Prefrontal cortical functional abnormality in major depressive disorder: a stereotactic meta-analysis

First, the objective was to test the hypothesis that prefrontal cortical regions most often reported to be maximally abnormal in studies of major depressive disorder, correspond to those regions reported maximally active when healthy subjects engage in diverse emotional tasks. Second, the objective was to determine whether such regions are reported typically to be either over or under-active.

Medline and Embase were used to search for neuroimaging studies of major depressive disorder from 1990 to 2005. Forty-two original studies using voxel based techniques were included, and compared with data from our previous meta-analysis on healthy subjects which included one hundred and eighty-one original studies [Steele, J.D., Lawrie, S.M., 2004b. Segregation of cognitive and emotional function in the prefrontal cortex: a stereotactic meta-analysis. Neuroimage 21, 868-875].

The medial prefrontal cortex is the region reported maximally abnormal most often when healthy subjects experience emotion. The region is centred on Broadmans Area (BA) 32 but extends into BA 25. Two further clusters of reported loci were identified in the lateral prefrontal cortex: one in the lateral orbitofrontal region reported active when healthy subjects experience emotion (BA 47); the other centred on a dorsolateral region (BA 46 and 9) associated with cognitive tasks. No reporting bias for overactivity or underactivity was identified.

This study pooled data from diverse studies deliberately. There were insufficient numbers of original studies to support sub-group analyses.

Despite the variability of reports in the literature, activity reported to be abnormal in depressive disorder is particularly localised to those brain regions that represent the substrate for normal emotional experience in healthy subjects.