Chapter 10 Brain Imaging in Psychopharmacology

Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

1 of 26

10/05/2009 15:53

Print Close Window

Ebrahim Haroon, Giuseppe Pagnoni, Christine M. Heim, Gregory S. Berns, Helen S. Mayberg: Chapter 10. Brain Imaging in

Psychopharmacology, in The American Psychiatric Publishing Textbook of Psychopharmacology, 4th Edition. Edited by Alan F.

Schatzberg, Charles B. Nemeroff. Copyright ©2009 American Psychiatric Publishing, Inc. DOI:

10.1176/appi.books.9781585623860.426202. Printed 5/10/2009 from www.psychiatryonline.com

Textbook of Psychopharmacology >

Chapter 10. Brain Imaging in Psychopharmacology

BRAIN IMAGING IN PSYCHOPHARMACOLOGY: INTRODUCTION

Functional brain imaging refers to a class of techniques that noninvasively measure correlates of neural

activity. Positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) are the

two technologies most commonly used today to study the human brain “in action.” The explosion of

information about human brain function occurring in the past decade has resulted in large part from

these two techniques. As will be described in this chapter, PET imaging has made considerable

contributions to our understanding of the mechanisms of drug action, mostly through application of

radiopharmaceutical labeling of neurotransmitter receptors. fMRI, on the other hand, has gained rapid

acceptance because of the widespread availability of magnetic resonance imaging (MRI) scanners, the

lack of radioactive exposure, and the better image resolution offered.

The advent of neuroimaging techniques for probing in vivo human brain function undoubtedly represents

a major milestone in the scientific endeavor of understanding the relationship between mental disorders

and the brain. The development of the specific tools employed in brain mapping, although fairly recent,

has already produced an impressive amount of experimental data, whose potential informational content

is most likely being underexploited at the present time (Van Horn and Gazzaniga 2002). The

neuroimaging approach offers the unique possibility of noninvasively investigating the

neurophysiological, neuroanatomical, and neurochemical correlates of the living, performing human

individual. As a complement to classical neuropsychology, this approach represents an unprecedented

break from the necessity of lesion studies for the inference of structure–function relationships in the

human brain. Furthermore, a neuroimaging assay can typically acquire data simultaneously from the

entire cerebral system, thereby allowing the study of the distributed processing properties of the brain

(Friston 2002). Those properties represent a fundamental and distinctive feature of massively parallel

processing systems but are not easily penetrated with the standard neurophysiological methods

employed in the animal, such as intracortical electrode recording, which can probe only a limited number

of sites simultaneously.

[Portions of this chapter are reprinted from Berns GS: “Functional Neuroimaging.” Life Sciences

65:2531–2540 1999. Copyright 1999, Elsevier Science. Used with permission.]

NEUROIMAGING TECHNIQUES

Neuroimaging is a generic term for a number of techniques and methods aimed at detecting meaningful

information through the acquisition of brain images of different kinds. The presentation of different

classification criteria may help the reader to get a sense of this rich but often confounding landscape. A

first classification may be technology based:

MRI

PET

Single photon emission computed tomography (SPECT)

Electroencephalography (EEG)

Magnetoencephalography (MEG)

Each of the above technologies requires different hardware, and each measures different physical

quantities. It should be noted that whereas PET and SPECT employ radioactive tracers that limit their

repeated use on the same subject, the other techniques are noninvasive and therefore allow more

latitude in regard to experimental design and the subsequent statistical treatment of the data. MRI and

PET/SPECT scanners are able to implement different imaging protocols, according to the specific

acquisition modality employed (MRI) or the nature of the injected radioisotope (PET/SPECT). More

recently, the availability of combined PET–computed tomography (PET–CT) and PET–MRI scanners hasPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

2 of 26

10/05/2009 15:53

yielded an added dimension to our capacity to understand human brain in action (Blodgett et al. 2007).

The neurally related variable that is actually imaged may provide an alternative classification framework:

Vascular (or hemodynamic) effects engendered by neural activity: PET (H2 15O), fMRI (blood oxygenation

level–dependent [BOLD] contrast, perfusion imaging), SPECT

Metabolic demand: PET (18fluorodeoxyglucose)

Receptor density: PET/SPECT (radioligands)

Neurochemistry: Magnetic resonance spectroscopy (MRS)

Connectivity pathways: MRI (diffusion tensor imaging), fMRI-based functional connectivity analysis

Surface electromagnetic effects of brain activity: EEG/MEG

Morphometry of brain structures: MRI

We begin this chapter with current thoughts on the physiological basis of functional imaging and the

phenomenon of neural activation it attempts to study (referred to as “neurovascular coupling”) and then

proceed to examine the two major functional imaging modalities of relevance to psychiatry—PET and

fMRI. The section on fMRI will also incorporate basic principles of MRI. We end the chapter with a brief

review of magnetic resonance (MR)–based structural imaging modalities and how they might confer

newer insights into the understanding of mental illness and development of newer treatments. The focus

is intended to be on how these imaging methods are used to better understand psychopharmacology.

More details on individual disorders will be available in their respective sections.

FUNCTIONAL IMAGING AND CORRELATES OF NEURAL ACTIVITY

It has been known for more than 100 years that blood flow to the brain increases in a regionally specific

manner, according to mental activity. The father of modern psychology, William James, was aware of

observations relating regional brain pulsation to mental activity (James 1890). Paul Broca, known

primarily for his observations on the effects of left frontal lesions on language, performed several

experiments relating regional brain temperature to cognitive function (Broca 1879). But it was not until

the 1950s, when Seymour Kety and Louis Sokoloff developed the autoradiographic technique for

quantitatively measuring regional blood flow, that specific cognitive functions could be directly mapped

in the living brain (Kety 1965).

Both PET and fMRI rely on the fact that blood flow increases in areas where neuronal activity increases,

and most studies implicitly assume the validity of this relationship. It is easy to see the link in terms of

an increased metabolic demand. The activation of a neural circuit is a complex network of

electrochemical processes that requires energy. The most demanding processes, in terms of energy

expenditure, are those related to synaptic activity, which results in breakdown of energy stores in the

neuron in the form of adenosine triphosphate (ATP) molecules. To replace the ATP energy stores

degraded by the increased metabolic demand, a surge in the concentration of glucose and oxygen is

necessary, which results in increased blood flow to the activated region. It is this vascular or

hemodynamic response to neural activity—that is, the variation in regional cerebral blood flow

(rCBF)—that is the quantity actually measured in the majority of brain activation studies with fMRI and

PET/SPECT (Arthurs and Boniface 2002; Jueptner and Weiller 1995). Thus, the hemodynamic response

represents an indirect assay of neural activity (Villringer and Dirnagl 1995). It is important to note that

the hemodynamic response lags behind the actual neural activity by a few seconds. The hemodynamic

response is also blurred in the spatial domain compared with the underlying neural activity. This imposes

limits on the spatiotemporal resolution of blood flow methods, independently of technology

improvement.

A special caveat should be noted concerning the interpretation of rCBF results. The measurement of

task-related variations of rCBF does not provide any clear indication about the nature of the underlying

neural activity (i.e., whether it is excitatory or inhibitory), although hypotheses have been proposed for

and argued against a bias favoring excitatory contributions (Heeger et al. 1999; Tagamets and Horwitz

2001; Waldvogel et al. 2000). The construction of specific inferences about the actual state of

activity—actively excited or actively inhibited—of brain regions showing an increase in rCBF during an

experimental task would require the integration of information from many different sources (e.g.,

electrophysiology, neurochemistry, cytoarchitectonics).

In summary, despite the fact that the physiological and biochemical processes linking the neural activity

and the hemodynamic response have not been clarified yet, the empirical relationship between thesePrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

3 of 26

10/05/2009 15:53

parameters appears both reliable and reproducible in a variety of contexts. The validity of fMRI

measurements of signal change as an assay of neural activity has been documented using

electrophysiological techniques such as neuronal field potentials (Logothetis et al. 2001).

POSITRON EMISSION TOMOGRAPHY

PET was developed from in vivo autoradiographic techniques. In an autoradiographic procedure, an

animal is typically injected with a biologically interesting compound synthesized with a radioisotope

(e.g., 3H). When the animal is sacrificed, the local tissue radioactivity is easily quantified. Although

autoradiography yields exquisitely detailed pictures of brain activity, it can only be applied in animals,

and the animals must be sacrificed to obtain the brain tissue. Although these techniques were in use in

the 1950s and 1960s, the development of an in vivo method applicable to humans awaited technological

advancements from the “silicon revolution”—namely, the availability of high-quality inexpensive crystal

detectors and the huge advancements in computing power realized in the late 1970s.

PET requires three basic technologies: the production of positron-emitting compounds, the ability to

detect simultaneously emitted gamma rays, and the computational power to reconstruct the sources of

emission. Positrons, or positively charged electrons (antimatter), have a particular advantage over other

radioactive compounds. When a positron encounters an electron, the two annihilate each other, and their

collective energy is transformed into two high-energy photons that are emitted in exactly opposite

directions. Because the photons travel 180° apart, it is easy to arrange a ring of detectors to determine

where the annihilation occurred. When two detectors are activated simultaneously, then one knows that

the emission occurred somewhere along the line connecting the two detectors. By collecting the counts

over a period of time, say 60 seconds, and over a full sphere surrounding the subject’s head, it becomes

possible to reconstruct the geometry of the source.

Positrons are produced indirectly, through the radioactive decay of particular isotopes. The most

commonly used isotopes (carbon-11 [11C], oxygen-15 [15O], fluorine-18 [18F], nitrogen-13 [13N]) are

produced in a cyclotron by the bombardment of targets with high-energy protons. This results in a gas

(e.g., 15O2), which can then be used in any chemical reaction (e.g., oxidation–reduction reaction with

product H2 15O). After appropriate purification procedures, these compounds can then be injected

intravenously into a human subject, and they flow to the brain in about 20 seconds. The isotope

undergoes radioactive decay by positron emission, and the half-life depends on the particular isotope

(e.g., 2 minutes for 15O).

Because the photons emitted during positron decay are fairly high in energy (511-keV gamma rays), they

tend to pass through matter with relative ease. A specialized detector, called a scintillation detector, is

required to accurately count the decays in a directional fashion. PET scanners consist of rings of these

detectors arranged in parallel planes. An individual detector would be constructed from a scintillating

crystal, either bismuth germanate (BGO) or lutetium oxyorthosilicate (LSO), and amplification

electronics. When a gamma ray enters the crystal, it loses its energy through either the photoelectric or

Compton effect, which results in the production of electrons. These electrons further interact with the

crystal, resulting in the production of visible wavelength photons. These photons are then detected and

amplified by a photomultiplier tube and converted into an electrical pulse. A “coincidence circuit” allows

for the identification of the detector that picks up the 180°-emitted gamma ray.

Depending on the injected molecule, a particular regional distribution will occur. In the case of H2 15O, it

will follow the rCBF. Other compounds will cross the blood–brain barrier and bind to specific receptors, in

which case the distribution of radioactivity will reflect receptor concentration. Fluorodeoxyglucose-18

( 18FDG), a commonly used tracer, is metabolized by hexokinase during glycolysis, like glucose. Unlike

glucose-6 phosphate, 18FDG is not metabolized further and thus accumulates intracellularly, yielding a

measurement of local metabolic activity (Kennedy et al. 1976; Reivich et al. 1979).

