Changing trends in the UK management of upper GI bleeding: is there evidence of reduced UK training experience?

Learning curve of achieving competency in emergency endoscopy in upper gastrointestinal bleeding: how much experience is necessary?

OBJECTIVES

The management of upper gastrointestinal bleeding (UGIB) has seen rapid advancements with revolutionising innovations. However, insufficient data exist on the necessary number of emergency endoscopies needed to achieve competency in haemostatic interventions.

DESIGN

We retrospectively analysed all oesophagogastroduodenoscopies with signs of recent haemorrhage performed between 2015 and 2022 at our university hospital. A learning curve was created by plotting the number of previously performed oesophagogastroduodenoscopies with signs of recent haemorrhage against the treatment failure rate, defined as failed haemostasis, rebleeding and necessary surgical or radiological intervention.

RESULTS

The study population included 787 cases with a median age of 66 years. Active bleeding was detected in 576 cases (73.2%). Treatment failure occurred in 225 (28.6%) cases. The learning curve showed a marked decline in treatment failure rates after nine oesophagogastroduodenoscopies had been performed by the respective endoscopists followed by a first plateau between 20 and 50 procedures. A second decline was observed after 51 emergency procedures followed by a second plateau. Endoscopists with experience of <10 emergency procedures had higher treatment failure rates compared with endoscopists with >51 emergency oesophagogastroduodenoscopies performed (p=0.039) or consultants (p=0.041).

CONCLUSIONS

Our data suggest that a minimum number of 20 oesophagogastroduodenoscopies with signs of recent haemorrhage is necessary before endoscopists should be considered proficient to perform emergency procedures independently. Endoscopists might be considered as advanced-qualified experts in managing UGIB after a minimum of 50 haemostatic procedure performed. Implementing recommendations on minimum numbers of emergency endoscopies in education programmes of endoscopy trainees could improve their confidence and competency in managing acute UGIB.

Upper endoscopy photodocumentation quality evaluation with novel deep learning system

OBJECTIVES

Visualization and photodocumentation during endoscopy procedures are suggested to be one indicator for endoscopy performance quality. However, this indicator is difficult to measure and audit manually in clinical practice. Artificial intelligence (AI) is an emerging technology that may solve this problem.

METHODS

A deep learning model with an accuracy of 96.64% was developed from 15,305 images for upper endoscopy anatomy classification in the unit. Endoscopy images for asymptomatic patients receiving screening endoscopy were evaluated with this model to assess the completeness of photodocumentation rate.

RESULTS

A total of 15,723 images from 472 upper endoscopies performed by 12 endoscopists were enrolled. The complete photodocumentation rate from the pharynx to the duodenum was 53.8% and from the esophagus to the duodenum was 78.0% in this study. Endoscopists with a higher adenoma detection rate had a higher complete examination rate from the pharynx to duodenum (60.0% vs. 38.7%, P < 0.0001) and from esophagus to duodenum (83.0% vs. 65.7%, P < 0.0001) compared with endoscopists with lower adenoma detection rate. The pharynx, gastric angle, gastric retroflex view, gastric antrum, and the first portion of duodenum are likely to be missed by endoscopists with lower adenoma detection rates.

CONCLUSIONS

We report the use of a deep learning model to audit endoscopy photodocumentation quality in our unit. Endoscopists with better performance in colonoscopy had a better performance for this quality indicator. The use of such an AI system may help the endoscopy unit audit endoscopy performance.

Deep learning-based endoscopic anatomy classification: an accelerated approach for data preparation and model validation

BACKGROUND

Photodocumentation during endoscopy procedures is one of the indicators for endoscopy performance quality; however, this indicator is difficult to measure and audit in the endoscopy unit. Emerging artificial intelligence technology may solve this problem, which requires a large amount of material for model development. We developed a deep learning-based endoscopic anatomy classification system through convolutional neural networks with an accelerated data preparation approach.

PATIENTS AND METHODS

We retrospectively collected 8,041 images from esophagogastroduodenoscopy (EGD) procedures and labeled them using two experts for nine anatomical locations of the upper gastrointestinal tract. A base model for EGD image multiclass classification was first developed, and an additional 6,091 images were enrolled and classified by the base model. A total of 5,963 images were manually confirmed and added to develop the subsequent enhanced model. Additional internal and external endoscopy image datasets were used to test the model performance.

