Neuropsychology/Language/Behavior: All Ages

1,2Selim R.Benbadis, and2LeanneHeriaud(1Neurology & Neurosurgery, University of South Florida, Tampa, FL; and2 Comprehensive Epilepsy Program, Tampa General Hospital, Tampa, FL ) Rationale: The Wada test is an important component of the presurgical evaluation for epilepsy, and is performed at most epilepsy centers. It is well known to be a very cursory test with many limitations. Occasionally, the results are such that the test is repeated, but there are very little data on the outcome of repeated Wada tests. Methods: We reviewed all Wada tests performed by a single epileptologist over a 6 year period, and identified patients who had more than one Wada test. Only patients in whom the entire procedure was repated were included. Tests were performed according to the Loring protocol, using either amobarbital or methohexital. All patients who underwent Wada testing were being considered for a temporal lobectomy. Results: Over the 6 year period, a total of 201 patients underwent a Wada test. Of those, 14(7%) patients had the test twice. Reasons for repeating the test were: 5 invalid results due to confounding factor (excessive sedation in 3, no drug effect in 1, anxiety attck in 1); 9 “reversed” memory scores without obvious cause, thus prohibiting surgery. Of the 5 tests repeated due to an invalid test, all had a successful repeat test allowing for a decision (4 had surgery, one did not). Of the 9 “reversed” Wadas, 8 repeat tests were no longer “reversed” and allowed surgery, and one remained reversed and prohibited surgery. Language representation never changed on repeat testing. Conclusions: Repeating Wada tests is well worthwhile, whether there is an obvious cause for invalid data or the results are “reversed,” as there is significant test‐retest variability for memory results. 1Laura A.Drea,1Joseph M.Cunningham,1George L.Morris III, and1Christopher M.Inglese(1 Regional Epilepsy Center, St. Luke's Medical Center, Milwaukee, WI ) Rationale: Duration of drug effect is a critical factor when measuring language and memory function during the intracarotid sodium amytal procedure (Wada test). Positive correlations have previously been reported between patient age and duration of drug effect. Prior studies have also indicated longer recovery times following the second injection if administered on the same day and greater sedation following dominant hemisphere injection. The purpose of this study was to identify potential correlations between age and recovery times following sodium amytal injection. Potential differences in order and side of injection, and gender were also explored. Methods: A retrospective database study was completed of 93 patients who had Wada testing between 2002 and 2005 as part of their pre‐surgical evaluation for medically intractable epilepsy. The Regional Epilepsy Center employs the Wada test protocol developed by the Medical College of Georgia. Patients whose injection amounts differed between injections were excluded from this study (36/93). A total of 57 patients were included in this analysis (114 hemispheric injections). The patient group included 36 female and 21 male patients ranging in age from 8 to 73 years (mean = 38.32; SD = 15.86). Recovery times for motor function and EEG were correlated with each other and with age using Spearman's rank correlation coefficient. Within subjects analysis of variance was conducted to explore effects of order and side of injection, and gender on recovery times. Results: Recovery time to baseline motor strength, drift, asterixis, and EEG were significantly correlated with one another (p < .0001). Age, however, did not correlate with any of the recovery variables (p > .05). There was also no significant effect of injection order, side of injection, or gender on any of the recovery time variables, based upon multiple ANOVA comparisons. Conclusions: Contrary to results of prior studies, this study found no correlation between age and recovery time for motor variables or baseline EEG. Additional analysis of our data using 10‐year age clusters similar to those utilized in the Segal et al (2002) study did not reveal similar results, in that our youngest age clusters did not reveal faster recovery times, and the reason for this descrepancy in results is unknown. In addition, prior research has suggested a cumulative effect of injections, but this study found no difference in recovery between the first and second injections. Furthermore, there were no noted differences in recovery times following left versus right injections or male versus female patients. The lack of significant correlations between age and recovery of function, between administration order, side of injection, and gender suggest that Wada testing may be more robust than previous studies have indicated. Further prospective research on these variables with larger samples is recommended to determine their relationship, if any, on recovery of function following injection. 