Anterior temporal electrodes in complex partial seizures

Anterior temporal electrodes in complex partial seizures

Electroencephalography and chnlcal Neurophysmlogy, 1988, 70 105-109 105 Elsevier Sclenufic Pubhshers Ireland, Ltd EEG03395 Anterior temporal elect...

305KB Sizes 2 Downloads 19 Views

Electroencephalography and chnlcal Neurophysmlogy, 1988, 70 105-109


Elsevier Sclenufic Pubhshers Ireland, Ltd


Anterior temporal electrodes in complex partial seizures R i c h a r d W. H o m a n , M a r k C. J o n e s a n d S u m a n t R a w a t Department of Neurology, Umoerslry of Texas Health Sctence Center and Neurology Sermce, Dallas Veterans Adm,mstratwn Medical Center, Dallas, TX (U S A ) (Accepted for p u b h c a t m n 11 December 1987)

Swnmary Anterior temporal (AT) and sphenoldal (SP) electrodes as well as standard 10-20 electrodes were used to evaluate pattents w~th recent onset complex parUal seizures A total of 58 eplleptiform focl were found m 80 patients A T electrodes stgmficantly improved detectaon of foca m comparison with 10-20 electrodes_ SP electrodes resulted m a small, non-slgmflcant increase in detectaon in cornpanson to A T electrodes A T electrodes m a y be substituted for basal electrodes for the purpose of &agnosls in the majority of pattents with recent onset complex parUal seizures

Key words: Epilepsy, E E G sptkes; Temporal lobe

Complex partial seizures (CPS) are the single most common type of epileptic seizure (Hauser and Kurland 1975). Epfleptfform EEG activity associated with CPS is found in limbic structures, particularly meslal temporal and mestal frontal cortex (Klass 1975). Basal electrodes, both nasopharyngeal (NP) and sphenoidal (SP) have been employed to define EEG abnormalities m patients w~th CPS (Maclean 1949; Maclean and Arellano 1950; Rowt et al, 1961; Moniem and Chuntamanee 1969). Anterior temporal (AT) electrodes were described by Silverman (1960) as an adjunct to the 10-20 system. Recent findings suggest that recording from this position would be effective for evaluation of patients w~th CPS. Wilkus and Thompson (1985) found that sphenoidal electrodes rmgrate laterally with time, yet still provide adequate recordings. Laxer (1984) found that 'mma-sphenoxdal' electrodes, employing a subcutaneous location, compared favorably to true

Correspondence to" Rachard W Homan, M D , Director, Regtonal Epilepsy Center, VA Medtcal Center, 4500 S_ Lancaster R d , Dallas, TX 75216 (U.S A.)

sphenoidal electrodes. In patients with CPS Morris et al. (1986) found that epileptiform activity was most often maximal in the D9, D10 electrodes, wluch are equivalent to AT electrodes. Sperhng and Engel (1985) have found that both ear and AT electrodes were as effective as NP electrodes for the detection of spike discharges in patients suspected of having epilepsy. These same investigators have also reported that SP electrodes are more sensitive than ear or NP electrodes (Sperling et al. 1985). A dtrect comparison of SP and AT electrodes has not been described, however. King et al. (1986) have extensively discussed the use of sphenoldal electrodes. They point out that there are 3 reasons for using sphenoidal electrodes: (1) documentation of an mterictal discharge in patients felt to have true seizures; (2) diagnosis of patients w~th questaonable events; and (3) localization of the primary site of discharge m patients being considered for surgery. The value of SP electrodes has been well documented for the last purpose (Rov~t et al. 1961; Flamgln et al. 1985). However, the first two are more relevant for new onset, previously undaagnosed setzure patients. In such patxents invasive techniques are not indicated until non-invaslve

0013-4649/88/$03 50 © 1988 Hsevmr Scientific Pubhshers Ireland, Ltd_

106 scalp electrodes have been found to be ineffectwe for documenting an abnormahty. We hypothestzed that, for detection of an interlctal discharge, A T electrodes m an appropriate montage would be equivalent to SP electrodes and improve diagnosUc capability in comparison to 10-20 electrodes used in a standard antero-postenor (A-P) temporal chain

