建立一个评估重症急性胰腺炎预后的预测模型  

Construction of a predictive model to assess the prognosis of severe acute pancreatitis

作　　者：康文松 蔡兆辉[1] 张旗[1] 魏国峰[1] 骆佳艳 李海山[1] KANG Wen-song;CAI Zhao-hui;ZHANG Qi(Department of Emergency Medicine,The Second Peoples Hospital of Hefei,Hefei Hospital Affiliated to Anhui Medical University,Hefei 230000,China)

机构地区：[1]合肥市第二人民医院(安徽医科大学附属合肥医院)急诊科,合肥230000

出　　处：《肝胆外科杂志》2024年第1期21-26,共6页Journal of Hepatobiliary Surgery

摘　　要：目的临床上评估重症急性性肺炎胰腺炎患者预后的评分工具有很多种,目前这些工具的有效性仍然未知。本研究的目的是探讨建立一个准确的模型来评估重症急性性肺炎胰腺炎的预后。方法选取合肥市第二人民医院急诊科2018年01月至2023年12月收治的诊断为重症急性胰腺炎患者62例。通过单因素和Logistic多因素分析出影响重症急性胰腺炎患者预后的因素,建立预测模型Nomogram,并通过bootstrap进行内部验证。结果昊Logistic多因素分析发生严重低蛋白血症、使用血管活性药物和血钙是影响重症急性胰腺炎患者预后的独立危险因素。内部验证经bootstrap验证,具有较好的判别程度。校准曲线表明,模型具有较好的一致性。采用模型、急性生理与慢性健康评分(Acute Physiology and Chronic Health E-valuation,APACHEII)、急性胰腺炎严重程度床边指数(Bedsideindexof severityacutepancreatic,BISAP)和急性胰腺炎Ranson评分系统预测重症急性胰腺炎死亡率,根据受试者工作特征曲线(ROC)曲线下面积(AUC)分析评分系统对重症急性胰腺炎预后预测的准确性。结论基于是否发生严重低蛋白血症、使用血管活性药物和血钙构建的nomogram预测模型可有效筛选出影响重症急性胰腺炎患者预后的因素,为重症急性胰腺炎患者的个体化治疗提供依据。Purpose There were various clinical scoring tools to assess the prognosis of patients with severe acute pneumonia pancreatitis,and the validity of these tools was still unknown.The aim of this study was to investigate the development of an accurate model to assess the prognosis of severe acute pneumonia pancreatitis.Methods 62 patients diagnosed with severe acute pancreatitis admitted to the emergency department of the Second Peoples Hospital of Hefei City from January 2018 to December 2023 were selected.The factors affecting the prognosis of patients with severe acute pancreatitis were analyzed by single-factor and Logistic multifactor anal-ysis,and the predictive model Nomogram was established and internally validated by bootstrap.Results Logistic multifactorial analysis of the occurrence of severe hypoproteinemia,use of vasoactive drugs,and blood calcium were independent risk factors affecting the prognosis of patients with severe acute pancreatitis.Internal validation was validated by bootstrap with a good degree of discrimination.Calibration curves showed that the model had good agreement.The model,Acute Physiology and Chronic Health Evaluation(APA-CHEII),Bedside index of severity acute pancreatic(BISAP)and Acute pancreatitis Ranson scoring system were predicted for mortali-ty in severe acute pancreatitis,and the accuracy of the scoring systems in predicting the prognosis of severe acute pancreatitis was ana-lyzed based on the area under the curve(AUC)of the subjects'work characteristic curve(ROC).Conclusion A nomogram prediction model constructed based on the occurrence of severe hypoproteinemia,the use of vasoactive drugs,and blood calcium was effective in screening out factors affecting the prognosis of patients with severe acute pancreatitis and pro-viding a basis for individualized treatment of patients with severe acute pancreatitis.

关 键 词：重症急性胰腺炎 预测模型 评分工具 预后 

分 类 号：R659[医药卫生—外科学]