Progress in the diagnosis and treatment of patients with meningiomas

Neuro-Oncology Review

Progress in the Diagnosis and Treatment of Patients with Meningiomas Part I: Diagnostic Imaging, Preoperative Embolization Herbert H. Engelhard, M.D., Ph.D., F.A.C.S. Department of Neurosurgery, The University of Illinois at Chicago, Chicago, Illinois

Engelhard HH. Progress in the diagnosis and treatment of patients with meningiomas. Part I: diagnostic imaging, preoperative embolization. Surg Neurol 2001;55:89 –101.

operative embolization of meningiomas is feasible, and seems to be reasonably safe.

BACKGROUND

MR imaging, CT scans, and cerebral angiography can currently be used in a complementary fashion to diagnose, evaluate, and treat patients with meningiomas, with a high degree of clinical certainty. Angiography is used to determine the sites of blood supply to the tumor, which can then be attacked first intraoperatively, making tumor removal easier. Preoperative embolization continues to have value in selected patients, including those in whom the blood supply to the tumor is difficult to access at the time of surgery. © 2001 by Elsevier Science Inc.

The clinical management of patients with meningiomas has changed over the past decade. Change has occurred because of a variety of factors including improved diagnostic imaging, better results with surgery and interventional neuroradiology, and the advent of radiosurgery. Recent clinical studies from several disciplines have provided new information on topics germane to the management of patients with meningiomas. Collecting this information into a series of review articles would have significant value, primarily for neurosurgeons. OBJECTIVE

The purpose of this first paper is to bring together and evaluate the available data on: 1) noninvasive diagnostic imaging of meningiomas, including magnetic resonance imaging (MRI), computed tomography (CT) scanning, and MR angiography, venography and spectroscopy; 2) the present role of cerebral angiography in patients with meningiomas; and 3) the current status of preoperative embolization for these tumors. RESULTS

With the advent of MR technology, the quality of diagnostic imaging for meningiomas has improved dramatically, and this is reflected in more sophisticated preoperative planning. MR imaging provides improved delineation of dura and sinus involvement, and even information about a tumor’s consistency. Meningiomas have characteristic neuroimaging features, yet other lesions can still mimic a meningioma. MR venography can be used to demonstrate sinus patency, but intra-arterial cerebral angiography gives the most precise information concerning the degree of tumor involvement of critical vascular structures, and the anatomy of arterial feeders. In trained hands, superselective catheterization for pre-

Address reprint requests to: Dr Herb Engelhard, Department of Neurosurgery, The University of Illinois at Chicago, 912 South Wood St., Chicago, IL 60612. Received October 3, 2000; accepted December 5, 2000. © 2001 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

CONCLUSIONS

KEY WORDS

Brain tumor treatment, cerebral angiography, computed tomography, embolization, magnetic resonance imaging, magnetic resonance spectroscopy, meningioma.

eningiomas are the most common benign brain tumor in adults, and as such, are encountered fairly often in neurosurgical practice. Surgical resection continues to be the most effective treatment for meningiomas [2,5,15,25]. Several excellent chapters and reviews are available which provide recent overviews of the topic of meningiomas [16,17,26,45,51]. There, one can find information about the historical aspects, epidemiology, intracranial distribution, etiology, presenting features, diagnostic imaging, pathology, molecular biology, surgical approaches, blood supply, and adjunctive treatment of this type of tumor. In contrast, this series of review articles will focus on a selection of key topics that have been evolving over the past several years. This first paper covers 1) advances that have been made in the diagnostic imaging of meningiomas, especially in MR imaging; 2) the implications these advances have for surgical planning; 3) the cur-

M

0090-3019/01/$–see front matter PII S0090-3019(01)00349-4

90

Surg Neurol 2001;55:89 –101

rent role of angiography and preoperative embolization of meningiomas. Neurosurgeons have to constantly integrate information from a variety of sources to make clinical judgments and care for their patients. The information comes from their training, from neurosurgical meetings and discussions with colleagues, and from books and journal articles. This basic fund of knowledge is then continually modified according to the individual neurosurgeon’s own experiences with patients. Very few “randomized” or “controlled” studies of meningioma patients are available or even feasible. Therefore, the answers to most clinical questions fall into the realm of clinical judgment and treatment options. Questions addressed in this paper include: What are the characteristic imaging features of meningiomas? What is the significance of a “dural tail”? Does the dura normally enhance? Can the consistency of a meningioma be predicted based on the preoperative MR images? What lesions can mimic a meningioma on MRI? Is magnetic resonance spectroscopy (MRS) useful for meningioma patients? When is the optimal time for obtaining an MRI scan postoperatively to assess whether or not residual meningioma is present? Do CT scans and cerebral angiography currently have any use for patients with meningiomas? What is the risk of cerebral angiography in a patient with a brain tumor? How important is preoperative embolization for a meningioma? What are the complications of this procedure? Has embolization been used alone—without surgery—in attempting to treat difficult meningiomas? Information pertaining to positron emission tomography (PET) and single photon emission computed tomography (SPECT) is not included in this review.

MR Imaging of Meningiomas: Characteristic Features The advent of CT and then MRI revolutionized the diagnosis of intracranial disease, including meningiomas [66]. In Cushing’s classic monograph, patients were described who presented with huge meningiomas and, often, associated deformities of the skull [12]. While large tumors can still be encountered in clinical practice [2,39,46,48 –50,66], advances in imaging have enabled physicians to detect progressively smaller meningiomas. Intracranial tumors can now be detected that have a diameter less than 3 mm (1.5T MRI scanner, with

