AI-assisted capsule endoscopy reading in suspected small bowel bleeding: a multicentre prospective study.

BACKGROUND: Capsule endoscopy reading is time consuming, and readers are required to maintain attention so as not to miss significant findings. Deep convolutional neural networks can recognise relevant findings, possibly exceeding human performances and reducing the reading time of capsule endoscopy. Our primary aim was to assess the non-inferiority of artificial intelligence (AI)-assisted reading versus standard reading for potentially small bowel bleeding lesions (high P2, moderate P1; Saurin classification) at per-patient analysis. The mean reading time in both reading modalities was evaluated among the secondary endpoints. METHODS: Patients aged 18 years or older with suspected small bowel bleeding (with anaemia with or without melena or haematochezia, and negative bidirectional endoscopy) were prospectively enrolled at 14 European centres. Patients underwent small bowel capsule endoscopy with the Navicam SB system (Ankon, China), which is provided with a deep neural network-based AI system (ProScan) for automatic detection of lesions. Initial reading was performed in standard reading mode. Second blinded reading was performed with AI assistance (the AI operated a first-automated reading, and only AI-selected images were assessed by human readers). The primary endpoint was to assess the non-inferiority of AI-assisted reading versus standard reading in the detection (diagnostic yield) of potentially small bowel bleeding P1 and P2 lesions in a per-patient analysis. This study is registered with ClinicalTrials.gov, NCT04821349. FINDINGS: From Feb 17, 2021 to Dec 29, 2021, 137 patients were prospectively enrolled. 133 patients were included in the final analysis (73 [55%] female, mean age 66·5 years [SD 14·4]; 112 [84%] completed capsule endoscopy). At per-patient analysis, the diagnostic yield of P1 and P2 lesions in AI-assisted reading (98 [73·7%] of 133 lesions) was non-inferior (p<0·0001) and superior (p=0·0213) to standard reading (82 [62·4%] of 133; 95% CI 3·6-19·0). Mean small bowel reading time was 33·7 min (SD 22·9) in standard reading and 3·8 min (3·3) in AI-assisted reading (p<0·0001). INTERPRETATION: AI-assisted reading might provide more accurate and faster detection of clinically relevant small bowel bleeding lesions than standard reading. FUNDING: ANKON Technologies, China and AnX Robotica, USA provided the NaviCam SB system.

Small-bowel capsule endoscopy in patients with Meckel's diverticulum: clinical features, diagnostic workup, and findings. A European multicenter I-CARE study.

BACKGROUND AND AIMS: Meckel's diverticulum (MD) may remain silent or be associated with adverse events such as GI bleeding. The main aim of this study was to evaluate indicative small-bowel capsule endoscopy (SBCE) findings, and the secondary aim was to describe clinical presentation in patients with MD. METHODS: This retrospective European multicenter study included patients with MD undergoing SBCE from 2001 until July 2021. RESULTS: Sixty-nine patients with a confirmed MD were included. Median age was 32 years with a male-to-female ratio of approximately 3:1. GI bleeding or iron-deficiency anemia was present in nearly all patients. Mean hemoglobin was 7.63 ± 1.8 g/dL with a transfusion requirement of 52.2%. Typical capsule endoscopy (CE) findings were double lumen (n = 49 [71%]), visible entrance into the MD (n = 49 [71%]), mucosal webs (n = 30 [43.5%]), and bulges (n = 19 [27.5%]). Two or more of these findings were seen in 48 patients (69.6%). Ulcers were detected in 52.2% of patients (n = 36). In 63.8% of patients (n = 44), a combination of double lumen and visible entrance into the MD was evident, additionally revealing ulcers in 39.1% (n = 27). Mean percent SB (small bowel) transit time for the first indicative image of MD was 57% of the total SB transit time. CONCLUSIONS: Diagnosis of MD is rare and sometimes challenging, and a preoperative criterion standard does not exist. In SBCE, the most frequent findings were double-lumen sign and visible diverticular entrance, sometimes together with ulcers.

The Effectiveness and Tolerability of a Very Low-Volume Bowel Preparation for Colonoscopy Compared to Low and High-Volume Polyethylene Glycol-Solutions in the Real-Life Setting.

