- Email: [email protected]
Ictal SPECT in the definition of the seizure onset zone
Presurgical Assessment of the Epilepsies with Clinical Neurophysiology and Functional Imaging Handbook of Clinical Neurophysiology, Vol. 3 Felix Rosenow and Hans O. Lüders (Eds.) © 2004 Elsevier B.V. All rights reserved
147
CHAPTER 2.10
Ictal SPECT in the definition of the seizure onset zone Gregory D. Cascinoa,∗ , Jeffrey R. Buchhaltera , Brian P. Mullanb and Elson L. Soa b
a Department of Neurology, Mayo Clinic, 200 First St., SW, Rochester, MN 55905, USA Department of Diagnostic Radiology (Nuclear Medicine), Mayo Clinic, Rochester, MN 55905, USA
1. Introduction Partial or localization-related epilepsy is characterized by recurrent and unprovoked focal seizure activity and is the most common seizure disorder (Dreifuss, 1987; Mattson, 1992; Cascino, 1996a). Over 90% of the incident cases of epilepsy in adults have focal seizures (Dreifuss, 1987). Individuals with partial epilepsy may experience simple partial seizures, complex partial seizures, or secondarily generalized tonic–clonic seizures (Mattson, 1992). The most epileptogenic region in patients with focal seizures is the amygdalohippocampal complex, associated with mesial temporal lobe epilepsy (Cascino, 1996a). Approximately 30–40% of newly diagnosed patients with epilepsy will experience medically refractory seizures that are physically and socially disabling (Dreifuss, 1987; Hauser, 1992; Mattson, 1992; Camfield and Camfield, 1996). Less than 10% of patients who have focal seizures refractory to their initial antiepileptic drug (AED) therapy will be rendered seizure-free with additional medication trials (Hauser and Hesdorffer, 1990; Hauser, 1992; Mattson, 1992; Camfield and Camfield, 1996). Surgical treatment is an effective therapy for highly selected patients with localizationrelated epilepsy (Crandall, 1987; Dreifuss, 1987; Awad et al., 1991; Cascino et al., 1993b; Engel and Ojemann, 1993; Cascino, 1996; Radhakrishnan et al., 1998; Wiebe et al., 2001). The goals of surgical treatment are to render the individual seizure-free and allow the patient to become a participating and productive member of society (Dreifuss, 1987; Cascino, 1996; Radhakrishnan et al., 1998). Surgery is most often performed for
∗
Correspondence to: Dr. Gregory D. Cascino. E-mail address: [email protected] Tel.: +1-507-284-2511; fax: +1-507-284-8686.
individuals with mesial temporal lobe epilepsy or focal seizures related to foreign-tissue lesional pathology (Awad et al., 1991; Cascino et al., 1993b; Wiebe et al., 2001). The surgically excised hippocampus in patients with mesial temporal lobe epilepsy usually reveals focal cell loss and an increase in astrogliosis, i.e. hippocampal sclerosis (Cascino et al., 1992, 1993a, 1996b; Cambier et al., 2001). The surgically excised pathology in approximately 65% of patients with intractable partial epilepsy is hippocampal sclerosis (Hauser, 1992; Wiebe et al., 2001). Patients with a chronic seizure disorder and a lesional epileptic syndrome may have a low-grade glioma, ganglioglioma, cavernous hemangioma, or focal cortical dysplasia (Awad et al., 1991; Palmini et al., 1991; Cascino et al., 1992, 1993a,b; Mosewich et al., 2000). A structural lesion underlying the epileptogenic zone is identified in approximately 30% of patients undergoing epilepsy surgery (Cascino et al., 1993b). Mesial temporal lobe epilepsy and focal seizures related to a lesional pathology are considered surgically remediable epileptic syndromes because of the significant reduction in seizure tendency following a focal cortical resection and excision of the pathological findings underlying the epileptogenic zone (Crandall, 1987; Awad et al., 1991; Berkovic et al., 1991; Cascino et al., 1992, 1993a,b, 1996; Engel and Ojemann, 1993; Cascino, 1996; Radhakrishnan et al., 1998; Cambier et al., 2001; Wiebe et al., 2001). Individuals with hippocampal sclerosis, lesional pathology, and other “macroscopic” structural abnormalities, e.g. focal encephalomalacia, usually have an abnormal MRI study (Palmini et al., 1991; Cascino et al., 1992, 1993b; Cascino, 1996, 2001). These patients are classified as having substrate-directed partial epilepsy. The MRI in these individuals is often a reliable indicator of the focal pathology and may suggest the region of seizure onset (Cascino, 2001). A comprehensive evaluation including ictal
148
video-EEG monitoring is required, however, because the MRI-identified structural abnormalities may not be contiguous or may be remote from the epileptogenic zone (Awad et al., 1991). There is a broad consensus that a MRI study using the “temporal lobe seizure protocol” has a high diagnostic yield in patients with mesial temporal lobe epilepsy related to hippocampal sclerosis (Berkovic et al., 1991; Cascino et al., 1992; Cambier et al., 2001). MRI shows hippocampal formation atrophy and an associated signal intensity alteration in over 90% of these individuals (Berkovic et al., 1991; Cascino et al., 1992; Cambier et al., 2001). The hippocampal atrophy is related to focal cell loss, while the signal intensity change may be associated with gliosis (Berkovic et al., 1991; Cascino et al., 1992; Cambier et al., 2001). MRI almost invariably identifies a foreign-tissue lesion in patients with tumors and vascular malformations (Cascino et al., 1993b; Engel and Ojemann, 1993). The most common primary brain tumors associated with intractable partial epilepsy include gangliogliomas, dysembryoplastic neuroepithelial tumors, and the low-grade glial neoplasms (Palmini et al., 1991; Cascino et al., 1992; Cascino, 2001). The sensitivity and specificity of MRI in patients with malformations of cortical development depend on several factors, including the histopathology and anatomical localization of the lesion (Palmini et al., 1991). Approximately 75% of patients with focal cortical dysplasia will have a localizing structural neuroimaging alteration (Palmini et al., 1991; Cascino, 2001). MRI significantly alters the preoperative evaluation and operative strategy in patients with hippocampal sclerosis and lesional pathology (Berkovic et al., 1991; Cascino et al., 1992, 1993b, 1996; Palmini et al., 1991; Cambier et al., 2001; Wiebe et al., 2001). The rationale for the presurgical evaluation is to identify the site of ictal onset and initial seizure propagation, i.e. epileptogenic zone, and determine the likely pathological findings underlying the epileptic brain tissue (Engel and Ojemann, 1993; Cascino et al., 1996). The demonstration of concordance between the MRI-identified pathological findings and the epileptic brain tissue indicates a highly favorable operative outcome. Approximately 80% of patients with unilateral hippocampal sclerosis, a low-grade primary brain neoplasm, or a cavernous hemangioma are rendered seizure-free following surgical treatment (Awad et al., 1991; Cascino et al., 1992, 1993a, 1996; Cascino, 1996; Mosewich et al., 2000; Cambier et al., 2001; Wiebe et al., 2001). Over 90%
G.D. CASCINO ET AL.
