Brain Metabolic Correlates of Depression and Recovery

July 1, 2006

Antidepressants may have a protective effect on the hippocampal atrophy seen in patients with severe, untreated depression. This atrophy may be caused by an overabundance of glucocorticoids.

July 2006, Vol. XXIII, No. 8

Depression is increasingly considered to have a biologic basis. The effectiveness of antidepressant medication suggested this decades ago, but now neuroimaging, neuropathology, and neurochemical studies can delineate brain pathology. The hippocampus (a bilateral subcortical temporal lobe structure) is one area of the brain where patients with major depression may experience loss of volume if their depression is prolonged and untreated. Antidepressant medication seems to protect against such hippocampal atrophy.1

Hippocampal and cortical atrophy

A hippocampal volume loss of up to 19% may occur in patients with severe, untreated depression, which may explain some of the memory impairment seen in such patients; the hippocampus has long been known as important for intact memory function.2 Some investigators have even speculated that the proximate cause of hippocampal atrophy involves an overabundance of glucococorticoids.3 The circumstantial evidence includes the following:

  • Chronic stress and serious depression are both associated with hypersecretion of cortisol (as demonstrated by abnormalities in the dexamethasone suppression test).

  • Glucocorticoids, secreted by the adrenal gland in times of stress, have known neurotoxic effects, including dendritic retraction and inhibition of neurogenesis.

  • Hippocampal volume loss and depression develop secondary to Cushing syndrome, in which a pituitary adenoma stimulates oversecretion of glucocorticoids.4

Even patients who have recovered clinically from depression show significantly greater levels of waking salivary cortisol than age- and gender-matched controls.5 Indeed, elevated levels of cortisol may be responsible for the high frequency of comorbid medical disorders, such as coronary heart disease, that are associated with depression. The degree to which ongoing stress, hippocampal abnormality, or primary hypothalamic-pituitary-adrenal axis functional abnormalities contribute to a person's vulnerability to future depressive episodes is still uncertain, however.

There is also evidence of cortical atrophy-particularly in the orbitofrontal cortex6 and dorsolateral prefrontal cortex7-associated with prolonged and repeated depression. These findings may also explain why depressed persons often have a significant impairment in cognition and working memory-functions usually associated with the frontal lobes.8

How can we bring together the new findings concerning the functions of the hippocampus and the cortex? In 1997, Mayberg9 proposed that depression results from a failure of the coordinated interactions of a distributed network of limbic-cortical pathways. Although her hypothesis was originally based on resting patterns of regional glucose metabolism seen in depressed patients, changes in metabolism with antidepressant treatment, and blood flow changes with induced sadness in healthy study participants, the theory is compatible with new findings showing different patterns of brain recovery resulting from different kinds of treatment.

Metabolic effects of depression treatment

Other researchers measured changes in regional glucose metabolism using positron emission tomography (PET) in patients in remission from depression after 15 to 20 sessions of cognitive behavioral therapy (CBT).10 As might be expected, treatment response was associated with significant metabolic changes. The pattern observed, however, is most interesting: decreases in glucose metabolism in the frontal cortex and increases in glucose metabolism in the hippocampus. This pattern is opposite from the pattern seen previously in clinical recovery from depression facilitated by paroxetine (Paxil). With paroxetine therapy, PET scans showed increases in glucose metabolism in the prefrontal cortex and decreases in glucose metabolism in the hippocampus.10

Taken together, the treatmentspecific changes in metabolic patterns with CBT and paroxetine therapy show that each treatment targets different primary sites with differential top-down (CBT) and bottom-up (paroxetine) effects. In other words, CBT seems to work primarily on the cortex (as one would expect with a form of psychotherapy), whereas pharmacotherapy targets lowerbrain limbic and subcortical areas. Mayberg9 and others make the point that it is the overall modulation of the complex prefrontal cortex-hippocampal pathway system that is important, rather than any single focal change.

