Epilepsia, 40(1 1):1551-1556, 1999 Lippincott Williams Kr Wilkins, Inc., Philadelphia 0 International League Against Epilepsy

Clinical Research

Gamma Knife Surgery for Mesial Temporal Lobe Epilepsy

Jean Rkgis, *?Fabrice Bartolomei, *Marc Rey, ?Pierre Genton, ?Charlotte Dravet, $Franck Semah, j- Jean-Louis Gastaut, ?Patrick Chauvel, and Jean-Claude Peragut

Stereotactic and Functional Neurosurgery Department, Timone Hospital: *Neurophysiology Department, INSERM CJF 9706, Timone Hospital: and f “Centre Saint Paul,” Marseilles; and $Service hositalier F. Joliot, INSERM U3334, CEA, Orsay, France

Summary: Purpose: Gamma knife radiosurgery (GK) allows precise and complete destruction of chosen target structures containing healthy andor pathologic cells, without significant concomitant or late radiation damage to adjacent tissues. All the well-documented radiosurgery of epilepsy cases are epilep- sies associated with tumors or arteriovenous malformations (AVMs). Results prompted the idea to test radiosurgery as a new way of treating epilepsy without space-occupying lesions.

Methods: To evaluate this new method, we selected seven patients with drug-resistant “mesial temporal lobe epilepsy” (MTLE).The preoperative evaluation program was the one we usually perform for patients selected for microsurgery of TLE [video-EEG analysis of seizures, foramen ovale electrode re- cording, magnetic resonance imaging (MRI) positron emission tomography (PET) scan, neuropsychological testing]. In lieu of microsurgery, the amygdalohippocampectomy was performed by using GK radiosurgery.

The idea of treating epilepsy by radiation is a dated one. Between 1955 and 1973, Talairach and colleagues (1) treated 44 patients with epilepsy by stereotactic im- plantation of yttrium 90 into the amygdala and the hip- pocampus. Difficulties related to the technical use of the radioelement led to the cessation of this approach, but the first clinical results were promising. Some years later, the cessation of seizures after radiosurgical treatment of arteriovenous malformations (AVMs) (2,3) led Linquist et al. (4) to advocate radiosurgery as a way of treating epilepsy. Gamma knife (GK) radiosurgery allows, in a single session, “precise and complete destruction of cho- sen target structures containing healthy and/or pathologi- cal cells, without significant concomitant or late radia- tion damage to adjacent tissues” (5). Until now, all the

Accepted April 28, 1999. Address correspondence and reprint requests to Dr. J. RCgis at Ser-

vice de Neurochirurgie Fonctiunnelle et SterCotaxique, C.H.U. La Ti- mone, 264 rue Saint Pierre, 13385 Marseille CEDEX 05, France. E-mail: jregis@ap-hm.fr

Results: Morphologic (MRI) signs of destruction of the target took place at 9 months after GK surgery. Since the treatment day, the first patient has been seizure free. Seizure improve- ment came more gradually for the following patients, and com- plete cessation of seizures occurred around the tenth month (range, 8-1 5 months). MRI shows that the amygdaloentorhi- nohippocampal target was selectively injured. No significant side effect (except one case of homologous quadrantanopia) or morbidity and no mortality was observed. The current follow- up is 24-61 months, and all (but one) patients are seizuiz free.

Conclusions: This initial experience proves clearly the short- to middle-term efficiency and safety of GK for MTLE surgery. These results need further confirmation of long-term efficiency, but the introduction of GK surgery into epilepsy surgery can reduce dramatically its invasiveness and morbidity. Key Words: Mesial temporal lobe epilepsy-Gamma knife- Radiosurgery-Amygdalohippocampectomy-Radionecrosis.

documented radiosurgical treatments of epilepsy cases in the literature were symptomatic cases with space- occupying lesions such as tumors or AVMs (5-9). How- ever, the results of these studies gave us the idea of testing radiosurgery as a new way of treating epilepsy without space-occupying lesions.