Engelhard

and without gadolinium enhancement). The threshold of detection depends not only on the strength of the magnetic field, but also on a number of other variables including acquisition time, slice thickness, and the size of the gap between slices. MRI machines with higher field strengths and improved software will decrease this threshold even further in the near future. At the University of Illinois at Chicago, one of the country’s first 3T clinical MRIs unit is now operational; the degree of resolution it provides is remarkable. Improved resolution of tumor and anatomic structures, displayed in multiple planes, allows the neurosurgeon to more accurately perceive threedimensional relationships between tumor, dura, brain (and cranial nerves), and cerebral vascular structures. Three-dimensional computer-generated reconstructions based on MR imaging can be striking [4]. Appreciation of anatomic relationships is also the key to the neuroradiologic diagnosis of meningiomas. On unenhanced T1-weighted images, most meningiomas are well-circumscribed extraaxial masses, which are usually (60 –90% of the time [15,22,30,66]) isointense with gray matter. Most other meningiomas (10 – 40% [15,22,30,66]) are slightly hypointense to gray matter. Because of this, they may be hard to appreciate on unenhanced T1-weighted images (Figure 1). On T2 imaging, meningiomas have a more variable appearance, which seems to relate to the consistency of the tumor. Dense calcifications may show up as darker areas both on T1- and T2-weighted images. Rapid growth may cause areas of central necrosis, which are hypointense on T1 and hyperintense on T2 [22,66]. Cyst formation, lipid production, and hemorrhage may occur in meningiomas, but are relatively rare [22]. MRI is particularly valuable (over CT) for diagnosing tumors of the orbit, posterior fossa and skull base [41,45]. With gadolinium-DTPA infusion, meningiomas usually show a marked, homogeneous enhancement pattern. This is maximal approximately five minutes after gadolinium administration [64]. Enhancement usually shows the tumor to be sharply demarcated from normal brain. Most often, meningiomas are rounded to lobular, but they may also present en plaque, that is, as a sheet of varying thickness adjacent to bone. A key feature is that they are usually based broadly on the dura (Figure 2). When gadolinium is used, the improved resolution of the newer MR scanners allows better delineation of the extent of tumor spread into dura adjacent to the tumor and the degree of tumor invasion into the dural sinuses. With infusion, meningiomas typically have a dural “tail” (Figure 2). The presence of such

Meningiomas: Imaging and Embolization

Surg Neurol 2001;55:89 –101

91

Gadolinium-DTPA-enhanced MR image from the edge of a convexity meningioma. This image illustrates: 1) how meningiomas are typically broadly – based on the dura, and 2) dural enhancement away from the tumor mass (arrow). Such dural enhancement has been called a “dural tail.” It is not pathgnomonic for meningioma, having been reported in association with a number of other conditions. When identified preoperatively adjacent to a meningioma, such enhancing dura usually— but not always— contains tumor cells. Because of this, resection of the dural tail has been advocated when possible to decrease the likelihood of tumor recurrence.

2

T1-weighted images of a 2-cm-diameter meningioma. A: Image without gadolinium enhancement. In this scan, the presence of the meningioma is very difficult to appreciate, since the tumor is isointense to brain. B: Gadolinium-enhanced image, which clearly demonstrates the tumor. The mass is well-circumscribed and enhances intensely and homogeneously, as is typical of a meningioma.

1

a tail, although suspicious, is not pathognomonic for meningioma, as was once believed. The meninges incorporated in the enhancing tail have been studied histologically and with electron microscopy. Enhancement usually results from tumor infiltration, but can also be because of nonneoplastic “reactive change.” This dural reaction has been variously attributed to “hyperemia,” tissue proliferation, hypervascularity, and increased

permeability and/or dilatation of the dural vessels [18,23,26,28,37,39,56]. Any process that causes meningeal irritation can produce enhancement [59]. Accordingly, dural tails have been found in association with other tumors, inflammatory processes and even aneurysms [24]. However, since the dural tail probably indicates tumor extension (about 2/3 of the cases [28]), it should be resected, if possible, to reduce the risk of tumor recurrence [15,28]. It remains to be seen whether or not gadolinium-enhanced MRI is superior to direct visual inspection under magnification for detecting tumor involvement of the dura. Both sources of information are used together to plan the dural resection. Another MR imaging feature of many meningiomas (and other tumors as well) is that of brain edema. Edema from a meningioma may produce a surrounding lower intensity (darker) signal on T1 images, but is better seen as a higher-intensity (whiter) signal on the T2 images [22]. It has been stated that as many as 2/3 of patients with (symp-

92

1

Surg Neurol 2001;55:89 –101

Engelhard

Characteristic MRI Features of Meningiomas

General:

T1-weighted images: T2-weighted images: Enhancement: Anatomy:

Well-circumscribed, extra-axial masses; usually occur in typical locations; calcification and/or edema may be present Isointense (to slightly hypointense) with gray matter Variable appearance; may relate to consistency Homogeneous, usually intense; may have a dural “tail” Look for invasion of dura and sinuses and displacement of neural and vascular structures

2

Intracranial Lesions That Can Mimic a Meningioma

Granulomas:

Primary tumors:

Metastatic tumors: Other lesions:

tomatic) meningiomas have at least some degree of peritumoral edema [40]. Edema from a meningioma can, at times, be abundant [39]. Lobato et al reported that meningiomas located along the frontal convexity or middle third of the falx were most likely to be associated with edema formation [40]. Ide et al reported that parasagittal, sphenoid ridge, and olfactory groove meningiomas induce edema more frequently than others [29]. The presence and duration of symptoms, tumor size, and degree of cortical damage (or invasion) are other factors that have been found to correlate with the formation of edema adjacent to meningiomas [29, 40]. The characteristic MRI features of meningiomas are summarized in Table 1. Though not pathognomonic, knowledge of these features can help tremendously in discriminating meningiomas from other tumors and brain lesions. Yet even with the improved tumor imaging afforded by MRI, other intracranial lesions can still mimic a meningioma, especially if there is meningeal involvement [5,20, 24,34,39,44,48,66]. These conditions are listed in Table 2. The post-gadolinium T1-weighted study may help in distinguishing meningiomas from sarcoidosis, other granulomas, and leptomeningeal lymphomatosis or carcinomatosis. In these other conditions, a subarachnoid gyral-like enhancement pattern may be seen. Additional acquisition techniques (pulse sequences) are being developed; their utility in the diagnosis of meningiomas is not yet clear [14,64]. Fat suppression images may help delineate the tumor’s extent, especially in areas near the scalp or bone marrow. For instance, for detection of optic sheath meningiomas, post-gadolinium T1-weighted fat suppression MRI is the best imaging technique [41]. FLAIR (fluid attenuation inversion recovery)

foreign body granuloma neurosarcoidosis infectious granuloma (e.g., tuberculosis) schwannoma cavernous hemangioma hemangiopericytoma plasmacytoma chordoma chondroma Langerhans’ cell histiocytosis Rosai–Dorfman disease sarcomas lymphoma breast carcinoma prostatic carcinoma aneurysm other inflammatory processes

images can be used to better delineate meningioma from surrounding cerebrospinal fluid.

