Endoscopic ultrasound guided interventions in the management of pancreatic cancer

There has been a growing interest in developing endoscopic ultrasound (EUS)-guided interventions for pancreatic cancer, some of which have become standard of care. There are two main factors that drive these advancements to facilitate treatment of patients with pancreatic cancer, ranging from direct locoregional therapy to palliation of symptoms related to inoperable pancreatic cancer. Firstly, an upper EUS has the capability to access the entire pancreas-lesions in the pancreatic head and uncinate process can be accessed from the duodenum, and lesions in the pancreatic body and tail can be accessed from the stomach. Secondly, there has been a robust development of devices that allow through-the-needle interventions, such as placement of fiducial markers, brachytherapy, intratumoral injection, gastroenterostomy creation, and ablation. While these techniques are rapidly emerging, data from a multicenter randomized controlled trial for some procedures are awaited prior to their adoption in clinical settings.

Management and treatment of decompensated hepatic fibrosis and severe refractory Schistosoma mansoni ascites with transjugular intrahepatic portosystemic shunt

This study aimed to report the first case of a patient with hepatosplenic schistosomiasis mansoni, refractory ascites and portal vein thrombosis treated with a transjugular intrahepatic portosystemic shunt (TIPS), at the Instituto de Radiologia, Hospital das Clinicas, Faculdade de Medicina, Universidade de Sao Paulo, Brazil. After the procedure, the patient recovered favorably and progressed with portal pressure reduction and no deterioration of the liver function. Endovascular shunt modification is a conservative medical approach that often helps in reducing symptoms significantly, making it a less invasive and a safer alternative to liver transplantation for the treatment of schistosomiasis with portal hypertension.

EUS imaging for the diagnosis of nonalcoholic fatty liver disease

BACKGROUND AND AIMS

EUS is increasingly used to evaluate patients with liver disease, but its role in assessing hepatic steatosis has not been reported. The goal of our study was to assess the accuracy of EUS for diagnosing hepatic steatosis.

METHODS

We identified all patients who underwent EUS-guided liver biopsy sampling at our institution. All digitally stored EUS liver images were reviewed by a single radiologist, who rated the severity of liver echogenicity using a 4-point US scale. Liver biopsy specimens for all study patients were reviewed by a single liver pathologist, who rated them for steatosis and fibrosis using Nonalcoholic Steatohepatitis Clinical Research Network criteria. Receiver operator characteristic curves were used to assess the diagnostic accuracy of EUS for hepatic steatosis for all patients and in a subgroup analysis for obese and nonobese patients.

RESULTS

During the study period, 76 patients underwent EUS-guided liver biopsy sampling. The average age of study patients was 56.5 years, 50% were women, and 43.2% were obese. The accuracy for EUS for the diagnosis of hepatic steatosis was .8 (95% confidence interval [CI], .7-.89). The accuracy of EUS for the diagnosis of hepatic steatosis in obese patients was .93 (95% CI, .8-.99) and in nonobese patients was .69 (95% CI, .54-.83). For obese patients, EUS had a positive predictive value of 89.7% and a negative predictive value of 75%. The finding of course echotexture on EUS had an accuracy of 79% for the diagnosis of grade 3 fibrosis or cirrhosis.

CONCLUSIONS

EUS is a useful tool for the diagnosis of hepatic steatosis, particularly in obese patients in whom abdominal US has modest accuracy.

Endoscopic management of subepithelial lesions including neuroendocrine neoplasms: European Society of Gastrointestinal Endoscopy (ESGE) Guideline

1: ESGE recommends endoscopic ultrasonography (EUS) as the best tool to characterize subepithelial lesion (SEL) features (size, location, originating layer, echogenicity, shape), but EUS alone is not able to distinguish among all types of SEL.Strong recommendation, moderate quality evidence. 2: ESGE suggests providing tissue diagnosis for all SELs with features suggestive of gastrointestinal stromal tumor (GIST) if they are of size > 20 mm, or have high risk stigmata, or require surgical resection or oncological treatment.Weak recommendation, very low quality evidence. 3: ESGE recommends EUS-guided fine-needle biopsy (EUS-FNB) or mucosal incision-assisted biopsy (MIAB) equally for tissue diagnosis of SELs ≥ 20 mm in size.Strong recommendation, moderate quality evidence. 4: ESGE recommends against surveillance of asymptomatic gastrointestinal (GI) tract leiomyomas, lipomas, heterotopic pancreas, granular cell tumors, schwannomas, and glomus tumors, if the diagnosis is clear.Strong recommendation, moderate quality evidence. 5: ESGE suggests surveillance of asymptomatic esophageal and gastric SELs without definite diagnosis, with esophagogastroduodenoscopy (EGD) at 3-6 months, and then at 2-3-year intervals for lesions < 10 mm in size, and at 1-2-year intervals for lesions 10-20 mm in size. For asymptomatic SELs > 20 mm in size that are not resected, ESGE suggests surveillance with EGD plus EUS at 6 months and then at 6-12-month intervals.Weak recommendation, very low quality evidence. 6: ESGE recommends endoscopic resection for type 1 gastric neuroendocrine neoplasms (g-NENs) if they grow larger than 10 mm. The choice of resection technique should depend on size, depth of invasion, and location in the stomach.Strong recommendation, low quality evidence. 7: ESGE suggests considering removal of histologically proven gastric GISTs smaller than 20 mm as an alternative to surveillance. The decision to resect should be discussed in a multidisciplinary meeting. The choice of technique should depend on size, location, and local expertise.Weak recommendation, very low quality evidence. 8: ESGE suggests that, to avoid unnecessary follow-up, endoscopic resection is an option for gastric SELs smaller than 20 mm and of unknown histology after failure of attempts to obtain diagnosis.Weak recommendation, very low quality evidence. 9: ESGE recommends basing the surveillance strategy on the type and completeness of resection. After curative resection of benign SELs no follow-up is advised, except for type 1 gastric NEN for which surveillance at 1-2 years is advised.Strong recommendation, low quality evidence. 10: For lower or upper GI NEN with a positive or indeterminate margin at resection, ESGE recommends repeating endoscopy at 3-6 months and another attempt at endoscopic resection in the case of residual disease.Strong recommendation, low quality evidence.