Types of PET Studies

Most PET neuroimaging studies can be grouped into one of three categories: metabolic, blood flow, or

receptor studies.

Metabolic PET Studies

Metabolic studies use 18FDG to measure regional glucose metabolism. 18FDG, like all 18F compounds, has Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

4 of 26

10/05/2009 15:53

the advantage of a relatively long half-life (110 minutes). This allows the synthesis to be performed in

one location, the subject injection in another location, and the scanning in yet another location. In fact,

one can have a subject doing a particular task in a location remote from the PET scanner and inject

18FDG, which will be trapped in brain regions according to the local metabolic rate. This has an obvious

advantage in situations in which placing the subject in the scanner would alter the conditions of the task.

For example, 18FDG is used commonly in sleep studies. The main disadvantage is that the long half-life

results in effectively no temporal resolution. This offers a time-averaged snapshot of a particular brain

state, and the state is averaged over 20–60 minutes. Figure 10–1 shows functional localization of an

epileptic focus in the right temporal lobe during presurgical workup.

FIGURE 10–1. Functional localization of an epileptic focus in the right temporal lobe during presurgical

workup using magnetic resonance imaging (MRI) and fluorodeoxyglucose positron emission

tomography (FDG-PET).

Source. Image courtesy of Carolyn C. Meltzer, M.D.

Most of the FDG uptake studies are based on the assumption that the glucose uptake and neural activity

at the synaptic level might be coupled. A caveat must be borne in mind when evaluating studies using

FDG-PET. Glutamate is the main excitatory neurotransmitter in the brain, and it is removed from the

synapse through a process of uptake by astroglial tissues, thus terminating neural activation (Magistretti

and Pellerin 1999). But recent studies have shown that uptake of glutamate by astroglia can by itself

stimulate glucose (and FDG) uptake (Magistretti 2006). In fact, deactivation might actually be coupled

with increased glucose uptake in a variety of conditions (Magistretti 2006). Thus, the same problems

that accompany studies of fMRI—i.e., whether the signal is actively excitatory versus actively

inhibitory—are present in FDG-PET studies as well.

Blood Flow PET Studies

Blood flow studies use H2 15O to measure changes in local brain blood flow (Herscovitch et al. 1983;

Mintun et al. 1984). As noted earlier, blood flow is an indirect measure of local synaptic activity. Because

15O has a short half-life (2 minutes), several administrations can be performed in one session. A typical

H2 15O study would have 8–16 injections and scans for each subject. The experimental design would

manipulate the task that the subject performs during each scan. Each scan lasts about 1 minute, with

8–10 minutes between scans (5 half-lives). H2 15O studies not only allow for multiple conditions to be

studied, but they also allow for the repetition of conditions, increasing statistical power. The main

disadvantage is that because of the short half-life, the H2 15O must be produced reliably and in close

proximity to the scanner.

Receptor-Mapping PET StudiesPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

5 of 26

10/05/2009 15:53

Receptor studies use radioligands—chemicals incorporating a positron-emitting isotope into a molecule

whose pharmacokinetics are already known. Ideally, these ligands bind specifically to one receptor type.

Most of these studies are of the mapping type, which shows the distribution of a particular receptor in

the brain (e.g., dopamine type 2 [D2] receptor). Here, the measured radioactivity reflects both the local

concentration of receptors (Bmax) and the affinity of the ligand for the receptor (measured by KD, the

equilibrium dissociation constant). If the ligand acts as a competitive antagonist, then the apparent

affinity is also affected by the concentration of the endogenous neurotransmitter. The analysis can be

simplified by considering the ratio of Bmax to KD, termed the binding potential (BP). Ligands undergo

both specific and nonspecific binding. Typically, one is interested only in the specific binding (i.e., to the

receptor of interest). Use of a reference tissue that is known to have a low receptor concentration allows

one to subtract out the nonspecific binding (e.g., the cerebellum has a low concentration of D2

receptors). In this case, the difference in distribution for the two tissues is directly proportional to the

1. Ligands require a more involved synthesis than either water or 18FDG, and their use is a race against

the clock as the isotope decays. The end product must meet several requirements: high specific activity

(amount of radioactivity per mole), high radiochemical purity, and sterility. 18F ligands are easier to

synthesize because of their long half-life, but 11C ligands (20-minute half-life) have a higher potential for

biological relevance. These will be discussed in the next section, which addresses psychopharmacological

applications of PET tracers and ligands.

Clinical Applications of PET and SPECT in Psychopharmacology

PET-based receptor imaging has provided us with a window to view the complex functioning of brain

systems involved in mediating treatment response to psychopharmacological agents. A snapshot of

available radionuclide-binding modalities is provided in Table 10–1. Using PET-based radioligands, it has

been possible to visualize density, distribution, and occupancy of neural receptors and transporters

before, during, and after drug therapy (Talbot and Laruelle 2002). Details about use of neuroimaging

methods in individual psychiatric disorders will be described in detail in later chapters pertaining to

these disorders. In the following sections, specific examples will be provided to illustrate the use of

imaging methods to answer pertinent questions of relevance to psychopharmacology.

TABLE 10–1. Pharmacoimaging in psychiatry

Imaging modality Labeling agent Binding site Clinical focus and type of

pharmacological probe

Dopamine

PET

[ 18F] DOPA

Aromatic L-amino

acid decarboxylase

Viability of

dopamine-synthesizing neurons

Probe type: enzyme labeling

ligand

PET

[ 11C] methylphenidate, [11C]

cocaine, [123I] -CIT, [11C]

WIN 35428

Dopamine uptake

receptor

Synaptic dopamine availability

and correlation with cognition

Probe type: reuptake transporter

ligand

PET

[ 11C] raclopride, [11C] FLB 457,

[ 18F] fallipiride, [11C] NPA

D2 receptor Binding and affinity and

occupancy of D2 receptors by

antipsychotics

Probe type: postsynaptic (PS)

receptor ligand

PET

[ 11C] NNC-112, [11C] SCH

23390

D1 receptor Role of dopamine in cognition

Probe type: PS receptor ligand

SPECT

[ 123I] iodobenzamide

D2 receptor Hyperresponse of dopamine

secretion in schizophrenia

Probe type: PS receptor ligand

SerotoninPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

6 of 26

10/05/2009 15:53

Imaging modality Labeling agent Binding site Clinical focus and type of

pharmacological probe

PET

[ 11C] methyl-L-tryptophan

5-HT synthesis A marker of 5-HT biosynthesis

Probe type: precursor ligand

PET

[ 18F] setoperone, [18F]

ketanserin, [18F] altanserin

5-HT2A receptor Serotonin turnover among

suicidal and depressed patients

Probe type: PS receptor ligand

PET

[ 11C] WAY-100685

5-HT1A receptor Antidepressant efficacy studies

Type: autoreceptor ligand

PET

[ 11C] McN-5652, [11C] DASB

SERT Antidepressant binding efficacy

Probe type: reuptake site ligand

SPECT

[ 123I] -CIT, [123I] ADAM

SERT; type: same

as above

Antidepressant binding efficacy

Probe type: reuptake site ligand

SPECT

[ 123I] 5-I-R91150

5-HT2A receptor Serotonin turnover

Probe type: PS receptor ligand

Amino acid

transmitters:

GABA/glutamate

SPECT

[ 123I] iomazenil

Benzodiazepine

receptor

GABA levels in anxiety states

Probe type: PS receptor ligand

Magnetic resonance

spectroscopy

None None Concentrations of GABA,

glutamate

Probe type: metabolomic

approach

Beta-amyloid imaging

PET

[ 18F] FDDNP, [11C] PIB, [11C]

SB

Beta-amyloid plaque Progression of senile plaques in

Alzheimer’s disease

Probe type: ligand of pathological

deposit

Note. 5-HT = serotonin (5-hydroxytryptamine); GABA = -aminobutyric acid; PET = positron emission tomography;

SERT = serotonin transporter; SPECT = single photon emission computed tomography.

Neurotransmitter Synthesizing Systems

Appropriate availability of neurotransmitters and neuromodulators is essential to normal neurological

and psychological function. Dysfunction or degeneration of neurons that synthesize these substances can

lead to various disorders. For example, Parkinson’s disease is caused by selective degeneration of the

dopamine-synthesizing neurons of the nigrostriatal system. Uptake of [18F] DOPA, which selectively

labels aromatic L-amino acid decarboxylase (AADC), a critical enzyme in the synthesis of dopamine, has

been used to estimate both the number of surviving cells and AADC activity among nigral neurons, thus

providing a tool to understand the connection between dopamine dysfunction and clinical symptom

evolution (Cropley et al. 2006; Ravina et al. 2005). PET scanning using [11C] methyl-L-tryptophan, which

is a marker of serotonin (5-HT) synthesis, is being used in identifying overactive serotonin-synthesizing

systems in differentiating epileptogenic from nonepileptogenic lesions in tuberous sclerosis (Luat et al.

2007) prior to neurosurgery. In a similar vein, recent functional imaging techniques have combined with

neurosurgical treatments such as deep brain stimulation to study brain imaging biomarkers of treatment

response (Carbon and Eidelberg 2002).

Neurotransmitter Binding Sites

PET imaging has been used to identify binding sites of neurotransmitters of relevance to psychiatric

disorders, in order to characterize patients and inform treatment decisions based on mechanisms of drug

action. For instance, studies with ligands that bind to D2 receptors have informed us that lower bindingPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

7 of 26

10/05/2009 15:53

affinity, faster dissociation, and optimal occupancy at usually prescribed doses of D2 receptor

antagonists might form the basis of atypical antipsychotic drug action (Kapur and Remington 2001).

Studies reporting increased binding of serotonin type 2A (5-HT2A) receptors using agents such as

18F-altanserin in frontal cortex (Brodmann area [BA] 9) and their association with increased pessimism

and self-injurious and suicidal behavior have added to our understanding of depression (Meyer et al.

2003). Alterations in the function of -aminobutyric acid (GABA) systems in posttraumatic stress

disorder (PTSD) and panic disorder have been reported based on decreased binding of (123I)iomazenil to

benzodiazepine receptors in the BA 9 region of these patients (Bremner et al. 2000).

Neurotransmitter Reuptake Site Binding Studies

Drugs that bind to and inhibit the serotonin reuptake transporter (SERT) have been shown to be

associated with symptomatic recovery from depression in a large number of patients (Nemeroff 2007).

Radioligands that show high specificity for binding to SERT, such as [11C]

3-amino-4-(2-dimethylaminomethyl-phenylsulfanyl)-benzonitrile (11C DASB), have been used to study

the site occupancy levels of selective serotonin reuptake inhibitors (SSRIs) and how these values

correlate with clinical efficacy (Meyer 2007). These methods have led to reformulation of clinical

algorithms in the management of depressive disorders. Suicide remains a serious risk among all

psychiatric patients, and the role of SERT in the expression of suicide has been studied using PET ligands

(Purselle and Nemeroff 2003). Dopamine transporter (DAT) ligands have also been used to study

cognitive and motor dysfunction in Parkinson’s disease (Cropley et al. 2006; Ravina et al. 2005), and the

same methods are currently being applied to investigate conditions such as

attention-deficit/hyperactivity disorder (ADHD).

