Intraoperative Basal Temporal Language Maps


 

by

John P. Brockway, Ph.D. Carolinas Epilepsy Center, and Neurosciences Institute, Carolinas Medical Center, Charlotte, NC 28283, USA; MemTestCorp™, Davidson, NC 28036, USA Ronald L. Follmer, M.D. Carolinas Epilepsy Center, Neurosciences Institute, and Department of Internal Medicine, Charlotte, NC 28232, USA Kristin A. Solsrud, B.A. Carolinas Epilepsy Center, and Neurosciences Institute, Carolinas Medical Center, Charlotte, NC 28283, USA Anthony L. Asher, M.D. Carolina Neurosurgery and Spine, Charlotte, NC 28207, USA Carolinas Epilepsy Center, and Neurosciences Institute, Carolinas Medical Center, Charlotte, NC 28232, USA Michael D. Heafner, Sr., M.D. Carolina Neurosurgery and Spine, Charlotte, NC 28207, USA Carolinas Epilepsy Center, and Neurosciences Institute, Carolinas Medical Center, Charlotte, NC 28232, USA

ABSTRACT

This paper reports two cases using a procedure which revealed a basal temporal language area (BTLA) in fusiform gyrus of language dominant hemisphere (L) by employing electrical stimulation and recording of the cortex, with grid electrodes, during computerized speech and object naming tasks. The procedure reduced the time necessary to map the inferior temporal, fusiform, and parahippocampal gyri by identifying the existence of and margins for BTLA during a single craniotomy for planned resection of the anterior temporal lobe in patients with medically intractable complex partial epileptic seizures. Both patients underwent intraoperative mapping for language while awake using low current, biphasic, bipolar electrical stimulation of adjacent pairs of subdural electrodes. During stimulation in the region of the fusiform gyrus, language deficits including speech arrest, dysnomia, and jargon aphasia were elicited. No interictal hot spots were observed within this region. Temporal lobe resection was immediately carried out using knowledge of the location of these language areas to guide resection. Prior reports of temporal lobe (including BTLA) mapping used chronically implanted electrodes (10-14 days), while our procedure considerably reduced mapping time, reduced from two to one the number of invasive surgical procedures and attempted to preserve this language area.