Although the risk of epilepsy surgery is low, surgical complications may occur, such as postoperative neuro- logic deficit due to unintended vascular compromise or other accidental damage (10). To avoid this risk, less- invasive techniques, such as radiosurgery, may be a promising alternative choice that is still to be investi- gated.

Mesial temporal lobe epilepsy (MTLE) is a recognized condition associated with hippocampal sclerosis and has a good outcome after limited TL resection (1 1). There is agreement that MTLE can be diagnosed in most patients without invasive tests (12-14). In March 1993, the first patient with medically refractory MTLE was treated with GK surgery in Marseille. In lieu of a microsurgical pro-

1551

1552 J. REGIS ET AL.

cedure, an entorhinoamygdalohippocampectomy was performed with GK and low marginal dose (25 Gy). The patient became seizure free from the treatment day with- out any clinical complication (15). This case suggests that GK surgery may be a promising tool for minimal invasive epilepsy surgery.

To determine the best technical parameters of treat- ment, we reproduced in three other patients with MTLE, the same procedure with slight changes: progressive de- crease of target volume and dose. The quality of the result decreased in parallel with dose and volume. The second case (20 Gy; 4.8 ml) only became seizure free several months after the procedure, the third case (20 Gy; 3.9 ml) improved gradually but still have seizures, and the fourth case (20 Gy; 3.5 ml), after a period of im- provement, went on having relatively frequent seizures and is considered a failure, However, after the follow-up of these four patients for 4 years, the safety and effi- ciency of this method above a doselvolume threshold encouraged us to organize a more extensive study ac- cording to the initial technical procedure.

The preliminary results of seven patients (including the first patient treated) with drug-resistant MTLE treated by this standard radiosurgical protocol are re- ported. The three patients treated initially with a different protocol (lower doses and volumes) are not included.

METHODS

Selection of patients Patients (four men, three women) were recruited from

two institutions (the Neurophysiology Department, Ti- mone Hospital and Centre Saint-Paul, Marseille, France). All had chronic drug-resistant partial epilepsy. Patients' data are summarized in Table 1. Mean duration of the disease was 27 years (range, 16-35 years), and mean age at GK surgery was 33 years (range, 2 5 4 0 years; Table I). All patients were given neurologic and medical ex- aminations and awake EEG. They all experienced sei- zures during surface video-EEG monitoring with scalp and bilateral foramen ovale electrodes (FO) (16). In all selected patients, ictal semiology was highly suggestive of MTLE (1 2,14) (Table 1 ). FO recordings demonstrated a unilateral mesiotemporal seizure onset in all the cases with delayed propagation to the lateral part of the tem- poral lobe. Patients were given neuropsychological evaluations and Wada tests, showing no mnesic disorder after internal carotid artery injection of amytal, ipsilateral to the epileptogenic zone. Magnetic resonance imaging (MRI) and [ '8F]-deoxyglucose PET (FDG-PET) were performed before the procedure. For all patients, MRI presurgical scans with coronal slices showed moderate to marked focal atrophy of the ipsilateral hippocampus (Fig. I). FDG-PET scans were performed with a high- resolution head-dedicated tomograph (ECAT SIEMENS

953-31B, 5-mm axial resolution, 31 slices at 3.37-mm intervals). In all the patients, FDG-PET showed an in- terictal hypometabolism that involved the hippocampal region, the temporal pole, and the anterior part of the lateral temporal cortex of the epileptogenic temporal lobe. This hypometabolism pattern is usually seen in MTLE, as previously reported (17). These neuroimaging findings were consistent with the electroclinical data and led to the diagnosis of MTLE associated with presum- able mesial temporal sclerosis. All patient gave their in- formed consent before receiving GK treatment.

