Risk of Metachronous Advanced Colorectal Neoplasia After Removal of Diminutive Versus Small Nonadvanced Adenomas: A Multicenter Study.

BACKGROUND: Current postpolypectomy guidelines treat 1-9 mm nonadvanced adenomas (NAAs) as carrying the same level of risk for metachronous advanced colorectal neoplasia (ACRN). AIMS: To evaluate whether small (6-9 mm) NAAs are associated with a greater risk of metachronous ACRN than diminutive (1-5 mm) NAAs. METHODS: We retrospectively evaluated 10,060 index colonoscopies performed from July 2011 to June 2019. A total of 1369 patients aged ≥ 40 years with index NAAs and having follow-up examinations were categorized into 5 groups based on size and number of index findings: Group 1, ≤ 2 diminutive NAAs (n = 655); Group 2, ≤ 2 small NAAs (n = 529); Group 3, 3-4 diminutive NAAs (n = 78); Group 4, 3-4 small NAAs (n = 65); and Group 5, 5-10 NAAs (n = 42). Size was classified based on the largest NAA. ACRN was defined as finding an advanced adenoma or colorectal cancer at follow-up. RESULTS: The absolute risk of metachronous ACRN increased from 7.2% in patients with all diminutive NAAs to 12.2% in patients with at least 1 small NAA (P = 0.002). Patients in Group 2 (adjusted odds ratio [AOR] 1.89; 95% confidence interval [CI], 1.21-2.95), Group 3 (AOR 2.40; 95% CI 1.78-4.90), Group 4 (AOR 2.77; 95% CI 1.35-5.66), and Group 5 (AOR 3.71; 95% CI 1.65-8.37) were associated with an increased risk of metachronous ACRN compared with Group 1. CONCLUSIONS: Patients with small NAAs have an increased risk of metachronous ACRN. Postpolypectomy guidelines should consider including risk stratification between small and diminutive adenomas.

Comparison of Right Colon Adenoma Miss Rates Between Water Exchange and Carbon Dioxide Insufflation: A Prospective Randomized Controlled Trial.

GOALS: To test the hypothesis that water exchange (WE), when compared with carbon dioxide (CO2) insufflation, significantly reduces the right colon adenoma miss rate (rAMR) in a blinded randomized controlled trial with cap-assisted colonoscopy. BACKGROUND: The unblinded consecutive group observational data showed that WE significantly decreased rAMR. The unblinded data are limited by potential bias. STUDY: Consecutive patients aged 45 years or more were randomized to undergo insertion with WE or CO2. Withdrawal and polypectomy were performed with CO2 in both groups to the hepatic flexure. The colonoscope was reinserted to the cecum. A second colonoscopist re-examined the right colon. The second colonoscopist was unaware, but made a guess, of the initial insertion method. The number of additional adenomas divided by the total number detected in both examinations equaled rAMR. RESULTS: Among 262 patients (131/group), demographic variables were similar. The body mass index was significantly higher in the WE group. Compared with CO2, WE significantly decreased rAMR [18.0% (33/183) vs. 34.6% (62/179), P=0.0025] and right colon serrated polyp miss rate [17.4% (27/155) vs. 39.3% (33/84), P=0.002]. Multivariate logistic regression analysis showed that WE was an independent predictor of rAMR (odds ratio, 0.42; 95% confidence interval, 0.21-0.86), and so was ≥2 adenomas in the right colon (odds ratio, 2.35; 95% confidence interval, 1.17-4.76). Whether the second colonoscopist guessed the insertion method correctly or not, and demographic and procedure variables were not associated with rAMR. CONCLUSIONS: The randomized controlled trial validated unblinded observational data showing that WE significantly decreased rAMR and right colon serrated polyp miss rate (clinical trial registration number: NCT03845933).

Proton Pump Inhibitors and Risk of Hepatocellular Carcinoma in Patients With Chronic Hepatitis B or C.

