EUS-guided confocal laser endomicroscopy: prediction of dysplasia in intraductal papillary mucinous neoplasms (with video).

BACKGROUND AND AIMS: Previous studies have validated EUS-guided needle-based confocal laser endomicroscopy (nCLE) diagnosis of intraductal papillary mucinous neoplasms (IPMNs). We sought to derive EUS-guided nCLE criteria for differentiating IPMNs with high-grade dysplasia/adenocarcinoma (HGD-Ca) from those with low/intermediate-grade dysplasia (LGD). METHODS: We performed a post hoc analysis of consecutive IPMNs with a definitive diagnosis from a prospective study evaluating EUS-guided nCLE in the diagnosis of pancreatic cysts. Three internal endosonographers reviewed all nCLE videos for the patients and identified potential discriminatory EUS-guided nCLE variables to differentiate HGD-Ca from LGD IPMNs (phase 1). Next, an interobserver agreement (IOA) analysis of variables from phase 1 was performed among 6 blinded external nCLE experts (phase 2). Last, 7 blinded nCLE-naïve observers underwent training and quantified variables with the highest IOA from phase 2 using dedicated software (phase 3). RESULTS: Among 26 IPMNs (HGD-Ca in 16), the reference standard was surgical histopathology in 24 and cytology confirmation of metastatic liver lesions in 2 patients. EUS-guided nCLE characteristics of increased papillary epithelial "width" and "darkness" were the most sensitive variables (90%; 95% confidence interval [CI], 84%-94% and 91%; 95% CI, 85%-95%, respectively) and accurate (85%; 95% CI, 78%-90% and 84%; 95% CI, 77%-89%, respectively) with substantial (κ = 0.61; 95% CI, 0.51-0.71) and moderate (κ = 0.55; 95% CI, 0.45-0.65) IOAs for detecting HGD-Ca, respectively (phase 2). Logistic regression models were fit for the outcome of HGD-Ca as predictor variables (phase 3). For papillary width (cut-off ≥50 µm), the sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) for detection of HGD-Ca were 87.5% (95% CI, 62%-99%), 100% (95% CI, 69%-100%), and 0.95, respectively. For papillary darkness (cut-off ≤90 pixel intensity), the sensitivity, specificity, and AUC for detection of HGD-Ca were 87.5% (95% CI, 62%-99%), 100% (95% CI, 69%-100%), and 0.90, respectively. CONCLUSIONS: In this derivation study, quantification of papillary epithelial width and darkness identified HGD-Ca in IPMNs with high accuracy. These quantifiable variables can be used in multicenter studies for risk stratification of IPMNs. (Clinical trial registration number: NCT02516488.).

Index admission cholecystectomy for acute biliary pancreatitis favorably impacts outcomes of hospitalization in cirrhosis.

BACKGROUND AND AIM: Despite higher rates of gallstones in patients with cirrhosis, there are no population-based studies evaluating outcomes of acute biliary pancreatitis (ABP). Therefore, we sought to evaluate the predictors of early readmission and mortality in this high-risk population. METHODS: We utilized the Nationwide Readmission Database (2011-2014) to evaluate all adults admitted with ABP. Multivariable logistic regression models were used to assess independent predictors for 30-day readmission, index admission mortality, and calendar year mortality. RESULTS: Among 184 611 index admissions with ABP, 4344 (2.4%) subjects had cirrhosis (1649 with decompensation). Subjects with cirrhosis, when compared with those without, incurred higher rates of 30-day readmission (20.9% vs 11.2%; P < 0.001), index mortality (2.0% vs 1.0%; P < 0.001), and calendar year mortality (4.2% vs 0.9%; P < 0.001). Decompensation in cirrhosis was associated with significantly fewer cholecystectomies (26.7% vs 60.2%; P < 0.001) and endoscopic retrograde cholangiopancreatographies (23.3% vs 29.9%; P < 0.001). Multivariate analysis revealed that severe acute pancreatitis (odds ratio [OR]: 14.8; 95% confidence interval [CI]: 5.3, 41.2), sepsis (OR: 12.6; 95% CI: 5.8, 27.4), and decompensation (OR: 3.1; 96% CI: 1.4, 6.6) were associated with increased index admission mortality. Decompensated cirrhosis (OR: 1.8; 95% CI: 1.1, 3.0) and 30-day readmission (OR: 5.6; 95% CI: 3.3, 9.5) were predictors of calendar year mortality. However, index admission cholecystectomy was associated with decreased 30-day readmissions (OR: 0.6; 95% CI: 0.4, 0.7) and calendar year mortality (OR: 0.44; 95% CI: 0.25, 0.78). CONCLUSIONS: The presence of cirrhosis adversely impacts hospital outcomes of patients with ABP. Among modifiable factors, index admission cholecystectomy portends favorable prognosis by reducing risk of early readmission and consequent calendar year mortality.

