Cavernous Malformations Presenting With Seizures Therapeutic Options and Outcome

Cavernous Malformations Presenting with Seizures: Therapeutic Options and Outcome

Kris A. S m i t h , M D , S a m J a v e d a n , M D , J o s e p h M. Z a b r a m s k i , M D , D a v i d E. B l u m , M D , a n d

R o b e r t F. S p e t z l e r , M D

Controversies exist regarding the appropriate initial management of patients with cavernous malformations associated with a seizure disorder. We retrospectively reviewed our experience with simple lesionectomy in the treatment of 82 patients with supratentorial cortical cavernous malformations who presented with seizures. Fifty-five patients had adequate follow up (at least 1 year; mean, 52 months) to assess outcome related to seizure control. Forty (73%)patients had single and 15 (27%) had multiple cavernous malformations. Eighteen (33%) patients un- derwent surgery within the first 2 months of their initial seizure. Thirty-seven (67%) patients had seizures for more than 6 months before their lesionectomy. All patients were treated with le- sionectomy alone as their initial management. Outcomes were evaluated based on Engel's criteria. Seizure outcome was based on the number of seizures pre- and postoperatively, lesion loca- tion, and number of lesions. Of the 55 patients with adequate follow up, 69% were seizure free (grade 1) and 13% were almost seizure free (grade II). All 18 patients who were operated on within 2 months of their initial seizure have remained seizure free. Of the patients with chronic epilepsy before surgery, 57% are seizure free, and 17% are almost seizure free. Patients with multiple cavernous malformations and well-localized seizure ac- tivity on preoperative testing responded well to lesionectomy. Lesion location was not significant in predicting seizure outcome after surgery. Lesionectomy is a valuable treatment for patients with cavernous malformations associated with epilepsy and rep- resents the best initial treatment option. Patients with chronic epilepsy who fail to become seizure free after lesionectomy may require further epilepsy monitoring and resection. Initial lesionec- tomy, however, will avoid unnecessary epilepsy surgery in most of these patients. Copyright 2002, Elsevier Science (USA). All rights reserved.

I n clinical studies, 34 to 81% of patients with cavernous malformations present with seizures. >~2 Because the risk of

hemorrhage is small in patients who present with seizures (less than 1%/lesion/yr), 2,1~ some authors question the need to treat lesions in this population surgically unless their epilepsy can- not be controlled with medications. 2,5 However, numerous re- ports document improved seizure control after the removal of cavernous malformations "lesionectomy" in patients with epi- lepsy. 1.3.4,6,8-1o.13-15

From the Divisions of Neurological Surgery and Neurology, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoe- nix, AZ

Address reprint requests to Kris A. Smith, MD, c/o Neuroscience Publications; Barrow Neurological Institute, 350 West Thomas Road; Phoenix, AZ 85013-4496; [email protected]

Copyright 2002, Elsevier Science (USA). All rights reserved. doi:l 0.1053/otns.2002.32483

In patients with epilepsy, health-related quality of life is most closely related to degree of seizure control. ~6 Surgery should maximize seizure control while minimizing the risks of the procedure. Lesion location may be a critical factor in determin- ing which lesions are best treated by lesionectomy rather than by more aggressive epilepsy operations. 6,z3a7 Cavernous mal- formations in the temporal lobe may kindle seizure loci in nearby limbic structures and thus reqmre a more aggressive surgical approach. 4,~Ba7 Because lesionectomy is less likely to be successful at seizure control when there is "dual pathol- ogy, "2-~8 it is important to rule out other causes of seizures (i.e., coexistence of mesial temporal sclerosis) before recommending a procedure:

In this study, we reviewed our experience with lesionectomy for seizure control in pauents with supratentorial cavernous malformations. 19-22

Methods

The clinical records of all patients who underwent resection of supratentorial cavernous malformations at the authors' institu- tion between 1984 and 1998 were retrospectively reviewed. We identified 82 patients who underwent resection lesionectomy of a supratentorial cortical cavernous malformation associated with at least one preoperative seizure. Of these 82 patients, 55 (22 males and 23 females) had adequate follow up (more than 1 year; median, 45 months; mean, 52 months) to assess out- come related to seizure control. All patients had histopatholog- ically confirmed cavernous malformations. Their mean age and median age were 30 and 32 years, respectively (range, 1 to 67 years).

