The lateral resolution of the image is 0.7 µm and optical slice thickness 7 µm. Confocal and endoscopy images can be generated at the same time. Before endoscopy, 20 000 U chymotrypsin was given to each participant to remove gastric mucus. One endoscopist (XMG) experienced with endomicroscopy carried out the CLE. After conventional endoscopy, 15 mL acriflavine hydrochloride (0.05%; Sigma Aldrich,

Germany) was applied topically by use of a spray catheter. The greater and lesser curvature of the antrum and corpus were carefully observed separately PD0332991 on CLE. At least 10 images were taken from each site. Specific mucosal epithelium changes in the body and antrum of the stomach were identified. H. pylori infection was diagnosed on the basis of CLE criteria while the image was generated. At least one site with positive changes related to H. pylori infection was considered H. pylori infection. If obvious lesions were seen, they were additionally scanned, and a biopsy was taken if necessary. CLE scanning was carried out far enough apart to avoid the influence of these lesions. All images www.selleckchem.com/products/abc294640.html were stored as digital files available for analysis after the procedure. Two biopsy specimens from the greater curvature of the gastric antrum and corpus were obtained for histopathology examination.

The biopsy samples were fixed in 10% formalin, processed routinely, and embedded in paraffin. Serial sections at 4-µm intervals were stained with hematoxylin and eosin and Giemsa. An experienced pathologist (CJZ) who was blinded to the CLE results

evaluated the biopsy specimens. One biopsy specimen was taken from the antrum within 3 cm of the pylorus for the rapid urease test. We defined a case with a positive rapid urease test and Giemsa staining results as H. pylori infection. If one of the results was negative, a further 13C-urea breath test was used to confirm infection. Data were collected by one of investigators (RJ) using a standardized collection form designed for the study. To assess interobserver agreement, three endoscopists (YQL, TY, and XLZ), who were blinded to the H. pylori diagnosis were asked to reassess the CLE images. We selected at random 50 digitally stored images check details from 50 patients. Each image was assessed by the CLE criteria. Interobserver agreement was determined by kappa value: values of 0.01–0.2 indicating slight agreement, 0.21–0.4 fair, 0.41–0.6 moderate, 0.61–0.8 substantial and 0.81–0.99 almost perfect. The chi-squared test was used to compare the observed sample distribution. The sensitivity, specificity, positive and negative predictive values of the CLE patterns were calculated. A P-value of < 0.05 was considered statistically significant. Calculations involved the use of SPSS v11.0 (SPSS Inc., Chicago, IL, USA). Confocal laser endomicroscopy images corresponded well with transverse sections of histopathology from the same sites.