MR Imaging of Meningiomas: Histology, Consistency, Extent of Resection Neurosurgeons, neuroradiologists, and neuropathologists have been interested in determining whether or not MR imaging of a meningioma could be used to predict its histologic type, vascularity, consistency, and/or biologic behavior. With respect to histology, studies of T1- and T2-weighted images have yielded variable results. Some authors have reported that hyperintensity on T2-weighted images is more consistent with an angioblastic, meningothelial, syncytial, and/or melanocytic subtype of meningioma [9,16,30,42,58], while others have not found a correlation [16,65]. Findings have also been variable regarding the prediction of tumor vascularity, including attempts to relate vascularity to the degree of gadolinium enhancement [9]. Malignant meningiomas more commonly display increased signal intensity on both T1- and T2weighted MR scans than do benign meningiomas, as well as features such as “mushrooming” (irregular projections of tumor into the brain surface), indistinct margins, and central necrosis [2,22]. However, correlation of radiologic features and biologic behavior, including the degree of malignancy, continues to be imperfect [2]. Whenever needed, a tissue diagnosis should be obtained. The preoperative MRI does seem to be helpful in

Meningiomas: Imaging and Embolization

predicting the consistency (i.e., firmness) of the tumor, found at the time of surgery. Suzuki et al studied 73 intracranial meningiomas and concluded that the lower intensity portions of the tumor on T2-weighted images were harder and more fibrous in character, whereas the higher intensity portions were softer [58]. Intraoperative blood loss did not vary with consistency, but harder tumors generally took longer to resect than softer ones [58]. Other authors have reported similar findings on T2weighted and also proton density images [9,42,65]. Degree of tumor calcification may also be a factor in consistency; this can be assessed by MRI or, better, by CT scan. For harder tumors, preoperative embolization may be of benefit [65]. Ildan et al reported that meningiomas with features including peritumoral edema, hyperintensity on T2 imaging, cortical penetration, and vascular supply from pialcortical arteries were predictive of “cleavage,” that is, how easily the tumor could be separated from brain [30]. Postoperatively, the question arises as to when is the best time to scan a patient to determine the extent (if any) of residual tumor. Some parts of the meninges—particularly around the anterior portions of the temporal lobes and in the parasagittal region—may normally show some fine linear enhancement [18,59]. The meninges may also enhance postoperatively, with or without tumor cells being present [7,18,62]. This is illustrated in Figure 3B. Postoperative enhancement in the absence of residual tumor has been attributed to a local inflammatory response, and may even be present on the first postoperative day [7, 18]. The presence of thick or nodular enhancement, however, has a high correlation with recurrent or residual meningioma [15,62]. Dolinskas and Simeone conducted a study in which enhanced MRI scans were obtained on 21 patients at four different time intervals: 1) during the first five postoperative days, 2) 3 to 8 weeks after surgery, 3) 3 months to 1 year after surgery, and 4) 1 year or more after surgery. They found that residual foci of meningioma were best detected during the very early postoperative period. Later, areas of membrane enhancement can occur and progressively thicken with time, which could mask the presence of residual meningioma. This enhancement in itself does not necessarily indicate the presence of residual tumor, because it can thin with time (usually at about 6 months) and again become confined to the craniotomy site [18]. Our usual practice is to obtain an MRI scan with and without gadolinium enhancement within three days of sur-

Surg Neurol 2001;55:89 –101

93

A: Contrast-enhanced CT scan of a large convexity meningioma, showing smooth contour, mass effect, and homogeneous intense enhancement. B: Gadoliniumenhanced MRI obtained immediately after total removal of the tumor seen in A. Note the wide area of dural enhancement on the left side (arrows), which is a “normal” postoperative finding and does not necessarily indicate the presence of residual tumor cells.

3

gery, usually on the first postoperative day. If residual tumor is present, there are four options: 1) follow-up with serial MRI scans, 2) radiosurgery, 3) radiation therapy, 4) reoperation [19].

94

3

Surg Neurol 2001;55:89 –101

Engelhard

Recent Developments in MR Imaging of Meningiomas

DEVELOPMENT New acquisition techniques e.g., fat suppression Magnetic resonance angiography/ venography Magnetic resonance spectroscopy 3D volumetric determinations MR-guided surgery Intraoperative MRI

SIGNIFICANCE Better appreciation of anatomy; improved differential diagnosis Better appreciation of vascular anatomy/invasion/occlusion More accurate preoperative diagnosis of pathology; evaluation of embolization Measuring response to treatment, e.g., radiosurgery, antiangiogenesis therapy More precise craniotomies Confirmation of tumor location and extent of resection (if needed)

MR Technology for Meningiomas: Recent Developments Table 3 lists recent developments in MR imaging that may have an impact upon the management of meningioma patients. MR-guided surgery and intraoperative MRI will be discussed in a subsequent article in this series. Personal computer-based software can now be used to obtain precise volume determinations for intracranial mass lesions. Such determinations may be useful when assessing the response of a meningioma to a treatment, such as radiosurgery or anti-angiogenesis therapy. A primary purpose in imaging a meningioma is to precisely define the relevant vascular anatomy including the location of feeding vessels, displacement of normal vascular structures, position of cortical draining veins, and degree of dural sinus invasion or occlusion. The nature of this information can have a significant impact on surgical planning [48]. In addition to MRI, MR angiography (MRA), and MR venography (MRV) can be very useful for this [60]. MRV has the ability to noninvasively detect the patency of the dural venous sinuses [25,45,60]. Yet non-visualization of a sinus on MRV may not be sufficient to prove total occlusion. Currently, conventional intra-arterial digital subtraction angiography gives more precise information concerning the degree of tumor involvement of critical venous structures and the anatomy of arterial feeders and vessels of passage, than does MR imaging at 1.5T. Detection of vessels with a diameter less than 1 mm has remained elusive for MRA, and MRA may overestimate the degree of stenosis [60]. With MRA, it is difficult to determine whether lack of visualization of an artery is the result of severe stenosis or complete occlusion [60]. However, vessels are more clearly shown with the 3T MRI unit, and this may impact the preoperative evaluation of patients in the near future. Functional MRI (fMRI) can be used to noninva-

sively localize areas of cortical function, such as the various speech and motor areas. Conceptually, it would seem to be important to know the relative proximity of any brain tumor to areas of critical cortical activity. Yet the use of fMRI for extra-axial tumors, although reported in the literature, has not been demonstrated to have clear value. Meningiomas (unless malignant) compress but don’t usually invade normal brain, and can be identified visually more easily than intra-axial tumors. Therefore, fMRI has not been as useful for meningiomas as it has been for gliomas, which can infiltrate cortical regions within and adjacent to eloquent areas of the brain. In the future, however, fMRI may become relatively more important for those meningioma patients considering radiosurgery (for instance for a tumor recurrence) near critical cortical areas, such as the motor strip or speech areas [63]. Clinical MR spectroscopy (MRS) is another advancement in MR technology that could potentially have a wider impact on meningioma patients in the near future. MRS studies of meningiomas have been conducted since the late 1980s, and 1H and 31P spectra reported. In general, 1H MRS has been more widely used [16,33]. As with other tumors, meningiomas usually demonstrate an increased choline signal on 1H MRS, with decreases in the N-acetylaspartate and phosphocreatine/creatine (“creatine”) peaks [16]. Many investigators consider the presence of alanine as characteristic of meningiomas [8]. The alanine/creatine ratio in meningeal cells is 3 to 4 times higher than in astrocytes, neurons, and oligodendrocytes. In meningiomas, pyruvate kinase is inhibited by L-alanine, resulting in an increased pool of pyruvate, which may then be converted to alanine [8]. A diagnosis of meningioma should be considered when the alanine peak is higher than the creatine peak. However, in some series alanine was not found, and necrotic areas may show less alanine [8]. MRS has also been performed on meningioma patients who underwent tumor embolization, with