of these patients will experience an excellent surgical outcome, i.e. auras only or rare nondisabling seizures (Radhakrishnan et al., 1998; Cambier et al., 2001). The efficacy of surgical treatment is reduced in patients with malformations of cortical development (Engel and Ojemann, 1993). Nearly half of the individuals, however, experience a worthwhile reduction in seizures following surgical treatment (Palmini et al., 1991). 2. Localization-related epilepsy with a normal MRI Patients with intractable localization-related seizure disorders and normal MRI studies, i.e. nonsubstratedirected partial epilepsy, may be considered for surgical treatment (O’Brien et al., 1996; Mosewich et al., 2000). Patients with normal MRI studies may have mesial temporal lobe epilepsy or extrahippocampal, neocortical, seizures (Jack et al., 1995; O’Brien et al., 1996; Mosewich et al., 2000). The diagnosis of mesial temporal lobe epilepsy in this group may be suggested by ictal semiology, interictal epileptiform discharges, or ictal EEG pattern (Radhakrishnan et al., 1998). The interictal PET study usually shows regional hypometabolism in patients with mesial temporal lobe epilepsy (Henry et al., 1994; Ho et al., 1995; Theodore et al., 1996). Approximately 50% of patients with normal MRI studies who undergo surgery for mesial temporal lobe epilepsy are rendered seizure-free (Jack et al., 1995). Most commonly, patients with focal seizures and normal MRI studies have extrahippocampal, neocortical, seizures of extratemporal origin (Cascino et al., 1992; Mosewich et al., 2000). The pathological findings underlying the epileptogenic zone in patients with a normal MRI and localization-related epilepsy include astrogliosis, focal cell loss, no histopathological alteration, or a focal cortical dysplasia (Cascino et al., 1992; Mosewich et al., 2000; Cascino, 2001). Unfortunately, patients with extratemporal seizures and a normal MRI study are less favorable candidates for surgical treatment (Cascino et al., 1992; Mosewich et al., 2000). An estimated 20–30% of these patients with extratemporal, mainly frontal lobe, seizures will enter a seizure remission following a focal cortical resection (Cascino et al., 1992). It may be difficult to localize or lateralize the epileptogenic zone in patients with neocortical seizures and a normal MRI. The epileptic brain tissue in patients with extrahippocampal, neocortical, seizures may represent a continuum that results in a subtotal resection of
ICTAL SPECT IN THE DEFINITION OF THE SEIZURE ONSET ZONE
the epileptogenic zone. The potential limitations of interictal and ictal extracranial and intracranial EEG monitoring in patients with partial seizures of extratemporal origin have been well defined (Cascino et al., 1992). There is a strong need for additional neurodiagnostic studies that can be carried out prior to implantation of intracranial electrodes to localize the epileptogenic zone in patients with extrahippocampal, neocortical, seizures and a normal MRI study. Interictal PET (2-fluorodeoxyglucose) has a low diagnostic yield in these patients (Theodore, 1996). Peri-ictal functional neuroimaging techniques have the capability to assist in localizing the epileptic brain tissue, i.e. identify the seizure-onset zone, in patients with intractable partial epilepsy (Marks et al., 1992; Henry, 1996; Newton and Berkovic, 1996; O’Brien et al., 1998a,b, 1999a,b; Brinkmann et al., 2000; So, 2000; So et al., 2000). The rationale for these diagnostic studies includes the selection of operative candidates. The results of peri-ictal functional neuroimaging may modify the preoperative evaluation and alter the surgical excision (So, 2000). Unfortunately, PET does not lend itself to ictal imaging in the epilepsy monitoring unit (Henry, 1996; Theodore, 1996). SPECT has the potential for peri-ictal imaging studies in patients undergoing video-EEG monitoring prior to surgical treatment (Marks et al., 1992; Newton and Berkovic, 1996; O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). 3. Single photon emission computed tomography SPECT is most appropriate for ictal imaging in patients with localization-related epilepsy being considered for epilepsy surgery to identify the seizure-onset zone (Marks et al., 1992; Newton and Berkovic, 1996; O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). Ictal SPECT studies are superior to interictal images in localization-related epilepsy (O’Brien et al., 1998a,b). SPECT involves blood-flow imaging using radiopharmaceuticals, e.g. 99m Tcbicisate, that have a rapid first pass brain extraction with maximum uptake being achieved within 30–60 s of an intravenous injection (O’Brien et al., 1998a,b, 1999b; Brinkmann et al., 2000). The SPECT images can be acquired up to 4 h after the termination of the seizure so that the individual patient can recover from the seizure episode prior to being transported to the
149
nuclear medicine laboratory. SPECT studies have an important clinical application in the potential identification of the epileptic brain tissue when the presurgical evaluation is indeterminate as to the localization of the ictal-onset zone (O’Brien et al., 1998a). This includes patients with nonsubstrate-directed partial epilepsy, widespread unilateral structural abnormalities, and multilobar lesions (O’Brien et al., 1999a). The initial blood-flow SPECT studies in patients with intractable partial epilepsy involved interictal imaging which variably detected a focal hypoperfusion in the region of the epileptogenic zone (Marks et al., 1992; Newton and Berkovic, 1996). Interictal SPECT images have proven to have a relatively low sensitivity and a relatively high false positive rate in temporal lobe epilepsy (Newton and Berkovic, 1996). Interictal SPECT has also been shown to have a low diagnostic yield in patients with extratemporal seizures. Ictal SPECT studies have been confirmed to be useful in patients with temporal lobe epilepsy to identify a region of focal hyperperfusion (Marks et al., 1992; Newton and Berkovic, 1996). The rationale for interictal SPECT imaging at present is to serve as a reference for a baseline study for the interpretation of ictal SPECT images. The diagnostic yield of ictal SPECT has been established to be superior to interictal SPECT in patients being considered for a focal cortical resection (Marks et al., 1992; Newton and Berkovic, 1996; O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). The recent development of stabilized radiotracers that do not require mixing immediately before injection, such as 99m Tc-bicisate, has made ictal SPECT more practical in patients with extratemporal seizures that often are not associated with an aura and may have a shorter seizure duration (O’Brien et al., 1999b). 4. Subtraction ictal SPECT co-registered to MRI (SISCOM) Subtraction peri-ictal SPECT co-registered to MRI (SISCOM) has been introduced in the evaluation of patients with localization-related epilepsy being considered for epilepsy surgery (O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). SISCOM may be useful in the evaluation of patients with focal seizures and normal MRI studies (O’Brien et al., 2000). A localized cerebral blood-flow alteration demonstrated using SISCOM might be intimately associated with
150
the epileptogenic zone in individuals with localizationrelated epilepsy (So, 2000). Subtracting normalized and co-registered ictal and interictal SPECT images, and then matching the resultant difference image to the high-resolution MRI for anatomical correlation, has been shown to be a reliable indicator of the localization of the epileptic brain tissue (O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). SISCOM is superior to the traditional visual analysis of the interictal and ictal images. SISCOM, in a series of 51 patients, had a higher rate of localization (88.2 versus 39.2%, P < 0.0001) and a better inter-observer agreement, and was a better predictor of surgical outcome than visual inspection of the interictal and ictal images (O’Brien et al., 1999a). The methodology used for SISCOM at Mayo Clinic involves co-registering of the interictal to the ictal SPECT image using voxel matching methods (O’Brien et al., 1998b; Brinkmann et al., 2000). The normalized interictal image is subtracted from the normalized ictal image to derive the difference (subtraction) in cerebral blood flow related to the partial seizure. Thresholding of the subtraction image to display only the pixels with intensities greater than two standard deviations above zero is performed. Finally, the images with intensities of more than two standard deviations are co-registered onto the MRI. Following implantation of subdural electrodes for chronic intracranial EEG monitoring, the electrode positions can be segmented from a spiral X-ray-computed axial tomography scan and co-registered with the SISCOM image (So, 2000). This allows the relationship between the localized peri-ictal blood-flow alteration and the ictal-onset zone to be determined. The SISCOM region of blood-flow alteration is a surrogate for the localization of the epileptogenic zone independent of the pathological finding (O’Brien et al., 2000). The clinical parameters that are significant in determining the diagnostic yield of SISCOM may include the duration of the seizure and the length of time of the injection from ictal onset (O’Brien et al., 1996, 1998a). Our experience suggests that to increase the chance of detecting a SISCOM focus, the seizure being studied should be at least 5–10 s in duration, and the time from seizure onset to injection should be less than 45 s (O’Brien et al., 1998a). The SISCOM findings also correlate with the operative outcome (O’Brien et al., 1998a, 2000). Patients with a SISCOM alteration concordant with the epileptogenic zone are most likely to experience a significant reduction in seizure tendency if
G.D. CASCINO ET AL.
