iew

arstra

Keywords:EpilepsySPECTPET

taltom

platform is used, which combines the functional imaging with the additional morphological information

genic region in the brain is necessary to identify the epileptogenicarea and minimize the side effects of the operation. This localiza-tion should be done ideally using several methods based on differ-ent pathophysiological principles. To plan the optimal surgicalapproach for the achievement of complete seizure control, an addi-

presurgical evaluation of epilepsies, as these methods are able todetect epileptogenic foci in morphologically inconspicuous areasand because they are, in contrast to EEG-based methods, not pri-marily dependent on the electrical activity of the brain. State-of-the-art imaging with PET and SPECT should be an integral part ofa multimodality imaging platform and can, if properly used, sub-stantially reduce the number of invasive EEG recordings. In addi-tion, these are important research tools to gain a betterunderstanding of the neurobiology of epilepsy.

* Corresponding author. Fax: +49 89 7095 7646.

Epilepsy & Behavior 15 (2009) 5055

Contents lists availab

&

.e lE-mail address: christian.lafougere@med.uni-muenchen.de (C. la Fougre).Epilepsy is a chronic disorder characterized by repeated sei-zures caused by the excessive electrical ring of a number of neu-rons. Antiepileptic drugs are useful in controlling the seizures, but20 to 30% of patients with epilepsy continue to have seizures de-spite treatment [1]. Surgical intervention is an important optionand has become an accepted treatment method in properly se-lected patients with intractable focal epilepsy, but accurate resec-tion of epileptogenic areas is imperative for successful seizurecontrol. For the selection of patients eligible for surgery, as wellas for best treatment response, precise localization of the epilepto-

Although long-term intracranial EEG evaluation remains thegold standard here, the invasiveness of this approach requires care-ful patient preselection. Structural imaging with magnetic reso-nance imaging (MRI) is necessary for presurgical investigation toidentify morphological abnormalities such as hippocampal sclero-sis and malformations of cortical development. In addition, MRIcan provide anatomical information for functional imaging modal-ities like single-photon emission computed tomography (SPECT)and positron emission tomography (PET).

Functional neuroimaging by PET and SPECT has a role in the1. Introduction1525-5050/$ - see front matter 2009 Elsevier Inc. Adoi:10.1016/j.yebeh.2009.02.025of magnetic resonance imaging (MRI), but was lower in extra temporal lobe epilepsy. Functional imagingwith SPECT and PET reects seizure related changes of cerebral perfusion, glucose-metabolism and neu-roreceptor status. In this review the usefulness of SPECT and PET imaging in clinical routine in epilepsy aswell as the role of different neuroreceptor PET-tracer, which were used in epilepsy are discussed. The useof perfusion SPECT tracer allows the investigation of ictal activations, but the low temporal resolution ofictal perfusion SPECT often results in the detection of both the ictal onset zone as well as the propagationpathways, an area that has not always need to be resected in order to render a patient seizure free. Theadditional use of interictal PET with uorine-18 uorodeoxyglucose which measures regional cerebralmetabolism or interictal perfusion SPECT enhance the informational value of ictal SPECT and were shownto be important tools to better dene the ictal onset and surround inhibition zones. In recent years PETimaging of different cerebral neuroreceptor-systems inter alia GABAA receptors, serotonin receptors(5-HT1A), opioid receptors as well as dopamine receptors was used to investigate the neurochemical basisof epilepsy, the role of these neurotransmitters for the epileptogenesis as well as the spread of epilepticactivity during seizures and partially entered in clinical routine. Currently some of these radioligands arealso used to investigate new treatment approaches.

2009 Elsevier Inc. All rights reserved.

tional important objective of presurgical investigation is denitionof the boundaries of the epileptogenic region.Accepted 19 February 2009Available online 21 February 2009

the presurgical evaluation of epileptic onset zone in patients with intractable partial epilepsy. In temporallobe epilepsy the sensitivity of these methods was shown to be excellent, in particular if a multimodalReview

