CLINICAL ROLE OF ENDOSCOPIC ULTRASONOGRAPHY FOR THE DIAGNOSIS OF EARLY COLORECTAL CANCER AND SELECTING THE TREATMENT PROCEDURE

Abstract

Background: Accurate evaluation of the depth of tumor invasion, including the degree of submucosal invasion, is a prerequisite to selecting the treatment procedure for early colorectal cancer (CRC). The purpose of the present study was to evaluate the significance of endoscopic ultrasonography (EUS) for diagnosing the depth of invasion of early CRC and selecting the treatment procedure. We concurrently estimated the usefulness of three-dimensional EUS (3-D-EUS) compared with that of conventional EUS. Methods: We studied 413 consecutive early CRC for which the depth of invasion was examined by EUS. They consisted of 239 lesions of mucosal cancers and 174 lesions of submucosal cancers (sm cancers). We divided sm cancers into two groups, sm-slight cancers (38 lesions) and sm-massive cancers (136 lesions), according to the degree of infiltration in the vertical direction in the submucosa. The diagnostic accuracy of the depth of cancerous invasion by EUS and the characteristics of tumors that were difficult to image by EUS were examined. For 59 lesions, the depth of invasion was concurrently evaluated by 3-D-EUS to compare the clinical usefulness of this diagnostic tool with that of conventional EUS. Results: In 364 lesions (88%) of early CRC, we could diagnose the depth of invasion by EUS. Differentiation between mucosal or sm-slight cancers, which were generally treated by endoscopic resection or local excision, and sm-massive cancers, which were suitable for radical operation, was possible in 90%. A total of 49 lesions (12%) could not be imaged by EUS. Difficulty in imaging often occurred with lesions located proximally to the transverse colon and with protruded-type lesions. The accuracy rate of 3-D-EUS for differentiating between mucosal or sm-slight cancers and sm-massive cancers, including difficult-to-image lesions, was 86%. This figure was slightly, but not significantly higher, than the accuracy rate of 73% for conventional EUS (P = 0.07). However, the concurrent application of 3-D-EUS was considered useful in 31 of the 59 lesions (53%) evaluated by both techniques. Conclusion: EUS is useful for evaluating the depth of tumor invasion and selecting the treatment procedure for early CRC. The concurrent use of 3-D-EUS may further improve diagnostic accuracy and decrease the number of difficult-to-image lesions.