Researchers have hypothesized that the long-term use of proton pump inhibitors (PPIs) can increase the risk of developing cancer. However, the association between PPI use and hepatocellular carcinoma (HCC) risk is unclear. Using data from the Taiwan National Health Insurance Research Database for the period between 2003 and 2013, we identified 35,356 patients with chronic hepatitis B virus (HBV) or hepatitis C virus (HCV) infections. One-to-one propensity score matching by gender, age, cohort entry year, comorbidity, and medication resulted in the inclusion of 7,492 pairs of patients (PPI users and non-PPI users) for analyses. We performed multivariate and stratified analysis using the Kaplan-Meier method and Cox proportional hazards models in order to estimate the association between PPI use and the risk of developing HCC. In the HBV cohort, 237 patients developed HCC during a median follow-up of 53 months. However, PPI use was not associated with an increased risk of developing HCC (adjusted hazard ratio [aHR], 1.25; 95% confidence interval [CI], 0.90-1.73; P = 0.18). In the HCV cohort, 211 patients developed HCC; but again, PPI use was not associated with an increase in the risk of developing HCC (aHR, 1.19; 95% CI, 0.88-1.61; P = 0.25). We observed no relationship between a dose-dependent effect of PPI use and HCC risk. Subgroup analysis also confirmed that PPI use was not correlated to an increased HCC risk. Conclusion: Based on a retrospective population-based cohort study throughout Taiwan, where the prescription of PPI is tightly regulated, PPI use is not associated with the risk of developing HCC among patients with chronic HBV or HCV infections.

Risk of Renal Injury After the Use of Polyethylene Glycol for Outpatient Colonoscopy: A Prospective Observational Study.

GOAL: The goal of this study was to estimate the risk of renal injury after the use of 3-L polyethylene glycol (PEG) before outpatient colonoscopy. BACKGROUND: Population-based studies showed that the use of PEG was associated with renal injury, but this association has not been confirmed by prospective study. STUDY: Patients ≥40 years of age with an estimated glomerular filtration rate ≥30 mL/min were screened for enrollment. Laboratory data were collected before, during, and after the colonoscopies. Patients with a ≥30% increase in baseline serum creatinine levels were followed until a peak level was detected. Renal injury included acute renal dysfunction (ARD) and acute kidney injury (AKI), defined as a 30% to 49% increase and ≥50% increase in creatinine levels compared with the baseline, respectively. RESULTS: A total of 1163 patients (mean age, 55.7 y) completed the study. Baseline and first postcolonoscopy laboratory data were obtained an average of 17.0 days before and 17.3 days after the colonoscopies were performed, respectively. Renal injury was identified in 32 patients; 26 patients (2.2%) had ARD, and 6 patients (0.5%) had AKI. All patients with renal injury recovered fully during follow-up. In the subgroup analysis, the use of nonsteroidal anti-inflammatory drugs (NSAIDs) before colonoscopy was statistically associated with the development of AKI (odds ratio, 6.5; 95% confidence interval, 1.2-35.5; P=0.03). CONCLUSIONS: This prospective study showed that the use of PEG was associated with a small risk of renal injury. NSAIDs use was statistically associated with AKI in the context of colonoscopy for which PEG was used for bowel preparation.

Water exchange colonoscopy decreased adenoma miss rates compared with literature data and local data with CO2 insufflation: an observational study.