Meningiomas: Imaging and Embolization

or without subsequent surgery [3,61]. In embolized tumors, infarction was indicated by increased levels of lactate and aliphatic lipids [61]. In patients who did not have surgery, long-term follow-up with MRS indicated the occurrence of fatty degeneration of the tumor [3]. MRS may also be helpful in diagnosing atypical or malignant meningiomas [57]. At present, MRS of meningiomas is still considered an evolving area of study. Meningiomas are currently diagnosed based on their characteristic MRI appearance rather than by MRS. MRS may have an expanded role in the future, such as in documenting the metabolic response of meningioma to embolization, radiosurgery, or other therapeutic interventions.

Current Role of CT Scans and Angiography in Patients with Meningiomas Although MR imaging currently dominates the diagnosis of intracranial disorders, including tumors, other neuroimaging techniques, such as CT scanning and intra-arterial cerebral angiography, continue to have a role in selected cases. All but the smallest meningiomas should be visible on a highquality, contrast-enhanced CT scan [25]. There are more CT scanners than MRI units; CT is available in the vast majority of hospitals in North America [64]. CT is less expensive than MRI, has superior tolerance of patient motion, and allows easier access to critically ill patients [64]. Because of this, a pre- and post-contrast CT may be the most feasible initial screening test for an intracranial mass lesion [64]. Detection of tumors in the posterior fossa, of course, is more difficult because of artifact produced by the dense bone. CT scans (bone windows) are useful for assessing the degree of bony involvement of a meningioma, especially at the skull base, and (if needed) the extent of tumor calcification. Highly calcified meningiomas probably have less potential for further growth. CT scans can also be helpful in establishing the relationship of the tumor to bony landmarks; for example, in the petroclival region [45]. Thinly sliced contrast-enhanced CT scans can be used to generate three-dimensional maps of vascular anatomy, which in some situations may be superior to MRA/MRV. Postoperatively, obtaining a CT scan is the fastest and most cost-effective way to screen for the presence of intracranial blood or air, or possible ventricular dilatation or brain swelling. On the non-enhanced CT scan, a meningioma is

Surg Neurol 2001;55:89 –101

95

typically isodense to slightly hyperdense relative to brain, homogeneous, and sharply marginated [26, 39,45]. As with MRI, on the noncontrast scan, a meningioma—particularly a small one—may be very difficult to discern. Tumor calcification may take a variety of forms from punctate to nodular to dense calcificaton of the entire tumor [39]. Bone changes may be caused by actual infiltration of bone by the tumor, erosion from direct pressure and/or bony reaction to the arterial feeding of the tumor [39]. Hyperostosis may occur with larger tumors near bone, or be the primary manifestation of a thin en plaque meningioma [39]. Bone windows are particularly helpful in viewing such changes, and helping to discriminate meningiomas from other tumors. Schwannomas, for example, may cause bone erosion and remodeling, but do not produce the hyperostosis typical of meningiomas, and almost never calcify [39]. A chordoma is an example of a tumor that might resemble a meningioma, but has a different pattern of bone destruction. On the post-infusion CT scan, meningiomas usually show homogeneous, intense enhancement (Figure 3A). Such enhancement results not from hypervascularity, but from iodinated contrast material passing from capillaries into the interstitial space of the tumor. Meningiomas do not have a blood– brain barrier [39]. As with MRI, the enhancement shows the tumor to be sharply demarcated from normal brain and broadly based on bone or dura. About 1⁄4 of meningiomas do not enhance [25]. Approximately 15% of benign meningiomas will have an unusual appearance because of necrosis, fat, and/or blood [26,39]. In a direct comparison of meningioma imaging by CT versus MRI, Kizana et al found that MRI was better at defining the cleavage planes between tumor and the brain, and also vascular features, whereas CT was better in identifying hyperostosis of the skull and tumor calcification [35]. As on MRI, malignant meningiomas are less likely to show homogeneous enhancement or calcification on CT scan, and more likely to be associated with moderate to severe edema than are benign meningiomas [2,25]. In the past, cerebral angiography was widely used to establish the diagnosis of a meningioma by demonstrating its arterial supply from meningeal vessels and the typical finding of a “delayed blush,” that is, contrast persisting into the late venous phase [4,32,66]. In the classic situation, the arterial pedicle enters the tumor at its meningeal attachment and supplies a radially arrayed tumor vascular pattern resulting in a “sunburst” or “spokeswheel” appearance [32,49], as illustrated in Figure

96

Surg Neurol 2001;55:89 –101

Engelhard

Angiograms depicting embolization of the meningioma seen in Figure 3A. A: Pre-embolization angiogram. The feeding vessel from the middle meningeal artery is clearly seen. B: The meningioma is partially embolized, as indicated by the decreased tumor vascularity. C: The meningioma is completely embolized and the feeding vessel is completely occluded.