PET and SPECT in epilepsy: A critical rev

C. la Fougre *, A. Rominger, S. Frster, J. Geisler, P. BDepartment of Nuclear Medicine, Ludwig Maximilian University of Munich, Marchionini

a r t i c l e i n f o

Article history:Received 18 February 2009

a b s t r a c t

Molecular imaging with icwell as positron emission

Epilepsy

journal homepage: wwwll rights reserved.tensteinsse 15, D-81377 Munich, Germany

and interictal single-photon emission computed tomography (SPECT) asography (PET) rank among the established functional imaging tests for

le at ScienceDirect

Behavior

sevier .com/locate /yebeh

2. SPECT

Although brain perfusion SPECT allows only quantitative esti-mations through tracer uptake ratios, it is widely used to measureregional cerebral blood ow (rCBF) and is an accepted adjunctivetechnique in the presurgical evaluation of patients with refractorypartial epilepsy [2]. In epilepsy, its application is based on theassumption that the increased ictal neuronal activity occurringduring epileptic seizures is associated with increased metabolismand regional cerebral blood ow. For brain perfusion SPECTinvestigations, iodine-123- and technetium-99m-labeled radio-pharmaceuticals are commercially available that differ in pharma-cokinetic behavior and physical characteristics [3]. However, theseradiopharmaceuticals have in common the ability to cross the in-tact bloodbrain barrier rapidly, because of their small molecularsize and their lipophilicity, to be distributed proportionally to the

Cortical and subcortical rCBF changes during seizures may be-gin with hyperperfusion in the epileptic zone followed by rapidextension to other regions through seizure spread and general-ization. However, because of its low temporal resolution, ictalSPECT hyperperfusion patterns often contain both the ictal onsetzone and the propagation pathways; however, the latter regiondoes not need to be resected to render a patient seizure free[11].

Ictal SPECT has been shown to be more accurate for localizationof temporal lobe seizures when the radiotracer is injected immedi-ately after the seizure [12], but usually, tracer injection is per-formed briey after the seizure starts, and moreover, a delaymust be overcome before the radiotracer gets to the brain. There-fore, ictal SPECT shows an increase in rCBF that is related to the sei-zure, but that also might indicate propagation from the area of ictalonset. Moreover, the so-called postictal switch phenomenon,

C. la Fougre et al. / Epilepsy & Behavior 15 (2009) 5055 51blood ow in the cerebral tissue and to be retained in the brainfor a sufcient time to permit image acquisition (>30 min) [4].

Initially 123I-labeled amines were used for this purpose, butthese tracers reach peak brain activity as late as 20 min after injec-tion and, additionally, show redistribution over time, resulting inreuptake into the cerebral cortex that is not proportional to rCBF[5]. Because of this, 99mTc-hexamethylpropyleneamine-oxime(99mTc-HMPAO) and 99mTc-ethyl cysteinate dimer (99mTc-ECD),which have several advantages over 123I-labeled amines, are cur-rently the most frequently used tracers for investigating rCBF.These tracers have a quicker initial uptake in brain and reach thepeak within 2 min of injection, without redistribution. Thus, theinitial tracer uptake and distribution correspond to rCBF at thetime of injection and remain unchanged up to at least 2 h, indepen-dent of rCBF variations occurring after the xation time. Conse-quently, the radiotracer can be injected into the patient outsidethe nuclear medicine facility, for example, during an epileptic sei-zure in the epilepsy ward; rCBF images reecting the distributionat the moment of seizure can be acquired later with a SPECT cam-era after recovery from the seizure. The appropriate half-time (6 h)of technetium, as well as the stability of the tracer (particularlyECD), renders this technique useful and affordable for clinicalroutine.

The superiority of ictal SPECT compared with interictal SPECTfor identication of the location or the lateralization of epilepticseizures has been demonstrated by several studies of patients withtemporal lobe epilepsy (TLE), indicating sensitivities between 73and 97% for ictal SPECT and only 50% for interictal SPECT [69].Although there are fewer studies dealing with the role of SPECTin extra-TLE, data suggest lower sensitivity (66%) for ictal SPECTcompared with the results found in TLE [8,10].Fig. 1. Patient with focal epilepsy in the left frontal lobe left. Ictal 99mTc-ECD SPECT scainterictal 99mTc-ECD SPECT scan (B) reveals no signicant asymmetry. The result of subinterictal scan (B), delineates ictal activation. In the corresponding interictal [18F]FDG-PETwhich results in false localization or lateralization of the ictal sei-zure, can be observed when the delay between seizure onset andtracer application is too long [13].

