Development of convolutional neural network models that recognize normal anatomic structures during real-time radial-array and linear-array EUS (with videos).

BACKGROUND AND AIMS: EUS is a high-skill technique that requires numerous procedures to achieve competence. However, training facilities are limited worldwide. Convolutional neural network (CNN) models have been previously implemented for object detection. We developed 2 EUS-based CNN models for normal anatomic structure recognition during real-time linear- and radial-array EUS evaluations. METHODS: The study was performed from February 2020 to June 2022. Consecutive patient videos of linear- and radial-array EUS videos were recorded. Expert endosonographers identified and labeled 20 normal anatomic structures within the videos for training and validation of the CNN models. Initial CNN models (CNNv1) were developed from 45 videos and the improved models (CNNv2) from an additional 102 videos. CNN model performance was compared with that of 2 expert endosonographers. RESULTS: CNNv1 used 45,034 linear-array EUS frames and 21,063 radial-array EUS frames. CNNv2 used 148,980 linear-array EUS frames and 128,871 radial-array EUS frames. Linear-array CNNv1 and radial-array CNNv1 achieved a 75.65% and 71.36% mean average precision (mAP) with a total loss of .19 and .18, respectively. Linear-array CNNv2 obtained an 88.7% mAP with a .06 total loss, whereas radial-array CNNv2 achieved an 83.5% mAP with a .07 total loss. CNNv2 accurately detected all studied normal anatomic structures with a >98% observed agreement during clinical validation. CONCLUSIONS: The proposed CNN models accurately recognize the normal anatomic structures in prerecorded videos and real-time EUS. Prospective trials are needed to evaluate the impact of these models on the learning curves of EUS trainees.

Assessing EUS-guided radiofrequency ablation in unresectable pancreatic ductal adenocarcinoma: a single-center historic cohort study.

BACKGROUND AND AIMS: EUS-guided radiofrequency ablation (EUS-RFA) has emerged as an alternative for the local treatment of unresectable pancreatic ductal adenocarcinoma (PDAC). We assessed the feasibility and safety of EUS-RFA in patients with unresectable PDAC. METHODS: This study followed an historic cohort compounded by locally advanced (LA-) and metastatic (m)PDAC-naïve patients who underwent EUS-RFA between October 2019 and March 2022. EUS-RFA was performed with a 19-gauge needle electrode with a 10-mm active tip for energy delivery. Study primary endpoints were feasibility, safety, and clinical follow-up, whereas secondary endpoints were performance status (PS), local control, and overall survival (OS). RESULTS: Twenty-six patients were selected: 15 with locally advanced pancreatic duct adenocarcinoma (LA-PDAC) and 11 with metastatic pancreatic duct adenocarcinoma (mPDAC). Technical success was achieved in all patients with no major adverse events. Six months after EUS-RFA, OS was seen in 11 of 26 patients (42.3%), with significant PS improvement (P = .03). Local control was achieved, with tumor reduction from 39.5 mm to 26 mm (P = .04). A post-treatment hypodense necrotic area was observed at the 6-month follow-up in 11 of 11 patients who were still alive. Metastatic disease was a significant factor for worsening OS (hazard ratio, 5.021; 95% confidence interval, 1.589-15.87; P = .004). CONCLUSIONS: EUS-RFA for the treatment of pancreatic adenocarcinoma is a minimally invasive and safe technique that may have an important role as targeted therapy for local treatment of unresectable cases and as an alternative for poor surgical candidates. Also, RFA may play a role in downstaging cancer with a potential increase in OS in nonmetastatic cases. Large prospective cohorts are required to evaluate this technique in clinical practice.

Short esophageal myotomy versus standard myotomy for treatment of sigmoid-type achalasia: results of an international multicenter study.

