The Locahng Value of Depth Electroencephalography in 32 Patients

with Refractory Epilepsy Susan S. Spencer, MD,' Dennis D. Spencer, MD,? Peter D. Williamson, MD,'

and Richard H. Mattson, MD"

~~ ~~

Clinical seizure manifestations, physical examination, radiological studies, neuropsychological tests, and scalp and depth electroencephalographic (EEG) studies were done to localize seizure foci in 32 patients, 23 of whom have undergone surgery with more than one year of follow-up. Of 16 patients with unlocalized scalp EEGs, depth EEG detected a consistent focal seizure onset in 3. Of 15 patients with localized scalp EEGs, depth EEG revealed multiple foci in 3 and inaccurate localization of the focus by scalp EEG in 4. Therefore, localization by scalp EEG was inaccurate in 10 of 31 patients. When depth EEG revealed a consistent focal seizure onset and this localization was used to determine the surgical resection site, good to excellent results were obtained in 12 of 13 patients. When depth EEG revealed additional foci of seizure origin, surgical results were fair or poor in all patients. No other localizing criteria predicted depth EEG results or surgical success with such accuracy. Furthermore, some criteria, including neuropsychological tests and radiological abnormalities, were falsely localizing at times. Therefore, of the presently available localizing criteria, depth EEG appears to be the most accurate.

Spencer SS, Spencer DD, Williamson PD, Mattson RH: The localizing value of depth electro- encephalography in 32 patients with refractory epilepsy. Ann Neurol 12:248-253, 1982

Surgical removal of epileptogenic cerebral tissue has been shown to benefit 60 to 80% of selected patients with medically refractory focal epilepsy [2, 4, 9-12, 20, 21, 29, 321. The 20 to 40% failure rate is partly attributed to lack of precise localization of the epileptogenic cerebral area [22, 241.

Studies of surgical prognostic factors in refractory focal epilepsy have not precisely defined a set of cri- teria which would best localize epileptogenic cerebral areas [l, 111. Commonly used localization criteria in- clude physical findings, clinical seizure history, radiological studies, neuropsychological testing, scalp electroencephalography (EEG), nasopharyngeal and sphenoidal recordings, and depth EEG. Depth EEG has been shown to provide increased information, but opinions differ as to whether it promotes better localization and increased surgical success ([7,22,26, 28, 301 and Spencer DD, Spencer SS, Williamson PD, et al: Surgery for intractable epilepsy: determi- nants for success in seizure control. Submitted for publication).

Since 1972, 32 patients have had extensive depth and scalp electrographic study as well as these other diagnostic localizing tests. Twenty-five have under-

gone surgical procedures, 23 with more than one year of follow-up. The value of depth EEG is exam- ined and related to other localizing diagnostic proce- dures.

Methods Patient Population Thirty-two patients evaluated since 1972, most since 1975, are included. They were initially selected by the following criteria [S, 16, 221:

1.

2. 3. 4.

5 . 6.

Seizures refractory to appropriate anticonvulsant medi- cation as judged by serum anticonvulsant levels Socially incapacitating attacks Possible surgical accessibility of the seizure focus Patient interest in and compliance with extensive evalu- ation Age between 10 and 50 years No major complicating medical or psychiatric illness

Patient Evaluation Further evaluation has been described in a previous publi- cation [2 51. Clinical assessment included careful investiga- tion for localizing information contained in the patient's

From the Departments of "Neurology and +Surgery (Section of Neurosurgery), Yale University School of Medicine and West Haven VA Medical Center, New Haven, CT.

Received Mar 24, 1981, and in revised form Dec 3, 1981, and Jan 11, 1982. Accepted for publication Jan 17, 1982.

Address reprint requests to Dr Susan S. Spencer, Epilepsy Center, VA Medical Center, West Haven, CT 065 16.

248 0364-5 13418210~0248-06$01.25 0 1982 by the American Neurological Association

description of seizures and in the physical examination. Formal mapping of visual fields was performed on all pa- tients. Detailed neuropsychological testing, including mea- sures of verbal and performance IQ, verbal and visual/ spatial memory, language function, motor integrative func- tion, and sensory perceptual function, was performed at least once during the preoperative evaluation.

