Search

OBJECTIVE Colloid cysts of the third ventricle are histologically benign lesions that can cause obstructive hydrocephalus and death. Historically, colloid cysts have been removed by open microsurgical approaches. More recently, minimally invasive endoscopic and port-based techniques have offered decreased complications and length of stay, with improved patient satisfaction. METHODS A single-center retrospective analysis of patients with colloid cysts who underwent surgery at a large tertiary care hospital was performed. The cohort was assessed based on the surgical approach, comparing endoscopic resection to open microsurgical resection. The primary endpoint was rate of perioperative complications. Univariate analysis was used to assess several procedure-related variables and the cost of treatment. Multivariate analysis was used to assess predictors of perioperative complications. Total inpatient cost for each case was extracted from the health system financial database. RESULTS The study included 78 patients with colloid cysts who underwent resection either via an endoscopic approach (n = 33) or through a craniotomy (n = 45) with an interhemispheric-transcallosal or transcortical-transventricular approach. Nearly all patients were symptomatic, and half had obstructive hydrocephalus. Endoscopic resection was associated with reduced operative time (3.2 vs 4.9 hours, p < 0.001); lower complication rate (6.1% vs 33.1%, p = 0.009); reduced length of stay (4.1 vs 8.9 days, p < 0.001); and improved discharge to home (100% vs 75.6%, p = 0.008) compared to microsurgical resection. Coagulated residual cyst wall remnants were more common after endoscopic resection (63.6% vs 19.0%, p < 0.001) although this was not associated with a significantly increased rate of reoperation for recurrence. The mean follow-up was longer in the microsurgical resection group (3.1 vs 4.9 years, p = 0.016). The total inpatient cost of endoscopic resection was, on average, one-half (47%) that of microsurgical resection. When complications were encountered, the total inpatient cost of microsurgical resection was 4 times greater than that of endoscopic resection where no major complications were observed. The increased cost-effectiveness of endoscopic resection remained during reoperation. CONCLUSIONS Endoscopic resection of colloid cysts of the third ventricle offers a significant reduction in perioperative complications when compared to microsurgical resection. Endoscopic resection optimizes nearly all procedure-related variables compared to microsurgical resection, and reduces total inpatient cost by > 50%. However, endoscopic resection is associated with a significantly increased likelihood of residual coagulated cyst wall remnants that could increase the rate of reoperation for recurrence. Taken together, endoscopic resection represents a safe and effective minimally invasive approach for removal of colloid cysts.

OBJECTIVE The internal high-dose volume varies widely for a given prescribed dose during stereotactic radiosurgery (SRS) to treat brain metastases (BMs). This may be altered during treatment planning, and the authors have previously shown that this improves local control (LC) for non–small cell lung cancer BMs without increasing toxicity. Here, they seek to identify potentially actionable dosimetric predictors of LC after SRS for melanoma BM. METHODS The records of patients with unresected melanoma BM treated with single-fraction Gamma Knife RS between 2006 and 2017 were reviewed. LC was assessed on a per-lesion basis, defined as stability or a decrease in lesion size. Outcome-oriented approaches were utilized to determine optimal dichotomization for dosimetric variables relative to LC. Univariable and multivariable Cox regression analysis was implemented to evaluate the impact of collected parameters on LC. RESULTS Two hundred eighty-seven melanoma BMs in 79 patients were identified. The median age was 56 years (range 31–86 years). The median follow-up was 7.6 months (range 0.5–81.6 months), and the median survival was 9.3 months (range 1.3–81.6 months). Lesions were optimally stratified by volume receiving at least 30 Gy (V30) greater than or equal to versus less than 25%. V30 was ≥ and < 25% in 147 and 140 lesions, respectively. For all patients, 1-year LC was 83% versus 66% for V30 ≥ and < 25%, respectively (p = 0.001). Stratifying by volume, lesions 2 cm or less (n = 215) had 1-year LC of 82% versus 70% (p = 0.013) for V30 ≥ and < 25%, respectively. Lesions > 2 to 3 cm (n = 32) had 1-year LC of 100% versus 43% (p = 0.214) for V30 ≥ and < 25%, respectively. V30 was still predictive of LC even after controlling for the use of immunotherapy and targeted therapy. Radionecrosis occurred in 2.8% of lesions and was not significantly associated with V30. CONCLUSIONS For a given prescription dose, an increased internal high-dose volume, as indicated by measures such as V30 ≥ 25%, is associated with improved LC but not increased toxicity in single-fraction SRS for melanoma BM. Internal dose escalation is an independent predictor of improved LC even in patients receiving immunotherapy and/or targeted therapy. This represents a dosimetric parameter that is actionable at the time of treatment planning and warrants further evaluation.