Neurotransmitter Autoreceptor Binding Methods

Synaptic turnover of serotonin secretion is regulated by two processes—by SERT-mediated reuptake and

by negative feedback mediated through serotonin type 1A (5-HT1A) autoreceptors. Downregulation of

5-HT1A receptors may be associated with antidepressant efficacy and might also explain the delay in the

onset of clinical response after being started on SSRIs (Blier and Ward 2003). Radioligands such as [11C]

WAY-100865 have been used to study the dynamics of 5-HT1 receptors (Fisher et al. 2006). This

methodology has also made contributions to better understanding how serotonergic system imbalances

disrupt mood-regulating neural circuitry, resulting in depressive disorders (Fisher et al. 2006).

Pathological Cerebral Deposits

Degenerative disorders such as Alzheimer’s disease have acquired critical relevance, given the larger

proportion of aging population in the modern society. Since the pathological changes associated with

neurodegeneration (such as beta-amyloid deposits, neurofibrillary tangles, and decreased metabolic

activity) often precede clinical disease by several decades, early identification of these changes is of

paramount importance. Three different ligands have been reported recently, all of them with the ability

to detect intracerebral beta-amyloid deposition with high accuracy using PET radiotracer technology:

[ 18F] FDDNP (Small et al. 2006), [11C] PIB (Pittsburgh Compound–B) (Klunk et al. 2004), and [11C] SB

(Ono et al. 2003). Such compounds have potential clinical applications in that they may help identify

patients with mild cognitive impairment (MCI) who are at increased risk of converting to dementia

(Mathis et al. 2005).

PET Tracers of Cerebral Metabolism and Blood Flow

PET has been used to assess functional activity of brain regions, both in the resting state and in response

to various stimuli. The methods used include use of FDG-PET and radioactive [15O] H2O-PET to study

metabolic activity and blood flow, respectively. Figure 10–2A shows a picture of increased cerebral blood

flow to paralimbic regions during a sad mood induction task (to be described later) using H2O-PET. In

contrast, Figure 10–2B shows metabolic activity differences among depressed versus healthy patients

using FDG-PET. These modalities have been effectively used to study a variety of mental phenomena and

have been of considerable benefit in enhancing our understanding of psychiatric disorders. Of particular

interest have been studies using PET to understand the biological basis of schizophrenia (Fujimoto et al.

2007; Lange et al. 2005), bipolar disorder (Post et al. 2003), depression (Mayberg 2003b; Neumeister et

1. 2004), substance abuse and craving (Kilts et al. 2004), PTSD (Bremner 2007), ADHD (Schweitzer et
2. 2003), and Alzheimer’s disease (Small 1996). Most of the studies of psychiatric disorders have shownPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

8 of 26

10/05/2009 15:53

generally similar patterns of resting blood flow and metabolic abnormalities in using PET. That said, such

identified patterns have not yet proven adequately consistent to warrant use of PET as a diagnostic

procedure in an individual patient. Notably, only resting-state FDG-PET scanning has undergone repeated

sensitivity and specificity testing to be considered useful and reliable for the diagnosis of Alzheimer’s

disease in patients with progressive cognitive disturbance and presumed dementia (Silverman et al.

2002). Nonetheless, research studies using these methods have provided considerable new insights into

disease pathophysiology as well as mechanisms mediating treatment response (Erritzoe et al. 2003;

Evans et al. 2006; Mayberg 2003a; Roffman et al. 2005).

FIGURE 10–2. (A) Task-induced increased cerebral blood flow using H2O-PET. (B) FDG-PET

resting-state contrasts among depressed patients versus healthy control subjects.

FDG = fluorodeoxyglucose; PET = positron emission tomography; Cg25 = subgenual cingulate; pCg = posterior

cingulate; Hth = hypothalamus.

Source. (A) Adapted from Mayberg HS, Liotti M, Brannan SK, et al.: “Reciprocal Limbic-Cortical Function and

Negative Mood: Converging PET Findings in Depression and Normal Sadness.” American Journal of Psychiatry

156:675–682, 1999. Copyright 1999, American Psychiatric Association. Used with permission. (B) Image courtesy

of Helen Mayberg, M.D.

MAGNETIC RESONANCE IMAGING

The basis of MRI technology rests on the magnetic properties of the hydrogen atom, which, as aPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

9 of 26

10/05/2009 15:53

component of water, is found ubiquitously in organic tissues (water constitutes roughly 80% of brain

weight). The nucleus of a hydrogen atom, a single proton, is characterized by an intrinsic magnetic

moment, called spin. The protons in tissue are normally oriented in random directions, but if an external

magnetic field is applied, they will tend to align with the field. Quantum mechanics and statistical physics

dictate that spins can orient either in the direction of the applied field (parallel) or in the direction

opposite to it (anti-parallel), but on average the parallel orientation will gain a weak majority. This

situation will result in a net magnetic moment induced by the external field in the tissue; in other words,

the tissue will become slightly magnetized. The external (or main) field in an MR environment is provided

by a powerful electromagnet whose intensity is typically 1.5 tesla for clinical scanners (for comparison,

the electromagnets used in car demolition lots have a similar strength), and up to 7 tesla for human

research scanners. The intensity of the induced magnetization is dependent on the proton distribution

(i.e., on the local molecular characteristic of the tissue). Although this induced magnetization could, in

principle, provide a contrast signal resolving different type of tissues (e.g., gray matter, white matter,

cerebrospinal fluid) into an image of anatomical detail, the magnitude of this effect is in practice so small

that it does not lend itself to direct measurement.

The way in which MRI actually recovers its signal is by first perturbing the examined system. This

instrumental perturbation is performed by applying short radiofrequency (RF) pulses that, when

appropriately tuned, are able to transiently tip the orientation of the spins away from the applied

magnetic field. However, the tendency of the spins is to return to the orientation coherent with the

applied magnetic field, given that the latter state is characterized by a lower energy. The relaxation

process occurs through the emission of an RF wave that is detected by the same RF hardware that emits

the excitation pulses. In MR terminology, this is referred to as the RF coil, which has the form of a small

cylindrical cage that surrounds the subject’s head in the scanner. The emitted RF wave—or, more

precisely, the temporal signature of its decay as the excited spins relax—constitutes the actual MR signal

and depends on the molecular characteristics of the local tissue, as well on the particular sequence of

excitation pulses employed. The details of the physics that specify how appropriate pulse sequences can

be engineered to acquire images of the brain system with different physiological meaning are beyond the

scope of this discussion, and the interested reader is directed elsewhere (e.g., Buxton 2002).

The spin relaxation measured with MRI can be decomposed into longitudinal and transverse components,

which are only partially related. The measurement of the relaxation time of the longitudinal component,

called T1, provides images in which the contrast between different types of tissue (notably, gray matter,

white matter, and cerebrospinal fluid) is maximized. Such T1-weighted images are capable of defining

the anatomy of the living brain with great precision and are therefore used as an anatomical reference

for most of the neuroimaging studies. An image of the entire brain, with a resolution, or voxel size, of 1

mm3 (voxel stands for “volume pixel,” the unitary element of a three-dimensional image), can be

acquired on a 1.5-tesla clinical scanner in less than 6 minutes.

The measurement of the relaxation time of the transverse component, which can be further split

between the T2 and the T2* characteristic times, provides images that are influenced by the local

inhomogeneity of the magnetic field, which in turn is increased by the local blood perfusion. In

particular, T2*-weighted images are characterized by a contrast that highlights changes in vascular

dynamics that accompany neural activity and are thus employed in functional mapping studies. The

advent of a very fast technique for the acquisition of T2*-weighted images, called echoplanar imaging

(EPI), allows the collection of an entire brain volume in 3–4 seconds and has been instrumental in the

rapid development of functional MRI. The ultrarapid acquisition of EPI images and the nature of the

detected T2* signal, which tends to become negligible when integrated over very small voxels, limit the

resolution of standard EPI images, which typically have a voxel size of 3–4 mm per side. The use of

particular technical and experimental arrangements has allowed, in special cases, the achievement of

submillimetric precision, such as that required for the imaging of the columnar organization of visual

cortex (Menon and Goodyear 1999). Table 10–2 includes the multiple applications of MR-based

technologies in psychopharmacology.

TABLE 10–2. Use of magnetic resonance imaging (MRI) in psychopharmacology researchPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

10 of 26

10/05/2009 15:53

Type of imaging Technique Method of

analysis

Used to measure

Structural MRI

(sMRI)

Voxel-based morphometry (VBM) Automated Volume of brain regions in brain

disorders and ischemic lesions

sMRI Region of interest (ROI) analysis Manual Same as above to measure volumes

and ischemic brain lesions

(hyperintensities)

Functional MRI

(fMRI)

BOLD technique (described in text) Computerized

algorithm

Area of activation in response to

cognitive/affective challenge

Functional

connectivity

analysis

Resting-state activity, independent

component analysis (ICA), structural

equation modeling

Computerized

algorithms,

statistical models

Connectivity between different

components of neural network during

various mental states

Diffusion-based

MRI

Diffusion-weighted,

perfusion-weighted, diffusion tensor

imaging (DTI)

Computerized

algorithms

Tissue integrity by imaging water

diffusion in restricted and free space,

used in diagnosis of stroke and

neurodegeneration

Diffusion tensor

tractography

DTI-based imaging and fiber tracking Computerized

algorithms

Anatomical white matter tract

connectivity between brain locations

Magnetic

resonance

spectroscopy

Detect concentration of specific

metabolites in cerebral regions

Automated and

voxel based

(manual)

Neuronal/glial metabolic

abnormalities in localized in single or

multiple voxels

Innovative in vivo magnetic resonance approaches

Neuroimaging

genomics

fMRI activations to challenges in

various genotypes

Computerized

genotyping

Identifying candidate

disease-associated genes among

genetically distinct populations

(COMT, BDNF, 5-HTTLPR

polymorphisms)

sMRI + fMRI Data fusion using joint independent

component analysis (jICA) of

simultaneously recorded structural

and functional data

Computerized

algorithms

Structural and functional

disconnection in mental disorders

Virtual reality

(VR)–fMRI

VR-based cognitive challenges Computerized Cerebral activation in real-life

scenarios (spatial recognition

memory)

Hyperscanning Online linkage of two fMRI scanners

in different locations

Web based Cerebral activation changes during

social interactions (social

neuroscience technique)

Note. 5-HTTLPR = serotonin transporter–linked polymorphic region; BDNF = brain-derived neurotrophic factor;

BOLD = blood oxygenation level–dependent; COMT = catechol-O-methyltransferase.

