INTRODUCTION:Nosocomial infections (NIs) in cirrhosis are associated with high mortality but could be preventable. Logistic regression (LR) models have failed to identify high-risk patients. We aimed to develop machine learning (ML) models to predict NI. METHODS:The CLEARED consortium consists of prospectively enrolled cirrhosis inpatients from >120 centers. Using day-of-admission clinical data, 3 ML approaches (random forest [RF], extreme gradient boosting, and neural networks [NNs]) were used to predict NI. Data were split 80:20 for training and testing stratified by the outcome. Models were compared using area under the receiver operating characteristic curve (AUC). RESULTS:In total, 8,263 patients (55.90 ± 13.34 years; 64.1% men) from 127 centers in 37 countries were included. NI developed in 869 (10.5%), a median of 6 (4-11) days of postadmission. Major NIs were respiratory (29.6%) and urinary tract infection (15.7%), and spontaneous bacterial peritonitis (13.5%). NIs occurred more frequently in patients from low/low-middle income countries and those with severe liver disease, alcohol etiology, and admission infections. NIs were associated with inpatient mortality (31.9% vs 8.1%, P < 0.001) and liver transplantation (4.8% vs 1.9%, P < 0.001). Although the RF model (AUC 0.69) showed good calibration (Brier score 0.09), outperforming extreme gradient boosting, neural network, and LR models (AUC 0.66 for all; LR comparison P = 0.043), no model achieved AUC ≥0.80 for clinical utility. At 10% predicted probability threshold, the RF model demonstrated only 75.4% sensitivity, 52.9% specificity, and 15.9% positive predictive value (PPV). DISCUSSION:NIs cannot be accurately predicted from day-of-admission data using ML models, even in a large, prospective, global cirrhosis cohort. Every hospitalized patient with cirrhosis should receive protocolized infection control measures.

The global profile of Chinese hepatology has expanded rapidly over the past four decades, driven by major scientific investment, increased global mobility and landmark clinical and translational discoveries. As a new generation of Chinese investigators emerges around the globe, initiatives that foster international collaboration, mentorship and scientific exchange will be essential for advancing the future of global liver research.

Acute-on-chronic liver failure is a complex condition with varied definitions, complicating risk stratification and targeted management. We apply unsupervised clustering to data from 1,256 patients with acute-on-chronic liver failure as defined by the North American Association for the Study of End-Stage Liver Disease. We systematically evaluate multiple algorithmic models to enable unbiased cohort stratification and to determine key clustering factors and clinical impacts across clusters. The optimal number of clusters is determined using the Partitioning Around Medoids with Ambiguous Clustering algorithm. The clusters from the best-performing nonnegative matrix factorization algorithm are selected, with the Lee's algorithm demonstrating the best performance. Two distinct clusters are identified, showing markedly different 30-day mortality rates (70.35% vs 26.06%). Importantly, acid-base balance-related variables, including bicarbonate, pH, base excess, lactate, and anion gap, are among the primary clustering drivers. An external validation cohort, a decompensated cirrhosis cohort, and a European Association for the Study of the Liver-Chronic Liver Failure Consortium defined Acute-on-chronic liver failure patients cohort confirm consistent distributions of key clustering variables and divergence in 30-day mortality rates, supporting the role of acid-base balance variables. In summary, we show that the unbiased clustering approach successfully identifies distinct acute-on-chronic liver failure clusters with different mortality risks and emphasizes the critical role of metabolic regulation in acute-on-chronic liver failure outcomes, as well as consistent validation.