BACKGROUND: Reports showed adenoma miss rates (AMRs) of 22.5-27% in the right colon and 23.4-33.3% in the proximal colon. Missed lesions could contribute to postcolonoscopy cancers. Water exchange (WE) with near-complete removal of infused water during insertion increased adenoma detection rate but the impact on AMR had not been reported. We hypothesized that WE could reduce AMRs. Study 1 compared the AMRs of WE with literature data. Study 2 developed local AMR data with CO2 insufflation. METHODS: The lead author attended a research seminar in 2017 on WE colonoscopy. For performance improvement, study 1 was undertaken. When data in study 1 confirmed WE produced a considerably lower AMRs in the right and proximal colon, study 2 with CO2 insufflation was performed. RESULTS: Eighty-six patients completed each study. In study 1, WE removed 89% of infused water upon arrival to the cecum. The AMRs of right colon (17.5%) and proximal colon (15.5%) were considerably lower than those in the literature. Upon completion of study 2, compared with local data of CO2 insufflation, WE showed a significantly lower AMR in the right (17.5% vs. 33.8%, P = 0.034) and proximal (15.5% vs. 30.4%, P = 0.018) colon, respectively. The major limitation was that the investigation consisted of two consecutive observational studies, not a randomized controlled trial (RCT). CONCLUSIONS: WE with near-complete (89%) removal of infused water during insertion significantly decreased AMRs in the right and proximal colon compared with literature data and those of CO2 insufflation in our hands. The provocative data warrant confirmation in a RCT. TRIAL REGISTRATION: NCT03832322 (Retrospectively registered on February 2, 2019).

Comparison of polyp detection during both insertion and withdrawal versus only withdrawal of colonoscopy: A prospective randomized trial.

BACKGROUND AND AIMS: Polyps seen and not removed during colonoscope insertion are sometimes unable to be found during withdrawal. We aimed to evaluate whether additional inspection and polypectomy during insertion increases adenoma detection rate (ADR) compared with inspection and polypectomy entirely during withdrawal. METHODS: A total of 421 patients aged ≥ 45 years and undergoing colonoscopy were prospectively randomized to receive inspection and polypectomy during both insertion and withdrawal (study group) or inspection and polypectomy entirely during withdrawal (control group). The primary outcome was the ADR. Secondary outcomes included other adenoma-related parameters and procedure-related measures. RESULTS: Baseline demographics, procedure indications, preparation quality, total procedure time, sedative doses, colonoscopy difficulty, and patient discomfort were similar between the groups. The insertion time was significantly longer in the study group (11.1 ± 4.8 vs 6.2 ± 4.7 min, P < 0.0001). The withdrawal time was significantly longer in the control group (29.2 ± 9.8 vs 23.1 ± 7.9 min, P < 0.0001). There was no significant difference in the ADR (63.5% [study group] vs 68.1% [control group]), the mean adenoma per procedure (1.6 ± 2.0 vs 1.9 ± 2.4), or the mean adenoma per positive procedure (2.5 ± 2.0 vs 2.7 ± 2.5) between groups. The proximal colon ADR was significantly higher in the control group compared with the study group (56.2% vs 46.0%, P = 0.041). CONCLUSIONS: Additional inspection and polypectomy during colonoscope insertion did not improve ADR compared with inspection and polypectomy entirely during withdrawal. These results do not support an additional role for routine inspection during insertion (clinical trial registration number: NCT03444090).

Optimal procedural sequence for same-day bidirectional endoscopy with moderate sedation: A prospective randomized study.

BACKGROUND AND AIM: Same-day bidirectional endoscopy (BDE) is a commonly performed procedure, but the optimal sequence for the procedure with moderate conscious sedation is not well established. This study investigated the optimal sequence for same-day BDE under moderate conscious sedation and carbon dioxide insufflation in terms of sedation doses, patient discomfort, and colonoscopy performance. METHODS: A prospective randomized controlled study of 120 patients who were scheduled for BDE examination was performed. Colonoscopy followed by esophagogastroduodenoscopy (EGD) examination was performed in 60 patients (colonoscopy-EGD group), and EGD followed by colonoscopy examination was performed in another 60 patients (EGD-colonoscopy group). Endoscopists and patients completed a questionnaire to assess objective and subjective discomfort. RESULTS: Baseline demographics, procedure indications, bowel preparation quality, cecal intubation rate/time, colonoscopy withdrawal time, endoscopic interventions, BDE procedure time, colon polyp/adenoma detection rates, patient discomfort, and adverse events were similar between the two study groups. The total doses of fentanyl and midazolam were significantly higher for the colonoscopy-EGD group than for the EGD-colonoscopy group (83.4 ± 17.7 vs 68.7 ± 18.6 µg and 6.3 ± 1.4 vs 5.2 ± 1.3 mg, P < 0.0001 and P < 0.0001, respectively). The recovery time to discharge was significantly longer for the colonoscopy-EGD group than for the EGD-colonoscopy group (43.5 ± 16.2 vs 34.5 ± 8.9 min, P = 0.0003). CONCLUSIONS: Esophagogastroduodenoscopy followed by colonoscopy is the optimal sequence for same-day BDE under moderate conscious sedation and carbon dioxide insufflation. Following this order allows for a reduction of sedation doses and for shorter recovery times.