The marksmanship of SPECT has been shown to be improved bydigital analysis of ictal and interictal examinations, that is, by sub-traction of ictal and interictal scans, as shown in Fig. 1, whichshould preferentially be co-registered to MRI (SISCOM) [1419].This multimodality imaging, which combines the structural andfunctional imaging information, improves the ability to detectand dene the extent of epileptogenic lesions and to regionalizepotentially epileptogenic regions in patients who have normalMRI scans. In addition, the remarkable predictive value of SISCOMwith respect to surgical outcome has been described, and could beof help in the decision making for epilepsy surgery: among patientswhose SISCOM ndings fell within the margins of the resected tis-sue or in the disconnected hemisphere, 75% were rendered seizurefree; among those whose SISCOM ndings fell outside the marginsof the surgery, 100% continued to have seizures [19].

3. Glucose metabolism with [18F]FDG

Positron emission tomography with uorine-18 uorodeoxy-glucose ([18F]FDG), as well as other receptor ligands, is an impor-tant tool to dene the ictal onset zone and to better understandthe neurobiology and functional alterations induced by variousforms of epilepsy.

The glucose analog [18F]FDG is an indirect marker of neuronalactivity and allows absolute quantication of cerebral glucosemetabolism when additional arterial (ideally) or venous bloodsamples are taken. Interictal [18F]FDG-PET has an established rolein the noninvasive localization of epileptogenic foci and also re-n (A) shows a discrete hyperperfusion in the left frontal lobe (red arrow), but thetraction of the ictal and interictal scans (blue area), which is superimposed on theinvestigation (C), a discrete hypometabolism was observed in this area (red arrow).

ects dynamic seizure-related changes in cerebral cellularfunctions.

The epileptogenic focus in the interictal phase usually appearsas a hypometabolic area on [18F]FDG-PET. In TLE, comparison withthe contralateral side has been shown to be useful for localizationof the epileptic zone. However, [18F]FDG uptake frequently extendsbeyond the seizure onset zone, which makes precise focus localiza-tion difcult. For example, many patients with TLE present with anadditional pronounced hypometabolism of the frontal lobe, whichmay represent inhibitory phenomena induced by the epileptogenicfocus.

Numerous studies, some of which consisted of large numbers ofpatients, have reported a sensitivity of 7085% for [18F]FDG-PET inpatients with TLE. In patients with extra-TLE, however, the popula-tions investigated were smaller, and the values for diagnostic sen-sitivity and accuracy reported for interictal studies weresignicantly lower: 3060% depending on the localization of thefocus [2025]. The highest clinical benet of [18F]FDG-PET can beachieved in patients with suspected TLE and normal MRI ndings.In this situation, PET correctly lateralizes the lesion in 80% of thecases [25].

Ictal [18F]FDG-PET cannot be routinely performed because of lo-gistic difculties due mainly to the short half-life of the 18F label(110 min). In addition, interpretation is difcult because of the pro-longed cerebral [18F]FDG uptake, which may lead to contaminationof the signal by seizure propagation [26].

4. Neuroreceptor PET

have been used to investigate the neurochemical basis of the epi-lepsies and have partially entered clinical routine.

4.1. GABAA receptors: [11C]Flumazenil

[11C]Flumazenil binds to the central benzodiazepine receptor(cBZR)-c-aminobutyric acid (GABAA) receptor complex. Measure-ments of GABAA receptor density with PET and [11C]umazenilhave been demonstrated to be promising methods of identifyingand localizing epileptogenic regions. In this context, GABA receptorbinding was signicantly lower in the epileptic focus than in thecontralateral homotopic reference region and the remaining neo-cortex [27]. Reduced binding can be seen at the epileptogenic focusand the seizure onset region, however, in a more restricted distri-bution than the corresponding area of [18F]FDG hypometabolism,and sometimes also in projection areas [28]. In addition, the degreeof GABA receptor reduction showed a positive correlation with sei-zure frequency.

