- Email: [email protected]
SPECT imaging in psychiatry
SPECT Imaging in Psychiatry A Review Omer Bonne, M.D., Yodphat Krausz, M.D., and Bernard Lerer, M.D.
Abstract: In the last two decades, brain imaging has become an integral part of clinical and research psychiatry. Single photon computed emission tomography (SPECT) is rapidly gaining acceptance as one of the major imaging techniques available, along with computed tomography (CT), magnetic resonance imaging (MRlJ, and positron emission tomography (PET). Each of these techniques has its assets and drawbacks. This review concerns SPECT, a highly prevalent imaging technique whose potential value in brain imaging has not been appreciated until recently. Its purpose is to expose practicing clinicians and research psychiatrists alike to the attributes of this instrument, which is available in most nuclear medicine departments today. An effort is made to provide a comprehensive account of this technique, including a brief summary of the basic principles, the various methods of its application, and recent findings in most psychiatric disorders. Analogies to its “aristocratic cousin,” PET, are presented to emphasize similarities and differences. Finally, directions for future development and implementation of SPECT are suggested.
Introduction The role of biological factors in the etiology and pathogenesis of major psychiatric disorders has come under increasing scrutiny over the past three decades [l]. Nevertheless, extensive research efforts, based on a wide range of experimental strategies, have not been able to demonstrate definitive and replicable abnormalities of brain structure or function in any of the disorders studied. In-vivo imaging of the human brain began to play a major role in this field beginning in the early
From the Departments of Psychiatry (O.M., B.L.) and Nuclear Medicine and Medical Biophysics (Y.K.), Hadassah University Hospital, Jerusalem, Israel. Address reprint requests to: Dr. Omer Bonne, Department of Psychiatry, Hadassah University Hospital, P.O. Box 12000, Jerusalem 91120, Israel.
296 ISSN 0163-8343/92/$5.00
1970s with the development of computed tomography (CT), and the ability to measure regional cerebral blood flow (rCBF) by the use of Xe-133. These modalities exemplify the two main branches of imaging technology today: “structural” imaging, i.e., CT scan and magnetic resonance imaging (MRI); and “functional” imaging, i.e., positron emission tomography (PET) and single photon emission computed tomography (SPECT). As the term implies, structural imaging provides information of an anatomical nature whereas functional techniques are designed to reflect the level of brain metabolic activity, either directly or by extrapolation from regional cerebral blood flow measurements. Each of these technologies has made a significant contribution to the expanding body of data available (21. The present review concerns SPECT, the newcomer to the field of brain imaging. Its purpose is to provide practicing clinicians with basic insights into the theoretical concepts of the technique, its practical application, possible diagnostic utility, and potential research contribution. As disorders at the interface of psychiatry and neurology are presently the most amenable to study with SPECT, these will be emphasized, but other potential applications will also be considered.
Background Efforts to directly evaluate brain function began with invasive procedures by Kety and Schmidt in 1948 [3] which involved intracarotid injection of nitrous oxide and internal jugular venous sampling, and provided gross estimates of whole brain perfusion. The advent of radioisotopes and collimated detectors allowed for regional measurement Gr,~rral Hospital Psychiatry 14, 296-306, 1992 0 1992 Elsevier Science Publishing Co., Inc. 655 Avenue of the Americas, New York, NY 10010
SPECT Imaging
of cerebral blood flow. Further advances in detector sensitivity and curve-fitting algorithms enabled replacement of intracarotid injections with either inhalation or intravenous injection of the radiotracer, most notably Xe-133, which for several years was the uniform radiotracer and is still frequently used today. These measurements all utilized twodimensional methods for assessing brain function. The landmark finding of decreased rCBF in the frontal lobes of schizophrenic patients by Ingvar and Franzen in 1974 [4] delineated the important potential role of functional imaging in the study of psychiatric conditions. Tomographic imaging of rCBF was first successfully done by Yamamoto et al. in 1977 [5] employing Kr-77, and soon afterward Xe-133 was also utilized. Notwithstanding these promising results, it was only several years later, and especially after the development of PET, that research in this direction was intensified. The potential value of PET imaging in psychiatry is beyond doubt [6,7]. However, its extremely high purchase and operating costs significantly limit its availability. Increasing efforts are therefore being made to establish the capability of SPECT, a much more common and less costly instrument, to provide reliable and comparable data.
Basic Principles SPECT is a three-dimensional method capable of reflecting the distribution of a radiotracer within the human body. It involves administration of a radiotracer and either simultaneous or delayed detection by a tomographic instrument. It is frequently used for purposes other than brain imaging, mostly cardiac assessment, not to be described in this review. Radiotracers emit gamma rays uniformly in all directions. In order to localize radiation coming from a specific brain area, a collimator, which is a lead sheet containing long apertures allowing only parallel rays to penetrate and blocking all others, is placed in front of the detector. As all parallel rays arriving from any brain depth along a line will pass through the collimator, this process will not suffice to fix the exact point from which a ray is emitted. This is the conventional planar nuclear medicine image. To expand two-dimensional information into a threedimensional picture, additional two-dimensional projections should be gathered from different angles. This is usually accomplished by rotating a gamma camera (complex of collimator and radiation detector) around the head of the examinee,
in Psychiatry
although other options are also possible. These reconstructed projections are displayed as crosssections through the brain. (A more detailed technical description of SPECT can be found in [8].)
Radiopharmaceuticals
(Radiotracers)
Radiopharmaceuticals used for SPECT imaging can be classified into three categories: diffusible, static, and neuroreceptor ligands. Diffusible tracers, as their name implies, do not remain attached to any specific brain area. They freely traverse the blood-brain barrier and do not chemically interact with brain parenchyma. This mode of study is designated dynamic or D-SPECT. The routinely available tracer for this method is Xe133. Information is drawn from detection of the transit of the radiotracer within the brain by the application of a mathematical model. The low gamma-ray energy and diffusibility of Xe-133 significantly limit the resolution of these studies. Lipophilic radiotracer complexes readily pass the blood-brain barrier. Once located within the brain they lose their lipophilic capacity and undergo an as yet unclear chemical reaction which traps them within the brain in a defined area. The first static radiotracers were used in the early 1980s ([9,10]; these were I-123 isotopes bound to an amine structure. Higher energy and fixed distribution permits the use of an ordinary gamma camera with improved resolution compared with D-SPECT. Presently available I-123 labeled radiotracers are I-123-IMP (iodoisopropylamphetamine) and 1-123HIDPM (hydroxyisopropyldiamine). Tc-99m has been the mainstay of nuclear medicine for many years. It has better dosimetry and physical properties than does 1-123. However, technetium chemistry is rather complex and it has only been in the past few years that the labeling of Tc-99m to suitable carriers has been achieved [ll]. Because of their better dosimetry characteristics, high availability, and low cost, Tc-99m tracers have become the routine radiotracers used today [12]. The most prevalent Tc-bound radiotracer and the only one that has been approved by the FDA for clinical application is Tc-99m-HMPAO (hexamethylpropyleneamineoxime). This agent is highly lipophilic but unstable. Once settled, HMPAO remains fixed, reflecting distribution at time of administration. Resolution of SPECT is also improving with the use of Tc-99m-bound tracers. For D-SPECT with Xe-133 the resolution of SPECT is about 17 mm, but new detectors in conjunction
297
0. Bonne et al.
with Tc-99m tracers are capable of resolution of 8 mm. These values are approaching PET values. The third class of radiotracers are neuroreceptor ligands. This group of radiotracers, more prevalent in PET, enable access to receptor sites within the brain and allow a certain degree of receptor imaging. The development of this type of radiotracer for SPECT is in its first stages. A few reports have appeared describing successful labeling of muscarinic cholinergic [13] and dopaminergic Dl [14] and D2 [15] receptors. A single report presented the possibility of producing SPECT benzodiazepine receptor binding [16]. All SPECT receptor binding agents are at this time I-123 labeled. This whole class of radiotransmitter ligand radiotracers is at present experimental.
Image Interpretation Reliable analysis of SPECT data remains a major challenge. Various factors have to be taken into consideration. A correlation between radiotracer distribution within the brain and cerebral blood flow has to be established. Studies do point to such a correlation [17], but opinion varies as to the mode of relationship. Comparison of SPECT studies to those performed with PET tracers disclosed a good correlation, both in animal [18] and human [19] subjects, in normal or pathological conditions. In fact, rCBF has even been reported to reflect neuronal activity more accurately than oxygen consumption in visual cortex stimulation [20]. SPECT image data thus reflect brain function. In order for this data to be useful, the anatomic location of the image must be as precisely defined as possible. Patients must be positioned carefully. Image slices are taken according to well-defined external skull anatomical landmarks, most commonly the orbitocanthal meatal line (OML). Additional parallel slices at specified distances from this imaginary line are taken, as needed. A brain atlas, CT scan, MRI, or simple skull x-ray films are used to define brain regions at each slice level. This method is only an approximation, as individual brain position relative to external landmarks may vary somewhat. When aimed at specific brain areas, SPECT (and PET) will define certain anatomic locations as regions of interest (ROI). Although the possibility of acquiring absolute numerical rCBF values exists, (with Xe-133), this method involves arterial sampling and is rarely used today. Most studies measure tracer uptake from the ROI, define the area,
298
and so receive mean uptake for each ROI. This value is related to a reference area, most often either mean whole slice uptake or a specified brain region (i.e., occipital lobe, cerebellum, and so forth). The ratio received from the division of the ROI value by the reference value is usually presented as “rCBF.” This value can give an indication of either hyper or hypoperfusion in a certain area and can be approximately compared between different studies. A comparison is also sometimes made between symmetrical ROIs from the two hemispheres (see [8].) Visual analysis, when carried out by multiple experienced observers, can supply useful information. It also compares ROIs to reference areas or to the contralateral hemisphere. Visual interpretation may sometimes be more sensitive than the quantitative computerized method described above [21]. An important aspect of functional in contrast to structural imaging is the condition of the patient at the time of examination. Structure of the brain is expected to remain constant regardless of brain activity whereas brain function, by definition, is expected to change. Grossly, functional studies are undertaken either in the “resting” or “activated” states. Resting condition ideally implies absence of all somatic, sensory, cognitive and affective stimuli. However, there is no standardized resting state, and issues such as eyes open vs eyes closed or ears open vs ears plugged during examination vary between studies. Moreover, the degree of anxiety experienced or the presence of delusions in psychotic patients may cause wide variations in this resting state. These elements (and many more not mentioned for reasons of brevity) complicate reproducibility of SPECT studies and may account for apparent conflicting results.
