Computer-aided diagnosis for optical diagnosis of diminutive colorectal polyps including sessile serrated lesions: a real-time comparison with screening endoscopists.

BACKGROUND : We aimed to compare the accuracy of the optical diagnosis of diminutive colorectal polyps, including sessile serrated lesions (SSLs), between a computer-aided diagnosis (CADx) system and endoscopists during real-time colonoscopy. METHODS : We developed the POLyp Artificial Recognition (POLAR) system, which was capable of performing real-time characterization of diminutive colorectal polyps. For pretraining, the Microsoft-COCO dataset with over 300 000 nonpolyp object images was used. For training, eight hospitals prospectively collected 2637 annotated images from 1339 polyps (i. e. publicly available online POLAR database). For clinical validation, POLAR was tested during colonoscopy in patients with a positive fecal immunochemical test (FIT), and compared with the performance of 20 endoscopists from eight hospitals. Endoscopists were blinded to the POLAR output. Primary outcome was the comparison of accuracy of the optical diagnosis of diminutive colorectal polyps between POLAR and endoscopists (neoplastic [adenomas and SSLs] versus non-neoplastic [hyperplastic polyps]). Histopathology served as the reference standard. RESULTS : During clinical validation, 423 diminutive polyps detected in 194 FIT-positive individuals were included for analysis (300 adenomas, 41 SSLs, 82 hyperplastic polyps). POLAR distinguished neoplastic from non-neoplastic lesions with 79 % accuracy, 89 % sensitivity, and 38 % specificity. The endoscopists achieved 83 % accuracy, 92 % sensitivity, and 44 % specificity. The optical diagnosis accuracy between POLAR and endoscopists was not significantly different (P = 0.10). The proportion of polyps in which POLAR was able to provide an optical diagnosis was 98 % (i. e. success rate). CONCLUSIONS : We developed a CADx system that differentiated neoplastic from non-neoplastic diminutive polyps during endoscopy, with an accuracy comparable to that of screening endoscopists and near-perfect success rate.

Comprehensive review of publicly available colonoscopic imaging databases for artificial intelligence research: availability, accessibility, and usability.

BACKGROUND AND AIMS: Publicly available databases containing colonoscopic imaging data are valuable resources for artificial intelligence (AI) research. Currently, little is known regarding the available number and content of these databases. This review aimed to describe the availability, accessibility, and usability of publicly available colonoscopic imaging databases, focusing on polyp detection, polyp characterization, and quality of colonoscopy. METHODS: A systematic literature search was performed in MEDLINE and Embase to identify AI studies describing publicly available colonoscopic imaging databases published after 2010. Second, a targeted search using Google's Dataset Search, Google Search, GitHub, and Figshare was done to identify databases directly. Databases were included if they contained data about polyp detection, polyp characterization, or quality of colonoscopy. To assess accessibility of databases, the following categories were defined: open access, open access with barriers, and regulated access. To assess the potential usability of the included databases, essential details of each database were extracted using a checklist derived from the Checklist for Artificial Intelligence in Medical Imaging. RESULTS: We identified 22 databases with open access, 3 databases with open access with barriers, and 15 databases with regulated access. The 22 open access databases contained 19,463 images and 952 videos. Nineteen of these databases focused on polyp detection, localization, and/or segmentation; 6 on polyp characterization, and 3 on quality of colonoscopy. Only half of these databases have been used by other researcher to develop, train, or benchmark their AI system. Although technical details were in general well reported, important details such as polyp and patient demographics and the annotation process were under-reported in almost all databases. CONCLUSIONS: This review provides greater insight on public availability of colonoscopic imaging databases for AI research. Incomplete reporting of important details limits the ability of researchers to assess the usability of current databases.

Real-time diagnostic accuracy of blue light imaging, linked color imaging and white-light endoscopy for colorectal polyp characterization.

