Colorectal cancer (CRC) is a serious health problem

Impact of Wide-Angle, High-Definition Endoscopy in the Diagnosis of Colorectal Neoplasia: A Randomized Controlled Trial MARIA PELLISÉ, GLÒRIA FERNÁNDEZ ESPARRACH, ANDRÉS CÁRDENAS, ORIOL SENDINO, ELENA RICART, EVA VAQUERO, ANTONIO Z. GIMENO GARCÍA, CRISTINA RODRÍGUEZ DE MIGUEL, MICHEL ZABALZA, ANGELS GINÈS, JOSEP M. PIQUÉ, JOSEP LLACH, and ANTONI CASTELLS Endoscopy Unit, Gastroenterology Department, Institut de Malalties Digestives i Metabòliques, Hospital Clínic, CIBERehd, IDIBAPS, University of Barcelona, Barcelona, Spain See editorial on page 1035. Background & Aims: It is essential to optimize standard technique to be able to increase polyp detection. We sought to compare the performance of using a high-definition, wide-angle endoscope (HDE) versus a standard colonoscope (SC) for the detection of colorectal neoplasia. Patients and Methods: All consecutive consenting adult patients referred from primary care centers were included and randomly assigned at a 1:1 ratio to undergo HDE or SC. Times to reach and withdraw from the cecum were measured. Morphology, size, location, and pathologic diagnosis of each polyp were recorded. Sample size calculation resulted in a total of 682 patients needed. Results: A total of 693 consecutive patients fulfilled all inclusion criteria (73 excluded owing to insufficient bowel preparation). Each arm included 310 patients with no baseline characteristic differences. Time to reach the cecum was slightly superior for SC (8.9 4.8 minutes vs 8.2 4.5 minutes; P.055). Pathology examination was feasible in 418 lesions (272 adenomas, 109 hyperplastic polyps, and 37 inflammatory lesions). Both techniques detected a similar number and type of lesions, and there were no differences in the distribution along the colon, in the degree of dysplasia, or morphology of adenomas. The per-patient basis analyses demonstrated that there were no differences between the 2 arms of the study in the detection rates of polyps (SC, 0.84 1.59; HDE, 0.83 1.30), adenomas (0.45 1.07 vs 0.43 0.87), small adenomas (0.22 0.71 vs 0.28 0.78), flat adenomas (0.30 0.91 vs 0.21 0.63), or hyperplastic polyps (0.16 0.50 vs 0.18 0.54). Conclusion: HDE did not detect significantly more colorectal neoplasia than SC. Colorectal cancer (CRC) is a serious health problem in the general population; it is the third most common neoplasia and the second leading cause of cancer death in Western countries. 1 It is considered that at least two thirds of CRC develop through the adenoma carcinoma pathway. 2 Consequently, screening for CRC and its precursor lesion has become an increasingly prevalent practice. Colonoscopy is the most effective diagnostic and therapeutic tool for secondary and tertiary prevention of CRC; it enables polyp identification as well as polyp removal. In fact, large prospective studies have shown substantial reductions in the expected risk of CRC during long-term follow-up after screening. 3 6 However, the benefit of varies among the studies. A recent population-based retrospective study with 35,975 patients demonstrates that, after a negative result from a performed in the usual clinical practice, the risk of developing CRC is quite low but persists to be still at most 60% 70% of the corresponding figure in the general population. 7 This lack of total efficacy of could lie in biological factors (rapid tumor growth or alternative pathway to the adenoma carcinoma sequence), but it may also be explained by technical limitations of endoscopy. In that sense, recent analyses of pooled data from polyp prevention trials suggest that the reduction in CRC incidence after polypectomy and clearing may be lower in real-world practice than what had been previously documented in controlled trials. 8,9 In fact, several comparative and back-to-back studies have demonstrated missed lesion rates of 15% 27% for small adenomas and 6% 12% for lesions 1 cm. 3,10 13 Abbreviations used in this paper: CRC, colorectal cancer; HDE, highdefinition endoscope; NBI, narrow-band imaging; SC, standard. 2008 by the AGA Institute 0016-5085/08/$34.00 doi:10.1053/j.gastro.2008.06.090 1062