Radiosurgical treatment With ethics committee agreement (ComitC Consultatif

de Protection des Personnes dans la Recherche BiomCdi- cale, Marseille I, 1994 and 1996), the radiosurgical treat- ment was carried out with the 201-source cobalt 60 Gamma Unit B model in Marseille (18). We used MR and computed tomography (CT) for target localization and the Kula System (Elekta Instrument AB, Stockholm, Sweden) for dose planning. Target choice was based on the Zurich experience of microsurgical amygdalohippo- campectomy, the demonstration by Wieser of the pre- dominant role of the parahippocampal cortex, and the requirement, for safety reasons of keeping the ciosimetry far enough from the brainstem and optic tract (15,19). The target included the head of the hippocampus and the anterior part of the hippocampal body, the amygdalofu- gal part of the amygdala, and the entorhinal area. For optimal interpretation of the imagery, the frame was placed in such a way that the base plan was parallel to that of temporal horn. The mean calculated volume of the target was 6,500 mm3 (range, 6,250-6,900 mm3). Ac- cording to the volume of the target, we used a marginal dose of 25 Gy at the 50% isodose line.

After GK surgery, patients were given medical exami- nations and awake EEGs every month. The clinical fol- low-up was performed simultaneously by the neurosur- geon practicing the radiosurgery and independently by a group of epileptologists to obtain a more objective evalu- ation. The patient kept a report of seizure frequency by using a diary. The follow-up time is the period after the treatment. Outcome of seizure was classified according to Engel's classification (20). MRIs were performed regularly after the GK surgery (3, 6, 9, 12, and 24 months).

RESULTS

Gamma-knife surgery efficiency

Control of accuracy in targeting The chronology of MRI changes was very reproduc-

ible from one patient to another (Fig. 3). The MRI ex- aminations showed only a slightly increased T, signal inside the target volume at 6 months. Morphologic

E p i / e p w , Vol. 40, N o . 11, 1999

GAMMA KNIFE RADIOSURGERY IN MTLE 1553

TABLE 1. Main patient data

FO FDG-PET recordings:

Age at Age temporo- initial ictal epilepsy at GK Previous Early clinical MRI mesial discharge Side of

onset surgery Previous AD ictal hippocampal hypometa- (n = number the KG Patient (yr) (yr) Sex history treatment symptoms atrophy bolism of seizure) surgery

1 3 25 M FC (age I y, CBZ followed by PHT left transient PB hemiparesis) VGB

2 15 31 M FC (age 9 and 11 CBZ mo) VGB

CZP

3 5 35 F FC (age 2 yr). PHT Six GTCS in VGB 1991

Staring, chewing, epigastric sensation

R R mesial temporal R (n = 8)

Behavioral arrest, chewing, simple hand automatisms

Epigastric pain (ascend to the chest), pleasure feeling, chewing

Epigastric sensation, chewing, right-hand automatisms, left-hand dystonia

Epigastric and thoracic sensation, look on the right, chewing, unresponsiveness

Epigastric pain with fearful sensation, verbal automatisms, agitation

sensation, fear, chewing, speech arrest

Epigastric

L L mesial temporal L (n = 3 )

+ (L)

L L mesial temporal L (n = 10)

4 7 34 M - VPA CBZ PB

R R mesial temporal R (n =4)

5 7 34 F FC with left hemiparesis (age 3 yr). One GTCS at age 7 yr

-

PB PHT CBZ

R R mesial temporal R (n =5)

6 5 40 F R R mesial temporal R (n =4)

VPA PB VGB CBZ LTG

+ (slight, R)

L L mesial temporal L (n = 10)

7 6 36 F Head trauma (8 mo)

PB CBZ VGB PGB VPA

GK, gamma knife; M, male; F, female; R, right; L, left; FC, febrile convulsion; GTCS, generalized tonic-chic seizure; CBZ, carbamazepine; PHT, phenytoin; PB, phenobarbital; VPA, valproate; VGB, bigabatrin; LTC, lamotrigine; PGB, progabide; CZP, clonazepan.