Endoscopic Ultrasound Evaluation of Portal Pressure

Portal hypertension is a complex syndrome with multiple clinical manifestations that develop in a variety of conditions and diseases. The spectrum of portal hypertension manifestations is wide and dependent on the physiologic site of increased portal resistance (pre-, post-, and sinusoidal or intrahepatic), as well as the presence of hepatocellular dysfunction.

Liver Biopsy in Patients With Alcohol-Associated Liver Disease With Acute-on-Chronic Liver Failure

Patients with alcohol-associated liver disease may develop severe forms of presentation of acute-on-chronic liver failure, with a high risk for short-term mortality. Alcoholic hepatitis should be suspected among patients with alcohol-associated liver disease who present with acute-on-chronic liver failure. In this review, we discuss the need and feasibility of liver biopsy in the diagnosis of alcoholic hepatitis and predicting its prognosis among decompensated patients with alcohol-associated liver disease and acute-on-chronic liver failure.

Laparoscopic Intragastric Cystgastrostomy: A Technical Report

Background: Management of symptomatic pancreatic pseudocysts poses a unique challenge to minimally invasive surgeons. Despite the predominance of endoscopic management of pancreatic pseudocysts, the laparoscopic approach remains a critical skill in the armamentarium of surgeons. Methods: This report details a laparoscopic intragastric approach to create a pancreatic cystgastrostomy using intraoperative ultrasound and endoscopy. Conclusion: Laparoendoscopic techniques for pancreatic pseudocysts are still required in selective cases when endoscopic management is not available or fails. Using this technique provides patients with same clinical benefits of an endoscopic approach.

Artificial Intelligence: A Shifting Paradigm in Cardio-Cerebrovascular Medicine

The future of healthcare is an organic blend of technology, innovation, and human connection. As artificial intelligence (AI) is gradually becoming a go-to technology in healthcare to improve efficiency and outcomes, we must understand our limitations. We should realize that our goal is not only to provide faster and more efficient care, but also to deliver an integrated solution to ensure that the care is fair and not biased to a group of sub-population. In this context, the field of cardio-cerebrovascular diseases, which encompasses a wide range of conditions-from heart failure to stroke-has made some advances to provide assistive tools to care providers. This article aimed to provide an overall thematic review of recent development focusing on various AI applications in cardio-cerebrovascular diseases to identify gaps and potential areas of improvement. If well designed, technological engines have the potential to improve healthcare access and equitability while reducing overall costs, diagnostic errors, and disparity in a system that affects patients and providers and strives for efficiency.

Role of endoscopic ultrasound in the field of hepatology: Recent advances and future trends

The role of endoscopic ultrasound (EUS) as a diagnostic and therapeutic modality for the management of various gastrointestinal diseases has been expanding. The imaging or intervention for various liver diseases has primarily been the domain of radiologists. With the advances in EUS, the domain of endosonologists is rapidly expanding in the field of hepatology. The ability to combine endoscopy and sonography in one hybrid device is a unique property of EUS, together with the ability to bring its probe/transducer near the liver, the area of interest. Its excellent spatial resolution and ability to provide real-time images coupled with several enhancement techniques, such as contrast-enhanced (CE) EUS, have facilitated the growth of EUS. The concept of “Endo-hepatology” encompasses the wide range of diagnostic and therapeutic procedures that are now gradually becoming feasible for managing various liver diseases. Diagnostic advancements can enable a wide array of techniques from elastography and liver biopsy for liver parenchymal diseases, to CE-EUS for focal liver lesions to portal pressure measurements for managing various liver conditions. Similarly, therapeutic advancements range from EUS-guided eradication of varices, drainage of bilomas and abscesses to various EUS-guided modalities of liver tumor management. We provide a comprehensive review of all the different diagnostic and therapeutic EUS modalities available for the management of various liver diseases. A synopsis of all the technical details involving each procedure and the available data has been tabulated, and the future trends in this area have been highlighted.