Functional MRI

Functional magnetic resonance imaging, or fMRI, refers to a variant of MRI that is sensitive to local

changes in deoxyhemoglobin concentration. The increase in regional blood flow, as engendered by

neural activation, overshoots the oxygen consumption. This results in an apparent decrease in

deoxyhemoglobin. In the 1930s, Linus Pauling had observed that the amount of oxygen carried by

hemoglobin is inversely proportional to the degree to which it perturbed a magnetic field. This property

of differential paramagnetism was finally demonstrated in vivo in the late 1980s, and fMRI was born

(Ogawa et al. 1992; Thulborn et al. 1982).

The BOLD Signal in Functional MRI

Functional MRI exploits the fact that deoxyhemoglobin has paramagnetic properties and oxyhemoglobinPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

11 of 26

10/05/2009 15:53

does not. Deoxyhemoglobin disturbs the local magnetic environment, causing the surrounding protons to

dephase even faster than they would otherwise (Figure 10–3).

FIGURE 10–3. Schematic diagram of the effect of hemoglobin (Hb) on the local magnetic field of brain

tissue.

Only deoxyhemoglobin (deoxyHb) has paramagnetic properties and locally distorts the field, leading to faster spin

dephasing.

Source. Reprinted from Pagnoni G, Berns GS: “Brain Imaging in Psychopharmacology,” in The American Psychiatric

Publishing Textbook of Psychopharmacology, 3rd Edition. Edited by Schatzberg AF, Nemeroff CB. Washington, DC,

2003, pp. 163–172. Copyright 2003, American Psychiatric Publishing, Inc. Used with permission.

Because neuronal activity leads to an overactive increase in blood flow, this actually decreases the

amount of deoxyhemoglobin relative to oxyhemoglobin. Because less deoxyhemoglobin means less rapid

spin dephasing, this increase in blood flow appears as an increase in MR signal. This is called the blood

oxygenation level–dependent (BOLD) signal. In response to a regionally specific neuronal activation, the

BOLD signal will usually increase by an amount on the order of 1% on a standard 1.5-tesla clinical

scanner. The intensity of the signal is proportional to the strength of the main magnetic field—for

example, it will double in the case of a 3-tesla scanner.

The temporal resolution of fMRI is determined both by the hemodynamic response and the physical

constraints of the scanner magnetic fields. The hemodynamic response generally lags the neural activity

by 3–5 seconds and may extend upward to 10–15 seconds (Figure 10–4). The rate at which the scanner

can acquire images is influenced by the desired resolution. Generally, the more slices and the finer the

resolution within each slice, the longer a whole-brain acquisition takes. While an individual slice can be

acquired in as little as 60 milliseconds, whole-brain imaging usually requires about 2–3 seconds.

FIGURE 10–4. Relative blood oxygenation level–dependent (BOLD) response to 1-second visual

stimulation.Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

12 of 26

10/05/2009 15:53

These functional magnetic resonance imaging (fMRI) data are from the occipital cortex and were obtained in a

healthy volunteer in a 3-tesla scanner. The amplitude of the signal is about 2%, with the peak 5–8 seconds after

the stimulus.

Source. Reprinted from Pagnoni G, Berns GS: “Brain Imaging in Psychopharmacology,” in The American Psychiatric

Publishing Textbook of Psychopharmacology, 3rd Edition. Edited by Schatzberg AF, Nemeroff CB. Washington, DC,

2003, pp. 163–172. Copyright 2003, American Psychiatric Publishing, Inc. Used with permission.

Whereas fMRI measurements are easy to perform, there are specific limitations with BOLD imaging.

The BOLD effect originates from venous vessels (capillaries, venules, and veins), so the signal is not exactly

collocated either with the locus of neural activity or with the arterial supply. This spatial error may, however,

be negligible for brain-mapping studies employing a standard spatial resolution (voxel size ~50 mm3 ).

1.  

Bulk head motion and physiological pulsation (heart pulse, respiration) artifacts. For the motion, head

movement should be restrained while maintaining a comfortable situation for the subject.

2.  

Susceptibility artifacts. The fact that BOLD detects local changes in magnetic susceptibility (due to the

variation in deoxyhemoglobin concentration) renders it vulnerable to the large discontinuity that exists at the

interfaces between bone/air and bone/liquid. In these regions, the steep variations in tissue density cause a

distortion of the local magnetic field, resulting in both a spatial distortion of the image and a drop in the BOLD

signal. This makes it difficult to detect the small changes associated with deoxyhemoglobin variations. The

problematic regions are notably the orbitofrontal cortex and the inferior part of the temporal lobes, which

unfortunately are the locus of many interesting neuropsychological processes.

3.  

Functional MRI Experimental Designs

The particular characteristics of fMRI (i.e., its noninvasiveness and its spatiotemporal resolution) allow

for its use in a variety of experimental designs. Two commonly employed schemas are block design and

event-related design.

In a block design experiment, the experimental conditions alternate in blocks of some tens of seconds,

each consisting of multiple repetitions of the same condition. In the subsequent statistical analysis, the

scans acquired during the same block are pooled together to give an average effect. In an event-related

design, each single experimental event/stimulus is modeled separately, which usually allows more

latitude both in terms of the analysis (one can, for instance, separate the scan corresponding to a correct

behavioral response from the scans corresponding to the subject’s errors) and of the experiment itself

(for instance, one can alternate between events of different types to avoid effects related to

habituation). The analysis of such studies relies on the assumption of linearity in the BOLD response: ifPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

13 of 26

10/05/2009 15:53

events of the same type are presented not too closely one to the other (the “linearity” seems to break

down with an interstimulus interval of 1–2 seconds), then the resulting BOLD signal is simply the sum of

the BOLD responses for the individual events.

Comparison of PET and Functional MRI

The advantages and disadvantages of PET versus fMRI are summarized in Table 10–3.

TABLE 10–3. Advantages and disadvantages of positron emission tomography (PET) versus functional

magnetic resonance imaging (fMRI)

Advantages of PET versus fMRI

Quiet (good for acoustic stimulation); fMRI may have noise >90 dB

Less sensitive to movement artifact

Allows metabolic and receptor mapping

Allows imaging of brain regions that are typically difficult to image with fMRI because of the presence of a

susceptibility artifact (orbitofrontal cortex, inferior temporal lobe) that causes both distortion and loss of signal

Allows the use of standard measurement devices (physiological, behavioral) inside the scanner (i.e., avoids the

complication for the need of specially designed MRI-compatible hardware)

(In the MRI environment, the presence of a very strong static magnetic field commands the use of diamagnetic

components; moreover, every electric device in the scanner room needs to be carefully shielded to prevent

interference problems to and from the scanner. Scanning is not used in patients who have pacemakers or

ferromagnetic metal parts in their bodies.)

Disadvantages of PET versus fMRI

Injection of a radioactive isotope precludes the use of PET for longitudinal studies in which the same subjects are

scanned repeatedly over an extended period of time.

PET provides an integral measure (over time) of brain activity (for activation techniques), with a temporal resolution

on the order of minutes because of the lifetime of the isotope. By comparison, fMRI has a temporal resolution on the

order of seconds. This prevents the use of sophisticated, event-related designs with PET. Also, the number of

images typically collected with PET on a single subject rarely exceeds a dozen, thereby limiting the statistical

treatment in the analysis of the data.

Spatial resolution is more limited with PET than with fMRI.

Cyclotron must be located nearby.

PET is more expensive than fMRI (utilization costs per hour: fMRI, ~$500; PET, ~$2,000).

The acquisition procedure is time-consuming and requires more resources. (One scan typically lasts ~3 hours [fMRI

typically lasts <1 hour]. In comparison, the MRI experimental setup is easier to perform and can be operated by

just one person.)

Applications of Functional MRI in Psychopharmacological Research

At the time of writing this chapter, most of the applications of fMRI methods have been

experimental—for example, helping investigators map areas of functional activation in response to

cognitive and affective tasks. Similarly, several novel and innovative uses of imaging methods are seen in

the literature, and the field has evolved considerably since the previous edition of this textbook. A few of

these advances will be reviewed in this section and are also included in Table 10–2.

Neural activation in response to tasks

Measurement of neural activation patterns in response to specific tasks, using fMRI, has been

particularly useful in elucidating the neural mechanisms contributing to the development of psychiatric

disorders. Thus, specific tasks can be constructed that target specific emotional or cognitive behaviors

that are altered in a given disorder. This research strategy allows delineation of neural pathways

involved in the generation of specific symptoms and behaviors. A few examples of commonly used

paradigms are discussed below.

Mood/self-referential paradigmsPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

14 of 26

10/05/2009 15:53

As an example, a core feature of depression involves negative bias and anhedonia, suggesting specific

alterations in neural pathways mediating salience, self-reference, and reward. Accordingly, several fMRI

studies reported failed activation of dorsomedial prefrontal cortex (BA 10) in depressed patients in

response to positive words or pictures (Le Bastard et al. 2003; Mitterschiffthaler et al. 2003).

Interestingly, activation of this region is also seen in tasks requiring self-referencing in healthy subjects

(Craik et al. 1999; Fossati et al. 2003). A depression-specific pattern has been described using an

emotional “Go/No-Go task” that identified blunted responses in reward circuits in response to neutral

words but exaggerated responses in self-referential areas of the rostral cingulate and medial frontal

cortex in response to sad words (Elliott et al. 2002). Such sensitive tasks can plausibly be used as

outcome measures in the evaluation of the efficacy of antidepressant treatments.

Facial expression–processing paradigms

Another task frequently used across a range of disorders involves the presentation of faces with different

emotional expressions, such as angry, sad, fearful, happy, and neutral. Such faces can be either overtly

presented or may be hidden behind a mask, depending on the study hypothesis. With overt presentation,

subjects typically perform a behavioral task related to classifying some aspect of the faces. Application of

such tasks in fMRI studies has produced a large amount of data regarding the neural basis of emotional

processing in healthy subjects, reporting robust activation of the amygdala in response to emotional

faces (Whalen et al. 1998). Of note, patients with depression, anxiety disorders, and PTSD exhibit

increased amygdala responses to the presentation of fearful or angry faces (Rauch et al. 2000; Shin et al.

2005; Whalen et al. 2002). The paradigm has also proven useful to evaluate treatment effects of

antidepressant drugs (Sheline et al. 2001), as well as temperamental or genetic contributions to

emotional processing (Hariri et al. 2002; Stein et al. 2007). Similar strategies have identified

antidepressant-induced changes with other tasks (Fu et al. 2007).

Cognitive “working memory” paradigms

Another common task used to identify neurocognitive changes is the so-called n-back task. The n-back

task is a working memory task that reflects prefrontal cortical function. The n-back task has been used in

fMRI studies to identify altered neural circuits in a variety of disorders, including schizophrenia and

depression (Harvey et al. 2005; Meyer-Lindenberg and Weinberger 2006). The “n-back” working memory

test (Owen et al. 2005) involves visual presentation of letter stimuli at previously chosen intervals and

epochs (e.g., 2-second interval for 30-second epochs). The baseline (control) condition is usually a

0-back condition in which subjects are required to press a button with the right index finger when the

stimulus (e.g., the letter “x”) appears. In the experimental condition (1-, 2-, or 3-back), subjects are

required to press a button if the presented stimulus is the same as a stimulus presented n trials

previously (n = 1, n = 2, or n = 3). The level of task difficulty and the condition are varied in a previously

specified order throughout the scan time. Subject performance during scanning in regard to accuracy

(number of target stimuli correctly identified) and response time (RT) is usually recorded. Interestingly,

findings vary across disorders, suggesting differing impacts of underlying pathophysiology on networks

mediating these behaviors. For example, increased prefrontal activity is seen in depressed patients

relative to control subjects performing this task, an effect amplified by task difficulty (Harvey et al.

