|
|
||||||||
Original Report |
1 Department of Radiologic Pathology, Armed Forces Institute of Pathology, 6825
16th St., NW, Washington, DC 20306-6000.
2 Department of Radiology and Nuclear Medicine, Uniformed Services University of
the Health Sciences, Bethesda, MD 20814-4799.
3 Department of Radiology, UDIAT Diagnostic Center, Parc Tauli, Sabadell 08208,
Spain.
4 Department of Soft Tissue Pathology, Armed Forces Institute of Pathology,
Washington, DC.
5 Department of Gastrointestinal Pathology, Armed Forces Institute of Pathology,
Washington, DC.
Received April 22, 2004;
accepted after revision June 8, 2004.
The opinions and assertions contained herein are the private views of the
authors and are not to be construed as official or as reflecting the view of
the Department of the Army or Defense.
Abstract
|
|
|---|
CONCLUSION. Gastrointestinal schwannomas are uncommon mesenchymal neoplasms that are uniquely different tumors from their soft-tissue and central nervous system counterparts. They are homogeneously attenuating, well-defined, mural masses on CT. The lack of low-attenuation hemorrhage, necrosis, and degeneration within the tumor may help distinguish these tumors from gastrointestinal stromal tumors on CT.
|
|
|---|
Initially reported in 1988 by Daimaru et al. [2], gastrointestinal schwannomas are benign tumors that are associated with an excellent prognosis after surgical resection. Gastrointestinal schwannomas are classified as mesenchymal or neuroectodermal neoplasms. These are a heterogeneous group of tumors arising from the wall of the gastrointestinal tract that include gastrointestinal stromal tumors (GISTs), leiomyomas, leiomyosarcomas, schwannomas, neurofibromas, ganglioneuromas, paragangliomas, lipomas, granular cell tumors, and glomus tumors. Of these, GISTs are by far the most common in the stomach and intestines.
Gastrointestinal schwannomas occur most commonly in the stomach (6070% of cases), followed by the colon and rectum [1, 46]. Esophageal and small-intestinal schwannomas have been rarely reported [2, 5]. To our knowledge, there have been two cases reported in the English-language radiology literature describing the CT features of gastrointestinal schwannomas [7, 8]. The purpose of our study was to describe the clinical, pathologic, and CT features of gastrointestinal schwannomas.
|
|
|---|
Clinical Data and Pathology Review
Clinical, pathology, and surgical data were reviewed for patient age, sex,
presenting signs and symptoms, and a medical history of neurofibromatosis or
malignancy. All patients underwent complete surgical excision of their tumors.
Pathology and surgery records were reviewed for the location of the tumor
within the gastrointestinal tract and its exact location within the bowel wall
(mucosal, mural, or extrinsic). Histopathology reports and slide material were
available for all patients.
Experienced gastrointestinal and soft-tissue pathologists reviewed the H and Estained slide material in each case to establish or reconfirm the diagnosis of gastrointestinal schwannoma. The morphologic diagnosis of schwannoma was based on the presence of a spindle cell tumor with a microtrabecular architectural pattern interspersed with fibers of collagen. The slides were also assessed for the presence of a lymphoid cuff, lymphoplasmacytic infiltrate, and germinal centers. Pathology reports were used to establish tumor size and immunoreactivity to S-100 protein.
CT Imaging and Review
CT images were available for review in all patients. Seven patients
underwent abdominal CT, and one patient underwent chest CT. Two patients had
IV contrastenhanced incremental CT scans, one with 8-mm and the other
with 10-mm slice thickness. The remaining patients (n = 6) had
helical CT scans (two unenhanced scans and four IV contrastenhanced
scans) with slice thickness ranging from 5 to 7 mm. Oral contrast material was
administered to six of the eight patients. Because our patients are referred
from many institutions, the CT techniques were not standardized. Their studies
had been performed on a variety of equipment over a 10-year period, and
differing protocols had been followed regarding slice thickness and contrast
injection technique.
One experienced abdominal radiologist and one fourth-year radiology resident reviewed all images retrospectively with the final interpretation by consensus. The final histopathologic diagnosis of schwannoma was known at the time of interpretation. The images were evaluated for the presence or absence of a mass in the gastrointestinal tract. If a mass was present, the size, shape (round, oval, or infiltrating), margins (well-defined or ill-defined), enhancement pattern (no enhancement, homogeneous, heterogeneous, or rimlike), and location with respect to the bowel wall (intraluminal, mural, or exophytic) were documented. The mass was also evaluated for the presence or absence of a capsule, ulcer formation, cystic change, cavity formation, calcification, or necrosis. The CT scans were evaluated for additional masses in the gastrointestinal tract, lymphadenopathy, ascites, and masses within the solid organs of the abdomen.