OBJECTIVEIntraoperative MRI (iMRI) is used in the surgical treatment of glioblastoma, with uncertain effects on outcomes. The authors evaluated the impact of iMRI on extent of resection (EOR) and overall survival (OS) while controlling for other known and suspected predictors.METHODSA multicenter retrospective cohort of 640 adult patients with newly diagnosed supratentorial glioblastoma who underwent resection was evaluated. iMRI was performed in 332/640 cases (51.9%). Reviews of MRI features and tumor volumetric analysis were performed on a subsample of cases (n = 286; 110 non-iMRI, 176 iMRI) from a single institution.RESULTSThe median age was 60.0 years (mean 58.5 years, range 20.5–86.3 years). The median OS was 17.0 months (95% CI 15.6–18.4 months). Gross-total resection (GTR) was achieved in 403/640 cases (63.0%). Kaplan-Meier analysis of 286 cases with volumetric analysis for EOR (grouped into 100%, 95%–99%, 80%–94%, and 50%–79%) showed longer OS for 100% EOR compared to all other groups (p < 0.01). Additional resection after iMRI was performed in 104/122 cases (85.2%) with initial subtotal resection (STR), leading to a 6.3% mean increase in EOR and a 2.2-cm3 mean decrease in tumor volume. For iMRI cases with volumetric analysis, the GTR rate increased from 54/176 (30.7%) on iMRI to 126/176 (71.5%) postoperatively. The EOR was significantly higher in the iMRI group for intended GTR and STR groups (p = 0.02 and p < 0.01, respectively). Predictors of GTR on multivariate logistic regression included iMRI use and intended GTR. Predictors of shorter OS on multivariate Cox regression included older age, STR, isocitrate dehydrogenase 1 (IDH1) wild type, no O6-methylguanine DNA methyltransferase (MGMT) methylation, and no Stupp therapy. iMRI was a significant predictor of OS on univariate (HR 0.82, 95% CI 0.69–0.98; p = 0.03) but not multivariate analyses. Use of iMRI was not associated with an increased rate of new permanent neurological deficits.CONCLUSIONSGTR increased OS for patients with newly diagnosed glioblastoma after adjusting for other prognostic factors. iMRI increased EOR and GTR rate and was a significant predictor of GTR on multivariate analysis; however, iMRI was not an independent predictor of OS. Additional supporting evidence is needed to determine the clinical benefit of iMRI in the management of glioblastoma.

Nonhemorrhagic neurological deficits are underrecognized symptoms of intracranial dural arteriovenous fistulas (dAVFs) having cortical venous drainage. These symptoms are the consequence of cortical venous hypertension and portend a clinical course with increased risk of neurological morbidity and mortality. One rarely documented and easily misinterpreted type of nonhemorrhagic neurological deficit is progressive dementia, which can result from venous hypertension in the cortex or in bilateral thalami. The latter, which is due to dAVF drainage into the deep venous system, is the less common of these 2 dementia syndromes. Herein, the authors report 4 cases of dAVF with venous drainage into the vein of Galen causing bithalamic edema and rapidly progressive dementia. Two patients were treated successfully with endovascular embolization, and the other 2 patients were treated successfully with endovascular embolization followed by surgery. The radiographic abnormalities and presenting symptoms rapidly resolved after dAVF obliteration in all 4 cases. Detailed descriptions of these 4 cases are presented along with a critical review of 15 previously reported cases. In our analysis of these 19 published cases, the following were emphasized: 1) the clinical and radiographic differences between dAVF-induced thalamic versus cortical dementia syndromes; 2) the differential diagnosis and necessary radiographic workup for patients presenting with a rapidly progressive thalamic dementia syndrome; 3) the frequency at which delays in diagnosis occurred and potentially dangerous and avoidable diagnostic procedures were used; and 4) the rapidity and completeness of symptom resolution following dAVF treatment.

