Ann Nucl Med (2012) 26:397–404 DOI 10.1007/s12149-012-0585-4
ORIGINAL ARTICLE
SISCOM technique with a variable Z score improves detectability of focal cortical dysplasia: a comparative study with MRI Yukio Kimura • Noriko Sato • Kimiteru Ito • Kouhei Kamiya • Yasuhiro Nakata • Yuko Saito • Hiroshi Matsuda • Kenji Sugai • Masayuki Sasaki • Hideharu Sugimoto
Received: 7 December 2011 / Accepted: 13 February 2012 / Published online: 17 March 2012 Ó The Japanese Society of Nuclear Medicine 2012
Abstract Objective Focal cortical dysplasia (FCD) is one of the causes of epilepsy, but its diagnosis by MRI remains difficult. The purpose of this study was to evaluate the use of subtraction ictal SPECT coregistered to MRI (SISCOM) and MRI to detect the epileptogenic focus in patients with FCD. Methods MRI and SISCOM findings of 20 patients with pathologically proven FCD were retrospectively reviewed. MRI was visually assessed for detecting FCD. SISCOM was evaluated by a new method selecting a higher standard deviation (Z score) area as the epileptogenic focus. We scored the detectability in both SISCOM and MRI while referring to the pathology.
Results Sixteen patients agreed with pathology on SISCOM and 14 patients on MRI. Although MRI could not point out foci in two cases of FCD type I, SISCOM could do so in both of them. A combined diagnosis of SISCOM and MRI agreed with the pathology in 18 patients. Conclusions Narrowing the target by elevating the Z score on SISCOM seems to be an appropriate method to detect the foci without the need for expertise of radiologists. We recommend this combined method of SISCOM and MRI for presurgical evaluation in patients with FCD. Keywords Focal cortical dysplasia SISCOM Ictal SPECT Epilepsy
Introduction Y. Kimura N. Sato (&) K. Ito K. Kamiya Y. Nakata Department of Radiology, National Center of Neurology and Psychiatry, 4-1-1 Ogawahigashi-cyo, Kodaira, Tokyo 187-8511, Japan e-mail:
[email protected] Y. Saito Department of Pathology and Laboratory Medicine, National Center of Neurology and Psychiatry, Kodaira, Tokyo, Japan H. Matsuda Department of Nuclear Medicine, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan K. Sugai M. Sasaki Department of Child Neurology, National Center of Neurology and Psychiatry, Kodaira, Tokyo, Japan H. Sugimoto Department of Radiology, Jichi Medical University, Shimotsuke, Tochigi, Japan
Focal cortical dysplasia (FCD) is a localized cerebral cortical malformation that causes epilepsy in both children and adults. To identify the epileptogenic focus is critical for planning surgery. The histopathologic features of FCD range from mild cortical dyslamination to more severe forms [1]. Although MRI can show the signal or morphological abnormalities of FCD, it is sometimes difficult to detect them when the signals are mild or the abnormalities are slight, even for a well-trained neuroradiologist, much less by a non-specialist. Single photon emission computed tomography (SPECT) is a noninvasive, functional neuroimaging method for the determination of epileptogenic foci. SPECT images usually demonstrate hypoperfusion interictally and hyperperfusion ictally, and they are traditionally analyzed visually [2]. Subtraction ictal SPECT coregistered to MRI (SISCOM) is a recently developed neuroimaging method that can be used to measure local differences in cerebral blood flow
123
398
Ann Nucl Med (2012) 26:397–404
caused by changed neuronal activity between the interictal and ictal states [3]. The accuracy of SISCOM in the localization of the seizure focus has been assessed by several previous studies [4–9]. The SISCOM technique makes it easier to find increased perfusion areas more than 2 standard deviations (Z scores) from the mean; however, if there are multiple increased perfusion areas more than 2 Z scores, detecting the epileptogenic focus is difficult. We thought that if we could regulate the Z scores, it would be possible to narrow the target. There are many previous studies about diagnosing focal epilepsy using SISCOM [4–9], but no original studies focusing on FCD have been reported. The purpose of this study was to evaluate SISCOM and MRI for their ability to detect the epileptogenic focus in patients with FCD.
