Lactated Ringer's or Normal Saline for Initial Resuscitation in Patients Hospitalized With Acute Pancreatitis: A Retrospective Database Analysis.

OBJECTIVES: Fluid resuscitation is required in acute pancreatitis (AP) to prevent hypovolemia and organ hypoperfusion. Lactated Ringer's (LR) is a buffered crystalloid with possible advantages in AP versus normal saline (NS). We aim to assess outcomes in patients hospitalized with AP based on fluid used for resuscitation. METHODS: In this retrospective analysis, we identified hospital admissions to Veterans Affairs facilities for AP from 2011 to 2017 and grouped by initial resuscitation fluid: LR versus NS. Outcomes included major complications and mortality at 30 and 365 days. Multivariable models were used to adjust for confounding variables. RESULTS: A total of 20,049 admissions were included in the study, of which 10% received LR as initial fluid. After adjustment for all available confounders, resuscitation with LR was associated with lower 1-year mortality compared with NS (adjusted odds ratio, 0.61 [95% confidence interval, 0.50-0.76]). Major complication and early mortality were similar between groups. CONCLUSIONS: In this study, we demonstrate an association between use of LR as initial resuscitation fluid and reduced 1-year mortality in a large retrospective sample of veterans hospitalized with AP. These results support the use of LR for resuscitation for most patients hospitalized with AP.

Intra-ampullary papillary-tubular neoplasm.

Video 1Cholangioscopic examination of the ampullary channel and extrahepatic bile duct.

Cholangioscopy in primary sclerosing cholangitis: a case series of benign features.

BACKGROUND AND AIMS: Cholangioscopy is useful in establishing a visual diagnosis of cholangiocarcinoma (CCA), but this is harder to achieve in primary sclerosing cholangitis (PSC) because of the stricture-forming nature of the disease. Furthermore, it can be harder to differentiate malignant from benign features of the underlying inflammation. This case series demonstrates the varied features of nonmalignant inflammatory findings in PSC. METHODS: A single experienced endoscopist performed cholangioscopy for PSC cases referred for ERCP. RESULTS: Cholangioscopy in these 5 cases without CCA demonstrated the features of acute and chronic inflammation, acute inflammatory mass, dominant stricture, acute cholangitis in a duct with features of chronic inflammation with a large pigmented stone, and fibrostenotic disease. Cholangioscopic maneuvers such as advancement across strictures after balloon dilation, targeted mucosal biopsy, and electrohydraulic lithotripsy (EHL) of impacted stones are demonstrated. The relevant radiographic and histopathologic features of the disease accompany each case description. Regarding long-term prognosis, 1 case of acute inflammatory mass and a case of worsening liver function required a liver transplant evaluation, whereas the other 3 cases remain stable. CONCLUSIONS: Cholangioscopic features of benign disease in PSC are varied. Knowledge of these features is essential in differentiating between benign and malignant findings. These features, combined with biopsy and cytology evaluation, can help in tailoring management in patients with benign PSC.

Cholangioscopy in primary sclerosing cholangitis: a case series of dominant strictures and cholangiocarcinoma.

BACKGROUND AND AIMS: Prompt and accurate differentiation of benign and malignant strictures in primary sclerosing cholangitis (PSC) is crucial. ERCP with brush cytology, the most common modality to achieve this, is hindered by a low diagnostic yield. Cholangioscopy can overcome this limitation by establishing a visual diagnosis based on the characteristic morphologic features of cholangiocarcinoma (CCA) and can aid in targeted biopsies of suspicious lesions. However, its role in PSC remains unclear. This case series demonstrates the performance of the latest generation of single-operator cholangioscope for this indication. METHODS: A single experienced endoscopist performed cholangioscopy for PSC cases referred for ERCP. RESULTS: Cholangioscopies of patients 1 to 3 demonstrate the features of extrahepatic duct dominant strictures (DS) and the cholangioscopic maneuvers undertaken in these cases, including advancement across the DS after balloon dilation, biopsy of the DS, and electrohydraulic lithotripsy of impacted stones. Cholangioscopies of patients 4 to 6 demonstrate the varied features of CCA ranging from focal stricture with tumor vessels, papillary frond-like projections, and features of an intraductal papillary biliary neoplasm. Also shown are the radiographic and histopathologic features of the disease. CONCLUSIONS: Cholangioscopy allowed us to identify morphologic features of both malignancy and benign disease in PSC in the setting of extrahepatic duct strictures, and we were able to obtain adequate targeted tissue samples for histopathologic confirmation.

