Luteal-phase deficiency and diminished ovarian reserve: a narrative review of interactions and clinical implications

Abstract

Diminished ovarian reserve (DOR) and luteal phase defect (LPD) are common endocrine disorders affecting the fertility of women of reproductive age. Traditionally, these conditions have been considered and treated independently. However, clinical observations frequently reveal that patients with DOR also exhibit features of LPD, suggesting a potential pathophysiological link between them. This review aims to explore the interplay between DOR and LPD from a novel perspective by integrating epidemiological data, current diagnostic and therapeutic practices, and recent insights into molecular mechanisms. Special emphasis is placed on the role of hypothalamic–pituitary–gonadal (HPG) axis dysfunction, oxidative stress–inflammatory microenvironment imbalance, and key signaling pathways, such as PI3K/Akt/mTOR, in mediating their interaction. We propose a central hypothesis: LPD may not merely be a complication of DOR; rather, intrinsic features of LPD—such as insufficient or prematurely withdrawn progesterone secretion—may exert negative feedback on the HPG axis and exacerbate oxidative damage within the ovarian microenvironment, thereby actively contributing to the onset or progression of DOR. Based on this hypothesis, we further suggest that treating LPD—particularly through luteal phase support therapy—may have benefits beyond improving endometrial receptivity. Such interventions could potentially modulate the endocrine milieu at both the systemic and local ovarian levels, thereby playing a role in managing DOR and possibly improving antral follicle count (AFC). These insights open up new directions for future therapeutic strategies targeting DOR.

Introduction: clinical challenge and potential link between DOR and LPD

Infertility affects approximately 12.6 to 17.5% of couples of reproductive age worldwide [1], with diminished ovarian reserve (DOR) being one of the key contributors to female infertility. DOR is characterized by a reduction in the quantity and/or quality of remaining ovarian follicles, typically manifested by decreased levels of anti-Müllerian hormone (AMH), reduced antral follicle count (AFC), and elevated basal follicle-stimulating hormone (FSH) levels [2]. The prevalence of DOR has significantly increased with the trend of delayed childbearing [3, 4]. And DOR is associated with markedly reduced assisted reproductive technology (ART) success rates [3, 5]. DOR not only reduces the number of oocytes retrieved, pregnancy rates, and embryo quality but also increases the risk of embryo aneuploidy and recurrent miscarriage [6, 7].

Concurrently, luteal-phase defect (LPD) has historically been described in two ways: as inadequate endometrial response or inadequate corpus luteum function [8]. In this review, we primarily use the term LPD to refer to inadequate corpus luteum function characterized by suboptimal progesterone secretion or shortened progesterone exposure, which in turn compromises endometrial receptivity for implantation and early pregnancy maintenance [9, 10]. LPD not only impairs endometrial receptivity but also disrupts immune tolerance, potentially contributing to implantation failure and early pregnancy loss [11, 12]. The incidence of LPD in natural cycles ranges from 3 to 10%, but is notably higher in ovulation induction cycles, making it a possible cause of infertility [13].

Notably, there are no published studies reporting the prevalence of LPD in DOR patients. However, in clinical settings, patients with DOR not only exhibit reduced follicular recruitment during the follicular phase, but some studies also suggest a trend toward shortened luteal phases. For example, Pfister observed that women with DOR had a slightly higher proportion of short luteal phases compared with women with normal ovarian reserve; however, after adjusting for age, this association was no longer statistically significant, suggesting a possible trend rather than a definitive link [14]. In addition, Harris demonstrated that lower AMH levels were independently associated with shorter luteal phases, further suggesting a mechanistic link between diminished ovarian reserve and luteal function [15]. Moreover, Guzel reported that young women with early stage of diminished ovarian reserve exhibited significantly shorter menstrual cycles at initial diagnosis, possibly due to luteal phase shortening, and this finding persisted at one-year follow-up [16]. These data further support the notion that early decline in ovarian reserve may be clinically accompanied by shortened luteal phases. Nonetheless, available evidence remains limited and inconsistent, and further studies are warranted.

This raises a critical clinical question: are DOR and LPD merely two independent reproductive endocrine disorders, or do they represent interconnected phenomena with shared or reciprocal mechanisms? This review aims to synthesize current evidence and explore the complex interplay between DOR and LPD. We propose a new perspective: LPD may not simply be a consequence of DOR, but may itself actively contribute to the pathogenesis and progression of DOR through multiple pathways. Recognizing and understanding this potential link is crucial for optimizing diagnostic and therapeutic strategies for both conditions.

