Surgery: All Ages

Abstract

1 Joshua R. Anderson, 1 Nazih Moufarrij, 1 John Gorecki, and 1 Kore Liow ( 1 Comprehensive Epilepsy Center, Via Christi Regional Medical Center, Wichita, KS ) Rationale: Anterior Medial Temoral Lobe Resection (AMTR) is a procedure widely performed for refractory mesial temporal lobe epilepsy syndrome at tertiary care university‐based medical centers. However, increasingly, more tertiary non‐university‐based community hospitals or university affiliated community hospitals are performing these procedures. There is little available outcome data for AMTR performed at community hospitals. Methods: Retrospective chart review performed on 33 patients who underwent AMTR at the 945 bed tertiary care Via Christi Regional Medical Center in Wichita, Kansas from 2002–2003. All patients were worked up at the Comprehensive Epilepsy Center including Video Epilepsy Monitoring, Specialized Epilepsy Magnetic Resonance Imaging Protocol, FD Glucose Positron Emission Tomography Scans, Intracarotid Amobarbitol Testings, Intracranial Recording by fellowship‐trained epileptologist. Neurosurgeon was performing AMTR at academic institution prior to praticing at community hospital. Results: The patient population of 33 people included 16 males and 17 females averaging 32 years of age. All had complex or simple partial seizures. Patients had waited for an average of 20 years before being referred to epilepsy center for surgical evaluation. Patients had failed a average of 3.7 seizure medications, and they were on 2 seizure medications at the time of surgery, and had avarage of 17.9 seizures/month prior to surgery. 21 patients had Magnetic Resonance Imaging (MRI) abnormalities with 11 mesial temporal sclerosis noted. 17 patients had right sided resection while 16 right sided resections. Post‐operatively, 31 of 33 (93%) patients had Engel Class I surgical outcome at average follow up duration of 6 months. 30 of 33 (90%) patients are gainfully employed or attending school/college as of last clinic follow up. 27 of the 31 (87%) adult patients who were not able to drive before surgery are now driving. Surgical complications included 1 chronic aphasia, 1 venous infarct in the posterior cerebral artery distribution and 1 scalp infection. Conclusions: AMTR epilepsy surgery performed at tertiary care community hospital in Kansas by fellowship‐trained epileptologist and neurosurgeon shows comparable favorable outcome. (Supported by Via Christi Comprehensive Epilepsy Center, Wichita, Kansas.) 1,2 Cristine M. Baldauf, 1,2 Arthur Cukiert, 1,2 Jose A. Burattini, 1,2 Pedro P. Mariani, 1 Lauro Seda, 1 Rodio Brandao, 1,2 Meire Argentoni‐Baldochi, 1,2 Carla Baise‐Zung, 1,2 Cassio R. Forster, and 1,2 Valeria A. Mello ( 1 Neurology and Neurosurgery, Hospital Brigadeiro, Sao Paulo, Sao Paulo, Brazil ; and 2 Neurology and Neurosurgery, Clinica de Epilepsia de Sao Paulo, Sao Paulo, Sao Paulo, Brazil ) Rationale: The use of valproic acid has been associated to bleeding in some centers and reports. On the other hand, there is no consensus on if valproate should be withdrawn before major surgical procedures, such as neurosurgical operations or not. This paper reviews our experience with refractory epileptic patients who underwent neurosurgical procedures and who were under high dosage of valproate which has not been withdrawn before surgery. Methods: One hundred consecutive refractory epileptic patients (children and adults) who have been receiving at least 1500mg of valproate daily (mean = 2250mg/day) and who have been submitted to epilepsy surgery have been retrospectively reviewed. All patients have been submitted to standard preoperative coagulation tests including platelet function and have no formal abnormal findings (patients with clear preoperative coagulopathies have been excluded from this analysis). Results: Six patients presented with extensive, clearly abnormal, bleeding during surgery; the mean transfusion volume was 3,0 liters. Only 1 patient presented with clinically significant postoperative hemathoma (epidural) that had to be evacuated. These 6 patients had more transfusions and longer hospital permanence, but no difference regarding long‐term outcome. All these patients have been receiving at least 2500mg of valproate daily (mean 3000mg/day) and had platelet counts at the lower normal range. The other patients did not differ from patients not receiving valproate regarding intraoperative bleeding, hospital stay, transfusion or outcome. Conclusions: Although not statistically significant, patients receiving 2500mg/day or more of valproic acid and low (although within normal range) platelet counts might be in risk for intraoperative bleeding during major surgical procedures. Considering the potential risk for major neurological complications, we recommend that valproate should be withdrawn in this selected patient population. 1 Jorge A. Gonzalez‐Martinez, 2 Teeradej Srikijvilaikul, 3 Imad M. Najm, 4 Christopher Kellinghaus, and 1 William E. Bingaman ( 1 Neurosurgery, The Cleveland Clinic Foundation, Cleveland, OH ; 2 Neurosurgery, King Chulalongkorn Memorial Hospital, Bangkok, Thailand ; 3 Neurology, The Cleveland Clinic Foundation, Cleveland, OH ; and 4 Neurology, University of Munster, Munster, Germany ) Rationale: Outcome after epilepsy surgery with histopathology correlations indicated that different pathological findings of focal cortical dysplasia (FCD) had different outcome. FCD and balloon cells demonstrated best outcome with 75% seizure‐free at least 1‐year follow‐up. Our objective is to correlate seizure outcome after surgery for FCD with pathological findings and variables associated with seizures outcome. Methods: We retrospectively reviewed patients who underwent surgery for drug‐resistant partial epilepsy due to FCD and confirmed by histopathologic analysis between 1990–2002 with a minimum of 1‐year follow‐up. Histological findings were classified as: type Ia, architectural abnormalities, type Ib, architectural abnormalities plus giant or immature neurons, type IIa, architectural abnormalities plus dysmorphic neurons, and type IIb, architectural abnormalities plus dysmorphic neurons plus balloon cells. Seizure outcome was classified according to Engel's classification. Results: Sixty‐two patients were identified, 53% of patients were seizure‐free (Engel Ia) and 79% had favorable outcome (Engel I &II) after surgery. FCD type Ia was the most common histologically identified (64.5%) and temporal lobe was the most common location of type Ia (67.5%). FCD type IIb was commonly found in frontal and central lobe (83.3%). Magnetic resonance imaging study (MRI) identified FCD in 52.5% type Ia and 91.6% in type IIb. Temporal lobe FCD had a more favorable outcome than extratemporal FCD and the presence of balloon cells (type IIb) predicts good seizure outcome (91.6% of the patients with type IIb FCD were seizure free versus 57.5% for patients with type Ia cortical dysplasia). Conclusions: Surgical treatment of FCD can achieve seizure‐free outcome in approximately half of the patients. Temporal lobe location and FCD type IIb predict good outcome. 