Value of C-reactive protein in predicting the progression of acute decompensation of cirrhosis to acute-on-chronic liver failure

Feng YANG; Jun GUO; Jiale SHEN; Ruiqi LI; Xixuan WANG; Yongfeng YANG

Return

Value of C-reactive protein in predicting the progression of acute decompensation of cirrhosis to acute-on-chronic liver failure

VernacularTitle:C反应蛋白对急性失代偿期肝硬化进展为慢加急性肝衰竭的预测价值
Author: Feng YANG ¹ ; Jun GUO ² ; Jiale SHEN ² ; Ruiqi LI ³ ; Xixuan WANG ³ ; Yongfeng YANG ¹
Author Information

1. School of Public Health， Nanjing Medical University， Nanjing 211166， China
2. Department of Hepatology， Nanjing Hospital Affiliated to Nanjing University of Chinese Medicine （Nanjing Second Hospital）， Nanjing 210000， China
3. School of Medicine， Southeast University， Nanjing 210000， China
Publication Type:Journal Article
Keywords: Acute Decompensation of Cirrhosis; Acute-on-Chronic Liver Failure; C-Reactive Protein; Prognostic Study
From: Journal of Clinical Hepatology 2026;42(7):1623-1631
CountryChina
Language:Chinese
Abstract: ObjectiveTo investigate the efficacy of C-reactive protein （CRP） in predicting the progression to acute-on-chronic liver failure （ACLF） within 90 days after disease onset in patients with acute decompensation （AD） of cirrhosis， and to provide a reference for the early identification of high-risk populations in clinical practice. MethodsA retrospective study was conducted among 906 patients with liver cirrhosis who were hospitalized due to AD in Nanjing Second Hospital from January 1， 2015 to October 31， 2024， and demographic features， AD type， and laboratory markers were collected on admission. The primary endpoint was progression to ACLF within 90 days after admission， and according to the presence or absence of ACLF， the patients were divided into non-ACLF group with 676 patients and ACLF group with 230 patients. The independent-samples t test was used for comparison of normally distributed continuous data between groups， and the Wilcoxon rank-sum test was used for comparison of non-normally distributed continuous data between groups； the chi-square test or the Fisher’s exact test was used for comparison of categorical data between groups. The univariate and multivariate Cox proportional-hazards regression models were used to identify independent predictive factors for ACLF. The receiver operating characteristic （ROC） curve was used to assess the predictive performance of CRP and existing scoring systems （Model for End-Stage Liver Disease ［MELD］， Child-Turcotte-Pugh ［CTP］ score， and Chronic Liver Failure Consortium Acute Decompensation ［CLIF-C AD］ score）， and the restricted cubic spline analysis was used to investigate the nonlinear relationship between CRP and the risk of ACLF. Youden index was used to determine the optimal predictive cut-off value for CRP， and the Kaplan-Meier survival curve was plotted for risk stratification， while the log-rank test was used for survival analysis. ResultsAmong the 906 AD patients， the incidence rate of ACLF was 25.39% within 90 days. The incidence rate of ACLF within 90 days was 45.1% in the 264 patients with bacterial infection and 17.3% in the 642 patients without bacterial infection （χ²=74.791， P<0.001）. Among the 906 patients， 312 patients （34.44%） had acute-on-chronic pre-liver failure （pre-ACLF）， with a significantly higher proportion of patients who progressed to ACLF than those in the non-pre-ACLF group ［60.58% （189/312） vs 6.90% （41/594）， χ²=310.234， P<0.001］. Among the 230 patients with ACLF， there were 102 patients with stable ACLF （44.35%） and 128 patients with unstable ACLF （55.65%）， with a significant difference in CRP between the two groups of patients （U=5 234.500， P<0.001）. The restricted cubic spline analysis showed a significant nonlinear relationship between CRP and the risk of ACLF （P<0.001）. The optimal cut-off value of CRP was determined as 10.16 mg/L based on the Youden index， with a sensitivity of 70.4% and a specificity of 64.5%. The Kaplan-Meier curve analysis showed that the high-risk group with CRP≥10.16 mg/L had a significantly higher event-free survival rate of ACLF within 90 days compared with the low-risk group （P<0.001）. The multivariate Cox regression analysis showed that AD type-hepatic encephalopathy （hazard ratio ［HR］=4.199， 95% confidence interval ［CI］： 3.056 — 5.770， P<0.001）， AD type-bacterial infection （HR=1.826， 95%CI： 1.356 — 2.459， P<0.001）， CRP （≥10.16 mg/L）（HR=2.356， 95%CI： 1.825 — 3.047， P<0.001）， white blood cell count （HR=1.021， 95%CI： 1.002 — 1.041， P=0.035）， hemoglobin （HR=0.994， 95%CI： 0.989 — 0.999， P=0.025）， international normalized ratio （HR=1.657， 95%CI： 1.372 — 2.001， P<0.001）， total bilirubin （HR=1.003， 95%CI： 1.002 — 1.004， P<0.001）， albumin （HR=0.956， 95%CI： 0.926 — 0.986， P=0.004）， creatinine （HR=1.005， 95%CI： 1.004 — 1.006， P<0.001）， serum sodium （HR=0.976， 95%CI： 0.956 — 0.997， P=0.025）， and lactate dehydrogenase （HR=1.001， 95%CI： 1.001 — 1.002， P<0.001） were all independent predictive factors for progression to ACLF within 90 days. The ROC curve analysis showed that CRP alone had an area under the ROC curve （AUC） of 0.724 （95%CI： 0.685 — 0.763） in predicting ACLF， and CRP+MELD score had the best predictive performance （AUC=0.870， 95%CI： 0.843 — 0.898）， while CRP+CTP score and CRP+CLIF-C AD score had an AUC of 0.852 （95%CI： 0.823 — 0.881） and 0.845 （95%CI： 0.812 — 0.877）， respectively. There was a significant increase in model performance after integration （P<0.05）. ConclusionSerum CRP level on admission is an independent predictive factor for progression to ACLF within 90 days in patients with AD of cirrhosis， and a CRP level of ≥10.16 mg/L can be used as a simple threshold for identifying high-risk patients. Incorporating CRP into existing assessment systems may enhance the early identification of patients at a high risk for ACLF， which provides a reference for clinical intervention.