New model predicts postpancreatectomy hemorrhage risk accurately

A novel predictive model using lasso-logistic regression has been developed to assess the risk of postpancreatectomy hemorrhage. Analyzing data from over 9,600 patients, the model identifies nine independent risk factors, including preoperative BMI and ASA scores. Its predictive accuracy reached an area under the ROC curve of 0.87 during external validation, outperforming previous models significantly. This advancement aims to enhance surgical safety and improve patient outcomes following pancreatectomy procedures.