image: Brown line (model) = ROC of our IPN prediction model, which had the best prediction scores. Cyan line (APACHE II) = ROC drawn based on APACHE II scores. Orange line (Marshall) = ROC of Marshall scores. Khaki line (SOFA) = ROC of SOFA scores. Blue line (BISAP) = ROC of the BISAP scores. The AUC of APACHE-II score, Marshall score, SOFA score and BISAP score are 0.64 (95%CI: 0.59–0.69), 0.62 (95%C1: 0.56–0.67), 0.57 (95%C1: 0.51–0.62), 0.76 (95%CI: 0.71–0.81), respectively. The performance of the IPN model differed significantly from that of the BISAP scores (p = 0.007). APACHE II, Acute Physiology and Chronic Health Evaluation II; AUC, area under the ROC curve; BISAP, Beside Index for Severity in Acute Pancreatitis; ROC, receiver operating characteristics; SOFA, Sepsis-related Organ Failure Assessment.
Credit: By Kai Song, Wenhua He, Zuoyan Wu, Jie Meng, Wei Tian, Shicheng Zheng, Dong Mu, Ruifeng Wang, Hongda Chen, Yin Zhu, Dong Wu
A new prediction model for infected pancreatic necrosis (IPN) in patients with acute pancreatitis (AP) offers a groundbreaking approach to improving patient outcomes. Developed by a team of researchers across eight Chinese hospitals, the model harnesses five early clinical indicators—respiratory rate, temperature, serum glucose, calcium, and blood urea nitrogen (BUN)—to identify high-risk patients within 24 hours of hospital admission.
The study, recently published in eGastroenterology, analyzed data from over 3,000 patients diagnosed with AP between 2017 and 2023. Researchers employed advanced statistical methods, including LASSO regression and multivariate analysis, to develop and validate the model. In the development phase, it achieved an area under the receiver operating characteristic curve (AUC) of 0.85, significantly outperforming widely used scoring systems like BISAP (AUC 0.76) and SOFA (AUC 0.57).
“Infected pancreatic necrosis is a serious complication of acute pancreatitis that increases mortality risk and hospital stays,” said Dr Dong Wu, a senior researcher from Peking Union Medical College Hospital. “Our model represents a practical and highly accurate tool for early risk stratification, ensuring timely intervention and better resource allocation in healthcare systems.”
Why Early Detection Matters
IPN occurs in approximately 6% of AP cases, with a markedly higher prevalence in patients suffering from severe acute pancreatitis (SAP). Left untreated, the condition can lead to systemic infections, multiple organ failure, and increased mortality. Timely and accurate prediction is critical for initiating appropriate treatments, including antibiotics and minimally invasive procedures.
The model focuses on variables readily available in routine clinical settings, such as vital signs and basic laboratory tests. This approach ensures ease of implementation and minimizes reliance on expensive or specialized diagnostic tools. “By focusing on universally available clinical data, our model can be adopted across diverse healthcare settings, including those with limited resources,” noted Dr. Yin Zhu from The First Affiliated Hospital of Nanchang University.
Implications for Clinical Practice
The study underscores the model’s utility in guiding personalized care strategies for AP patients. For high-risk individuals identified by the model, clinicians can prioritize closer monitoring and early initiation of targeted therapies. This proactive approach has the potential to significantly reduce complications and associated healthcare costs.
“Decision curve analysis revealed that the model offers a positive net benefit across a wide range of clinical thresholds,” explained Dr. Hongda Chen from Peking Union Medical College Hospital. “This means clinicians can confidently use it to balance the risks of overtreatment against the dangers of missed diagnoses.”
Global Reach and Future Directions
Although validated in a Chinese population, the researchers aim to expand the model’s application globally. Further studies are planned to adapt it for Western populations, where alcohol-related pancreatitis is predominant and other demographic factors may influence outcomes. The model also opens avenues for further research into preventive measures and early interventions for IPN.
“This innovation is a step forward in personalized medicine,” added Dr. Dong Wu. “We hope it will serve as a catalyst for future advancements in acute pancreatitis care, particularly in regions with high disease burden.”
See the article:
Song K, He W, Wu Z, et al. Early clinical predictors of infected pancreatic necrosis: a multicentre cohort study. eGastroenterology 2024;2:e100095. doi:10.1136/egastro-2024-100095
About eGastroenterology
eGastroenterology is a new, open-access, and open peer-reviewed BMJ Journal, which focuses on basic, clinical, translational, and evidence-based medicine research in all areas of gastroenterology (including hepatology, pancreatology, esophagology, and gastrointestinal surgery). eGastroenterology is now indexed by DOAJ, Scopus, Dimensions, OpenAlex, ROAD, and COPE, with more to come!
For more information, please visit: egastroenterology.bmj.com and follow us on Twitter (@eGastro_BMJ).
Sign-up to Email Alerts for eGastroenterology: https://emails.bmj.com/k/Bmj/jausu/egastroenterology
Journal
eGastroenterology
Article Title
Early clinical predictors of infected pancreatic necrosis: a multicentre cohort study