. Collectively, these findings highlight the necessity of patient-tailored therapeutic strategies and further high-quality research to optimise clinical and reproductive outcomes in POR management.