Neuro-Oncology, Oct 2012

Patients with cancer are at high risk of developing venous thromboembolism (VTE). The purpose of this study was to review VTE development in patients undergoing craniotomy for a neoplasm, and to further analyze risk based on multiple pre-, intra-, and post-operative variables. We analyzed all consecutive patients at our institution during the years 1999-2010 who were admitted for craniotomy and had a histological diagnosis of intra-cranial neoplasm. Data points including patient demographics, length of stay, surgical time, surgical position, pre-existing comorbidities, results of extremity ultrasounds for deep venous thrombosis (DVT), and results of diagnostic pulmonary embolus (PE) studies were collected. This study was reviewed and approved by our institutional review board. A total of 1147 patients met the inclusion criteria. Nineteen percent of our patients were diagnosed with a DVT and 4.2% were diagnosed with a PE during their hospitalization. According to logistic multivariate regression analysis, male gender (OR = 14.2, p < .001), ethnicity other than Caucasian (OR = 2.0, p = .04), prolonged ICU stay (OR = 5.0, p = .003), prior VTE (OR = 7.6, p = .01), and tumor histology type (p = .04) (29.6% with lymphomas, 23.8% with metastatic brain tumors, 21.4% with high grade gliomas, and 8.2% of those with meningiomas) were all significant risk factors, after controlling for post-operative VTE medical and mechanical prophylaxis. These data indicate that VTE occurs frequently in neurosurgical patients undergoing surgery for intracranial neoplasms. Overall, data on the prevention of and the best treatment for VTE after craniotomy are limited. Establishing safe standard of care is the next step and is of utmost importance.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:


NEURO-ONCOLOGY † O C T O B E R Abstracts Shouwei Li 1 2 3 Zhong Zhang 1 2 3 Xuzhu Chen 0 1 2 3 Gan You 1 2 3 Pei Yang 1 2 3 Wei Yan 1 2 3 Zhaoshi Bao 1 2 3 Kun Yao 1 2 3 4 Yanwei Liu 1 2 3 Lei Wang 1 2 3 Tao Jiang 1 2 3 Department of Neurosurgery 1 2 3 Beijing Tiantan Hospital 1 2 3 Capital Medical University 1 2 3 Beijing 1 2 3 China 1 2 3 Department of Neurosurgery 1 2 3 0 Department of Neuroimaging, Beijing Tiantan Hospital, Capital Medical University , Beijing , China 1 Beijing Sanbo Brain Hospital, Capital Medical University , Beijing , China 2 NS-03. NEVER SAY NEVER: RISK FACTORS FOR, AND CONSEQUENCES OF HOSPITAL ACQUIRED NEVER EVENTS IN NEURO-ONCOLOGIC SURGERY Brad E. Zacharia , Christopher Deibert, Hani R. Malone, Adam M. Sonabend, Alfred I. Neugut, Benjamin Spencer , and Jeffrey N. Bruce; Columbia University Medical Center , New York, NY , USA 3 NS-01. A RETROSPECTIVE REVIEW OF VTE DEVELOPMENT IN 1147 PATIENTS ADMITTED FOR SURGICAL TREATMENT OF INTRACRANIAL NEOPLASM Allan D. Nanney, III , Joseph G. Adel, Timothy R. Smith , and James P. Chandler; Northwestern University , Chicago, IL , USA 4 Department of Pathology, Beijing Sanbo Brain Hospital, Capital Medical University , Beijing , China Published by Oxford University Press on behalf of the Society for Neuro-Oncology 2012. - NEUROSURGICAL TREATMENTS Patients with cancer are at high risk of developing venous thromboembolism (VTE). The purpose of this study was to review VTE development in patients undergoing craniotomy for a neoplasm, and to further analyze risk based on multiple pre-, intra-, and post-operative variables. We analyzed all consecutive patients at our institution during the years 1999-2010 who were admitted for craniotomy and had a histological diagnosis of intracranial neoplasm. Data points including patient demographics, length of stay, surgical time, surgical position, pre-existing comorbidities, results of extremity ultrasounds for deep venous thrombosis (DVT), and results of diagnostic pulmonary embolus (PE) studies were collected. This study was reviewed and approved by our institutional review board. A total of 1147 patients met the inclusion criteria. Nineteen percent of our patients were diagnosed with a DVT and 4.2% were diagnosed with a PE during their hospitalization. According to logistic multivariate regression analysis, male gender (OR ¼ 14.2, p , .001), ethnicity other than Caucasian (OR ¼ 2.0, p ¼ .04), prolonged ICU stay (OR ¼ 5.0, p ¼ .003), prior VTE (OR ¼ 7.6, p ¼ .01), and tumor histology type (p ¼ .04) (29.6% with lymphomas, 23.8% with metastatic brain tumors, 21.4% with high grade gliomas, and 8.2% of those with meningiomas) were all significant risk factors, after controlling for post-operative VTE medical and mechanical prophylaxis. These data indicate that VTE occurs frequently in neurosurgical patients undergoing surgery for intracranial neoplasms. Overall, data on the prevention of and the best treatment for VTE after craniotomy are limited. Establishing safe standard of care is the next step and is of utmost importance. NS-02. RISK FACTORS FOR VTE IN PATIENTS UNDERGOING CRANIOTOMY FOR NEOPLASTIC DISEASE Kristopher T. Kimmell and Kevin Walter; University of Rochester Medical Center, Rochester, NY BACKGROUND: Patients undergoing neurosurgical procedures for neoplasia have historically been considered at higher risk for developing venous thromboemboli (VTE). This has been suggested in previously published randomized control trials but without statistical significance. METHODS: We retrospectively reviewed a prospectively collected patient cohort from a national surgical quality database (NSQIP, All patients who underwent non-emergent surgery were analyzed. Craniotomy for neoplastic indications was based on CPT codes. Various clinical factors were analyzed to identify possible risk factors for development of VTE events. RESULTS: 3098 patients who underwent non-emergent craniotomy were identified, and 1741 patients underwent procedures for neoplastic disease (56.2%). The rate of DVT in these patients was 3.2% compared to 1.4% in other neurosurgical patients (OR 2.298, CI 2.293-2.303). The rate of pulmonary embolism (PE) was 1.8% compared to 0.5% (OR 3.611, CI 3.599-3.623). Pre-operative characteristics in patients undergoing craniotomy for neoplasm associated with VTE were history of COPD, CHF, HTN, stroke, bleeding disorder, the presence of impaired