Endoscopic mucosal resection of flat and depressed types of early colorectal cancer

The problem of de novo colorectal carcinoma

From April 1985 to March 1995, colonoscopy was carried out at our institution in 24,059 patients, 31,800 times in symptomatic and/or asymptomatic average risk persons. 184 submucosal invasive carcinomas were detected. Unlike protruding-type lesion, the depressed-type invades the submucosal layer, even though the size is within 10 mm. The depressed type of invasive carcinoma accounted for 20 lesions, and represented 10.9% (20 of 184) of all the invasive carcinomas. The pit pattern of depressed-type lesions shows a small round pit (type IIIs pit pattern) and that of carcinoma lesions shows the irregular pit and non-structure (type V pit pattern).

Endoscopic assessment of invasion of colorectal tumors with a new high-frequency ultrasound probe

The present study was carried out to clarify the usefulness of and problems associated with a new ultrasound probe in the assessment of invasion of colorectal tumors. Normal colorectal wall of 23 resected specimens from colorectal cancer cases was examined in vitro with the probe. Fifty-one patients with colorectal carcinoma and 16 patients with rectal carcinoid tumor were presurgically examined using the probe via the biopsy channel of a conventional colonoscope. After endoscopic and ultrasonographic examination, 27 patients underwent endoscopic resection and 40 underwent surgical resection. The ultrasonographic findings were compared with histologic findings in all cases. Carcinomas and carcinoid tumors were visualized as an echo-poor region with the probe. The overall accuracy rate for depth of invasion was 76% (39/51) for colorectal cancer. The accuracy rates for tumors limited to the mucosa and tumors invading the submucosa were 83% and 90%, respectively; these rates were higher than those for tumors invading the muscularis propria (50%) and beyond the muscularis propria (73%). In all 16 cases of carcinoid tumor, the depth of invasion was accurately shown to be limited to the submucosa. This probe is useful to assess invasion of colorectal tumors, especially small and flat lesions limited to the mucosa or submucosa.

A clinicopathologic study of small flat colorectal carcinoma

BACKGROUND

Recent studies have noted that small flat carcinomas do exist in the human colon, but their clinicopathologic features have not been fully delineated.

METHODS

A clinicopathologic study of 62 flat colorectal carcinomas (FCC) and 80 polypoid colorectal carcinomas (PCC) was performed. FCC were defined for the study as colorectal adenocarcinomas confined to the mucosa or submucosa, macroscopically characterized as slightly elevated, often nearly flat, and sometimes with central shallow depressions.

RESULTS

In intramucosal and submucosal invasive carcinomas, the mean size of FCCs was significantly smaller than that of PCCs, (P < 0.001). When compared with PCCs, FCCs were found more often in the proximal colon, less frequently well differentiated, and had fewer adenomatous remnants. In submucosal invasive cases, FCCs showed more frequent deep invasion (67 versus 32%, P < 0.05) and lymphovascular permeation (41% versus 16%, P < 0.05) compared with PCCs.

CONCLUSIONS

These results suggest that FCCs are characterized by frequent deep invasion with lymphovascular permeation even when small, and that they may be more prevalent away from the rectosigmoid area.