Published online Dec 14, 2005. doi: 10.3748/wjg.v11.i46.7335
Revised: April 23, 2005
Accepted: April 30, 2005
Published online: December 14, 2005
AIM: To clarify the risk factors for bleeding after endoscopic mucosal resection (EMR).
METHODS: A total of 297 consecutive patients who underwent EMR were enrolled. Some of the patients had multiple lesions. Bleeding requiring endoscopic treatment was defined as bleeding after EMR. Odds ratios (OR) with 95% confidence intervals (CI), calculated by logistic regression with multivariate adjustments for covariates, were the measures of association.
RESULTS: Of the 297 patients, 57 (19.2%) patients with bleeding after EMR were confirmed. With multivariate adjustment, the cutting method of EMR, diameter, and endoscopic pattern of the tumor were associated with the risk of bleeding after EMR. The multivariate-adjusted OR for bleeding after EMR using endoscopic aspiration mucosectomy was 3.07 (95%CI, 1.59-5.92) compared with strip biopsy. The multiple-adjusted OR for bleeding after EMR for the highest quartile (16-50 mm) of tumor diameter was 5.63 (95%CI, 1.84-17.23) compared with that for the lowest (4-7 mm). The multiple-adjusted OR for bleeding after EMR for depressed type of tumor was 4.21 (95%CI, 1.75-10.10) compared with elevated type.
CONCLUSION: It is important to take tumor characteristics (tumor size and endoscopic pattern) and cutting method of EMR into consideration in predicting bleeding after EMR.