Object The purpose of aneurysm surgery is complete aneurysm obliteration while sparing associated arteries. Indocyanine green (ICG) videoangiography is a new technique that allows for real-time evaluation of blood flow in the aneurysm and vessels. The authors performed a retrospective study to compare the accuracy of ICG videoangiography with intraoperative angiography (IA), and determine if ICG videoangiography can be used without follow-up IA. Methods From June 2007 through September 2009, 155 patients underwent craniotomies for clipping of aneurysms. Operative summaries, angiograms, and operative and ICG videoangiography videos were reviewed. The number, size, and location of aneurysms, the ICG videoangiography and IA findings, and the need for clip adjustment after ICG videoangiography and IA were recorded. Discordance between ICG videoangiography and IA was defined as ICG videoangiography demonstrating aneurysm obliteration and normal vessel flow, but post-IA showing either an aneurysmal remnant and/or vessel occlusion requiring clip adjustment. Results Thirty-two percent of patients (49 of 155) underwent both ICG videoangiography and IA. The post-ICG videoangiography clip adjustment rate was 4.1% (2 of 49). The overall rate of ICG videoangiography–IA agreement was 75.5% (37 of 49) and the ICG videoangiography–IA discordance rate requiring post-IA clip adjustment was 14.3% (7 of 49). Adjustments were due to 3 aneurysmal remnants and 4 vessel occlusions. These adjustments were attributed to obscuration of the residual aneurysm or the affected vessel from the field of view and the presence of dye in the affected vessel via collateral flow. Although not statistically significant, there was a trend for ICG videoangiography–IA discordance requiring clip adjustment to occur in cases involving the anterior communicating artery complex, with an odds ratio of 3.3 for ICG videoangiography–IA discordance in these cases. Conclusions These results suggest that care should be taken when considering ICG videoangiography as the sole means for intraoperative evaluation of aneurysm clip application. The authors further conclude that IA should remain the gold standard for evaluation during aneurysm surgery. However, a combination of ICG videoangiography and IA may ultimately prove to be the most effective strategy for maximizing the safety and efficacy of aneurysm surgery.

Object The aim of this study was to define the optimal treatment for patients with symptomatic intraluminal carotid artery thrombus (ICAT). Methods The authors performed a retrospective chart review of patients who had presented with symptomatic ICAT at their institution between 2001 and 2011. Results Twenty-four patients (16 males and 8 females) with ICAT presented with ischemic stroke (18 patients) or transient ischemic attack ([TIA], 6 patients). All were initially treated using anticoagulation with or without antiplatelet drugs. Eight of these patients had no or only mild carotid artery stenosis on initial angiography and were treated with medical management alone. The remaining 16 patients had moderate or severe carotid stenosis on initial angiography; of these, 10 underwent delayed revascularization (8 patients, carotid endarterectomy [CEA]; 2 patients, angioplasty and stenting), 2 refused revascularization, and 4 were treated with medical therapy alone. One patient had multiple TIAs despite medical therapy and eventually underwent CEA; the remaining 23 patients had no TIAs after treatment. No patient suffered ischemic or hemorrhagic stroke while on anticoagulation therapy, either during the perioperative period or in the long-term follow-up; 1 patient died of an unrelated condition. The mean follow-up was 16.4 months. Conclusions Results of this study suggest that initial anticoagulation for symptomatic ICAT leads to a low rate of recurrent ischemic events and that carotid revascularization, if indicated, can be safely performed in a delayed manner.

Object The goal of this study was to determine the clinical course of Borden-Shucart Type I cranial dural arteriovenous fistulas (DAVFs) and to calculate the annual rate of conversion of these lesions to more aggressive fistulas that have cortical venous drainage (CVD). Methods A retrospective chart review was conducted of all patients harboring DAVFs who were seen at the authors' institution between 1997 and 2009. Twenty-three patients with Type I DAVFs who had available clinical follow-up were identified. Angiographic and clinical data from these patients were reviewed. Neurological outcome and status of presenting symptoms were assessed during long-term follow-up. Results Of the 23 patients, 13 underwent endovascular treatment for intolerable tinnitus or ophthalmological symptoms, and 10 did not undergo treatment. Three untreated patients died of unrelated causes. In those who were treated, complete DAVF obliteration was achieved in 4 patients, and palliative reduction in DAVF flow was achieved in 9 patients. Of the 19 patients without radiographic cure, no patient developed intracranial hemorrhage or nonhemorrhagic neurological deficits (NHNDs), and no patient died of DAVF-related causes over a mean follow-up of 5.6 years. One patient experienced a spontaneous, asymptomatic obliteration of a partially treated DAVF in late follow-up, and 2 patients experienced a symptomatic conversion of their DAVF to a higher-grade fistula with CVD in late follow-up. The annual rate of conversion to a higher-grade DAVF based on Kaplan-Meier cumulative event-free survival analysis was 1.0%. The annual rate of intracranial hemorrhage, NHND, and DAVF-related death was 0.0%. Conclusions A small number of Type I DAVFs will convert to more aggressive DAVFs with CVD over time. This conversion to a higher-grade DAVF is typically heralded by a change in patient symptoms. Follow-up vascular imaging is important, particularly in the setting of recurrent or new symptoms.