Furthermore, [11C]umazenil PET has been shown to distin-guish patients with frequent seizures. Therefore, the method maynot only be suitable for noninvasive localization of the seizure fo-cus, but may also provide a biochemical marker of epileptogenicity,strengthening the hypothesis that inhibitory mechanisms are dis-turbed in the epileptic focus [29]. Moreover, [11C]umazenil PETwas shown to be more accurate than [18F]FDG-PET for the detec-tion of cortical regions of seizure onset and frequent spiking in pa-tients with extra-TLE, whereas both [18F]FDG and [11C]umazenilPET showed low sensitivity in the detection of cortical areas of ra-pid seizure spread [30]. Even if [11C]umazenil PET did not provesuperior to [18F]FDG-PET in assessing the extent of the ictal onset

52 C. la Fougre et al. / Epilepsy & Behavior 15 (2009) 5055Neurotransmitters are directly responsible for modulating syn-aptic activity and PET also allows quantication of specic ligandreceptor relationships that seem to be important for epileptogene-sis and spread of epileptic activity. Several PET receptor ligandsFig. 2. [18F]FDG (A: coronal, B: transversal) and [18F]ethylumazenil (C: coronal, D: tranmetabolism and reduced binding to the cBZR-GABAA receptor complex in the mesial temzone, as dened by intracranial EEG recordings, some authors sta-ted that it may provide useful data complementary to those of MRIand [18F]FDG PET [24], as shown in Fig. 2. However, the use ofsversal) PET scans of a patient with right TLE showing relatively decreased glucoseporal lobe (red arrows).

y &interictal [11C]umazenil PET in patients with idiopathic general-ized epilepsy has yielded nonuniform results: thus, a slight reduc-tion in the neocortex of patients with idiopathic generalizedepilepsy has been reported in comparison with patients with par-tial seizures [31]; in another study, the authors reported a wide-spread increase in benzodiazepine receptors in cerebralneocortex, thalamus, and cerebellar cortex [32]. Also observedwas increased benzodiazepine receptor density in the cerebellarnuclei and decreased density in the thalamus [33].

In unilateral hippocampal sclerosis, reduction of binding of[11C]umazenil has been shown to be conned to the hippocampusand to be over and above that caused by neuron loss and hippo-campal atrophy [34]. In malformations of cortical development,abnormalities of benzodiazepine receptors, as demonstrated with[11C]umazenil PET, were more extensive than the structuralabnormality revealed with MRI [35]. There often were areas of in-creased benzodiazepine receptors, a pattern that may reect bothfunctional and structural anomalies [36].

Currently, the clinical role of [11C]umazenil and its SPECTderivative [123I]iomazenil for the detection of temporal and extra-temporal foci is viewed controversially in the literature [37,38].Even though [11C]umazenil binding has revealed focal abnormal-ities in 80% of patients with refractory TLE and normal high-qualityMRI [38], this method has not been considered consistently helpfulin localizing epileptic foci. In addition, the short half-time of 11C(20 min) hampers the wide use of [11C]umazenil for clinical rou-tine. The possible implementation of [18F]umazenil in clinicalroutine will most probably provide a more denite answer aboutthe clinical benet of benzodiazepine receptor imaging [39]. Thisradiopharmaceutical has ideal properties for PET imaging and willallow with state-of-the-art equipment to image benzodiazepineGABAA receptor binding with a spatial resolution of approximately3 mm [40].

4.2. Serotonin receptors (5-HT1A): [18F]MPPF

Experimental data in animals indicate that 5-HT1A receptors arelocated predominantly in limbic areas and that serotonin, via thesereceptors, mediates an antiepileptic and anticonvulsant effect.

[18F]MPPF (20-methoxyphenyl-(N-20-pyridinyl)-p-[18F]uoro-benzamidoethylpiperazine), an antagonist of the 5-HT1A receptors,was used to assess the extent of 5-HT1A receptor binding changesin patients with TLE and hippocampal ictal onset. A signicantlydecreased binding potential (BPND) was detected ipsilateral to theepileptogenic zone in the hippocampus, temporal pole, insula,and temporal neocortex. In patients with normal hippocampus vol-ume, the BP decrease was restricted to the temporal pole [41]. Thedecrease was more pronounced in the seizure onset zone and in re-gions where the discharge propagated than in regions where onlyinterictal paroxysms or no epileptic activity was recorded [42].Therefore, these results indicate that [18F]MPPF binding not onlyreects structural changes or neuronal loss, but can also be consid-ered a marker of the epileptogenic zone.

The reduction in 5-HT1a BPND in patients with mesial TLE maybe explained, on the one hand, by a decrease in receptor densitydue to depletion of serotonin, which leads to downregulationand hyperexcitability, or, on the other hand, by an increase inendogenous serotonin as a reactional process to modulatehyperexcitability.