Methods We studied previously undlagnosed patients referred for evaluatxon of possible seizures_ Referring physicians included both staff and house staff from a variety of medical specialties. Fifty-eight patients had no diagnostic epdeptiform discharges on initial routine waking EEG, which was performed employmg standard 10-20 system electrode placement with both bipolar and referential montages. The follow-up E E G on these patients was performed following overnight sleep deprlvauon and included both waking and sleep portions_ For this recordmg we employed a combination of 10-20, AT and SP electrodes. A T electrodes were placed accordmg to Sllverman (1960). SP electrodes of local manufacture were employed. Informed consent was obtamed for the placement of SP electrodes These were inserted as described by Sperlmg and Engel (1986). Based on the imtlal findings, we studied an additional 68 newly referred patients with AT and 10-20 electrodes only. Simultaneous comparison of A T and SP electrodes was accomplished by employing the same montage for both: A1-X1 (X = A T or SP), X1-Nz (nose), Nz-X2, X2-A2. This montage was modified from one recommended by Sperhng and Engel (1985) and is, in essence, a subtemporal bipolar chain. We also compared A T and SP electrodes to a standard bipolar antero-postenor temporal chaan incorporating 10-20 electrode placements In addition, m selected pataents we sequentially compared AT, SP, and 10-20 electrodes in the above montages to one employing a common average reference electrode. Recordings were mterpreted by 2 of us (M J, RH) independently, and judgments made as to whether detection of epdeptfform actwaty was

R w_ HOMAN ET ,~L accomphshed by the 10-20 system, or, if first AT, then SP electrodes were necessary. A third electroencephalographer was employed to settle any disagreements An electrode type was considered necessary for detection ff uneqmvocal epdeptfform discharges were recorded by that type but not by preceding electrode types m the sequence 10 20, AT, SP. Fmally, while bhnded to E E G data, we retlospectlvely evaluated in detml the patient histories to determine the clinical probability of the presence of complex partial seizures at the time of lmtial referral. Patients were classified as possibly having complex partial setzures (group A) according to current international criteria (Commission on Classification and Terminology 1981) if they had experienced two or more identical events conslsting of alteration of consciousness with postictal confusion and anmesm for the event. Appropriate simple partial onset, secondary generalization, and the presence of automatisms were considered supportive, but not essentml If the htstory suggested seizures other than CPS, patients were classified m group B. A thtrd group was composed of patients who were thought to probably not have seizures (group C)_ A chi-square analysis was performed for comparison of 10-20, AT, and SP electrodes for detection of spikes and sharp waves in all groups.

Results For the entire group of 126 patients, 80 were classified in group A by chmcal h~story alone, while 30 were classxfied m group B, and 16 in group C. In group A, 37 patients had 52 foci For groups B and C, the eqmvalent numbers were 12/16 and 3/3, respectively. Table I compares the number of patients and f o o detected by the 10-20 and A T electrodes in each group of patients In group A, A T electrodes were slgruficantly superior to 10-20 electrodes for detection of both patients and f o o at the P < 0.001 level The lack of significant difference between 10-20 and A T electrodes for groups B and C may be attributable both to the smaller numbers in these groups and to the


107 AI-S ! ~




Detectton of epdeptogemc discharges by 10-20 and AT electrodes 10-20

AT detections




11 19

37 a 52 b

Group A (patzent N Patients FOCl

= 80)

26 ~ 33 b

Group B (pattent Patxents Foct


N = 30)


10 14

Group C (panent

N = 16)

Pauents Foc]

3 3

2 2

12 16

0 0

3 3

T4 P3 "01 P4 - 0 2

~.b P < 0001

fact that patients w~th CPS by history were excluded from these groups. Of the 58 patients who had both A T and SP electrodes, 46 were in group A. In this group, 30 independent epileptfform foca were found m 24 patients. Table II indicates the effectiveness of AT and SP electrodes for detection of patients and foe1 in group A patients. AT electrodes were sigmficantly superior to 1 0 - 2 0 electrodes, while the total number of patients and loci detected by the AT and SP electrodes was not significantly different. Comparison of A T and SP electrodes for detection in patient groups B and C also revealed no significant differences. Fig. 1 lUustrates a case in which AT electrodes were necessary for detection of epileptiform dzscharges. The event seen in the left 1 0 - 2 0 electrodes was considered suggestive, but not defi-




10-20 detections

AT deteCtions Add1-


tmnal 12 " 13 b

~.b P < 0001.

