Advances in Neuroimaging: Impact on Psychiatric Practice

Neuroimaging is often used in clinical psychiatry to rule out medical and neurological conditions that can mimic psychiatric disease rather than for the diagnosis of specific psychiatric disorders. Indeed, no known primary psychiatric disorder can be definitively diagnosed on the basis of neuroimaging alone.¹ Brain imaging can be grossly divided into 2 separate categories: structural imaging and functional imaging. Structural imaging uses modalities such as CT and MRI, whereas functional imaging modalities include positron emission tomography (PET), single-photon emission CT (SPECT), magnetic resonance spectroscopy (MRS), functional MRI (fMRI), and diffusion tensor MRI tractography (DT-MRI or DTI).

Traditionally, the structural and functional divide has fallen along the lines of clinical and research applications: structural imaging is involved in the former and functional imaging is concerned with the latter. However, with developing research, the applicability of functional modalities such as fMRI is continually expanding. In fact, it is not unreasonable to envision a time in which functional neuroimaging could yield critical information about a patient’s specific diagnosis or the likelihood of a patient responding to certain therapeutic interventions. This review discusses the indications for structural imaging in patients presenting with psychiatric symptoms. Following brief descriptions of currently available functional neuroimaging modalities, the clinical and research utility of functional neuroimaging in psychiatric populations is discussed.

INDICATIONS FOR STRUCTURAL IMAGING
To rule out comorbidities

One large analysis across diverse populations found evidence of cortical atrophy in 30% of psychiatric patients who underwent CT imaging.² In another study of 253 patients who presented with psychiatric symptoms, 15% had a change in their treatment regimen as a result of undergoing a structural brain MRI.³ However, the question of whether and when it is worthwhile to image patients with psychiatric symptoms remains unresolved.

Multiple studies have attempted to establish guidelines for the indications for neuroimaging in psychiatry. In one study, the presence of focal neurological signs and advanced patient age were the only reliable predictors of abnormalities on imaging in psychiatric inpatients.⁴ Dougherty and Rauch¹ have proposed guidelines for structural neuroimaging in psychiatric populations. They suggest imaging for patients with abrupt changes in mental status associated with 1 of 3 criteria:

• Age over 50.

• Abnormal findings on neurological examination.

• A history of significant head trauma.

They also include new-onset psychosis and new-onset delirium of unknown cause as criteria for neuroimaging. In addition, they recommend structural imaging before an initial course of electroconvulsive therapy.¹

CT has the following advantages compared with MRI: faster acquisition time and no contraindications in patients who have metallic implants. However, in the absence of contraindications and strict time constraints, MRI is the preferred modality because it provides better differentiation of gray from white matter, better evaluation of white matter pathology, better overall spatial resolution, and better ability to detect pathology in the posterior fossa.

As a primary tool to diagnose psychiatric illness

Several reports have indicated mild structural abnormalities associated with neuropsychiatric diseases. MRI has been used in Alzheimer disease to establish volume loss in critical medial temporal lobe structures (such as the hippocampus and the entorhinal cortex), as well as to predict progression from mild cognitive impairment to Alzheimer disease.^5,6

In schizophrenia, common structural changes include enlargement of the lateral and third ventricles and volume loss in the dorsolateral prefrontal cortex, the medial temporal lobe, the thalamus, and the superior temporal gyrus.^7,8 There are less consistent findings of changes in cortical volume in patients with mood disorders. Despite the above discoveries, findings on structural imaging are too variable and nonspecific to be used in isolation to diagnose psychiatric disorders.