Neuropsychological and Activation Studies Introduction of neuropsychological testing in conjunction with SPECT studies is becoming common procedure. Correspondence between neuropsychological deficits and SPECT pattern is evaluated. Correlation between findings serves as indirect proof of the capability of SPECT to generate meaningful data reflecting damaged brain areas. In schizophrenics, for example, Paulman et al. [21] used the Wisconsin Card Sorting Test (WCST) and the Luria-Nebraska battery to describe association of left frontal hypoperfusion with impairment, and Weinberger et al. [22] found failure in activation of
SPECT Imaging
in Psychiatry
the dorsolateral prefrontal cortex when performing the WCST. Activation studies involve either pharmacological intervention or administration of a predetermined task or stimulus. While activated, the patient is surveyed by a functional imaging technique, and deviation from the resting state is assessed. Pharmacological agents utilized include acetazolamide, respiration of 5% CO,, and psychotropic active compounds. The tasks or stimuli may also be sensory, somatic, or cognitive. An extensive account of the technique, application, and findings has been recently presented by Tikofsky and Hellman [23]. Activation studies and neuropsychological testing in combination with functional imaging are invaluable both in providing insights into normal brain performance and detection of brain pathology which may not be evident in the resting state.
SPECT Findings in Psychiatric Disorders Measurements of rCBF have been made for about 40 years IS], but it is due to the development of PET that functional imaging has begun to assume a leading role in psychiatric research. SPECT is not expected to supersede PET’s diagnostic capabilities or research potential. However, it is widely available and relatively inexpensive, allowing access to almost any suitable psychiatric case. As mentioned, most SPECT studies carried out today employ HMPAO, approved by the FDA and introduced into clinical practice only in 1988. Most SPECT studies published in recent years, although this is not always explicitly stated, have tried to replicate findings obtained by PET and the older Xe-133 rCBF two-dimensional studies. SPECT findings in psychiatric disorders will be reviewed. Not every SPECT study concerning psychiatric disorders will be discussed, as some sporadic reports do not significantly contribute to an understanding of the instrument or the disorder under study. Relevant data from PET, structural imaging, and electroencephalographic studies will be noted when appropriate.
Dementia Dementia is the most common neuropsychiatric disorder in middle-aged and elderly people. Its prevalence ranges between 10 1241 and 15% [8] in populations over 65 years of age, although it may
occur in younger age groups. Over 50% [8,16] of patients with dementia are believed to suffer from Alzheimer’s disease (AD), a diagnosis usually made by exclusion, which can only be definitively confirmed at autopsy. Another 20-30% suffer from multi-infarct dementia (MID), and in a substantial number of patients (around 20%) both disorders may be present. Ten percent of patients presenting with dementias are believed to be actually suffering from depression, a condition sometimes termed “pseudodementia” which is reversible upon adequate antidepressant treatment. Other and rarer causes of dementia-metabolic and endocrine abnormalities, those associated with movement disorders (Huntington’s and Pick’s), Korsakoff’s psychosis and others-account for the remaining part. SPECT studies of AD have yielded unequivocal findings. All studies report symmetrically reduced uptake in the parietal lobes [24-291, as demonstrated in Figure 1. Temporal lobe involvement is also commonly encountered [24-27,291 Frontal and occipital lesions are sometimes observed in severe, advanced cases [25,28]. These findings are consistent with those reported for PET [28,30-321. Promising results have been achieved in several studies [23,26,29] correlating neuropsychological deficits and cognitive impairments with areas of localized brain damage. These studies lend support to the
299
0. Bonne et al.
opinion that the cognitive deterioration noted in AD can be attributed to specified brain areas, and does not reflect diffuse brain damage. In MID, in contrast to AD, the SPECT image may show asymmetrical patchy areas of reduced uptake , which often correspond to the distribution of major cerebral arteries [24,25,27,28]. In Huntington’s disease [27], reduced uptake was observed in the caudate nuclei, whereas impairment in frontal lobe function was noted in patients suffering from Korsakoff’s psychosis that was correlated to cognitive deficits as measured by neuropsychological testing [29]. Severe compromise in frontal lobe perfusion was prominent in Pick’s dementia and progressive supranuclear palsy (PSP) [25].
Epilepsy Symptomatology in epilepsy, particularly temporal lobe (partial-complex, psychomotor) epilepsy (TLE) can mimic the whole range of psychiatric phenomena [33]. Disturbances in thought form and content, perception and level of consciousness are commonly observed. Cognitive ability and memory may be compromised. Behavior may be confused or automatic. These phenomena may occur in the ictal, postictal, or interictal phases of the disease. Patients with no known history of convulsive disorder are therefore frequently suspected to be suffering from an epileptic disorder, especially if the observed syndrome is atypical or resistant to standard medication. Success in controlling psychiatric symptoms has been achieved by the administration of the anticonvulsant carbamazepine to patients with diagnosed temporal lobe epilepsy [34,35]. Furthermore, patients without diagnosed TLE may respond to carbamazepine particularly when affective symptoms or behavioral dyscontrol are prominent [34]. Whether these patients have an epileptic focus that cannot be demonstrated by conventional means is unclear, as TLE frequently eludes electroencephalographic (EEG) confirmation, even when nasopharyngeal leads are used as well as assessments by CT and MRI. SPECT findings in TLE have been most consistent (Fig. 2) and are comparable to those achieved by PET [28,36]. In fact, ictal SPECT studies may be simpler to carry out due to the stability and longer half-life of static SPECT tracers. Epileptogenic foci are detected in the ictal and interictal phases by hyper and hypoperfusion, respectively [37,38]. 300
Figure 2. HMPAO brain SPECT of temporal lobe epilepsy. Transaxial slice disclosirrg perfusion deficit is in the left temporal area (arrow).
Moreover, low flow areas, remote from the seizure focus and occasionally in the contralateral hemisphere, have been identified by SPECT[37,39]. These areas correspond to patterns of cognitive impairment as located by neuropsychological testing [22,38,40]. SPECT may also prove helpful in monitoring brain response to anticonvulsant therapy
[411.
Schizophrenia Schizophrenia, the most extensively studied of all psychiatric disorders, has been studied by numerous functional imaging techniques which have set out to reproduce the finding of reduced cerebral blood flow in frontal lobes of schizophrenic patients described by Ingvar and Franzen in 1974 [4]. The large number of studies (primarily PET) has served to emphasize the complexity of this disorder. It is beyond the scope of this review to go into a detailed description of all studies in this field so a general overview will be presented. In a recent review, Buchsbaum [42] summarized findings from PET studies according to three brain areas: frontal lobes, basal ganglia, and (mainly left) temporal lobes. Frontal lobe function is usually presented as a ratio between frontal and either occipital or whole brain metabolism or blood flow. The value thus attained is compared to a parallel assessment of normal brain function. In Buchsbaum’s review, most studies showed lower ratios in schizo-
SPECT Imaging
phrenic patients, some of them significantly. However, a minority of studies found no difference between schizophrenics and normal controls, and a few even showed contradictory results. It should be noted that studies differ in methodology as to sensory or cognitive stimulation, medication status, exact ROIs, and other factors that might influence findings. Results from the other two brain areas examined generally follow the same trend and suffer from the same limitations. The effect of neuroleptic treatment may be more crucial to basal ganglia studies. In schizophrenic patients scanned both on and off medication, a greater increase in metabolic rate was noted in basal ganglia compared with cortical regions in the medicated state [43]. Are there three types of schizophrenia, each conforming to a distinct anatomical area? Are these lesions found to a certain extent in all schizophrenic patients? Does the basic flaw reside in faulty connections between brain regions, as others assume [44]? In the resolution of these and other fundamental issues, functional imaging should play a major part. Few SI’ECT studies are available on schizophrenia. The most methodologically sound report [21] published to date is a D-SPECT Xe-133 inhalation
in Psychiatry
study on a group of 40 medicated and nonmedicated, paranoid and nonparanoid schizophrenic patients, combined with neuropsychological testing. Absolute bilateral hemispheric CBF was elevated in schizophrenic patients compared with a normal control group (a finding not consistent with PET findings [45-471 which show no significant difference between patient and control populations). Higher CBF values were observed in the right hemiphere in both patient and control groups. Reduced relative left frontal rCBF was associated with neuropsychological impairment, and increased whole brain CBF with relative bilateral hypofrontal and hypotemporal CBF was correlated with paranoid symptoms. Another report [48] presented a group of 28 female schizophrenic patients, divided into subgroups of “positive” and “negative” symptom, “acute” and “chronic” patients. HMPAO uptake was normal in the positive-acute group (mostly paranoid) and reduced in temporal regions and basal ganglia in the negative-acute and chronic group. Type of hallucination was reported to determine rCBF distribution in a third HMPAO SPECT study [49]. Increased rCBF was found in hippocampal regions of patients suffering from auditory hallucinations, whereas patients with tactile hallucinations manifested reduction in rCBF in inferior temporal regions. Less methodologically sound studies reported multifocal sites of increased and decreased perfusion in all but one of 20 patients [50] or completely negative findings in a group of seven patients in another HMPAO study [51]. A pilot HMPAO SPECT study [52] examined 18 patients diagnosed as either schizophrenic, schizoaffective, or depressed. Hypofrontality was significantly more pronounced in schizophrenic as compared with depressed patients, with schizoaffective patients occupying an intermediate position. All three groups, however, showed some evidence of hypofrontality. Overall, SPECT studies in schizophrenia have not yielded a clear pattern. However, most of the work reported thus far has been of a preliminary nature, and methodological issues noted earlier in this review remain to be effectively and consistently addressed.
Affective Disorders Even though depression is a highly prevalent disorder, imaging studies on this disorder have been relatively neglected. Two-dimensional blood flow
301
0. Bonne et al.
studies reported either normal flow patterns [53] in depressive patients or displayed contradictory results showing increased [54] or decreased [55] flow in left hemispheric structures and opposite findings in the other hemisphere. Spatial and verbal activation enabled discrimination between depressive and control subjects who were nondistinguishable before activation in another rCBF study [56]. PET studies have not proved more conclusive. One study reported normal whole-brain glucose metabolism with a specific reduction in the inferior left frontal cortex [57]. Several studies made a distinction between subtypes of affective disorder. Patients were grouped into unipolar and bipolar depression, bipolar mixed, and bipolar manic. Buchsbaum et al., [58] reported a decrease in anterior-posterior ratio in bipolar patients while receiving painful electrical shocks to the forearm. Baxter et al. [59] described lower whole-brain supratentorial glucose metabolic rates in bipolar depressed and bipolar mixed patients than in unipolar depressed patients. Metabolic rate significantly increased after clinical improvement. The same group reported (on a small number of patients) in a later study [60] a lower ratio of glucose metabolic rates in the caudate nucleus divided by metabolic rate in the ipsilateral supratentorial hemisphere in unipolar-depressed but not in bipolar-depressed patients; these findings pertain only to the left side. The findings from the last three studies suggest a supratentorial-prefrontal pathology in bipolar depressed patients and caudate nucleus pathology in unipolars. Baxter et al., [61] however, in a later PET study, noted a decrease in prefrontal to ipsilateral hemisphere glucose metabolic ratio in all categories of depressive patients studied. A significant increase in this ratio was noted as patients responded to medication. Another SPECT study [62] disclosed significantly lower CBF in unipolar depressive patients compared with bipolar depressive patients or control subjects. CBF was higher than normal in bipolar manic and mixed patients in the same study. A subsequent study by the same group 1631 localized diminished flow to the right parietal and temporal lobes in unipolar depressive subjects and contralateral hyperperfusion in the left temporal and parieta1 lobes in bipolar manic or mixed patients. In a recent controlled study [64] on a group of 41 dea reduction in global cortical pressed patients, blood flow was noted, which was attributed mostly to reductions in selective frontal, central, superior
302
temporal, and anterior parietal regions. Unipolar and bipolar depressed patients were not separated except for the exclusion of rapid cycling bipolar patients. Measurement of CBF in conjunction with electroconvulsive therapy (ECT) has intrigued researchers and there are several reports on this entity. CBF was measured by the two-dimensional intracarotid Xe-133 method [65], two-dimensional Xe-133 inhalation method [66], Xe-133 D-SPECT [67], HMPAO SPECT [68], and PET 1691. Pretreatment CBF values were most often within normal limits. Posttreatment findings include decrease in CBF [67], increase in CBF [70], or redistribution of tracer uptake [68]. No distinction between subtypes of depression is mentioned in all the above studies. Finally, a recent HMPAO SPECT study [71] found a high correlation between severity of depressive symptoms and volume of ischemic lesion in patients suffering from poststroke depression.