The location of all resected lesions was recorded to assess differences in outcome. Two patients had lesions in the occip- ital lobe, 12 had lesions in the parietal lobe, 17 had lesions in the temporal lobe, and 24 had lesions in the frontal lobe.

Of the 55 patients, 40 (73%) had single cavernous malforma- tions and 15 (27%) had more than one lesion. Eighteen (33%) patients presented with a acute history of new seizures onset and underwent surgery within 2 months of their first seizure. Thirty-seven (67%) patients had more than a 6-month history of seizures before undergoing resection. No patients fell be- tween these two categories. Overall, patients had experienced seizures before surgery for a mean of 72 months (median, 12 months; range, 1 to 576 months).

Initially, all patients underwent simple lesionectomy of their cavernous malformation without extra-lesional cortical resec- tion or any attempt at electrocorticography (ECoG) to guide the resection. Hemosiderin-stained cortex adjacent to the le- sion was removed although such tissue was removed more judiciously in eloquent locations. The value of removing all

Operative Techniques in Neurosurgery, Vol 5, No 3 (September), 2002: pp 161-165 161

TABLE 1. Engers Classification of Postoperative Outcome

Class Description

I Free of disabling seizures* II Rare disabling seizures "almost seizure-free" III Worthwhile improvement-i- IV No worthwhile improvement1-

TABLE 3. Outcome of Lesionectomy as a Function of Number of Cavernous Malformations

No. lesions Class I Class II Class III Class IV

1 29 5 3 3 >1 10 1 2 2

*Excludes early postoperative seizures (first few weeks). -i-Determination of "worthwhile improvement" will require quantitative

analyses of additional data such as percent seizure reduction, cognitive function, and quality of life.

Modified from Engel J Jr: Surgical Treatment of the Epilepsies, 2nd edition, New York, Raven Press, 1993. With permission from Lippincott- Raven.

hemosiderin-stained tissue is unclear. 23 Since 1992, all cases have been performed with the assistance of frameless stereotac- tic guidance. The goal has been to minimize the size of the craniotomy directly over the lesion and to operate through cortical incisions that were as small as possible while safely and completely resecting the entire lesion. When possible, the le- sion was resected through an approach using an adjacent or overlying sulcus to minimize cortical trauma.

Clinical follow up of seizure control was obtained by review- ing the office charts of the treating neurosurgeons and referring neurologists. Phone interviews with patients and family mem- bers were conducted to determine the patients' status in terms of seizure occurrence and anticonvulsant medication usage, if any. Seizure outcomes were categorized according to Engel's criteria (Table 1).24 Outcomes based on chronicity of epilepsy, number of lesions, and lesion location were subjected to two- sided chi-square analysis.

patients who presented with one or two seizures and were operated on within 2 months of presentation were seizure-free at last follow up (Table 2). One had seizures in the initial postoperative period but now has rare auras 'only. Only two patients remain on antiepileptic medications. Of the 37 patients with chronic epilepsy (>6 months) before lesionectomy, 21 (57%) were grade 1, 6 (17%) were grade 2, 5 (14%) were grade 3, and 5 (14%) were grade 4 after surgery.

Of the 5 patients who faile d lesionectomy (Engel grade 4), one required subdural grid mapping and underwent further epilepsy surgery and had a good outcome. One patient re- mained seizure-free for 10 years after a lesionectomy of a right temporal malformation only to become symptomatic from a new lesion that developed in the left frontal Rolandic area. She continues to have focal motor seizures after lesionectomy of the second malformation. The remaining three patientg continue to be treated medically. There were no significant differences in outcome as a function of number of malformations (P = 0.7508, Table 3). Nor did outcome differ significantly (P = 0.4855) as a function of lesion location (Table 4) although patients with temporal and parietal lesions tended to have less favorable outcomes than patients with frontal or occipital le- sions.

Results

Of the 55 patients available for follow up, 38 (69%) were sei- zure-free (grade 1), 7 (13%)Were almost seizure-free (grade 2), 5 (9%) improved (grade 3), and 5 (9%) were unchanged (grade 4) after surgery (Table 2). No patients died in the perioperative period, and no major perioperative complications such as reop- eration for hematoma or infection were encountered. One pa- tient required reoperadon after the lesion Could not be localized at the initial surgery.

