Development and Validation of a Machine Learning-based, Point-of-Care Risk Calculator for Post-ERCP Pancreatitis and Prophylaxis Selection.

BACKGROUND AND AIMS: A robust model of post-ERCP pancreatitis (PEP) risk is not currently available. We aimed to develop a machine learning-based tool for PEP risk prediction to aid in clinical decision-making related to periprocedural prophylaxis selection and post-procedural monitoring. METHODS: Feature selection, model training, and validation were performed using patient-level data from 12 randomized controlled trials. A gradient-boosted machine (GBM) model was trained to estimate PEP risk and the performance of the resulting model was evaluated using the area under the receiver operating curve (AUC) with 5-fold cross-validation. A web-based clinical decision-making tool was created, and a prospective pilot study was performed using data from ERCPs performed at the Johns Hopkins Hospital over a one-month period. RESULTS: A total of 7389 patients were included in the GBM with an 8.6% rate of PEP. The model was trained on twenty PEP risk factors and 5 prophylactic interventions (rectal non-steroidal anti-inflammatory drugs [NSAID], aggressive hydration, combined rectal NSAID and aggressive hydration, pancreatic duct [PD] stenting, and combined rectal NSAID and PD stenting). The resulting GBM model had an AUC of 0.70 (65% specificity, 65% sensitivity, 95% negative predictive value, 15% positive predictive value). A total of 135 patients were included in the prospective pilot study, resulting in an AUC of 0.74. CONCLUSIONS: This study demonstrates the feasibility and utility of a novel machine learning-based PEP risk estimation tool with high negative predictive value to aid in prophylaxis selection and identify patients at low risk who may not require extended post-procedure monitoring.

Effective, safe and efficient porcine model of Forrest Ib bleeding gastric and colonic ulcers.

BACKGROUND: Developing effective gastrointestinal (GI) bleeding animal models is necessary to advance endoscopic hemostasis methods and train endoscopists on their use. Our aim, therefore, was to develop an effective and safe porcine GI bleeding model in the stomach and colon of large and small-sized oozing-type ulcers. METHODS: Gastric and colonic bleeding ulcers were created using either a hybrid endoscopic submucosal dissection (ESD) technique or a cap-assisted endoscopic mucosal resection (EMR-C) technique in 14 pigs. Prior to ulcer creation, animals were treated with either oral apixaban or intravenous (IV) unfractionated heparin anticoagulation in combination with clopidogrel and aspirin. The primary outcome was the technical success of inducing oozing-type Forrest Ib bleeding ulcers. Secondary outcomes included ulcer diameter, number, creation time and the number of complications associated with each technique. RESULTS: Using hybrid ESD and IV heparin anticoagulation, bleeding was observed in 21/23 (91.3%) gastric ulcers and 6/7 (85.7%) colonic ulcers created. The mean diameter and ulcer creation time were 2.3 ± 0.3 cm and 5.3 ± 0.5 min, respectively, for gastric ulcers and 2.2 ± 0.4 cm and 4.06 ± 0.6 min, respectively, for colonic ulcers. Using EMR-C and IV heparin anticoagulation, bleeding was observed in 14/15 (93.3%) gastric ulcers and 6/6 (100%) colonic ulcers created. The mean diameter and ulcer creation time were 0.8 ± 0.2 cm and 2.1 ± 0.5 min, respectively, for gastric ulcers and 0.7 ± 0.2 cm and 1.7 ± 0.3 min, respectively, for colonic ulcers. None of the ulcers created in animals anticoagulated with apixaban developed bleeding. None of the 14 pigs developed any complications. CONCLUSION: We have demonstrated the effectiveness and safety of a porcine GI bleeding model utilizing IV heparin anticoagulation and either hybrid ESD or EMR-C techniques to create oozing-type bleeding ulcers in the stomach and colon with customizable size.

EUS-directed transenteric ERCP-assisted internalization of a percutaneous biliary drain in Roux-en-Y anatomy.

Video 1EUS-directed transenteric ERCP-assisted internalization of percutaneous biliary drain in Roux-en-Y hepaticojejunostomy anatomy.

Rectal INdomethacin, oral TacROlimus, or their combination for the prevention of post-ERCP pancreatitis (INTRO Trial): Protocol for a randomized, controlled, double-blinded trial.

