| Objective:To analyze the clinical characteristics of patients with acute pancreatitis(AP)complicated by infected pancreatic necrosis(IPN),to explore the predictive value of serological indicators for the occurrence of IPN in AP,and to develop a data model.Methods:The clinical data of 366 patients with AP admitted to the First Affiliated Hospital of Southwest Medical University from April2020 to April 2022 were selected and divided into infected necrosis group(n=70)and non-infected necrosis group(n=296)according to whether the patients had infected pancreatic necrosis(IPN)or not.SPSS 25.0software was used to analyze and compare the non-serological data and relevant serological indicators of the two groups,such as:red blood cell distribution width(RDW),Apo-B to Apo-A1 ratio(Apo B/A1),Apo-A1(Apo-A1),Serum calcium(SC),Red blood cell distribution width to serum calcium ratio(RDW/SC),etc.Multi-factor logistic regression model was established by data analysis software to analyze the relevant factors affecting the occurrence of pancreatic infectious necrosis in patients with acute pancreatitis;subsequently,the work characteristics of subjects(ROC)curve was used to test the predictive value of relevant indicators for the occurrence of IPN in patients with AP;finally,the best sensitive indicators obtained were used to construct a predictive data model and to evaluate and validate it.Results:A total of 366 patients with AP(222 males and 144 females)were included in this study,70 in the infected necrosis group and 296 in the non-infected necrosis group.The average age of the two groups was(51.86±15.80)years old.SPSS 25.0 software analysis showed that there was no significant difference in general clinical data between the two groups(P>0.05),but there were significant differences in serological indicators Apo-A1,Apo-B/A1,RDW,RDW/SC and SC between the two groups(P<0.05).The results of the multifactorial logistic regression model showed that Apo-A1 was an independent protective factor for the development of IPN in AP,and Apo-B/A1,RDW,RDW/SC,and low serum calcium were independent risk factors for the development of IPN in AP.ROC curve was plotted and the area under the curve(AUC)RDW/SC>Apo-A1>Apo-B/A1>serum calcium>RDW,where RDW/SC and Apo-A1 had better predictive value.RDW/SC with the following parameters:AUC=0.871,critical value of 6.056,sensitivity of 0.786,specificity of 0.861,95%CI:0.822 to 0.919;the parameters of Apo-A1were as follows:AUC=0.782,critical value 1.160 g/L,sensitivity 0.814,specificity 0.696,95%CI:0.721 to 0.844,P<0.001.RDW/SC,Apo-A1combined predictors were included to establish data and nomogram models,and the evaluation yielded:The optimal critical value of the model is 0.224,the sensitivity is 0.829,the specificity is 0.899,and the C index is 0.905(95%CI:0.864～0.946),which indicates that the model has good discrimination;Hosmer-Lemeshow good of fit test resultsχ~2is9.626,P>0.05,indicating that the calibration of the model is good;DCA curve analysis showed that the model established in this study had a net benefit between 0.00～0.88 and 0.91～0.93,and the model had certain clinical effectiveness.Conclusions:1.Apo-B/A1,RDW,and RDW/SC were significantly higher in patients with AP complicated by IPN compared with the non-infected necrosis group,and Apo-A1 and serum calcium were lower in patients in the infected necrosis group compared with the non-infected necrosis group;2.Apo-A1 is an independent protective factor for acute pancreatitis complicated by IPN,Apo-B/A1,RDW,RDW/SC,and low serum calcium are independent risk factors for the occurrence of IPN,all of which can be used as predictors of IPN,among which RDW/SC and Apo-A1 are good predictors;3.The discrimination,calibration and clinical utility of the clinical prediction model constructed based on RDW/SC and Apo-A1 are good,so this risk model can be used to predict the occurrence of IPN at an early stage. |
PDF Full Text Request