Sturge-Weber syndrome

* Background.-Surgical management of intractable epilepsy continues to be important in select cases to achieve seizure control.

Design.-This study retrospectively reviews the pathologic findings in 133 consecutive cases of extratemporal Lobe epilepsy experienced during a 17-year period.

Results.-The study group consists of 133 patients (78 mates) who underwent extratemporal lobe resection for epilepsy at a mean age of 21.1 years (range, 3 months to 57 years). In 50 patients (37.6%), cortical dysplasia (neuronal migration abnormalities) was identified. The most common patterns of dysplasia observed included diffuse architectural disorganization in 46 cases, neuronal cytomegaly in 30 cases, increased numbers of molecular layer neurons in 30 cases, and balloon cells in 18 cases, Tumors were identified in 37 cases (27.8%) and included 13 astrocytomas, 7 gangliogliomas, 6 dysembryoplastic neuroepithelial tumors, 6 glioneuronal hamartomas, 4 oligodendrogliomas, and 1 oligoastrocytoma (mixed glioma). Twen

ty-four resections (18%) showed evidence of remote ischemic damage or infarct. Neuronal heterotopia was identified in 59 resection specimens (44.4!). Other less common findings included vascular malformations in 4 patients (3.0%), Sturge-Weber malformations in 3 patients (2.3%), and Rasmussen encephalitis in 2 patients (1.5!). Two patients were known to have tuberous sclerosis. in 23 resection specimens (17.3%), no significant pathologic finding was identified. Coexistent cortical dysplasia and tumor were seen in 10 cases and coexistent dysplasia and infarct or remote ischemic damage in 11 cases.

Conclusion.-This series demonstrates that most patients with extratemporal lobe epilepsy have significant histopathologic findings, which most frequently include cortical dysplasia, tumor, or evidence of remote ischemic damage or infarct. Coexistent pathologic findings were present in a significant minority of cases (16.5!).

(Arch Pathol tab Med. 2000;124:545-549)

Since its inception a century ago, resection of seizure foci has become an accepted and effective means of treating pharmacoresistent epilepsy. Surgical resection of sei zure foci has become more common in the last decade.1,2 Most resections tend to involve the temporal lobe. Most cases result in good clinical outcome, defined as a signif icant reduction in postoperative seizure frequency. A number of series have reported worse outcome in patients with extratemporal seizure foci compared with temporal lobe-based epilepsy which is believed to be due in part to the more diffuse nature of extratemporal lobe lesions.2-6

Because of the relatively smaller numbers of cases of extratemporal lobe resection compared with temporal labe resections seen in most epilepsy centers, series examining the histopathologic findings in extratemporal-based chronic epilepsy have been significantly smaller. We reviewed our experience at the Cleveland Clinic Foundation with regard to the histopathologic findings in extratemporal lobe tissues resected in patients with pharmacoresistent epilepsy generated during a 17-year period. Particular attention is paid to the presence of coexistent lesions that have been independently associated with epileptogenesis.

MATERIALS AND METHODS

A computerized search of the surgical pathology files at the Cleveland Clinic Foundation was performed during a 17-year period (1981-1997). The search was for surgical resection specimens in patients with medically intractable epilepsy (ie, a number of unsuccessful attempts at various pharmacologic regimens were made before surgical intervention). Resection specimens in 133 consecutive patients were identified. All available histologic material was examined in each case, The number of slides reviewed ranged from 1 to 48 (mean, 12 slides). Half or more of the tissue submitted to the Department of Pathology were examined microscopically in 102 cases (77%); in 11 resections, less than half of the material submitted to the Department of Pathology was examined histologically. In 12 resections, the extent of tissue sampling histologically was not ascertainable from the pathology repast. Histologic sections were routinely generated from formalinfixed, paraffin-embedded tissues, sectioned 4 (mu)m thick and stained with hematoxylin-eosin.

