Prospective study of the evaluation of the usefulness of tumor typing by narrow band imaging for the differential diagnosis of gastric adenoma and well-differentiated adenocarcinoma.

2011 
Background and Aim:  Presently, the differential diagnosis of gastric adenoma and well-differentiated adenocarcinoma by endoscopy is very difficult. We carried out magnifying endoscopy with narrow band imaging (NBI) in lesions that required discrimination between gastric adenoma and well-differentiated adenocarcinoma, and prospectively evaluated whether the tumor typing that we propose is useful for their differential diagnosis. Methods:  The materials were 93 lesions that required differential diagnosis between gastric adenoma and well-differentiated adenocarcinoma among the gastric epithelial tumors for which endoscopic treatment was planned at three facilities during the 14 months between November 2008 and December 2009. According to the typing method proposed by our facility based on images of the mucosal ultrastructure and microvessels obtained by magnified endoscopy combined with NBI, type I-II and type III-V lesions were diagnosed as gastric adenoma and well-differentiated adenocarcinoma, respectively, before endoscopic treatment, and the accuracy of the diagnoses were prospectively examined by comparing them with the postoperative pathological findings. Results:  Of the 93 lesions, 87 could be typed into the five types of our typing method, but six lesions could not be classified. The 87 lesions consisted of 16 type I, 12 type II, 29 type III, 27 type IV, and three type V lesions. The percentages of accurate preoperative diagnoses of types I-II as adenoma and types III-V as well-differentiated adenocarcinoma were 79% and 93%, respectively. Conclusions:  The tumor typing based on NBI was useful for the endoscopic differentiation of gastric adenoma and well-differentiated adenocarcinoma.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    8
    References
    30
    Citations
    NaN
    KQI
    []