Role of Endoscopic Ultrasound in Liver Disease: Where Do We Stand?

As the burden of liver disease in the general populace steadily increases, so does the need for both advanced diagnostic and treatment options. Endoscopic ultrasound is a reliable diagnostic and therapeutic method that has an established role, foremost in pancreatobiliary pathology. This paper aims to summarize the growing role of endoscopic ultrasound in hepatology based on the search of the current literature. A number of applications of endoscopic ultrasound are reviewed, including both noninvasive methods and tissue acquisition in focal and diffuse liver disease, portal hypertension measurement, detection and management of gastric and esophageal varices, treatment of focal liver lesions and staging of pancreatobiliary malignancies, treatment of cystic and solid liver lesions, as well as liver abscess drainage. Both hepatologists and endoscopists should be aware of the evolving role of endoscopic ultrasound in liver disease. The inherent invasive nature of endoscopic examination limits its use to a targeted population identified using noninvasive methods. Endoscopic ultrasound is one the most versatile methods in gastroenterology, allowing immediate access with detection, sampling, and treatment of digestive tract pathology. Further expansion of its use in hepatology is immanent.

Imputation of missing values for electronic health record laboratory data

Laboratory data from Electronic Health Records (EHR) are often used in prediction models where estimation bias and model performance from missingness can be mitigated using imputation methods. We demonstrate the utility of imputation in two real-world EHR-derived cohorts of ischemic stroke from Geisinger and of heart failure from Sutter Health to: (1) characterize the patterns of missingness in laboratory variables; (2) simulate two missing mechanisms, arbitrary and monotone; (3) compare cross-sectional and multi-level multivariate missing imputation algorithms applied to laboratory data; (4) assess whether incorporation of latent information, derived from comorbidity data, can improve the performance of the algorithms. The latter was based on a case study of hemoglobin A1c under a univariate missing imputation framework. Overall, the pattern of missingness in EHR laboratory variables was not at random and was highly associated with patients' comorbidity data; and the multi-level imputation algorithm showed smaller imputation error than the cross-sectional method.

Biliopancreatic Endoscopy in Altered Anatomy

Background and Objectives: Anatomical post-surgical alterations of the upper gastrointestinal (GI) tract have always been challenging for performing diagnostic and therapeutic endoscopy, especially when biliopancreatic diseases are involved. Esophagectomy, gastrectomy with various reconstructions and pancreaticoduodenectomy are among the most common surgeries causing upper GI tract alterations. Technological improvements and new methods have increased the endoscopic success rate in these patients, and the literature has been rapidly increasing over the past few years. The aim of this systematic review is to identify evidence on the available biliopancreatic endoscopic techniques performed in the altered post-surgical anatomy of upper GI tract. Materials and Methods: We performed a systematic search of PubMed, MEDLINE, Cochrane Library, and SCOPUS databases. Study-level variables extracted were the last name of the first author, publication year, study design, number of patients, type of post-surgical anatomical alteration, endoscopic technique, success rate and endoscopic-related adverse events. Results: Our primary search identified 221 titles, which was expanded with studies after the citation search. The final full-text review process identified 52 articles (31 retrospective studies, 8 prospective studies and 13 case reports). We found several different techniques developed over the years for biliopancreatic diseases in altered anatomy, in order to perform both endoscopic ultrasound (EUS) and endoscopic retrograde cholangiopancreatography (ERCP). They included enteroscopy-assisted ERCP (double and single balloon enteroscopy-ERCP, spiral enteroscopy-ERCP) laparoscopic assisted ERCP, EUS-Directed transgastric ERCP, EUS-directed transgastric intervention, gastric access temporary for endoscopy, and percutaneous assisted trans prosthetic endoscopic therapy. The success rate was high (most of the techniques showed a success rate over 90%) and a low rate of adverse events were reported. Conclusions: We suggest the considerationof the novel techniques when approaching patients with altered anatomy who require biliopancreatic endoscopy, focusing on the surgery type, success rate and adverse events reported in the literature.

Neuroendocrine Tumors of the Gastrointestinal Tract: A Focused Review and Practical Approach for Gastroenterologists

Neuroendocrine tumors (NETs) are rare tumors derived from the neuroendocrine cell system, and more commonly found in the gastrointestinal (GI) tract. Over the last decades, the incidence of GI-NETs has been steadily increasing, partly due to the expanding indications for endoscopy. Most patients with NETs are asymptomatic, and their NETs are noticed during screening examinations; thus, endoscopists are on the frontline of the diagnosis of GI-NETs. Since GI-NETs are less frequent than other malignancies, the natural history, diagnosis, and management of these tumors may not be fully understood. In this review, we aim to update the endoscopist on key clinical features and management of patients with gastric, duodenal, and rectal NETs.

