Construction of a predictive model for concurrent infection in liver failure patients based on prognostic nutritional...

This study aimed to explore the relationship between the Prognostic Nutritional Index (PNI, a composite indicator of albumin and lymphocyte count reflecting nutritional and immune status) and inflammatory cytokines in predicting infections among liver failure patients, and to construct a predictive model based on these indicators.

A retrospective analysis was conducted on 163 patients with liver failure admitted to our hospital between January 2020 and December 2023. Patients were categorized into an Infection group and a Non-infection group based on the presence of concurrent infections. Clinical data and laboratory parameters were collected and compared between the two groups. Indicators with significant differences were evaluated for collinearity. Non-collinear factors were selected for a logistic regression model to identify infection predictors. Statistically significant variables were used to create a risk prediction nomogram using R software, with internal validation performed.

Statistically Significant differences (P < 0.05) were observed between the two groups in terms of C-reactive protein (CRP), soluble triggering receptor expressed on myeloid cells-1 (sTREM-1), Systemic Inflammatory Response Index (SIRI, a novel inflammatory biomarker), PNI, and Acute Physiology and Chronic Health Evaluation II (APACHE II). No collinearity was detected1 (VIF ≤ 10, tolerance ≥ 0.1). Logistic regression analysis identified CRP, sTREM-1, SIRI, and APACHE II as risk factors for infection (OR > 1, P < 0.05), while PNI was a protective factor (OR < 1, P < 0.05). These five variables were incorporated into a nomogram-based predictive model. The model demonstrated excellent performance, with an area under the ROC curve (AUC) of 0.960 (95% CI: 0.927-0.993), indicating high predictive accuracy.

CRP, sTREM-1, SIRI, PNI, and APACHE II scores are independent predictors of infection in liver failure patients. These indicators can be used to identify high-risk populations, providing a theoretical basis for implementing appropriate clinical interventions.