4727 Reassessment of usefulness of eus in differentiation of benign and malignant gastric stromal tumors which were diagnosed according to pathologic guidelines of ackerman's surgical pathology.

Abstract

Backgrounds: GISTs originate from mesenchymal tissue within the GI tract, and it is difficult to differentiate between benign and malignancy by endoscopy. Although recent development of EUS allows a trial of differentiation from benign and malignant stromal tumors, its diagnostic accuracy is controversial. The aim of this study is to determine whether EUS can differentiate between benign and malignant gastric stromal tumors. Materials and Methods: EUS was performed prior to endoscopic resection or surgical resection of 38 gastric stromal tumors. EUS features such as tumor size(diameter>4 cm), irregular extraluminal border, echogenic foci, and cystic spaces, which were proposed by Chak A et al. as EUS findings of malignant GISTs(Gastrointest Endosc 1997) were evaluated. Pathologic diagnosis was based on guidelines of Ackerman's surgical pathology (Ackerman's surgical pathology 8th ed, 1996). Results: Subjects consisted of 12 males and 26 females. 22 cases has been endoscopically resected, 16 cases surgically removed. The results of immunohistochemical staining showed 12 cases of smooth muscle type, 7 cases of neural type, 1 case of mixed type and 18 cases of uncommitted type. On the basis of above immunohistochemical results and histologic findings, 12 cases were regarded as benign, 17 cases as potentially malignant or malignant, and 9 cases as malignant. Analyses of EUS features according to the criteria of Chak et al. are shown in the following table. Conclusion: When gastric stromal tumors were diagnosed with criteria from Ackerman's surgical pathology, EUS features, which were proposed by Chak et al, were assessed to be less useful in differentiating between benign and malignant stromal tumors. Backgrounds: GISTs originate from mesenchymal tissue within the GI tract, and it is difficult to differentiate between benign and malignancy by endoscopy. Although recent development of EUS allows a trial of differentiation from benign and malignant stromal tumors, its diagnostic accuracy is controversial. The aim of this study is to determine whether EUS can differentiate between benign and malignant gastric stromal tumors. Materials and Methods: EUS was performed prior to endoscopic resection or surgical resection of 38 gastric stromal tumors. EUS features such as tumor size(diameter>4 cm), irregular extraluminal border, echogenic foci, and cystic spaces, which were proposed by Chak A et al. as EUS findings of malignant GISTs(Gastrointest Endosc 1997) were evaluated. Pathologic diagnosis was based on guidelines of Ackerman's surgical pathology (Ackerman's surgical pathology 8th ed, 1996). Results: Subjects consisted of 12 males and 26 females. 22 cases has been endoscopically resected, 16 cases surgically removed. The results of immunohistochemical staining showed 12 cases of smooth muscle type, 7 cases of neural type, 1 case of mixed type and 18 cases of uncommitted type. On the basis of above immunohistochemical results and histologic findings, 12 cases were regarded as benign, 17 cases as potentially malignant or malignant, and 9 cases as malignant. Analyses of EUS features according to the criteria of Chak et al. are shown in the following table. Conclusion: When gastric stromal tumors were diagnosed with criteria from Ackerman's surgical pathology, EUS features, which were proposed by Chak et al, were assessed to be less useful in differentiating between benign and malignant stromal tumors.