FUNCTIONAL NEUROIMAGING MODALITIES

PET can be used to assess cerebral blood flow and cerebral glucose metabolism and to characterize neurotransmitter receptors. This technology uses injected unstable isotopes (eg, ¹⁸F, ¹⁵O, ¹¹C) that emit positrons, which, in turn, collide with electrons to produce gamma ray radiation. The PET scanners detect this gamma ray radiation and the resulting information is fed to a computer, which produces an image. A common tracer used to measure cerebral glucose metabolism is ¹⁸F fluorodeoxyglucose. ¹⁵O-labeled H₂O or CO₂ is the tracer traditionally used in the assessment of cerebral blood flow. In addition to its usefulness in assessing cerebral blood flow and metabolism, PET remains the gold standard in studies of neurotransmitter receptors and transporters. Several radioligands are available for PET characterization of different receptors, including dopamine, serotonin, benzodiazepine, and opioid receptors.

SPECT is used to image regional cerebral blood flow that

reflects cerebral metabolic activity. Like PET, SPECT scanning uses radiation from unstable isotopes to construct images. Unlike PET, SPECT does not yield a direct measurement of cerebral glucose metabolism. In addition, because of the technical differences between positron emission and single-photon emission, SPECT has slightly poorer spatial resolution than PET.

fMRI uses MRI machines with specific acquisition parameters and higher-speed scanning to assess cerebral blood flow and cerebral blood volume. fMRI accomplishes this by detecting changes in the paramagnetic properties of hemoglobin. This technique produces blood oxygen level–dependent signals of blood flow that are tightly coupled with neuronal activity. fMRI has a slightly better spatial resolution and a far better temporal resolution than either PET or SPECT (Figure).

MRS uses special MRI acquisition parameters to quantify various chemical substances within select brain areas (or regions of interest). Traditional molecular signatures measured with MRS include N-acetylaspartate, creatine, choline, and lactate. The quantification of these chemicals and the quantification of the ratios of one chemical to another yield information that can be clinically useful. For example, N-acetylaspartate levels are a marker of neuronal integrity, lactate levels are a measure of anaerobic metabolism, and creatine and choline levels can be used to assess cellular membrane turnover. MRS has therefore been used in certain neurological conditions, such as multiple sclerosis, CNS lymphoma, and mitochondrial disorders. In psychiatry, recent research has made possible the assessment of drug concentrations in the brain (eg, lithium and fluoxetine) by MRS.^9,10

DT-MRI uses yet another special set of MRI acquisition parameters to enable reconstruction of white matter tracts and assessment of white matter tract integrity.¹¹ Clinically, this technology has been used to gauge the integrity of white matter pathways (eg, in patients with diffuse axonal injury). In research settings, DT-MRI can be used to anatomically map the trajectory of white matter bundles.

FUNCTIONAL NEUROIMAGING IN PSYCHIATRYDementing disorders

The most common use of functional neuroimaging in neuropsychiatry is in the evaluation of dementia. Cerebral perfusion studies (typically using SPECT) and cerebral glucose metabolism studies (using PET) have become an essential part of the diagnostic armamentarium for evaluating patients with dementia. Various dementias often show a characteristic pattern of hypoperfusion. For example, Alzheimer disease is associated with bilateral hypoperfusion in temporoparietal areas. Subdivisions of the frontotemporal dementias can also show differing patterns of cerebral hypoperfusion and hypometabolism. fMRI has recently been used to augment the ability to distinguish between healthy subjects and those with Alzheimer disease. Gazzaley and Small¹² cite numerous studies that have detected blood oxygen level–dependent fMRI signal differences in the hippocampi of subjects with Alzheimer disease and mild cognitive impairment compared with age-matched controls. Thus, fMRI has promise in detecting preclinical Alzheimer disease in those with memory complaints.

Interesting work is also being done with the use of DT-MRI. Persson and colleagues¹³ have shown a possible role for DT-MRI in predicting the progression to Alzheimer disease in those who are genetically susceptible. For instance, aberrations have been shown in white matter integrity in the posterior corpus callosum and the medial temporal lobe in nondemented apolipoprotein e4 homozygotes but not in persons who do not carry the gene.