1,2 Jose A. Burattini, 1,2 Arthur Cukiert, 1,2 Pedro P. Mariani, 1 Lauro Seda, 1 Rodio Brandao, 1,2 Cristine M. Baldauf, 1,2 Meire Argentoni‐Baldochi, 1,2 Carla Baise‐Zung, 1,2 Cassio R. Forster, and 1 Valeria A. Mello ( 1 Neurology and Neurosurgery, Hospital Brigadeiro, Sao Paulo, Sao Paulo, Brazil ; and 2 Neurology and Neurosurgery, Clinica de Epilepsia de Sao Paulo, Sao Paulo, Sao Paulo, Brazil ) Rationale: Extratemporal resection is often larger then temporal lobe resection in patients with refractory epilepsy, especially in patients with normal MRI findings. Cortical resection margins could be restricted by the presence of eloquent areas within the planned resection or due to individual vascular, especially venous, anatomy. This paper discusses our experience while performing extratemporal resections in patients with normal MRI regarding the intraoperative factors limiting cortical resection. Methods: Sixty‐one patients with extratemporal refractory epilepsy and normal MRI were studied (both children and adults). All patients have been submitted to invasive neurophysiological investigation by means of chronically implanted subdural electrodes. Planned cortical resection necessarily included the ictal‐onset region and the region from which habitual seizures were elicited by stimulation and, whenever possible, the interictal zone. Results: In 11 patients cortical resection has been limited by the presence of eloquent areas within the planned procedure; in 28 patients the cortical resection has been limited by the patient's venous drainage pattern and in 6, it has been limited by both. There was no mortality and no unexpected neurological morbidity. Conclusions: Despite the fact the both the presence of eloquent areas and vascular anatomy could limit the actual cortical resection in extratemporal epileptic patients, the venous' drainage pattern appears to be the most important single factor. In 64% of these patients, the cortical resection suggested by invasive neurophysiological findings would be limited by one of these factors, or both. 1 Mar Carreño, 1 Antonio Donaire, 2 Inés Barceló, 1 Nuria Bargalló, and 1 Jordi Rumiá ( 1 Neurology, Hospital Clínic, Barcelona, Barcelona, Spain ; and 2 Neurology, Hospital de Son Dureta, Palma de Mallorca, Mallorca, Spain ) Rationale: Linear scleroderma in coup de sabre (LSCS) presents as band‐like sclerotic skin lesions, usually involving the frontoparietal area of the head. Progression of the disease may result in hemifacial syndrome (identical to idiopathic Parry Romberg syndrome ‐PRS‐). LSCS and PRS may be accompanied focal epilepsy, which is often refractory to medical treatment. Methods: We report two patients, one with LSCs and other with idiopathic PRS who presented with a clinical and pathological picture identical to Rasmussen encephalitis (RE). Results: Patient #1 started to have seizures at the age of 3, consisting of clonic twitching of the right arm and bilateral asymmetric tonic seizures. Seizures were pharmacorresistant. Progressive hemiparesis and cognitive decline were noticed by the age of 5, in addition to slight atrophy of the left side of the face, which progressed over the next years. MRIs showed marked and progressive atrophy of the left hemisphere. Presurgical evaluation at the age of 10 showed seizures arising from the left frontocentral region. She underwent a left functional hemispherectomy. Pathology showed massive neuronal cortical loss, reactive astrocitosis, perivascular lymphocytic infiltrates and microglial nodules, compatible with RE. She has had a 75% seizure reduction.Patient #2 started to have seizures at the age of 23. Seizures consisted of paresthesias in the left thigh extending to the left arm and left hemiface, sometimes progressing into clonic movements of the left arm. At the age of 18 he had been diagnosed with LSCS (focal lesion over the right frontoparietal area), and left hemidystonia, succesfully treated with botulinum toxin. Seizures were initially controlled with carbamazepine. Seizure frequency suddenly increased at the age of 30 and he developed epilepsia partialis continua involving the left arm. Fluctuating hemiparesis was noticed, first after clusters of seizures and later in between seizures. MRIs showed progressive atrophy of the right hemisphere involving also the right basal ganglia. He underwent presurgical evaluation at the age of 33. Seizures were seen to arise from the right parietal region. He underwent invasive video‐EEG monitoring with subdural electrodes over the right parietal region. He had a focal resection of the right parietal area. He has been seizure free for three months. Pathology showed focal perivascular lymphocytic infiltrates, microglial nodules, focal neuronal loss, and reactive astrocytosis, all suggestive of RE. Conclusions: LSCs and PRS seem to be part of the same spectrum of diseases characterized by progressive facial atrophy and neurological symptoms. Refractory focal epilepsy is common and surgical treament may be indicated. Pathological findings may be identical to chronic encephalitis (RE), suggesting a common autoimmune pathogenesis. 1 Dongwoo John Chang, and 1 Christi DeLemos ( 1 Neurological Surgery, University of California Davis Medical Center, Sacramento, CA ) Rationale: Decades of successful empirical experience with epilepsy surgery have demonstrated the safety and efficacy of surgical intervention on selected seizure syndromes. Recently, the beneficial effect of anterior temporal lobe resection has been reaffirmed by a prospective randomized clinical trial for patients with mesial temporal sclerosis (New England Journal of Medicine 2001). Despite these findings, only a small minority of potential surgical candidates undergo epilepsy surgery in the United States. This study attempts to determine the attitudes and practice patterns of general neurologists vis‐a‐vis epileptologists regarding epilepsy treatment in the State of California and in the State of New York. Methods: Confidential e‐mail questionnaires were sent to all neurologists in the State of California and in the State of New York. Two groups were queried as comparison cohorts: epileptologists and all other neurologists, including those who have other declared subspecialty interests in neurology. The survey covers all facets of epilepsy management, including questions regarding the definition of medical intractability, attitudes and practice patterns regarding surgical treatment of seizure disorders, the necessity and feasibility of further randomized clinical trials for other surgical procedures for epilepsy, and the perceived relative value of specialized epilepsy treatment centers. Results: A complete analysis of the findings of this study will be presented at the December 2005 meeting of the American Epilepsy Society. The final results of this survey are unavailable at the time of the writing of this abstract. Conclusions: Published data demonstrate the benefit of epilepsy surgery in selected patient subgroups. However, because surgery is perceived to be invasive, it is important to determine the burden of proof that needs to be satisfied in order to allow selected surgical procedures to become primary therapy in epilepsy management as opposed to remaining as options of last resort. 