BACKGROUND: Acute pancreatitis remains the most common and morbid complication of endoscopic retrograde cholangiopancreatography (ERCP). The use of rectal indomethacin and pancreatic duct stenting has been shown to reduce the incidence and severity of post-ERCP pancreatitis (PEP), but these interventions have limitations. Recent clinical and translational evidence suggests a role for calcineurin inhibitors in the prevention of pancreatitis, with multiple retrospective case series showing a reduction in PEP rates in tacrolimus users. METHODS: The INTRO trial is a multicenter, international, randomized, double-blinded, controlled trial. A total of 4,874 patients undergoing ERCP will be randomized to receive either oral tacrolimus (5 mg) or oral placebo 1-2 h before ERCP, and followed for 30 days post-procedure. Blood and pancreatic aspirate samples will also be collected in a subset of patients to quantify tacrolimus levels. The primary outcome of the study is the incidence of PEP. Secondary endpoints include the severity of PEP, ERCP-related complications, adverse drug events, length of hospital stay, cost-effectiveness, and the pharmacokinetics, pharmacodynamics, and pharmacogenomics of tacrolimus immune modulation in the pancreas. CONCLUSIONS: The INTRO trial will assess the role of calcineurin inhibitors in PEP prophylaxis and develop a foundation for the clinical optimization of this therapeutic strategy from a pharmacologic and economic standpoint. With this clinical trial, we hope to demonstrate a novel approach to PEP prophylaxis using a widely available and well-characterized class of drugs. TRIAL REGISTRATION: NCT05252754, registered on February 14, 2022.

Weight and organ specific immune cell profiling of sleeve gastrectomy in mice.

OBJECTIVE: Sleeve gastrectomy (SG) has profound, immediate weight-loss independent effects on obesity related diabetes (T2D). Our prior studies have shown that immunologic remodeling may play a part in this metabolic improvement. However, to date, little is known about how the major immune cell populations change following SG and whether these are weight loss dependent. METHODS: Using mass cytometry with time of flight analysis (CyTOF), we broadly quantified the organ-specific immune cell repertoire induced by SG from splenic, jejunal, ileal, colonic, and hepatic lymphocyte fractions. Surgeries were performed in both diet-induced obese (DIO), insulin resistant mice and lean mice, which leads to sustained and non-sustained weight loss in SG animals compared to shams, respectively. Intergroup comparisons allow understanding of the relative contribution of diet, weight-loss, and surgery on immune profiling. Conserved immune changes represent surgery-specific, weight-independent, and diet-independent phenotypic changes. RESULTS: Initial analysis by way of visualization of t-distributed stochastic neighbor embedding analysis revealed changes in the B cell compartment following SG in both DIO and lean mice compared to Sham animals. In depth, traditional gating showed a shift within the splenic B cell compartment toward innate-like phenotype. There was a 1.3-fold reduction in follicular B cells within DIO SG (14% absolute reduction; p = 0.009) and lean SG (15% absolute reduction; p = 0.031) animals with a significant increase in innate-like B cell subsets in DIO SG mice(2.2 to 4.3-fold increase; p < 0.05). There was a similar trend toward increased innate B cell subsets in lean SG mice. There was a concomitant increase in multiple circulating immunoglobulin classes in both models. Further, lean (p = 0.009) and DIO SG animals (p = 0.015) had a conserved 5.5-fold and 5.7-fold increase, respectively, in splenic neutrophils and tendency toward M2 macrophage polarization. CONCLUSIONS: SG induces surgery-specific, weight-loss independent immune cells changes that have been previously linked to improved glucose metabolism. This immune phenotype may be a major contributor to post SG physiology. Characterizing the complex immune milieu following SG is an important step toward understanding the physiology of SG and the potential therapies therein.

Critical Adverse Impact of IL-6 in Acute Pneumovirus Infection.

