What can be drawn from new neuroimaging results concerning anxiety disorders and their interactions with depression?

MRCPsych, Diploma in Clinical Psychotherapy, MD
Psychopharmacology Unit
University of Bristol
Bristol, UK

What can be drawn from new neuroimaging results concerning anxiety disorders and their interactions with depression?

by A. L . Malizia, United Kingdom

Neuroimaging, be it based on emission tomography or magnetic resonance, has provided insight into brain changes associated with depression and anxiety, leading to further questions with regard to preclinical models and healthy volunteers. Not all findings, however, are consistent. Further investment is required toward the development of genotype and phenotype assessment in scanned subjects and toward increasing robustness of techniques, minimizing variation in results from one institute to another. Such variation in findings from different laboratories is likely to continue, as in some aspects of animal research, unless this investment occurs.

Medicographia. 2012;34:283-288 (see French abstract on page 288)

The advent of human brain imaging tools has allowed clinical neuroscientists to investigate brain processes and structures in vivo in a way that was unachievable 20-30 years ago. There are many tools to choose from that essentially use one of three technologies: emission tomography, magnetic resonance, and encephalography. Results from these investigations have extended psychiatric knowledge of brain function and dysfunction in anxiety and major depressive disorders. After an introduction that describes unresolved sources of possible confound, this review will briefly discuss the most important findings, in the author’s opinion, in different domains. It will finally focus on describing results from molecular imaging as an example of why a detailed understanding of methodology and a complementary experimental medicine cycle are important. The last point is emphasized because if we achieve these, we are likely to build solid foundations that will allow us to draw strong conclusions about brain changes in psychiatric disorders. A thorough reading of neuroimaging in psychiatry reveals that for many key areas there are very different results from seemingly similar experiments. While authors can claim to have obtained similar results to other research groups, this is so because of a great deal of approximation. For all imaging, attention to methodological detail is very important as seemingly small technical and experimental variation can generate and explain very different results. Almost all imaging techniques have evolved greatly, but many have not yet reached the stage whereby the strength of findings is guaranteed. To this extent, we may have to consider a proportion of the results as preliminary. The most robust results seem to be the findings that confirm theories based on experimental animals, yet the true power of the techniques in the long term is likely to be the demonstration that not all human brain processes map directly on animal equivalents and that change secondary to human disease may not be a variation of changes observed with similar processes in health.

Furthermore, all the investigations are carried out in specialized centers and very unusual environments, which may have diverse effects on patients and healthy volunteers, potentially changing the ecological validity, nature, or magnitude of the observed parameters. Finally, it is likely that small variations in clinical characteristics of those taking part in experiments have the potential to alter results; sometimes these differences are not detected by experimental protocols and measuring instruments, losing “fine grain” information that may have explanatory power.

For example, most studies try to recruit people with “pure” syndromes, but how different are depressed people who do not have prominent anxiety symptoms from the equally prevalent depressed patients with comorbid anxiety disorders or subthreshold anxiety disorders?1 What are the differences between depressed people with overvalued guilt and reference ideas from patients who do not have these and from depressed people who have delusions of reference or guilt? How does prominent suicidal ideation change brain structure and function in depression and in anxiety disorders?2 Is suicidality a separate variable that should be teased out independently? If so, what is the boundary that would divide it from people who have no suicidal ideation? Is wishing to be dead truly different from having thoughts of killing oneself?

Finally, on a global scale, there are differences between nations, which are not only due to the diversity of any one sample or genetic background, but also to local socioeconomic factors. For example, does research carried out with patients from the public health sector produce the same results as research carried out in people who only get free health care if they take part in research?

In spite of the above caveats, neuroimaging has produced some important results; even if some or many of these will be contradicted by subsequent research, brain imaging is one of the investigative tools that has put psychiatric illness back where it is generated: the brain. This in itself has already been a worthwhile result.

Anatomical brain mapping

Anatomical magnetic resonance imaging (MRI) and magnetic resonance (MR) tractography can be used to measure anatomical variables. The former can measure gray and white matter and cerebrospinal fluid (CSF) distribution while the latter can measure the thickness and direction of white matter tracts that connect brain areas. Both rely on signals that are relative and where boundaries define categories, so variability is expected in the results from different centers using different scanners and analytical tools. In addition, while spatial “normalization” (the process of stretching any brain volume to make it fit with standard templates) is essential for volumetric comparisons, this does not fully avoid partial volume effects (whereby a signal is diluted and smoothed out at boundaries between structures3); so caution has to be exercised when interpreting results, making certain that appropriate comparisons have been made.

