Intradural schwannomas of the spine: MR findings with emphasis on contrast-enhancement characteristics.

Intradural extramedullary schwannomas are nerve sheath neoplasms that consist of focal proliferations of Schwann cells involving a spinal nerve. We reviewed the MR findings in seven patients with pathologically proved intradural schwannomas. The contrast-enhancement characteristics on MR images were determined and compared with the histologic features of the tumor. Six lesions were variably hyperintense on T2-weighted images and one was uniformly hypointense compared with the signal intensity of the spinal cord. Signal on T1-weighted images ranged from hypointense to isointense. All seven tumors showed heterogeneous enhancement; in five, the enhancement involved only the periphery of the lesion. The pattern of enhancement did not correlate with the signal characteristics noted on unenhanced T1- and T2-weighted images. Pathologically, hyaline thickening of vessel walls and cyst formation were prevalent in the peripherally enhancing lesions. However, enhancement did not correlate with the relative proportion of Antoni type A and type B tissue. Recognition of the MR characteristics of intradural extramedullary schwannomas may be helpful in the differential diagnosis of spinal tumors. In particular, peripheral contrast enhancement of an intradural extramedullary tumor on MR images should suggest the diagnosis of schwannoma.     

Spinal intradural capillary hemangioma: MR findings

We report a case of a spinal intradural capillary hemangioma. On MR images, a well-circumscribed intradural mass was detected at the T8-T9 level. The signal intensity of the mass relative to the spinal cord was isointense on T1-weighted images, hyperintense on T2-weighted images, and showed homogeneous, strong enhancement on contrast-enhanced T1-weighted images. The mass had both extramedullary and intramedullary components.     

MR imaging of intradural extramedullary tumors.

Thirty-one consecutive intradural extramedullary spinal tumors examined with MR at 0.3 T were reviewed. In 13 of the patients myelography had been performed. There were 11 patients with meningeoma, 14 with neuroma, one ependymoma, 3 metastases, and 2 lipomas. All tumors were surgically removed and verified by histology. The intradural extramedullary location of the tumors was accurately assessed by MR imaging in all patients and by myelography in 10 of 13. The MR diagnoses were in accordance with the histologic findings in 74% of cases. Compression of the spinal cord or cauda equina with widening of the subarachnoid space above and below the mass or outward displacement of epidural fat was characteristic of the intradural extramedullary tumors. The signal intensity of meningeoma as well as of neuroma was slightly lower or equal to that of the cord on T1-weighted images, and equal to or higher than cord signal on T2-weighted images. Neuroma had a lower signal intensity on T1-weighted images and a higher signal intensity on T2-weighted images than meningeoma. Meningeoma appeared more homogeneous than neuroma and had a broad base towards the dura.     

Radiological findings of spinal schwannomas and meningiomas: focus on discrimination of two disease entities

This study evaluates the effectiveness of CT and MR imaging in differentiating intradural extramedullary spinal schwannomas and meningiomas in a large group of patients. In addition, the study correlates tumour location, morphologic characteristics and enhancement pattern. From January 2000 to June 2007, we retrospectively reviewed 128 consecutive patients (51 male, 77 female; mean age at admission 53.8 years; range 17–83 years) with spinal intradural extramedullary tumours (92 schwannomas, 36 meningiomas) at our institution. Fifty-one of ninety-two schwannomas (55.4%) showed fluid signal intensity on T2-weighted MR images. Twenty-two of thirty-six meningiomas (61.1%) showed hyperintense signal intensity and thirteen of thirty-six meningiomas (36.1%) showed isointense signal on T2-weighted MR images. Fifty-four schwannomas (58.7%) showed rim enhancement and thirty-three meningiomas (91.7%) showed diffuse enhancement on contrast-enhanced T1-weighted MR imaging. Twenty-one meningiomas (58.3%) showed dural tail sign in contrast-enhanced T1-weighted MR imaging. Twenty-one meningiomas (58.3%) showed calcification on CT images. MR and CT imaging results are therefore useful for the differentiation of schwannomas from meningiomas of the spine.     