Diagnosis and current treatment strategies for DOR and LPD

The number of primordial follicles in women is finite and declines steadily with age. When the number of antral follicles in both ovaries falls markedly, it is indicative of DOR [17, 18]. Diagnosis of DOR primarily relies on patient age, ovarian response history, and ovarian reserve markers such as basal FSH levels, and the more sensitive indicators, AMH levels, and AFC. The Bologna criteria proposed by the European Society of Human Reproduction and Embryology (ESHRE) remain among widely adopted diagnostic standards [19–21].

Currently, treatment options for DOR are limited. Hormone replacement therapy (HRT) is commonly used to alleviate symptoms associated with hormonal insufficiency but remains controversial due to its ability to carries associated risks [22]. Emerging interventions, such as testosterone, DHEA, antioxidants like resveratrol, and traditional Chinese medicine therapies, have been explored for their potential to enhance ovarian responsiveness or delay the decline of ovarian reserve; however, their efficacy and underlying mechanisms require further investigation [21, 23, 24].

The diagnosis of LPD is more complex, requiring multidimensional assessment. The diagnostic approach is mid-luteal serum progesterone measurement, with levels below 10 ng/mL commonly suggesting insufficiency; however, repeated measurements are necessary to confirm the diagnosis [25]. Although endometrial biopsy was historically considered the gold standard for diagnosis, its invasive nature has limited its use [26]. Importantly, other underlying disorders must be ruled out when diagnosing LPD [25, 27]. The mainstay of LPD treatment is progesterone supplementation. Common formulations include vaginal suppositories, oral dydrogesterone, or intramuscular progesterone injections. In patients with concurrent follicular development defects, ovulation induction using clomiphene citrate or letrozole, and luteal-phase support with human chorionic gonadotropin (hCG) may be employed. Furthermore, coexisting endocrine disorders, such as hypothyroidism or hyperprolactinemia, should be addressed [28–30].

Despite the existence of distinct diagnostic and treatment pathways for DOR and LPD, viewing them as isolated conditions may overlook their intrinsic connections. Current therapeutic strategies often target a single disorder without fully considering their mutual influences, particularly the potential impact of LPD status on ovarian function in DOR patients.

Overlapping pathophysiological mechanisms and interactions between DOR and LPD

DOR and LPD intersect across multiple pathophysiological dimensions, suggesting a bidirectional relationship that may perpetuate a vicious cycle.

Dysfunction of the hypothalamic–pituitary–gonadal (HPG) axis and bidirectional feedback

Clinical and experimental evidence suggests a bidirectional interaction between DOR and LPD via disruption of the hypothalamic–pituitary–gonadal (HPG) axis. In DOR, the decline in follicle number reduces estradiol and inhibin B secretion, thereby altering GnRH pulsatility and the LH/FSH ratio, which impairs corpus luteum formation and predisposes to luteal insufficiency [31]. Conversely, in women with LPD, premature withdrawal or insufficient secretion of progesterone fails to provide adequate negative feedback to the hypothalamus and pituitary. This results in disturbed GnRH/LH/FSH rhythms and may impair subsequent cycle dynamics. In assisted reproduction, inadequate luteal-phase progesterone is linked to suboptimal cycle outcomes and therefore necessitates enhanced luteal support [32].

Beyond impaired recruitment, progesterone deficiency may also compromise the follicular microenvironment. Abnormal gonadotropin rhythms weaken granulosa cell support and reduce AMH secretion, while impaired follicle survival further aggravates the decline of ovarian reserve [33].

The kisspeptin–GnRH pathway provides additional mechanistic insight. Kisspeptin is a potent stimulator of GnRH release, and studies have shown dynamic changes in endogenous kisspeptin secretion across the menstrual cycle, consistent with regulation by ovarian steroids including progesterone [34]. In mice, selective deletion of progesterone receptors in kisspeptin neurons leads to disrupted estrous cyclicity, progressive fertility decline, and a marked reduction or absence of corpora lutea formation [35], although other data suggest that nuclear PGR loss in these neurons does not completely abolish feedback [36]. While these experimental findings are based on animal models, they offer valuable mechanistic insight that may inform hypotheses regarding human reproductive physiology though direct extrapolation remains uncertain. In humans, exogenous kisspeptin administration can acutely stimulate LH pulsatility [37].