Severe respiratory virus infections feature robust local host responses that contribute to disease severity. Immunomodulatory strategies that limit virus-induced inflammation may be of critical importance, notably in the absence of antiviral vaccines. In this study, we examined the role of the pleiotropic cytokine IL-6 in acute infection with pneumonia virus of mice (PVM), a natural rodent pathogen that is related to respiratory syncytial virus and that generates local inflammation as a feature of severe infection. In contrast to Influenza A, PVM is substantially less lethal in IL-6 -/- mice than it is in wild-type, a finding associated with diminished neutrophil recruitment and reduced fluid accumulation in lung tissue. Ly6Chi proinflammatory monocytes are recruited in response to PVM via a CCR2-dependent mechanism, but they are not a major source of IL-6 nor do they contribute to lethal sequelae of infection. By contrast, alveolar macrophages are readily infected with PVM in vivo; ablation of alveolar macrophages results in prolonged survival in association with a reduction in virus-induced IL-6. Finally, as shown previously, administration of immunobiotic Lactobacillus plantarum to the respiratory tracts of PVM-infected mice promoted survival in association with diminished levels of IL-6. We demonstrated in this study that IL-6 suppression is a critical feature of the protective mechanism; PVM-infected IL-6 -/- mice responded to low doses of L. plantarum, and administration of IL-6 overcame L. plantarum-mediated protection in PVM-infected wild-type mice. Taken together, these results connect the actions of IL-6 to PVM pathogenesis and suggest cytokine blockade as a potential therapeutic modality in severe infection.

A flow-cytometric method to evaluate eosinophil-mediated uptake of probiotic Lactobacillus reuteri.

Eosinophils are resident leukocytes of gut mucosa. Here we present a combined flow cytometric-antibiotic protection assay to identify mouse eosinophils capable of bacterial uptake, specifically, Gram-positive Lactobacillus reuteri, in studies performed ex vivo. The assay may be adapted for use in vivo.

SiglecF+Gr1hi eosinophils are a distinct subpopulation within the lungs of allergen-challenged mice.

Although eosinophils as a group are readily identified by their unique morphology and staining properties, flow cytometry provides an important means for identification of subgroups based on differential expression of distinct surface Ags. Here, we characterize an eosinophil subpopulation defined by high levels of expression of the neutrophil Ag Gr1 (CD45+CD11c-SiglecF+Gr1hi). SiglecF+Gr1hi eosinophils, distinct from the canonical SiglecF+Gr1- eosinophil population, were detected in allergen-challenged wild-type and granule protein-deficient (EPX-/- and MBP-1-/-) mice, but not in the eosinophil-deficient ΔdblGATA strain. In contrast to Gr1+ neutrophils, which express both cross-reacting Ags Ly6C and Ly6G, SiglecF+Gr1hi eosinophils from allergen-challenged lung tissue are uniquely Ly6G+ Although indistinguishable from the more-numerous SiglecF+Gr1- eosinophils under light microscopy, FACS-isolated populations revealed prominent differences in cytokine contents. The lymphocyte-targeting cytokines CXCL13 and IL-27 were identified only in the SiglecF+Gr1hi eosinophil population (at 3.9 and 4.8 pg/106 cells, respectively), as was the prominent proinflammatory mediator IL-13 (72 pg/106 cells). Interestingly, bone marrow-derived (SiglecF+), cultured eosinophils include a more substantial Gr1+ subpopulation (∼50%); Gr1+ bmEos includes primarily a single Ly6C+ and a smaller, double-positive (Ly6C+Ly6G+) population. Taken together, our findings characterize a distinct SiglecF+Gr1hi eosinophil subset in lungs of allergen-challenged, wild-type and granule protein-deficient mice. SiglecF+Gr1hi eosinophils from wild-type mice maintain a distinct subset of cytokines, including those active on B and T lymphocytes. These cytokines may facilitate eosinophil-mediated immunomodulatory responses in the allergen-challenged lung as well as in other distinct microenvironments.

Signaling via pattern recognition receptors NOD2 and TLR2 contributes to immunomodulatory control of lethal pneumovirus infection.