4

4. Meningiomas have specific arterial feeders which are characteristic according to the location of the tumor [11,32,45,49,55]. Currently, if the diagnosis is somehow in doubt despite MRI, demonstration of the spokeswheel-like intense blush from an external carotid blood supply is highly suggestive of meningioma. Other disease processes can occasionally have a similar appearance [32,48]. More importantly, cerebral angiography continues to provide valuable information for planning

the surgical attack on a meningioma [32,45,50]. Recent technical improvements have increased the speed, quality, and safety of diagnostic cerebral angiography [53]. Angiography gives the most precise information concerning the degree of tumor involvement of critical vascular structures and the anatomy of arterial feeders. Included in this information is: 1) the arterial supply and site of dural attachment of the tumor; 2) the location/displacement of key vascular structures in relation to the

Meningiomas: Imaging and Embolization

Surg Neurol 2001;55:89 –101

Features of Meningiomas Affecting Surgical Planning/Ease of Resection That Can Be Evaluated by Preoperative Imaging Studies

4

FEATURE

STUDY

Tumor location Tumor size Tumor consistency Tumor vascularity Adherence to neural structures Surgical accessibility Relationship to structures: brain cranial nerves arterial structures veins, sinuses bone Invasion and/or encasement of: brain cranial nerves arterial structures veins, sinuses bone

MRI MRI MRI (T2) MRI, MRA, angiography MRI MRI, CT MRI MRI MRI MRI MRI, CT MRI MRI MRI, MRA, angiography MRI, MRV, angiography CT

tumor, such as cortical draining veins and “arteries of passage,” that is, major cerebral arteries supplying brain beyond the tumor; 3) the degree of invasion, encasement and/or occlusion of vascular structures, such as venous sinuses and arteries of passage; 4) the nature of the arterial blood supply of the tumor: dural, pial or mixed; and 5) the precise degree of tumor vascularity [4,25,32,55]. Precise determination of the sources of blood supply to the tumor (Figure 4) continues to be the major reason for performing preoperative angiography. By attacking the blood supply first, the removal of the rest of the tumor becomes virtually bloodless and easier. For tumors in some locations, conclusively establishing the lack of patency of a venous sinus may also be critical, so that it can be removed along with the tumor to achieve a total resection. Angiography can also be used to perform a trial (i.e., balloon) occlusion of a potentially functional artery that may need to be sacrificed to completely resect a particular tumor, such as a medial sphenoid wing meningioma [2,11,17,49]. Before the occlusion is conducted, the angiogram gives information about the collateral circulation of the brain. Angiography is also useful for assessing the feasibility of preoperative embolization. Intravenous digital subtraction angiography and MR angiography (at 1.5T) have not been as reliable in duplicating the angiographic findings of meningiomas [48]. Table 4 summarizes the various features of meningiomas that are important for surgical planning, and the neurodiagnostic studies that can be used to evaluate them.

97

In weighing whether or not to recommend angiography, knowledge of the risks of the procedure is important. Complications from angiography can be divided into three categories: local (such as groin hematoma), systemic (such as renal failure and allergic reaction to the contrast agent), and neurologic [53]. While the incidence of complications of contemporary cerebral angiography overall has been reported, most of the angiograms in such series were not performed specifically in patients with brain tumors. Heiserman et al in a study of 1,000 consecutive patients undergoing cerebral angiography found a 1% incidence of neurologic deficits related to the procedure, with about half of these being persistent. All the complications occurred in patients presenting with a history of stroke, transient ischemic attack, and/or carotid bruit. However, only 40 of the 1000 angiograms were performed in patients with tumors [27]. A recent meta-analysis of cerebral angiography performed in neurovascular patients put the risk of permanent neurologic deficit at only 0.07%, and serious non-neurologic complications at 0.6%, rates that were much lower than expected [10]. Factors that correlated with adverse events included age, volume of contrast used, length of the procedure, use of multiple catheters, and the presence of systolic hypertension [10]. In the 1970s, Mani and Eisenberg analyzed 5,000 catheter cerebral arteriograms, and found that the complication rate was lower for patients with tumors than for those with cerebrovascular occlusive disease or subarachnoid hemorrhage [43]. Gelman et al studied 48 patients who underwent 392 consecutive angiograms for intra-arterial chemotherapy for brain tumors. Ten groin hematomas, two asymptomatic carotid arterial dissections, and seven transient neurologic events were among the complications encountered; the overall complication rate (8.4%, including those attributable to the chemotherapy itself) was judged to be low [21]. Taking this information together, it would seem that the risk of a significant complication from angiography in a tumor patient—in trained hands—is extremely low, but not nonexistent.

Preoperative Embolization of Meningiomas: Current Status Meningiomas typically have a rich blood supply from the dura and bone adjacent to them [2]. Accordingly, embolization at the time of cerebral an-

98

Surg Neurol 2001;55:89 –101

giography has been used for many years to reduce the blood supply to meningiomas and other vascular tumors. Recent studies continue to illustrate the benefits of preoperative embolization of meningiomas [6,13,47]. Preoperative embolization can be used to make a meningioma smaller, softer, and less bloody; accordingly, it becomes easier to manipulate and remove at the time of surgery. Theoretically, this should shorten the operative time, reduce blood loss, and reduce the potential for brain injury related to brain retraction and manipulation [2,17,25,45,49,55]. Angiography is the most useful test for assessing the feasibility of preoperative embolization because it provides the best visualization of feeding vessels and potentially dangerous anastomoses [55]. At present, medical centers vary widely in their use of cerebral angiography, and in many locations embolization might not be available at all. In major centers where large numbers of interventional procedures are performed, mainly for aneurysms and arteriovenous malformations, angiography may be commonplace and embolization routinely performed. In smaller hospitals, cerebral angiography may no longer be performed at all, given the progress that has been made in diagnosing and localizing tumors by CT and MR imaging. How, then, does one decide whether or not to recommend preoperative embolization? This topic continues to be an area of controversy in neurosurgery [54]. Most meningiomas do not need to be embolized—a proximal vascular occlusion can be performed intraoperatively, if it is needed. For a given patient, the benefit of potentially devascularizing the tumor preoperatively has to be weighed against the possible complications of embolization. At our institution, each patient’s preoperative studies are reviewed with the interventional neuroradiologist to ascertain the feasibility of embolizing a particular meningioma and the risks involved. Refinements in interventional neuroangiography techniques now allow superselective catheterization of tumor vessels [45]. For meningiomas, these usually originate from the external carotid artery. Feeding vessels can be occluded with a variety of agents including Gelfoam powder, polyvinyl alcohol (PVA) foam (with particles of various diameters), lyophilized dura, Silastic microspheres, n-butyrylcyanoacrylate (acrylic glue), fibrin glue, and/or detachable coils [1,11,45,52,61]. The goal of the endovascular approach is to reach the tumor’s capillary bed, then obliterate the arterial and arteriolar feeders while preserving normal brain, cranial nerves and scalp vessels [2,49,55]. If the vessel

