Metabolites and depressive symptoms: Network- and longitudinal analyses from the Finnish Depression and Metabolic Syndrome in Adults (FDMSA) Study

Abstract

BackgroundDepression is associated with metabolic abnormalities linked to metabolic syndrome and tissue inflammation, but the interplay between metabolic markers and their association with subsequent depression is unknown. Therefore, we aimed to describe the network of metabolites and their prospective association with depressive symptoms. MethodsThe Finnish Depression and Metabolic Syndrome in Adults (FDMSA) cohort, originally a prospective case-control study, comprised a group with Beck Depression Inventory (BDI)-I scores ≥10 at baseline, and controls (n = 319, BDI-I < 10); mean (sd) follow-up time: 7.4 (0.7) years. Serum metabolic biomarkers were determined by proton nuclear magnetic resonance (NMR), and depressive symptoms sum-score by using the BDI-I. We examined the prospective associations between metabolites at baseline and BDI score at follow-up utilizing multivariate linear regression, parsimonious predictions models and network analysis. ResultsSome metabolites tended to be either negatively (e.g. histidine) or positively associated (e.g. glycoprotein acetylation, creatinine and triglycerides in very large high density lipoproteins [XL-HDL-TG]) with depressive symptoms. None of the associations were significant after correction for multiple testing. The network analysis suggested high correlation among the metabolites, but that none of the metabolites directly influenced subsequent depressive symptoms. LimitationsAlthough the sample size may be considered satisfactory in a prospective context, we cannot exclude the possibility that our study was underpowered. ConclusionsOur results suggest that the investigated metabolic biomarkers are not a driving force in the development of depressive symptoms. These findings should be confirmed in studies with larger samples and studies that account for the heterogeneity of depressive disorders.