It is, therefore, essential to optimize technique to be able to increase polyp detection and thus reduce substantially the onset of CRC, resulting in an improvement of survival. There are several actions that are aimed at achieving this objective, for example, ensuring a perfect preparation of the colon, 14,15 performing a complete up to the cecum, 16 and carrying out a thorough examination of the mucosa, including a long withdrawal inspection time. 17 Moreover, in the last years, several technologies have been developed to improve performances. 18 30 Chromoendoscopy with contrast dyes is the best-studied method to improve detection of flat lesions during, but it is time consuming and impractical. 23 30 High-definition coupled with chromoscopy has also been tested and shown to increase purely flat adenoma and hyperplastic polyp detection, but not the overall adenoma detection rate. 30,31 Recently, a newly commercialized colonoscope with advanced features has been developed aimed at improving polyp detection rate. This colonoscope has 4 distinct features: (1) Narrow-band imaging (NBI): A modification in the light beam enhances visualization of the network of the mucosa providing contrast and acting as a substitute of chromoendoscopy. This system offers the possibility to switch from conventional white light to blue NBI light alternatively; (2) High definition: The system uses a high-definition 1080-line television and a highresolution CCD with up to 1 million pixels, which provides images double the quality of normal television; (3) Wide angle: The scope has a field of vision of 170 (30% more than the conventional model) that is supposed to improve the detection of lesions hidden behind the folds of the colon; and (4) Close focus (2 mm) and electronic zoom ( 1.5): This achieves a 70 magnification effect, useful for mucosal structure enhancement. Recently, Rex et al 19 conducted a randomized controlled trial using this new endoscope and comparing withdrawal in white light versus NBI in 434 patients. The results demonstrate that NBI did not result in better detection of adenomas. However, it is important to note that, in this trial, the adenoma detection rate with both techniques was substantially higher than that previously reported in most of the studies. The authors suggested that this could be explained by the fact that a wide-angle, high-definition endoscope (HDE) was used in both arms of the study. It is logical to assume that a higher resolution endoscope could provide better results but, to date, this hypothesis has not been validated. We conducted a randomized, controlled trial aimed at comparing the performance of using a wideangle, HDE versus a conventional endoscope for the detection of colorectal neoplasia. This trial was carried Figure 1. Enrollment of subjects (following the STARD initiative 33 ). out in a nonselected population and in a clinical setting representative of real-life conditions. Patients and Methods General Study Design This study was conducted in patients undergoing at a community-based hospital. The study was approved by the Institutional Review Board and the Ethics Committee at Hospital Clinic de Barcelona. The study population consisted of consecutive consenting adult patients (age 18) referred from primary care centers for outpatient between September 2006 and July 2007. Patients were excluded if they had polyposis syndromes or hereditary nonpolyposis CRC, previous surgical resection of the colon or rectum, or inflammatory bowel disease. Patients were randomly assigned at a 1:1 ratio to undergo with wide-angle, HDE, or standard (SC). Both HDE and SC were performed by the same team of endoscopists (see below). No stratification was done according to the indication for. Colonoscopy Technique Colonoscopies were performed during standard 40-minutes time slots. Seven full-time, board-certified gastroenterologists who spend 50% of their daily activity in endoscopy performed the procedures. In the SC group, was performed with a standard-resolution, standard-angle (140 ) videocolonoscope (Olympus Q160, Olympus Europe, Hamburg, Germany) with a highdefinition 1080-line screen. In the HDE group, was performed with a wide-angle (170 ), high-resolution videoendoscope (Olympus prototype XCF H160AY2L corresponding to Olympus H180 series, Olympus Europe) with a high-definition 1080-line screen. Bowel preparation included ingestion of 2 5 L of polyethylene glycol 1063