spin echo T2-weighted images with TR/TE/FA/NEX of 3,800 ms/95 ms/150°/1 or 5,000 ms/81 ms/140°/1, 5.0 mm thickness with 1.8 mm gap or 3.0 mm thickness with 1.2 mm gap, 20 or 35 slices, 291 9 512 or 348 9 512 matrix, 25 cm FOV; (c) transverse and coronal fluidattenuated inversion recovery (FLAIR) images with TR/TE/TI/FA/NEX of 9,000 ms/101 ms/2,500 ms/170°/1 or 12,000 ms/94 ms/2,700 ms/150°/1, 5.0 mm thickness with 1.8 mm gap or 3.0 mm thickness with 1.2 mm gap, 20 or 35–42 slices, 179 9 256 or 202 9 320 matrix, 25 cm FOV; and (d) coronal turbo short-tau inversion recovery (STIR) images with TR/TE/TI/FA/NEX of 4,200 ms/81 ms/ 180 ms/180°/1 or 5,000 ms/10 ms/230 ms/150°/1, 5.0 mm thickness with 1.0 mm gap or 3.0 mm thickness with 0.9 mm gap, 20 or 42 slices, 224 9 512 or 284 9 448 matrix, 25 cm FOV.
Materials and methods
SPECT
Patients
All patients underwent ictal and interictal SPECT scans with 99mTc ethyl cysteinate dimer (99mTc-ECD). All ictal and interictal SPECT examinations were performed within 2 weeks before or after MRI examination. On ictal SPECT examination, 99mTc-ECD was immediately injected by a specialized pediatrician team after confirmed seizure onset based on the clinical symptoms and/or EEG results. Interictal SPECT scans were performed when the patients had been seizure free over 24 h. The tracer of ECD was injected at a maximum dose of 600 MBq for adults, and patients younger than 18 years received age-adjusted doses. SPECT scans were acquired by a two-head rotating gamma camera system (Siemens E-CAM), with a low-energy high-resolution collimator, photo peak centered on 140 keV and acceptance window of 20 %, 30 projections per head over 180° on a 128 9 128 matrix, acquisition time of 20 min. Projections were filtered with a Butterworth filter (order 8, cutoff frequency 0.5) and reconstructed by the filtered backprojection method in transaxial slices parallel to the orbitomeatal line and parallel to the long axis of the temporal lobe, from which coronal sections were produced. For attenuation correction we used the Chang method.
We retrospectively reviewed the MRI and SISCOM findings in 20 patients with FCD that had been pathologically proven. Our local ethics committee did not require committee approval or patient informed consent for this retrospective review. A review of cases from April 2005 to March 2011 revealed 38 patients who were pathologically diagnosed as FCD in our institution. Eighteen patients who did not have both ictal and interictal SPECT before surgery were excluded from the study, although all 38 patients had received MR examinations. As a consequence, 20 patients with FCD were enrolled. Nine were males and 11 were females, from 3 months to 27 years of age, with a mean age of 6.17 ± 7.09 years. Surgical sites were determined from the results of comprehensive analysis of clinical findings, scalp electroencephalography (EEG), intracranial EEG, MRI, SPECT, and positron emission tomography (PET). All cases were pathologically diagnosed with FCD. MR examinations Brain MR examinations were performed with 1.5 Tesla scanners (Siemens Magnetom Symphony) (patients 1–17) and 3.0 Tesla scanners (Siemens Magnetom Verio) (patients 18–20), according to the following protocols specifically designed for epilepsy studies: (a) high-resolution threedimensional (3D) sagittal T1-weighted magnetization prepared rapid acquisition with gradient echo sequence (MPRAGE), with repetition time (TR)/echo time (TE)/ flip angle(FA)/number of excitations (NEX) of 1600 ms/ 2.64 ms/15°/1 or 1800 ms/2.26 ms/9°/1, 1.23 mm or 0.80 mm thickness with no gap, 144 or 208 slices, 256 9 256 or 388 9 320 matrix, 26 cm FOV; (b) transverse turbo
123
SISCOM procedure Ictal and interictal SPECT and T1-weighted 3D MR images were transferred to the same computer. Each patient’s ictal and interictal SPECT scans were registered to the same patient’s MRI scan, using an automated image registration (AIR) program [10]. The AIR software was used to align the SPECT to the MRI scans of each subject using a 6-parameter rigid-body transformation. Before coregistration of SPECT and MRI, the outer scalp was removed from MRI by applying a binary mask for the whole brain to
Ann Nucl Med (2012) 26:397–404
