| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 Department of Diagnostic Radiology, Kangnam St. Mary's Hospital, The Catholic
University of Korea, 505 Banpodong, Seocho-gu, Seoul 137-701, South
Korea.
2 Department of Diagnostic Radiology, Yale University School of Medicine, New
Haven, CT.
3 Department of Radiology, Kyung-Hee University, Seoul, South Korea.
4 Department of Radiology, Yonsei University, Seoul, South Korea.
5 Department of Radiology, Korea Cancer Center Hospital, Seoul, South
Korea.
6 Department of Pathology, Kangnam St. Mary's Hospital, The Catholic University
of Korea, Seoul 137-701, South Korea.
7 Department of Orthopedic Surgery, The Catholic University of Korea, Seoul,
South Korea.
Received October 28, 2003; accepted after revision March 16, 2004.
Address correspondence to W.-H. Jee
(whjee{at}catholic.ac.kr).
OBJECTIVE. The purpose of our study was to evaluate whether MRI can discriminate between extraaxial neurofibromas and neurilemmomas.
MATERIALS AND METHODS. MR images of 52 patients with a pathologically proven extraaxial neurofibroma or neurilemmoma were retrospectively reviewed by observers who were unaware of the surgical results, regarding the presence or absence of individual imaging criteria. MRI findings in 12 patients with a localized neurofibroma and 40 patients with a neurilemmoma were compared using the chi-square test or Fisher's exact test.
RESULTS. MRI findings suggestive of neurofibroma (p < 0.05) were a target sign on T2-weighted images (58% in neurofibromas vs 15% in neurilemmomas), central enhancement (75% vs 8%), and a combination of both findings (63% vs 3%). MRI findings suggestive of a neurilemmoma (p < 0.05) were a fascicular appearance on T2-weighted images (25% vs 63%), a thin hyperintense rim on T2-weighted images (8% vs 58%), a combination of both findings (8% vs 48%), and diffuse enhancement (13% vs 67%). No significant difference was seen between neurofibromas and neurilemmomas for a centrally entering and exiting nerve (42% in neurofibromas vs 23% in neurilemmomas), a peripherally entering and exiting nerve (58% vs 77%), a cystic area (38% vs 64%), a low-signal margin (100% vs 100%), peripheral enhancement (13% vs 26%), or a target sign on contrast-enhanced images (11% vs 31%).
CONCLUSION. MRI shows features helpful for differentiating extraaxial neurofibromas from neurilemmomas; however, no single finding or combination of findings allows definitive differentiation.
Neurofibromas and neurilemmomas (schwannomas) are benign peripheral nerve sheath tumors. For neurilemmomas, marginal excision can usually spare the parent nerve because a neurilemmoma is generally separable from the underlying nerve, allowing the native nerve and its function to be preserved. Partial resection can be performed to spare the nerve and recurrence is unusual [1]. For neurofibromas, a resection of the nerve with the tumor is necessary because a neurofibroma cannot be separated from the nerve fibers [1–5]. Neurofibromas involving major nerves may be treated conservatively or may be treated with debulking because of the need to sacrifice the nerve for complete tumor removal [1]. Differentiation of neurofibromas from neurilemmomas can help the surgeon determine the preoperative plan and advise the patient about the likelihood of loss of nerve function as a result of surgery.
The purpose of our study was to evaluate the MRI findings for discriminating between extraaxial neurofibromas and neurilemmomas.
Materials and Methods
The MRI examinations of 52 patients with pathologically proven extraaxial neurofibromas or neurilemmomas were included in this study. The cases were collected from multiple centers and were identified by reviewing pathology databases. Our institution did not require institutional review board approval for the review of retrospective studies. The patients were 21 males and 31 females ranging in age from 12 to 68 years (mean age, 41 years). MRI was performed with various 1.5-T (n = 31) or 1.0-T (n = 21) MR units. MRI consisted of T1-weighted (TR range/TE range, 350–760/12–30) and T2-weighted (1,800–4,200/70–127) sequences performed in at least two orthogonal planes. In 21 patients, conventional spin-echo T2-weighted (1,800–2,400/60–100) imaging was performed; and in 31 patients, a turbo spin-echo (n = 23) or fast spin-echo (n = 8) pulse sequence was used to obtain T2-weighted (3,000–4,485/75–132) images. In two patients, the fat-suppression option with selective suppression by a chemical shift selective pulse was combined with turbo spin-echo (n = 1) or fast spin-echo (n = 1) T2-weighted images. In 47 patients, contrast-enhanced T1-weighted spin-echo imaging was performed after the administration of 0.1 mmol/kg of body weight of gadopentetate dimeglumine (Magnevist, Schering) with (n = 19) or without (n = 28) selective fat suppression by a chemical shift selective pulse.