1 Peter Chin, 2 Anne Berg, 3 Susan Spencer, 4 Michael Sperling, 5 Shlomo Shinnar, 6 John Langfitt, 7 Carl Bazil, 8 Thaddeus Walczak, 9 Steven Pacia, 1 Martin Lee, and 1 Barbara Vickrey ( 1 Neuro, UCLA ; 2 BIOS, NIU ; 3 Neuro, Yale ; 4 Neuro, Thomas Jefferson ; 5 Neuro, Montefiore Med. Ctr. ; 6 Neuro, U. of Rochester ; 7 Neuro, Columbia ; 8 MINCEP ; and 9 Neuro, NYU ) Rationale: Patients often suffer detrimental effects of longstanding epilepsy with respect to social adjustment, education and employment before undergoing resective surgery as a last resort. Post‐operative employment gains have been shown to enhance the impact of epilepsy surgery, so studying employment patterns may inform interventions aimed at reintegrating patients into society. Methods: At 7 US centers, 396 patients with focal epilepsy enrolled in a cohort study and underwent resective surgery. Eligible patients had refractory epilepsy for at least 2 years, failed 2 or more AED's, and had at least 20 seizures in the 24 months prior to enrollment. Data were collected from interviews and medical records at baseline and at 3, 12, and 24 months following surgery. Descriptive data on employment pre‐ and post‐surgery were examined. Associations between employment patterns and disability status, race (white vs. non‐white) and age < 35 were assessed using the Wilcoxon rank sum test. Results: The cohort was 48% male; average age was 37 years. Mean duration of intractable epilepsy was 13 years. Prior to surgery 45% of the cohort was regularly employed, including 148 (38%) employed full‐time and 27 (7%) part‐time. Other categories at baseline included 102 (26%) disabled and unemployed; 45 (11%) unemployed; 27 (7%) in school; 22 (6%) homemakers; 12 (3%) irregular or seasonal work; 7 (2%) retired; 3 (1%) volunteer work only; and 1 other. Two years after surgery 52% of the cohort was employed, representing a net employment gain of 7%. Twenty‐two percent of those who were disabled and unemployed before surgery obtained regular work, compared with 23% of those who were unemployed but not disabled (p = .93). Whites were more likely than non‐whites to be employed before surgery (49% vs 25%, p = .002) and 24 months after surgery (56% vs. 37%, p = .007) but were not significantly more likely to change employment status at 24 months (p = .35). Age under 35 did not influence work status before surgery (p = .55) and 24 months after surgery (p = .77) or change in employment at 24 months (p = .07). Conclusions: Improved employment status is seen in all categories following epilepsy surgery except the retired (0 of 7). However, overall employment gains remain low after 2 years because 12% of the cohort found work but 5% concurrently lost their jobs. Vocational rehabilitation efforts should be enhanced and should include those who were working prior to surgery but unemployed at follow up. The potential for post‐operative employment among those who were formerly disabled from epilepsy does not appear to differ from that of the non‐disabled unemployed. Race and age below 35 did not affect post‐operative change in employment. (Supported by NIH NINDS RO1 32375–06 and the RWJ Foundation.) 1 Walter J. Hader, 1 Neelan Pillay, 1 S. Terence Myles, 1 Lisa Partlo, and 1 Samuel Wiebe ( 1 Department of Clinical Neurosciences, University of Calgary, Calgary, AB, Canada; Department of Clinical Neurosciences, University of Calgary, Calgary, AB, Canada; Department of Clinical Neurosciences, University of Calgary, Calgary, AB, Canada; Department of Clinical Neuropsychology, University of Calgary, Calgary, AB, Canada; and Department of Clinical Neurosciences, Universtiy of Calgary, Calgary, AB, Canada ) Rationale: Selective resection of mesial temporal structures in patients with intractable temporal lobe epilepsy has been advocated in an attempt to minimize temporal neocortical removal associated with standard temporal resections. The benefits of such a strategy have not been not well documented. Methods: A retrospective review of all patients who underwent either selective amygdalo‐hippocampectomy (SAH) or a standard cortico‐amygdalo‐hippocampectomy (CAH) for their intractable temporal lobe epilepsy at the University of Calgary was completed. Patients had either a transcortical SAH or CAH, extending to 4.5 cm on the dominant middle temporal gyrus and 6 cm on the non‐dominant temporal lobe Results: A total of 72 patients were identified. Forty‐eight (67%)had SAH and 24(33%)CAH. No difference in age at onset of epilepsy, side of surgery, seizure types, duration of epilepsy or age at surgery was identified between the two groups Patients undergoing SAH had preoperative evidence of MTS on MRI (81 vs 33%) while 67% of MRIs were normal in patients undergoing CAH. Forty‐five (63%) of 72 patients were free of disabling seizures at last follow up. Clinical and formal visual field deficits were more common in CAH than SAH. The duration of stay in hospital was shorter in patients with SAH. Detailed pre and post operative neuropsychological testing was available in 36 patients and revealed no significant differences in those undergoing SAH or CAH Conclusions: Patients treated with SAH for intractable temporal lobe epilepsy demonstrate greater evidence of mesial temporal lobe pathology, suffer fewer visual complications and require shorter duration of stay in hospital than patients undergoing more standard CAH. Neuropsychological consequences of temporal lobe surgery in this study were similar in those patients treated with both SAH and CAH. 1,2 Leigh R. Hochberg, 1 Selim Suner, 1 Matthew R. Fellows, 3 Gerhard M. Friehs, and 1,4 John P. Donoghue ( 1 Neuroscience, Brown Univ., Providence, RI ; 2 Neurology, Mass. Gen. Hosp./Brigham & Women's/Spaulding Rehab., Harvard Medical School, Boston, MA ; 3 Clinical Neurosciences, Brown Univ., Providence, RI ; and 4 Cyberkinetics Neurotechnology Systems, Inc., Foxboro, MA ) Rationale: Cortical physiology has been studied in animals for more than 40 years using single electrodes to record action potentials under different stimulus or behavioral conditions. Microwires or microelectrode arrays have now made it possible to record the activity of dozens to hundreds of individual neurons and local field potentials (LFPs) simultaneously. An array of 96 silicon/platinum microelectrodes has been used to record chronically (months to years) from macaque primary motor cortex (M1). This technology, BrainGate, forms the basis for a human neuromotor prosthesis for persons with tetraparesis, now in pilot clinical trial. If the array also provides useful neural signals acutely – minutes to hours after placement in human cortex – it could prove valuable for applications in seizure detection, intraoperative monitoring, and neurocritical care. Methods: 4x4 mm arrays of 96 microelectrodes (Cyberkinetics, Inc.) were implanted into macaque M1 or visual cortex (V1, data courtesy of D.L. Ringach). In the M1 experiments, recordings were first attempted in the awake animal the following day. In V1, recordings were obtained beginning 10–20 minutes after array insertion while the animal was sedated with propofol/sufentanil. Single units, multi‐unit waveforms, and LFPs were collected while the animal was sitting in a primate chair (M1) or while retinotopic mapping was performed (V1). Neural signals were transferred to a connector; amplified signals from each electrode were manually separated into data streams and stored digitally. Results: In both animals receiving a M1 array, action potentials, multi‐unit waveforms, and LFPs were recorded simultaneously on post‐operative day 1. 