RESULTS

The base model achieved total accuracy of 96.29%. For the enhanced model, the total accuracy was 96.64%. The overall accuracy improved with the enhanced model compared with the base model for the internal test dataset without narrowband images (93.05% vs. 91.25%, p < 0.01) or with narrowband images (92.74% vs. 90.46%, p < 0.01). The total accuracy was 92.56% of the enhanced model on the external test dataset.

CONCLUSIONS

We constructed a deep learning-based model with an accelerated approach that can be used for quality control in endoscopy units. The model was also validated with both internal and external datasets with high accuracy.

Forrest Classification for Bleeding Peptic Ulcer: A New Look at the Old Endoscopic Classification

The management of peptic ulcer bleeding is clinically challenging. For decades, the Forrest classification has been used for risk stratification for nonvariceal ulcer bleeding. The perception and interpretation of the Forrest classification vary among different endoscopists. The relationship between the bleeder and ulcer images and the different stages of the Forrest classification has not been studied yet. Endoscopic still images of 276 patients with peptic ulcer bleeding for the past 3 years were retrieved and reviewed. The intra-rater agreement and inter-rater agreement were compared. The obtained endoscopic images were manually drawn to delineate the extent of the ulcer and bleeding area. The areas of the region of interest were compared between the different stages of the Forrest classification. A total of 276 images were first classified by two experienced tutor endoscopists. The images were reviewed by six other endoscopists. A good intra-rater correlation was observed (0.92–0.98). A good inter-rater correlation was observed among the different levels of experience (0.639–0.859). The correlation was higher among tutor and junior endoscopists than among experienced endoscopists. Low-risk Forrest IIC and III lesions show distinct patterns compared to high-risk Forrest I, IIA, or IIB lesions. We found good agreement of the Forrest classification among different endoscopists in a single institution. This is the first study to quantitively analyze the obtained and explain the distinct patterns of bleeding ulcers from endoscopy images.

Current Status and Future Perspective of Artificial Intelligence in the Management of Peptic Ulcer Bleeding: A Review of Recent Literature

With the decreasing incidence of peptic ulcer bleeding (PUB) over the past two decades, the clinician experience of managing patients with PUB has also declined, especially for young endoscopists. A patient with PUB management requires collaborative care involving the emergency department, gastroenterologist, radiologist, and surgeon, from initial assessment to hospital discharge. The application of artificial intelligence (AI) methods has remarkably improved people's lives. In particular, AI systems have shown great potential in many areas of gastroenterology to increase human performance. Colonoscopy polyp detection or diagnosis by an AI system was recently introduced for commercial use to improve endoscopist performance. Although PUB is a longstanding health problem, these newly introduced AI technologies may soon impact endoscopists' clinical practice by improving the quality of care for these patients. To update the current status of AI application in PUB, we reviewed recent relevant literature and provided future perspectives that are required to integrate such AI tools into real-world practice.

The Role of Haemostasis Course in Increasing Knowledge and Skills in Managing Upper Gastrointestinal Bleed of the Delegates: A British Society of Gastroenterology's Endoscopy Quality Improvement Programme, Yorkshire Project

Introduction

An acute upper gastrointestinal bleed (AUGIB) is a fatal and prevalent medical emergency if not appropriately treated in a timely fashion.

Aim

The aim of this project was to compare the knowledge and skills of the participants in managing upper gastrointestinal bleeding (UGIB) before and after a one-day UGIB haemostasis course.

Methods

A one-day haemostasis course in line with the British Society of Gastroenterology’s Endoscopy Quality Improvement Project Initiative was organised at the Sheffield Teaching Hospitals National Health Service (NHS) Trust. The course included lectures on UGIB and its management, which was followed by hands-on training on adrenaline injection, variceal banding, clip placement, thermal therapy, Hemospray® use, Sengstaken-Blakemore tube placement, and Danis stent placement via porcine or plastic models. Pre- and post-course feedback questionnaires consisting of self-assessed ratings related to knowledge, skills, and behaviour relevant to UGIB were offered to all delegates. Two-tailed Wilcoxon signed-rank test was used to compare the results.

