Cortical Dysplasia in Temporal Lobe Epilepsy: Magnetic Resonance Imaging Correlations Ruben Kuzniecky, MD,”Julio H. Garcia, MD,f Edward Faught, MD,” and Richard B. Morawetz, MD1‘

Cortical dysplasia has been documented in histological specimens surgically removed for treatment of refractory temporal lobe epilepsy. We studied 10 patients with cortical dysplasia and complex partial seizures who underwent temporal lobectomy. Magnetic resonance imaging revealed abnormalities in 5 of the patients who had microscopically detectable major abnormalities. Magnetic resonance imaging revealed an abnormal cortical-white matter architectonic pattern in 2 patients with moderate cortical dysplasia. I n the remaining 3 patients, magnetic resonance imaging findings were unremarkable. These observations suggest that magnetic resonance imaging is sensitive in the d e t e d o n of certain dysplastic lesions in temporal lobe epilepsy. Preoperative identification of these abnormalities by magnetic resonance imaging may permit early and optimal surgical treatment in patients with refractory epilepsy. Kuzniecky R, Garcia JH, Faught E, Morawetz RB. Cortical dysplasia in temporal lobe epilepsy: magnetic resonance imaging correlations. Ann Neurol 1991;29:293-298

Temporal lobectomy is an effective, yet underused, method of treatment for most patients with medically resistant temporal lobe epilepsy [ 1). Although electroencephalographic (EEG) information remains crucial in defining the epileptogenic area, there is considerable controversy regarding the techniques used to identify the epileptogenic focus. The presence of abnormalities-on-imaging procedures, in congruence with electrographic and other localizing data, often permits a rational decision for surgical treatment. Computed tomographic (CT) scanning is usually unrevealing in patients with temporal lobe epilepsy [2-4). Magnetic resonance imaging (MRI) is unquestionably superior to CT scanning in the detection of small epileptogenic structural lesions such as tumors or small vascular malformations [5 , 6). Although controversial evidence exists [7), we have reported that with appropriate techniques, MRI is sensitive in detecting mesial temporal sclerosis [ S ] , which is present in about 60% of patients with temporal lobe epilepsy fs]. More recently, with further refinement of our techniques, we have demonstrated hippocampal atrophy in association with other abnormalities in patients with mesial temporal sclerosis

r91. Over the past several years, many studies have described a variety of histological abnormalities in temporal lobe epilepsy [10-12]. In 1971, Taylor and colleagues [ 13) reported on the histological abnormalities found in surgical specimens removed from a group of patients with intractable partial epilepsy. They described “cortical dyslamination” associated with abnorFrom the Univcrsity of Alabama at Birmingham Epilepsy Center, Deparrments of *Neurology, ?Neurosurgery, and $Pathology, University of Alabama at Birmingham, AL.

mal giant neurons. They termed this abnormality “focal cortical dysplasia” and ascribed ir to a developmental malformation. More recently, this pathological entity has been revived and expanded to include mild forms of cortical and subcortical neuronal derangements 114-17). In recent studies, we reported on the usefulness of MRI in the diagnosis and management of various epileptic disorders caused by developmental migration disorders {18, 19). In this report, we describe the MRI findings in a group of patients with histological focal dysplasia of the temporal lobe and complex partial seizures.

Methods Ten patients were entered into this study. They were selected from a group of consecutive patients who underwent temporal lobe resections for medically intractable temporal lobe epilepsy at the University of Alabama at Birmingham Epilepsy Center (Birmingham,AL). The patients were selected on the basis of ( I ) a histological verification of cortical dysplasia and (2) a high-quality preoperative MRI scan of the brain. All patients were evaluated with prolonged EEG-video monitoring (scalp = 10; intracranial electrodes = 7) and neuropsychological studies including bilateral intracarotid sodium Amytal testing. All patienrs had CT scans with and without contrast agents (GE 9800, General Electric, Milwaukee, WI).

MRI Studies The MRI examinations were performed on two different units. A Picker (Cleveland, OH) operating at a field strength of 0.5 T and a Siemens Magneton GBS-2 (Erlangen, Germany) operating at a field strength of Received May 30, 1990, and in revised form Aug 10. Accepted for publication Sep 9. 1990.

