T HEORETICALLY, the results of cortical resection

Transcription

1 J Neurosurg 60: , 1984 Subdural strip electrodes for localizing epileptogenic foci ALLEN R. WYLER, M.D., GEORGE A. OJEMANN, M.D., ETTORE LETTICH, R.E.E.G.T., AND ARTHUR A. WARD, JR., M.D. Department of Neurological Surgery, University of Washington, School of Medicine, Seattle, Washington v" Surgical removal of epileptogenic brain is indicated for treatment of many medically refractory focal seizure disorders. One of the important factors in providing good results from surgery is the accuracy of identifying the epileptogenic focus. However, accurate localization may be difficult when only standard scalp recordings are used. Many epilepsy centers have used intracranial recording techniques to better define regions of cortical epileptogenicity. Although subdural strip electrodes were first utilized many years ago, the more popular method of intracranial recording has been by intracortical depth electrodes. The authors present their method of placing subdural strip electrodes for extensive recordings from the cortex. To date, this method has been used to provide continuous monitoring of the electrocorticogram in 28 patients for periods up to 3 weeks, with only two minor complications. This procedure is relatively safe and a valuable alternative to placing intracortical depth electrodes. KEY WORDS 9 epilepsy 9 epilepsy surgery 9 cortical electrodes 9 electrocorticography 9 electroencephalography 9 depth electrodes T HEORETICALLY, the results of cortical resection for the treatment of partial (focal) epilepsy should be excellent. However, various surgical series have shown that such surgery abolishes seizures in only 60% to 80% of casesj 1 In many cases, a poor result is due to inaccurate preoperative localization of the epileptogenic foci. This is a risk, particularly when patients have bilateral frontal or temporal epileptiform discharges. Therefore, some surgeons have used various methods of recording directly from the surface or depths of the brain to better localize foci that are otherwise ambiguous. Penfield and Jasper 9 were the first to use intracranial recordings. Wires as well as a type of"strip" electrode were placed in the epidural space. Widespread use of intracortical depth electrodes was popularized by Bancaud and Talairach. 1,~2.13 In contrast to Penfield and Jasper's technique, they implanted multi-contact shaft electrodes which penetrated the brain parenchyma to reach both mesial cortical surfaces and deep gray matter structures such as the amygdala. Although depth electrodes have been successfully used for localizing epileptogenic foci, 3-5 the objective may be equally well realized in many cases from the somewhat simpler technique of subdural strip electrodes. We describe our methods for manufacturing these electrodes and our clinical experience with them. We were successful in monitoring seizures in 28 patients with complicated epileptic foci who were being evalu- ated at the University of Washington Regional Epilepsy Center as candidates for epilepsy surgery. Clinical Material and Methods Design of Electrode Strip All patients were referred to the University of Washington Epilepsy Center for the possibility of surgical treatment of their medically intractable seizure disorder. Each patient had already undergone intensive electroencephalographic (EEG) and video monitoring with conventional scalp and sphenoidal electrodes without clear localization of the epileptogenic foci. Therefore, they were given the option to be monitored with intracranial strip electrodes. We have used newly manufactured electrodes for each individual patient. After removal, each set of electrodes is discarded. This procedure is followed because Jakob-Creutzfeldt disease may be transmitted by intracranial electrodes. 2 Each multicontact strip electrode is made of Silastic, Teflon, and stainless steel components. These materials were selected because they are known to be nontoxic to brain tissue and yet capable of withstanding steam autoclave sterilization without deterioration. Each strip electrode consists of four stainless steel discs (0.07 mm thick and 5 mm in diameter) with Teflon-coated multistranded stainless steel wires spot- J. Neurosurg. / Volume 60/June,

2 A. R. Wyler, et al. FIG. 1. Drawing of strip electrode. Upper: The surface with electrode contacts exposed, showing that the Silastic over the electrodes has been punched out. Lower: The dorsal surface illustrating the electrodes with leads with connectors crimped on. The dorsal surface of the electrode strip has a "pocket" at its tip to house the electrode-inserting instrument (not shown). FIG. 3. Drawing showing two strip electrodes in place. The dura is closed and the wires from one strip electrode lead out through the scalp. The No. 11 needle is positioned to bring the wires from the second strip electrode through the scalp. Each wire is passed through the needle and the needle is then withdrawn leaving only the wires exiting through the needle puncture wound. The