Clinical Data for Early Prediction of Hypertriglyceridemic Acute Pancreatitis: A Retrospective Cohort Study
Description
This dataset contains de-identified clinical, laboratory, physiological, and outcome data from a single-center retrospective cohort study conducted at West China Hospital, Sichuan University. The study investigated whether routinely available baseline characteristics, in addition to triglyceride concentration, could predict the development of hypertriglyceridemic acute pancreatitis (HTG-AP) in patients with persistent severe hypertriglyceridemia. Adults with two triglyceride measurements >5.6 mmol/L between 2010 and 2020 were included. Baseline data included demographic characteristics, medical history, vital signs, blood counts, biochemical tests, lipid profiles, glucose-related indicators, urinalysis, and abdominal ultrasonography. The primary outcome was incident HTG-AP during follow-up. A total of 287 patients were included, of whom 57 (19.9%) developed HTG-AP. Triglyceride concentration, lymphocyte percentage, diastolic blood pressure, and heart rate were identified as independent predictors. The resulting four-variable model showed good discrimination and calibration, with an AUC of 0.811 and an optimism-corrected AUC of 0.798 after bootstrap validation. This dataset can be used to reproduce the reported analyses, evaluate the prediction model, and explore additional predictors or alternative statistical approaches. As the data are derived from a retrospective single-center cohort, associations should not be interpreted as causal, and external validation is recommended before clinical application.