Premenstrual dysphoric disorder (PMDD) is defined as a depressive disorder in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (American Psychiatric Association [APA], 2013). Women with PMDD undergo diverse changes in mood, cognitive function, and appetite and experience disturbances in sleep patterns during the luteal phase of the menstrual cycle. The prevalence of PMDD ranges from 1% to 8% (Yonkers and Simoni, 2018). The characteristic symptoms of PMDD during the late-luteal (LL) phase and their alleviation after menstruation indicate a potential correlation between ovarian hormones, such as estrogen and progesterone, and the pathogenesis of PMDD (Rapkin and Akopians, 2012). Studies have suggested that PMDD exacerbates stress reactions, implying a potential dysregulation of the hypothalamic–pituitary–adrenal (HPA) axis (Beddig et al., 2019; Kiesner and Granger, 2016). Another review underscored the role of inflammatory reactions in women with PMDD (Sultana et al., 2022). Cortisol and brain-derived neurotrophic factor (BDNF) serve as markers of stress reactivity (Zorn et al., 2017) and inflammatory responses (Carniel & da Rocha, 2021), respectively. Therefore, a comprehensive investigation of the interplay between ovarian hormones, stress, and inflammatory markers may reveal the pathogenesis of PMDD.

Around the middle of the menstrual cycle, a surge in luteinizing hormone triggers ovulation. Following ovulation, the ruptured follicle transforms into the corpus luteum, marking the commencement of the luteal phase. The corpus luteum then becomes the primary source of progesterone secretion. Both estrogen and progesterone levels increase, peaking around the mid-luteal (ML) phase, after which they rapidly decline during the LL phase, playing a role in the onset of menstruation (Silberstein and Merriam, 2000). The onset of PMDD symptoms is often observed during the LL phase, which tend to diminish after menstruation (APA, 2013). Notably, spontaneous anovulation was demonstrated to alleviate PMDD symptoms (Hammarback et al., 1991). Furthermore, PMDD symptoms tend to remit when endogenous ovarian hormone secretion is suppressed by treatment with gonadotropin-releasing hormone (GnRH) agonists (Schmidt et al., 1998) or hormonal contraception (Yonkers et al., 2017). Therefore, menstruation or menstruation-related processes may trigger PMDD among women susceptible to dynamic variations in ovarian hormones (Halbreich, 2003).

Schmidt et al. (2017) demonstrated that the add-back of reproductive steroid hormones following GnRH agonist–induced ovarian suppression triggered PMDD symptoms in women with PMDD. However, these symptoms exhibited a time-limited nature, subsiding after 1 month of stable add-back. This experiment indicated that changes in postovulatory estrogen and progesterone levels might play a role in the precipitation of PMDD symptoms (Schweizer-Schubert et al., 2020). Another review study suggested that prolonged exposure to or withdrawal from high levels of progesterone may exacerbate PMDD symptoms (Sundström Poromaa et al., 2003).

Earlier animal models have suggested that progesterone withdrawal plays a role in the precipitation of depressive or anxiety symptoms in women with PMDD (Li et al., 2012; Smith et al., 2006). If we hypothesized that progesterone withdrawal contributes to PMDD symptoms, women experiencing a more pronounced decline in progesterone levels might also experience more severe PMDD symptoms during the LL phase. Accordingly, the present study evaluated the association between the severity of PMDD symptoms and a decrease in progesterone from the ML to LL phase.

Exposure to estrogen or progesterone has been suggested to be a contributor to PMDD symptoms (Sundström Poromaa et al., 2003). Segebladh et al. (2009) demonstrated that a combination of estradiol and progesterone add-back treatment was associated with more pronounced symptoms compared with estradiol-only treatment among women with PMDD. This finding implies that exposure to high doses of estrogen or progesterone, but not withdrawal from them, triggers PMDD symptoms. According to this implication, women who experience elevated estrogen or progesterone levels during the luteal phase might also experience more pronounced PMDD symptoms. Further, Redei & Freeman (1995) found that progesterone levels were associated with PMDD symptoms in a time-lagged manner. They claimed that ovulation-dependent increases in progesterone levels could contribute to PMDD symptoms. Given that the progesterone level peaks during the ML phase, the level of progesterone during this phase or the extent of its increase from the preovulation (PO) phase (the baseline for progesterone level; Fig. 1) to the ML phase might serve as an indicator of the degree of progesterone exposure during the luteal phase. Therefore, the levels of these indicated hormones should be evaluated during the PO (baseline), ML (highest concentration), and LL (lowest concentration) phases to determine potential dynamic correlations between hormone levels and PMDD symptoms.