Adequate bowel cleansing is essential for high-quality colonoscopy. Recently, a new very low-volume 1 litre (1L) polyethylene glycol (PEG) plus ascorbate solution (ASC) has been introduced. Our aims were to assess the effectiveness and tolerability of this product compared to low-volume 2L PEG-ASC and high-volume 4L PEG solutions, in a real-life setting. In six endoscopy units in Sweden, outpatients undergoing colonoscopy were either prescribed solutions according to local routines, or the very low-volume solution in split dose regimen. Bowel cleansing effectiveness and patient experience was assessed using the Boston Bowel preparation scale (BBPS) and a patient questionnaire. A total of 1098 patients (mean age 58 years, 52% women) were included. All subsegment and the total BBPS scores were significantly greater for 1L PEG-ASC in comparison to other solutions (p < 0.05 for 1L PEG-ASC and 4L PEG for transverse and left colon, otherwise p < 0.001). Nausea was more frequent with 1L PEG-ASC compared to 2L PEG-ASC (p < 0.001) and vomiting were more often reported compared to both other solutions (p < 0.01 and p < 0.05 for 2L PEG-ASC and 4L PEG, respectively). Smell, taste, and total experience was better for 1L PEG-ASC compared to 4L PEG (p < 0.001), and similar compared to the 2L PEG-ASC. In conclusion, 1L PEG-ASC leads to better bowel cleansing compared to 2L PEG-ASC or 4L PEG products, with similar or greater patient satisfaction.

Small bowel capsule endoscopy in obscure gastrointestinal bleeding: A matched cohort comparison of patients with normal vs surgically altered gastric anatomy.

BACKGROUND: Little is known about small bowel capsule endoscopy (SBCE) outcomes in patients with surgically altered anatomy. AIMS: To assess the feasibility and diagnostic yield of orally ingested SBCE to investigate obscure gastrointestinal bleeding (OGIB) in patients with surgically altered gastric anatomy, compared to native gastric anatomy. METHODS: 207 patients with OGIB were selected from an open, multicenter, retrospective cohort (SAGA study) and match-paired according to age, gender and bleeding type (overt/occult) to 207 control patients from a randomized controlled trial (PREPINTEST). Primary outcomes were the diagnostic yield (P1 or P2 findings), completion rate, adverse events rate, and small bowel transit time (SBTT). RESULTS: The diagnostic yield was not statistically different between groups (44.9% in SAGA vs 42.5% in control patients). Inflammatory/ulcerated lesions were significantly more frequent in patients with SAGA (43.0% vs 29.3%). The median SBTT was significantly longer in the SAGA group than in control patients (283 vs 206 minutes), with a significantly lower completion rate (82.6% vs 89.9%); Adverse events were scarce (0.5% vs 0.0%). CONCLUSION: Patients with surgically altered gastric anatomy should benefit from SBCE investigation for OGIB as much as non-operated patients.

The Effectiveness of a Very Low-Volume Compared to High-Volume Laxative in Colon Capsule Endoscopy.

Colon capsule endoscopy (CCE) is a promising modality for colonic investigations, but completion rates (CR) and adequate cleansing rates (ACR) must be improved to meet established standards for optical colonoscopy. Improvements should be made with patient acceptability in mind. We aimed to compare a very low-volume polyethylene glycol (PEG) laxative to a conventional high-volume laxative. We carried out a single-center retrospective comparative cohort study including patients referred for CCE. One hundred and sixty-six patients were included in the final analysis, with eighty-three patients in each group. We found a CR and ACR of 77% and 67% in the high-volume group and 72% and 75% in the very low-volume group, respectively. In the high-volume group, 54% had complete transit and adequate cleansing, whereas this was the case for 63% in the very low-volume group. No statistically significant difference in CR, ACR, or a combination of the two was found. A very low-volume bowel preparation regimen was non-inferior to a high-volume regimen before CCE in terms of CR and ACR.

Capsule endoscopy in young patients with iron deficiency anaemia and negative bidirectional gastrointestinal endoscopy.