Predicting the Probability of Chlamydia Reinfection in African American Women Using Immunologic and Genetic Determinants in a Bayesian Model.

BACKGROUND: African Americans have the highest rates of Chlamydia trachomatis (CT) infection in the United States and also high reinfection rates. The primary objective of this study was to develop a Bayesian model to predict the probability of CT reinfection in African American women using immunogenetic data. METHODS: We analyzed data from a cohort of CT-infected African American women enrolled at the time they returned to a clinic in Birmingham, AL, for the treatment of a positive routine CT test result. We modeled the probability of CT reinfection within 6 months after treatment using logistic regression in a Bayesian framework. Predictors of interest were presence or absence of an HLA-DQB1*06 allele and CT-specific CD4+ IFN-γ response, both of which we had previously reported were independently associated with CT reinfection risk. RESULTS: Among 99 participants evaluated, the probability of reinfection for those with a CT-specific CD4+ IFN-γ response and no HLA-DQB1*06 alleles was 14.1% (95% credible interval [CI], 3.0%-45.0%), whereas the probability of reinfection for those without a CT-specific CD4+ IFN-γ response and at least one HLA-DQB1*06 allele was 61.5% (95% CI, 23.1%-89.7%). CONCLUSIONS: Our model demonstrated that presence or absence of an HLA-DQB1*06 allele and CT-specific CD4+ IFN-γ response can have an impact on the predictive probability of CT reinfection in African American women.

Effects of Image Quantity and Image Source Variation on Machine Learning Histology Differential Diagnosis Models.

AIMS: Histology, the microscopic study of normal tissues, is a crucial element of most medical curricula. Learning tools focused on histology are very important to learners who seek diagnostic competency within this important diagnostic arena. Recent developments in machine learning (ML) suggest that certain ML tools may be able to benefit this histology learning platform. Here, we aim to explore how one such tool based on a convolutional neural network, can be used to build a generalizable multi-classification model capable of classifying microscopic images of human tissue samples with the ultimate goal of providing a differential diagnosis (a list of look-alikes) for each entity. METHODS: We obtained three institutional training datasets and one generalizability test dataset, each containing images of histologic tissues in 38 categories. Models were trained on data from single institutions, low quantity combinations of multiple institutions, and high quantity combinations of multiple institutions. Models were tested against withheld validation data, external institutional data, and generalizability test images obtained from Google image search. Performance was measured with macro and micro accuracy, sensitivity, specificity, and f1-score. RESULTS: In this study, we were able to show that such a model's generalizability is dependent on both the training data source variety and the total number of training images used. Models which were trained on 760 images from only a single institution performed well on withheld internal data but poorly on external data (lower generalizability). Increasing data source diversity improved generalizability, even when decreasing data quantity: models trained on 684 images, but from three sources improved generalization accuracy between 4.05% and 18.59%. Maintaining this diversity and increasing the quantity of training images to 2280 further improved generalization accuracy between 16.51% and 32.79%. CONCLUSIONS: This pilot study highlights the significance of data diversity within such studies. As expected, optimal models are those that incorporate both diversity and quantity into their platforms.s.

Whole exome sequencing of a gut-associated lymphoid tissue neoplasm points to precursor or early form of sporadic colon carcinoma.

Gut-associated lymphoid tissue (GALT) carcinoma is a colorectal neoplasm characterized by cystically dilated neoplastic glands that extend into prominent, well-circumscribed submucosal lymphoid tissue. Although often subtle, lamina propria between and around the neoplastic glands (identified by plasma cells, scattered eosinophils, etc.) is frequent in cases with classic morphology, arguing (at least in such cases) in favor of adenoma extending into lymphoglandular complexes rather than true invasive carcinoma. Some have postulated that the tumor arises from M-cells, specialized epithelial cells overlying GALT, and others have suggested it represents a unique pathway to carcinoma, specific to the environmental conditions of epithelium overlying lymphoid tissue. Although both hypotheses are intriguing, definitive phenotypic and genetic support is currently lacking. To address these possibilities, we undertook whole exome sequencing and immunohistochemical characterization of a GALT neoplasm recently identified on our clinical service. We discovered well-known mutations in both APC and KRAS, as well as mutations in several Wnt pathway components (MED12, BCL9L, RFX4, DACT3). No immunohistochemical expression of GP2, a marker of M-cell differentiation, was identified. Expression of CDX2, SATB2, and the DNA mismatch repair proteins was observed, while expression of both CK7 and CK20 was absent. No PD-L1 expression was present on tumor cells, but PD-L1 expression was noted in a subset of tumor-adjacent mononuclear cells. Overall, the findings suggest that GALT neoplasms, although morphologically distinct, may be a precursor or early form of typical sporadic colon carcinoma.