INTRODUCTION

Intraoperative mapping of the basal temporal language area (BTLA) is an efficacious procedure that locates requisite language and memory functions. This mapping procedure served multiple purposes: to localize more specifically those gyri producing interictal epileptiform EEG activity (“hot spots”); to identify the margins of language/memory functions in lateral neocortex; to identify any basal temporal language areas and their associated margins prior to resection; and to reduce from two to one, the number of craniotomies necessary for anterior temporal lobectomy within language dominant hemisphere. Reports have agreed that BTLA is a language area that can be localized through chronic grid implantation (Luders, Lesser et al., 1986). However, the consequences of removing BTLA remain controversial, with some reports suggesting no lasting language effects although there are often reports of short term deficits (Kluin, Abou-Khalil, et al., 1988; Burnstine, Lesser, et al., 1990; Luders, Lesser, et al., 1991; Suzuki, Shimizu, et al. 1992; Abou-Khalil, Welch, et al., 1994) or of mild deficits (Krauss, Fisher et al., 1996). No definitive body of extant literature highlights the importance of removing this area. In a regional epilepsy center, over six years of observations, we developed several concerns: 1) that with BTLA removal immediate post-op naming deficits are more severe than reported, even when these deficits are relatively short-lived (<6 mo) and showed moderate rates of recovery; 2) that BTLA resection may adversely affect the quality of life for more than a few months in moderate and high functioning individuals, particularly those who are best able to qualify for and cooperate during intraoperative mapping procedures; and 3) that the rates of recovery may be lower (take longer, do not recover as fully) than indicated. Our prior experience with patients having chronically implanted electrodes for recording and stimulation, whose BTLA was subsequently resected, revealed through repeated testing either longer lasting (*14 mo) or more severe naming deficits (*54-71% loss) than extant literature suggested. Evidence which supported BTLA existence had largely been produced through chronically implanted grid electrical stimulation of the fusiform gyrus, (Luders, Lesser, et al. 1986; Luders, Lesser, et al. 1991). Electrical stimulation of BTLA has mimicked the pathological states of the language dominant deep temporal lobe, inferior temporal lobe, and Brodmann's area 37, frequently producing aphasias, such as amnestic aphasia (Mills, Martin 1912; Nielson, 1946; Lhermitte, Gautier, 1969; Mohr, 1992). Malow, Bookheimer, et al. (1992) found anomia and dysnomia when stimulating BTLA using visual confrontation naming and auditory responsive naming tasks. Other deficits have been demonstrated during BTLA stimulation such as anomia during the Boston Naming Test, impaired semantic clustering using the California Verbal Learning Test, and overall impaired vocabulary skills (Burnstine, Lesser et al., 1990). Suzuki, Shimizu, et al. (1992) reported the location of BTLA in a pt. with seizures localized to a lesion in the fusiform gyrus. The pt. experienced speech arrest and transient aphasia during these spells, consistent with electrical disturbance in BTLA. A similar case was presented by Abou-Khalil, Welch, et al. (1994). This patient experienced speech arrest during seizures and electrical stimulation of the seizure focus (fusiform gyrus) which they reported as evidence of BTLA involvement. There have been varying reports on whether or not BTLA existed in one form or another in all patients. The proportion of patients exhibiting BTLA varies by report; 13/29 (Schaffler, Luders et al., 1996), 5/10 (Kluin, Abou-Khalil, et al., 1988), 5/5 (Burnstine, Lesser, et al., 1990), 6/12 (Abou-Khalil, Welch, et al., 1994), 20/25 (Krauss, Fisher et al., 1996). Why it existed in some and not others is not yet known. The variance in these reports may have been due to anatomical or physiological variation, varying methodology, and/or patient selection criteria. When found, BTLA has been mapped in dominant hemisphere for language typically in the fusiform gyrus and sometimes involving the parahippocampal and inferior temporal gyri. Schaffler, Luders et al.(1996) used chronically implanted grids and reported the anterior and posterior borders of BTLA as 3.5 cm and 7.5 cm from the anterior temporal tip, respectively. The lateral and mesial borders were reported as 2.5-3 cm and 4-4.5 cm from the lateral edge of the temporal lobe, respectively. The majority of language interruptions, dysnomias, speech arrests in this general area were found directly in the middle of the fusiform gyrus. Krauss et al. (1996 located a center of confrontational naming deficits situated a mean of 4.6±0.5 cm (SEM) from the temporal tip, while word reading sites were situated a mean of 5.4± 0.5 cm from the tip. They have reported schematized maps for six kind language function location. Considering the nature of prior reports of limited language and memory deficits created by stimulating BTLA, there has been controversy over how and whether resection of this area will affect language and memory functions. Several reports suggested little or no noticeable lasting deficits produced by BTLA resection. The most recent and extensive report, (Krauss et al., 1996), involved 20 of 25 pts. monitored for 10-14 days who had a basal temporal language area. When BTLA was resected in 13 of their patients, they performed significantly worse on confrontation naming (pre-post, Fig 3, p. 479 and worse compared to those with BTLA not resected Fig 4. p 480), but did not decline significantly on performance IQ, verbal IQ, or recognition memory. Krauss et al. pointed out however that “only 3 of 13 with BTLA removed had *10% decreases in naming.” Burnstine, et al. (1990) reported one patient who suffered dyslexia for one year after surgical resection of BTLA. Luders, et al. (1991) reported two patients who experienced no decline in language performance following BTLA resection. The pt. reported by Abou-Khalil et al. (1994) had deficits in recalling and naming that resolved within four months following resection of the lesion in the fusiform gyrus. Our own experience has been that the patients who have a BTLA removed have more trouble with language, particularly object naming, for a longer period of time than most have reported, up to two years. We find that there is no uniform agreement about the risk of impairment after resection of this area. There exists a great deal of idiopathic heterogeneity of this region and therefore there exist uncertainty and concern over removing the temporal lobe without mapping BTLA. It is also possible and we suggest that resection of BTLA may cause more lasting deficits in language functions than have been reported, especially in high functioning patients. Extensive repeated testing with the Boston Naming Task can lead to improvement in scores due to learning effects which themselves may mask naming deficits over time. Results from earlier studies may have been under-reported due to the pooling of non-BTLA with BTLA patients. We suggest that diminished object naming performances speak to a quality of life issue and relate directly to the duty to inform patients of these possible outcomes. Our intraoperative speech mapping procedure allowed BTLA to be quickly and economically located, thus possibly spared. The time spent mapping the temporal lobe using this procedure was reduced from 10-14 days to two hours in the operating room. The testing was reliable in that it was computerized; each task presentation was controlled for onset and duration time, spatial frequency and contrast, brightness and distance. Testing was tailored to specific functions for which BTLA seems responsible, and results were replicable. For example, random continuous computer object naming task was tailored for this procedure while the timing of and between presentation of the objects was held invariant, as were all conditions of visual presentations (e.g., contrast, familiarity, size, luminance, etc.). During intraoperative mapping procedures performed while these two patients were awake, we found speech arrest, jargon aphasia, and anomia while stimulating BTLA. Considering the severity of language disturbances that were produced by electrical stimulation of BTLA, prior experience with BTLA patients following resection, and lack of interictal “hot spots” generated from this region during this testing phase, BTLA was spared in both of these cases. The procedure was efficacious because it required one surgery rather than two, required for typical chronic speech mapping procedures. This procedure allowed us to locate and spare BTLA in both patients we report.