the focal cortical resection includes the region of periictal blood-flow change (O’Brien et al., 1998a, 2000). The disadvantages of a SISCOM study include the need for hospitalization and long-term EEG monitoring, the use of radioisotopes for two imaging procedures, and the occurrence of habitual seizure activity. The indications for SISCOM in patients undergoing a presurgical evaluation include: nonsubstrate-directed partial epilepsy and conflicting findings in the noninvasive evaluation. SISCOM may be used to identify a “target” for placement on intracranial EEG electrodes (So, 2000). The presence of a SISCOM alteration may obviate the need for intracranial EEG recordings in selected patients. For example, patients with nonsubstrate-directed partial epilepsy of temporal lobe origin may not require chronic intracranial EEG monitoring if the extracranial ictal EEG pattern and peri-ictal SPECT studies are concordant. SISCOM also improves the diagnostic yield of postictal SPECT studies in patients with intractable partial epilepsy (O’Brien et al., 1999a). The superiority of SISCOM in localizing the epileptogenic zone, particularly in extratemporal epilepsy, has been previously demonstrated (O’Brien et al., 2000). The prognostic importance of the SISCOM focus in patients undergoing a focal cortical resection for partial epilepsy of extratemporal origin has been evaluated (O’Brien et al., 2000). In a Mayo Clinic study, the operative outcome in 36 patients with extratemporal epilepsy who had a SISCOM study prior to surgery was evaluated (O’Brien et al., 2000). The presence of a localizing SISCOM alteration concordant with the ictal-onset zone was a favorable predictor of an excellent surgical outcome (P < 0.05; O’Brien et al., 2000). Eleven of 19 patients (57.9%) with a concordant SISCOM focus and 3 of 17 patients (17.6%) with a nonlocalizing or discordant SISCOM were rendered seizure-free or experienced only nondisabling seizures. Approximately three-quarters of the patients with a localized SISCOM abnormality had a normal structural MRI. In addition, this study demonstrated that the extent of resection of the SISCOM focus was also of prognostic importance (P < 0.05; O’Brien et al., 2000). Failure to resect the neocortical region intimately associated with the localized blood-flow change concordant with the ictal-onset zone was a predictor of an unfavorable operative outcome (O’Brien et al., 2000). Finally, the feasibility and diagnostic yield of SISCOM have been evaluated in pediatric patients with seizure disorders at Mayo Clinic (O’Brien et al.,
151
ICTAL SPECT IN THE DEFINITION OF THE SEIZURE ONSET ZONE
1998c). Fifty-nine patients were admitted on 71 occasions to the epilepsy-monitoring unit for video-EEG monitoring and SISCOM studies (O’Brien et al., 1998c). The mean age was 12 years (range, 1.5–17 years). A SPECT study was performed in 67.6% of admissions and in 72.9% of patients. An ictal injection was obtained in 65.2% of patients. The SPECT injection was postictal in 34.8% of patients. The SISCOM revealed a localized abnormality in 92.3% of patients. Only two patients were unable to have a scan performed after the SPECT injection. 5. Indications for SISCOM The potential indications for SISCOM include patients with intractable partial epilepsy and a normal MRI scan, multilobar pathology, and the presence of a discordant noninvasive presurgical evaluation (O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). SISCOM has also been demonstrated to be of use in patients with widespread unilateral hemispheric epileptogenic lesions. The importance of SISCOM in surgical failures with intractable epilepsy being considered for re-operation has also been evaluated (Fessler et al., 2000). Identification of a localized seizure-onset zone may be used to select operative candidates and alter the diagnostic evaluation. SISCOM may determine the strategy for placement of subdural or depth electrodes for chronic intracranial EEG monitoring or the extent of the focal cortical resection, or both. A SISCOM localized abnormality is also of prognostic importance and may be used to counsel patients regarding surgical treatment. 6. Mayo Clinic protocol The current protocol at the Mayo Clinic in Rochester, MN is for the patients with intractable partial epilepsy being considered for epilepsy surgery to undergo initially a neurological history and examination, MRI head seizure protocol, and routine EEG prior to admission to the epilepsy monitoring unit. Review of the outpatient evaluation and the initial video-EEG findings will be used to determine the potential candidates for a SISCOM study. The peri-ictal neuroimaging studies are performed from 7 a.m. to 3 a.m. 7 days of the week. The radioligand technetium-99m-ethyl cysteinate diethylester (99m Tc-ECD) is injected as soon after seizure onset as possible. The interictal injection
and scan are performed 24 h or later after the ictal imaging study. Patients are treated with antiepileptic drug medication (including benzodiazepine therapy, if necessary) to suppress seizure activity during the period between the two injections. Individuals undergo continuous EEG monitoring at the time of the ictal and interictal SPECT injections. If appropriate, the patients are sedated at the time of the neuroimaging studies. A nurse accompanies the patient to nuclear medicine. The patients ultimately are discussed at a surgical epilepsy conference, during which the comprehensive presurgical evaluation is reviewed. Selected candidates may proceed with chronic intracranial EEG recordings, epilepsy surgery, or both. Neuropsychometry, visual perimetry, and cerebral arteriography with a sodium amobarbital study are performed prior to surgical treatment. 7. Illustrated case reports 7.1. Patient # 936 A 22-year-old patient presented with a history of medically refractory partial seizures. The age at seizure onset was 18 months. The patient had exclusively nocturnal clinical events associated with a hypermotor seizure lasting 30–60 s in duration. Neurological examination revealed a mild chronic global static encephalopathy. The seizures occurred every time the patient fell asleep. Interictal EEG had shown only mild diffuse slowing. There were no epileptiform discharges. The MRI head was normal. The scalp-recorded ictal EEG revealed a myogenic artifact with no definite seizure pattern. An ictal SPECT was performed. The SISCOM study showed a region of focal hyperperfusion in the left frontal lobe (Fig. 1). The patient had a subdural grid of electrodes placed on the dorsolateral frontal convexity. The ictal-onset zone was intimately related to the localized SISCOM abnormality. A focal cortical resection was performed. The pathological findings were consistent with focal cortical dysplasia. The patient has been seizure-free over five years following surgery on a reduced amount of antiepileptic drug medication. 7.1.1. Comment This patient presented with an intractable seizure disorder and a normal MRI. The ictal semiology and electroclinical correlation may suggest the presence of an extratemporal localization of the epileptogenic
152
G.D. CASCINO ET AL.
Fig. 1. Subtracted peri-ictal SPECT co-registered to structural MRI (SISCOM) in a patient with partial seizures and a normal MRI study. (A, B) The SISCOM revealed a localized abnormality in the left frontal lobe. (Note: the left cerebral hemisphere is on the left-hand side of the figure.)
zone. The identification of the seizure-onset zone using SISCOM altered the preoperative evaluation. The presence of a SISCOM focus may indicate that the patient is a favorable candidate for surgical treatment. Chronic intracranial EEG monitoring was necessary to localize the ictal-onset zone and determine the extent of focal cortical resection. The patient may not have been considered a candidate for intracranial EEG recordings without a SISCOM focus. 7.2. Patient # 323 A 13-year-old patient presented with an intractable seizure disorder. The patient had predominantly one type of clinical event associated with “discomfort” in his feet followed by blinking of the right eye, bicycling movements in his legs, and posturing in both upper extremities, maximal left. The seizures typically emerged from sleep and were associated with apparent altered awareness. The age at seizure onset was approximately 3 years. There were rare generalized tonic–clonic seizures. The neurological examination was normal. The interictal EEG showed midline spike discharges during sleep. The MRI head was normal. The scalp-recorded ictal EEG revealed a subtle midline and bilateral central region seizure pattern with no definite lateralization. The SISCOM study showed a region of focal hyperperfusion in the mesial aspect of the right cerebral hemisphere (Fig. 2). PET revealed focal hypometabolism in the posterior right frontal lobe. Chronic intracranial EEG monitoring was performed with the electrodes placed over the SISCOM abnormality. A focal cortical resection was performed. The pathological findings confirmed the presence of focal cortical dysplasia.
Fig. 2. Subtracted peri-ictal SPECT co-registered to structural MRI (SISCOM) in a patient with right supplementary sensorimotor area seizures. MRI study was normal. (A, B) SISCOM showing a focal hyperperfusion abnormality. (C) PET scan revealing a right posterior frontal lobe region of focal hypometabolism. (D) Intracranial EEG electrodes placed over the SISCOM and PET abnormalities. (E) Chronic intracranial EEG monitoring used to delineate the ictal-onset zone that was concordant with the functional neuroimaging studies. (F, G) Pathological findings underlying the epileptogenic zone revealing focal cortical dysplasia with large, abnormal neurons and disrupted lamination. (Note: the right cerebral hemisphere is on the left-hand side of the figures.)
The patient has remained seizure-free following surgery. 7.2.1. Comment This patient has a symptomatic partial epilepsy of probable right supplementary sensorimotor area origin. The SISCOM was used to detect the seizure-onset zone. The concordance of the PET, SISCOM and intracranial EEG findings indicated that the patient may be a favorable operative candidate.