A recent study showed that, in individual patients with TLE, vi-sual analysis of [18F]MPPF PET data, blinded to clinical informationand data from other presurgical investigations, permits identica-tion of the epileptogenic lobe with a sensitivity of 90% (38 of 42 pa-

C. la Fougre et al. / Epilepstients), even in those without hippocampal sclerosis, and provedusable in clinical practice for preoperative evaluation of patientswith TLE [43]. Comparison between [18F]FDG and [18F]MPPF PETdata showed the clear-cut superiority of the latter for localizingvisually the epileptogenic area in patients with mesial TLE. Thespecicity of [18F]FDG PET was lower than that of [18F]MPPF PET,with only 23% of patients showing hypometabolism restricted tohippocampus, amygdala, and temporal pole, versus 38% demon-strating such abnormalities with [18F]MPPF PET [43].

4.3. Opioid receptors

PET studies using opioid receptor ligands with different selec-tivity support ndings from animal experiments suggesting a pre-dominantly anticonvulsant effect of opioid peptides (for review seeHenriksen et al. [44]). The earliest human PET study indicatingselective modulation of the opiate system in partial epilepsy wasthat of Frost et al. [45], who demonstrated, with the l-selective li-gand [11C]carfentanyl, increased l-opioid receptor availability inthe areas of the epileptogenic temporal lobe exhibiting interictalhypometabolism. In contrast to this nding no comparable asym-metry was detectable with the nonspecic opioid receptor ligand[11C]diprenorphine and the l- and j-selective ligand [18F]cyclof-oxy [4648]. A study with the d-selective opioid receptor ligand[11C]methylnaltrindole also showed increased receptor availabilityin the epileptogenic temporal lobe, but with a different regionalpattern [49]. The in vivo demonstration in humans of selective re-gional alterations of l- and j-receptor binding in TLE supports thehypothesis derived from animal studies that endogenous ligandbinding to these receptor subtypes plays a role in the tonic anti-convulsive mechanism that limits the spread of seizure activityfrom the epileptogenic focus [44].

The in vivo demonstration of a generalized displacement of opi-oid receptor ligands during absence seizures [50] and a focal dis-placement during seizures of reading epilepsy [51] providesstrong support for the prevailing opinion that endogenous opioidsare released following generalized and partial seizures [52]. Resultsof a study with [11C]diprenorphine by Hammers et al. [53] suggesta role for the opioid system in the postictal rise in seizure thresh-old. This study demonstrated a postictal increase in [11C]diprenor-phine binding. The authors of the study suggested that synapticopioid levels increase at the time of seizures, leading to a reductionin [11C]diprenorphine binding, and that this is followed by a grad-ual recovery of available surface receptors with an overshoot overbasal levels, which is detected by PET about 8 h after seizures, anda gradual return to normal or low-normal levels during the interic-tal phase.

4.4. Dopamine receptors: [18F]Fallypride

The role of dopamine in the pathophysiology of focal epilepsy isa matter of controversy. The dopaminergic inuence on epilepticseizures arising from mesial temporal structures [54] might beinhibitory (i.e., via dopaminergic hippocampal projections enhanc-ing Ca2+-dependent K+ conductance) [55]. Data obtained in animalssuggest that activation of the dopaminergic D1- and D2-receptorshas diverging effects on the regulation of seizure threshold, D2being anticonvulsant and D1 activation more proconvulsant[54,56]. Further studies in animals suggest a neuroprotective roleof dopamine through the inhibitory control of glutamate neuro-transmission and excitotoxicity in epilepsy [57].

The high-afnity dopamineD2/D3-receptor antagonist [18F]fally-pride can be used to characterize striatal as well as extrastriatal D2/D3-receptor binding. In patients with mesial TLE and hippocampalsclerosis, D2/D3-receptor binding was reduced at the pole and thelateral aspects of the epileptogenic temporal lobe, but there was

Behavior 15 (2009) 5055 53no change in hippocampal binding although all patients had hippo-campal atrophy. This selective decrease inD2/D3-receptor binding inthe pole and lateral temporal lobe, which show hypometabolism on

y &[18F]FDG-PET, is an indication of a modulation of the dopaminergicsystem in the irritative zone, indicating that the dopaminergicsystemmight also be part of the endogenous anticonvulsant mech-anism that prevents seizure generalization [58].