sensorium, hemiplegia, chronic steroid use, and chemotherapy within 30 days of surgery. Post-operative events associated with VTE included pneumonia, UTI, stroke, coma, sepsis, and septic shock. CONCLUSION: Patients undergoing craniotomy for neoplasm have a baseline increased risk of developing VTE. This risk appears to be magnified in the setting of pre-operative medical comorbidities, in particular cardiopulmonary, cerebrovascular, and neurologic impairments. Post-operative complications also strongly correlate with these events. More work needs to be done to develop pre-operative risk stratification for patients undergoing craniotomy for neoplasm and to identify causative agents intrinsic to neoplasms that play a role in VTE formation. The ever-growing cost burden of the U.S. healthcare system poses concerns regarding its long-term viability. Following the publication of the landmark IOM study “To Err is Human,” effort has been made to institute programs that improve quality and reduce costs. In 2007, in response to growing safety concerns and to minimize misaligned financial incentives, the Centers for Medicare and Medicaid Services (CMS) began prohibiting reimbursement for specified hospital-acquired conditions (HACs), also known as never events. Data regarding HACs in the neuro-oncologic literature are limited. We sought to develop benchmark data, evaluate the risk factors for, and complications of HACs following surgery for common neurosurgical oncologic procedures. We utilized the Nationwide Inpatient Sample (NIS) covering the years 2002-2009 and selected adults admitted for resection of a primary or metastatic brain tumor, meningioma, or acoustic neuroma. Baseline demographics, medical comorbidity, and perioperative neurological complications were assessed, as were hospital-level variables such as teaching-status and surgical volume. A generalized-estimating-equation, multivariable-logistic model was utilized with CMS HACs as the outcome variable. Overall, 310,133 patients were analyzed and 5.4% experienced an HAC. Patients with acoustic neuroma were half as likely to suffer an HAC relative to those with primary brain tumors (OR 0.44; p , 0.01). Increasing comorbidities and a neurosurgical complication substantially increased one’s risk of having an HAC (ORs 1.56 and 2.48; p , 0.01). Patients at urban teaching hospitals and those with high surgical volume were more likely to suffer an HAC (OR 1.33 and 1.16; p , 0.01). The CMS Never Events initiative was designed to better align payment with quality of care. While there are faults with such a system, it appears that it is here to stay and is only being expanded and increasingly adopted. Benchmark data and thorough analysis of the relationship between patient, procedure, and never event incidence will be important as we adapt to and implement such quality initiatives. Multimodal treatments have shown survival benefits in patients with glioblastoma (GBM); however, tumor volume resection remains the most effective treatment. In this prospective observational study, we analyzed the clinical values of established predictors in the context of extensive tumor resection in 234 newly diagnosed GBM patients who underwent near-total resection. Common survival factors, including adjuvant therapy, O6-methylguanine DNA methyltransferase (MGMT), and isocitrate dehydrogenase 1 (IDH1), were analyzed. At a median follow-up of 48.5 months, the median progression-free survival (PFS) and overall survival (OS) was 11.1 and 14.1 months, respectively. Multivariate analysis identified that independent prognosticators for longer OS were only gross-total resection (GTR) (P ¼ 0.003; relative risk [RR], 0.36) and higher KPS score (P ¼ 0.008; RR, 0.41); independent prognosticators for favorable PFS were GTR (P ¼ 0.050; RR, 0.56), IDH1 mutation (P ¼ 0.011; RR, 0.42), and higher KPS (P ¼ 0.028; RR, 0.52). Age at diagnosis, initial alkylating radiochemotherapy, and MGMT promoter methylation were correlated with improved survival in univariate analysis, but their independent predictive values were reduced or lost in multivariate analysis. Interestingly, in the subgroup of patients receiving GTR, neither MGMT nor IDH1 effectively predicted both OS and PFS; and although alkylating radiochemotherapy showed strongly survival advantage, radiotherapy alone did not significantly improve the outcome. Overall, surgical extent significantly impacted the value of prognostic markers. We therefore conclude that surgical resection should be taken into full account when assessing stratification variables or the benefit of adjuvant therapeutic modalities in patients with GBM. NS-05. ANTERIOR TRANSPETROSAL (KAWASE) APPROACH TO THE PETROCLIVAL REGION: INDICATIONS, SURGICAL TECHNIQUE, EXPECTATIONS AND RESTRICTIONS Mohamed K. Farhoud; Alexandria Faculty of Medicine, Neurosurgery Department, Egypt, Alexandria, Egypt OBJECTIVES: This study was designed to study the anterior transpetrosal approach for petroclival lesions with respect to microanatomy, access to the vascular supply of the tumor, and the amount of bone drilling needed to achieve a better working angle, together with its limitations. MATERIAL: Fourteen patients were included in this study. Ten were suffering from petroclival meningiomas, while 4 had a trigeminal schwannoma. METHODS: All patients were operated on in the lateral position; the amount of bone drilling was calculated using the AMIRA software with a 3D reconstruction postoperative CT of the petrous bone. RESULTS: Subtotal removal was achieved in 4 patients, whereas total excision was achieved in the remaining 10. DISCUSSION: Tumors involving the petroclival region have been a surgical challenge for many years. It is difficult to safely remove tumors in this area because of their proximity to the vertebrobasilar system, brainstem, cavernous sinus, and cranial nerves. The anterior transpetrosal approach has been developed and refined to gain access to this challenging anatomic area. CONCLUSION: The anterior transpetrosal approach allows direct access to the petroclival region through a keyhole created by drilling the anterior petrous apex. Bloodless tumor removal can be achieved by attacking the main blood supply to the tumor