MATERIALS AND METHODS

Patient 148 A 30y, RtH, WM was admitted for intraoperative mapping to determine the specific location of language areas prior to resection of his left temporal lobe. He had meningitis as an infant and developed seizures at 13 years of age. Seizures were complex partial with secondary generalization. He was left hemisphere dominant for language (Intracarotid Sodium Amobarbital Procedure or IAP), and had failed the following anti-epileptic drugs: valproic acid, phenytoin, lamotrigine, carbamazepine, gabapentin, and phenobarbitol. Scalp EEGs localized the seizure focus to the left temporal lobe. His magnetic resonance imaging (MRI) showed left hippocampal atrophy. A PET scan showed hypometabolic activity in the left anterior temporal lobe. A SPECT scan showed hypometabolic activity in both temporal lobes interictally and hypermetabolic activity in the left temporal region ictally. The EEG revealed spikes in the left anterior temporal area. After lateralization and localization of seizures through video and EEG monitoring with scalp and sphenoidal electrodes, the patient was considered as a candidate for surgery. Following the intraoperative mapping procedure, his left temporal lobe was resected, measured from the anterior pole, extending posteriorly 3-4 cm along the superior temporal gyrus, approximately 5 cm along the middle temporal gyrus and approximately 5 cm along the inferior temporal gyrus. The patient was tested with the Brockway Memory Assessment Battery (BMAB 3.0) (23 computerized memory tasks) preceding and one month following the surgery. Patient 149 A 42y, LtH, WF was admitted for intraoperative mapping to determine the specific location of language and memory functions prior to resection of her left temporal lobe. She developed seizures at 13 years of age. Seizures were complex partial. She had mixed dominance for language (Intracarotid Sodium Amobarbital Procedure or IAP), and had failed the following anti-epileptic drugs: lamictal, carbamazepine, phenobarbitol, phenytoin, valproic acid, primidone, and gabapentin. Scalp EEGs localized the seizure focus to the left temporal lobe. Her magnetic resonance imaging (MRI) showed left mesiotemporal sclerosis. A PET scan showed moderate hypometabolism in the left anterior temporal lobe. A SPECT scan showed hypometabolic activity in the left temporal lobe interictally and hypermetabolic activity in the left temporal region ictally. The EEG revealed spikes in the left anterior temporal area. After lateralization and localization of seizures through video and EEG monitoring with scalp and sphenoidal electrodes, the patient was considered as a candidate for surgery. Following the intraoperative mapping procedure, her left temporal lobe was resected, measured from the anterior pole, extending posteriorly 4.0 cm along the superior temporal gyrus, approximately 4-4.5 cm along the middle temporal gyrus and approximately 4.5-5 cm along the inferior temporal gyrus. The fusiform gyrus was spared posterior to the 3.0 cm mark. The patient was tested with the BMAB 3.0 (23 memory tasks) preceding and one month following the surgery, (Brockway et al., in press). Intraoperative Mapping Procedure Surgical drapes were placed so the patient could see and hear various phrases and pictures which were projected on the computer screen. IV sedation was induced with a rapidly acting hypnotic agent (Propofol), and the scalp was anesthetized with local anesthetic mixture of 0.5% Lidocaine and 0.25% Marcaine. A question-mark shaped incision was made in the left temporal region and the scalp and temporalis muscles were reflected anteriorly. A temporal bone flap was then raised with the drill and craniotomy. The dura was opened using a U-shaped incision and was reflected posteriorly. The floor of the middle fossa and the temporal tip were identified. We measured back approximately 5.0 cm from the temporal tip along the middle and inferior temporal gyri and identified anatomic structures, i.e., large cortical veins, which were used during the mapping portion of the procedure. The patient was allowed to awaken from his/her Propofol sedation. An AD-TECH 5x4 grid electrode array was placed over lateral temporal neocortex. A 2x8 strip electrode was inserted lateral to medially beneath the temporal lobe with the anterior row approximately 4.5 cm from the anterior pole of the temporal lobe for Pt. 148 and 3-4 cm back for Pt. 149. Each round contact measured approximately 2.36 mm in diameter, with distance between the electrodes, CTC=1.0 cm. Two neurosurgeons concurred on placement and