Background and objective The Atlanta classification of acute pancreatitis enabled standardised reporting of research and aided communication between clinicians. Deficiencies identified and improved understanding of the disease make a revision necessary. Methods A web-based consultation was undertaken in 2007 to ensure wide participation of pancreatologists. After an initial meeting, the Working Group sent a draft document to 11 national and international pancreatic associations. This working draft was forwarded to all members. Revisions were made in response to comments, and the web-based consultation was repeated three times. The final consensus was reviewed, and only statements based on published evidence were retained. Results The revised classification of acute pancreatitis identified two phases of the disease: early and late. Severity is classified as mild, moderate or severe. Mild acute pancreatitis, the most common form, has no organ failure, local or systemic complications and usually resolves in the first week. Moderately severe acute pancreatitis is defined by the presence of transient organ failure, local complications or exacerbation of co-morbid disease. Severe acute pancreatitis is defined by persistent organ failure, that is, organ failure >48 h. Local complications are peripancreatic fluid collections, pancreatic and peripancreatic necrosis (sterile or infected), pseudocyst and walled-off necrosis (sterile or infected). We present a standardised template for reporting CT images. Conclusions This international, web-based consensus provides clear definitions to classify acute pancreatitis using easily identified clinical and radiologic criteria. The wide consultation among pancreatologists to reach this consensus should encourage widespread adoption.

/pdf/classification-of-acute-pancreatitis-2012-revision-of-the-608d2rjeud.pdf

Classification of acute pancreatitis—2012: revision of the Atlanta classification and definitions by international consensus

https://www.spg.pt/wp-content/uploads/2015/11/2013-ACG_Guideline_AcutePancreatitis_September_2013.pdf

American College of Gastroenterology guideline: management of acute pancreatitis.

Acute pancreatitis is inflammation of the pancreas; it is sometimes associated with a systemic inflammatory response that can impair the function of other organs or systems. The inflammation may settle spontaneously or may progress to necrosis of the pancreas or surrounding fatty tissue. The distant organ or system dysfunction may resolve or may progress to organ failure. Thus there is a wide spectrum of disease from mild (80%), where patients recover within a few days, to severe (20%) with prolonged hospital stay, the need for critical care support, and a 15-20% risk of death. If patients have organ failure during the first week in hospital, it is usually already present on the first day in hospital. This early organ failure may resolve in response to treatment. The diagnosis of severe acute pancreatitis depends on the presence of persistent organ failure (>48 hours) either during the first week or at a later stage, and also on the presence of local complications (usually apparent after the first week).

Acute pancreatitis

https://biblio.ugent.be/publication/8582111/file/8582114.pdf

IAP/APA evidence-based guidelines for the management of acute pancreatitis

Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient’s record. We propose a representation of patients’ entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two US academic medical centers with 216,221 adult patients hospitalized for at least 24 h. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting: in-hospital mortality (area under the receiver operator curve [AUROC] across sites 0.93–0.94), 30-day unplanned readmission (AUROC 0.75–0.76), prolonged length of stay (AUROC 0.85–0.86), and all of a patient’s final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios. In a case study of a particular prediction, we demonstrate that neural networks can be used to identify relevant information from the patient’s chart.

/pdf/scalable-and-accurate-deep-learning-with-electronic-health-eu4vc3mdb6.pdf

Scalable and accurate deep learning with electronic health records

https://www.gastroenterologyandhepatology.net/files/2013/10/gh1211_Banks1.pdf

Bechien U. Wu

Papers

The value of formal clinical research training in initiating a career as a clinical investigator.

Editorial: fluid resuscitation in acute pancreatitis: striking the right balance.

Endoscopic Versus Surgical Necrosectomy for Patients With Symptomatic Pancreatic Necrotic Collections: A Retrospective Cohort Study

955 A Prospective Validation of the Bedside Index for Severity in Acute Pancreatitis (Bisap) Score in Assessing Mortality in Acute Pancreatitis

Derivation and External Validation of Machine Learning-Based Model for Detection of Pancreatic Cancer