Dynamic Positron Emission Tomography/Computed Tomography Imaging Correlate of Nonalcoholic Steatohepatitis.

Nonalcoholic steatohepatitis (NASH) is a severe form of nonalcoholic fatty liver disease characterized by lobular inflammation and hepatocyte injury and is a key determinant of clinical outcome.1 Liver biopsy remains the gold standard for diagnosis but is limited by risks of the procedure and interobserver variability. Although magnetic resonance imaging (MRI)-based technology may provide novel means to identify NASH,2 there remains a significant need for other modalities to diagnose NASH noninvasively. Glucose transport, an integral tissue process altered in NASH,3 is measurable with 18F-fluorodeoxyglucose positron emission tomography (FDG PET). Because unenhanced computed tomography (CT) scan can detect hepatic steatosis quite reliably,4 and PET combines unenhanced CT for attenuation correction, we hypothesized that measurement of the combination of glucose transport by PET and steatosis by CT could yield a reliable radiologic correlate of NASH.

Mycoplasma genitalium Infection in Young Women Without Urogenital Symptoms Presenting to a Community-Based Emergency Department in Birmingham, Alabama.

ABSTRACT: We used the Food and Drug Administration-cleared Aptima Mycoplasma genitalium assay to evaluate for M. genitalium infection among young women without urogenital symptoms presenting to a community-based emergency department in Birmingham, Alabama, between August 2016 to August 2019 for evaluation of nongynecological concerns. M. genitalium was detected in 23 (14.8%) of 155 women.

Medical Student Attitudes toward USMLE Step 1 and Health Systems Science - A Multi-Institutional Survey.

Phenomenon: Because of its importance in residency selection, the United States Medical Licensing Examination Step 1 occupies a critical position in medical education, stimulating national debate about appropriate score use, equitable selection criteria, and the goals of undergraduate medical education. Yet, student perspectives on these issues and their implications for engagement with health systems science-related curricular content are relatively underexplored. Approach: We conducted an online survey of medical students at 19 American allopathic medical schools from March-July, 2019. Survey items were designed to elicit student opinions on the Step 1 examination and the impact of the examination on their engagement with new, non-test curricular content related to health systems science. Findings: A total of 2856 students participated in the survey, representing 23.5% of those invited. While 87% of students agreed that doing well on the Step 1 exam was their top priority, 56% disagreed that studying for Step 1 had a positive impact on engagement in the medical school curriculum. Eighty-two percent of students disagreed that Step 1 scores should be the top item residency programs use to offer interviews. When asked whether Step 1 results should be reported pass/fail with no numeric score, 55% of students agreed, while 33% disagreed. The majority of medical students agreed that health systems science topics were important but disagreed that studying for Step 1 helped learn this content. Students reported being more motivated to study a topic if it was on the exam, part of a course grade, prioritized by residency program directors, or if it would make them a better physician in the future. Insights: These results confirm the primacy of the United States Medical Licensing Examination Step 1 exam in preclinical medical education and demonstrate the need to balance the objectives of medical licensure and residency selection with the goals of the broader medical profession. The survey responses suggest several potential solutions to increase student engagement in health systems science curricula which may be especially important after Step 1 examination results are reported as pass/fail.

Improving the accuracy of gastrointestinal neuroendocrine tumor grading with deep learning.

The Ki-67 index is an established prognostic factor in gastrointestinal neuroendocrine tumors (GI-NETs) and defines tumor grade. It is currently estimated by microscopically examining tumor tissue single-immunostained (SS) for Ki-67 and counting the number of Ki-67-positive and Ki-67-negative tumor cells within a subjectively picked hot-spot. Intraobserver variability in this procedure as well as difficulty in distinguishing tumor from non-tumor cells can lead to inaccurate Ki-67 indices and possibly incorrect tumor grades. We introduce two computational tools that utilize Ki-67 and synaptophysin double-immunostained (DS) slides to improve the accuracy of Ki-67 index quantitation in GI-NETs: (1) Synaptophysin-KI-Estimator (SKIE), a pipeline automating Ki-67 index quantitation via whole-slide image (WSI) analysis and (2) deep-SKIE, a deep learner-based approach where a Ki-67 index heatmap is generated throughout the tumor. Ki-67 indices for 50 GI-NETs were quantitated using SKIE and compared with DS slide assessments by three pathologists using a microscope and a fourth pathologist via manually ticking off each cell, the latter of which was deemed the gold standard (GS). Compared to the GS, SKIE achieved a grading accuracy of 90% and substantial agreement (linear-weighted Cohen's kappa 0.62). Using DS WSIs, deep-SKIE displayed a training, validation, and testing accuracy of 98.4%, 90.9%, and 91.0%, respectively, significantly higher than using SS WSIs. Since DS slides are not standard clinical practice, we also integrated a cycle generative adversarial network into our pipeline to transform SS into DS WSIs. The proposed methods can improve accuracy and potentially save a significant amount of time if implemented into clinical practice.

