Infectious Complications of Acute Pancreatitis Is Associated with Peripheral Blood Phagocyte Functional Exhaustion

Construction and validation of machine learning models for sepsis prediction in patients with acute pancreatitis

BACKGROUND

This study aimed to construct predictive models for the risk of sepsis in patients with Acute pancreatitis (AP) using machine learning methods and compared optimal one with the logistic regression (LR) model and scoring systems.

METHODS

In this retrospective cohort study, data were collected from the Medical Information Mart for Intensive Care III (MIMIC III) database between 2001 and 2012 and the MIMIC IV database between 2008 and 2019. Patients were randomly divided into training and test sets (8:2). The least absolute shrinkage and selection operator (LASSO) regression plus 5-fold cross-validation were used to screen and confirm the predictive factors. Based on the selected predictive factors, 6 machine learning models were constructed, including support vector machine (SVM), K-nearest neighbour (KNN), multi-layer perceptron (MLP), LR, gradient boosting decision tree (GBDT) and adaptive enhancement algorithm (AdaBoost). The models and scoring systems were evaluated and compared using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, and the area under the curve (AUC).

RESULTS

A total of 1, 672 patients were eligible for participation. In the training set, 261 AP patients (19.51%) were diagnosed with sepsis. The predictive factors for the risk of sepsis in AP patients included age, insurance, vasopressors, mechanical ventilation, Glasgow Coma Scale (GCS), heart rate, respiratory rate, temperature, SpO2, platelet, red blood cell distribution width (RDW), International Normalized Ratio (INR), and blood urea nitrogen (BUN). The AUC of the GBDT model for sepsis prediction in the AP patients in the testing set was 0.985. The GBDT model showed better performance in sepsis prediction than the LR, systemic inflammatory response syndrome (SIRS) score, bedside index for severity in acute pancreatitis (BISAP) score, sequential organ failure assessment (SOFA) score, quick-SOFA (qSOFA), and simplified acute physiology score II (SAPS II).

CONCLUSION

The present findings suggest that compared to the classical LR model and SOFA, qSOFA, SAPS II, SIRS, and BISAP scores, the machine learning model-GBDT model had a better performance in predicting sepsis in the AP patients, which is a useful tool for early identification of high-risk patients and timely clinical interventions.

Fire in the belly: A scoping review of the immunopathological mechanisms of acute pancreatitis

Introduction

Acute pancreatitis (AP) is characterised by an inflammatory response that in its most severe form can cause a systemic dysregulated immune response and progression to acute multi-organ dysfunction. The pathobiology of the disease is unclear and as a result no targeted, disease-modifying therapies exist. We performed a scoping review of data pertaining to the human immunology of AP to summarise the current field and to identify future research opportunities.

Methods

A scoping review of all clinical studies of AP immunology was performed across multiple databases. Studies were included if they were human studies of AP with an immunological outcome or intervention.

Results

205 studies met the inclusion criteria for the review. Severe AP is characterised by significant immune dysregulation compared to the milder form of the disease. Broadly, this immune dysfunction was categorised into: innate immune responses (including profound release of damage-associated molecular patterns and heightened activity of pattern recognition receptors), cytokine profile dysregulation (particularly IL-1, 6, 10 and TNF-α), lymphocyte abnormalities, paradoxical immunosuppression (including HLA-DR suppression and increased co-inhibitory molecule expression), and failure of the intestinal barrier function. Studies including interventions were also included. Several limitations in the existing literature have been identified; consolidation and consistency across studies is required if progress is to be made in our understanding of this disease.

Conclusions

AP, particularly the more severe spectrum of the disease, is characterised by a multifaceted immune response that drives tissue injury and contributes to the associated morbidity and mortality. Significant work is required to develop our understanding of the immunopathology of this disease if disease-modifying therapies are to be established.

Circulating monocytes in acute pancreatitis

Acute pancreatitis is a common gastrointestinal disease characterized by inflammation of the exocrine pancreas and manifesting itself through acute onset of abdominal pain. It is frequently associated with organ failure, pancreatic necrosis, and death. Mounting evidence describes monocytes - phagocytic, antigen presenting, and regulatory cells of the innate immune system - as key contributors and regulators of the inflammatory response and subsequent organ failure in acute pancreatitis. This review highlights the recent advances of dynamic change of numbers, phenotypes, and functions of circulating monocytes as well as their underling regulatory mechanisms with a special focus on the role of lipid modulation during acute pancreatitis.

Significant Succession of Intestinal Bacterial Community and Function During the Initial 72 Hours of Acute Pancreatitis in Rats

Acute pancreatitis (AP) is followed by structural and functional changes in the intestine, resulting from microbiome dysbiosis. However, it remains unclear how gut microbiome changes within the initial 72h of onset. In this study, severe acute pancreatitis (SAP), mild acute pancreatitis (MAP), and sham operation (SO) were replicated in rat models. 16S ribosomal RNA gene sequencing was used to explore the gut bacteria community. The predicted Cluster of Orthologous Genes (COG) and Kyoto Encyclopedia of Genes and Genomes (KEGG) metabolic pathways were associated with the 16S rRNA profiles. Compared to the SO group, significant community succession was found during the initial 72h in AP group. At 72 h after AP induction, the Firmicutes/Bacteroidetes (F/B) ratios were significantly different, with the highest ratio in SAP group and the lowest in MAP group. Lactobacillus was the most abundant genus, but it nearly disappeared in SAP rats at 72 h. Clostridiaceae 1 and Clostridium sensu stricto 1 were significantly enriched in AP group. Bacteroidales S24-7 and Bacteroidales S24-7 group norank were enriched in MAP group, while Collinsella, Morganella, and Blautia were enriched in SAP group. Lactobacillus was significantly correlated with nine COGs. Nine COGs showed significant differences between AP group and SO group. Moreover, four COGs showed significant differences between the MAP and SAP groups. KEGG Level_3 pathways propanoate metabolism (Ko00640) in AP group was significantly higher than that in SO group. The aspartate‒ammonia ligase and four KEGG orthology terms of the AP group were lower than that in the SO group, respectively. All these results suggest that the intestinal bacterial community structure and function was changed during the initial 72h in AP rats. The intestinal F/B ratio and the relative abundance of Lactobacillus could be potential markers for early diagnosis of MAP and SAP. The genus Clostridium sensu stricto 1 was the most enriched genus in AP, and may be an important marker for AP.

