Unraveling the hormonal causality in postpartum depression: findings from a two-sample Mendelian randomization study

Abstract

OBJECTIVES: Postpartum depression (PPD) profoundly affects maternal health and disrupts mother-infant bonding, with lasting effects on infant development. The causal roles of hormones that change most dramatically during the perinatal period, including estradiol, cortisol, and progesterone, in PPD pathogenesis remain unresolved. To address this, we conducted a two-sample Mendelian randomization (MR) study to assess bidirectional causal relationships between these hormones and PPD risk. MATERIAL AND METHODS: Genetic instruments were derived from genome-wide association studies: estradiol from UK Biobank (n = 163,985), cortisol (n = 3,366), and progesterone (n = 1,261) from German LIFE study, and PPD data from a meta-analysis of 18 European cohorts (17,339 cases/53,426 controls). We utilized seven methods to investigate the causal effects and conducted sensitivity analyses for robustness. RESULTS: MR analyses revealed that per one standard-deviation (SD) increase in genetically predicted estradiol levels was associated with a 20% higher risk of PPD (odds ratio [OR] = 1.20, 95% confidence interval [CI] = 1.03-1.40, false discovery rate [FDR] p = 0.0250). Conversely, per one SD increase in genetically predicted cortisol levels was linked to a 19% lower risk of PPD (OR = 0.81, 95% CI = 0.67-0.96, FDR p = 0.0250). No significant causal association was found between genetically predicted progesterone levels and PPD risk. In reverse MR, PPD showed no causal effect on estradiol, cortisol, or progesterone levels. CONCLUSIONS: Our study clarifies the distinct causal roles of estradiol and cortisol in PPD risk and may inform future mechanistic studies and clinical strategies for PPD prevention and treatment.