1ChristianElger,2HermannStefan,3CarlosPerdomo, and3SantiagoArroyo(1Klinik fur Epileptologie, Bonn, Germany;2Neurologische Klinik, Erlangen, Germany; and3 Eisai Global Clinical Development, Ridgefield Park, NJ ) Rationale: This study assessed the dose‐range relationships of rufinamide versus placebo for efficacy, safety, tolerability, and pharmacokinetics in patients with inadequately controlled partial seizures. Methods: Patients (age range, 15–65 y) with inadequately controlled partial seizures, with or without secondary generalization, and on 1 to 3 concomitant antiepileptic drugs (AEDs) were enrolled in this double‐blind, randomized, placebo‐controlled, parallel‐group study. Patients with ≥9 seizures on stable AED dosages during baseline (3 months) were eligible for a 3‐month double‐blind treatment phase (DB). These patients were randomized equally to 1 of 5 treatment groups: rufinamide 200 mg/d, 400 mg/d, 800 mg/d, and 1600 mg/d, or placebo. Primary efficacy was seizure frequency per 28 days in DB; secondary efficacy included seizure frequency ratio and response to treatment (≥25% reduction in seizure frequency per 28 days vs baseline). Safety and tolerability were evaluated by adverse events (AEs), vital signs, and laboratory tests. Trough levels of AEDs were obtained at baseline; plasma sampling for AEDs and rufinamide levels was done before randomization and at scheduled visits. Results: Of the 737 patients enrolled in the study, 647 were randomized. A statistically significant linear trend of dose response for seizure frequency per 28 days in DB (p = 0.003) and for treatment responders (p = 0.0035) was observed in favor of rufinamide. Median seizure frequency ratio was significantly reduced in patients treated with 400 mg/d (11%, p = 0.0274), 800 mg/d (16%, p = 0.0123), and 1600 mg/d (17%, p = 0.0163) of rufinamide compared to placebo. AEs were generally similar to placebo for all rufinamide doses ≤800 mg/d. At 1600 mg/d, dizziness, somnolence, diplopia, and nystagmus were reported ≥2 times as frequently as placebo and the lower doses of rufinamide. Serious AEs occurred in 21 rufinamide‐treated (4.1%) and 5 placebo (3.8%) patients. Plasma rufinamide levels increased dose‐proportionally with doses ≤800 mg/d, but a slight reduction was observed with doses >800 mg/d. Concomitant use of phenytoin, primidone, or phenobarbital resulted in an approximate 25% increase in plasma clearance (CL) of rufinamide; patients using valproate experienced an approximate 22% decrease in rufinamide plasma CL. Rufinamide did not significantly alter plasma levels of other concomitant AEDs. Conclusions: Efficacy of rufinamide was demonstrated in a dose‐dependent manner and rufinamide dosages up to 1600 mg/d were generally well tolerated. Plasma levels of rufinamide increased in a dose‐proportional manner up to 800 mg/d. Rufinamide concentrations were slightly altered by certain concomitantly administered AEDs, however a change in rufinamide dosage may not be required. (Supported by Eisai Inc.) 1AnnHempel,1Rosette A.Jabbour,1Gail L.Risse,1,2John R.Gates,1,2Michael D.Frost, and1,2Frank J.Ritter(1Epilepsy, Minnesota Epilepsy Group, P.A. of United Hospital and Children's Hospitals and Clinics ‐ St. Paul, St. Paul, MN; and2 Department of Neurology, University of Minnesota, Minneapolis, MN ) Rationale: It is well known that focal resection involving the language dominant hemisphere can result in mild decrements in language functions, such as confrontation naming, even when cortical language areas are mapped during electrical stimulation studies prior to surgery. However, it is unclear to what extent postoperative language change occurs following resection of frontal or temporal cortex in the “nondominant” or minor language hemisphere when patients have a bilateral capacity for language. Methods: Three males and 1 female (ages 10–34) underwent the intracarotid amobarbital procedure (IAP) and cortical language mapping prior to focal resection for intractable seizures. In each case, surgery involved frontal and/or temporal cortex in the minor language hemisphere. All patients were classified with bilateral language, three of whom were primarily left dominant with one being primarily right dominant. Cortical language areas were identified during stimulation studies in the frontal region in all patients and in the temporal region in three patients. In no case did the surgery include resection of mapped language cortex. Postoperative interval ranged from four to seven months. Pre‐ and postoperative cognitive test performance was compared on tests of verbal fluency and confrontation naming. Results: Two patients displayed declines in verbal fluency and confrontation naming post operatively. For one of these, language decline was part of a more generalized cognitive decline possibly attributable to a medication change. In the second patient, initial postoperative deficits had recovered to baseline levels when the patient was re‐evaluated at 1 year, 3 months postoperatively. Stable verbal fluency and confrontation naming were observed in the other two patients. Conclusions: These limited data do not support the likelihood of persistent language impairment following “nondominant” frontal or temporal lobe surgery in patients with bilateral language who underwent language mapping. Nonetheless, these preliminary findings are insufficient to suggest that language cortex in the minor language hemisphere can be routinely resected in this population. A single individual in a sample of four patients displayed decline on formal tests during the early postoperative interval, which later resolved. This suggests the possibility that postoperative language decrements, when they do occur, may be transient in these cases. 