scalp is then closed in layers. FIG. 2. Strip electrode being inserted through a burr hole. A linear incision is made through the dura, and tack-up sutures hold the dural edges open for the electrode that is passed over the cortical surface within the subdural space. welded on one of their surfaces. The discs are then attached either 1 or 1.5 cm apart along a mm thick strip of Silastic sheeting by means of Silastic medical adhesive. This is sealed with another strip of Silastic sheeting. A length of wire is also sandwiched between the two sheetings after having been threaded through a 4-cm long Silastic tubing. After the Silastic strips have bonded together, holes 3 mm in diameter are punched out from the Silastic sheeting (on the opposite side from the welded wires), thus providing electrical contact at the center of each disc (Fig. 1). Each strip is then trimmed to 8 mm in width and 5 or 6.5 cm in length for strips with interelectrode distances of 1 and 1.5 cm, respectively. Each wire (approximately 20 cm long) is then terminated by first threading on two short, colored Teflon sleeves. These color sleeves are used to identify electrode contacts after implantation is completed. A subminiature gold-plated pin connector is then crimped to the end of the wire. The pin connectors are small enough to be threaded easily through a No. 11 needle (2.4 mm in inner diameter) 1196 J. Neurosurg. / Volume 60/June, 1984

3 Localization ofepileptogenic foci Fic. 4. Anteroposterior (left) and lateral (right) skull films documenting the position of the strip electrodes. Strip electrodes lie over the mesial and lateral frontal lobes via frontal burr holes. Through bilateral temporal burr holes, strip electrodes have been passed medially under each temporal lobe while two others pass over the temporal lobes laterally. Thus, both frontal and temporal lobes are sampled for mesial and lateral cortical acuwty. that was made for the specific purpose of passing the wires through the scalp. On the dorsal tip of the electrode is glued a Silastic "pocket" to accept the electrodepassing instrument (Fig. 2). Electrode Placement and Monitoring Subdural strip electrodes have been inserted in patients under either local or general anesthesia. However, most of the patients have been operated on under general anesthesia because of the possibility that dehydrating agents and hyperventilation may be needed during electrode insertion if the brain is not slack. A linear incision is made over the proposed insertion site and carried down to the periosteum. A burr hole is placed and enlarged slightly. A linear incision is then made in the dura and "tack-up" sutures placed to provide maximal dural opening. As shown in Fig. 2, the moistened strip electrode is passed between the dura and brain in the desired direction. Up to four electrode strips have been inserted through each burr hole, but it is more common to insert only two. The recording quality of each electrode contact is checked and verified before closure of the burr hole. The modified No. 11 needle is then tunneled between scalp and galea (Fig. 3) for a distance of 4 to 5 cm from the incision. Each electrode wire is brought through the scalp via the needle, which is then removed. Next, the dura is approximated and a lock suture passed around the Silastic tube passing through the dura to secure the electrode. The burr hole is filled with Gelfoam, and the scalp closed in layers. A simple dressing is placed over the incisions exposing as much scalp as possible for simultaneous scalp recordings if needed. We have not had a consistent policy as to whether our patients have been placed on antibiotics postoperatively. Before 24-hour electrocorticographic (ECoG) recordings are started, plain anteroposterior and lateral skull films are obtained to confirm electrode locations (see Fig. 4). More recently we have also been obtaining a computerized tomography (CT) scan to document electrode placement with respect to cortical structures such as the mesial temporal lobe. The EEG recordings are usually done with a referential montage depending on the location of the focus and previous recordings. Usually the reference has been to the scalp and not the neck. Results Twenty-eight patients have undergone 30 procedures for placement of subdural strip electrodes (Table 1). All patients have been placed on simultaneous EEG and video monitoring for periods of up to 3 weeks. We have elected to record for periods no longer than 3 weeks to minimize the risk of intracranial infections. In three cases insufficient spontaneous seizures were recorded for a definitive decision to be made as to epileptogenic foci, and two patients were monitored a second time several months after the first electrode implantation. Each patient had a minimum of four and the majority had eight strip electrodes (32 contacts) placed. Strip electrodes were implanted bitemporally in 13 cases. In all these patients a mesial and lateral temporal strip electrode was used in each temporal burr hole. In six cases electrodes were placed bilaterally along the mesial and lateral frontal regions as well as bitemporally. In five patients only bilateral mesial and lateral frontal electrodes were implanted, and in four other cases specialized arrays were used for more difficult localizations. For example, one patient had seizures which on scalp monitoring clearly originated from the fight hemisphere, yet it could not be determined whether the J. Neurosurg. / Volume 60 / June,