In conclusion, SPECT and PET are useful and reliable tools forimaging epilepsies in clinical routine as well as for research pur-poses. Their unique ability to image in vivo changes in brain bio-chemistry has provided valuable insights into the mechanisms ofepileptogenesis, seizure propagation, and termination. Progress inthe sensitivity and spatial resolution achieved by modern instru-mentation will probably further increase the clinical role of func-tional imaging in the preoperative assessment of patients withepilepsy.

References

[1] Engel Jr J. Introduction to temporal lobe epilepsy. Epilepsy Res1996;26:14150.

[2] OBrien TJ, So EL, Mullan BP, et al. Subtraction ictal SPECT co-registered to MRIimproves clinical usefulness of SPECT in localizing the surgical seizure focus.Neurology 1998;50:44554.

[3] Catafau AM. Brain SPECT in clinical practice. Part I: perfusion. J Nucl Med2001;42:25971.

[4] Neirinckx RD, Canning LR, Piper IM, et al. Technetium-99m dl-HMPAO: a newradiopharmaceutical for SPECT imaging of regional cerebral blood perfusion. JNucl Med 1987;28:191202.

[5] Nishizawa S, Tanada S, Yonekura Y, et al. Regional dynamics of N-isopropyl-(123I)p-iodoamphetamine in human brain. J Nucl Med 1989;30:1506.

[6] Spanaki MV, Spencer SS, Corsi M, MacMullan J, Seibyl J, Zubal IG. Sensitivityand specicity of quantitative difference SPECT analysis in seizure localization.J Nucl Med 1999;40:7306.

[7] Devous Sr MD, Thisted RA, Morgan GF, Leroy RF, Rowe CC. SPECT brain imagingin epilepsy: a meta-analysis. J Nucl Med 1998;39:28593.

[8] Weil S, Noachtar S, Arnold S, Yousry TA, Winkler PA, Tatsch K. Ictal ECD-SPECTdifferentiates between temporal and extratemporal epilepsy: conrmation byexcellent postoperative seizure control. Nucl Med Commun 2001;22:2337.

[9] Zaknun JJ, Bal C, Maes A, et al. Comparative analysis of MR imaging, ictal SPECTand EEG in temporal lobe epilepsy: a prospective IAEA multi-center study. EurJ Nucl Med Mol Imaging 2008;35:10715.

[10] Lee JJ, Lee SK, Lee SY, et al. Frontal lobe epilepsy: clinical characteristics,surgical outcomes and diagnostic modalities. Seizure 2008;17:51423.

[11] Van Paesschen W, Dupont P, Sunaert S, Gofn K, Van Laere K. The use of SPECTand PET in routine clinical practice in epilepsy. Curr Opin Neurol2007;20:194202.

[12] Van Paesschen W. Ictal SPECT. Epilepsia 2004;45(Suppl. 4):3540.[13] Newton MR, Berkovic SF, Austin MC, Rowe CC, McKay WJ, Bladin PF. Postictal

switch in blood ow distribution and temporal lobe seizures. J NeurolNeurosurg Psychiatry 1992;55:8914.

[14] Ahnlide JA, Rosen I, Linden-Mickelsson Tech P, Kallen K. Does SISCOMcontribute to favorable seizure outcome after epilepsy surgery? Epilepsia2007;48:57988.

[15] Kaiboriboon K, Lowe VJ, Chantarujikapong SI, Hogan RE. The usefulness ofsubtraction ictal SPECT coregistered to MRI in single- and dual-headed SPECTcameras in partial epilepsy. Epilepsia 2002;43:40814.

[16] OBrien TJ. SPECT: methodology. Adv Neurol 2000;83:1132.[17] OBrien TJ, So EL, Cascino GD, et al. Subtraction SPECT coregistered to MRI in

focal malformations of cortical development: localization of the epileptogeniczone in epilepsy surgery candidates. Epilepsia 2004;45:36776.

[18] OBrien TJ, So EL, Mullan BP, et al. Subtraction peri-ictal SPECT is predictive ofextratemporal epilepsy surgery outcome. Neurology 2000;55:166877.

[19] Wichert-Ana L, de Azevedo-Marques PM, Oliveira LF, et al. Interictalhyperemia correlates with epileptogenicity in polymicrogyric cortex.Epilepsy Res 2008;79:3948.

[20] Drzezga A, Arnold S, Minoshima S, et al. 18F-FDG PET studies in patients withextratemporal and temporal epilepsy: evaluation of an observer-independentanalysis. J Nucl Med 1999;40:73746.