Endoscopic intervention and cholecystectomy in pregnant women with acute biliary pancreatitis decrease early readmissions.

BACKGROUND AND AIMS: Acute biliary pancreatitis (ABP) is associated with increased rates of morbidity in pregnancy. Because there is a paucity of population-based studies evaluating ABP in pregnancy, we sought to investigate clinical outcomes in hospitalized pregnant women on a national level. METHODS: By using the Nationwide Readmission Database (2011-2014), we identified all women (age ≥18 years) with an index admission for ABP in the United States. Multivariate and propensity-score matched analyses were performed to evaluate the impact of pregnancy on the clinical outcomes of early readmission and severe acute pancreatitis (SAP) in ABP. RESULTS: There were 7787 hospitalizations for ABP in pregnant women during the study period. The rate of 30-day readmission was 16.26%; 57% of these early readmissions were due to adverse events of ABP. Compared with nonpregnant women with ABP, ERCP (21.1% vs 25.2%; P < .001) and cholecystectomy (52.8% vs 55.2%; P = .02) were performed less frequently during pregnancy. Propensity-score matched analysis revealed an increased risk of 30-day readmissions in pregnancy (odds ratio [OR], 1.96; 95% confidence interval [CI], 1.67-2.30), whereas there was no difference in the risk of SAP (OR, 1.09; 95% CI, 0.76-1.57). Multivariate analysis demonstrated that weekend admission (OR, 1.40; 95% CI, 1.10-1.79) and >1 week of hospitalization (OR, 1.75; 95% CI, 1.24-2.48) increased the risk of 30-day readmission, whereas ERCP (OR, 0.40; 95% CI, 0.27-0.57) and cholecystectomy (OR, 0.13; 95% CI, 0.10-0.18) reduced the odds of early readmission in pregnancy. CONCLUSIONS: Pregnant women with ABP less frequently undergo timely endoscopic biliary decompression and cholecystectomy. These modifiable factors can potentially lower early readmissions in pregnant women.

Improving Pancreatic Cyst Management: Artificial Intelligence-Powered Prediction of Advanced Neoplasms through Endoscopic Ultrasound-Guided Confocal Endomicroscopy.

Despite the increasing rate of detection of incidental pancreatic cystic lesions (PCLs), current standard-of-care methods for their diagnosis and risk stratification remain inadequate. Intraductal papillary mucinous neoplasms (IPMNs) are the most prevalent PCLs. The existing modalities, including endoscopic ultrasound and cyst fluid analysis, only achieve accuracy rates of 65-75% in identifying carcinoma or high-grade dysplasia in IPMNs. Furthermore, surgical resection of PCLs reveals that up to half exhibit only low-grade dysplastic changes or benign neoplasms. To reduce unnecessary and high-risk pancreatic surgeries, more precise diagnostic techniques are necessary. A promising approach involves integrating existing data, such as clinical features, cyst morphology, and data from cyst fluid analysis, with confocal endomicroscopy and radiomics to enhance the prediction of advanced neoplasms in PCLs. Artificial intelligence and machine learning modalities can play a crucial role in achieving this goal. In this review, we explore current and future techniques to leverage these advanced technologies to improve diagnostic accuracy in the context of PCLs.

A Comparison of Machine Learning Methods and Conventional Logistic Regression for the Prediction of In-Hospital Mortality in Acute Biliary Pancreatitis.

