Lithium monotherapy associated longitudinal effects on resting state brain networks in clinical treatment of bipolar disorder

Objectives Lithium is one of the most effective and specific treatments for bipolar disorder (BP), but the neural mechanisms by which lithium impacts symptoms remain unclear. Past research has been limited by a reliance on cross-sectional designs, which does not allow for identification of within-person changes due to lithium and has not examined communication between brain regions (ie, networks). In the present study, we prospectively investigated the lithium monotherapy associated effects in vivo on the brain connectome in medication-free BP patients. In particular, we examined the within-person impact of lithium treatment on connectome indices previously linked to mania and depression in bipolar disorder. Methods Thirty-nine medication-free subjects - 26 BP (13 (hypo)manic and 13 depressed) and 13 closely matched healthy controls (HC) - were included. fMRI data were obtained at 3 timepoints: baseline, after 2 weeks, and after 8 weeks (total of 117 scans: 78 BP and 39 HC scans). BP subjects were clinically treated with lithium for 8 weeks while HC were scanned at the same time points but not treated. Graph theory metrics and repeated measures GLM were used to analyze lithium treatment associated effects. Results Consistent with hypotheses, lithium treatment was associated with a normalizing effect on mania-related connectome indices. Furthermore, shifts in both mania- and depression-related connectome indices were proportional to symptom change. Finally, lithium treatment-associated impact on amygdala function differed depending on baseline mood. Conclusions Present findings provide deeper insight into the therapeutic neural mechanisms associated with lithium treatment.