Address correspondence to Dr Kuzniecky, Deparcment of Neurology, UAB Station, Birmingham, AL 35294.

Copyright (0 1391 by the American Neurological Association

293

1.5 T were used. Spin-echo sequences with echo times (TE) of 25 to 100 msec and repetition times (TR) of 2,000 to 2,500 msec were performed in the axial and coronal planes. T1-weighted images were obtained with T R of 50 msec and TE of 15 msec in 3 patients. Three patients underwent repeated studies in both units. Gadolinium-diethylenetriamine pentaacetic acid (DTPA)-enhanced MRls were performed in 3 patients. Coronal and axial images with a T R of 70 msec and TE of 20 msec were obtained after the administration of 0.2 ml/kg of intravenous gadolinium-DTPA.

Clinical Feuturex Age at onset (yr) Full scale IQ Seizure frequency Partial complex seizures Secondary generalization

4.1 (range, 0.7-13 years) 90 (range, 70-112) 8.8 per month (range, 6-20 months) 10 patients 8 patients

able pregnancy and birth histories. There were 6 females and 4 males aged 7 to 41 years (mean, 24 years). Additional information is presented in the Table.

Histological Studies The surgical resection was done by the subpial aspiration technique. The resection included a standard, en bloc neocorticectomy, sparing the superior temporal gyrus. This technique has been described in detail elsewhere 1201. The mesial temporal structures including uncus, amygdala, and anterior hippocampal formation were also resected. Because of the surgical technique used, the middle and inferior temporal gyri were the best available anatomical structures for histological examination. After gross examination, the specimens were fixed in a buffered aldehyde and embedded in paraffin. Histological sections were stained with hematoxylin-eosin and cresyl violet (Nissl). Axons were demonstrated by silver impregnation (Sevier-Munger technique). All specimens were reacted with antisera for glial fibrillary acidic protein and neuron-specific enolase. In some patients, additional immunohistochemical reactions included antiserum for synaptophysin. The histological diagnosis was done following the criteria established by previous studies 113-17). All tissue sections were examined and arbitrarily classified into two groups. Group 1 consisted of tissue sections with mild dysplastic changes and Group 2 included specimens with severe dysplasia. This subdivision was based on the degree of histological abnormalities found on the specimens. The abnormalities were graded from slight to severe as follows: (A) multiple clusters of 5 to 10 neuronal aggregates in white matter (high-power field), (Bj changes in orientation of neurons with poor cortical lamination, ( C ) cortical dyslamination without giant neurons, (Dj neuronal clustering with bare areas within cortex, (E) cortical dyslamination with abnormal giant neurons and large reactive astrocytes. Specimens with abnormalities A, B, or C, or any Combination of A, B, and C, were classified in Group 1. Specimens with additional abnormalities such as D and E were included in Group 2.

Results Clinical Features Febrile convulsions were identified by history in 4 patients as a possible eciological factor. All had unremark294 Annals of Neuroloev Vol 29 No

3,

March 1991

Electroencephalographic Studies Scalp EEG recordings showed predominantly unilateral temporal lobe interictal abnormalities in 8 patients. In 1 patient, bilateral independent but predominantly unilateral epileptiform abnormalities were recorded. In the remaining patient, secondary bilateral synchronous spike and wave discharges in conjunction with a temporal focus were observed. The distribution of epileptiform discharges revealed that most abnormalities were located over anterior-lateral temporal and frontal neocortex and less frequently over the mesial sphenoidal electrodes. Intracranial studies were performed in 7 of the 10 patients (6 epidural electrodes, 1 foramen ovale electrode). In 3 patients, predominantly mesial EEG ictal onsets were observed; in another 3 patients, the ictal pattern was widespread over the lateral temporal neocortex; and in the last patient, a unilateral widespread predominantly frontotemporal pattern was observed. The interictal and ictal EEG studies indicated a left temporal focus in c) of the 10 patients. MRI Ahnormalities MRI studies revealed abnormalities in 7 of the 10 patients. The pattern of changes was variable between patients but was restricted to one of the temporal lobes. In 4 patients, the MRI abnormalities consisted of nonhomogeneous increased signal-intensity lesions within the anterior temporal, inferior and middle temporal gyri, and extending into the white matter. In 2 of these patients, the lesions extended into the mesial structures. T1-weighted images demonstrated an abnormal organization of the cortical and white matter pattern in certain patients (Fig 1). These abnormalities were clearly detected in the T2-weighted images (Figs 2A, 3A). Gadolinium-enhanced scans did not reveal any changes in these patients (Fig 2B). In 2 patients, the MRI abnormalities were characterized by increased signal-intensity lesions on the inferior temporal gyrus with limited extension into the adjacent mesial temporal structures. The lesions were nonhomogeneous, without mass-effect, and with the cortex focally thickened (Fig 4A). The MRI in the seventh patient revealed a less-defined cortical-white matter