4 A. R. Wyler, et al. origin of the seizures was posterior temporal or central parietal. Thus, arrays were spread over the lateral and mesial parietal and temporal lobes on the right hemisphere. The recordings both interictally and ictally have been excellent in quality, and in many cases have disclosed loci not easily identified on scalp recordings because of various problems, one of which is an electromyographic artifact. Figure 5 is an example of a recording at the start of and during a complex partial seizure that began from the right inferior frontal region. In this 32-yearold man, electrodes had been placed through superior and inferior frontal areas so that recordings could be made from the mesial, lateral, and inferior frontal cortex bilaterally. During the seizure, the patient was having tonic-clonic movements but no artifacts are seen in the recordings. The results of the monitoring are as follows. In 20 patients the recordings localized an EEG epileptogenic focus; 18 of these patients have subsequently undergone epilepsy surgery and have been followed for 1 year or longer. Two patients have not been followed long enough to determine the outcome of their epilepsy surgery. In one patient, monitoring was inconclusive and no surgical decision could be made. Seven patients were found to have either primary generalized epilepsy or bilateral foci (either frontal or temporal); these patients were not thought to be appropriate candidates for a cortical excision procedure. Interestingly, one patient, who was found to have bilateral frontal discharges that could not be lateralized, subsequently underwent an anterior corpus callosum section and has since been significantly improved. Therefore, in this case, the monitoring did not localize a focus yet contributed to the surgical decision anyway. On the other hand, one patient who had previously undergone a right temporal lobectomy on the basis of scalp recordings and had not done well postoperatively had strip electrode monitoring. The strip recordings demonstrated a left temporal focus, and so, although the monitoring localized a focus, it was decided not to operate. Of the 18 patients who have undergone surgery and have been followed over 1 year, five (28%) are seizurefree, six (33%) are significantly improved, and seven (39%) are not improved. Therefore, if both the seizurefree and significantly improved patients are combined, 11 (61%) of the patients with surgery have been helped. This represents 39% of the total number of patients monitored. From these 30 strip electrode implantation procedures, two complications have occurred, for a complication rate of 8%. One patient developed a small brain abscess which was treated successfully with twist drill aspiration and antibiotic therapy. The second patient developed a small cortical contusion which resolved without neurological deficit. In all patients who were subsequently operated on, visual inspection of the brain which underlay the strip electrodes revealed no gross pathological changes. There was no evidence of in- flammatory response and no adhesions. In seven patients brain directly underlying the burr holes was sent for pathological examination and no abnormalities were found. Discussion For many epileptic patients who are considered for cortical resection of their epileptogenic foci, conventional scalp recordings cannot identify the location of their loci. This has led to the development of other strategies directed at two goals. The first is to determine if a focus is present and its location. The second is to confirm the location of a focus that is suspected from other diagnostic data. The strategies may be somewhat different for these two goals. It is known that epileptogenic foci reside in gray matter and not in white matter since the pathophysiology arises from neuronal somata and/or dendrites rather than from their axons. Thus, localizing techniques should provide maximal information regarding electrical activity from cortical gray matter, while causing minimal risk to the patient. Regarding the first goal, the recording technique should preferably not be too discrete since the goal is to scan relatively broad expanses of cortex looking for a focus of which the location is otherwise not known. Thus, electrodes should be larger rather than smaller, and the goal should be to sample the activity of as large an area of cortex as possible. TABLE 1 Placement of subdural strip electrodes in 28 patients Case Electrode No. Locations Findings Outcome 1 bifrontal, bitemporal generalized no surgery 2 bifrontal no localization no surgery 3 