Oral Presentations
2024; Wiley; Volume: 36; Issue: S1 Linguagem: Italiano
10.1111/den.14862
ISSN1443-1661
AutoresTo compare the diagnostic accuracy of an EUS-based AI model AIWorks-EUS, mdconsgroup, Guayaquil, Ecuador against that of expert and nonexpert endoscopists' visual impression, in the identification of normal anatomical structures during EUS procedures.,
ResumoDigestive EndoscopyVolume 36, Issue S1 p. 10-30 AbstractFree Access Oral Presentations First published: 02 July 2024 https://doi.org/10.1111/den.14862AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onEmailFacebookTwitterLinkedInRedditWechat THURSDAY, 4 JULY 2024 09:00–10:30 OPENING CEREMONY WITH BEST ABSTRACTS BAA1 COMPARATIVE EVALUATION OF ARTIFICIAL INTELLIGENCE VS ENDOSCOPISTS' DIAGNOSTIC ACCURACY IN ENDOSCOPIC ULTRASOUND FOR IDENTIFYING NORMAL ANATOMICAL STRUCTURES: A MULTI-INSTITUTIONAL, CROSS-SECTIONAL, PROSPECTIVE STUDY C. Robles-Medranda1, J. Baquerizo-Burgos1,2, M. Puga-Tejada1, I. Raijman3,4, T.M. Berzin5, J.A. Nebel6, J. Iglesias7, R. Kunda8, R. Del Valle1, J. Alcivar-Vasquez1, J.C. Mendez2, A. Chilan-Pincay2, M. Sanchez-Cepeda2, G. Lara1, V. Oregel-Aguilar9, I. Boston5, C. Pattni5, M. Egas-Izquierdo1, D. Cunto1, M. Arevalo-Mora1,10, H. Pitanga-Lukashok1, D. Tabacelia11,12 1Instituto Ecuatoriano de Enfermedades Digestivas (IECED), Guayaquil, Ecuador, 2mdconsgroup, Guayaquil, Ecuador, 3Houston Methodist Hospital, Houston, United States, 4Baylor Saint Luke's Medical Center, Houston, United States, 5Beth Israel Deaconess Medical Center and Harvard Medical School, Center of Advanced Endoscopy, Boston, United States, 6Gastrolife, Barra Life Medical Center, Rio de Janeiro, Brazil, 7Hospital Clínico Universitario. Universidad de Santiago de Compostela (USC), Santiago de Compostela, Spain, 8Universitair Ziekenhuis Brussel (UZB), Vrije Universiteit Brussel (VUB), Department of Surgery, Department of Gastroenterology-Hepatology, Department of Advanced Interventional Endoscopy, Brussels, Belgium, 9Hospital Civil Morelia, Morelia, Mexico, 10Larkin Community Hospital, South Miami, United States, 11Elias Emergency University Hospital, Bucharest, Romania, 12Carol Davila University of Medicine and Pharmacy, Bucharest, Romania AIMS: To compare the diagnostic accuracy of an EUS-based AI model (AIWorks-EUS, mdconsgroup, Guayaquil, Ecuador) against that of expert and nonexpert endoscopists' visual impression, in the identification of normal anatomical structures during EUS procedures. METHODS: A multi-institutional, cross-sectional, prospective study designed to assess the diagnostic accuracy of an EUS-based AI model (AIWorks-EUS, mdconsgroup, Guayaquil, Ecuador), in identifying normal anatomical structures through EUS, compared to that of expert and nonexpert endoscopists. Consecutive patients' videos of linear-array EUS were recorded and separated into individual clips of the mediastinal, gastric, or duodenal window. The video dataset was reviewed by nine endoscopists: 5/9 experts and 4/9 nonexperts. Their visual impression was compared to the observation outputs provided by the EUS-based AI model. RESULTS: A total of 39 videos of linear EUS array were obtained from 21 patients. Mean age 50 ± 20 years and 81% female. The diagnostic accuracy of the EUS-based AI model in terms of sensitivity, specificity, positive (PPV) and negative predictive value (NPV), and observed agreement (OA) were 95.2%, 99.3%, 97.9%, 98.4%, and 98.2%, respectively. The EUS-based AI model obtained a significant higher diagnostic accuracy than experts (OA 94%; P < 0.001) or nonexperts (OA 90%; P < 0.001). When analyzing per-structure OA obtained by the EUS-based AI model, experts and nonexperts, it noticed a 