6957 Evaluation of usefulness of endoscopic ultrasonography for diagnosis of villous tumors

• Published in: Gastrointestinal endoscopy 2000-04, Vol.51 (4), p.AB230-AB230
• Main Authors: Konishi, Kazuo, Akita, Yasushi, Kaneko, Kazuhiro, Kurahashi, Toshinori, Yoshikawa, Nozomi, Mitamura, Keiji
• Format: Article
• Language: English
• ISSN: 0016-5107; 1097-6779
• DOI: 10.1016/S0016-5107(00)14628-0

Background: Villous tumors are potentially or substantially malignant lesions. A detailed examination of this tumor is important for determining therapeutic strategy. We used to frequently observe a peculiar finding in villous tumors during endoscopic ultrasonography (EUS). The study of villous tumor based on EUS is not enough performed, although reports of usefulness of EUS for pretherapeutic staging of colorectal cancer have increased. Aim of this study is to clarify the characteristics of EUS findings of villous tumors. Methods: We performed EUS for 52 lesions histologically diagnosed as mucosal lesion (adenoma and intramucosal cancer) or submucosal cancer at Showa University Hospital from 1993 to 1999. The instrument used for EUS is the CF-UM200 with a 7.5 MHz radial-scan transducer (Olympus Optical Co.Ltd., Tokyo, Japan). These lesions were histologically classified into three categories: non-villous type; 24 cases (less than 25% of villous component), tubulovillous type; 21 cases (25-75% of villous component) and villous type; 7 cases (more than 75% of villous component). In this study, both tubulovillous type and villous type were defined as villous tumor. We compared EUS findings with histologic findings of the lesions. Results: By EUS, the non-villous tumor was demonstrated as a hypoechoic mass and its echo level was intermediate between the third hyperechoic layer and the fourth hypoechoic layer of colorectal wall. The echo-pattern of villous tumor contained multiple cystic area. The incidence of this echo-pattern was significantly different between villous tumors (18/28; 64%) and non-villous tumors (5/24; 21%)(p=0.0211). In the further investigation, cystic area was diffusely distributed in 7/7 (100%) of villous type and 5/21 (24%) of tubulovillous type (p=0.0007). But, there was no diffuse distribution in non-villous type. The depth of invasion was correctly determined by EUS as follows: mucosal lesion; 58% of villous tumors and 80% of non-villous tumors, submucosal cancers; 50% of villous tumors and 78% of non-villous tumors, respectively. Conclusions: EUS is useful for the diagnosis of villous tumors, however it is difficult to determine the depth of invasion of villous tumor by only EUS.