Researchers have developed and validated a nomogram based on the high-sensitivity C-reactive protein-to-high-density lipoprotein cholesterol ratio, known as HCHR, to predict the risk of secondary bacterial infections in patients with acute-on-chronic liver failure. This retrospective study, published in BMC Infectious Diseases, addresses a critical gap in clinical care, as secondary bacterial infections are common in this patient population and are associated with poor clinical outcomes and increased short-term mortality. Despite the severity of these infections, reliable tools for early risk stratification have remained limited. The study evaluates the predictive ability of HCHR and constructs an individualized prediction model to assist in identifying patients at higher risk during hospitalization.
The study utilized a retrospective design, enrolling 524 patients with acute-on-chronic liver failure. These patients were randomly divided into two distinct groups: a training set comprising 367 patients and a validation set consisting of 157 patients. The primary outcome of interest was the occurrence of in-hospital secondary bacterial infections.
To identify the most relevant predictors for the model, the research team employed univariate analysis followed by Least Absolute Shrinkage and Selection Operator, or LASSO, regression. The researchers then compared seven different machine learning algorithms to determine which one offered the best balance of discrimination, calibration, and clinical utility.
The results indicated that secondary bacterial infections occurred in 51.0% of the patients in the study cohort. The HCHR was significantly higher in the group of patients who developed infections compared to those who did not. Statistical analysis revealed a strong association between HCHR and infection risk, with an odds ratio of 1.028 and a 95% confidence interval ranging from 1.023 to 1.034. The p-value was less than 0.001, indicating a statistically significant relationship. Furthermore, the data showed a dose-response relationship, suggesting that as the HCHR increased, the likelihood of developing a secondary bacterial infection also increased.
After comparing the various machine learning algorithms, the researchers selected a logistic regression model based on HCHR as the final prediction tool. This model was chosen because of its favorable performance metrics, including an area under the curve, or AUC, of 0.868, with a 95% confidence interval of 0.804 to 0.920. Additionally, the model demonstrated good calibration, as indicated by a Brier score of 0.142. The final model incorporated four key variables: HCHR, the presence of ascites, hepatic encephalopathy, and hepatorenal syndrome.
The study found that the new model significantly outperformed the MELD score in predicting secondary bacterial infections, with a p-value of less than 0.001. The model enabled effective risk stratification, categorizing patients into different risk groups based on their predicted probability of infection.
In terms of risk stratification, the model demonstrated a clear distinction between high-risk and low-risk groups. Patients classified in the high-risk group had an infection rate of 89.1%, whereas those in the low-risk group had an infection rate of 24.7%.
The authors conclude that HCHR is a readily available biomarker that is associated with the risk of in-hospital secondary bacterial infection in patients with acute-on-chronic liver failure. The HCHR-based nomogram demonstrated promising performance in internal validation, suggesting that it may serve as a preliminary tool for individualized risk stratification. While the study provides strong evidence for the utility of this model, it is important to note that the findings are based on a retrospective design and internal validation.
Subject of Research: Development of a predictive nomogram for secondary bacterial infections in acute-on-chronic liver failure using HCHR.
Article Title: Development and validation of a nomogram based on high-sensitivity C-reactive protein-to-high-density lipoprotein cholesterol ratio (HCHR) for predicting secondary bacterial infections in patients with acute-on-chronic liver failure: a retrospective study
Article References: Pan, L., Lou, Z., Jiang, S., Zeng, Y., Pan, L., Yan, Z., He, Z., & Ruan, B. (2026). Development and validation of a nomogram based on high-sensitivity C-reactive protein-to-high-density lipoprotein cholesterol ratio (HCHR) for predicting secondary bacterial infections in patients with acute-on-chronic liver failure: a retrospective study. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14735-z
Image Credits: AI Generated
DOI: 10.1186/s12879-026-14735-z
Keywords: acute-on-chronic liver failure, secondary bacterial infections, HCHR, nomogram, risk stratification, biomarker, retrospective study, machine learning, liver disease, clinical prediction, Development, validation
