摘要
目的探讨中度重症急性胰腺炎(MSAP)、重症急性胰腺炎(SAP)并发胰源性门静脉高压(PPH)的影响因素并构建预测模型。方法纳入2016年1月1日至2022年12月31日于昆明医科大学第一附属医院诊断为MSAP、SAP的患者1095例,以是否并发PPH分为PPH组(145例)和非PPH组(950例)。收集两组患者的一般资料(如性别、急性胰腺炎病因、实际住院天数等),并发症(如门静脉血栓、胰腺假性囊肿、胰腺包裹性坏死等),实验室检查指标(如白蛋白、D-二聚体等),以及改良的计算机断层扫描严重指数(MCTSI)评分等。采用最小绝对收缩与选择算子(LASSO)和多因素logistic回归分析MSAP、SAP并发PPH的独立危险因素并构建列线图预测模型。计算受试者操作特征曲线的曲线下面积以评估模型的区分度,采用校准曲线和Hosmer-Lemeshow拟合优度检验评估模型的预测精度,采用临床决策曲线分析(DCA)评估模型的临床实用性。结果LASSO和多因素logistic回归分析显示,门静脉血栓、胰腺假性囊肿、胰腺包裹性坏死、实际住院天数、MCTSI、白蛋白降低均是MSAP、SAP并发PPH的独立危险因素(OR=7.013、2.085、1.846、1.030、1.235、0.955,95%置信区间分别为4.061~12.112、1.255~3.463、1.066~3.199、1.013~1.047、1.123~1.357、0.927~0.983;均P<0.05)。模型的曲线下面积为0.820(95%置信区间0.780~0.859),校准曲线接近参考曲线,Hosmer-Lemeshow拟合优度检验表明该模型具有良好的拟合度(χ^(2)=9.82,P=0.278)。DCA结果显示,模型在较广泛的风险阈值范围(阈概率为0.1~0.9)具有较高的净获益,有一定临床实用价值。结论门静脉血栓、胰腺假性囊肿、胰腺包裹性坏死、实际住院天数、MCTSI、白蛋白降低是MSAP、SAP并发PPH的独立危险因素,构建的列线图模型具有良好的区分度、校准度和临床实用性。
Objective To investigate the risk factors of moderately severe acute pancreatitis(MSAP)and severe acute pancreatitis(SAP)complicated with pancreatogenic portal hypertension(PPH)and to establish a prediction model.Methods From January 1,2016 to December 31,2022,a total of 1095 patients diagnosed with MSAP or SAP at the First Affiliated Hospital of Kunming Medical University were enrolled and divided into PPH group(145 cases)and non-PPH group(950 cases)according to the presence or absence of concomitant PPH.The general data(gender,etiology of acute pancreatitis,days of hospitalization,etc.),complications(portal vein thrombosis,pancreatic pseudocyst,pancreatic encapsulated necrosis,etc.),laboratory indicators(albumin,D-dimer,etc.),and scores of modified computed tomography severity index(MCTSI)were collected in the two groups.The least absolute shrinkage and selection operator(LASSO)and multivariate logistic regression analysis were performed to analyze the independent risk factors of MSAP and SAP complicated with PPH,and the nomogram prediction model was established.The area under the curve of the receiver operating characteristic curve was calculated to evaluate the discrimination of the calibration curve and Hosmer-Lemeshow goodness of fit test were used to assess the predictive accuracy of the model,and clinical decision curve analysis(DCA)was used to evaluate the clinical practicability of the model.Results The results of LASSO and multivariate logistic regression analysis showed that portal vein thrombosis,pancreatic pseudocyst,pancreatic encapsulated necrosis,days of hospitalization,MCTSI and decreased albumin were independent risk factors of MSAP and SAP complicated with PPH(OR=7.013,2.085,1.846,1.030,1.235 and 0.955;95%confidence interval 4.061 to 12.112,1.255 to 3.463,1.066 to 3.199,1.013 to 1.047,1.123 to 1.357 and 0.927 to 0.983;all P<0.05).The area under the curve of the model was 0.820(95%confidence interval 0.780 to 0.859),the calibration curve was close to the reference curve,and the Hosmer-Lemeshow goodness-fit test showed that the model had a good fit(χ^(2)=9.82,P=0.278).The result of DCA indicated that the model had a high net benefit in a wide range of risk threshold(threshold probability 0.1 to 0.9),and had certain clinical practicability.Conclusions Portal vein thrombosis,pancreatic pseudocyst,pancreatic encapsulated necrosis,days of hospitalization,MCTSI and decreased albumin are the independent risk factors of MSAP and SAP complicated with PPH.The established nomogram model has good differentiation,calibration and clinical practicability.
作者
杨佳妮
张奇瑞
刘妍
廖宇航
田秋燕
胡万宇
缪应雷
马岚青
张海蓉
Yang Jiani;Zhang Qirui;Liu Yan;Liao Yuhang;Tian Qiuyan;Hu Wanyu;Miao Yinglei;Ma Lanqing;Zhang Hairong(Department of Gastroenterology,the First Affiliated Hospital of Kunming Medical University,Kunming 650032,China)
出处
《中华消化杂志》
CSCD
北大核心
2024年第9期598-604,共7页
Chinese Journal of Digestion
关键词
急性胰腺炎
胰源性门静脉高压
危险因素
列线图
Acute pancreatitis
Pancreatogenic portal hypertension
Risk factors
Nomograms
