Mo r e than 90% of meningiomas are histologically

Transcription

1 J Neurosurg 113: , 2010 Treatment decision making based on the published natural history and growth rate of small meningiomas A review and meta-analysis Mi c h a e l E. Su g h r u e, M.D., Mar t i n J. Ru t k o w s k i, B.A., De r i c k Ara n d a, M.D., Ig o r J. Ba r a n i, M.D., Mi c h a e l W. McDe r m o t t, M.D., a n d And r e w T. Pa r sa, M.D., Ph.D. Brain Tumor Research Center, Department of Neurological Surgery, University of California, San Francisco, California Object. Definitive data allowing clinicians to predict which meningioma patients will fail to respond to conservative management are lacking. To address this need, the authors systematically reviewed the published literature regarding the natural history of small, untreated meningiomas. Methods. The authors performed a systematic review of the existing literature on untreated meningiomas that were followed with serial MR imaging. They summarize the published linear rates of tumor growth, and the risk factors for development of new or worsened symptoms during follow-up by using a stratified chi-square test. Results. The search methods identified 22 published studies reporting on 675 patients with untreated meningiomas followed by serial MR imaging. Linear growth rates varied significantly: no growth was the most common rate, although reports of more aggressive tumors noted growth rates of up to a 93% linear increase in size per year. The authors found that few patients with initial tumor diameters < 2 cm went on to develop new or worsened symptoms over a median follow-up period of 4.6 years. Patients with initial tumor diameters of cm demonstrated a marked difference in the rate of symptom progression if their tumors grew > 10% per year, compared with those tumors growing 10% per year (42% vs 0%; p < 0.001, chi-square test). Patients with tumors between > 2.5 and 3 cm in initial size went on to develop new or worsened symptoms 17% of the time. Conclusions. This systematic review of the literature regarding the clinical behavior of untreated meningiomas suggests that most meningiomas 2.5 cm in diameter do not proceed to cause symptoms in the approximately 5-year period following their discovery. Those that do cause symptoms can usually be predicted with close radiographic follow-up. Based on these findings, the authors suggest the importance of observation in the early course of treatment for small asymptomatic meningiomas, especially those with an initial diameter < 2 cm. (DOI: / JNS091966) Ke y Wo r d s meningioma observation management conservative management Abbreviation used in this paper: SRS = stereotactic radiosurgery. Mo r e than 90% of meningiomas are histologically benign lesions. 26 In the pre-ct/mr imaging era, these tumors were typically discovered when they were large and symptomatic, making excision the obvious therapy for most of these lesions. Subsequent improvements in imaging led to an increased ability to detect small meningiomas, often prior to the onset of significant symptoms. 28 Increased experience with these smaller lesions revealed that although some meningiomas never progressed to cause symptoms, many others did. This naturally raised questions regarding treatment for these lesions, including proper selection of patients and appropriate treatment modality. 2,3,26 Although surgery remains the first option for large meningiomas or those immediately adjacent to radiosensitive structures such as the optic apparatus, 26 one appealing option for the management of small meningiomas is SRS. 4,10,12,13 A number of groups have demonstrated excellent results with SRS as the sole therapy for small meningiomas, with tumor control rates exceeding 90%, and a favorable side-effect profile. 4,10,12,13 The idea that SRS provides an acceptable alternative to microsurgical resection for the treatment of selected patients with small meningiomas is difficult to refute. 26 Although it is well established that patients treated with SRS seldom experience tumor progression necessitating conventional surgery, we believe the more important issue is whether these patients need treatment of any kind. 30 Given that most meningiomas are not malignant tumors, the principal reason to treat WHO Grade I meningiomas is to prevent symptoms related to tumor compression of surrounding neurological structures, or to 1036 J Neurosurg / Volume 113 / November 2010