through detachment of the tentorial artery prior to tumor resection. Direct access to the space anterior to the internal auditory canal can minimize injury to the lower cranial nerves. We consider this approach to be a good alternative to the classical subtemporal approach, but it requires a lot of training and it is limited to those lesions that do not extend below the ponto-medullary junction. NS-06. IS THERE STILL A PLACE FOR STEREOTACTIC BRACHYTHERAPY FOR BRAIN TUMORS IN MODERN NEURO-ONCOLOGY? Maximilian I. Ruge; University Clinics of Cologne, Cologne, Germany Stereotactic implantation of irradiation sources (so-called stereotactic brachytherapy [SBT]) has been applied for intrinsic brain tumors and metastases for more than four decades in numerous patients. The majority of studies reported about the application of high-dose-rate Iodine-125 implants (40-70 cGy/h) for high-grade gliomas, including two prospective randomized trials, which compared standard treatment + SBT. This approach, however, was associated with high incidence of radiation-induced adverse effects requiring repeated surgery and failed to provide any significant oncological benefit as compared to standard treatment regiments. Another approach using SBT was the application of low-dose-rate implants (3-8 cGy/ h) for slow-growing low-grade gliomas or brain metastases, which demonstrated in several very recent publications to be associated with only little permanent deficits and almost an absence of radiation-induced necrosis. SBT represents a highly localized treatment option for well-circumscribed tumors (≤ 4 cm) and aims as well as micro-neurosurgical resection towards the devitalisation/removal of the dense (visible) portion of tumor cells by delivering a lethal irradiation dose (≥ 200 Gy in the vicinity of the implanted source) from within the tumor while optimally sparing surrounding tissue. This makes SBT a minimal invasive and safe technique and highly attractive, especially for inoperable lesions in highly eloquent locations. Furthermore, SBT can be repeated in case of local/regional recurrence and does not hinder or limit the application of EBRT. Even though class I evidence is yet lacking, recent available data for low-grade gliomas, metastases, and other rare CNS tumors demonstrate that progression-free and overall survival are well comparable to this after microsurgical (complete) resection or stereotactic radiosurgery. Local tumor response is at least as effective as after EBRT or chemotherapy. Thus, SBT still represents, as one of the oldest neurosurgical and highly sophisticated stereotactic techniques, a strictly local neuro-oncological treatment option for selected patient populations deserving further attention. INTRODUCTION: Treatment options for glioblastoma at recurrence are various despite the limited efficacy. Surgical resection has been used both for confirmation of recurrent disease as well as for debulking in order to provide relief of symptoms. Therefore, the role surgical resection for recurrent glioblastoma has not been completely clarified. METHODS: A retrospective analysis was made for glioblastoma patients followed between 01/2005 and 06/2010. Eligibility criteria for the study were: age ≥ 18 years; PS: 0-2; chemotherapy at disease progression after RT/TMZ; and availability of data regarding second progression. RESULTS: 232 patients with recurrent glioblastoma (mean age: 52 years, range: 18-77 years, MGMT methylated/ unmethylated: 62 [37.6%]/103 [62.4%]) were evaluated. At progression after RT/TMZ, 102 patients (44%) were treated with re-surgery followed by chemotherapy, and 130 patients (56%) with chemotherapy alone. Overall survival from first surgery was 22.4 months (95% CI: 20 - 24.7), 25.8 months (95% CI: 20.6-31) in patients who received second surgery at recurrence, and 18.6 months (95%CI: 17-20.1; p ¼ 0.003) in patients treated without surgery. However, in multivariate analysis, no significant effect of re-surgery was found (p ¼ 0.11), age (p ¼ 0.001), MGMT methylation (p ¼ 0.002), and PFS6 (p ¼ 0.0001) being the only significant prognostic factors. Moreover, median time between first and second surgery was 13.1 months, being significantly longer in patients with MGMT methylated than in patients with MGMT unmethylated (19.3 vs. 13 months, p ¼ 0.001). Median survival time calculated from first recurrence was 8.6 months (95% CI: 7.4-9.8), and 9.6 months (95% CI: 7.5-11.6) and 7.5 months (95% CI: 5.7-9.3) in patients who received second surgery or not, respectively (p ¼ 0.3). CONCLUSIONS: Our data suggested that second surgery may have a limited impact in the clinical course of recurrent glioblastoma patients. MGMT methylation status, as well other clinical factors (ie, age), remain the major prognostic determinants of the outcome. NS-08. PREDICTION OF SURVIVAL AFTER SURGICAL RESECTION OF GLIOBLASTOMA: IMPACT OF PRE-OPERATIVE AND RESIDUAL TUMOR VOLUME Pablo F. Recinos, Matthew M. Grabowski, Amy S. Nowacki, Nicolas Thompson, and Michael A. Vogelbaum; Cleveland Clinic, Cleveland, OH INTRODUCTION: Extent of resection (EOR) previously has been associated with length of survival in patients with glioblastoma (GBM). However, the role of post-surgical residual tumor volume (RTV) on clinical outcome is not clearly understood. We hypothesized that the pathobiology of GBM is primarily dependent on RTV, rather than EOR, and sought to evaluate the effect of RTV on overall survival with use of a new volumetric analysis tool. METHODS: A retrospective analysis of patients who underwent primary surgical resection for a supratentorial GBM followed by standard radiation and chemotherapy at Cleveland Clinic between 2003 and2011 was performed. Patients were excluded if they were ,18 years old, had preoperative chemotherapy/radiation therapy, had other intracranial malignancies, could not undergo MRI, or if they did not have post-operative imaging within 48 hours. A quantitative, semi-automated volumetric analysis (BrainLab iPlan software) tool was used to measure the pre-operative tumor volume (PTV) and RTV. Clinical and outcome data were collected and compared to PTV, EOR, and RTV. RESULTS: There were 133 patients identified who met the study entry criteria. The median