Studying placebo response is also informative in that it helps explain what is going on in the brain of depressed patients. Patients who receive placebo in medication trials have increases in cortical glucose metabolism and decreases in limbic system glucose metabolism that are similar to the metabolic changes in patients taking paroxetine or fluoxetine (Prozac) who have had a clinical response.11 Active drug responders, however, showed more robust changes, especially in the hippocampus, where the metabolic effects of antidepressant medication treatment were clearly seen by week 1. Similar early responses in the hippocampus did not show up in placebo-treated patients or in patients who did not demonstrate a clinical response to antidepressant medication.

Although we might have guessed that trials of CBT against placebo would have produced CBT-like but less robust changes in the placebo group, that turned out not to be the case. There was no overlap between the pattern of brain changes seen with psychotherapy and placebo. As Mayberg9 notes, “Such differences not only support the obvious conclusion that psychotherapy is an active treatment with modality-specific effects on regional brain function but also provide evidence that placebo response is not simply a nonspecific type of psychotherapy.”12

Another line of research seeks to explain the brain atrophy seen in severe depression by observing the effects of neurotrophic growth factors, such as brain-derived neurotrophic factor (BDNF). Support for this research comes from the observation that the amount of BDNF was decreased in the plasma of depressed persons13 but was up-regulated in postmortem hippocampal tissue of patients receiving antidepressant treatment at the time of death.14 In addition, electroconvulsive therapy, long known as the most effective treatment of depression in humans, increases the amount of BDNF in rat brains.15 Infusions of BDNF into either the midbrain or hippocampus produce an antidepressant-like effect in behavioral models of depression.16

The exploration of brain atrophy (and recovery from brain atrophy) seen in major depression is an ongoing enterprise, with a variety of hypotheses seeking to explain the phenomenon. The results so far are encouraging and may point to better design and targeted effects of new-generation antidepressants.17

Classifying depression by brain metabolism?

A related avenue of brain research seeks to distinguish one type of depression from another. This may foreshadow a future classification system based directly on brain pathology (derived from imaging and other studies) rather than on clinical symptomatology alone. Patients with different forms of depression might get the precise treatment they need without the sometimes difficult trial-and-error approach currently in use.

Indeed, a recent article in the American Journal of Psychiatry begins to suggest such a possibility by distinguishing lithium-responsive bipolar patients from their healthy siblings on the basis of brain responses to emotional challenge. The methodology employed in this study by Krüger and associates18 may be just as important as the findings. Changes in regional cerebral blood flow (rCBF) were measured with [15O] water PET after induction of transient sadness in 9 euthymic lithium responders and 9 healthy siblings.

The transient sadness was induced by a standardized method in which participants were requested to draft a short autobiographic script describing a sad life event. Each script was tested before brain scanning to be sure it reliably reproduced a sad mood. During the PET study, the script was projected onto a computer screen to facilitate recall, while the participant underwent the brain scan. Such a technique seems able to find rCBF patterns in lithiumresponsive patients that distinguish them from their unaffected siblings and even from valproate (Depakote)-responsive patients with bipolar disorder, patients with unipolar depression, and normal controls.

Such studies begin to take us beyond passive morphologic or metabolic brain studies to find possible differences in response to emotional challenge. Clinicians know that many patients do quite well when things are going well for them. The cascade of feelings and symptoms that eventually results in a clinical syndrome begins when one loses a significant relationship or experiences an insult to self-esteem. Resilience to emotional happenings may distinguish illness from successful coping. As Swann19 points out in an editorial accompanying the article by Krüger and colleagues,18 “we need to deconstruct the susceptibility, protective, and illness-course mechanisms that underlie the illness.”

In applying modern functional imaging methods to the brain in an attempt to understand the mind, one must remember that what is really being measured derives from changes in blood flow, glucose consumption, and glucose oxidation, which are at best physiologic measures of brain energy consumption. That still leaves us wanting an explanation of mind in terms of chemistry and connectivity of neurons.20 Indeed, how these findings about brain morphology and function tie in or supersede the current emphasis on the role of neurotransmitters, such as serotonin, norepinephrine, dopamine, and glutamate/glutamine, is still a matter of debate.