circumscribed abnormalities in 2 of the 5 patients (see Fig 4 A , B).

Follow-Up

All

A

B

Fig 1. (A,B) TI -weighted image demomtrates an abnormal cortical white matter organization. Note thick gyrus and white matter pattern i n the 1eJt anterior temporal lobe (arrows).

architectonic pattern in the affected temporal lobe, with a minimally increased abnormal signal intensity from the cortex (Fig 4B).

M R I Pathological Correlations Based on pathological subgroups, 5 patients were classified in Group 2 (see Figs 2C, 3B). In this group, MRI revealed pronounced abnormalities in all patients (see Figs 2A, 3A). The remaining 5 patients were classified in Group 1. MRI correlation in this group revealed

Fig 2. (A)Magnetic resonance irnaging-T2-~~eightedsequence showing increased signal in the Left anterior temporal lobe. (B) Gadolinium-diethylenetriamine pentaacetir acid scan oJthe same tecpatient produced no abnormal mhanctmnt. (C) Micro~~ropiition showing a cluster of neurons i n deep white matter. (Hematoxylin-eoJin; original magnification, x 400.) Other abnormalities included cortical &lamination with bare areas within cortex.

patients underwent temporal lobe resections as previously described. Follow-up ranged from 6 to 40 months, with a mean of 24 months. Six patients have been seizure free since surgery. Three patients have had 3 greater than 80% reduction in frequency of attacks. One patient was seizure free for 8 months, but seizures have recurred. Postoperatively, 1 patient developed a mild hemiparesis that has improved. Discussion The term cortical dysplasia was first introduced by Taylor and colleagues in 1971 [13], but it was Roncoroni {Zl} who made the first comments on the relation between epilepsy and maturational brain disturbances. Over the last several years, numerous studies have described neuronal microdysgenesis in patients with primary generalized epilepsy {22) and partial epilepsies { 15- 17). Several quahtative studies have reported focal cortical dysplasia and milder forms of dysplasia in patients with refractory complex partial seizures [l4-16, 233. More recently, quantitative studies have revealed increased neuronal microdysgenesis in patients with temporal lobe epilepsy than in age-matched controls { 171. Our patients represent examples of mild to severe cortical dysplasia. We specifically excluded specimens with minimal degrees of neuronal ectopia because, without cell-density counts, the significance of these findings is questionable. Hippocampal pyramidal ectopic neuronal aggregates were present in one speci-

Kuzniecky et al: Dysplasia, MRI, and Temporal Lobe Epilepsy

295

A

B

Fi g 3. (Aj Magnetic resonance imaging-T2-weighted image showing a nonhomogeneous high-intensity J.ignal lesion in the left mesial and anterior temporal lobe. (B) Example of one of the several abnomzalities in this patient; cortical dyslamination including large neurons and astrocjites. (Hematoxylin-eosin;original magnification, x 400.)