Taken together, while direct clinical evidence in women with LPD is lacking, converging findings from menstrual cycle observations, animal models, and interventional studies suggest that progesterone–kisspeptin signaling may contribute to GnRH/LH dysregulation and impaired follicular development, warranting further investigation (Fig. 1).

Fig. 1.

Role of HPG axis in DOR and LPD

Oxidative stress and inflammation: common disruptors of the microenvironment

Excessive production of reactive oxygen species (ROS) or impaired ROS clearance, leading to oxidative stress and chronic low-grade inflammation, is key shared pathological mechanisms underlying both DOR and LPD [38].

In DOR, elevated ROS expression, largely mediated by NF-κB activation, directly impairs mitochondrial function in oocytes, induces granulosa cell apoptosis, accelerates follicular atresia, and promotes depletion of the ovarian reserve [33, 39–41].

Similarly, oxidative stress is a core pathogenic factor in LPD. Inadequate luteal blood perfusion can cause ischemia–reperfusion injury, resulting in massive ROS production, which suppresses the expression of steroidogenic acute regulatory protein (StAR) and reduces progesterone synthesis. Concurrently, decreased activity of antioxidant enzymes, such as superoxide dismutase (SOD) and glutathione peroxidase (GPx), diminishes ROS scavenging capacity, rendering luteal cells more vulnerable to oxidative damage and premature apoptosis [42, 43].

Recent studies further demonstrate that vascular regression and impaired angiogenesis within the corpus luteum are strongly associated with hypoxia-driven ROS accumulation and subsequent luteal insufficiency [44]. Likewise, novel imaging has revealed that abnormal endometrial microvascular perfusion around the mid-luteal phase may predispose to local hypoxia and oxidative stress [45]. These findings reinforce the concept that perfusion-related mechanisms and oxidative stress are complementary rather than contradictory as both converge on progesterone deficiency as the central driver of LPD [44, 46].

Furthermore, declining hormone levels during the late luteal phase weaken the antioxidant capacity of the Nrf2 pathway while activating inflammatory signals, such as the COX-2/PGE2 axis, thus establishing a vicious cycle of "low hormone levels–high oxidative stress–strong inflammation" [47]. Clinical data, although limited, support this model. Doppler ultrasound and other imaging modalities have reported altered—often reduced or higher-impedance—luteal and endometrial blood flow in women undergoing infertility evaluation [48, 49]. Moreover, follicular fluid studies in women with DOR have consistently revealed elevated oxidative and inflammatory markers [33].

Importantly, this oxidative stress-inflammation state may form a self-amplifying vicious cycle between DOR and LPD. The ovarian dysfunction associated with DOR often entails aggravated oxidative stress, while persistent oxidative and inflammatory microenvironments under LPD conditions not only impair luteal function but may also further damage the ovarian stroma, compromising the survival and development of remaining follicles, and thereby exacerbating DOR progression (Fig. 2).

Fig. 2.

Role of oxidative stress and inflammation in DOR and LPD

PI3K/Akt/mTOR signaling pathway: a dysregulated regulatory hub

The PI3K/Akt/mTOR pathway plays a critical, albeit contrasting, role in follicle activation and luteal function.

In DOR, the loss of phosphatase and tensin homolog (PTEN) function leads to hyperactivation of PI3K/Akt signaling, resulting in the phosphorylation and inactivation of FOXO3a. This promotes premature and excessive activation of primordial follicles, accelerating depletion of the ovarian reserve [50, 51]. While these mechanistic findings are largely based on experimental and animal models, they offer valuable insight that may inform hypotheses regarding human follicular activation and depletion though their direct relevance to clinical physiology remains uncertain.

Hyper-activation of PI3K can also overstimulate mTORC1, enhancing ribosomal RNA synthesis and protein translation, thus further driving primordial follicle activation. Simultaneously, mTORC1 over-activity reduces autophagy pathways and accelerates oxidative damage, leading to increased follicular apoptosis (Fig. 3).

Fig. 3.

Role of PI3K/Akt in DOR

Conversely, in LPD, the PI3K/Akt/mTOR pathway appears to be suppressed (Fig. 4). Proper activation of this pathway is essential for granulosa cell proliferation, luteinization, and subsequent progesterone synthesis. Suppression of PI3K/Akt/mTOR signaling leads to reduced cyclin D expression, impairing cell cycle progression, while diminished mTORC1 activity reduces progesterone synthesis, ultimately resulting in defective luteal development [52].

