We thank Dr. Matsushita et al. for their comments on our article "Resection Depth and Layer of Cold Snare Polypectomy versus Endoscopic Mucosal Resection". However, it appears that their arguments are based on the perspective of mucosal lesions. In mucosal lesions without submucosal invasion, we agree that the submucosal thickness of the resected specimen is not an important matter because it does not affect treatment outcome. In the clinical setting, however, we have encountered diminutive polyps that were apparently mucosal lesions as well as small polyps of 6–9 mm in diameter. In another study, small polyps contained 4% of category 4 or 5 polyps; of those, 6% were deep invasive cancers with lymph node metastasis requiring further surgical treatment [1]. Moreover, some of the malignant small polyps were nonpolypoid and not easily recognized as invasive before endoscopic resection.

Resected specimens require histopathological evaluation, especially at the vertical margin. For polyps, specimens from CSP contained significantly more unevaluable samples than those resected by EMR in our report. Only precise histopathological assessment at vertical margins can identify patients with invasive tumors for proper additional treatment. Simple endoscopic removal of numerous colorectal polyps may indeed reduce the incidence and mortality of CRC, but possibly at the expense of inadequately resected submucosal invasive cancers due to insufficient sampling.

It appears true that EMR specimens tend to have a thicker submucosa as a result of saline injection. However, we maintain that the EMR technique has the ability to resect deeper, and more importantly evaluable, layers than does CSP. This may improve the evaluation of vertical margins in cases of invasive cancer appearing as benign polyps.