Background and study aims Fujifilm has developed a novel ELUXEO 7000 endoscope system that employs light-emitting diodes (LEDs) at four different wavelengths as light sources that enable blue light imaging (BLI), linked color imaging (LCI), and high-definition white-light endoscopy (HD-WLE). The aim of this study was to address the diagnostic accuracy of real-time polyp characterization using BLI, LCI and HD-WLE (ELUXEO 7000 endoscopy system). Patients methods This is a prespecified post-hoc analysis of a prospective study in which 22 experienced endoscopists (> 2,000 colonoscopies) from eight international centers participated. Using a combination of BLI, LCI, and HD-WLE, lesions were endoscopically characterized including a high- or low-confidence statement. Per protocol, digital images were created from all three imaging modalities. Histopathology was the reference standard. Endoscopists were familiar with polyp characterization, but did not take dedicated training for purposes of this study. Results Overall, 341 lesions were detected in 332 patients. Of the lesions, 269 histologically confirmed polyps with an optical diagnosis were included for analysis (165 adenomas, 27 sessile serrated lesions, and 77 hyperplastic polyps). Overall, polyp characterization was performed with high confidence in 82.9 %. The overall accuracy for polyp characterization was 75.1 % (95 % confidence interval [CI] 69.5-80.1 %), compared with an accuracy of 78.0 % (95 % CI 72.0-83.2 %) for high confidence assignments. The accuracy for endoscopic characterization for diminutive polyps was 74.7 % (95 %CI 68.4-80.3 %), compared with an accuracy of 78.2 % (95 % CI 71.4-84.0 %) for high-confidence assignments. Conclusions The diagnostic accuracy of BLI, LCI, and HD-WLE by experienced endoscopist for real-time polyp characterization seems limited (NCT03344289).

Linked Colour imaging for the detection of polyps in patients with Lynch syndrome: a multicentre, parallel randomised controlled trial.

OBJECTIVE: Despite regular colonoscopy surveillance, colorectal cancers still occur in patients with Lynch syndrome. Thus, detection of all relevant precancerous lesions remains very important. The present study investigates Linked Colour imaging (LCI), an image-enhancing technique, as compared with high-definition white light endoscopy (HD-WLE) for the detection of polyps in this patient group. DESIGN: This prospective, randomised controlled trial was performed by 22 experienced endoscopists from eight centres in six countries. Consecutive Lynch syndrome patients ≥18 years undergoing surveillance colonoscopy were randomised (1:1) and stratified by centre for inspection with either LCI or HD-WLE. Primary outcome was the polyp detection rate (PDR). RESULTS: Between January 2018 and March 2020, 357 patients were randomised and 332 patients analysed (160 LCI, 172 HD-WLE; 6 excluded due to incomplete colonoscopies and 19 due to insufficient bowel cleanliness). No significant difference was observed in PDR with LCI (44.4%; 95% CI 36.5% to 52.4%) compared with HD-WLE (36.0%; 95% CI 28.9% to 43.7%) (p=0.12). Of the secondary outcome parameters, more adenomas were found on a patient (adenoma detection rate 36.3%; vs 25.6%; p=0.04) and a colonoscopy basis (mean adenomas per colonoscopy 0.65 vs 0.42; p=0.04). The median withdrawal time was not statistically different between LCI and HD-WLE (12 vs 11 min; p=0.16). CONCLUSION: LCI did not improve the PDR compared with HD-WLE in patients with Lynch syndrome undergoing surveillance. The relevance of findings more adenomas by LCI has to be examined further. TRIAL REGISTRATION NUMBER: NCT03344289.

Definition of competence standards for optical diagnosis of diminutive colorectal polyps: European Society of Gastrointestinal Endoscopy (ESGE) Position Statement.

