Magnifying endoscopy combined with narrow band imaging system for early gastric cancer: correlation of vascular pattern with histopathology (including video).

BACKGROUND AND STUDY AIMS: The narrow band imaging (NBI) system consists of a sequential electronic endoscope system and a source of light equipped with new narrow band filters, yielding very clear images of microvessels on mucosal surfaces. The aim of this prospective study was to measure the correlation between the magnified images obtained with the NBI system and the histological findings, especially with regard to the vascular pattern. In addition, three-dimensional images of microvessels were reconstructed using a laser scanning microscope.

PATIENTS AND METHODS: Between July 2001 and August 2003, 165 patients with depressed-type early gastric cancer lesions were enrolled in the study. The lesions were carefully observed with magnification using the NBI system. The images, the pathological characteristics of the lesions, and three-dimensionally reconstructed images of the microvascular networks in biopsied specimens were carefully analyzed. The microvascular patterns were classified into three groups: A, fine network; B, corkscrew; and C, unclassified pattern. The endoscopic images were compared with the histological findings.

RESULTS: Of the three types of filter available for use with the NBI system, microvascular formation was best enhanced in B mode images produced using short wavelengths, which focus on the superficial mucosal layer. Among 109 cases of differentiated adenocarcinoma, the group A microvascular pattern was observed in 72 cases (66.1 %). Among 56 cases of undifferentiated adenocarcinoma, the group B pattern was observed in 48 cases (85.7 %; P = 0.0011) The microvascular structure observed using the NBI system corresponded with the superficial mucosal layer in the three-dimensional images obtained using laser scanning microscopy and the resected specimens.

CONCLUSIONS: Magnifying endoscopy performed in combination with the NBI system is not sufficient to replace conventional histology, but is capable of predicting the histological characteristics of gastric cancer lesions.