BACKGROUND: Recent data imply young patients (age ≤50 years) undergoing small-bowel (SB) capsule endoscopy (CE) for iron deficiency anaemia (IDA) show higher diagnostic yield (DY) for sinister pathology. We aimed to investigate DY of CE in a large cohort of young IDA patients, and evaluate factors predicting significant SB pathology. MATERIALS AND METHODS: This was a retrospective, multicentre study (2010-2015) in consecutive, young patients (≤50 years) from 18 centres/12 countries, with negative bidirectional gastrointestinal (GI) endoscopy undergoing SBCE for IDA. Exclusion criteria: previous/ongoing obscure-overt GI bleeding; age <19 or >50 years; comorbidities associated with IDA. Data retrieved: SBCE indications; prior investigations; medications; SBCE findings; final diagnosis. Clinical and laboratory data were analysed by multivariate logistic regression. RESULTS: Data on 389 young IDA patients were retrieved. In total, 169 (43.4%) were excluded due to incomplete clinical data; data from 220 (122F/98M; mean age 40.5 ± 8.6 years) patients were analysed. Some 71 patients had at least one clinically significant SBCE finding (DY: 32.3%). They were divided into two groups: neoplastic pathology (10/220; 4.5%), and non-neoplastic but clinically significant pathology (61/220; 27.7%). The most common significant but non-neoplastic pathologies were angioectasias (22/61) and Crohn's disease (15/61). On multivariate analysis, weight loss and lower mean corpuscular volume(MCV) were associated with significant SB pathology (OR: 3.87; 95%CI: 1.3-11.3; p = 0.01; and OR: 0.96; 95%CI: 0.92-0.99; p = 0.03; respectively). Our model also demonstrates association between use of antiplatelets and significant SB pathology, although due to the small number of patients, definitive conclusions cannot be drawn. CONCLUSION: In IDA patients ≤50 years with negative bidirectional GI endoscopy, overall DY of SBCE for clinically significant findings was 32.3%. Some 5% of our cohort was diagnosed with SB neoplasia; lower MCV or weight loss were associated with higher DY for SB pathology.

Colonic obstruction caused by video capsule entrapment in a metal stent.

Video capsule endoscopy (VCE) has become the method of choice for visualizing the small bowel mucosa and is generally considered to be a safe method. Although uncommon, the most feared complication of VCE is capsule retention that can potentially lead to life-threatening bowel obstruction. Herein, we present for the first time a case of capsule retention in a colonic stent. The patient had known Crohn's disease with colonic involvement and underwent an uneventful but incomplete small bowel VCE for assessment of disease activity and extension for optimizing medical treatment. Five months later, the patient presented with intestinal obstruction due to a Crohn's-stricture in the sigmoid colon, which was successfully decompressed with a self-expandable metal stent. Nonetheless, two days later the patient showed signs of bowel obstruction again and abdominal X-ray showed that the capsule was trapped in the metal stent in the sigmoid colon. Subsequently, emergency surgery was performed and the patient fully recovered. Intestinal capsule retention necessitating interventional removal is rare. This report describes a unique case of capsule retention in a colonic metal stent and highlights the potential risk of performing capsule endoscopy examinations in patients with gastrointestinal stents.

Macroscopic findings in collagenous colitis: a multi-center, retrospective, observational cohort study.

BACKGROUND: Collagenous colitis (CC) is by definition a histological diagnosis. However, colonoscopy often reveals characteristic endoscopic findings. The aim of this study was to evaluate the frequency and type of endoscopic findings in patients diagnosed with CC in 4 participating centers. METHODS: This was a retrospective study; the databases of 2 university hospitals in Edinburgh (Scotland) and Malmö (Sweden), and 2 district general hospitals in Tomelloso (Spain) and Gateshead (England) were interrogated for patients diagnosed with CC between May 2008 and August 2013. Endoscopy reports and images were retrieved and reviewed; data on lesions, sedation, bowel preparation and endoscopist experience were abstracted. Categorical data are reported as mean±SD. Fischer's exact, chi-square and t (unpaired) tests were used to compare datasets. A two-tailed P-value of <0.05 was considered statistically significant. RESULTS: 607 patients (149 male, mean age 66.9±12.25 years) were diagnosed with CC. A total of 108/607 (17.8%) patients had one or more suggestive endoscopy findings: i.e., mucosal erythema/edema, 91/607 (15%); linear colonic mucosal defects, 12/607 (2%); or mucosal scarring, 5/607 (0.82%). For colonic mucosa erythema, there was no difference in the odds of finding erythema with the use of different bowel preparation methods (P=0.997). For colonic mucosal defects there was some evidence (P=0.005) that patients colonoscoped by experienced endoscopists had 87% less odds of developing such defects. Moreover, there was evidence that analgesia reduced the odds of developing mucosal defects by 84%. CONCLUSION: A significant minority of patients with CC have endoscopic findings in colonoscopy. The description of such findings appears to be related to the endoscopist's experience.

Association Between Fecal Calprotectin Levels and Small-bowel Inflammation Score in Capsule Endoscopy: A Multicenter Retrospective Study.

