FULL-LENGTH ORIGINAL RESEARCH

Ictal single photon emission computed tomography in epileptic auras *†Sherif A. Elwan, ‡Guiyun Wu, ‡Steve S. L. Huang, *Imad M. Najm, and *Norman K. So Epilepsia, 55(1):133–136, 2014 doi: 10.1111/epi.12475

SUMMARY

Sherif Elwan is a part of the Neurology Staff at Ain Shams University, Cairo, Egypt.

Objectives: Little is known about whether ictal single photon emission computed tomography (SPECT) during an isolated aura can localize the epileptogenic zone (EZ). This study seeks to evaluate the yield of ictal SPECT injection in isolated epileptic auras. Methods: We identified 20 patients with focal epilepsy studied during 26 isolated auras by ictal interictal subtraction SPECT coregistered to magnetic resonance imaging (SISCOM). Studies were rated by two readers who blindly scored the images for presence or absence of an area of dominant hyperperfusion and the lateralization and localization of ictal hyperperfusion; kappa statistics were calculated. Results are correlated with the localization or lateralization of the EZ, time of injection, and electroencephalography (EEG) findings during aura. Results: Fourteen (53%) of 26 injections in 13 patients were rated by both readers as having an area of dominant hyperperfusion with poor interobserver agreement (k = 0.128). Nine of 26 injections in eight patients were correctly lateralized to the side of the EZ (j = 0.46), but only one of 21 injections in one patient was correctly localized (j = 0.146). No difference was found when comparing temporal and extratemporal cases. Studies obtained in auras with ictal EEG change were no more likely to be correctly localized than in ones without (p = 0.19). The timing of injection was not a predictor of success. Significance: Ictal SPECT injection during an isolated aura has a low yield of correct localization of the EZ and cannot be relied on alone during presurgical evaluation. A repeat injection during a seizure with clinical signs and ictal EEG accompaniment is recommended. KEY WORDS: Epileptic auras, Ictal single photon emission computed tomography, Epileptogenic zone.

Subtraction ictal single photon emission computed tomography (SPECT) coregistered with magnetic resonance imaging (MRI) (SISCOM) is a valuable technique to help localize the epileptogenic zone (EZ) in patients who are being evaluated for epilepsy surgery.1–4 An earlier SPECT isotope injection produces more localized findings as comAccepted October 13, 2013; Early View publication December 6, 2013. *Epilepsy Center, Neurological Institute, Cleveland Clinic, Cleveland, Ohio, U.S.A.;†Department of Neurology, Ain Shams University, Cairo, Egypt; and ‡Department of Nuclear Medicine, Cleveland Clinic, Cleveland, Ohio, U.S.A. Address correspondence to Norman K. So, Epilepsy Center, S51, Cleveland Clinic, 9500 Euclid Ave, Cleveland, OH 44195, U.S.A. E-mail: [email protected] Wiley Periodicals, Inc. © 2013 International League Against Epilepsy

pared to a later one.2,5 Two previous reports of five and three patients with temporal lobe epilepsy studied by SISCOM during isolated auras, gave conflicting localization rates of 33% and 100%, respectively.6,7 We now report on a series of 20 patients with focal epilepsy, both temporal and extratemporal, studied by ictal SPECT injections during isolated auras.

Methods We identified 20 patients with focal epilepsy who had ictal SPECT during isolated auras as part of their presurgical evaluation. They were from a series of 529 ictal SPECT studies at our institution between 2003 and 2011. This retro-

133

134 S. A. Elwan et al. spective study was approved by the Cleveland Clinic institutional review board. We excluded patients who had nonlateralized or nonlocalized epilepsy or multiple epileptic foci. The methodology for ictal injection of SPECT isotope 99mTc-ECD (ethyl-cysteinate dimer) during scalp video– electroencephalography (EEG) monitoring and subsequent generation of SISCOM images has been described previously.8 The isotope was injected when the patient indicated an aura, whether or not accompanied by a scalp ictal discharge. Two nuclear medicine investigators (GW, SH) blinded to the patient’s identity and other presurgical data assessed the SISCOM images independently using a z-score of 2 as threshold with MatLab 7.5 (The MathWorks, Natick, MA, U.S.A.) and statistical parametric mapping (SPM2) software (Welcome Institute of Neuroscience, University College London, London, United Kingdom), and rated each study for (1) presence or absence of one or more areas of dominant hyperperfusion; (2) lateralized hyperperfusion, left or right; and (3) localized hyperperfusion at a lobar level, where possible. We calculated the interobserver agreement between readers by Kappa statistics and correlated the rate of concordant lateralization or localization of the SISCOM findings to the lateralization or localization of the EZ and to the presence or absence of ictal EEG, and time of injection from clinical onset.

