A functional magnetic resonance imaging investigation of emotion processing in remitted major depressive disorder: towards the identification of neurobiological trait markers and clinical implications for treatment

Emotional dysregulation is a core feature of Major Depressive Disorder (MDD), manifested by perceptual biases in processing positive and negative emotional stimuli. In support, a constellation of neuroimaging studies have consistently reported abnormalities in brain areas during the processing of negative and positive emotional stimuli in acutely depressed patients during a major depressive episode. These areas, which are pivotal for bottom-up emotion generation processes and top-down cognitive control of emotion through reciprocal connections, include lateral prefrontal cortical (PFC) regions (dorsolateral PFC, ventrolateral PFC), medial PFC regions (orbitofrontal cortex) and subcortical regions (amygdala, ventral striatum). However despite considerable progress into identifying the neural circuitry involved in MDD, a neural circuitry marker for the MDD trait has not yet been defined. Studying individuals in remission from MDD who are euthymic and medication-free, allows for the examination of enduring abnormalities in the neural circuitry supporting emotion processing and its regulation that may represent trait markers of the illness. Therefore the aims of this thesis were to examine, by means of two functional magnetic resonance imaging (fMRI) studies, neural activation in key areas subserving emotion processing and higher-order cognitive control processes, to determine whether abnormalities in the circuit persist into remission and thus represent trait-like markers of MDD. Using blood oxygen level dependent (BOLD) fMRI, 19 remitted medication-free individuals with recurrent MDD (rMDD) (mean age 33.6 ± 13.64) and 20 healthy controls (HC) (mean age 35.8 ± 12.10) completed an implicit gender-labelling emotion processing task (study 1), and a working memory task with emotional distracter stimuli (study 2). In each study faces depicting negative (fearful), positive (happy) and neutral emotions were used. Brain imaging data were analyzed in SPM2 using a whole-brain and region-of-interest approach corrected for multiple comparisons using AlphaSim (PFWE<.05) for the following a priori hypothesized regions: DLPFC (BA 9/46), VLPFC (BA 45/47), OFC (BA 11), subgenual ACC (BA 25), ventral striatum and amygdala. Exploratory, post-hoc analyses were performed in each study, to examine the association between brain activation during the processing of negative and positive emotional stimuli and clinical variables in the rMDD group. Findings indicate altered left-sided medial (OFC) (t37= 3.85, P<0.001, PFWE<0.05) and dorsolateral prefrontal (DLPFC) (t37= 2.55, P=0.008, PFWE<0.05) cortical responses to negative emotional stimuli in the rMDD group compared to HC (study 1). Furthermore, during cognitive load, rMDD patients exhibited greater neural activity in DLPFC during performance of the working memory-load condition in the context of negative emotional distracters, (t37= 2.71, p=.005, PFWE<.05) but reduced activity in DLPFC (t37= 2.43, P=.01, PFWE<.05) and VLPFC (t37= 2.82, P=.01, PFWE<.05) in the context of positive emotional stimuli (study 2). Finally, duration of euthymia was significantly associated with right-sided DLPFC functioning such that the longer rMDD individuals were in recovery, the greater DLPFC functioning (r(19)= .69, P<0.001 ) (study 1), although this finding was not replicated in study 2. These findings demonstrate enduring deficits in OFC, DLPFC and VLPFC functioning in response to predominantly negative emotional stimuli in rMDD individuals. These findings may reflect neurobiological trait markers of the illness and have widespread implications for treatment development and prevention of illness onset in at-risk individuals.