MR images were retrospectively reviewed by an experienced musculoskeletal radiologist who was unaware of the surgical results, for the presence or absence of individual imaging criteria. On all imaging sequences, tumors were categorized as homogeneous or heterogeneous and as hypointense, of intermediate intensity, or hyperintense relative to skeletal muscle on T1-weighted images and relative to fat on T2-weighted images. Intermediate intensity meant that the signal intensity was approximately equal to the signal intensity of skeletal muscle on T1-weighted images and to that of fat on T2-weighted images. Imaging features analyzed on T2-weighted images included the target sign, a fascicular appearance, and a thin hyperintense rim. The reviewer classified lesions as showing a target sign on T2-weighted images when there was a central hypointense focus with a peripheral hyperintense rim, the central focus making up less than three quarters of the diameter of the lesion. The reviewer classified the lesions as having a thin hyperintense rim when the rim was less than one quarter the diameter of the lesion. The tumor was categorized as having a fascicular appearance if multiple small ringlike structures (with peripheral higher signal intensity) were seen on either T2- or intermediate-weighted images, as has been described previously [6]. On contrast-enhanced T1-weighted images, the enhancement pattern was evaluated, including whether a target configuration was present. In addition, the presence of a centrally or peripherally entering and exiting nerve, cystic area, low-signal margin, or split fat sign, and the shape of the lesion were assessed. The tumor was classified as having a fusiform, dumbbell, or ovoid shape. Statistical analysis was performed using the chi-square or Fisher's exact test. For all statistical comparisons, significance was defined as a p value of less than 0.05.
Results
On T1-weighted images, all tumors were of hypointense to intermediate signal intensity. On T2-weighted images, neurofibromas showed hyperintense signal intensity with a heterogeneous (n = 11) or a homogeneous (n = 1) pattern, whereas neurilemmomas showed a hyperintense signal with a heterogeneous (n = 31) or a homogeneous (n = 9) pattern. The target sign on T2-weighted images was observed in seven of 12 patients with neurofibromas (Figs. 1A, 1B, and 1C) and in six of 40 patients with neurilemmomas (Figs. 2 and 3A, 3B, 3C). A fascicular appearance was observed in 25 of 40 patients with neurilemmomas (Figs. 4 and 5) and in three of 12 patients with neurofibromas on T2-weighted images. A thin hyperintense rim was observed in 23 of 40 patients with neurilemmomas (Fig. 5) and in one of 12 neurofibromas on T2-weighted images.
|
|
|
|
|
|
|
|
|
Among eight patients with neurofibromas who underwent contrast-enhanced MRI, central enhancement (Figs. 1A, 1B, and 1C) was observed in six patients (75%), diffuse enhancement in one patient, and peripheral enhancement in one patient. Among 39 patients with neurilemmomas who underwent contrast-enhanced MRI, diffuse enhancement (Figs. 6A, 6B, and 6C) was seen in 26 patients, heterogeneous enhancement in 19, homogeneous enhancement in eight, peripheral enhancement in 10, and central enhancement in three patients. Five patients who had neurilemmomas with a target sign on T2-weighted images underwent contrast-enhanced MRI, and only one of those showed central enhancement, with four showing diffuse enhancement. A target configuration was observed in 12 of 39 patients with neurilemmomas and in one of nine patients with neurofibromas on contrast-enhanced T1-weighted images. Neurofibromas showed a centrally entering and exiting nerve in five patients and a peripherally entering and exiting nerve in seven patients. Neurilemmomas showed a peripherally entering and exiting nerve in 30 patients and a centrally entering and exiting nerve in nine patients, with the nerve not visible in one patient.
|
|
|
MRI findings suggestive of neurofibroma (p < 0.05) were as follows: target sign on T2-weighted images, central enhancement, and a combination of both findings (63% for neurofibromas vs 3% for neurilemmomas, p < 0.001) (Table 1). MRI findings suggestive of a neurilemmoma (p < 0.05) were as follows: a fascicular appearance on T2-weighted images, a thin hyperintense rim on T2-weighted images, a combination of both findings (8% for neurofibromas vs 48% for neurilemmomas, p < 0.05), and diffuse enhancement.
|
No significant difference was seen between neurofibromas and neurilemmomas for a centrally or peripherally entering and exiting nerve, cystic area, low-signal margin, peripheral enhancement, target configuration on contrast-enhanced images, or split-fat sign (Table 1). No significant difference was seen between neurofibromas and neurilemmomas for fusiform, dumbbell, or ovoid shape (Table 1).