19% difference between the structure with the highest (aorta; 100%) and the lowest OA (ampulla; 81%). CONCLUSIONS: The EUS-based AI model achieved higher diagnostic accuracy than expert and nonexpert endoscopists in the identification of normal anatomical structures. This model provides endoscopic assistance during live procedures, regardless endoscopic experience. Further studies must evaluate this EUS-based AI model application in EUS training programs. DISCLOSURE: Carlos Robles-Medranda is a key opinion leader and consultant for Pentax Medical, Steris, Medtronic, Motus, Micro-tech, G-Tech Medical Supply, CREO Medical, and mdconsgroup. Issac Raijman is a speaker for Boston Scientific, ConMed, Medtronic, and GI Supplies; advisory board member for Micro-Tech; co-owner of EndoRx. Tyler M. Berzin is a consultant for Medtronic, Boston Scientific, Wision, A.I., Magentiq Eye, and RSIP Vision. The other authors declare no conflicts of interest. BAA2 SIMULTANEOUS DRAINAGE VS STEP-UP THERAPY FOR DRAINAGE OF MULTIPLE INFECTED PANCREATIC NECROTIC COLLECTIONS IN PATIENTS WITH ACUTE NECROTIZING PANCREATITIS: A RANDOMIZED TRIAL B. Venkat Siddharda1, J. Shah1, P. Gupta2, H. Bhujade2, Y.R. Sakaray3, S. Ancil1, V. Jearth1, A.K. Singh1, H.S. Mandavdhare1, V. Sharma1, S. Rana1, U. Dutta1 1Postgraduate Institute of Medical Education and Research, Gastroenterology, Chandigarh, India, 2Postgraduate Institute of Medical Education and Research, Radiodiagnosis and Imaging, Chandigarh, India, 3Postgraduate Institute of Medical Education and Research, General Surgery, Chandigarh, India AIMS: Minimally invasive step-up approach is the recommended modality of treatment in patients with infected pancreatic necrosis(IPN). Majority of patients with pancreatic collections are having a single IPN, however, some patients with necrotizing pancreatitis have multiple IPNs. In such scenario, whether drainage of all collections simultaneously has better clinical outcome compared to draining collections sequentially as per clinical response has not been explored earlier. We performed a single center, open label, randomized trial to compare sequential drainage vs simultaneous drainage of multiple IPNs in patients with acute necrotizing pancreatitis(ANP). METHODS: All consecutive patients of acute pancreatitis with multiple confirmed or clinically suspected IPN were screened for inclusion criteria. In simultaneous group(group A), all independent collections were intervened simultaneously using either endoscopic or percutaneous approach depending on feasibility. In sequential group(group B), only collection with larger size or gas configuration was intervened. Additional interventions in either group were done as per pre-defined clinical criteria. Primary outcome was the score on Comprehensive Complication Index(CCI) till clinical success. Secondary outcomes were number of interventions required for clinical success, new onset organ failure, major disease/procedure related complications and mortality.(CTRI/2022/07/043878). RESULTS: 60 patients with multiple IPNs were enrolled (29 patients in group A and 31 in group B). All patients were having an ongoing SIRS and 66.6%(n = 40) patients were having an ongoing organ failure. Mean CCI was 72.48 ± 28.28 in group A and 64.43 ± 34.91 in group B(P = 0.332). Total interventions were lower in group B(4.55 ± 2.21 vs 3.23 ± 2.14; P = 0.022). Development of new onset organ failure(34.5% vs 38.7%; P = 0.734), requirement of surgical intervention(27.6% vs 22.6%; P = 0.655) and mortality(41.3% vs 38.7%; P = 0.833)were equal amongst both groups. CONCLUSIONS: Sequential drainage tailored according to the clinical response has equivalent clinical outcome with