bifrontal, bitemporal generalized no surgery 4 bitemporal unilat temporal no surgery 5 bitemporal unilat temporal signif improvemt 6 bifrontal unilat frontal signif improvemt 7 bitemporal bitemporal no change 8 bitemporal, frontal unilat frontal seizure-free 9 bitemporal, bifrontal generalized no surgery 10 bitemporal rt temporal improved 11 bitemporal unilat temporal signif improvemt 12 It temporal unilat temporal no change 13 bitemporal, frontal no seizures no surgery 14 bifrontal, bitemporal It frontal seizure-free 15 bitemporal unilat temporal pending 16 rt temporal & parietal rt temporal no change 17 bifrontal, bitemporal generalized no surgery 18 bitemporal unilat temporal pending 19 bitemporal bitemporal no surgery 20 bifrontal unilat frontal signif improvemt 21 bitemporal unilat temporal no change 22 bifrontal unilat frontal seizure-free 23 bifrontal failure signif improvemt 24 bitemporal, frontal unilat temporal seizure-free 25 bifrontal unilat frontal no change 26 bifrontal, bitemporal bifrontal no change 27 bitemporal unilat temporal seizure-free 28 bitemporal bilat temporal no surgery 1198 J. Neurosurg. / Volume 60 / June, 1984

5 Localization of epileptogenic foci These goals would appear to be best met by a technique that places the electrodes directly on the cortex. Although areas of cortex are clearly infolded and buried, at least one-third of it lies exposed on the external surface of the brain and, furthermore, cortical electrodes will also sample a significant volume of cortex in the superficial banks of sulci. These are some reasons why subdural recording has been utilized. When hunting for foci that are not localized by scalp recording, the flexibility provided for subdural strip arrays can be very productive. The risks are small and there are no compelling reasons for limiting the number of subdural strips that are inserted. Thus, the area of cortex studied is much greater than that provided by probe depth electrodes. On any given electrode track, only a portion of the electrode contacts will reside in gray matter. The solid angle of the individual contacts on the shafts is relatively small so that the volume of brain that is recorded is close to the contact. In addition, if any significant sampling of different areas is to be obtained, this requires multiple electrode tracks. Although clinical sequelae are rare, the biological facts are clear that many neurons are destroyed with each electrode insertion. On the other hand, if the second goal is to be achieved, recording via stereotaxically implanted electrodes may be preferable. In this instance, evidence is available regarding the possible location of a focus and the goal is to confirm this hypothesis. Depth electrodes provide localized recordings and the recording sites can be precisely placed by stereotaxic implantation at the site desired. The general criteria to support the decision to monitor patients with depth electrodes have been discussed by Crandall, 4 and more recently by Walter. 14 In a recent review, Spencer tl discussed the results of depth EEG in the selection of patients with medically refractory epilepsy for surgery. She estimated that from the available data, depth recordings enabled 36% of patients to be operated on who otherwise would have been diagnosed as having nonfocal epilepsy. Furthermore, she stated that such recordings could have prevented surgery in another 18% of cases by demonstrating that multiple independent foci were present. She FIG. 5. Complex partial seizure recorded from subdural electrodes implanted bilaterally around the frontal lobes and on the anterior mesial surface of the temporal lobes of a 32-year-old man. This recording shows focal epileptiform activity arising from the right inferior frontal region (a) followed by a short period of attenuation of all activities bilaterally (b), before clinical manifestations became evident (c), and then developed into a generalized seizure with tonic clonic movements (d and e). Note the lack of movement and electromyographic artifacts in the latter phase. J. Neurosurg. / Volume 60 / June,

6 A. R. Wyler, et al. concluded that depth electrode recording has an important role in evaluating surgical candidates since it may have the potential to alter the surgical decision in more than 50% of patients. However, she also pointed out that only large epilepsy centers are performing these recordings, and by the nature of their clinical practices, they attract more difficult cases for surgical consideration. Although the technique of subdural strip electrodes was originally reported by Penfield and Jasper, 9 it has not gained wide acceptance. With the advent of stereotaxic surgery, there was enthusiasm to utilize this new technology to record from sites inside the skull. It is true that depth electrodes provide the best information regarding the electrical activity of gray matter at some distance from the pial surface including subcortical nuclei. Some believe that it provides the best technique for recording from the amygdala in patients with foci