Post hoc validation of a tool that accurately predicts the outcome of endoscopic therapy in Bouveret syndrome

Background

Bouveret syndrome is characterized by gallstone impaction in the upper gastrointestinal tract causing gastric outlet obstruction. In Bouveret syndrome, endoscopic gallstone removal can avert the need for surgery. However, in cases in which endoscopic therapy is unlikely to succeed, endoscopic attempts delay definitive treatment and compound patient risks. We previously developed a model that predicts endoscopic outcomes from data derived through a systematic review. This tool uses gallstone length, site of impaction, and the number of planned methods of lithotripsy to predict the likelihood of endoscopic success with an accuracy of 81.0%. This study aimed to evaluate our tool performance in an independent, non-training data set and assess endoscopic and surgical outcomes.

Methods

Systematic searches of the PubMed, Scopus, and Cochrane databases were performed for articles published between 16 April 2018 and 1 June 2021. The data reported after our previous study were harvested and inputted into the tool to evaluate their ability to accurately predict outcomes when compared with actual outcomes.

Results

Newly collated data in fields of interest showed no significant statistical differences compared with previous training data sets. Endoscopic therapy was successful in 41.9% of cases. Gallstones of ≤4 cm had a higher chance of successful endoscopic intervention (odds ratio 6.7, 95% confidence interval 1.7–25.8, P < 0.01). Complications of surgery were reported in 29.5%; there was one fatality reported. Post hoc evaluation of our predictive tool demonstrated an AUROC score of 0.80.

Conclusions

We have demonstrated in an independent data set that the tool can be used to accurately predict outcomes of endoscopic therapy. Patients in whom endoscopic therapy is most likely to fail should be offered an early surgical opinion.

Predicting short and long-term mortality after acute ischemic stroke using EHR

OBJECTIVE

Despite improvements in treatment, stroke remains a leading cause of mortality and long-term disability. In this study, we leveraged administrative data to build predictive models of short- and long-term post-stroke all-cause-mortality.

METHODS

The study was conducted and reported according to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) guideline. We used patient-level data from electronic health records, three algorithms, and six prediction windows to develop models for post-stroke mortality.

RESULTS

We included 7144 patients from which 5347 had survived their ischemic stroke after two years. The proportion of mortality was between 8%(605/7144) within 1-month, to 25%(1797/7144) for the 2-years window. The three most common comorbidities were hypertension, dyslipidemia, and diabetes. The best Area Under the ROC curve(AUROC) was reached with the Random Forest model at 0.82 for the 1-month prediction window. The negative predictive value (NPV) was highest for the shorter prediction windows - 0.91 for the 1-month - and the best positive predictive value (PPV) was reached for the 6-months prediction window at 0.92. Age, hemoglobin levels, and body mass index were the top associated factors. Laboratory variables had higher importance when compared to past medical history and comorbidities. Hypercoagulation state, smoking, and end-stage renal disease were more strongly associated with long-term mortality.

CONCLUSION

All the selected algorithms could be trained to predict the short and long-term mortality after stroke. The factors associated with mortality differed depending on the prediction window. Our classifier highlighted the importance of controlling risk factors, as indicated by laboratory measures.

The Need for Alternatives to Liver Biopsies: Non-Invasive Analytics and Diagnostics

Histology remains essential for the diagnosis and management of many disorders affecting the liver. However, the biopsy procedure itself is associated with a low risk of harm to the patient and cost to the health services; samples may not be adequate and are subject to sampling variation. Furthermore, interpretation often depends on the skill of the pathologist. Increasingly, new techniques are becoming available that are altering the indications for liver biopsy. Many diseases of the liver can be diagnosed and managed using serological and radiological techniques; the degree of fibrosis and fat can often be assessed by serological or imaging techniques and the nature of space occupying lesions defined by serology, imaging and use of liquid biopsy. However, these techniques, too, are subject to limitations: sensitivity and specificity is not always adequate for diagnosis or management; some techniques are expensive and often also require expert interpretation. Although there may be less need for liver biopsy today, histology remains the gold standard as well as an essential tool for the diagnosis and management of many conditions, especially where there are multiple pathologies, or where a diagnosis cannot or has not been made by alternative approaches. Until less invasive techniques become more reliable and accessible, liver histology will remain a key investigation.

Comparison of Two Specialized Histology Needles for Endoscopic Ultrasound (EUS)-Guided Liver Biopsy: A Pilot Study

BACKGROUND

EUS-guided liver biopsy (EUS-LB) has been shown to be a safe and effective alternative to percutaneous liver biopsy. The optimal needle device and technique for EUS-LB is still evolving. The aim of this study was to compare the efficacy of two second-generation 19G fine-needle biopsy (FNB) (Franseen- and Fork-tip) devices for EUS-LB.

METHODS

This is a repeated-measure crossover study with a prospectively maintained cohort of patients. We performed EUS-LB with a one-pass and single-actuation method using two 19G FNB needles in 22 consecutive patients between 10/2018 and 9/2019. Patients were randomized to left vs right liver lobes to be biopsied as well as the needle sequence. The specimens obtained were evaluated for adequacy for histologic diagnosis. The primary outcome was number of complete portal tracts (CPTs), post-fix aggregate, and longest specimen length. Secondary outcomes were prefix aggregate specimen length and the specimen adequacy judged by two expert pathologists.