Imaging of tumors of the spine and spinal cord

Imaging of the spine and spinal cord has traditionally been accomplished with plain radiography, myelography, and CT. Recently, MR imaging has become the technique of choice in the assessment of lesions of the spine and spinal cord. MR imaging provides accurate localization of intramedullary, intradural extramedullary, and extradural tumors. Ependymomas and low-grade astrocytomas are the most common intramedullary tumors. MR imaging findings are distinguishable by the delineation and size of the lesion, and the signal intensity on T2-weighted images. Other less common tumors include malignant astrocytomas, hemangioblastomas, and intramedullary metastasis. Numerous foci of high-velocity signal loss are seen in the hemangioblastomas. Metastasis, meningiomas, and schwannomas are the most common intradural extramedullary tumors. Meningiomas are characterized by dural enhancement on postcontrast T1-weighted images. Schwannomas and neurofibromas often erode bony structures and appear to be dumbbell-shaped. Epidural metastasis accounts for the majority of extradural tumors. Primary malignant extradural tumors include lymphomas, chordomas, and so on. The most common primary benign extradural tumor is hemangioma, which often appears to be hyperintense on both T1-weighted and T2-weighted images. Intramedullary non-neoplastic lesions include demyelinating, vascular, and infectious diseases. Diffuse, peripheral, or speckled contrast enhancement, and lack of contrast enhancement may suggest non-neoplastic lesions.     

Application of turbo fluid-attenuated inversion-recovery MRI in evaluation of intraspinal tumors

PURPOSE: Although fluid-attenuated inversion-recovery (FLAIR) magnetic resonance imaging (MRI) is widely applied to diagnose central nervous system diseases, its role in diagnosis of intraspinal tumors is unclear. In this study, we evaluated the potential clinical application of a turbo FLAIR sequence for imaging of intraspinal tumors. MATERIALS AND METHODS: Forty-eight consecutive patients with intraspinal tumors underwent MRI with turbo FLAIR and turbo spinal echo (TSE) sequences. Turbo FLAIR images were then qualitatively and quantitatively compared with T2-weighted TSE images. RESULTS: Turbo FLAIR images were evaluated as superior to T2-weighted TSE images for image artifact, extradural tumor conspicuity, and intradural extramedullary tumor conspicuity and detection. Intramedullary tumor conspicuity with turbo FLAIR was less than T2-weighted TSE. Similar capabilities in detection of extradural and intramedullary tumors were found between turbo FLAIR and T2-weighted TSE. Turbo FLAIR and T2-weighted TSE displayed similar normal spinal cord signal-noise ratio (SNR) and tumor-to-cerebrospinal fluid (CSF) contrast-to-noise ratio (CNR). In addition, turbo FLAIR yielded significantly higher tumor-to-CSF contrast than T2-weighted TSE. However, tumor SNR, tumor-to-normal spinal cord contrast and CNR with turbo FLAIR images were lower than those with T2-weighted TSE images. CONCLUSION: This study demonstrated (a) a superiority of turbo FLAIR to T2-weighted TSE in displaying and detecting intradural extramedullary tumors, (b) a superiority of turbo FLAIR to T2-weighted TSE in demonstrating extradural tumors, and (c) less usefulness in displaying intramedullary tumors with turbo FLAIR than with T2-weighted TSE.     

MR imaging findings in spinal infections: rules or myths?

PURPOSE: To systematically evaluate magnetic resonance (MR) imaging findings described as being indicative of spinal infection in patients with proven spinal infection. MATERIALS AND METHODS: Contrast material-enhanced spinal MR images obtained in 46 consecutive patients (22 women, 24 men; mean age, 58.2 years) with culture or histologic examination results positive for spinal infection were systematically evaluated by two observers. Tuberculous and postoperative infections were excluded. Disk signal intensity and disk height, presence of the nuclear cleft, vertebral signal intensity alterations, endplate erosions on T1-weighted MR images, and presence of paraspinal or epidural inflammation were evaluated. Patient charts and surgical reports were reviewed. RESULTS: In the 44 patients with disk infection, MR imaging criteria with good to excellent sensitivity included presence of paraspinal or epidural inflammation (n = 43, 97.7% sensitivity), disk enhancement (n = 42, 95.4% sensitivity), hyperintensity or fluid-equivalent disk signal intensity on T2-weighted MR images (n = 41, 93.2% sensitivity), and erosion or destruction of at least one vertebral endplate (n = 37, 84.1% sensitivity). Effacement of the nuclear cleft was only applicable in 18 patients (n = 15, 83.3% sensitivity). Criteria with low sensitivity included decreased height of the intervertebral space (n = 23, 52.3% sensitivity) and disk hypointensity on T1-weighted MR images (n = 13, 29.5% sensitivity). Involvement of several spinal levels occurred in seven (16%) patients. Other spinal infections included isolated vertebral osteomyelitis (n = 1) and primary epidural abscess (n = 1). CONCLUSION: Most MR imaging criteria commonly used to diagnose disk infections offer good to excellent sensitivity. In atypical manifestations of proven spinal infections, however, some of the classically described MR imaging criteria may not be observed.     