Other Neuropsychiatric Disorders Only sparse SPECT assessments of other neuropsychiatric disorders are found. These provide only preliminary insights into the cerebral blood flow of these conditions and are briefly alluded to for the sake of completeness. Structural imaging of severe eating disorders (emaciated anorexia as well as normal-weight bulimia) has shown morphological brain alteration in these disorders. Findings were characterized by enlargement of cerebrospinal fluid (CSF) spaces and/ or ventricular dilatation [72,73]. A deficit in neuropsychological test performance was noted in conjunction with morphological abnormalities [73]. A PET study [74] and a D-SPECT study [75] found no evidence for a reduced functional capacity of the brain in the anorectic state, reflected either in glucose metabolism 1741 or in cerebral blood flow ]751. Few attempts to evaluate the classical “neuroses” have been reported. Woods et al. [76], in an activation HMPAO SPECT study, observed rCBF differences following yohimbine administration compared with placebo injection in panic disorder (I’D) patients, not evidenced in normal controls. An IMP SPECT study [77] compared PD patients before and after pharmacological treatment, and found nonsignificant rCBF changes. Reduced uptake in the right anterior frontal area was found, however, in panic disorder patients but not in nor-
SPECT Imaging
ma1 controls. A comprehensive review of anxiety and CBF is presented by Mathew and Wilson [78], emphasizing physiological considerations of alterations in CBF. In obsessive-compulsive disorder (OCD), PET findings [79] have included increased glucose metabolic rates in the left orbital gyrus and bilateral caudate nuclei. These findings were distinguishable from both normal controls and unipolar depressed patients. A follow-up study by the same group [80] found bilateral increased glucose metabolic rates in orbital gyri. The inconsistency from the first report is explained by methodological faults in the earlier study. Two National Institute of Mental Health (NIMH) groups have reported findings in OCD. The first [Bl] reported increased metabolic rates in both orbital gyri. The second group [82] found increased absolute glucose metabolism in the left orbital and right sensorimotor regions and bilaterally in the anterior cingulate gyri and lateral prefrontal area. Moreover, patients who responded to clomipramine therapy had significantly lower right anterior cingulate and right orbital metabolism than did nonresponders. CBF has been measured during the acute stage and after recovery from alcohol withdrawal reaction. Hemmingsen et al. [83] noted in a D-SPECT study a significant decrease in rCBF in alcoholwithdrawn patients with visual hallucinations and psychomotor agitation. Decreases were not significant in withdrawal patients not manifesting these disturbances. Other studies, however, showed prerecovery low CBF values [84,85] that normalized after recovery.
Conclusions SPECT imaging in psychiatry has not yet entered routine clinical use, although the instrumentation needed for its application is available in most nuclear medicine departments. Psychiatric diagnosis relies at present almost solely on clinical expertise. SPECT image patterns may prove to be of value in differentiating neuropsychiatric syndromes with similar clinical presentation, as well as in clarifying the etiology of atypical, seemingly obscure behavioral disorders. Implementation of SPECT in the workup of patients with cognitive deterioration should become common practice. Differential diagnosis between dementia and depression (presenting as pseudodementia) can be substantially facilitated. Discrimination between the various types of dementia can
in Psychiatry
also be assisted, enabling determination of prognosis and the planning of treatment. Patients manifesting disturbances in level of consciousness or conduct and atypical psychiatric symptomatology may benefit from SPECT study, particularly if CT, MRI, and EEG are negative and a diagnosis of temporal lobe epilepsy is suspected. For the major psychoses (schizophrenia and affective disorders), SPECT does not yet have a clear clinical application. This is even more true for the neurotic disorders. SPECT does offer, however, considerable research potential in these disorders. Systematic study of large groups of subjects may yet reveal consistent pathological findings. Some of the most intriguing issues in clinical psychiatry today could benefit from the introduction of SPECT into research protocols: Is schizophrenia a single disease or are there several autonomous entities which we wrongly designate as schizophrenia? Is schizoaffective disorder schizophrenia or a separate disorder, or is it a subtype of schizophrenia, or of the effective disorders? Are major depression and psychotic depression separate entities? Are there pathological differences between unipolar and bipolar depressed patients? Is it possible to demonstrate brain pathology in the neuroses? These are just a few examples of the clinical research potential of SPECT. It may also contribute to other areas of research. For instance, it could be effectively utilized in the study of normal brain function, such as the ongoing effort to establish a relationship between cognitive faculties and regional brain activity. Activation studies and neuropsychological testing in conjunction with SPECT are thus on the way to becoming common research practice. Further development of receptor binding radiotracers and their research application could result in the elucidation of biochemical pathways in the normal human brain and in psychopathological states. This capacity of SPECT could also enable objective drug response monitoring in psychiatric patients. SPECT is a newcomer to the field of psychiatry; it has yet to mature. Validation and sensitivity/ specificity studies are urgently needed if it is to manifest its full potential. The relatively wide availability of SPECT permits the inclusion in research protocols of a sufficient number of subjects to reliably substantiate research findings. Particular attention should be given to standardizing methodology and to achieving consistency with regard
303
0. Bonne et al.
to
the state in which subjects are evaluated. Once these objectives are achieved, SPECT could prove to be a valuable clinical and research tool in modern psychiatry.
17.
18.
References 1. Meltzer HY (ed): Psychopharmacology: The Third Generation of Progress. New York, Raven Press, 1987 2. Andreasen NC: Brain imaging: applications in psychiatry. Science 239:1381-1388, 1988 3. Kety SS, Schmidt CF: The nitrous oxide method for quantitative determination of cerebral blood flow in man: theory, procedure and normal values. J Clin Invest 27~475-483, 1948 4. Ingvar DH, Franzen G: Abnormalities of cerebral blood flow distribution in patients with chronic schizophrenia. Acta Psychiatr Stand 50:425-464, 1974 5. Yamamoto YL, Thompson CJ, Meyer E, et al: Dynamic positron emission tomography for a study of cerebral hemodynamics in a cross section of the head using positron-emitting 68-Ga-EDTA and 77-Kr. J Comput Assit Tomogr 1:43-56, 1977 6. Cohen RM, Semple WE, Gross M: Positron emission tomography. Psychiatr Clin North Am 9:1:63-79, 1986 7. Jacobson HG: Positron emission tomography-a new approach to brain chemistry. JAMA 260:27042710, 1988 8. Devous MD: Imaging brain function by single photon computed emission tomography. In Andreasen NC (ed), Brain Imaging: Applications in Psychiatry. Washington DC, American Psychiatric Press, 1989 9. Winchell HS, Baldwin RM, Lin TH: Development of I-123 labeled amines for brain studies: localization of I-123 iodophenylalkylamines in rat brain. J Nucl Med 21:940-947, 1980 10. Kung HF, Tramposch KM, Blau M: A new brain perfusion agent: I-123-HIDPM. J Nucl Med 24:66-72, 1983 11. Kung HF: New Technetium 99m-labeled brain perfusion imaging agents. Semin Nucl Med 2:150-158, 1990 12. Lassen NA, Blasberg RG: Technetium-99m-d, lHM-PAO, the development of a new class of Tc-99m labeled tracers: an overview. J Cereb Blood Flow Metab S:Sl-S3, 1988 13. Thonoor CM, Couch MW, Geer DM, et al: Biodistribution and radiation dosimetry of radioiodinated SCH23982, a potential D-l receptor imaging agent. J Nucl Med 29:1668-1674, 1988 14. Kung HK, Pan S, Kung MI’, et al: In vitro and in vivo evaluation of [1123]IBZM: a potential CNS D-2 dopamine receptor imaging agent. J Nucl Med 30:8892, 1989 15. Eckelman WC, Reba RC, Rzesotarsky WJ, et al: External imaging of cerebral muscarinic achetylcholine receptors. Science 223:291-293, 1984 16. Saji H, Nakatsuka I, IIda Y, et al: Radioiodinated
304
19.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
30.
31.
32.
33.
diazepam derivative for SPECT studies of benzodiazepine receptor. J Nucl Med 30:803-804, 1989 Neirinckx RD, Canning LR, Piper IM, et al: Technetium 99m d, I-HMPAO: a new radiopharmaceutical for SPECT imaging of regional cerebral blood perfusion. J Nucl Med 28:191-202, 1987 Sokoloff L: Localization of functional activity in the central nervous system by measurement of glucose utilization with radioactive deoxyglucose. J Cereb Blood Flow Metab 1:7-36, 1981 Lebrun-Gradie P, Baron JC, Soussaline F, et al: Coupling between regional blood and oxygen utilization in the normal brain. Arch Neurol 40:230-236, 1983 Fox PT, Raichle ME, Mintun MA, Dence C: Nonoxidative glucose consumption during focal physiologic neural activity. Science 241:462-464, 1988 Paulman RG, Devous MD, Gregory RR, et al: Hypofrontality and cognitive impairment in schizophrenia: dynamic single photon tomography and neuropsychoiogical assessment of schizophrenic brain function. Biol Psychiatry 22377-399, 1990 Weinberger DR, Berman KF, Zec RF: Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia. Arch Gen Psychiatry 43:114-124, 1986 Tikofsky RS, Hellman RS: Brain single photon emission tomography: newer activation and intervention studies. Semin Nucl Med 21:40-57, 1991 Kirn TF: Dementias appear to have individual profiles on single photon emission tomography. JAMA 261:965-967, 1989 Bonte FJ, Horn RT, Weiner MF: Single photon tomography in Alzheimer’s disease and the dementias. Semin Nucl Med 20342-352, 1990 Goldenberg G, Podreka I, Suess E, et al: The cerebral localization of neuropsychological impairment in Alzheimer’s disease: a SPECT study. J Neurol 236: 131-138, 1989 Smith FW, Besson JAO, Gemmell HG, Sharp PF: The use of technetium-99m-HM-PA0 in the assessment of patients with dementia and other neuropsychiatric conditions. J Cereb Blood Flow Metab 8:5116S122, 1988 Maurer AH: Nuclear medicine: SPECT comparisons to PET. Radio1 Clin North Am 26:5:1059-1074, 1988 Hunter R, McCluskie R, Wyper J, et al: The pattern of function-related regional cerebral blood flow investigated by single photon emission tomography with 99mTC - HMPAO in patients with presenile Alzheimer’s disease and Korsakoff’s psychosis. Psychol Med 19:87-855, 1989 Duara R, Grady CL, Haxby JV, et al: Positron emission tomography in Alzheimer’s disease. Neurology 36:879-887, 1986 Cohen RM, Semple, WE, Gross M: Positron emission tomography. Psychiatr Clin North Am 9:1:63-79, 1986 Jacobson HG: Positron emission tomography-a new approach to brain chemistry. JAMA 260:18: 2704-2710, 1988 Fenton GW: Psychiatric disorders of epilepsy: classification and phenomenology. In Reynolds EH,
SPECT Imaging
34.