ICTAL SPECT IN THE DEFINITION OF THE SEIZURE ONSET ZONE
8. Conclusions The rationale for neurosurgical treatment of epilepsy is to significantly reduce seizure tendency and allow the individual to become a participating and productive member of society (Dreifuss, 1987; Mattson, 1992; Cascino, 1996). Favorable operative candidates include patients with mesial temporal lobe epilepsy or foreign-tissue lesions. These surgically remediable epileptic syndromes are typically associated with abnormal MRI findings revealing substrate-directed disease (Cascino et al., 1993b). The results of epilepsy surgery in patients with nonsubstrate-directed partial epilepsy of extratemporal origin have been disappointing (Cascino et al., 1992). In patients with intractable partial epilepsy and a normal MRI, there are significant concerns regarding the localization of the epileptogenic zone prior to consideration of surgical treatment. Chronic intracranial EEG monitoring may prove necessary in these patients, especially with extratemporal epilepsy. Identification of a localized SISCOM focus may be a reliable indicator of the seizure-onset zone. SISCOM may reveal a localized region of cerebral hyperperfusion or hypoperfusion in up to 80% of patients with intractable partial epilepsy (O’Brien et al., 1998a). The SISCOM findings are also predictive of operative outcome (O’Brien et al., 2000). Ultimately, a decision regarding surgical treatment must be based on a convergence of the neurodiagnostic evaluation. Electrophysiological studies invariably need to be performed to localize the ictal-onset zone in these patients. Resection of the SISCOM focus is necessary to significantly reduce the seizure tendency in patients with a localized SISCOM abnormality that is concordant with the epileptic brain tissue. References Awad, IA, Rosenfeld, J, Ahl, H, Hahn, JF and L¨uders, H (1991) Intractable epilepsy and structural lesions of the brain: mapping, resection strategies, and seizure outcome. Epilepsia, 32: 179–186. Berkovic, SF, Andermann, F, Olivier, A et al. (1991) Hippocampal sclerosis in temporal lobe epilepsy demonstrated by magnetic resonance imaging. Ann. Neurol., 29: 175–182. Brinkmann, BH, O’Brien, TJ, Webster, DB et al. (2000) Voxel significance mapping using local image variances in subtraction ictal SPET. Nucl. Med. Commun., 21: 545– 551.
153 Cambier, DM, Cascino, GD, So, EL and Marsh, WR (2001) Video–EEG monitoring in patients with hippocampal atrophy. Acta. Neurol. Scan., 103: 1–7. Camfield, P and Camfield, C (1996) Antiepileptic drug therapy: when is epilepsy truly intractable? Epilepsia, 37(1): S60–S65. Cascino, GD (1996) Selection of candidates for surgical treatment of epilepsy. In: GD Cascino and CR Jack, Jr (Eds.), Neuroimaging in Epilepsy: Principles and Practice. Butterworth-Heinemann, Boston, pp. 209–218. Cascino, GD (2001) Advances in neuroimaging: surgical localization. Epilepsia, 42: 3–12. Cascino, GD, Jack, CR, Jr, Parisi, JE et al. (1992) MRI in the presurgical evaluation of patients with frontal lobe epilepsy and children with temporal lobe epilepsy: pathological correlation and prognostic importance. Epilepsy Res., 11: 51–59. Cascino, GD, Boon, PAJM and Fish, DR (1993a) Surgically remediable lesional syndromes. In: J Engel, Jr (Ed.), Surgical Treatment of the Epilepsies, 2nd edn. Raven Press, New York, pp. 77–86. Cascino, GD, Jack, CR, Parisi, J et al. (1993b) Operative strategy in patients with MRI-identified dual pathology and temporal lobe epilepsy. Epilepsy Res., 14: 175–182. Cascino, GD, Trenerry, MR, So, E et al. (1996) Routine EEG and temporal lobe epilepsy: relation to long-term EEG monitoring, quantitative MRI, and operative outcome. Epilepsia, 37: 651–656. Crandall, PH (1987) Postoperative management and criteria for evaluation. In: DP Purpura, JK Penry and RD Walter (Eds.), Advances in Neurology. Raven Press, New York, pp. 31–49. Dreifuss, FE (1987) Goals of surgery for epilepsy. In: J Engel, Jr (Ed.), Surgical Treatment of the Epilepsies, 1st edn. Raven Press, New York, pp. 31–49. Engel, J, Jr and Ojemann, GA (1993) The next step. In: J Engel, Jr (Ed.), Surgical Treatment of the Epilepsies, 2nd edn. Raven Press, New York, pp. 319–329. Fessler, JA, Cascino, GD, So, EL et al. (2000) Subtraction ictal SPECT co-registered to MRI (SISCOM) in the evaluation for repeat epilepsy surgery. Neurology, 54(Suppl. 3): A4. Hauser, A and Hesdorffer, D (1990) Prognosis. In: WA Hauser and DC Hesdorffer (Eds.), Epilepsy: Frequency, Causes and Consequences. Demos, New York, pp. 197–243. Hauser, W (1992) The natural history of drug resistant epilepsy: epidemiologic considerations. Epilepsy Res. Suppl., 5: 25–28. Henry, TR (1996) Functional neuroimaging with positron emission tomography. Epilepsia, 37: 1141–1154. Henry, TR, Babb, TL, Engel, J, Jr et al. (1994) Hippocampal neuronal loss and regional hypometabolism in temporal lobe epilepsy. Ann. Neurol., 36: 925–927.
154 Ho, SS, Berkovic, SF, Berlangieri, SU et al. (1995) Comparison of ictal SPECT and interictal PET in the presurgical evaluation of temporal lobe epilepsy. Ann. Neurol., 37: 738–745. Jack, CR, Jr, Trenerry, MR, Cascino, GD, Sharbrough, FW, So, EL and O’Brien, PC (1995) Bilaterally symmetric hippocampi and surgical outcome. Neurology, 45: 1353–1358. Marks, DA, Katz, A, Hoffer, P et al. (1992) Localization of extratemporal epileptic foci during ictal single photon emission computed tomography. Ann. Neurol., 31: 250–255. Mattson, RH (1992) Drug treatment of uncontrolled seizures. Epilepsy Res. Suppl., 5: 29–35. Mosewich, RK, So, EL, O’Brien, TJ et al. (2000) Factors predictive of the outcome of frontal lobe epilepsy surgery. Epilepsia, 41: 843–849. Newton, MR and Berkovic, SF (1996) Interictal, ictal, and postictal single photon emission computed tomography. In: GD Cascino and CR Jack, Jr (Eds.), Neuroimaging in Epilepsy: Principles and Practice. Butterworth-Heinemann, Boston, pp. 177–192. O’Brien, TJ, So, EL, Mullan, BP et al. (1996) Extent of resection of the ictal subtraction SPECT focus is an important determinant of epilepsy surgery outcome. Epilepsia, 37(Suppl. 5): S182. O’Brien, TJ, So, EL, Mullan, BP et al. (1998a) Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology, 55: 445–454. O’Brien, TJ, O’Connor, MK, Mullan, BP et al. (1998b) Subtraction ictal SPECT co-registered to MRI in partial epilepsy: Description and technical validation of the method with phantom and patients studies. Nucl. Med. Commun., 19: 31–45. O’Brien, TJ, Zupanc, ML, Mullan, BP et al. (1998c) The practical utility of performing peri-ictal SPECT in the
G.D. CASCINO ET AL.
evaluation of children with partial epilepsy. Pediatric Neurol., 19: 15–22. O’Brien, TJ, So, EL, Mullan, BP et al. (1999a) Subtraction SPECT co-registered to MRI improves postictal localization of seizure foci. Neurology, 52: 137–146. O’Brien, TJ, Brinkmann, BH, Mullan, BP et al. (1999b) Comparative study of 99mTc-ECD and 99m Tc-HMPAO for SPECT: qualitative and quantitative analysis. J. Neurol. Neurosurg. Psychiatry, 66: 331–339. O’Brien, TJ, So, EL, Mullan, BP et al. (2000) Subtraction peri-ictal SPECT is predictive of extratemporal epilepsy surgery outcome. Neurology, 55: 1668–1677. Palmini, A, Andermann, F, Olivier, A et al. (1991) Focal neuronal migrational disorders and intractable partial epilepsy: results of surgical treatment. Ann. Neurol., 30: 750–757. Radhakrishnan, K, So, EL, Silbert, PL et al. (1998) Predictors of outcome of anterior temporal lobectomy for intractable epilepsy: a multivariate study. Neurology, 51: 465– 471. So, EL (2000) Integration of EEG, MRI and SPECT in localizing the seizure focus for epilepsy surgery. Epilepsia, 41(Suppl. 3): S48–S54. So, EL, O’Brien, TJ, Brinkmann, BH et al. (2000) The EEG evaluation of single photon emission computed tomography abnormalities in epilepsy. J. Clin. Neurophysiol., 17: 10–28. Theodore, WH (1996) Positron emission tomography in the evaluation of epilepsy. In: GD Cascino and CR Jack, Jr (Eds.), Neuroimaging in Epilepsy. Principles and Practice. Butterworth-Heinemann, Boston, pp. 165– 175. Wiebe, S, Blume, WT, Girvin, JP and Eliasziw, M (2001) Effectiveness and efficiency of surgery for temporal lobe epilepsy study group. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N. Engl. J. Med., 345: 311–318.