Importance of absence of CSF pulsation artifacts in the MR detection of significant myelographic block at 1.5 T

MR imaging was performed on 21 patients who had high-grade myelographic block due to various diseases in all spinal compartments (extradural, intradural/extramedullary, and intramedullary) and in all portions of the spinal canal (cervical, thoracic, and lumbosacral). Loss of CSF pulsation artifacts due to significant compression of the spinal cord was demonstrated on non-motion-compensated T2-weighted examinations in each case. We believe that the absence of such artifacts on these sequences indicates significant spinal cord compression in patients without classic signs and symptoms of cord compression but with intraspinal disease identified on T1-weighted studies.     

MR imaging of intraspinal tumors—Capability in histological differentiation and compartmentalization of extramedullary tumors

Summary Magnetic resonance (MR) images of 29 consecutive patients with intraspinal neoplasms (9 intramedullary tumors, 20 extramedullary tumors) were reviewed to evaluated the utility of MR imaging in distinguishing the intraspinal compartmental localisation and signal characteristics of each lesion. Compartment and histology of all neoplasms were surgically proven. MR correctly assigned one of three compartments to all lesions, 9 intramedullary, 14 intradural extramedullary (6 schwannomas, 3 neurofibromas, 5 meningiomas), and 6 extradural (3 schwannomas, 1 meningioma, 1 cavernous hemangioma, 1 metastatic renal cell carcinoma). All intramedullary tumors showed swelling of the spinal cord itself. In all five extradural tumors a low intensity band was visualized between the spinal cord and tumor. On the other hand, a low intensity band was demonstrated in no cases with intradural tumors. Visualization of this low intensity band is important in differentiating extradural from intradural-extramedullary lesions. We call this low intensity band, the extradural sign. Signal intensity of intradural tumors varied with histology. In extramedullary tumors, signal intensity of schwannomas was similar to that of the cerebrospinal fluid (CSF) both on T1 weighted (inversion recovery) and T2 weighted spin echo (SE) images. On the other hand, meningiomas tended to be isointense to the spinal cord on both T1 and T2 weighted SE images. We found relatively reliable signal characteristics to discriminate meningioma from schwannoma.     

Intradural spinal metastases in pediatric patients with primary intracranial neoplasms: Gd-DTPA enhanced MR vs CT myelography

Recently, the application of intravenous gadolinium-diethylenetriamine pentaacetic acid (Gd-DTPA) has been shown to improve the detection of intradural extramedullary spinal disease in adults. The ability of Gd-DTPA enhanced magnetic resonance (MR) to detect intradural extramedullary spinal metastases was studied in pediatric brain tumor patients. Spinal MR images before and after intravenous injection of Gd-DTPA were analyzed retrospectively in eight pediatric patients with known intracranial neoplasms and clinically suspected subarachnoid tumor seedings. Contrast enhanced spinal MR was compared with CT myelography in four of these patients. In our pediatric population Gd-DTPA enhanced images revealed tumor seeding not appreciable on noncontrast images. Although CT myelography has been the accepted standard investigation in the evaluation of suspected spinal metastases in children, we found that contrast enhanced MR is equal or superior in sensitivity to CT myelography. Spinal MR also provided information not obtainable via CT myelography. In the future, Gd-DTPA enhanced spinal MR should be considered in the initial evaluation of suspected subarachnoid spinal metastases in pediatric patients with known primary brain tumors.     

Spinal epidural extraskeletal Ewing sarcoma: MR findings in two cases

SUMMARY: We report the CT myelography and MR findings of two cases of extraskeletal Ewing sarcoma involving the spinal epidural and paravertebral spaces in a middle-aged man (case 1) and a young woman (case 2). In both cases CT myelography showed epidural and paravertebral masses on one side, with widening of the ipsilateral neural foramina at the C5-C6 level in case 1 and at the C7-T1 level in case 2. On MR images, the masses were isointense to muscle on T1-weighted images, hyperintense on T2-weighted images, and showed moderate enhancement on contrast-enhanced T1-weighted images. In one case, all pulse sequences showed linear signal voids, representing the vertebral artery encasement within the mass. The intradural component connected with the main mass was detected in the other case.     

