IntroductionIn the past 2 decades, functional and structural neuroimaging techniques have been developing rapidly, yielding a large body of research addressing neural correlates of major depressive disorder (MDD). Neuroimaging has substantially shaped our understanding of depressive etiopathology, indicating its potential to become an objective tool of diagnostic, therapeutic and prognostic value.1 In MDD, first neuroimaging studies examined cerebral metabolism by means of positron emission tomography (PET). Predominantly left-sided decreased regional cerebral blood flow (rCBF) in the dorsolateral prefrontal cortex (DLPFC), medial prefrontal cortex (MPFC) and anterior cingulate cortex (ACC) were found.2,3 Comparing periods of acute illness with periods of recovery, remission of depressive symptoms was associated with a significant increase in rCBF in the left DLPFC cortical midline regions, such as the MPFC and ACC, suggesting state-relatedness of metabolic alterations. 4 Pretreatment metabolism in the rostral (pregenual) part of the ACC was reported to predict positive treatment response, 5-7 whereas hypermetabolism in the subcallosal part of the ACC was related to treatment resistance.8-11 Different treatment modalities have been shown to target distinct regions of the brain in patients with MDD. For instance, cognitive behavioural therapy was found to be associated with increased metabolism in the hippocampus and dorsal ACC together with metabolic decreases in prefrontal areas. In contrast, pharmacotherapy with paroxetine was related to prefrontal metabolism increases and subcortical decreases.12 Based on a rapidly growing body of neuroimaging evidence, specific neurobiological Background: Abnormal regional cerebral blood flow (rCBF) and grey matter volume have been frequently reported in patients with major depressive disorder (MDD). However, it is unclear to what extent structural and functional change co-occurs in patients with MDD and whether markers of neural activity, such as rCBF, can be predicted by structural change. Methods: Using MRI, we investigated resting-state rCBF and brain structure in patients with MDD and healthy controls between July 2008 and January 2013. We acquired perfusion images obtained with continuous arterial spin labelling, used voxel-based morphometry to assess grey matter volume and integrated biological parametric mapping analyses to investigate the impact of brain atrophy on rCBF. Results: We included 43 patients and 29 controls in our study. Frontotemporal grey matter volume was reduced in patients compared with controls. In patients, rCBF was reduced in the anterior cingulate and bilateral parahippocampal areas and increased in frontoparietal and striatal regions. These abnormalities were confirmed by analyses with brain volume as a covariate. In patients with MDD there were significant negative correlations between the extent of depressive symptoms and bilateral parahippocampal rCBF. We found a positive correlation between depressive symptoms and rCBF for right middle fron...