62 and 29 discriminated units with “fair” (signal:noise ratio approx. 2 to 4) or better waveforms were recorded in each animal, respectively. LFPs were also observed from an additional 32 and 31 electrodes, respectively. In the 10 V1 animals, LFPs were seen consistently on all electrodes as soon as recording was initiated. Multi‐unit waveforms were seen on most electrodes, and a range of 15–70 single units was isolated in each experiment. Conclusions: Microelectrode arrays can record from cortical neuronal ensembles as soon as 10–20 minutes after insertion. It thus may be possible to record human cortical neuronal activity in a variety of acute and subacute settings. Potential studies include high spatial resolution recording of epileptigenic tissue in patients undergoing subdural grid placement (including further investigation of interictal spike physiology and seizure prediction paradigms) and neuronal/LFP activity patterns in tissue at risk for ischemia secondary to subarachnoid hemorrhage‐related vasospasm. (Supported by NINDS, VA, DARPA, ONR, Cyberkinetics.) 1 Seo‐Young Lee, 1 Ji Hoon Phi, and 1 Chun Kee Chung ( 1 Neurosurgery, Seoul National University Hospital, Seoul, Seoul, Korea ) Rationale: Proper time, extent, and prognosis of Epilepsy surgery for low grade tumor are not established. This study is aimed at investigation of the surgical outcome and the prognostic factors of epilepsy patients with a brain tumor. Methods: We acquired the data from the epilepsy surgery cohort since May 1995. Inclusion criteria were as follows: patients who had 1) a hemispheric intra‐axial neuroepithelial tumor, 2) seizure as an only presenting feature, 3) no neurological deficit at presentation, 4) at least 1 year of follow‐up after surgery. We excluded highly malignant tumors (world health organization grade IV). We evaluated the seizure outcome according to Engel's classification. The degree of resection was assessed on postoperative magnetic resonance imaging (MRI). Results: 75 patients fulfilled the criteria. Median follow‐up was 3.3 years (range 1.0–9.3 years). Ganglioglioma (25 cases) was the most common histopathological entity and dysembryoplastic neuroepithelial tumor (18 cases), oligodendroglioma (13 cases) and diffuse astrocytoma (6 cases) followed. Eight cases of world health organization (WHO) grade III tumors were included.One year after surgery, 77.3% of the patients were seizure‐free (class I). Overall seizure‐free rate at last follow‐up rose up to 85.7% and 72% of those patients were free of anticonvulsants. Favorable seizure outcome was associated with the degree of resection, pathological grade, tumor location that required awake anesthesia, intractability and EEG concordance in univariate analysis. Among theses factors, only degree of resection and pathological grade gained statistical significance in multivariate analysis (p = 0.023, 0.020 respectively).Concomitant hippocampal sclerosis (dual pathology) was found in 9 patients (12%). Resection of both the tumor and mesial structures yielded better outcome than resection of the tumor alone (p = 0.048). Conclusions: Degree of surgical resection and pathological grade of the tumor are the most important factors for seizure outcome in epileptic patients with a brain tumor. In the case of dual pathology, both the tumor and mesial structures should be resected for better seizure outcome. 1 George L. Morris, 1 Christopher M. Inglese, 2 Wade M. Mueller, and 3 Sara J. Swanson ( 1 Regional Epilepsy Center, St Luke's Medical Center, Milwaukee, WI ; 2 Department Neurosurgery, MCW, Milwaukee, WI ; and 3 Department of Neuropsychology, MCW, Milwaukee, WI ) Rationale: To determine the effect of normal MRI findings on seizure free outcomes following epilepsy surgery Methods: We reviewed the results of epilepsy surgeries performed in Milwaukee from 1991 to 2001. In 362 evaluations of epilepsy, we found 214 surgical resections. We found 176 cases with valid MRI results, surgery locations, and 6 month seizure‐free outcomes. We divided patients into seizure‐free versus non‐seizure‐free at their 6 month assessment. Surgery locations were defined as lobar or multi‐lobar. MRIs were collapsed into normal or abnormal resuilts. We performed an analsis using Pearson's Chi‐squared technique on the relationship between normal versus abnormal MRI results and the patient's seizure outcome, including analysis by surgery location. Results: Of the 176 surgeries performed, 136 were temporal and 22 were frontal. Seizure freedon coccurred in 103 of temporal resections (76%) and 8 of frontal resections (37%). Abnormal MRIs were found in 92 temporal resections (67%) and 15 frontal resections (68%). Seizure freedom was no different in patients with normal versus abnormal MRIs (p = 0.991). Seizure freedom rates for temporal resections were 70% in normal MRI versus 77% in abnormal MRI (p = 0.43). Seizure freedom rates for frontal resections in normal MRI were 42% versus 33% in abnormal MRIs. Conclusions: Normal MRI in the evaluation of localization‐related epilepsies did not influence seizure‐free rates in our surgical experience. Previous studies suggest the presence of MRI abnormalities improves surgical outcomes and we have previously reported abnormalities of prognostic significance. Patients presenting for epilepsy surgery should not be discouraged that normal MRI findings may influence their outcome. 1,2 Johan Pallud, 1 Ute Haussler, 1 Sophie Hamelin, 2 Bertrand Devaux, and 1 Antoine Depaulis ( 1 U704, Inserm‐UJF, Grenoble, France ; and 2 Neurosurgery, Sainte‐Anne Hospital, Paris, France ) Rationale: Surgical resection of mesial temporal structures is an effective treatment of medically intractable mesial temporal lobe epilepsy (MTLE). However, dominant hippocampus resection may ensue with postoperative verbal memory deficit that warrants surgical alternatives to resection attempts. Based on the principle of multiple subpial transections, we developed in the kainate (KA) model of MTLE in mice, a technique of intrahippocampal transection, to disrupt longitudinally projecting axonal circuits that may play a role in generation and propagation of seizures. Methods: C57BL/6 adult mice were stereotactically injected with KA (1 nmol in 50 nl) in the right dorsal hippocampus and equiped for hippocampal and cortical EEG recording. Two hippocampal electrodes were placed along the longitudinal axis of the hippocampus, 1.4 mm apart, with the anterior one located at the site of KA injection. Three weeks later, when spontaneous recurrent focal discharges occurred, an intrahippocampal transection was stereotactically performed in the transversal plan between the 2 hippocampal electrodes. Seizures were quantified on EEG recordings before transection and during the following 3 weeks. Animals with such transections were compared to sham operated animals and animals with electrolytic lesion of the hippocampus. Upon completion of experiments, histological controls were performed to ascertain the hippocampal sclerosis (neuronal loss, gliosis, mossy fiber sprouting and granule cell dispersion), the position of the electrodes and the extent