(MRI) signs of injury of the target were observed 9 months after GK surgery (Fig. 4). Inside the target vol- ume, there was a focally swollen hippocampus with heterogeneous T, signal hyperintensity. A contrast-en- hancement ring corresponding exactly to the periphery of the target region (50% isodose line) appeared (Fig. 2). Outside the target volume, there was a larger area of homogeneously increased T,W signal, which predomi- nantly affected the white-matter tracts of the temporal lobe. A corticosteroid treatment was systematically ini- tiated orally at this time (methylprednisolone, 0.5 mg/ kglday) and progressively stopped depending on MRI evolution. At the twelfth month, the MR scans showed changes similar to the previous scan. At month 24, MRI examination (Fig. 3 ) no longer showed contrast enhance- ment or high T, signal (for more detail, see 21).

0 2 4 6 8 10 1 2 14 16 FIG. 1. Kaplan-Meier plot displaying the progressive onset of seizure cessation depending on time. Patient 4 has been seizure free 13 months after gamma-knife surgery. Seven months after surgery, she had a new short seizure, and 8 month later, another one. All the other patients are seizure free (IA).

Epilepiu, Vol. 40, N o . 11, 1999

1.5.54 J. REGIS ET AL.

Seizure outcome The current duration of follow-up is 19-61 months.

The first patient has been seizure free since the treatment day. All the other patients had a gradual decrease of seizure frequency and seizure intensity and duration. A dramatic effect on seizure frequency took place around postoperative month 10 (range, 8-15; mean, 10.16). In- deed, from this date, most of the patients became seizure free (class IA of Engel’s classification). This date corre- sponds approximately to the appearance of morphologic (MRI) signs of destruction of the amygdaloentorhinohip- pocampal target that was selectively injured after GK surgery. To date, antiepileptic drugs (AEDs) have been reduced to a monotherapy for patients 1 and 6.Treatment was stopped for patient 4. Table 2 shows the current

FIG. 2. At 9 months (A), inside the target volume is a heterogeneous T, signal hypointensity and a con- trast enhancement ring correspond- ing exactly to the periphery of the target region. Outside the target vol- ume, a larger area of homoge- neously decreased T, signal is ob- served. At month 24 (B), the MRI examination demonstrated the dis- appearance of the radiologic signs.

follow-up and Engel’s classification for the seven pa- tients. All the patients in this series are now seizure free (IA of the Engel classification) except one case (patient 4) with rare persisting seizures (IIA).

Gamma-knife surgery safety No major side effect, morbidity, or mortality was ob-

served. Three patients experienced slight transient head- ache around the tenth month. However it disappeared immediately after introduction of the corticosteroid treat- ment. The neuropsychological outcome is currently un- der investigation, but no patient complained of any memory or cognitive worsening during follow-up. Neu- rologic examination, including visual field, remained normal except in patient 4, who had a nondisabling hom-

FIG. 3. Coronal T2 MR images of the mesial structures before (A) and 3 years after (6) the gamma-knife surgery. A: A very clear mesial tem- poral lobe atrophy on the side treated and the long-term control where the structures appear as in the preoperative view (no more con- trast enhancement, no more edema, no macroscopic signs of the destruction of the mesial temporal lobe structures).

E p i l e p w Vol 40, N o . 11, 1999

GAMMA KNIFE RADIOSURGERY IN MTLE 1555

m - 8

FIG. 4. Delay of onset of the MRI LI I

changes. For the majority of the pa- tients, the MRI changes appear -9-1 0 months; for some patients, they occur

f, g k

.- 5 s g5 *

only -16-18 months. 2 .C

Time since GK (Months)

onymous superior quadrantanopia defect. No toxicity due to corticosteroid treatment was observed.

