A longitudinal functional magnetic resonance imaging study of verbal working memory in depression after antidepressant therapy

Background: Impairments in the neural circuitry of verbal working memory are evident in depression. Factors of task demand and depressive state might have significant effects on its functional neuroanatomy. Methods: Two groups underwent functional magnetic resonance imaging while performing a verbal working memory task of varying cognitive load (n-back). The patient group comprised 20 medication-free individuals in an acute episode of unipolar major depression and the control group comprised 20 healthy individuals. Scans were acquired at weeks 0 (baseline), 2, and 8. Patients received treatment with fluoxetine after the baseline scan. Cerebral activations were measured for mean overall activation as well as the linear and quadratic load-response activity with increasing task demand (1-, 2-, 3-back). Results: There were no significant differences in performance accuracy between groups. However, a main effect of group was observed in the load-response activity in frontal and posterior cortical regions within the verbal working memory network in which patients showed a greater load-response relative to control subjects. Group by time effects were revealed in the load-response activity in the caudate and thalamus. As a marker of treatment response, a lower linear load-response at baseline in the dorsal anterior cingulate, left middle frontal, and lateral temporal cortices was associated with an improved clinical outcome. Conclusions: Maintenance of performance accuracy inpatients was accompanied by a significant increase in the load-response activity in frontal and posterior cortical regions within the verbal working memory network. These data also provide further support for resilience of activity in the anterior cingulate as a predictor of treatment response in depression.