Pattern recognition receptors (PRRs) engage microbial components in the lung, although their role in providing primary host defense against respiratory virus infection is not fully understood. We have previously shown that Gram-positive Lactobacillus plantarum (Lp) administered to the respiratory tract promotes full and sustained protection in response to an otherwise lethal mouse pneumovirus (PVM) infection, a robust example of heterologous immunity. While Lp engages PRRs TLR2 and NOD2 in ex vivo signaling assays, we found that Lp-mediated protection was unimpaired in single gene-deleted TLR2(-/-) and NOD2(-/-) mice. Here we demonstrate substantial loss of Lp-mediated protection in a double gene-deleted NOD2(-/-)TLR2(-/-) strain. Furthermore, we demonstrate protection against PVM infection by administration of the bi-functional NOD2-TLR2 agonist, CL-429. The bi-functional NOD2-TLR2 ligand CL-429 not only suppresses virus-induced inflammation, it is significantly more effective at preventing lethal infection than equivalent amounts of mono-molecular TLR2 and NOD2 agonists. Interestingly, and in contrast to biochemical NOD2 and/or TLR2 agonists, Lp remained capable of eliciting primary proinflammatory responses from NOD2(-/-)TLR2(-/-) mice in vivo and from alveolar macrophages challenged ex vivo. Taken together, we conclude that coordinate engagement of NOD2 and TLR2 constitutes a key step in the genesis of Lp-mediated protection from a lethal respiratory virus infection, and represents a critical target for modulation of virus-induced inflammatory pathology.

Immortalized MH-S cells lack defining features of primary alveolar macrophages and do not support mouse pneumovirus replication.

The SV-40-transformed MH-S cell line maintains some, but not all, features of primary alveolar macrophages (AMs) from BALB/c mice. We show here that MH-S cells produce inflammatory cytokines IL-6 and CXCL10 in response to challenge with Gram-positive Lactobacillus reuteri, and to TLR2 and NOD2 ligands Pam3CSK4 and MDP, respectively. In contrast, although wild-type AMs are infected in vivo by pneumonia virus of mice (PVM), no virus replication was detected in MH-S cells. Interestingly, the surface immunophenotype of MH-S cells (CD11c(+)Siglec F(-)) differs from that of wild-type AMs (CD11c(+) Siglec F(+)) and is similar to that of immature AMs isolated from granulocyte macrophage-colony stimulating factor (GM-CSF) gene-deleted mice; AMs from GM-CSF(-/-) mice also support PVM replication. However, MH-S cells do not express the GM-CSF receptor alpha chain (CD116) and do not respond to GM-CSF. Due to these unusual features, MH-S cells should be used with caution as experimental models of AMs.

Priming of the Respiratory Tract with Immunobiotic Lactobacillus plantarum Limits Infection of Alveolar Macrophages with Recombinant Pneumonia Virus of Mice (rK2-PVM).

UNLABELLED: Pneumonia virus of mice (PVM) is a natural rodent pathogen that replicates in bronchial epithelial cells and reproduces many clinical and pathological features of the more severe forms of disease associated with human respiratory syncytial virus. In order to track virus-target cell interactions during acute infection in vivo, we developed rK2-PVM, bacterial artificial chromosome-based recombinant PVM strain J3666 that incorporates the fluorescent tag monomeric Katushka 2 (mKATE2). The rK2-PVM pathogen promotes lethal infection in BALB/c mice and elicits characteristic cytokine production and leukocyte recruitment to the lung parenchyma. Using recombinant virus, we demonstrate for the first time PVM infection of both dendritic cells (DCs; CD11c(+) major histocompatibility complex class II(+)) and alveolar macrophages (AMs; CD11c(+) sialic acid-binding immunoglobulin-like lectin F(+)) in vivo and likewise detect mKATE2(+) DCs in mediastinal lymph nodes from infected mice. AMs support both active virus replication and production of infectious virions. Furthermore, we report that priming of the respiratory tract with immunobiotic Lactobacillus plantarum, a regimen that results in protection against the lethal inflammatory sequelae of acute respiratory virus infection, resulted in differential recruitment of neutrophils, DCs, and lymphocytes to the lungs in response to rK2-PVM and a reduction from ∼ 40% to <10% mKATE2(+) AMs in association with a 2-log drop in the release of infectious virions. In contrast, AMs from L. plantarum-primed mice challenged with virus ex vivo exhibited no differential susceptibility to rK2-PVM. Although the mechanisms underlying Lactobacillus-mediated viral suppression remain to be fully elucidated, this study provides insight into the cellular basis of this response. IMPORTANCE: Pneumonia virus of mice (PVM) is a natural mouse pathogen that serves as a model for severe human respiratory syncytial virus disease. We have developed a fully functional recombinant PVM strain with a fluorescent reporter protein (rK2-PVM) that permits us to track infection of target cells in vivo. With rK2-PVM, we demonstrate infection of leukocytes in the lung, notably, dendritic cells and alveolar macrophages. Alveolar macrophages undergo productive infection and release infectious virions. We have shown previously that administration of immunobiotic Lactobacillus directly to the respiratory mucosa protects mice from the lethal sequelae of PVM infection in association with profound suppression of the virus-induced inflammatory response. We show here that Lactobacillus administration also limits infection of leukocytes in vivo and results in diminished release of infectious virions from alveolar macrophages. This is the first study to provide insight into the cellular basis of the antiviral impact of immunobiotic L. plantarum.

