Abstract
PURPOSE: The relationship between women's reproductive traits and postpartum depression (PPD) has not been clarified. We reveal the association between genetically predicted modifiable women's reproductive traits and PPD using two-sample Mendelian randomization (MR). METHODS: We used genome-wide association studies (GWASs) to obtain instrumental variables (IVs) of 9 women's reproductive traits. Univariate and multivariate MR analyses were used to examine the association between traits and the risk of PPD (13,657 cases and 236,178 controls). The primary causal effect assessment employed the IVW method. Heterogeneity was assessed using Cochran's Q test. Multiple horizontal effects were assessed using the MR-PRESSO and MR-Egger intercept. Leave-one-out and LASSO regression analyses were used to check the robustness of the UVMR and MVMR results, respectively. RESULTS: In the UVMR result, genetic prediction showed that age at first sexual intercourse (AFS) (OR = 0.474, 95% CI 0.396-0.567; p = 4.6 × 10-16), age at first birth (AFB) (OR = 0.865, 95% CI 0.805-0.930; p = 8.02 × 10(-5)), and age at last live birth (ALLB) (OR = 0.296, 95% CI 0.138-0.636; p = 0.002) were significantly inversely associated with PPD, while a higher lifetime number of sexual partners (LNSP) (OR = 1.431, 95% CI 1.009-2.031; p = 0.045) and a greater number of spontaneous miscarriages (OR = 1.519, 95% CI 1.021-2.262; p = 0.039) are suggested to be associated with an increased risk of PPD. In the MVMR result, only AFB (OR = 0.804, 95% CI 0.661-0.978; p = 0.029) retained a direct causative relationship with PPD. CONCLUSION: The study indicates that AFB is a significant risk factor for PPD. Furthermore, the likelihood of developing PPD appears to decrease with increasing gestational age at the time of the first childbirth.