location using MRI and in situ. measurements. In order to locate and preserve both language and memory functions in medial cortex, biphasic, bipolar electrical stimulation was conducted (Grass S12 stimulator). This experimental approach was approved by the Institutional Review board as part of a three year study of the medial lobe structures and their relationship to complex role of memory, and patient signed appropriate consent forms. Stimulation. Unilateral stimulation of the region of hippocampus and fusiform gyrus was conducted using adjacent pairs of contacts on each grid electrode array. Stimulation was delivered by a Grass S12 Isolated biphasic stimulator which delivered a 0.3 ms square wave pulse of alternating polarity with frequency of 50 Hz. Trains of stimuli of five seconds were ordinarily delivered beginning with a 0.5 mA current. These trains of stimuli were increased in stepwise fashion until an afterdischarge or a positive symptom occurred. A positive symptom was defined as an auditory, visual, gustatory, olfactory, or kinesthetic sensation as reported by the patient. These symptoms could also include subjective pain (face, eye, head). Electrical intensity used for testing was delivered 0.5 mA below afterdischarge or positive symptom threshold. Language & Memory Tasks Using this experimental approach, over the course of two hours, the patient was asked to perform a series of language and memory tasks on a monitor driven by a laptop computer. The monitor was rotated 90o so that the patient could see the screen comfortably while on her/his side. The three principal tasks were the reading of a story, continuous object recognition task, and a word-picture paired-associates (PA) task. The reading task involved the presentation of a paragraph of a story on the screen for the patient to read. The PA task presented a word and a picture, controlled for imagery, pronouncability, concreteness, and frequency. The PA task entailed looking at a simultaneously presented picture and word for 5 seconds and naming each of them. The object recognition task used continuous pictures that appeared on the monitor for two seconds each, no inter-stimulus interval. Stimulation was always delivered during the encoding phase of the task, beginning two seconds prior to visual onset of the PA or object recognition task. Duration of the stimulation pulse stream was 10.0 sec. The tasks were run consecutively, in alternating fashion as testing of each pair of electrodes proceeded, and were invariantly ordered, first reading the story, then PA, or first reading the story and then object recognition. All intratask timing, ITI's, visual presentations, warning signals, etc., were presented and controlled by microcomputer, in the surgical suite. Digitized objects were shown in 256 levels of grayscale and were equated for spatial frequency and visual contrast. Digitized sound was adjusted for patient comfort. Twenty-one testing trials were conducted in left hippocampus for Pt. 148, fifteen testing trials for Pt. 149. Testing and stimulation parameters were held to a minimum to guard against possible effects of kindling. Photographs of grid in place were taken to record position accurately. Schematic computer generated maps were used to record findings from each electrode site. Repetition of testing at various contact pairs was performed to replicate electrode results to insure accuracy for margins for resection. BTLA Grid Placement The 2x8 strip was placed perpendicular to the long axis of and underneath the temporal lobe with the distal pair of electrodes on or near the parahippocampal gyrus. The electrodes were numbered 1 through 16 with 1 as the most medial anterior electrode. Measurements were made from the anterior pole, and L/M distance calculated from coronal T1 MRI. Resections Approximately 4.0 to 5.0cm of lateral temporal cortex measured from the anterior pole was then resected using cautery and sharp dissection. Care was taken to spare areas that were associated with language function. Once the lateral neo-cortical resection was complete, the medial structures (para-hippocampal gyrus and hippocampus) were then removed en bloc using a subpial dissection technique. The mesial temporal structures were then sent for pathologic analysis. The pial edges were cauterized and the wound was lined with hemostatic material, absorbable collagen. The dura was closed in a watertight fashion and the bone flap is secured with plates. The temporalis fascia and galea were closed with interrupted absorbable stitches and the skin was closed with staples.