BACKGROUND : The European Society of Gastrointestinal Endoscopy (ESGE) has developed a core curriculum for high quality optical diagnosis training for practice across Europe. The development of easy-to-measure competence standards for optical diagnosis can optimize clinical decision-making in endoscopy. This manuscript represents an official Position Statement of the ESGE aiming to define simple, safe, and easy-to-measure competence standards for endoscopists and artificial intelligence systems performing optical diagnosis of diminutive colorectal polyps (1 - 5 mm). METHODS : A panel of European experts in optical diagnosis participated in a modified Delphi process to reach consensus on Simple Optical Diagnosis Accuracy (SODA) competence standards for implementation of the optical diagnosis strategy for diminutive colorectal polyps. In order to assess the clinical benefits and harms of implementing optical diagnosis with different competence standards, a systematic literature search was performed. This was complemented with the results from a recently performed simulation study that provides guidance for setting alternative competence standards for optical diagnosis. Proposed competence standards were based on literature search and simulation study results. Competence standards were accepted if at least 80 % agreement was reached after a maximum of three voting rounds. RECOMMENDATION 1: In order to implement the leave-in-situ strategy for diminutive colorectal lesions (1-5 mm), it is clinically acceptable if, during real-time colonoscopy, at least 90 % sensitivity and 80 % specificity is achieved for high confidence endoscopic characterization of colorectal neoplasia of 1-5 mm in the rectosigmoid. Histopathology is used as the gold standard.Level of agreement 95 %. RECOMMENDATION 2: In order to implement the resect-and-discard strategy for diminutive colorectal lesions (1-5 mm), it is clinically acceptable if, during real-time colonoscopy, at least 80 % sensitivity and 80 % specificity is achieved for high confidence endoscopic characterization of colorectal neoplasia of 1-5 mm. Histopathology is used as the gold standard.Level of agreement 100 %. CONCLUSION : The developed SODA competence standards define diagnostic performance thresholds in relation to clinical consequences, for training and for use when auditing the optical diagnosis of diminutive colorectal polyps.

[Weight loss as a presenting clinical feature of malignancy]. / Gewichtsverlies als uiting van een maligniteit.

Unexpected weight loss presents a diagnostic challenge as it is associated with a wide range of benign and serious conditions. Although it is readily associated with malignancy, the risk of cancer in adults with unexpected weight loss presenting to primary care is 2% or less. In male patients aged 60 years or older and in patients with concurrent clinical symptoms, signs and abnormal blood test, the risk of cancer increases. Initial testing should include a history including medication review, physical examination and blood tests. Recommended blood tests include a complete blood count, basic metabolic panel, liver function tests, thyroid function tests, C-reactive protein level, erythrocyte sedimentation rate and lactate dehydrogenase measurement. Imaging and invasive testing may be considered based on initial evaluation. When the initial evaluation is unremarkable, an observation period is recommended for young patients in particular.

Dye-Based Chromoendoscopy in Patients With Lynch Syndrome: An Individual Patient Data Meta-Analysis of Randomized Trials.

INTRODUCTION: The additional diagnostic value of dye-based chromoendosocpy (CE) for surveillance of patients with Lynch syndrome is subject of debate. METHODS: To clarify this debate, we performed an individual patient data meta-analysis of randomized studies that compared CE with WLE for the detection of adenomas in patients with Lynch syndrome. RESULTS: Three randomized studies comprising 533 patients were included. The adenoma detection rate was 74/265 (28%) in patients randomized to WLE compared with 83/266 (31%) in patients randomized to CE (odds ratio 1.17; 95% confidence interval 0.81-1.70). DISCUSSION: Based on low-quality evidence, CE showed no apparent increase in adenoma detection compared to WLE during surveillance of patients with Lynch syndrome.

Guidance for setting easy-to-adopt competence criteria for optical diagnosis of diminutive colorectal polyps: a simulation approach.