Affective disorders

Several strategies are available to investigate the neural basis of mood disorders with functional neuroimaging. While being imaged, subjects perform a cognitive activation paradigm (such as a test of working memory). Alternatively, subjects can be imaged during a mood challenge paradigm, in which mood states are induced by emotionally valenced pictures, by reading emotionally laden vignettes, or by recalling emotional autobiographical events. By use of mood challenge and cognitive paradigms, areas of activation can be compared between patients with mood disorders and matched normal controls.¹⁴

Mayberg and colleagues¹⁵ have put forward an influential view in which depression results from a dysfunction of limbic-cortical circuits. In their model, cognitive and attentional disturbances in depression result from decreased functioning of the dorsal compartment (the dorsolateral prefrontal cortex, the dorsal anterior cingulate, and the posterior cingulate), whereas vegetative and emotional symptoms result from increased activity in the ventral compartment (the subgenual anterior cingulate, the ventral prefrontal cortex, the insula, the hippocampus, and the amygdala).^14,16

The investigators also describe a rostral compartment (consisting of the rostral anterior cingulate) that serves to regulate the other 2 compartments in this circuit. Neuroimaging studies appear to be consistent with their model. For example, numerous studies of cerebral glucose metabolism and regional cerebral blood flow have shown hypometabolism and hypoperfusion in the dorsal compartment prefrontal areas.¹⁵

fMRI studies in which sad states are induced in healthy controls also show decreased activity in these areas, specifically the dorsal anterior cingulate cortex and the dorsolateral prefrontal cortex.¹⁷ The same studies have also shown increased activity in ventral compartment areas, specifically the subgenual prefrontal, the pregenual anterior cingulate, and the ventral prefrontal cortices and the insula.¹⁷ The amygdala is also implicated in depression, and the intensity of amygdala activation in PET has been correlated with the severity of depression.¹⁸ Interestingly, other research has shown normalization of frontal metabolism with the treatment of depression.¹⁹

Involvement of specific brain areas in mania and hypomania are less well characterized than they are in unipolar depression.¹⁴ Functional imaging studies of individuals with bipolar illness have demonstrated dysfunction in frontal and striatal regions. In the resting state, there appears to be decreased regional cerebral blood flow in the ventromedial and orbitofrontal cortex.²⁰ Cognitive activation tasks have shown alterations in the recruitment of ventral prefrontal, dorsolateral prefrontal, and dorsal anterior cingulate regions.¹⁷ In the basal ganglia, increased activation during manic episodes has been reported in the caudate head.²¹ MRS studies have also implicated anomalies in second messenger systems in prefontal and striatal areas in bipolar patients.²² Specifically, such studies have revealed abnormalities in the concentration of choline and myoinositol (both critical in second messenger signaling) in the brains of bipolar patients.^23,24

Anxiety disorders

Models of the pathophysiology of obsessive-compulsive disorder (OCD) have implicated a cortical-striatal circuit, in which projections from the orbitofrontal and cingulate cortex to the striatum synapse in the globus pallidus and thalamus on their way back to the frontal cortex.¹⁴ Functional neuroimaging studies have revealed increased regional brain activity in the orbitofrontal cortex, caudate, thalamus, and anterior cingulate cortex in patients with OCD, both at rest and during symptom provocation.^14,25,26 These findings are consistent with theories that posit deficient gating of striatal input at the thalamic level causing increased activity in both the orbitofrontal cortex and anterior cingulate cortex in patients with OCD.^17,26

Functional imaging studies in patients with panic disorder that use lactate to induce panic attacks have shown widespread abnormalities in monoaminergic projection systems as well as more focal abnormalities in the medial temporal lobe.¹⁷ Studies of specific phobias have yielded more inconsistent results but suggest involvement of the medial temporal lobe.¹⁷

The functional imaging of posttraumatic stress disorder (PTSD) has demonstrated exaggerated amygdala activation and decreased activation of the medial frontal lobe during symptom provocation studies.^14,27 In addition, the extent of amygdala activation has been correlated with the severity of symptoms in PTSD.¹⁴