Engelhard

feeding the tumor cannot be selectively catheterized, embolization is usually not attempted. Meningiomas in specific locations have specific arterial feeders, which have been carefully described and tabulated [11,45,49,55]. Internal carotid artery branches supplying the tumor are usually not embolized because of the risk of stroke [2]. The interventional neuroradiologist must have a thorough knowledge of the vascular supply to the meninges, the anatomy of potentially dangerous anastomoses between the external and internal carotid artery vascular territories [45], and functional neuroanatomy in relation to the vascular supply of the brain. Usually such knowledge is obtained during the course of an interventional neuroradiology fellowship. “Endovascular surgical neuroradiology” is now recognized as a subspecialty by the Accreditation Council for Graduate Medical Education. Even in the major centers, angiography is not without its risks, as described above. Embolization procedures by their very nature carry additional risk. Neurologic deficit is a possibility if the embolic material gets into vessels feeding cranial nerves or into the brain itself [55]. Embolization of scalp vessels may cause ischemic necrosis of the scalp flap. The microcatheters can get caught or break during the procedure if acrylic glue is used [31]. Other reported complications have included pain, trismus, and various types of hemorrhage (subarachnoid, peritumoral, intratumoral) [49,55]. Despite all these possibilities, the overall risk of a complication in embolizing a meningioma is low [1]. In the study published by Probst et al, two out of 80 (2.5%) patients had neurologic deficits after the procedure. Both of these involved cranial nerves; one of these had resolved by 1 year after the procedure, and one persisted beyond one year [52]. Preoperative embolization of meningiomas is therefore still considered an option for use in selected cases. Some authors feel it is of benefit for treating convexity and skull base lesions [4]. Others feel it is particularly useful for surgically difficult tumors, such as those at the skull base [17,38,49] and for tumors predicted to be very vascular or firm based on T2 MR imaging [38,65]. This is provided embolization can be accomplished without undue risk of infarcting normal brain and/or cranial nerves; predominant vascular supply from the external carotid artery is preferred [38]. Optic nerve meningiomas, for example, which are usually supplied by branches of the ophthalmic artery, are usually not embolized because of the risk to the optic nerves [55]. The middle meningeal artery may provide blood supply to the eye and/or give collaterals to the seventh cranial nerve at the level of the

Meningiomas: Imaging and Embolization

foramen spinosum. To avoid complications, the microcatheter must be advanced far past any dangerous anastomoses [G. Debrun, personal communication]. Blood supply from the pia can limit the effectiveness of embolization [49]. Considering all of this, at our institution embolization is likely to be recommended for a meningioma if it is highly vascular and most of the vascular supply comes from meningeal vessels safely accessible with superselective catheterization. Embolization is especially useful for larger tumors or if it is anticipated that the blood supply will be reached only at the end of the operation [52]. Embolization is conducted the day before surgery with solid particles, not acrylic glue, because we feel particles are safer. Rarely does the tip of the microcatheter reach the tumor itself [G. Debrun, personal communication]. Embolization is usually conducted under general anesthesia to prevent movement. The particles have to be injected slowly to avoid dangerous reflux. The most effective embolization (Figure 4) occurs with the most distal loading of the vascular bed [52]. Wedging the catheter into the tumor vessels should be avoided because it can cause the tumor to rupture and/or hemorrhage [55]. One must be particularly cautious when embolizing very large meningiomas associated with edema and/or shift of the intracranial contents. The embolization procedure could cause additional swelling and subsequent neurologic deficit. The patient must be optimized in terms of his or her hydration status and steroid administration [49,61]. It may be preferable to perform the surgery immediately after embolization for these large meningiomas with mass effect [G. Debrun, personal communication]. Recurrent and malignant meningiomas are usually more difficult to embolize [49]. Observation in the Neurosurgical Intensive Care Unit after the embolization procedure is performed is recommended. Authors differ as to when embolization should be conducted before surgery [1,52,55]. At our Institution, surgery is usually performed the day after the embolization procedure. Others have recommended waiting 3 to 5 days [17,49,55] or even 1 to 2 weeks [1]. These recommendations might depend on the type of material used for the embolization [17]. After embolization, the success of the procedure can be estimated by the post-embolization angiogram (Figure 4) and also by post-contrast MRI [49,61]. MRI may underestimate the effect of the procedure on the tumor, however. If more precise information is needed, mapping the relative regional cerebral blood volume seems to be a superior technique [6]. If embolization is not used or is unsuccessful, interruption of the tumor’s blood supply is often an

Surg Neurol 2001;55:89 –101

99

initial step in its surgical removal. Embolization alone, without subsequent surgery, has been attempted to slow tumor growth in cases in which surgical treatment is contraindicated [1–3,36,55]. Embolization—perhaps combined with chemotherapy and/or medications that inhibit angiogenesis— might someday develop into a primary treatment modality for selected tumors [1,6].

Conclusions Significant advances have been made in the preoperative diagnosis and characterization of meningiomas over the past decade. While no imaging feature is as yet pathognomonic for meningioma, the diagnosis and surgical anatomy are almost always known preoperatively with a high degree of certainty. Each of the three major imaging modalities continues to have a role in evaluating meningioma patients: CT for patient screening and evaluating bony anatomy, MRI for discerning anatomic relationships, and intra-arterial angiography for answering critical questions related to vascular anatomy. T2-weighted MR images may be useful for predicting tumor consistency, with the presence of hyperintensity indicating a softer tumor. The correlation of radiologic features with histology— including the degree of malignancy— continues to be imperfect [2]. Postoperative screening for residual foci of meningioma is accomplished by obtaining a pre- and post-contrast MRI within 3 days of surgery. The role of the newer MR techniques continues to evolve, as technological advances continue to be made. MRA can be helpful in indicating the tumor’s blood supply, while MRV can be used to demonstrate sinus patency. Cerebral angiography is still superior to MRI at 1.5T for defining vascular structures [35]. Angiography can be used to precisely identify the sources of blood supply to the tumor, thus providing valuable information for planning the surgical attack. While the complication rate of cerebral angiography in patients with tumors is not precisely known, based on the available literature it should certainly be very low (less than 1%). Preoperative embolization can be used to reduce a tumor’s blood supply preoperatively, potentially making it less vascular, softer, and easier to resect. Neurosurgeons who have not operated on an embolized meningioma may not be aware of the benefits of this procedure [11]. Embolization is usually not needed, however, if the blood supply can be easily interrupted during the approach to the tumor.