AGER1 downregulation associates with fibrosis in nonalcoholic steatohepatitis and type 2 diabetes.

Type 2 diabetes is clinically associated with progressive necroinflammation and fibrosis in nonalcoholic steatohepatitis (NASH). Advanced glycation end-products (AGEs) accumulate during prolonged hyperglycemia, but the mechanistic pathways that lead to accelerated liver fibrosis have not been well defined. In this study, we show that the AGEs clearance receptor AGER1 was downregulated in patients with NASH and diabetes and in our NASH models, whereas the proinflammatory receptor RAGE was induced. These findings were associated with necroinflammatory, fibrogenic, and pro-oxidant activity via the NADPH oxidase 4. Inhibition of AGEs or RAGE deletion in hepatocytes in vivo reversed these effects. We demonstrate that dysregulation of NRF2 by neddylation of cullin 3 was linked to AGER1 downregulation and that induction of NRF2 using an adeno-associated virus-mediated approach in hepatocytes in vivo reversed AGER1 downregulation, lowered the level of AGEs, and improved proinflammatory and fibrogenic responses in mice on a high AGEs diet. In patients with NASH and diabetes or insulin resistance, low AGER1 levels were associated with hepatocyte ballooning degeneration and ductular reaction. Collectively, prolonged exposure to AGEs in the liver promotes an AGER1/RAGE imbalance and consequent redox, inflammatory, and fibrogenic activity in NASH.

Squamous Cell Papilloma of the Esophagus: A Case Series Highlighting Endoscopic and Histologic Features.

Esophageal squamous papillomas are rare epithelial lesions typically discovered incidentally during EGD. Their prevalence is estimated to be less than 0.01% in the general population. We present three cases of esophageal squamous papillomas identified histologically. It may be possible to identify these lesions macroscopically. One study provided a positive predictive value of 88% for squamous papilloma utilizing the triad of exophytic growth, wart-like projections, and surface vessel crossing seen on narrow band imaging during endoscopy. The etiology is unclear. Chronic mucosal irritation from GERD or esophagitis is the prevailing theory of pathogenesis, but HPV has been detected in some lesions. The malignant potential of these lesions is considered controversial. There are documented cases demonstrating complications with squamous cell carcinoma, so we recommend removal of all esophageal squamous papillomas; however, the small absolute number of cases documented in the literature makes drawing any associations or conclusions between esophageal squamous papillomas and squamous cell carcinoma difficult. Further research is needed regarding treatment and surveillance. This case series helps contribute to the small but growing literature of this rare finding.

Limited Number of Cases May Yield Generalizable Models, a Proof of Concept in Deep Learning for Colon Histology.

BACKGROUND: Little is known about the effect of a minimum number of slides required in generating image datasets used to build generalizable machine-learning (ML) models. In addition, the assumption within deep learning is that the increased number of training images will always enhance accuracy and that the initial validation accuracy of the models correlates well with their generalizability. In this pilot study, we have been able to test the above assumptions to gain a better understanding of such platforms, especially when data resources are limited. METHODS: Using 10 colon histology slides (5 carcinoma and 5 benign), we were able to acquire 1000 partially overlapping images (Dataset A) that were then trained and tested on three convolutional neural networks (CNNs), ResNet50, AlexNet, and SqueezeNet, to build a large number of unique models for a simple task of classifying colon histopathology into benign and malignant. Different quantities of images (10-1000) from Dataset A were used to construct >200 unique CNN models whose performances were individually assessed. The performance of these models was initially assessed using 20% of Dataset A's images (not included in the training phase) to acquire their initial validation accuracy (internal accuracy) followed by their generalization accuracy on Dataset B (a very distinct secondary test set acquired from public domain online sources). RESULTS: All CNNs showed similar peak internal accuracies (>97%) from the Dataset A test set. Peak accuracies for the external novel test set (Dataset B), an assessment of the ability to generalize, showed marked variation (ResNet50: 98%; AlexNet: 92%; and SqueezeNet: 80%). The models with the highest accuracy were not generated using the largest training sets. Further, a model's internal accuracy did not always correlate with its generalization accuracy. The results were obtained using an optimized number of cases and controls. CONCLUSIONS: Increasing the number of images in a training set does not always improve model accuracy, and significant numbers of cases may not always be needed for generalization, especially for simple tasks. Different CNNs reach peak accuracy with different training set sizes. Further studies are required to evaluate the above findings in more complex ML models prior to using such ancillary tools in clinical settings.