Residual Gastric Volume After Bowel Preparation With Polyethylene Glycol for Elective Colonoscopy: A Prospective Observational Study.

GOAL: To examine the residual gastric volume (RGV) in colonoscopy after bowel preparations with 3-L polyethylene glycol (PEG). BACKGROUND: Obstacles to high-volume bowel preparation by anesthesia providers resulting from concerns over aspiration risk are common during colonoscopy. STUDY: Prospective measurements of RGV were performed in patients undergoing esophagogastroduodenoscopy (EGD) and morning colonoscopy with split-dose PEG preparation, patients undergoing EGD and afternoon colonoscopy with same-day PEG preparation, and patients undergoing EGD alone under moderate conscious sedation. Colonoscopy patients were allowed to ingest clear liquids until 2 hours before the procedure. Patients undergoing EGD alone were instructed to eat/drink nothing after midnight. RESULTS: There were 860 evaluated patients, including 330 in the split-dose preparation group, 100 in the same-day preparation group, and 430 in the EGD-only group. Baseline demographics and disease/medication factors were similar. The mean RGV in patients receiving the same-day preparation (35.4 mL or 0.56 mL/kg) was significantly higher than that in patients receiving the split-dose preparation (28.5 mL or 0.45 mL/kg) and in patients undergoing EGD alone (22.8 mL or 0.36 mL/kg) (P=0.023 and P<0.0001, respectively). Within the bowel-preparation groups, patients with fasting times of 2 to 3 hours had similar RGV compared with patients who had fasting times >3 hours. The shape of the distribution and the range of RGV among the 3 study groups were similar. No aspiration occurred in any group. CONCLUSIONS: PEG bowel preparations increase RGV mildly, but seem to have no clinical significance. These results support the current fasting guidelines for colonoscopy.

Predictors of Suboptimal Bowel Preparation Using 3-l of Polyethylene Glycol for an Outpatient Colonoscopy: A Prospective Observational Study.

BACKGROUND: A 3-l polyethylene glycol (PEG) solution provided better bowel cleansing quality than a 2-l solution for outpatient colonoscopy. Predictors of suboptimal preparation using a 3-l PEG have not been previously reported. AIMS: To investigate the possible predictors of suboptimal bowel preparation using 3-l of PEG. METHODS: We analyzed a database of 1404 consecutive colonoscopies during a 27-month period at a community hospital. A split-dose PEG regimen was provided for morning colonoscopies, and a same-day PEG regimen was provided for afternoon colonoscopies. The level of bowel cleansing was prospectively scored according to the Boston Bowel Preparation Scale (BBPS). Possible predictors of suboptimal colon preparation, defined as a BBPS score <7, were analyzed using univariate statistics and multivariate logistic regression models. RESULTS: The mean age of the study population (46.7 % men) was 52.5 years (range 20-80 years, SD 11.1 years), and the majority of patients (77.6 %) underwent morning colonoscopies. A suboptimal bowel preparation was reported in 17.2 % of the observed colonoscopies. In the multivariate regression analysis, constipation (odds ratio [OR] 1.60, 95 % confidence interval [CI] 1.15-2.22), male gender (OR 1.68, 95 % CI 1.25-2.25), obesity (OR 1.76, 95 % CI 1.29-2.41), and inadequate (<80 %) PEG consumption (OR 5.4, 95 % CI 2.67-10.89) were independent predictors of a suboptimal colon preparation. CONCLUSIONS: This prospective study identified that constipation, male gender, obesity, and inadequate intake of PEG were significant risk factors for suboptimal bowel preparation using a 3-l PEG solution for outpatient colonoscopy. Interventions of optimized colonoscopy preparation should be targeted at these patient populations.

