Abstract
Cumulative meta-analyses are used to evaluate the extent to which further studies are needed to confirm or refute a hypothesis. We used this approach to assess observational evidence on systemic inflammation in individuals with major depressive disorder. We identified 58 studies of four common inflammatory markers in a literature search of PubMed, Embase and PsychInfo databases in May 2014. Pooled data from the earliest eight studies already showed an association between interleukin-6 concentrations and major depression; 23 more recent studies confirmed this finding (d=0.54, p<0.0001). A significant association between C-reactive protein levels and major depression was noted after 14 studies and this did not change after addition of six more studies (d=0.47, p<0.0001). For these two inflammatory markers, there was moderate heterogeneity in study-specific estimates, subgroup differences were small, and publication bias appeared to be an unlikely explanation for the findings. Sensitivity analyses including only high-quality studies and subjects free of antidepressant medication further verified the associations. While there was a link between tumour necrosis factor-α levels and major depression (d=0.40, p=0.002), the cumulative effect remained uncertain due to the extensive heterogeneity in study-specific estimates and inconsistencies between subgroups. No evidence was found for the association between interleukin-1β levels and major depression (d=-0.05, p=0.86). In conclusion, this cumulative meta-analysis confirmed higher mean levels of interleukin-6 and C-reactive protein in patients with major depression compared to non-depressed controls. No consistent association between tumour necrosis factor-α, interleukin-1β and major depression was observed. Future studies should clarify the specific immune mechanisms involved as well as continue testing anti-inflammatory therapies in patients suffering from major depression.