of the transection. Results: After a complete hippocampal transection (n = 6), no seizures were recorded at the hippocampal electrode located posterior to the transection, whereas discharges persisted at the KA focus, with a reduction of their occurrence and an increase of their duration. After partial hippocampal transection (n = 6) discharges, without modification of their occurrence and duration, were recorded at both hippocampal electrodes, as well as in the sham group (n = 6). After hippocampal electrolytic lesion (n = 6), no discharges were recorded at both hippocampal electrodes. Conclusions: Complete intrahippocampal transversal transection stops diffusion of discharges within the hippocampus. Moreover, such transections also modify the generation of focal discharges, which are less frequent and of a longer duration as compared to controls. These results suggest a role of longitudinally projecting axonal circuits of the hippocampus in the generation and the diffusion of seizures in this model of MTLE. (Supported by Inserm, Fondation pour la Recherche Medicale, Societe Francaise de Neurochirurgie and Fondation Francaise pour la Recherche sur l'Epilepsie.) 1 Freedom F. Perkins, 1 Dave F. Clarke, 2 Amy Waller, 1 Jeremy D. Slater, 1 Amy L. McGregor, 3 Mark H. McManis, 3 Eduardo M. Castillo, 1 James W. Wheless, and 2 James E. Baumgartner ( 1 Department of Neurology/Texas Comprehensive Epilepsy Program, The University of Texas Health Science Center at Houston, Houston, TX ; 2 Department of Surgery ; and 3 Department of Neurosurgery ) Rationale: The corpus callosum is the principle pathway linking the cerebral corticies, anatomically and neurophysiologically. Corpus callosotomy (CC) is a palliative surgical procedure used in patients with generalized seizure types. It interrupts bilateral synchrony and is most effective in the treatment of both tonic and atonic seizures. With the introduction of the vagus nerve stimulator, CC has seemingly declined in many epilepsy surgical centers. At our center, though the number of CCs declined in the years 2000 and 2001, in recent years it has regained prominence. Methods: We retrospectively reviewed the number of CCs performed between 1995 and March 2005 by a single surgeon (JB). The number of surgeries done on a yearly basis, extent of the procedure (partial versus complete), repeat operations (completion of CC and focal resection after CC) and acute/subacute surgical complications were assessed. Results: The patient's ages ranged from 18 months to 23 years. Eighty‐nine patients had a total of ninety‐seven surgeries (8 in 1995; 10 in 1996; 14 in 1997; 12 in 1998; 14 in 1999; 1 in 2000; 0 in 2001; 8 in 2002; 15 in 2003; 12 in 2004; and 4 thus far in 2005). Fifty‐five patients had partial CCs, thirty patients had complete CCs, and information was unavailable in four. Seven patients with partial CCs returned for completion after their seizures were not adequately controlled; one of whom had to have two further surgeries. Four cases returned for focal resection after a primary focus was identified post CC. Rare complications included a bone flap infection in 2 children; transient diabetes insipidus in 7 patients; and small frontal lobe infarcts in 3 patients. The operating time from incision to complete closure decreased over time (mean duration of approximately 60 minutes). The duration of the hemi‐body neglect/disconnection syndrome seen post CC, has also declined over time. Information on time to discharge was available in patients from 1999. The time to discharge decreased over time from a mean of 8.3 days in 1999 to 5.4 in 2004, suggesting a decrease in complication rate. Conclusions: Our institution has seen a resurgence in the number of CCs performed in recent years. It has been effective in rapidly decreasing or eliminating both tonic and atonic seizures. The overall morbidity has been remarkably low. With good patient selection, using an experienced surgeon, the CC should still be entertained as a palliative procedure in patients with tonic and atonic seizures, primarily in patients with frequent seizure inflicted injuries. 1 Vicenta Salanova, 1 Omkar Markand, 1 Robert Worth, and 1 Hema Patel ( 1 Neurology and Neurosurgery, Indiana University, Indianapolis, IN ) Rationale: Objectives: To determine whether, there are significant differences in the clinical history, evaluation, complication rate and outcome, between younger and older patients undergoing surgery for temporal lobe epilepsy Methods: Sixty‐four patients 40 years of age or older (group 1) with refractory temporal lobe epilepsy (TLE) underwent surgery following a comprehensive presurgical evaluation. We compared the history, clinical manifestations, surgical outcome, pathologic diagnosis, surgical outcome and the complication rate with 28 patients with TLE treated surgically at < 18 years of age (group 2, and noted any significant differences) Results: Group 1; 64 patients (40 yrs or >): The age of seizure onset ranged from 1 yr to 46 yrs, the age at surgery from 40 yrs to 59 yrs. 23% had a history of febrile seizures (FS), 81% had abnormal head MRIs, and 73% had abnormal FDG‐PET scans, 18% required invasive recordings. Thirty seven patients underwent right and 27 had left sided resections. There was no surgical mortality, one patient (1.5%) had a mild hemiparesis. 63% became seizure free and 23% had rare seizures (follow‐up 1yr to 18 yrs). Group 2; 28 patients (<18 yrs): The age of seizure onset ranged from 1 yr to 13 yrs, and the age at surgery from 8 yrs to 17 yrs. 54% had a history of FS, 62% had abnormal head MRIs, and 94% abnormal FDG‐PET scans, 10% required invasive recordings. 54% had mesial temporal sclerosis.There was no surgical mortality, or complications. 79% became seizure free, and 7% had rare seizures (follow‐up 1 yr to 18 yrs) Conclusions: While the older patients had a favorable outcome, the younger patients who had a higher incidence of febrile convulsions, and less demand for invasive studies, fared even better, and had no complications, emphasizing the need for earlier surgical treatment 1 Janel Schneider, and 1 Daniela Minecan ( 1 Department of Neurology, University of Michigan Medical Center, Ann Arbor, MI ) Rationale: Rasmussen's encephalitis (RE) is a relentlessly progressive disease with onset most commonly in childhood. Clinically, RE is characterized by the development of refractory partial seizures and progressive hemiparesis. The exact etiology of RE is not known, although recent scientific studies have implicated an underlying immune mechanism. These studies have led to trials of various immunomodulatory therapies, some of which have shown potential benefit, with improved seizure frequency and delayed progression of neurologic deficit. Yet, this benefit is variable and often short‐lived. The only proven curative therapy for RE is functional hemispherectomy. This radical surgery is typically performed after significant neurologic deficit has occurred. Affected children may have the potential for recovery of the motor and language functions of the involved hemisphere. However, a dilemma is faced in deciding on the appropriate treatment course for adolescent or adult onset cases of RE which often have a more variable course and slower progression. In these