2 The natural history of meningiomas arrest growth in small asymptomatic tumors documented to be growing. Conceptually, this implies preventing the tumor from growing to a size at which it causes symptoms, and avoiding compression or invasion of adjacent neurological structures. It is known that the growth rate is variable in these tumors, and after radiographic diagnosis many meningiomas frequently plateau in size, and either stop growing or grow very slowly. 30 Thus, data regarding tumor control following SRS must be interpreted in light of the natural history of the disease. In the current era of modern imaging, it is fairly straightforward to follow a slow-growing meningioma over a number of years, and to intervene prior to the onset of tumor mass effect and subsequent symptoms, or if the tumor begins to grow on serial imaging. This approach ideally reduces the number of patients undergoing unnecessary treatment and delays the risks of surgery or SRS as long as possible. A principal limitation of this conservative approach is omission of a histological diagnosis. Consequently, WHO Grades II and III meningiomas might progress if untreated, whereas definitive tissue diagnosis would target these high-grade lesions for treatment. 26 An additional limitation is the theoretical risk that a certain percentage of patients with untreated tumors will develop symptoms between follow-up imaging studies. 7,20,26,30 Alternately, if a patient s tumor grows so rapidly that it passes the suitable size threshold for SRS, such a delay limits the treatment options available to the patient. 12,13 Although data regarding the natural history of untreated meningiomas have been published in a number of studies, we have not encountered any clear, evidencebased guidelines for treatment of patients with small meningiomas. It remains unclear which tumors are suitable for observation and which are at risk for growth or development of symptoms necessitating early treatment. To address this, we performed a systematic review of the literature, summarized the growth rates of untreated meningiomas, and determined which factors might suggest a more aggressive tumor progression, to help identify patients in whom earlier treatment will clearly be beneficial. Methods Article Selection Articles were identified via a PubMed search performed using the key words meningioma, observation, natural history, untreated, conservative management, and meningiomas, alone and in combination. We then searched all references in these papers. All data from these papers regarding patients treated conservatively were analyzed for the inclusion criteria. Inclusion criteria for articles included the following: 1) data for untreated meningiomas had to be presented separately from data for patients undergoing surgery or radiosurgery; 2) data for untreated meningiomas had to be presented separately from patients with other tumor types; 3) patients could not have undergone any therapy for their tumor prior to the beginning of observation; and 4) tumor growth must have been analyzed by serial radiographic assessment. J Neurosurg / Volume 113 / November 2010 Data Extraction Data from individual and aggregated case series were extracted from each paper. A percentage of the tumors that grew during the observation period were extracted from all papers and stored as a binary variable. When available, a linear annual fractional growth rate was extracted from the articles. If linear annual growth rate was not available, when possible, this value was calculated using the starting and final tumor size, and the length of follow-up. For simplicity of data presentation, when initial tumor sizes were presented as tumor volumes, these numbers were converted to tumor diameter by assuming that the tumor was perfectly spherical and solving for d in the equation V = (L W H)/2 = d 3 /2. We chose to study linear growth rates because it seemed inappropriate to assume that the largest tumor diameter could be used to estimate the tumor volume by assuming that this diameter reflected the tumor size in the other dimensions. Symptom progression was defined as any new symptoms, or any worsening of existing ones. Statistical Analysis Comparisons between binary variables were made using the Pearson chi-square test. Cutoffs used in our analysis of the effect of tumor growth rate and initial size on outcome were determined empirically by analyzing a number of smaller subgroups to see at what point the outcome trends began to change, and combining subgroups that seemed homogeneous. For all tests, the p value was considered significant at the 5% (that is, p < 0.05) level, after correcting for multiple comparisons. All continuous values are presented as the mean ± SD. Results Results of the Systematic Review Our search methods identified 22 published studies 1 3, 5,7 9,11,14 16,18 25,27,29,30 reporting on 675 patients with untreated meningiomas that were followed by serial MR imaging, for which the initial tumor size and serial growth rate were provided. The demographic characteristics of these patients are summarized in Table 1. The median length of follow-up was 4.6 years. Data regarding size and growth rate separated by location were available for 357 patients (53%). Effect of Tumor Location on Outcome of Untreated Meningiomas A wide variety of tumor locations were presented in the literature. Not surprisingly, the cerebral convexity and falx were the most common sites of tumor origin in our series (Table 2). Whereas most sites of origin demonstrated a similar rate of symptom progression, the rate was significantly higher for cavernous sinus meningiomas (61%). One potential explanation for this is that the cavernous sinus tumors in this series were larger than tumors in other regions at the time of diagnosis. Table 2 also summarizes the sizes and rates of linear growth for other sites of origin in the series. 1037

3 M. E. Sughrue et al. TABLE 1: Demographic and clinical data in 675 patients with untreated meningiomas Factor Value demographic data no. of patients 675 age (yrs) ± SD 63 ± 9 sex (M/F) 16%/84% % w/ presenting Sxs cranial neuropathy 12 headache 9 incidental 6 vertigo/gait disturbance 5 blurry vision 5 seizures 1 mental status changes 1 not stated 60 % w/ initial largest diameter <1 cm cm 6 >1.5 2 cm 42 >2 2.5 cm 6 >2.5 3 cm 23 >3 cm 19 % w/ linear growth rate 0 10% per yr % per yr % per yr 5 >33% per yr 4 % w/ initial MRI traits T2 hyperintense 13 calcification 37 Rates of Growth of Untreated Meningiomas and the Effect on Outcome Our analysis found that 56% of all untreated meningiomas, and 51% of meningiomas with an initial diameter 2.5 cm that were reported in the literature did not demonstrate any evidence of growth on serial imaging obtained over the median follow-up period of 4.6 years. Linear growth rates varied significantly and ranged from no growth, which was the most common rate, up to a 93% linear increase in size per year. 7 Using data from studies that presented both size and growth rate, we were able to subanalyze patient outcomes based on these 2 factors. In general, high rates of tumor growth gradually diminished with increasing tumor size. Slightly more than one-third of tumors between 2 and 3 cm continued to grow > 10% per year, and no tumors > 2.5 cm continued growing at this relative rate. We found that despite the more frequent occurrence of high relative rates of growth, very few patients with initial tumor sizes < 2 cm went on to develop new or worsened symptoms over the 4.6-year median follow-up period (Table 3). This implies that not all tumor growth is necessarily symptomatic. Analyzing the population beginning with tumors cm, we found a marked difference in the rate of symptom progression between patients with tumors growing > 10% per year compared with patients with tumors growing 10% per year (42% vs 0%; p < 0.001, chi-square test). Effect of Tumor T2 Hyperintensity on Outcome We analyzed the prognostic value of initial intrinsic tumor T2 hyperintensity in predicting the eventual development of new or worsened symptoms in patients with untreated meningiomas (Table 4). Interestingly, 52% of patients with tumors between 2 and 2.5 cm that had T2 hyperintensity