BACKGROUND AND AIMS: One reason the optical diagnosis strategy for diminutive colorectal polyps has not yet been implemented is that the current competence criteria (Preservation and Incorporation of Valuable Endoscopic Innovation [PIVI] initiative) are difficult to use in daily practice. To provide guidance for setting alternative easy-to-adopt competence criteria, we determined the lowest proportion of diminutive polyps that should have a correct optical diagnosis to meet the PIVI. METHODS: For this simulation study, we used datasets from 2 prospectively collected cohorts of patients who underwent colonoscopy in either a primary colonoscopy or fecal immunochemical test (FIT) screening setting. In the simulation approach, virtual endoscopists or computer-aided diagnosis systems performed optical diagnosis of diminutive polyps with a fixed diagnostic performance level (strategy) on all individuals in the cohort who had ≥1 diminutive polyp. Strategies were defined by systematically varying the proportion of correct optical diagnoses for each polyp subtype (ie, adenomas, hyperplastic polyps, sessile serrated lesions). For each strategy, we determined whether PIVI-1 (≥90% agreement with U.S. or European Society for Gastrointestinal Endoscopy [ESGE] surveillance guidelines) and PIVI-2 (≥90% negative predictive value [NPV] for neoplastic lesions in the rectosigmoid) were met using Monte Carlo sampling with 1000 repetitions, with histology as reference. RESULTS: The level of overall diagnostic accuracy to achieve the PIVI differed significantly depending on the clinical setting and guidelines used. In the colonoscopy screening setting, all diagnostic strategies in which 92% of all diminutive polyps (regardless of histology) were diagnosed correctly led to 90% or more agreement with U.S. surveillance intervals (ie, PIVI-1). For all diagnostic strategies in which ≥89% of all diminutive polyps were correctly diagnosed, at least 90% NPV was achieved (ie, PIVI-2). For the FIT screening setting, values were respectively ≥77% and ≥94%. When using ESGE guidelines, PIVI-1 was in both settings already met when 40% of all diminutive polyps were diagnosed correctly. CONCLUSIONS: In contrast to the fixed PIVI criteria, our simulation study shows that different thresholds for the proportion of correctly diagnosed diminutive polyps lead to different clinical consequences depending on guidelines and clinical setting. However, this target proportion of diminutive colorectal polyps correctly diagnosed with optical diagnosis represents easier-to-adopt competence criteria.

Are adenoma and serrated polyp detection rates correlated with endoscopists' sensitivity of optical diagnosis?

INTRODUCTION : Endoscopists with a high adenoma detection rate (ADR) and proximal serrated polyp detection rate (PSPDR) detect these polyps more frequently, which may be attributable to better recognition of their endoscopic features. Little is known about the association between endoscopic lesion detection and differentiation skills. Therefore, we evaluated the correlation between the ADR, PSPDR, and the sensitivity of optical diagnosis for adenomas and serrated polyps. METHODS: We performed an exploratory post-hoc analysis of the DISCOUNT-2 study, including complete colonoscopies after a positive fecal immunochemical test (FIT) performed by endoscopists who performed ≥ 50 colonoscopies. The correlations between the ADR, PSPDR, and the sensitivity of optical diagnosis were calculated using Pearson's rho correlation coefficient. RESULTS: 24 endoscopists performed ≥ 50 colonoscopies, resulting in a total of 2889 colonoscopies. The overall ADR was 84.5 % (range 71.4 % - 95.3 %) and overall PSPDR was 13.7 % (4.3 % - 29.0 %). The sensitivity of optical diagnosis for adenomas and serrated polyps were 94.5 % (83.3 % - 100 %) and 74.0 % (37.5 % - 94.1 %), respectively. No correlation could be demonstrated between the ADR and the sensitivity of optical diagnosis for adenomas (-0.20; P = 0.35) or between the PSPDR and the sensitivity of optical diagnosis for serrated polyps (-0.12; P = 0.57). CONCLUSIONS: In a homogeneous FIT-positive population, no correlation between the ADR, PSPDR, and the sensitivity of optical diagnosis for adenomas and serrated polyps could be demonstrated. These exploratory results suggest that lesion detection and differentiation require different endoscopic skills. Further prospective studies are needed; until then, monitoring of both performance indicators is important to secure optimal efficacy of FIT-based colorectal cancer screening.

Colorectal cancer.