The difference in outcome between patients with acute and chronic epilepsy was statistically significant (P < 0.01). All 18

TABLE 2. Seizure Outcome after Lesionectomy in 55 Pat ients :

Outcomes : No. (%)

All patients Seizure-free (Engel 1) 38 (69) Rare seizures (Engel 2) 7 (13) Improved: frequency (Engel 3) 5 (9) No improvement (Engel 4) 5 (9)

Sporadic seizures* Seizure:free 18 (100) Off antiepileptic drugs 16 (89)

Chronic epi!epsyl Seizure~free 21 (57) Rare seiZure 6 (17) improved'frequency 5 (14) No improvement 5 (14)

* = <2' months before Surgery, 18 patients; 1 = >6,rnohths before surgery, 37 patients.

Discuss ion

Although cavernous ,malformations often manifest with sei- zures, their natural history is not well established. Recent arti- cles have focused primarily on mode of presentation, the risk ot hemorrhage, and subsequent complications associated with cavernous malformations as diagnosed by MRI. 2,r,~I Some au- thors have differentiated between the risks of future hemor- rhage in patients whose lesions manifest with hemorrhagic events and those who present with seizures or whose lesions are discovered incidentally. 2,5,r The risk of hemorrhage in the former group appears to be substantially higher than in the latter two groups, at least in the first year or tWO after hemor- rhage. 2,5,r Moriarty and coworkers 25 found that the overall an- nual risk of seizures was 4.8%/patient-year and 2.4%/patient- year in patients with no previous history of seizures at presentation.

Several articles on the management of cavernous malforma- tions associated with seizures have been published. 1,3-6,8-1~ The extent of surgical resection needed to afford the best chance of curing epilepsy caused by cavernous malformations

TABLE 4. Outcome of Lesionectomy as a Function of Lesion Location in 55 Patients

Location Class I Class II Class III Class IV

Frontal 20 3 0 1 Temporal 10 2 3 2 Parietal 6 2 2 2 Occipital 2 0 0 0

1 6 2 SMITH ET AL

Fig 1. (A) T2-weighted coronal and Tl-weighted (B) coronal and (C) sagittal enhanced magnetic resonance (MR) images of a frontal cavernous malformation in a 17-year-old girl who presented with a secondarily generalized seizure. The patient also had a large posterior fossa cavernous malformation as seen in (D) coronal T2-weighted and (E) Tl-weighted (MR) images, Both lesions were treated under a single anesthetic using frameless stereotactic microsurgery, The patient has remained seizure-free and neurologically normal and needed no anticonvulsant medications at a 4-year follow up,

is controversial. Some authors propose that in most cases sim- ple lesionectomy is effective and associated with less risk of neurologic morbidity than more extensive epilepsy operations that resect adjacent cortex based on ECoG. 4 In contrast, simple

lesionectomy to control seizures with loci in cortex adjacent to or at sites distant from the cavernous malformations is unlikely to be successful. 1Bar

Robinson and coworkers 1~ reported 21 patients with intrac-

CAVERNOUS MALFORMATIONS PRESENTING WITH SEIZURES 163

table epilepsy and cavernous malformations who underwent video-EEG telemetry. Ictal onset localized to the lesion's loca- tion in only three cases. Ictal onset localized to the lesion and surrounding areas in seven cases and was indeterminate or bilateral in 7 cases, The authors concluded that lesion removal often improved seizure control but recommended that patients undergo the same extensive preoperative evaluation as other patients undergoing epilepsy surgery. They also suggested that removal of an extralesional "epileptogenic zone" may be neces- sary in many cases.

Acciarri and coworkers ~ reported a cure or reduction in the use of antiepileptic medication in 20 of 36 patients (55%) who underwent lesionectomy and improved seizure control on a preexisting medical regimen in the remainder (16 of 36). Their surgical approach, similar to ours, involved lesionectomy with removal of surrounding gliosis when possible but did not in- volve corticography or preoperative ictal localization. The same group has since reported 100~ improved seizure control in children undergoing lesionectomy for epilepsy. They have pos- tulated that lesionectomy may help prevent the adjacent tissue from converting to an autonomous epileptogenic focus.