Schizophrenia

Schizophrenia is associated with deficits in multiple cognitive functions. Thus, cognitive activation paradigms can be used to elucidate deficits in patients with schizophrenia as they undergo functional neuroimaging. The most consistent brain area implicated in the functional neuroimaging of schizophrenia is the dorsolateral prefrontal cortex. Working memory tasks (such as the N-back test) have consistently demonstrated reduced activation on PET and fMRI in the dorsolateral prefrontal cortex of patients with schizophrenia compared with healthy controls.^28,29 Reduced dorsolateral prefrontal cortex activation has also been found in tests of mental flexibility and set shifting, most notably the Wisconsin Card Sort Test (which typically activates the dorsolateral prefrontal cortex in healthy individuals).^7,29

Other studies have discovered abnormalities in prefrontal activation during procedural learning tasks and tasks of skill automatization.^7,30 However, Callicott⁷ has cited some studies in which dorsolateral prefrontal cortex activation is actually increased. In addition to investigating dysfunction in specific brain regions, aberrant functional connectivity between regions has been investigated using fMRI. Such studies have shown reduced coupling between medial temporal and prefrontal regions in schizophrenic patients.³¹ Other neuroimaging studies have probed the underpinnings of positive symptoms in schizophrenia. Studies of patients actively experiencing auditory hallucinations have had variable results, including reduced temporal lobe activation,³² a decrease in the response of the primary auditory cortex to external auditory stimuli,¹⁰ and an activation of primary auditory cortex during subjective reporting of auditory hallucinations.³³

As mentioned above, PET studies have the particular advantage of characterizing neuroreceptors through the use of radioligands. This technique has been thoroughly applied to the study of dopamine receptor binding in schizophrenia. PET and SPECT studies in schizophrenia generally fall into 1 of 2 categories: clinical studies, in which neurotransmitter binding is examined between schizophrenic patients and healthy controls; and occupancy studies that probe the action of neuroleptics in terms of their receptor binding.³⁴ Such studies have suggested overactivity or increased density of striatal D₂ dopamine receptors and underactivity or decreased density of cortical D₁ receptors.³⁵ Recent research has also implicated aberrant glutamatergic transmission in N-methyl d-aspartate receptors in the pathogenesis of the disease.^35,36

There is also an increasing amount of literature involving the use of DT-MRI in schizophrenic patients. Such studies have suggested abnormalities in white matter tracts that connect prefrontal and temporal areas, including the uncinate fasciculus, the cingulum bundle, and the arcuate fasciculus.³⁷ Although some authors have argued that there are inconsistent findings in such studies,³⁸ many believe that with the proper methodological refinements, DT-MRI can be a useful tool in understanding the pathophysiology of schizophrenia.

Finally, Callicott⁷ and others have suggested a role for functional neuroimaging in psychiatric genetic studies. By using functional imaging, an intermediate phenotype can be identified in individuals who are genetically similar to patients with schizophrenia. In this way, if abnormalities detected during the functional neuroimaging of patients with schizophrenia are also found in their healthy relatives, that functional abnormality can be inferred to be heritable. For example, Callicott and colleagues³⁹ found that cognitively normal siblings of patients with schizophrenia showed exaggerated activation of the right dorsolateral prefrontal cortex during a working memory task. Although the siblings appeared cognitively intact, they showed an activation pattern that was different from that of healthy controls and similar to that of patients with schizophrenia. It is hoped that such intermediate phenotypes can pave the way for the continued discovery of susceptibility genes for schizophrenia.

CONCLUSIONS

Structural neuroimaging is sometimes indicated to rule out an occult medical or neurological disorder in patients who present with psychiatric symptoms. Although formal guidelines have yet to be established, many investigators agree that acute alterations in mental status, the presence of focal neurological signs, and a history of significant head trauma or epilepsy are all indications for neuroimaging. When imaging is warranted, MRI is the preferred modality unless it is contraindicated. Although abnormalities on structural imaging have been associated with many psychiatric diseases, these changes are both too subtle and too variable to be used as reliable diagnostic tools. PET, SPECT, fMRI, and other techniques have been invaluable research tools to identify regional brain abnormalities in psychiatric disease. In the future, these technologies may also guide clinical diagnosis of psychiatric conditions and the choice of appropriate treatment.

