In patients with progressive neurologic signs and symptoms suggestive of a vestibular schwannoma and cerebellopontine angle pathology, neuroimaging with MRI is the most critical diagnostic test (Jackler and Pitts 1990; Mafee et al 1990; Curtin and Hirsch 1992; Smirniotopoulos et al 1993; McKenzie 1994; Macfarlane and King 1995; Mafee 1995; Slattery et al 2003). These studies should be performed with and without contrast media for the most accurate visualization of the mass, in order to assess the relationship of the mass to the internal auditory canal and other cerebellopontine angle anatomy, as well as to assist in differential diagnosis. In general, both CT and MRI have excellent sensitivity for tumors with a significant component beyond the plane of the porus acusticus and within the cerebellopontine angle cistern (Curtin and Hirsch 1992). However, MRI is consistently more sensitive than CT for small intracanalicular lesions, and is now considered the imaging modality of choice for screening patients for the presence of a vestibular schwannoma (Jackler and Pitts 1990; Curtin and Hirsch 1992; Smirniotopoulos et al 1993; McKenzie 1994; Macfarlane and King 1995; Slattery et al 2003). An MRI scan that does not show enhancement or a mass within the internal auditory canal rules out the diagnosis of a vestibular schwannoma.
On T1-weighted unenhanced MRI, vestibular schwannomas appear isointense or slightly hypointense relative to brain and are hyperintense compared to cerebrospinal fluid (Curtin and Hirsch 1992; Smirniotopoulos et al 1993; Macfarlane and King 1995). After administration of contrast, intracanalicular and small cisternal tumors enhance diffusely. Most authors agree that the use of contrast significantly improves the detection rate for intracanalicular schwannomas (Welling et al 1990; Macfarlane and King 1995). Tumors as small as 2 to 3 mm can be clearly delineated from surrounding structures. Large cisternal tumors may have heterogeneous enhancement due to the presence of cystic regions and hemorrhage. On T2-weighted MRI, the tumor is hyperintense relative to brain, and may be isointense with cerebrospinal fluid. Some authors feel fast spin echo T2-weighted MRI is an equivalent screening method for acoustic schwannomas, compared to T1-enhanced images. However, a study by Zealley and colleagues demonstrates that fast spin echo T2-weighted imaging only has a 56% confidence rate, compared to T1-enhanced imaging (Zealley et al 2000). Small intracanalicular tumors are often difficult to visualize using only fast spin echo technique. Vestibular schwannomas are usually centered over the internal auditory canal when there is a cisternal component. Tapering of the cisternal mass toward the internal auditory canal and porus acusticus is usually present. Intracanalicular tumors always have their longitudinal axis along the path of the internal auditory canal. The multiplanar capability of MRI enables the tumor and its surrounding anatomical structures to be visualized in much greater detail than CT. Other advantages over CT include the lack of beam-hardening artifact, the ability to identify vascular structures in close proximity to the tumor, and superior contrast resolution. However, CT is superior to MRI for evaluating the anatomy of the internal auditory canal and petrous temporal bone. Some data suggest that serial MRI scanning at the same facility and between different facilities is reliable, with a minimum detectable change in diameter of 1.1 mm and enhancing volume of 0.15 cm2 (Slattery et al 2003).
On unenhanced CT, vestibular schwannomas appear isodense or hypodense, compared to the brain. After contrast administration, most tumors show homogeneous enhancement. Virtually all cisternal tumors and most intracanalicular tumors can be detected with enhanced CT (Curtin and Hirsch 1992; Smirniotopoulos et al 1993; Macfarlane and King 1995). Small intracanalicular tumors may remain undetected due to partial volume averaging from surrounding dense bone. Similar to MRI, the mass is centered on the internal auditory canal and usually has a component tapering into the porus acusticus. In 70% to 90% of cases, erosion and enlargement of the internal auditory canal are evident on bone windows. Calcification within the tumor is usually negligible.
Because the symptoms of hearing loss and vertigo are common in the general population and often not related to a vestibular schwannoma (only 5% to 10% of patients evaluated have a tumor), several nonimaging diagnostic tests have been employed to screen patients for neuroimaging. Most authors recommend a battery of pure tone audiometry, speech discrimination assessment, and auditory evoked brainstem responses (Jackler and Pitts 1990; Selesnick and Jackler 1992; Macfarlane and King 1995). Pure tone audiometry is abnormal in the majority of patients with a vestibular tumor. In 60% to 70%, high-frequency hearing loss is present. Larger tumors are more likely to cause audiometric abnormalities. Although pure tone audiometry may remain intact with up to 75% of cranial nerve fiber loss, speech processing is usually impaired (Selesnick and Jackler 1992). Speech discrimination deficits have been detected in 45% to 80% of patients with vestibular schwannomas. The most sensitive audiological screening method is the auditory evoked brainstem response (Jackler and Pitts 1990; Selesnick and Jackler 1992; Macfarlane and King 1995). The auditory evoked brainstem response is more sensitive and specific than all other nonimaging screening tests. Response latencies can be delayed with cochlear nerve stretching, even when hearing remains normal. The most consistent abnormality of the auditory evoked brainstem response is an interaural difference of greater than 0.3 msec in the latency of wave V (Jackler and Pitts 1990; Selesnick and Jackler 1992; Macfarlane and King 1995). The sensitivity of auditory evoked brainstem response for detecting a vestibular schwannoma is 93% to 98%, with a specificity of 90%. Intracanalicular and small cisternal tumors tend to cause delays in wave V, whereas large cisternal tumors often abolish wave V completely.
