Schwannomas (cellular neurilemomas) are predominantly benign tumors derived from the Schwann cells that form the neural sheath. The incidence of submucosal schwannoma has been estimated at 2 to 6 percent of all submucosal tumors of the intestine [¹]. The largest published series documents that the most frequent site of primary gastrointestinal schwannoma is the stomach [¹]. Isolated schwannomas of the colon without associated neurofibromatosis are extremely rare [¹,²]. Here, we present two cases of laparoscopically resected schwannoma of the colon together with a review of the literature.

Schwannoma is a rare neoplasm that originates from Schwann cells and which commonly occurs in the peripheral nerve of the limbs or body, spinal cord, and central nervous system. When it occurs in the gastrointestinal tract (GIT), the colon is the least likely site [²]. The stomach has been reported to be the most common site of GIT involvement [¹,³].

Although the exact incidence of colonic schwannoma is unknown, some studies have estimated the incidence indirectly. Schwannomas are vastly less common than GISTs, being outnumbered by approximately 50-100:1 [⁴]. Hou et al. [³] determined that benign schwannomas represented 2.9% of gastrointestinal mesenchymal tumors in their investigations. Another study similarly identified twenty (3.3%) colorectal schwannomas among 600 mesenchymal tumors of the colon and rectum [⁵]. Inagawa et al. [⁶] reviewed reports of benign schwannomas arising from the colon and rectum in a Japanese literature search and found rectum (45.7%) to be the most frequent site of occurrence. They also calculated the incidence of schwannoma arising from the right colon, excluding the appendix, at 19.6% (9 cases among forty-six) of cases of colon and rectal schwannomas.

The extremely rare incidence of colonic schwannoma may be explained at least in part in light of the result of Hou et al. In their review of 25 cases of schwannoma, only twelve (48%) were primarily diagnosed as schwannomas; the other thirteen cases were originally diagnosed as leiomyoma, cellular leiomyoma, leiomyosarcoma or neurofibroma [³]. Studies in 1988 and 2001 by two different groups defined and fully documented the morphological and phenotypic characteristics of digestive schwannomas, which included a prominent lymphoid cuff with or without germinal center formation [¹,⁵]. Schwannomas of the GIT evaluated on a purely morphologic basis can be misdiagnosed as GISTs with neural differentiation [²,³]. Owing to advances in immunohistochemical staining, however, the number of reports of intestinal schwannoma has gradually increased, and its differential diagnosis from other spindle cell tumors such as leiomyoma can now be clearly made [³,⁶].

The most important immunohistochemical markers for diagnosis of schwannoma are S-100 protein, CD117 (KIT) and CD34. Cells from neurofibromas are 30 to 40% immunoreactive with S-100 protein, in contrast to 100% immunoreactivity in schwannomas [⁴]. In addition, schwannomas are consistently negative for CD117 and are usually negative for CD34, while in contrast GISTs exhibit high expression of CD117 and CD34 (70%) but are negative for S-100 protein [⁴,⁵]. A combinatorial assay of S-100, CD34 and CD117 proteins, therefore, is definitive in differentiating schwannoma from GISTs. Glial fibrillary acidic protein (GFAP) serves as another useful marker in immunohistochemical studies, allowing differentiation of schwannomas from gastrointestinal autonomic nerve tumors (GANTs). Schwannomas of the GIT express GFAP (63.6%), whereas GANTs do not express GFAP [³].

We made the preoperative diagnoses of both cases as submucosal tumor such as GISTs. Inagawa et al. [⁶] demonstrated that the rate of accurate preoperative diagnosis for benign schwannomas was only 15.2% (seven of forty-six cases) and that the most common preoperative diagnosis was in fact submucosal tumor (39.1%). The reasons for inaccurate preoperative diagnosis can include an insufficient volume of biopsy specimen for pathologic confirmation and a limited knowledge of such tumors. Levy et al. [⁷] described schwannomas evaluated by CT as homogeneously attenuating, well-defined, mural masses. These CT findings were also observed in our case. Contrary to these CT findings of gastrointestinal schwannomas, most of GISTs show a much more heterogeneous appearance resulted from the presence of hemorrhage, necrosis, and cystic change [⁷].

We did not have a preoperative diagnosis of schwannoma and could not confirm as being benign tumor. So, we performed right hemicolectomy and segmental left colectomy on oncologic basis, laparoscopically. Although surgical resection with adequate margins is the treatment of choice, the question still persists as to whether or not radical resection according to oncologic principles should be performed for all GIT schwannomas. Controversy over the optimal extent of resection for these cases arises partly from the rarity of the disease itself and partly from confusion between schwannomas and GANTs. Some authors have proposed that radical resection is necessary for the management of schwannoma, but they seem to have considered schwannomas as merely a subset of GANTs [⁸]. Although both schwannomas and GANTs originate from nerve tissue, however, their clinical courses are quite different. While GANTs are invariably accompanied by aggressive clinical behavior [⁹], digestive schwannomas generally pursue a benign course regardless of the highly variable extent of resection employed for their treatment, ranging from simple enucleation to radical operations [¹,³,⁶]. Braumann et al. [¹⁰] suggested that in the absence of adequate preoperative diagnosis, the extent of excision should be determined by the characteristics of the tumor such as size, location, and consistency. Needless to say, radical excision according to oncologic principles is warranted for an optimal outcome when the tumor shows a higher aggressiveness on pathologic result. However, the complete local excision with a sufficient surgical margin is generally considered to be a correct treatment of colonic schwannomas because the risk of malignant transformation is extremely low [³,⁶].

In summary, because of the uncertainties inherent in traditional methods of diagnosis, we strongly recommend the use of immunohistochemmical assays in order to differentiate schwannomas from other types of mesenchymal tumors such as GISTs and GANTs. Correct preoperative diagnosis, in addition of course to the operative findings or characteristics of the tumor itself, can be important in governing the extent of resection in digestive schwannoma.

No potential conflict of interest relevant to this article was reported.