the MRI in the manner described previously [11]. The coregistered ictal and interictal perfusion SPECT images were normalized according to the global mean voxel counts. Normalized ictal and interictal SPECT images were subtracted to obtain ictal–interictal difference images. Binary mask was generated at threshold of 35 % of the maximum value for each SPECT image. Final mask was generated from filling inner holes of implicit intersection of two binary mask images. Mean and standard deviations of subtracted image of normalized ictal and interictal SPECT images were calculated. In the routine examinations, 2 standard deviations were selected as the Z score in the subtraction image to be regarded as significantly increased perfusion areas during the seizure and superimposed on the tomographic and surface rendering images of the patient’s MRI. In this study, we changed Z scores 0.1 by 0.1 to narrow the epileptogenic foci. MR imaging review Two neuroradiologists (Y.N. and N.S. with 15 and 20 years of experience with MR imaging, respectively) did not know the patients’ clinical information except for having independently evaluated the MR imaging findings of the patients with FCD. Following the procedure described by Matsuda et al. [4], the cerebrum was divided into 18 sites, that is, the either right or left frontal, temporal, parietal, occipital, frontotemporal, frontoparietal, occipitoparietal, occipitotemporal, or temporoparietal sites. They chose only one lesion, where the FCD seemed to be located, out of 18 sites. The abnormal MRI findings were classified as follows: (1) Definite: abnormal lesion obviously detected; (2) Probable: abnormal lesion was suspected; (3) Uncertain: no abnormal lesion was detected. These findings were scored as 2, 1, and 0, respectively. If epileptogenic focus pointed out by MRI agreed with the pathologically proven site, its score was reflected in the results. On the other hand, if the two did not agree, the score was set at 0. In cases with FCD spreading over two lobes, agreement with pathology occurred if both or either one of them were pointed out. When reviewers found different results, they discussed the case and came to a decision by consensus. There were no other epileptogenic diseases besides FCD, such as hippocampal sclerosis or neoplasms, in any of the study patients. SISCOM image analysis By changing the Z score, we were able to select the highest and the second-highest Z score areas from the 18 areas. We assigned a score of 2 at the area of the highest Z score and 1 at the area of the second-highest Z score (Fig. 1). If more than two sites had the same highest Z score, we assigned a score 2 to the largest area and 1 to the second-largest area.
399
If epileptogenic focus pointed out by SISCOM agreed with the pathologically proven site, its score was reflected in the results. On the other hand, if it did not agree, its score was set to 0. In cases with FCD spreading over two lobes the results would be determined to agree with pathology when both or either one of them were pointed out. Deep gray matter and infratentorial regions were excluded from selection of the top two increased hyperperfusion sites, because generally they did not have epileptogenicity. Pathology All patients were subjected to surgery, and the sites were histopathologically confirmed as FCD. The histopathology of FCD was classified using Palmini’s classification, as follows: IA, architectural abnormality only; IB, architectural abnormality plus giant or immature neurons; IIA, architectural abnormality with dysmorphic neurons without balloon cells; IIB, architectural abnormality with dysmorphic neurons and balloon cells [1]. Surgical outcome Surgical outcome was classified into four groups according to Engel’s classification [12]. Class I means seizure free or only nondisabling auras; class II, almost seizure free or [75 % seizure reduction; class III, 50–75 % seizure reduction; and class IV, \50 % seizure reduction or no improvement. Patients who achieved outcome classes I or II were considered to have shown good improvement from epilepsy surgery. We evaluated patients who had at least 12 months of follow-up. Therefore, 6 patients whose follow-up period was less than 12 months were not classified.