individuals, a temporizing treatment is desirable as the clinical consequences of functional hemispherectomy may hasten the development of permanent neurologic deficit. There are few reports on the outcome in older RE patients after other, less radical, surgical interventions. Methods: We report a case of late‐onset RE, in which the patient had become progressively disabled by her seizures, specifically the development of epilepsia partialis continua (EPC) and the adverse effects of high doses of multiple anti‐epileptic medications. Despite the severity of her seizures, neurologically, she had no evidence of permanent motor deficit. Results: Intracranial EEG monitoring identified the area involved in her focal motor seizures and focal resection of this area resulted in dramatic improvement of seizure frequency and allowed for the initiation of a medication taper. This resection resulted in minimal motor deficits of the corresponding side and overall the patient experienced a significant improvement in her quality of life. Conclusions: Based on this experience, it may be proposed that focal surgical resection should be considered in older RE patient's who are disabled primarily by their seizures and have relative preservation of neurologic function, thus avoiding the certain deficits from functional hemispherectomy. The duration of this benefit is unclear, although in our patient improvments have been sustained for a period of months. Long‐term follow‐up will be necesary to compare the benefits of focal surgical resection to other less radical therapies. 1 Teeradej Srikijvilaikul, 2 Tayard Deesudchit, 3 Jarturon Tantivatana, and 4 Chaichon Locharernkul ( 1 Neurosurgery, Chulalongkorn Comprehensive Epilepsy Program, King Chulalongkorn Memorial Hospital, Patumwan, Bangkok, Thailand ; 2 Pediatrics, Chulalongkorn Comprehensive Epilepsy Program, King Chulalongkorn Memorial Hospital, Patumwan, Bangkok, Thailand ; 3 Radiology, Chulalongkorn Comprehensive Epilepsy Program, King Chulalongkorn Memorial Hospital, Patumwan, Bangkok, Thailand ; and 4 Neurology, Chulalongkorn Comprehensive Epilepsy Program, King Chulalongkorn Memorial Hospital, Patumwan, Bangkok, Thailand ) Rationale: Intracarotid amobarbital procedure (WADA test) has been widely used to determine the hemisphere for speech dominance and memory function. Recently, there has been a worldwide shortage of the amobarbital despite a need for the test in some epilepsy surgery candidates. There were reports of effective use of propofol for speech and memory testing with safety. We report our experiences in intracarotid propofol test during epilepsy preoperative evaluation. Methods: Between October 2004‐March 2005, 4 patients, 1 male and 3 females, underwent intracarotid propofol test for epilepsy preoperative evaluation. The patients' ages were 13–23 years. Two patients had left hemispheric epilepsy with congenital right hemiparesis who were considered candidates for functional hemispherectomy. One patient who was ambidexterous had failed left temporal lobectomy and was planning for reoperation. The fourth patient had daily atonic seizures and bisynchronous EEG discharges from a right frontal lobe tumor.The intracarotid propofol test was performed to determine speech dominant hemisphere in the first three patients and to suppress epileptiform discharges in the fourth. The propofol preparation was 10 mg in 10 ml of normal saline solution. Results: The dose of propofol was between 15–18 mg for each injection. We were able to identify speech dominant hemisphere in all 3 patients. Two patients who underwent left functional hemispherectomy had no postoperative language deficit. Another patient showed left sided speech dominance after left sided injection. He then underwent subdural electrode recording. After cortical stimulation mapping, the speech areas were found on the same hemisphere shown by propofol test. In the patient with right frontal lobe tumor, bisynchronous spike‐and‐waves were completely suppressed after right sided injection, but unchanged after injection to the left. After tumor resection, she had fewer seizures then became seizure free on the fifth month. There were no complications during and after the tests in all subjects. Conclusions: Intracarotid propofol test can be used for epilepsy preoperative evaluation. The results of the test were confirmed by surgical outcome in our cases. 1 Kenou Van Rijckevorsel, 1 Cécile Grandin, 1 Marianne De Tourtchaninoff, 1 Geraldo Vaz, and 1 Christian Raftopoulos ( 1 Neurology, Université Catholique de Louvain, Brussels, Belgium; Neuroradiology, Université Catholique de Louvain, Brussels, Belgium; and Neurosurgery, Université Catholique de Louvain, Brussels, Belgium ) Rationale: Mesiotemporal epilepsy refractory to antiepileptic drugs can be treated by surgery. Temporal lobectomy and selective amygdalo‐hippocampectomy are the two main procedures. Both techniques give same seizure outcome. However, there are no studies that compare the seizure outcome with the amount of the resection and the etiology of the epilepsy. This is the aim of this report. Methods: Among 37 temporal epilepsy surgery cases, 26 patients benefit from a selective trans‐sylvian amygdalo‐hippocampectomy. Each patient has a pre‐ and postsurgical evaluation with neurological examination, cognitive testing, RMI and visual field exploration. The amount of resection was calculated on RMI T2 coronal slides (3 mm thickness, 0.3 mm gap) acquired perpendicularly to the hippocampus. For the hippocampus, the% of resected length was calculated and for the amygdala, the% of resected volume was visually evaluated.Seizure outcome was evaluated according to the Engel's classification. Results: 1 patient, with per‐surgery hemorrhage, had temporary arm paresis and speech difficulties, left temporal hemianopsia and right quadranopsia, 8 additional patients have a partial visual field defect (superior quadranopsia) at the short term exam.Among the 26 patients, 22 patients have one year of follow‐up or more. 14 were in Engel class I, 6 in class II and 2 in class III. There was no correlation between the seizure outcome and the extent of removal which varied from less than 1/3 to more than 2/3 of the hippocampus length and from less than half to complete removal for the amygdala.The two patients with class III outcome have a complete resection of the lesion, the amygdala and the hippocampus. For both patients, the etiology was tumoral: for one patient it was the third surgery and for the other patient, there was a 20‐year evolution of refractory seizures. Conclusions: There is no correlation between the amount of hippocampal resection and the seizure outcome. This could be partially explained by the fact that for some patients, the epileptic focus is located into the amygdala or has an extent outside the limbic structures and/or the macroscopic lesion. 