Magnetic Resonance Imaging Characteristics of Primary Intraspinal Peripheral Primitive Neuroectodermal Tumour

Purpose

The purpose of our study was to describe the salient magnetic resonance imaging (MRI) findings in primary intraspinal peripheral primitive neuroectodermal tumour (PNET).

Methods

A retrospective review of the clinical and MRI images of 7 pathologically proven cases of intraspinal peripheral PNETs was performed. The various parameters, such as vertebral level of involvement; tumour location, size, focality, and margin; signal intensity of the lesion; the presence of hemorrhage or calcification; any signal voids; assessment of the adjacent cord for cord compression; cord dilatation; the presence of paraspinal tissue mass; or vertebral or other bony changes, were analysed.

Results

All 7 patients had lesions in the thoracolumbar region. Three patients had extradural lesions, 4 had intradural extramedullary lesions, and none had intramedullary lesions. Six lesions were well circumscribed. Only 1 patient had multifocal involvement. All lesions were of hypointense or isointense signal on T1-weighted imaging, whereas all but one were hyperintense on T2-weighted imaging. Lesions enhanced heterogeneously except 1 intradural extramedullary lesion, which enhanced homogeneously. A paraspinal mass was noticed in 2 patients. Vertebral collapse was present in 1 patients.

Conclusion

Intraspinal peripheral PNETs are rare spinal tumours. Although imaging characteristics are not specific, a focal circumscribed lesion in a young individual at the intramedullary, extramedullary intradural, or extradural spinal location that shows hypointense and hyperintense signal on T1- and T2-weighted images, respectively, requires PNET to be considered in the differentials.     

Acute spontaneous spinal epidural hematomas.

BACKGROUND AND PURPOSE: Although previous reports have characterized MR imaging features of spinal epidural hematomas (EDH), few cases have been reported during the acute or hyperacute phase within the first 48 hours. Our goal in this investigation was to correlate the MR imaging features of acute (< or =48 hours) spontaneous EDH with clinical management and outcome. METHODS: Eight patients with acute spontaneous EDH (five men and three women; age range, 31-81 years) underwent MR imaging at 1.5 T (T1-weighted, n = 8; T1-weighted after the administration of 0.1 mmol/kg contrast material, n = 6; T2-weighted, n = 8; and T2-weighted, n = 4). The interval from symptom onset to hospital admission ranged from immediate to 5 days. Two neuroradiologists reviewed the MR images for signal characteristics, contrast enhancement, and cord compression. Treatment and clinical outcome were correlated with the imaging findings. RESULTS: The EDH were located in the cervical (n = 3), cervicothoracic (n = 2), thoracolumbar (n = 2), and lumbar (n = 1) regions. On T1-weighted images, the signal intensity of the EDH was isointense to spinal cord in five cases, hyperintense in two cases, and hypointense in one case and did not correlate with time to imaging. Isointensity on T1-weighted images persisted for 5 days in one case. On T2-weighted images, all EDHs were hyperintense with focal, heterogeneous hypointensity. Cord compression was severe in six patients, moderate in one patient, and minimal in one patient. Four cases were treated conservatively with complete resolution or improvement of symptoms within 1 to 3 weeks. CONCLUSION: MR imaging findings were useful in establishing the diagnosis of EDH but did not influence management or predict outcome in this series. Heterogeneous hyperintensity to cord with focal hypointensity on T2-weighted images should suggest the diagnosis of acute spinal EDH. Severity of neurologic impairment had the greatest impact on management and outcome. Nonoperative treatment may be successful in cases with minimal neurologic deficits, despite cord compression revealed by MR imaging.     

Imaging of the spine and spinal cord

Recent technologic advances in MR imaging of the spine and spinal cord have been aimed at shortening examination time and suppressing artifacts. Gradient-recalled fast T2-weighted imaging is advocated for evaluating cervical radiculopathy. Better signal-to-noise ratios are achieved with three-dimensional Fourier transform gradient-recalled T2-weighted imaging and with decreased bandwidth acquisition. Obtaining high-contrast images of the spine is often complicated by the appearance of truncation artifacts. In outpatient radiculography, both iopamidol and iohexol appear safe, but iohexol seems better tolerated. In trauma, MR imaging is definitive, and signal abnormalities can help in evaluating neurologic recovery. Intraoperative spinal sonography may be helpful in evaluating acute injury. Enhancement with gadolinium diethylenetriamine penta-acetic acid is useful in the evaluation of disk space infections, osteomyelitis, and epidural abscess and in the study of spinal cord sarcoidosis. In patients with the acquired immunodeficiency syndrome, MR imaging may show hyperintense spine on T2-weighted images. MR imaging with gadolinium diethylenetriamine penta-acetic acid or gadolinium tetra-azacyclododecane tetra-acetic acid enhancement can be useful in evaluating intramedullary and intradural extramedullary tumors and for determining the extent of spinal leptomeningeal metastases. MR imaging can also be used to differentiate benign from pathologic fractures.     