Results Table 1 shows the clinical findings of 20 FCD patients. Histopathological findings were 11 patients (55 %) with FCD type IIA, five patients (25 %) with FCD type IIB, one patient (5 %) with FCD type IA, and one patient (5 %) with FCD type IB (Table 1). Two patients (10 %) were non-classifiable. Table 2 shows the results of MRI and SISCOM evaluations. On SISCOM, the number of patients whose score was 2 and agreed with pathology was fourteen, the number whose score was 1 and agreed with pathology was two, and the number whose score was 0 (without agreement) was four. On MRI, the number of patients whose score was 2 and agreed with pathology was eight, the number whose score was 1 and agreed with pathology was six, and the number whose score was 0 (without agreement) was six. All sites pointed out by MRI were diagnosed with FCD by pathological study. The total
123
400
Ann Nucl Med (2012) 26:397–404
Fig. 1 Narrowing the target by elevating Z score on SISCOM. a Z score 2.0. Multiple increased hyperperfusion areas are noted in the left frontal, temporal, and parietal lobes. b Z score 2.5. The number of hyper perfusion areas is decreased to two points, in the left frontal and temporal lobes. c Z score 2.8. Only the left frontal lobe remains as a
hyper perfusion area. The result of SISCOM evaluation of this case is that the left frontal lobe is the highest Z score area (SISCOM score 2) and the left temporal lobe is the second-highest Z score area (SISCOM score 1)
whose score agreed with pathology in the case of SISCOM was 30, which was higher than the case of MRI at 22. If we set a cutoff value at 2 for the score to agree with pathology, the number of cases was 14 for SISCOM and 8 for MRI. If the results of both SISCOM and MRI were summed up, eighteen cases showed correlation. Six patients scored 2 for
both MRI and SISCOM (Fig. 2). SISCOM detected foci in four patients (including two all FCD type I) in whom MRI could not detect foci (Fig. 3). On the other hand, MRI pointed out foci in three patients in whom SISCOM could not detect foci. In two patients, neither MRI nor SISCOM could show foci.
123
Ann Nucl Med (2012) 26:397–404
401
Table 1 Clinical findings of 20 focal cortical dysplasia (FCD) patients Patient no.
Age
Sex
Surgical site
Pathology (type of FCD)
Outcome (Engel’s classification)
Follow-up (months)
1
3m
M
lt. frontal
IIA
–
4
2
1y
M
rt. frontal
IIB
I
12
3
3m
F
lt. parieto-temporal
IIA
–
5
4
2y
F
lt. parieto-occipital
IIA
III
46
5
8m
F
lt. occipital
IIA
I
48
6
10 y
F
rt. parietal
IIB
II
16
7 8
20 y 12 y
F F
lt. frontal lt. frontal
IIA IIA
I II
32 71
9
3m
M
rt. parieto-temporal
IIA
–
5
10
2y
F
rt. frontal
IIB
–
1
11
6y
F
rt. frontal
IIA
I
12
12
6y
M
lt. frontal
IB
I
59
13
8y
F
lt. parieto-frontal
IIA
III
12
14
8y
M
lt. frontal
IA
–
11
15
27 y
M
lt. occipital
Not classified
III
66
16
8y
M
lt. frontal
IIB
–
9
17
2y
F
lt. parietal
IIA
I
61
18
11 m
M
rt. frontal
Not classified
I
42
19
1y
M
lt. temporal
IIB
III
32
20
8y
F
rt. temporal
IIA
I
12
Outcome is listed only for patients who had at least 12 months of follow-up Age: m months, y years
Surgical outcome The duration of postsurgical follow-up was from 1 to 71 months, with a mean age of 27.8 ± 23.5 months. Fourteen patients (70 %) had at least 12 months of followup; of these, 8 (57 %) had a seizure-free outcome (Engel class I), 2 (14 %) had a favorable outcome (Engel class II), and 4 (28 %) had a nonfavorable outcome (Engel class III).