RESULTS

A total of 44 liver biopsies were performed in 22 patients. The CPTs were higher in the Franseen-tip needle group compared to the Fork-tip needle group (14.4 vs 9.5, p = 0.043). Post-fix aggregate specimen length (44.9 mm vs 34.6 mm, p = 0.097), the post-fix longest specimen length (19.9 mm vs 13.7 mm, p = 0.175), and prefix aggregate specimen length (51.7 mm vs 45 mm, p = 0.265) were not significantly different. Both needles showed similarly high histologic adequacy (100% vs 95.5%, p = 0.312). Interestingly, the right of the liver showed higher yield of CPTs with both needles (Franseen, 16.2 vs. 12.8, p = 0.003, the Fork-tip, 12.8 vs. 7.0, p < 0.0001).

CONCLUSION

EUS-guided liver biopsy using the 19G Franseen-tip needle may provide more CPTs than 19G Fork-tip needle on a single-pass, single-actuation comparison.

Machine Learning-Enabled 30-Day Readmission Model for Stroke Patients

Background and Purpose: Hospital readmissions impose a substantial burden on the healthcare system. Reducing readmissions after stroke could lead to improved quality of care especially since stroke is associated with a high rate of readmission. The goal of this study is to enhance our understanding of the predictors of 30-day readmission after ischemic stroke and develop models to identify high-risk individuals for targeted interventions. Methods: We used patient-level data from electronic health records (EHR), five machine learning algorithms (random forest, gradient boosting machine, extreme gradient boosting-XGBoost, support vector machine, and logistic regression-LR), data-driven feature selection strategy, and adaptive sampling to develop 15 models of 30-day readmission after ischemic stroke. We further identified important clinical variables. Results: We included 3,184 patients with ischemic stroke (mean age: 71 ± 13.90 years, men: 51.06%). Among the 61 clinical variables included in the model, the National Institutes of Health Stroke Scale score above 24, insert indwelling urinary catheter, hypercoagulable state, and percutaneous gastrostomy had the highest importance score. The Model's AUC (area under the curve) for predicting 30-day readmission was 0.74 (95%CI: 0.64-0.78) with PPV of 0.43 when the XGBoost algorithm was used with ROSE-sampling. The balance between specificity and sensitivity improved through the sampling strategy. The best sensitivity was achieved with LR when optimized with feature selection and ROSE-sampling (AUC: 0.64, sensitivity: 0.53, specificity: 0.69). Conclusions: Machine learning-based models can be designed to predict 30-day readmission after stroke using structured data from EHR. Among the algorithms analyzed, XGBoost with ROSE-sampling had the best performance in terms of AUC while LR with ROSE-sampling and feature selection had the best sensitivity. Clinical variables highly associated with 30-day readmission could be targeted for personalized interventions. Depending on healthcare systems' resources and criteria, models with optimized performance metrics can be implemented to improve outcomes.

Prediction of Long-Term Stroke Recurrence Using Machine Learning Models

Background: The long-term risk of recurrent ischemic stroke, estimated to be between 17% and 30%, cannot be reliably assessed at an individual level. Our goal was to study whether machine-learning can be trained to predict stroke recurrence and identify key clinical variables and assess whether performance metrics can be optimized. Methods: We used patient-level data from electronic health records, six interpretable algorithms (Logistic Regression, Extreme Gradient Boosting, Gradient Boosting Machine, Random Forest, Support Vector Machine, Decision Tree), four feature selection strategies, five prediction windows, and two sampling strategies to develop 288 models for up to 5-year stroke recurrence prediction. We further identified important clinical features and different optimization strategies. Results: We included 2091 ischemic stroke patients. Model area under the receiver operating characteristic (AUROC) curve was stable for prediction windows of 1, 2, 3, 4, and 5 years, with the highest score for the 1-year (0.79) and the lowest score for the 5-year prediction window (0.69). A total of 21 (7%) models reached an AUROC above 0.73 while 110 (38%) models reached an AUROC greater than 0.7. Among the 53 features analyzed, age, body mass index, and laboratory-based features (such as high-density lipoprotein, hemoglobin A1c, and creatinine) had the highest overall importance scores. The balance between specificity and sensitivity improved through sampling strategies. Conclusion: All of the selected six algorithms could be trained to predict the long-term stroke recurrence and laboratory-based variables were highly associated with stroke recurrence. The latter could be targeted for personalized interventions. Model performance metrics could be optimized, and models can be implemented in the same healthcare system as intelligent decision support for targeted intervention.