Results All 20 patients had a lateralized EZ (left 8, right 12), and 16 were further localized to a lobe or sublobar region (eight temporal, eight extratemporal). Localization of the EZ was based on congruence of presurgical test results and one of (1) postresection Engel grade I and II seizure outcomes in 10 patients, (2) location of MRI lesion in 4 patients, and (3) stereotactic depth EEG localization in 2 patients. The remaining four patients had seizures with ictal EEG recordings lateralized to one hemisphere, and other supportive findings: lateralized MRI encephalomalacia (two patients), Rasmussen’s syndrome (one patient), and lateralized somatosensory aura and seizure semiology (one patient). Eleven of the 20 patients underwent epilepsy surgery, 10 of whom had an Engel grade I or II outcome, and all had a localized EZ, whereas one had a lateralized EZ but did not have a favorable outcome. Of the auras reported, seven were unclassifiable, five somatosensory, three abdominal, three mixed, one olfactory, and one autonomic. Twenty-six ictal SPECT injections were performed in 20 patients (six patients were injected in two separate auras). Fourteen (53%) of 26 injections in 13 patients were rated by both readers as having an area of dominant hyperperfusion with poor interobserver agreement (k = 0.128). Dominant hyperperfusion was rated as absent by both readers in 2 injections, and in 10 there was discordance for presence or absence of hyperperfusion between them. Subsequent correlation of lateralization and localization with the EZ Epilepsia, 55(1):133–136, 2014 doi: 10.1111/epi.12475

required the consensus of both readers. Testing SISCOM lateralization in the 20 patients, there was concordant and correct lateralization in 9 (35%) of 26 injections in 8 of 20 patients; concordant but incorrect lateralization in 3, concordant but absent lateralization in 2, at least one incorrect lateralization in 6, and one lateralized and one nonlateralized result in 6 injections. SISCOM lateralization with the side of the EZ showed moderate interobserver agreement (j = 0.46). Testing SISCOM localization only one (5%) of 21 injections in one of 16 patients with a localized EZ was correctly localized with poor interobserver agreement (j = 0.146). Lack of concordance resulted from disagreement for lobes in the correct hemisphere, or disagreement on the degree of localization, with one reader giving a lobar localization while the other gave only lateralization within the same hemisphere. In six injections, at least one reader placed the site of hyperperfusion in the incorrect hemisphere. No difference was found when comparing temporal and extratemporal cases. We looked at the presence or absence of scalp ictal EEG change during auras in the 16 patients with a localized EZ studied by 21 injections. One of four injections with an ictal EEG (duration from 5 to 90 s) was correctly localizing; as compared to none in 17 without (p = 0.19). We studied the time of injection from first clinical indication (mean = 19.5 s, median = 15 s, range 3–55 s), comparing earlier ( 0.7) in the rate of correct lateralization or localization (Table 1). Six patients had two separate injections during separate auras, and in four patients one of the auras showed concordant lateralization to the correct hemisphere, two with and two without EEG change. The remaining two patients did not have concordant lateralizing or localizing findings in either injection. Four patients were injected during an aura, and separately in a habitual seizure with motor or dyscognitive signs (Table 2 and Fig. 1). In each case the habitual seizure was correctly lateralizing or localizing as compared to only one of four auras. Table 1. Time of injection from clinical onset and ictal SPECT hyperperfusion

Early injections (3–10 s) Late injections (13–55 s) Statistical significance (p)

Concordant dominant hyperperfusion

Concordant correct lateralization

Concordant correct localization

6/12 8/14 1

3/12 6/14 0.7

0/8 1/13 1

Discussion An early injection is generally held to be advantageous when performing ictal SPECT. One study showed that injections within the first 20 s of clinical or EEG onset were more likely to be correctly localizing compared to later injections.2 In frontal lobe epilepsy, an injection time of 5 s

135 Ictal SPECT in Epileptic Auras Table 2. Ictal SPECT in four patients who had injections during auras and seizures Pt 1 Pt 2 Pt 3 Pt 4

Aura/seizure

Scalp EEG

SSA SSA ? complex motor Abdominal aura Automotor seizure Unclassified aura Right head/arm tonic ? GTCS Unclassified aura Unclassified aura ? automotor ? right version ? GTCS

No EEG change Lateralized right No EEG change Right temporal No EEG change Lateralized left No EEG change Vertex

Injection time (s) 17 12 24 51 7 11 10 17

Ratings by two readers Correct lateralization Correct localization/correct lateralization No dominant focus Correct localization Correct localization/no dominant focus Correct localization Correct lateralization/wrong side Correct lateralization

SSA, somatosensory aura; GTCS, generalized tonic–clonic seizure.