Fig 4. (AJMagnetic resonance imaging (MRl)-T2-~eightedimage revealing a well circumscribed high-intensity abnomality in the left posterior mesial temporal lobe with extension into white matter (arrows). (Bi MRI-T2-weighted image, axial. The normal right temporal lobe cortical-white mutter definition is indistinct in comparison with contralateralside. Loss of u?hite matter digitations ir noted (arrows).

men. In another 2 patients, structural abnormality was not found in the mesial structures. Because of the limitations imposed by the surgical technique, which preclude preservation of topographical relations, we cannot make meaningful comments on the exact distribution of the dysplastic changes among gyri. The data derived from MRI, however, makes it plausible that the histological abnormalities are more extensive in some patients and circumscribed in others. MRI revealed abnormalities in 7 of our patients. Nonhomogeneous high-signal-intensity lesions were

296 Annals of Neurology Vol 29 No 3 March 1991

detected in 5 patients; however, variability was observed with some lesions extending into the white matter, whereas others were more localized (see Fig 4A). In some patients, the temporal cortical ribbon appeared thick. T2-weighted images provided best visualization of the lesions. Gadolinium-DTPA added no information to the images, suggesting but not excluding nonneoplastic lesions. The normal cortical-white matter architectonic pattern of the temporal lobe was affected in some of our patients. This abnormality consisted of poor whiteigray matter demarcation with focal thickening of the cortex (see Fig 1). The significance of these findings was controversial at first, but became consistent when several MRIs revealed similar abnormalities. Furthermore, temporal lobe atrophy was not seen in these patients. Consideration of the specificity of the MRI findings in our patients is important. Previous studies reported nonspecific high-intensity signals from the mesial structures in patients with temporal lobe epilepsy. Pathological correlations demonstrated a variety of histological abnormalities ranging from mesial temporal sclerosis to small tumors [S-7). More recently, however, our studies have shown that certain distinctive imaging findings have emerged in these patients. It is now clear that mesial temporal sclerosis, the most common pathological abnormality in temporal lobe seizures, is distinguished by the presence of hippocampal atrophy associated with increased signal intensity on T2-weighted images [ 5 , 91. Small structural lesions may appear as high-intensity abnormalities, however, other features such as calcifications and variable mass-effect are usually present, and we have observed the absence of focal atrophy or temporal horn dilatation in these patients. Although variable nonspecific high-intensity ab-

Fig 5 . Mild dysplaJia in a patient with n o w 1 magnetic resonance imaging: neuronal clustering and dyslamination in hjier I-II of the neocortex. (Hematoxylin and eosin; originul rnagnz3cation, x 400.1

normalities were observed among our patients, certain imaging features such as an abnormal corticalwhite matter architectonic pattern and the presence of gyral thickening may suggest the underlying pathological substrate. Further studies are necessary to confirm these preliminary observations. In this study, we attempted to correlate the histological observations with the MRI abnormalities. It is important to mention, however, that because of the surgical technique, limited histological material was available for study, and therefore, accurate correlation was not possible. We found a moderate correlation between the severity of dysplastic features in the histological specimens and the MRI findings. Patients with severe dysplastic lesions (Group 2) (see Figs 2, 3) had major MRl abnormalities, whereas, patients with mild to moderate dysplasia (Fig 5 ) correlated with more subtle and restricted lesions on imaging or with normal MRIs (3 patients with group 1 histology had normal MRIs). The preceding findings are expected. Tl-