Table 1. Demographics, Indications, and Colonoscopy Baseline Characteristics in Both Arms of the Study Standard High-definition P value No. of patients 310 310 Male, no. (%) 147 (47) 151 (49).81 Age, years a 58.6 16.0 58.5 17.1.95 Indication CRC screening and 82 (26) 70 (23).53 surveillance, no. (%) Colorectal symptoms, 201 (65) 211 (68) no. (%) Others, no. (%) 27 (9) 29 (9) NSAID use, b no. (%) 60 (19) 53 (17).53 Previous, no. (%) 92 (30) 84 (27).48 Family history of CRC, no. (%) 79 (25) 61 (20).08 Personal history of adenomas, 50 (16) 40 (13).25 no. (%) Bowel cleansing Excellent, no. (%) 247 (80) 248 (80).58 Good, no. (%) 53 (17) 56 (18) Fair, no. (%) 10 (3) 6 (2) Complete, no. (%) 289 (93) 292 (94).62 Time to cecum, min a 8.9 4.8 8.2 4.5.05 Time for withdrawal, min a 6.3 2 6.2 1.9.69 CRC, colorectal cancer; NSAID, nonsteroidal anti-inflammatory drug. a Expressed as mean values standard deviation. b Treatment with any NSAID was defined as the use of a drug of this class at any time during the 7 days before. electrolyte solution before the examination and a dietary restriction of solid food 1 2 days before the procedure. The quality of preparation was graded by the endoscopist as follows: (1) excellent, no solid or liquid residue; (2) good, complete mucosal examination after aspiration; (3) fair, persistence of residue despite aspiration, thus preventing correct examination of 5% 20% of the mucosa; and (4) poor, persistence of residue despite aspiration, thus preventing correct examination of 20% of the mucosa. Patients with poor preparation were excluded from the study. All colonoscopies were performed with the patient under conscious sedation with intravenous midazolam plus fentanyl. Times to reach and withdraw from the cecum were measured using a stopwatch. The stopwatch was started when the first endoscopic view of the rectum was obtained and stopped when the cecal caput was entered with the tip of the colonoscope. The success of cecal intubation was verified by identification of the usual landmarks (ileocecal valve, triradiate cecal folds, and appendix orifice). Insertion time includes time for passing the instrument from the rectum to the cecum with the exception of time taken for any polypectomy performed during insertion. No attempt was made to systematically examine for polyps during insertion. During extubation, each segment was carefully examined by using standardized procedure, at first with maximal insufflation to detect color changes and, then, with minimal insufflation Table 2. Characteristics of Endoscopic Findings in Both Arms of the Study Standard High-definition P value No. of evaluated lesions 201 217 Location, no. (%) Proximal 148 (57) 135 (51).41 Distal 78 (30) 92 (35) Rectal 34 (13) 35 (13) Pathology, no. (%) Adenoma 119 (59) 125 (58).04 Hyperplastic 51 (25) 58 (27) Carcinoma 19 (9) 9 (4) Others 12 (6) 25 (11) No. of adenomas 119 125 Degree of dysplasia, no. (%) Low grade 115 (97) 120 (96).94 High grade 4 (3) 5 (4) Morphology, no. (%) Flat 33 (28) 26 (21).26 Sessile or pedunculate 86 (72) 99 (79) Size, no. (%) 5 mm 69 (58) 88 (70).06 5 mm 50 (42) 37 (30) Advanced adenomas,* no. (%) 49 (41) 42 (34).27 *Advanced adenomas were defined as adenomas 10 mm, with villous component or high-grade dysplasia. to detect small mucosal surface abnormalities. Targeted indigo carmine chromoscopy was used for a better definition of suspicious lesions, but not for increasing the yield of polyp detection. The watch was restarted as soon as examination of the cecum began and was stopped when the scope was withdrawn from the anus. It was stopped whenever a polyp was identified until the polyp had been removed and retrieved, for all suctioning and washing of the colon, and for any biopsy specimens Table 3. Detection Rate of Significant Lesions in Both Study Arms Standard High-definition P value No. of evaluated patients 310 310 No. of polyps per patient* 0.84 1.59 0.83 1.30.96 No. of adenomas per patient* 0.45 1.07 0.43 0.87.87 No. of adenomas 5 mmper 0.22 0.71 0.28 0.78.30 patient* No. of purely flat adenomas per 0.30 0.91 0.21 0.63.42 patient* No. of hyperplastic polyps per 0.16 0.50 0.18 0.54.64 patient* Patients with 1 polyp (%) 119 (38) 134 (43).22 Patients with 1 adenoma (%) 79 (25) 82 (26).85 Patients with 3 adenomas (%) 16 (5) 10 (3).32 Patients with 1 hyperplastic 39 (12) 42 (13).81 polyp (%) Patients with HGD adenoma or carcinoma (%) 13 (4) 14 (4) 1.00 HGD, high-grade dysplasia. *Expressed as mean value standard deviation. 1064