Several decades ago, colorectal cancer was infrequently diagnosed. Nowadays, it is the world's fourth most deadly cancer with almost 900 000 deaths annually. Besides an ageing population and dietary habits of high-income countries, unfavourable risk factors such as obesity, lack of physical exercise, and smoking increase the risk of colorectal cancer. Advancements in pathophysiological understanding have increased the array of treatment options for local and advanced disease leading to individual treatment plans. Treatments include endoscopic and surgical local excision, downstaging preoperative radiotherapy and systemic therapy, extensive surgery for locoregional and metastatic disease, local ablative therapies for metastases, and palliative chemotherapy, targeted therapy, and immunotherapy. Although these new treatment options have doubled overall survival for advanced disease to 3 years, survival is still best for those with non-metastasised disease. As the disease only becomes symptomatic at an advanced stage, worldwide organised screening programmes are being implemented, which aim to increase early detection and reduce morbidity and mortality from colorectal cancer.

CD31-positive microvessel density within adenomas of Lynch Syndrome patients is similar compared to adenomas of non-Lynch patients.

Background and study aims Microsatellite instability accelerates colorectal cancer development in patients with Lynch syndrome (LS). Previous research showed that virtual chromoendoscopy increases detection of adenomas during colonoscopy surveillance of patients with LS. Because previous research revealed that Lynch patients have an increased vascular network in the oral mucosa, we hypothesized that increased vascularization of LS-associated adenomas is the cause of better detection with virtual chromoendoscopy. Patients and methods In this pilot study, patients with LS having a proven germline mutation were selected from two tertiary referral hospitals and non-LS patients from an outpatient colonoscopy center. Adenomas from patients with LS were exactly matched in size and histology with adenomas from non-LS patients. Initial adenoma diagnosis was confirmed by a specialist pathologist. All adenomas were stained with CD31 and adenomatous tissue was annotated by the specialist pathologist. Image analysis of CD31-positive microvessel density was conducted using FIJI software. Results Colonoscopy of 63 patients with LS and 24 non-LS patients provided 40 adenomas that could be exactly matched in size and histology. In image-analysis, the CD31-positive microvessel density (2.49 % vs. 2.47 %, P  = 0.96), the average size of CD31-positive structures (514 µm 2 vs. 523 µm 2 , P  = 0.26) nor the amount of vascular structures per mm 2 (183 vs. 176, P  = 0.50) differed between adenomas of LS patients and non-Lynch patients. Conclusion The outcomes of this pilot case-control study did not provide further insights into the mechanism of increased adenoma detection in LS patients using virtual chromoendoscopy techniques.

Diminutive Polyps With Advanced Histologic Features Do Not Increase Risk for Metachronous Advanced Colon Neoplasia.

BACKGROUND & AIMS: With advances in endoscopic imaging, it is possible to differentiate adenomatous from hyperplastic diminutive (1-5 mm) polyps during endoscopy. With the optical Resect-and-Discard strategy, these polyps are then removed and discarded without histopathology assessment. However, failure to recognize adenomas (vs hyperplastic polyps), or discarding a polyp with advanced histologic features, could result in a patient being considered at low risk for metachronous advanced neoplasia, resulting in an inappropriately long surveillance interval. We collected data from international cohorts of patients undergoing colonoscopy to determine what proportion of patients are high risk because of diminutive polyps advanced histologic features and their risk for metachronous advanced neoplasia. METHODS: We collected data from 12 cohorts (in the United States or Europe) of patients undergoing colonoscopy after a positive result from a fecal immunochemical test (FIT cohort, n = 34,221) or undergoing colonoscopies for screening, surveillance, or evaluation of symptoms (colonoscopy cohort, n = 30,123). Patients at high risk for metachronous advanced neoplasia were defined as patients with polyps that had advanced histologic features (cancer, high-grade dysplasia, ≥25% villous features), 3 or more diminutive or small (6-9 mm) nonadvanced adenomas, or an adenoma or sessile serrated lesion ≥10 mm. Using an inverse variance random effects model, we calculated the proportion of diminutive polyps with advanced histologic features; the proportion of patients classified as high risk because their diminutive polyps had advanced histologic features; and the risk of these patients for metachronous advanced neoplasia. RESULTS: In 51,510 diminutive polyps, advanced histologic features were observed in 7.1% of polyps from the FIT cohort and 1.5% polyps from the colonoscopy cohort (P = .044); however, this difference in prevalence did not produce a significant difference in the proportions of patients assigned to high-risk status (0.8% of patients in the FIT cohort and 0.4% of patients in the colonoscopy cohort) (P = .25). The proportions of high-risk patients because of diminutive polyps with advanced histologic features who were found to have metachronous advanced neoplasia (17.6%) did not differ significantly from the proportion of low-risk patients with metachronous advanced neoplasia (14.6%) (relative risk for high-risk categorization, 1.13; 95% confidence interval 0.79-1.61). CONCLUSION: In a pooled analysis of data from 12 international cohorts of patients undergoing colonoscopy for screening, surveillance, or evaluation of symptoms, we found that diminutive polyps with advanced histologic features do not increase risk for metachronous advanced neoplasia.