2005). On the other hand, several studies have reported that schizophrenic patients demonstrate deficits

in activation of the prefrontal cortex during this task, thought to reflect alterations in dopamine

functioning. Normalization of neural activation patterns by antipsychotic drugs is associated with

response to treatment. The n-back task has also been useful in identifying genetic contributions to

schizophrenia risk (Meyer-Lindenberg and Weinberger 2006). Bookheimer et al. (2000) used a verbal

memory paradigm, in which patients memorized unrelated pairs of words during scanning, to study

hippocampal activation in patients at risk of developing Alzheimer’s disease. Not only did the carriers of

the apolipoprotein epsilon 4 (ApoE- 4)++ allele (higher risk of dementia) show greater hippocampal

activation, but this baseline activation pattern predicted longitudinal cognitive decline.

Reward-processing paradigms

Reward-processing paradigms are commonly used in studying substance use and craving, but also in

studying anhedonia in both depression and schizophrenia. In addiction studies, typically the patient,

while lying in a scanner, is presented with multiple contexts associated with drug abuse, and activation

of reward-processing circuitry is studied. Zink et al. (2006) used fMRI to study activation of the basalPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

15 of 26

10/05/2009 15:53

ganglia (a key component of reward circuitry) of healthy volunteers in response to salient stimuli with

high motivational relevance. Figure 10–5 illustrates differential activation patterns in the ventral

striatum in response to wins versus monetary losses during a gambling task, as measured with fMRI. The

concept of “intertemporal discounting” is often used to describe choosing between discounted immediate

reward to delayed reward at a rate that is much higher in value. Patients with impulse-control disorders

such as pathological gambling show considerable variation in their tendency to use intertemporal

discounting (Dixon et al. 2006). A recent study using fMRI reported that increased activation of

paralimbic cortex was associated when choosing smaller/earlier rewards while fronto/parietal activation

was seen when making the larger/later (opposite) type of choice (McClure et al. 2004). Studies such as

these might help us develop brain activation–based biomarkers of complex disorders such as

pathological gambling and substance abuse. A recent innovation in this field has been the development

of “hyperscanning”—a method by which a person in an fMRI scanner at one location can interact with

another person in another scanner using World Wide Web–based connectivity (Montague et al. 2002).

Thus, the brain structures that are activated during social interactions could be studied using this

technology.

FIGURE 10–5. Ventral striatum activation for wins versus monetary losses during a gambling task, as

visualized with functional magnetic resonance imaging (fMRI).

Source. Image courtesy of Giuseppe Pagnoni, Ph.D.

Summary of neural activation paradigms

In conclusion, these examples illustrate the potential of fMRI activation studies in probing

brain–behavior relationships in psychiatric disorders, owing to the flexibility of fMRI paradigm design as

well as the large variety of standardized tasks and possibility to develop novel tasks.

Neuroimaging genomics

Despite the substantial advances in the pharmacotherapy of psychiatric disorders, response rates of

patients remain unsatisfactory. For example, only 40%–70% of depressed patients adequately respond

to 6 weeks of treatment with a single antidepressant drug. The factors that determine whether or not a

patient will favorably respond to a specific drug are poorly understood, and there are no establishedPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

16 of 26

10/05/2009 15:53

strategies for clinicians to decide whether a patient might benefit from psychotherapy rather than

pharmacotherapy (Binder and Holsboer 2006). Similarly, there is considerable individual variability in the

response to antipsychotic drugs in patients with schizophrenia, which remains unexplained, and some

but not other patients experience debilitating side effects (Reynolds et al. 2006). In concert with the

rapid advances in the field of molecular genetics, there has been increasing recognition that genes

influence individual differences in treatment response, as studied in the field of pharmacogenetics (see

Chapter 3, “Genetics and Genomics”). Neuroimaging techniques have substantial potential to further the

field of pharmacogenetics. There has been a surge of studies in recent years using neuroimaging to

describe genetically determined differences in brain structure and function. Research strategies include

neuroimaging studies in twin populations, unaffected individuals with high familial risk for a specific

disorder (i.e., healthy siblings of patients), and groups of individuals with different variants of functional

polymorphisms in genes involved in brain development or behavior (Hariri et al. 2006; Meyer-Lindenberg

and Weinberger 2006; Peper et al. 2007). Such studies not only advance our understanding of individual

differences in cognitive and emotional behaviors, and the mechanisms involved in mental disorders, but

also elucidate genetically determined variations in neural systems that are targeted by specific

treatments. For example, one recent study found that the functional Val66Met polymorphism in the

brain-derived neurotrophic factor (BDNF) gene moderates the association between depression and

hippocampal volume loss. Carriers of the Met allele had smaller hippocampal volumes than subjects who

were homozygous for the Val allele (Frodl et al. 2007). These findings suggest that the Met allele

conveys increased risk for smaller hippocampi and susceptibility to depression. Because antidepressants

stimulate neurogenesis in the hippocampus, carriers of the Met allele may be particularly responsive to

antidepressant drugs.

In the area of schizophrenia research, it has been shown that a genetic variation (Val[108/158]Met) of

catechol-O-methyltransferase (COMT), an important enzyme that degrades cortical dopamine at the

synapse, is associated with deficits in working memory and prefrontal cortical activation in response to a

working memory task. Specifically, carriers of the Val allele exhibit more deficits and less prefrontal

cortical activation compared with subjects homozygous for the Met allele, likely reflecting low synaptic

dopamine availability due to greater COMT activity, potentially increasing risk of developing

schizophrenia (Meyer-Lindenberg and Weinberger 2006). Interestingly, the same polymorphism predicts

effects of antipsychotic treatment on prefrontal cortical function in schizophrenic patients, as measured

by fMRI, contributing to individual differences in treatment response (Bertolino et al. 2004). Based on

such findings, COMT inhibitors are being investigated in the treatment of cognitive symptoms in

schizophrenia.

Of course, in studying the effects of genes on neural responses to drugs, factors that interact with genes

in shaping brain structure and function—for example, environmental factors across development—must

be considered as well. In another example, it has been found that a functional polymorphism in the

promoter region of the serotonin transporter gene—5-HTTLPR polymorphism (SCL6A4)—moderates the

relationship between stress, including child maltreatment, and depression (Caspi et al. 2003; Kaufman et

1. 2004; Kendler et al. 2005). Carriers of the short (S) allele had increased depression risk in relation to

stress, whereas subjects homozygous for the long (L) allele were resilient even in the context of severe

stress. Remarkably, S allele carriers demonstrate increased amygdala activation as well as decreased

functional connectivity between amygdala and inhibitory prefrontal cortical regions in response to

emotional stimuli, compared with L/L allele carriers (Hariri et al. 2002; Heinz et al. 2005; Pezawas et al.

2005). Of note, both the SCL6A4 polymorphism and developmental stress have been associated with

treatment response to serotonergic drugs (Nemeroff et al. 2003; Reimold et al. 2007). Accordingly,

neuroimaging is evolving as a valuable tool to evaluate gene–environment–drug interactions at the

neural level. Such studies, taken together, have the potential to yield diagnostic markers to guide

treatment decisions and identify targets for the prevention of manifest disorders in subjects at risk.

Default or resting-state functional imaging

Most studies reviewed so far have used challenges or probes that elicit neural activation and blood flow

responses that could be captured by fMRI or PET scanners. Several authors have consistently found

evidence for the existence of a “default” neural network that preferentially shows greater activity during

restful or passive cognitive states (Buckner and Vincent 2007; Gusnard et al. 2001). High activity of

these regions (posterior cingulate; inferior parietal and medial frontal cortices) during periods of wakefulPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

17 of 26

10/05/2009 15:53

rest and passive self-reflection have led some to hypothesize that this activity may “consolidate the past,

stabilize brain ensembles, and prepare us for the future” (Buckner and Vincent 2007, p. 1066).

Abnormalities in the functional connectivity of this network in psychiatric disorders such as

schizophrenia (Bluhm et al. 2007), depression (Greicius et al. 2007), dementia (Rombouts et al. 2005),

autism (Cherkassky et al. 2006), and multiple sclerosis (Lowe et al. 2002) have been reported. It is

possible that further research into these default networks might yield key diagnostic or pathological

information about various psychiatric disorders.

Application of virtual reality–based techniques to functional imaging

Most of the recently available challenge tasks used in functional imaging are based on tasks or tests

standardized for clinical testing. A major criticism of this approach has been that it might not be

reflective of real-life situations. Consequently, some authors use virtual reality (VR)–based techniques to

overcome this criticism. Ability to remember places and navigate through a VR-based city (similar to

spatial memory testing in rodents) has been used in an fMRI environment to study activity of parietal

and temporal cortices (Spiers and Maguire 2006).

Statistical modeling methods

A complex array of data is now available using functional and structural imaging methods in multiple

states of activation within subjects and between groups, necessitating the use of elegant multivariate

and multifactorial models (such as independent component analysis). A detailed description of these

techniques is beyond the scope of this chapter and has been undertaken elsewhere (Pearlson and

Calhoun 2007). Thus, using the related approach of structural equation modeling, it has been possible to

compare functional activation changes across multiple scans and states: depressed subjects versus

control subjects, treatment responders versus nonresponders, medication responders versus

cognitive-behavioral therapy responders, and so forth (Chen et al. 2007). Use of such models has

provided Chen and colleagues with the opportunity to identify and classify depression phenotypes at the

level of neural systems, with future implications for evidence-based treatment selection.

Structural MRI: Overview and Use in Psychopharmacological Research

As indicated in Table 10–2, MR technology has also provided important information on structural

abnormalities in various psychiatric disorders. Use of these methods in the understanding of psychiatric

disorders and their treatment will be addressed in the chapters on individual disorders. Only a general

introduction will be provided here.

Volumetric Methods

Magnetic resonance–based volumetric methods involve estimating volumes of cerebral structures using

MR images. They may be divided into two different types—automated (voxel-based morphometry [VBM])

and manual (region of interest [ROI]) methods. A detailed review of these methods is provided

elsewhere (Pearlson and Calhoun 2007). Using volumetric studies of hippocampus in depression, Sheline

et al. (2002) described a subtype of depression associated with hippocampal volume loss, memory

impairment, and hippocampal loss of 5-HT2A receptors. The alteration of cerebral structure by

medications has been reported as well. G. J. Moore et al. (2000) reported that lithium administration led

to a 3% increase in the volume of gray matter within a period of about 4 weeks. Psychiatric disorders are

clinically heterogeneous entities and such structural MRI–based studies have helped us to characterize

more specific pathological subtypes of depression. T2-weighted MRI studies have also been used to

identify and rate subcortical hyperintensities, the increase of which is believed to result in late-life

cerebrovascular disease and late-life depression (Alexopoulos et al. 1997; Kumar et al. 2002; Parsey and

Krishnan 1997).