fewer requirement of interventions compared to simultaneous drainage of all collections in patients with multiple IPNs. BAA3 ARTIFICIAL INTELLIGENCE IN COLONOSCOPY: EVALUATING BENEFITS AND HARMS OF OPTICAL DIAGNOSIS IN A MULTICENTER CLINICAL STUDY Y. Mori1, N. Halvorsen1, I. Barua1, S. Gulati2, M. Misawa3, K. Mori4, B. Hayee2, O. Olabintan2, J.A. Nilsen5, S.O. Frigstad5, J. East6, A. Rastogi7, C. Hassan8, M. Kalager1, M. Løberg1, Ø. Holme1, A. Haji2, M. Bretthauer1 1University of Oslo, Oslo, Norway, 2King's College Hospital, London, United Kingdom, 3Showa University Northern Yokohama Hospital, Yokohama, Japan, 4Nagoya University, Nagoya, Japan, 5Baerum Hospital, Oslo, Norway, 6University of Oxford, Oxford, United Kingdom, 7Kansas University, Kansas city, United States, 8Humanitas University, Milan, Italy AIMS: Artificial intelligence enables real-time optical distinction between neoplastic versus non-neoplastic colonic polyps with computer-aided diagnosis (CADx). We quantified how this novel technology affects patient burden and healthcare costs and resources. METHODS: We analyzed data from a clinical trial investigating CADx-assisted optical diagnosis for diminutive polyps (<=5 mm). We simulated how the leave-in-situ optical diagnosis strategy would change colonoscopy-related cost, surveillance interval agreement, and untreated neoplasm as compared with the standard care (i.e., remove-all-polyp). Leave-in-situ is defined as diminutive rectosigmoid polyps predicted as non-neoplastic are not removed. We used public reimbursement rates and clinical guidelines in the UK for analysis. RESULTS: We analyzed 1,134 patients (59% male, median age 67 years). The CADx-assisted leave-in-situ optical diagnosis strategy reduced the number of histopathological assessments by 25.8% (95% confidence interval: 23.8% to 28.0%). This decreased the average colonoscopy-related cost by 17 £. Conversely, optical diagnosis resulted in 0.02 (0.01 to 0.03) neoplastic polyps per patient not removed due to misdiagnosis, corresponding to 2.1% of all included neoplastic polyps. Disagreement in surveillance interval prediction was found in 0.2% (0.0% to 0.6%). CONCLUSIONS: CADx-assisted optical diagnosis could save colonoscopy-related costs with a substantial reduction of histopathological assessments. Deviation from surveillance interval recommendations and left neoplasm appeared marginal. DISCLOSURE: Olympus (consulting, lecture fee, and device loan). Cybernet (license fee) SATURDAY, 6 JULY 2024 14:00–15:00 BEST ABSTRACTS SESSION BOP1 POLYSACCHARIDE HEMOSTATIC POWDER VERSUS HEMOCLIPPING IN MANAGING POST-ENDOSCOPIC SPHINCTEROTOMY BLEEDING: A PILOT RANDOMIZED CLINICAL TRIAL J. Zuo, H. Li, P. Li, S. Zhang Beijing Friendship Hospital, Capital Medical University, Beijing, China AIMS: Hemoclipping has been considered an effective approach for managing post-endoscopic sphincterotomy (EST) bleeding, however, endoscopists often encounter obstacles when dealing with post-EST bleeding using a side-viewing duodenoscope. In recent years, polysaccharide hemostatic powder (PHP) has demonstrated promising outcomes in handling gastrointestinal bleeding. Our objective is to assess the clinical efficacy of PHP in patients following EST. METHODS: Forty adult patients in our center presenting with significant oozing bleeding under endoscopy after EST were randomized 1:1 to PHP or hemoclipping. The efficacy of achieving initial hemostasis, delayed bleeding rate, and other major complication were analyzed between two group. RESULTS: Immediate hemostasis was achieved in 100% of patients treated initially with PHP versus 90% in the hemoclipping group (P = 0.371). In the PHP group, 3 cases had medium