35.
36. 37.
38.
39.
40.
41.
42.
43.
44. 45.
46.
47.
48.
49.
Trimble MR (eds), Epilepsy and Psychiatry. New York, Churchill Livingstone, 1981 Dalby MA: Behavioral effects of carbamazepine. In Penry JK, Daly D (eds), Complex Partial Seizures and Their Treatment: Advances in Neurology, Vol. II. New York, Raven Press, 1975 Kuhn R: The psychotropic effect of carbamazepine in nonepileptic adults, with particular reference to the drug’s possible mechanism of action. In Birkmayer W (ed), Epileptic Seizures-Behavior-Pain. Bern, Hans Huber, 1975 Editorial: SPECT and PET in epilepsy. Lancet 135137 January 21, 1989 Bonte FJ, Devous MD, Stokely EM, Homan RW: Single photon tomographic determination of regional cerebral blood flow in epilepsy. AJNR 4:544-546, 1983 Devous MD, Leroy RF, Homan RW: Single photon computed emission tomography in epilepsy. Semin Nucl Med 20:325-341, 1990 Lee BI, Markland ON, Wellman HN, et al: HIDPMSPECT in medically intractable complex partial seizures: ictal study. Arch Neurol 45:397-402, 1988 Homan RW, Paulman RC, Devous MD, et al: Neuropsvchologic function and regional cerebral blood flow m epilepsy. In Porter JR et al (eds), Advances in Epileptology: XVth Epilepsy International Symposium. New York, Raven, 1984 Homan RW, Devous MD, Stokely EM, Bonte FJ: Cerebellar blood flow in partial seizures (suppl 1) Neurology 37~327-334, 1987 Buchsbaum MS: The frontal lobes, basal ganglia and temporal lobes as sites for schizophrenia. Schizophr Bull 16:379-389, 1990 Buchsbaum MS, Wu JC, DeLisi LE, et al: Positron emission tomography studies of basal ganglia and somatosensorp cortex neuroleptic drug effects: differences between normal controls and schizophrenic paticnth. Biol Psychiatry 22:479-494, 1987 Robbins TW: The case f&r fronto-striatal dysfunction in schizophrenia. Schizophr Bull 16:391-402, 1990 DeLisi LE, Buchsbaum MS, Holcomb HH, et al: Clinical correlates of decreased anteroposterior metabolic gradients in Positron Emission Tomography (PET) of schizophrenic patients. Am J Psychiatry 142:7881, 1985 Sheppard G, Gruzelier J, Manchanda R, et al: 015 Positron emission tomographic scanning in predominantly never-treated schizophrenic patients. Lancet 2:1448-1452, 1983 Cohen RM, Semple WE, Gross M: From syndrome to illness: delineating the pathophysiology of schizophrenia with PET. Schizophr Bull 14:2:169-176, 1988 Bajic M, Medved V, Basic M, et al: Cerebral perfusion inhomogeneities in schizophrenia demonstrated with single photon emission tomography and Tc99m-HMPAO. Acta Psychiatr Stand 80:427-433, 1989 Musalek M, Podreka I, Walter H, et al: Regional brain function in hallucinations: a study of rCBF with 99mTc-HMPAO-SPECT in patients with auditory hallucinations, tactile hallucinations, and normal controls. Compr Psychiatry 30:1:99-108, 1989
in Psychiatry
50. Abdel-Dayem HM, El-Hilu S, Sehweil A, et al: Cerebral perfusion changes in schizophrenic patients using Tc-99m hexamethylpropylene amineoxime (HMPAO). Clin Nucl Med 15:7:468-472, 1990 51. Smith FW, Besson JAO, Gemmell HG, Sharp PF: The use of technetium-99m-HM-PA0 in the assessment of patients with dementia and other neuropsychiatric conditions. J Cereb Blood Flow Metab 8:Sl16S122, 1988 52. Schroeder J, Sauer K, Wilhelm KR, et al: Regional cerebral blood flow in endogeneous psychoses: A Te-99m HMPAO-SPECT Pilot Study. Psychiatry Res 29:331-333, 1989. 53. Risberg J: Regional cerebral blood flow measurements by 133-Xenon inhalation: methodology and applications in neuropsychology and psychiatry. Brain Lang 9:934-936, 1980 54. Uytdenhoef P, Portlange P, Jacguy J, et al: Regional cerebral blood flow and lateralized and hemispheric dysfunction in depression. Br J Psychiatry 143:128132, 1983 55. Mathew RJ, Meyer JS, Francis DJ, et al: Cerebral blood flow in depression. Am J Psychiatry 137:14491450, 1980 56. Gur RE, Skolnick BE, Gur RC, et al: Brain function in psychiatric disorders II: Regional cerebral blood flow in medicated unipolar deprc>ssives. Arch Gen Psychiatry 41:695-699, 1984 57. Kuhl DE, Metter EJ, Riege WH, et al: Local cerebral glucose utilization in elderly patients with depression, multi-infarct dementia and Alzheimer’s disease. J Cereb Blood Flow Metab 35:494-495, 1983 58. Buchsbaum MS, Detisi LE, Holcvmb HH, et al: Anteroposterior gradients in cerebral glucose use in schizophrenia and affective disorders. Arch Cen Psychiatry 41:1159-1166, 1984 59. Baxter LR, Phelps ME, Maziotta JC, et al: Cert>bral metabolic rates for glucose in mood disorders. Arch Gen Psychiatry 42~441-447, 1985 60. Schwartz JM, Baxter LR, Mazziotta JC, et al: The differential diagnosis of depression. JAMA 258: 10:1368-1373, 1987 61. Baxter LR, Schwartz JM, Phelps ME, et al: Reduction of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen Psvchiatrv 46: 243-250, 1989 62. Rush AJ, Schlesser MA, Stokely EM, et al: Cerebral blood flow in depression and mania. Psycophamacol Bull 18:6-8, 1982 63. Devous MD, Rush AJ, Schlesser MA, et al: Single photon tomographic determination of regional cerebral blood flow in psychiatric disorders. J Nucl Med 25:P57, 1984 64. Sackeim HA, Prohovnik I, Moeller JR, et al: Regional cerebral blood flow in mood disorders. Arch Gen Psychiatry 47:60-70, 1990 65. Broderson P, Paulson OB, Bolwig TG, et al: Cerebral hyperemia in induced epileptic seizures. Arch Neurol 9:334-338, 1973 66. Prohovnik I, Hakansson K, Risberg J: Observations of the functional significance of regional cerebral blood flow in resting normal patients. Neuropsychologia 18:203-217, 1980
305
0. Bonne et al.
67. Rosenberg R, Vorstrup S, Andersen A, Bolwig MD: Effect of ECT on cerebral blood flow in melancholia assessed with SPECT. Convulsive Ther 4:62-73, 1988 68. Bajc M, Medved V, Basic M, et al: Acute effect of electroconvulsive therapy on brain perfusion assessed by Tc-99m-hexamethylpropyleneamineoxim and single photon emission computed tomography. Acta Psyciatr Stand 80:421-426, 1989 69. Ackerman RF, Engel J, Baxter L: Positron emission tomography and autoradiographic studies of glucose utilization following electroconvulsive seizures in humans and rats. Ann NY Acad Sci 462~263-269, 1986 70. Silfverskiold I’, Gustafson L, Risberg J, Rosen I: Acute and late effects of electroconvusive therapy. Ann NY Acad Sci 462~236-248, 1986 71. Schwartz JA, Speed NM, Mountz JM, et al: 99mTc-Hexamethylpropyleneamineoxim single photon emission CT in poststroke depression. Am J Psychiatry 1471242-244, 1990 72. Enzmann DR, Lane B: Cranial computed tomography findings in anorexia nervosa. J Comput Assist Tomogr 1:410-414, 1977 73. Kohlmeyer K, Lehmkul G, Poutska F: Computer tomography of anorexia nervosa. AJNR 4:437-438, 1983 74. Herholz K, Krieg JC, Emrich HM, et al: Regional cerebral glucose metabolism in anorexia nervosa measured by positron emission tomography. Biol Psychiatry 22:43-51, 1987 75 Krieg JC, Lauer C, Leinsinger G, et al: Brain morphology and regional cerebral blood flow in anorexia nervosa. Biol Psychiatry 25:1041-1048, 1989 76. Woods SW, Koster K, Krystal JK, et al: Yohimbine
306
77.
78. 79.
80.
81.
82.
83.
84.
85.
alters regional cerebral blood in panic disorder. Lancet 21678, 1988 Tikofsky RS, Goldstein MD, Hellman RS, et al: Quantitiation of CBF (SPECT/IMP) in treated and untreated panic-disordered patients: comparison with normal controls (abstract) J Nucl Med 31:880 1990 Mathew RJ, Wilson WH: Anxiety and cerebral blood flow. Am J Psychiatry 147:838-849, 1990 Baxter LR, Phelps ME, Mazziotta JC, et al: Local cerebral glucose metabolic rates in obsessivecompulsive disorder: a comparison with rates in unipolar depression and in normal controls. Arch Gen Psychiatry 44:211-218, 1987 Baxter LR, Schwartz JM, Guze BH: Pet imaging in obsessive compulsive disorder with and without depression (suppl) J Clin Psychiatry 51:61-69, 1990 Nordahl TE, Benkelfat C, Semple WE, et al: Cerebral glucose metabolic rates in obsessive-compulsive disorder. Neuropsychopharmacology 2:23-28, 1989 Swedo SE, Schapiro MB, Grady CL, et al: Cerebral glucose metabolism in childhood-onset obsessivecompulsive disorder. Arch Gen Psychiatry 46:518523, 1989 Hemmingsen R, Vorstrup S, Clemmesen L, et al: Cerebral blood flow during delirium tremens and related clinical states studied with Xenon-133 inhalation tomography. Am J Psychiatry 145:1384-1390, 1988 Marx I’, Neundorfer B, Potz G, et al: Cerebral blood flow and aminoacid metabolism. In Harper AM (ed), Blood Flow and Metabolism in the Brain. Edinburgh, Churchill Livingstone, 1975 Berglund M, Risberg J: Regional cerebral blood flow during alcohol withdrawl. Arch Gen Psychiatry 38: 351-355, 1981
Abstract: In the last two decades, brain imaging has become an integral part of clinical and research psychiatry. Single photon computed emission tomography (SPECT) is rapidly gaining acceptance as one of the major imaging techniques available, along with computed tomography (CT), magnetic resonance imaging (MRlJ, and positron emission tomography (PET). Each of these techniques has its assets and drawbacks. This review concerns SPECT, a highly prevalent imaging technique whose potential value in brain imaging has not been appreciated until recently. Its purpose is to expose practicing clinicians and research psychiatrists alike to the attributes of this instrument, which is available in most nuclear medicine departments today. An effort is made to provide a comprehensive account of this technique, including a brief summary of the basic principles, the various methods of its application, and recent findings in most psychiatric disorders. Analogies to its “aristocratic cousin,” PET, are presented to emphasize similarities and differences. Finally, directions for future development and implementation of SPECT are suggested.