Current Practices in Ileal Pouch Surveillance for Patients With Ulcerative Colitis: A Multinational, Retrospective Cohort Study.

BACKGROUND AND AIMS: There are no universally accepted guidelines regarding surveillance of ulcerative colitis [UC] patients after restorative proctocolectomy and ileal pouch-anal anastomosis [IPAA]. There also exists a lack of validated quality assurance standards for performing pouchoscopy. To better understand IPAA surveillance practices in the face of this clinical equipoise, we carried out a retrospective cohort study at five inflammatory bowel disease [IBD] referral centres. METHODS: Records of patients who underwent IPAA for UC or IBD unclassified [IBDU] were reviewed, and patients with <1-year follow-up after restoration of intestinal continuity were excluded. Criteria for determining the risk of pouch dysplasia formation were collected as well as the use of pouchoscopy, biopsies, and completeness of reports. RESULTS: We included 272 patients. Median duration of pouch follow-up was 10.5 [3.3-23.6] years; 95/272 [35%] had never undergone pouchoscopy for any indication; 191/272 [70%] had never undergone pouchoscopy with surveillance as the specific indication; and 3/26 [12%] high-risk patients had never undergone pouchoscopy. Two cases of adenocarcinoma were identified, occurring in the rectal cuff of low-risk patients. Patients under the care of surgeons appeared more likely to undergo surveillance, but rates of incomplete reporting were higher among surgeons [78%] than gastroenterologists [54%, p = 0.002]. CONCLUSIONS: We observed wide variation in surveillance of UC/IBDU-IPAA patients. In addition, the rate of neoplasia formation among 'low-risk' patients was higher than may have been expected. We therefore concur with previous recommendations that pouchoscopy be performed at 1 year postoperatively, to refine risk-stratification based on clinical factors alone. Reports should document findings in all regions of the pouch and biopsies should be taken.

Optical diagnosis expanded to small polyps: post-hoc analysis of diagnostic performance in a prospective multicenter study.

BACKGROUND: Optical diagnosis can replace histopathology of diminutive (1 - 5 mm) polyps if surveillance intervals based on optical diagnosis of polyps have ≥ 90 % agreement with intervals based on polyp histology and if the negative predictive value (NPV) for predicting neoplastic histology in the rectosigmoid is ≥ 90 %. This study aims to assess whether small (6 - 9 mm) polyps can be included in optical diagnosis strategies. METHOD: This is a post-hoc analysis of a prospective multicenter study in which 27 endoscopists, all performing endoscopies for the Dutch screening program, were trained in optical diagnosis. For 1 year, endoscopists recorded the predicted histology for all lesions detected using narrow-band imaging during 3144 consecutive colonoscopies after a positive fecal immunochemical test, along with confidence levels. Surveillance interval agreement and NPV were calculated for high confidence predictions for polyps of 1 - 9 mm and compared with histopathology. Surveillance interval agreement was calculated using the European Society of Gastrointestinal Endoscopy surveillance guideline. RESULTS: Surveillance interval agreement was 95.4 % (confidence interval [CI] 94.2 % - 96.4 %), and NPV for predicting neoplastic histology in the rectosigmoid 90.0 % (CI 87.3 % - 92.2 %). The reduction in histology (45.9 % vs. 30.5 %) and the proportion of patients who could have received direct surveillance advice (15.6 % vs. 7.3 %) was higher when small polyps were included (P < 0.001). T1 cancer was found in seven small polyps (0.33 %), five of which would have been discarded without histopathology. CONCLUSION: Including small polyps in the optical diagnosis strategy improves its efficacy while maintaining performance thresholds. However, there is a small risk of missing T1 cancers when small polyps are included in the optical diagnosis strategy.

