Urinary phytoestrogens and depression in perimenopausal US women: NHANES 2005–2008

Abstract

BackgroundFluctuating hormonal levels observed during the menopausal transition may increase vulnerability to depression in susceptible women. Thus, it is of interest to examine the effect of natural estrogens such as phytoestrogens on the risk of depression in perimenopausal women. MethodsOur analysis included 193 perimenopausal women of the National Health and Nutrition Survey (NHANES) 2005–2008 aged 45–55 years. Urinary concentrations of phytoestrogens (isoflavones and lignans) were measured by HPLC-APPI-MS/MS. Depression was assessed using the Patient Health Questionnaire-9 (PHQ-9). Logistic regression models examined the association of phytoestrogens concentrations (creatinine-standardized and log-transformed) with depression (yes/no). ResultsUnadjusted odds ratios (OR) of the associations between urinary phytoestrogen concentrations and perimenopausal depression were below 1; however, only lignans were significantly inversely associated with depression. The latter findings were not attenuated in multivariate analysis including age, race, body mass index, poverty income ratio, smoking, alcohol consumption, cancer, diabetes, and cardiovascular disease (lignans: OR=0.66; 95% confidence intervals (CI) 0.50–0.87, enterodiol: OR=0.63; 95% CI 0.51–0.78, enterolactone: OR=0.75; 95% CI 0.60–0.93). LimitationsOur cross-sectional study design does not allow for causal inferences. Because information to precisely assess perimenopausal symptoms was missing, we defined perimenopause based on women's age. ConclusionsLower lignans but not isoflavones concentrations were statistically significantly associated with an increased risk of depression in perimenopausal women. Because of medical risks associated with the use of hormone therapy, further investigation on the effect of lignans on the risk of depression in perimenopausal women is warranted.