taken. The vast majority of polyps were removed during extubation. A minimum of 6 minutes in withdrawal was advised. 17 The morphology, size, and location of each polyp were recorded. The polyp morphology was determined following the Paris classification 32 and the polyp size was estimated by placing full open biopsy forceps next to the lesion before resection. All visible lesions were resected, except some polyps in the rectosigmoid, 4 mm that seemed to be hyperplastic (pit pattern type 1 or type 2). Lesions 5 mm were resected using biopsy forceps. Pedunculated polyps or polyps of 5 mm were resected by polypectomy snare. Protusive nonpedunculated polyps and flat lesions 5 mm were resected by mucosectomy. Statistical Analysis Demographic data, personal and family case histories of cancer, information from any previous colonoscopies, and the indication for the present were recorded. The primary outcome measure was the rate of neoplasia detection per patient in each group. A sample size calculation was performed assuming from the most recent previous data 31 that the mean number of adenomas per patient in the control group would be 0.5 0.9. The study was powered to establish a 30% significant increase in the rate of adenoma detection in the HDE group. With these parameters, we calculated that a total of 340 patients in each group were required to achieve significance ( error 0.05, error 0.1, and dropouts 10%). Data are reported as mean values standard deviation unless otherwise noted. Quantitative and qualitative variables were compared by means of the Student s t test and the 2 test, respectively. Statistical analysis has been performed with the use of SPSS software, version 12.0 (SPSS, Chicago, IL). Results A total of 693 consecutive patients referred from primary care centers and attending for outpatient fulfilled inclusion criteria and were enrolled in the study. Patients were randomly assigned at a 1:1 ratio to undergo with the wide-angle, highdefinition colonoscope (HDE group) or with the conventional colonoscope (SC group). There were no differences with respect to the number of both procedures performed by each of the 7 endoscopists involved in the study (data not shown). Seventy-three patients had to be excluded from further analysis owing to insufficient bowel preparation that precluded satisfactory examination; 620 patients completed the study protocol (HDE, 310 patients; SC, 310 patients; Figure 1). 33 Patient characteristics in both arms of the study, including age, gender, indication for the procedure, previous study, family history of CRC, and personal history of adenomatous polyps, as well as quality of bowel preparation, completion of the, insertion time, and withdrawal time are shown in Table 1. There were no differences between the 2 arms for any of these variables. Nevertheless, time to reach the cecum was slightly superior for SC than for HDE (8.9 4.8 min vs 8.2 4.5 minutes, respectively), but the difference was not significant (P.055). A total of 518 lesions were identified in the entire study population and pathology examination was feasible in 418 of these lesions (81%). Of them, 272 were adenomas (65%), 109 hyperplastic polyps (26%), and 37 inflammatory lesions (9%). Two hundred thirty-five adenomas had low-grade dysplasia (86%), whereas 37 had high-grade dysplasia or carcinoma (14%). Approximately half of the adenomas were 5 mm (52%), but most were sessile or pedunculated (68%). Table 2 summarizes characteristics of endoscopic findings in both arms of the study. Both techniques detected a similar number and type of lesions, and there were no differences in the distribution along the colon. Moreover, there were no differences in both arms of the study in the degree of dysplasia or morphology of adenomas. However, HDE detected more small adenomas than SC (69 vs 88, respectively), but the difference was not significant (P.06; Table 2). A per-patient basis analyses was performed demonstrating that there was no difference between the 2 arms of the study in the detection rates of polyps, adenomas, small adenomas, flat adenomas, or hyperplastic polyps per patient (Table 3). Moreover, the number of patients with 1 adenoma was similar in the SC and the HDE groups (79 vs 82, respectively; P.85). When considering those individuals at higher risk for developing CRC, the number of patients with 3 adenomas (SC group, 16 vs HDE group, 10; P.32) or with high-grade adenomas (SC group, 13 vs HDE group, 14; P 1.00) were also similar. Finally, the number of patients with 1 hyperplastic polyp was also similar in both groups (SC, 39 vs HDE, 42; P.81). Discussion To our knowledge, this is the first report of a randomized controlled trial in a nonselected population comparing a new colonoscope with wide-angle and highdefinition capabilities with the conventional colonoscope in the diagnosis of colorectal neoplasia. In this study, HDE did not detect significantly more adenomas than SC. In addition, no statistically significant difference was detected in the overall detection of polyps, in the ratio of flat or small adenomas, or in the identification of patients at higher risk for developing CRC. The main strength of this study relies on the fact that it reflects the true diagnostic yield of this new technology 1065

