Colorectal polyps are eminence lesions that protruded from mucosa into lumen,including adenomatous polyps and non-adenomatous polyps. Early detection and resection of adenoma has significance in prevention of colorectal cancer,and the surveillance afterwards is also crucial. Interval cancer is the colorectal cancer that develops between initial negative colonoscopy or all the polyps having cleared and the next colonoscopy. It is an important duty for endoscopists to reduce the incidence of interval cancer. This article reviewed the advances in studies on surveillance after resection of colorectal adenoma and the development of interval cancer.

Magnetic endoscopic imaging (MEI)is a non-radiographic imaging technique that has been developed in recent years. MEI is capable of displaying real-time three dimensional images of the colonoscope shaft within the abdominal cavity. MEI system has been shown to be beneficial in increasing the cecal intubation rate,helping remove colonoscope insertion tube loops,reducing the duration of looping,assisting training of colonoscopy when compared with traditional colonoscopy. This article reviewed the progress in research on MEI.

0.05). Conclusion Though the noninvasive MRCP offers a diagnostic means equivalent to that of ERCP it is still too early to say it will take the place of ERCP.

Objective To compare the detection rates of capsule endoseopies with push endoseo-pies. Methods From May 2002 through January 2003 , thirty - nine patients with suspected small bowel diseases, particularly the gastrointestinal bleeding of unknown origin were examined by capsule endoscopies. Of the 39 patients. 32 complained of obscure recurrent gastrointestinal bleeding. From January 1993 to October 1996, 36 patients suffered from unexplained GI bleeding underwent push endoscopies. All patients had prior normal results on gastroseopy, colonoscopy, small bowel barium radiography, seintigraphy and/ or angiogra-phy. Results M2A capsule endoscopies disclosed abnormal small bowel findings in 26 out of 32 patients (81% ). Twenty one of 26 patients had significant pathological findings in explaining their clinical disorders with diagnostic yield of 66% (21 of 32 patients). Definite bleeding sites diagnosed by capsule endoscopies in 21 patients including angiodysplasia 8, inflammatory small bowel diseases 5, small bowel polyps 4, GI stro-mal tumor 2,earcinoid tumor and lipoma 1 , and bemorrhagie gastritis 1. Push endoscopies detected the definite sources of bleeding in 9 of 36 patients (25%). Definite bleeding sources included angiodysplasia 2, leiomyosareoma 2, leiomyoma 1 , lymphoma 1 , Grohns disease 1 , small bowel polyps 1 ,and adenocareinoma of ampulla 1. Suspected bleeding sources were seen with push endoscopies in two additional patients, and other five patients with capsule endoscopies. Conclusion Gapsule endoseopy is superior to push endoscopy in detecting obscure GI bleeding ( P

Objective To evaluate the diagnostic efficacy of non-magnified NBI in distinguishing neoplastic from non-neoplastic colorectal polyps,using a simple classification system(NBI international colo-rectal endoscopic[NICE]classification).Methods A total of 181 lesions detected by white light colonosco-py were enrolled in this prospective study.Each lesion was assessed by NBI,and the histology was predicted according to characteristics of lesion color,microvascular architecture,and surface pattern.The results were compared with actual histologic findings.Results The diagnostic sensitivity,specificity,PPV,NPV,and accuracy of NBI were 95. 8%(114/119),91. 9%(57/62),95. 8%(114/119),91. 9%(57/62),and 94. 5%(171/181),respectively.Diagnostic accuracy was 90. 0%(72/80)in the 1 to 5 mm group,95. 9%(47/49)in the 6 to 9 mm group,98. 1%(51/52)in the more than 10 mm group,with no statistically sig-nificant differences(P=0. 175).The sensitivity,specificity,PPV,NPV,and accuracy of NBI for diminu-tive rectosigmoid neoplasms were 87. 5%(14/16),95. 2%(20/21),93. 3%(14/15),90. 9%(20/22), and 91. 9%(34/37).Conclusion NICE classification with non-magnified NBI is effective in distinguishing neoplastic from non-neoplastic colorectal lesions.

Objective To compare the efficacy of rabeprazole and hydrotalcite in treatment of patients with bile-reflux gastritis after cholecystectomy.MethodsPatients,who underwent cholecystectomy and were confirmed with bile reflux gastritis by 24 h gastric bilirubin monitoring,were enrolled in the study.Patients were randomly assigned into control group (n=30),rabeprazole group (n= 30,20 mg daily),hydrotalcite group (n= 29,1.0 g three times daily) and rabeprazole combined with hydrotalcite group (combination group,n= 31) and treated for 8 weeks.Dyspeptic symptoms of abdominal pain,bloating,heartburn and bitter taste were observed.The endoscopic and histological examination were performed 2 weeks after treatment to evaluate the improvement of inflammation and histological activity.The 24 h bilirubin monitoring was used to assess the total per cent of bilirubin absorption (value of 0.14 units or greater) time,the number of reflux episodes and the number of reflux episodes lasting longer than 5 min.Results The dyspeptic symptoms were relieved in three groups after treatment.However,the endoscopic oedema (2.11 ±0.77 vs 1.50 ±0.67,P<0.05) and the histological activity (2.87±0.72 vs 1.97±0.78,P<0.05) as well as the number of reflux episodes> 5 min (18.26+ 1.80 vs 9.70± 1.20) were improved most significantly in combination group after treatment.There was no statistical difference in rabeprazole and hydrotalcite groups before and after treatment.Whereas the total percent time of bilirubin absorption value >0.14 decreased in three treatment groups after treatment (P<0.05).Conclusion Rabeprazole combined with hydrotalcite is effective in the treatment of patients with bile reflux gastritis after cholecystectomy.