Neurofibromas were observed in the arms (n = 8), the legs (n = 2), and the neck (n = 2). Neurilemmomas were observed in the legs (n = 19), the arms (n = 13), the feet (n = 4), the neck (n = 2), the trunk (n = 1), and the hand (n = 1).
Discussion
Previous reports [1, 2, 7] have indicated that some MRI signs may be helpful for differentiating neurilemmomas from neurofibromas. Our study was limited to extraaxial neurofibromas and neurilemmomas because intraspinal neurilemmomas have necrosis and cyst formation much more frequently than their intracranial and peripheral nerve counterparts [8, 9]. Only the localized type of neurofibroma or neurilemmoma was included. Diffuse or plexiform neurofibromas, neurofibromatosis, and schwannomatosis were excluded.
In this study, the reviewer found MRI findings helpful for differentiating neurofibromas from neurilemmomas; however, no single finding or group of findings allowed definitive diagnosis of either. The target sign on T2-weighted images, central enhancement, and a combination of both findings were different for neurofibromas versus neurilemmomas. Previous studies indicate that the target sign on T2-weighted images corresponds pathologically to central fibrocollagenous tissue and peripheral predominately myxoid tissue [1, 3, 7, 10]. In our study, the target sign on T2-weighted images was observed in seven (58%) of 12 neurofibromas, which was consistent with previous reports of 50–70% [3, 5, 7, 10]. According to Varma et al. [7], seven (54%) of 13 neurilemmomas had the target pattern, unlike our prevalence of 15% (6/40). This discrepancy could be related in part to our strict criteria for the target sign on T2-weighted images, with the central focus making up less than three quarters of the diameter of the lesion. Our results are also different from those of Suh et al. [3], who found the target pattern only in neurofibromas; however, their study included only four patients with neurilemmomas. Unlike the target sign on T2-weighted images, contrast-enhanced target configuration was not helpful for discriminating the two tumors, with no significant difference between neurilemmomas (31%) and neurofibromas (11%). Contrast-enhanced target configuration was not classified as the target sign in this study, unlike in previous reports [7, 11–14]. To our knowledge, this sign has not been previously investigated.
A thin hyperintense rim on T2-weighted images was more common in neurilemmomas (58%) than in neurofibromas (8%). Diffuse enhancement was more common in neurilemmomas (67%) than in neurofibromas (13%), whereas central enhancement was more common in neurofibromas (75%) than neurilemmomas (8%). Our patients with extraaxial neurilemmomas showed a different enhancement pattern than shown in a previous report [8] that five of seven spinal neurilemmomas showed peripheral enhancement. Previous reports on enhancement pattern have included approximately 10 patients.
Our prevalence of a peripherally located nerve in 77% (30/39) of neurilemmomas was lower than that in the previous report of Cerofolini et al. [5], which found the nerve to be located peripherally in 90% (9/10) of neurilemmomas. More important, our reviewer found a peripherally located nerve in more than half the neurofibromas, with no significant difference between neurofibromas and neurilemmomas for a centrally versus a peripherally entering and exiting nerve. This finding is in contrast to prior reports [2, 5] that suggested that a peripherally entering nerve is indicative of a neurilemmoma.
Limitations of our study include that the reviewer of the MR images knew that all patients had a pathologically confirmed neurofibroma or neurilemmoma, which may have increased sensitivity for detecting each MRI finding. Another limitation is the small number of neurofibromas present in our patient population. The relatively high number of schwannomas relative to neurofibromas in this study may have affected our results.
In conclusion, the target sign on T2-weighted images or central enhancement favors the diagnosis of a neurofibroma, whereas a fascicular appearance, a thin hyperintense rim, or diffuse enhancement suggests a neurilemmoma. MRI shows helpful features for differentiation of neurofibromas from neurilemmomas; however, no single sign or combination of signs allows the definitive differentiation of neurofibromas from neurilemmomas.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A Handra-Luca, D Vidaud, M-P Vullierme, N Colnot, D Henin, P Ruszniewski, P Bedossa, and A Couvelard Plexiform neurofibroma mimicking a pancreatic cystic tumour Gut, November 1, 2005; 54(11): 1658 - 1659. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Endo, F. Murayama, S.-i. Otani, K. Tetsuka, T. Hasegawa, Y. Sato, and Y. Sohara Alternative Surgical Approaches for Apical Neurinomas: A Thoracoscopic Approach Ann. Thorac. Surg., July 1, 2005; 80(1): 295 - 298. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