Immunobiotic Lactobacillus administered post-exposure averts the lethal sequelae of respiratory virus infection.

We reported previously that priming of the respiratory tract with immunobiotic Lactobacillus prior to virus challenge protects mice against subsequent lethal infection with pneumonia virus of mice (PVM). We present here the results of gene microarray which document differential expression of proinflammatory mediators in response to PVM infection alone and those suppressed in response to Lactobacillus plantarum. We also demonstrate for the first time that intranasal inoculation with live or heat-inactivated L. plantarum or Lactobacillus reuteri promotes full survival from PVM infection when administered within 24h after virus challenge. Survival in response to L. plantarum administered after virus challenge is associated with suppression of proinflammatory cytokines, limited virus recovery, and diminished neutrophil recruitment to lung tissue and airways. Utilizing this post-virus challenge protocol, we found that protective responses elicited by L. plantarum at the respiratory tract were distinct from those at the gastrointestinal mucosa, as mice devoid of the anti-inflammatory cytokine, interleukin (IL)-10, exhibit survival and inflammatory responses that are indistinguishable from those of their wild-type counterparts. Finally, although L. plantarum interacts specifically with pattern recognition receptors TLR2 and NOD2, the respective gene-deleted mice were fully protected against lethal PVM infection by L. plantarum, as are mice devoid of type I interferon receptors. Taken together, L. plantarum is a versatile and flexible agent that is capable of averting the lethal sequelae of severe respiratory infection both prior to and post-virus challenge via complex and potentially redundant mechanisms.

ATP-binding cassette transporter 1 attenuates ovalbumin-induced neutrophilic airway inflammation.

Apolipoprotein A-I (apoA-I) is an important component of high-density lipoprotein particles that mediates reverse cholesterol transport out of cells by interacting with the ATP-binding cassette transporter 1 (ABCA1). apoA-I has also been shown to attenuate neutrophilic airway inflammation in experimental ovalbumin (OVA)-induced asthma by reducing the expression of granulocyte colony-stimulating factor (G-CSF). Here, we hypothesized that overexpression of the ABCA1 transporter might similarly attenuate OVA-induced neutrophilic airway inflammation. Tie2-human ABCA1 (hABCA1) mice expressing human ABCA1 under the control of the Tie2 promoter, which is primarily expressed by vascular endothelial cells, but can also be expressed by macrophages, received daily intranasal OVA challenges, 5 d/wk for 5 weeks. OVA-challenged Tie2-hABCA1 mice had significant reductions in total bronchoalveolar lavage fluid (BALF) cells that reflected a decrease in neutrophils, as well as reductions in peribronchial inflammation, OVA-specific IgE levels, and airway epithelial thickness. The reduced airway neutrophilia in OVA-challenged Tie2-hABCA1 mice was associated with significant decreases in G-CSF protein levels in pulmonary vascular endothelial cells, alveolar macrophages, and BALF. Intranasal administration of recombinant murine G-CSF to OVA-challenged Tie2-hABCA1 mice for 5 days increased BALF neutrophils to a level comparable to that of OVA-challenged wild-type mice. We conclude that ABCA1 suppresses OVA-induced airway neutrophilia by reducing G-CSF production by vascular endothelial cells and alveolar macrophages. These findings suggest that ABCA1 expressed by vascular endothelial cells and alveolar macrophages may play important roles in attenuating the severity of neutrophilic airway inflammation in asthma.