Objective To investigate the status of Helicobacter pylori(H.pylori) infection and the accuracy of the two commonly used diagnostic methods, the 14 C urease breath test( 14 C UBT) and the rapid urease test(RUT) in patients with partial gastrectomy due to peptic ulcer or gastric cancer. Methods Culture, histology, RUT, combined with 14 C UBT, were carried out to diagnose the H. pylori infection in patients with partial gastrectomy. The Giemsa staining and H.pylori culture served as the ‘golden reference’ in evaluating the diagnostic accuracy of the RUT and 14 C UBT. The prevalence of H.pylori infection of patients with partial gastrectomy was determined with the dyspeptic controls without surgery. Results 37 patients (17 Billroth Ⅰ resection, 20 Billroth Ⅱ resection) were included in the study. The overall prevalence in the Giemsa staining and culture proved as H.pylori infection was 29.7%, and no statistical difference was seen between patients will Billroth Ⅰ(29.4%) and Billroth Ⅱ(30.0%) resections. The sensitivity of the RUT was 72.7%, the specificity was 57.7%, and the overall accuracy was 62.2%. The sensitivity of the 14 C UBT was 63.6%, the specificity was 100.0%. The negative predictive value was 86.7%, and the overall accuracy was 89.2%. The total prevalence of H.pylori infection in the controls was 71.4%. Conclusion Prevalence of H.pylori infection in patients with partial gastric resection is low. The two diagnostic methods, 14 C UBT and RUT, cannot be recommended as accurate diagnostic tools for H.pylori infection in patients after partial gastrectomy due to low specificity and low sensitivity, respectively.

0.05).②Gabexate mesilate significantly reduced the frequency of contraction (P0.05).③High dose gabexate mesilate could markedly reduce the motility index ( P

Objective To evaluate the effectiveness of wireless capsule endoscopies in patients with suspected CD of the small bowel.Methods From May 2002 through April 2003, we prospectively examined 20 suspected CD patients by capsule endoscopies with normal results in other conventional examinations. It includes the presence of the following symptoms and signs: abdominal pain, weight loss, positive fecal occult blood test, iron deficiency anemia, diarrhea and fever. Mean duration of symptoms before diagnosis was 6 5 years.Results Of the 20 participants, 13 (65%) were diagnosed as having CD of the small bowel according to the findings of the M2A Given Capsule. Among the findings detected by the capsule were mucosal erosions (2 patients), aphthae (5 patients), granulomatous nodularity (1 patient), large ulcers (2 patients), and ulceration with incomplete obstruction (3 patients). All of the 13 patients who received medications showed good clinical improvement.Conclusion Wireless capsule endoscopies were the effective device for diagnosing patients with suspected CD, undetected by conventional diagnostic measures.It may be more likely to detect early lesions in the small bowel of patients with CD.

Objective To determine the detection rate of duodenal papilla and the diagnostic rate of small bowel diseases by capsule endoscopy(CE)and its possible influencing factors.Methods A total of 369 cases with suspected small bowel disease who had undergone CE were retrospectively reviewed. The de- tection rate of duodenal papilla and that of small bowel disease were calculated respectively. The detection rate of duodenal papilla in another 369 patients having received gastroscopy was also counted in order to findout the miss detection rate of forward-viewing endoscopy. The CE findings were reviewed at the rate of 1 im-age and 15 images per second by hand-motion respectively. The detection rate of duodenal papilla was calcu-lated,and the number of frames showing duodenal papilla was counted. Differences of detection rate of duo-denal papilla were also analyzed by different age and gender groups. In addition,the potential difference in detection rate of duodenal papilla between different bowel preparations before CE of two different reviewing approaches was also investigated respectively. Results The miss detection rate of duodenal papilla with for-ward-viewing gastroscopy was 22. 0%. The detection rate of duodenal papilla reviewed at the rate of 1 image by hand-motion,where less frames were needed to visualize duodenal papilla(3. 7vs7. 0),was significantly higher than that at the rate of 15 images per second(32. 2%vs13. 6%,P=0. 001). The diagnostic rate of small bowel diseases was also significantly higher than that of duodenal papilla by CE(67. 2%vs32. 2%,P=0. 000). Age and gender were not significantly correlated with the detection rate of duodenal papilla. No significant correlation was found between various kinds of bowel preparation before CE at different viewing rate and the detection rate of duodenal papilla. Conclusion It iS relatively difficult to identify duodenal pa- pilla by CE,as they are peculiarly positioned in anatomy,so the detection of duodenal papilla does not nec- essarily represent the detection rate of small bowel diseases. The major factors that influencing the detection of duodenal papilla by CE include local anatomy,viewing rate and technical limitations of CE.