Histopathologic findings in each case here documented. Particular attention was paid to the presence of cortical dysplasia, tumors, evidence of remote ischemic injury or infarct, the presence of heterotopic white matter neurons, vascular malformations, Sturge-Weber malformations, and Rasmussen encephalitis. Cortical dysplasia was defined as a maldevelopmental abnormality of the cortex relative to the neuronal component.7 Classification of the histologic patterns of cortical dysplasia was based in part on the schema outlined by Mischel et al and included the following: laminar cortical architectural disorganization that consisted of a loss of the normal laminar architecture of the cortex and/or a haphazard arrangement of neuronal cells within the cortex; neuronal cytomegaly characterized by enlargement of the cell body; increased numbers of molecular layer neurons; presence of the balloon cells characterized by abundant eosinophilic cytoplasm and displaced nuclei; white matter heterotopia marked by the presence of foci of white matter within the cortex; and glioneuronal clustering marked by aggregations of both neuronal and filial cells within the cortex.8 In addition, evidence of pialglioneuronal tissue was documented when present, and heterotopic neurons situated in the deep white matter were also noted when identified:

Tumors, when identified, were classified according to definitions outlined by the World Health Organization (WHO).9 Glialneuronal hamartomatous lesions were characterized by a circumscribed, disorganized collection of neuronal and filial cells that demonstrated minimal cytologic atypia.10 Mixed gliomas were defined as tumors consisting of 2 or more glioma cell types in which the minor glioma component comprised at least 20/ of the total tumor examined histologically.11

Vascular malformations, when identified, were classified according to definitions outlined by Burger and Scheithauer.12 Cases of Sturge-Weber disease were marked by proliferation of small to medium-sized venous and capillary vessels in the leptomeninges and underlying parenchyma, with extensive calcification and gliosis of the parenchyma; as previously described.13,14 Cases of Rasmussen encephalitis were marked by microglial nodule formation associated with perivascular chronic inflammation and mild gliosis as previously described.15,16

RESULTS

A total of 133 patients who underwent extratemporal lobe resections for medically intractable epilepsy comprise the study group. The patients who underwent surgical resection ranged in age from 3 months to 57 years at the time of surgery (mean age, 21.1 years; median age, 19 years). Seventy-eight patients were male, and 55 patients were female. Two patients each were known to have tuberous sclerosis and Davidoff-Dyke-Mason syndrome. One patient was known to have Coffin-Lawry syndrome. One patient had a concomitant Arnold-Chiari malformation.

Major pathologic findings are summarized in the Table. Fifty resections (37.6%) showed evidence of cortical dysplasia. More than half of these lesions were situated in the frontal lobe (n = 28, 56%). Smaller numbers were localized to the parietal lobe (n = 7) and occipital lobe (n = 4. In 11 additional cases, multiple lobes were involved by the dysplastic process. Histologically, the most common pattern of dysplasia observed was that of cortical laminar disorganization in 46 patients (92%) (Figure 1). Neuronal cytomegaly (Figure 2) and increased numbers of molecular layer neurons were observed in 30 patients each (60%). Balloon cells were seen in 18 cases (36%) (Figure 3). Less common forms of dysplasia observed included white matter heterotopia in 3 cases (6%) and glioneuronal clustering in 3 cases (6%). Pial-glioneuronal tissue was seen in 2 resections (4%). Many lesions were riot monomorphic in that more than one pattern of dysplasia was observed in most cases. In 26 patients (52%), 3 of more patterns of cortical dysplasia were observed. In 12 additional patients (24%), 2 patterns of cortical dysplasia were observed. Multiple neurons were observed in the deep white matter (neuronal heterotopia in 59 resections (44.4%).

Tumors were identified in 37 resections (27.8%). Twenty-six tumors arose in the frontal lobe, 5 in the parietal lobe, and 1 in the occipital lobe; in 5 tumors, more than 1 lobe was involved. The most commonly encountered tumors were astrocytomas (n = 13, 35.1%). The astrocytic tumors were classified as low-grade astrocytomas (WHO grades I and II lesions in 10 cases, and 3 tumors were classified as anaplastic astrocytoma (WHO grade III lesions). Gangliogliomas were observed in 7 instances (18.9%), dysembryoplastic neuroepithelial tumors in 6 cases (16.2%); glioneuronal hamartomas in 6 cases (16.2%), low-grade oligodendrogliomas in 4 cases (10.8%, and low-grade oligoastrocytoma (mixed glioma) in 1 case (2.7%. In 10 neoplasms, coexistent cortical dysplasia was identified. These tumors included 5 glioneuronal hamartomas, 3 dysembryoplastic neuroepithelial tumors, 1 ganglioglioma, and 1 low-grade fibrillary astrocytoma.

Remote infarct or ischemic lesions were identified in 24 resections (18.0%). In 20 cases, the pathologic lesion was confined to the frontal lobe; 2 lesions involved the parietal lobe and occipital lobe each. Histologically these lesions showed features of ischemia and infarct, including variable amounts of cystic change, gliosis, and neuronal loss. Areas of adjacent cortical architectural disorganization, at times indistinguishable from cortical dysplasia; were observed in 11 of these 24 resections. In most of these cases, these changes were felt to be related to the ischemic event.