While diffusion tensor imaging has so far not produced any significant findings in the field of anxiety and depression except for helping to separate the connective anatomy of some of the putative circuits, volumetric analysis has demonstrated that the hippocampus, amygdalae, anterior cingulate, basal ganglia, and parts of the frontal cortex have smaller volumes in people with depression; ventricular space is also increased.4,5 This is consistent with neuropathological findings which have noted that with recurrent depression there is loss of brain tissue,6,7 and with resting metabolic studies which show a change in metabolism in some of the same areas. Of greatest interest is the fact that hippocampal atrophy seems reversible on successful antidepressant treatment or on cessation of the depressive episode. In the hippocampal area, this could be due not only to glial proliferation and synaptic enrichment, but also to neurogenesis.

Anatomical studies have also revealed that patients with major depression have increased white matter abnormalities, especially when older. The treatment implication of these findings are that people who have these vascular lesions may be harder to treat successfully and that medicines that increase an orthostatic drop in blood pressure may contribute to the problem.8 No such difference has been found for anxiety disorders, except for panic disorder where there is an increase in white matter abnormalities in the temporal lobes.9 Very similar brain areas have been implicated in morphological changes in anxiety disorders. However, the only disorder where a decrease in volume has been shown to be reversed by treatment is posttraumatic stress disorder, where hippocampal volume increases with antidepressant treatment and is accompanied by improvements in specific memory tests.10 In addition, obsessive-compulsive disorder (OCD) seems to have a different pattern of volume changes from other disorders, whereby thalamus and basal ganglia have an increase in volume, and the anterior cingulate and the orbitofrontal cortex show a decrease in volume.11 These studies emphasize that some structural changes, eg, in the hippocampus, may be common to a number of psychiatric disorders while others, eg, basal ganglia, may help differentiation between disorders.

Functional brain mapping

Functional MR, MR arterial spin labeling, water or fluorodeoxyglucose positron emission tomography (PET), hexamethylpropyleneamine oxime single-photon emission computed tomography (HMPAO SPECT), electroencephalography (EEG), and magneto encephalography (MEG) are used to investigate resting or task-related activity in populations of neurons. Understanding that the signal is generated by summation of hundreds of thousands of neurons is important as it can address apparent discrepancies in results when compared with invasive preclinical studies, whereby electrodes can be so accurate as to record the activity of single neurons in specific brain structures. Encephalographic techniques have very good temporal resolution, but poor anatomical resolution, while tomographic techniques have reasonable spatial resolution and poorer (MR) or very poor (PET/SPECT) temporal resolution. Some experiments now combine them in order to gain precision in the spatial and temporal domains.

Encephalographic techniques record the summation of electrical or magnetic neuronal signals through the skull. The other techniques use metabolic surrogate measures of multineuronal synaptic activity such as blood perfusion or tissue metabolic rate. The general assumption is that the local level of neuronal activity is highly correlated with glucose consumption (thus with fluorodeoxyglucose trapping) and/or with local perfusion; however, there have been a number of studies where recorded changes in local fluorodeoxyglucose concentration do not map on all the perfusion changes and vice versa, demanding that we better understand the nature of the signals.12 As well as transient localized changes, some of these techniques can measure the change in temporospatial correlation of activity from different brain areas at rest or during experimental or control conditions.

Initial and subsequent studies in depression have demonstrated decreases in activity in the dorsolateral prefrontal cortex, although this may be more closely related to symptoms of slowing and psychomotor retardation. The approach of trying to map individual symptoms has been less popular compared with those of investigating network changes with recovery, changes in emotion in healthy volunteers, and healthy volunteer– patient differences elicited by carrying out specific tasks in the scanner. The main finding associated with depression recovery has been alteration in function in the subgenual cingulate in response to physical treatments and placebo13; this has not been a universal finding, but has been useful in developing a successful target for deep brain stimulation in treatment of refractory depression.14,15 In addition, this same area has been reported to be within the network implicated in the experience of sadness in healthy volunteers, lending further credence to its pivotal implication.

