Etiology and long-term rebleeding of endoscopic ulcerative lesions in the small bowel in patients with obscure gastrointestinal bleeding: A multicenter cohort study

Video capsule endoscopy in overt and occult obscure gastrointestinal bleeding: Insights from a single-center, observational study in Japan

Objective

This study aimed to evaluate the use of video capsule endoscopy (VCE) in patients with obscure gastrointestinal bleeding (OGIB), compare cases of overt and occult OGIB, assess the rates of balloon-assisted enteroscopy (BAE) interventions and rebleeding, and identify predictive markers of positive VCE findings.

Methods

Medical records of 430 patients who underwent VCE for OGIB between 2004 and 2022 were analyzed. Occult OGIB was defined as IDA or positive fecal occult blood, whereas overt OGIB was defined as clinically imperceptible bleeding. We retrospectively analyzed demographics, VCE findings based on Saurin classification (P0, P1, and P2), outcome of BAE interventions, and rebleeding rates.

Results

A total of 253 patients with overt OGIB and 177 with occult OGIB were included. P1 findings were predominant in both groups, with a similar distribution. The percentage of patients receiving conservative therapy was higher in P1 than in P2 for both overt and occult OGIB. BAE was more frequently performed in P2 than in P1 VCE (83.0% vs. 35.3% in overt OGIB, 84.4% vs. 24.4% in occult OGIB). The percentage of positive findings and intervention in total BAE performed patients were comparable in P1 and P2 of overt OGIB, whereas these percentages in P2 were more than P1 of occult OGIB.

Conclusion

VCE effectively identified OGIB lesions requiring intervention, particularly occult OGIB lesions, potentially reducing unnecessary BAE. Rebleeding rates varied according to the VCE findings, emphasizing the importance of follow-up in high-risk patients.

Automated detection of ulcers and erosions in capsule endoscopy images using a convolutional neural network

Capsule endoscopy (CE) is an important tool in the management of patients with known or suspected inflammatory bowel disease. Ulcers and erosions of the enteric mucosa are prevalent findings in these patients. They frequently occur together, and their identification in CE is crucial for an accurate evaluation of disease severity. Nevertheless, reviewing CE images is a time-consuming task, and the risk of overlooking lesions is significant.Over the last decade, artificial intelligence (AI) has emerged as a means for overcoming these pitfalls. Of all AI methods, convolutional neural networks (CNN), due to their complex multilayer architecture present the best results in medical image analysis, particularly capsule endoscopy. Therefore, we aimed to develop a CNN for the automatic identification of ulcers and erosions in the small bowel mucosa. A total of 1483 CE exams (PillCam SB3®) performed at a single center between 2015 and 2020 were analysed. From these exams, a total of 6130 frames of the enteric mucosa were obtained, 4233 containing enteric ulcers and erosions, and the remaining containing normal mucosa or other findings. Ulcers and erosions were stratified according to Saurin's classification for bleeding potential: P1E-erosions with intermediate bleeding risk; P1U-ulcers with intermediate bleeding risk; P2U-ulcers with high bleeding risk. For automatic identification of these lesions, these images were inserted into a CNN model with transfer learning. The pool of images was divided for constitution of training and validation datasets, comprising 80% and 20% of the total number of images, respectively. The output provided by the CNN was compared to the classification provided by a consensus of specialists. After optimizing the neural architecture of the algorithm, our model was able to automatically detect and distinguish ulcers and erosions (any bleeding potential) in the small intestine mucosa with an accuracy of 95.6%, sensitivity of 90.8%, and a specificity of 97.1%. We believe that our study lays the foundation for the development and application of effective AI tools to CE. These techniques should improve diagnostic accuracy and reading efficiency. Schematic representation of the workflow and summary of the results.

Automatic detection of various abnormalities in capsule endoscopy videos by a deep learning-based system: a multicenter study

BACKGROUND AND AIMS

A deep convolutional neural network (CNN) system could be a high-level screening tool for capsule endoscopy (CE) reading but has not been established for targeting various abnormalities. We aimed to develop a CNN-based system and compare it with the existing QuickView mode in terms of their ability to detect various abnormalities.

METHODS

We trained a CNN system using 66,028 CE images (44,684 images of abnormalities and 21,344 normal images). The detection rate of the CNN for various abnormalities was assessed per patient, using an independent test set of 379 consecutive small-bowel CE videos from 3 institutions. Mucosal breaks, angioectasia, protruding lesions, and blood content were present in 94, 29, 81, and 23 patients, respectively. The detection capability of the CNN was compared with that of QuickView mode.

RESULTS

The CNN picked up 1,135,104 images (22.5%) from the 5,050,226 test images, and thus, the sampling rate of QuickView mode was set to 23% in this study. In total, the detection rate of the CNN for abnormalities per patient was significantly higher than that of QuickView mode (99% vs 89%, P < .001). The detection rates of the CNN for mucosal breaks, angioectasia, protruding lesions, and blood content were 100% (94 of 94), 97% (28 of 29), 99% (80 of 81), and 100% (23 of 23), respectively, and those of QuickView mode were 91%, 97%, 80%, and 96%, respectively.