Discussion To the best of our knowledge, this is the first study to compare SISCOM and MRI in FCD patients. We demonstrated a new method to point out epileptogenic foci by searching for the highest Z score area on SISCOM, and we were able to show better concordance of pathology than MRI. This SISCOM method leads to detecting the pathological area without the need for the expertise of radiologists. Furthermore, diagnosis using both SISCOM and MRI resulted in higher detectability of foci, up to 90 %. FCD was first described in detail by Taylor et al. [13]. They reported 10 patients with drug-resistant epilepsy who underwent surgical resection. Microscopic examination revealed a peculiar histopathology including cortical
disorganization, large bizarre neurons, and, in half of the patients, balloon cells. Since then, the term ‘‘FCD’’ has been widely used for a large spectrum of lesions comprising cortical dyslamination, cytoarchitectural lesions, and underlying abnormalities of white matter [1]. With ongoing advances in presurgical neuroimaging techniques such as high-resolution MRI, more subtle cortical abnormalities can be identified as potential epileptic foci. However, diagnosis by visual inspection of MRI can be difficult, and subtle dysplastic lesions often remain unrecognized. MRI findings of FCD vary from obvious to unclear, and the interpretation of MRI is sometimes difficult, especially for general radiologists. Colombo et al. [14] reported that they were unable to detect any MR signs of FCD in 39 % of patients. In this study, 30 % of patients had uncertain findings of MRI by our reviewers. The visual comparison between ictal and interictal SPECT studies is particularly difficult because of variability in the intensity of the ictal and interictal images. SISCOM has been reported to reliably display the epileptogenic focus in patients with localization-related epilepsy. SISCOM may demonstrate a localized region of cerebral hyperperfusion in approximately 80 % of patients with intractable partial epilepsy [15]. The accuracy of SISCOM in the localization of the seizure focus has been assessed in
123
402
Ann Nucl Med (2012) 26:397–404
Table 2 The results of MRI and SISCOM evaluations in 20 patients with focal cortical dysplasia Patient no.
Surgical site
SISCOM The site of the highest Z score (score 2)
MRI The site of the second-highest Z score (score 1)
Score agreed with pathology
The site judged as definite (score 2)
MRI ? SISCOM The site judged as probable (score 1)
Score agreed with pathology
Score agreed with pathology
1
lt. frontal
lt. frontal
lt. temporal
2
lt. frontal
–
2
4
2
rt. frontal
rt. frontal
lt. occipital
2
rt. frontal
–
2
4
3
lt. parieto-temporal
lt. temporal
lt. frontal
2
lt. parietotemporal
–
2
4
4
lt. parieto-occipital
lt. parietal
lt. occipital
2
lt. parietooccipital
–
2
4
5
lt. occipital
lt. occipital
lt. parietal
2
lt. occipital
–
2
4
6
rt. parietal
rt. parietal
lt. parietal
2
rt. parietal
–
2
4
7
lt. frontal
lt. frontal
rt. occipital
2
–
lt. frontal
1
3
8
lt. frontal
lt. frontal
rt. frontal
2
–
lt. frontal
1
3
9
rt. parieto-temporal
rt. temporal
rt. parietal
2
–
rt. parietotemporal
1
3
10
rt. frontal
rt. frontal
rt. parietal
2
–
rt. frontal
1
3
11
rt. frontal
rt. frontal
lt. parietal
2
–
–
0
2
12
lt. frontal
lt. frontal
lt. temporal
2
–
–
0
2
13 14
lt. parieto-frontal lt. frontal
lt. frontal lt. frontal
lt. temporal lt. parietal
2 2
– –
– –
0 0
2 2
–
lt. occipital
1
2
lt. frontal
1
2
15
lt. occipital
rt. occipital
lt. occipital
1
16
lt. frontal
lt. parietal
lt. frontal
1
17
lt. parietal
lt. temporal
lt. occipital
0
lt. parietal
–
2
2
18
rt. frontal
lt. occipital
lt. frontal
0
rt. frontal
–
2
2
19
lt. temporal
rt. frontal
rt. occipital
0
–
–
0
0
20
rt. temporal
rt. frontal
rt. occipital
0
–
–
0
0
Italicized entries agree with pathologically proven site
several previous studies [4–9], but when there are multiple areas with a higher than fixed Z score of 2, it is difficult to interpret them. Our SISCOM method was easily detectable and highly reproducible for epileptogenic foci by selecting the highest and largest Z score area. The diagnosis of FCD by MRI relies on the detailed analysis of several features, including cortical thickness, blurring of the grey/white matter junction, grey matter signal changes, and focal and/or lobar hypoplasia/atrophy. Before myelination, the hypointensity on T2-weighted images of FCD is usually obvious in contrast to the hyperintensity of the unmyelinated white matter, as shown in Fig. 2. It is usually more difficult to point out abnormal regions on MRI in FCD type I than type II [14]. In this study, there were two cases of FCD type I (patients 11 and 13). Abnormal lesions on MRI were not pointed out in either of them, but we were able to point out epileptogenic foci on SISCOM, as shown in Fig. 3. SISCOM may be useful, especially in FCD type I, which is usually invisible on MRI.