Changes in amygdalae reactivity to presentation of emotional faces has been used both as a marker of successful treatment16 and of early changes in brain function with antidepressants.17 Medial and dorsolateral frontal activation have also been found to be altered in a number of comparative experiments, including studies of response to social stimuli18 or to emotional distractors during a working memory task19 in remitted unmedicated patients, which could be trait abnormalities. Studies of functional connectivity suggest decreased connectivity between limbic structures and frontal areas,20 which may be more severe in treatment of refractory depression21 and which are consistent with the effects described in the previous three paragraphs. In anxiety disorders, key findings have been the overactivity of fronto-striatal-thalamic circuits in OCD and differences in amygdalae reactivity in social anxiety and panic disorder and posttraumatic stress disorder.22-26 These are emerging as differences which map across a number of different disorders and may be regarded as a future way to classify diseases.

Receptor, neurotransmitter, and transporter mapping

Other radioligand PET/SPECT and MR spectroscopy (MRS) measure the concentration of specific biochemical molecules. While MRS measures these directly, provided that adequate calibration of the instrument has been achieved, emission tomography relies on radioactive emission from appropriate “tagged” ligands to record changes in concentration of these molecules in various parts of the brain over a period of time. Since this signal represents the totality of the signal from a particular small volume of the brain (voxel), a number of assumptions have to be made in order to finally extract the signal of interest. These only work if the ligand has been administered in tracer doses, if vascular effects can be excluded, if an appropriate, robust and unconfounded reference measurement can be achieved to account for transport into the brain and for nonspecific binding, and if a temporary equilibrium has been reached. The main PET findings in terms of receptor/ transporter changes in affective and anxiety disorders are summarized in Table I (page 286). SPECT findings are not summarized here, as for a variety of reasons they are more difficult to interpret.

Table I
Table I. Summary of PET receptor and transporter studies in anxiety and affective disorders.

Abbreviations: –, no published peer reviewed studies; 0, no difference from controls; Acut, acute; BPD, bipolar depression; BPM, bipolar mania; Chr, chronic; DACC, dorsal accumbens; DAT, dopamine transporter; DRD1, dopamine D1 receptor; DRD2, dopamine D2 receptor; GAD, generalized anxiety disorder; MAO, monoamine oxidase; MDD, major depressive disorder; mGluR5, metabotropic glutamate receptor 5; MPFCx, medial prefrontal cortex; OCD, obsessive-compulsive disorder; Older, older patients, usually >65; PAG, periaqueductal gray; PD/AG, panic disorder/agoraphobia; PTSD, posttraumatic stress disorder; S, state; SAnD, social anxiety disorder; SERT, serotonin transporter; T, trait.

As can be seen, there are a number of issues:
_ To date relatively few brain receptors and transporters can be studied. This is because synthesizing PET compounds is not very difficult, but finding ones that have the characteristics that allow signal recovery is very difficult. Useful brain PET ligands have to be able to cross the blood-brain barrier (ie, be lipophilic) yet have relatively little nonspecific binding.
_ However, even with existing compounds, there have been a limited number of investigations. These studies are considered to be expensive and often require a considerable team effort. Only a few centers exist in the world that are able to carry out this work to a high standard and there has never been a combined and considerable industrial-governmental investment in the area, with some of the promising laboratories not managing to achieve sustainability.
_ Results can be contradictory; there are many potential reasons for this, some being the clinical presentation and state of the study sample, and some due to technical issues, discussed in part below.
_ For some ligands the interpretation can be twofold: for example, a decrease in tracer binding at the D2 receptor can occur because the density of receptors is decreased or because there is an increase in synaptic dopamine. In this respect, it is, for instance, interesting that D2 binding is unaffected in people who are currently experiencing hypomania. This may be true or an artifact because people who are likely to stay in the scanner must have a mild form of the disease or because the ligand measures the total of functioning and internalized receptors! Here I will discuss two examples that I think are conceptually important: receptor changes in anxiety disorders and 5-HT1A binding in depression.

The most intriguing finding in anxiety disorders is that when γ-aminobutyric acid ionotropic receptor family A (GABAA) benzodiazepine receptors have been measured, there are localized or even global decreases27 in receptor expression; in complementary studies, 5-HT1A binding is also decreased,28 and this can be partially reversible with drug response. When these findings are reviewed29 their importance is that they map to receptor knockout studies in mice, whereby underexpression of GABAA receptors or of 5-HT1A receptors leads to increases in anxiety and decreases in expression of the other receptor, a situation that can be replicated with increases in corticosterone and with other neurochemical phenomena associated with increased anxiety30,31 (see also discussion in reference 27). This is an important mapping of human findings on preclinical studies and vice versa, which has not yet been followed up to its full extent.