9 14

SP detections Ad&-


tzonal 21 " 27 b

3 3

24 30




e ~" " ~ " ..I,-



Fig. 1. Pattent with possible CPS. EEG reveals a sptke m the left AT electrode while the pattern seen m the left 10-20 electrodes was not considered to be of diagnostic slgmficance

mtely diagnost]c of an eplleptiform discharge. Absence of well-defined discharges in the SP derivation suggests a lateral focus. Fig. 2 compares the subtemporal chain, shown in a, to an average reference montage (b) for each b Nz_l~f



~ •


S 1 - Nz Nz - 8 2

.-;~m O~M=

S2 -

S2 _ A 2 ~ A 1 - T1 "------4



T1 T2A1 -

T1 - N z

A2 -

Nz-T 2


T2 - A 2


Fpl-F T F7 _ T3 ~ , ~ . ~ l ~ . ~ . ~ . ~ _ . _ - _ , , ~ .

T3T4 -

T3 - T5 ~


F p 2 - F8

Comparison of 10-20, AT and SP eleCtrodes for deteCtmn of epfleptogemc &scharges m CPS pauents (group A)

Pataents FOCl


FT-Ta Ta-'rs



T4 ~



P 3 - O1

F36 m


c~43 ~ • 5 0 I.NI 1 s a c P4- 02 ~ " EKG











J ----..~-


Fig 2 The subJeCt had possible CPS. Spikes were deteCted by the 10-20, AT, and SP electrodes and were locah2.ed by the left SP electrode a subtemporal bipolar chaan b c o m m o n average reference



of the electrode types. I n a, discharges are more frequent in the SP electrode, suggesting a mesial or basal focus I n b, discharges are seen in b o t h the A1 a n d Nz electrodes, c o n f i r m i n g that these are active sites. The a m p h t u d e of the discharge supports the m e s l a l / b a s a l location The s u b t e m poral chain a n d average reference montages were equivalent in defining the discharges. Both A T a n d SP electrodes detected eplleptlform foci n o t detected b y the other electrode or b y the 1 0 - 2 0 electrodes. SP electrodes detected 4 focl n o t detected by A T a n d 1 0 - 2 0 electrodes (3 in group A, 1 in group B) However, in these cases mterrrattent theta frequency focal slowing was seen in 1 0 - 2 0 or A T electrodes

p l a c e m e n t over the 1 0 - 2 0 placements Daly (1979) has advised against the use of a n ear reference electrode for e v a l u a t i o n of temporal discharges because the ear participates m these discharges a n d such a referential m o n t a g e (with 1 0 - 2 0 electrodes) would be subject to m - p h a s e cancellation W e did not find this to be a p r o b l e m with the s u b t e m p o r a l chain As shown in Fig 2, the subtemporal chain a n d a n average reference electrode m o n t a g e were equally effective. Use of the A T electrodes in either a s u b t e m p o r a l chain or an average reference m o n t a g e will allow the majority of E E G laboratories to i m p r o v e diagnosis of new onset seizures without subjecting patients to the discomfort of basal electrodes.


The authors wish to thank Dr Joan Relsch for advice concerning statistical analysis, and Jennifer Reed, REEGT, for technical assistance

Our findings c o n f i r m previous work a n d extend the c o m p a r i s o n of basal a n d lateral surface electrodes. Sperhng a n d Engel (1985) f o u n d that ear a n d A T electrodes were as effective as N P electrodes, b u t that SP electrodes were more sensitive t h a n ear or N P electrodes (Sperling et al. 1985). W e confirm the greater sensiavity of SP, in comp a r i s o n to A T electrodes, b u t also find that this is rarely i m p o r t a n t for detection of an event for the purpose of diagnosis A T electrodes are located over the lateral surface of the temporal lobe ( H o m a n et al. 1987). Laxer (1984) speculated that because of this location, A T electrodes would be of hmited value for detecting mesial focl. Fig. 2 dlustrates the abihty of the A T electrode to detect e p d e p t i f o r m discharges a p p a r e n t l y arising i n mesial or basal surfaces. W e have shown that A T electrodes significantly i m p r o v e detectxon of epileptiform activity in c o m p a r i s o n to 1 0 - 2 0 electrodes i n s t a n d a r d A P temporal chains. F u r t h e r i m p r o v e m e n t in spike detection by SP electrodes occurred i n a small n u m b e r of cases i n which i n t e r m i t t e n t focal slow actlvaty was seen m lateral surface electrodes. The m o n t a g e we employed for the A T electrode was used m this case to provide direct c o m p a r i s o n to SP electrodes a n d maximize assessment of the inferior aspect of the temporal lobe. It u n d o u b t e d l y c o n t r i b u t e d to the advantage of the A T

