Association of serum individual and mixed aldehydes with depressive symptoms in the general population: A machine learning study

Humans have many opportunities to be exposed to aldehydes which have potential mechanisms for causing depression. We aimed to explore the relationships between serum individual and mixed aldehydes with depressive symptoms in general population. The data was extracted from the National Health and Nutrition Examination Survey 2013–2014. Depressive symptoms were assessed by Patient Health Questionnaire-9. Weighted binomial logistic regression and Bayesian kernel machine regression (BKMR) model were used to explore the association of six individual aldehyde and mixed aldehydes with depressive symptoms, respectively. Sex stratification analysis and sensitivity analysis were conducted. A total of 701 participants were included. We found a positive association between the highest (Q4) versus lowest quartile (Q1) of butyraldehyde with depressive symptoms (OR: 2.86, 95 % CI: 1.22–6.68), and a negative association between the Q3 versus Q1 of benzaldehyde (0.21, 0.07–0.60) and isopentanaldehyde (0.28, 0.08–0.90) with depressive symptoms in multivariate-adjusted model. The mixed aldehydes were positively associated with depressive symptoms using BKMR model, and butyraldehyde and heptanaldehyde were the dominant aldehydes. Several aldehydes, such as butyraldehyde and benzaldehyde, interacted with each other in their effects on depressive symptoms. The results of gender stratification analysis showed that butyraldehyde was the major contributor to the total effect of aldehydes on depressive symptoms in males, while heptanaldehyde was the dominant aldehyde in females. Causality cannot be inferred in this cross-sectional study. Our study indicated that mixed aldehydes can increase the risk of depressive symptoms, of which butyraldehyde and heptanaldehyde were the major contributing aldehydes.