5-HT1A receptors are also thought to be important in depression and in mediating the cerebral effects of stress responses. Since the development of WAY-100635 as a ligand for PET studies, exquisitely delineating 5-HT1A receptors (Figure 1), the issue of 5-HT1A expression in depression has been studied by a number of groups. Until these studies, the only data available came from ex vivo measurement in patients who had undergone neurosurgery for intractable depression whereby both a decrease in Bmax and Kd had been observed.32 The majority of reported PET studies showed a decrease in cortical and raphe binding in depression; however, one group at Columbia showed the opposite and demonstrated that if they used a different analytical technique—the one used by the other groups—the finding of a decrease in binding would be replicated. Since the technique used by Columbia relies on fewer assumptions, it questions the validity of a conclusion that had been based on a number of concordant previous studies (for review and discussion see reference 33). This is an important issue to resolve as the credibility of PET studies depends on the robustness of findings, allowing the science community to disprove or accept findings from lab animals in order to better understand disease. In this area of endeavor, the other points that are worth noting are that measured increases in monoamine oxidase (MAO) during depression are potentially of great interest, that so far dopamine and serotonin system findings in OCD are consistent with theory of how effective pharmacological treatments work, and that MR spectroscopy has reawakened interest in GABA in depressive disorders and in the mechanism of panic attacks.34,35 5-HT transporter and 5-HT2 studies have been more variable than expected, probably because of the characteristics of ligands initially used and the confound of suicidality.

Figure 1
Figure 1. Delineation of 5-HT1A receptors
by WAY-100635 PET.

Note how the raphe is visible even though it is a structure
which is difficult to dissect in brains. Also note the
intense signal in the hippocampus, anterior cingulate,
inferior temporal lobe, and insula.

Table II
Table II. Occupancy by or receptor effects of specific treatments.

Abbreviations: DAT, dopamine transporter; DRD2, dopamine D2 receptor; ECT,
electroconvulsive therapy; MAOA, monoamine oxidase A; RIMA, reversible inhibitors
of monoamine oxidase A; SERT, serotonin transporter; SNRI, serotoninnorepinephrine
reuptake inhibitors; SSRI, selective serotonin reuptake inhibitor.

Findings from studies of occupancy or in response to physical treatments (Table II) have demonstrated that MAO inhibitors (MAOIs) and selective serotonin reuptake inhibitors (SSRIs)36,37 achieve their effects at occupancies/binding above 70%-80% and that pindolol, which was studied as an antidepressant augmenter, did not achieve 5-HT1A occupancies that were high enough to consistently test the hypothesis that raphe 5-HT1A blockade would augment antidepressant response. In addition, one of these studies demonstrated that pindolol achieves different occupancies in the raphe and the cortex, a finding that is not easily explained given that pure antagonists do not do this.38 Unfortunately, this finding has never been further investigated in preclinical models, even though it may inform on serotonergic system regulation. Another interesting finding is that electroconvulsive therapy (ECT) down regulates 5-HT2 receptors in man,39 which is not what is seen with electroconvulsive shock (ECS) in rats, but is consistent with the effects in primates and its powerful antidepressant effects. These studies should be repeated across chemical systems to ensure that downstream effects of treatment are also well understood.


Neuroimaging remains an important and powerful tool in the investigation of human anxiety and depression. However, the findings need to be scrutinized with a critical eye and funding for this area should be made a priority so as to be able to obtain robust measures (thus far lacking) of brain function in disease. The apparent discrepancies are either due to unsophisticated clinical categorization or to methodological issues that will be resolved with time. _