weighted images were best to demonstrate the abnormal cortical-white pattern. T2-weighted images demonstrated abnormal increased intensity signal lesions, however, in those patients with major histological changes. The abnormality on T2-weighted scans may be related to increased water content in the abnormal cells, which consequently induces an increase in the T 2 relaxation time. Previous studies have suggested an unfavorable postoperative outcome for patients with cortical dysplasia. Bruton [l 1) recently reviewed 249 patients with temporal lobectomy; cortical dysplasia was found in 8 patients (3.2%), and the outcome was unfavorable in all. In Taylor’s original publication 1 131, however, only 1 of the 5 patients who underwent temporal lobectomy failed to improve, whereas the other 4 had experienced seizure relief. Goldring [24} and others {25}, as well as our data, suggest a more favorable clinical outcome in most patients with temporal lobe epilepsy and cortical dysplasia who are treated surgically. MRI has made a significant impact on the investigation and management of candidates for epilepsy surgery [ 5 , 6, 18, 261.In addition to being more sensitive than CT scanning in the detection of small epileptogenic structural lesions, MRI has the ability to demonstrate mesial temporal sclerosis in the vast majority of patients with temporal lobe epilepsy if the appropriate imaging techniques are used 15, 91. MRI has shown to be very valuable in the investigation of developmental abnormalities of the brain often associated with severe epilepsy [19, 27-29]. We previously reported on the ability of MRI to detect focal cortical dysplasia in patients with developmental Rolandic lesions [ 181. Others have also reported abnormal MR images in isolated patients with cortical dysplasia of extratemporal origin [30, 31). Chugani and colleagues [32} recently reported 5 patients with cryptogenic infantile spasms and focal dysgenesis. MRI revealed subtle abnormalities in 1 patient. Although the pathological data was limited by the surgical technique, the present study suggests that MRI is sensitive in the detection of certain dysplastic lesions in the temporal lobes. Furthermore, we have found that the present MRI resolution permits identification of lesions with moderate to severe histological abnormalities but not of those patients with mild neuronal and cortical abnormalities (Group 1). The preoperative identification of dysplastic lesions by MRI is important because these histological changes have been found consistently in brain specimens surgically removed for the treatment of refractory partial epilepsy [16-18]. These findings are, perhaps, most important if one considers that these epileptogenic developmental lesions are highly resistant to medical treatment and are common in the pediatric population [18, 24, 27, 33, 341. Visualization of these lesions by MRI should allow early diagnosis and surgical interven-

Kuzniecky et al: Dysplasia, MRI, and Temporal Lobe Epilepsy 297

tion in children with refractory focal epilepsy. In those patients without obvious MRI abnormalities, although not diagnostically specific, positron emission tomography (PET) may reveal functional metabolic disturbance in the regions of cytoarchitectonic abnormalities, as suggested by Chugani et al(32). As noted in their study [ 3 2 ) and demonstrated in some of our patients, however, subtle MRI abnormalities such as an abnormal cortical-white matter architectonic pattern may be present and may be easily overlooked. Studies combining MRI, PET, and microscopy are necessary to define the usefulness and the limitations of each technique and to understand the biology of these developmental lesions.

Presented in part at the 42nd Annual Meeting of the American Academy of Neurology, Miami, FL, May 1990. We thank Yvonne Zelenka K. for her helpful criticism, and Or S. Berkovic for reviewing the manuscript.

References 1. Ojeman G. Surgical therapy for medically intractable epilepsy. J Neuroszq 1987;66:489-499 2. Blom R, Vinuela F, Fox AJ, et al. Computed tomography in temporal lobe epi1epsy.jComput Assist Tornogv 1984;8:10 1-405 3. Bozdanoff B, Stafford C, Green L, et al. CT in the evaluation of patients with focal epilepsy. Neurology 1975;25:1013-1017 4. Jabbari B, Huott AD, DiChiro G, et al. Surgically correctable lesions detected by CT in I43 patients with chronic epilepsy. Surg Neurol 1978;10:319-322 5. Kuzniecky R, de la Sayette V, Ethier R, et al. Magnetic resonance imaging in temporal lobe epilepsy: pathological correlations. Ann Neurol 1987;22:341-347 6. Heinz E, Heinz T, Radke R, et al. Efficacy of MR vs CT in epilepsy. AJNR 1988;9:1123-1 128 7. Sperling M, Wilson G, Engel J, et al. Magnetic resonance imaging in intractable partial epilepsy: correlative studies. Ann Neurol 1986;20:57-62 8. Babb TL, Brown WJ. Pathological findings in epilepsy. In: Engel J Jr, ed. Surgical treatment of the epilepsies. New York: Raven Press, 1987:511-540 9. Berkovic S, Andermann F, Ethier R, et al. Hippocampal sclerosis in temporal lobe epilepsy demonstrated by MRI. Ann Neurol (in press) 10. Mathieson G. Pathology of temporal lobe foci. In: Penry JR, Daly DD, eds. Complex partial seizures and their treatment. New York: Raven Press, 1975:163-185 11. Bruton CJ. The neuropathology of temporal lobe epilepsy. Oxford: Oxford University Press, 1988:l-158 12. Bahb T, Brown W, PretoriousJ, et al. Temporal lobe volunietric cell densities in temporal lobe epilepsy. Epilepsia 1984;6: 729-740 13. Taylor D, Falconer M, Bruton C, et al. Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiatry 197 1;14:369-387