Impact of Bowel Preparation with Low-Volume (2-Liter) and Intermediate-Volume (3-Liter) Polyethylene Glycol on Colonoscopy Quality: A Prospective Observational Study.

BACKGROUND/AIMS: Two-liter polyethylene glycol (PEG) is the most commonly used bowel-cleansing regimen in Taiwan, but its efficacy is unsatisfactory. The aim of this study was to compare 2-liter and 3-liter PEG in terms of their impact on colonoscopy quality among an average-risk population. METHODS: Two-liter PEG was provided between August 2012 and May 2013, while 3-liter PEG was provided between June 2013 and March 2014. A split-dose regimen was provided for morning colonoscopy and a same-day regimen was provided for afternoon colonoscopy. The level of bowel cleansing was prospectively scored. RESULTS: A total of 407 consecutive subjects completed the 2-liter regimen, and another 407 consecutive subjects completed the 3-liter regimen. The 3-liter group had a significantly higher rate of excellent or good preparations, equivalent to a Boston bowel preparation scale of ≥7, than the 2-liter group (90 vs. 73%, p < 0.0001). More subjects in the 3-liter preparation group compared with the 2-liter group had overall adenoma (70 vs. 54%, p < 0.0001), proximal adenoma (47 vs. 35%, p = 0.0006), sessile serrated adenoma (28 vs. 6%, p < 0.0001), and advanced adenoma (21 vs. 9%, p < 0.0001). CONCLUSIONS: Three-liter PEG provided better cleansing quality and higher adenoma detection rate than 2-liter PEG.

Recurrence Pattern Is an Independent Surgical Prognostic Factor for Long-Term Oncological Outcomes in Patients with Hepatocellular Carcinoma.

BACKGROUND: The perioperative outcomes of a partial hepatectomy for hepatocellular carcinoma (HCC) have improved. However, high recurrence rates after a curative hepatectomy for HCC is still an issue. This study aimed to analyze the difference between various recurrence patterns. METHODS: We retrospectively reviewed 754 patients with HCC who underwent a curative hepatectomy between January 2012 and March 2021. Patients with recurrent events were categorized into three types: regional recurrence (type I), multiple intrahepatic recurrence (type II), or presence of any distant metastasis (type III). RESULTS: The median follow-up period was 51.2 months. Regarding recurrence, 375 (49.7%) patients developed recurrence, with 244 (32.4%), 51 (6.8%), and 80 (10.6%) patients having type I, II, and III recurrence, respectively. Type III recurrence appeared to be more common in male patients and those with major liver resection, vascular invasion, a large tumor size (>5 cm), a higher tumor grade, and higher levels of AST and AFP (p < 0.05). Patients who had distant metastasis at recurrence had the shortest recurrence time and the worst overall survival (p < 0.001 and p < 0.001). CONCLUSIONS: our study demonstrated that recurrence with distant metastasis occurred earliest and had the worst outcome compared to regional or multiple intrahepatic recurrences.

Comparison of Rockall and Blatchford scores to assess outcome of patients with bleeding peptic ulcers after endoscopic therapy.