1 Mary Ann Werz, 2 Mark L. Cohen, 3 Robert Maciunas, 4 Barbara E. Swartz, 1 Pamela Lang, 1 Michael Schoenberg, 1 Gaotami Rao, and 5 Mark S. Scher ( 1 Neurology, Case Western Reserve Univ; University Hospitals of Cleveland, Cleveland, OH ; 2 Pathology, Case Western Reserve Univ; University Hospitals of Cleveland, Cleveland, OH ; 3 Neurosurgery, Case Western Reserve Univ; University Hospitals of Cleveland, Cleveland, OH ; 4 Neurology, Hoag Hospital Epilepsy Center, Newport Beach, CA ; and 5 Pediatrics, Case Western Reserve Univ; University Hospitals of Cleveland, Cleveland, OH ) Rationale: The pathogenesis of partial epilepsy is poorly understood. We have reviewed our adult epilepsy surgical cases for potential perinatal and childhood contributors to the development of epilepsy. Methods: Retrospective review of surgical cases since 2000 included age of seizure onset, age at surgery, seizure type and frequency, lobe(s) resected, pathology, MRI/PET, neuropsychology, and surgical outcome. Patients, and when possible their mothers, were called for childhood history. Birth records were requested from outside hospitals. Queries included survey of risk factors in the mother, fetus, neonate, infant, and child. Results: Forty‐six epilepsy surgery cases, excluding malignant tumors, were reviewed. The median age in years at surgery was 45 (range 16–74). The median duration of intractable epilepsy was 15 years with a range of several months to 60 years. Average full scale IQ was 95 (SD = 15.2). Resections were of the temporal lobe in 33 (29 standard, 4 neocortical), frontal lobe in 5, parietal lobe in 3, and multilobar in 5. Pathology revealed dysplastic etiologies in 20 patients: focal cortical dysplasia (FCD) in 12 (Palmini classification Ia n = 4, Ib n = 3, IIa n = 5), DNET in 6 and ganglioglioma in 2. Other lesions identified were: vascular malformations (7), pilocytic astrocytoma (1), meningioangiomatosis (1), osteoma (1), and pathogen‐free granuloma (1). Mesial temporal sclerosis was co‐identified in 7 cases of FCD (Type I, IIa) and 9 cases in isolation. MTS was associated with febrile seizures, status epilepticus and/or meningitis in 9 of 17 cases (as infants in 6). Infantile spasms was associated with one case of Type IIa FCD. Significant maternal risk factors were not identified though a maternal history of neurofibramatosis (NF1) was identified in one case of MTS. Isolated MTS was associated with low birth weight (1800 g) in one and with high birth weight (above 4000 g) in 4 patients. Conclusions: Presumed congenital structural abnormalities of the brain were identified in 60% of our adult surgical series: 26% FCD (15% Type I and 11% Type IIa), 17%“extreme FCD” (DNET/ganglioglioma), 15% vascular malformations, and 2% meningioangiomatosis. Isolated MTS or non‐specific gliosis occurred in 34%. These data suggest that developmental dysgenesis of cerebral cortex may be a major cause of adult epilepsy. The timing of microdysgenesis in development as well as genetic and environmental factors in FCD need to be identified. 1 Alan J. Wilensky, 1 Carl B. Dodrill, and 2 George A. Ojemann ( 1 Department of Neurology/Epilepsy Center, University of Washington, Harborview Medical Center, Seattle, WA ; and 2 Department of Neurological Surgery, University of Washington, Seattle, WA ) Rationale: A patient's outcome after epilepsy surgery is usually apparent by one year. Those who have been seizure‐free since surgery usually remain so unless they attempt to stop all of their antiepileptic medication. About half of those who stop their medications do so successfully and remain seizure free. However, there is a very small group of patients who are seizure free, off or on medication, who have recurrent seizures many years after their surgery. This study was undertaken to see if there were any common characteristics to this small group of patients. Methods: Various data bases at the University of Washington were queried to identify patients who had been seizure free for at least five years after surgery. We then identified any patients in that group for whom there were data indicating they had had a recurrence of their seizures. Actual clinical records were then reviewed to confirm long‐term seizure freedom and recurrence and data were abstracted from those records. Results: Nine patients were seizure free for at least 9 years after surgery and then relapsed. One additional patient relapsed in the 6th year year after surgery. An 11th patient developed simple partial seizures after stopping medications after year 4 and had complex partial seizures starting in year 10. Age at onset ranged from 2 months to 19 years and surgery occurred 5 to 29 years later. Seizures recurred as late as 38 years after surgery in one patient and and 22, and 23 years later in two others. Although 58% of all of our resections are left‐sided, only 37% of this group had left sided operations. Two of those had right ‐sided language so that 82% of this group were operated on in the non‐language dominant hemisphere. In a comparable group of 12 patients who remained seizure free only 42% had non‐dominasntr resections.All of these patients had temporal resections as part or all of their surgery. However, atypical for our temporal resection patients as a whole, 9 of the 11 did not have hippocampal resections. Long term 9 of 11 of these patients have continued to have seizures at last follow‐up. Two have had a second surgery and one of those still has frequent seizures. Conclusions: In this small group of patients with late recurrence of seizures after apparent successful surgery two findings stand out. First, this group has an excess proportion, compared to our population as a whole, of non‐dominant hemisphere resections. Second, this group had only 2 patients in whom hippocampal resections were undertaken whereas the majority of our temporal operations include hippoxcampal resection. This finding is consistent with other evidence that extent of hippocampal resection correlates with outcome. These patients may have remained seizure free if their original surgeries had been more complete. 1 Chong H. Wong, 1 Julie Birkett, 1,2 Mark Dexter, 1 Ernest Somerville, 1,2 Raymond Chaseling, 1 Michael Fearnside, and 1,2 Andrew Bleasel ( 1 Department of Neurology and Neurosurgery, Westmead Hospital, Westmead, NSW, Australia ; and 2 Department of Neurology and Neurosurgery, The Children's Hospital at Westmead, Westmead, NSW, Australia ) Rationale: Intracranial EEG recording is necessary in a subset of patients undergoing epilepsy surgery. This study evaluates the risk factors associated with morbidity and mortality during video‐EEG monitoring with subdural electrodes in patients with intractable partial epilepsy. Methods: Retrospective audit of all patients undergoing monitoring with intracranial electrodes at Westmead Hospital and The Children's Hospital at Westmead between 1988–2004. The following variables were recorded: epilepsy syndrome, anticonvulsant medication, duration of implantation surgery and intracranial monitoring, side and site of grid implantation, number of grids and electrodes, size of grids, frequency of clinical seizures, previous craniotomy, other medical conditions and length of hospital stay. Results: 296 patients with refractory epilepsy underwent epilepsy surgery during this period. 79 intracranial monitoring sessions were performed in 71 patients (49 males; median age: 24 years; IQR 17; 35, range 6–54 years) to further localize the ictal onset. 69% had extratemporal epilepsy and 44% were non‐lesional. Median monitoring duration 11 days (IQR 7; 14). 82% of the monitored patients underwent epilepsy resection. 