RESULTS

Pt. 148 Electrodes 9 and 10: These electrodes were stimulated for 5 seconds beginning at 4.0 mA with 2.0 mA steps increased until 15.0 mA was reached without an afterdischarge. Testing was performed at 15.0 mA for 10 seconds. First, the reading task was performed. The patient began to read and as soon as the electrodes were stimulated, there was total speech arrest. This was repeated at 15.0 mA and speech arrest occurred again. There was no afterdischarge. The next task was object naming. The patient could not name any objects (1/sec randomized continuous presentation) during the time of the stimulation of electrodes 9 and 10, exhibiting complete anomia. Electrodes 2 and 3: These electrodes were stimulated in the exact manner that was used for 9 and 10. Here the patient had speech hesitation at 15.0 mA but no speech arrest. Testing was repeated and the patient again had significant speech hesitations. The patient performed the object naming task next and could name the objects before and directly after the stimulation, however experienced complete anomia during the time of stimulation. Electrodes 3 and 4: These electrodes were tested at 15.0 mA and the tasks were performed satisfactorily. Electrodes 10 and 11: These electrodes were stimulated in the same fashion and testing was performed at 15.0 mA. When the electrodes were stimulated, the patient had very halting, hesitating speech. This was repeated and the patient had the same halting, hesitating speech. During the object naming task, the patient could not name any of the objects while being stimulated. Electrodes 11 and 12: These electrodes were stimulated in the same fashion and testing was performed at 15.0 mA. Here the patient could read fluently while being stimulated. However, during the object naming task, he could only name four out of eight objects presented. Pt. 149 Electrodes 1 and 9: These electrodes were stimulated for 5 seconds beginning at 4.0 mA and increasing until 7.5 mA was reached without an afterdischarge. The patient was feeling tingling in her face that began to become painful. Testing was performed at 7.5 mA for 10 seconds. The tasks were performed satisfactorily. However, the patient reported a feeling of déjà vu. Electrodes 2 and 10: These electrodes were stimulated in the exact manner that was used for 1 and 9. The pt. reported tingling in her bottom lip at 7.0 mA. When stimulated at 8.0 mA, the pt. had a "far away" feeling like before a spell. At a second stimulation at 8.0 mA, stimulation made the pt. laugh out loud. At 9.0 mA, the pt. heard a tune during stimulation. The pt. could not name the tune but moved her lips and tongue. At a second stimulation at 9.0 mA, the pt. again heard the tune but could not say the name when asked for it. At 9.5 mA, the pt.'s mouth drew to the left. At a second stimulation at 9.5 mA, there was no drawing of the mouth. Testing was performed at 9.5 mA. During the reading task, the pt. experienced speech arrest followed by jargon aphasia. During the object naming task, the pt. experienced complete anomia during stimulation. Electrodes 2 and 3: These electrodes were stimulated in the same fashion and an afterdischarge was seen at 11.0 mA. Testing was performed at 10.5 mA. When the electrodes were stimulated during the reading task, the patient had jargon aphasia. During the object naming task, the patient could not name any of the objects while being stimulated. There was an afterdischarge during this task so it was performed again. The pt. could not name the objects the second time with no afterdischarge. Electrodes 10 and 11: These electrodes were stimulated in the same fashion and testing was performed at 15.0 mA. The reading task was performed twice. Here the patient read slowly once and read fluently once while being stimulated. The object naming task was performed twice. She missed two objects the first time and one object the second time. Electrodes 3 and 4: These electrodes were stimulated in the same fashion and at 2.0 mA, the pt. experienced shooting pain in her head. Testing was performed at 1.0 mA. The tasks were performed satisfactorily. Electrodes 11 and 12: These electrodes were stimulated in the same fashion and at 1.5 mA, the pt. experienced pain behind her ear. Testing was performed at 1.0 mA. The tasks were performed satisfactorily. Memory Scores The patients were given the BMAB 3.0 before and one month following the surgery. Their scores on language and memory functions did not decline significantly following the surgery. A comparison of these scores revealed no general decline in either recognition or recall functions; and only one specific deficit, in paired associates. No deficits were observed in object recognition (object anomia as seen during intraoperative stimulation). Procedural irregularities led to unreliable variability in paired associate data. These data will not be presented here.

