796. Basal cortisol levels across the menstrual cycle in premenstrual dysphoric disorder and premenstrual syndrome: A systematic review and meta-analysis evaluating methodological heterogeneity

Abstract

Abstract Background Premenstrual dysphoric disorder (PMDD) and premenstrual syndrome (PMS) are characterized by cyclical affective and somatic symptoms. The hypothalamic-pituitary-adrenal (HPA) axis has been extensively investigated as a potential biological substrate for these conditions. However, the existing literature regarding basal cortisol levels remains highly inconsistent. These discrepancies may stem from methodological variations, particularly the imprecise definition of menstrual cycle phases and the use of different biological sampling matrices (e.g., serum versus saliva) across studies. Aims & Objectives This study aimed to systematically evaluate basal cortisol levels in individuals with PMDD/PMS compared to healthy controls. Furthermore, we sought to determine the extent to which methodological factors, specifically the biological specimen type and cycle phase classification, contribute to the heterogeneity of findings in the existing literature. Method Following PRISMA guidelines and PROSPERO registration, a systematic review and meta-analysis of longitudinal studies from 1984 to 2025 was conducted. To address prior methodological inconsistencies, raw menstrual cycle data from included studies were reclassified into a standardized 5-phase model. The analyses focused on contrasting the asymptomatic mid-to-late follicular phase with the symptomatic premenstrual (late luteal) phase. Subgroup analyses were performed to evaluate the moderating effects of specimen type and diagnostic category. Results Twenty-two studies comprising 23 independent samples met the inclusion criteria. The pooled meta-analysis indicated no significant differences in basal cortisol levels between PMDD/PMS patients and healthy controls during either the mid-to-late follicular or the late luteal phases. Notably, subgroup analysis by specimen type revealed a distinct pattern: while studies utilizing serum cortisol exhibited substantial heterogeneity, the analysis of salivary cortisol yielded zero heterogeneity (I2=0%) in both cycle phases, albeit within a smaller subset of studies. Leave-one-out sensitivity testing and publication bias analyses supported the stability of these null findings. Discussion & Conclusions The results of this meta-analysis do not support the hypothesis that basal cortisol alterations serve as a reliable biological marker for PMDD or PMS. The marked reduction in heterogeneity observed within the salivary cortisol subgroup suggests that prior inconsistencies in the literature may be largely attributable to methodological variance, specifically the choice of sampling matrix. These findings indicate that future psychoendocrine research in PMDD should shift focus from basal cortisol measurements toward dynamic assessments of HPA-axis reactivity or the investigation of alternative neurosteroid mechanisms.