Prognostic Value of Systemic Immune-Inflammation Index in Patients with Hepatitis B Virus-Related Acute-on-chronic Liver Failure
薛红,刘甜甜,郑建铭,张文宏
DOI: https://doi.org/10.3760/cma.j.cn311365-20220326-00101
2023-01-01
Abstract:Objective:To investigate the prognostic value of systemic immune-inflammation index (SII) in patients with hepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF).Methods:The clinical data, including age, gender, complications, laboratory examination results post-admission, SII, model for end-stage liver disease (MELD) score, MELD-Na score, Child-Turcotte Pugh (CTP) score of HBV-ACLF patients treated in Huashan Hospital, Fudan University from January 2016 to August 2021 were retrospectively analyzed. The patients were divided into survival group and death group according to the outcome at 90 days of follow-up.Paired sample t test, Mann-Whitney U test and chi-square test were used for statistical analysis.Pearson correlation was used to analyze the correlation between SII and the prognosis prediction model of HBV-ACLF. The area under the curve (AUC) was used to analyze the clinical efficacies of SII, MELD score, MELD-Na score and CTP score in predicting the prognosis of HBV-ACLF patients, and the optimal cut-off value of SII for predicting the prognosis of HBV-ACLF was calculated. Kaplan-Meier method was used for survival analysis. Results:A total of 140 patients with HBV-ACLF were included. There were 88 patients in the survival group, including 65 males and 23 females, with the age of (47.69±11.96) years. There were 52 cases in the death group, including 40 males and 12 females, with the age of (52.73±12.22) years. The age, aspartate aminotransferase, total bilirubin, serum creatinine, international normalized ratio, neutrophil-to-lymphocyte ratio, monocyte-to-lymphocyte ratio, SII, MELD score, MELD-Na score, CTP score and the incidence of infection in the death group were all significantly higher than those in the survival group, and albumin, lymphocyte count, platelet count, prognostic nutritional index in the death group were all significantly lower than those in the survival group, and the differences were all statistically significant ( t=-2.39, Z=-2.84, t=-4.81, Z=-2.15, Z=-4.91, Z=-3.47, Z=-3.36, Z=-3.83, Z=-4.69, Z=-4.56, Z=-6.31, χ2=24.96, t=3.06, t=3.03, Z=-7.57 and t=4.12, respectively, all P<0.05). Pearson correlation analysis showed that SII was positively correlated with CTP score ( r=0.272 7, P=0.001), MELD score ( r=0.365 8, P<0.001) and MELD-Na score ( r=0.381 1, P<0.001). The AUC of SII was the largest of 0.80, and 0.76 for MELD score, 0.74 for MELD-Na score and 0.73 for CTP score. The optimal cut-off value of SII was 447.49. Kaplan-Meier analysis showed that the 90 days survival rate of patients with SII≥447.49(38.60%(22/57)) was lower than that of SII<447.49 group (79.52%(66/83)), and the difference between the two groups was significant ( χ2=23.80, P<0.001). Conclusions:SII can be used to assess the severity and prognosis of HBV-ACLF patients. SII ≥447.49 indicates poor prognosis.