Assessing the quality of the endoscopic abstracts accepted at the United European Gastroenterology Week (UEGW) meeting is al− ways necessarily subjective and may reflect the progress of en− doscopic research as well as peculiarities of the selection com− mittees. Nevertheless, the author's subjective impression for 2005 was that less original work primarily sent to UEGW was available. In fact, many large studies recently presented were not seen at UEGW ± e. g. papers on double−balloon endoscopy and reports on some new colonoscopes. Instead, a few new tech− niques were sometimes reported without clear−cut results, al− though the authors told us how happy they were. Let's hope this will change in the years to come.
Reflux and Barrett's Complex; Early Cancers and ImagingSurveillance of Barrett's esophagus by conventional means, with close observation using modern video endoscopes and four− quadrant biopsies [1, 2], is obviously still thought to be so un− bearable and imperfect that many efforts are being undertaken to improve the situation and avoid the biopsies ± although the percentage of the daily workload taken up by Barrett's esopha− gus surveillance in an average endoscopy center is probably neg− ligible. Staining with various solutions [2, 3] has been suggested as one way of improving lesion detection (although the results have been quite variable [4], and narrow−band imaging (NBI) was developed to more or less replace staining [5, 6]. The Amster− dam group presented the results of a randomized cross−over study in 28 patients with Barrett's esophagus who were exam− ined with high−resolution endoscopy with either staining or NBI within 6 weeks, with no appreciable differences between the two groups [7]. NBI was also tested in the colon on various polyps, and different vascular patterns were described, although clear accuracy rates or correlation values with histology were not given [8].There are currently two ways of imaging early neoplastic lesions: lesion identification and tissue characterization (enthusiastically termed "endoscopic histology"). For identifying lesions, auto− fluorescence [9] is now available in a video format, and ± so far as I understand the abstract ± a group in Japan showed that auto− fluorescence identified five secondary lesions in patients sent for mucosectomy of early gastric cancer, whereas conventional en− doscopy found one [10]. Barrett's esophagus is another area for autofluorescence imaging [11]. In the colon, autofluorescence was used with and without NBI in a variety of lesions, but con− clusions regarding the rate of adenoma detection with autofluor− escence could not be made due to small numbers. NBI prediction of adenomas according to the Kudo classification achieved a sen− sitivity and specificity of 85 % and 90 %, respectively [12].The other area for imaging, tissue characterization, is represen− ted by various techniques such as confocal laser microscopic en− doscopy (CLME) [13 ± 15] and endocytoscopy [16,17]. The latter was investigated in fresh human specimens to test...