[21] Arnold S, Schlaug G, Niemann H, et al. Topography of interictal glucosehypometabolism in unilateral mesiotemporal epilepsy. Neurology1996;46:142230.

[22] Gambhir S, Singh Sethi R, Deswal S. Incidental detection of a single brainmetastasis from breast carcinoma during Tc-99m MIBI scintimammography.Clin Nucl Med 2001;26:8834.

[23] Juhasz C, Chugani DC, Muzik O, et al. Relationship of umazenil and glucosePET abnormalities to neocortical epilepsy surgery outcome. Neurology2001;56:16508.

[24] Ryvlin P, Bouvard S, Le Bars D, et al. Clinical utility of umazenil-PET versus[18F]uorodeoxyglucose-PET and MRI in refractory partial epilepsy: aprospective study in 100 patients. Brain 1998;121(Pt. 11):206781.

54 C. la Fougre et al. / Epileps[25] Won HJ, Chang KH, Cheon JE, et al. Comparison of MR imaging with PET andictal SPECT in 118 patients with intractable epilepsy. AJNR Am J Neuroradiol1999;20:5939.[26] Barrington SF, Koutroumanidis M, Agathonikou A, et al. Clinical value of ictalFDG-positron emission tomography and the routine use of simultaneous scalpEEG studies in patients with intractable partial epilepsies. Epilepsia1998;39:75366.

[27] Savic I, Persson A, Roland P, Pauli S, Sedvall G, Widen L. In-vivo demonstrationof reduced benzodiazepine receptor binding in human epileptic foci. Lancet1988;2:8636.

[28] Savic I, Thorell JO, Roland P. [11C]Flumazenil positron emission tomographyvisualizes frontal epileptogenic regions. Epilepsia 1995;36:122532.

[29] Savic I, Svanborg E, Thorell JO. Cortical benzodiazepine receptor changes arerelated to frequency of partial seizures: a positron emission tomographystudy. Epilepsia 1996;37:23644.

[30] Muzik O, da Silva EA, Juhasz C, et al. Intracranial EEG versus umazenil andglucose PET in children with extratemporal lobe epilepsy. Neurology2000;54:1719.

[31] Savic I, Widen L, Thorell JO, Blomqvist G, Ericson K, Roland P. Corticalbenzodiazepine receptor binding in patients with generalized and partialepilepsy. Epilepsia 1990;31:72430.

[32] Koepp MJ, Richardson MP, Brooks DJ, Cunningham VJ, Duncan JS. Centralbenzodiazepine/gamma-aminobutyric acid A receptors in idiopathicgeneralized epilepsy: an [11C]umazenil positron emission tomographystudy. Epilepsia 1997;38:108997.

[33] Savic I, Pauli S, Thorell JO, Blomqvist G. In vivo demonstration of alteredbenzodiazepine receptor density in patients with generalised epilepsy. JNeurol Neurosurg Psychiatry 1994;57:797804.

[34] Koepp MJ, Labbe C, Richardson MP, et al. Regional hippocampal[11C]umazenil PET in temporal lobe epilepsy with unilateral and bilateralhippocampal sclerosis. Brain 1997;120(Pt. 10):186576.

[35] Hammers A, Koepp MJ, Richardson MP, et al. Central benzodiazepine receptorsin malformations of cortical development: a quantitative study. Brain2001;124:155565.

[36] Duncan JS. Positron emission tomography receptor studies. Adv Neurol1999;79:8939.

[37] Kaneko K, Sasaki M, Morioka T, et al. Pre-surgical identication ofepileptogenic areas in temporal lobe epilepsy by 123I-iomazenil SPECT: acomparison with IMP SPECT and FDG PET. Nucl Med Commun 2006;27:8939.

[38] Koepp MJ, Hammers A, Labbe C, Woermann FG, Brooks DJ, Duncan JS. 11C-Flumazenil PET in patients with refractory temporal lobe epilepsy and normalMRI. Neurology 2000;54:3329.

[39] Ryzhikov NN, Seneca N, Krasikova RN, et al. Preparation of highly specicradioactivity [18F]umazenil and its evaluation in cynomolgus monkey bypositron emission tomography. Nucl Med Biol 2005;32:10916.