1. van Tol MJ, Demenescu LR, van der Wee NJ, et al. Functional magnetic resonance imaging correlates of emotional word encoding and recognition in depression and anxiety disorders. Biol Psychiatry. 2012;71(7):593-602.
2. Wagner G, Koch K, Schachtzabel C, Schultz CC, Sauer H, Schlösser RG. Structural brain alterations in patients with major depressive disorder and high risk for suicide: evidence for a distinct neurobiological entity? Neuroimage. 2011; 54(2):1607-1614.
3. McGonigle J, Mirmehdi M, Malizia AL. Methodological issues in comparing brain connectivity between groups. Magn Reson Med. 2012 submitted.
4. Koolschijn PC, van Haren NE, Lensvelt-Mulders GJ, Hulshoff Pol HE, Kahn RS. Brain volume abnormalities inmajor depressive disorder: ameta-analysis ofmagnetic resonance imaging studies. Hum Brain Mapp. 2009;30(11):3719-3735.
5. Sacher J, Neumann J, Fünfstück T, Soliman A, Villringer A, Schroeter ML. Mapping the depressed brain: a meta-analysis of structural and functional alterations in major depressive disorder. J Affect Disord. 2012;140(2):142-148.
6. Rajkowska G, Miguel-Hidalgo JJ. Gliogenesis and glial pathology in depression. CNS Neurol Disord Drug Targets. 2007;6(3):219-233.
7. Altshuler LL, Abulseoud OA, Foland-Ross L, et al. Amygdala astrocyte reduction in subjects with major depressive disorder but not bipolar disorder. Bipolar Disord. 2010;12(5):541-549.
8. Colloby SJ, Vasudev A, O’Brien JT, Firbank MJ, Parry SW, Thomas AJ. Relationship of orthostatic blood pressure to white matter hyperintensities and subcortical volumes in late-life depression. Br J Psychiatry. 2011;199:404-410.
9. Dantendorfer K, Prayer D, Kramer J, et al. High frequency of EEG and MRI brain abnormalities in panic disorder. Psychiatry Res. 1996;68(1):41-53.
10. Vermetten E, Vythilingam M, Southwick SM, Charney DS, Bremner JD. Longterm treatment with paroxetine increases verbal declarative memory and hippocampal volume in posttraumatic stress disorder. Biol Psychiatry. 2003;54(7): 693-702.
11. Radua J, van den Heuvel OA, Surguladze S, Mataix-Cols D. Meta-analytical comparison of voxel-based morphometry studies in obsessive-compulsive disorder vs other anxiety disorders. Arch Gen Psychiatry. 2010;67(7):701-711.
12. Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH. Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry. 2008;64(6):461-467.
13. Mayberg HS, Liotti M, Brannan SK, et al. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am J Psychiatry. 1999;156(5):675-682.
14. Mayberg HS, Lozano AM, Voon V, et al. Deep brain stimulation for treatmentresistant depression. Neuron. 2005;45(5):651-660.
15. Holtzheimer PE, Kelley ME, Gross RE, et al. Subcallosal cingulate deep brain stimulation for treatment-resistant unipolar and bipolar depression. Arch Gen Psychiatry. 2012;69(2):150-158.
16. Sheline YI, Barch DM, Donnelly JM, Ollinger JM, Snyder AZ, Mintun MA. Increased amygdala response to masked emotional faces in depressed subjects resolves with antidepressant treatment: an fMRI study. Biol Psychiatry. 2001;50 (9):651-658.
17. Browning M, Holmes EA, Harmer CJ. The modification of attentional bias to emotional information: A review of the techniques, mechanisms, and relevance to emotional disorders. Cogn Affect Behav Neurosci. 2010;10(1):8-20.
18. Elliott R, Lythe K, Lee R, et al. Reduced medial prefrontal responses to social interaction images in remitted depression. Arch Gen Psychiatry. 2012;69(1): 37-45.
19. Kerestes R, Ladouceur CD, Meda S, et al. Abnormal prefrontal activity subserving attentional control of emotion in remitted depressed patients during a working memory task with emotional distracters. Psychol Med. 2012;42(1):29-40.
20. Carballedo A, Scheuerecker J, Meisenzahl E, et al. Functional connectivity of emotional processing in depression. J Affect Disord. 2011;134(1-3):272-279.
21. Lui S, Wu Q, Qiu L, et al. eResting-state functional connectivity in treatment-resistant depression. Am J Psychiatry. 2011;168(6):642-648.