Introduction The role of biological factors in the etiology and pathogenesis of major psychiatric disorders has come under increasing scrutiny over the past three decades [l]. Nevertheless, extensive research efforts, based on a wide range of experimental strategies, have not been able to demonstrate definitive and replicable abnormalities of brain structure or function in any of the disorders studied. In-vivo imaging of the human brain began to play a major role in this field beginning in the early
From the Departments of Psychiatry (O.M., B.L.) and Nuclear Medicine and Medical Biophysics (Y.K.), Hadassah University Hospital, Jerusalem, Israel. Address reprint requests to: Dr. Omer Bonne, Department of Psychiatry, Hadassah University Hospital, P.O. Box 12000, Jerusalem 91120, Israel.
296 ISSN 0163-8343/92/$5.00
1970s with the development of computed tomography (CT), and the ability to measure regional cerebral blood flow (rCBF) by the use of Xe-133. These modalities exemplify the two main branches of imaging technology today: “structural” imaging, i.e., CT scan and magnetic resonance imaging (MRI); and “functional” imaging, i.e., positron emission tomography (PET) and single photon emission computed tomography (SPECT). As the term implies, structural imaging provides information of an anatomical nature whereas functional techniques are designed to reflect the level of brain metabolic activity, either directly or by extrapolation from regional cerebral blood flow measurements. Each of these technologies has made a significant contribution to the expanding body of data available (21. The present review concerns SPECT, the newcomer to the field of brain imaging. Its purpose is to provide practicing clinicians with basic insights into the theoretical concepts of the technique, its practical application, possible diagnostic utility, and potential research contribution. As disorders at the interface of psychiatry and neurology are presently the most amenable to study with SPECT, these will be emphasized, but other potential applications will also be considered.
Background Efforts to directly evaluate brain function began with invasive procedures by Kety and Schmidt in 1948 [3] which involved intracarotid injection of nitrous oxide and internal jugular venous sampling, and provided gross estimates of whole brain perfusion. The advent of radioisotopes and collimated detectors allowed for regional measurement Gr,~rral Hospital Psychiatry 14, 296-306, 1992 0 1992 Elsevier Science Publishing Co., Inc. 655 Avenue of the Americas, New York, NY 10010
SPECT Imaging
of cerebral blood flow. Further advances in detector sensitivity and curve-fitting algorithms enabled replacement of intracarotid injections with either inhalation or intravenous injection of the radiotracer, most notably Xe-133, which for several years was the uniform radiotracer and is still frequently used today. These measurements all utilized twodimensional methods for assessing brain function. The landmark finding of decreased rCBF in the frontal lobes of schizophrenic patients by Ingvar and Franzen in 1974 [4] delineated the important potential role of functional imaging in the study of psychiatric conditions. Tomographic imaging of rCBF was first successfully done by Yamamoto et al. in 1977 [5] employing Kr-77, and soon afterward Xe-133 was also utilized. Notwithstanding these promising results, it was only several years later, and especially after the development of PET, that research in this direction was intensified. The potential value of PET imaging in psychiatry is beyond doubt [6,7]. However, its extremely high purchase and operating costs significantly limit its availability. Increasing efforts are therefore being made to establish the capability of SPECT, a much more common and less costly instrument, to provide reliable and comparable data.
Basic Principles SPECT is a three-dimensional method capable of reflecting the distribution of a radiotracer within the human body. It involves administration of a radiotracer and either simultaneous or delayed detection by a tomographic instrument. It is frequently used for purposes other than brain imaging, mostly cardiac assessment, not to be described in this review. Radiotracers emit gamma rays uniformly in all directions. In order to localize radiation coming from a specific brain area, a collimator, which is a lead sheet containing long apertures allowing only parallel rays to penetrate and blocking all others, is placed in front of the detector. As all parallel rays arriving from any brain depth along a line will pass through the collimator, this process will not suffice to fix the exact point from which a ray is emitted. This is the conventional planar nuclear medicine image. To expand two-dimensional information into a threedimensional picture, additional two-dimensional projections should be gathered from different angles. This is usually accomplished by rotating a gamma camera (complex of collimator and radiation detector) around the head of the examinee,
in Psychiatry
although other options are also possible. These reconstructed projections are displayed as crosssections through the brain. (A more detailed technical description of SPECT can be found in [8].)
Radiopharmaceuticals
(Radiotracers)
Radiopharmaceuticals used for SPECT imaging can be classified into three categories: diffusible, static, and neuroreceptor ligands. Diffusible tracers, as their name implies, do not remain attached to any specific brain area. They freely traverse the blood-brain barrier and do not chemically interact with brain parenchyma. This mode of study is designated dynamic or D-SPECT. The routinely available tracer for this method is Xe133. Information is drawn from detection of the transit of the radiotracer within the brain by the application of a mathematical model. The low gamma-ray energy and diffusibility of Xe-133 significantly limit the resolution of these studies. Lipophilic radiotracer complexes readily pass the blood-brain barrier. Once located within the brain they lose their lipophilic capacity and undergo an as yet unclear chemical reaction which traps them within the brain in a defined area. The first static radiotracers were used in the early 1980s ([9,10]; these were I-123 isotopes bound to an amine structure. Higher energy and fixed distribution permits the use of an ordinary gamma camera with improved resolution compared with D-SPECT. Presently available I-123 labeled radiotracers are I-123-IMP (iodoisopropylamphetamine) and 1-123HIDPM (hydroxyisopropyldiamine). Tc-99m has been the mainstay of nuclear medicine for many years. It has better dosimetry and physical properties than does 1-123. However, technetium chemistry is rather complex and it has only been in the past few years that the labeling of Tc-99m to suitable carriers has been achieved [ll]. Because of their better dosimetry characteristics, high availability, and low cost, Tc-99m tracers have become the routine radiotracers used today [12]. The most prevalent Tc-bound radiotracer and the only one that has been approved by the FDA for clinical application is Tc-99m-HMPAO (hexamethylpropyleneamineoxime). This agent is highly lipophilic but unstable. Once settled, HMPAO remains fixed, reflecting distribution at time of administration. Resolution of SPECT is also improving with the use of Tc-99m-bound tracers. For D-SPECT with Xe-133 the resolution of SPECT is about 17 mm, but new detectors in conjunction
297
0. Bonne et al.
with Tc-99m tracers are capable of resolution of 8 mm. These values are approaching PET values. The third class of radiotracers are neuroreceptor ligands. This group of radiotracers, more prevalent in PET, enable access to receptor sites within the brain and allow a certain degree of receptor imaging. The development of this type of radiotracer for SPECT is in its first stages. A few reports have appeared describing successful labeling of muscarinic cholinergic [13] and dopaminergic Dl [14] and D2 [15] receptors. A single report presented the possibility of producing SPECT benzodiazepine receptor binding [16]. All SPECT receptor binding agents are at this time I-123 labeled. This whole class of radiotransmitter ligand radiotracers is at present experimental.
Image Interpretation Reliable analysis of SPECT data remains a major challenge. Various factors have to be taken into consideration. A correlation between radiotracer distribution within the brain and cerebral blood flow has to be established. Studies do point to such a correlation [17], but opinion varies as to the mode of relationship. Comparison of SPECT studies to those performed with PET tracers disclosed a good correlation, both in animal [18] and human [19] subjects, in normal or pathological conditions. In fact, rCBF has even been reported to reflect neuronal activity more accurately than oxygen consumption in visual cortex stimulation [20]. SPECT image data thus reflect brain function. In order for this data to be useful, the anatomic location of the image must be as precisely defined as possible. Patients must be positioned carefully. Image slices are taken according to well-defined external skull anatomical landmarks, most commonly the orbitocanthal meatal line (OML). Additional parallel slices at specified distances from this imaginary line are taken, as needed. A brain atlas, CT scan, MRI, or simple skull x-ray films are used to define brain regions at each slice level. This method is only an approximation, as individual brain position relative to external landmarks may vary somewhat. When aimed at specific brain areas, SPECT (and PET) will define certain anatomic locations as regions of interest (ROI). Although the possibility of acquiring absolute numerical rCBF values exists, (with Xe-133), this method involves arterial sampling and is rarely used today. Most studies measure tracer uptake from the ROI, define the area,
298
and so receive mean uptake for each ROI. This value is related to a reference area, most often either mean whole slice uptake or a specified brain region (i.e., occipital lobe, cerebellum, and so forth). The ratio received from the division of the ROI value by the reference value is usually presented as “rCBF.” This value can give an indication of either hyper or hypoperfusion in a certain area and can be approximately compared between different studies. A comparison is also sometimes made between symmetrical ROIs from the two hemispheres (see [8].) Visual analysis, when carried out by multiple experienced observers, can supply useful information. It also compares ROIs to reference areas or to the contralateral hemisphere. Visual interpretation may sometimes be more sensitive than the quantitative computerized method described above [21]. An important aspect of functional in contrast to structural imaging is the condition of the patient at the time of examination. Structure of the brain is expected to remain constant regardless of brain activity whereas brain function, by definition, is expected to change. Grossly, functional studies are undertaken either in the “resting” or “activated” states. Resting condition ideally implies absence of all somatic, sensory, cognitive and affective stimuli. However, there is no standardized resting state, and issues such as eyes open vs eyes closed or ears open vs ears plugged during examination vary between studies. Moreover, the degree of anxiety experienced or the presence of delusions in psychotic patients may cause wide variations in this resting state. These elements (and many more not mentioned for reasons of brevity) complicate reproducibility of SPECT studies and may account for apparent conflicting results.
