Clinical Factors Predicting Outcomes After Surgical Resection for Sporadic Cerebellar Hemangioblastomas

Transcription

1 Clinical Factors Predicting Outcomes After Surgical Resection for Sporadic Cerebellar Hemangioblastomas Masafumi Fukuda, Tetsuro Takao, Tetsuya Hiraishi, Junichi Yoshimura, Naoki Yajima, Akihiko Saito, Yukihiko Fujii Key words - Cerebellar hemangioblastoma - Recurrence - Solid tumor - Surgery Abbreviations and Acronyms MRI: Magnetic resonance imaging VHL: Von Hippel-Lindau Department of Neurosurgery, Brain Research Institute, University of Niigata, Niigata-City, Japan To whom correspondence should be addressed: Masafumi Fukuda, M.D. [ mfuku529@bri.niigata-u.ac.jp] Citation: World Neurosurg. (2014) 82, 5: Journal homepage: Available online: /$ - see front matter ª 2014 Elsevier Inc. All rights reserved. INTRODUCTION Hemangioblastomas are benign neoplasms of the central nervous system that often occur in the posterior fossa and especially in the cerebellar hemispheres (2). Theymani- fest as highly vascular, solid, or cystic tumors. Hemangioblastomas of the brain may occur sporadically, but in about 20%e30% of cases, they are associated with von Hippel-Lindau (VHL) disease (14). Despite similar histologic characteristics, syndromic hemangioblastomas show systemic manifestations different from those of sporadic tumors. Hemangioblastomas associated with VHL disease may manifest with aggressive development through rapidly evolving or multifocal lesions and may have a strong effect on long-term patient outcomes (24). However, in some clinical studies, patients with sporadic hemangioblastomas and patients with hemangioblastomas associated with VHL disease are considered together as the sample (4, 20), and the studies specifically report the outcomes of syndromic hemangioblastomas (10, 14, 15, 17). In the present study, we retrospectively investigated the clinical and radiologic characteristics of patients who underwent - OBJECTIVE: To determine whether various clinical factors are related to longterm outcomes of patients with sporadic cerebellar hemangioblastomas. - METHODS: Subjects included 36 patients (19 men and 17 women) who underwent resection of sporadic hemangioblastoma in the cerebellum. Age at surgery ranged from 17e79 years (mean, 49.7 years). The tumor size, which was defined as the largest diameter of the lesion including the extratumoral cyst, ranged from 10e67 mm (mean, 36.4 mm). Obstructive hydrocephalus secondary to mass effect on the fourth ventricle was present in 21 (58.3%) patients preoperatively. - RESULTS: Total tumor removal was achieved in 31 of 36 patients (86%). In 4 (11%) patients with solid tumors, postoperative hematoma occurred in the removal cavity, and hematoma removal was required immediately after surgery. We followed 30 patients for >12 months after the initial surgery (mean, 72.9 months; range, 12e274 months). Recurrence of hemangioblastoma developed in 4 of 30 patients (13%) at 6 months, 17 months, 6 years, and 22 years after surgery. At the final follow-up examination, 9 (30%) of 30 patients showed some residual neurologic symptoms (poor group), whereas the remaining 21 patients showed no deficits (good group). Using univariate analysis, both age at surgery and tumor characteristics (cystic or solid) were significantly related to long-term patient outcomes (P < 0.05). However, in a multiple logistic regression analysis, only tumor characteristics were correlated with outcomes (P [ 0.017). At the final follow-up examination, patients with solid tumors more frequently showed poor outcomes than patients with cystic tumors. - CONCLUSIONS: The solid configuration observed on preoperative images of sporadic cerebellar hemangioblastomas is one of the most important clinical factors related to both immediate and long-term outcomes after surgery. surgical resection of sporadic cerebellar hemangioblastomas to determine the factors that can predict long-term outcomes. METHODS Patients The study included 36 patients (19 men and 17 women) with sporadic cerebellar hemangioblastomas who were treated surgically at the University of Niigata during the period March 1990eFebruary The criterion for VHL disease diagnosis in the absence of a family history of central nervous system or retinal hemangioblastoma was presence of at least 2 central nervous system or retinal hemangioblastomas or 1 hemangioblastoma and 1 of the following