Illness characteristics and functional connectivity alterations across large-scale brain networks in bipolar disorder: the BipOld Study

Background

Bipolar disorder alters large-scale brain networks such as the default mode, salience, and frontoparietal control networks. Early illness characteristics and clinical course determinants influence these connection patterns, although how is unclear. These associations can help to clarify neuroprogressive mechanisms. Aims: To investigate associations between clinical measures of early illness course and resting-state functional connectivity across major brain networks in patients with bipolar disorder.

Methods

The present study is part of the BipOld Study, a prospective longitudinal cohort study that investigates neuroprogression and dementia risk in individuals aged 40 years and older with bipolar disorder and combines detailed retrospective data about the clinical course with multimodal neuroimaging and cognitive assessments. Resting-state functional connectivity analyses were done with CONN (22.v2407). Seed-to-voxel analyses were conducted with regions of interest from the default mode network (DMN) and salience network (SN). We examined functional connectivity and clinical factors (age at first episode, duration until maintenance treatment, total number of episodes, and psychotic symptom severity) using multiple linear regressions in the general linear model framework. We determined statistical significance at a voxel-wise level of p < 0.001 (uncorrected) and a cluster-level family-wise error correction of p < 0.05

Results

The current analysis included 19 participants (11 women (57.9%) and 8 men (42.1%)), with a mean age of 59.3 years (SD=9.7). Younger age at first episode was associated with reduced functional connectivity between the left DMN’s lateral parietal region and the right frontoparietal control network’s prefrontal cortex (t(15)=5.42, pFWE=0.005), as well as between the right DMN’s lateral parietal region and the left frontoparietal control network’s prefrontal cortex (t(15)=6.90, pFWE=0.002). The left SN’s rostral prefrontal cortex and the right frontoparietal control network’s prefrontal cortex both decreased (t(15)=8.20, pFWE=0.011). Longer duration until initiation of maintenance treatment was positively associated with functional connectivity between the left SN’s supramarginal gyrus and the left limbic network’s temporal pole (t(15)=5.83, pFWE=0.001). A higher total number of episodes was associated with decreased functional connectivity between the left anterior insula of the salience network (SN) and the cerebellar Crus II of the left frontoparietal control network (t(15)=11.77, pFWE=0.002) and between the medial prefrontal cortex of the default mode network (DMN) and the left sensorimotor network (t(15)=6.35, pFWE= Higher psychotic symptom intensity was linked with enhanced DMN connectivity, notably between the posterior cingulate cortex and inferior temporal cortex (t(15)=5.65, pFWE=0.020).

Conclusions

Our findings point to the early onset of illness, the delay in the initiation of maintenance treatment, and the cumulative burden of episodes as correlates of distinctive patterns of dysconnectivity in core large-scale brain networks in bipolar disorder. Reduced coupling of the default mode, salience, and frontoparietal control networks may indicate impaired cognitive control and self-referential processing, especially in those with earlier onset and higher illness burden. In contrast, increased within-network connectivity of the default mode network in relation to psychotic symptoms may indicate maladaptive network hyperintegration. The results support a hypothesis of network-level rearrangement associated with illness development and suggest that early and ongoing treatment may mitigate connection changes.

References

No references