CONCLUSIONS

We developed and tested a CNN-based detection system for various abnormalities using multicenter CE videos. This system could serve as an alternative high-level screening tool to QuickView mode.

Non-steroidal anti-inflammatory drug-induced enteropathy as a major risk factor for small bowel bleeding: a retrospective study

Background

Small bowel (SB) bleeding accounts for 5% of all gastrointestinal (GI) bleeding cases and 80% of obscure GI bleeding cases. Although angioectasia is the common etiology of SB bleeding, nonsteroidal anti-inflammatory drug (NSAID)-induced SB lesions are also reported as a major cause in studies from Eastern countries. Herein, we assessed the frequency of occurrence of NSAID-induced SB lesions in Korean patients with obscure GI bleeding.

Methods

We retrospectively analyzed medical records of all consecutive patients aged ≥18 years who underwent capsule endoscopy from March 2018 to February 2019 at Ulsan University Hospital and Kosin University Gospel Hospital.

Results

Of the 83 subjects (all Korean; mean age ± standard deviation: 59 ± 18 years; age range: 18–84 years; men: n = 52; women: n = 31), 55 (66.2%) had stool with clear blood and 28 (33.8%) had normal stool with iron deficiency anemia. The detection rate of SB bleeding and lesions using capsule endoscopy was 72.3% (60 of 83 patients). A significantly higher frequency (40 of 51) of ulcerative/erosive lesions than other causes was observed in patients with inactive bleeding but visible SB lesions. As a result, NSAID-induced enteropathy accounted for 30.1% of 83 patients with obscure GI bleeding (25 of the all 60 SB bleeding cases).

Conclusions

Contrary to what is reported for patients in Western countries, this study in Korean patients showed an improved diagnostic yield of capsule endoscopy for obscure GI bleeding and that NSAID-induced enteropathy was the most common etiology of SB bleeding. Aggressive small intestine examination is required for patients with unexplained GI bleeding.

Clinical usefulness of a deep learning-based system as the first screening on small-bowel capsule endoscopy reading

BACKGROUND AND AIM

To examine whether our convolutional neural network (CNN) system based on deep learning can reduce the reading time of endoscopists without oversight of abnormalities in the capsule-endoscopy reading process.

METHODS

Twenty videos of the entire small-bowel capsule endoscopy procedure were prepared, each of which included 0-5 lesions of small-bowel mucosal breaks (erosions or ulcerations). At another institute, two reading processes were compared: (A) endoscopist-alone readings and (B) endoscopist readings after the first screening by the proposed CNN. In process B, endoscopists read only images detected by CNN. Two experts and four trainees independently read 20 videos each (10 for process A and 10 for process B). Outcomes were reading time and detection rate of mucosal breaks by endoscopists. Gold standard was findings at the original institute by two experts.

RESULTS

Mean reading time of small-bowel sections by endoscopists was significantly shorter during process B (expert, 3.1 min; trainee, 5.2 min) compared to process A (expert, 12.2 min; trainee, 20.7 min) (P < 0.001). For 37 mucosal breaks, detection rate by endoscopists did not significantly decrease in process B (expert, 87%; trainee, 55%) compared to process A (expert, 84%; trainee, 47%). Experts detected all eight large lesions (>5 mm), but trainees could not, even when supported by the CNN.

CONCLUSIONS

Our CNN-based system for capsule endoscopy videos reduced the reading time of endoscopists without decreasing the detection rate of mucosal breaks. However, the reading level of endoscopists should be considered when using the system.

Clinical Utility of Emergency Capsule Endoscopy for Diagnosing the Source and Nature of Ongoing Overt Obscure Gastrointestinal Bleeding

Background and Aims

In patients with ongoing overt obscure gastrointestinal bleeding (OGIB), prompt detection of the bleeding source is crucial to treatment success. However, there is no consensus on the optimal timing of diagnostic capsule endoscopy (CE). We investigated the clinical utility of emergency CE for detecting the source of ongoing overt OGIB.

Methods

We retrospectively evaluated 146 consecutive patients who, between February 2009 and July 2018, underwent emergency CE at Hiroshima University Hospital to detect the source of ongoing overt OGIB. Patients with a bleeding source located outside the small bowel were excluded. The remaining 127 patients were stratified according to the timing of CE relative to the onset of bleeding: patients in group A (n = 15, 12 men; mean age: 75 years; age range: 62–83 years) underwent CE within 48 hours of bleeding onset, whereas patients in group B (n = 112, 73 men; mean age: 65 years; age range: 17–88 years) underwent CE at >48 hours after bleeding onset. All patients underwent double-balloon endoscopy, and the final diagnosis was compared against the CE findings.