Diffusion Tensor Imaging and Tractography

Most of the previously described volumetric methods focus on gray matter structures as opposed to

diffusion tensor imaging (DTI) and associated techniques that image white matter. DTI technologies

image water diffusion to study the directionality and integrity of white matter connections between

different locations of the brain (Kubicki et al. 2007). These studies advance the hypothesis that

psychiatric disorders are characterized by altered connectivity (“disconnection syndromes”). These

studies are beginning to help us understand the basis of cerebral organization and connectivity and howPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

18 of 26

10/05/2009 15:53

it might be affected in psychiatric disorders such as autism (Alexander et al. 2007), late-life depression

(Taylor et al. 2007), and schizophrenia (Nestor et al. 2007).

Magnetic Resonance Spectroscopy

MRS technology is based on the fact that MR acquisition involves receiving echoed RF waves of multiple

cellular chemical constituents. The individual chemical and metabolite constituents could be measured by

suppressing the resonance frequency of water molecules. A detailed exposition of the various types of

MRS techniques is beyond the scope of this chapter, and the reader is referred to excellent reviews on

the topic (Mason and Krystal 2006; C. M. Moore et al. 1999). Among the markers currently being

researched are N-acetylaspartate (NAA), glutamate/glutamine (glx), myoinositol (mI), choline (Cho),

glutathione, creatine, GABA, phosphomonoester (PME), and phosphodiester (PDE) (Lyoo and Renshaw

2002). An example of an MRS spectrum from a healthy control subject is provided in Figure 10–6. Using

this approach, Frye et al. (2007)were able not only to demonstrate elevated glutamate/glutamine (glx)

in anterior cingulate/medial prefrontal areas of patients with bipolar depression but also to document

the reduction of glutamine among patients who showed clinical response to treatment with lamotrigine.

Using proton MRS, C. M. Moore et al. (2007) measured multiple metabolites in the anterior cingulate

cortex (ACC) of children with bipolar disorder and reported decreased glutamine concentrations in the

presence of normal glutamate levels. Using another sample of adult patients, these same authors

reported that anterior cingulate cortex glutamine levels were elevated rapidly following administration of

the anticonvulsant topiramate (C. M. Moore et al. 2006). Using MRS technology, Sanacora et al. (2003)

reported that cortical GABA concentrations increased following a course of electroconvulsive therapy and

used this information to hypothesize that this increase in GABA might be associated with clinical

recovery. Thus, MRS might provide very useful biomarkers of neural functioning and disease states

among patients with psychiatric disorders.

FIGURE 10–6. Proton magnetic resonance spectroscopy (MRS) spectrum from right dorsolateral

prefrontal cortex voxel of a healthy individual.Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

19 of 26

10/05/2009 15:53

MM = macromolecules; NAA = N-acetylaspartate; Glx = glutamate/glutamine; Cr/PCr = creatine/phosphocreatine;

Ch = choline; mI = myoinositol.

Source. Reprinted from Haroon E, Watari K, Thomas MA, et al.: “Prefrontal Myo-Inositol Concentration and

Visuospatial Functioning Among Diabetic Depressed Patients.” Psychiatry Research: Neuroimaging 171:10–19,

2009. Copyright 2009, Elsevier Ltd. Used with permission.

ELECTROMAGNETIC MEASURES OF BRAIN ACTIVITY

The recording of the group electrical activity of neuronal assemblies is possible in human subjects with

the techniques of EEG and MEG. In these techniques, electrical activity is recorded extracranially from

the scalp of the individual in a noninvasive way (Bearden 2007). The appeal of these techniques is the

extremely high temporal resolution (~msec), which allows one to follow the measurement of global

patterns of neural activity in real time. Also, the electrical signal is directly related to the neural activity

(in contrast to methods such as PET and fMRI, which rely on indirect assessment of neural activity

through vascular effects), although the recorded electrical signals still represent an average over

extended regions of the brain. This high temporal resolution makes them ideal candidates for studying

synchrony phenomena (Tononi and Edelman 1998; Varela et al. 2001). Although considerable limitations

and constraints exist in regard to spatial localization of activity across brain regions, these have mostly

been overcome through technical and mathematical means. In spite of this, the spatial resolution, which

depends also on the number of electrodes at one’s disposal, remains quite poor in comparison with that

achievable with fMRI and is on the order of 1 centimeter. Also, the signals are quite weak, so the task

has to be repeated a large number of times (tens or even hundreds).

The electromagnetic techniques record the scalp distribution of the field produced by the sum of mainly

postsynaptic potentials of the neurons, each of which can be conceptualized as an electric dipole. EEG

detects the electrical potential of the field, and MEG detects the magnetic component. The twoPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

20 of 26

10/05/2009 15:53

techniques have very different technical requirements. EEG uses relatively simple equipment, basically a

multielectrode helmet, an amplifying and filtering device, and a computer (Ebner et al. 1999). MEG, by

contrast, employs cutting-edge technology (Ioannides 2006), because the detection of a magnetic field

intensity as weak as the one produced by the brain (~20,000 billion times weaker than the intensity of

the Earth’s magnetic field) requires the use of superconducting coil units based on superconducting

quantum interference device (SQUID) technology. These units need to be specially cooled to a

temperature near absolute zero. To avoid the intrusion of electromagnetic interference from the

environment, the recording takes place in a room that has been appropriately shielded.

A minimum of 32 scalp electrodes is needed to localize the sources of the recorded potential EEG, but

high-density arrays of 128 or 256 electrodes are now common. MEG commonly employs arrays of

100–300 detectors. The intensity of the EEG or MEG response is typically very low, and it takes generally

a large number of repetitions (~100) of the task/stimulus presentation (trial) to achieve a sufficient

power. The time series from each of the recording units is then segmented according to the stimulus

delivery sequence, and these temporally realigned segments are averaged together to obtain a waveform

that represents the mean time course of the response to the trial for that recording location (evoked

potentials). By combining these waveforms from all of the detectors, one can obtain a map of the surface

field distribution at each time interval inside the trial window. Both EEG and MEG are essentially surface

techniques: because the intensity of the electromagnetic field decreases rapidly with distance, the

detection of neural signals is restricted to the sources closest to the detectors—that is, the neocortex.

The activity of subcortical regions is very difficult to detect.

Use of Electroencephalography and Magnetoencephalography in

Psychopharmacology

While these techniques carry considerable potential, their use in psychopharmacology remains elusive.

The use of EEG and related technology of evoked potentials has helped contribute considerably to our

understanding of the pathophysiology of schizophrenia (Ford and Mathalon 2005; Turetsky et al. 1998).

Tucker et al. (2003) used dense-array EEG to study frontolimbic responsiveness in patients with

moderate depression. A potential for the use of EEG technology used in conjunction with fMRI in study of

analgesia and pain management has been recommended by some authors (Wise and Tracey 2006).

Pizzagalli et al. (2003) combined PET and EEG technologies using a source localization technique to

identify disruption of frontocingulate connectivity among patients with depression. Studies such as these

provide critical information that can complement and enhance our understanding of functional imaging

approaches.

CONCLUSION

Contemporary brain imaging methods provide a variety of strategies to probe structural, functional, and

chemical abnormalities in specific neural circuits relevant to psychiatric disease. Such studies are having

a considerable impact on our conceptualization of these disorders, with potential impacts on diagnosis

(Mayberg 2003a), clinical management (monitoring occupancy, or PIB changes with treatment [Klunk et

1. 2004]), and novel treatment development (Mayberg et al. 2005). Brain imaging in

psychopharmacology can be categorized both by the scanning technology (e.g., MRI, PET, or EEG) and by

its purpose (e.g., activation, resting state, behavioral, biochemical, or receptor mapping).

Receptor-mapping studies have clearly added to our ability to understand mechanisms of action of

psychopharmacological agents and their side-effect profiles. Activation studies, which indirectly measure

neuronal activity vis-à-vis changes in cerebral blood flow, have become widely used with fMRI

technology, providing new insights into behaviorally specific subcircuits. Structural MRI studies have

also begun to yield considerable data enabling a better understanding of how the disease processes are

regionally localized, where to target for functional imaging, and where to obtain specimens for

postmortem histopathological analysis. Multimodal imaging through the combination of fMRI, PET,

structural MRI, MRS, and electromagnetic measurements (EEG, MEG) offers the promise of identifying

both neuronal and chemical changes related to brain function.

REFERENCES

Alexander AL, Lee JE, Lazar M, et al: Diffusion tensor imaging of the corpus callosum in autism.

Neuroimage 34:61–73, 2007 [PubMed]

Alexopoulos GS, Meyers BS, Young RC, et al: “Vascular depression” hypothesis. Arch Gen PsychiatryPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

21 of 26

10/05/2009 15:53

54:915–922, 1997 [PubMed]

Arthurs OJ, Boniface S: How well do we understand the neural origins of the fMRI BOLD signal? Trends

Neurosci 25:27–31, 2002 [PubMed]

Bearden S: EEG reviewing/recording strategy. Am J Electroneurodiagnostic Technol 47:1–19, 2007

[PubMed]

Bertolino A, Caforio G, Blasi G, et al: Interaction of COMT (Val(108/158)Met) genotype and olanzapine

treatment on prefrontal cortical function in patients with schizophrenia. Am J Psychiatry 161:1798–1805,

2004 [Full Text] [PubMed]

Binder EB, Holsboer F: Pharmacogenomics and antidepressant drugs. Ann Med 38:82–94, 2006 [PubMed]

Blier P, Ward NM: Is there a role for 5-HT1A agonists in the treatment of depression? Biol Psychiatry

53:193–203, 2003 [PubMed]

Blodgett TM, Meltzer CC, Townsend DW: PET/CT: form and function. Radiology 242:360–385, 2007

[PubMed]

Bluhm RL, Miller J, Lanius RA, et al: Spontaneous low-frequency fluctuations in the BOLD signal in

schizophrenic patients: anomalies in the default network. Schizophr Bull 33:1004–1012, 2007 [PubMed]

Bookheimer SY, Strojwas MH, Cohen MS, et al: Patterns of brain activation in people at risk for

Alzheimer’s disease. N Engl J Med 343:450–456, 2000 [PubMed]

Bremner JD: Functional neuroimaging in post-traumatic stress disorder. Expert Rev Neurother

7:393–405, 2007 [PubMed]

Bremner JD, Innis RB, White T, et al: SPECT [I-123]iomazenil measurement of the benzodiazepine

receptor in panic disorder. Biol Psychiatry 47:96–106, 2000 [PubMed]

Broca P: Sur les températures morbides locales. Bulletin de l’Académie de. Medecine 8:1331–1347, 1879

Buckner RL, Vincent JL: Unrest at rest: default activity and spontaneous network correlations.

Neuroimage 37:1091–1096, 2007 [PubMed]

Buxton RB: Introduction to Functional Magnetic Resonance Imaging: Principles and Techniques.