manifestations of VHL disease: renal carcinoma, pheochromocytoma, pancreatic cyst, or papillary cystadenoma of epididymis. In a patient with a family history of central nervous system or retinal hemangioblastoma, diagnosis of VHL syndrome required only the presence of 1 hemangioblastoma or any other manifestation of the VHL syndrome (4, 13). Patients who did not fulfill the aforementioned criteria were excluded from this study. Characteristics of patients are summarized in Table 1. Patient age at surgery ranged from 17e79 years (mean, 49.7 years). Mean duration between first symptoms and diagnosis was 9.7 months (range, 0.5e67 months). The presenting symptoms before operation are reported in Table 2. Radiologic Data Patients were classified as 1 of 2 types according to the tumor characteristics WORLD NEUROSURGERY 82 [5]: , NOVEMBER

2 Table 1. Characteristics of Patients with Sporadic Hemangioblastoma Patient Number Age (years)/sex History Duration (months) Tumor Size (mm) Tumor Characteristics (S or C) Tumor Location (P or A) Hydrocephalus (y or n) Surgery Postoperative Hemorrhage (y or n) Final Periods (months) Regrowth (y or n) Final Outcome 1 63/F 5 38 C P y Total n 64 n Good 2 70/M C P n Total n 62 n Poor 3 33/F 2 25 C A n Total n 110 n Good 4 28/M C P y Total n 55 n Good 5 17/M 1 40 C P y Total n 1 NA NA 6 69/F C P y Total n 96 n Good 7 36/M 3 40 C P y Total n 143 n Good 8 36/M 2 60 C P y Total n 24 n Good 9 33/F 2 40 C P n Total n 38 n Good 10 22/F 3 50 C P y Total n 85 n Good 11 30/F C A n Total n 58 n Good 12 62/F C A y Total n 46 n Good 13 41/M 1 33 C A y Total n 29 y Good 14 55/M 1 67 C A y Total n 17 n Good 15 66/M 2 50 C P n Total n 13 n Good 16 65/F C P n Total n 4 NA NA 17 30/F 4 50 C P y Total n 109 n Good 18 49/M 9 30 S A y Partial* y 274 y Poor 19 65/M S P n Total n Death NA NA 20 45/M 6 20 S A y Total n 1 NA NA 21 41/F 2 35 S P y Total n 88 n Good 22 60/M 2 40 S P n Total n 12 n Good 23 57/F 5 25 S A n Subtotal n 240 y Poor 24 61/F S P y Total n 70 n Poor 25 25/F 2 25 S A y Total n 4 NA NA 26 52/M 2 25 S A n Partial y y Death NA NA 27 33/F 2 40 S A y Total n 144 n Poor 28 59/F 3 40 S P y Total n 36 n Good 29 39/M 6 35 S A y Total n 96 n Good 30 61/F S P n Total n 122 n Good WORLD NEUROSURGERY,

3 31 43/M S P n Total n 47 n Good 32 64/M S P y Total y 47 n Poor 33 79/M 6 30 S A n Subtotal n 19 y Poor 34 70/M 5 53 S P n Total n 17 n Good 35 63/M S A n Subtotal y 13 n Poor 36 67/F 5 50 S A y Total n 12 n Poor F, female; M, male; S, solid; C, cyst; P, posterior; A, anterior; y, yes; n, no; NA, not available. *Total resection of residual tumors was performed with hematoma removal 2 months later. ytotal resection of residual tumors was performed with hematoma removal, simultaneously. determined by magnetic resonance imaging (MRI). The cystic tumor type was defined as having cysts with a mural nodule. The solid tumor type was defined as having no extratumoral cysts but included a partial intratumoral microcystic form. We assigned 17 patients to the cystic tumor group and 19 to the solid tumor group (Table 1). Tumor size was defined as the largest diameter of the lesion, including the extratumoral cyst, and ranged from 10e67 mm (mean, 36.4 mm 13.4) (Table 1). In patients with a cystic tumor, the mean mural nodule size was 13.2 mm (range, 5e23 mm). Patients were further assigned into anterior or posterior groups based on the location of the hemangioblastoma. In the anterior group, the tumor center was located anterior to the line linking the left and right posterior edges ofthejunctionbetweenthetransverseand sigmoid sinuses on axial MRI (Figure 1A). In the posterior group, the tumor was located posterior to this line (Figure 1B). In cystic tumors, the locations were defined as the center of the mural nodule (Figure 1C and D). Compared with tumors in the posterior group, tumors in the anterior group involved the cranial nerves or brainstem more frequently. The anterior group contained 15 patients, and the posterior group contained 21 patients (Table 1). Preoperatively, obstructive hydrocephalus secondary to mass effect over the fourth ventricle was present in 21 patients (58.3%) (Table 1). Surgical Procedures In 5 patients with solid tumor, preoperative embolization was performed, and the patients showed no complications after their procedures. Based on the tumor locations, 3 surgical approaches were considered for each patient. The lateral suboccipital approach