seen on initial imaging developed new or worsened symptoms during the follow-up period. Further subanalysis of these data by growth rate found that patients with tumors in this size range that had T2 hyperintensity and linear growth rates > 10% per year had a 92% chance of new or worsened symptoms during the follow-up period. Conversely, patients without intrinsic tumor T2 hyperintensity in the same size range did not develop new or worsened symptoms during follow-up (p < 0.001). Indeed, others have reported hypointensity with microcalcification and psammomatous meningiomas. Discussion In this study, we systematically reviewed the published literature regarding the natural history of untreated meningiomas in the MR imaging era. We found that 51% of untreated meningiomas 2.5 cm in largest diameter in patients selected for observation demonstrate no growth over a follow-up period of 4.6 years, and that an additional 26% grow 10% per year. More importantly, we found that a very small fraction (approximately 2%) of patients with initial tumor diameters < 2 cm will go on to develop new or worsened symptoms over the several years following diagnosis. The rate is much higher for patients with an initial tumor diameter of 2 cm, especially in tumors with a high growth rate; patients with such lesions go on to develop progressive symptoms in 42% of cases, which probably results from the fact that volumetric growth of these tumors is exponentially larger than growth associated with much smaller tumors, leading to a larger degree of brain compression than would be expected for a smaller tumor. It should be acknowledged, however, that there is inherent bias in any analysis of patients selected for observation as opposed to those selected for treatment, because the authors of the other reports probably elect to treat most very symptomatic or fast-growing meningiomas. Are we overtreating small meningiomas A comparison of our results with published series of outcomes for meningiomas treated with SRS suggests that this may be true. 4,6,10,12,13 Many of these studies include patients with tumors < 2 cm, 4,6,10,12,13 and our data suggest that these only rarely become more symptomatic within 5 years of diagnosis, even when demonstrating a fair amount of growth in the intervals between serial imaging. Thus, while SRS may slow the growth of these small tumors, it 1038 J Neurosurg / Volume 113 / November 2010

4 The natural history of meningiomas TABLE 2: Summary of rates of new symptoms and distributions of growth rates organized by tumor locations* Location No. Tumor Size & Annual Linear Growth Rate Sx Progression <2 cm cm >2.5 cm No Yes 10% >10% 10% >10% 10% >10% convexity % 11% 35% 38% 10% 12% 2% 3% falx/parasagittal 54 87% 13% 18% 22% 22% 20% 10% 8% CPA 12 60% 40% 8% 25% 25% 42% 0% 0% cavernous sinus 31 39% 61% 9% 4% 0% 0% 0% 87% sphenoid wing 32 95% 5% 24% 31% 7% 17% 7% 14% petroclival 42 72% 28% 0% 5% 5% 5% 5% 79% jugular foramen 7 83% 17% 17% 0% 0% 0% 0% 83% parasellar 15 86% 14% 80% 0% 0% 0% 0% 20% tentorium 12 90% 10% 0% 33% 0% 33% 0% 33% orbit % 0% 100% 0% 0% 0% 0% 0% other 21 67% 33% 33% 17% 0% 0% 25% 25% total % of total 28% 22% 8% 11% 5% 25% * CPA = cerebellopontine angle. remains unclear if these tumors would have ever caused a clinical problem; at the very least, treatment might be delayed. Furthermore, our analysis found that a select group of larger tumors in the 2- to 3-cm range are slow growing and will not cause symptoms in the near future. Thus, although SRS treatment yields an impressive rate of tumor stasis, 4,12,13 it is unclear how much better this clinical control rate is than what would have been accomplished without subjecting the patient to radiation. Our analysis found that the patients at highest risk for developing progressive symptoms were those with tumors between 2 and 2.5 cm in initial diameter, tumors demonstrating a linear growth rate > 10% per year, or those that were hyperintense to brain on T2-weighted MR imaging. Interestingly, these faster-growing tumors were in the minority of the ones in this size range, suggesting that growth rate might begin to plateau in some tumors. Lesions that continue to grow rapidly may be more biologically aggressive