Object The yield of CT angiography (CTA) and MR angiography (MRA) in patients with subarachnoid hemorrhage (SAH) who have a negative initial catheter angiogram is currently not well understood. This study aims to determine the yield of CTA and MRA in a prospective cohort of patients with SAH and a negative initial catheter angiogram. Methods From January 1, 2005, until September 1, 2010, the authors instituted a prospective protocol in which patients with SAH—as documented by noncontrast CT or CSF xanthochromia and a negative initial catheter angiogram— were evaluated using CTA and MRA to assess for causative cerebral aneurysms. Two neuroradiologists independently evaluated the noncontrast CT scans to determine the SAH pattern (perimesencephalic or not) and the CT and MR angiograms to assess for causative cerebral aneurysms. Results Seventy-seven patients were included, with a mean age of 52.8 years (median 54 years, range 19–88 years). Fifty patients were female (64.9%) and 27 male (35.1%). Forty-three patients had nonperimesencephalic SAH (55.8%), 29 patients had perimesencephalic SAH (37.7%), and 5 patients had CSF xanthochromia (6.5%). Computed tomography angiography demonstrated a causative cerebral aneurysm in 4 patients (5.2% yield), all of whom had nonperimesencephalic SAH (9.3% yield). Mean aneurysm size was 2.6 mm (range 2.1–3.3 mm). Magnetic resonance angiography demonstrated only 1 of these aneurysms. No causative cerebral aneurysms were found in patients with perimesencephalic SAH or CSF xanthochromia. Conclusions Computed tomography angiography is a valuable adjunct in the evaluation of patients with nonperimesencephalic SAH who have a negative initial catheter angiogram, demonstrating a causative cerebral aneurysm in 9.3% of patients.

Object Large cerebral arteriovenous malformations (AVMs) are often not amenable to direct resection or stereotactic radiosurgery (SRS) treatment. An alternative treatment strategy is staged endovascular embolization followed by SRS (Embo/SRS). The object of this study was to examine the experience at Washington University in St. Louis with Embo/SRS for large AVMs and review the results in earlier case series. Methods Twenty-one cases involving patients with large AVMs treated with Embo/SRS between 1994 and 2006 were retrospectively evaluated. The AVM size (before and after embolization), procedural complications, radiological outcome, and neurological outcome were examined. Radiological success was defined as AVM obliteration as demonstrated by catheter angiography, CT angiography, or MR angiography. Radiological failure was defined as residual AVM as demonstrated by catheter angiography, CT angiography, or MR angiography performed at least 3 years after SRS. Results The maximum diameter of all AVMs in this series was > 3 cm (mean 4.2 cm); 12 (57%) were Spetzler-Martin Grade IV or V. Clinical follow-up was available in 20 of 21 cases; radiological follow-up was available in 19 of 21 cases (mean duration of follow-up 3.6 years). Forty-three embolization procedures were performed; 8 embolization-related complications occurred, leading to transient neurological deficits in 5 patients (24%), minor permanent neurological deficits in 3 patients (14%), and major permanent neurological deficits in none (0%). Twenty-one SRS procedures were performed; 1 radiation-induced complication occurred (5%), leading to a permanent minor neurological deficit. Of the 20 patients with clinical follow-up, none experienced cerebral hemorrhage. In the 19 patients with radiological follow-up, AVM obliteration was confirmed by catheter angiography in 13, MR angiography in 2, and CT angiography in 1. Residual nidus was found in 3 patients. In patients with follow-up catheter angiography, the AVM obliteration rate was 81% (13 of 16 cases). Conclusions Staged endovascular embolization followed by SRS provides an effective means of treating large AVMs not amenable to standard surgical or SRS treatment. The outcomes and complication rates reported in this series compare favorably to the results of other reported therapeutic strategies for this very challenging patient population.