Extensive electrographic analysis for localization of epileptogenic foci included routine scalp EEG, naso- pharyngeal and sphenoidal leads, and sleep and sleep- deprivation recordings. An attempt was made to record at least three spontaneous seizures from each patient. Con- trolled withdrawal of medication was sometimes necessary to precipitate ictal events for analysis. Medication with- drawal in chronic refractory epilepsy does not provide misleading information during determination of the number of sites of seizure onset [27]. All recording was done with continuous audiovisual and EEG monitoring.

Radiological investigation was performed to detect mass lesions that could have required surgery without further investigation, to document ventricular asymmetries and abnormal vasculature, and to establish anatomical land- marks for insertion of depth probes. Skull series, angio- grams, and pneumoencephalograms were performed on all patients. CT scans have been done on all patients since 1975.

Three to six depth probes were inserted. The probes were of the multicontact, 22-gauge variety designed by Charles Ray [23]. They were implanted in frontal and tem- poral locations as dictated by clinical and electrical infor- mation [25]. The frontal electrode was entirely within the frontal lobe with a depth target of the medial orbital frontal lobe. The midtemporal electrode traversed the frontal lobe en route to its anterior temporal depth target. The poste- rior electrode traversed the occipital lobe en route to a hippocampal target. Electrodes were inserted through stainless-steel guidepins that were stereotactically im- planted under general anesthesia.

Continuous audiovisual and EEG monitoring was per- formed until at least three and often more seizures were recorded. Ictal onset by EEG findings and clinical seizure manifestations were reviewed by three experienced elec- troencephalographers.

Surgery involved resection of the focal cerebral area identified by this evaluation as being responsible for the initiation of habitual clinical seizures. Most often this en- tailed en bloc frontal or temporal lobectomy, with cortical mapping under local anesthesia when necessary.

Follow-up Patients were followed at three-month intervals initially, and later at six-month intervals. All anticonvulsant medica- tions were continued for at least three years postopera- tively. Results were coded at longest follow-up using gen- erally accepted criteria [4, 13, 191 (Table I), with Class 1 representing excellent results and Classes 4 and 5, poor re- sults. Neuropsychological testing, physical examination, and scalp EEG recordings including overnight monitoring were repeated at six-month intervals for the first two years, and thereafter were performed yearly.

Table 1. Surgical Outcome Classes

Result Class Description Classification

1 No seizures, rare aura, or greater Excellent than 95% reduction in seizure frequency

2 Greater than 75% reduction in Good seizure frequency

3 50 to 75% reduction in seizure Fair frequency

4 Less than 50% reduction in sei- Poor zure frequency or no change

5 Increase in seizure frequency Poor

Analysis Clinical seizure manifestations, physical examination, radiological studies, neuropsychological evaluations, and scalp EEG were analyzed with regard to localization of ab- normal results. Concerning the clinical seizure, initial tonic or clonic motor activity of an arm, a leg, or both was at- tributed to contralateral frontal lobe seizure activity. Ictal or postictal aphasias were localized to dominant frontal or temporal lobe. Formed visual hallucinations in one visual field were considered to be of contralateral temporal lobe origin and unformed visual hallucinations of occipital lobe origin. Abnormalities on physical examination were as- sessed in terms of predicted anatomical location of disease. Motor system abnormalities were considered to have frontal lobe localization, aphasias were classified as with- in the dominant frontal o r temporal lobe, and cognitive deficits were regarded as diffuse and bilateral. Extensive neuropsychological testing was performed preoperatively on 20 of the 23 patients with one year of surgical follow-up. Degree of impairment on standardized tests was deter- mined, and abnormalities were localized to cerebral re- gions. Localization on the basis of the scalp EEG was de- termined for 22 of the 23 patients with more than one year of surgical follow-up. For the remaining patient, in- sufficient EEG records had been preserved to allow ade- quate assessments comparable to those of the other pa- tients. The scalp EEG was considered “localized” if at least 80% of all ictal and interictal abnormalities arose in one area. Scalp EEG localization was most often done from in- terictal abnormalities. At least three ictal events were studied but rarely provided good localization or even lateralization. Interictal records included numerous standard EEG recordings with varied montages as well as overnight recordings through all stages of sleep. The mon- tage for the overnight recording was selected to maximize interictal abnormalities. Sleep deprivation and medication withdrawal were employed as necessary. The precise depth EEG localization of seizure onset and interictal abnor- malities were analyzed for each patient. Multiple seizures were recorded from each patient, and most attention was given to ictal onset by depth EEG for determination of final localization. Localizing abnormalities from the different testing procedures were tabulated with respect to the re- sults of surgery and depth EEG localization.