than those that begin to plateau at this size. Our analysis does not allow us to determine if this subgroup contains higher-grade lesions, although this hypothesis is worthy of further analysis. Additionally, we were very interested in studying the prognostic significance of T2/FLAIR imaging changes in the surrounding brain in these tumors. This may suggest the presence of brain edema (faster-growing tumor) versus no TABLE 3: Summary of rates of symptom development or progression for various sized meningiomas* Tumor Size (cm) Growth Rate < >2.5 10%/yr 17.1% >10%/yr 2.2% 42.3% * = insufficient data to draw conclusions. J Neurosurg / Volume 113 / November 2010 edema (slow-growing tumor); however, these data were not presented consistently enough to study this question in any rigorous fashion, and it deserves further attention. Regardless, our data suggest that tumors with these characteristics are more likely to grow during observation and to cause symptoms in the next 4 5 years. When observing a probable meningioma on MR imaging, the principal concern of many clinicians is that the tumor will grow rapidly during the interval between imaging studies and cause worsened symptoms, or exceed the size at which SRS is an option. 4,12,13 Although inevitably if all small meningiomas are left untreated, some TABLE 4: Initial tumor size versus T2 signal features and observed annual linear growth rates* Tumor Size & Growth Rate No T2 Hyperintensity T2 Hyperintensity p Value summary of rates of Sx development or progression, categorized by size of tumor <2 cm 1.9% NS cm 52.4% <0.001 >2.5 cm NS subanalysis w/ patients subdivided by size & growth rate of tumor <2 cm 10% >10% cm 10% >10% >2.5 cm 10% >10% * NS = not significant. 2.2% 91.7% NS <

5 M. E. Sughrue et al. tumors will grow and cause symptoms, our analysis suggests these cases are predictable, and that smaller meningiomas are unlikely to cause problems for several years, if ever. Table 5 and Fig. 1 depict a series of calculations for untreated meningiomas of various sizes growing at the range of linear growth rates reported in the existing literature. Most notably, if one decides to observe a patient with a 1- or 1.5-cm meningioma and it proceeds to grow at the fastest annual growth rate we found reported in the literature, 7 it will still be a suitable size for SRS when the patient receives the first annual MR imaging. This of course assumes that the tumor lacks imaging features that are a cause for concern, or a clinical history that suggests that a histopathological diagnosis is necessary. Thus, waiting a year in these situations to observe a tumor s biological behavior is very unlikely to limit a patient s ensuing treatment options or to put that patient at risk for progressive symptoms between imaging studies. In many clinical settings, the first two imaging follow-ups are done 6 months apart to exclude rapid growth, and, once a 1-year interval has elapsed with no radiographically confirmed linear growth, and no new clinical symptoms attributable to the tumor site have developed, then repeat imaging in another year seems reasonable. It is important to consider these results in the context of individual tumor and patient characteristics. More specifically, the location of tumors should have a significant impact on the threshold for intervention, because it is probably best to avoid significant increases in the size of tumors in more surgically challenging areas, such as the cerebellopontine angle, the anterior clinoid process, and the petroclival junction, where the risk associated with resection of a larger tumor might be made more complex by a modest increase in size. Similarly, recent data suggest that the risk of having a higher-grade meningioma is not equally distributed between meningiomas arising from all locations, but rather is lower for skull base meningiomas and higher for convexity meningiomas. 