Object The aim of this study was to determine the efficacy of venous ultrasonography in screening for deep venous thrombosis (DVT) after subarachnoid hemorrhage (SAH). A large cohort of patients who had suffered SAH was evaluated with the primary end points of ascertaining the incidence of asymptomatic DVT with venous Doppler ultrasonography screening and of identifying risk factors for the development of DVT and subsequent pulmonary embolism. Methods Data from patients with aneurysmal SAH who had been admitted to the neurosurgical intensive care unit (ICU) between December 2002 and October 2006 were retrospectively evaluated. Patients who had undergone surgical or endovascular treatment of an aneurysm following SAH and survived ≥ 15 days were included in the study. Results The overall incidence of DVT among the entire study cohort was 18%. A subgroup analysis identified all patients, with or without symptoms for DVT, who had undergone venous Doppler ultrasonography screening. The incidence of asymptomatic DVT was 24%. Univariate analysis of all patients revealed a significant correlation between the risk of DVT and Hunt and Hess grade (r = 0.38, p < 0.0001), Fisher grade (r = 0.31, p < 0.0001), total hospital stay (r = 0.49, p < 0.0001), and number of days in the ICU (r = 0.48, p < 0.0001). Multivariate analysis demonstrated that the total hospital stay and number of days in the ICU were significant predictors of DVT in all patients (p < 0.0001 and p < 0.0002, respectively). In the subgroup of screened patients, Hunt and Hess grade, total hospital stay, and number of days in the ICU were significant predictors of DVT. Although screened patients were more likely to have DVT (χ2 = 6.0976, p < 0.02), there was no significant difference in the incidence of DVT or pulmonary embolism between patients who did and those who did not undergo routine lower-extremity Doppler ultrasonography screening. Conclusions Routine compressive venous Doppler ultrasonography is an efficient, noninvasive means of identifying DVT as a screening modality in both symptomatic and asymptomatic patients following aneurysmal SAH. The ability to confirm or deny the presence of DVT in this patient population allows one to better identify the indications for chemoprophylaxis. Prophylaxis for venous thromboembolism in neurosurgical patients is common. Emerging literature and anecdotal experience have exposed risks of complications with prophylactic anticoagulation protocols. The identification of patients at high risk—for example, those with asymptomatic DVT—will allow physicians to better assess the role of prophylactic anticoagulation.

✓ Definitive endovascular treatment of dural arteriovenous fistulas (DAVFs) requires obliteration of the site of the fistula: either the diseased dural sinus or the pial vein. Access to this site is often limited by occlusion of the sinus proximal and distal to the segment containing the fistula. The authors describe a technique in which the mastoid emissary vein is used to gain access to a Borden–Shucart Type II DAVF in the transverse–sigmoid sinus. Recognition of this route of access, if present, may facilitate endovascular treatment of these lesions. Access to the transverse sinus via this approach can be straightforward and may be underused.

Object. Cells that lose their ability to undergo apoptosis may promote the development of neoplasms and result in resistance to clinical treatment with DNA-damaging modalities such as radio- and chemotherapy. Four established human glioma cell lines that are resistant to apoptosis were transfected with the proapoptotic gene bax and assessed for their sensitivity to a proapoptotic stimulus.Methods. Two cell lines had a wild-type p53 genotype (U87 and D247MG) and two had mutant p53 genotypes (U138 and U373). Constitutive overexpression of murine bax was achieved in U138 and U373 only, which resulted in an increased sensitivity of these lines to the apoptosis-inducing effect of cytosine arabinoside (ara-C). Multiple attempts to produce constitutive overexpression of bax in U87 and D247MG cells resulted in spontaneous, near-complete cell loss. Vector-only control transfections were successful in all four cell lines. Inducible overexpression of bax was achieved in the U87 cells and elevated levels of BAX were observed as early as 6 hours after gene induction. This overexpression of BAX resulted in the spontaneous induction of apoptosis in these cells.Conclusions. Overexpression of BAX in four human glioma cell lines resulted in increased sensitivity to apoptosis. In the two lines that had a wild-type p53 genotype, overexpression of BAX produced spontaneous apoptosis. In contrast, the lines that had mutant, nonfunctional P53 did not undergo spontaneous apoptosis, but they were rendered more sensitive to the apoptosis-inducing effect of ara-C. Modulation of BAX expression may be a useful therapeutic modality for gliomas, regardless of p53 genotype.