Other less common, yet significant, findings that were felt to be causative of seizures included 4 examples of vascular malformations (3.0%), including 2 arteriovenous malformations, 1 cavernous angioma, and 1 malformation with sclerosed vessels that was not further classifiable as to type. Sturge-Weber malformations were identified in 3 instances (2.3%) and changes consistent with Rasmussen encephalitis in 2 patients (i.5%).

Nonspecific meningeal fibrosis was observed in 29 (21.8%) of 133 cases. Focal acute and/or chronic leptomeningeal inflammation was encountered in 48 resections (36.1%). These changes were felt to be related to invasive seizure monitoring (eg, implanted subdural electrode grids) performed before surgery.

COMMENT

This study is the largest series to date that examines extratemporal resections for epilepsy. The largest series published before this consisted of 63 extratemporal corticectomies, lobectomies, and functional hemispherectomies assembled by Wolf et al.17 In their series, structural lesions were identified in 85.7% of specimens, which is comparable to the results in this study (82.7%). Identification of lesions in most of these cases is probably due to the selection of candidates for epilepsy surgery; patients with an identifiable focal lesion on imaging studies are more likely to be offered surgery as a treatment option. The most common significant pathologic finding in Wolf's series was malformative in nature, consisting of a variety of lesions, including cortical dysplasia. The incidence of cortical dysplasia, however, in this series appears to be somewhat higher than that identified bv Wolf: 15.9% of cases in the Wolf series versus 37.6% in the current series. Differences with regard to the incidence of dysplasia between these 2 populations may in part be due to definitions used for cortical dysplasia. The descriptions of so-called glioneuronal hamartia described by Wolf is somewhat general and was defined as "focal defect in tissue organization during development."17 These lesions were described as generally consisting of neuronal and filial elements. The neuronal elements were oriented randomly and demonstrated cytologic pleomorphism. Six of the 9 hamartia lesions described presumably fit this pathologic profile. Specific descriptions of the 3 remaining cases were not forthcoming. In addition, one example of microgyria was also noted. As has been already evidenced in the literature, the spectrum of pathologic findings included under the guise of dysplasia varies.

The spectrum of cortical dysplasia observed in the 50 patients in this series were described along the histomorphologic lines outlined by Mischel efi al.8 Mischel et al reviewed the pathologic findings in 77 patients with cortical dysplasia and categorized the dysplastic findings into 9 specific microscopic abnormalities.8 Twenty-six of the patients underwent a hemispherectomy that involved 3 or more lobes, and 19 patients underwent a partial hemispherectomy that involved 2 or more lobes. The most common pathologic finding with regard to dysplasia in their experience involved cortical laminar disorganization, which likewise was the most commonly observed finding in the current series. Interestingly, they included single heterotopic neurons in the white matter as a distinct pattern of dysplasia of cortical dysplasia. Although a frequently observed pathologic finding in this series as well, neuronal heterotopia was not included in our study as a "cortical" dysplasia, since the primary defect is white matter based and of dubious significance with regard to epileptogenesis. Of particular significance was the fact that in most patients (76%) 2 or more patterns of dysplasia were observed in the same resection specimen. Compared with other series of temporal lobectomies,18-21 the incidence of cortical dysplasia in the extratemporal lobe resections appears to be a somewhat more frequent observance. This may be due in part to the fact that the most common pathologic finding of temporal lobe resections involve hippocampal or mesial temporal sclerosis and neoplasia. The presence of dysplasia in multiple lobes in most cases implies that; at least in a subset of patients with epilepsy and cortical dysplasia, the pathologic findings may be more diffuse than a single faces. Obviously, the extent of dysplasia or multifocality of dysplasia may have serious ramifications with regard to the relative effectiveness of the surgical procedure to improve seizures in this subset of patients.

The second most significant pathologic finding observed in extratemporal lobe resections involved neoplasms, which were encountered in slightly more than a quarter of the cases. If one includes the hamartoma described by Wolf et al under the heading of neoplasm, a somewhat similar incidence of 20.6!o was observed by Wolf and colleagues.17 The incidence of neoplasia in most of the temporal lobectomy series has been somewhat lower by comparison18-21; in contrast, Wolf et al reported a higher incidence of neoplasia (34.7%).18 The types of neoplasms most commonly observed in this clinical setting appear in general to be somewhat similar between these 2 groups. Most of the tumors fall into the categories of low-grade astrocytoma (WHO grades I and II), ganglioglioma, and dysembryoplastic neuroepithelial tumor. In our series, we also observed a significant number of glioneuronal hamartomas. These are somewhat unusual lesions that represent a circumscribed malformation comprised of neuronal and glial cells that demonstrate minimal cytologic atypia. In contrast to cortical dysplasia, these lesions represent a grossly apparent mass, although admittedly the distinction from dysplasia microscopically may be difficult.