BACKGROUND/AIMS: To determine the accuracy of Rockall and Blatchford scores for predicting outcome after endoscopic treatment in two groups of patients with bleeding peptic ulcers: those who initially presented with upper gastrointestinal (UGI) bleeding (Group A) and those who developed UGI bleeding during hospital treatment for another condition (Group B). METHODOLOGY: A total of 593 patients who had had endoscopic treatment for bleeding peptic ulcers from January 2009 to July 2010 were divided into Groups A and B. Endoscopic therapy including monotherapy (thermal therapy or hemoclipping) and combination therapy was applied. The Blatchford and complete Rockall scores for the two subgroups were calculated. Predictive statistics for the use of the two scoring systems were then compared for Groups A and B. RESULTS: Thirty-day re-bleeding and mortality rates increased with increased Rockall and Blatchford scores. Rockall scores were more accurate than the Blatchford scores for predicting mortality. However, neither the Rockall nor the Blatchford score could accurately predict recurrence of bleeding. When the results in Group B and Group A subgroups were compared, the average Rockall score for Group A was lower than that for Group B (5.6 vs. 6.3, p < 0.001). CONCLUSIONS: In high-risk patients with peptic ulcer bleeding, the Rockall score can better predict 30-day mortality than can the Blatchford score; this was particularly true for Group B patients.

Rapid Polyp Classification in Colonoscopy Using Textural and Convolutional Features.

Colorectal cancer is the leading cause of cancer-associated morbidity and mortality worldwide. One of the causes of developing colorectal cancer is untreated colon adenomatous polyps. Clinically, polyps are detected in colonoscopy and the malignancies are determined according to the biopsy. To provide a quick and objective assessment to gastroenterologists, this study proposed a quantitative polyp classification via various image features in colonoscopy. The collected image database was composed of 1991 images including 1053 hyperplastic polyps and 938 adenomatous polyps and adenocarcinomas. From each image, textural features were extracted and combined in machine learning classifiers and machine-generated features were automatically selected in deep convolutional neural networks (DCNN). The DCNNs included AlexNet, Inception-V3, ResNet-101, and DenseNet-201. AlexNet trained from scratch achieved the best performance of 96.4% accuracy which is better than transfer learning and textural features. Using the prediction models, the malignancy level of polyps can be evaluated during a colonoscopy to provide a rapid treatment plan.

The Vessels That Encapsulate Tumor Clusters (VETC) Pattern Is a Poor Prognosis Factor in Patients with Hepatocellular Carcinoma: An Analysis of Microvessel Density.

The outcomes of patients with hepatocellular carcinoma (HCC) are unsatisfactory because of its high recurrence rate. The Vessels that encapsulate tumor clusters (VETC) pattern is a unique vascular structure. In this study, we investigated the clinical−pathological features of HCC patients with the VETC pattern. We retrospectively reviewed patients with HCC who underwent curative hepatectomy at Chang Gung Memorial Hospital between 2007 and 2013. The form of the VETC pattern was established using an anti-CD31 stain. The results were classified into positive (VETC+) and negative (VETC−) patterns. We investigated and compared demographic data between these two groups. Overall, 174 patients were classified into either the VETC+ or VETC− groups. The median followed-up period was 80.5 months. There were significant differences in the number of hepatitis B carriers, the occurrence of vascular invasion, tumor size, TNM staging, microvessel density, and recurrence (all p < 0.05). Regarding the prediction of disease-free survival, after COX regression multivariate analysis, VETC+ remained independently associated with recurrent episodes (p = 0.003). The intra-tumoral microvessel density, demonstrated by CD-31, was the only clinical−pathological feature independently associated with VETC+. Our study demonstrated that the VETC pattern is an independent factor of poor prognosis for DFS. Higher intra-tumoral microvessel density was significantly associated with the VETC pattern. Further studies are needed to validate our findings.

Correlation between drug sensitivity profiles of circulating tumour cell-derived organoids and clinical treatment response in patients with pancreatic ductal adenocarcinoma.