100 Surg Neurol 2001;55:89 –101

The next article in this series will review the topics of the “incidental” meningioma and its natural history, radiosurgery for meningiomas, and the occurrence and management of seizures in meningioma patients. The author wishes to thank Dr. Mahmood Mafee and Dr. Gerard Debrun for their thoughtful suggestions regarding the content of this manuscript, and Jill Hohbein for helping with the references.

references 1. Ahuja A, Gibbons KJ, Hopkins LN. Endovascular techniques to treat brain tumors. In: Youmans JR, ed. Neurological Surgery. Philadelphia: W.B. Saunders Co. 1996:2826 – 40. 2. Akeyson EW, McCutcheon IE. Management of benign and aggressive intracranial meningiomas. Oncology 1996;10:747–56. 3. Bendszus M, Martin–Schrader I, Warmuth–Metz M, Hoffman E, Solymosi L. MR imaging and MR spectroscopy-revealed changes in meningiomas for which embolization was performed without subsequent surgery. Am J Neuroradiol 2000;21:666 –9. 4. Black PM. Meningiomas. Neurosurgery 32; 1993:643– 57. 5. Braunstein JB, Vick NA. Meningiomas: the decision not to operate. Neurology 1997;48:1459 – 62. 6. Bruening R, Wu RH, Yousry TA, Berchtenbreiter C, Weber J, Peller M, Steiger HJ, Reiser M. Regional relative blood volume MR maps of meningiomas before and after partial embolization. J Computer Assisted Tomography 1998;22:104 –10. 7. Burke JW, Podransky AE, Bradley WG. Meninges: benign postoperative enhancement on MR images. Radiology 1990;174:99 –102. 8. Castillo M, Kwock L. Clinical applications of proton magnetic resonance spectroscopy in the evaluation of common intracranial tumors. Top Magn Reson Imaging 1999;10:104 –13. 9. Chen TC, Zee C-S, Miller CA, et al. Magnetic resonance imaging and pathological correlates of meningiomas. Neurosurgery 1992;31:1015–22. 10. Cloft HJ, Joseph GJ, Dion JE. Risk of cerebral angiography in patients with subarachnoid hemorrhage, cerebral aneurysm, and arteriovenous malformation: a meta-analysis. Stroke 1999;30:317–20. 11. Connors III JJ, Wojack JC. Meningiomas. In: Connors III JJ, Wojack JC, eds. Interventional Neuroradiology: Strategies and Practical Techniques. Philadelphia: W.B. Saunders Co. 1999:100 –21. 12. Cushing H, Eisenhardt L. Meningiomas: their classification, regional behavior, life history and surgical end results. Springfield, IL: Charles C. Thomas, 1938. 13. Dean BL, Flom RA, Wallace RC, Khayata MH, Obuchowski NA, Hodak JA, Zabramski JM, Spetzler RF. Efficacy of endovascular treatment of meningiomas: evaluation with matched samples. Am J Neuroradiol 1994;15:1675– 80. 14. DeLaPaz RL. Advances in brain tumor diagnostic imaging. Curr Opin Neurol 1995;8:430 – 6. 15. DeMonte F. Current management of meningiomas. Oncology 1995;9:83–96.

Engelhard

16. DeMonte F, Al–Mefty O. Meningiomas. In: Kaye AH, Laws ER, eds. Brain Tumors: An Encyclopedic Approach. New York: Churchill Livingstone, 1995:675– 704. 17. Desai R, Bruce J. Meningiomas of the cranial base. J Neuro-Oncol 1994;20:255–279. 18. Dolinskas CA, Simeone FA. Surgical site after resection of a meningioma. Am J Neuroradiol 1998;19:419 – 26. 19. Engelhard HH. Current role of radiation therapy and radiosurgery in the management of meningiomas. Neurosurgical Focus 1997;2(4):1– 4. 20. Feldman RP, Marcovici A, Suarez M, Goodrich JT. Foreign body granuloma mimicking intracranial meningioma: case report and review of the literature. Neurosurgery 1999;44:855– 8. 21. Gelman M, Chakeres DW, Newton HB. Brain tumors: Complications of cerebral angiography accompanied by intra-arterial chemotherapy. Radiology 1999;213: 135– 40. 22. Ginsberg LE, Moody DM. Meningiomas: imaging. In: Wilkins RH, Rengachary SS, eds. Neurosurgery, 2nd Ed. New York: McGraw–Hill, 1996:855–72. 23. Goldsher D, Litt AW, Pinto RS, Bannon KR, Kricheff II. Dural “tail” associated with meningiomas on GdDTPA-enhanced MR images: characteristics, differential diagnostic value and possible implications for treatment. Radiology 1990;176:447–50. 24. Good CD, Kingsley DP, Taylor WJ, Harkness WF. “Dural tail” adjacent to a giant posterior cerebral artery aneurysm: case report and review of the literature. Neuroradiology 1997;39:577– 80. 25. Greenberg HS, Chandler WF, Sandler HM. Brain Tumors. New York: Oxford University Press, 1999:275– 87. 26. Haddad G, Al–Mefty O. Meningiomas: an overview, in Wilkins RH, Rengachary SS, eds. Neurosurgery, 2nd Ed. New York: McGraw–Hill, 1996:833– 41. 27. Heiserman JE, Dean BL, Hodak JA, et al. Neurologic complications of cerebral angiography. Am J Neuroradiol 1994;15:1401–7. 28. Hutzelmann A, Palmie A, Buhl R, Freund M, Heller M. Dural invasion of meningiomas adjacent to the tumor margin on Gd-DPTA-enhanced MR images: histopathologic correlation. Eur Radiol 1998;8:746 – 8. 29. Ide M, Jimbo M, Kubo O, et al. Peritumoral brain edema and cortical damage by meningioma. Acta Neurochir 1994 (Suppl.)60:369 –72. 30. Ildan F, Tuna M, Gocer AI, Boyar B, Bagdatoglu H, Sen O, Haciyakupoglu S, Burgut HR. Correlation of the relationships of brain-tumor interfaces, magnetic resonance imaging, and angiographic findings to predict cleavage of meningiomas. J Neurosurg 1999;91:384 – 90. 31. Inci S, Ozcan OE, Benli K, Saatci I. Microsurgical removal of a free segment of microcatheter in the anterior circulation as a complication of embolization. Surg Neurol 1996;46:562– 6. 32. Jacobs JM, Harnsberger HR. Diagnostic angiography and meningiomas. In: Al–Mefty O, ed. Meningiomas. New York: Raven Press 1991:225– 40. 33. Kaibara T, Tyson RL, Sutherland GR. Human cerebral neoplasms studied using MR spectroscopy: a review. Biochem Cell Biol 1998;76:477– 86. 34. Kattner KA, Stroink AR, Roth TC, Lee JM. Rosai–

Meningiomas: Imaging and Embolization

35. 36.

37. 38. 39. 40.

41. 42.

43.

44.

45. 46. 47.

48. 49.

50. 51.