age was 60 years (range, 23-86 years), and the median Karnofsky Performance Status (KPS) at presentation was 90 (range, 30-100). The median overall survival was 14 months (range, 1-84.2 months). In a multivariate Cox proportional hazards analysis, age (p ¼ 0.0099), KPS (p ¼ 0.0004), and RTV (p ¼ 0.0387) were significantly associated with overall survival. EOR was not a statistically significant predictor of survival. Given that significant colinearity was seen between PTV and RTV, a separate multivariate model without RTV demonstrated that PTV also significantly predicted survival (p ¼ 0.0117). CONCLUSION: Both PTV and RTV are predictive of survival in patients with GBM. In addition, RTV may be a more sensitive and meaningful predictor of survival than EOR. NS-07. DOES RE-SURGERY HAVE A ROLE IN RECURRENT GLIOBLASTOMA TREATMENT? Alba A. Brandes1, Mario Ermani2, Antonio Fioravanti3, Alvaro Andreoli3, Eugenio Pozzati3, Antonella Bacci4, Stefania Bartolini1, Rosalba Poggi1, Girolamo Crisi5, and Enrico Franceschi1; 1Medical Oncology Department, Bellaria-Maggiore Hospital, Bologna, Italy; 2Department of Neurosciences, Statistic and Informatic Unit, Azienda Ospedale-Universita` , Padova, Italy; 3Neurosurgery Department, Bellaria-Maggiore Hospital, Bologna, Italy; 4Neuroradiology Department, Bellaria-Maggiore Hospital, Bologna, Italy; 5Neuroradiology Department, Azienda Ospedaliero-Universitaria, Parma, Italy NS-09. DIRECT MEDICAL COSTS FOR PATIENTS WITH TUBEROUS SCLEROSIS COMPLEX AND SURGICAL RESECTION OF SUBEPENDYMAL GIANT CELL ASTROCYTOMA Peter Sun1, Darcy Krueger2, Zhimei Liu3, and Michael Kohrman4; 1Kailo Research Group, Fishers, IN; 2University of Cincinnati, Cincinnati, OH; 3Novartis Pharmaceutical Corporation, East Hanover, NJ; 4University of Chicago, Chicago, IL INTRODUCTION: Previous studies demonstrated that surgical resection of subependymal giant cell astrocytoma (SEGA) was associated with repeated SEGA surgery, surgical complications, and increased prevalence of SEGA-related clinical conditions, including seizure, hydrocephalus, headache, stroke, or hemiparesis in a real-world setting. This study examines the direct medical costs among patients with tuberous sclerosis complex and SEGA surgery. METHODS: Three large US national healthcare claims databases (2000-2009) were used with a single cohort panel study design. Patients with health insurance coverage one year before and one year after their first observed SEGA surgery at age 35 or younger were included in the study. Average inpatient, outpatient, medication, and total medical costs per patient without inflation adjustment were estimated for six periods either before or after the surgery. Cross-period cost differences as well as annual cost differences before and after the surgery were assessed using repeated measures analysis and bootstrapping methods to adjust for potential biases from repeated measures, small sample size, and abnormal distribution of medical costs. RESULTS: The identified patients (N ¼ 47) had a mean age of 11.6 years at their first observed SEGA surgery, and 66% were male. The average total medical costs per patient per year were $8,543.1 (inpatient: $3,770.3; outpatient: $3,472.9; medication: $1,300.0) in the pre-surgery year and $85,397.0 (inpatient: $71,562.1; outpatient: $11,497.2; medication: $2,337.7) in the post-surgery year. Pre-post cost differences for inpatient ($67,791.8), outpatient ($8,024.3), medication ($1,037.7), and total medical costs ($76,853.9) were all statistically significant at p , 0.05. Most medical cost increases occurred within the first half post-surgery year; however, the medical costs in the second half post-surgery year were still significantly higher than those in the second half pre-surgery year. CONCLUSIONS: In the real world, in addition to clinical consequences, SEGA surgery was associated with significant increases in inpatient, outpatient, medication, and total medical costs. NS-10. 5-AMINOLEVULINIC ACID GUIDED RESECTION OF HIGH GRADE GLIOMA Andrew B. Dagens; University of Glasgow, Glasgow, United Kingdom Each year in the United Kingdom, 1860 cases of high-grade glioma are diagnosed. The prognosis for newly diagnosed high-grade glioma remains very poor. Median survival for anaplastic astrocytoma is 3 years; for glioblastoma multiforme, median survival is 1 year. There is no curative therapy for highgrade glioma. Treatment strategy is surgical resection followed by adjuvant radiotherapy. Research has suggested that maximal resection of tumour leads to increased survival time, yet complete resection is technically difficult due to the diffuse nature of the tumour and the risk of incurring neurological damage. 5-aminolevulinic acid (‘gliolan’) is used in porphyrin synthesis, and pre-operative administration leads to the accumulation of fluorescent porphyrins, which under ultraviolet light can be used to guide resection. We report preliminary results from a cohort of patients with high-grade glioma who underwent 5-ala-guided resection in order to assess the safety and efficacy of using 5-ala. Nine patients underwent gliolan-guided resection between 2009 and 2012 at the INS in Glasgow. For 8 patients this was the first resection. Each patient was given oral gliolan 1 hour before induction of anaesthesia. CT-guided craniotomy under ultraviolet microscopy was performed. Each patient received adjuvant radiotherapy. Surveillance imaging was performed 1 month postoperatively. Measured outcomes included percent of tumour resected and complications of the procedure. Complete resection was achieved in 60% of cases. No neurological defects were incurred as a result of the procedure. There were no complications from the dye or from the procedure. Gliolan-guided resection could potentially help the resection of gliomas safely. The impact on survival and time to progression are unclear NS-11. CYSTIC CRANIOPHARYNGIOMAS: MICROSURGICAL OR STEREOTACTIC TREATMENT? Walter Rachinger, Mathias Kunz, Sabina Eigenbrod, J u¨rgen Lutz, Jo¨ rg-Christian Tonn, and Friedrich-Wilhelm Kreth; Klinikum Großhadern, Munich, Germany OBJECTIVE: The value of microsurgical and stereotactic treatment particularly in cystic craniopharyngiomas is poorly defined. We present