INTRODUCTION: Pancreatic ductal adenocarcinoma (PDAC) is highly aggressive and has poor prognosis. There are few biomarkers to inform treatment decisions, and collecting tumour samples for testing is challenging. METHODS: Circulating tumour cells (CTCs) from patients with PDAC liquid biopsies were expanded ex vivo to form CTC-derived organoid cultures, using a laboratory-developed biomimetic cell culture system. CTC-derived organoids were tested for sensitivity to a PDAC panel of nine drugs, with tests conducted in triplicate, and a weighted cytotoxicity score (CTS) was calculated from the results. Clinical response to treatment in patients was evaluated using Response Evaluation Criteria in Solid Tumours (RECIST) version 1.1 criteria at the time of blood sampling and 3 months later. The correlation between CTS and clinical response was then assessed. RESULTS: A total of 41 liquid biopsies (87.8% from patients with Stage 4 disease) were collected from 31 patients. The CTC-derived organoid expansion was achieved in 3 weeks, with 87.8% culture efficiency. CTC-derived organoid cultures were positive for EpCAM staining and negative for CD45 staining in the surface marker analysis. All patients had received a median of two lines of treatment prior to enrolment and prospective utility analysis indicated significant correlation of CTS with clinical treatment response. Two representative case studies are also presented to illustrate the relevant clinical contexts. CONCLUSIONS: CTCs were expanded from patients with PDAC liquid biopsies with a high success rate. Drug sensitivity profiles from CTC-derived organoid cultures correlated meaningfully with treatment response. Further studies are warranted to validate the predictive potential for this approach.

Endoscopic metal stents for the palliation of malignant upper gastroduodenal obstruction.

BACKGROUND/AIMS: Gastric outlet obstruction (GOO) is frequently caused by tumor. Recently, endoscopic implantation of self-expanding metallic stents (SEMS) has been introduced as an improved palliative treatment for GOO. This study aims to study the effect of SEMS placement on nutrient intake in patients with GOO and correlate different SEMS positions with postoperative clinical outcomes. METHODOLOGY: Fifty six non-operable patients with GOO were enrolled. Obstruction of the duodenum (n=23) or gastric outlet (n=33) were commonly found. Either Wallstent Enteral Stents, WallFlex Enteral Duodenal or partially covered Ultraflex esophageal stents were placed under endoscopic and fluoroscopic guidance. The Gastric Outlet Obstruction Score (GOOSS) was used as the main outcome measurement. RESULTS: The procedure was technically feasible in 100% of patients and gave satisfactory clinical results in 98.2% (55/56). The patients had a median survival time of 97.5 days (range 9-380). Median stent patency was 72 days with a range of 8 to 267 days. The average GOOSS, measuring oral intake, was significantly improved, regardless of obstruction site (p<0.05). We also found that the site of SEMS placement did not affect the clinical outcome. CONCLUSIONS: Palliation with SEMS is a safe and effective method for restoring gastric intake in patients with malignant GOO.

Exploring and predicting mortality among patients with end-stage liver disease without cancer: a machine learning approach.

OBJECTIVE: End-stage liver disease is a global public health problem with a high mortality rate. Early identification of people at risk of poor prognosis is fundamental for decision-making in clinical settings. This study created a machine learning prediction system that provides several related models with visualized graphs, including decision trees, ensemble learning and clustering, to predict mortality in patients with end-stage liver disease. METHODS: A retrospective cohort study was conducted: the training data were from patients enrolled from January 2009 to December 2010 and followed up to December 2014; validation data were from patients enrolled from January 2015 to December 2016 and followed up to January 2019. Hospitalized patients with noncancer-related chronic liver disease were identified from the hospital's electrical medical records. RESULTS: In traditional multivariable logistic regression and Cox proportional hazard model, prothrombin time of international normalized ratio, which was significant with P value = 0.002, odds ratio = 2.790 and hazard ratio 1.363. Besides, blood urea nitrogen and C-reactive protein were also significant, with P value <0.001 and 0.026. The area under the curve was 0.771 in the receiver operating characteristic curve. In machine learning, blood urea nitrogen and age were regarded as the primary factors for predicting mortality. Creatinine, prothrombin time of international normalized ratio and bilirubin were also significant mortality predictors. The area under the curve of the random forest and AdaBoost was 0.838 and 0.792. CONCLUSION: The machine learning techniques provided a comprehensive assessment of patient conditions; it could help physicians make an accurate diagnosis of chronic liver disease and improve healthcare management.

Deep learning for abdominal ultrasound: A computer-aided diagnostic system for the severity of fatty liver.

