Gastrointestinal endoscopy最新文献

Background and aims: Gastric variceal bleeding occurs less commonly than bleeding from esophageal varices (EVs), although it is associated with higher morbidity and mortality. Bleeding from gastroesophageal varices type 1 (GOV1) is treated like EVs. In contrast, other forms of gastric variceal bleeding, including gastroesophageal varices type 2 (GOV2) and isolated gastric varices types 1 (IGV1) and 2 (IGV2), are treated with varying endoscopic approaches. Nonendoscopic methods include transjugular intrahepatic portosystemic shunt (TIPS) or balloon-occluded retrograde transvenous obliteration (BRTO). This technology report focuses on endoscopic management of gastric varices (GVs).

Methods: The MEDLINE database was searched through August 2022 for relevant articles by using key words such as gastric varices, glue, cyanoacrylate, thrombin, sclerosing agents, band ligation, topical hemostatic spray, coils, EUS, TIPS, and BRTO. The article was drafted, reviewed, and edited by the American Society for Gastrointestinal Endoscopy (ASGE) Technology Committee and approved by the Governing Board of the ASGE.

Results: Endoscopic injection with cyanoacrylate (CYA) glue has been the primary endoscopic method to treat GVs. EUS-guided angiotherapy with CYA glue and coil embolization has emerged as an alternative method enabling improved detection of GVs with a high technical success for targeting and obliterating GVs. Combining CYA glue with coil therapy allows the coil to act as a scaffold for the glue, reducing the risk of glue embolization and improving outcomes. Alternative injectates or topical treatments have been described but remain poorly studied.

Conclusions: The mainstay paradigm for the endoscopic management of gastric variceal bleeding is the injection of CYA glue. The published success of EUS-guided angiotherapy using CYA glue with or without embolization coils has increased our treatment armamentarium.

Background and aims: Liver fibrosis staging is challenging in patients with obesity and metabolic dysfunction-associated steatotic liver disease (MASLD). Liver biopsies are invasive, whereas non-invasive tests such as vibration-controlled transient elastography (VCTE) can be inaccurate in patients with obesity. We hypothesized that endoscopic ultrasound shear wave elastography (EUS-SWE) is more accurate for liver fibrosis staging in patients with MASLD and obesity, and in this pilot study we aimed to test this hypothesis and establish optimal fibrosis stage cutoffs for EUS-SWE.

Methods: This was a multicenter, cross-sectional study from prospectively collected data. Consecutive patients who underwent EUS-SWE with subsequent liver biopsy were included. EUS-SWE was compared to Fibrosis-4 Index (FIB-4) and VCTE. Area under the receiver operator characteristic (AUROC) curve analysis was performed, and 90% sensitivity and specific cutoffs were calculated to determine optimal cutoffs.

Results: 62 patients were included. Mean body mass index was 40.74kg/m². EUS-SWE was superior to FIB-4 in discriminating significant fibrosis (F2; AUROC 0.87 vs 0.61, p<0.0048) and advanced fibrosis (F3; AUROC 0.93 vs 0.63 p<0.0001), but not cirrhosis (F4; AUROC 0.95 vs 0.81, p=0.099). EUS-SWE was superior to VCTE in predicting advanced fibrosis and cirrhosis (p=0.0067 and 0.0022 respectively). 90% sensitivity cutoffs for EUS-SWE were 7.50, 8.48, and 11.30 for F2-F4 respectively, and 90% specificity cutoffs were 9.82, 10.20, and 14.60 respectively.

Conclusions: In this pilot study, EUS-SWE was superior to FIB-4 and VCTE for liver fibrosis staging in patients with MASLD and obesity (Clinical trial registration number: NCT05728697).

Background and aims: Accurate reporting of polyp characteristics is crucial for effective resource allocation in endoscopic resection referrals. We present our experience with pre-referral management of challenging colorectal lesions.

Methods: We reviewed a prospectively collected database of consecutive referrals for endoscopic resection of challenging colorectal lesions. The database included details of prior colonoscopies. We assessed pre-referral management using established guidelines.

Results: Among 1,826 referred lesions in 1826 patients, size estimates were missing for 421 (24%) lesions; 56 (3.2%) were found twice as large as previously estimated, while 65 (3.7%) were half the previous estimate. Morphological descriptions were absent for 376 (22%) polyps. Tattooing was performed for 822 (47%) lesions, with 247 (30%) placed correctly. Of the 1,103 (77%) polyps biopsied, only 11 (1.1%) were classified as high-risk by the referring endoscopist.

Conclusions: Errors in pre-referral management of challenging colorectal lesions are frequent. Improved adherence to national guidelines could enhance practice.

Background and aims: Adenoma detection rate (ADR) is a key quality metric in colonoscopy, reflecting the ability to detect adenomas. However, concerns remain regarding the robustness of ADR as a benchmark. In particular, "one and done" phenomenon may exists where physicians are less motivated to find additional adenoma following the first adenoma. To investigate this further, we aim to understand the relationship between single and multiple adenoma detection.

Methods: We conducted a retrospective cohort study using data from 89,587 screening colonoscopies performed by 32 endoscopists within the Sutter Health-Palo Alto Medical Foundation between 2015 and 2020. We analyzed ADR, single adenoma detection rate (ADR1), and multiple adenoma detection rate (ADR2+) using multivariate logistic regression and linear regression models. Endoscopists were then fit into 4 categories based on median ADR1 and ADR2+ (low ADR1, high ADR2+ ["all or none"], low ADR1, low ADR2+ ["none and done"], high ADR1, high ADR2+ ["all and done"], high ADR1, low ADR2+ ["one and done"].

Results: The overall ADR was 45.4%. ADR, ADR1, and ADR2+ were significantly associated with similar factors including older age, male sex, higher body mass index, smoking status, high-risk colonoscopies, Medicare insured, use of mucosal assist devices, longer withdrawal times, adequate preparation, and procedures performed by female, high-volume endoscopists with a longer duration in practice. The median ADR1 and ADR2+ were 23.95% and 21.29%, respectively. ADR1 and ADR2+ were positively correlated (Pearson correlation coefficient (r), 0.701, p<0.001). Only 4 of 32 endoscopists fit our "one and done" category.

Conclusion: Our findings suggest that ADR1 correlates with ADR2+. Despite concerns, the "one and done" phenomenon is not commonly seen in clinical practice. ADR remains a good surrogate marker for multiple adenoma detection.