17 Study Limitations Although these findings represent a helpful summary of the published literature on this topic, an analysis of published data is only as good as its composite studies, and may reflect source study biases. Most notably, data presentation methods differ significantly between different studies, potentially biasing some data sets. Furthermore, few studies published disaggregated data regarding different imaging features, such as irregular borders, edema in the adjacent brain, and heterogeneous contrast enhancement, which would allow for a critical analysis if some of these rapidly growing tumors were potentially higher grade, or would not be observed by many clinicians. It is impossible for us to control for the quality of the data reported in the literature, and in some studies, TABLE 5: A series of calculations for untreated meningiomas of various sizes growing at the range of linear growth rates reported in the existing literature Growth Rate Projected Tumor Size (cm) Growth Projected Tumor Size (cm) Initial Yr 1 Yr 2 Yr 3 Yr 4 Yr 5 Rate Initial Yr 1 Yr 2 Yr 3 Yr 4 Yr 5 3% % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % J Neurosurg / Volume 113 / November 2010

6 The natural history of meningiomas Fig. 1. Graph representing the data presented in Table 5. Data are presented for 6 growth rates for tumors with initial sizes of 1 cm (A), 1.5 cm (B), and 2 cm (C). The horizontal dashed lines depict the size threshold for radiosurgery. The vertical dashed lines indicate the timing of the first annual MR image after the decision was made to observe a tumor with an unknown growth rate. systematic differences in clinical assessment may lead to incorrect estimates of rates of tumor growth and the rate of symptom evolution, which we cannot assess critically or control for. Finally, due to the diverse range of data presentation, the number of variables able to be studied and controlled for is limited. Variables that might be of interest that are inconsistently presented across studies cannot be reviewed. For example, we used linear growth rates due to the limited available data about volumetric growth rates, and it is possible that different results might have been obtained if more volumetric data had been available. Conclusions In this study, we present a systematic summary of the literature regarding the relationship between size, serial growth, and imaging features and the rate of symptom development in small, untreated meningiomas. Most importantly, we believe that more systematic efforts to predict the failure of observation alone are warranted. Furthermore, it is important to determine if current radiosurgery treatment patterns reduce the incidence of large symptomatic meningiomas requiring surgery. Disclosure Dr. Parsa is supported in part by the Reza and Georgianna Khatib Endowed Chair in Skull Base Tumor Surgery. Dr. McDermott is supported in part by the Robert and Ruth Halperin Endowed Chair in Meningioma Research. Dr. Sughrue is supported by the American Association of Neurological Surgeons Neurosurgery Research and Education Foundation. Mr. Rutkowski is supported by the Doris Duke Charitable Foundation. The authors declare that they are not involved in any relationships with companies that make products related to this study. Author contributions to the study and manuscript preparation in clude the following. Conception and design: Parsa, Sughrue, Ar an da. Acquisition of data: Sughrue, Rutkowski, Aranda. Analysis and interpretation of data: Sughrue, Rutkowski. Drafting the article: Sug hrue, Rutkowski, McDermott. Critically revising the article: Par sa, Sughrue, Aranda, Barani, McDermott. Reviewed final version of the manuscript and approved it for submission: Parsa, Sug hrue, Rutkowski, Aranda. Statistical analysis: Sughrue, Barani, Mc Dermott. Administrative/technical/material support: Sughrue. Study supervision: Parsa. J Neurosurg / Volume 113 / November 2010 References 1. Bindal R, Goodman JM, Kawasaki A, Purvin V, Kuzma B: The natural history of untreated skull base meningiomas. Surg Neurol 59:87 92, Braunstein JB, Vick NA: Meningiomas: the decision not to operate. Neurology 48: , Firsching RP, Fischer A, Peters R, Thun F, Klug N: Growth rate of incidental meningiomas. J Neurosurg 73: , Flickinger JC, Kondziolka D, Maitz AH, Lunsford LD: Gamma knife radiosurgery of imaging-diagnosed intracranial meningioma. Int J Radiat Oncol Biol Phys 56: , Go RS, Taylor BV, Kimmel DW: The natural history of asymp tomatic meningiomas in Olmsted