in a routine practice with a nonselected population and nondedicated endoscopists. This information is important to critically assess the role of this commercially available colonoscope in clinical practice. We are aware, however, of some limitations of the study. First, our study is underpowered to detect differences between groups 30% in the rate of adenoma detection. Nevertheless, this figure is similar to the one used in studies of analogous characteristics 31 and somehow represents the limit of a clinically meaningful difference. Second, bowel preparation was not excellent/good in all individuals. However, considering that the proportion of patients with fair cleansing was minor and similar in both groups, it seems that this factor may not influence the results. Finally, instead of using chromoendoscopy systematically, targeted indigo carmine staining was only used for a better definition of suspicious lesions. The rationale behind this approach relies on the fact that the study was specifically designed to ascertain the pure contribution of HDE rather than optimizing endoscopic strategies. A critical issue to interpret this study is the quality of procedures in both arms. In that sense, the overall findings of in this report are similar to those described previously. Whereas, the adenoma per patient rate in the control group is higher than those reported in the control group of various comparative studies with chromoscopy or high-resolution techniques, 17,20,27 31,34,35 the adenoma rate in the study group is lower than some previous reports with high-resolution colonoscopes. The multicenter French study 31 comparing high-resolution with chromoscopy versus SC, found an adenoma rate slightly superior to our results (0.6 1vs 0.43 0.87). It is to be noted that these authors were using a higher resolution colonoscope (Fujinon EC485ZW) with the addition of panchromoendoscopy. On the other hand, in the recent trial by Rex et al 19 comparing withdrawal with white light versus NBI using the wide-angle, high-resolution colonoscope, the adenoma rate in the control group was sharply higher than in our series (1.8 2.2 vs 0.43 0.87). Technical shortcomings could partially explain this difference because quality of bowel preparation was fair in 20% of cases and the rate of complete colonoscopies was not total (94%). Moreover, it has to be noted that the population included in our study was not a selected or a high-risk population because only 25% were referred for screening or surveillance reasons, only 15% had a personal history of CRC including adenomas, and the mean age was 60 years old. However, the most important explanation for the excellent results in the study by Rex et al is that only 1 very experienced endoscopist was involved in the study. In that regard, our results probably reflect more accurately the true diagnostic yield of the tested technique when performed in conditions of routine practice. In fact, data from a European study by Adler et al 20 using the same wide-angle, HDE and comparing withdrawal with white light versus NBI detected only 0.33 0.73 adenomas per patient in the control group. Two capabilities of this new colonoscope were tested in this study: wide-angle vision and high-definition imaging. Our results are in accordance with 2 previous backto-back trials evaluating the efficiency on adenoma miss rates with a wide-angle endoscope prototype very similar to the one used in this trial. In those studies the use of a wide-angle colonoscope was associated with a reduction in the insertion time to the cecum, as well as in the examination time during withdrawal, but no difference was found in adenoma miss rates. 