LONG-TERM EFFECTS OF SHAM SURGERY ON PHAGOCYTE FUNCTIONS IN RATS

Animal models of inflammatory disorders, including those of the nervous system are commonly used to explore the pathophysiological role of immune cell response in disease triggering and course and to develop biotechnology products for therapeutic use. Modeling some of these disorders, particularly neurodegenerative diseases, implies surgical manipulations for the intracerebral introduction of disease-initiating substances (toxins, amyloids etc.). Design of these experiments involves the use of sham-operated animals as a control of non-specific intrinsic side-effects elicited by surgical manipulations per se, including local and systemic inflammation, where phagocytic cells are key participants. Short-term post-surgical immunomodulatory effects are widely reported. However, no study thus far has examined the long term effects of sham-surgery on phagocyte functions. The purpose of this study was to evaluate the effect of sham-surgery, commonly used for modeling neurodegenerative diseases, on phagocyte functions in the far terms after the surgical manipulations. Materials and Methods. Adult male Wistar rats were used in the study. Sham surgery consisted of stereotactic unilateral injection of saline solution into the median forebrain bundle (sham-operated 1, SO1) or directly into the substantia nigra (sham-operated 2, SO2). Before the placebo surgery, animals were anaesthetized using nembutal and ketamine/xylazine correspondingly. Functional characteristics (phagocytic activity, oxidative metabolism, CD80/86 and CD206 expression) of phagocytes (microglia, peritoneal macrophages, circulating monocytes and granulocytes) were examined by flow cytometry. Differential leukocyte count was conducted using hematological analyzer. Results. Phagocytes from animals underwent of different protocols of placebo surgery, demonstrated various patterns of functional changes on day 29 after the manipulations. In animals from SO1 group, we observed signs of residual neuroinflammation (pro-inflammatory shift of microglia functional profile) along with ongoing resolution of systemic inflammation (anti-inflammatory metabolic shift of circulating phagocytes and peritoneal macrophages). In rats from SO2 group, pro-inflammatory polarized activation of peritoneal phagocytes was registered along with anti-inflammatory shift in microglia and circulating phagocytes. Conclusions. Sham surgery influences functions of phagocytic cells of different locations even in the far terms after the manipulations. These effects can be considered as combined long-term consequences of surgical brain injury and the use of anesthetics. Our observations evidences, that sham associated non-specific immunomodulatory effects should always be taken into consideration in animal models of inflammatory central nervous system diseases.

[Etiology and pathogenesis of infected pancreatic necrosis]

Modern literature data confirm the central role of intestinal barrier complex not only as a target in acute necrotizing pancreatitis, but also as a trigger for septic complications. Intra-abdominal hypertension, endothelial dysfunction and gut microbiome changes following necrotizing pancreatitis might have an independent impact on acute intestinal distress syndrome and bacterial translocation. Monitoring of these conditions and early target therapy can improve the outcomes in patients with severe acute pancreatitis. Adverse outcomes of infected pancreatic necrosis including high mortality and morbidity are largely due to the prevalence of multidrug-resistant bacterial pathogens.

Circulating CitH3 Is a Reliable Diagnostic and Prognostic Biomarker of Septic Patients in Acute Pancreatitis

Purpose

Acute pancreatitis (AP) is an inflammatory disease. AP starts with sterile inflammation and is often complicated with critical local or systemic infection or sepsis in severe cases. Septic AP activates peptidyl arginine deiminase (PAD) and citrullinates histone H3 (CitH3), leading to neutrophil extracellular trap (NET) formation. Investigating the role of NETs and underlying mechanisms in septic AP may facilitate developing diagnostic and therapeutic approaches. In this study, we sought to identify the expression of CitH3 in septic AP patients and to analyze the correlation of CitH3 concentration with NET components as well as clinical outcomes.

Methods

Seventy AP patients with or without sepsis (40 septic cases, 30 nonseptic cases) and 30 healthy volunteers were recruited in this study. Concentration of NET components (CitH3 and double-strain DNA) and key enzymes (PAD2/4) were measured. Clinical and laboratory characteristics of patients were recorded and analyzed.

Results

Levels of CitH3 were elevated significantly in septic AP patients compared with those in nonseptic AP and healthy volunteers. The area under the curve (AUC, 95% confidence interval) for diagnosing septic AP was 0.93 (0.86–1.003), and the cutoff was 43.05 pg/ml. Among septic AP cases (n = 40), the concentration of CitH3 was significantly increased in those who did not survive or were admitted to the intensive care unit, when compared with that in those who survived or did not require intensive care unit. Association analysis revealed that CitH3 concentration was positively correlated with PAD2, PAD4, dsDNA concentration, and Sequential Organ Failure Assessment scores.

Conclusion

CitH3 concentration increased in septic AP patients and was closely correlated with disease severity and clinical outcomes. CitH3 may potentially be a diagnostic and prognostic biomarker of septic AP.