75% had seizure outcome of Engel 1 or 2 with a median follow up period of 31 months. Overall complication rate was 31.6%. 26.6% were directly related to subdural electrode implantation; 13.9% transient complications requiring no treatment, 8.9% transient complications requiring treatment, 1.3% permanent and there were 2 deaths (2.5%). Specific complications included osteomyelitis (2.5%), haemorrhage (7.6%), cerebral infarction (3.8%), transient neurological deficit (5.1%) and pulmonary emboli or deep venous thrombosis (3.8%). The 2 deaths were both related to raised intracranial pressure (1 venous infarction, 1 unexplained) and occurred within 48 hrs of implantation. Complications were associated with greater number of electrodes (p < 0.001), greater number of strip/grid arrays (0.039), 8x8 grids (0.001) and longer duration of monitoring (0.029). Only the total number of electrodes was found to be statistically significant (p < 0.001) when fitted into a multiple logistic regression. No significant correlation existed between complications and seizure frequency, interhemispheric site, repositioning of electrodes or age. Conclusions: The transient complication rate in this series is comparable to previously reported series1. We found increased morbidity and mortality were associated with a greater number of electrodes. The results of this audit are being used to modify our implantation and monitoring protocols.REFERENCE1. Hamer HM, Morris HH, Mascha EJ et al . Neurology 2002: 58; 97–103 (Supported by Sanofi Aventis Australia.) 1 Megdad M. Zaatreh, 1 Albert Hinn, 1 Bradley Vaughn, 1 Melanie Adams, 1 Spencer Weig, 1 Michael Tennison, 1 Robert Greenwood, and 2 Eldad Hadar ( 1 Department of Neurology, The University of North Carolina, Chapel Hill, NC ; and 2 Department of Neurosurgery, The University of North Carolina, Chapel Hill, NC ) Rationale: Epilepsy Surgery offers a cure for selective patients with refractory seizures. After successful epilepsy surgery some patients can be successfully tapered off anticonvulsants potentially improving their quality of life further. Although Epileptologists tend to withdraw anticonvulsants many months or even years after successful surgery, systemic studies about the best timing for anticonvulsant withdrawal are lacking. We present our experience with rapid withdrawal of anticonvulsants in a group of patients after epilepsy surgery Methods: Patients with epilepsy surgery followed in our Epilepsy Surgery Program were offered a trail of anticonvulsants withdrawal 6 weeks after epilepsy surgery with complete withdrawal at 12 weeks. Recurrence of seizures was monitored. Patients with early postoperative seizures, more than 4 anticonvulsants, history of generalized status epilepticus and patients on phenobarbital were excluded. Results: Among 27 patients with epilepsy surgery rapid withdrawal of anticonvulsants was attempted in 15 (9 Females). Age ranged between 6–57 years (mean 28.3 years). Eight patients had temporal lobectomy, 6 extartemporal resections and 1 with hemishperectomy. Pathology was consistent with mesiotemporal sclerosis in 4, focal cortical dysplasia in 4, low grade tumor in 2, atrophy and gliosis is 2 and normal in 3. Seizures recurred in 8 of the 15 patients, 4 during and 4 after complete drug withdrawal. Among the 8 patients with seizure recurrence 4 continued to have seizures after reinstating anticonvulsants while seizures stopped in the other 4 after restarting or increasing anticonvulsants. Seven patients remained seizure free with a mean follow‐up of 9 months postsurgery (6–19 months). Pathological examination of the patients who remained seizure free was cortical dysplasia in 4, low grade tumor in 2 and one patient with mesiotemporal sclerosis. Drug withdrawal was not successful in any patient with normal postoperative histopathology. Conclusions: In selective patients with epilepsy surgery pathological findings may be a factor in successful early withdrawal of anticonvulsants. Further studies are needed to define patients who could benefit from early anticonvulsant withdrawal after epilepsy surgery and the effects of that on quality of life. 1 Tammy N. Tsuchida, 1 Heidi E. Kirsch, 2 Charles E. McCulloch, 1 Gilbert A.W. Woo, and 1 Paul A. Garcia ( 1 Neurology, UC San Francisco, San Francisco, CA ; and 2 Biostatistics, UC San Francisco, San Francisco, CA ) Rationale: Many epileptologists advocate early surgical treatment of refractory epilepsy in children and young adults. This study was performed to determine if there is a relationship between epilepsy duration prior to epilepsy surgery and successful surgical outcomes in patients with childhood‐onset epilepsy. Methods: We performed secondary data analysis of a tertiary care epilepsy surgery database selecting for patients with onset of epilepsy < 18 years old and with their first resective epilepsy surgery. Outcomes at one or more years after surgery were classified as good if the patient was rated Engel Class (EC) 1and poor if rated EC 2–4. Predictors included race, sex, age at onset of epilepsy, age at time of surgery, duration of epilepsy prior to surgery, IQ prior to surgery, MRI findings, bilateral or unilateral MRI findings, Pathology, and Diagnosis. Diagnosis was determined by review of the mri and pathology reports by raters who were blinded to the surgical outcome. Results: 188 patients met criteria for inclusion and had sufficient information for analysis. Mean age at onset 7.46 +/− 5.73 y. Mean age at surgery 30 +/− 11 y. Diagnoses were MTS (60%), cortical malformation (8.5%), tumor (4.2%), and gliosis (7.4%). Patients with cortical malformation, tumor or vascular malformation had a median duration of 13 years. Patients with MTS or gliosis had a median duration of 25 years. If data for patients with one year post surgical followup was analyzed utilizing chi‐square analysis, unilateral mri abnormalities were predictive of good outcome (p = 0.035). Duration, age at onset, and age at surgery were divided into quartiles for analysis. If only the diagnosis MTS was analyzed, the oldest quartile of age at surgery (40–61y) was associated with poorer outcomes (p = 0.034). Cox survival analysis of time to EC3 for patients with longer follow‐up showed that patients with intermediate disease durations had lower hazards than the shortest durations (0.5y to 13y compared to 14y to 22y, HR = .52, p = 0.047; 0.5y to 13y compared to 23y to 32y, HR = 0.60, p = 0.094). The longest disease durations (32.85y to 51y) had hazards comparable to the shortest (HR = 1.02, p = 0.95). When restricted to MTS diagnosis, this relationship persisted (p = 0.03). Conclusions: Our study suggests that children with MTS are less likely to become seizure free than young adults with the same diagnosis. We suspect that the relationship is not causative; rather, young children may have MTS on a different pathophysiological basis. In contrast, it is possible that patients with very long epilepsy duration may have worse outcomes due to the long‐term effect of seizures. Along with prior studies of clinical and radiological findings and surgical outcome, this study should be taken into consideration in counseling patients on the likelihood of surgical success. (Supported by NS01692NIH/NINDS Neurological Sciences Academic Development Award (NSADA).