was applied in 21 patients, the midline suboccipital approach was applied in 14 patients, and the occipital transtentorial approach was applied in 1 patient. All patients were placed in a lateral oblique position. In the lateral suboccipital approach, the head was fixed in a neutral position, whereas the face was rotated toward the floor in the midline suboccipital approach. In the occipital transtentorial approach, the head was fixed in a neutral position with the chin slightly elevated. In 5 patients with obstructive hydrocephalus, ventricular drainage was performed before tumor resection. Follow-up We determined the outcomes of patients examined with follow-up for >12 months after the initial surgery based on their medical records. These patients were assigned into 1 of 2 groups depending on whether they presented with some neurologic symptoms (poor group) or no signs (good group) at the final examinations. The symptoms included residual preoperative signs and new postoperative deficits. In addition, we determined whether the tumors recurred during the follow-up periods. Statistical Analysis Statistical analyses were performed using SPSS Statistics version 21 (IBM Corporation, Armonk, New York, USA). In patients with follow-up data, patient age, duration of history, preoperative tumor size, tumor types (cystic vs. solid tumor), tumor location (anterior vs. posterior location), and association of hydrocephalus (yes or no) were tested as independent factors in a final outcome analysis using both univariate and multivariate logistic regression analyses. The c 2, Fisher exact, or Mann-Whitney U test was used to compare factors between the good and poor groups. A P value < 0.05 was considered significant. RESULTS Surgical Results Total tumor removal (mural nodule or solid tumor) was achieved in 31 of 36 patients (86%). In 2 patients (patients 18 and 26 in Table 1) with solid tumors, only partial resection was possible because of severe tumor bleeding or acute intraoperative cerebellar swelling. The remaining 3 patients (patients 23, 33, and 35 in Table 1) with solid tumors in the anterior cerebellum underwent subtotal removal, in which small pieces of the tumors were left at locations adjacent to the lower cranial nerves. Histologic examination showed that all tumors were consistent with hemangioblastoma and revealed a rich vascular plexus with polygonal, lipid-laden stromal cells. There was no difference in the pathologic findings between solid tumors and mural nodules in cystic tumors. Short-Term Outcomes In 4 patients (patients 18, 26, 32, and 35 in Table 1) with solid tumor, postoperative WORLD NEUROSURGERY 82 [5]: , NOVEMBER

4 Table 2. Preoperative and Postoperative Symptoms of Patients with Cerebellar Hemangioblastomas Symptoms Preoperative Frequency (n [ 36) hematomas of the surgical bed were noted on computed tomography performed immediately after surgery. In 3 of the 4 patients who underwent subtotal or partial removal, hemorrhages appeared to be present from the residual tumors. For the 1 remaining patient (patient 32 in Table 1; Figure 2A and B), it was thought that the patient s large tumor with solid components had been completely removed during surgery. However, the patient experienced a sudden increase in blood pressure (170 mm Hg) followed by consciousness disturbance 3 hours after surgery; postoperative computed tomography scanning showed a massive hematoma in the removal cavity (Figure 2C). In all patients presenting with a hematoma, the removal was performed immediately. In 2 patients (patients 18 and 26) who underwent partial removal, resection of the residual tumors were performed with the hematoma removal, either simultaneously (patient 26) or 2 months later (patient 18). Postoperative deficits included hemiparesis in 3 patients (8%), swallowing disturbance in 3 patients (8%), cerebellar ataxia in 2 patients (6%), and sensory impairment in 2 patients (6%). In 3 patients (8%) with postoperative hematoma, postoperative ventriculoperitoneal shunt was required for hydrocephalus. Two patients (patients 26 and 19) died 1 month and 3 months after surgery of unrelated causes (acute myocardial infarction and bladder cancer). Postoperative Frequency (n [ 30) Present Preoperatively Present Postoperatively Headaches 18 (50%) Dizziness/vertigo 18 (50%) Nausea/vomiting 9 (25%) Cerebellar ataxia 8 (22%) 7 (23%) Diplopia 2 (6%) Visual disturbances 1 (3%) 1 (3%) Numbness in extremities 1 (3%) Dysphagia 3 (10%) Sensory impairments 1 (3%) Hearing disturbances 1 (3%) Regrowth of Tumor Further follow-up examinations were not performed in 3 patients (patients 5, 20, and 25) who were referred to other hospitals. Additionally, for 1 patient (patient 16), the follow-up period was too short (4 months after surgery) to evaluate long-term outcome. These 4 patients, in addition to the 2 patients who died, were excluded from the analysis of long-term outcomes. For the 30 patients observed with follow-up for >12 months after the initial surgery, mean follow-up was 72.9 months (range, 12e274 months). Regrowth of hemangioblastomas in the operative regions occurred in 4 of 30 patients (13%) at 6 months (patient 13), 17 months (patient 33), 6 years (patient 23), and 22 years (patient 18) after surgery. Subtotal removal was performed in 2 of 4 patients (patients 23 and 33), whereas complete tumor resection was performed in the remaining 2 patients (patients 13 and 18). Two patients needed a second operation at 1 year (patient 13) and 22 years (patient 18) after surgery, and Gamma Knife surgery was performed in 1 patient (patient 23) at 6.5 years after surgery. These patients showed no regrowth at their final examinations. The 1 remaining patient (patient 33) received no additional treatment because he was diagnosed with malignant lymphoma with poor prognosis on the basis of MRI, which revealed regrowth of his residual tumor. Long-Term Outcomes At the final examinations, 9 of 30 patients (30%) showed some neurologic symptoms (poor group). The symptoms were assigned into 2 categories depending on whether they 1) manifested preoperatively and remained or deteriorated postoperatively, or 2) they developed postoperatively (Table 2). The former symptoms included cerebellar ataxia with varying degree in 7 patients (23%) and visual loss in 1 patient (3%). The latter symptoms included dysphagia in 3 patients (10%), hearing disturbance in 1 patient (3%), and sensory impairment in 1 patient (3%). Our findings for univariate analysis of factors related to long-term outcomes are shown in Table 3. Patient age and tumor characteristics (solid or cystic) were related to the postoperative outcomes for cerebellar hemangioblastomas. Patient age (46.5 years 15.4) in the good group was significantly younger compared with the poor group (60.3 years 13.2, P < 0.05). The solid tumor rate in the good group was 33% (7 of 21 patients) but was 89% (8 of 9 patients) in the poor group (P < 0.05). Duration of history, preoperative tumor size, tumor location (anterior vs. posterior location), and hydrocephalus (yes or no) were not significant contributing factors. Results of the multivariate logistic regression analysis are summarized in Table 4. Patient age, tumor location, and tumor characteristics were associated with outcomes (the level of statistical significance determined with the log-likelihood ratiotestwassetatp < 0.1). In the stepwise multiple logistic regression analysis, only the tumor characteristics (solid or cystic) predicted the long-term outcomes after surgery (P ¼ 0.017; odds ratio, ; 95% confidence interval, 1.656e154.60). DISCUSSION In the present study, 4 (11%) patients with solid tumors among 36 patients with sporadic cerebellar hemangioblastomas exhibited postoperative hematoma in the removal cavity and required hematoma removal immediately after surgery. During follow-up examinations, 4 (13%) of 30 patients exhibited tumor regrowth. At the final examinations, 9 (30%) of 30 patients showed some neurologic symptoms remaining that had persisted from the preoperative period. Patients with solid tumors had poorer longterm outcomes than patients with cystic WORLD NEUROSURGERY,

5 Figure 1. Axial-enhanced T1-weighted magnetic resonance images demonstrating solid (A, B) and cystic (C, D) hemangioblastomas in the cerebellum. Patients were classified into 2 groups depending on whether the center of the tumors were anterior (A, C) or posterior (B, D) to the line linking the posterior edges of the junction of the transverse to the sigmoid sinus (dotted lines). tumors. These findings suggest that the solid configuration observed on preoperative images of sporadic cerebellar hemangioblastomas is one of the most important clinical factors related to both immediate and long-term outcomes after surgery. Postoperative Hemorrhages In previous reports on hemangioblastomas, the