Results

A total of 36 individuals attended the course. Delegates had an average endoscopy procedure count of 583. The cohort ranged from different fields of medicine, including gastroenterology consultants and junior doctors. Ten of the delegates were Joint Advisory Group-certified in upper gastrointestinal endoscopy. Feedback datasheets were returned by 22 delegates. Significant improvements were reported post-course (p < 0.001), especially in the hands-on and behavioural areas.

Conclusion

Overall, there was a significant improvement in the knowledge, procedural skills, and confidence of the delegates in the management of an AUGIB post-course. We recommend not only to include this course in gastrointestinal training but also to conduct a course such as this for consultants and junior doctors who wish to undergo gastrointestinal training in the future.

Performance Comparison of the Deep Learning and the Human Endoscopist for Bleeding Peptic Ulcer Disease

Purpose

Management of peptic ulcer bleeding is clinically challenging. Accurate characterization of the bleeding during endoscopy is key for endoscopic therapy. This study aimed to assess whether a deep learning model can aid in the classification of bleeding peptic ulcer disease.

Methods

Endoscopic still images of patients (n = 1694) with peptic ulcer bleeding for the last 5 years were retrieved and reviewed. Overall, 2289 images were collected for deep learning model training, and 449 images were validated for the performance test. Two expert endoscopists classified the images into different classes based on their appearance. Four deep learning models, including Mobile Net V2, VGG16, Inception V4, and ResNet50, were proposed and pre-trained by ImageNet with the established convolutional neural network algorithm. A comparison of the endoscopists and trained deep learning model was performed to evaluate the model’s performance on a dataset of 449 testing images.

Results

The results first presented the performance comparisons of four deep learning models. The Mobile Net V2 presented the optimal performance of the proposal models. The Mobile Net V2 was chosen for further comparing the performance with the diagnostic results obtained by one senior and one novice endoscopists. The sensitivity and specificity were acceptable for the prediction of “normal” lesions in both 3-class and 4-class classifications. For the 3-class category, the sensitivity and specificity were 94.83% and 92.36%, respectively. For the 4-class category, the sensitivity and specificity were 95.40% and 92.70%, respectively. The interobserver agreement of the testing dataset of the model was moderate to substantial with the senior endoscopist. The accuracy of the determination of endoscopic therapy required and high-risk endoscopic therapy of the deep learning model was higher than that of the novice endoscopist.

Conclusions

In this study, the deep learning model performed better than inexperienced endoscopists. Further improvement of the model may aid in clinical decision-making during clinical practice, especially for trainee endoscopist.

Training in endotherapy for acute upper gastrointestinal bleeding: a UK-wide gastroenterology trainee survey

INTRODUCTION

Competence in endoscopic haemostasis for acute upper gastrointestinal bleeding (AUGIB) is typically expected upon completion of gastroenterology training. However, training in haemostasis is currently variable without a structured training pathway. We conducted a national gastroenterology trainee survey on haemostasis exposure and on attitudes and barriers to training.

METHODS

A 24-item electronic survey was distributed to UK gastroenterology trainees covering the following domains: demographics, training setup, attitudes and barriers, confidence in managing AUGIB independently and exposure to individual haemostatic modalities (supervised and independent). Responses were analysed by region and training grade to assess potential variation in training.

RESULTS

A total of 181 trainees completed the questionnaire (response rate 33.5%). There was significant variation in AUGIB training setup across the UK (p<0.001), with 22.7% of trainees declaring no access to structured or ad hoc training. 31.5% expressed confidence in managing AUGIB independently; this varied by trainee grade (0% of first-year specialty trainees (ST3s) to 60.7% of final-years (ST7s)) and by training setup (p=0.001). ST7 trainees reported lack of experience with independently applying glue (86%), Hemospray (54%), heater probe (36%) and variceal banding (36%). Overall, 88% of trainees desired additional haemostasis training and 89% indicated support for a national certification process to ensure competence in AUGIB.

CONCLUSION

AUGIB training in the UK is variable. The majority of gastroenterology trainees lacked confidence in haemostasis management and desired additional training. Training provision should be urgently reviewed to ensure that trainees receive adequate haemostasis exposure and are competent by completion of training.