BACKGROUND AND AIMS: Patients with sigmoid-type achalasia can be challenging to treat with peroral endoscopic myotomy (POEM). A short myotomy improves technical success; however, outcomes have not previously been evaluated. METHODS: This was a multicenter, international, retrospective study of patients who underwent POEM with short (≤4 cm) or standard esophageal myotomy. Outcomes included clinical and technical success, procedural adverse events, and reflux rates. RESULTS: A total of 109 patients with sigmoid achalasia (sigmoid, n = 74; advanced sigmoid, n = 35) underwent POEM across 13 centers (short myotomy, n = 59; standard, n = 50). Technical success was 100% across both groups. Patients who underwent short myotomy had a significantly shorter mean procedure time (57.7 ± 27.8 vs 83.1 ± 44.7 minutes, P = .0005). A total of 6 adverse events were recorded in 6 patients (5.5%; 4 mild, 2 moderate); the adverse event rate was not significantly different between short and standard groups. Ninety-eight patients had follow-up data (median, 3.6 months; interquartile range, 1-14 months). Clinical success was 94% (short, 93%; standard, 95%; P = .70) and did not differ based on achalasia subtype or sigmoid achalasia severity. Twenty-one (22%) patients reported post-POEM reflux and 44% (16 of 36) had objective evidence of pathologic reflux. Rates of pathologic reflux were significantly increased in the standard versus short group (odds ratio, 18.0; 95% confidence interval, 2.0-159.0; P = .009). CONCLUSIONS: POEM with short myotomy is effective and safe for the short-term treatment of sigmoid and advanced sigmoid achalasia. Short myotomy may lead to less reflux than standard myotomy.

Short-Term Clinical and Technical Outcomes of a Modified Zenker's Peroral Endoscopic Myotomy (Z-POEM) With Mucosal Flap Incision.

BACKGROUND: Zenker's diverticulum (ZD) per-oral endoscopic myotomy (Z-POEM) is an effective treatment for symptomatic (ZD). A modification to Z-POEM is the mucosal flap incision (MFI). We describe the technical and clinical success of patients who underwent Z-POEM with MFI. METHODS: We included patients who underwent Z-POEM with MFI for ZD at 8 international centers. The primary outcome was the rate of clinical success, assessed by post-procedure Kothari-Haber symptom score (KHSS) without retreatment. Secondary outcomes included technical success, serious adverse events, and clinical recurrence with need for retreatment. RESULTS: 36 patients (69% male, age 69 + 9 years) underwent Z-POEM with MFI for symptomatic ZD. Mean ZD size was 32 mm (+ 14). Clinical success was achieved in 35 (97%) cases. Median pre-procedure KHSS was 6 and median post-procedure KHSS was 1 (P <0.0001). Technical success was achieved in all cases. Mean procedure time was 57 minutes (+ 32), median follow up time was 196 days (IQR 39-499). There was one adverse event (3%) which was treated endoscopically. CONCLUSION: Z-POEM with MFI has a high rate of technical and clinical success. Prospective evaluation is needed to further validate this technique in patients with large ZD.

Impact of COVID-19 Infection on Pancreato-Biliary Diseases Requiring Endoscopic Retrograde Cholangiopancreatography.

BACKGROUND: Viral infections are known to impact the pancreato-biliary system; however, there are limited data showing that the same is true of COVID-19. Endoscopic retrograde cholangiopancreatography (ERCP) can safely be performed in patients with COVID-19 infection, but outcomes of patients with COVID-19 infections and concomitant pancreatic and biliary disease requiring endoscopic intervention are unknown. AIMS: This study aims to evaluate the severity of pancreaticobiliary diseases and post-ERCP outcomes in COVID-19 patients. METHODS: Patients with pancreato-biliary disease that required inpatient ERCP from five centers in the United States and South America between January 1, 2020, and October 31, 2020 were included. A representative cohort of patients from each month were randomly selected from each site. Disease severity and post-ERCP outcomes were compared between COVID-19 positive and COVID-19 negative patients. RESULTS: A total of 175 patients were included: 95 COVID positive and 80 COVID negative. Mean CTSI score for the patients who had pancreatitis was higher in COVID-positive cohort by 3.2 points (p < .00001). The COVID-positive group had more cases with severe disease (n = 41) versus the COVID-negative group (n = 2) (p < .00001). Mortality was higher in the COVID-19 positive group (19%) compared to COVID-negative group (7.5%) even though the COVID-19-negative group had higher incidence of malignancy (n = 17, 21% vs n = 7, 7.3%) (p = 0.0455). CONCLUSIONS: This study shows that patients with COVID infection have more severe pancreato-biliary disease and worse post-ERCP outcomes, including longer length of stay and higher mortality rate. These are important considerations when planning for endoscopic intervention. CLINICALTRIALS: gov: (NCT05051358).