123
There was a report of a study that included a large number of epilepsy patients in which Matsuda et al. [4] reviewed SISCOM evaluations of 123 patients with intractable epilepsy who had undergone epilepsy surgery. However, their diseases were various and the etiology was unclear in some patients. For example, the break-down of patients was to malformation of cortical development (MCD), hippocampal sclerosis, tumors, and cavernous hemangiomas. FCD was considered to be contained in MCD, but the details were not shown. O’Brien et al. [3] reviewed SISCOM evaluations of 22 patients with focal MCDs, including many varieties of diseases such as FCD, polymicrogyria, hippocampal sclerosis, and tuberous sclerosis. The present work is the first SISCOM study focusing on FCD. SISCOM evaluation showed better concordance of the epileptogenic focus than did MRI. We were able to show the superiority of SISCOM over MRI. Of course, diagnosis using both SISCOM and MRI was better than diagnosis using only SISCOM or MRI. Similar to the authors of a
Ann Nucl Med (2012) 26:397–404
Fig. 2 T2-weighted MR image and SISCOM of a patient with FCD (patient 5). a T2-weighted image shows a hypointensity lesion in the left occipital lobe (white arrow). The abnormal region is obvious, and FCD is highly suspected (MRI score 2). Before myelination, T2weighted images show FCD as a hypointensity area, which is in good contrast to the high-intensity unmyelinated white matter. b SISCOM shows the highest Z score area (Z score 2.0) only in the left occipital lobe (SISCOM score 2). The left occipital region was pathologically diagnosed as FCD type IIA
previous study [9], we think that the reading of MRI guided by SISCOM can help to identify subtle lesions in clinical practice. The results of the present study demonstrate that the SISCOM technique provides valuable information for localizing the epileptogenic zone in medically refractory epilepsy in FCD, even when lesions are not apparent on MRI. Compared with the interpretations of MRI that require a high level of knowledge and experience, this new SISCOM method is easy and highly reproducible. This study had several limitations. The first was that not all patients were examined by 3-Tesla MRI. Only 3 of 20 patients were examined by 3-Tesla MRI, and the remaining 17 patients were examined by 1.5-Tesla MRI. There was a report that 3-Tesla MRI was superior to 1.5-Tesla MRI in the detection and accurate characterization of structural brain abnormalities in patients with epilepsy [16]. If we
403
Fig. 3 T2-weighted MR image and SISCOM of a patient with FCD (patient 14). a The T2-weighted image does not detect an abnormal lesion, and neither do other sequences of MRI (MRI score 0). The patient underwent corpus callosotomy 2 years ago. b SISCOM shows the highest Z score area (Z score 2.8) only in the left frontal lobe (SISCOM score 2). The left frontal lesion was pathologically diagnosed as FCD type IA
had examined all patients with 3-Tesla MRI, the MRI scores might have been better. The second limitation was that we had a limited number of patients. In particular, there were only two cases of FCD type I, which is usually difficult to point out by MRI. If there had been a larger number of patients, we might have been able to evaluate the difference between FCD type I and type II and show the obvious superiority of SISCOM over MRI in FCD type I. In conclusion, we retrospectively reviewed the MRI and SISCOM findings in patients with FCD. By selecting higher Z score areas, we were able to show the superiority of SISCOM over MRI. SISCOM may be more useful than MRI for diagnosing FCD type I, which is more difficult to observe via MRI. We conclude that it is better to use this new method of SISCOM together with MRI in presurgical evaluations of FCD.