References Classification and Terminology of the International League Against Epilepsy Proposal for revased chmcal and electroencephalograpluc classification of epflepnc seizures Epllepsla, 1981, 22:489-501 Daly, D_D Use of the EEG for chagnosls and evaluataon of epllepUc seizures and nonepdeptlc episodic disorders_ In D W_ Klass and D D. Daly (Eds), Current Practice of Chmcal Electroencephalography Raven Press, New York, 1979 221-268 Flamgm, H, Kang, D and Gallagher, B_ Surgical treatment of epdepsy In T Pedley and B_ Meldrum (Eds.), Advances m Eplleptology, Vol 2 Churchill Llxangstone, Edinburgh, 1985 297-339 Hauser, W.A and Kurland, L T The epldermologyof epilepsy in Rochester, Minnesota, 1935 through 1967 Epdepsia, 1975, 16 1-66 Homan, R W_, Herman, J and Purdy, P Cerebral locauon of international 10-20 system electrode placement Electroenceph chn Neurophyslol, 1987, 66:376-382 King, D W, So, E L, Marcus, R and Gallagher, B B Techmques and apphcataons of sphenoldal recording J Chn Neurophyslol, 1986, 3 51-65 Klass, D W Electroencephalographac mamfestatlons of complex parnal seizures In J K Pertry and D_D Daly (Eds), Advances in Neurology, Vol 11 Raven Press, New York, 1975 Laxer, K D Mlm-sphenoldal electrodes in the Investigation of seizures Electroenceph chn Neurophyslol, 1984, 58 127-129 Maclean, P D A new nasopharyngeal lead Electroenceph chn Neurophyslol, 1949, 1 110-112 Commission on

ANTERIOR TEMPORAL ELECTRODES IN SEIZURES Maclean, P D and Arellano, Z A P Basal lead studaes m epdepUc automatisms_ Electroenceph chn. Neurophyslol_, 1950, 2 1-16. Momem, S.A and Chuntamanee, C. The use of sphenoldal and pharyngeal electrodes m epdepsy Electroenceph clm. Neurophyslol, 1969, 26:340-341 Morns, III, H.H, Luders, H., Lesser, R P , Dlrmer, D S. and Klem, G_H_ The value of closely spaced scalp electrodes m the locahzaaon of epileptlform foo. a study of 26 pauents with complex partial setzures Electroenceph clin Neurophyslol, 1986, 63 107-111 Rovit, R R, Gloor, P and Rasmussen, T Sphenoldal electrodes m the electrographtc study of pauents with temporal lobe epilepsy J. Neurosurg, 1961, 18: 151-158_ Sdverman, D_ The anterior temporal electrode and the ten-

109 twenty system Electroenceph_ chn Neurophyslol, 1960, 12 735-737 Sperhng, M R. and Engel, J. Electroencephalographlc recordmg from the temporal lobes a compartson of ear, antertor temporal, and nasopharyngeal electrodes. Ann Neurol_, 1985, 17- 510-513 Sperlmg, M R. and Engel, J Sphenoldal electrodes J Chll Neurophyslol, 1986, 3:67-73 Sperlmg, M.R., Men&us, J R. and Engel, J The detecaon of temporal lobe spikes: a comparison of sphenoidal, nasopharyngeal, and ear electrodes Neurology, 1985, 35 (Suppl_ 1) 225. Wdkus, R J. and Thompson, P M. Sphenoldal electrode positions and basal EEG during long term momtormg. Epllepsla, 1985, 26 137-142