Exophytic syrinx, an extreme form of syringomyelia: CT, myelographic, and MR imaging features.

Magnetic resonance (MR) imaging and computed tomography (CT) were used to study severe syringomyelia, in which the syrinx was so eccentric relative to the center of the spinal cord that at initial examination it appeared to be an exophytic extramedullary mass, in five of six patients with type 2 Arnold-Chiari malformation and one patient with an intramedullary tumor. Sagittal and axial images were routinely obtained at 1.5 T; CT was performed after intrathecal injection of contrast material. On sagittal MR images, the exophytic component of the syrinx typically displaced the spinal cord peripherally and mimicked an extramedullary intradural lesion such as an arachnoid cyst or meningioma. On axial MR images, the gradual appearance and disappearance of this component could be traced as it compressed the spinal cord, which had a signet-ring appearance at the equator of the syrinx. It is concluded that both MR imaging and postmyelographic CT can reveal the characteristics of this lesion on axial images, but MR imaging is superior because it enables direct sagittal imaging.     

Gadolinium-DTPA-Enhanced MR Imaging of Spinal Neoplasms: Preliminary Investigation and Comparison with Unenhanced Spin-Echo and STIR Sequences

Unenhanced T1- and T2-weighted spin-echo, short inversion time inversion recovery (STIR), and gadolinium-DTPA (Gd-DTPA)-enhanced spin-echo and STIR imaging techniques were used in 20 patients as part of a multicenter study to assess the safety and efficacy of Gd-DTPA in spinal imaging. Five patients had normal MR scans. Of those with lesions, both Gd-DTPA-enhanced T1-weighted spin-echo and unenhanced STIR scans improved detection and evaluation of spinal tumors over conventional spin-echo methods, particularly T2-weighted spin echo, by providing higher tissue contrast in shorter imaging times. The Gd-DTPA-enhanced T1-weighted spin-echo scans were most helpful in evaluating intradural tumors, whereas STIR sequences were most effective for extradural tumors and bone metastases. In most cases, Gd-DTPA-enhanced T1-weighted spin-echo scans best delineated tumor margins, and the enhancement was helpful in suggesting a cellular or active nature of the lesions. In some cases, the enhancement resulted in a more homogeneous and thus less abnormal-appearing marrow in vertebrae involved by tumor; therefore, a precontrast T1-weighted spin-echo scan is necessary in all patients who are to be studied with Gd-DTPA.A combined approach that uses T1-weighted spin-echo, Gd-DTPA-enhanced T1-weighted spin-echo, and STIR images currently appears optimal for MR imaging of spinal neoplasms. T2-weighted spin-echo images add information only in occasional cases.     

Syringomyelia as a consequence of compressive extramedullary lesions: postoperative clinical and radiological manifestations

MR imaging was performed to determine the cause of the onset of new neurologic symptoms in five patients who had previously undergone surgical excision of extramedullary masses. Syringomyelia and the absence of recurrent or residual lesions were documented in all cases. Three patients showed long cysts (multiseptated in two and smooth in one) with low signal intensity on both T1- and T2-weighted images. The flow-void phenomenon related to fluid motion in these three cysts, which were enlarging clinically, was responsible for the hypointensity on the T2-weighted images. In two patients the fluid within the lesions behaved similarly to normal nonpulsatile CSF and may have represented syrinx cavities in a state of "arrested growth." Three patients had surgical decompression under real-time intraoperative sonographic control, which showed the presence of intramedullary cyst-fluid pulsations in two cases and the absence of cyst-fluid pulsations in one case. These sonographic observations correlated with the MR findings. We postulate that these syrinx cavities form as a result both of the effect that the original extramedullary lesion had upon the underlying spinal cord and the subsequent postoperative alterations in the CSF dynamics at the level of prior surgery. Syringomyelia should be considered in patients with recurrent or new symptoms who previously had surgery for extramedullary lesions.     