Prophylactic EUS-guided gallbladder drainage prevents acute cholecystitis in patients with malignant biliary obstruction and cystic duct orifice involvement: a randomized trial (with video).

BACKGROUND AND AIMS: Patients with distal malignant biliary obstruction (MBO) and cystic duct orifice tumoral involvement have an increased risk for the development of acute cholecystitis after self-expandable metallic stent (SEMS) placement. We aimed to determine whether primary EUS-guided gallbladder drainage prevents acute cholecystitis in these patients. METHODS: This was a single-center, randomized control trial in patients with distal MBO enrolled from July 2018 to July 2020. Patients were randomized into 2 groups: an interventional group treated with conventional ERCP biliary drainage with SEMS placement and subsequent primary EUS-guided gallbladder drainage (EUS-GBD) and a control group treated with conventional biliary drainage alone. The primary outcome of the study was the occurrence of post-treatment acute cholecystitis, assessed for ≤12 months or until death. The secondary outcomes were hospitalization length and median survival time. RESULTS: Forty-four patients were included in the study: 22 in each group. Five patients in the control group (22.7%) and none in the intervention group experienced acute cholecystitis. The median hospitalization time was significantly lower in the interventional group than in the control group (2 days vs 1 day, P = .017). There was no difference in the observed median survival rates in the primary EUS-GBD group (2.9 months) and the control group (2.8 months) (P = .580). CONCLUSION: In this single-center study of patients with unresectable MBO and occlusion of the cystic duct orifice, prophylactic EUS-GBD demonstrated a reduced incidence of acute cholecystitis.

Artificial intelligence for diagnosing neoplasia on digital cholangioscopy: development and multicenter validation of a convolutional neural network model.

BACKGROUND: We aimed to develop a convolutional neural network (CNN) model for detecting neoplastic lesions during real-time digital single-operator cholangioscopy (DSOC) and to clinically validate the model through comparisons with DSOC expert and nonexpert endoscopists. METHODS: In this two-stage study, we first developed and validated CNN1. Then, we performed a multicenter diagnostic trial to compare four DSOC experts and nonexperts against an improved model (CNN2). Lesions were classified into neoplastic and non-neoplastic in accordance with Carlos Robles-Medranda (CRM) and Mendoza disaggregated criteria. The final diagnosis of neoplasia was based on histopathology and 12-month follow-up outcomes. RESULTS: In stage I, CNN2 achieved a mean average precision of 0.88, an intersection over the union value of 83.24 %, and a total loss of 0.0975. For clinical validation, a total of 170 videos from newly included patients were analyzed with the CNN2. Half of cases (50 %) had neoplastic lesions. This model achieved significant accuracy values for neoplastic diagnosis, with a 90.5 % sensitivity, 68.2 % specificity, and 74.0 % and 87.8 % positive and negative predictive values, respectively. The CNN2 model outperformed nonexpert #2 (area under the receiver operating characteristic curve [AUC]-CRM 0.657 vs. AUC-CNN2 0.794, P < 0.05; AUC-Mendoza 0.582 vs. AUC-CNN2 0.794, P < 0.05), nonexpert #4 (AUC-CRM 0.683 vs. AUC-CNN2 0.791, P < 0.05), and expert #4 (AUC-CRM 0.755 vs. AUC-CNN2 0.848, P < 0.05; AUC-Mendoza 0.753 vs. AUC-CNN2 0.848, P < 0.05). CONCLUSIONS: The proposed CNN model distinguished neoplastic bile duct lesions with good accuracy and outperformed two nonexpert and one expert endoscopist.