Dorfman disease mimicking parasagittal meningioma: case presentation and review of literature. Surg Neurol 2000; 53:452–7. Kizana E, Lee R, Young N, Dorsch NW, Soo YS. A review of the radiological features of intracranial meningiomas. Australas Radiol 1996;40:454 – 62. Koike T, Sasaki O, Tanaka R, Arai H. Long-term results in a case of meningioma treated by embolization alone— case report. Neurol Med Chir (Tokyo) 1990; 30:173–7. Larson JJ, Tew JM, Wiot JG, deCourten–Myers GM. Association of meningiomas with dural “tails”: surgical significance. Acta Neurochir 1992;114:59 – 63. Latchaw RE. Preoperative intracranial meningioma embolization: technical considerations affecting riskto-benefit ratio. Am J Neuroradiol 1993;14:583– 6. Latchaw RE, Hirsch WL. Computerized tomography of intracranial meningiomas. In: Al–Mefty O, ed. Meningiomas. New York: Raven Press 1991:195–207. Lobato RD, Alday R, Gomez PA, Rivas JJ, Dominguez J, Cabrera A, Madero S, Ayerbe J. Brain oedema in patients with intracranial meningioma. Correlation between clinical, radiological, and histological factors and the presence and intensity of oedema. Acta Neurochir 1996;138:485–93. Mafee MF, Goodwin J, Dorodi S. Optic nerve sheath meningiomas: role of MR imaging. Radiol Clin North Am 1999;37:37–58. Maiuri F, Iaconetta G, de Divitiis O, Cirillo S, Di Salle F, De Caro ML. Intracranial meningiomas: correlations between MR imaging and histology. Eur J Radiol 1999;31:69 –75. Mani RL, Eisenberg RL. Complications of catheter cerebral angiography: analysis of 5,000 procedures. II. Relation of complication rates to clinical and arteriographic diagnoses. Am J Neuroradiol 1978;131: 867–9. McCutcheon TM, Weber P, Smith T, Patel S. Skull base lesions mimicking meningioma: can surgery be avoided? Presented at: Annual Meeting of the Congress of Neurological Surgeons; October 3– 8, 1998; Seattle, Washington. McDermott MW, Wilson CB. Meningiomas. In: Youmans JR, ed. Neurological Surgery, Fourth Edition. Philadelphia: W. B. Saunders Co. 1996:2782– 825. Mighell AJ, Stassen LF, Soames JV. Meningioma—an unusual forehead swelling: a case report. Br J Oral Maxillo Surg 1994;32:253– 6. Morimura T, Takeuchi J, Maeda Y, Tani E. Preoperative embolization of meningiomas: its efficacy and histopathological findings. Noshuyo Byori 1994;11: 123–9. Murtagh R, Linden C. Neuroimaging of intracranial meningiomas. In: Ransohoff J, ed. Neurosurg Clin North Am 1994;5:217–33. Nelson PK, Setton A, Choi IS, et al. Current status of interventional neuroradiology in the management of meningiomas. In: Ransohoff J, ed. Meningiomas. Neurosurg Clin North Am 1994;5:235–59. Ojemann R. Meningiomas. Neurosurg Clin North Am 1990;1:181–97. Ojemann RG. Supratentorial meningiomas: clinical features and surgical management. In: Wilkins RH,

Surg Neurol 101 2001;55:89 –101

52.

53.

54.

55. 56. 57.

58.

59.

60. 61.

62.

63.

64.

65.

66.

Rengachary SS, eds. Neurosurgery, 2nd Ed. New York: McGraw–Hill, 1996:873–90. Probst EN, Grzyska U, Westphal M, Zeumer H. Preoperative embolization of intracranial meningiomas with a fibrin glue preparation. Am J Neuroradiol 1999; 20:1695–702. Pryor JC, Setton A, Nelson PK, Berenstein A. Complications of diagnostic cerebral angiography and tips on avoidance. Neuroimaging Clin North Am 1996;6: 751– 8. Rao JG, Bendzus M, Schaller C, Burger R, Scheinemann K, Schramm J, Roosen K, Solymosi L. Is there a benefit in the preoperative embolization of meningiomas? Presented at: American Association of Neurological Surgeons Annual Meeting; April 8 –13, 2000; San Francisco, California. Rodesch G, Lasjaunias P. Embolization and meningiomas In: Al–Mefty O, ed. Meningiomas. New York: Raven Press 1991:285–97. Sato M, Matsumoto M, Kodama N. Meningeal enhancement surrounding meningiomas on Gd-DPTA MRI. Fukushima J MedSci 1998;44:1–11. Shino A, Nakasu S, Matsuda M, Handa J, Morikawa S, Inubushi T. Noninvasive evaluation of the malignant potential of intracranial meningiomas performed using proton magnetic resonance spectroscopy. J Neurosurg 1999;91:928 –34. Suzuki Y, Sugimoto T, Shibuya M, Sugita K, Patel SJ. Meningiomas: correlation between MRI characteristics and operative findings including consistency. Acta Neurochir 1994;129:39 – 46. Sze G, Solitsky S, Bronen R, Krol G. MR imaging of the cranial meninges with emphasis on contrast enhancement and meningeal carcinomatosis. Am J Neuroradiol 1989;10:965–75. Van Hemert RL. MRA of cranial tumors and vascular compressive lesions. Clin Neurosci 1997;4:146 –52. Wakhloo AK, Juengling FD, Van Velthoven V, Schumacher M, Hennig J, Schwechheimer K. Extended preoperative polyvinyl alcohol microembolization of intracranial meningiomas: assessment of two embolization techniques. Am J Neuroradiol 1993;14: 571– 82. Weingarten K, Ernst RJ, Jahre C, Zimmerman RD. Detection of residual or recurrent meningioma after surgery: value of enhanced vs. unenhanced MR imaging. Am J Radiol 1992;158:645–50. Witt TC, Kondziolka D, Baumann SB, Noll DC, Small SL, Lunsford LD. Preoperative cortical localization with functional MRI for use in stereotactic radiosurgery. Stereotact Funct Neurosurg 1996;66:24 –9. Yamada K, Sorensen AG. Diagnostic imaging. In: Bernstein M, Berger MS, eds. Neuro-Oncology: The Essentials. New York: Thieme Medical Publishers, Inc., 2000:56 –77. Yamaguchi N, Kawase T, Sagoh M, Ohira T, Shiga H, Toya S. Prediction of consistency of meningiomas with preoperative magnetic resonance imaging. Surg Neurol 1997;48:579 – 83. Zimmerman RD. MRI of intracranial meningiomas. In: Al–Mefty O, ed. Meningiomas. New York: Raven Press 1991:209 –23.