Fig. 4.

Role of PI3K/Akt in LPD

Although the direction of pathway dysregulation differs, hyper-activation in DOR versus suppression in LPD, this underscores the sensitivity of the PI3K/Akt/mTOR axis as a central regulatory node. The contrasting dysregulation highlights the complex imbalance of ovarian function at different stages and suggests a deeper molecular connection between DOR and LPD.

Other potential linking factors

Vitamin A metabolism may also participate in ovarian function regulation by modulating antioxidant capacity and steroidogenesis (Fig. 5). The active form of vitamin A, retinoic acid, exerts antioxidant effects that can protect ovarian function. In DOR, retinoic acid has been shown, primarily in experimental studies, to alleviate oxidative stress and reduce follicular damage. Moreover, the correlation between retinol-binding protein 4 (RBP4), oxidative stress markers, and AMH levels suggests that vitamin A may help preserve ovarian reserve by maintaining antioxidant defenses [53].

Fig. 5.

Role of Vitamin A in DOR and LPD

In the context of LPD, vitamin A influences progesterone production by regulating genes related to steroid hormone biosynthesis; however, clinical studies have not consistently confirmed a direct link between vitamin A deficiency and LPD, indicating that vitamin A might not be a primary pathogenic factor in LPD [54, 55].

In summary, while vitamin A metabolism may have potential impacts on the pathogenesis of both DOR and LPD, the specific mechanisms remain to be fully elucidated. Appropriate vitamin A supplementation could support ovarian health, but should be evaluated comprehensively alongside other reproductive regulatory factors.

Clinical implications and a new perspective: could treating LPD become a therapeutic strategy for improving DOR?

Drawing upon the aforementioned mechanistic insights, DOR and LPD should not be viewed as isolated reproductive disorders. Rather, they are intricately connected and mutually influential through multiple pathways, including HPG axis dysregulation, oxidative stress and inflammation, and disordered signaling pathways. This gives rise to a clinically significant hypothesis: LPD—particularly characterized by insufficient progesterone secretion—may exert persistent negative regulatory effects on the residual follicular pool by destabilizing the HPG axis and perpetuating an oxidative microenvironment, thereby accelerating or exacerbating DOR manifestations.

From this perspective, we propose a potentially valuable therapeutic strategy that merits further exploration: luteal phase support in DOR patients with coexisting LPD. At present, no clinical studies have directly demonstrated that luteal phase support can improve ovarian reserve markers such as AMH, AFC, or FSH in women with DOR. However, progesterone supplementation could offer dual benefits—not only correcting luteal insufficiency and improving endometrial receptivity for potential conception, but more importantly, stabilizing HPG axis feedback, mitigating endocrine disruption associated with low progesterone levels, and possibly reducing local ovarian oxidative stress. These changes may indirectly support follicular development and partially improve ovarian reserve markers [56].

Conclusion and future directions

DOR and LPD represent two major challenges in the field of female reproductive health. The interplay between these conditions appears to be far more complex than previously recognized. This review elucidates their potential interactions from the perspectives of HPG axis regulation, oxidative stress–inflammation dynamics, and shared molecular signaling pathways. Our central hypothesis posits that LPD may not merely be a comorbidity or symptom of DOR but rather an active contributor to its progression.

This emerging perspective encourages clinicians to adopt a more comprehensive evaluation of luteal function when managing patients with DOR. Future research should prioritize the following directions: first, elucidating the molecular mechanisms and signaling networks underlying the interaction between DOR and LPD. Second, conducting prospective clinical trials to assess the long-term impact of luteal phase support on ovarian reserve markers, follicular development, and pregnancy outcomes in DOR patients. Third, exploring individualized, molecular subtype-based treatment strategies that integrate interventions targeting shared pathological pathways, such as oxidative stress and inflammation.

In summary, recognizing LPD management as an integral component of DOR treatment may open new therapeutic avenues and improve reproductive outcomes for affected women. Progesterone supplementation or other strategies aimed at correcting luteal dysfunction may in turn support follicular-phase ovarian function, offering new hope to women struggling with DOR—a hypothesis that deserves deeper scientific investigation.

Author contributions

MG: project development and Manuscript editing; CZ: Data collection and Manuscript writing. All authors read and approved the final manuscript.

Funding

The authors did not receive support from any organization for the submitted work.

Data availability

Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical approval and consent to participate

Not applicable.

Consent for publication

Not applicable.