Gastric linitis plastica due to signet-ring cell carcinoma with Krukenberg tumors diagnosed by endoscopic ultrasound-guided fine-needle aspiration

Gastric linitis plastica is a rare condition characterized by the stiffening and thickening of the gastric wall, presenting a "leather bottle" appearance. It is generally associated with infiltration of poorly differentiated gastric cancer cells, but similar findings can arise due to metastases from other cancers and benign inflammatory processes. Despite advances in imaging modalities, pathological evaluation is required for diagnosis. As endoscopic mucosal biopsies may not be diagnostic, endoscopic ultrasound-guided fine-needle aspiration of the thickened stomach wall presents a potential diagnostic option. We report a case which presented with bilateral ovarian tumors and was ultimately diagnosed as gastric signet-ring cell carcinoma with Krukenberg tumors by endoscopic ultrasound-guided fine-needle aspiration of the stomach wall. The preoperative diagnosis allowed the patient to avoid surgery and to start chemotherapy for gastric cancer. We also review the existing literature on endoscopic ultrasound-guided fine-needle aspiration in the context of gastric linitis plastica.

Is the Mediterranean Diet the Best Approach to NAFLD Treatment Today?

Non-alcoholic fatty liver disease (NAFLD) is one of the most prevalent diseases worldwide, involving about 20-30% of the general population [...].

Early Detection of Septic Shock Onset Using Interpretable Machine Learners

BACKGROUND

Developing a decision support system based on advances in machine learning is one area for strategic innovation in healthcare. Predicting a patient's progression to septic shock is an active field of translational research. The goal of this study was to develop a working model of a clinical decision support system for predicting septic shock in an acute care setting for up to 6 h from the time of admission in an integrated healthcare setting.

METHOD

Clinical data from Electronic Health Record (EHR), at encounter level, were used to build a predictive model for progression from sepsis to septic shock up to 6 h from the time of admission; that is, T = 1, 3, and 6 h from admission. Eight different machine learning algorithms (Random Forest, XGBoost, C5.0, Decision Trees, Boosted Logistic Regression, Support Vector Machine, Logistic Regression, Regularized Logistic, and Bayes Generalized Linear Model) were used for model development. Two adaptive sampling strategies were used to address the class imbalance. Data from two sources (clinical and billing codes) were used to define the case definition (septic shock) using the Centers for Medicare & Medicaid Services (CMS) Sepsis criteria. The model assessment was performed using Area under Receiving Operator Characteristics (AUROC), sensitivity, and specificity. Model predictions for each feature window (1, 3 and 6 h from admission) were consolidated.

RESULTS

Retrospective data from April 2005 to September 2018 were extracted from the EHR, Insurance Claims, Billing, and Laboratory Systems to create a dataset for septic shock detection. The clinical criteria and billing information were used to label patients into two classes-septic shock patients and sepsis patients at three different time points from admission, creating two different case-control cohorts. Data from 45,425 unique in-patient visits were used to build 96 prediction models comparing clinical-based definition versus billing-based information as the gold standard. Of the 24 consolidated models (based on eight machine learning algorithms and three feature windows), four models reached an AUROC greater than 0.9. Overall, all the consolidated models reached an AUROC of at least 0.8820 or higher. Based on the AUROC of 0.9483, the best model was based on Random Forest, with a sensitivity of 83.9% and specificity of 88.1%. The sepsis detection window at 6 h outperformed the 1 and 3-h windows. The sepsis definition based on clinical variables had improved performance when compared to the sepsis definition based on only billing information.

CONCLUSION

This study corroborated that machine learning models can be developed to predict septic shock using clinical and administrative data. However, the use of clinical information to define septic shock outperformed models developed based on only administrative data. Intelligent decision support tools can be developed and integrated into the EHR and improve clinical outcomes and facilitate the optimization of resources in real-time.

Interventional Endoscopic Ultrasound: Current Status and Future Directions

The evolution of endoscopic ultrasound (EUS) from a diagnostic to a therapeutic procedure has resulted in a paradigm shift toward endoscopic management of disease states that previously required percutaneous or surgical approaches. The past few years have seen additional techniques and devices that have enabled endoscopists to expand its diagnostic and therapeutic capabilities. Some of these techniques initially were reported more than a decade ago; however, with further device development and refinement in techniques there is potential for expanding the application of these techniques and new technologies to a broader group of interventional gastroenterologists. Lack of formalized training, devices, and prospective data regarding their use in addition to a scarcity of guidelines on implementation of these technologies into clinical practice are contributing factors impeding the growth of the field of interventional EUS. In April 2019, the American Gastroenterological Association's Center for Gastrointestinal Innovation and Technology conducted its annual Tech Summit and a key session focused on interventional EUS. This article is a White Paper generated from the conference, discusses the published literature pertaining to the topic of interventional EUS, and outlines a proposed framework for the implementation of interventional EUS techniques into clinical practice. Three primary areas of interventional EUS are addressed: (1) EUS-guided access; (2) EUS-guided tumor ablation; and (3) endohepatology. There was general agreement among participants on several key components. The introduction of these novel interventions requires better tools, more data on safety/outcomes, and improved training for endoscopists. Participants also agreed that widespread implementation and use of these techniques will require support from Gastrointestinal Societies and other key stakeholders including payers. Continued work by the Gastrointestinal Societies and manufacturers to provide training programs, appropriate equipment/work environments, and policies that motivate endoscopists to adopt new techniques is essential for growing the field of interventional EUS.

