The Sad Cingulate Before CBT

The latest sad cingulate news is an fMRI study that examined the responsiveness of this region (subgenual cingulate cortex, aka Brodmann area 25) to emotional stimuli as a predictor of recovery in depressed patients receiving cognitive behavior therapy (CBT).

Fourteen depressed individuals were scanned while they rated the personal relevance of words that were positive, negative, or neutral. Then the subjects received 16 sessions of CBT over a 12 week period. To detect brain regions that were associated with recovery, the authors (Siegle et al.) ran whole-brain voxelwise analyses to find regions that correlated with changes in symptom severity after therapy. Lo and behold, the only brain area that was associated with improvement was BA 25.

In the graph above, I've drawn in dotted red lines to illustrate the participants who did show a treatment response. These 7 people (in the lower left-hand corner of the graph) got better after CBT (change in Beck Depression Inventory score is negative) AND showed a decrease in BA 25 activity after viewing negative words ("sustained" BA 25 signal is also negative). The graph is correct in predicting the outcome of 11 out 14 patients (78.6%).

OK, what kind of neural responsiveness are we really looking at here? The "sad cingulate" hemodynamic response was measured in an interval 6-10 sec after subjects rated the self-relevance of a negative word, compared to a pre-stimulus baseline. SO the subgenual cingulate was more active in the baseline "resting" state than in the "rate words" condition. This would be consistent with the elevated "resting state" activity in BA 25 seen by Mayberg et al., which reverted to control levels in those who responded to SSRIs.

In the current study, we don't know what happened to neural activity in BA 25 after the full course of CBT. Furthermore, the task-related decrease in BA 25 activity was not specific to negative words... it also occurred to positive and neutral words in the CBT responders, so I'm not sure why the authors claim that the responders showed deficient emotional control processes (since the same pattern of neural responsiveness was obtained to neutral words).

Use of fMRI to Predict Recovery From Unipolar Depression With Cognitive Behavior Therapy.
Greg J. Siegle, Ph.D., Cameron S. Carter, M.D. and Michael E. Thase, M.D.
American Journal of Psychiatry 163:735-738, April 2006

OBJECTIVE: In controlled treatment trials, 40%–60% of unmedicated depressed individuals respond to cognitive behavior therapy (CBT). The authors examined whether pretreatment neural reactivity to emotional stimuli accounted for this variation. METHOD: Unmedicated depressed individuals (N=14) and never depressed comparison subjects (N=21) underwent fMRI during performance of a task sensitive to sustained emotional information processing. Afterward, depressed participants completed 16 sessions of CBT. RESULTS: Participants whose sustained reactivity to emotional stimuli was low in the subgenual cingulate cortex (Brodmann’s area 25) and high in the amygdala displayed the strongest improvement with CBT. CONCLUSIONS: The presence of emotion regulation disruptions, which are targeted in CBT, may be the key to recovery with this intervention.

SUMMARY from The Neurocritic: the authors' interpretation of the BA 25 results seemed a little convoluted to me. Perhaps more noteworthy (or at least straightforward) is what happened in the amygdala (using a region of interest analysis): it showed a huge response to negative words in the depressed people but not the controls. This is nothing new, but the huge and specific amygdala response to negative words was larger in CBT responders than non-responders, and it correlated with self-report on a rumination scale.