1M.Jones‐Gotman,1V.Sziklas,1J.Djordjevic,1F.Dubeau, and1J.Gotman(1 Neurology and Neurosurgery, McGill University, Montreal, QC, Canada ) Rationale: We previously reported our experience using etomidate in place of sodium amobarbital for intracarotid speech and memory procedures 1 . Using our new etomidate speech and memory test ( e SAM), we have solved the problem of frequent shortages of amobarbital, and have also introduced an important change from the traditional intracarotid amobarbital procedure (IAP) in that the anesthetic effect is maintained by infusion after the initial injection. This impacts especially on the memory application of these procedures, because the short duration of the hemianesthesia in IAP often necessitated cautious conclusions about memory when some items were introduced after the drug was no longer active. We report a description of our procedure and a comparison of memory‐test outcomes in e SAM vs. those in a retrospective sample of the last 40 patients to undergo IAP in our center before ambobarbital became unavailable in 2003. Methods: Our behavioral procedure for e SAM and IAP is the same except for the issues of urgency and timing. We developed it (in 5 versions) based on a 4‐year retrospective analysis of our data (259 injections), and tested it in 25 patients without injection to provide “norms.” Speech tests comprise naming, comprehension, serial speech, spelling, reading and repetition (words, sentences), presented in rotation at two items per task to ensure sampling of all tasks. Memory tests use 24 real objects: 8 before injection, 8 “critical” items shown beginning 60 sec after injection, and 8 new “foil” items added to the other 16 for recognition testing. Timing and interpretations from the 49 e SAM tests (26 patients) performed to date are compared with the last 75 IAP tests (40 patients) done at our site. Results: In IAP, “critical” memory items were shown on average between 90 and 202 sec after injection in dominant‐, and between 83 and 160 sec in nondominant‐hemisphere tests: on average 3.3 of 7 speech tasks were sampled before memory item presentation began. Rapid return of function according to EEG resulted in 30% of dominant and 25% of nondominant‐hemisphere memory tests being interpreted cautiously; speech results were unambiguous. In e SAM we no longer aim for presentation of memory items to begin at 60 sec, but instead present them after all speech tasks have been sampled. Thus, in e SAM, “critical” memory items were shown between 201 and 326 sec after injection of the bolus in dominant‐, and between 201 and 297 sec in nondominant‐hemisphere tests. Function did not return before speech and memory had been tested adequately, and unambiguous results were obtained for both. Conclusions: For IAP, our behavioral procedure produced 100% interpretable speech results and 70 to 75% unambiguous memory results. The same procedure using e SAM presents a further advantage by allowing all tasks to be sampled during full anesthesia and without rushing. To date, all e SAM tests have yielded clearly interpretable results. REFERENCE 1.  Jones‐Gotman et al Epilepsia ( 2005 ); 45 : 345 .1Lorna V.Myers, and1MichaelCohen(1 Neuropsychology, Northeast Regional Epilepsy Group, White Plains, NY ) Rationale: Demographic changes in the US have led to increasing referrals of Hispanic American patients to neuropsychologists. An important factor associated to the assessment of Hispanic‐Americans include issues of language proficiency . According to Centeno and Obler (2001), “the major concern in a clinical setting is to evaluate a bilingual speaker's language dominance or balance in order to determine the language to be used during the diagnostic process... A second concern is to determine the extent to which lack of proficiency can falsely mimic or exaggerate neuropsychological deficits.” A relatively common current testing practice, involves administering psychological tests in English to patients for whom this is not a dominant language. A subgroup that is especially at risk of being inadequately assessed is that of patients who appear to “speak well enough to be tested in English.” Other current testing practices include assessments through non‐verbal measures and with the use of interpreters. Most of these practices are outside the boundaries of accepted American Psychological Association ethical standards because they can easily result in inaccurate measurements. However, in the last nine months, three cases were evaluated in our center that had recently been tested at other centers using the above mentioned techniques. We present the neuropsychological profiles of these cases when tested in their non‐dominant English or through an interpreter and when tested in their dominant language by a bilingual neuropsychologist. Methods: Three presurgical epilepsy patients referred to our service were tested in their dominant language, Spanish, utilizing instruments that were standardized on a Spanish‐speaking, Hispanic American population (Neuropsychological Screening battery for Hispanics and Woodcock‐Muñoz Batería Cognitiva). A comparison between the results (intellectual functioning, language, verbal and visual memory, and visual spatial skills) obtained from this assessment and from a previous one conducted in English using interpreters and/or predominantly non‐verbal measures one year earlier was performed. Results: The major differences noted between the evaluations were that when performed in English many language and language‐based tests could not be administered leading to missing data. Moreover, the use of visual spatial (non‐verbal) tests that were administered in English and through an interpreter and later interpreted using mainstream English (American) norms also led to incongruent results. Results in English were inconsistent with imaging studies and electroencephalographic results. Conclusions: Neuropsychological testing in the patient's non‐dominant language produces equivocal findings that may be partial and incomplete. Accurate assessments will include testing in the patient's dominant language (which will be determined through language proficiency assessment) using tests that have been standardized for use with Spanish, Hispanic American populations. 