in the temporal lobe. Our experience has been that subdural strips that are passed beneath the temporal lobe provide excellent recording from the uncus, most of the amygdala, and the hippocampus. Goldring 6 has used epidural recording as a part of the technique he employs for surgical therapy. Rather than undertake a craniotomy under local anesthesia and then locally map the discrete extent of the focus, he prefers to undertake such mapping as a preliminary procedure: He thus performs a craniotomy under general anesthesia and places one or more sheets containing grids of electrodes epidurally over the areas he wishes to monitor. These sheets of electrodes provide the opportunity to examine the electrical activity of a fairly broad area of cortex electively, with the patient awake in the recording laboratory. Extended periods of recording are then possible and the observations can be extended by electrical stimulation. The second-stage operation is then performed, again under general anesthesia, and is restricted to only the resection, which is a much shorter procedure than when done de novo under local anesthesia. Lueders, et al., 7 have used similar sheets of electrodes for seizure focus localization as well as investigating the cortical areas involved in somatosensory evoked responses. Marsan and Van Buren 8 have also used strip electrodes very similar to the ones described here for localizing difficult foci. A significant element of the choice of recording techniques relates to the risks involved. In our series, the subdural strip electrode technique caused two complications, but neither resulted in residual morbidity. The actual complication rate for implanted depth electrodes has not been reported except in anecdotal form. Furthermore, Rasmussen 1~ has pointed out that "the ultimate risks of inserting electrodes into the good hemisphere of a brain that has already demonstrated its ability to generate a clinically significant seizure tendency still await assessment." Therefore, it is appealing to consider a technique such as that reported here. In conclusion, we have found the use of subdural strip electrodes to be an important technique for localizing epileptogenic foci in those cases where adequate localization cannot be obtained by extensive scalp EEG monitoring. We have used this technique in 28 patients considered for epilepsy surgery with 0% mortality and an 8% morbidity rate. These two cases with complications occurred earlier in the series when certain details of the technique were being evolved. We use subdural strip recording to replace the technique of stereotaxic implantation of intracerebral electrodes. We believe that the reintroduction of this technique will prove to be a useful tool in the evaluation of candidates for epilepsy surgery. References 1. Bancaud J, Talairach J, Bonis A, et al: La St6r6o-l~lectroenc6phalographie dans l'epilepsie. Informations NeurophysiopathologiquesApport6es par l'investigation Fonctionelle St6r6otaxique. Paris: Masson, Bernoulli C, Siegfried J, Baumgartner G, et al: Danger of accidental person-to-person transmission by Creutzfeldt- Jakob disease by surgery. Lancet 1: , 1977 (Letter) 3. Brazier MAB: Depth recordings from the amygdaloid region in patients with temporal lobe epilepsy. Eleetroencephalogr Clin Neurophysiol 8: , 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 Press, 1973, pp Crandall PH, Walter RD, Rand RW: Clinical applications of studies on stereotactically implanted electrodes in temporal-lobe epilepsy. J Neurosurg 20: , Goldring S: A method of surgical management of focal epilepsy, especially as it relates to children. J Neurosurg 49: , Lueders H, Lesser RP, Hahn J, et al: Cortical somatosensory evoked potentials in response to hand stimulation. J Neurosurg 58: , Marsan CA, Van Buren JM: Epileptiform activity in cortical and subcortical structures in the temporal lobe of man, in Baldwin M, Bailey P (eds): Temporal Lobe Epilepsy. Springfield, Ill: Charles C Thomas, 1958, pp Penfield W, Jasper H: Epilepsy and the Functional Anatomy of the Human Brain. Boston: Little, Brown & Co, 1954, p Rasmussen T: The place of surgery in the treatment of epilepsy, in Morley TP (ed): Current Controversies in Neurosurgery. Philadelphia: WB Saunders, 1976, pp Spencer SS: Depth electroencephalography in selection of refractory epilepsy for surgery. Ann Neurol 9: , Talairach J, Bancaud J: Stereotactic approach to epilepsy. Prog Neurol Surg 5: , Talairach J, Bancaud J: Stereotaxic exploration and therapy in epilepsy, in Vinken PJ, Bruyn GW (eds): The Epilepsies. Handbook of Clinical Neurology, Vo115. Amsterdam: North-Holland, 1974, pp Walter RD: Tactical considerations leading to surgical treatment of limbic epilepsy, in Brazier MAB (ed): Epilepsy: Its Phenomena in Man. New York: Academic Press, 1973, pp Manuscript received July 29, Accepted in final form December 30, Address reprint requests to: Allen R. Wyler, M.D., Department of Neurological Surgery, RI 20, University of Washington, Seattle, Washington J. Neurosurg. / Volume 60/June, 1984