Objective: Recently, oral vancomycin prophylaxis (OVP) has been suggested for the prevention of Clostridium difficile infection (CDI). We conducted a systematic review and meta-analysis to investigate the efficacy and safety of this approach.Design: Systematic review and meta-analysis.Methods: We conducted a computerized search of MEDLINE, EMBASE, and Cochrane databases from inception to March 2019 for publications investigating OVP for CDI prevention. Results were screened for eligibility. Relevant data were extracted and analyzed. Publication bias was assessed using the Egger test.Results: Ultimately, 8 retrospective studies and 1 prospective study examining 2174 patients, published between 2016 and 2019 were included in the review. OVP was associated with decreased CDI (odds ratio, 0.263; 95% confidence interval, 0.13–0.52) with considerable heterogeneity (I² = 61%). Meta-regression showed that total daily dose of OVP correlated with CDI, explaining 100% of heterogeneity between studies. Furthermore, 3 studies evaluated the risk of vancomycin-resistant enterococci (VRE) infection after OVP and found no significant increase.Conclusion: Our results suggest that OVP might decrease CDI rates in at-risk populations, although this conclusion should be interpreted with caution. Higher daily doses of OVP might increase CDI. Although the use of OVP in high-risk patients may reduce CDI, this suggestion has yet to be validated by prospective blinded randomized controlled trials.

Primary gastrointestinal lymphoma is the most common type of extra-nodal lymphoma, representing about 30%-50% of all extra-nodal involvement. The stomach is the most common site, with the colon and rectum accounting for a minority of occurrences. Primary colorectal lymphoma is uncommon, representing only 0.3% of all large intestinal malignancies and approximately 3% of gastrointestinal (GI) lymphomas, with the majority of these being B-cell non-Hodgkin lymphoma and diffuse large B-cell lymphoma (DLBCL) being the most common subtype. We present a case of an 85-year-old male who presented with symptoms suggestive of bowel obstruction, who, after further evaluation, was diagnosed with primary non-Hodgkin lymphoma of the colon, DLBCL subtype.

Hepatocellular carcinoma (HCC) is the most common primary tumor of the liver resulting in approximately 800,000 deaths annually. A growing body of research investigating statin use and HCC risk has shown conflicting results. We aim to evaluate the current evidence of statin impact on HCC risk. We performed a comprehensive literature search in PubMed, PubMed Central, Embase, and ScienceDirect databases from inception through May 2019 to identify all studies that evaluated the association between statin use and HCC. We included studies that presented an odds ratio (OR) with a 95% confidence interval (CI) or presented data sufficient to calculate the OR with a 95% CI. Statistical analysis was performed using the Comprehensive Meta-Analysis (CMA), Version 3 software, and a Forrest plot was generated. We assessed for publication bias using conventional techniques. Twenty studies (three randomized controlled trials, six cohorts, and 11 case-controls) with 2,668,497 patients including 24,341 cases of HCC were included in the meta-analysis. Our findings indicate a significant risk reduction of HCC among all statin users with a pooled odds ratio of 0.573 (95% CI: 0.491-0.668, I2= 86.57%) compared to non-users. No publication bias was found using Egger’s regression test or on visual inspection of the generated Funnel plot. The results indicate that statin use was associated with a 43% lower risk of HCC compared to statin non-users. Further prospective randomized research is needed to confirm the association.

This study evaluated the effectiveness of large language models (LLMs) and vision-language models (VLMs) in gastroenterology. We used board-style multiple-choice questions to assess the performance of both proprietary and open-source LLMs and VLMs—including GPT, Claude, Gemini, Mistral, Llama, Mixtral, Phi, and Qwen, across different interfaces, computing environments, and levels of compression (quantization). Among the proprietary models, o1-preview (82.0%) and Claude3.5-Sonnet (74.0%) had the highest accuracy, outperforming the top open-source models: Llama3.3-70b (65.7%) and Qwen-2.5-72b (61.0%). Among the small quantized open-source models, the 8-bit Llama 3.2-11b (51.7%) and 6-bit Phi3-14b (48.7%) performed the best, with scores comparable to their full-precision counterparts. Notably, VLM accuracy on image-containing questions improved (~10%) when given human-generated captions, remained unchanged with original images, and declined with LLM-generated captions. Further research is warranted to evaluate model capabilities in real-world clinical decision-making scenarios.

Outcomes of total pancreatectomy with islet autotransplantation: A systematic review and meta-analysis