or large incisions, while the hemoclipping group had 4 cases. Immediate hemostasis failed in 1 case of medium incision and 1 case of large incision in the hemoclipping group, successfully managed with PHP as rescue therapy. The mean procedural time was significantly shorter for the PHP (P = 0.002). No delayed bleeding was observed in both groups. Two cases of intraoperative duct blockage occurred in the PHP group. No significant difference in the rates of post-ERCP pancreatitis between the two groups. PHP n (%) Hemoclipping n (%) P-value Total patients 20 20 Na Immediate hemostasis achieved 20 18 0.371 Immediate hemostasis achieved after crossover Na 2 Na Delayed bleeding 0 0 Na Mean hemostasis time (s) 66.6 ± 38.0 125.0 ± 39.8 0.002 Post-ERCP Pancreatitis 1 2 0.840 CONCLUSIONS: PHP is a promising modality for managing post-EST bleeding in achieving easier hemostasis for immediate bleeding and preventing delayed bleeding. BOP2 LONG-TERM SURVIVAL AFTER ENDOSCOPIC RESECTION FOR GASTRIC CANCER: REAL-WORLD EVIDENCE FROM A NATIONWIDE MULTICENTER PROSPECTIVE COHORT STUDY IN JAPAN H. Suzuki1,2, T. Hirasawa3, H. Ono4, Y. Takeuchi5,6, K. Ishido7, S. Hoteya8, T. Yano9, S. Tanaka10, M. Endo11, M. Nakagawa12, T. Toyonaga13, H. Doyama14, K. Hirasawa15, T. Shimazu16, K. Takizawa4, S. Tanabe17,18, H. Kondo19, H. Iishi5,20, M. Ninomiya21, I. Oda1 1National Cancer Center Hospital, Endoscopy Division, Tokyo, Japan, 2Nihon University School of Medicine, Division of Gastroenterology and Hepatology, Department of Medicine, Tokyo, Japan, 3Cancer Institute Hospital, Department of Gastroenterology, Tokyo, Japan, 4Shizuoka Cancer Center, Division of Endoscopy, Shizuoka, Japan, 5Osaka International Cancer Institute, Department of Gastrointestinal Oncology, Osaka, Japan, 6Gunma University Hospital, Department of Endoscopy and Endoscopic Surgery, Maebashi, Japan, 7Kitasato University School of Medicine, Department of Gastroenterology, Kanagawa, Japan, 8Toranomon Hospital, Department of Gastroenterology, Tokyo, Japan, 9National Cancer Center Hospital East, Department of Gastroenterology and Endoscopy, Chiba, Japan, 10Hiroshima University Hospital, Department of Endoscopy, Hiroshima, Japan, 11School of Medicine, Iwate Medical University, Division of Gastroenterology, Department of Internal Medicine, Iwate, Japan, 12Hiroshima City Hiroshima Citizens Hospital, Department of Endoscopy, Hiroshima, Japan, 13Kobe University Hospital, Department of Endoscopy, Hyogo, Japan, 14Ishikawa Prefectural Central Hospital, Department of Gastroenterology, Ishikawa, Japan, 15Yokohama City University Medical Center, Division of Endoscopy, Kanagawa, Japan, 16National Cancer Center, Behavioral Science Division, Center for Public Health Sciences, Tokyo, Japan, 17Kitasato University School of Medicine, Department of Advanced Medicine, Research and Development Center for New Medical Frontiers, Kanagawa, Japan, 18Ebina General Hospital, Department of Gastroenterology, Ebina, Japan, 19Tonan Hospital, Center for Digestive Diseases, Hokkaido, Japan, 20Itami City Hospital, Department of Gastroenterology, Hyogo, Japan, 21Yuuai Medical Center, Digestive Disease Center, Okinawa, Japan AIMS: We aimed to clarify the long-term outcomes of endoscopic resection (ER) for early gastric cancers (EGCs) based on pathological curability in a multicenter prospective cohort study in Japan. METHODS: We analyzed the long-term outcomes of 9054 patients with 10,021 EGCs undergoing ER between July 2010 and June 2012. Primary endpoint was the 5-year overall survival (OS). The hazard ratio for all-cause mortality was calculated using the Cox proportional hazards model. Pathological curability was categorized into en bloc