Variation in exposure to endoscopic haemostasis for acute upper gastrointestinal bleeding during UK gastroenterology training

INTRODUCTION

Gastroenterologists are typically expected to be competent in endoscopic haemostasis for acute upper gastrointestinal bleeding (AUGIB), with the Certificate of Completion of Training (CCT) often heralding the onset of participation in on-call AUGIB rotas. We analysed the volume of haemostasis experience recorded by gastroenterology CCT holders on the Joint Advisory Group on Gastrointestinal Endoscopy Training System (JETS) e-portfolio, the UK electronic portfolio for endoscopy, and assessed for variations in exposure to haemostasis.

METHODS

UK gastroenterologists awarded CCT between April 2014 and April 2017 were retrospectively identified from the specialist register. Credentials were cross-referenced with JETS to retrieve AUGIB haemostasis procedures prior to CCT. Procedures were collated according to variceal versus non-variceal therapies and compared across training deaneries.

RESULTS

Over the 3-year study period, 241 gastroenterologists were awarded CCT. 232 JETS e-portfolio users were included for analysis. In total, 12 932 haemostasis procedures were recorded, corresponding to a median of 42 (IQR 21-71) per gastroenterologist. Exposure to non-variceal modalities (median 28, IQR 15-52) was more frequent than variceal therapies (median 11, IQR 5-22; p<0.001). By procedure, adrenaline injection (median 12, IQR 6-23) and variceal band ligation (median 10, IQR 5-20) were most commonly recorded, whereas sclerotherapy experience was rare (median 0, IQR 0-1). Exposure to haemostasis did not differ by year of CCT (p=0.130) but varied significantly by deanery (p<0.001), with median procedures ranging from 20-126.

CONCLUSION

Exposure to AUGIB haemostasis during UK gastroenterology training varied across deaneries and procedural modalities which should prompt urgent locoregional review of access and delivery of training. Endoscopy departments should ensure the availability of supportive provisions in haemostasis (i.e. training/upskilling, supervision, mentorship) during the early post-CCT period.

Trends in UK endoscopy training in the BSG trainees' national survey and strategic planning for the future

BACKGROUND

Improvements in the structure of endoscopy training programmes resulting in certification from the Joint Advisory Group in Gastrointestinal Endoscopy have been acknowledged to improve training experience and contribute to enhanced colonoscopy performance.

OBJECTIVES

The 2016 British Society of Gastroenterology trainees' survey of endoscopy training explored the delivery of endoscopy training - access to lists; level of supervision and trainee's progression through diagnostic, core therapy and subspecialty training. In addition, the barriers to endoscopy training progress and utility of training tools were examined.

METHODS

A web-based survey (Survey Monkey) was sent to all higher specialty gastroenterology trainees.

RESULTS

There were some improvements in relation to earlier surveys; 85% of trainees were satisfied with the level of supervision of their training. But there were ongoing problems; 12.5% of trainees had no access to a regular training list, and 53% of final year trainees had yet to achieve full certification in colonoscopy. 9% of final year trainees did not feel confident in endoscopic management of upper GI bleeds.

CONCLUSIONS

The survey findings provide a challenge to those agencies tasked with supporting endoscopy training in the UK. Acknowledging the findings of the survey, the paper provides a strategic response with reference to increased service pressures, reduced overall training time in specialty training programmes and the requirement to support general medical and surgical on-call commitments. It describes the steps required to improve training on the ground: delivering additional training tools and learning resources, and introducing certification standards for therapeutic modalities in parallel with goals for improving the quality of endoscopy in the UK.

Acute upper gastrointestinal bleeding

Acute upper gastrointestinal bleeding (AUGIB) is a frequently encountered medical emergency with an incidence of 84-160/100000 and associated with mortality of approximately 10%. Guidelines from the National Institute for Care and Care Excellence outline key features in the management of AUGIB. Patients require prompt resuscitation and risk assessment using validated tools. Upper gastrointestinal endoscopy provides accurate diagnosis, aids in estimating prognosis and allows therapeutic intervention. Endoscopy should be undertaken immediately after resuscitation in unstable patients and within 24 hours in all other patients. Interventional radiology may be required for bleeding unresponsive to endoscopic intervention. Drug therapy depends on the cause of bleeding. Intravenous proton pump inhibitors should be used in patients with high-risk ulcers. Terlipressin and broad-spectrum antibiotics should be used following variceal haemorrhage. Hospitals admitting patients with AUGIB need to provide well organised services and ensure access to relevant services for all patients, and particularly to out of hours endoscopy.