BACKGROUND: Accurate inflammation reporting in capsule endoscopy (CE) is important for diagnosis and monitoring of treatment of inflammatory bowel disease (IBD). Fecal calprotectin (FC) is a highly specific biomarker of gut inflammation. Lewis score (LS) was developed to standardize quantification of inflammation in small-bowel (SB) CE images. GOALS: Multicenter retrospective study aiming to investigate correlation between LS and FC in a large group of patients undergoing CE for suspected or known small-bowel IBD, and to develop a model for prediction of CE results (LS) based on FC levels. STUDY: Five academic centers and a district general hospital offering CE in UK, Finland, Sweden, Canada, and Israel. In total, 333 patients were recruited. They had small-bowel CE and FC done within 3 months. RESULTS: Overall, correlation between FC and LS was weak (r s: 0.232, P < 0.001). When two clinically significant FC thresholds (100 and 250 µg/g) were examined, the r s between FC and LS was 0.247 (weak) and 0.337 (moderate), respectively (P = 0.307). For clinically significant (LS ≥ 135) or negative (LS < 135) for SB inflammation, ROC curves gave an optimum cutoff point of FC 76 µg/g with sensitivity 0.59 and specificity 0.41. LIMITATIONS: Retrospective design. CONCLUSIONS: LS appears to show low correlation with FC as well as other serology markers of inflammation. FC does not appear to be a reliable biomarker for significant small-bowel inflammation. Nevertheless, FC level ≥ 76 µg/g may be associated with appreciable visual inflammation on small-bowel CE in patients with negative prior diagnostic workup.

Dissecting Lewis score under the light of fecal calprotectin; an analysis of correlation of score components with calprotectin levels in capsule endoscopy.

BACKGROUND: Lewis Score (LS) is an inflammatory score in small-bowel capsule endoscopy (SBCE). Fecal calprotectin (FC) is considered the non-invasive, 'gold standard' marker of gastrointestinal (GI) inflammation. Recently, we reported that LS shows only a moderate correlation with FC. In this study, we aim to evaluate which LS parameters have greater correlation with FC. METHODS: A retrospective, two-center study; 74 patients who underwent SBCE within 7 (median 1.5) days from a FC measurement. LS was calculated; univariate and multivariate analyses were performed, investigating LS correlation with FC, and which LS parameters had stronger correlation coefficient (rs) with FC. RESULTS: 74 patients had an FC measurement within 7 days of their SBCE examination (median 22 time-interval: 1.5 days; IQR: 5). Coefficient rs between LS and FC was moderate (0.454). In univariate analysis, the variables that gave the strongest association with FC were: the higher tertile subscore for ulcer, the summative ulcer subscore, the higher tertile ulcer score (only with descriptors of ulcer size and number), the summative ulcer score (only with descriptors of ulcer size and number), and subscores including various combinations of the stenosis descriptors. In multivariate analysis, the only positive predictor for FC was the higher tertile ulcer subscore (only with descriptors of ulcer size and number). CONCLUSION: LS shows only moderate correlation to FC. This is due to a) an inherent limitation of LS, and b) the notion of correlating the 2 parameters, and consideration should be given to development of a new, simplified (or composite) inflammation score/index for SBCE.

Underwater endoscopic mucosal resection of large colorectal lesions.

In this prospective study, 11 consecutive patients with neoplastic colorectal lesions (median size 20 mm, range 15 - 25 mm) underwent endoscopic polyp removal by underwater endoscopic mucosal resection (EMR). Six lesions were removed en bloc and five lesions were removed by piecemeal resection. Pathological examination revealed seven R0 resections, and in four cases the pathology could not be determined. Two cases of procedure-related bleeding occurred but these were easily managed using hemostatic forceps and clip application. No perforations or delayed bleedings were observed. Underwater EMR is a relatively simple, safe, and useful method for the removal of large colorectal lesions.

Small bowel capsule endoscopy in the management of established Crohn's disease: clinical impact, safety, and correlation with inflammatory biomarkers.

