Multimodal functional and structural neuroimaging investigation of major depressive disorder following treatment with duloxetine.

BACKGROUND: Longitudinal neuroimaging studies of major depressive disorder (MDD) have most commonly assessed the effects of antidepressants from the serotonin reuptake inhibitor class and usually reporting a single measure. Multimodal neuroimaging assessments were acquired from MDD patients during an acute depressive episode with serial measures during a 12-week treatment with the serotonin-norepinephrine reuptake inhibitor (SNRI) duloxetine. METHODS: Participants were medication-free MDD patients (n = 32; mean age 40.2 years) in an acute depressive episode and healthy controls matched for age, gender, and IQ (n = 25; mean age 38.8 years). MDD patients received treatment with duloxetine 60 mg daily for 12 weeks with an optional dose increase to 120 mg daily after 8 weeks. All participants had serial imaging at weeks 0, 1, 8, and 12 on a 3 Tesla magnetic resonance imaging (MRI) scanner. Neuroimaging tasks included emotional facial processing, negative attentional bias (emotional Stroop), resting state functional MRI and structural MRI. RESULTS: A significant group by time interaction was identified in the anterior default mode network in which MDD patients showed increased connectivity with treatment, while there were no significant changes in healthy participants. In the emotional Stroop task, increased posterior cingulate activation in MDD patients normalized following treatment. No significant group by time effects were observed for happy or sad facial processing, including in amygdala responsiveness, or in regional cerebral volumes. Reduced baseline resting state connectivity within the orbitofrontal component of the default mode network was predictive of clinical response. An early increase in hippocampal volume was predictive of clinical response. CONCLUSIONS: Baseline resting state functional connectivity was predictive of subsequent clinical response. Complementary effects of treatment were observed from the functional neuroimaging correlates of affective facial expressions, negative attentional bias, and resting state. No significant effects were observed in affective facial processing, while the interaction effect in negative attentional bias and individual group effects in resting state connectivity could be related to the SNRI class of antidepressant medication. The specificity of the observed effects to SNRI pharmacological treatments requires further investigation. TRIAL REGISTRATION: Registered at clinicaltrials.gov ( NCT01051466 ).

Effects of antidepressant therapy on neural components of verbal working memory in depression.

Impairments in verbal working memory are evident in major depression. Verbal working memory is comprised of the components of encoding, maintenance and retrieval. Whether the neural impairments are expressed in specific components, and how pharmacological therapy could modify the neural correlates are not well understood. We investigated the neural correlates of verbal working memory components in depression using the Sternberg task in a longitudinal magnetic resonance imaging study. Serial scans were acquired in 23 patients (mean age 39.8 years) during an acute depressive episode and following 12 weeks of pharmacological therapy with duloxetine and in 22 matched healthy controls (mean age 39.1 years) at the same time points. A significant group by time interaction was evident during the long maintenance phase, extending from the left middle frontal to the middle temporal and caudate regions, in which there was reduced activation in healthy participants at the follow -up scan but there were no changes in patients. Persistent neural engagement during the maintenance phase following treatment was revealed in major depression. The findings emphasize that impairments in verbal working memory may be initiated in the maintenance phase in major depression in order to sustain performance. Further research with larger sample size and using randomized, placebo-controlled double-blind studies are required to confirm our results.