Educational Case: Infectious Esophagitis.

The following fictional case is intended as a learning tool within the Pathology Competencies for Medical Education (PCME), a set of national standards for teaching pathology. These are divided into three basic competencies: Disease Mechanisms and Processes, Organ System Pathology, and Diagnostic Medicine and Therapeutic Pathology. For additional information, and a full list of learning objectives for all three competencies, see http://journals.sagepub.com/doi/10.1177/2374289517715040.1.

Educational Case: Gastric High-Grade B-Cell Lymphoma With MYC and BCL2 Gene Rearrangement (Double-Hit Lymphoma).

The following fictional case is intended as a learning tool within the Pathology Competencies for Medical Education (PCME), a set of national standards for teaching pathology. These are divided into three basic competencies: Disease Mechanisms and Processes, Organ System Pathology, and Diagnostic Medicine and Therapeutic Pathology. For additional information, and a full list of learning objectives for all three competencies, see http://journals.sagepub.com/doi/10.1177/2374289517715040.1.

A Novel Approach for Management of Bleeding Stomal Varices: A Case Report of Ultrasound-Guided Percutaneous Sclerotherapy.

Ileostomy variceal bleeds can be a serious complication in patients with cirrhosis and ileostomy but make up a small portion of total variceal bleeds. Multiple modalities have been described as therapeutic options for stomal variceal bleeding, but an optimal intervention has yet to be established. We present a case of a 51-year-old patient with preserved ejection fraction heart failure, hepatitis C cirrhosis, recent esophageal varices banding, and colectomy with ileostomy who developed bleeding ileostomy varices that were effectively treated under direct ultrasound-guided percutaneous injection of sodium tetradecyl sulfate to the feeding superior mesenteric venous flow. The patient did not have a recurrence of bleeding at 7-month follow-up. We consider direct ultrasound-guided percutaneous injection of sodium tetradecyl sulfate of acute bleeding stomal varices to be safe and effective in decompensated cirrhotic patients.

Mycoplasma genitalium Coinfection in Women With Chlamydia trachomatis Infection.

We evaluated the prevalence of Mycoplasma genitalium coinfection in 302 chlamydia-infected women seen at a sexually transmitted disease clinic in Birmingham, AL. M genitalium coinfection was detected in 22 (7.3%). No participant characteristics predicted coinfection. Among coinfected women, M genitalium was detected again in 6 (28.6%) of 21 women returning for a 3-month follow-up visit after azithromycin treatment.

Structural and practical identifiability of dual-input kinetic modeling in dynamic PET of liver inflammation.

Dynamic 18F-FDG PET with tracer kinetic modeling has the potential to noninvasively evaluate human liver inflammation using the FDG blood-to-tissue transport rate K 1. Accurate kinetic modeling of dynamic liver PET data and K 1 quantification requires the knowledge of dual-blood input function from the hepatic artery and portal vein. While the arterial input function can be derived from the aortic region on dynamic PET images, it is difficult to extract the portal vein input function accurately from PET images. The optimization-derived dual-input kinetic modeling approach has been proposed to overcome this problem by jointly estimating the portal vein input function and FDG tracer kinetics from time activity curve fitting. In this paper, we further characterize the model properties by analyzing the structural identifiability of the model parameters using the Laplace transform and practical identifiability using computer simulation based on fourteen patient datasets. The theoretical analysis has indicated that all the kinetic parameters of the dual-input kinetic model are structurally identifiable, though subject to local solutions. The computer simulation results have shown that FDG K 1 can be estimated reliably in the whole-liver region of interest with reasonable bias, standard deviation, and high correlation between estimated and original values, indicating of practical identifiability of K 1. The result has also demonstrated the correlation between K 1 and histological liver inflammation scores is reliable. FDG K 1 quantification by the optimization-derived dual-input kinetic model is promising for assessing liver inflammation.