As has been previously described in the literature, many epilepsy-associated neoplasms appear to be cortical dysplasia-associated lesions. In this particular series, evidence of coexistent cortical dysplasia was found in 26.3% of the tumors and most notably involved gangliogliomas, glioneuronal hamartomas, and the dysembryoplastic neuroepithelial tumor: All 3 of these lesions have been previous reported to be associated with dysplasia.10,22-27 Although not encountered in this series, the pleomorphic xanthoastrocytoma, which is more commonly temporal and parietal in location, also has been recently described in association with cortical dysplasia.28 The significance of this association of neoplasia with cortical dysplasia still remains somewhat debated. Whether these tumors arise from cortical dysplasia, whether they represent some peculiar tumoral form of dysplasia, or whether they merely coexist with dysplasia has yet to be ascertained. Nevertheless, the association implies that, similar to dysplasia, many of these neoplasms may be maldevelopmental in nature. Most of these lesions generally have an excellent prognosis and are potentially curable with a complete excision.

In a significantly few cases; epilepsy appeared to be associated with remote ischemic or infarctive events. Many of these events are presumably due to posttraumatic or prenatal or perinatal injury. Similar pathologic findings were observed in slightly more than 10% of the resection specimens examined by Wolf et al.17 The precise cause of these ischemic or infarctive events is not evident from examination of the tissue. Interestingly, a coexistent pattern of cortical dysplasia-like changes and infarct was seen in 11 cases. It is not unexpected that a disorganized cortical architecture may be the result of a localized ischemic event that occurs when the brain is developing.

Many of the other less frequently observed findings in this series, including vascular malformations, Sturge-Weber malformations, and Rasmussen encephalitis, have all been described in similarly small numbers in other series of resections for chronic epilepsy. Although these lesions are not as commonly encountered as the dysplasia, tumors, and ischemic and infarctive events, it is important to recognize that in a small percentage of cases the origins of the epilepsy may rest elsewhere.

In a significantly few cases (17.31o in this series), no obvious pathologic finding was discerned to explain the epilepsy. This may indicate that the origin for this seizure lies at the cellular or biochemical level, thereby making it difficult to detect by light microscopic examination. Alterations in neuronal structure, synaptic structure, interface between neuronal processes, or the biochemical milieu of a neuron or its environment may all potentially result in seizures.

Special thanks to Denise Egleton for her assistance in the preparation of the manuscript.

References

1. Engel J Jr: Current concepts: surgery for seizures: N Engl J Med. 1996;334: 647-652.

2. Cross JH, Jackson GD, Neville BGR, et al. Early detection of abnormalities in partial epilepsy using magnetic resonance. Arch Dis Child. 1994;69:104-109.

3. Silander HC, Blom S, Malmgren K, Rosen I, Uvebrant P. Surgical treatment of epilepsy: a retrospective Swedish multicenter study. Acta Neurol.Scand. 1997; 95:321-330.

4. Fish DR, Smith SJ, Quesney LF, Andermann F, Rasmussen T. Surgical treatment of children with medically intractable epilepsy: results and highlights of 40 years' experience. Epilepsia. 1993;34:244-247.

5. Holmes MD, Dodrill CB, Ojemann LM, Ojemann GA. Five-year outcome after epilepsy surgery in nonmonitored an

6. MacKenzie RA, Matheson JM, Smith JS, Dwyer M. Surgery for refractory epilepsy. Med )Aust. 1993;153:69-76.

7. Prayson RA, Estes ML. Cortical dysplasia: a histopathologic study of 52 cases of partial lobectomy in patients with epilepsy. J Neuropathol Exp Neural. 1995, 54:137-153.

8. Mischel PS. Nguyen LP, Vinters HV. Cerebral cortical dysplasia associated

with pediatric epilepsy: review of neuropathological features and proposal far a grading system. J Neuropathol Exp Neurol. 1995;54:137-153.