resection, negative margins, and negative lymphovascular invasion: differentiated-type, pT1a, ulcer negative, ≤2 cm (Category A1); differentiated-type, pT1a, ulcer negative, >2 cm or ulcer positive, ≤3 cm (Category A2); undifferentiated-type, pT1a, ulcer negative, ≤2 cm (Category A3); differentiated-type, pT1b (SM1), ≤3 cm (Category B); or noncurative resections (Category C). RESULTS: Overall, the 5-year OS was 89.0% (95% CI, 88.3%–89.6%). In a multivariate analysis, no significant differences were observed when the hazard ratio of Categories A2, A3, and B were compared with that of A1. In addition, gastric cancer-related death of Category A1, Category A2, Category A3, Category B and Category C were found in 5 (0.1%), 5 (0.2%), 0 (0%), 0 (0%) and 30 patients (1.7%), respectively. Among the 10 patients of gastric cancer-related deaths in Category A1 or A2, 7 patients died because of metachronous gastric cancers (n = 6) or synchronous gastric cancer (n = 1). The remaining 3 patients died of gastric cancer lesions treated by ER. In contrast, among the 30 patients of gastric cancer-related deaths in Category C, one patient died because of metachronous gastric cancers. The remaining 29 patients died of gastric cancer lesions treated by ER. CONCLUSIONS: ER can be recommended as a standard treatment for patients with EGCs fulfilling Category A2, A3, and B, as well as A1. BOP3 THE EFFICACY OF PREVENTIVE DEFECT CLOSURE AFTER ENDOSCOPIC PAPILLECTOMY E. Khon, I. Nedoluzhko, K. Shishin A.S. Loginov Moscow Clinical Scientific Center, Department of Health of Moscow, Operative Endoscopy, Moscow, Russia AIMS: To decrease the postoperative complication rate after endoscopic papillectomy. METHODS: The study involved 59 patients who underwent endoscopic papillectomy in our institution. 2 groups were retrospectively formed - Group I (n = 32) patients who underwent standard endoscopic papillectomy. Group II (n = 27) patients who underwent modified endoscopic papillectomy with the closure of the defect with haemostatic clips. The closure of the defect was performed after standard snare papillectomy with subsequent pancreatic stent placement. Resection defect was closed with haemostatic clips leaving 2 mm between the last clip and the stented pancreatic duct. Retrospective analysis was performed in order to evaluate the effectiveness of the method. RESULTS: Delayed bleeding rate was 25% (n = 8) in the Group I and 3.7 % (n = 1) in the Group II (P = 0.018). The source of the bleeding in the Group II was located closely to the common bile duct orifice - the area which was not closed with the clips to save the bile outflow. Perforation rate was 9.4% (n = 3) in the Group I and 0% in the Group II (P = 0.092). The pancreatitis rate was 9.4% (n = 3) in the Group I and 11.1% (n = 3) in the Group II (P = 0.827). The average period of hospital stay was 4.0 in the Group I and 2.0 in the Group II (P = <0.01). CONCLUSIONS: The preventive defect closure after endoscopic papillectomy is safe, effective and accessible measure to reduce the rate of the delayed bleeding and perforation without increasing the risks of postpapillectomy pancreatitis. The method significantly reduces the period of hospital stay (P = <0.01). BOP4 CLINICAL OUTCOMES OF ENDOSCOPIC RESECTION IN PATIENTS WITH COLORECTAL T1 CANCER ≤ 15 MM; A MULTICENTER RETROSPECTIVE REAL-WORLD STUDY H. Tanaka1, S. Kuribayashi1, Y. Fukai2, K. Takahashi3, K. Furuya4, M. Sekiguchi5, Y. Takeuchi1, T. Uraoka1 1Gunma University Graduate School of Medicine, Department of Gastroenterology and Hepatology, Maebashi, Japan, 2Maebashi Red Cross Hospital, Department of Gastroenterology, Maebashi, Japan, 3Haramachi Red Cross Hospital, Internal