outcome data in terms of functional results and progression-free survival (PFS) after individualized treatment strategies. Solid tumors were considered eligible for open tumor resection, whereas predominantly cystic lesions were treated by stereotactic catheter placement for internal drainage. The target point for catheter implantation was the prepontine cisterne, thereby enabling drainage in both directions the ventricular space (upstream) and the prepontine cisterne (downstream). METHODS: Patients with craniopharyngioma treated between 2001 and 2010 either by open tumor resection or stereotactic drainage were analyzed retrospectively. Reference point of the study was the date of surgical treatment. PFS was estimated with the Kaplan Meier method; prognostic factors were obtained from multivariate regression vi88 models including all relevant patient-, radiological-, and functional-related data. Adjuvant radiation therapy was used as second-line treatment in case of unresectable recurrences. RESULTS: 68 consecutively treated patients (median age: 53 years) were included. Patients of the microsurgical (N ¼ 44) and stereotactic (N ¼ 24) groups did not differ in terms of age, gender, tumor size, functional deficits, and duration of symptoms. Complete tumor resection and complete cyst drainage was obtained in 23/44 and 21/24 patients, respectively. Median PFS after surgical and stereotactic treatment was 51 months and 46 months, respectively (p ¼ 0.1). Endocrinological outcome was significantly better after stereotactic treatment (p ¼ 0.004). The overall transient complication rate was 4.4 %, being lower in the stereotactic group (8.1% vs. 0 %, p , 0.05). However, one patient of the stereotactic group experienced a space occupying bleeding leading to permanent morbidity (overall permanent morbidity rate: 1.5 %). CONCLUSIONS: Stereotactic treatment of cystic craniopharyngiomas is effective and safe. This hypothesis requires further investigation in a prospective randomized study for cystic craniopharyngiomas. NS-12. MELANOMA BRAIN METASTASES TREATED WITH SURGERY AND MULTIMODALITY THERAPY Huy T. Duong1, Charlene Chaloner1, Gal Bordo1, Amy Eisenberg1, Katherine Rosenthal2, Myung-Shin Sim1, Peter Boasberg2, Mark B. Faries1, Omid Hamid2, and Daniel F. Kelly1; 1John Wayne Cancer Institute at Saint John’s Health Center, Santa Monica, CA; 2The Angeles Clinic, Santa Monica, CA BACKGROUND: Control of brain metastases in patients with malignant melanoma remains a challenge. Herein we present a series of patients treated with surgical resection of symptomatic metastases. METHOD: Over 4.5 years, all consecutive patients with symptomatic brain melanoma metastases who had craniotomy and tumor resection were included. All patients had adjuvant radiotherapy with chemotherapy or immunotherapy. Resection rates, neurological recovery, survival rates using Kaplan-Meier estimates, and complications were noted. RESULTS: The cohort consisted of 40 patients (31 males, 9 females; mean age 59 years + 13 years). At time of surgery, 15 patients had 1 metastasis, 11 patients had 2-3, and 14 patients had .3 metastases. Eight patients (20%) had prior radiation (6 SRS, 2 WBRT); 2 had prior craniotomies. Patients underwent 48 surgeries for resection of 57 metastases (9 had 2 metastases removed at same operation). Mean maximal tumor diameter was 30 + 12 mm. Gross total, near total, and subtotal resection was achieved in 82%, 16%, and 2% of metastases, respectively. Five patients had 6 re-resections for recurrent disease. Of 30 patients who presented with focal neurological deficits, seizures, or headaches, 27 (90%) had improvement or resolution of these symptoms or deficits. One patient had post-operative monoparesis; 1 had fatal pulmonary embolus 4 days after uneventful surgery. There were no postoperative hematomas, CSF leaks, or vascular injuries. Median hospital stay was 3 days (range, 2-14). Of 39 (98%) patients with survival data, median survival was 8 months; the 12-month and 24-month survival rates were 43% and 20%, respectively. CONCLUSION: Surgery for symptomatic melanoma brain metastases generally has a favorable short-term outcome with significant improvement in neurologic status and headache. While overall survival remains poor, there are a minority of long-term survivors. Prognostic factors related to AJCC stage and impact of adjuvant therapies are discussed. NS-13. COMPLETE, BUT NOT PARTIAL RESECTION IS IMPORTANT IN THE ERA OF RADIOTHERAPY, CONCOMITANT AND ADJUVANT TEMOZOLOMIDE FOR GLIOBLASTOMA PATIENTS Friedrich-Wilhelm Kreth1, Niklas Thon1, Matthias Simon2, Manfred Westphal3, Gabriele Schackert4, Guido Nikkhah5, Bettina Hentschel6, Torsten Pietsch7, Guido Reifenberger8, Michael Weller9, and J o¨rg-Christian Tonn1; 1Department of Neurosurgery at the University of Munich, Munich, Germany; 2Department of Neurosurgery at the University of Bonn, Bonn, Germany; 3Department of Neurosurgery at the University of Hamburg, Hamburg, Germany; 4Department of Neurosurgery at the University of Dresden, Dresden, Germany; 5Department of Stereotactic Neurosurgery at the University of Freiburg, Freiburg, Germany; 6Institute of Medical Informatics, Statistics and Epidemiology at the University of Leipzig, Leipzig, Germany; 7Department of Neuropathology at the University of Bonn, Bonn, Germany; 8Department of Neuropathology at the University of Du¨ sseldorf, Du¨ sseldorf, Germany; 9Department of Neurology at the University of Z u¨rich, Zu¨ rich, Switzerland Resection followed by radiotherapy (RT) plus concomitant and adjuvant temozolomide (TMZ) is considered standard of care for glioblastoma. Treatment, however, is highly variable and the benefit of partial resection remains unclear. This multicenter observational study was conducted to assess both current treatment strategies and prognostic factors, including extent of resection and MGMT promoter methylation in patients with newly diagnosed glioblastoma. Histology, MGMT promoter methylation status, and all clinical data were centrally analysed. Survival analyses were performed with the Kaplan-Meier method. Endpoints were overall survival (OS) and progression-free survival (PFS). Prognostic factors were assessed with proportional hazards models. Overall, 326 patients were included between 2006 and 2010. Postoperative MRI revealed 36.8% complete and 40.8% partial resections. 