|
李丽茵.重症胰腺炎患者28天预后的列线图预测模型构建[J].内科急危重症杂志,2026,32(1):30-33
|
| DOI：10.11768/nkjwzzzz20260107 |
| 中文关键词: 重症胰腺炎 血清标志物 预后 预测 列线图 |
| 英文关键词: |
| 基金项目: |
|
| 摘要点击次数: 56 |
| 全文下载次数: 60 |
| 中文摘要: |
| 摘要 目的：构建重症急性胰腺炎（SAP）患者28 d预后的列线图预测模型。方法：选取SAP患者280例，根据28 d预后情况，分为预后良好组和预后不良组，比较2组一般资料、血清标志物水平，通过Lasso-Cox回归分析SAP患者预后不良的影响因素，构建预后不良的列线图预测模型。另选取SAP患者280例，开展外部验证。结果：SAP患者28 d预后不良率为30.71%（86/280）；Lasso-Cox回归分析显示，年龄、入院时呼吸频率、SAP严重程度床边指数（BISAP）、急性生理与慢性健康状况评估评分（APACHEⅡ）、机械通气、血清C反应蛋白（CRP）、降钙素原（PCT）、白介素-6（IL-6）、载脂蛋白B与载脂蛋白A比值（ApoB/ApoA1）、血管紧张素转化酶2（ACE2）、血小板活化因子（PAF）、前白蛋白（PAB）水平为SAP患者预后不良的影响因素（P均＜0.05）；列线图预测模型预测SAP患者预后不良的AUC为0.962，且该模型预测结果与实际观测结果一致性较好；外部验证显示，列线图预测模型预测SAP患者预后不良的灵敏度为92.77%，特异性为83.76%，准确度为86.43%。结论：基于血清标志物构建SAP患者28 d预后的列线图预测模型具有较高预测效能和校准度，可为临床识别预后不良高危患者提供可靠依据。 |
| 英文摘要: |
| Abstract Objective:To construct a nomogram prediction model for the 28-day prognosis of patients with severe acute pancreatitis (SAP). Methods: A total of 280 patients with SAP were selected. According to the 28-day prognosis, the patients were divided into good prognosis group and poor prognosis group. The general data and serum marker levels of the two groups were compared. The factors affecting the poor prognosis of SAP patients were analyzed using Lasso-Cox regression analysis, and a nomogram prediction model for poor prognosis was constructed based on the influencing factors.In addition, 280 SAP patients were selected for external validation.Results: The 28-day poor prognosis rate of patients with SAP was 30.71% (86/280); Lasso-Cox regression analysis showed that age, respiratory rate at admission, bedside index of severity of acute pancreatitis (BISAP), acute physiology and chronic health evaluation II (APACHE II) score, mechanical ventilation, serum C-reactive protein (CRP), procalcitonin (PCT), interleukin-6 (IL-6), apolipoprotein B to apolipoprotein A ratio (ApoB/ApoA1), angiotensin converting enzyme 2 (ACE2), platelet activating factor (PAF), and prealbumin (PAB) levels were all factors that affected the poor prognosis of SAP patients (P<0.05). The AUC of the nomogram prediction model for predicting poor prognosis in SAP patients was 0.962, and the model's prediction results were in good agreement with the actual observed results. External validation showed that the nomogram prediction model had a sensitivity of 92.77%, specificity of 83.76%, and accuracy of 86.43% for predicting poor prognosis in SAP patients. Conclusion: The 28-day prognosis prediction model of SAP patients based on serum markers has high predictive efficiency and calibration degree, and can provide a reliable basis for clinical identification of high-risk patients with poor prognosis. |
|
|
|
|