Endoscopic ultrasound guided hepatic interventions

Intervention for liver disease has predominantly been performed through the percutaneous approach. However, as endoscopic ultrasound (EUS) applications have expanded, there have emerged various EUS-guided interventions for liver disease, a space we call "Endo-Hepatology". EUS-guided liver biopsy can be considered the "forerunner" of Endo-Hepatology and has become a clinical option for patients requiring histologic diagnosis and staging of their liver disease. EUS also enables direct access to the portal vein. Subsequently, many procedures are being explored, such as angiography, measurement of the portosystemic pressure gradient, portal vein sampling to detect cancer cell or DNA, and EUS-guided transhepatic intrahepatic portosystemic shunt creation. Since the transducer is close to the liver, especially the left and caudate lobes, EUS can be used as a rescue when the percutaneous approach is not favorable and EUS-guided treatments of liver tumor, cyst and abscess have been reported. This review summarizes the available studies of EUS-guided intervention in the liver.

Pancreatic Trauma: Imaging Review and Management Update

Traumatic injuries of the pancreas are uncommon and often difficult to diagnose owing to subtle imaging findings, confounding multiorgan injuries, and nonspecific clinical signs. Nonetheless, early diagnosis and treatment are critical, as delays increase morbidity and mortality. Imaging has a vital role in diagnosis and management. A high index of suspicion, as well as knowledge of the anatomy, mechanism of injury, injury grade, and role of available imaging modalities, is required for prompt accurate diagnosis. CT is the initial imaging modality of choice, although the severity of injury can be underestimated and assessment of the pancreatic duct is limited with this modality. The time from injury to definitive diagnosis and the treatment of potential pancreatic duct injury are the primary factors that determine outcome following pancreatic trauma. Disruption of the main pancreatic duct (MPD) is associated with higher rates of complications, such as abscess, fistula, and pseudoaneurysm, and is the primary cause of pancreatic injury-related mortality. Although CT findings can suggest pancreatic duct disruption according to the depth of parenchymal injury, MR cholangiopancreatography and endoscopic retrograde cholangiopancreatography facilitate direct assessment of the MPD. Management of traumatic pancreatic injury depends on multiple factors, including mechanism of injury, injury grade, presence (or absence) of vascular injury, hemodynamic status of the patient, and associated organ damage. ^©RSNA, 2020 See discussion on this article by Patlas.

A State-of-the-Art Review on the Evolving Utility of Endoscopic Ultrasound in Liver Diseases Diagnosis

Liver diseases are amongst the most common diseases worldwide and manifest as a parenchymatic and/or biliary injury due to several causes as well as focal liver lesions, ranging from benign to malignant ones. The diagnosis of liver diseases is based mainly on biochemical and advanced imaging studies and, when required, on liver biopsy. Endoscopic ultrasound (EUS), which combines endoscopy and ultrasonography, is one of the main examination techniques used in gastroenterology as it is applied to evaluate abnormalities in the lumen of the upper and lower gastrointestinal tract and to define pancreatic and hepato-biliary features, often in chronic patients. Given its high spatial resolution and its proximity to the liver, EUS is gaining popularity in the diagnostic work up of liver diseases. This is a comprehensive overview of the current literature on the diagnostic indications for EUS use in patients with liver diseases. We performed a MEDLINE\PubMed and Embase search, and all articles that were relevant, after reviewing abstracts, were assessed and the full text was analyzed to extract data regarding technical success, diagnostic yield, bioptic characteristics, and complications rate. EUS-guided imaging and biopsy techniques in liver diseases have shown consistent favorable promising results among the reports through the literature, with an excellent diagnostic yield and safety profile, especially in the context of focal lesions and portal hypertension. The application of EUS in the diagnosis of liver diseases is a promising technique and should be considered as a first-line therapeutic option in selected cases.

Systematic review: management of localised low-grade upper gastrointestinal neuroendocrine tumours

BACKGROUND

Neuroendocrine tumours (NETs) of the stomach and duodenum are rare, but are increasing in incidence. Optimal management of localised, low-grade gastric and duodenal NETs remains controversial.

AIMS

To systematically review recent literature that has evaluated the management of localised low-grade gastric and duodenal NETs.

METHODS

A systematic literature search was conducted. Articles were screened and eligible articles fully assessed. Additional articles were identified through the included articles' reference lists.

RESULTS

Several relevant retrospective case series were identified, but there was considerable heterogeneity between studies and they reported a variety of parameters. Type I gastric NETs had an excellent prognosis and conservative management approaches such as endoscopic surveillance/resection were appropriate in most cases. Many type III gastric NETs were low grade and appeared to have a better prognosis than has previously been appreciated. Endoscopic rather than surgical resection was therefore effective in some patients who had small, low-grade tumours. Duodenal NETs were more heterogenous. Endoscopic resection was generally safe and effective in patients who had small, low-grade, nonfunctional, non-ampullary tumours. However, some patients, especially those with larger or ampullary duodenal NETs, required surgical resection.