Spencer et al: Localizing Value of Depth EEG 249

Results Depth E E G wersus Scalp E E G Analysis of these patients allows the group to be di- vided into 15 patients having localizing scalp EEGs and 16 patients with nonlocalizing scalp EEGs, all 3 1 of whom were further studied by implantation of depth electrodes. Of the 15 patients with well- localized scalp EEG abnormalities, 3 had poor locali- zation by depth EEG study, i.e., fewer than 80% of seizures were found to arise in a single well-defined area. Two of these 3 patients have had resection of the area producing the greatest number of seizures with at least one year of follow-up, and both have had poor surgical results. Thus, in a group of patients with localized epileptogenic foci by scalp EEG, study by depth EEG identified those with inconsistent lo- cation of seizure onset. These findings predicted a poor surgical result. The remaining 12 patients with well-localized scalp EEGs had well-localized foci on depth EEG. Eleven have had surgery with more than one year of follow-up, and results are good to excel- lent in 10.

Of these 15 patients whose scalp EEGs showed a localized abnormality, 14 had well-defined anterior temporal lobe foci. Seven (50%) had confirmation of the anterior temporal lobe seizure focus by depth EEG. In the remaining 7 patients, depth EEG showed the focus to be elsewhere in 1 patient (frontal), be- yond the usual limits of temporal lobe resection in 3 patients (posterior temporal), or unlocalized or mul- tifocal in 3 patients. When a depth EEG diagnosis of multifocal seizure onset was ignored in favor of a plan to proceed with surgical resection of the site respon- sible for the greatest number of seizures, the results were poor (2 patients). When the depth EEG evalu- ation indicated a consistent onset of seizures from a more posterior temporal location, results were good but never excellent ( 3 patients). When depth EEG confirmed the scalp EEG localization with consistent anterior temporal seizure onset, results were excel- lent (7 patients). Similarly, when the depth EEG de- termined a focus elsewhere, i.e., the frontal lobe, and that focus was resected, the results were good (1 patient).

The 16 patients with unlocalized scalp EEGs fell into two groups when studied with depth EEG. Thirteen had unlocalized depth EEGs (fewer than 80% of seizures arising in one location). The 7 pa- tients in this category who underwent resection of the anatomical site responsible for the greatest number of seizures as determined by depth EEG had only fair or poor surgical results. Three patients in the group with unlocalized scalp EEGs had 100% of seizures arising from one well-defined location on depth EEG, and the 2 who have undergone surgery with more than one year of follow-up have had good

and excellent results, respectively. The percentage yield of depth EEG in this regard is low in our pa- tients (3 of 16), in contrast to the yield reported elsewhere [ 181.

Specialized Electrode Recordings Sphenoidal and nasopharyngeal recordings have been suggested as alternatives to depth EEG. In 2 patients with anterior temporal scalp foci, the nasopharyngeal recording revealed abnormalities on the appropriate side. One of these patients had a medial orbital fron- tal focus and the other a posterior temporal focus by depth EEG. Both had good results following resec- tion of the seizure focus determined by the depth re- cording. Nasopharyngeal recording would not have predicted a seizure focus outside of the anterior tem- poral region in either case. In 1 patient, a well- localized focus was found on both nasopharyngeal and sphenoidal recordings despite bilateral abnor- malities on scalp EEG, but depth EEG showed all typical seizures to arise consistently from the con- tralateral temporal lobe. Another patient, who had bilateral abnormalities on scalp, nasopharyngeal, and sphenoidal recordings, was shown by depth EEG to have a consistent anterior temporal focus on one side initiating all clinical seizures. This patient underwent surgery with excellent results. These other tech- niques therefore were not able to provide the localizing information available from the depth EEG.