References:

References

1. Dougherty DD, Rauch SL. Brain imaging in psychiatry. In: Dougherty DD, Rauch SL, eds. Essentials of Neuroimaging for Clinical Practice. Arlington, Va: American Psychiatric Publishing; 2004:1-10.
2. Renshaw PF, Rauch SL. Neuroimaging in clinical psychiatry. In: Nicholi AM Jr, ed. The Harvard Guide to Psychiatry. 3rd ed. Cambridge, Mass: Belknap Press; 1999:84-97.
3. Erhart SM, Young AS, Marder SR, Mintz J. Clinical utility of magnetic resonance imaging radiographs for suspected organic syndromes in adult psychiatry.
J Clin Psychiatry. 2005;66:968-973.
4. Mueller C, Rufer M, Moergeli H, Bridler R. Brain imaging in psychiatry-a study of 435 psychiatric in-patients at a university clinic. Acta Psychiatr Scand. 2006;114:91-100.
5. Jack CR, Petersen RC, Xu Y, et al. Rates of hippocampal atrophy correlate with change in clinical status in aging and AD. Neurology. 2000;55:484-489.
6. Mungas D, Reed BR, Jagust WJ, et al. Volumetric MRI predicts rate of cognitive decline related to AD and cerebrovascular disease. Neurology. 2002;59: 867-873.
7. Callicott JH. Schizophrenia. In: D’Esposito M, ed. Functional MRI: Applications in Clinical Neurology and Psychiatry. United Kingdom: Informa Healthcare Publishing; 2006:181-195.
8. Wright IC, Rabe-Hesketh S, Mellers J, et al. Meta-analyses of regional brain volumes in schizophrenia. Am J Psychiatry. 2000;157:16-25.
9. Renshaw PF, Wicklund S. In vivo measurement of lithium in humans by nuclear magnetic resonance spectroscopy. Biol Psychiatry. 1988;23:465-475.
10. David AS, Woodruff PW, Howard R, et al. Auditory hallucinations inhibit exogenous activation of auditory association cortex. Neuroreport. 1996;7:932-936.
11. Catani M. Diffusion tensor magnetic resonance imaging tractography in cognitive disorders. Curr Opin Neurol. 2006;19:599-606.
12. Gazzaley A, Small S. Alzheimer’s disease. In:
D’Esposito M, ed. Functional MRI: Applications in Clinical Neurology and Psychiatry. United Kingdom: Informa Healthcare Publishing; 2006:25-34.
13. Persson J, Lind J, Larsson A, et al. Altered brain white matter integrity in healthy carriers of the APOE epsilon4 allele: a risk for AD? Neurology. 2006;66: 1029-1033.
14. Deckersbach T, Dougherty DD, Rauch SL. Functional imaging of mood and anxiety disorders. J Neuroimaging. 2006;16:1-10.
15. 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;156:675-682.
16. Mayberg HS. Limbic-cortical dysregulation: a proposed model of depression. J Neuropsychiatry Clin Neurosci. 1997;9:471-481.
17. Deckersbach T, Dougherty DD, Rauch SL. Mood and anxiety disorders. In: D’Esposito M, ed. Functional MRI: Applications in Clinical Neurology and Psychiatry. United Kingdom: Informa Healthcare Publishing; 2006:115-135.
18. Abercrombie HC, Schaefer SM, Larson CL, et al. Metabolic rate in the right amygdala predicts negative affect in depressed patients. Neuroreport. 1998;9:
3301-3307.
19. Brody AL, Saxena S, Silverman DH, et al. Brain metabolic changes in major depressive disorder from pre- to post-treatment with paroxetine. Psychiatry Res. 1999;91:127-139.
20. Rubin E, Sackeim HA, Prohovnik I, et al. Regional cerebral blood flow in mood disorder: IV. Comparison of mania and depression. Psychiatry Res. 1995;61:
1-10.
21. Blumberg HP, Stern E, Martinez D, et al. Increased anterior cingulate and caudate activity in bipolar mania. Biol Psychiatry. 2000;48:1045-1052.
22. Strakowski SM, Delbello MP, Adler CM. The functional neuroanatomy of bipolar disorder: a review of neuroimaging findings. Mol Psychiatry. 2005;10:
105-116.
23. Malhi GS, Valenzuela M, Wen W, Sachdev P. Magnetic resonance spectroscopy and its applications in psychiatry. Aust N Z J Psychiatry. 2002;36:31-43.
24. Silverstone PH, McGrath BM, Kim H. Bipolar disorder and myo-inositol: a review of the magnetic resonance spectroscopy findings. Bipolar Disord. 2005;
7:1-10.
25. Breiter HC, Rauch SL, Kwong KK, et al. Functional magnetic resonance imaging of symptom provocation in obsessive-compulsive disorder. Arch Gen Psychiatry. 1996;53:595-606.
26. Rauch SL, Whalen PJ, Dougherty DD, et al. Neurobiological models of obsessive compulsive disorders. In: Jenike MA, Baer L, Minichiello WE, eds. Obsessive-Compulsive Disorders: Practical Management. Boston: Mosby; 1998:222-253.
27. Rauch SL, van der Kolk BA, Fisler RE, et al. A symptom provocation study of posttraumatic stress disorder using positron emission tomography and script-driven imagery. Arch Gen Psychiatry. 1996; 53:380-387.
28. Callicott JH, Bertolino A, Mattay VS, et al. Physiological dysfunction of the dorsolateral prefrontal cortex in schizophrenia revisited. Cereb Cortex. 2000;
10:1078-1092.
29. Riehemann S, Volz HP, Stutzer P, et al. Hypofrontality in neuroleptic-naive schizophrenic patients during the Wisconsin Card Sorting Test-an fMRI study. Eur Arch Psychiatry Clin Neurosci. 2001; 251:66-71.
30. Jansma JM, Ramsey NF, Slagter HA, et al. Functional anatomical correlates of controlled and automatic processing. J Cogn Neurosci. 2001;13:730-743.
31. Stephan KE, Baldeweg T, Friston KJ. Synaptic plasticity and dysconnection in schizophrenia. Biol Psychiatry. 2006;59:929-939.
32. Van de Ven VG, Formisano E, Röder CH, et al. The spatiotemporal pattern of auditory cortical responses during verbal hallucinations. Neuroimage. 2005;27: 644-655.
33. Dierks T, Linden DE, Jandl M, et al. Activation of Heschl’s gyrus during auditory hallucinations. Neuron. 1999;22:615-621.
34. Frankle WG, Laruelle M. Neuroreceptor imaging in psychiatric disorders. Ann Nucl Med. 2002;16:437-446.
35. Abi-Dargham A, Laruelle M. Mechanisms of action of second generation antipsychotic drugs in schizophrenia: insights from brain imaging studies. Eur Psychiatry. 2005;20:15-27.
36. Abou-Saleh MT. Neuroimaging in psychiatry: an update. J Psychosom Res. 2006;61:289-293.
37. Kubicki M, McCarley R, Westin CF, et al. A review of diffusion tensor imaging studies in schizophrenia. J Psychiatr Res. 2007;41:15-30.
38. Kanaan RA, Kim JS, Kaufmann WE, et al. Diffusion tensor imaging in schizophrenia. Biol Psychiatry. 2005;58:921-929.
39. Callicott JH, Egan MF, Mattay VS, et al. Abnormal fMRI response of the dorsolateral prefrontal cortex in cognitively intact siblings of patients with schizophrenia [published correction appears in Am J Psychiatry. 2004;161:1145]. Am J Psychiatry. 2003;160: 709-719.