Cambridge, UK, Cambridge University Press, 2002

Carbon M, Eidelberg D: Modulation of regional brain function by deep brain stimulation: studies with

positron emission tomography. Curr Opin Neurol 15:451–455, 2002 [PubMed]

Caspi A, Sugden K, Moffitt TE, et al: Influence of life stress on depression: moderation by a polymorphism

in the 5-HTT gene. Science 301:386–389, 2003 [PubMed]

Chen CH, Ridler K, Suckling J, et al: Brain imaging correlates of depressive symptom severity and

predictors of symptom improvement after antidepressant treatment. Biol Psychiatry 62:407–414, 2007

[PubMed]

Cherkassky VL, Kana RK, Keller TA, et al: Functional connectivity in a baseline resting-state network in

autism. Neuroreport 17:1687–90, 2006 [PubMed]

Craik FIM, Moroz TM, Moscovitch M, et al: In search of the self: a positron emission tomography

investigation. Psychological Science 10:26–34, 1999

Cropley VL, Fujita M, Innis RB, et al: Molecular imaging of the dopaminergic system and its association

with human cognitive function. Biol Psychiatry 59:898–907, 2006 [PubMed]

Dixon MR, Jacobs EA, Sanders S: Contextual control of delay discounting by pathological gamblers. J Appl

Behav Anal 39:413–422, 2006 [PubMed]

Ebner A, Sciarretta G, Epstein CM, et al: EEG instrumentation. The International Federation of Clinical

Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 52:7–10, 1999 [PubMed]

Elliott R, Rubinsztein JS, Sahakian BJ, et al: The neural basis of mood-congruent processing biases in

depression. Arch Gen Psychiatry 59:597–604, 2002 [PubMed]Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

22 of 26

10/05/2009 15:53

Erritzoe D, Talbot P, Frankle WG, et al: Positron emission tomography and single photon emission CT

molecular imaging in schizophrenia. Neuroimaging Clin N Am 13:817–832, 2003 [PubMed]

Evans KC, Dougherty DD, Pollack MH, et al: Using neuroimaging to predict treatment response in mood

and anxiety disorders. Annu Clin Psychiatry 18:33–42, 2006 [PubMed]

Fisher PM, Meltzer CC, Ziolko SK, et al: Capacity for 5-HT1A-mediated autoregulation predicts amygdala

reactivity. Nat Neurosci 9:1362–1363, 2006 [PubMed]

Ford JM, Mathalon DH: Corollary discharge dysfunction in schizophrenia: can it explain auditory

hallucinations? Int J Psychophysiol 58:179–189, 2005 [PubMed]

Fossati P, Hevenor SJ, Graham SJ, et al: In search of the emotional self: an FMRI study using positive and

negative emotional words. Am J Psychiatry 160:1938–1945, 2003 [Full Text] [PubMed]

Friston K: Beyond phrenology: what can neuroimaging tell us about distributed circuitry? Annu Rev

Neurosci 25:221–250, 2002 [PubMed]

Frodl T, Schule C, Schmitt G, et al: Association of the brain-derived neurotrophic factor Val66Met

polymorphism with reduced hippocampal volumes in major depression. Arch Gen Psychiatry 64:410–416,

2007 [PubMed]

Frye MA, Watzl J, Banakar S, et al: Increased anterior cingulate/medial prefrontal cortical glutamate and

creatine in bipolar depression. Neuropsychopharmacology 32:2490–2499, 2007 [PubMed]

Fu CH, Williams SC, Brammer MJ, et al: Neural responses to happy facial expressions in major depression

following antidepressant treatment. Am J Psychiatry 164:599–607, 2007 [Full Text] [PubMed]

Fujimoto T, Takeuch K, Matsumoto T, et al: Abnormal glucose metabolism in the anterior cingulate cortex

in patients with schizophrenia. Psychiatry Res 154:49–58, 2007 [PubMed]

Greicius MD, Flores BH, Menon V, et al: Resting-state functional connectivity in major depression:

abnormally increased contributions from subgenual cingulate cortex and thalamus. Biol Psychiatry

62:429–437, 2007 [PubMed]

Gusnard DA, Raichle ME, Raichle ME: Searching for a baseline: functional imaging and the resting human

brain. Nat Rev Neurosci 2:685–694, 2001 [PubMed]

Hariri AR, Mattay VS, Tessitore A, et al: Serotonin transporter genetic variation and the response of the

human amygdala. Science 297:400–403, 2002 [PubMed]

Hariri AR, Drabant EM, Weinberger DR: Imaging genetics: perspectives from studies of genetically driven

variation in serotonin function and corticolimbic affective processing. Biol Psychiatry 59:888–897, 2006

[PubMed]

Harvey PO, Fossati P, Pochon JB, et al: Cognitive control and brain resources in major depression: an

fMRI study using the n-back task. Neuroimage 26:860–869, 2005 [PubMed]

Heeger DJ, Boynton GM, Demb JB, et al: Motion opponency in visual cortex. J Neurosci 19:7162–7174,

1999 [PubMed]

Heinz A, Braus DF, Smolka MN, et al: Amygdala-prefrontal coupling depends on a genetic variation of the

serotonin transporter. Nat Neurosci 8:20–21, 2005 [PubMed]

Herscovitch P, Markham J, Raichle ME: Brain blood flow measured with H2(15)O, I: theory and error

analysis. J Nucl Med 24:782–789, 1983 [PubMed]

Ioannides AA: Magnetoencephalography as a research tool in neuroscience: state of the art.

Neuroscientist 12:524–544, 2006 [PubMed]

James W: The Principles of Psychology. New York, Henry Holt, 1890

Jueptner M, Weiller C: Does measurement of regional cerebral blood flow reflect synaptic activity?

Implications for PET and fMRI. Neuroimage 2:148–156, 1995 [PubMed]

Kapur S, Remington G: Dopamine D(2) receptors and their role in atypical antipsychotic action: still

necessary and may even be sufficient. Biol Psychiatry 50:873–883, 2001 [PubMed]Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

23 of 26

10/05/2009 15:53

Kaufman J, Yang BZ, Douglas-Palumberi H, et al: Social supports and serotonin transporter gene

moderate depression in maltreated children. Proc Natl Acad Sci U S A 101:17316–17321, 2004 [PubMed]

Kendler KS, Kuhn JW, Vittum J, et al: The interaction of stressful life events and a serotonin transporter

polymorphism in the prediction of episodes of major depression: a replication. Arch Gen Psychiatry

62:529–535, 2005 [PubMed]

Kennedy C, Des Rosiers MH, Sakurada O, et al: Metabolic mapping of the primary visual system of the

monkey by means of the autoradiographic [14C]deoxyglucose technique. Proc Natl Acad Sci U S A

73:4230–4234, 1976 [PubMed]

Kety SS: Biochemistry and mental function. Nature 208:1252–1257, 1965 [PubMed]

Kilts CD, Gross RE, Ely TD, et al: The neural correlates of cue-induced craving in cocaine-dependent

women. Am J Psychiatry 161:233–241, 2004 [Full Text] [PubMed]

Klunk WE, Engler H, Nordberg A, et al: Imaging brain amyloid in Alzheimer’s disease with Pittsburgh

Compound-B. Ann Neurol 55:306–319, 2004 [PubMed]

Kubicki M, McCarley R, Westin CF, et al: A review of diffusion tensor imaging studies in schizophrenia. J

Psychiatr Res 41:15–30, 2007 [PubMed]

Kumar A, Mintz J, Bilker W, et al: Autonomous neurobiological pathways to late-life major depressive

disorder: clinical and pathophysiological implications. Neuropsychopharmacology 26:229–236, 2002

[PubMed]

Lange C, Kracht L, Herholz K, et al: Reduced glucose metabolism in temporo-parietal cortices of women

with borderline personality disorder. Psychiatry Res 139:115–126, 2005 [PubMed]

Le Bastard G, Fossati P, Pochon JB, et al: Self-referential processing and emotional bias in depression: a

fMRI study (poster #49). Presented at: Ninth International Conference of Functional Mapping of the

Human Brain. New York City, NY, June 2003

Logothetis NK, Pauls J, Augath M, et al: Neurophysiological investigation of the basis of the fMRI signal.

Nature 412:150–157, 2001 [PubMed]

Lowe MJ, Phillips MD, Lurito JT, et al: Multiple sclerosis: low-frequency temporal blood oxygen

level-dependent fluctuations indicate reduced functional connectivity initial results. Radiology

224:184–192, 2002 [PubMed]

Luat AF, Makki M, Chugani HT: Neuroimaging in tuberous sclerosis complex. Curr Opin Neurol

20:142–150, 2007 [PubMed]

Lyoo IK, Renshaw PF: Magnetic resonance spectroscopy: current and future applications in psychiatric

research. Biol Psychiatry 51:195–207, 2002 [PubMed]

Magistretti PJ: Neuron-glia metabolic coupling and plasticity. J Exp Biol 209:2304–2311, 2006 [PubMed]

Magistretti PJ, Pellerin L: Cellular mechanisms of brain energy metabolism and their relevance to

functional brain imaging. Philos Trans R Soc Lond B Biol Sci 354:1155–1163, 1999 [PubMed]

Mason GF, Krystal JH: MR spectroscopy: its potential role for drug development for the treatment of

psychiatric diseases. NMR Biomed 19:690–701, 2006 [PubMed]

Mathis CA, Klunk WE, Price JC, et al: Imaging technology for neurodegenerative diseases: progress

toward detection of specific pathologies. Arch Neurol 62:196–200, 2005 [PubMed]

Mayberg HS: Modulating dysfunctional limbic-cortical circuits in depression: towards development of

brain-based algorithms for diagnosis and optimised treatment. Br Med Bull 65:193–207, 2003a

Mayberg HS: Positron emission tomography imaging in depression: a neural systems perspective.

Neuroimaging Clin N Am 13:805–815, 2003b

Mayberg HS, Lozano AM, Voon V, et al: Deep brain stimulation for treatment-resistant depression. Neuron

45:651–660, 2005 [PubMed]

McClure SM, Laibson DI, Loewenstein G, et al: Separate neural systems value immediate and delayedPrint: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

24 of 26

10/05/2009 15:53

monetary rewards. Science 306:503–507, 2004 [PubMed]

Menon RS, Goodyear BG: Submillimeter functional localization in human striate cortex using BOLD

contrast at 4 Tesla: implications for the vascular point-spread function. Magn Reson Med 41:230–235,

1999 [PubMed]

Meyer-Lindenberg A, Weinberger DR: Intermediate phenotypes and genetic mechanisms of psychiatric

disorders. Nat Rev Neurosci 7:818–827, 2006 [PubMed]

Meyer JH: Imaging the serotonin transporter during major depressive disorder and antidepressant

treatment. J Psychiatry Neurosci 32:86–102, 2007 [PubMed]

Meyer JH, McMain S, Kennedy SH, et al: Dysfunctional attitudes and 5-HT2 receptors during depression

and self-harm. Am J Psychiatry 160:90–99, 2003 [Full Text] [PubMed]

Mintun MA, Raichle ME, Kilbourn MR, et al: A quantitative model for the in vivo assessment of drug

binding sites with positron emission tomography. Ann Neurol 15:217–227, 1984 [PubMed]

Mitterschiffthaler MT, Kumari V, Malhi GS, et al: Neural response to pleasant stimuli in anhedonia: an

fMRI study. Neuroreport 14:177–182, 2003 [PubMed]

Montague PR, Berns GS, Cohen JD, et al: Hyperscanning: simultaneous fMRI during linked social

interactions. Neuroimage 16:1159–1164, 2002 [PubMed]

Moore CM, Frederick BB, Renshaw PF: Brain biochemistry using magnetic resonance spectroscopy:

relevance to psychiatric illness in the elderly. J Geriatr Psychiatry Neurol 12:107–117, 1999 [PubMed]

Moore CM, Wardrop M, deB Frederick B, et al: Topiramate raises anterior cingulate cortex glutamine

levels in healthy men: a 4.0 T magnetic resonance spectroscopy study. Psychopharmacology (Berl)

188:236–243, 2006 [PubMed]

Moore CM, Frazier JA, Glod CA, et al: Glutamine and glutamate levels in children and adolescents with

bipolar disorder: a 4.0 T proton magnetic resonance spectroscopy study of the anterior cingulate cortex.