[40] de Jong HW, van Velden FH, Kloet RW, Buijs FL, Boellaard R, Lammertsma AA.Performance evaluation of the ECAT HRRT: an LSO-LYSO double layer highresolution, high sensitivity scanner. Phys Med Biol 2007;52:150526.

[41] Merlet I, Ryvlin P, Costes N, et al. Statistical parametric mapping of 5-HT1Areceptor binding in temporal lobe epilepsy with hippocampal ictal onset onintracranial EEG. NeuroImage 2004;22:88696.

[42] Merlet I, Ostrowsky K, Costes N, et al. 5-HT1A receptor binding andintracerebral activity in temporal lobe epilepsy: an [18F]MPPF-PET study.Brain 2004;127:90013.

[43] Didelot A, Ryvlin P, Lothe A, Merlet I, Hammers A, Mauguiere F. PET imaging ofbrain 5-HT1A receptors in the preoperative evaluation of temporal lobeepilepsy. Brain 2008;131:275164.

[44] Henriksen G, Willoch F. Imaging of opioid receptors in the central nervoussystem. Brain 2008;131:117196.

[45] Frost JJ, Mayberg HS, Fisher RS, et al. Mu-opiate receptors measured bypositron emission tomography are increased in temporal lobe epilepsy. AnnNeurol 1988;23:2317.

[46] Bartenstein PA, Prevett MC, Duncan JS, Hajek M, Wieser HG. Quantication ofopiate receptors in two patients with mesiobasal temporal lobe epilepsy,before and after selective amygdalohippocampectomy, using positronemission tomography. Epilepsy Res 1994;18:11925.

[47] Mayberg HS, Sadzot B, Meltzer CC, et al. Quantication of mu and non-muopiate receptors in temporal lobe epilepsy using positron emissiontomography. Ann Neurol 1991;30:311.

[48] Theodore WH, Carson RE, Andreasen P, et al. PET imaging of opiate receptorbinding in human epilepsy using [18F]cyclofoxy. Epilepsy Res 1992;13:12939.

[49] Madar I, Lesser RP, Krauss G, et al. Imaging of delta- and mu-opioid receptorsin temporal lobe epilepsy by positron emission tomography. Ann Neurol1997;41:35867.

[50] Bartenstein PA, Duncan JS, Prevett MC, et al. Investigation of the opioid systemin absence seizures with positron emission tomography. J Neurol NeurosurgPsychiatry 1993;56:1295302.

[51] Koepp MJ, Richardson MP, Brooks DJ, Duncan JS. Focal cortical release ofendogenous opioids during reading-induced seizures. Lancet 1998;352:9525.

[52] Bajorek JG, Lee RJ, Lomax P. Neuropeptides: anticonvulsant and convulsantmechanisms in epileptic model systems and in humans. Adv Neurol1986;44:489500.

[53] Hammers A, Asselin MC, Hinz R, et al. Upregulation of opioid receptor bindingfollowing spontaneous epileptic seizures. Brain 2007;130:100916.

Behavior 15 (2009) 5055[54] Barone P, Palma V, DeBartolomeis A, Tedeschi E, Muscettola G, Campanella G.Dopamine D1 and D2 receptors mediate opposite functions in seizuresinduced by lithiumpilocarpine. Eur J Pharmacol 1991;195:15762.

[55] Benardo LS, Prince DA. Dopamine modulates a Ca2+-activated potassiumconductance in mammalian hippocampal pyramidal cells. Nature1982;297:769.

[56] Al-Tajir G, Starr MS. Anticonvulsant effect of striatal dopamine D2 receptorstimulation: dependence on cortical circuits? Neuroscience 1991;43:517.

[57] Bozzi Y, Vallone D, Borrelli E. Neuroprotective role of dopamine againsthippocampal cell death. J Neurosci 2000;20:86439.

[58] Werhahn KJ, Landvogt C, Klimpe S, et al. Decreased dopamine D2/D3-receptorbinding in temporal lobe epilepsy: an [18F]fallypride PET study. Epilepsia2006;47:13926.

C. la Fougre et al. / Epilepsy & Behavior 15 (2009) 5055 55

PET and SPECT in epilepsy: A critical reviewIntroductionSPECTGlucose metabolism with [18F]FDGNeuroreceptor PETGABAA receptors: [11C]FlumazenilSerotonin receptors (5-HT1A): [18F]MPPFOpioid receptorsDopamine receptors: [18F]Fallypride

References