Neuropsychological and Activation Studies Introduction of neuropsychological testing in conjunction with SPECT studies is becoming common procedure. Correspondence between neuropsychological deficits and SPECT pattern is evaluated. Correlation between findings serves as indirect proof of the capability of SPECT to generate meaningful data reflecting damaged brain areas. In schizophrenics, for example, Paulman et al. [21] used the Wisconsin Card Sorting Test (WCST) and the Luria-Nebraska battery to describe association of left frontal hypoperfusion with impairment, and Weinberger et al. [22] found failure in activation of
SPECT Imaging
in Psychiatry
the dorsolateral prefrontal cortex when performing the WCST. Activation studies involve either pharmacological intervention or administration of a predetermined task or stimulus. While activated, the patient is surveyed by a functional imaging technique, and deviation from the resting state is assessed. Pharmacological agents utilized include acetazolamide, respiration of 5% CO,, and psychotropic active compounds. The tasks or stimuli may also be sensory, somatic, or cognitive. An extensive account of the technique, application, and findings has been recently presented by Tikofsky and Hellman [23]. Activation studies and neuropsychological testing in combination with functional imaging are invaluable both in providing insights into normal brain performance and detection of brain pathology which may not be evident in the resting state.
SPECT Findings in Psychiatric Disorders Measurements of rCBF have been made for about 40 years IS], but it is due to the development of PET that functional imaging has begun to assume a leading role in psychiatric research. SPECT is not expected to supersede PET’s diagnostic capabilities or research potential. However, it is widely available and relatively inexpensive, allowing access to almost any suitable psychiatric case. As mentioned, most SPECT studies carried out today employ HMPAO, approved by the FDA and introduced into clinical practice only in 1988. Most SPECT studies published in recent years, although this is not always explicitly stated, have tried to replicate findings obtained by PET and the older Xe-133 rCBF two-dimensional studies. SPECT findings in psychiatric disorders will be reviewed. Not every SPECT study concerning psychiatric disorders will be discussed, as some sporadic reports do not significantly contribute to an understanding of the instrument or the disorder under study. Relevant data from PET, structural imaging, and electroencephalographic studies will be noted when appropriate.
Dementia Dementia is the most common neuropsychiatric disorder in middle-aged and elderly people. Its prevalence ranges between 10 1241 and 15% [8] in populations over 65 years of age, although it may
occur in younger age groups. Over 50% [8,16] of patients with dementia are believed to suffer from Alzheimer’s disease (AD), a diagnosis usually made by exclusion, which can only be definitively confirmed at autopsy. Another 20-30% suffer from multi-infarct dementia (MID), and in a substantial number of patients (around 20%) both disorders may be present. Ten percent of patients presenting with dementias are believed to be actually suffering from depression, a condition sometimes termed “pseudodementia” which is reversible upon adequate antidepressant treatment. Other and rarer causes of dementia-metabolic and endocrine abnormalities, those associated with movement disorders (Huntington’s and Pick’s), Korsakoff’s psychosis and others-account for the remaining part. SPECT studies of AD have yielded unequivocal findings. All studies report symmetrically reduced uptake in the parietal lobes [24-291, as demonstrated in Figure 1. Temporal lobe involvement is also commonly encountered [24-27,291 Frontal and occipital lesions are sometimes observed in severe, advanced cases [25,28]. These findings are consistent with those reported for PET [28,30-321. Promising results have been achieved in several studies [23,26,29] correlating neuropsychological deficits and cognitive impairments with areas of localized brain damage. These studies lend support to the
299
0. Bonne et al.
opinion that the cognitive deterioration noted in AD can be attributed to specified brain areas, and does not reflect diffuse brain damage. In MID, in contrast to AD, the SPECT image may show asymmetrical patchy areas of reduced uptake , which often correspond to the distribution of major cerebral arteries [24,25,27,28]. In Huntington’s disease [27], reduced uptake was observed in the caudate nuclei, whereas impairment in frontal lobe function was noted in patients suffering from Korsakoff’s psychosis that was correlated to cognitive deficits as measured by neuropsychological testing [29]. Severe compromise in frontal lobe perfusion was prominent in Pick’s dementia and progressive supranuclear palsy (PSP) [25].
Epilepsy Symptomatology in epilepsy, particularly temporal lobe (partial-complex, psychomotor) epilepsy (TLE) can mimic the whole range of psychiatric phenomena [33]. Disturbances in thought form and content, perception and level of consciousness are commonly observed. Cognitive ability and memory may be compromised. Behavior may be confused or automatic. These phenomena may occur in the ictal, postictal, or interictal phases of the disease. Patients with no known history of convulsive disorder are therefore frequently suspected to be suffering from an epileptic disorder, especially if the observed syndrome is atypical or resistant to standard medication. Success in controlling psychiatric symptoms has been achieved by the administration of the anticonvulsant carbamazepine to patients with diagnosed temporal lobe epilepsy [34,35]. Furthermore, patients without diagnosed TLE may respond to carbamazepine particularly when affective symptoms or behavioral dyscontrol are prominent [34]. Whether these patients have an epileptic focus that cannot be demonstrated by conventional means is unclear, as TLE frequently eludes electroencephalographic (EEG) confirmation, even when nasopharyngeal leads are used as well as assessments by CT and MRI. SPECT findings in TLE have been most consistent (Fig. 2) and are comparable to those achieved by PET [28,36]. In fact, ictal SPECT studies may be simpler to carry out due to the stability and longer half-life of static SPECT tracers. Epileptogenic foci are detected in the ictal and interictal phases by hyper and hypoperfusion, respectively [37,38]. 300
Figure 2. HMPAO brain SPECT of temporal lobe epilepsy. Transaxial slice disclosirrg perfusion deficit is in the left temporal area (arrow).
Moreover, low flow areas, remote from the seizure focus and occasionally in the contralateral hemisphere, have been identified by SPECT[37,39]. These areas correspond to patterns of cognitive impairment as located by neuropsychological testing [22,38,40]. SPECT may also prove helpful in monitoring brain response to anticonvulsant therapy
[411.
Schizophrenia Schizophrenia, the most extensively studied of all psychiatric disorders, has been studied by numerous functional imaging techniques which have set out to reproduce the finding of reduced cerebral blood flow in frontal lobes of schizophrenic patients described by Ingvar and Franzen in 1974 [4]. The large number of studies (primarily PET) has served to emphasize the complexity of this disorder. It is beyond the scope of this review to go into a detailed description of all studies in this field so a general overview will be presented. In a recent review, Buchsbaum [42] summarized findings from PET studies according to three brain areas: frontal lobes, basal ganglia, and (mainly left) temporal lobes. Frontal lobe function is usually presented as a ratio between frontal and either occipital or whole brain metabolism or blood flow. The value thus attained is compared to a parallel assessment of normal brain function. In Buchsbaum’s review, most studies showed lower ratios in schizo-
SPECT Imaging
phrenic patients, some of them significantly. However, a minority of studies found no difference between schizophrenics and normal controls, and a few even showed contradictory results. It should be noted that studies differ in methodology as to sensory or cognitive stimulation, medication status, exact ROIs, and other factors that might influence findings. Results from the other two brain areas examined generally follow the same trend and suffer from the same limitations. The effect of neuroleptic treatment may be more crucial to basal ganglia studies. In schizophrenic patients scanned both on and off medication, a greater increase in metabolic rate was noted in basal ganglia compared with cortical regions in the medicated state [43]. Are there three types of schizophrenia, each conforming to a distinct anatomical area? Are these lesions found to a certain extent in all schizophrenic patients? Does the basic flaw reside in faulty connections between brain regions, as others assume [44]? In the resolution of these and other fundamental issues, functional imaging should play a major part. Few SI’ECT studies are available on schizophrenia. The most methodologically sound report [21] published to date is a D-SPECT Xe-133 inhalation
in Psychiatry
study on a group of 40 medicated and nonmedicated, paranoid and nonparanoid schizophrenic patients, combined with neuropsychological testing. Absolute bilateral hemispheric CBF was elevated in schizophrenic patients compared with a normal control group (a finding not consistent with PET findings [45-471 which show no significant difference between patient and control populations). Higher CBF values were observed in the right hemiphere in both patient and control groups. Reduced relative left frontal rCBF was associated with neuropsychological impairment, and increased whole brain CBF with relative bilateral hypofrontal and hypotemporal CBF was correlated with paranoid symptoms. Another report [48] presented a group of 28 female schizophrenic patients, divided into subgroups of “positive” and “negative” symptom, “acute” and “chronic” patients. HMPAO uptake was normal in the positive-acute group (mostly paranoid) and reduced in temporal regions and basal ganglia in the negative-acute and chronic group. Type of hallucination was reported to determine rCBF distribution in a third HMPAO SPECT study [49]. Increased rCBF was found in hippocampal regions of patients suffering from auditory hallucinations, whereas patients with tactile hallucinations manifested reduction in rCBF in inferior temporal regions. Less methodologically sound studies reported multifocal sites of increased and decreased perfusion in all but one of 20 patients [50] or completely negative findings in a group of seven patients in another HMPAO study [51]. A pilot HMPAO SPECT study [52] examined 18 patients diagnosed as either schizophrenic, schizoaffective, or depressed. Hypofrontality was significantly more pronounced in schizophrenic as compared with depressed patients, with schizoaffective patients occupying an intermediate position. All three groups, however, showed some evidence of hypofrontality. Overall, SPECT studies in schizophrenia have not yielded a clear pattern. However, most of the work reported thus far has been of a preliminary nature, and methodological issues noted earlier in this review remain to be effectively and consistently addressed.
Affective Disorders Even though depression is a highly prevalent disorder, imaging studies on this disorder have been relatively neglected. Two-dimensional blood flow
301
0. Bonne et al.
studies reported either normal flow patterns [53] in depressive patients or displayed contradictory results showing increased [54] or decreased [55] flow in left hemispheric structures and opposite findings in the other hemisphere. Spatial and verbal activation enabled discrimination between depressive and control subjects who were nondistinguishable before activation in another rCBF study [56]. PET studies have not proved more conclusive. One study reported normal whole-brain glucose metabolism with a specific reduction in the inferior left frontal cortex [57]. Several studies made a distinction between subtypes of affective disorder. Patients were grouped into unipolar and bipolar depression, bipolar mixed, and bipolar manic. Buchsbaum et al., [58] reported a decrease in anterior-posterior ratio in bipolar patients while receiving painful electrical shocks to the forearm. Baxter et al. [59] described lower whole-brain supratentorial glucose metabolic rates in bipolar depressed and bipolar mixed patients than in unipolar depressed patients. Metabolic rate significantly increased after clinical improvement. The same group reported (on a small number of patients) in a later study [60] a lower ratio of glucose metabolic rates in the caudate nucleus divided by metabolic rate in the ipsilateral supratentorial hemisphere in unipolar-depressed but not in bipolar-depressed patients; these findings pertain only to the left side. The findings from the last three studies suggest a supratentorial-prefrontal pathology in bipolar depressed patients and caudate nucleus pathology in unipolars. Baxter et al., [61] however, in a later PET study, noted a decrease in prefrontal to ipsilateral hemisphere glucose metabolic ratio in all categories of depressive patients studied. A significant increase in this ratio was noted as patients responded to medication. Another SPECT study [62] disclosed significantly lower CBF in unipolar depressive patients compared with bipolar depressive patients or control subjects. CBF was higher than normal in bipolar manic and mixed patients in the same study. A subsequent study by the same group 1631 localized diminished flow to the right parietal and temporal lobes in unipolar depressive subjects and contralateral hyperperfusion in the left temporal and parieta1 lobes in bipolar manic or mixed patients. In a recent controlled study [64] on a group of 41 dea reduction in global cortical pressed patients, blood flow was noted, which was attributed mostly to reductions in selective frontal, central, superior
302
temporal, and anterior parietal regions. Unipolar and bipolar depressed patients were not separated except for the exclusion of rapid cycling bipolar patients. Measurement of CBF in conjunction with electroconvulsive therapy (ECT) has intrigued researchers and there are several reports on this entity. CBF was measured by the two-dimensional intracarotid Xe-133 method [65], two-dimensional Xe-133 inhalation method [66], Xe-133 D-SPECT [67], HMPAO SPECT [68], and PET 1691. Pretreatment CBF values were most often within normal limits. Posttreatment findings include decrease in CBF [67], increase in CBF [70], or redistribution of tracer uptake [68]. No distinction between subtypes of depression is mentioned in all the above studies. Finally, a recent HMPAO SPECT study [71] found a high correlation between severity of depressive symptoms and volume of ischemic lesion in patients suffering from poststroke depression.