Predictors of adverse events after endoscopic ultrasound-guided through-the-needle biopsy of pancreatic cysts: a recursive partitioning analysis.

BACKGROUND AND STUDY AIMS : Endoscopic ultrasound-guided through-the-needle biopsy (TTNB) of pancreatic cystic lesions (PCLs) is associated with a non-negligible risk for adverse events (AEs). We aimed to identify the hierarchic interaction among independent predictors for TTNB-related AEs and to generate a prognostic model using recursive partitioning analysis (RPA). PATIENTS AND METHODS : Multicenter retrospective analysis of 506 patients with PCLs who underwent TTNB. RPA of predictors for AEs was performed and the model was validated by means of bootstrap resampling. RESULTS : Mean cysts size was 36.7 mm. Most common diagnoses were intraductal papillary mucinous neoplasm (IPMN, 45 %), serous cystadenoma (18.8 %), and mucinous cystadenoma (12.8 %). Fifty-eight (11.5 %) AEs were observed. At multivariate analysis, age (odds ratio [OR] 1.32, 1.09-2.14; p = 0.05), number of TTNB passes (OR from 2.17, 1.32-4.34 to OR 3.16, 2.03-6.34 with the increase of the number of passes), complete aspiration of the cyst (OR 0.56, 0.31-0.95; p = 0.02), and diagnosis of IPMN (OR 4.16, 2.27-7.69; p < 0.001) were found to be independent predictors of AEs, as confirmed by logistic regression and random forest analyses. RPA identified three risk classes: high-risk (IPMN sampled with multiple microforceps passes, 28 % AEs rate), low-risk (1.4 % AE rate, including patients < 64 years with other-than-IPMN diagnosis sampled with ≤ 2 microforceps passes and with complete aspiration of the cyst) and middle-risk class (6.1 % AEs rate, including the remaining patients). CONCLUSION : TTNB should be selectively used in the evaluation of patients with IPMN. The present model could be applied during patient selection as to optimize the benefit/risk of TTNB.

Cholangioscopy-based convoluted neuronal network vs. confocal laser endomicroscopy in identification of neoplastic biliary strictures.

Background and study aims Artificial intelligence (AI) models have demonstrated high diagnostic performance identifying neoplasia during digital single-operator cholangioscopy (DSOC). To date, there are no studies directly comparing AI vs. DSOC-guided probe-base confocal laser endomicroscopy (DSOC-pCLE). Thus, we aimed to compare the diagnostic accuracy of a DSOC-based AI model with DSOC-pCLE for identifying neoplasia in patients with indeterminate biliary strictures. Patients and methods This retrospective cohort-based diagnostic accuracy study included patients ≥ 18 years old who underwent DSOC and DSOC-pCLE (June 2014 to May 2022). Four methods were used to diagnose each patient's biliary structure, including DSOC direct visualization, DSOC-pCLE, an offline DSOC-based AI model analysis performed in DSOC recordings, and DSOC/pCLE-guided biopsies. The reference standard for neoplasia was a diagnosis based on further clinical evolution, imaging, or surgical specimen findings during a 12-month follow-up period. Results A total of 90 patients were included in the study. Eighty-six of 90 (95.5%) had neoplastic lesions including cholangiocarcinoma (98.8%) and tubulopapillary adenoma (1.2%). Four cases were inflammatory including two cases with chronic inflammation and two cases of primary sclerosing cholangitis. Compared with DSOC-AI, which obtained an area under the receiver operator curve (AUC) of 0.79, DSOC direct visualization had an AUC of 0.74 ( P = 0.763), DSOC-pCLE had an AUC of 0.72 ( P = 0.634), and DSOC- and pCLE-guided biopsy had an AUC of 0.83 ( P = 0.809). Conclusions The DSOC-AI model demonstrated an offline diagnostic performance similar to that of DSOC-pCLE, DSOC alone, and DSOC/pCLE-guided biopsies. Larger multicenter, prospective, head-to-head trials with a proportional sample among neoplastic and nonneoplastic cases are advisable to confirm the obtained results.