147
CHAPTER 2.10
Ictal SPECT in the definition of the seizure onset zone Gregory D. Cascinoa,∗ , Jeffrey R. Buchhaltera , Brian P. Mullanb and Elson L. Soa b
a Department of Neurology, Mayo Clinic, 200 First St., SW, Rochester, MN 55905, USA Department of Diagnostic Radiology (Nuclear Medicine), Mayo Clinic, Rochester, MN 55905, USA
1. Introduction Partial or localization-related epilepsy is characterized by recurrent and unprovoked focal seizure activity and is the most common seizure disorder (Dreifuss, 1987; Mattson, 1992; Cascino, 1996a). Over 90% of the incident cases of epilepsy in adults have focal seizures (Dreifuss, 1987). Individuals with partial epilepsy may experience simple partial seizures, complex partial seizures, or secondarily generalized tonic–clonic seizures (Mattson, 1992). The most epileptogenic region in patients with focal seizures is the amygdalohippocampal complex, associated with mesial temporal lobe epilepsy (Cascino, 1996a). Approximately 30–40% of newly diagnosed patients with epilepsy will experience medically refractory seizures that are physically and socially disabling (Dreifuss, 1987; Hauser, 1992; Mattson, 1992; Camfield and Camfield, 1996). Less than 10% of patients who have focal seizures refractory to their initial antiepileptic drug (AED) therapy will be rendered seizure-free with additional medication trials (Hauser and Hesdorffer, 1990; Hauser, 1992; Mattson, 1992; Camfield and Camfield, 1996). Surgical treatment is an effective therapy for highly selected patients with localizationrelated epilepsy (Crandall, 1987; Dreifuss, 1987; Awad et al., 1991; Cascino et al., 1993b; Engel and Ojemann, 1993; Cascino, 1996; Radhakrishnan et al., 1998; Wiebe et al., 2001). The goals of surgical treatment are to render the individual seizure-free and allow the patient to become a participating and productive member of society (Dreifuss, 1987; Cascino, 1996; Radhakrishnan et al., 1998). Surgery is most often performed for
∗
Correspondence to: Dr. Gregory D. Cascino. E-mail address: [email protected] Tel.: +1-507-284-2511; fax: +1-507-284-8686.
individuals with mesial temporal lobe epilepsy or focal seizures related to foreign-tissue lesional pathology (Awad et al., 1991; Cascino et al., 1993b; Wiebe et al., 2001). The surgically excised hippocampus in patients with mesial temporal lobe epilepsy usually reveals focal cell loss and an increase in astrogliosis, i.e. hippocampal sclerosis (Cascino et al., 1992, 1993a, 1996b; Cambier et al., 2001). The surgically excised pathology in approximately 65% of patients with intractable partial epilepsy is hippocampal sclerosis (Hauser, 1992; Wiebe et al., 2001). Patients with a chronic seizure disorder and a lesional epileptic syndrome may have a low-grade glioma, ganglioglioma, cavernous hemangioma, or focal cortical dysplasia (Awad et al., 1991; Palmini et al., 1991; Cascino et al., 1992, 1993a,b; Mosewich et al., 2000). A structural lesion underlying the epileptogenic zone is identified in approximately 30% of patients undergoing epilepsy surgery (Cascino et al., 1993b). Mesial temporal lobe epilepsy and focal seizures related to a lesional pathology are considered surgically remediable epileptic syndromes because of the significant reduction in seizure tendency following a focal cortical resection and excision of the pathological findings underlying the epileptogenic zone (Crandall, 1987; Awad et al., 1991; Berkovic et al., 1991; Cascino et al., 1992, 1993a,b, 1996; Engel and Ojemann, 1993; Cascino, 1996; Radhakrishnan et al., 1998; Cambier et al., 2001; Wiebe et al., 2001). Individuals with hippocampal sclerosis, lesional pathology, and other “macroscopic” structural abnormalities, e.g. focal encephalomalacia, usually have an abnormal MRI study (Palmini et al., 1991; Cascino et al., 1992, 1993b; Cascino, 1996, 2001). These patients are classified as having substrate-directed partial epilepsy. The MRI in these individuals is often a reliable indicator of the focal pathology and may suggest the region of seizure onset (Cascino, 2001). A comprehensive evaluation including ictal
148
video-EEG monitoring is required, however, because the MRI-identified structural abnormalities may not be contiguous or may be remote from the epileptogenic zone (Awad et al., 1991). There is a broad consensus that a MRI study using the “temporal lobe seizure protocol” has a high diagnostic yield in patients with mesial temporal lobe epilepsy related to hippocampal sclerosis (Berkovic et al., 1991; Cascino et al., 1992; Cambier et al., 2001). MRI shows hippocampal formation atrophy and an associated signal intensity alteration in over 90% of these individuals (Berkovic et al., 1991; Cascino et al., 1992; Cambier et al., 2001). The hippocampal atrophy is related to focal cell loss, while the signal intensity change may be associated with gliosis (Berkovic et al., 1991; Cascino et al., 1992; Cambier et al., 2001). MRI almost invariably identifies a foreign-tissue lesion in patients with tumors and vascular malformations (Cascino et al., 1993b; Engel and Ojemann, 1993). The most common primary brain tumors associated with intractable partial epilepsy include gangliogliomas, dysembryoplastic neuroepithelial tumors, and the low-grade glial neoplasms (Palmini et al., 1991; Cascino et al., 1992; Cascino, 2001). The sensitivity and specificity of MRI in patients with malformations of cortical development depend on several factors, including the histopathology and anatomical localization of the lesion (Palmini et al., 1991). Approximately 75% of patients with focal cortical dysplasia will have a localizing structural neuroimaging alteration (Palmini et al., 1991; Cascino, 2001). MRI significantly alters the preoperative evaluation and operative strategy in patients with hippocampal sclerosis and lesional pathology (Berkovic et al., 1991; Cascino et al., 1992, 1993b, 1996; Palmini et al., 1991; Cambier et al., 2001; Wiebe et al., 2001). The rationale for the presurgical evaluation is to identify the site of ictal onset and initial seizure propagation, i.e. epileptogenic zone, and determine the likely pathological findings underlying the epileptic brain tissue (Engel and Ojemann, 1993; Cascino et al., 1996). The demonstration of concordance between the MRI-identified pathological findings and the epileptic brain tissue indicates a highly favorable operative outcome. Approximately 80% of patients with unilateral hippocampal sclerosis, a low-grade primary brain neoplasm, or a cavernous hemangioma are rendered seizure-free following surgical treatment (Awad et al., 1991; Cascino et al., 1992, 1993a, 1996; Cascino, 1996; Mosewich et al., 2000; Cambier et al., 2001; Wiebe et al., 2001). Over 90%
G.D. CASCINO ET AL.