Object. Genes known to be involved in the regulation of apoptosis include members of the bcl-2 gene family, such as inhibitors of apoptosis (bcl-2 and bcl-xl) and promoters of apoptosis (bax). The authors investigated a potential approach for the treatment of malignant gliomas by using a gene transfection technique to manipulate the level of an intracellular protein involved in the control of apoptosis. Methods. The authors transfected the murine bax gene, which had been cloned into a mammalian expression vector, into the C6 rat glioma cell line. Overexpression of the bax gene resulted in a decreased growth rate (average doubling time of 32.96 hours compared with 22.49 hours for untransfected C6, and 23.11 hours for clones transfected with pcDNA3 only), which may be caused, in part, by an increased rate of spontaneous apoptosis (0.77 ± 0.15% compared with 0.42 ± 0.08% for the vector-only transfected C6 cell line; p = 0.038, two-tailed Student's t-test). Treatment with 1 µM cytosine arabinoside (ara-C) resulted in significantly more cells undergoing apoptosis in the cell line overexpressing bax than in the vector-only control cell line (23.57 ± 2.6% compared with 5.3 ± 0.7% terminal deoxynucleotidyl transferase—mediated biotinylated—deoxyuridine triphosphate nick-end labeling technique—positive cells; p = 0.007). Furthermore, measurements of growth curves obtained immediately after treatment with 0.5 µM ara-C demonstrated a prolonged growth arrest of at least 6 days in the cell line overexpressing bax. Conclusions. These results can be used collectively to argue that overexpression of bax results in increased sensitivity of C6 cells to ara-C and that increasing bax expression may be a useful strategy, in general, for increasing the sensitivity of gliomas to antineoplastic treatments.

Preoperative differentiation between dysembryoplastic neuroepithelial tumor (DNT) and low-grade glioma is often not possible. Dysembryoplastic neuroepithelial tumor is a recently described entity of uncertain origin; however, the diagnosis has important clinical implications. Clinical and radiological findings of DNT and low-grade glioma, especially oligodendroglioma, may be similar. Treatment options and prognosis differ significantly between these two lesions; consequently, accurate diagnosis is imperative. The authors describe two individuals who presented simultaneously at their institution: one patient with an oligodendroglioma and a second patient with DNT. The natural history, neurodiagnostic, and pathological features of each are reviewed with special emphasis on the potential utility of magnetic resonance spectroscopy in differentiating these lesions.

✓ The clinical usefulness of intraoperative electroencephalographic (EEG) monitoring of cerebral perfusion during aneurysm surgery has received little attention, primarily due to problems with electrode placement over the operative site, which is the area of maximal risk. In this report, 27 patients undergoing surgery for anterior circulation aneurysms were monitored intraoperatively with a two-channel computerized EEG complex using a bilateral frontal-occipital montage. In 16 patients, a normal EEG pattern was observed throughout surgery; all 16 awoke neurologically intact and their postoperative angiograms did not reveal cerebral vasospasm. In the other 11 patients, one of two patterns of persistent EEG abnormalities was identified. 1) In six of these patients a marked attenuation of EEG activity was observed ipsilaterally which coincided with various intraoperative events including brain retraction, hypotension, and aneurysm dissection/clipping. Five of these six patients awoke with new neurological deficits which persisted beyond 12 hours in two, both of whom had angiographically proven vasospasm 24 hours after surgery. 2) In the remaining five patients, a distinct abnormal EEG pattern consisting of marked hyperactivity in the delta, theta, and alpha frequency ranges was observed ipsilaterally. Four of these five patients had a poor neurological outcome and vasospasm on their angiogram 24 hours after surgery. Thus, EEG monitoring which spans the operative area during aneurysm surgery is practicable and appears to be of value in the detection of compromised cerebral perfusion during aneurysm surgery. The possible significance of the two abnormal EEG patterns identified in this report is discussed.