DEPTH EEG VERSUS OTHER LOCALIZING CRITERIA. Further analysis of the 14 patients with anterior temporal foci on scalp EEG was done to determine whether any other criteria might supply the same localizing information as the depth EEG. In Table 2, the patients in this group are tabulated according to localization by clinical seizure pattern, physical ex- amination, radiological investigation, neuropsycho- logical testing, scalp and depth EEG, and overall localization without depth EEG. None of the local- izing items was consistently predictive of the depth EEG localization. Furthermore, misleading informa- tion was obtained, especially from radiological pro- cedures. Three patients in whom surgical resec- tion was performed on the basis of depth EEG with excellent or good results had radiological tests localizing abnormalities to the contralateral hemi- sphere. Analysis of the 3 patients with unlocalized scalp EEGs and well-localized depth EEGs also showed that no other studies could predict the depth EEG localization (Table 3) .

Discussion Improved ability to localize epileptogenic foci pre- cisely is likely to be the best way to improve surgical results in patients with refractory epilepsy [22, 24,

250 Annals of Neurology Vol 12 No 3 September 1982

Table 2. Results Localizing Tests in Patients with Anterior Temporal Foci on Scalp E E G

Overall Clinical Physical Neuro- Localization

Patient Seizure Exami- X-ray psychological Scalp Depth Resection without No. Description nation Studies Testing EEG EEG Site Outcome" Depth EEG

1 Nonlocalizing 2 L frontal 3 R frontal 4 R frontal 5 Multifocal 6 L frontal 7 Muldfocal 8 Nonlocalizing 9 L temp.

10 L frontal 11 Muldfocal 12 L temp. 13 Nonlocalizing

Normal L hem. Normal Normal R eye Normal Normal Normal L temp. Normal Normal Normal Normal

L hem. L hem. Normal Normal R occip. Normal R temp. Normal L temp. L temp. Normal L temp. R temp.

R frontotemp. L frontotemp. Bilateral Bilateral R post. temp. Bilateral L frontotemp. L frontotemp. L frontotemp. Bilateral L temp. Bilateral Bilateral

R ant. temp. L ant. temp. R ant. temp. R ant. temp. R ant. temp. R ant. temp. L ant. temp. L ant. temp. L ant. temp. L ant. temp. R ant. temp. L ant. temp. L ant. temp.

R frontal Bilateral Bilateral Bilateral R post. temp. R post. temp. L post. temp. L ant. temp. L ant. temp. L ant. temp. R ant. temp. L ant. temp. L ant. temp.

R frontal L temp. R frontal R temp. R post. temp. R temp. L temp. L temp. L temp. L temp. R temp. L temp. L temp.

2 4 4 Too soon Too soon 2 2 1 1 1 1 1 1

Bilateral L hem. Bilateral Bilateral R hem. Bilateral Bilateral L frontotemp. L fronrotemp. Bilateral Bilateral Bilateral Bilateral

14 Nonlocalizing Normal R temp. R hem. R ant. temp. R ant. temp. R temp. 1 R hem.

"See Table 1 for classification of surgical results.

Table 3. Results of Localizing Tests in Patients with Nonlocalizing Scalp E E G s and Consistent Depth E E G Localization

Overall Clinical Physical Neuro- Localization

Patient Seizure Exami- X-ray psychological Scalp Depth Resection without No. DescriDtion nation Studies Testing EEG EEG Site Outcomea Depth EEG

15 Nonlocalizing L temp. L temp. L frontotemp. Nonlocalizing L ant. temp. None . . . L frontotemp. 16 L frontal R frontal Normal L frontal Nonlocalizing R ant. temp. R temp. 1 Bilateral 17 R occip. temp. Normal R front. L temp. Nonlocalizing R post. temp. R temp. 2 Bilateral

"See Table 1 for classification of surgical results.