22.4% underwent biopsy. Surgery-related morbidity was lower after biopsy (1.4% vs. 12.3%, p ¼ 0.003). 64.1% of patients received RT + TMZ, 19.6% RT, 4.6% TMZ, and 11.7% supportive care. Patients ≤ 60years and KPS ≥ 90 were more likely to receive open tumor resection and RT + TMZ (p , 0.01). Overall, median PFS and OS were 6.5 months (95% CI, 6.0-7.0) and 12.9 months (95% CI, 11.2-14.5). Median OS (PFS) was as high as 28.1 (13.8) months for patients with MGMT methylated tumors after complete resection followed by RT + TMZ to as low as 3.0 (2.4) months for biopsied patients receiving supportive treatment only. Favorable prognostic factors for OS were a methylated MGMT promoter (RR ¼ 0.43; p , 0.001), RT + TMZ treatment (RR ¼ 0.18, p , 0.001), age ≤ 60 years (RR ¼ 0.53; p , 0.001), a KPS ≥ 70 (RR ¼ 0.40; p , 0.001), and complete resection (RR ¼ 0.59; p ¼ 0.002). Incomplete resection was as good as biopsy only, both in all patients and in a large subgroup of 209 patients receiving RT + TMZ. MGMT methylation and complete resection continued to be of independent prognostic importance. Our study confirms superior OS in case of MGMT promoter methylation, RT + TMZ treatment, and complete resection. Incomplete resection does not significantly improve OS as compared to biopsy NS-14. FACTORS INFLUENCING OVERALL SURVIVAL SPECIFIC TO ADULT LOW-GRADE ASTROCYTOMA: A POPULATION-BASED STUDY Sarah Ironside1, James Perry2, May Tsao2, Todd Mainprize2, Julia Keith2, Norman Laperrierre3, Lawrence Paszat2, and Arjun Sahgal2; 1University of Toronto, Toronto, ON, Canada; 2Sunnybrook Health Sciences Centre, Toronto, ON, Canada; 3Princess Margaret Hospital, Toronto, ON, Canada BACKGROUND: We report a population-based overall survival and prognostic factor analysis specific to adult patients with WHO grade 2 astrocytoma. METHODS: Electronic medical records of all histologically confirmed cases of grade 2 astrocytoma (not centrally reviewed) diagnosed between 1992 and 1996 in the province of Ontario, Canada, were identified from the Ontario Cancer Registry (OCR). Baseline patient and tumor characteristics and cause/date of death were summarized. Univariate and multivariate analyses were used to assess the prognostic association of age, gender, astrocytoma sub-type, year of diagnosis, and initial surgical management (excision vs. biopsy) on patient survival using the Cox proportional hazards model. RESULTS: 182 patients met study inclusion criteria (mean age, 50 years, SD ¼ 17 years). Mean survival time was 4.1 years (SD ¼ 5.1 years). Overall survival was inversely associated with increasing age. 54% of patients had a surgical resection and 46% were biopsied alone. 73% of patients had a consultation with a radiation oncologist (ROnc). Univariate and multivariate analyses confirmed that patients age ,30 (n ¼ 27) were significantly more likely to undergo an excision as compared to a biopsy alone (OR ¼ 4.26, 95% CI, 1.54-11.77), and a consult with a ROnc (OR ¼ 4.85, 95% CI, 1.27-18.54). For those having had a tumor excision, patients younger than age 50 had a lower hazard of dying compared to older patients, and the HR for patients age 50-59 was 3.30 (95% CI, 1.4-6.53) and for age ≥ 60 was 5.26 (95% CI, 2.56-10.82). A similar association for those biopsied alone was not observed. Gender and year of diagnosis were not prognostic for overall survival. CONCLUSIONS: Patients younger than age 60 had a favorable overall survival as compared to older patients with grade 2 astrocytoma. In patients having had a tumor resection, age ,50 was associated with a longer overall survival; however, a similar association was not observed when patients were biopsied alone. NS-15. COMPLICATIONS AND OUTCOMES IN RECURRENT GLIOMA SURGERY Jason M. Hoover, Macaulay Nwojo, Ross Puffer, and Ian F. Parney; Mayo Clinic, Rochester, MN In order to assess perioperative outcomes and complications in recurrent intracranial glioma surgery, we retrospectively reviewed all adult intracranial glioma patients undergoing more than 1 surgery (biopsy or resection) at our institution between 1995 and 2010. Demographics, type of surgery, adjuvant therapies, neurological complications, regional (wound) complications, and systemic complications were recorded along with Karnofsky Performance Score (KPS). Odds ratios were determined with each subsequent surgery relative to the first surgery. A total of 131 patients (90 men, 41 women) underwent 2-5 surgeries. The majority underwent resection (74%) as the initial procedure. Neurological complications occurred in 4.8% at first surgery, 12.1% at second (OR ¼ 2.7; P ¼ 0.0437), 8.2% at third (OR ¼ 1.75; P ¼ NS), and 11.1% at 4 or more surgeries (OR ¼ 2.4583; P ¼ NS). Regional complications occurred in 6.2% after first surgery, 9.9% after second surgery (OR ¼ 2.30; P ¼ 0.095), 13.7% after third surgery (OR ¼ 3.31; P ¼ 0.015), and 22.2% after 4 or more surgeries (OR ¼ 5.95; P ¼ 0.056). Systemic complications occurred in 3.2% after first surgery, in 7.3% after second surgery (OR ¼ 2.3; P ¼ 0.1399), in 4.1% after third surgery (OR ¼ 1.3; P ¼ NS), and 0% after 4 or more surgeries. Reduced KPS occurred in 0% after first surgery, 8.1% after second surgery (OR ¼ 3.13; P ¼ 0.0018), 10.2% after third surgery (OR ¼ 5.52; P , 0.0001), and 11.1% after 4 or more surgeries (OR ¼ 1.037; P ¼ NS). We conclude that complication risk increases with multiple glioma surgeries. The largest increase in neurological risk occurs between the first and second surgery. In contrast, regional complication risk increases consistently with each procedure. Systemic complication risk is not significantly altered with increasing surgeries. Importantly, increasing complications with recurrent surgery only result in a modestly increased risk of functional decline. Repeated surgeries for gliomas remain a viable option but must be undertaken with an accurate assessment of the risks involved. These findings may help counsel patients considering multiple glioma surgeries. NS-16. THE IMPACT OF GLIOBLASTOMA RESECTION ON PATIENT’S SURVIVAL Shingo Tanaka, Mitsutoshi Nakada, Yutaka Hayashi, and