--------------------------------------------- Insert Table 1 about here ---------------------------------------------------------------------------------------------

DISCUSSION

This intraoperative mapping procedure may represent a more efficient way of locating a BTLA in cooperative patients who are candidates for temporal lobe resection. These patients may also be the higher functioning patients who can cooperate during this procedure and may benefit from it most. During this procedure, the patient is awake while grid or strip electrodes are placed and subsequently electrically stimulated as memory and language functions are tested. In our patients, BTLA was demonstrated through replicated total speech arrest and complete object anomia. BTLA was surgically preserved in these patients and their testing scores reflect the consistency of their object naming scores before and after surgery. We found high individual variability of results to be the norm. “Bipolar stimulation of sites separated by only 1 cm could produce entirely different deficits in different patients.” (Krauss et al. 1996). We concur that intraoperative procedures are useful for a limited number of patients such as those who are able to withstand the ordeal. However, these patients are usually the very patients in whom sparing the BTLA may make a significant difference, as they may be individuals who rely more on these visual/verbal recognition skills. These patients were admitted to the hospital for 4 days and experienced only one surgical procedure which was significantly more efficacious and less costly than the other reported procedures. Craniotomy and implantation of a grid, 10-14 days of electrical stimulation, combined with the lobectomy and several days of recovery is a much more arduous pathway. Considering the inconsistency and variance of reports concerning the recoverability of function following resection of BTLA, our procedure would help ensure the preservation of a BTLA with a shortened hospital stay and a single surgical procedure with reduced costs. This procedure has several possible disadvantages. Electrode sites are usually tested during a single phase, rather than over several days. As cortical resistance, impedance, and afterdischarge thresholds are known to change over time, it is possible that these results are highly variable. It was always the case, however, that when we found complete speech arrest or total dysnomia, in BTLA, it was dramatic, replicable, and easily noticed and recorded. This suggests that it gives reliable indication of the region’s certain involvement in language functions.

ACKNOWLEDGMENTS We would like to acknowledge our gratitude to Dr. John G. Malone and Dr. Kenneth Ashkin, whose patients we tested. We are indebted to Regina Cooke, REEG Technician. We are appreciative of the IAP information generously provided by Dr. Bryan Connell, who together with Dr. Howard performed the procedures.