of these patients will experience an excellent surgical outcome, i.e. auras only or rare nondisabling seizures (Radhakrishnan et al., 1998; Cambier et al., 2001). The efficacy of surgical treatment is reduced in patients with malformations of cortical development (Engel and Ojemann, 1993). Nearly half of the individuals, however, experience a worthwhile reduction in seizures following surgical treatment (Palmini et al., 1991). 2. Localization-related epilepsy with a normal MRI Patients with intractable localization-related seizure disorders and normal MRI studies, i.e. nonsubstratedirected partial epilepsy, may be considered for surgical treatment (O’Brien et al., 1996; Mosewich et al., 2000). Patients with normal MRI studies may have mesial temporal lobe epilepsy or extrahippocampal, neocortical, seizures (Jack et al., 1995; O’Brien et al., 1996; Mosewich et al., 2000). The diagnosis of mesial temporal lobe epilepsy in this group may be suggested by ictal semiology, interictal epileptiform discharges, or ictal EEG pattern (Radhakrishnan et al., 1998). The interictal PET study usually shows regional hypometabolism in patients with mesial temporal lobe epilepsy (Henry et al., 1994; Ho et al., 1995; Theodore et al., 1996). Approximately 50% of patients with normal MRI studies who undergo surgery for mesial temporal lobe epilepsy are rendered seizure-free (Jack et al., 1995). Most commonly, patients with focal seizures and normal MRI studies have extrahippocampal, neocortical, seizures of extratemporal origin (Cascino et al., 1992; Mosewich et al., 2000). The pathological findings underlying the epileptogenic zone in patients with a normal MRI and localization-related epilepsy include astrogliosis, focal cell loss, no histopathological alteration, or a focal cortical dysplasia (Cascino et al., 1992; Mosewich et al., 2000; Cascino, 2001). Unfortunately, patients with extratemporal seizures and a normal MRI study are less favorable candidates for surgical treatment (Cascino et al., 1992; Mosewich et al., 2000). An estimated 20–30% of these patients with extratemporal, mainly frontal lobe, seizures will enter a seizure remission following a focal cortical resection (Cascino et al., 1992). It may be difficult to localize or lateralize the epileptogenic zone in patients with neocortical seizures and a normal MRI. The epileptic brain tissue in patients with extrahippocampal, neocortical, seizures may represent a continuum that results in a subtotal resection of
ICTAL SPECT IN THE DEFINITION OF THE SEIZURE ONSET ZONE
the epileptogenic zone. The potential limitations of interictal and ictal extracranial and intracranial EEG monitoring in patients with partial seizures of extratemporal origin have been well defined (Cascino et al., 1992). There is a strong need for additional neurodiagnostic studies that can be carried out prior to implantation of intracranial electrodes to localize the epileptogenic zone in patients with extrahippocampal, neocortical, seizures and a normal MRI study. Interictal PET (2-fluorodeoxyglucose) has a low diagnostic yield in these patients (Theodore, 1996). Peri-ictal functional neuroimaging techniques have the capability to assist in localizing the epileptic brain tissue, i.e. identify the seizure-onset zone, in patients with intractable partial epilepsy (Marks et al., 1992; Henry, 1996; Newton and Berkovic, 1996; O’Brien et al., 1998a,b, 1999a,b; Brinkmann et al., 2000; So, 2000; So et al., 2000). The rationale for these diagnostic studies includes the selection of operative candidates. The results of peri-ictal functional neuroimaging may modify the preoperative evaluation and alter the surgical excision (So, 2000). Unfortunately, PET does not lend itself to ictal imaging in the epilepsy monitoring unit (Henry, 1996; Theodore, 1996). SPECT has the potential for peri-ictal imaging studies in patients undergoing video-EEG monitoring prior to surgical treatment (Marks et al., 1992; Newton and Berkovic, 1996; O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). 3. Single photon emission computed tomography SPECT is most appropriate for ictal imaging in patients with localization-related epilepsy being considered for epilepsy surgery to identify the seizure-onset zone (Marks et al., 1992; Newton and Berkovic, 1996; O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). Ictal SPECT studies are superior to interictal images in localization-related epilepsy (O’Brien et al., 1998a,b). SPECT involves blood-flow imaging using radiopharmaceuticals, e.g. 99m Tcbicisate, that have a rapid first pass brain extraction with maximum uptake being achieved within 30–60 s of an intravenous injection (O’Brien et al., 1998a,b, 1999b; Brinkmann et al., 2000). The SPECT images can be acquired up to 4 h after the termination of the seizure so that the individual patient can recover from the seizure episode prior to being transported to the
149
nuclear medicine laboratory. SPECT studies have an important clinical application in the potential identification of the epileptic brain tissue when the presurgical evaluation is indeterminate as to the localization of the ictal-onset zone (O’Brien et al., 1998a). This includes patients with nonsubstrate-directed partial epilepsy, widespread unilateral structural abnormalities, and multilobar lesions (O’Brien et al., 1999a). The initial blood-flow SPECT studies in patients with intractable partial epilepsy involved interictal imaging which variably detected a focal hypoperfusion in the region of the epileptogenic zone (Marks et al., 1992; Newton and Berkovic, 1996). Interictal SPECT images have proven to have a relatively low sensitivity and a relatively high false positive rate in temporal lobe epilepsy (Newton and Berkovic, 1996). Interictal SPECT has also been shown to have a low diagnostic yield in patients with extratemporal seizures. Ictal SPECT studies have been confirmed to be useful in patients with temporal lobe epilepsy to identify a region of focal hyperperfusion (Marks et al., 1992; Newton and Berkovic, 1996). The rationale for interictal SPECT imaging at present is to serve as a reference for a baseline study for the interpretation of ictal SPECT images. The diagnostic yield of ictal SPECT has been established to be superior to interictal SPECT in patients being considered for a focal cortical resection (Marks et al., 1992; Newton and Berkovic, 1996; O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). The recent development of stabilized radiotracers that do not require mixing immediately before injection, such as 99m Tc-bicisate, has made ictal SPECT more practical in patients with extratemporal seizures that often are not associated with an aura and may have a shorter seizure duration (O’Brien et al., 1999b). 4. Subtraction ictal SPECT co-registered to MRI (SISCOM) Subtraction peri-ictal SPECT co-registered to MRI (SISCOM) has been introduced in the evaluation of patients with localization-related epilepsy being considered for epilepsy surgery (O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). SISCOM may be useful in the evaluation of patients with focal seizures and normal MRI studies (O’Brien et al., 2000). A localized cerebral blood-flow alteration demonstrated using SISCOM might be intimately associated with
150
the epileptogenic zone in individuals with localizationrelated epilepsy (So, 2000). Subtracting normalized and co-registered ictal and interictal SPECT images, and then matching the resultant difference image to the high-resolution MRI for anatomical correlation, has been shown to be a reliable indicator of the localization of the epileptic brain tissue (O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). SISCOM is superior to the traditional visual analysis of the interictal and ictal images. SISCOM, in a series of 51 patients, had a higher rate of localization (88.2 versus 39.2%, P < 0.0001) and a better inter-observer agreement, and was a better predictor of surgical outcome than visual inspection of the interictal and ictal images (O’Brien et al., 1999a). The methodology used for SISCOM at Mayo Clinic involves co-registering of the interictal to the ictal SPECT image using voxel matching methods (O’Brien et al., 1998b; Brinkmann et al., 2000). The normalized interictal image is subtracted from the normalized ictal image to derive the difference (subtraction) in cerebral blood flow related to the partial seizure. Thresholding of the subtraction image to display only the pixels with intensities greater than two standard deviations above zero is performed. Finally, the images with intensities of more than two standard deviations are co-registered onto the MRI. Following implantation of subdural electrodes for chronic intracranial EEG monitoring, the electrode positions can be segmented from a spiral X-ray-computed axial tomography scan and co-registered with the SISCOM image (So, 2000). This allows the relationship between the localized peri-ictal blood-flow alteration and the ictal-onset zone to be determined. The SISCOM region of blood-flow alteration is a surrogate for the localization of the epileptogenic zone independent of the pathological finding (O’Brien et al., 2000). The clinical parameters that are significant in determining the diagnostic yield of SISCOM may include the duration of the seizure and the length of time of the injection from ictal onset (O’Brien et al., 1996, 1998a). Our experience suggests that to increase the chance of detecting a SISCOM focus, the seizure being studied should be at least 5–10 s in duration, and the time from seizure onset to injection should be less than 45 s (O’Brien et al., 1998a). The SISCOM findings also correlate with the operative outcome (O’Brien et al., 1998a, 2000). Patients with a SISCOM alteration concordant with the epileptogenic zone are most likely to experience a significant reduction in seizure tendency if
G.D. CASCINO ET AL.