22. Menzies L, Chamberlain SR, Laird AR, Thelen SM, Sahakian BJ, Bullmore ET. Integrating evidence from neuroimaging and neuropsychological studies of obsessive- compulsive disorder: the orbitofronto-striatal model revisited. Neurosci Biobehav Rev. 2008;32(3):525-549.
23. Shin LM, Handwerger K. Is posttraumatic stress disorder a stress-induced fear circuitry disorder? J Trauma Stress. 2009;22(5):409-415.
24. Fusar-Poli P, Placentino A, Carletti F, et al. Functional atlas of emotional faces processing: a voxel-based meta-analysis of 105 functional magnetic resonance imaging studies. J Psychiatry Neurosci. 2009;34(6):418-432.
25. Freitas-Ferrari MC, Hallak JE, Trzesniak C, et al. eNeuroimaging in social anxiety disorder: a systematic review of the literature. Prog Neuropsychopharmacol Biol Psychiatry. 2010;34(4):565-580.
26. Kim MJ, Loucks RA, Palmer AL, et al. The structural and functional connectivity of the amygdala: from normal emotion to pathological anxiety. Behav Brain Res. 2011;223(2):403-410.
27. Malizia AL, Cunningham VJ, Bell CJ, Liddle PF, Jones T, Nutt DJ. Decreased brain GABAA-benzodiazepine receptor binding in panic disorder: preliminary results from a quantitative PET study. Arch Gen Psychiatry. 1998;55(8):715-720.
28. Nash JR, Sargent PA, Rabiner EA, et al. Serotonin 5-HT1A receptor binding in people with panic disorder: positron emission tomography study. Br J Psychiatry. 2008;193(3):229-234.
29. Nikolaus S, Antke C, Beu M, Müller HW. Cortical GABA, striatal dopamine and midbrain serotonin as the key players in compulsive and anxiety disorders— results from in vivo imaging studies. Rev Neurosci. 2010;21(2):119-139.
30. Bailey SJ, Toth M. Variability in the benzodiazepine response of serotonin 5- HT1A receptor null mice displaying anxiety-like phenotype: evidence for genetic modifiers in the 5-HT mediated regulation of GABAA receptors. J Neurosci. 2004;24(28):6343-6351.
31. Vinkers CH, Oosting RS, van Bogaert MJ, Olivier B, Groenink L. Early-life blockade of 5-HT1A receptors alters adult anxiety behavior and benzodiazepine sensitivity. Biol Psychiatry. 2010;67(4):309-316.
32. Francis PT, Pangalos MN, Stephens PH, et al. Antemortem measurements of neurotransmission: possible implications for pharmacotherapy of Alzheimer’s disease and depression. J Neurol Neurosurg Psychiatry. 1993;56(1):80-84.
33. Shrestha S, Hirvonen J, Hines CS, et al. Serotonin-1A receptors in major depression quantified using PET: controversies, confounds, and recommendations. Neuroimage. 2012;59(4):3243-3251.
34. Bhagwagar Z, Wylezinska M, Jezzard P, et al. Reduction in occipital cortex γ-aminobutyric acid concentrations in medication-free recovered unipolar depressed and bipolar subjects. Biol Psychiatry. 2007;61(6):806-812.
35. Layton ME, Friedman SD, Dager SR. Brain metabolic changes during lactateinduced panic: effects of gabapentin treatment. Depress Anxiety. 2001;14(4): 251-254.
36. Sacher J, Houle S, Parkes J, et al. Monoamine oxidase A inhibitor occupancy during treatment of major depressive episodes with moclobemide or St. John’s wort: an [11C]-harmine PET study. J Psychiatry Neurosci. 2011;36(6):375-382.
37. Voineskos AN, Wilson AA, Boovariwala A, et al. Serotonin transporter occupancy of high-dose selective serotonin reuptake inhibitors during major depressive disorder measured with [11C]DASB positron emission tomography. Psychopharmacology (Berl). 2007;193(4):539-545.
38. Martinez D, Hwang D, Mawlawi O, et al. eDifferential occupancy of somatodendritic and postsynaptic 5HT1A receptors by pindolol: a dose-occupancy study with [11C]WAY 100635 and positron emission tomography in humans. Neuropsychopharmacology. 2001;24(3):209-229.
39. Yatham LN, Liddle PF, Lam RW, et al. Effect of electroconvulsive therapy on brain 5-HT2 receptors in major depression. Br J Psychiatry. 2010;196(6):474-479.

Keywords: bipolar; brain imaging; major depression; MRI; OCD; panic disorder; PET; posttraumatic stress disorder; social anxiety disorder