More recently, Casazza and coworkers 4 reported seizure-free outcomes in 23 of 26 patients who presented with sporadic seizures and in 13 of 21 patients who presented with chronic epilepsy. They used simple lesionectomy and found no, rela- tionship between residual hemosiderin on postoperat{ve MRI and the likelihood of a seizure relapse. Results improved if lesion location correlated with electroclinical data. Cohen and coworkers 6 found lesionectomy to be most successful in pa- tients with few preoperative seizures and short intervals be- tween seizure onset and surgery. They recommended lesionec- tomy as a reasonable strategy for patients with a brief history of seizures but suggested that patients with a longer preoperative history undergo more extensive evaluation before surgical treatment is pursued.

In our series, seizures were eliminated in all patients whose history of seizures was brief and in 57% of patients with chronic epilepsy treated with lesionectomy alone. Seizure control im- proved after lesionectomy in an additional 31~ of patients with chronic epilepsy. Results of lesionectomy were poorest in pa- tients with a history of intractable chronic epilepsy before sur- gery. None of our patients developed recurrent seizures after undergoing early lesionectomy. There were no deaths and no permanent neurological morbidity attributable to lesionec- tomy.

The results in this series suggest that lesionectomy for a cavernous malformation that presents with a the new onset of seizure a~tivity may be a reasonable alternative to medical man- agement to prevent epilepsy as long as the lesion is in a readily accessible-location and can be removed with minimal morbid- ity. It is unknown, however, how many of these patients would have become seizure-flee if they had received medical manage- ment alone. Extrapolating from clinical trials of anticonvul- sants in new or recent onset epilepsy suggests this number would be between 40 and 60%. 2~ More importantly, prophylac- tic lesionectomy for epilepsy may reduce these patients' need for long-term antiepileptic drugs and their associated side ef- fects. Of the 18 patients with medically controlled seizures before surgery, 16 (89%) were free of seizures and medications after surgery, and an attempt at weaning the other 2 patients from antiepileptic drugs is planned.

For patients with chronically uncontrolled epilepsy associ-

ated with cavernous malformations, lesionectomy alone helped most patients. Fifty-seven percent became seizure-flee and an additional 13% were almost seizure-flee after lesionectomy. Overall, these results are inferior to those reported for epilepsy surgery compared with lesionectomy. In many recent epilepsy surgery series, the seizure-flee outcome rates approximate 80%. 27'28 These outcomes suggest that patients with long- standing epilepsy, poorly controlled seizures, and/or a temporal lobe lesion should be evaluated in a more comprehensive fash- ion. If necessary, surgery should be planned to include more extensive resection than the limited lesionectomy used in our series. In particular, our study supports previous studies that patients with dual pathology should not undergo a simple le- sionectomy. However, a prospective, randomized, controlled trial would be needed to answer this question. A reasonable approach is to offer lesionectomy as the initial treatment to all patients with the realization that a significant minority will later require further investigation and resection of extralesional tis- sue for complete seizure control. However, more than 50% of the patients will have been rendered seizure-free without an invasive evaluation or, more importantly, unnecessary tissue resection.

Conclus ion

Lesionectomy is a valuable treatment option for patients with cavernous malformations associated with epilepsy, successful control of seizures is greatest in patients who undergo resection soon after the onset of symptoms. Patients with chronic epi- lepsy who fail to become seizure free after lesionectomy may require further epilepsy monitoring and resection. Initial le- sionectomy, however, will avoid unnecessary epilepsy surgery in most of these patients.

References 1. Acciarri N, Giulioni M, Padovani R, et al: Surgical management of

cerebral cavernous angiomas causing epilepsy. J Neurosurg Sci 39:13-20, 1995

2. Aiba T, Tanaka R, Koike T, et al: Natural history of intracranial cavernous malformations. J Neurosurg 83:56-59, 1995

3. Buckingham MJ, Crone KR, Ball WS, et al: Management of cerebral cavernous angiomas in children presenting with seizures. Childs Nerv Syst 5:347-349, 1989

4. Casazza M, Broggi G, Franzini A, et al: Supratentorial cavernous angiomas and epileptic seizures: preoperative course and postoper- ative outcome. Neurosurgery 39:26-34, 1996

5. Churchyard A, Khangure M, Grainger K: Cerebral cavernous angi- oma: a potentially benign condition? Successful treatment in 16 cases. J Neurol Neurosurg Psychiatry 55:1040-1045, 1992