the focal cortical resection includes the region of periictal blood-flow change (O’Brien et al., 1998a, 2000). The disadvantages of a SISCOM study include the need for hospitalization and long-term EEG monitoring, the use of radioisotopes for two imaging procedures, and the occurrence of habitual seizure activity. The indications for SISCOM in patients undergoing a presurgical evaluation include: nonsubstrate-directed partial epilepsy and conflicting findings in the noninvasive evaluation. SISCOM may be used to identify a “target” for placement on intracranial EEG electrodes (So, 2000). The presence of a SISCOM alteration may obviate the need for intracranial EEG recordings in selected patients. For example, patients with nonsubstrate-directed partial epilepsy of temporal lobe origin may not require chronic intracranial EEG monitoring if the extracranial ictal EEG pattern and peri-ictal SPECT studies are concordant. SISCOM also improves the diagnostic yield of postictal SPECT studies in patients with intractable partial epilepsy (O’Brien et al., 1999a). The superiority of SISCOM in localizing the epileptogenic zone, particularly in extratemporal epilepsy, has been previously demonstrated (O’Brien et al., 2000). The prognostic importance of the SISCOM focus in patients undergoing a focal cortical resection for partial epilepsy of extratemporal origin has been evaluated (O’Brien et al., 2000). In a Mayo Clinic study, the operative outcome in 36 patients with extratemporal epilepsy who had a SISCOM study prior to surgery was evaluated (O’Brien et al., 2000). The presence of a localizing SISCOM alteration concordant with the ictal-onset zone was a favorable predictor of an excellent surgical outcome (P < 0.05; O’Brien et al., 2000). Eleven of 19 patients (57.9%) with a concordant SISCOM focus and 3 of 17 patients (17.6%) with a nonlocalizing or discordant SISCOM were rendered seizure-free or experienced only nondisabling seizures. Approximately three-quarters of the patients with a localized SISCOM abnormality had a normal structural MRI. In addition, this study demonstrated that the extent of resection of the SISCOM focus was also of prognostic importance (P < 0.05; O’Brien et al., 2000). Failure to resect the neocortical region intimately associated with the localized blood-flow change concordant with the ictal-onset zone was a predictor of an unfavorable operative outcome (O’Brien et al., 2000). Finally, the feasibility and diagnostic yield of SISCOM have been evaluated in pediatric patients with seizure disorders at Mayo Clinic (O’Brien et al.,
151
ICTAL SPECT IN THE DEFINITION OF THE SEIZURE ONSET ZONE
1998c). Fifty-nine patients were admitted on 71 occasions to the epilepsy-monitoring unit for video-EEG monitoring and SISCOM studies (O’Brien et al., 1998c). The mean age was 12 years (range, 1.5–17 years). A SPECT study was performed in 67.6% of admissions and in 72.9% of patients. An ictal injection was obtained in 65.2% of patients. The SPECT injection was postictal in 34.8% of patients. The SISCOM revealed a localized abnormality in 92.3% of patients. Only two patients were unable to have a scan performed after the SPECT injection. 5. Indications for SISCOM The potential indications for SISCOM include patients with intractable partial epilepsy and a normal MRI scan, multilobar pathology, and the presence of a discordant noninvasive presurgical evaluation (O’Brien et al., 1998a,b,c, 1999a,b, 2000; Brinkmann et al., 2000; Fessler et al., 2000; So, 2000; So et al., 2000). SISCOM has also been demonstrated to be of use in patients with widespread unilateral hemispheric epileptogenic lesions. The importance of SISCOM in surgical failures with intractable epilepsy being considered for re-operation has also been evaluated (Fessler et al., 2000). Identification of a localized seizure-onset zone may be used to select operative candidates and alter the diagnostic evaluation. SISCOM may determine the strategy for placement of subdural or depth electrodes for chronic intracranial EEG monitoring or the extent of the focal cortical resection, or both. A SISCOM localized abnormality is also of prognostic importance and may be used to counsel patients regarding surgical treatment. 6. Mayo Clinic protocol The current protocol at the Mayo Clinic in Rochester, MN is for the patients with intractable partial epilepsy being considered for epilepsy surgery to undergo initially a neurological history and examination, MRI head seizure protocol, and routine EEG prior to admission to the epilepsy monitoring unit. Review of the outpatient evaluation and the initial video-EEG findings will be used to determine the potential candidates for a SISCOM study. The peri-ictal neuroimaging studies are performed from 7 a.m. to 3 a.m. 7 days of the week. The radioligand technetium-99m-ethyl cysteinate diethylester (99m Tc-ECD) is injected as soon after seizure onset as possible. The interictal injection
and scan are performed 24 h or later after the ictal imaging study. Patients are treated with antiepileptic drug medication (including benzodiazepine therapy, if necessary) to suppress seizure activity during the period between the two injections. Individuals undergo continuous EEG monitoring at the time of the ictal and interictal SPECT injections. If appropriate, the patients are sedated at the time of the neuroimaging studies. A nurse accompanies the patient to nuclear medicine. The patients ultimately are discussed at a surgical epilepsy conference, during which the comprehensive presurgical evaluation is reviewed. Selected candidates may proceed with chronic intracranial EEG recordings, epilepsy surgery, or both. Neuropsychometry, visual perimetry, and cerebral arteriography with a sodium amobarbital study are performed prior to surgical treatment. 7. Illustrated case reports 7.1. Patient # 936 A 22-year-old patient presented with a history of medically refractory partial seizures. The age at seizure onset was 18 months. The patient had exclusively nocturnal clinical events associated with a hypermotor seizure lasting 30–60 s in duration. Neurological examination revealed a mild chronic global static encephalopathy. The seizures occurred every time the patient fell asleep. Interictal EEG had shown only mild diffuse slowing. There were no epileptiform discharges. The MRI head was normal. The scalp-recorded ictal EEG revealed a myogenic artifact with no definite seizure pattern. An ictal SPECT was performed. The SISCOM study showed a region of focal hyperperfusion in the left frontal lobe (Fig. 1). The patient had a subdural grid of electrodes placed on the dorsolateral frontal convexity. The ictal-onset zone was intimately related to the localized SISCOM abnormality. A focal cortical resection was performed. The pathological findings were consistent with focal cortical dysplasia. The patient has been seizure-free over five years following surgery on a reduced amount of antiepileptic drug medication. 7.1.1. Comment This patient presented with an intractable seizure disorder and a normal MRI. The ictal semiology and electroclinical correlation may suggest the presence of an extratemporal localization of the epileptogenic
152
G.D. CASCINO ET AL.
Fig. 1. Subtracted peri-ictal SPECT co-registered to structural MRI (SISCOM) in a patient with partial seizures and a normal MRI study. (A, B) The SISCOM revealed a localized abnormality in the left frontal lobe. (Note: the left cerebral hemisphere is on the left-hand side of the figure.)
zone. The identification of the seizure-onset zone using SISCOM altered the preoperative evaluation. The presence of a SISCOM focus may indicate that the patient is a favorable candidate for surgical treatment. Chronic intracranial EEG monitoring was necessary to localize the ictal-onset zone and determine the extent of focal cortical resection. The patient may not have been considered a candidate for intracranial EEG recordings without a SISCOM focus. 7.2. Patient # 323 A 13-year-old patient presented with an intractable seizure disorder. The patient had predominantly one type of clinical event associated with “discomfort” in his feet followed by blinking of the right eye, bicycling movements in his legs, and posturing in both upper extremities, maximal left. The seizures typically emerged from sleep and were associated with apparent altered awareness. The age at seizure onset was approximately 3 years. There were rare generalized tonic–clonic seizures. The neurological examination was normal. The interictal EEG showed midline spike discharges during sleep. The MRI head was normal. The scalp-recorded ictal EEG revealed a subtle midline and bilateral central region seizure pattern with no definite lateralization. The SISCOM study showed a region of focal hyperperfusion in the mesial aspect of the right cerebral hemisphere (Fig. 2). PET revealed focal hypometabolism in the posterior right frontal lobe. Chronic intracranial EEG monitoring was performed with the electrodes placed over the SISCOM abnormality. A focal cortical resection was performed. The pathological findings confirmed the presence of focal cortical dysplasia.
Fig. 2. Subtracted peri-ictal SPECT co-registered to structural MRI (SISCOM) in a patient with right supplementary sensorimotor area seizures. MRI study was normal. (A, B) SISCOM showing a focal hyperperfusion abnormality. (C) PET scan revealing a right posterior frontal lobe region of focal hypometabolism. (D) Intracranial EEG electrodes placed over the SISCOM and PET abnormalities. (E) Chronic intracranial EEG monitoring used to delineate the ictal-onset zone that was concordant with the functional neuroimaging studies. (F, G) Pathological findings underlying the epileptogenic zone revealing focal cortical dysplasia with large, abnormal neurons and disrupted lamination. (Note: the right cerebral hemisphere is on the left-hand side of the figures.)
The patient has remained seizure-free following surgery. 7.2.1. Comment This patient has a symptomatic partial epilepsy of probable right supplementary sensorimotor area origin. The SISCOM was used to detect the seizure-onset zone. The concordance of the PET, SISCOM and intracranial EEG findings indicated that the patient may be a favorable operative candidate.