REFERENCES

Abou-Khalil B, Welch L, Blumenkopf B, Newman K, Whetsell WO Jr. Global aphasia with seizure onset in the dominant basal temporal region. Epilepsia 1994; 35(5):1079-1084.

Brockway JP, Follmer RL, Preuss LA, Prioleau CE, Burrows GS, Solsrud KA, Cooke CN, Greenhoot JH, Howard J. Memory, Simple and Complex Language and the Temporal Lobe. Brain and Language 1998, 29; 1-29.

. Burnstine TH, Lesser RP, Hart J, et al. Characterization of BTLA in patients with left temporal lobe epilepsy. Neurology 1990;40:966-970. Fedio P, August A, Sato S, Kufta C. Neuropsychological characteristics of patients with basolateral temporal language. Epilepsia 1992;33(suppl 3):120-121.

Kluin K, Abou-Khalil B, Hood T. Inferior speech are in patients with temporal lobe epilepsy [Abstract]. Neurology 1988;38(suppl 1):277.

Krauss G, Fisher R, Plate C, Hart J, Uematsu S, Gordon B, Lesser RP. Cognitive effects of resecting basal temporal language areas. Epilepsia 1996;37:476-483.

Lhermitte T, Gautier JC. Aphasia. In: Vinken PJ, Bruyn GW, eds. Handbook of Clinical Neurology. vol.4. Amsterdam: North Holland Publishing Company, 1969:84-104.

Luders H, Lesser RP, Hahn J, et al. Basal temporal language area demonstrated by electrical stimulation. Neurology 1986;36:505-510.

Luders H, Lesser RP, Hahn J, et al. Basal temporal language area. Brain 1991;114:743-754.

Malow S, Bookheimer T, Blaxton W, et al. Lateral versus basal temporal electrical stimulation: dissociations in auditory and verbal language tasks. Epilepsia 1992;33(suppl 3): 21.

Mills CK, Martin E. Aphasia and agraphia in some practical surgical relations. JAMA 1912;59:1513-1518.

Mohr JP. Acquired language disorders. In: Asbury AK, McKhann GM, McDonald WI, eds. Diseases of the nervous system: clinical neurobiology. vol.1. Philadelphia: W.B. Saunders Company, 1992:729-740.

Nielsen JM. Agnosia, apraxia, aphasia. Their value in cerebral localization. New York: Paul B. Hoeber, 1946.

Schaffler L, Luders HO, Beck GJ. Quantitative comparisons of language deficits produced by extraoperative electrical stimulation of Broca's, Wernicke's, and Basal Temporal Language Areas. Epilepsia 1996;37(5):463-475.

Suzuki I, Shimizu H, Ishijima B, Tanu K, Sugishita M, Adachi N. Aphasic seizure caused by focal epilepsy in the left fusiform gyrus. Neurology 1992;42:2207-2210.

LEGEND TO THE FIGURE Topotonological brain maps for Pts. 148 and 149 of lateral neocortex including inferior and fusiform gyrus showing existence of basal temporal language area during intraoperative mapping. Subdural grids (2-4x5, 2x8) show localization of language effects during low current biphasic bipolar stimulation.

LEGEND TO THE TABLE Table 1. Memory and language performance scores (percentage correct for 26 specific tasks) for Pts. 148 and 149, prior and subsequent to tailored anterior temporal lobectomy (left--language dominant lobe) following intraoperative mapping of basal temporal language area. RUNNING

HEAD: Basal Temporal Language Area

HOME

Information requests too John P. Brockway, Ph.D. President & CEO Memory Testing Corporation P.O. Box 1621 Davidson NC 28036-1621 USA Pager: 704 356 8625 Cellular: 704 576 1844 Fax: 704 896 7593 Home: 704 892 3665 Work: 704 384 5472 Conference Software Address Specific Directory Server Additional Information: Last Name Brockway, Ph.D. First Name John P. Version 2.1