Prediction model for bleeding after endoscopic submucosal dissection of gastric neoplasms from a high-volume center

Modified BEST-J Score Model Predicts Bleeding after Endoscopic Submucosal Dissection with Fewer Factors

This study constructed a simplified post-endoscopic submucosal dissection (ESD) prediction model with a prognostic nutritional index (PNI). A total of 449 patients who underwent gastric ESD was included, divided with a ratio of 2:1, and assigned to the model or validation cohort. A prediction model of post-ESD (modified BEST-J score) was constructed using the model cohort. The modified BEST-J score was evaluated by comparing its accuracy to the BEST-J score in the validation cohort. Within 4 weeks of ESD, melena, hematemesis, or a 2 g/dL or greater decrease in hemoglobin level that required esophagogastroduodenoscopy was defined as post-ESD bleeding. In the model cohort, 299 patients were enrolled and 25 (8.4%) had post-ESD bleeding. Independent risk factors for post-ESD bleeding were use of P2Y12RA, tumor size > 30 mm, location of lesion at lower one-third of the stomach, and PNI ≤ 47.9. Constructing the modified BEST-J score based on these variables, the sensitivity, specificity, and positive likelihood ratio were 73.9%, 78.1%, and 3.37. When comparing the modified BEST-J score to the BEST-J score in the validation cohort, no significant difference was observed by ROC-AUC (0.77 vs. 0.75, p = 0.81). Modified BEST-J score can predict post-ESD bleeding more simply, with the same accuracy as the BEST-J score.

Reopenable clip-over-the-line method for closing large mucosal defects following gastric endoscopic submucosal dissection: Prospective feasibility study

OBJECTIVES

Large mucosal defects following gastric endoscopic submucosal dissection (ESD) cause postoperative bleeding. To address this limitation and ensure closure of large mucosal defects, we developed the reopenable clip-over-the-line method (ROLM) using a reopenable clip and nylon line. The purpose of this study was to evaluate the feasibility of the ROLM for closure of large mucosal defects following gastric ESD in a prospective, consecutive series of cases.

METHODS

We performed the ROLM on 50 consecutive patients with gastric mucosal defects at the Ise Red Cross Hospital and Mie Prefectural Shima Hospital. The time to complete the ROLM, numbers of clips and lines required, size of defect, and closure success rate were measured, and postoperative adverse events were recorded.

RESULTS

In all, 50 lesions were included in this study period between July 2021 and March 2022. The success rates of defect closure and defect closure without submucosal dead space of the ROLM were both 100% (50/50), with a median ROLM time of 30 (range, 14-35) min and a median resected specimen major axis of 45 (range, 31-73) mm. The median number of reopenable clips used was 31 (range, 10-93). Following gastric ESD, two cases of post-ESD bleeding were observed during the follow-up periods.

CONCLUSION

Our results suggest that ROLM is a feasible strategy for complete closure of mucosal defects post-ESD without submucosal dead space. Future comparative studies with post-ESD bleeding rate as the main outcome are desirable to evaluate the efficacy of ROLM.

Utility of a deep learning model and a clinical model for predicting bleeding after endoscopic submucosal dissection in patients with early gastric cancer

BACKGROUND

Bleeding is one of the major complications after endoscopic submucosal dissection (ESD) in early gastric cancer (EGC) patients. There are limited studies on estimating the bleeding risk after ESD using an artificial intelligence system.

AIM

To derivate and verify the performance of the deep learning model and the clinical model for predicting bleeding risk after ESD in EGC patients.

METHODS

Patients with EGC who underwent ESD between January 2010 and June 2020 at the Samsung Medical Center were enrolled, and post-ESD bleeding (PEB) was investigated retrospectively. We split the entire cohort into a development set (80%) and a validation set (20%). The deep learning and clinical model were built on the development set and tested in the validation set. The performance of the deep learning model and the clinical model were compared using the area under the curve and the stratification of bleeding risk after ESD.

RESULTS

A total of 5629 patients were included, and PEB occurred in 325 patients. The area under the curve for predicting PEB was 0.71 (95% confidence interval: 0.63-0.78) in the deep learning model and 0.70 (95% confidence interval: 0.62-0.77) in the clinical model, without significant difference (P = 0.730). The patients expected to the low- (< 5%), intermediate- (≥ 5%, < 9%), and high-risk (≥ 9%) categories were observed with actual bleeding rate of 2.2%, 3.9%, and 11.6%, respectively, in the deep learning model; 4.0%, 8.8%, and 18.2%, respectively, in the clinical model.