County, Minnesota. Neu rology 51: , Han JH, Kim DG, Chung HT, Park CK, Paek SH, Kim CY, et al: Gamma knife radiosurgery for skull base meningiomas: long-term radiologic and clinical outcome. Int J Radiat Oncol Biol Phys 72: , Hashiba T, Hashimoto N, Izumoto S, Suzuki T, Kagawa N, Maruno M, et al: Serial volumetric assessment of the natural history and growth pattern of incidentally discovered meningiomas. Clinical article. J Neurosurg 110: , Herscovici Z, Rappaport Z, Sulkes J, Danaila L, Rubin G: Natural history of conservatively treated meningiomas. Neurology 63: , Igarashi T, Saeki N, Yamaura A: Long-term magnetic resonance imaging follow-up of asymptomatic sellar tumors their natural history and surgical indications. Neurol Med Chir (Tokyo) 39: , Iwai Y, Yamanaka K, Ikeda H: Gamma Knife radiosurgery for skull base meningioma: long-term results of low-dose treatment. Clinical article. J Neurosurg 109: , Iwai Y, Yamanaka K, Morikawa T, Ishiguro T, Honda Y, Matsusaka Y, et al: [The treatment for asymptomatic meningiomas in the era of radiosurgery.] No Shinkei Geka 31: , 2003 (Jpn) 12. Kondziolka D, Levy EI, Niranjan A, Flickinger JC, Lunsford LD: Long-term outcomes after meningioma radiosurgery: physician and patient perspectives. J Neurosurg 91:44 50, Kondziolka D, Mathieu D, Lunsford LD, Martin JJ, Madhok R, Niranjan A, et al: Radiosurgery as definitive management of intracranial meningiomas. Neurosurgery 62:53 60, Kuratsu J, Kochi M, Ushio Y: Incidence and clinical features of asymptomatic meningiomas. J Neurosurg 92: , Kuroiwa T, Hirai T, Ohta T: Meningioma followed up for ra- 1041

7 M. E. Sughrue et al. diological findings before and after radiosurgery: case report. Minim Invasive Neurosurg 42:44 46, Mani H, Peyman GA, Leff SI: Intraocular meningothelial meningiomas. Int Ophthalmol 12: , McGovern SL, Aldape KD, Munsell MF, Mahajan A, Demonte F, Woo SY: A comparison of World Health Organization tumor grades at recurrence in patients with non-skull base and skull base meningiomas. J Neurosurg [epub ahead of print October 2, DOI: / JNS09617] 18. Nabika S, Kiya K, Satoh H, Mizoue T, Oshita J, Kondo H: [Strategy for the treatment of incidental meningiomas.] No Shinkei Geka 35:27 32, 2007 (Jpn) 19. Nakamura M, Roser F, Michel J, Jacobs C, Samii M: The natural history of incidental meningiomas. Neurosurgery 53: 62 71, Nakasu S, Fukami T, Nakajima M, Watanabe K, Ichikawa M, Matsuda M: Growth pattern changes of meningiomas: longterm analysis. Neurosurgery 56: , Niiro M, Yatsushiro K, Nakamura K, Kawahara Y, Kuratsu J: Natural history of elderly patients with asymptomatic meningiomas. J Neurol Neurosurg Psychiatry 68:25 28, Olivero WC, Lister JR, Elwood PW: The natural history and growth rate of asymptomatic meningiomas: a review of 60 patients. J Neurosurg 83: , Park CK, Jung HW, Kim JE, Paek SH, Kim DG: The selection of the optimal therapeutic strategy for petroclival meningiomas. Surg Neurol 66: , Pitz S, Becker G, Schiefer U, Wilhelm H, Jeremic B, Bamberg M, et al: Stereotactic fractionated irradiation of optic nerve sheath meningioma: a new treatment alternative. Br J Ophthalmol 86: , Pless M, Lessell S: Spontaneous visual improvement in orbital apex tumors. Arch Ophthalmol 114: , Pollock BE: Stereotactic radiosurgery of benign intracranial tumors. J Neurooncol 92: , Van Havenbergh T, Carvalho G, Tatagiba M, Plets C, Samii M: Natural history of petroclival meningiomas. Neurosurgery 52:55 64, Vernooij MW, Ikram MA, Tanghe HL, Vincent AJ, Hofman A, Krestin GP, et al: Incidental findings on brain MRI in the general population. N Engl J Med 357: , Yano S, Kuratsu J: Indications for surgery in patients with asymptomatic meningiomas based on an extensive experience. J Neurosurg 105: , Yoneoka Y, Fujii Y, Tanaka R: Growth of incidental meningiomas. Acta Neurochir (Wien) 142: , 2000 Manuscript submitted December 30, Accepted March 16, Please include this information when citing this paper: published online April 30, 2010; DOI: / JNS Address correspondence to: Andrew T. Parsa, M.D., Ph.D., De - partment of Neurological Surgery, University of California, San Fran cisco, 505 Parnassus Avenue, San Francisco, California parsaa@neurosurg.ucsf.edu J Neurosurg / Volume 113 / November 2010