Results

The CE lesion detection rate was significantly higher in group A (12/15 patients, 80%) than in group B (53/112 patients, 47%) (p = 0.0174). There was no significant difference between the two groups regarding the patients' background characteristics. Vascular lesions were the most frequent finding in both groups. The diagnostic concordance rate between emergency CE and double-balloon endoscopy was 100% in group A and 92.9% in group B. Rebleeding after endoscopic treatment was confirmed in only one patient in group B.

Conclusions

Emergency CE represents a useful diagnostic modality in patients with ongoing overt OGIB, potentially improving detection rates and reducing rebleeding risk.

Small bowel ulcers: when is it not inflammatory bowel disease ?

PURPOSE OF REVIEW

Increasing use of small bowel endoscopy unravels ulcers in a relevant number of patients. Although often attributed to inflammatory bowel disease (IBD), these ulcers may be unspecific or caused by a variety of other diseases. This review summarizes the recent literature related to differential diagnosis of small bowel ulcers.

RECENT FINDINGS

NSAID enteropathy is the major differential diagnosis to IBD in patients with small bowel ulcers. Intestinal tuberculosis must be considered in patients at risk. Rare causes for small bowel ulcers are autoinflammatory, neoplastic, vascular diseases, or nontuberculous infections. Morphology of small bowel ulcers cannot provide a reliable differentiation, and even histology is not specific in all cases. History with special focus on NSAID medication and clinical symptoms not typical for IBD, laboratory tests in the search of systemic disease, and microbiologic testing of biopsies can be helpful. Genetic testing allows identification of few rare entities like defects in prostaglandin metabolism.

SUMMARY

Due to a massive overlap in the endoscopic appearance of small bowel ulcers between possible causes, diagnosis should be based on broad information also including history, histology, imaging, and laboratory tests.

Automatic detection of erosions and ulcerations in wireless capsule endoscopy images based on a deep convolutional neural network

BACKGROUND AND AIMS

Although erosions and ulcerations are the most common small-bowel abnormalities found on wireless capsule endoscopy (WCE), a computer-aided detection method has not been established. We aimed to develop an artificial intelligence system with deep learning to automatically detect erosions and ulcerations in WCE images.

METHODS

We trained a deep convolutional neural network (CNN) system based on a Single Shot Multibox Detector, using 5360 WCE images of erosions and ulcerations. We assessed its performance by calculating the area under the receiver operating characteristic curve and its sensitivity, specificity, and accuracy using an independent test set of 10,440 small-bowel images including 440 images of erosions and ulcerations.

RESULTS

The trained CNN required 233 seconds to evaluate 10,440 test images. The area under the curve for the detection of erosions and ulcerations was 0.958 (95% confidence interval [CI], 0.947-0.968). The sensitivity, specificity, and accuracy of the CNN were 88.2% (95% CI, 84.8%-91.0%), 90.9% (95% CI, 90.3%-91.4%), and 90.8% (95% CI, 90.2%-91.3%), respectively, at a cut-off value of 0.481 for the probability score.

CONCLUSIONS

We developed and validated a new system based on CNN to automatically detect erosions and ulcerations in WCE images. This may be a crucial step in the development of daily-use diagnostic software for WCE images to help reduce oversights and the burden on physicians.

Management of occult obscure gastrointestinal bleeding patients based on long-term outcomes

BACKGROUND

There is no consensus regarding the management of occult obscure gastrointestinal bleeding (OGIB) patients without a confirmed bleeding source. This study aimed to consider the management of occult OGIB patients based on their long-term outcomes.

METHODS

We retrospectively enrolled 357 consecutive occult OGIB patients (203 men; mean age: 59.7 years) who underwent capsule endoscopy (CE) at Hiroshima University Hospital, Japan and were followed up for more than 12 months (mean follow-up period; 50.2 months). Patients were divided into three groups as follows: Group A consisted of 98 of 157 patients who had positive findings and indication for treatment, Group B consisted of 59 of 157 patients who had positive findings but no indication for treatment, and Group C consisted of 200 patients who had negative small-bowel findings. We examined the rate of positive CE findings, detection rate and details of bleeding sources, overt bleeding rate, the rate of anemia exacerbation, 5-year anemia exacerbation rate, and overall survival rate.

RESULTS

The positive CE findings rate was 44% (157/357) and detection rate of bleeding source was 27% (98/357). The details of Group A were as follows: angioectasia (n = 61), nonspecific ulceration (n = 10), nonsteroidal anti-inflammatory drug-induced ulcer (n = 8), and others (n = 19). The details of Group B were as follows: erythema (n = 31), angioectasia (n = 25), and others (n = 3). There were no patients with overt bleeding in Group B. Although six patients had anemia exacerbation in Group B, they had angioectasia without a bleeding source.

CONCLUSION

The long-term outcomes of occult OGIB patients were good. Occult OGIB patients without bleeding source lesions may not require follow-up CE.