CONCLUSION

A deep learning model can predict and stratify the bleeding risk after ESD in patients with EGC.

A narrative review of postoperative bleeding in patients with gastric cancer treated with endoscopic submucosal dissection

Endoscopic submucosal dissection (ESD) is now considered a standard treatment for selected patients with early gastric cancer. Compared with endoscopic mucosal resection (EMR), ESD provides a higher complete resection rate (R0), and therefore, a lower local recurrence rate. However, ESD is a more time-consuming procedure, creating a wider and deeper ulcer floor which may cause complications. Post-ESD bleeding is one of them. Although most post-ESD bleedings can be controlled by endoscopic hemostasis at the time of operation, some bleeding after ESD may result in serious conditions such as hemorrhagic shock. Even with preventive methods such as ulcer closure, the application of fibrin glue and polyglycolic acid shielding, acid secretion inhibitors and hemostasis on second-look endoscopy, our experiences told us that post-ESD bleeding cannot be entirely avoidable, especially for patients with big size ulcer bed, anticoagulants/antithrombosis and chronic kidney diseases. The present review first defined post-ESD bleeding, then the incidence, the risk factors, such as the location of operative lesion, the size and depth, chronic kidney diseases, the impacts of anticoagulant and antithrombotic agents. We finally reviewed the managements of post-ESD bleeding, including approaches of coagulating potential bleeding spots during the procedure, lesion closure, lesion shielding and the application of gastric acid secretion inhibitors.

Gray-Level Co-occurrence Matrix Analysis for the Detection of Discrete, Ethanol-Induced, Structural Changes in Cell Nuclei: An Artificial Intelligence Approach

Gray-level co-occurrence matrix (GLCM) analysis is a contemporary and innovative computational method for the assessment of textural patterns, applicable in almost any area of microscopy. The aim of our research was to perform the GLCM analysis of cell nuclei in Saccharomyces cerevisiae yeast cells after the induction of sublethal cell damage with ethyl alcohol, and to evaluate the performance of various machine learning (ML) models regarding their ability to separate damaged from intact cells. For each cell nucleus, five GLCM parameters were calculated: angular second moment, inverse difference moment, GLCM contrast, GLCM correlation, and textural variance. Based on the obtained GLCM data, we applied three ML approaches: neural network, random trees, and binomial logistic regression. Statistically significant differences in GLCM features were observed between treated and untreated cells. The multilayer perceptron neural network had the highest classification accuracy. The model also showed a relatively high level of sensitivity and specificity, as well as an excellent discriminatory power in the separation of treated from untreated cells. To the best of our knowledge, this is the first study to demonstrate that it is possible to create a relatively sensitive GLCM-based ML model for the detection of alcohol-induced damage in Saccharomyces cerevisiae cell nuclei.

Strategies that Reduce Post-endoscopic Submucosal Dissection Bleeding

Bleeding after endoscopic submucosal dissection (ESD), one of the most common postprocedural adverse events, is the main cause of increased readmission rates and hospital costs. Generally, the incidence of post-ESD bleeding is estimated to be about 5%. However, the incidence of bleeding is particularly increased in high-risk patients. In particular, it has an incidence of over 50% in patients that use antithrombotic agents. The well-known risk factors for post-ESD bleeding include antithrombotic therapy, lesions in the proximal stomach, specimen size >4 cm, and concomitant renal disease. Currently, the number of patients at a high risk of post-ESD bleeding has been increasing. This may be due to the aging society and the increase in the usage of antithrombotic agents. Therefore, several strategies have been employed to prevent post-ESD bleeding. These strategies include acid inhibition therapy, preventive hemostasis using Doppler endoscopic ultrasound and artery-selective clipping, second look endoscopy, the closings method, and the shield methods. However, these methods are technically demanding, which hinders their wide usage in clinical practice. Recently, several hemostatic powders have been developed and clinically used in the treatment of gastrointestinal bleeding. In this article, we review the risk factors for post-ESD bleeding and the recently introduced prevention methods. Moreover, we aimed to explore realistic and appropriate strategies for the prevention of post-ESD bleeding.