DISCUSSION

In this preliminary study, the seven patients repre- sented a homogeneous group of patients with clinical, neuroradiologic (MRI, PET), and electrophysiologic evi- dence of MTLE, in the absence of any space-occupying lesion (1 2,14). MTLE is the most frequent type of drug- resistant TLE and is well known as being an epileptic condition highly remediable by microsurgical temporal cortectomies.

In patients with epilepsy given temporal lobe micro- surgery, the retrospective comparative studies of Guldov et al. (22,23) and Spencer (10) emphasized the risk of some neurologic deficit such as transient hemiparesis, transient aphasia, homonyinous campimetric field defi- cits, and memory impairment.

Radiosurgery allows operation on the patient’s brain without opening the skull and can thus dramatically re- duce the invasiveness of the neurosurgical procedure. The minimal invasiveness, safety, and efficiency of GK surgery have been well demonstrated for treating AVMs ( 2 ) , acoustic neurinomas (24,25), meningiomas (26), and metastasis (27). The benefits for the patient undergoing this extensive experience are numerous: absence of mor- tality, important decrease of morbidity, improved com- fort (the procedure is performed under local anesthesia), reduction of the hospital stay (48-72 h and ambulatory

TABLE 2. Seizure outcome and follow-up

Frequency of seizure Date of Delay in before GK the GK Follow-up seizure Engel’s

Patients Mean (/mo) surgery (mo) cessation class

I 4 March93 61 I IA 2 10 May95 42 10 IA 3 I Dec 95 22 8 IA 4 14 Apr96 34 15 IIB 5 9 July 96 29 16 IA 6 18 July 96 21 12 IA I 19 Oct 96 24 10 IA

treatment is possible), improvement of the rate and delay of return to work (18,27). In addition, several recent series have enhanced the dramatic improvement of the cost-effectiveness rate (27).

These facts suggest that application of this method to epilepsy surgery can greatly benefit the patient (decreas- ing invasiveness of surgical treatment) and society (de- creasing the cost of epilepsy surgery).

This report is the first to investigate the efficiency and safety of this method for the surgical treatment of severe resistant epilepsy without space-occupying lesions.

These preliminary results confirm the very good tol- erance of the method. Because of the short duration of the follow-up in this series, further long-term evaluation is needed.

However, > 100,000 patients have been treated throughout the world by using GK surgery for other pa- thologies during several decades of follow-up. In par- ticular, no long-term neoplasia genesis has been reported after GK surgery. What makes radiosurgery completely different from radiotherapy is the importance of the vol- ume receiving radiation and the fractionation of the dose. The occurrence of neoplasm after conventional radio- therapy is strongly suspected to be related to fractionna- tioii (28).

The two microsurgical approaches proposed in MTLE are either a selective amygdalohippocampectomy (1 1) or a “standard” temporal cortectomy enlarged to the polar region and to a lesser degree to the lateral cortex (29). Some authors claimed that the more selective approach, sparing more nonepileptogenic tissue, led to a better neu- ropsychologic outcome (12,30).

Results of classic epilepsy surgery show that nearly 70% of patients with TLE become seizure free after sur- gery (10,20,3 1,32).

Definite conclusions cannot be drawn concerning GK efficiency because of the few patients treated and the short-term follow-up. However, our preliminary results seem to show that GK “arnygdalohippocampectomy” is as effective as the microsurgical approach.

Indeed all the patients had experienced cessation of

Epdepszu, Vol. 40, NO. 11, 199Y

1556 J. REGIS ET AL.

seizure 10 months (range, 8-15 months) after radiosur- gery. This period corresponds to the delay of onset of the MRI signs. All patients are currently seizure free (class IA, patients 1-3 and 5-7) except patient 4, who had some infrequent seizures (three seizures during the past year) after a period of complete remission.