Objective Cirrhosis is a progressive liver disease caused by chronic inflammation. The serum albumin-to-creatinine ratio (CAR) and glucose-to-potassium ratio (GPR) are emerging prognostic biomarkers, but their value in liver cirrhosis remains unclear. Our study, based on the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, aims to explore the association between CAR, GPR, and the prognosis of cirrhosis, providing quantitative evidence for clinical decision-making, optimizing treatment, and improving patient outcomes. Methods Based on clinical data from 3,649 cirrhosis patients in the MIMIC-IV database, the cohort was divided into two groups based on survival status, and baseline statistical descriptions were provided for each group. The CAR and GPR were categorized into quartiles, and their associations with prognosis were analyzed using Cox regression, Kaplan-Meier (K-M) curves, restricted cubic splines (RCS), and subgroup analysis. Predictive models were developed using machine learning algorithms, and performance was evaluated through the area under the curve (AUC), calibration curves, and decision curve analysis (DCA). Results Baseline characteristics results showed that patients in the non-survivor group (n = 846) had significantly higher CAR and GPR values, among other indicators, compared to those in the survivor group (n = 2,803) ( p < 0.05). Cox regression analysis found that elevated CAR was significantly associated with an increased 90-day mortality rate, whereas higher GPR was significantly associated with a decreased 90-day mortality rate. K-M curves revealed significant differences in 90-day all-cause mortality between different CAR and GPR groups ( p < 0.005). RCS analysis showed a significant non-linear relationship between CAR, GPR, and 90-day mortality ( p non-linear < 0.001). Subgroup analysis showed that CAR had a more pronounced prognostic impact in married individuals, White patients, and those with Medicaid insurance, while the protective effect of GPR was particularly notable in hypertensive patients. The Gradient Boosting Machine for Survival Analysis (GBM) model demonstrated the best predictive performance (AUC: 0.802-0.806), with superior discrimination, calibration, and clinical net benefit compared to other models. Conclusion CAR and GPR are important prognostic indicators for outcomes in patients with cirrhosis and have significant clinical application value.

BACKGROUND:The prevalence of chronic kidney disease (CKD), defined as a glomerular filtration rate (GFR) of <60 mL/min/1.73 m2 for >3 months, is rising in the global population. OBJECTIVE:To assess the global prevalence of CKD in cirrhosis and how it impacts the prognosis of these patients. DESIGN:The Chronic Liver Disease Evolution and Registry for Events and Decompensation consortium prospectively enrolled non-electively admitted cirrhosis patients from 127 sites globally, each with up to 100 patients. Data collected were demographics, comorbid conditions, cirrhosis history, hospital course and patient outcomes. Patients were divided into those with (CKD+) and without CKD (CKD-) and compared. We also compared patients from different World Bank income strata. RESULTS:Of 7040 inpatients enrolled, the global prevalence of CKD was 18.17%, with the highest prevalence observed in high-income countries (HICs), which paralleled their higher prevalence of metabolic syndrome. CKD+ patients had lower median enrolment GFR (32 (21, 44) mL/min/1.73 m2) when compared with CKD- patients (88 (63, 117) mL/min/1.73 m2, p<0.0001), associated with a more complex history of cirrhosis complications, with ascites occurring in 76.5% of CKD+ versus 61.1% of patients with CKD- (p<0.0001). The most common in-hospital complication was the development of AKI (59.4%) in CKD+ versus 27% in CKD- patients (p<0.0001). CKD was associated with higher in-hospital and 30-day postdischarge mortality (both p<0.0001). CONCLUSIONS:The presence of CKD negatively impacts the prognosis of admitted patients with cirrhosis in a global cohort. Meticulous management of ascites and lifestyle changes, especially in HICs, may improve the outcome of these patients.