A

B

Figure 1. Ictal SPECT Patient 2. (A) abdominal aura: no EEG change. Ictal SPECT: no dominant hyperperfusion. (B) automotor seizure: EEG right temporal onset. Ictal SPECT: right temporal hyperperfusion. Epilepsia ILAE

correctly localized the EZ in 61% of 18 patients.9 Examples of false contralateral hyperperfusion in mesial temporal lobe epilepsy occurred when the average injection time from ictal EEG onset was 30 s or later.10 For this reason, when a patient indicates the onset of a seizure by an aura, isotope may be injected before knowing if there will be further clinical or EEG progression. Our findings agree with Van Paesschen6 that ictal SPECT studies in isolated auras cannot be relied on, as concordant and correct localization with the EZ was obtained in only 5%, and lateralization in 35%. In a related observation of ictal SPECT in mild seizures characterized by motionless stare in surgically proven temporal lobe epilepsy, no significant hyperperfusion could be found by statistical parametric voxel-based group analysis, even when seizures showed ictal EEG changes.11 In practice, this means that when an isolated aura is the first to

be injected in the course of presurgical workup, it is insufficient, and a more fully developed seizure with clinical signs and ictal EEG change is still needed. It is well known that an aura activates only a small subset of neurons in the EZ,12 which may not be sufficient to increase blood perfusion above the threshold of SPECT detection. Limitations of the present report include small numbers, the inherent need to trust in the patient’s self-identification that the symptoms experienced were typical of their habitual aura instead of something nonspecific, and the inherent inaccuracy of using the self-report time given by the patient as the start of a seizure.

Disclosure None of the authors has any conflict of interest to disclose. This study received no intramural or extramural funding. We confirm that we have Epilepsia, 55(1):133–136, 2014 doi: 10.1111/epi.12475

136 S. A. Elwan et al. read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References 1. O’Brien TJ, So EL, Mullan BP, Hauser MF, Brinkmann BH, Bohnen NI, Hanson D, Cascino GD, Jack CR Jr, Sharbrough FW. Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology 1998;50:445–454. 2. Lee SK, Lee S-Y, Yun C-H, Lee H-Y, Lee J-S, Lee D-S. Ictal SPECT in neocortical epilepsies: clinical usefulness and factors affecting the pattern of hyperperfusion. Neuroradiology 2006;48:678–684. 3. Von Oertzen TH, Mormann F, Urbach H, Reichmann K, Koenig R, Clusmann H, Biersack HJ, Elger CE. Prospective use of subtraction ictal SPECT coregistered to MRI (SISCOM) in presurgical evaluation of epilepsy. Epilepsia 2011;52:2239–2248. 4. Desai A, Bekelis K, Thadani VM, Roberts DW, Jobst BC, Duhaime AC, Gilbert K, Darcey TM, Sudholme C, Siegel A. Interictal PET and ictal subtraction SPECT: sensitivity in the detection of seizure foci in patients with medically intractable epilepsy. Epilepsia 2013;54:341– 350. 5. Van Paesschen W. Ictal SPECT. Epilepsia 2004;45(Suppl. 4):35–40.

Epilepsia, 55(1):133–136, 2014 doi: 10.1111/epi.12475

6. Van Paesschen W, Dupont P, Van Heerden B, Vanbilloen H, Mesotten L, Maes A, Van Driel G, Mortelmans L. Self-injection ictal SPECT during partial seizures. Neurology 2000;54:1994–1997. 7. Shin WC, Hong SB, Tae WS, Kim SE. Ictal hyperperfusion patterns according to the progression of temporal lobe seizures. Neurology 2002;58:373–380. 8. Schneider F, Wang I, Alexopoulos AV, Almubarak S, Kakisaka Y, Jin K, Nair D, Mosher JC, Najm IM, Burgess RC. Magnetic source imaging and ictal SPECT in MRI-negative neocortical epilepsies: additional value and comparison with intracranial EEG. Epilepsia 2013;54:359–369. 9. Fukuda M, Masuda H, Honma J, Kameyama S, Taneka R. Ictal SPECT analyzed by three-dimensional stereotactic surface projection in frontal lobe epilepsy patients. Epilepsy Res 2006;68:95–102. 10. Cho J-W, Hong SB, Lee JH, et al. Contralateral hyperperfusion and ipsilateral hypoperfusion by ictal SPECT in patients with mesial temporal lobe epilepsy. Epilepsy Res 2010;88:247–254. 11. Chassagnon S, Namer IJ, Armspach JP, Nehlig A, Kahane P, Kehrli P, Valenti MP, Hirsch E. SPM analysis of ictal -interictal SPECT in mesial temporal lobe epilepsy: relationships between ictal semiology and perfusion changes. Epilepsy Res 2009;85:252–260. 12. Babb TL, Wilson CL, Isokawa-Akesson M. Firing patterns of human limbic neurons during stereoencephalography (SEEG) and clinical temporal lobe seizures. Electroencephalogr Clin Neurophysiol 1987;66:467–482.

Ictal single photon emission computed tomography in epileptic auras.

Little is known about whether ictal single photon emission computed tomography (SPECT) during an isolated aura can localize the epileptogenic zone (EZ...
184KB Sizes 0 Downloads 0 Views