Other Neuropsychiatric Disorders Only sparse SPECT assessments of other neuropsychiatric disorders are found. These provide only preliminary insights into the cerebral blood flow of these conditions and are briefly alluded to for the sake of completeness. Structural imaging of severe eating disorders (emaciated anorexia as well as normal-weight bulimia) has shown morphological brain alteration in these disorders. Findings were characterized by enlargement of cerebrospinal fluid (CSF) spaces and/ or ventricular dilatation [72,73]. A deficit in neuropsychological test performance was noted in conjunction with morphological abnormalities [73]. A PET study [74] and a D-SPECT study [75] found no evidence for a reduced functional capacity of the brain in the anorectic state, reflected either in glucose metabolism 1741 or in cerebral blood flow ]751. Few attempts to evaluate the classical “neuroses” have been reported. Woods et al. [76], in an activation HMPAO SPECT study, observed rCBF differences following yohimbine administration compared with placebo injection in panic disorder (I’D) patients, not evidenced in normal controls. An IMP SPECT study [77] compared PD patients before and after pharmacological treatment, and found nonsignificant rCBF changes. Reduced uptake in the right anterior frontal area was found, however, in panic disorder patients but not in nor-
SPECT Imaging
ma1 controls. A comprehensive review of anxiety and CBF is presented by Mathew and Wilson [78], emphasizing physiological considerations of alterations in CBF. In obsessive-compulsive disorder (OCD), PET findings [79] have included increased glucose metabolic rates in the left orbital gyrus and bilateral caudate nuclei. These findings were distinguishable from both normal controls and unipolar depressed patients. A follow-up study by the same group [80] found bilateral increased glucose metabolic rates in orbital gyri. The inconsistency from the first report is explained by methodological faults in the earlier study. Two National Institute of Mental Health (NIMH) groups have reported findings in OCD. The first [Bl] reported increased metabolic rates in both orbital gyri. The second group [82] found increased absolute glucose metabolism in the left orbital and right sensorimotor regions and bilaterally in the anterior cingulate gyri and lateral prefrontal area. Moreover, patients who responded to clomipramine therapy had significantly lower right anterior cingulate and right orbital metabolism than did nonresponders. CBF has been measured during the acute stage and after recovery from alcohol withdrawal reaction. Hemmingsen et al. [83] noted in a D-SPECT study a significant decrease in rCBF in alcoholwithdrawn patients with visual hallucinations and psychomotor agitation. Decreases were not significant in withdrawal patients not manifesting these disturbances. Other studies, however, showed prerecovery low CBF values [84,85] that normalized after recovery.
Conclusions SPECT imaging in psychiatry has not yet entered routine clinical use, although the instrumentation needed for its application is available in most nuclear medicine departments. Psychiatric diagnosis relies at present almost solely on clinical expertise. SPECT image patterns may prove to be of value in differentiating neuropsychiatric syndromes with similar clinical presentation, as well as in clarifying the etiology of atypical, seemingly obscure behavioral disorders. Implementation of SPECT in the workup of patients with cognitive deterioration should become common practice. Differential diagnosis between dementia and depression (presenting as pseudodementia) can be substantially facilitated. Discrimination between the various types of dementia can
in Psychiatry
also be assisted, enabling determination of prognosis and the planning of treatment. Patients manifesting disturbances in level of consciousness or conduct and atypical psychiatric symptomatology may benefit from SPECT study, particularly if CT, MRI, and EEG are negative and a diagnosis of temporal lobe epilepsy is suspected. For the major psychoses (schizophrenia and affective disorders), SPECT does not yet have a clear clinical application. This is even more true for the neurotic disorders. SPECT does offer, however, considerable research potential in these disorders. Systematic study of large groups of subjects may yet reveal consistent pathological findings. Some of the most intriguing issues in clinical psychiatry today could benefit from the introduction of SPECT into research protocols: Is schizophrenia a single disease or are there several autonomous entities which we wrongly designate as schizophrenia? Is schizoaffective disorder schizophrenia or a separate disorder, or is it a subtype of schizophrenia, or of the effective disorders? Are major depression and psychotic depression separate entities? Are there pathological differences between unipolar and bipolar depressed patients? Is it possible to demonstrate brain pathology in the neuroses? These are just a few examples of the clinical research potential of SPECT. It may also contribute to other areas of research. For instance, it could be effectively utilized in the study of normal brain function, such as the ongoing effort to establish a relationship between cognitive faculties and regional brain activity. Activation studies and neuropsychological testing in conjunction with SPECT are thus on the way to becoming common research practice. Further development of receptor binding radiotracers and their research application could result in the elucidation of biochemical pathways in the normal human brain and in psychopathological states. This capacity of SPECT could also enable objective drug response monitoring in psychiatric patients. SPECT is a newcomer to the field of psychiatry; it has yet to mature. Validation and sensitivity/ specificity studies are urgently needed if it is to manifest its full potential. The relatively wide availability of SPECT permits the inclusion in research protocols of a sufficient number of subjects to reliably substantiate research findings. Particular attention should be given to standardizing methodology and to achieving consistency with regard
303
0. Bonne et al.
to
the state in which subjects are evaluated. Once these objectives are achieved, SPECT could prove to be a valuable clinical and research tool in modern psychiatry.
17.
18.
References 1. Meltzer HY (ed): Psychopharmacology: The Third Generation of Progress. New York, Raven Press, 1987 2. Andreasen NC: Brain imaging: applications in psychiatry. Science 239:1381-1388, 1988 3. Kety SS, Schmidt CF: The nitrous oxide method for quantitative determination of cerebral blood flow in man: theory, procedure and normal values. J Clin Invest 27~475-483, 1948 4. Ingvar DH, Franzen G: Abnormalities of cerebral blood flow distribution in patients with chronic schizophrenia. Acta Psychiatr Stand 50:425-464, 1974 5. Yamamoto YL, Thompson CJ, Meyer E, et al: Dynamic positron emission tomography for a study of cerebral hemodynamics in a cross section of the head using positron-emitting 68-Ga-EDTA and 77-Kr. J Comput Assit Tomogr 1:43-56, 1977 6. Cohen RM, Semple WE, Gross M: Positron emission tomography. Psychiatr Clin North Am 9:1:63-79, 1986 7. Jacobson HG: Positron emission tomography-a new approach to brain chemistry. JAMA 260:27042710, 1988 8. Devous MD: Imaging brain function by single photon computed emission tomography. In Andreasen NC (ed), Brain Imaging: Applications in Psychiatry. Washington DC, American Psychiatric Press, 1989 9. Winchell HS, Baldwin RM, Lin TH: Development of I-123 labeled amines for brain studies: localization of I-123 iodophenylalkylamines in rat brain. J Nucl Med 21:940-947, 1980 10. Kung HF, Tramposch KM, Blau M: A new brain perfusion agent: I-123-HIDPM. J Nucl Med 24:66-72, 1983 11. Kung HF: New Technetium 99m-labeled brain perfusion imaging agents. Semin Nucl Med 2:150-158, 1990 12. Lassen NA, Blasberg RG: Technetium-99m-d, lHM-PAO, the development of a new class of Tc-99m labeled tracers: an overview. J Cereb Blood Flow Metab S:Sl-S3, 1988 13. Thonoor CM, Couch MW, Geer DM, et al: Biodistribution and radiation dosimetry of radioiodinated SCH23982, a potential D-l receptor imaging agent. J Nucl Med 29:1668-1674, 1988 14. Kung HK, Pan S, Kung MI’, et al: In vitro and in vivo evaluation of [1123]IBZM: a potential CNS D-2 dopamine receptor imaging agent. J Nucl Med 30:8892, 1989 15. Eckelman WC, Reba RC, Rzesotarsky WJ, et al: External imaging of cerebral muscarinic achetylcholine receptors. Science 223:291-293, 1984 16. Saji H, Nakatsuka I, IIda Y, et al: Radioiodinated
304
19.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
30.
31.
32.
33.
diazepam derivative for SPECT studies of benzodiazepine receptor. J Nucl Med 30:803-804, 1989 Neirinckx RD, Canning LR, Piper IM, et al: Technetium 99m d, I-HMPAO: a new radiopharmaceutical for SPECT imaging of regional cerebral blood perfusion. J Nucl Med 28:191-202, 1987 Sokoloff L: Localization of functional activity in the central nervous system by measurement of glucose utilization with radioactive deoxyglucose. J Cereb Blood Flow Metab 1:7-36, 1981 Lebrun-Gradie P, Baron JC, Soussaline F, et al: Coupling between regional blood and oxygen utilization in the normal brain. Arch Neurol 40:230-236, 1983 Fox PT, Raichle ME, Mintun MA, Dence C: Nonoxidative glucose consumption during focal physiologic neural activity. Science 241:462-464, 1988 Paulman RG, Devous MD, Gregory RR, et al: Hypofrontality and cognitive impairment in schizophrenia: dynamic single photon tomography and neuropsychoiogical assessment of schizophrenic brain function. Biol Psychiatry 22377-399, 1990 Weinberger DR, Berman KF, Zec RF: Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia. Arch Gen Psychiatry 43:114-124, 1986 Tikofsky RS, Hellman RS: Brain single photon emission tomography: newer activation and intervention studies. Semin Nucl Med 21:40-57, 1991 Kirn TF: Dementias appear to have individual profiles on single photon emission tomography. JAMA 261:965-967, 1989 Bonte FJ, Horn RT, Weiner MF: Single photon tomography in Alzheimer’s disease and the dementias. Semin Nucl Med 20342-352, 1990 Goldenberg G, Podreka I, Suess E, et al: The cerebral localization of neuropsychological impairment in Alzheimer’s disease: a SPECT study. J Neurol 236: 131-138, 1989 Smith FW, Besson JAO, Gemmell HG, Sharp PF: The use of technetium-99m-HM-PA0 in the assessment of patients with dementia and other neuropsychiatric conditions. J Cereb Blood Flow Metab 8:5116S122, 1988 Maurer AH: Nuclear medicine: SPECT comparisons to PET. Radio1 Clin North Am 26:5:1059-1074, 1988 Hunter R, McCluskie R, Wyper J, et al: The pattern of function-related regional cerebral blood flow investigated by single photon emission tomography with 99mTC - HMPAO in patients with presenile Alzheimer’s disease and Korsakoff’s psychosis. Psychol Med 19:87-855, 1989 Duara R, Grady CL, Haxby JV, et al: Positron emission tomography in Alzheimer’s disease. Neurology 36:879-887, 1986 Cohen RM, Semple, WE, Gross M: Positron emission tomography. Psychiatr Clin North Am 9:1:63-79, 1986 Jacobson HG: Positron emission tomography-a new approach to brain chemistry. JAMA 260:18: 2704-2710, 1988 Fenton GW: Psychiatric disorders of epilepsy: classification and phenomenology. In Reynolds EH,
SPECT Imaging
34.