After imaging review, clinical data, surgical records, pathology reports, and photographs of the gross pathologic specimens (n = 6) were analyzed with the available imaging studies. The appearance of the mass on gross photography (size, shape, margins, presence of internal hemorrhage, cystic change, ulceration, or cavitation) was compared with its imaging appearance.
|
|
|---|
Five patients (63%) had a gastric schwannoma; two patients, small intestinal; and one patient, esophageal. Two of the patients with gastric schwannoma presented with abdominal pain; one patient, gastrointestinal bleeding; and one patient, chest pain. In the fifth patient with a gastric schwannoma, the tumor was incidentally found on a CT scan obtained for bladder cancer staging. Both patients with small-intestinal schwannomas complained of abdominal pain, one of whom had a small-bowel obstruction secondary to adhesions unrelated to the tumor. The single patient with an esophageal schwannoma complained of dysphagia.
None of the patients had a history of neurofibromatosis. One patient had a concurrent malignancy (transitional cell carcinoma of the bladder). All patients underwent complete surgical excision of their tumors.
Histopathologic and Immunophenotypic Features
The histopathologic diagnosis of schwannoma was confirmed in all patients
(Figs. 1A and
1B). Evidence of a lymphoid
cuff, lymphoplasmacytic infiltrate, or germinal centers was present in all
tumors (Fig. 1A).
Immunohistochemical evidence of S-100 protein positivity was documented in
seven of the eight patients. The single patient without documented S-100
protein data was included in the study on the basis of typical light
microscopic features.
|
|
CT Features
All five patients with gastric schwannomas had well-defined, round, mural
masses with homogeneous attenuation on CT (two on unenhanced CT, three on
contrast-enhanced CT). Two of the three tumors on contrast-enhanced CT showed
minimal to no enhancement during the portal venous phase of contrast
enhancement (Figs. 2A and
2B), whereas a single tumor
showed homogeneous contrast enhancement during the equilibrium phase of
contrast enhancement (Fig. 3).
Two of the gastric schwannomas (40%) projected into the lumen of the stomach
simulating polypoid masses, and two (40%) had a predominantly exophytic growth
pattern. The tumor had both significant luminal and exophytic components in
only one case (20%). Two of the gastric schwannomas (40%) had ulcer craters
identified on CT (Figs. 4A and
4B).
|
|
|
|
|
There was no CT evidence of tumor capsule, cystic change, cavity formation, necrosis, or calcification in any of the tumors. There were no cases with additional gastrointestinal masses, pathologically sized lymphadenopathy, or ascites. The only patient who had a mass in another organ was the patient with transitional cell carcinoma of the bladder; that patient had focal thickening of the right anterior bladder wall.
Two patients had small-intestinal schwannomas. One patient had a 1.3-cm ileal schwannoma. The other patient had a 2.5-cm jejunal schwannoma and small-bowel obstruction, which was secondary to an adhesion distal to the schwannoma. Both patients underwent contrast-enhanced CT, which showed a round, homogeneous enhancing mass that projected into the lumen of the intestine in each case (Figs. 5A and 5B).
|
|
The single patient with an esophageal schwannoma was scanned during the arterial phase of CT contrast enhancement. The mass was located in the mid esophagus, extending from the level of the top of the aortic arch to the inferior margin of the arch. It extended medially into the mediastinum and deviated the esophageal lumen to the right. The tumor was well defined, round, and homogeneous in CT attenuation (Fig. 6).
|
Cut-surface photographs of gross pathologic specimens were available for comparison with CT scans in six patients (gastric, n = 5; small bowel, n = 1). The gross photographs of all six tumors showed a round, well-defined, unencapsulated pale pink or yellow tumor located in the wall of the stomach or small intestine. The cut surface of the tumor was homogeneous in each case and corresponded to the uniform tumor homogeneity on CT scans (Figs. 2B, 4B, and 5B). No evidence of hemorrhage, necrosis, or cystic change was detected on the cut surface of any of the specimens. In the two gastric cases that showed a focal ulcer on CT, there was a corresponding ulcer present in the mucosa overlying the tumor on the gross photograph (Fig. 4B).
|
|
|---|
Although small in number, our series shows that there is remarkable uniformity to the clinical presentation and CT appearance of gastrointestinal schwannomas. All of our patients except one were women, and all of our patients were over the age of 50. Although pain was the most common clinical complaint, the clinical presentation is also site-specific as in the case of our patient with an esophageal schwannoma who presented with symptoms of dysphagia. In addition to abdominal pain, three of our patients with gastric schwannomas also had symptoms directly referable to the stomach: nausea, postprandial pain, and early satiety.