incidence of postoperative hemorrhages ranged from 3%e16%, with some patients dying as a result of massive postoperative bleeding (12, 14, 20, 25). One previous study showed that the incidence of patient deaths with solid tumors (5 of 14; 36%) was higher than that with cystic tumors (2 of 38; 5%) (25). In the present study, 4 of 36 (11%) patients presented with postoperative hematomas and underwent hematoma removal immediately after surgery. All 4 patients with postoperative hematomas showed solid configurations based on preoperative MRI. Even when the surgeon confirmed complete removal of a solid hemangioblastoma in 1 patient, postoperative hemorrhage developed immediately after surgery and rapidly increased blood pressure (Figure 2C). Blood pressure should be carefully controlled to avoid surges of hypertension in cerebellar hemangioblastomas with solid configurations, especially in the first 24 hours after surgery (25). Regrowth of Tumors TherecurrencerateinpatientswithVHL disease ranged from 17%e75% (4,6,15,23), Figure 2. Large solid hemangioblastoma in the right cerebellum revealed on imaging. (A) Axial-enhanced T1-weighted magnetic resonance image showing marked contrast enhancement of the tumor. (B) Digital subtraction angiography showing rich feeding arteries with staining of the tumor. (C) Axial postoperative computed tomography scan showing hematoma in the removal cavity. WORLD NEUROSURGERY 82 [5]: , NOVEMBER

6 Table 3. Characteristics of Patients with Long-Term Outcomes Characteristics whereas the recurrence rate in patients without VHL disease ranged from 5%e17% (4, 12, 15). Regrowth rates in our patients (4 of 30 patients; 13%) with sporadic cerebellar hemangioblastomas are consistent with rates in previous reports (4, 12, 15). Additionally, 1 of the 4 patients in the present study redeveloped hemangioblastomas 22 years after surgery; however, the residual tumor was totally resected with hematoma removal. Patients with sporadic hemangioblastomas as well as patients with VHL disease should be carefully monitored over long follow-up periods using MRI. Some authors reported that stereotactic radiosurgery for hemangioblastomas was associated with a higher tumor control rate (3, 11, 16, 18). In cases with relatively short periods after radiation, it is difficult to determine whether absence of growth in a tumor implied a quiescent phase of tumor growth or a positive response to the radiation therapy. We continue to follow 1 patient who underwent stereotactic radiosurgery for regrowth of hemangioblastoma 13 years after the treatment. Long-Term Outcomes In a previous study on long-term outcomes in patients with VHL disease and cerebellar hemangioblastomas, only 15% of the patients showed improvement after surgery. In contrast, 75% showed symptoms, and 10% died of VHL-related disease (10). Other authors showed that half of their patients with hemangioblastomas in the posterior fossa had some deficits at 5 years after the first operation (20). To our knowledge, there are no reports regarding long-term outcomes in patients having cerebellar hemangioblastomas without VHL. In the present study, the percentage of patients with some symptoms was 30%, with mean follow-up periods of about 6 years. A study showed that patients with intracranial sporadic hemangioblastomas experienced good quality of life as assessed Table 4. Relationship Between Long-Term Outcomes and Clinical Variables Variable P Value OR Lower 95% CI for OR Upper Age at surgery Duration of history Tumor location (anterior vs. posterior) Tumor size Tumor characteristics (cystic vs. solid) Preoperative hydrocephalus (yes vs. no) OR, odds ratio; CI, confidence interval. Outcomes Good (n [ 21; 70%) Poor (n [ 9; 30%) P Value Age at surgery (years)* y Duration of history (years)* Tumor location (anterior/posterior) 6/15 6/ Tumor size (mm)* Tumor characteristics (cystic/solid) 14/7 1/ y Preoperative hydrocephalus (yes/no) 13/8 5/ *Mean SD. ystatistically significant. by a questionnaire at a mean postoperative follow-up of 40 months (12). These findings indicate that patients with sporadic hemangioblastomas have better long-term outcomes than patients with hemangioblastomas associated with VHL disease. Factors Predicting Long-Term Outcomes In the present study, symptoms persisting at the final follow-up examinations were cerebellar ataxia (23%) and dysphagia (10%). In univariate analysis, patient age