J Am Acad Child Adolesc Psychiatry 46:524–534, 2007 [PubMed]

Moore GJ, Bebchuk JM, Wilds IB, et al: Lithium-induced increase in human brain gray matter. Lancet

356:1241–1242, 2000 [PubMed]

Nemeroff CB: Prevalence and management of treatment-resistant depression. J Clin Psychiatry 68 (suppl

8):17–25, 2007

Nemeroff CB, Heim CM, Thase ME, et al: Differential responses to psychotherapy versus pharmacotherapy

in patients with chronic forms of major depression and childhood trauma. Proc Natl Acad Sci U S A

100:14293–14296, 2003 [PubMed]

Nestor PG, Kubicki M, Spencer KM, et al: Attentional networks and cingulum bundle in chronic

schizophrenia. Schizophr Res 90:308–315, 2007 [PubMed]

Neumeister A, Nugent AC, Waldeck T, et al: Neural and behavioral responses to tryptophan depletion in

unmedicated patients with remitted major depressive disorder and controls. Arch Gen Psychiatry

61:765–773, 2004 [PubMed]

Ogawa S, Tank DW, Menon R, et al: Intrinsic signal changes accompanying sensory stimulation:

functional brain mapping with magnetic resonance imaging. Proc Natl Acad Sci U S A 89:5951–5955,

1992 [PubMed]

Ono M, Wilson A, Nobrega J, et al: 11C-labeled stilbene derivatives as Abeta-aggregate-specific PET

imaging agents for Alzheimer’s disease. Nucl Med Biol 30:565–571, 2003 [PubMed]

Owen AM, McMillan KM, Laird AR, et al: N-back working memory paradigm: a meta-analysis of normative

functional neuroimaging studies. Hum Brain Mapp 25:46–59, 2005 [PubMed]

Parsey RV, Krishnan KR: A new MRI ratio method for in-vivo estimation of signal hypointensity in aging

and Alzheimer’s disease. Prog Neuropsychopharmacol Biol Psychiatry 21:1257–1267, 1997 [PubMed]

Pearlson GD, Calhoun V: Structural and functional magnetic resonance imaging in psychiatric disorders.

Can J Psychiatry 52:158–166, 2007 [PubMed]Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

25 of 26

10/05/2009 15:53

Peper JS, Brouwer RM, Boomsma DI, et al: Genetic influences on human brain structure: a review of

brain imaging studies in twins. Hum Brain Mapp 28:464–473, 2007 [PubMed]

Pezawas L, Meyer-Lindenberg A, Drabant EM, et al: 5-HTTLPR polymorphism impacts human

cingulate-amygdala interactions: a genetic susceptibility mechanism for depression. Nat Neurosci

8:828–834, 2005 [PubMed]

Pizzagalli DA, Oakes TR, Davidson RJ: Coupling of theta activity and glucose metabolism in the human

rostral anterior cingulate cortex: an EEG/PET study of normal and depressed subjects. Psychophysiology

40:939–949, 2003 [PubMed]

Post RM, Speer AM, Hough CJ, et al: Neurobiology of bipolar illness: implications for future study and

therapeutics. Ann Clin Psychiatry 15:85–94, 2003 [PubMed]

Purselle DC, Nemeroff CB: Serotonin transporter: a potential substrate in the biology of suicide.

Neuropsychopharmacology 28:613–619, 2003 [PubMed]

Rauch SL, Whalen PJ, Shin LM, et al: Exaggerated amygdala response to masked facial stimuli in

posttraumatic stress disorder: a functional MRI study. Biol Psychiatry 47:769–776, 2000 [PubMed]

Ravina B, Eidelberg D, Ahlskog JE, et al: The role of radiotracer imaging in Parkinson disease. Neurology

64:208–215, 2005 [PubMed]

Reimold M, Smolka MN, Schumann G, et al: Midbrain serotonin transporter binding potential measured

with [(11)C]DASB is affected by serotonin transporter genotype. J Neural Transm 114:635–639, 2007

[PubMed]

Reivich M, Kuhl D, Wolf A, et al: The [18F]fluorodeoxyglucose method for the measurement of local

cerebral glucose utilization in man. Circulation Research 44:127–137, 1979 [PubMed]

Reynolds GP, Templeman LA, Godlewska BR: Pharmacogenetics of schizophrenia. Expert Opin

Pharmacother 7:1429–1440, 2006 [PubMed]

Roffman JL, Marci CD, Glick DM, et al: Neuroimaging and the functional neuroanatomy of psychotherapy.

Psychol Med 35:1385–1398, 2005 [PubMed]

Rombouts SA, Barkhof F, Goekoop R, et al: Altered resting state networks in mild cognitive impairment

and mild Alzheimer’s disease: an fMRI study. Hum Brain Mapp 26:231–239, 2005 [PubMed]

Sanacora G, Mason GF, Rothman DL, et al: Increased cortical GABA concentrations in depressed patients

receiving ECT. Am J Psychiatry 160:577–579, 2003 [Full Text] [PubMed]

Schweitzer JB, Lee DO, Hanford RB, et al: A positron emission tomography study of methylphenidate in

adults with ADHD: alterations in resting blood flow and predicting treatment response.

Neuropsychopharmacology 28:967–973, 2003 [PubMed]

Sheline YI, Barch DM, Donnelly JM, et al: Increased amygdala response to masked emotional faces in

depressed subjects resolves with antidepressant treatment: an fMRI study. Biol Psychiatry 50:651–658,

2001 [PubMed]

Sheline YI, Mittler BL, Mintun MA: The hippocampus and depression. Eur Psychiatry 17 (suppl

3):300–305, 2002

Shin LM, Wright CI, Cannistraro PA, et al: A functional magnetic resonance imaging study of amygdala

and medial prefrontal cortex responses to overtly presented fearful faces in posttraumatic stress

disorder. Arch Gen Psychiatry 62:273–281, 2005 [PubMed]

Silverman DH, Cummings JL, Small GW, et al: Added clinical benefit of incorporating

2-deoxy-2-[18F]fluoro-D-glucose with positron emission tomography into the clinical evaluation of

patients with cognitive impairment. Mol Imaging Biol 4:283–293, 2002 [PubMed]

Small GW: Neuroimaging and genetic assessment for early diagnosis of Alzheimer’s disease. J Clin

Psychiatry 57 (suppl 14):9–13, 1996

Small GW, Kepe V, Ercoli LM, et al: PET of brain amyloid and tau in mild cognitive impairment. N Engl J

Med 355:2652–2663, 2006 [PubMed]Print: Chapter 10. Brain Imaging in Psychopharmacology http://www.psychiatryonline.com/popup.aspx?aID=426206&print=yes…

26 of 26

10/05/2009 15:53

Spiers HJ, Maguire EA: Spontaneous mentalizing during an interactive real world task: an fMRI study.

Neuropsychologia 44:1674–1682, 2006 [PubMed]

Stein MB, Simmons AN, Feinstein JS, et al: Increased amygdala and insula activation during emotion

processing in anxiety-prone subjects. Am J Psychiatry 164:318–327, 2007 [Full Text] [PubMed]

Tagamets M-A, Horwitz B: Interpreting PET and fMRI measures of functional neural activity: the effects

of synaptic inhibition on cortical activation in human imaging studies. Brain Res Bull 54:267–273, 2001

[PubMed]

Talbot PS, Laruelle M: The role of in vivo molecular imaging with PET and SPECT in the elucidation of

psychiatric drug action and new drug development. Eur Neuropsychopharmacol 12:503–511, 2002

[PubMed]

Taylor WD, Bae JN, MacFall JR, et al: Widespread effects of hyperintense lesions on cerebral white matter

structure. AJR Am J Roentgenol 188:1695–1704, 2007 [PubMed]

Thulborn KR, Waterton JC, Matthews PM, et al: Oxygenation dependence of the transverse relaxation

time of water protons in whole blood at high field. Biochimugia Biophysica Acta 714:265–270, 1982

[PubMed]

Tononi G, Edelman GM: Consciousness and complexity. Science 282:1846–1851, 1998 [PubMed]

Tucker DM, Luu P, Frishkoff G, et al: Frontolimbic response to negative feedback in clinical depression. J

Abnorm Psychol 112:667–678, 2003 [PubMed]

Turetsky BI, Colbath EA, Gur RE: P300 subcomponent abnormalities in schizophrenia, I: physiological

evidence for gender and subtype specific differences in regional pathology. Biol Psychiatry 43:84–96,

1998 [PubMed]

Van Horn JD, Gazzaniga MS: Databasing fMRI studies towards a “discovery science” of brain function.

Nat Rev Neurosci 3:314–318, 2002

Varela F, Lachaux JP, Rodriguez E, et al: The brainweb: phase synchronization and large-scale

integration. Nature Reviews Neuroscience 2:229–239, 2001 [PubMed]

Villringer A, Dirnagl U: Coupling of brain activity and cerebral blood flow: basis of functional

neuroimaging. Cerebrovasc Brain Metab Rev 7:240–276, 1995 [PubMed]

Waldvogel D, van Gelderen P, Muellbacher W, et al: The relative metabolic demand of inhibition and

excitation. Nature 406:995–998, 2000 [PubMed]

Whalen PJ, Rauch SL, Etcoff NL, et al: Masked presentations of emotional facial expressions modulate

amygdala activity without explicit knowledge. J Neurosci 18:411–418, 1998 [PubMed]

Whalen PJ, Shin LM, Somerville LH, et al: Functional neuroimaging studies of the amygdala in depression.

Semin Clin Neuropsychiatry 7:234–242, 2002 [PubMed]

Wise RG, Tracey I: The role of fMRI in drug discovery. J Magn Reson Imaging 23:862–876, 2006

[PubMed]

Zink CF, Pagnoni G, Chappelow J, et al: Human striatal activation reflects degree of stimulus saliency.

Neuroimage 29:977–983, 2006 [PubMed]

Copyright © 2009 American Psychiatric Publishing, Inc. All Rights Reserved.