3 11. Electrical localization is generally regarded as most valid [l, 141. The literature, however, is con- fusing. Localization of ictal onset by scalp EEG has been said to reflect the results of depth EEG reliably [15]. Often such localization is not possible [26]. Furthermore, the reliability of the scalp EEG, and even sphenoidal EEG, in identifying ictal onset has been questioned {5]. Electrical localization of the sei- zure focus was studied in detail in our patients. Fif- teen of them had well-localized scalp EEGs, usually by interictal records, with more than 80% of abnor- malities arising in a single well-defined location. Six- teen patients had unlocalized scalp EEGs. Compari- son with depth EEG findings, area of resection, and surgical results indicates that foci in one-half of our patients were inappropriately localized by scalp EEG. These misleading scalp EEGs most often localized the focus to the anterior temporal lobe, whereas depth EEGs showed a multifocal seizure onset in 3 patients, a frontal lobe seizure onset in 1 patient, o r a more posterior temporal onset of seizures in 3 pa- tients. In these patients with misleading localization in the anterior temporal lobe by scalp EEG, good re- sults were sometimes obtained by standard temporal lobectomy when seizures consistently arose more

posteriorly in the temporal lobe, but not a single ex- cellent result was attained. This group was identifi- able on the basis of depth EEG alone. Besides depth EEG, no other criteria, including neuropsychological testing, radiographic investigation, specialized EEG techniques, or historical information, were adequate to distinguish any of these 7 patients from the 8 with localized scalp EEGs.

If the 14 patients with scalp EEGs well localized to the anterior temporal lobe had undergone surgery without depth EEG study, results would have been good or excellent in 9 of the 12 with sufficient follow-up for the result to be assessed (75%). This number is well within the generally reported range of surgical success in refractory focal epilepsy [2, 4 , 9- 12, 20, 21, 28, 29, 321. Most often, however, depth EEG was not used in these surgical series. The use of depth EEG for selection could have raised the surgical success to 100% in this group. Furthermore, the degree of response to surgery in terms of percent of seizure reduction can be better predicted by depth EEG. If 75% reduction in seizure frequency is acceptable, patients with more posterior temporal foci should have standard temporal lobectomies; if not, a surgical procedure can be avoided with better knowledge that

Spencer et al: Localizing Value of Depth EEG 251

complete control of seizures cannot be effected. Similarly, the 3 patients with multifocal seizure onset and the patient with frontal lobe seizure onset, all detected by depth EEG evaluation alone, could have been spared unsuccessful surgical procedures.

Of the 16 patients with unlocalized scalp EEGs, depth EEG confirmed a multifocal seizure onset in 13 and demonstrated focal and consistent electrically localized seizure onset in only 3. Of the latter, 2 have undergone surgery with 1 good and 1 excellent re- sult. Of the 13 patients in whom depth EEG confirmed the scalp EEG findings of an unlocalized epileptogenic area, 7 underwent surgical resection of the cerebral area apparently responsible for initiating the greatest number of seizures; this was the origin of fewer than 80% of seizures as determined by depth EEG. Of this group, all have had fair or poor surgical results. Thus, depth EEG was able to identify some appropriate surgical candidates from the group of patients not being considered for surgery on the basis of scalp EEG, but its yield in this respect was not high. A previous review [28] indicates that other centers have had a higher detection rate of localized seizure foci on depth EEG among patients with un- localized scalp records [18]. Perhaps this is due to patient selection, number of depth sites studied, or number of seizures recorded.