CONCLUSIONS

Most type I gastric NETs behave indolently and surgical resection is only rarely indicated. Some type III gastric and duodenal NETs have a worse prognosis, but selected patients who have small, localised, nonfunctional, low-grade tumours are adequately and safely treated by endoscopic resection. Due to the complexity of this area, a multidisciplinary approach to management is strongly recommended.

The band and slough technique is effective for management of diminutive type 1 gastric and duodenal neuroendocrine tumors

Background and study aims  Endoscopic resection is recommended as initial treatment for early-stage gastric and duodenal neuroendocrine tumors (G-NETs and D-NETs). However, it can cause serious adverse events. We aimed to evaluate the efficacy and safety of the band and slough (BAS) technique as a novel and less aggressive endoscopic therapy for management of such tumors. Four patients, three diagnosed with < 10-mm D-NET and one with 10-mm type I G-NET, were treated with the BAS technique without endoscopic resection. Initial follow-up endoscopy at 3 months was done to assess for residual tumor. Subsequent endoscopic surveillance was performed. After one session of banding, all patients achieved complete remission at 3-month follow-up. No tumor recurrence was detected on repeat biopsy at 12-month surveillance endoscopy. None of the patients developed any adverse events including bleeding or perforation. The BAS technique may prove to be a safe and effective endoscopic therapy for diminutive, non-metastatic type 1 G-NETs and D-NETs. Studies of larger scale and longer follow-up periods are needed to corroborate these findings.

Natural History of Non-Alcoholic Fatty Liver Disease: A Study With Paired Liver Biopsies

BACKGROUND

Although there is unequivocal evidence for progression of nonalcoholic steatohepatitis (NASH) to cirrhosis, there is uncertainty with regard to the progression to nonalcoholic fatty liver (NAFL) and NASH.

AIMS

We investigated the rate of progression to NASH and advanced fibrosis in patients with nonalcoholic fatty liver disease (NAFLD) and assessed the factors associated with such progression.

METHODS

Histological assessment was performed in 36 patients with NAFLD with paired liver biopsies (≥1 year apart; median, 3.8 years; range, 1-10.33 years). NASH Clinical Research Network (NASH CRN) criteria were used to assess NAFLD Activity Score (NAS).

RESULTS

At baseline, 26 (72%) patients had NAFL and 10 (28%) patients had NASH. At follow-up, 27% NAFL progressed to NASH (NAS score ≥5), and 50% of patients with NASH no longer met the criteria of NASH. Fibrosis progressed in 15 (42%), regressed in 9 (25%), and remained stable in 12 (33%) patients overall. Thirty-five percent of patients with NAFL had fibrosis progression. The incidence of type 2 diabetes mellitus (T2DM) was higher in patients with NASH versus NAFL (40% vs. 27%). Both at the time of baseline and follow-up, liver biopsies, composite models of noninvasive scores such as Fibrosis-4 (FIB-4) score and NAFLD fibrosis score, and ratio of aspartate aminotransferase (AST) to alanine aminotransferase (ALT) were all significantly higher in progressors than in nonprogressors.

CONCLUSIONS

NAFLD is a dynamic liver disease with varying degrees of progression and regression. T2DM was strongly associated with fibrosis progression. Noninvasive fibrosis scores such as AST/ALT ratio, FIB-4 score, and NAFLD fibrosis score can identify those at risk of fibrosis progression.

Overview of the Pathogenesis, Genetic, and Non-Invasive Clinical, Biochemical, and Scoring Methods in the Assessment of NAFLD

Nonalcoholic fatty liver disease (NAFLD) is the most prevalent chronic liver disease worldwide. It represents a range of disorders, including simple steatosis, nonalcoholic steatohepatitis (NASH), and liver cirrhosis, and its prevalence continues to rise. In some cases, hepatocellular carcinoma (HCC) may develop. The develop;ment of non-invasive diagnostic and screening tools is needed, in order to reduce the frequency of liver biopsies. The most promising methods are those able to exclude advanced fibrosis and quantify steatosis. In this study, new perspective markers for inflammation, oxidative stress, apoptosis, and fibrogenesis; emerging scoring models for detecting hepatic steatosis and fibrosis; and new genetic, epigenetic, and multiomic studies are discussed. As isolated biochemical parameters are not specific or sensitive enough to predict the presence of NASH and fibrosis, there is a tendency to use various markers and combine them into mathematical algorithms. Several predictive models and scoring systems have been developed. Current data suggests that panels of markers (NAFLD fibrosis score, Fib-4 score, BARD score, and others) are useful diagnostic modalities to minimize the number of liver biopsies. The review unveils pathophysiological aspects related to new trends in current non-invasive biochemical, genetic, and scoring methods, and provides insight into their diagnostic accuracies and suitability in clinical practice.