Usefulness of NBI for Complete Endoscopic Resection of SSA/P

This study is a prospective randomized multi-center study to evaluate the role of NBI for improving complete resection rate of sessile serrated adenoma/polyp (SSA/P).

The authors will enroll consecutive patients who underwent colon polypectomy for suspicious SSA/P during colonoscopy.

All patients who undergo polypectomy will be enrolled, but excluded as follows; 1. poor bowel preparation according to Boston bowel preparation scale calculation, 2. inflammatory bowel disease, 3. failure of cecal insertion, 4. continuous anti-plate agents or anti-coagulation agents users.

First, the authors will inspect polyp using the arbitrary endoscopic criteria by inspection of white light endoscopy (WLE) and if that lesion is satisfied with 3 or more than 3 out of 5 categories (1. located in the right colon, 2. flat shape, 3. covered with mucus cap or debris, 4. 1 cm or larger than 1 cm in size, 5. varicose superficial vessel was seen on the surface after submucosal injection), endoscopic resection will be performed.

And then, the authors will inspect the resection margin carefully using NBI or WLE by pre-randomization before endoscopic resection. If the remnant tissue is detected, additional endoscopic resection will be performed.

ed margin as two groups (NBI inspection group and WLE inspection

And then, 4 biopsies from 4 quadrants of margin for evaluation of complete resection.

Primary end-point is comparing the rate of additional resection for remnant lesion, the SSA/P diagnosis rate of the remnant lesions after the additional resection, and the complete resection rate of the target lesion between two groups.

Secondary aim is evaluate the accuracy for SSA/P using arbitrary WLE criteria. For evaluation of the accuracy of arbitrary WLE criteria, the authors will compare the pathologic diagosis rate of SSA/P for each criterion, and calculate the sensitivity, the specificity, the positive predictive value, the negative predictive value, and the accuracy with area under curve of ROC.

The authors hypothesized complete resection rate of SSA/P as 90% in the NBI inspection group and 70% in the WLE inspection group based on previous studies including CARE study.

The sample size was calculated as 118 (each 59) patients of SSA/P. Study duration will be 2 years.

Inclusion criteria

• Colonoscopy examinee with suspicious SSA/P

Exclusion criteria

• Poor bowel preparation (total Boston bowel preparation scale < 6 or any region score < 2)
• Inflammatory bowel disease
• failed cecal insertion of colonoscopy
• continuous taking anticoagulation or anti-platelet drugs until on the procedure day