Depth EEG has been the most accurate criterion in our hands for localization of epileptogenic brain tis- sue as judged by surgical outcome. The procedure has traditionally been used to detect single foci in pa- tients whose seizures cannot be localized by other means, particularly by scalp EEG [7, 181. We have confirmed that depth EEG indeed can select some of these patients, though its yield in this respect was low (3 of 16 patients). More impressive was the ability of depth EEG to identify an area of epileptogenesis other than that predicted by the localized scalp EEG or any other localizing criterion, removal of which produced a high rate of surgical success. This use of the technique, although not previously promoted, has had a considerably higher yield (7 of 15 patients).

If our objective in the surgical management of re- fractory focal epilepsy is to attempt to “cure” sei- zures, or to be able to predict with some confidence the degree of reduction in seizure frequency to be offered as a result of a specified surgical procedure, then depth EEG study is a necessary part of the evaluation of each patient since it provides the most accurate means of localization. If some patients with multifocal or unlocalized scalp EEG recordings and medically refractory partial seizure types are to be offered a chance for surgical treatment, depth EEG is absolutely necessary. However, depth EEG eval- uation represents a major expense [26], a moderate time investment [26], and a minor but definite risk

[3, 17, 251. Accordingly, surgical therapy could be offered to some patients without prior depth elec- trode study. For example, in 70 to 80% of patients showing scalp EEG abnormalities localized to the temporal lobe, one could predict at least a 75% re- duction in seizure frequency from temporal lobec- tomy. In contrast, using depth EEG as part of the preoperative localization, a similar patient could be told more clearly that the expected reduction in sei- zure frequency was either 95%, greater than 75% but less than 95%, or less than 7596. This informa- tion may be crucial to some patients in making deci- sions about surgical treatment.

From this series, it appears that extensive testing in the form of radiological investigation and neu- ropsychological measures should not be used for localizing purposes. Also, unless further analyses show that they correlate better with surgical outcome than do depth EEG results, the various specialized EEG recordings done at many epilepsy centers ap- pear to add little to the evaluation or treatment of the patient with refractory epilepsy. The results of newer techniques, such as positron emission tomog- raphy, may alter these conclusions [6].

References 1.

2.

3.

4 .

5.

6.

7.

8.

9.

10.

11.

12.

Bengzon ARA, Rasmussen T, Gloor P, et al: Prognostic fac- tors in the surgical treatment of temporal lobe epileptics. Neurology (Minneap) 18:717-731, 1968 Bhatia R, Kollevold T: A followup study of 91 patients oper- ated on for focal epilepsy. Epilepsia 17:61-66, 1976 Crandall PH: Developments in direct recordings from epileptogenic regions in the surgical treatment of partial epilepsies. In Brazier MAB (ed): Epilepsy: Its Phenomena in Man. New York, Academic, 1973, pp 288-3 14 Crdnddl PH: Postoperative management and criteria for evaluation. In Purpura DP, Penry JK, Walter R D (eds): Neurosurgical Management of the Epilepsies. Adv Neurol

Engel J, Crandall PH, Brown WJ: Consistent false lateraliza- tion of seizure onset with sphenoidal and scalp telemetered EEG recordings in two patients with partial complex epilepsy. Electroencephalogr Clin Neurophysiol 50:60P, 1980 Engel J, Kuhl DE, Phelps ME, et al: Pathological correlates of focal temporal lobe hypvmetabolism in partial epilepsy. Epilepsia 22:236, 1981 Gloor P: Contributions of electroencephalography and elec- trocorticography to the neurosurgical treatment of the epilep- sies. In Purpura DP, Penry JK, Walter R D (eds): Neurosurgi- cal Management of the Epilepsies. Adv Neurol 8:59-106, 1975 Green JR: Temporal lobectomy, with special reference to selection of epileptic patients. J Neurosurg 26:584-593, 1967 Green JR, Scheetz DG: Surgery of epileptogenic lesions of the temporal lobe. Arch Neurol 10:33-46, 1964 Greenwood J, Kellaway P: Surgical treatment of epilepsy. Texas Med 65:64-7 1, 1969 Jensen I: Proceedings of the Scandinavian Neurosurgical So- ciety. Pathology and prognostic factors in temporal lobe epilepsy. Followup after temporal lobe resection. Acta Neurochir (Wien) 31:261-262, 1975 Jensen I: Temporal lobe surgery around the world: results,