Dysregulation of the HPA axis is associated with both depressive disorder and stress reactivity. Notably, PMDD is classified as a depressive disorder in the DSM-5 (APA, 2013). During the LL phase, women with PMDD experience relatively severe depressive symptoms (Ko et al., 2013). Additionally, they experience high stress levels in their daily lives and increased arousal of negative emotions in response to stressors during this phase (Beddig et al., 2019). Consequently, stress has been associated with cortisol response in women with PMDD (Beddig et al., 2019) because cortisol may be secreted in response to acute stress during the premenstrual phase. Furthermore, a blunted cortisol reaction was reported in individuals with PMDD in the Trier Social Stress Test (Huang et al., 2015). However, no HPA dysregulation was observed in adrenocorticotropic hormone and cortisol responses to dexamethasone or corticotrophin-releasing hormone among women with PMDD (Lee et al., 2012). Furthermore, one study reported that cortisol levels did not differ significantly between women with PMDD and controls without PMDD (Watanabe and Shirakawa, 2015), whereas another study reported the attenuation of cortisol-awakening responses among women with PMDD (Hou et al., 2019). Despite the inconsistent results reported previously, a recent review emphasized the importance of assessing symptom specificity, results of within-person analyses, and multiple cortisol parameters to comprehensively understand cortisol response in women with PMDD (Kiesner and Granger, 2016).

Notably, BDNF is crucial in modulating the cortisol response to stress (Linz et al., 2019). Additionally, BDNF was demonstrated to protect against stress pathophysiology (Buselli et al., 2019). One study documented the compensatory secretion of high levels of luteal BDNF among women with PMDD (Oral et al., 2015). However, fluctuations in stress and BDNF levels during the luteal phase, a critical period for developing PMDD symptoms, have not been adequately analyzed.

Vascular endothelial growth factor (VEGF) constitutes a family of secreted polypeptides and plays crucial roles in both vasculogenesis and angiogenesis (Holmes and Zachary, 2005; Melincovici et al., 2018). Moreover, VEGF plays a regulatory role in angiogenesis and vascular function in the human endometrium (Nardo, 2005). The levels of VEGF in lutein extracts were high in the early luteal and ML phases and then tended to decrease as the LL phase approached (Otani et al., 1999). Furthermore, VEGF has been proposed as a potential biomarker of depression, possibly exerting its effects through a mechanism related to neuroplasticity (Carvalho et al., 2015a, Carvalho et al., 2015b). PMDD is a depressive disorder characterized by fluctuating symptoms across the menstrual cycle. An evaluation of the association between fluctuations in VEGF levels and PMDD symptoms during the menstrual cycle could reveal the role of VEGF in the development of PMDD.

Based on previous intervention studies, the present study hypothesized that increased postovulatory progesterone or estrogen contributes to developing PMDD symptoms during the luteal phase (Schmidt et al., 2017; Segebladh et al., 2009). Accordingly, we investigated the dynamic changes in the levels of estrogen and progesterone among individuals with PMDD and the potential correlations between these changes and various PMDD symptoms occurring during the PO, ML, and LL phases. In addition, we hypothesized that cortisol levels become elevated among women with PMDD in response to stress (Beddig et al., 2019) and are positively correlated with PMDD symptoms. Furthermore, we postulated that BDNF and VEGF function as potential buffers against PMDD symptoms and are negatively correlated with such symptoms. Accordingly, we evaluated the levels of cortisol, BDNF, VEGF, and perceived stress across the three menstrual phases in women with MPDD to determine the dynamic associations between these hormones and PMDD symptoms.