Background & Aims: In patients with acutely decompensated cirrhosis (ADC) who present with clinically apparent precipitants (i.e., infections, acute liver injury), alterations in blood markers of inflammation associate with progression toward severe phenotypes (e.g., acute-on-chronic liver failure [ACLF]). However, it is unclear whether alterations in blood inflammatory markers associate with progression of ADC independently of precipitants. Methods: We prospectively enrolled 394 patients admitted for ADC who were classified into four phenotypes of increasing severity: no organ dysfunction (n = 168), organ dysfunction alone (n = 72), organ failure without ACLF (n = 91), and ACLF (n = 63). Clinical blood cell counts and serum levels of inflammatory markers (including soluble markers related to type-1, type-2, and type-3 inflammation) were obtained at enrollment. Ordinal regression with adjacent categories logit model adjusted for confounders (including precipitants) was used to analyze associations between changes in each blood inflammatory marker and the worsening of ADC. Results: Inflammatory markers that were associated with a higher risk of progressing to the next more severe stage were as follows: increasing neutrophil counts (adjusted common odds ratio [cOR] 1.17, 95% CI 1.06-1.28); increasing levels of the type-2 cytokine interleukin (IL)-25 (cOR 1.21, 95% CI 1.06-1.39), type-3 cytokines IL-6 (cOR 1.15, 95% CI 1.02-1.28) and IL-22 (cOR 1.16, 95% CI 1.03-1.30), or anti-inflammatory soluble CD163 (cOR 1.94, 95% CI 1.58-2.38); decreasing lymphocyte counts (cOR 0.77, 95% CI 0.68-0.87); or decreasing levels of the type-1 cytokine IFN-c (cOR 0.85, 95% CI 0.75-0.95). Conclusions: Among patients with ADC, alterations in blood levels of cytokines related to type-1, type-2 and type-3 inflammation, together with neutrophilia, lymphopenia and elevated anti-inflammatory signals were individually associated with an increased risk of progressing toward ACLF, independently of the presence of clinically apparent precipitants. (c) 2024 The Authors. Published by Elsevier B.V. on behalf of European Association for the Study of the Liver. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

BACKGROUND & AIMS:Cirrhosis is a major global burden requiring frequent hospitalizations and has high inpatient mortality. Traditional prognostic tools focused on inpatient mortality are affected by global disparities, which impacts timely management. We aimed to deploy machine learning (ML) approaches to enhance inpatient mortality prognostication. METHODS:Using the prospective Chronic Liver Disease Evolution and Registry for Events and Decompensation (CLEARED) cohort, which enrolled inpatients with cirrhosis globally, we used admission-day data to predict inpatient mortality with ML approaches vs logistic regression. Internal validation (75/25 split) and subdivision using World Bank income status (low/low-middle-income countries, upper-middle income countries, and high-income countries) were performed. The ML model with the best area under the receiver operating characteristic curve (AUROC) was externally validated in a US veteran inpatient population with cirrhosis. RESULTS:The CLEARED cohort included 7239 inpatients with cirrhosis (64% were men; mean [SD] age, 56 [13] years; median Model for End-Stage Liver Disease-Sodium score = 25) from 115 centers globally; 22.5% were from low/low-middle-income countries, 41% were from upper-middle-income countries, and 34% were from high-income countries; 11.1% of patients (n = 808) died in the hospital. Random forest (RF) showed the best AUROC (0.815) with high calibration, which was significantly better vs parametric logistic regression and LASSO models (AUROC, 0.774; P < .001 and AUROC, 0.787; P = .004, respectively). RF was the ML method with the highest AUROC and remained better than logistic regression, regardless of country income level (high-income countries: AUROC, 0.806; upper middle-income countries: AUROC, 0.867; and low/low-middle-income countries: AUROC, 0.768). External validation was performed in 28,670 veterans (96% were men, mean [SD] age, 67.8 [10.3] years, median Model for End-Stage Liver Disease-Sodium score = 15) with 4% (n = 1158) inpatient mortality. The AUROC using the CLEARED-derived RF model was 0.859. CONCLUSIONS:RF analysis trained on a global prospective cirrhosis cohort enhances mortality prediction over traditional methods, was consistent across country income levels, and was successfully validated externally in a US veteran population.