Jun-ichiro Hamada; Kanazawa University, Kanazawa, Japan It remains to be elucidated whether the extent of glioblastoma (GBM) resection is associated with patient survival, although the extent of surgical resection tends to improve survival in several literatures. We assessed whether the surgical extent of GBM relates to progression-free survival (PFS) and overall survival (OS) in our institute. We retrospectively reviewed 62 patients with histologically proven GBM who underwent tumor resection at our institute between January 1997 and January 2012. Tumor removal rate was calculated by pre- and postoperative gadolinium-enhanced magnetic resonance imaging (MRI). We analyzed the following items: 1) relation between surgical extent and PFS or OS using the Kaplan-Meier method, and 2) multivariate analysis of prognostic factors for PFS and OS; surgical extent, temozolomide (TMZ), MGMT promoter methylation status, Karnofsky performance status, and patient age by the Cox proportional hazard model. When the extent of resection is over 90% of the tumor volume, PFS and OS are significantly prolonged (P ¼ 0.04, 0.003). For PFS, independent prognostic factors were MGMT promoter methylation (P ¼ 0.001) and TMZ therapy (P ¼ 0.004). For OS, independent prognostic factors were MGMT promoter methylation (P ¼ 0.006), TMZ therapy (P ¼ 0.007), surgical extent more than 90% (P ¼ 0.02), and age less than 70 years old (P ¼ 0.01). It might be expected that survival time would be prolonged when the surgical extent exceeds 90% of enhanced lesions on MRI. NS-17. INTRAOPERATIVE MRI IMPROVES BRAIN TUMOR EXTENT OF RESECTION: HENRY FORD HOSPITAL INITIAL EXPERIENCE Ian Y. Lee, Tafsheen Ekram, Rajan Jain, Lisa Scarpace, Melvin Omodon, Jack Rock, Mark Rosenblum, and Steven Kalkanis; Henry Ford Hospital, Detroit, MI Maximal safe resection is the goal of brain tumor surgery. Recently, intraoperative magnetic resonance imaging (iMRI) has been used at brain tumor centers to improve tumor resection. This study determines the initial impact of iMRI on brain tumor resection and operative time. Seventeen patients who underwent craniotomy for intraaxial tumors (16 gliomas, 1 metastases) with iMRI between 2/2011 and 7/2011 were retrospectively evaluated for tumor volumes using post-contrast T1-weighted imaging at three time points: prior to resection, intra-operatively following first resection, and after final resection. Clinical and neurological outcomes were recorded. In addition, surgeons were polled contemporaneously about the utility of the iMRI. Mean tumor reduction prior to intraoperative imaging was 72% (range, 0-100%, mean volume reduction, 13.6 ml). In two cases, small residual enhancing nodules were identified by iMRI and were resected subsequently. Fourteen patients (83%) underwent a post-iMRI resection, which resulted in further additional average tumor volume reduction of 25% (range, 3-100%, mean volume reduction, 2.6 ml). With the iMRI-assisted and second resection, patients with gross total resection (GTR) increased from 24% to 47%, and with only 6% of patients having a partial resection (,50%). Overall, combining GTR and Majority Resection (90-99%), which are the clinically significant extents, the percentage improved from 35% to 76%. Operative time was minimally increased, with an average iMRI scan time of 22 minutes. None of the patients experienced neurological deficits or complications attributable to the second resection. In 57%, the surgeon felt the iMRI improved the extent of resection. In conclusion, iMRI improved safe resection of brain tumors in a majority of cases without significantly prolonging operative time. Real-time ability to identifying residual enhancing tumor after the first resection which improved the total tumor resection volume and percentage of patients with clinically significant tumor resection. NS-18. THE USE OF INTRA-OPERATIVE MRI DURING AWAKE CRANIOTOMIES FOR GLIOMAS IN ELOQUENT REGIONS: AN EXTENT OF RESECTION ANALYSIS Nduka M. Amankulor, Joon-Hyung Kim, Viviane Tabar, Kyung K. Peck, Andrei I. Holodny, and Philip H. Gutin; Memorial Sloan-Kettering Cancer Center, New York, NY INTRODUCTION: There has been increased utilization of intra-operative imaging during glioma surgery over the last decade. Solid evidence exists to support the use of intra-operative ultrasonography and intra-operative MRI (ioMRI) as tools to achieve greater extent of resection (EOR) for gliomas in non-eloquent brain. For gliomas in eloquent brain regions, however, there is far less evidence to compel the use of intra-operative imaging. METHODS: 1) We performed a retrospective analysis of EOR in 60 consecutive patients with gliomas affecting eloquent motor/language cortices and in whom ioMRI was used for tumor resection (biopsy excluded). We compared these results to historical controls at MSKCC who underwent identical procedures in the years immediately preceding our institution’s purchase of ioMRI. 2) 3D volumetric tumor measurements were performed by a radiologist (A.H.) and a neurosurgeon (N.A.) who were blinded to patient’s identifiers. 3) Statistical analysis was performed using the Wilcoxon rank-sum (2-tailed) and chi-squared tests where appropriate. RESULTS: A total of 23 low-grade gliomas (LGGs) and 37 high-grade gliomas (HGGs) were analyzed and compared to 35 historical matched controls. A statistically significant improvement in EOR was noted in HGG patients for whom ioMRI was used (p , 0.05). There was also a significantly higher proportion of HGG ioMRI patients for whom EOR greater than 95% was achieved compared to controls. While there was a trend toward a higher EOR% in LGG patients undergoing resection in the ioMRI setting compared to controls, these results were not significant. CONCLUSION: We present the first large series of ioMRI use during awake craniotomy for glioma surgery. Remarkably, our results parallel the data from ioMRI use in non-eloquent glioma. Furthermore, we show that ioMRI likely improves EOR in patients undergoing awake craniotomy for glioma resection, especially in the context of HGG. NS-19. MANAGEMENT OF LATERAL VENTRICLE MENINGIOMAS: SURGICAL APPROACHES AND OUTCOMES Chae-Yong Kim1, Young-Hoon Kim1, Tackeun Kim1, In Kyung Kim2, Jin Wook