间质水肿性胰腺炎发生坏死的危险因素分析和早期预测模型的构建  

Risk factors analysis and early prediction model construction for necrosis in interstitial oedematous pancreatitis

作　　者：曹波[1] 朱建国[1] 郭文文[2] 杨帆[3] 苏晟[1] 王之悦 郭浩东 汪琼 李海歌[1] CAO Bo;ZHU Jianguo;GUO Wenwen;YANG Fan;SU Sheng;WANG Zhiyue;GUO Haodong;WANG Qiong;LI Haige(Department of Medical Imaging,the Second Affiliated Hospital of Nanjing Medical University,Nanjing 210011,China;Department of Pathology,the Second Affiliated Hospital of Nanjing Medical University,Nanjing 210011,China;Department of Clinical Laboratory,the Second Affiliated Hospital of Nanjing Medical University,Nanjing 210011,China)

机构地区：[1]南京医科大学第二附属医院医学影像科,江苏南京210011 [2]南京医科大学第二附属医院病理科,江苏南京210011 [3]南京医科大学第二附属医院检验科,江苏南京210011

出　　处：《实用放射学杂志》2024年第11期1818-1822,共5页Journal of Practical Radiology

摘　　要：目的探讨间质水肿性胰腺炎(IOP)发生坏死的危险因素,并构建早期预测IOP发生坏死的列线图模型。方法回顾性分析306例IOP患者,根据随访CT增强检查有无胰腺坏死,分为坏死组和水肿组。使用logistic回归分析筛选出IOP发生坏死的独立预测因子,并构建列线图模型,通过受试者工作特征(ROC)曲线、校准曲线和决策曲线分析(DCA)评估模型的区分度、准确性和实用性。结果最终确定了Balthazar CT严重指数(CTSI)、性别、乳酸脱氢酶(LDH)和甘油三酯(TG)作为构建列线图模型的4个独立预测因子。列线图模型曲线下面积(AUC)为0.800[95%置信区间(CI)0.731~0.869]。校准曲线显示,模型在预测IOP发生坏死概率与实际概率之间有良好的一致性(P=0.737)。DCA表明,在3%~66%和75%~96%的阈概率范围内,列线图模型具有较高的实用性。结论基于Balthazar CTSI、性别、LDH和TG构建的列线图模型在早期预测IOP发生坏死方面表现出良好的效能。Objective To investigate the risk factors associated with necrosis in interstitial oedematous pancreatitis(IOP)and to develop a nomogram model for the early prediction of necrosis in IOP.Methods A retrospective analysis was conducted on 306 patients diagnosed with IOP.Patients were stratified into necrosis and edema groups based on the presence or absence of pancreatic necrosis through follow-up CT-enhanced examinations.Logistic regression analysis was employed to identify independent predictive factors for necrosis in IOP.Subsequently,a nomogram model was developed,and its discriminative ability,accuracy,and practicality were assessed through receiver operating characteristic(ROC)curve,calibration curve,and decision curve analysis(DCA).Results Balthazar computed tomography severity index(CTSI),gender,lactate dehydrogenase(LDH),and triglyceride(TG)were finally identified as four independent predictors for constructing the nomogram model.The area under the curve(AUC)of the nomogram model was 0.800[95%confidence interval(CI)0.731-0.869].The calibration curve indicated good consistency between the predicted probability and the actual probability of necrosis in IOP(P=0.737).DCA suggested high practicality of the nomogram model within the threshold probability range of 3%to 66%and 75%to 96%.Conclusion The nomogram model based on Balthazar CTSI,gender,LDH,and TG demonstrates good efficacy in early prediction of necrosis in IOP.

关 键 词：胰腺炎 坏死 列线图 计算机体层成像 

分 类 号：R657.5[医药卫生—外科学]