This initial experience clearly proves the short- to middle-term efficiency and safety of GK for MTLE sur- gery. The delay in efficacy and the onset of MR changes leading to prolonged steroid use are, in our opinion, the only two disadvantages of this approach. To improve the latter, the “marginal dose” or the volume should be re- duced or the delineation of the supposed anatomic target should be more precisely limited to the theoretical ana- tomic target. Unfortunately, until now, all those attempts have systematically ended in failure. Dose and volume are crucial parameters for radiosurgery safety. For this reason, we do not believe that radiosurgery should be used for epilepsy of another location (without a space- occupying lesion) because of the large amount of cortex that usually constitutes the epileptogenic zone.

These results require further confirmation as regards long-term efficiency; however, introduction of the GK method to epilepsy surgery can, in the future, dramati- cally reduce invasiveness and morbidity.

Acknowledgment: We thank the “Assistance Publique des HBpitaux de Marseille” for supporting this work.

REFERENCES I .

2.

3.

4.

5.

6

7

8

9

10

Talairach J, Bancaud J, Szikla G, Bonk A, Geler S, Vedrenne C. Approche nouvelle de la neurochirurgie de 1 ’epilepsie. Mithod- ologie ste‘riotaxique et re‘sultats thtrapeutigues, Congris Annuel de la Soci&te de Lungue Frunyaise, Marseille 25-28 Juin 1974, 1974. Vol. 20. Paris: Masson. Steiner L, Lindquist C, Adler J, Tomer J , Alves W, Steiner M. Clinical outcome of radiosurgery for cerebral arteriovenous mal- formations. J Neurosurg 1992;77: 1-8. Heikkinen ER, Konnov B, Melnikow L. Relief of epilepsy by radiosurgery of cerebral arteriovenous malformations. Stereotact Funct Neurosurg 1989;53: 157-66. Lindquist C, Kilstrom L, Hellstrand E. Functional neurosurgery: a future for the gamma knife? Stereotacr Funct Neurosurg 1991;57: 72-8 I . Alexander E 111, Lindquist C. Special indications: radiosurgery for functional neurosurgery and epilepsy. In: Alexander E 111, Loeffler J S , Lunsford LD, eds. Stereotactic radiosurgev. New York: Mc- Graw-Hill, 1993:221-5. Whang C, Kim C. Short-term follow-up of stereotactic gamma- knife radiosurgery in epilepsy. In: Ganz JC, Gildenberg PhL, Franklin PO, eds. Stereotactic and functional neurosurgery, Lek- sell Gamma Knife Society, Kyoto, Japan, 1994. Vol. 64 (suppl 1). New York: Karger. Alexander E, Loeffler JS. Radiosurgery using a modified linear accelerator. Neurosurg Clin North A m 1992;3: 167-90. Rossi GF, Scerrati M, Roselli R. Epileptogenic cerebral low grade tumors: effect of interstitial stereotactic irradiation on seizures. Appl Neurophysiol 1985;48: 127-32. Loeffler J S , Alexander E. The role of stereotactic radiosurgery in the management of intracranial tumors. Oncology 1990;4:2 1-3 1. Spencer S. Long term outcome after epilepsy surgery. Epilepsia 1996;37: 807- 1 3.

I I .