9. Kleihues P, Burger PC, Scheithauer BW. World Health Organization Histological Typing of Tumours of the Central Nervous System. New York, NY: SpringerVerlag; 1998.

10. Volk EE, Prayson RA. Hamartomas in the setting of chronic epilepsy: a clinicopathologic study of 13 cases. Hum PathoL. 1997;28:227-232.

11. Beckmann MJ, Prayson RA. A clinicopathologic study of 30 cases of oligoastrocytoma including X53 immunohistochemistry. Pathology. 1997;29:159164.

12, Burger PC, Scheithauer BW. Tumors of the Central Nervous System. Washington, DC: Armed Forces Institute of Pathology; 1994:287-292.

13. DiTrapani Cs, DiRocco C, Abbamondi AL, Caldarelli M, Pocchiari M. Light microscopic and ultrastructural studies of Sturge-Weber disease: Child Brain. 1982;9:23-36.

14. Prayson RA, Grewal ID, McMahon IT, Barna BP, Estes hlL. Leukocyte adhesion molecules and x-ray energy dispersive spectroscopy in Sturge-Weber disease. ) Pediatr Neuron 1996;15:332-336.

15. Rasmussen T, Olszewski J, Lloyd-Smith D. Focal seizures due to chronic localized encephalitis. Neurology 1958;8:435-445.

16. Farrell MA, Droogan O, Secor DL, Poukens V, Quinn B, Vinters H4: Chronic encephalitis associated with epilepsy: immunohistochemical and ultrastructural studies, Acta Neuropathol. 1995;89:313-321.

17. Wolf HK, Zentner j, Hufnagel A, et al. Surgical pathology of chronic epileptic seizure disorders: experience with 63 specimens from extratemporal corticectomies, lobectomies, and functional hemispherectomies. Acta Neuropathol 1993;86:466-472:

18. Wolf HK, Campos !vlC;, Zentner J, et al. Surgical pathology of temporal lobe epilepsy: experience with 216 cases. J Neuropathol Exp Neurol. 1993;52: 499-506.

19. Mathieson G. Pathology of temporal lobe foci. Adv Neurol. 1975;11:163185.

20. Babb TL, Brown WI. Pathological findings in epilepsy. In: Engle J, ed. Surgi(al Treatment of the Epilepsies. New York, IVY: Raven Press; 1987:511-540.

21. Bruton CI. The Neuropathology of Temporal Lobe Epilepsy: Oxford, England: Oxford University Press; 1988:

22. Prayson RA, Morris HH, Estes ML, Comair YG. Dysembryoplastic neuroepithelial tumor: a clinicopathologic and immunohistochemical study of 11 tumors including MIB-1 immunoreactivity. Clin Neuropathol. 1996;15:47-53.

23. Morris Hti, Matkovic Z, Estes Mt. et al. Ganglioglioma and intractable epilepsy: clinical and neurophysiologic features and predictions of outcome after surgery. Epilepsia. 1998;39:307-313.

24. Dumas-Duport C:. Dysembryoplastic neuroepithelial tumours. Brain Pathol. 1993:3:283-295.

25. Prayson RA, Estes ML, Morris fits. Co-existence of neoplasia and cortical dysplasia in patients presenting with seizures. Epilepsia, 1993;34:609-615.

26. Prayson RA, Khajavi K, Comair YG. Cortical architectural abnormalities and M tB-i immunoreactivity in gangliogliomas: a study of 60 patients with intracranial tumors. J Neuropathol Exp Neurol. 1995;54:513-520.

27. Wolf HK, Muller 418, Spanle M, Zentner I, Schramm ), Wiestler OD. Ganglioglioma: a detailed histopathological and immunohistochemical analysis of 61 cases. Acta Neuropathol. 1994;88:166-173_

28. Lach 8, Duggal N, DaSilva VE, Benoit BG. Association of pleomorphic xanthoastrocytoma with cortical dysplasia and neuronal tumors: a report of three cases. Cancer. 1996;78:2551-2563.

Accepted for publication September 9, 1999.

From the Departments of Anatomic Pathology (Drs frater and Prayson), Neurology (Dr Morris), and Neurosurgery (Dr Bingaman), Cleveland Clinic Foundation, Cleveland, Ohio.

Presented in part at the American Association of Neuropathologists Meeting, Minneapolis, Minn, June 1998.

Reprints: Richard A. Prayson, MD, Department of Anatomic Pathology IL25?, Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, Ot! 44195.

Copyright College of American Pathologists Apr 2000
Provided by ProQuest Information and Learning Company. All rights Reserved