Correction: Outcomes of lumen apposing metal stent placement in patients with surgically altered anatomy: Multicenter international experience.

[This corrects the article DOI: 10.1055/a-2411-1814.].

Outcomes of lumen apposing metal stent placement in patients with surgically altered anatomy: Multicenter international experience.

Background and study aims Although outcomes of lumen-apposing metal stents (LAMS) placement in native anatomy have been reported, data on LAMS placement in surgically altered anatomy (SAA) are sparse. We aimed to assess outcomes of LAMS placement in patients with SAA for different indications. Patients and methods This was an international, multicenter, retrospective, observational study at 25 tertiary care centers through November 2023. Consecutive patients with SAA who underwent LAMS placement were included. The primary outcome was technical success defined as correct placement of LAMS. Secondary outcomes were clinical success and safety. Results Two hundred and seventy patients (125 males; average age 61 ± 15 years) underwent LAMS placement with SAA. Procedures included EUS-directed transgastric ERCP (EDGE) and EUS-directed transenteric ERCP (EDEE) (n = 82), EUS-guided entero-enterostomy (n = 81), EUS-guided biliary drainage (n = 57), EUS-guided drainage of peri-pancreatic fluid collections (n = 48), and EUS-guided pancreaticogastrostomy (n = 2). Most cases utilized AXIOS stents (n = 255) compared with SPAXUS stents (n = 15). Overall, technical success was 98%, clinical success was 97%, and the adverse event (AE) rate was 12%. Using AGREE classification, five events were rated as Grade II, 21 events as Grade IIIa, and six events as IIIb. No difference in AEs were noted among stent types ( P = 0.52). Conclusions This study shows that placement of LAMS is associated with high technical and clinical success rates in patients with SAA. However, the rate of AEs is noteworthy, and thus, these procedures should be performed by expert endoscopists at tertiary centers.

Spanish translation and validation of a hybrid, health-related quality of life questionnaire for chronic esophageal conditions: NEQOL-S.

BACKGROUND: Chronic esophageal conditions (CEC) are associated with significant disease-related burden, disability, and costs. Health-related quality of life (HRQOL) constructs are intended to capture the physical, mental, social, and emotional aspects of a patient's life and how health status impacts these domains. The Northwestern Esophageal Quality of Life (NEQOL) can be used among esophageal diseases while maintaining sensitivity to specific conditions. We aimed to translate, cross-cultural adapt, and validate the NEQOL into Spanish. METHODS: After language and cross-cultural adaptation, the NEQOL was applied to an outpatient clinic-based population in a single tertiary center. We analyzed the internal consistency, construct, criterion validity, and test-retest reliability of the questionnaire. The criterion validity was tested against the SF-12 questionnaire. KEY RESULTS: After completing the translation process, no item was considered problematic. A total of 385 patients were included in the validation study. The internal consistency (Cronbach's alpha) for the total NEQOL-S score was 0.89. The NEQOL-S questionnaire showed moderate test-retest reliability (ICC = 0.828; 95% CI 0.755-0.881; p < 0.001). Criterion validity showed good coherence when correlated with the SF-12 survey (R2  = 0.538; 95% CI 0.491-0.585, p < 0.001). CONCLUSIONS AND INFERENCES: The translated and cross-culturally adapted NEQOL-S showed good psychometric properties that allow its use in Spanish-speaking patients suffering from CEC.

