Development and Validation of a Clinical Predictive Nomogram for Assessing the Risk of Recurrence of Acute Pancreatitis in Combined Hypertriglyceridemia

• Published in: Digestive diseases and sciences Vol. 69; no. 9; pp. 3426 - 3435
• Main Authors: Wen, Shuaiyong, Zhang, Yu, Zhao, Guijie, Zhang, Kun, Cui, Yunfeng
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.09.2024; Springer Nature B.V
• Subjects: Acute Disease; Adult; Biochemistry; Calibration; Female; Gallstones - complications; Gallstones - diagnosis; Gastroenterology; Hepatology; Humans; Hypertriglyceridemia - blood; Hypertriglyceridemia - complications; Hypertriglyceridemia - diagnosis; Male; Medicine; Medicine & Public Health; Middle Aged; Nomograms; Oncology; Original Article; Pancreatitis; Pancreatitis - blood; Pancreatitis - diagnosis; Predictive Value of Tests; Recurrence; Regression analysis; Reproducibility of Results; Risk Assessment - methods; Risk Factors; Transplant Surgery; Triglycerides - blood; Recurrence; Hypertriglyceridemia; Nomogram; Pancreatitis
• ISSN: 0163-2116; 1573-2568; 1573-2568
• DOI: 10.1007/s10620-024-08578-4

Background The objective of this study is to develop and validate a new nomogram-based scoring system for anticipating the recurrence of acute pancreatitis (AP) in combined hypertriglyceridemia (HTG). Methods A total of 292 patients diagnosed with AP combined with HTG participated in this research. Among them, 201 patients meeting the inclusion criteria were randomly divided into training and validation sets at a ratio of 7:3. Clinical data were collected for all patients. In the training set, predictive indicators were chosen through backward stepwise multivariable logistic regression analysis. Subsequently, a nomogram was developed based on the selected indicators. Finally, the model’s performance was validated in both the training and validation sets. Results By employing backward stepwise multivariable logistic regression analysis, we identified diabetes, gallstones, alcohol consumption, and triglyceride levels as predictive indicators. Subsequently, a clinical nomogram that incorporates these four independent risk factors was constructed. Model validation demonstrated an AUC of 0.726 (95% CI 0.644–0.809) in the training set and an AUC of 0.712 (95% CI 0.583–0.842) in the validation set, indicating a good discriminative ability. The Hosmer–Lemeshow test yielded P -values of 0.882 and 0.536 in the training and validation sets, respectively, suggesting good calibration. Calibration curves further confirmed good agreement. Ultimately, decision curve analysis (DCA) emphasized the clinical utility of our model. Conclusion We have developed a nomogram for predicting the recurrence of AP combined with HTG in patients, and this nomogram demonstrates good discriminative ability, calibration, and clinical utility. This tool holds the potential to assist clinicians in offering more personalized treatment strategies for AP combined with HTG.