ICTAL SPECT IN THE DEFINITION OF THE SEIZURE ONSET ZONE
8. Conclusions The rationale for neurosurgical treatment of epilepsy is to significantly reduce seizure tendency and allow the individual to become a participating and productive member of society (Dreifuss, 1987; Mattson, 1992; Cascino, 1996). Favorable operative candidates include patients with mesial temporal lobe epilepsy or foreign-tissue lesions. These surgically remediable epileptic syndromes are typically associated with abnormal MRI findings revealing substrate-directed disease (Cascino et al., 1993b). The results of epilepsy surgery in patients with nonsubstrate-directed partial epilepsy of extratemporal origin have been disappointing (Cascino et al., 1992). In patients with intractable partial epilepsy and a normal MRI, there are significant concerns regarding the localization of the epileptogenic zone prior to consideration of surgical treatment. Chronic intracranial EEG monitoring may prove necessary in these patients, especially with extratemporal epilepsy. Identification of a localized SISCOM focus may be a reliable indicator of the seizure-onset zone. SISCOM may reveal a localized region of cerebral hyperperfusion or hypoperfusion in up to 80% of patients with intractable partial epilepsy (O’Brien et al., 1998a). The SISCOM findings are also predictive of operative outcome (O’Brien et al., 2000). Ultimately, a decision regarding surgical treatment must be based on a convergence of the neurodiagnostic evaluation. Electrophysiological studies invariably need to be performed to localize the ictal-onset zone in these patients. Resection of the SISCOM focus is necessary to significantly reduce the seizure tendency in patients with a localized SISCOM abnormality that is concordant with the epileptic brain tissue. References Awad, IA, Rosenfeld, J, Ahl, H, Hahn, JF and L¨uders, H (1991) Intractable epilepsy and structural lesions of the brain: mapping, resection strategies, and seizure outcome. Epilepsia, 32: 179–186. Berkovic, SF, Andermann, F, Olivier, A et al. (1991) Hippocampal sclerosis in temporal lobe epilepsy demonstrated by magnetic resonance imaging. Ann. Neurol., 29: 175–182. Brinkmann, BH, O’Brien, TJ, Webster, DB et al. (2000) Voxel significance mapping using local image variances in subtraction ictal SPET. Nucl. Med. Commun., 21: 545– 551.
153 Cambier, DM, Cascino, GD, So, EL and Marsh, WR (2001) Video–EEG monitoring in patients with hippocampal atrophy. Acta. Neurol. Scan., 103: 1–7. Camfield, P and Camfield, C (1996) Antiepileptic drug therapy: when is epilepsy truly intractable? Epilepsia, 37(1): S60–S65. Cascino, GD (1996) Selection of candidates for surgical treatment of epilepsy. In: GD Cascino and CR Jack, Jr (Eds.), Neuroimaging in Epilepsy: Principles and Practice. Butterworth-Heinemann, Boston, pp. 209–218. Cascino, GD (2001) Advances in neuroimaging: surgical localization. Epilepsia, 42: 3–12. Cascino, GD, Jack, CR, Jr, Parisi, JE et al. (1992) MRI in the presurgical evaluation of patients with frontal lobe epilepsy and children with temporal lobe epilepsy: pathological correlation and prognostic importance. Epilepsy Res., 11: 51–59. Cascino, GD, Boon, PAJM and Fish, DR (1993a) Surgically remediable lesional syndromes. In: J Engel, Jr (Ed.), Surgical Treatment of the Epilepsies, 2nd edn. Raven Press, New York, pp. 77–86. Cascino, GD, Jack, CR, Parisi, J et al. (1993b) Operative strategy in patients with MRI-identified dual pathology and temporal lobe epilepsy. Epilepsy Res., 14: 175–182. Cascino, GD, Trenerry, MR, So, E et al. (1996) Routine EEG and temporal lobe epilepsy: relation to long-term EEG monitoring, quantitative MRI, and operative outcome. Epilepsia, 37: 651–656. Crandall, PH (1987) Postoperative management and criteria for evaluation. In: DP Purpura, JK Penry and RD Walter (Eds.), Advances in Neurology. Raven Press, New York, pp. 31–49. Dreifuss, FE (1987) Goals of surgery for epilepsy. In: J Engel, Jr (Ed.), Surgical Treatment of the Epilepsies, 1st edn. Raven Press, New York, pp. 31–49. Engel, J, Jr and Ojemann, GA (1993) The next step. In: J Engel, Jr (Ed.), Surgical Treatment of the Epilepsies, 2nd edn. Raven Press, New York, pp. 319–329. Fessler, JA, Cascino, GD, So, EL et al. (2000) Subtraction ictal SPECT co-registered to MRI (SISCOM) in the evaluation for repeat epilepsy surgery. Neurology, 54(Suppl. 3): A4. Hauser, A and Hesdorffer, D (1990) Prognosis. In: WA Hauser and DC Hesdorffer (Eds.), Epilepsy: Frequency, Causes and Consequences. Demos, New York, pp. 197–243. Hauser, W (1992) The natural history of drug resistant epilepsy: epidemiologic considerations. Epilepsy Res. Suppl., 5: 25–28. Henry, TR (1996) Functional neuroimaging with positron emission tomography. Epilepsia, 37: 1141–1154. Henry, TR, Babb, TL, Engel, J, Jr et al. (1994) Hippocampal neuronal loss and regional hypometabolism in temporal lobe epilepsy. Ann. Neurol., 36: 925–927.
154 Ho, SS, Berkovic, SF, Berlangieri, SU et al. (1995) Comparison of ictal SPECT and interictal PET in the presurgical evaluation of temporal lobe epilepsy. Ann. Neurol., 37: 738–745. Jack, CR, Jr, Trenerry, MR, Cascino, GD, Sharbrough, FW, So, EL and O’Brien, PC (1995) Bilaterally symmetric hippocampi and surgical outcome. Neurology, 45: 1353–1358. Marks, DA, Katz, A, Hoffer, P et al. (1992) Localization of extratemporal epileptic foci during ictal single photon emission computed tomography. Ann. Neurol., 31: 250–255. Mattson, RH (1992) Drug treatment of uncontrolled seizures. Epilepsy Res. Suppl., 5: 29–35. Mosewich, RK, So, EL, O’Brien, TJ et al. (2000) Factors predictive of the outcome of frontal lobe epilepsy surgery. Epilepsia, 41: 843–849. Newton, MR and Berkovic, SF (1996) Interictal, ictal, and postictal single photon emission computed tomography. In: GD Cascino and CR Jack, Jr (Eds.), Neuroimaging in Epilepsy: Principles and Practice. Butterworth-Heinemann, Boston, pp. 177–192. O’Brien, TJ, So, EL, Mullan, BP et al. (1996) Extent of resection of the ictal subtraction SPECT focus is an important determinant of epilepsy surgery outcome. Epilepsia, 37(Suppl. 5): S182. O’Brien, TJ, So, EL, Mullan, BP et al. (1998a) Subtraction ictal SPECT co-registered to MRI improves clinical usefulness of SPECT in localizing the surgical seizure focus. Neurology, 55: 445–454. O’Brien, TJ, O’Connor, MK, Mullan, BP et al. (1998b) Subtraction ictal SPECT co-registered to MRI in partial epilepsy: Description and technical validation of the method with phantom and patients studies. Nucl. Med. Commun., 19: 31–45. O’Brien, TJ, Zupanc, ML, Mullan, BP et al. (1998c) The practical utility of performing peri-ictal SPECT in the
G.D. CASCINO ET AL.
evaluation of children with partial epilepsy. Pediatric Neurol., 19: 15–22. O’Brien, TJ, So, EL, Mullan, BP et al. (1999a) Subtraction SPECT co-registered to MRI improves postictal localization of seizure foci. Neurology, 52: 137–146. O’Brien, TJ, Brinkmann, BH, Mullan, BP et al. (1999b) Comparative study of 99mTc-ECD and 99m Tc-HMPAO for SPECT: qualitative and quantitative analysis. J. Neurol. Neurosurg. Psychiatry, 66: 331–339. O’Brien, TJ, So, EL, Mullan, BP et al. (2000) Subtraction peri-ictal SPECT is predictive of extratemporal epilepsy surgery outcome. Neurology, 55: 1668–1677. Palmini, A, Andermann, F, Olivier, A et al. (1991) Focal neuronal migrational disorders and intractable partial epilepsy: results of surgical treatment. Ann. Neurol., 30: 750–757. Radhakrishnan, K, So, EL, Silbert, PL et al. (1998) Predictors of outcome of anterior temporal lobectomy for intractable epilepsy: a multivariate study. Neurology, 51: 465– 471. So, EL (2000) Integration of EEG, MRI and SPECT in localizing the seizure focus for epilepsy surgery. Epilepsia, 41(Suppl. 3): S48–S54. So, EL, O’Brien, TJ, Brinkmann, BH et al. (2000) The EEG evaluation of single photon emission computed tomography abnormalities in epilepsy. J. Clin. Neurophysiol., 17: 10–28. Theodore, WH (1996) Positron emission tomography in the evaluation of epilepsy. In: GD Cascino and CR Jack, Jr (Eds.), Neuroimaging in Epilepsy. Principles and Practice. Butterworth-Heinemann, Boston, pp. 165– 175. Wiebe, S, Blume, WT, Girvin, JP and Eliasziw, M (2001) Effectiveness and efficiency of surgery for temporal lobe epilepsy study group. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N. Engl. J. Med., 345: 311–318.