was a significant factor that predicted whether these symptoms persisted, although this finding was insignificant by multivariate analysis. Aging-related declines in cerebellar or swallowing functions (22) may be related to the results that older patients more frequently have some persisting symptoms compared with younger patients. Poor outcomes were shown by 6 of 9 patients (67%) with tumors located in the anterior cerebellum, although this finding did not reach a significant level. These patients had cranial nerve symptoms such as dysphagia, hearing disturbances, sensory impairments, or cerebellar ataxia. In anterior cerebellar tumors, especially in the cerebellopontine angle with cranial nerve or brainstem involvement, meticulous dissection of tumors from the vital structures should be performed under electrophysiologic monitoring (somatosensory evoked potentials, auditory evoked potentials, or cranial nerve motor evoked potentials) (8, 9). In our study, a cystic tumor was measured as a combined mural nodule and cyst diameter. Preoperatively, peritumoral cysts were one of the most important factors contributing to increased mass effect (10, 24). However, the tumor size did not correlate with postoperative outcomes in our patients. In patients with cystic tumors, removal of only the mural nodule was sufficient to control cerebellar hemangioblastomas if the cyst wall was not enhanced on MRI (1). Even in cases with large peritumoral cysts and small mural nodules, resection of only the mural nodule probably affects the surrounding structures less. Some authors showed that patients with solid hemangioblastomas had a poorer prognosis than patients with cystic tumors, regardless of presence of VHL disease (19, 23, 25). In contrast, other studies showed no difference in outcomes between patients with cystic tumors and solid tumors (4). In the present study, the tumor characteristics WORLD NEUROSURGERY,

7 were the only significant factors predicting long-term outcomes. In solid hemangioblastomas, an internal decompression of tumors may cause abundant blood loss and requires excessive bipolar coagulation (19, 23). En bloc resection of solid hemangioblastomas is recommended as the surgical management of arteriovenous malformations, even if the tumor is relatively large (10, 19, 23). However, surgical resection becomes more difficult with increasing tumor size, and the outcomes worsen, especially in giant tumors with diameters >4.0 cm (19). Endovascular embolization may be a helpful procedure that allows tumor resection with reduced blood loss and produces less damage to surrounding structures in cases with solid tumors (21). In this study,5 patients with solid tumors were treated by endovascular embolization and showed no complications. Because bleeding and tumor swelling have been reported as complications of adjuvant endovascular treatments (5, 7), obstruction of draining veins causing congestion of tumor vessels should never be performed (19). In benign tumors such as hemangioblastomas, outcomes depend on the extent of resection and biologic behavior of the tumor itself. In the current study, all 5 patients in whom total tumor removal was not achieved exhibited a solid configuration. Of these 5 patients, 4 had poor longterm outcomes. It is important to resect solid tumors completely to decrease morbidity. However, the biologic behavior of the hemangioblastoma remains unclear. Further studies are needed to clarify longterm outcomes related to both clinical and biologic aspects of hemangioblastomas. CONCLUSIONS Patient age at surgery and tumor characteristics are the most important factors predicting long-term outcomes in patients with sporadic cerebellar hemangioblastomas. Older patients and patients with solid tumors are more likely to have residual symptoms, such as cerebellar ataxia or dysphagia, over long-term postoperative periods. In addition, some patients with solid lesions develop postoperative hematomas in the removal cavity immediately after surgery. It is important to manage patients with solid hemangioblastomas carefully both intraoperatively and postoperatively. Even in patients without VHL disease, long-term follow-up MRI is needed to detect late recurrences of cerebellar hemangioblastomas. REFERENCES 1. Bishop FS, Liu JK, Chin SS, Fults DW: Recurrent cerebellar hemangioblastoma with enhancing tumor in the cyst wall: case report. Neurosurgery 62: E1378-E1379, Chakraborti PR, Chakrabarti KB, Doughty D, Plowman PN: Stereotactic multiple area radiotherapy. IV. Haemangioblastoma. Br J Neurosurg 11: , Chang SD, Meisel JA, Hancock SL, Martin DP, McManus M, Adler JR Jr: Treatment of hemangioblastomas in von Hippel-Lindau disease with lineal accelerator-based radiosurgery. Neurosurgery 43:28-35, Conway JE, Chou D, Clatterbuck RE, Brem H, Long DM, Rigamonti D: Hemangioblastomas of the central nervous system in von Hippel-Lindau syndrome and sporadic disease. Neurosurgery 48: 55-63, Cornelius JF, Saint-Maurice JP, Bresson D, George B, Houdart E: Hemorrhage after particle embolization of hemangioblastomas: comparison of outcomes in spinal and cerebellar lesions. J Neurosurg 106: , de la MonteSM, Horowitz SA: Hemangioblastomas: clinicalandhistopathological factors correlatedwith recurrence. Neurosurgery 25: , Eskridge JM, McAuliffe W, Harris B, Kim DK, Scott J, Winn HR: Preoperative endovascular embolization of craniospinal hemangioblastomas. AJNR Am J Neuroradiol 17: , Fukuda M, Oishi M, Hiraishi T, Saito A, Fujii Y: Pharyngeal motor evoked potentials elicited by transcranial electrical stimulation for intraoperative monitoring during skull base surgery. J Neurosurg 116: , Fukuda M, Oishi M, Takao A, Saito A, Fujii Y: Facial nerve motor-evoked potential monitoring during skull base surgery predicts facial nerve outcome. J Neurol Neurosurg Psychiatry 79: , Jagannathan J, Lonser RR, Smith R, DeVroom HL, Oldfield EH: Surgical management of cerebellar hemangioblastomas in patients with von Hippel- Lindau disease. J Neurosurg 108: , Kano H, Niranjan A, Mongia S, Kondziolka D, Flickinger JC, Lunsford LD: The role of stereotactic radiosurgery for intracranial hemangioblastomas. Neurosurgery 63: , Le Reste PJ, Henaux PL, Morandi X, Carsin- Nicol B, Brassier G, Riffaud L: Sporadic intracranial haemangioblastomas: surgical outcome in a single institution series. Acta Neurochir (Wien) 155: , Melmon KL, Rosen SW: Lindau s disease. Review of the literature and study of a large kindred. Am J Med 36: , Neumann HP, Eggert HR, Weigel K, Friedburg H, Wiestler OD, Schollmeyer P: Hemangioblastomas of the central nervous system. J Neurosurg 70: 24-30, Niemela M, Lemeta S, Summanen P, Bohling T, Sainto M, Kere J, Poussa K, Sankila R, Haapasalo H, Kaariainen E, Pukkala E, Jaaskelainen J: Long-term prognosis of haemangioblastoma of the CNS: impact of von Hippel-Lindau disease. Acta Neurochir (Wien) 141: , Niemela M, Lim YJ, Soderman M, Jaaskelainen J, Lindquist C: Gamma knife radiosurgery in 11 haemangioblastomas. J Neurosurg 85: , Padhi S, Sarangi R, Challa S, Bussary P, Panigrahi M, Purohit A: A 10-year retrospective study of hemangioblastomas of the central nervous system with reference to von Hippel-Lindau (VHL) disease. J Clin Neurosci 18: , Patrice SJ, Sneed PK, Flickinger JC, Shrieve DC, Pollock BE, Alexander E 3rd, Larson DA, Kondziolka DS, Gutin PH, Wara WM, McDermott MW, Lunsford LD, Loeffler JS: Radiosurgery for hemangioblastoma: results of a multiinstitutional experience. Int J Radiat Oncol Biol Phys 35: , Rachinger J, Buslei R, Prell J, Strauss C: Solid haemangioblastomas of the CNS: a review of 17 consecutive cases. Neurosurg Rev 32:37-48, Symon L, Murota T, Pell M, Bordi L: Surgical management of haemangioblastoma of the posterior fossa. Acta Neurochir (Wien) 120: , Takeuchi S, Tanaka R, Fujii Y, Abe H, Ito Y: Surgical treatment of hemangioblastomas with presurgical endovascular embolization. Neurol Med Chir (Tokyo) 41: , WadhwaR, Toms J,Chittiboina P, TawfikT, GlennC, Caldito G, Guthikonda B, Nanda A: Dysphagia following posterior fossa surgery in adults. World Neurosurg 82: , Wan J, Cui H, Wang Y: Surgical management of large solid hemangioblastomas of the posterior fossa. J Clin Neurosci 18:39-42, Wanebo JE, Lonser RR, Glenn GM, Oldfield EH: The natural history of hemangioblastomas of the central nervous system in patients with von Hippel-Lindau disease. J Neurosurg 98:82-94, Young S, Richardson AE: Solid haemangioblastomas of the posterior fossa: radiological features and results of surgery. J Neurol Neurosurg Psychiatry 50: , Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Received 21 November 2013; accepted 10 June 2014; published online 14 June 2014 Citation: World Neurosurg. (2014) 82, 5: Journal homepage: Available online: /$ - see front matter ª 2014 Elsevier Inc. All rights reserved. WORLD NEUROSURGERY 82 [5]: , NOVEMBER