BACKGROUND: The prevalence of nonalcoholic fatty liver disease is increasing over time worldwide, with similar trends to those of diabetes and obesity. A liver biopsy, the gold standard of diagnosis, is not favored due to its invasiveness. Meanwhile, noninvasive evaluation methods of fatty liver are still either very expensive or demonstrate poor diagnostic performances, thus, limiting their applications. We developed neural network-based models to assess fatty liver and classify the severity using B-mode ultrasound (US) images. METHODS: We followed standards for reporting of diagnostic accuracy guidelines to report this study. In this retrospective study, we utilized B-mode US images from a consecutive series of patients to develop four-class, two-class, and three-class diagnostic prediction models. The images were eligible if confirmed by at least two gastroenterologists. We compared pretrained convolutional neural network models, consisting of visual geometry group (VGG)19, ResNet-50 v2, MobileNet v2, Xception, and Inception v2. For validation, we utilized 20% of the dataset resulting in >100 images for each severity category. RESULTS: There were 21,855 images from 2,070 patients classified as normal (N = 11,307), mild (N = 4,467), moderate (N = 3,155), or severe steatosis (N = 2,926). We used ResNet-50 v2 for the final model as the best ones. The areas under the receiver operating characteristic curves were 0.974 (mild steatosis vs others), 0.971 (moderate steatosis vs others), 0.981 (severe steatosis vs others), 0.985 (any severity vs normal), and 0.996 (moderate-to-severe steatosis/clinically abnormal vs normal-to-mild steatosis/clinically normal). CONCLUSION: Our deep learning models achieved comparable predictive performances to the most accurate, yet expensive, noninvasive diagnostic methods for fatty liver. Because of the discriminative ability, including for mild steatosis, significant impacts on clinical applications for fatty liver are expected. However, we need to overcome machine-dependent variation, motion artifacts, lacking of second confirmation from any other tools, and hospital-dependent regional bias.

Interleukin-2 receptor alpha as a biomarker for nonalcoholic fatty liver disease diagnosis.

BACKGROUND: Two recent studies in the adult and pediatric Nonalcoholic Steatohepatitis-Clinical Research Network (NASH-CRN) cohorts have shown that soluble interleukin-2 receptor alpha (IL2RA) levels increased with fibrosis severity. However, no hepatic study has been conducted in Asian morbidly obese patients who underwent bariatric surgery. In this study, we proposed IL2RA as a biomarker for nonalcoholic fatty liver disease (NAFLD) diagnosis and performed immunohistochemistry (IHC) staining of IL2RA. METHODS: This prospective cohort study enrolled 123 morbidly obese patients who underwent bariatric surgery at Taipei Medical University Hospital from October 2016 to June 2018. During bariatric surgery, all patients underwent a wedge liver biopsy under laparoscopic guidance. The diagnoses of NASH and liver fibrosis were made histologically. In IHC of IL2RA, the number of lymphocytes with IL2RA immunoreactivity was counted in five high-power fields (×400, total: 1.19 mm2). RESULTS: Among the 123 patients, the mean age was 35.5 years, mean body mass index (BMI) was 40.6 kg/m2, 87 (70.7%) were female, 25 (20.7%) had diabetes mellitus, and 57 (46.3%; 11 with non-NAFLD and 46 with steatosis) and 66 (53.7%) were included in the non-NASH and NASH groups, respectively. The NASH group had higher IHC of IL2RA than the non-NASH group. In multivariate analysis, IHC of IL2RA (odds ratio, 1.025; 95% confidence interval, 1.006-1.045; p = 0.011) and alanine aminotransferase (ALT; odds ratio, 1.045; 95% confidence interval, 1.018-1.073; p = 0.001) were the independent factors associated with NASH. The area under the receiver operating curve of IL2RA IHC for NASH was 0.627 at the cutoff value of 82 (p = 0.0113). CONCLUSION: IL2RA is significantly associated with NASH in morbidly obese patients and would be a useful biomarker for NASH diagnosis.