8:265-280, 1975

252 Annals of Neurology Vol 12 No 3 September 1982

complications, and mortality. Acta Neurol Scand 52:354- 373, 1975

13. Jensen I: Temporal lobe epilepsy with special reference to surgical results, neuropathology, and social conditions. Dan Med Bull 24:235-251, 1977

14. Jensen I, Vaernet K: Temporal lobe epilepsy. Followup in- vestigation of 74 temporal lobe resected patients. Acta Neurochir (Wien) 37:173-200, 1977

15. Lieb JP, Walsh GO, Babb TL, et al: A comparison of EEG seizure patterns recorded with surface and depth electrodes in patients with temporal lobe epilepsy. Epilepsia 17:137-160, 1976

16. McNaughton FL, Rasmussen T: Criteria for selection of pa- tients for neurosurgical treatment. In Purpura DP, Penry JK, Walter R D (eds): Neurosurgical Management of the Epilep- sies. Adv Neurol 8 3 - 4 8 , 1975

17. Ojemann GA: The future role of surgery in the treatment of epilepsy. In Wada JA (ed): Modern Perspectives in Epilepsy. Eden Press, Great Britain, 1978, pp 209-226

18. Olivier A, GIoor P, Ives J: Stereotaxic seizure monitoring in patients with bitemporal epilepsy. Presented at the Annual Meeting of the American Association of Neurological Sur- geons, Toronto, April 1977

19. Penfield W, Steelman H: The treatment of focal epilepsy by cortical excision. Ann Surg 126:740-761, 1947

20. Rapport RL, Ojemann GA, Wyler AR, et al: Surgical man- agement of epilepsy. West J Med 127:185-189, 1977

21. Rasmussen T: Cortical resection in the treatment of focal epilepsy. In Purpura DP, Penry JK, Walter RD (eds): Neurosurgical Management of the Epilepsies. Adv Neurol 8:139-154, 1975

22. Rasmussen T: The place of surgery in the treatment of epilepsy. In Morley TP (ed): Current Controversies in Neurosurgery. Philadelphia, Saunders, 1976, pp 463-477

23. Ray CD: A new multicontact multi-purpose depth probe: first experimental results. Mayo Clin Proc 40:781-790, 1965

24. Rossi GF: Considerations on the principles of surgical treat- ment of partial epilepsy. Brain Res 95:395-402, 1975

25. Soloway SS, Williamson PD, Spencer DD, et al: Surgery for epilepsy: role of depth electroencephalography. Conn Med

26. Spencer SS: Depth electroencephalography in selection of refractory epilepsy for surgery. Ann Neurol9:207-2 14, 1981

27. Spencer SS, Spencer DD, Williamson PD, et al: Ictal effects of anticonvulsant medication withdrawal in epileptic patients. Epilepsia 22:297-307, 1981

28. Spencer SS, Williamson PD, Spencer DD, et al: Value of depth electroencephalography in 2 1 refractory epileptic pa- tients. Trans Am Neurol Assoc 104:241-243, 1980

29. Talairach J, Bancaud J: Stereotaxic exploration and therapy in epilepsy. In Vinken PJ, Bruyn CW (eds): Handbook of Clini- cal Neurology. Vol 15. The Epilepsies. Amsterdam, North Holland, 1974, pp 758-782

30. Talairach J, Bancaud J, Bonis A, et al: Functional stereotaxic exploration of epilepsy. Confin Neurol 22:328-330, 1962

31. Van Buren JM, Ajmone-Marsan C, Mutsuga N: Temporal lobe seizures with additional foci treated by resection. J Neurosurg 43:596-607, 1975

32. Van Buren JM, Ajmone-Marsan C, Mutsuga N, et al: Surgery of temporal lobe epilepsy. In Purpura DP, Penry JK, Walter RD (eds): Neurosurgical Management of the Epilepsies. Adv Neurol 8:155-196, 1975

44~70-75, 1980

Spencer et al: Localizing Value of Depth EEG 253