OBJECTIVES: For population databases, multivariable regressions are established analytical standards. The utilization of machine learning (ML) in population databases is novel. We compared conventional statistical methods and ML for predicting mortality in biliary acute pancreatitis (biliary AP). METHODS: Using the Nationwide Readmission Database (2010-2014), we identified patients (age ≥18 years) with admissions for biliary AP. These data were randomly divided into a training (70%) and test set (30%), stratified by the outcome of mortality. The accuracy of ML and logistic regression models in predicting mortality was compared using 3 different assessments. RESULTS: Among 97,027 hospitalizations for biliary AP, mortality rate was 0.97% (n = 944). Predictors of mortality included severe AP, sepsis, increasing age, and nonperformance of cholecystectomy. Assessment metrics for predicting the outcome of mortality, the scaled Brier score (odds ratio [OR], 0.24; 95% confidence interval [CI], 0.16-0.33 vs 0.18; 95% CI, 0.09-0.27), F-measure (OR, 43.4; 95% CI, 38.3-48.6 vs 40.6; 95% CI, 35.7-45.5), and the area under the receiver operating characteristic (OR, 0.96; 95% CI, 0.94-0.97 vs 0.95; 95% CI, 0.94-0.96) were comparable between the ML and logistic regression models, respectively. CONCLUSIONS: For population databases, traditional multivariable analysis is noninferior to ML-based algorithms in predictive modeling of hospital outcomes for biliary AP.

Review of current and evolving clinical indications for endoscopic ultrasound.

For the first several years after its development, endoscopic ultrasound (EUS) was primarily limited to identification of pancreatic malignancies. Since this time, the field of EUS has advanced at a tremendous speed in terms of additional clinical diagnostic and therapeutic uses. The combination of ultrasound with endoscopy provides a unique interventional modality that is a minimally invasive alternative to various surgical interventions. Given the expanding recommended indications for EUS, this article will serve to review the most common uses with supporting evidence, while also exploring innovative endeavors that may soon become common clinical practice.

A Prospective Blinded Study Evaluating the Role of Endoscopic Ultrasound before Endoscopic Retrograde Cholangiopancreatography in the Setting of "Positive" Intraoperative Cholangiogram during Cholecystectomy.

During laparoscopic cholecystectomy, intraoperative cholangiography (IOC) is used to identify common bile duct (CBD) stones. In patients whose IOC is suspicious for stones, endoscopic retrograde cholangiopancreatography (ERCP) is the modality of choice for stone removal. However, IOC has a false positive rate of 30 to 60 per cent, and ERCP adverse events may occur in 11 per cent of patients. Endoscopic ultrasound (EUS) may serve as a noninvasive means of diagnosing suspected CBD stones. This study sought to assess the role of EUS in predicting the likelihood of choledocholithiasis at ERCP in patients found to have a positive IOC. This was a prospective blinded study of EUS before ERCP in patients with a positive IOC. Recruited subjects who underwent cholecystectomy and had an IOC with suspicion for obstruction were referred for ERCP within one month of their procedure. In patients with a positive IOC, EUS had a positive predictive value of 95 per cent in detecting choledocholithiasis. IOC with single or multiple filling defects more often correlated to the presence of CBD stones. At ERCP, choledocholithiasis was present in 65 per cent of patients who had an IOC suspicious for CBD stones. EUS should be used as a noninvasive method to correctly identify retained CBD stones in low-to-moderate risk patients with a positive IOC.

Endoscopic closure of persistent gastric leak and fistula following laparoscopic sleeve gastrectomy.

INTRODUCTION: Conservative management for gastric leak and fistulae after laparoscopic sleeve gastrectomy (LSG) often results in prolonged hospitalization as well as requirement of TPN or recurrent surgery (Casella et al., 2009) [1]. Endoscopically-placed stents are an additional non-invasive method, but are associated with the complication of stent migration in up to 50% of cases (Casella and co-workers, 2009) [1,4]. As other non-invasive means of treatment are absent, we believe this case demonstrates a new technique for multiple gastric leaks following LSG in patients without sepsis or peritonitis. PRESENTATION OF CASE: A patient developed a staple line gastric leak that persisted for 10 weeks following LSG despite multiple modalities of treatment. She refused to undergo stent placement, so via esophagogastroduodenoscopy (EGD), fistula margins were cauterized with argon plasma coagulation and a fibrin sealant was injected to include the surrounding area. Endoclips were placed along the fistula tracts. A repeat procedure was required. Follow up imaging confirmed resolution of gastric leak and patient did not experience additional complications. DISCUSSION: The patient was able to discontinue TPN and return to an oral diet. Both procedures were well tolerated and did not require hospitalization. CONCLUSION: Endoscopic management of multiple gastric leaks and fistulae using fibrin seal, endoclips, and cauterization appears to be a promising noninvasive form of treatment with a lower associated morbidity and shortened hospitalization.