No abstract

No abstract

Background & aim: Although acute lower GI bleeding (LGIB) represents a significant healthcare burden, prospective real-life data on management and outcomes are scanty. Present multicentre, prospective cohort study was aimed at evaluating mortality and associated risk factors and at describing patient management. Methods: Adult outpatients acutely admitted for or developing LGIB during hospitalization were consecutively enrolled in 15 high-volume referral centers. Demographics, comorbidities, medications, interventions and outcomes were recorded. Results: Overall 1,198 patients (1060 new admissions;138 inpatients) were included. Most patients were elderly (mean-age 74 ±15 years), 31% had a Charlson-Comorbidity-Index ≥3, 58% were on antithrombotic therapy. In-hospital mortality (primary outcome) was 3.4% (95%CI 2.5-4.6). At logistic regression analysis, independent predictors of mortality were increasing age, comorbidity, inpatient status, hemodynamic instability at presentation, and ICU-admission. Colonoscopy had a 78.8% diagnostic yield, with significantly higher hemostasis rate when performed within 24-hours than later (21.3% vs.10.8%, p = 0.027). Endoscopic hemostasis was associated with neither in-hospital mortality nor rebleeding. A definite or presumptive source of bleeding was disclosed in 90.4% of investigated patients. Conclusion: Mortality in LGIB patients is mainly related to age and comorbidities. Although early colonoscopy has a relevant diagnostic yield and is associated with higher therapeutic intervention rate,

Granular mixed laterally spreading tumors (GM-LSTs) have an intermediate level of risk for submucosal invasive cancer (SMICs) without clear signs of invasion (covert); the optimal resection method is uncertain. We aimed to determine the risk of covert SMIC in GM-LSTs based on clinical and endoscopic factors. METHODS:We collected data from 693 patients (50.6% male; median age, 69 years) with colorectal GM-LSTs, without signs of invasion, who underwent endoscopic resection (74.2%) or endoscopic submucosal dissection (25.2%) at 7 centers in Italy from 2016 through 2019. We performed multivariate and univariate analyses to identify demographic and endoscopic factors associated with risk of SMIC. We developed a multivariate model to calculate the number needed to treat (NNT) to detect 1 SMIC. RESULTS:Based on pathology analysis, 66 patients (9.5%) had covert SMIC. In multivariate analyses, increased risk of covert SMIC were independently associated with increasing lesion size (odds ratio per mm increase, 1.02, 95% CI, 1.01-1.03; P [ .003) and rectal location (odds ratio, 3.08; 95% CI, 1.62-5.83; P [ .004). A logistic regression model based on lesion size (with a cutoff of 40 mm) and rectal location identified patients with covert SMIC with 47.0% sensitivity, 82.6% specificity, and an area under the curve of 0.69. The NNT to identify 1 patient with a nonrectal SMIC smaller than 4 cm was 20; the NNT to identify 1 patient with a rectal SMIC of 4 cm or more was 5. CONCLUSIONS:In an analysis of data from 693 patients, we found the risk of covert SMIC in patients with GM-LSTs to be approximately 10%. GM-LSTs of 4 cm or more and a rectal location are high risk and should be treated by en-bloc resection. ClinicalTrials.gov, Number: NCT03836131.

The thulium laser system (TLS) is an emerging surgical tool. The 2-μm wavelength provides a confined coagulation depth (0.2 - 0.4 mm) to reduce the potential for inadvertent injuries. For the first time ever, we assessed TLS feasibility for endoscopic hemostasis ex vivo in pigs. In addition, we performed the first in vivo hemostatic treatments in humans. Tissue damage induced by TLS using different settings and optical fibers was compared to that from argon plasma coagulation (APC) in established ex vivo animal models. Three consecutive patients with complex nonvariceal upper gastrointestinal bleedings were treated and followed up. No deep submucosal injury was observed in animal models. The TLS showed a progressive penetration depth with increased power outputs and tissue exposures but very limited vertical tissue injury (0.1 - 2.0 mm) and lateral spreading damage (0.1 - 0.3 mm and 0.2 - 0.7 mm using the 365-µm and 550-µm fibers, respectively). In vivo, endoscopic hemostasis with TLS was always successful without complications. The TLS has proven to be very precise and easy to use. This novel technique appears to be a promising tool for advanced interventional endoscopy.