Dr Pomerantz practices psychiatry in Longmeadow, Mass, and is assistant clinical professor of psychiatry at Harvard Medical School in Boston.


1. Sheline YI, Gado MH, Kraemer HC. Untreated depression and hippocampal volume loss. Am J Psychiatry. 2003;160:1516-1518.
2. Bremner JD, Narayan M, Anderson ER, et al. Hippocampal volume reduction in major depression. Am J Psychiatry. 2000;157:115-118.
3. Sapolsky RM. Depression, antidepressants and the shrinking hippocampus. Proc Natl Acad Sci U S A. 2001;98:12320-12322.
4. Starkman MN, Gebarski SS, Berent S, Schteingart DE. Hippocampal formation volume, memory dysfunction, and cortisol levels in patients with Cushing's syndrome. Biol Psychiatry. 1992;32:756-765.
5. Bhagwagar Z, Hafizi S, Cowen PJ. Increase in concentration of waking salivary cortisol in recovered patients with depression. Am J Psychiatry. 2003;160:1890-1891.
6. Bremner JD, Vythilingam M, Vermetten E, et al. Reduced volume of orbitofrontal cortex in major depression. Biol Psychiatry. 2002;51:273-279.
7. Rajkowska G, Miguel-Hidalgo JJ, Wei J, et al. Morphometric evidence for neuronal and glial prefrontal cell pathology in major depression. Biol Psychiatry. 1999;45:1085-1098.
8. Merriam EP, Thase ME, Haas GL, et al. Prefrontal cortical dysfunction in depression determined by Wisconsin Card Sorting Test performance. Am J Psychiatry. 1999;156:780-782.
9. Mayberg HS. Limbic-cortical dysregulation: a proposed model of depression. J Neuropsychiatry Clin Neurosci. 1997;9:471-481.
10. Goldapple K, Segal Z, Garson C, et al. Modulation of cortical-limbic pathways in major depression: treatment- specific effects of cognitive behavior therapy. Arch Gen Psychiatry. 2004;61:34-41.
11. Mayberg HS, Silva JA, Brannan SK, et al. The functional neuroanatomy of the placebo effect. Am J Psychiatry. 2002;159:728-737.
12. Mayberg HS. Defining neurocircuits in depression. Psychiatr Ann. 2006;11:259-267.
13. Karege F, Perret G, Bondolfi P, et al. Decreased serum brain-derived neurotrophic factor levels in major depressed patients. Psychiatry Res. 2002;109: 143-148.
14. Chen B, Dowlatshahi D, MacQueen GM, et al. Increased hippocampal BDNF immunoreactivity in subjects treated with antidepressant medication. Biol Psychiatry. 2001;50:260-265.
15. Angelucci F, Aloe L, Jimenez-Vasquez P, Mathe AA. Electroconvulsive stimuli alter the regional concentrations of nerve growth factor, brain-derived neurotrophic factor, and glial cell line-derived neurotrophic factor in adult rat brain. J ECT. 2002;18:138-143.
16. Shirayama Y, Chen AC, Nakagawa S, et al. Brainderived neurotrophic factor produces antidepressant effects in behavioral models of depression. J Neurosci. 2002;22:3251-3261.
17. Thomas RM, Peterson DA. A neurogenic theory of depression gains momentum. Mol Interv. 2003;3:441-444.
18. Krüger S, Alda M, Young LT, et al. Risk and resilience markers in bipolar disorder: brain responses to emotional challenge in bipolar patients and their healthy siblings. Am J Psychiatry. 2006;163:257-264.
19. Swann AC. Editorial: what is bipolar disorder? Am J Psychiatry. 2006;163:177-178.
20. Shulman RG. Functional imaging studies: linking mind and basic neuroscience. Am J Psychiatry. 2001;158:11-20.