BACKGROUND: Multiple studies have established the superior diagnostic accuracy of video capsule endoscopy (VCE) for the diagnosis of small bowel (SB) Crohn's disease (CD). However, data on the clinical impact of VCE in patients with established CD are scarce. The aim of this study was to examine the impact and safety of VCE on the clinical management of patients with established CD. METHODS: A retrospective, multicenter, cross-sectional study. The study cohort included consecutive patients with established SB CD who underwent VCE in 4 tertiary referral centers (1 Canada, 1 Sweden, and 2 United Kingdom) from January 2008 to October 2013. Patients were excluded if VCE was performed as a part of the initial diagnostic workup. The presence of SB mucosal inflammation was quantified using the Lewis score. Inflammatory biomarkers (C-reactive protein and fecal calprotectin) were measured and correlated with the Lewis score. RESULTS: The study included 187 patients. No SB inflammation was observed in 28.4%, mild-to-moderate inflammation in 26.6%, and moderate-to-severe inflammation in 45% of the patients (median Lewis score, 662; range, 0-6400). A change in management was recommended in 52.3% of the patients based on VCE findings. Elevated C-reactive protein, fecal calprotectin, or the combination of both were poorly correlated with significant SB inflammation. SB capsule retention occurred in 4 patients (2.1%). CONCLUSIONS: VCE in patients with established CD is safe, and the results often have a significant clinical impact. VCE should not be limited to CD patients with positive inflammatory markers because their predictive value for significant SB inflammation is poor.

Cystic neuroendocrine tumor in the pancreas detected by endoscopic ultrasound and fine-needle aspiration: a case report.

BACKGROUND: Pancreatic neuroendocrine tumors are typically solid neoplasms but in very rare cases present as cystic lesions. The diagnosis of cystic tumors in the pancreas is extremely difficult and the use of endoscopic ultrasound and fine-needle aspiration might be helpful in the work-up of patients with cystic neuroendocrine tumors in the pancreas. CASE PRESENTATION: A 78-year-old Caucasian man was admitted with a history of epigastric pain. Laboratory tests were normal. The patient underwent transabdominal ultrasound, computed tomography and magnetic resonance cholangiopancreatography demonstrating an unclear cystic mass in the head of the pancreas. The patient was referred for endoscopic ultrasound. Endoscopic ultrasound showed a hypoechoic lesion (42 × 47 mm) in the head of the pancreas with regular borders and large cystic components. The main pancreatic duct was normal without any connection to the cystic process. The lesion underwent fine-needle aspiration (22 Gauge). Cytological examination demonstrated cohesive groups of plasmacytoid cells staining positively for synaptophysin and chromogranin A, which is suggestive of a neuroendocrine tumor. CONCLUSIONS: Differential diagnosis of cystic lesions in the pancreas is very difficult with conventional radiology, such as computed tomography and magnetic resonance imaging. This unusual case with a pancreatic cystic neuroendocrine tumor highlights the clinical importance of endoscopic ultrasound in the work-up of patients with unclear lesions in the pancreas.

Second-generation colon capsule endoscopy compared with colonoscopy.

BACKGROUND: Colon capsule endoscopy (CCE) represents a noninvasive technology that allows visualization of the colon without requiring sedation and air insufflation. A second-generation colon capsule endoscopy system (PillCam Colon 2) (CCE-2) was developed to increase sensitivity for colorectal polyp detection compared with the first-generation system. OBJECTIVE: To assess the feasibility, accuracy, and safety of CCE-2 in a head-to-head comparison with colonoscopy. DESIGN AND SETTING: Prospective, multicenter trial including 8 European sites. PATIENTS: This study involved 117 patients (mean age 60 years). Data from 109 patients were analyzed. INTERVENTION: CCE-2 was prospectively compared with conventional colonoscopy as the criterion standard for the detection of colorectal polyps that are ≥6 mm or masses in a cohort of patients at average or increased risk of colorectal neoplasia. Colonoscopy was independently performed within 10 hours after capsule ingestion or on the next day. MAIN OUTCOME MEASUREMENTS: CCE-2 sensitivity and specificity for detecting patients with polyps ≥6 mm and ≥10 mm were assessed. Capsule-positive but colonoscopy-negative cases were counted as false positive. Capsule excretion rate, level of bowel preparation, and rate of adverse events also were assessed. RESULTS: Per-patient CCE-2 sensitivity for polyps ≥6 mm and ≥10 mm was 84% and 88%, with specificities of 64% and 95%, respectively. All 3 invasive carcinomas were detected by CCE-2. The capsule excretion rate was 88% within 10 hours. Overall colon cleanliness for CCE-2 was adequate in 81% of patients. LIMITATIONS: Not unblinding the CCE-2 results at colonoscopy; heterogenous patient population; nonconsecutive patients. CONCLUSION: In this European, multicenter study, CCE-2 appeared to have a high sensitivity for the detection of clinically relevant polypoid lesions, and it might be considered an adequate tool for colorectal imaging.