6. Cohen DS, Zubay GP, Goodman RR: Seizure outcome after le- sionectomy fo r cavernous malformations. J Neurosurg 83:237-242, 1995

7. Del Curling O Jr, Kelly DL Jr, Elster AD, et al: An analysis of the natural history of cavernous angiomas. J Neurosurg 75:702-708, 1991

8. Giulioni M, Acciarri N, Padovani R, et at: Results of surgery in children with cerebral cavernous angiomas causing epilepsy. Br J Neurosurg 9:135-141, 1995

9. Kraemer DL, Awad IA: Vascular malformations and epilepsy: clinical considerations and basic mechanisms. Epilepsia 35:$30-$43, 1994

10. Montes JL, Rosenblatt B, Farmer J-P, et al: Lesionectomy of MRI detected lesions in children with epilepsy. Pediatr Neurosurg 22:167- 173, 1995

11. Robinson JR, Awad IA, Little JR: Natural history of the cavernous angioma. J Neurosurg 75:709-714, 1991

164 SMITH ET AL

12. Ryvlin P, Maugui~re F, Sindou M, et al: Interictal cerebral metabolism and epilepsy in cavernous angiomas. Brain 118:677-687, 1995

13. Jooma R,Yeh H-S, Privitera MD, et al: Lesionectomy versus elec- trophysiologically guided resection for temporal lobe tumors mani- festing with complex partial seizures. J Neurosurg 83:231-236, 1995

14. Robinson JR Jr, Awad IA, Magdinec M, et al: Factors predisposing to clinical disability in patients with cavernous malformations of the brain. Neurosurgery 32:730-736, 1993

15. Scott RM, Barnes P, Kupsky W, et al: Cavernous angiomas of the central nervous system in children. J Neurosurg 76:38-46, 1992

16. Leidy NK, Elixhauser A, Vickrey B, et al: Seizure frequency and the health-related quality of life of adults with epilepsy. Neurology 53: 162-166, 1999

17. Yeh H-S, Tew JM Jr, Gartner M: Seizure control after surgery on cerebral arteriovenous malformations. J Neurosurg 78:12-18, 1993

18. Fisher RS, Blum D: Epilepsy surgery where there is dual pathology. Lancet 354:267-268, 1999

19. McLachlan RS, Rose K J, Derry PA, et al: Health-related quality of life and seizure control in temporal lobe epilepsy. Ann Neurol 41:482- 489, 1997

20. Devinsky O, Vickrey BG, Cramer J, et al: Development of the quality of life in epilepsy inventory. Epilepsia 36:1089-1104, 1995

21. Vickrey BG, Hays RD, Engel JP Jr, et al: Outcome assessment for

24.

�9 25.

epilepsy surgery: the impact of measuring health-related quality of life. Ann Neurol 37:158-166, 1995

22, Vickrey BG, Hays RD, Graber J, et al: A health-related quality of life instrument for patients evaluated for epilepsy surgery. Med Care 30:299-319, 1992

23. Zevgaridis D, van Velthoven V, Ebeling U, et al: Seizure control following surgery in supratentorial cavernous malformations: a ret- rospective study in 77 patients. Acta Neurochir 138:672-677, 1996 Engel J Jr: Surgical Treatment of the Epilepsies. New York, Raven Press, 1993 Moriarity JL, Wetzel M, Clatterbuck RE, et al: The natural history of cavernous malformations: a prospective study of 68 patients. Neu- rosurgery 44:1166,1173, 1999

26. Mattson RH, Cramer JA, Collins JF, et al: Comparison df carbamaz- epine, phenobarbital, phenytoin, and primidone in partial and sec- ondarily generalized tonic-clonic seizures. N Engl J Med 313:145- 151, 1985

27. Sperling MR, O'Connor M J, Saykin A J, et al: Temporal Iobectomy for refractory epilepsy. JAMA 276:470-475, 1996

28. Radhakrishnan K, So EL, Silbert PL, et al: Predictors of outcome of anterior ~emporal Iobectomy for intractable epilepsy: a multivariate study. Neurology 51:465-471, 1998

CAVERNOUS MALFORMATIONS PRESENTING WITH SEIZURES 165