BACKGROUND Sepsis often causes severe liver injury and leads to poor patient outcomes. Early detection of sepsis-induced liver injury (SILI) and early treatment are key to improving outcomes. AIM To investigate the clinical characteristics of SILI patients and analyze the associated risk factors, to identify potential sensitive biomarkers. METHODS Retrospective analysis of clinical data from 546 patients with sepsis treated in the intensive care unit of the 908th Hospital of Chinese People’s Liberation Army Joint Logistic Support Force between May 2018 and December 2022. The patients were divided into the sepsis group (n = 373) and SILI group (n = 173) based on the presence of acute liver injury within 2 hours of admission. We used the random forest algorithm to analyze risk factors and assessed potential diagnostic markers of SILI using the area under the receiver operating characteristic curve, Kaplan-Meier survival curves, subgroup analysis and correlation analysis. RESULTS Compared with the sepsis group, tissue plasminogen activator-inhibitor complex (t-PAIC) levels in serum were significantly higher in the SILI group (P < 0.05). Random forest results showed that t-PAIC was an independent risk factor for SILI, with an area under the receiver operating characteristic curve of 0.862 (95% confidence interval: 0.832-0.892). Based on the optimal cut-off value of 11.9 ng/mL, patients at or above this threshold had significantly higher levels of lactate and Acute Physiology and Chronic Health Evaluation II score. The survival rate of these patients was also significantly worse (hazard ratio = 2.2, 95% confidence interval: 1.584-3.119, P < 0.001). Spearman’s correlation coefficients were 0.42 between t-PAIC and lactate, and 0.41 between t-PAIC and aspartate transaminase. Subgroup analysis showed significant differences in t-PAIC levels between patients with different severity of liver dysfunction. CONCLUSION T-PAIC can serve as a diagnostic indicator for SILI, with its elevation correlated with the severity of SILI.
Read full abstract