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29

30

31

Wieser HG. Selective amygdalohippocampectomy as a surgical treatment of mediobasal limbic epilepsy. Surg Neurol 1984;17: 145-57. Wieser H, Engel JJ, Williamson P, Babb T, Gloor P. Surgically remediable temporal lobes syndromes. In: Engel JE Jr, ed. Surgical treatment of the epilepsies. 2nd ed. New York: Raven Press, 1993: 49-63, Williamson P, French J , Thadani V, et al. Characteristics of medial temporal lobe epilepsy: 11. Interictal and ictal scalp electroencepha- lography, neuropsychological testing, neuroimaging, surgical re- sults and pathology. Ann Neurol 1993;34:781-7. French J, Williamson P, Thadani V, et al. Characteristics of medial temporal lobe epilepsy: I. Results of history and physical exami- nation. Ann Neurol 1993;34:774-80. Regis J, Peragut JC, Rey M, et al. First selective amygdalohippo- campic radiosurgery for mesial temporal lobe epilepsy. Stereotact Funct Neurosurg 1994;64: 191-201. Wieser HG, Elger CE, Stodieck SRG. The foramen ovale elec- trodes: a new recording method for the preoperative evaluation of patients suffering from mesiobasal temporal lobe epilepsy. Elec- troencephalogr Clin Neurophysiol 1985;61:314-22. Semah F, Baulac M, Hasboun D, et al. Is interictal temporal hy- pometabolism related to mesiotemporal sclerosis? A positron emission tomographyhagnetic resonance imaging confrontation. Epilepsia 1995 ;36:447-56. Steiner L, Lindquist C, Steiner M. Radiosurgery: advances and technical standards in neurosurgery. Vol 19. New York: Springer- Verlag, 1992:2O-l02. Wieser HG, Siege1 AM, Yasargil GM. The Zurich amygdalohip- pocampectomy series: a short up-date. Acta Neurochir Suppl ( Wien) 1990;50: 122-7. Engel J, VanNess P, Rasmussen T, Ojemann L. Outcome with respect to epileptic seizures. In: Engel J, ed. Surgical treatment of the epilepsies. New York: Raven Press, 1993:609-22. RCgis J, Semah F, Bryan RN, et al. Early and delayed MR and PET changes after selective temporomesial radiosurgery in mesial tem- poral lobe epilepsy. AJNR A m J Neurorudiol 1999;20213-6. Guldvog B, Loyning Y, Hauglie-Hanssen E. Surgical versus medi- cal treatment of epilepsy. I. Outcome related to survival, seizures and neurological deficit. Epilepsia 1991 ;32:375-88. Guldvog B, Loyning Y, Hauglie-Hanssen E. Surgical versus medi- cal treatment of epilepsy. 11. Outcome related to social areas. Epi- lepsia 1991 ;32:477-86. Noren G, Arndt J , Hindmarsh T. Stereotactic radiosurgery in cases of acoustic neurinoma: further experience. Neurosurgery 198>;13: 12-22. Norin G. Gamma knife radiosurgery for acoustic neurinomas. In: Gildenberg PL, Tasker RR, eds. Textbook of stereotactic and,func- tional neurosurgery. Vol 1. New York: McGraw-Hill, 1996335- 44. Kondziolka D, Lunsford L, Coffey R, Flickinger J. Stereotactic radiosurgery of meningiomas. J Neurosurg 1991 ;74:552-9. Rutigliano MJ, Lunsford LD, Kondziolka D, Strauss MJ, Khanna V, Green M. The cost effectiveness of stereotactic radiosurgery versus surgical resection in the treatment of solitary metastatic brain tumors. Neurosurgety 1995;37:445-55. Tubiana M, Dutreix J, Wambersie A. Time and fractionation in radiotherapy. In: Francis T, ed. Introduction to radiobiolugy. Vol 1. London: Wiley, 1990:225-51. Spencer D, Insemi J . Temporal lobectomy. In: Luders HO, ed. Epilepsy surgery. Vol 1. New York Raven Press, 1991:53345. Arruda F, Cendes F, Andermann F, et al. Mesial atrophy and outcome after amygdalohippocampectomy or temporal lobe re- moval. Ann Neurol 1996;40:446-50. Rougier A, Dartigues J , Commenges D, Claverie B, Loiseau P, Cohadan F. A longitudinal assessment of seizure outcome and overall benefit from 100 corticectomies for epilepsy. J Neurol Neurosurg Psychiatty 1992;55:762-7.

32. Sperling M, O’Connor M, Saykin A. Five year outcome of tem- poral lobectomy for refractory epilepsy. Neurology 1995;45:395S.

Epilepsio, Vol. 40, No. 11, 1999