298

Annals of Neurology

Vol 29 No 3 March 1991

4. Andermann F, Olivier A, Melanson D, et al. Epilepsy due to focal cortical dysplasia with macrogyria and the forma fruste of tuberous sclerosis: a study of 15 patients. In: Wolf P, Dam M, Janz D, Dreifuss E, eds. Advances in epileptology, vol 16. New York: Raven Press, 1987:35-38 5. Nordberg C, Sourander P, Silvenius H , et al. Mild cortical dysplasia in patients with intractable partial seizures. A histological study. In: Wolf P, Dam M, Janz D, Dreifus E, eds. Advances in epileptology, vol 16. New York: Raven Press, 1987:29-

33 16. Armstrong D, Bruton C. Postscript: What terminology is appropriate for tissue pathology: how does it predict outcome? In: Engel J Jr, ed. Surgical treatment of the epilepsies. New York: Raven Press, 1987541-552 17. Hardiman 0, Burke T, Phillips J, et al. Microdysgenesis in resected teniporal neocortex: incidence and clinical significance in focal epilepsy. Neurology 1988;38:1041-1047 18. Kuzniecky K, Berkovic S, Andermann F, et al. Focal cortical myoclonus and Rolandic cortical dysplasia: clarification by MRI. Ann Neurol 1988;23:317-325 19. Kuzniecky R, Andermann F, Tampieri D, et al. Bilateral ccntral macrogyria: epilepsy, pseudobulbar palsy and mental retardation-a recognizable neuronal migration disorder. Ann Neurol 1989;25:547-554 20. Kutniecky R, Faught E. Pre-operative extracranial and epidural electrodes in temporal lobe epilepsy. Neurology 1989;39:112 (Abstract) 2 1. Roncoroni G. Die Histologie der Stirnlapperr bei Verbrechen und Epileptikern. Wien Klin Rundsch 1897;6:7-8 22. Meencke HJ, Janz D. Neuropathological findings in primary generalized epilepsy: a study of eight cases. Epilepsia 1984; 25:8-21 23. Leitenmaier K, Neimanis G. Umschriebene Dysplasien der Tempordlappens Lei Psychomotorischer Epilepsie. 2 Neurol 197 3 ;204:23-42 24. Goldring S. Surgical management of epilepsy in children. In: Engel J Jr. ed. Surgical treatment of the epilepsies. New York: Raven Press, 1987:445-464 25. Moreland DB, Glasaver FE, Egnatchick JG, et al. Focal cortical dysplasia. J Neurosurg 1988;68:487-490 26. Schorner W, Meencke J, Felix R. Temporal lobe epilepsy: comparison of CT and MRI. AJR 1987;149:1231-1239 27. Krawinkel M, Steen H, Terwey B. MRI in lissencephaly. Eur J Pediatr 1987;146:2O5-208 28. Barkovich J, Chuang S, Norman D. MR of neuronal migration anomalies. AJNR 1987;8:1009-1017 29. Van der Knaap M, Valk J. Classification of congenital abnormalities of the CNS. AJNR 1988;9:315-326 30. Nowell M, Grossman R, Packer R, et al. Focal cortical dysplasia on MRI: a case report. J Comput Tomogr 1988;12:61-

63 3 I . Sellier N, Kalifa G, Lalande G, et al. Focal cortical dysplasia: a rare cause of epilepsy. Ann Radio1 1987;30:439-445 32. Chugani H, Shields D, Shewmon A, et al. Infantile spasms: PET identifies focal cortical dysgenesis in cryptogenic cases for surgical treatment, Ann Neurol 1990;27:406-413 33. Engel J Jr, Andermann F, Duchowny M, et al. Epilepsy surgery workshop. Clevc Clin J Med 1983;56(suppl I):S84-S91 34. Madsen P, Peacock W. Pathology of lesions resected during surgery for intractable seizures of pediatric patients. Cleve Clin J Med 1989;56(supplIkS282