21,22 In our series, intubation time was slightly superior in the control group (8.9 4.8 vs 8.2 4.5 minutes), but this difference did not reach significance (P.05). Withdrawal time was similar in both groups in relation to the study design that advised for a minimal extubation time of 6 minutes. Finally, the rate of adenoma per patient was similar in both groups even in the right colon, and for smaller or flat lesions where a wider angle colonoscope could represent a potential advantage. On the other hand, a highresolution colonoscope has been compared with standard colonoscope in the multicenter French study where, in accordance with our results, the new technique did not represent any advantage in terms of adenoma detection rate or identification of patients at high risk. 31 In fact, in domestic as well as in medical devices, the latest technological advances get incorporated in practice even if their advantage has not been scientifically demonstrated. In this case, as occurred with the introduction of the videoendoscopes, it is foreseeable that high-definition scopes are likely to become standard of care and quality in 10 years. Although this study did not show any major differences between the 2 colonoscopes, it is an important reference for future endoscopic trials because it permits one to advocate that high-definition capability is probably not necessarily a plus when other capabilities added to high resolution are going to be evaluated. On the other hand, it is important to note that, even if the efficiency is not higher in terms of adenoma rate detection, other factors not directly assessed in this study could be of importance. Indeed, the new colonoscope probably facilitates the endoscopist s work; the wideangle facility has been demonstrated to shorten endoscopic time without decreasing the diagnostic efficiency, 21,22 and the high-definition screen provides a high-quality and clear image that is restful for the endoscopist s eyes. In summary, the present randomized controlled trial comparing the use of a new wide-angle, high-definition videocolonoscope with a conventional colonoscope for the purpose of identifying adenomatous polyps in general population demonstrates that the wide-angle HDE 1066

did not detect significantly more adenomas than SC. However, it is important to note that this recently commercialized colonoscope could represent a simple and available tool to facilitate. Finally, it seems that up until now performance depended on maintaining excellent standards of quality and on an exquisite endoscopist dedication. References 1. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005;55:10 30. 2. Morson BC. The polyp-cancer sequence in the large bowel. Proc R Soc Med 1974;67:451 457. 3. Winawer SJ, Zauber AG, O Brien MJ, et al. Randomized comparison of surveillance intervals after colonoscopic removal of newly diagnosed adenomatous polyps. The National Polyp Study Workgroup. N Engl J Med 1993;328:901 906. 4. Muller AD, Sonnenberg A. Prevention of colorectal cancer by flexible endoscopy and polypectomy. A case-control study of 32,702 veterans. Ann Intern Med 1995;123:904 910. 5. Citarda F, Tomaselli G, Capocaccia R, et al. Italian Multicentre Study Group. Efficacy in standard clinical practice of colonoscopic polypectomy in reducing colorectal cancer incidence. Gut 2001; 48:812 815. 6. Thiis-Evensen E, Hoff GS, Sauar J, et al. Population-based surveillance by : effect on the incidence of colorectal cancer. Telemark Polyp Study I. Scand J Gastroenterol 1999;34: 414 420. 7. Singh H, Turner D, Xue L, et al. Risk of developing colorectal cancer following a negative examination: evidence for a 10-year interval between colonoscopies. JAMA 2006;295: 2366 2373. 8. Robertson DJ, Greenberg ER, Beach M, et al. Colorectal cancer in patients under close colonoscopic surveillance. Gastroenterology 2005;129:34 41. 9. Pickhardt PJ, Nugent PA, Mysliwiec PA, et al. Location of adenomas missed by optical. Ann Intern Med 2004;141: 352 359. 10. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Workgroup. N Engl J Med 1993;329:1977 1981. 11. Hixson LJ, Fennerty MB, Sampliner RE, et al. Two-year incidence of colon adenomas developing after tandem. Am J Gastroenterol 1994;89:687 691. 12. Hixson LJ, Fennerty MB, Sampliner RE, et al. Prospective study of the frequency and size distribution of polyps missed by. J Natl Cancer Inst 1990;82:1769 1772. 13. Rex DK, Cutler CS, Lemmel GT, et al. Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies. Gastroenterology 1997;112:24 28. 14. Harewood GC, Sharma VK, de Garmo P. Impact of preparation quality on detection of suspected colonic neoplasia. Gastrointest Endosc 2003;58:76 79. 