Medicine, Higashi Agatsuma, Japan, 4National Hospital Organization Shibukawa Medical Center, Department of Gastroenterology, Shibukawa, Japan, 5Isesaki Municipal Hospital, Internal Medicine, Isesaki, Japan AIMS: Endoscopic submucosal dissection (ESD) has been known to achieve a higher en-bloc resection rate for T1 cancer in the colorectum. In contrast, endoscopic mucosal resection (EMR) is the standard procedure for the resection of small and medium-sized colorectal tumors. However, it is not clear whether endoscopic mucosal resection (EMR) is still the first choice of resection procedure for such T1 cancers. This study aims to investigate the treatment outcomes of EMR for colorectal T1 cancers ≤ 15mm. METHODS: This multicenter retrospective cohort study enrolled 102 patients with 102 colorectal T1 cancers (≤ 15mm, excluding 0-Ip) who underwent EMR or polypectomy (EMR group: 93 cases) and ESD (ESD group: 9 cases) between April 2009 and August 2019 at five institutions. We investigated the outcomes of colorectal endoscopic resection. RESULTS: The median age was 69 years in both groups. En-bloc and curative resection rates showed no significant differences between the two groups (EMR group: 91.3%, 32.3%, and ESD group: 88.9% 11.1%, respectively). Among the 71 cases of non-curative resection (non-Cura), 35 cases (29 EMR and 6 ESD) underwent additional surgical resection (non-Cura+OPE), while 36 cases (34 EMR and 2 ESD) were observed due to age or comorbidities (non-Cura+Obs). Median follow-up periods were 2.2 and 1.3 years (EMR and ESD groups, respectively) in all cases. There was no recurrence in Cura and non-Cura+OPE cases. However, one recurrence case was found in the non-Cura+Obs in the EMR group. This recurrent case which had a positive vertical margin and vascular invasion, had been under observation due to compromised cardiac function. After the recurrence, the patient underwent surgical intervention and remained recurrence-free for 3 years after the operation. CONCLUSIONS: This study suggested that EMR and polypectomy are still standard procedures for T1 cancers ≤ 15mm in clinical practice and are considered acceptable regarding patient burden and healthcare costs. BOP5 ADVANCING SMALL COLORECTAL POLYP CLASSIFICATION THROUGH CADX: ENHANCING PATHOLOGICAL TRIAGE C.-Y. Kuo1, C.-T. Shun2, N.N. Yeh3, W.-F. Hsu4, W.-Y. Chang4, H.-H. Lin5, L.-C. Chang4, H.-M. Chiu4 1Fu Jen Catholic University Hospital, Fu Jen Catholic University, Department of Internal Medicine, New Taipei City, Taiwan, 2National Taiwan University Hospital, Department of Pathology, Taipei, Taiwan, 3aetherAI Co., Ltd., Taipei, Taiwan, 4National Taiwan University Hospital, Department of Internal Medicine, Taipei, Taiwan, 5National Taiwan University Hospital Hsinchu Branch, Department of Internal Medicine, Hsinchu, Taiwan AIMS: To reduce the workload of pathologists burdened by the high volume of small, low-risk colorectal polyps (Hsu et al., J Gastroenterol Hepatol, 2020), this study develops a deep neural network (DNN)-based Computer-Aided Diagnosis (CADx) system to distinguish between non-neoplastic, conventional adenoma, and sessile serrated lesions (SSLs) in colorectal polyp pathology. METHODS: Within the TACOS trial framework (Chang et al., Ann Intern Med, 2023) (July 2018-July 2020), 979 whole slide images (WSIs) of polyp specimens from National Taiwan University Hospital were used to train (654 slides), validate (120 slides), and test (205 slides) the model in a 7:1:2 ratio. Each WSI was categorized by a single GI pathologist into non-neoplastic, conventional adenoma, or SSL. The ResNet-50 GMP with fixup initialization was the DNN framework used, processing WSIs at ×5 magnification. After 