The most consistent CT feature in our series of patients is the homogeneous pattern of tumor attenuation on both unenhanced and IV contrastenhanced scans. The homogeneous CT attenuation pattern directly corresponds to the homogeneity of the tumor cut surface seen grossly. Tumor enhancement was difficult to assess in our series because of the lack of standardization of the IV contrast injection protocols in our referral population. Of the three patients with gastric schwannomas who underwent contrast-enhanced CT, the tumor in two patients showed minimal to no enhancement during the portal venous phase of contrast enhancement and, in one patient, showed marked enhancement during the equilibrium phase. This may suggest that enhancement of gastrointestinal schwannomas occurs over time with peak enhancement occurring during the equilibrium phase.
Gastrointestinal schwannomas are hypothesized to arise from the myenteric plexus within the gastrointestinal tract wall because of their immunophenotypic similarities. Both schwannomas and the cells of the myenteric plexus express S-100 protein and GFAP (glial fibrillary acidic protein) positivity [2]. Grossly, the schwannomas in our series were located in the wall of the gastrointestinal tract, but on CT they had a primarily intraluminal location in four cases, extraluminal component in three cases, and both intraluminal and extraluminal in one case. Similar growth patterns have been observed in other mesenchymal neoplasms such as GISTs [10, 11].
Our study has limitations on the basis of our case material. Our patients were referred from many institutions over at 10-year period. The lack of standardization of imaging parameters and injection techniques influenced our ability to accurately analyze the contrast enhancement pattern of the tumors. We had the advantage, however, of being able to review surgical and pathology records in all patients and photographs of gross specimens in six patients to ensure that our CT interpretation was correct.
The main differential diagnosis for a mass arising in the wall of the gastrointestinal tract is a GIST because it is the most common mesenchymal neoplasm of the gastrointestinal tract. The most striking difference in the CT features of GISTs compared with schwannomas is the presence of hemorrhage, necrosis, and cystic change, which give GISTs a much more heterogeneous appearance on CT. In our previously reported data on gastric GISTs, 92% of cases had a peripheral enhancement pattern due to hemorrhage, necrosis, or cystic change within the tumor, whereas only 8% of cases showed homogeneous enhancement [11]. Similarly, Burkill et al. [10] reported 13% of their patients with malignant GISTs had a homogeneous tumor on CT. Thus, the CT features of smaller and nonmalignant GISTs overlap with those of schwannomas.
Primary and secondary lymphomas of the gastrointestinal tract may have a similar CT appearance to schwannomas because they arise in the wall of the gastrointestinal tract and tend to be homogeneous in CT attenuation before therapy. Adenopathy is a helpful distinguishing feature for lymphoma because gastrointestinal lymphomas are commonly accompanied by adenopathy in the supporting mesenteries and retroperitoneum. Adenopathy is not a feature of gastrointestinal schwannomas.
Although gastrointestinal adenocarcinomas originate from mucosa, they may extend beyond the wall of the intestine simulating a mesenchymal neoplasm. Typically, however, the margins of an adenocarcinoma on CT are spiculated rather than smooth, and they are commonly associated with regional adenopathy.
A special note should be made regarding the differential diagnosis of esophageal schwannomas. The most common mesenchymal neoplasm in the esophagus is an intramural leiomyoma [12], which is indistinguishable from schwannoma on the basis of imaging alone.
In summary, gastrointestinal schwannomas are uncommon benign neoplasms that are uniquely different from conventional soft-tissue and central nervous system schwannomas. They arise in the wall of the gastrointestinal tract and are most commonly found in the stomach. On CT, their homogeneous attenuation pattern may aid in differentiation from GISTs.
|
|
|---|
This article has been cited by other articles:
![]() |
K S MADHUSUDHAN, D N SRIVASTAVA, N R DASH, C GUPTA, and S D GUPTA Schwannoma of both intrahepatic and extrahepatic bile ducts: a rare case Br. J. Radiol., October 1, 2009; 82(982): e212 - e215. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |