Abstract
BACKGROUND: The direction of the association between inflammation and depressive symptoms remains inconsistent. The objective of this study was to evaluate the temporal relationship between inflammation and depressive symptoms, and to assess the role of genetic factors on this association. METHODS: In this longitudinal cross-lagged twin difference study, we examined 166 (83 pairs) middle-aged male twins recruited from the Vietnam Era Twin Registry, who were assessed at baseline and after 7 years of follow-up. We assayed plasma levels of two inflammatory biomarkers, interleukin-6 (IL-6) and high sensitivity C-reactive protein (CRP) and measured depressive symptoms using the Beck Depression Inventory-II (BDI). To evaluate the direction of the association, we constructed multivariable mixed-effects regression models and calculated standardized beta-coefficients to compare the strength of the within-pair association for both pathways. We then conducted a stratified analysis by zygosity and assessed the associations in monozygotic and dizygotic twin pairs separately. RESULTS: The 166 twins were 95% white and had a mean (SD) age of 54 (3) years at baseline. The cross-lagged analysis showed significant and positive associations from visit 1 IL-6 to visit 2 BDI across all models (beta-coefficients ranging from 0.18 to 0.22). However, the opposite pathway (visit 1 BDI to visit 2 IL-6) was not significant after adjusting for confounding factors. In contrast, visit 1 BDI was significantly associated with visit 2 CRP in all models (beta-coefficients ranging from 0.23 to 0.33), while the opposite pathway (visit 1 CRP to visit 2 BDI) showed no significant association. When stratifying by zygosity, significant associations from IL-6 to depression were only seen in monozygotic twins, but associations from depression to CRP were more robust in dizygotic twins, which implies that genetic factors may play a role in this association. CONCLUSIONS: The association between inflammation and depression may be bidirectional. Elevated IL-6 levels are more likely to be a risk factor of depression rather than a consequence, while the opposite may be true for elevated CRP. The biological underpinnings of these bidirectional pathways need further evaluation.