1V.Sziklas,1L.Francis,2M.Iordanova‐Maximov, and1M.Jones‐Gotman(1Neurology and Neurosurgery, McGill University, Montreal, QC, Canada; and2 Psychology, Concordia University, Montreal, QC, Canada ) Rationale: Traditional stereognosis tests are useful for demonstrating deficits associated with certain large lesions but not with detecting smaller, subtle ones. We devised a haptic search task similar to stereognosis tests but designed to be more sensitive. It requires tactile discrimination among objects that are identical except for size. As this skill should depend on the integrity of parietal cortex, our test is expected to help identify dysfunction in that brain region. Methods: 137 healthy volunteers (ages 18–92 years) and 18 patients with focal epilepsy completed sensory and motor control tasks (manual dexterity and two‐point discrimination), followed by two experimental tasks using wooden balls that increased or decreased in increments of 1/32 inches from the diameter of a reference ball (1¼ inches). The first task determined size‐discrimination thresholds. Subjects were then tested at their threshold on the second task, in which they located one target (odd ball) among other balls, all identical in size. Both tasks had unimanual and bimanual conditions. Reaction times and errors were measured. Results: A hand‐efficiency index for the bimanual search task produced a normal curve, with most healthy subjects showing no hand advantage, whereas the right hand of right‐handed subjects showed a moderate advantage on the unimanual task. A significant increase in reaction times with age was observed on the discrimination tasks but not on the search tasks. Only patients without primary sensory deficits were tested. Patients with focal lesions in parietal cortex showed increased reaction times on the contralateral hand for both discrimination and search. By contrast, patients with frontal‐lobe abnormalities were slower with both hands and normal performance was observed in patients with temporal‐lobe abnormality. Conclusions: Results from a large group of healthy individuals provide normative data for this task, which is meant for use with patients. Age‐related decline in healthy subjects warns that age must be taken into account when interpreting a patient's performance. Our initial results with patients suggest that the task is sensitive to lateralized damage in parietal regions, but performance is also affected by frontal‐lobe lesions. (Supported by G.W. Stairs Memorial Fund.) 1BrigidWaldron,1DavidGlosser,1MichaelSperling,1MaromiNei,1AndroZangaladze,1AshwiniSharan, and1JosephTracy(1 Thomas Jefferson University/Jefferson Medical College, Jefferson Comprehensive Epilepsy Center, Philadelphia, PA ) Rationale: Speech and language are established early in life, typically in the left hemisphere. Individuals who have suffered early injury to the brain from disorders such as epilepsy are known to have higher rates of atypical lateralization possibly through cognitive reorganization. An important unknown is the degree to which language functions reorganize en masse or whether shifts in laterality can occur independently for the separate language skills. The present study tested for potential independence among language functions by determining the laterality of five distinct language skills (auditory comprehension, speech, sentence repetition, visual confrontation naming, and single word reading) in a sample of temporal epilepsy patients. Methods: A total of 63 subjects (30 males, 33 females) underwent the IAP proceedure. Language ability on each of the skills was rated on a three‐point scale (faultless, attempt with error, and failure). The resulting scores were utilized to classify performance on each skill as left dominant, right dominant, bilateral representation or failed performance. We only identified dominance or bilateral representation through the demonstration of competency. Analysis of variance on this laterality index score was conducted utilizing side of seizure focus and age of onset group as between‐subject factors. Chi square analyses determined the association between laterality pattern and seizure focus/age of onset group. Results: A total of 38 patients (60% or 42 of 71) showed atypical language representation (bilateral representation or right hemispheric dominance) on at least one measure. Interestingly, both patients who showed atypical representation in all five language components together were left handed, indicating that left‐handedness may be related to complete atypical dominance but not partial atypical dominance. Conclusions: When language reorganization occurs in an epilepsy population it is most likely to occur in 3 domains or more. Though not significant, the early left hemisphere lesion group showed a higher contribution to atypical lateralization from what could be thought of as “truer” atypicality (bilateral and right dominance) as opposed to the late right group which shows a higher contribution from the left hemisphere but some contribution from the right. The atypical representation shown may reflect reorganization, but congenital atypical lateralization is possible. This study met the aim of capturing even subtle contributions of the non‐dominant hemisphere to language and was successful in assessing the patterns of atypical language dominance in an epilepsy population.