The CT and MRI appearance of schwannomas of other cranial nerves are similar to vestibular tumors except for their location (Lye et al 1987; Rigamonti et al 1987; Yuh et al 1988; Weber and McKenna 1994; Samii et al 1995; Strauss and Post 1995; Chung et al 1998; Zhang et al 2009). Trigeminal schwannomas are seen as enhancing masses that usually arise near Meckel cave in the middle fossa, posterior fossa, or both. On CT using bone windows, depending on the location of the tumor, erosion may be found affecting the foramen ovale or foramen spinosum, anteromedial portion of the petrous apex, lateral aspect of the sella turcica, anterior clinoid process, dorsum sellae, or superior orbital fissure. Schwannomas of the facial nerve are difficult to distinguish from vestibular tumors when they develop from the cisternal or canalicular segments of the nerve and are misdiagnosed in 36% of patients (Strauss and Post 1995). In more typical cases, the presence of an enhancing soft tissue mass within 1 or more segments of the fallopian canal is diagnostic of facial tumor (Chung et al 1998). The fallopian canal is often enlarged, which is seen best on CT with bone windows. Schwannomas of the jugular complex (cranial nerves IX, X, XI) are seen on CT and MRI as enhancing masses centered on the jugular foramen (Weber and McKenna 1994; Strauss and Post 1995; Rapana et al 1997). Enlargement of the jugular foramen is common and best visualized on CT with bone windows. Hypoglossal schwannomas present on CT and MRI as enhancing masses ventral to the lower brainstem in the region of the hypoglossal canal (Strauss and Post 1995; Tucker et al 2007). Erosion of the hypoglossal canal is present in some cases. Ocular nerve schwannomas (cranial nerves III, IV, VI) are well visualized with CT or MRI as enhancing masses within the cisternal space ventral to the brainstem or the cavernous sinus. CT with bone windows often demonstrates erosion or scalloping of the clinoids, sella, petrous apex, and superior orbital fissure.
Schwannomas (and neurofibromas) of the spine and peripheral nerves can be visualized with either CT or MRI, but are more clearly delineated with MRI (Schroth et al 1987; Demachi et al 1990; Friedman et al 1992; Hu and Huang 1992; Sanguinetti et al 1993; Newton et al 1995; Lin and Martel 2001; Colosimo et al 2003; Gupta and Maniker 2007). The tumor presents as a round or lobulated, intradural (extradural components are common), extramedullary mass that often compresses the spinal cord. In some cases, the tumor may have a plaquelike appearance, covering several spinal segments. An intradural location is suggested by widening of the anterior subarachnoid space at the margins of the mass. Although uncommon, some spinal schwannomas can be large and invasive, extending into the vertebral bodies and surrounding soft tissues (Sridhar et al 2001). On T1-weighted MRI the tumor is usually isointense or slightly hypointense relative to spinal cord and hyperintense relative to cerebrospinal fluid. Enhancement is often diffuse after contrast administration, but may be heterogeneous in tumors with cystic degeneration, necrosis, or hemorrhage (Schroth et al 1987; Demachi et al 1990; Friedman et al 1992). With T2-weighted images, the signal intensity is variable, depending on the relative amounts of Antoni A and B zones, cystic degeneration, and hemorrhage (Hu and Huang 1992). Most tumors have high signal intensity relative to spinal cord, similar to cerebrospinal fluid. However, tumors that contain hemorrhage or old blood products may have low signal intensity. Although uncommon, cystic schwannomas and neurofibromas have been described and delineated by MRI (Parmar et al 2001). The presence of internal septa, irregularity of walls, differences in the thickness of the walls, and hyperintensity of the cystic contents are suggestive of a nerve sheath tumor. MRI can clearly delineate schwannomas and neurofibromas of the peripheral nerves and differentiate them from surrounding normal soft tissues (Lin and Martel 2001). The rare intramedullary schwannoma can be suspected on MRI by the presence of a small to medium-sized mass that is well circumscribed, is associated with central cord edema and without syringomyelia, and demonstrates marked gadolinium enhancement (Colosimo et al 2003). Benign neurofibromas and malignant peripheral nerve sheath tumors can also be detected with the use of FDG-PET scanning (Gupta and Maniker 2007; Son et al 2007; Ferner et al 2008). Another group has shown that FDG-PET results can be predictive of future growth of plexiform neurofibromas. In a study of 18 patients, tumors with standardized uptake values greater than 2 were significantly more likely to grow in the subsequent year (p = 0.016) in comparison with tumors with lower standardized uptake values. The authors suggest that FDG-PET can be used to predict neurofibroma growth rates (Fisher et al 2008).
A more recent positron emission tomography study using 2-deoxy-2-fluoro-(18)F-D-glucopyranose (18F-FDG) attempted to analyze the factors involved in glucose transport and vascular formation in a series of patients with cranial and spinal nerve schwannomas (Hamada et al 2009). The standardized uptake value correlated with tumor size (< 5 cm vs. > 5 cm; p < 0.05) and microvascular density (negative vs. positive; p < 0.05). The retention index positively correlated with the expression of vascular endothelial growth factor in the tumors (negative vs. positive; p < 0.05). Glucose transporter protein expression (ie, Glut-1, Glut-3) did not correlate with the standardized uptake value or the retention index.
Angiography and myelography are less useful in the era of CT and MRI, and are usually not required for small intracranial or spinal schwannomas (Jackler and Pitts 1990; Samii et al 1995; Seppala et al 1995b; Strauss and Post 1995). Angiography may be helpful preoperatively if an aneurysm or malformation is suspected in the differential diagnosis. For large tumors, angiography is often necessary to delineate regional vascular anatomy.