35.
36. 37.
38.
39.
40.
41.
42.
43.
44. 45.
46.
47.
48.
49.
Trimble MR (eds), Epilepsy and Psychiatry. New York, Churchill Livingstone, 1981 Dalby MA: Behavioral effects of carbamazepine. In Penry JK, Daly D (eds), Complex Partial Seizures and Their Treatment: Advances in Neurology, Vol. II. New York, Raven Press, 1975 Kuhn R: The psychotropic effect of carbamazepine in nonepileptic adults, with particular reference to the drug’s possible mechanism of action. In Birkmayer W (ed), Epileptic Seizures-Behavior-Pain. Bern, Hans Huber, 1975 Editorial: SPECT and PET in epilepsy. Lancet 135137 January 21, 1989 Bonte FJ, Devous MD, Stokely EM, Homan RW: Single photon tomographic determination of regional cerebral blood flow in epilepsy. AJNR 4:544-546, 1983 Devous MD, Leroy RF, Homan RW: Single photon computed emission tomography in epilepsy. Semin Nucl Med 20:325-341, 1990 Lee BI, Markland ON, Wellman HN, et al: HIDPMSPECT in medically intractable complex partial seizures: ictal study. Arch Neurol 45:397-402, 1988 Homan RW, Paulman RC, Devous MD, et al: Neuropsvchologic function and regional cerebral blood flow m epilepsy. In Porter JR et al (eds), Advances in Epileptology: XVth Epilepsy International Symposium. New York, Raven, 1984 Homan RW, Devous MD, Stokely EM, Bonte FJ: Cerebellar blood flow in partial seizures (suppl 1) Neurology 37~327-334, 1987 Buchsbaum MS: The frontal lobes, basal ganglia and temporal lobes as sites for schizophrenia. Schizophr Bull 16:379-389, 1990 Buchsbaum MS, Wu JC, DeLisi LE, et al: Positron emission tomography studies of basal ganglia and somatosensorp cortex neuroleptic drug effects: differences between normal controls and schizophrenic paticnth. Biol Psychiatry 22:479-494, 1987 Robbins TW: The case f&r fronto-striatal dysfunction in schizophrenia. Schizophr Bull 16:391-402, 1990 DeLisi LE, Buchsbaum MS, Holcomb HH, et al: Clinical correlates of decreased anteroposterior metabolic gradients in Positron Emission Tomography (PET) of schizophrenic patients. Am J Psychiatry 142:7881, 1985 Sheppard G, Gruzelier J, Manchanda R, et al: 015 Positron emission tomographic scanning in predominantly never-treated schizophrenic patients. Lancet 2:1448-1452, 1983 Cohen RM, Semple WE, Gross M: From syndrome to illness: delineating the pathophysiology of schizophrenia with PET. Schizophr Bull 14:2:169-176, 1988 Bajic M, Medved V, Basic M, et al: Cerebral perfusion inhomogeneities in schizophrenia demonstrated with single photon emission tomography and Tc99m-HMPAO. Acta Psychiatr Stand 80:427-433, 1989 Musalek M, Podreka I, Walter H, et al: Regional brain function in hallucinations: a study of rCBF with 99mTc-HMPAO-SPECT in patients with auditory hallucinations, tactile hallucinations, and normal controls. Compr Psychiatry 30:1:99-108, 1989
in Psychiatry
50. Abdel-Dayem HM, El-Hilu S, Sehweil A, et al: Cerebral perfusion changes in schizophrenic patients using Tc-99m hexamethylpropylene amineoxime (HMPAO). Clin Nucl Med 15:7:468-472, 1990 51. Smith FW, Besson JAO, Gemmell HG, Sharp PF: The use of technetium-99m-HM-PA0 in the assessment of patients with dementia and other neuropsychiatric conditions. J Cereb Blood Flow Metab 8:Sl16S122, 1988 52. Schroeder J, Sauer K, Wilhelm KR, et al: Regional cerebral blood flow in endogeneous psychoses: A Te-99m HMPAO-SPECT Pilot Study. Psychiatry Res 29:331-333, 1989. 53. Risberg J: Regional cerebral blood flow measurements by 133-Xenon inhalation: methodology and applications in neuropsychology and psychiatry. Brain Lang 9:934-936, 1980 54. Uytdenhoef P, Portlange P, Jacguy J, et al: Regional cerebral blood flow and lateralized and hemispheric dysfunction in depression. Br J Psychiatry 143:128132, 1983 55. Mathew RJ, Meyer JS, Francis DJ, et al: Cerebral blood flow in depression. Am J Psychiatry 137:14491450, 1980 56. Gur RE, Skolnick BE, Gur RC, et al: Brain function in psychiatric disorders II: Regional cerebral blood flow in medicated unipolar deprc>ssives. Arch Gen Psychiatry 41:695-699, 1984 57. Kuhl DE, Metter EJ, Riege WH, et al: Local cerebral glucose utilization in elderly patients with depression, multi-infarct dementia and Alzheimer’s disease. J Cereb Blood Flow Metab 35:494-495, 1983 58. Buchsbaum MS, Detisi LE, Holcvmb HH, et al: Anteroposterior gradients in cerebral glucose use in schizophrenia and affective disorders. Arch Cen Psychiatry 41:1159-1166, 1984 59. Baxter LR, Phelps ME, Maziotta JC, et al: Cert>bral metabolic rates for glucose in mood disorders. Arch Gen Psychiatry 42~441-447, 1985 60. Schwartz JM, Baxter LR, Mazziotta JC, et al: The differential diagnosis of depression. JAMA 258: 10:1368-1373, 1987 61. Baxter LR, Schwartz JM, Phelps ME, et al: Reduction of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen Psvchiatrv 46: 243-250, 1989 62. Rush AJ, Schlesser MA, Stokely EM, et al: Cerebral blood flow in depression and mania. Psycophamacol Bull 18:6-8, 1982 63. Devous MD, Rush AJ, Schlesser MA, et al: Single photon tomographic determination of regional cerebral blood flow in psychiatric disorders. J Nucl Med 25:P57, 1984 64. Sackeim HA, Prohovnik I, Moeller JR, et al: Regional cerebral blood flow in mood disorders. Arch Gen Psychiatry 47:60-70, 1990 65. Broderson P, Paulson OB, Bolwig TG, et al: Cerebral hyperemia in induced epileptic seizures. Arch Neurol 9:334-338, 1973 66. Prohovnik I, Hakansson K, Risberg J: Observations of the functional significance of regional cerebral blood flow in resting normal patients. Neuropsychologia 18:203-217, 1980
305
0. Bonne et al.
67. Rosenberg R, Vorstrup S, Andersen A, Bolwig MD: Effect of ECT on cerebral blood flow in melancholia assessed with SPECT. Convulsive Ther 4:62-73, 1988 68. Bajc M, Medved V, Basic M, et al: Acute effect of electroconvulsive therapy on brain perfusion assessed by Tc-99m-hexamethylpropyleneamineoxim and single photon emission computed tomography. Acta Psyciatr Stand 80:421-426, 1989 69. Ackerman RF, Engel J, Baxter L: Positron emission tomography and autoradiographic studies of glucose utilization following electroconvulsive seizures in humans and rats. Ann NY Acad Sci 462~263-269, 1986 70. Silfverskiold I’, Gustafson L, Risberg J, Rosen I: Acute and late effects of electroconvusive therapy. Ann NY Acad Sci 462~236-248, 1986 71. Schwartz JA, Speed NM, Mountz JM, et al: 99mTc-Hexamethylpropyleneamineoxim single photon emission CT in poststroke depression. Am J Psychiatry 1471242-244, 1990 72. Enzmann DR, Lane B: Cranial computed tomography findings in anorexia nervosa. J Comput Assist Tomogr 1:410-414, 1977 73. Kohlmeyer K, Lehmkul G, Poutska F: Computer tomography of anorexia nervosa. AJNR 4:437-438, 1983 74. Herholz K, Krieg JC, Emrich HM, et al: Regional cerebral glucose metabolism in anorexia nervosa measured by positron emission tomography. Biol Psychiatry 22:43-51, 1987 75 Krieg JC, Lauer C, Leinsinger G, et al: Brain morphology and regional cerebral blood flow in anorexia nervosa. Biol Psychiatry 25:1041-1048, 1989 76. Woods SW, Koster K, Krystal JK, et al: Yohimbine
306
77.
78. 79.
80.
81.
82.
83.
84.
85.
alters regional cerebral blood in panic disorder. Lancet 21678, 1988 Tikofsky RS, Goldstein MD, Hellman RS, et al: Quantitiation of CBF (SPECT/IMP) in treated and untreated panic-disordered patients: comparison with normal controls (abstract) J Nucl Med 31:880 1990 Mathew RJ, Wilson WH: Anxiety and cerebral blood flow. Am J Psychiatry 147:838-849, 1990 Baxter LR, Phelps ME, Mazziotta JC, et al: Local cerebral glucose metabolic rates in obsessivecompulsive disorder: a comparison with rates in unipolar depression and in normal controls. Arch Gen Psychiatry 44:211-218, 1987 Baxter LR, Schwartz JM, Guze BH: Pet imaging in obsessive compulsive disorder with and without depression (suppl) J Clin Psychiatry 51:61-69, 1990 Nordahl TE, Benkelfat C, Semple WE, et al: Cerebral glucose metabolic rates in obsessive-compulsive disorder. Neuropsychopharmacology 2:23-28, 1989 Swedo SE, Schapiro MB, Grady CL, et al: Cerebral glucose metabolism in childhood-onset obsessivecompulsive disorder. Arch Gen Psychiatry 46:518523, 1989 Hemmingsen R, Vorstrup S, Clemmesen L, et al: Cerebral blood flow during delirium tremens and related clinical states studied with Xenon-133 inhalation tomography. Am J Psychiatry 145:1384-1390, 1988 Marx I’, Neundorfer B, Potz G, et al: Cerebral blood flow and aminoacid metabolism. In Harper AM (ed), Blood Flow and Metabolism in the Brain. Edinburgh, Churchill Livingstone, 1975 Berglund M, Risberg J: Regional cerebral blood flow during alcohol withdrawl. Arch Gen Psychiatry 38: 351-355, 1981