Abstract
BACKGROUND: Perinatal depression has been associated with unfavorable pregnancy and childhood development outcomes; however, no objective markers exist to identify perinatal mood disorders. We investigated whether metabolites in maternal urine during pregnancy can predict increased depressive symptoms in late pregnancy and postpartum among pregnant women at risk for perinatal depression. METHODS: We evaluated metabolomic markers in urine collected at 12-20 and 34-36 weeks' gestation. We analyzed 49 urinary metabolites using ion pairing reversed-phase liquid chromatography-mass spectrometry. Depressive symptom severity was identified using Beck Depression Inventory (BDI) scores from 105 participants at 12-20 and 34-36 weeks' gestation, and 6-8 weeks' postpartum. Mixed model repeated measures analysis evaluated associations between changes in maternal urinary metabolites and BDI scores across pregnancy. RESULTS: Increases in urinary xanthine and hypoxanthine were positively associated with increases in maternal depressive symptoms throughout pregnancy (p = 0.03 and 0.02, respectively). This finding did not persist after false discovery rate correction. None of the urinary metabolites examined were significantly associated with development of postpartum depressive symptoms. LIMITATIONS: This study is an exploratory secondary biologic sample analysis from a trial whose sample size was determined by a different primary outcome and expected effect size, which may have limited statistical power to detect associations between urinary metabolites, depressive symptoms, and mood trajectory over time. CONCLUSIONS: Increasing concentrations of xanthine and hypoxanthine were associated with increasing depressive symptoms throughout pregnancy. Further research is needed to evaluate the utility of these metabolic markers in identifying women at risk for perinatal depressive symptoms.