Endoscopic Ultrasound-Guided Gallbladder Drainage Versus Percutaneous Drainage in Patients With Acute Cholecystitis Undergoing Elective Cholecystectomy.

INTRODUCTION: Cholecystectomy (CCY) is the gold standard treatment of acute cholecystitis (AC). Nonsurgical management of AC includes percutaneous transhepatic gallbladder drainage (PT-GBD) and endoscopic ultrasound-guided gallbladder drainage (EUS-GBD). This study aims to compare outcomes of patients who undergo CCY after having received EUS-GBD vs PT-GBD. METHODS: A multicenter international study was conducted in patients with AC who underwent EUS-GBD or PT-GBD, followed by an attempted CCY, between January 2018 and October 2021. Demographics, clinical characteristics, procedural details, postprocedure outcomes, and surgical details and outcomes were compared. RESULTS: One hundred thirty-nine patients were included: EUS-GBD in 46 patients (27% male, mean age 74 years) and PT-GBD in 93 patients (50% male, mean age 72 years). Surgical technical success was not significantly different between the 2 groups. In the EUS-GBD group, there was decreased operative time (84.2 vs 165.4 minutes, P < 0.00001), time to symptom resolution (4.2 vs 6.3 days, P = 0.005), and length of stay (5.4 vs 12.3 days, P = 0.001) compared with the PT-GBD group. There was no difference in the rate of conversion from laparoscopic to open CCY: 5 of 46 (11%) in the EUS-GBD arm and 18 of 93 (19%) in the PT-GBD group ( P value 0.2324). DISCUSSION: Patients who received EUS-GBD had a significantly shorter interval between gallbladder drainage and CCY, shorter surgical procedure times, and shorter length of stay for the CCY compared with those who received PT-GBD. EUS-GBD should be considered an acceptable modality for gallbladder drainage and should not preclude patients from eventual CCY.

Novel EUS-guided microwave ablation of an unresectable pancreatic neuroendocrine tumor.

Video 1EUS-guided microwave ablation of an unresectable pancreatic mass using a novel generator platform and a specialized 19.5-gauge needle antenna.

Impact of Antibiotic Prophylaxis on Infection Rate after Endoscopic Ultrasound Through-the-Needle Biopsy of Pancreatic Cysts: A Propensity Score-Matched Study.

BACKGROUND: Despite weak evidence, antibiotic prophylaxis prior to endoscopic ultrasound-guided through-the-needle biopsy (EUS-TTNB) of pancreatic cystic lesions (PCLs) is routinely used in clinical practice. We aim to compare a group of patients treated with antibiotics before EUS-TTNB of PCLs and a group who did not undergo antimicrobial prophylaxis. METHODS: Out of 236 patients with pancreatic cystic lesions referred to two high-volume centers between 2016 and 2021, after propensity score matching, two groups were compared: 98 subjects who underwent EUS-TTNB under antibiotic prophylaxis and 49 subjects without prophylaxis. RESULTS: There was no difference in terms of baseline parameters between groups. Final diagnosis was serous cystadenoma in 36.7% of patients in the group not treated with prophylaxis and in 37.7% of patients in the control group, whereas IPMN and mucinous cystadenoma were diagnosed in 3 (6.1%) and 16 (32.6%) versus 6 (6.1%) and 32 (32.6%) patients in the two groups, respectively (p = 0.23). Overall, the adverse event rate was 6.1% in the group not treated with antibiotic prophylaxis and 5.1% in the control group (p = 0.49). Only a single infectious adverse event occurred in each group (p = 0.48). The diagnostic yields were 89.7% and 90.8% in the two groups (p = 0.7), and the diagnostic accuracy rate was 81.6% in both groups (p = 1.0). CONCLUSIONS: Prophylactic antibiotics do not seem to influence the risk of infection, and their routine use should be discouraged.