Diagnostic Accuracy of Endoscopic Trimodal Imaging and Chromoendoscopy for Lesion Characterization in Ulcerative Colitis.

BACKGROUND: During surveillance colonoscopy of patients with long-standing ulcerative colitis [UC], a variety of dysplastic and non-dysplastic lesions are detected. The aim of this study was to address the diagnostic accuracy of endoscopic characterization of endoscopic trimodal imaging [ETMI] and chromoendoscopy [CE]. ETMI includes the combination of autofluorescence imaging [AFI], narrow band imaging [NBI] and white light endoscopy [WLE]. METHODS: This is a pre-specified additional analysis of a multi-centre, randomized controlled trial that compared AFI with CE for dysplasia detection in 210 patients with long-standing UC [FIND-UC trial]. In the AFI arm, endoscopists used the ETMI system to record AFI colour, Kudo pit pattern using NBI and WLE for lesion characterization. For AFI, purple colour and ambiguous colour combined with pit pattern type III-V on NBI was considered dysplastic. Kudo pit pattern was described in the CE arm. For pit pattern description using NBI and CE, type III-V was considered dysplastic. Histology was the reference standard. RESULTS: In total, 52 dysplastic and 255 non-dysplastic lesions were detected. Overall sensitivity for real-time prediction of dysplasia was 76.9% (95% confidence interval [CI] 46.2-95.0) for ETMI, and 81.6% [95% CI 65.7-92.3] for CE. Overall negative predictive value [NPV] for ETMI was 96.9% [95% CI 92.0-98.8] and 94.7% [90.2-97.2] for CE. CONCLUSIONS: Sensitivity for endoscopic differentiation of dysplastic lesions detected during surveillance of patients with long-standing UC seems limited using ETMI and CE. Future research is warranted as the high NPV indicates that these techniques are valuable for the exclusion of dysplastic lesions [NTR4062].

Chromoendoscopy versus autofluorescence imaging for neoplasia detection in patients with longstanding ulcerative colitis (FIND-UC): an international, multicentre, randomised controlled trial.