Kim2, Yong Hwy Kim3, Jung Ho Han1, Chul-Kee Park2, Dong Gyu Kim2, and Hee-Won Jung2; 1Seoul National University Bundang Hospital, Seongnam-si, Republic of Korea; 2Seoul National University Hospital, Seoul, Republic of Korea; 3Seoul National University Hospital, Seongnam-si, Republic of Korea BACKGROUND: Meningiomas located in lateral ventricle (LVMNGs) are notably rare but surgically challenging intracranial tumors. This retrospective study was performed to analyze the clinico-radiological findings, surgical outcomes, and postoperative neurological results of the surgically treated LVMNGs. METHODS: Of the 3856 patients with surgically treated MNGs, 41 patients (1.1%) and 23 patients (0.6%) underwent microsurgery and radiosurgery as a primary treatment, respectively, for LVMNGs between 1991 and 2011 in our 2 institutions. The mean patient age was 46 years. The mean diameter of 41 tumors with microsurgery was 5.1 cm. Of the 41 tumors, the transcortical approach was used in 39 patients (95%) via temporal (21 patients; 51%), perieto-occipital (17 patients; 41%), or frontal route (1 patient; 2%). The mean size and marginal dose of 23 tumors with radiosurgery were 2.6 cm and 14.7 Gy. The median follow-up duration was 55.1 months. RESULTS: Gross or near-total removal was achieved in 36 patients (88%). Nineteen patients (46%) experienced postoperative neurological deficits including 7 hemipareses, 6 visual field defects, and 4 language problems. One patient died of postoperative hemorrhagic complication. The 1-, 5-, and 10-year recurrence-free survival rates were 100, 84, and 66%, respectively. Left-side tumor was the only significant risk factor for postoperative neurological deficit (p ¼ 0.019; OR ¼ 7.7), and trigonal tumor (p ¼ 0.061) and larger tumor (. 5cm) (p ¼ 0.071) were relative factors in the multivariate analysis. Surgical approach did not show statistical significance. The clinical and radiological control rates following radiosurgery for LVMNGs were 100% and 96%, respectively. CONCLUSIONS: LVMNGs were surgically challenging tumors and vulnerable to postoperative neurological risks. A tailored multidisciplinary treatment protocol and individualized surgical approach is needed for improving treatment outcomes. NS-20. OVERCOMING BRAIN SHIFT DURING NEURONAVIGATION-GUIDED GLIOMA SURGERY Masahiro Nonaka, Yohei Bamba, Yonehiro Kanemura, and Shin Nakajima; Osaka National Hospital, Osaka, Japan Recent development of imaging technology enables visualization of information about tumor and neuroanatomy; for example, metabolic images of glioma obtained from PET and neural fibers acquired from diffusion tensor images. Many of these images are possible to integrate to navigation system. However, by dural opening and by surgical resection of the tumor, induce brain shift, which makes preoperative image obtained for navigation unreliable. To overcome this problem, we developed implantable marker probes made from shunt tubes. These markers were embedded into the planned target under navigation guidance immediately after craniotomy before brain shift and were considered as the landmark for the tumor resection. The purpose of this study is to assess the usefulness of this technique by evaluating resected tumor volume and perioperative performance status. From 2009, 45 glioma operations were carried out using this method. Brain mapping or motor evoked potential were also used in some cases. Resection rates of Gd-enhanced areas were used for evaluation for tumors with enhanced lesion (n ¼ 25, group A) and resection rates of FLAIR-high signal areas were used for nonenhanced lesions (n ¼ 20, group B). Preoperative and postoperative performance status were measured based on Karnofsky performance scale (KPS). Mean resection rate of the group A was 97.6%. Mean preoperative KPS was 68.4%, and postoperative KPS showed improvement up to 76.4% (p , 0.05) in group A. Mean resection rate for group B was 84.4%; mean preoperative KPS was 95.0% while postoperative KPS was 91.8% in group B. There was no procedure-related complication in either group. The marker insertion under navigation guidance is a safe method. Using this method, higher resection rate with improvement of performance status may be achieved.

This is a preview of a remote PDF:

Allan D. Nanney III, Joseph G. Adel, Timothy R. Smith, James P. Chandler, Kristopher T. Kimmell, Kevin Walter, Brad E. Zacharia, Christopher Deibert, Hani R. Malone, Adam M. Sonabend, Alfred I. Neugut, Benjamin Spencer, Jeffrey N. Bruce, Yongzhi Wang, Shouwei Li, Zhong Zhang, Xuzhu Chen, Gan You, Pei Yang, Wei Yan, Zhaoshi Bao, Kun Yao, Yanwei Liu, Lei Wang, Tao Jiang, Mohamed K. Farhoud, Maximilian I. Ruge, Alba A. Brandes, Mario Ermani, Antonio Fioravanti, Alvaro Andreoli, Eugenio Pozzati, Antonella Bacci, Stefania Bartolini, Rosalba Poggi, Girolamo Crisi, Enrico Franceschi, Pablo F. Recinos, Matthew M. Grabowski, Amy S. Nowacki, Nicolas Thompson, Michael A. Vogelbaum, Peter Sun, Darcy Krueger, Zhimei Liu, Michael Kohrman, Andrew B. Dagens, Walter Rachinger, Mathias Kunz, Sabina Eigenbrod, Jürgen Lutz, Jörg-Christian Tonn, Friedrich-Wilhelm Kreth, Huy T. Duong, Charlene Chaloner, Gal Bordo, Amy Eisenberg, Katherine Rosenthal, Myung-Shin Sim, Peter Boasberg, Mark B. Faries, Omid Hamid, Daniel F. Kelly, Friedrich-Wilhelm Kreth, Niklas Thon, Matthias Simon, Manfred Westphal, Gabriele Schackert, Guido Nikkhah, Bettina Hentschel, Torsten Pietsch, Guido Reifenberger, Michael Weller, Jörg-Christian Tonn, Sarah Ironside, James Perry, May Tsao, Todd Mainprize, Julia Keith, Norman Laperrierre, Lawrence Paszat, Arjun Sahgal, Jason M. Hoover, Macaulay Nwojo, Ross Puffer, Ian F. Parney, Shingo Tanaka, Mitsutoshi Nakada, Yutaka Hayashi, Jun-ichiro Hamada, Ian Y. Lee, Tafsheen Ekram, Rajan Jain, Lisa Scarpace, Melvin Omodon, Jack Rock, Mark Rosenblum, Steven Kalkanis, Nduka M. Amankulor, Joon-Hyung Kim, Viviane Tabar, Kyung K. Peck, Andrei I. Holodny, Philip H. Gutin, Chae-Yong Kim, Young-Hoon Kim, Tackeun Kim, In Kyung Kim, Jin Wook Kim, Yong Hwy Kim, Jung Ho Han, Chul-Kee Park, Dong Gyu Kim, Hee-Won Jung, Masahiro Nonaka, Yohei Bamba, Yonehiro Kanemura, Shin Nakajima. NEUROSURGICAL TREATMENTS, Neuro-Oncology, 2012, vi86-vi90, DOI: 10.1093/neuonc/nos230