Purpose. To evaluate the level and influencing factors of shared decision making in patients with chronic hepatitis B and explore its effects on antiviral treatment adherence. Methods. A cross-sectional study was conducted with 239 chronic hepatitis B patients using a general information and disease-related information questionnaire, the 9-item Shared Decision Making Questionnaire, the Self-Efficacy for Appropriate Medication Use Scale, and the Morisky Medication Adherence Scale-8. A structural equation model was built to analyze the pathways through which shared decision making affects medication adherence. Results. The standardized shared decision-making score for patients with chronic hepatitis B was 71.99 ± 9.22, with higher levels of shared decision making observed in males and patients with cirrhosis. Shared decision making significantly affected self-efficacy for appropriate medication use (β = 0.568, P < 0.001) and adherence (β = 0.413, P < 0.001). Moreover, self-efficacy for appropriate medication mediated the relationship between shared decision making and adherence (β = 0.219, P = 0.002). Conclusions. Patients with chronic hepatitis B show above-average levels of shared decision making, with considerable individual differences. Promoting the use of shared decision making can significantly improve patients' medication self-efficacy and adherence to treatment. Highlights:Shared decision making (SDM) in chronic hepatitis B (CHB) patients is at an above-average level, with significant individual differences.Gender and cirrhosis status are key independent factors influencing SDM levels.Higher SDM levels positively affect both medication self-efficacy and adherence, with self-efficacy serving as a mediator.

Metabolic dysfunction-associated steatotic liver disease (MASLD) remains a prevalent condition with limited diagnostic and therapeutic options. This study aims to identify metabolic signatures of disease progression and develop non-invasive diagnostic models through three independent cohorts (including two cohorts confirmed by biopsy and one cohort confirmed by ultrasound) involving 293 participants for detecting significant fibrosis (≥F2) and mild to severe inflammatory activity (≥I2) using multiple machine learning techniques. The fibrosis panel shows area under the receiver operating characteristic curve (AUROC) of 0.928 (95% confidence interval [CI]: 0.835-0.978), 0.829 (0.732-0.902), and 0.806 (0.724-0.872) in the discovery cohort, validation cohort 1, and validation cohort 2, respectively, outperforming the fibrosis-4 index (FIB-4), aspartate aminotransferase-to-platelet ratio index (APRI), non-alcoholic fatty liver disease fibrosis score (NFS), liver stiffness measurement (LSM), and combination of hoMa, Ast and CK18 (MACK-3). The inflammation panel achieves AUROCs of 0.894 (0.791-0.957) and 0.776 (0.673-0.859) in the discovery cohort and validation cohort 1, respectively. The key metabolites guanidinoacetic acid (GAA) and sebacic acid (SA) demonstrate therapeutic efficacy in mice. These validated panels provide accurate stratification of MASLD severity, and GAA/SA offer therapeutic potential, advancing both diagnosis and treatment strategies.

BACKGROUND:Hepatitis E virus (HEV) infection is endemic in China. However, there are scarce data of HEV infection among hospital attendees seeking medical treatment or examination for various reasons. OBJECTIVE:We aim to investigate the prevalence and incidence of HEV infection by time, age, sex, and across departments in a tertiary hospital. METHODS:Paired results of anti-HEV immunoglobulin G (IgG) and IgM of 31,181 unique subjects during 2021-2022 were analysed. RESULTS:Overall seropositivity (95% confidence interval) of anti-HEV IgG and IgM was 41.25% (40.71%-41.80%) and 2.35% (2.19%-2.53%), respectively. Acute hepatitis E was more prevalent during winter-early spring and among adults aged 31-70. Anti-HEV IgG seroprevalence increased with age, levelling off at > 60 years of age. Not only the seropositivity, but also the levels of anti-HEV IgG were significantly lower in women than men of middle and old age. Young patients from the Department of Neurology had a significantly higher ratio of past HEV infection, while patients with manifestations of hepatitis, gastrointestinal symptoms or hematological diseases had higher seropositivity of anti-HEVIgM and should have high priority to HEV screening. CONCLUSION:Heterogeneity of HEV seroprevalence was noted at different times of the year, between sexes, among age groups and across departments in general hospital. The concentration of HEV-infected patients in a few departments supports a more focused screening strategy in health-care settings.