Adiponectin levels in patients with bipolar disorder: A systematic review and meta-analysis.

Bipolar disorder (BD) is associated with high prevalence rates of obesity-related conditions and subclinical inflammation. Adiponectin is produced by adipose tissue and exerts anti-inflammatory activities. We aimed to perform a systematic review and meta-analysis of studies investigating adiponectin levels in BD patients and healthy controls. Electronic databases were searched from their inception until 15th Jan 2019. Random-effects models with the Hedges' g as the effect size (ES) estimate were used. We included 11 studies, representing 477 patients and 380 controls. Pooled data analysis revealed no significant differences in adiponectin levels between BD patients and controls (ES = 0.28, 95%CI: -0.34 - 0.90, p = 0.372). The levels of adiponectin were significantly higher during euthymia (ES = 1.09, 95%CI: 0.03-2.16, p = 0.044). The levels of adiponectin in depressed patients were lower, but this result did not reach statistical significance (ES = -0.90, 95%CI: -1.85 - 0.05, p = 0.063). Due to low number of studies, the subgroup analysis of manic patients was not performed; however, a severity of manic symptoms was not associated with the ES estimates. Longer illness duration and a higher percentage of BD type I (BD-I) patients were associated with higher ES estimates. A higher severity of depressive symptoms was associated with lower ES estimates. Heterogeneity was significant in all analyses. Results of the Egger's test were insignificant, showing no publication bias. Our results indicate that adiponectin might be a state marker of BD as it appears to be elevated in euthymia and decreased in depression. Illness progression and a diagnosis of BD-I might contribute to higher adiponectin levels.