ABSTRACT
ObjectiveTo determine the features of antrum gastritis by magnified narrow band imaging (M-NBI) and to evaluate the feasibility of target biopsy with M-NBI.MethodsA total of 69 consecutive patients who underwent M-NBI were recruited in the study.The gastric mucosa of antrum was observed under M-NBI and the gastric microstructure patterns were classified into 5 type.M-NBI findings were compared with the conventional histological findings.In addition,the interobserver agreement was assessed.ResultsA total of 173 gastric antral sites from 69 patients were examined by M-NBI.Type A was mainly normal sites (89.8％,44/49),while type B and C were chronic inflammation (53.3％,24/45 )and active inflammation (74.3％,26/35).Compared with the type A group,the inflammatory degree in type B (P＜0.001 ) group and type C group (P＜0.001) was significantly higher.The sensitivity,specificity,accuracy of the type B and type C pattern for predicting antrum gastritis were 87.5％,78.0％ and 81.5％.The sensitivity,specificity,accuracy of the type D and type E pattern for predicting intestinal metaplasia of antrum were 83.8％,90.4％ and 77.5％,respectively.The kappa value for the interobserver agreement was 0.734 [ 95％ CI(0.591-0.977 ) ].ConclusionMicrostructure features of antrum gastritis determined by M-NBI are highly correlated with histopathology,and can evidently help diagnosis.
ABSTRACT
Objective To investigate the relationship of body mass index with hiatal hernia (HH) and reflux esophagitis (RE).Methods Two hundreds and twenty seven gastroesophageal reflux disease (GERD) patients with typical acid regurgitation and heartburn were enrolled and categorized into three groups according to body mass index (BMI, kg/m2) as normal weight (18.5≤BMI ＜24), overweight (24≤BMI＜28), and obesity (BMI≥28).RE, non-erosive reflux disease (NERD) and HH were diagnosed by gastroscopy.All the patients underwent ambulatory 24-hour pH monitoring and the pathological acid reflux was considered when the DeMeester score≥15.Effects of BMI on RE and HH were estimated by using logistic regression analysis.Results The percentages of RE and HH were 30.0%(68/227) and 5.7%(13/227), respectively.76.9% (10/13) HH patients had RE. Proportions of RE and HH increased significantly with increasing BMI (P＜0.05), so was that of RE above grade B in three groups (6.4%, 16.9% and 31.6%,P=0.003).DeMeester scores of the three groups were 15.9, 19.8 and 36.9, respectively (P＜0.05).The average 24-hour intra-esophagus pH value of overweight group, was significantly lower than that of normal weight patients in the afternoon and midnight (P＜0.01).Multivariate analysis showed obesity was a risk factor for HH with OR 7.058 (95% CI: 1.294～38.488, P=0.024), male (OR: 2.537, 95% CI: 1.350～4.766, P=0.004), overweight (OR: 1.921, 95% CI: 1.005～3.670, P=0.048), obesity (OR: 3.305, 95% CI: 1.123～9.724, P=0.030) and HH (OR: 6.879, 95% CI: 1.695～27.913, P=0.007) were risk factors for RE.Conclusion BMI has a significant association with HH and RE, obesity is a common risk factor for both HH and RE, HH may induce the development of RE.
ABSTRACT
Objective To analyze the differences of morphological and microvascular characteristics between hyperplastic polyps and colorectal adenoma (CA) under narrow band imaging (NBI) without magnification endoscopy,and to evaluate the value of NBI in differential diagnosis.Methods Patients with rectal polyps diagnosed by common endoscopy and pathologically confirmed CA and hyperplastic polyps were recruited in this study and under NBI examination.The pit pattern was divided into type A and B according to modified Kudo pit pattern classification.And the vascular pattern was classified into three types,type Ⅰ with invisible microvascular,type Ⅱ with even microvascular arranged along pit and type Ⅲ with uneven microvascular and irregular arranged.The differences of morphological and microvascular characteristics between hyperplastic polyps and CA were compared and the inter-observer consistency of NBI without magnification endoscopy was evaluated.Results Overall,87 patients with 107 polyps (73 CAs,34 hyperplastic polyps) underwent NBI without magnification endoscopy examination.The maximum diameter and the proportion of polyps with sublobe was higher in CA group than that of hyperplastic polyps group (P =0.0023 and 0.0047).In CA group,most pit shapes were type B (86.3％,63/73),and most vascular pattern types were Ⅱ/Ⅲ (82.2％,60/73).The sensitivity,specificity and accuracy of CA diagnosed with features of type B pit shape or Ⅱ/Ⅲ vascular pattern type was 97.3％,82.4％ and 92.5％.The sensitivity,specificity and accuracy of CA diagnosed with combined features of type B pit shape and Ⅱ/Ⅲ vascular pattern type was 71.2％,91.2％ and 77.6％.The mean kappa value of inter-observer consistency was 0.761.Conclusions There are differences in pit shapes and vascular pattern characteristics between CA and hyperplastic polyps.According to these two facts,CA and hyperplastic polyps can be initially differential diagnosed by NBI without magnification endoscopy.
ABSTRACT
Objective To investigate the value of colonoscopy with narrow band imaging (NBI) in predicting the histology of colorectal polypoid lesions. Methods A total of 173 colorectal polypoid lesions from 125 patients were included in the study. The lesions were diagnosed as neoplastic (adenoma or cancer)or non-neoplastic based on pit patterns and/or capillary patterns (CP), which was compared with the results of pathology, and the sensitivity, specificity and accuracy of each method were evaluated. Results In differentiation between neoplastic and non-neoplastic lesions, the sensitivity, specificity and accuracy of CP (94. 83% , 91. 23% and 93.64% , respectively), and those of CP combined with pit patterns (95. 69% ,96.49% and 95. 59% , respectively) were significantly higher than those of conventional colonoscopy (80. 17% , 84.21% and 81.50% , respectively, P<0.05). In differentiation between adenoma and cancer,the sensitivity, specificity and accuracy of CP were 86.90% , 100.00% and 87.93% , respectively.Conclusion NBI is superior to conventional colonoscopy in differentiation between neoplastic and non-neoplastic lesions, as well as in differentiation between adenoma and carcinoma.
ABSTRACT
Objective To discuss the value of transcatheter artery embolization for the treatment of GI massive hemorrhage. Methods Seventy-eight GI massive bleeding cases underwent emergency angiography. Intraarterial embolization was performed in 86 arteries of the 78 patients. All patients were followed up for 1 to 10 years. Results Bleeding stopped immediately in 100% of the patients. Bleeding recurred in 16 cases from 48 hours to 37 monthes. Conclusion Arterial embolization for digestive tract bleeding is safe and effective during emergency angiography to buy a time for definite treatment for some patients.