123
404 Conflict of interest
Ann Nucl Med (2012) 26:397–404 We declare that we have no conflict of interest.
References 1. Palmini A, Najm I, Avanzini G, Babb T, Guerrini R, FoldvarySchaefer N, et al. Terminology and classification of the cortical dysplasias. Neurology. 2004;62:S2–8. 2. Andreasen AR, Lassen NA. Single-photon emission computed tomography in temporal lobe epilepsy. In: Dam M, Gran L, editors. Comprehensive epileptology. New York: Raven Press; 1991. p. 375–84. 3. O’Brien TJ, So EL, Mullan BP, Hauser MF, Brinkmann BH, Bohnen NI, et al. Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology. 1998;50:445–54. 4. Matsuda H, Matsuda K, Nakamura F, Kameyama S, Masuda H, Otsuki T, et al. Contribution of subtraction ictal SPECT coregistered to MRI to epilepsy surgery: a multicenter study. Ann Nucl Med. 2009;23:283–91. 5. Wichert-Ana L, Mazzoncini P, Oliverira LF, Fernandes RMF, Velaso TR, Santos AC, et al. Ictal technetium-99m ethyl cysteinate dimer single-photon emission tomographic findings in epileptic patients with polymicrogyria syndromes: a subtraction of ictal-interictal SPECT coregistered to MRI study. Eur J Nucl Med Mol Imaging. 2008;35:1159–70. 6. Valenti MP, Froelich S, Armspach JP, Chenard MP, Dietemann JL, Kerhli P, et al. Contribution of SISCOM imaging in the presurgical evaluation of temporal lobe epilepsy related to dysembryoplastic neuroepithelial tumors. Epilepsia. 2002;43:270–6. 7. O’Brien TJ, So EL, Cascino GD, Hauser MF, Marsh WR, Meyer FB, et al. Subtraction SPECT coregistered to MRI in focal
123
8.
9.
10.
11.
12.
13.
14.
15. 16.
malformations of cortical development: localization of the epileptogenic zone in epilepsy surgery candidates. Epilepsia. 2004;45:367–76. Kaiboriboon K, Lowe VJ, Chantarujikapong SI, Hogan RE. The usefulness of subtraction ictal SPECT coregistered to MRI in single- and dual-headed SPECT cameras in partial epilepsy. Epilepsia. 2002;43:408–14. Ahnlide JA, Rose´n I, Tech PLM, Ka¨lle´n K. Dose SISCOM contribute to favorable seizure outcome after epilepsy surgery? Epilepsia. 2007;48:579–88. Woods RP, Cherry SR, Mazziotta JC. Rapid automated algorithm for aligning and reslicing PET images. J Comput Assist Tomogr. 1992;16:620–33. Matsuda H, Ohnishi T, Asada T, Li ZJ, Kanetaka H, Imabayashi E, et al. Correction for partial-volume effects on brain perfusion SPECT in healthy men. J Nucl Med. 2003;44:1243–52. Engel J Jr, Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In: Engel Jr J, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1993. p. 609–21. Taylor DC, Falconer MA, Burton CJ, Corsellis JAN. Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiatry. 1971;34:369–87. ¨ zkara C, Barkovich AJ. Colombo N, Salamon N, Raybaud C, O Imaging of malformations of cortical development. Epileptic Disord. 2009;11:194–205. Van Paesschen W. Ictal SPECT. Epilepsia. 2004;45:35–40. Phal PM, Usmanov A, Nesbit GM, Anderson JC, Spencer D, Wang P, et al. Qualitative comparison of 3-T and 1.5-T MRI in the evaluation of epilepsy. Am J Roentgenol. 2008;191:890–5.