BACKGROUND: Patients with longstanding ulcerative colitis undergo regular dysplasia surveillance because they have an increased colorectal cancer risk. Autofluorescence imaging and chromoendoscopy improve dysplasia detection. The aim of this study was to determine whether autofluorescence imaging should be further studied as an alternative method for dysplasia surveillance in patients with longstanding ulcerative colitis. METHODS: This prospective, international, randomised controlled trial included patients from an ulcerative colitis-dysplasia surveillance cohort from five centres in the Netherlands and the UK. Eligible patients were aged 18 years or older who were undergoing dysplasia surveillance after being diagnosed with extensive colitis (Montreal E3) at least 8 years before study start or with left-sided colitis (Montreal E2) at least 15 years before study start. Randomisation (1:1) was minimised for a previous personal history of histologically proven dysplasia and concomitant primary sclerosing cholangitis. The coprimary outcomes were the proportion of patients in whom at least one dysplastic lesion was detected and the mean number of dysplastic lesions per patient. The relative dysplasia detection rate, calculated as the ratio of the detection rates by autofluorescence imaging and chromoendoscopy, needed to be more than 0·67 (using an 80% CI) for both primary outcomes to support a subsequent large non-inferiority trial. Outcomes were analysed on a per-protocol basis. The trial is registered at the Netherlands Trial Register, number NTR4062. FINDINGS: Between Aug 1, 2013, and March 10, 2017, 210 patients undergoing colonoscopy surveillance for longstanding ulcerative colitis were randomised for inspection with either autofluorescence imaging (n=105) or chromoendoscopy (n=105). Dysplasia was detected in 13 (12%) patients by autofluorescence imaging and in 20 patients (19%) by chromoendoscopy. The relative dysplasia detection rate of autofluorescence imaging versus chromoendoscopy for the proportion of patients with ulcerative colitis with at least one dysplastic lesion was 0·65 (80% CI 0·43-0·99). The mean number of detected dysplastic lesions per patient was 0·13 (SD 0·37) for autofluorescence imaging and 0·37 (1·02) for chromoendoscopy (relative dysplasia detection rate 0·36, 80% CI 0·21-0·61). Adverse events were reported for two patients in the autofluorescence imaging group (one patient had intraprocedural mild bleeding, and one patient had abdominal pain) and for three patients in the chromoendoscopy group (two patients had intraprocedural mild bleeding, and one patient had perforation). INTERPRETATION: Autofluorescence imaging did not meet criteria for proceeding to a large non-inferiority trial. Therefore, existing autofluorescence imaging technology should not be further investigated as an alternative dysplasia surveillance method. FUNDING: Olympus Europe and Olympus Keymed.

Effects of Training and Feedback on Accuracy of Predicting Rectosigmoid Neoplastic Lesions and Selection of Surveillance Intervals by Endoscopists Performing Optical Diagnosis of Diminutive Polyps.

BACKGROUND & AIMS: Real-time differentiation of diminutive polyps (1-5 mm) during endoscopy could replace histopathology analysis. According to guidelines, implementation of optical diagnosis into routine practice would require it to identify rectosigmoid neoplastic lesions with a negative predictive value (NPV) of more than 90%, using histologic findings as a reference, and agreement with histology-based surveillance intervals for more than 90% of cases. METHODS: We performed a prospective study with 39 endoscopists accredited to perform colonoscopies on participants with positive results from fecal immunochemical tests in the Bowel Cancer Screening Program at 13 centers in the Netherlands. Endoscopists were trained in optical diagnosis using a validated module (Workgroup serrAted polypS and Polyposis). After meeting predefined performance thresholds in the training program, the endoscopists started a 1-year program (continuation phase) in which they performed narrow band imaging analyses during colonoscopies of participants in the screening program and predicted histological findings with confidence levels. The endoscopists were randomly assigned to groups that received feedback or no feedback on the accuracy of their predictions. Primary outcome measures were endoscopists' abilities to identify rectosigmoid neoplastic lesions (using histology as a reference) with NPVs of 90% or more, and selecting surveillance intervals that agreed with those determined by histology for at least 90% of cases. RESULTS: Of 39 endoscopists initially trained, 27 (69%) completed the training program. During the continuation phase, these 27 endoscopists performed 3144 colonoscopies in which 4504 diminutive polyps were removed. The endoscopists identified neoplastic lesions with a pooled NPV of 90.8% (95% confidence interval 88.6-92.6); their proposed surveillance intervals agreed with those determined by histologic analysis for 95.4% of cases (95% confidence interval 94.0-96.6). Findings did not differ between the group that did vs did not receive feedback. Sixteen endoscopists (59%) identified rectosigmoid neoplastic lesions with NPVs greater than 90% and selected surveillance intervals in agreement with those determined from histology for more than 90% of patients. CONCLUSIONS: In a prospective study following a validated training module, we found that a selected group of endoscopists identified rectosigmoid neoplastic lesions with pooled NPVs greater than 90% and accurately selected surveillance intervals for more than 90% of patients over the course of 1 year. Providing regular interim feedback on the accuracy of neoplastic lesion prediction and surveillance interval selection did not lead to differences in those endpoints. Monitoring is suggested, as individual performance varied. ClinicalTrials.gov no: NCT02516748; Netherland Trial Register: NTR4635.