重症急性胰腺炎列线图预测模型的建立与验证

Establishment and Validation of a Nomogram Prediction Model for Severe Acute Pancreatitis

摘要: 目的：确定重症急性胰腺炎(Severe Acute Pancreatitis, SAP)的危险因素，并构建列线图预测模型来早期预测重症急性胰腺炎的发生。方法：筛选2019年1月至2023年7月于安徽医科大学第二附属医院明确诊断为急性胰腺炎(Acute Pancreatitis, AP)的住院患者完整资料499例，按病情严重程度分为非重症急性胰腺炎组(Non-Severe Acute Pancreatitis, NSAP)(n = 349)和重症急性胰腺炎组(Severe Acute Pancreatitis, SAP)(n = 150)。收集所有患者发病时间在48小时内且未经治疗的首次发病的血浆及一般数据，两组通过比较分析之后，采用单因素逻辑回归分析、LASSO回归分析识别SAP的潜在危险因素，使用多因素逻辑回归分析确定的独立危险因素用于构建列线图。使用受试者操作特征曲线、Hosmer-Lemeshow检验、校准曲线和决策曲线分析来评估列线图的判别能力、校准能力和临床实用性。结果：共纳入499例AP患者，其中SAP患者150例。D-D、CRP、Ca²⁺及APACHE II评分是SAP的独立预测指标。构建的列线图预测模型受试者工作曲线图形下面积(AUC)为0.951，显示出出色的判别能力。Hosmer-Lemeshow检验和校准曲线表明预测模型具有令人满意的校准能力。决策曲线分析显示列线图模型在预测SAP方面具有良好的临床实用性。结论：该列线图预测模型可准确预测重症急性胰腺炎的发生。

Abstract: Methods: A total of 499 inpatients diagnosed with Acute Pancreatitis (AP) at the Second Affiliated Hospital of Anhui Medical University between January 2019 and July 2023 were enrolled. They were categorized into a Non-Severe Acute Pancreatitis (NSAP) group (n = 349) and a Severe Acute Pancreatitis (SAP) group (n = 150) based on disease severity. Plasma and general data collected within 24 hours of onset, prior to any treatment, were obtained from all patients. After comparative analysis between the two groups, univariate logistic regression and LASSO regression analyses were employed to identify potential risk factors for SAP. Independent risk factors identified via multivariate logistic regression analysis were used to construct the nomogram. The discriminative ability, calibration, and clinical utility of the nomogram were assessed using the Receiver Operating Characteristic (ROC) curve, the Hosmer-Lemeshow test, calibration curves, and Decision Curve Analysis (DCA). Results: Among the 499 included AP patients, 150 were diagnosed with SAP. D-dimer (D-D), C-reactive Protein (CRP), Calcium ions (Ca²⁺), and APACHE II score were identified as independent predictors of SAP. The constructed nomogram prediction model achieved an Area Under the Curve (AUC) of 0.951, demonstrating excellent discriminative ability. The Hosmer-Lemeshow test and calibration curves indicated satisfactory calibration of the prediction model. Decision Curve Analysis showed that the nomogram model had good clinical utility for predicting SAP. Conclusion: This nomogram prediction model can accurately predict the occurrence of Severe Acute Pancreatitis.