Gadolinium-DTPA-enhanced MR imaging of spinal neoplasms: preliminary investigation and comparison with unenhanced spin-echo and STIR sequences

Unenhanced T1- and T2-weighted spin-echo, short inversion time inversion recovery (STIR), and gadolinium-DTPA (Gd-DTPA)-enhanced spin-echo and STIR imaging techniques were used in 20 patients as part of a multicenter study to assess the safety and efficacy of Gd-DTPA in spinal imaging. Five patients had normal MR scans. Of those with lesions, both Gd-DTPA-enhanced T1-weighted spin-echo and unenhanced STIR scans improved detection and evaluation of spinal tumors over conventional spin-echo methods, particularly T2-weighted spin echo, by providing higher tissue contrast in shorter imaging times. The Gd-DTPA-enhanced T1-weighted spin-echo scans were most helpful in evaluating intradural tumors, whereas STIR sequences were most effective for extradural tumors and bone metastases. In most cases, Gd-DTPA-enhanced T1-weighted spin-echo scans best delineated tumor margins, and the enhancement was helpful in suggesting a cellular or active nature of the lesions. In some cases, the enhancement resulted in a more homogeneous and thus less abnormal-appearing marrow in vertebrae involved by tumor; therefore, a precontrast T1-weighted spin-echo scan is necessary in all patients who are to be studied with Gd-DTPA. A combined approach that uses T1-weighted spin-echo, Gd-DTPA-enhanced T1-weighted spin-echo, and STIR images currently appears optimal for MR imaging of spinal neoplasms. T2-weighted spin-echo images add information only in occasional cases.     

Central scars in primary liver tumors: MR features, specificity, and pathologic correlation

Tumor scars were identified at pathologic study and magnetic resonance (MR) imaging in ten of 17 (59%) primary liver tumors (nine hepatocellular carcinomas, four giant hemangiomas, two hepatic adenomas, and two cases of focal nodular hyperplasia). Histopathologic examination revealed three types of scar tissue. Inflammatory scars (n = 4), with edema, necrosis, hypercellularity, and loose fibrous tissue, appeared hypointense relative to liver on T1-weighted images and hyperintense on T2-weighted images. Vascular scars (n = 3), predominantly composed of vascular channels traversing collagenous tissue, showed MR features similar to those of inflammatory scars. Collagenous scars (n = 3) appeared hypointense relative to liver on both T1-weighted and T2-weighted images. Central tumor scars are a frequent but nonspecific feature of both benign and malignant primary liver tumors.     

MR imaging of the cauda equina in hereditary motor sensory neuropathies: correlations with sural nerve biopsy

BACKGROUND AND PURPOSE: Although spinal root abnormalities are known to occur, spinal MR examination is seldom performed in hereditary motor and sensory neuropathies (HMSN). The following work was undertaken to assess the MR imaging spectrum of lumbosacral spinal nerve root abnormalities and determine whether intradural nerve root involvement could be related to any biopsy feature. METHODS: Ten consecutive patients (eight male, two female; age range, 28-65 yrs) with Charcot-Marie-Tooth (CMT) (type I = 5, type II = 2) and Déjèrine-Sottas disease (DSD) (n = 3) underwent a contrast-enhanced lumbosacral MR examination. Sural nerve biopsy was performed in all patients. Atypical clinical features were present in two patients. The MR scans of each patient were reviewed for possible causes of myeloradiculopathy, spinal nerve root and ganglia dimensions, signal change, and abnormal enhancement. RESULTS: In the seven patients with CMT, abnormal MR findings were intradural nerve root hypertrophy (n = 2), signal abnormalities (n = 2), and enhancement (n = 3). Two of three patients with DSD had the abnormal MR finding of intradural nerve root enhancement. In both patients with atypical clinical features, MR imaging showed nerve root hypertrophy and enhancement. Both findings were related to an increased number of onion bulbs at sural nerve biopsy. Inflammatory infiltrates were not observed in any patients. CONCLUSION: In patients with HMSN enhancement of intradural spinal nerve roots, whether or not associated with marked thickening, may be found on lumbosacral MR examinations. Spinal nerve root thickening may be responsible for atypical symptoms, and its visibility on MR images represents a useful adjunct to diagnosis. Lumbosacral spinal nerve root abnormalities were related to an extremely high number of onion bulbs (indicating active demyelination) at sural nerve biopsy. Nerve root enhancement does not seem to be related to inflammatory infiltrates.