15. Froehlich F, Wietlisbach V, Gonvers JJ, et al. Impact of colonic cleansing on quality and diagnostic yield of : the European Panel of Appropriateness of Gastrointestinal Endoscopy European multicenter study. Gastrointest Endosc 2005;61: 378 384. 16. Rex DK. Maximizing detection of adenomas and cancers during. Am J Gastroenterol 2006;101:2866 2877. 17. Barclay RL, Vicari JJ, Doughty AS, et al. Colonoscopic withdrawal times and adenoma detection during screening. N Engl J Med 2006;355:2533 2541. 18. Matsushita M, Hajiro K, Okazaki K, et al. Efficacy of total with a transparent cap in comparison with without the cap. Endoscopy 1998;30:444 447. 19. Rex DK, Helbig CC. High yields of small and flat adenomas with high-definition colonoscopes using either white light or narrow band imaging. Gastroenterology 2007;133:42 47. 20. Adler A, Pohl H, Papanikolaou IS, et al. Randomized study on narrow-band imaging versus conventional for adenoma detection: does NBI induce a learning effect? Gut 2008; 57:59 64. 21. Deenadayalu VP, Chadalawada V, Rex DK. 170 degrees wideangle colonoscope: effect on efficiency and miss rates. Am J Gastroenterol 2004;99:2138 2142. 22. Rex DK, Chadalawada V, Helper DJ. Wide angle with a prototype instrument: impact on miss rates and efficiency as determined by back-to-back colonoscopies. Am J Gastroenterol 2003;98:2000 2005. 23. Stoffel EM, Stockwell DH, Normolle DP, et al. Chromoendoscopic detects more adenomas than conventional : a randomized trial of back-to-back colonoscopies. Gastroenterology 2006;130:A48. 24. Lecomte T, Cellier C, Meatchi T, et al. Chromoendoscopic for detecting preneoplastic lesions in hereditary nonpolyposis colorectal cancer syndrome. Clin Gastroenterol Hepatol 2005;3:897 902. 25. Axelrad AM, Fleischer DE, Geller AJ, et al. High-resolution chromoendoscopy for the diagnosis of diminutive colon polyps: implications for colon cancer screening. Gastroenterology 1996;110: 1253 1258. 26. Kiesslich R, von Bergh M, Hahn M, et al. Chromoendoscopy with indigocarmine improves the detection of adenomatous and nonadenomatous lesions in the colon. Endoscopy 2001;33:1001 1006. 27. Hurlstone DP, Cross SS, Slater R, et al. Detecting diminutive colorectal lesions at : a randomised controlled trial of pan-colonic versus targeted chromoscopy. Gut 2004;53:376 380. 28. Brooker JC, Saunders BP, Shah SG, et al. Total colonic dye-spray increases the detection of diminutive adenomas during routine : a randomized controlled trial. Gastrointest Endosc 2002;56:333 338. 29. Hurlstone DP, Cross SS, Adam I, et al. Efficacy of high magnification chromoscopic for the diagnosis of neoplasia in flat and depressed lesions of the colorectum: a prospective analysis. Gut 2004;53:284 290. 30. Lapalus MG, Helbert T, Napoleon B, et al. Does chromoendoscopy with structure enhancement improve the colonoscopic adenoma detection rate? Endoscopy 2006;38:444 448. 31. Le Rhun M, Coron E, Parlier D, et al. High resolution with chromoscopy versus standard for the detection of colonic neoplasia: a randomized study. Clin Gastroenterol Hepatol 2006;4:349 354. 32. The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon. Gastrointest Endosc 2003;58:S3 43. 33. Bossuyt PM, Reitsma JB, Bruns DE, et al. Standards for reporting of diagnostic accuracy. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. Clin Radiol 2003;58:575 580. 34. Sonwalkar S, Rotimi O, Rembacken BJ. Characterization of colonic polyps at conventional (nonmagnifying) after spraying with 0.2% indigo carmine dye. Endoscopy 2006; 38:1218 1223. 35. Rembacken BJ, Fujii T, Cairns A, et al. Flat and depressed colonic 1067

neoplasms: a prospective study of 1000 colonoscopies in the UK. Lancet 2000;355:1211 1214. Received February 11, 2008. Accepted June 19, 2008. Address requests for reprints to: Dr Maria Pellisé, Endoscopy Unit, Gastroenterology Department, Hospital Clinic, Villaroel 170, 08036 Barcelona, Spain. e-mail: mpellise@clinic.ub.es; fax number: 34 93 227 93 87. Supported in part by the Fondo de Investigaciones Sanitarias (FIS- ETES 07) P107/90174. Michel Zabalza received a research grant from the Fondo de Investigaciones Sanitarias. Cristina Rodríguez de Miguel is a research nurse supported by Olympus Medical Systems, Europe. Prototype endoscopic equipment was supplied by Olympus Medical Systems, Europe. CIBERehd is funded by the Instituto de Salud Carlos III. 1068