20 epochs of training, the model with the best performance based on validation results was selected for internal testing. RESULTS: The CADx model achieved 93.7% accuracy and 0.945 AUC for non-neoplastic polyps, 94.1% accuracy and 0.988 AUC for adenomatous polyps, and 94.6% accuracy and 0.974 AUC for SSLs in internal testing of 205 slides. Table Performance of the Deep Neural Network Model in Classifying Colorectal Polyps on Internal Test Sets Polyp types of internal test set AUC Accuracy, % Sensitivity, % Specificity, % PPV, % NPV, % Non-neoplastic (n = 15) 0.945 93.7 66.7 95.8 55.6 97.3 Adenomatous (n = 168) 0.988 94.1 94.0 94.6 98.8 77.8 Sessile serrated lesion (n = 22) 0.974 94.6 86.4 95.6 70.4 98.3 CONCLUSIONS: The DNN model shows significant promise in colorectal polyp classification, poised to streamline pathological diagnostics and reduce pathologists' workload. Future research will focus on external validation, comparing its efficacy against GI expert and non-expert pathologists to confirm its clinical applicability, identification of advanced histology, and judgment of completeness of resection. BOP6 EUS GUIDED LIVER BIOPSY VERSUS PERCUTANEOUS LIVER BIOPSY FOR THE EVALUATION OF LIVER DISEASES, A PROSPECTIVE RANDOMIZED STUDY FROM A TERTIARY HOSPITAL IN INDIA H. Raina1, A. Prakash1, R. Sodani2 1Ivy Hospital and Reasearch Institute, Gastroenterology and Hepatology, Amritsar, India, 2Ivy Hospital and Reasearch Institute, Radiology, Amritsar, India AIMS: Liver biopsy (LB) has historically been performed percutaneously. A newer method of obtaining an LB is by endoscopic ultrasound (EUS)guidance. To compare the safety and efficacy of this new method with the standard technique. METHODS: It was a prospective randomized study done from July 2018 to August, 2023. All adult patients with undiagnosed abnormal liver enzymes who consented were included and randomized into two groups. The EUS-guided liver biopsy (EUS-LB) procedure was done using 19 G Shark Core FNB needle (Medtronic). Two to three passes were taken. Percutaneous liver biopsy (PC-LB) was done by a single experienced radiologist under USG guidance with 18 Guage CorVocet needle (Meritmedica). RESULTS: 82 Patients in EUS-LB and 86 Patients in PC-LB group were analysed. Mean total core length (TCL) in EUS-LB = 20.9 mm (IQR 15.3–26.5); PC-LB = 22.6 mm (IQR 19.0-27.2) P = 0.4. Number of complete portal tracts (CPT) in EUS-LB = 24 (17–31) and in PC-LB = 26 (15–36); (P = 0.20).). Histologic diagnosis = 97.4% in the EUS-LB group, 98.7% in the PC-LB group (P = 0.541). Fragmented specimens were seen in 23.17% and 10.46 % in EUS-LB and PC -LB groups. Mean Hospital stay in EUS-LB = 2hrs (1–3 h) vs. 3.9 h (3–6 h) in PS-LB (P = 0.02). Pain requiring analgesics = 3.65% in EUS-LB and 87.20% in PC-LB group (P = 0.0005).Patient satisfaction and willing to repeat the procedure in EUS-LB and PC-LB was seen in 85.36% and 61.62% (P = 0.01), respectively. CONCLUSIONS: EUS-LB was safer, with less hospital stay, no pain and more patient satisfaction than PC-LB, with comparable efficacy. FRIDAY, 5 JULY 2024 09:00–10:30 FREE PAPER SESSION (PB) O001 EUS-GUIDED GASTROENTEROSTOMY VS. ENTERAL STENTING FOR MALIGNANT GASTRIC OUTLET OBSTRUCTION: AN UPDATED META-ANALYSIS WITH RANDOMIZED AND MATCHED-CONTROL SUBGROUP ANALYSIS M.V. Fernandes1, V. Antunes1, N.J. Milioli2, T.L. Correa3, O.C. Martins4, C.F. de Mesquita5, S. Baraldo6 1Federal University of Health Sciences of Porto Alegre (UFCSPA), Porto Alegre, Brazil, 2Pontifical Catholic University of Rio Grande do Sul, Porto Alegre, Brazil, 3Beth Israel Deaconess Medical Center, Boston, United States, 4Federal University of Juiz de Fora, Juiz de Fora, Brazil, 5Federal