Predictive Value Of Systemic Immune-inflammation Index In Severe Acute Pancreatitis

Objective:The aim of this study was to investigate the predictive power and advantage of the Systemic Immune-inflammation Index(SII)on the severity of patients with Acute pancreatitis(AP).Methods:This study retrospectively analyzed AP patients admitted to our gastroenterology department for inpatient treatment from January 1,2013 to January 1,2021,and the study subjects were screened strictly according to the inclusion and exclusion criteria,and 406 patients were finally included.Patient demographic information was collected through the hospital information system,including general information such as age,gender,body mass index(BMI),smoking,alcohol consumption,AP etiology,past medical history,length of hospitalization;laboratory indices within 24 h of admission such as white blood cell count(WBC),neutrophil ratio(NEU%),lymphocyte ratio(LYM%),monocyte count(MONT),erythrocyte pressure product(HCT),C-reactive protein(CRP),albumin(ALB),aspartate aminotransferase(AST),alanine aminotransferase(ALT),glutamyl transpeptidase(GGT),alkaline phosphatase(ALP),retinol binding protein(RBP),superoxide dismutase(SOD),lactate dehydrogenase(LDH),urea nitrogen(BUN),creatinine(Cr),random glucose(GLU),glycosylated hemoglobin(HbA1c),cholesterol(CHO),triglycerides(TG),lipoprotein a(Lp),APOA1,APOB,etc.;the commonly used clinical AP severity scores include Marshall score,BISAP score,CTSI score.Patients were divided into SAP and non-SAP groups according to the Atlanta classification,or into ABP or HTG-AP groups according to etiology.Statistical methods such as t-test,Kruskal-Wallis test,chi-square test or Fisher exact test,and multi-factor binary logistic regression analysis were used to analyze the variability according to the data characteristics as appropriate.ROC curves were plotted to analyze the predictive value of SII on the severity of AP.Results:Analysis of demographic and clinical characteristics of the study population.(1)A total of 406 patients were included in this study,including 353 cases in the nonSAP group and 53 cases in the SAP(Severe acute pancreatitis,SAP)group.Compared with patients in the non-SAP group,the percentage of males and age were higher in the SAP group,and the difference was statistically different(P<0.05).In addition,a multifactorial binary logistic regression analysis suggested that combined diabetes was an independent risk factor for the development of severe disease in patients with AP,with an OR of 2.935.(2)In the etiological analysis,the results showed that about 54.1%of AP patients had a lipogenic etiology,and the biliary etiology about 23.1%.Compared with patients with Acute biliary pancreatitis(ABP),patients with Hypertriglyceridemic acute pancreatitis(HTG-AP)are more likely to be male,older,have higher BMI,smoke and drink alcohol,and are more likely to have diabetes mellitus,hypertension,fatty liver,and high blood pressure.The differences were statistically significant(P<0.05).(3)The hematological findings within 24 h of admission suggested that patients in the SAP group had higher NEU%,LDH,GLU,WBC,CRP,MONT,AST,ALP,BUN,Cr,HbAlc,and Lp,and lower LYM%,RBP,SOD,and APOA1,with statistically significant differences(P<0.05).2.Analysis of the predictive value of SII and other blood inflammatory indicators and common clinical scoring systems for SAP.(1)SII,NLR,PLR,CAR,CTSI scores,Marshall scores,and BISAP scores were higher in the SAP group compared with patients in the non-SAP group,with statistically significant differences(P<0.05).(2)The AUC of SII for predicting SAP was 0.754(best cutoff value=1705.83,sensitivity 75.47%,specificity 69.12%),which was not significantly different from the CTSI score,Marshall score,and BISAP score in predicting the severity of AP.(3)There was no significant difference between SII and NLR in predicting AP severity,and SII and NLR were better at predicting AP patient severity compared with PLR and CAR(P<0.05).3.Analysis of the predictive value of SII，NLR,and PLR in the severity of ABP group and HTG-AP group.(1)SII predicted ABP severity with an AUC=0.788(sensitivity=85.71%,specificity=68.75%),with PLR(AUC=0.715,sensitivity=78.57%,specificity=63.75%)and NLR(AUC=0.833,sensitivity=100%,specificity=63.75%)with no statistically differences.(2)AUC=0.735(sensitivity=65.38%,specificity=78.35%)for SII prediction of HTGAP severity,which was comparable to the predictive ability of PLR(AUC=0.699,sensitivity=65.38%,specificity=72.68%)and NLR(AUC=0.712,sensitivity=61.54%,specificity=84.02%)There was no statistical difference.Conclusion:AP patients with SII values≥1719.093 were more likely to have SAP(AUC=0.752,sensitivity=76.9%,specificity=68.2%),and their predictive ability was similar to CTSI score,Marshall score,BISAP score,NLR,and better than PLR and CAR.and SII,NLR,and PLR were better at predicting biliary pancreatitis severity may be more advantageous.