Does brain structure differ between women with PMDD and healthy controls?

Up to 8 % of women world-wide are affected by premenstrual dysphoric disorder (PMDD), a condition in which patients experience impairing psychological, physiological, and cognitive symptoms in the days prior to menstruation (Halbreich et al., 2003). Despite the temporal association between symptom occurrence and menstrual cycle phase, absolute levels of ovarian hormones are similar in patients with PMDD and healthy controls. As endogenous ovarian hormones are powerful modulators of brain structure and function (e.g., for estradiol, see Ruehr et al., 2025), the altered sensitivity to normal hormonal fluctuations across the menstrual cycle in patients with PMDD (Schmidt et al., 2017) bears the potential for alterations in brain structure and function in this patient group. Recently, interest in the neuronal underpinnings of PMDD has increased, as demonstrated by a systematic review summarising 24 multimodal neuroimaging studies (Monteiro et al., 2024). Yet, it remains unclear whether changes in brain structure occur only in the symptomatic late luteal phase, or function as a trait-like characteristic of PMDD. Due to the limited number of studies looking at brain structure and function in PMDD, often limited to follicular versus luteal phase comparisons or single phase assessments, we will also consider neuroimaging studies across the menstrual cycle and in relation to ovarian hormones, to inform our hypotheses. Grey matter volume Findings on grey matter volume in patients with PMDD are heterogeneous. While some studies report reduced grey matter volume in patients, either at the whole-brain level (Dubol et al., 2022) or in the cerebellum and the cuneus (Stiernman et al., 2025), others suggest increased grey matter volume in the cerebellum (Berman et al., 2013). However, these studies are restricted to either group differences assessed at single timepoints (Dubol et al., 2022; Berman et al., 2013), or include phase comparisons without ovarian hormone analyses (Stiernman et al., 2025). In health, studies report grey matter volume modulation by cycle phase, as well as by endogenous ovarian hormones. For example, studies report an increased global grey matter volume (Hagemann et al., 2011), as well as increased inferior parietal lobe (Meeker et al., 2020) and bilateral hippocampal (Pletzer et al., 2018) grey matter volumes around the time of ovulation. However, while some studies find an underlying association between brain structure and estradiol (Pletzer et al., 2018, Lisofsky et al., 2015), others do not find underlying hormonal drivers (Hagemann et al., 2011). Grey matter density Findings on grey matter density in patients with PMDD are scarce. One study showed increased grey matter density in the hippocampal, yet decreased grey matter density in the parahippocampal cortex compared to healthy controls (Jeong et al., 2012), but tested participants only once during the luteal phase, without assessing ovarian hormones. In health, grey matter density has been positively associated with estradiol levels (e.g., Barth et al., 2016). Surface-based metrics Studies report thinner cortices in patients with PMDD compared to healthy controls (Dubol et al., 2022; Dubol et al., 2024). Findings regarding other surface-based metrics are mixed, suggesting either no difference in gyrification, sulcal depth, and cortical complexity during the late luteal phase between groups (Dubol et al., 2022), or report lower gyrification and decreased cortical folding in patients with PMDD independent of cycle phase (Dubol et al., 2024). In health, whole-brain cortical thickness has been positively associated with FSH, a hormone that is increasing before ovulation (Rizor et al., 2024). At a trend level, left grey matter thickness was observed in the periovulatory phase, compared to the follicular and luteal phases (Meeker et al., 2020). White matter microstructure Findings on white matter microstructure in patients with PMDD are scarce. One study found greater fractional anisotropy (FA) in the left uncinate fasciculus and higher right uncinate fasciculus volume in patients with PMDD in the luteal phase (Gu et al., 2022). Yet, this study exclusively assessed individuals during the luteal phase, without analysing ovarian hormones. In a dense-sampling study using a healthy individual, estradiol levels were positively correlated with FA in the bilateral hippocampus and inversely correlated with radial diffusivity (RD), suggesting estradiol-associated changes in hippocampal white matter microstructure (Barth et al., 2016). Further, while mean diffusivity of the fornix has been negatively associated with LH and estradiol levels (De Bondt et al., 2013), FA across widespread regions has been positively associated with LH and estradiol levels (Rizor et al., 2024). White matter volume Findings on white matter volume in patients with PMDD are scarce. One study found greater white matter volumes in the uncinate fasciculus, fornix and inferior fronto-occipital fasciculus across the menstrual cycle in patients with PMDD (Stenhammar et al., 2025). In healthy individuals, Meeker and colleagues (2020) observed greater white matter volume in the left inferior parietal lobule during the periovulatory and luteal phases compared to menstrual and follicular phases in healthy individuals, without any associations with sex hormone levels. Relationship between brain structure and (1) premenstrual symptoms, (2) immune markers, and (3) the cortisol awakening response Premenstrual symptom severity has been associated with grey matter volume (Dubol et al., 2022), but not with grey matter density (Jeong et al., 2012). In healthy individuals, overall premenstrual symptom severity was positively correlated with the volume of the left posterior cingulate cortex, while specific symptom domains have been associated with volumes of different regions (Marjanovic et al., 2025). In patients with PMDD, overall symptom severity correlated negatively with bilateral amygdala volume. Further, positive associations were found between the sulcal depth of the fusiform gyrus and overall premenstrual symptom severity, concentration, and energy loss, as well as between irritability and sulcal depth of the left posterior cingulate cortex (Dubol et al., 2022). Premenstrual symptom severity has also been positively correlated with FA in the right superior longitudinal fasciculus (Gu et al., 2022). While in healthy individuals, a negative correlation was found between hippocampal grey matter and IL6 concentrations (Marsland et al., 2008), the connection between grey matter structure and immune markers remains unexplored in PMDD. One single study showed a positive association between inflammatory TNFa levels and white matter microstructure, in specific with MD, AD, and RD in both patients with PMS and healthy controls (Duan et al., 2025). A smaller cortisol awakening response (CAR), measured by cortisol increase from awakening to 30-minutes post-awakening, has been associated with reduced grey matter volume in the perigenual anterior cingulate cortex (Boeringer et al., 2015). Moreover, a higher CAR, measured as maximum subtracted by mean morning cortisol concentration, has been associated with right relative to left cingulum fractional anisotropy, as well as the degree of fractional anisotropy asymmetry (Madsen et al., 2012). Summary Research looking at the brain structure in PMDD is still limited. The aim of this study is to investigate alterations in brain structure in patients with PMDD compared to healthy controls across the menstrual cycle. Based on previous work, hippocampal white matter constitutes our primary a priori region of interest, whereas whole-brain analyses are considered exploratory. Specifically, we will compare outcomes between the periovulatory phase and the symptomatic premenstrual phase, while examining associations with underlying endogenous ovarian hormone concentrations. As exploratory analyses, we will examine the relationships between brain structure and premenstrual symptoms, immune markers, and indices of the cortisol awakening response.

Reproduced under CC-BY.