University of Pernambuco, Recife, Brazil, 6Barretos Cancer Hospital, Endoscopy, Barretos, Brazil AIMS: To compare the efficacy and safety of endoscopic ultrasound-guided gastroenterostomy (EUS-GE) versus endoscopic stenting (ES) in malignant gastric outlet obstruction (GOO). METHODS: We performed a systematic review and meta-analysis using the PubMed, Embase, and Cochrane databases. We searched for studies comparing EUS-GE with ES in patients with malignant GOO and reporting at least one of the outcomes of interest. The primary outcomes assessed were clinical and technical success. The odds ratio (OR) was used for binary outcomes and the mean difference (MD) for continuous outcomes with their respective 95% confidence interval (CI). Heterogeneity was assessed using the Cochran Q test and I2 statistics. RESULTS: This study included one randomized controlled trial (RCT), two propensity-matched studies, and five retrospective studies, amounting to 1,088 patients. There was a statistically significant difference favoring EUS-GE in the clinical success (OR 3.13; 95% CI 1.95 to 5.00, P < 0.001; I2 = 0%), the recurrence of symptoms (OR 0.06; 95% CI 0.02 to 0.20, P < 0.001; I2 = 0%), the need for reintervention (OR 0.09; 95% CI 0.04 to 0.23, P < 0.001; I2 = 0%) and the adverse events rate (OR 0.44; 95% CI 0.23 to 0.83; P = 0.012; I2 = 68%). CONCLUSIONS: EUS-GE is a reasonable option in the treatment of malignant GOO with higher clinical success, fewer recurrences of symptoms, a lower re-intervention rate and fewer AEs. Standardization and wider dissemination of the EUS-GE technique are required to reduce technical challenges. O002 ENDOSCOPIC ULTRASOUND-GUIDED FINE NEEDLE BIOPSY WITH MACROSCOPIC ON-SITE EVALUATION VS FINE NEEDLE ASPIRATION WITH RAPID ON-SITE EVALUATION FOR SOLID PANCREATIC LESIONS: A MULTI-CENTRED PROSPECTIVE RANDOMIZED CONTROLLED TRIAL (MORE TRIAL) C. Chong1, S. Lakhtakia2, C.-S. Chung3, T.L. Ang4, S.-C. Liao5, Y.-t. Kuo6, H.-P. Wang6, N. Pausawasdi7, K. Hara8, A. Leow9, J.-H. Chen10, H. Leung11, A. Kwek4, S. Haba8, A. Chan11, R.S. Tang12, N. Nguyen13 1Prince of Wales Hospital, The Chinese University of Hong Kong, Department of Surgery, Hong Kong, Hong Kong, SAR of China, 2Asian Institute of Gastroenterology, Department of Gastroenterology, India, India, 3Far Eastern Memorial Hospital, Division of Gastroenterology and Hepatology, Department of Internal Medicine, Taipei, Taiwan, 4Changi General Hospital, Department of Gastroenterology and Hepatology, Singapore, Singapore, 5Taichung Veterans General Hospital, Department of Internal Medicine, Division of Gastroenterology, Taichung, Taiwan, 6National Taiwan University College of Medicine and Hospital, Department of Internal Medicine, Taipei, Taiwan, 7Faculty of Medicine Siriraj Hospital, Mahidol University, Division of Gastroenterology, Department of Medicine, Bangkok, Thailand, 8Aichi Cancer Center, Department of Gastroenterology, Nagoya, Japan, 9University of Malaya Medical Centre, University of Malaya, Department of Medicine, Faculty of Medicine, Malaysia, Malaysia, 10Taipei Tzu Chi Hospital, Department of Internal Medicine, Taipei, Taiwan, 11Prince of Wales Hospital, The Chinese University of Hong Kong, Department of Anatomical and Cellular Pathology, Hong Kong, Hong Kong, SAR of China, 12Prince of Wales Hospital, The Chinese University of Hong Kong, Department of Medicine and Therapeutics, Hong Kong, Hong Kong, SAR of China, 13Royal Adelaide Hospital, Department of Gastroenterology and Hepatology, Adelaide, Australia AIMS: Endoscopic ultrasound (EUS)-guided tissue is an invaluable tool in diagnosis of pancreatic mass lesion. EUS-FNA with ROSE is the gold standard for EUS-FN
Referência(s)