Predicting 90-day mortality in patients with HBV-ACLF using machine learning tools DOI Creative Commons
Juan Liu, Wentao Zhu, Ting Deng

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 4, 2024

Abstract Background Acute chronic liver failure (ACLF) is characterized by a systemic inflammatory response, mainly associated with hepatitis B virus (HBV) in the Asia-Pacific region, and has high mortality rate. We aimed to develop stable feasible prognostic prediction model based on machine learning (ML) tools predict 90-day patients virus-associated acute-on-chronic (HBV-ACLF). Method Clinical data from 573 HBV-ACLF across two hospitals were retrospectively collected. Prognostic models of constructed using support vector (SVM), decision tree (DT), random forest (RF), K nearest neighbour (KNN), least absolute shrinkage selection operator (LASSO), logistic regression (LR). Model performance metrics included accuracy, area under (AUC) receiver operating characteristic curve, sensitivity, specificity, positive predictive value (PPV), negative (NPV). Results In training cohort, RF demonstrated significantly higher AUC, PPV, NPV than LASSO, LR, SVM, DT, KNN models. However, AUC validation cohort was 0.728, decline PPV 0.688, 0.545, 0.655, respectively. LASSO had lowest at 0.739, while sensitivity 0.694. testing cohorts, SVM DT exhibited both 0.581. Although LR performed less effectively it outperformed cohorts. Conclusions summary, demonstrates efficacy greater stability, making more practical for clinical treatment decision-making.

Language: Английский

Recent advances in the prevention and treatment of decompensated cirrhosis and acute-on-chronic liver failure (ACLF) and the role of biomarkers DOI Creative Commons
Jonel Trebicka, Rubén Hernáez, Debbie L. Shawcross

et al.

Gut, Journal Year: 2024, Volume and Issue: 73(6), P. 1015 - 1024

Published: March 25, 2024

The progression of cirrhosis with clinically significant portal hypertension towards decompensated remains challenging and the evolution acute-on-chronic liver failure (ACLF), one or more extrahepatic organ failures, is associated very high mortality. In last decade, progress has been made in understanding mechanisms leading to decompensation ACLF. As advances, bacterial translocation across an impaired gut barrier culminates endotoxaemia, systemic inflammation cirrhosis-associated immune dysfunction (CAID). Gut-derived CAID have become logical targets for innovative therapies that prevent hepatic episodes Furthermore, classification disease biomarker discovery personalise care advanced field. This review discusses personalisation treatment

Language: Английский

Citations

11
Acute on chronic liver failure DOI
Jonel Trebicka, Michael Praktiknjo, Kai‐Henrik Peiffer

et al.

Deutsches Ärzteblatt international, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 18, 2025

Cirrhosis is the end stage of chronic liver disease. causes portal hypertension, which, in turn, can lead to acute on failure (ACLF), which defined as decompensation combined with liver, coagulation system, kidneys, lungs, and/or circulatory or hepatic encephalopathy. This review based a selective literature search for international publications NCBI database using keywords "liver cirrhosis" and "ACLF." Valid guidelines (up June 2024) were also included. present approximately 1% all hospital admissions Germany, complications cirrhosis cause an estimated 1 million deaths worldwide each year. ACLF, most severe form decompensated cirrhosis, bears 28-day mortality 45% affects 35% patients hospitalized cirrhosis. Its precipitating factors are infection, alcohol overuse, bleeding, drug-induced encephalopathy (benzodiazepines opioids). No drugs other treatments ACLF have been approved; only its etiology amenable treatment. Liver transplantation currently curative option but not suitable because narrow therapeutic window common presence contraindications. Prospective data lacking that would aid selection so post-transplantation survival rates be improved.

Language: Английский

Citations

1
A nomogram model based on albumin-bilirubin score for predicting 90-day prognosis in patients with acute-on-chronic liver failure DOI Creative Commons
Wei Ding,

Jiandong Shen,

Li Zhang

et al.

Frontiers in Medicine, Journal Year: 2025, Volume and Issue: 11

Published: Jan. 6, 2025

To develop a nomogram model based on the albumin-bilirubin (ALBI) score for predicting 90-day prognosis of patients with acute-on-chronic liver failure (ACLF) and to evaluate its predictive efficacy. Clinical data 290 ACLF at Third People's Hospital Nantong City, collected from December 2020 2023, were analyzed. The divided into training set (n = 200) validation 90), August 2022 as cut-off date. Patients in categorized an improvement group 133) mortality 67) their outcomes. power baseline parameters was assessed using univariate multivariate logistic regression construct model. Model performance receiver operating characteristic (ROC) curves, calibration decision curve analysis (DCA) Hosmer-Lemeshow test. Creatinine (CR) [odds ratio (OR) 1.013, 95% confidence interval (CI): 1.004-1.022], ALBI (OR 10.831, CI: 4.009-33.247), Gender 1.931, 0.973-3.870) ascites 3.032, 1.249-8.178) identified independent prognostic factors. formula derived index (PI) -0.591 + 0.658 × 1.109 0.012 CR 2.382 ALBI. area under (AUC) 0.804 (95% 0.741-0.866), specificity 85.0% sensitivity 65.7% 0.425. AUC 0.811 0.697-0.926). test indicated good fit p-value 0.287 0.423 set. Calibration curves demonstrated accuracy model, DCA results suggested that clinically useful when threshold below 0.6. incorporating CR, can predict patients, potentially helping optimize treatment strategies improve patient

Language: Английский

Citations

0
Acute and acute‐on‐chronic liver failure: Pathogenesis, management and perspectives DOI Open Access
Jonel Trebicka, Qing Xie

Liver International, Journal Year: 2025, Volume and Issue: 45(2)

Published: Jan. 28, 2025

Acute liver failure (ALF) and acute-on-chronic (ACLF) represent critical junctures in the spectrum of diseases, characterized by rapid deterioration function often multi-organ dysfunction. Despite advances medical care, they remain significant challenges clinical practice, necessitating a deeper understanding their pathophysiology development effective therapeutic strategies. This special issue intents to address these topics 15 selected reviews. The first two reviews are ALF pathogenesis therapy focus.1, 2 is rare but life-threatening condition sudden loss individuals with no pre-existing disease. It associated high mortality rate, typically ranging from 50% 80%. aetiology can be diverse, including viral hepatitis, drug-induced injury, autoimmune hepatitis acute ischaemic injury among others (Figure 1). Regardless underlying cause, hallmark massive hepatocyte death, leading impaired synthetic metabolic functions liver. presentation vary widely, common features include jaundice, coagulopathy, hepatic encephalopathy, often, deterioration. management involves intensive supportive measures maintain hemodynamic stability, correct coagulopathy manage complications such as cerebral oedema encephalopathy.1 Liver transplantation remains only definitive treatment option for many patients ALF, offering possibility long-term survival. However, availability donor organs timing crucial factors that significantly impact outcomes. In recent years, there has been growing interest artificial support systems bridge or promote regeneration ALF. These aim remove toxins, imbalances provide temporary while awaiting recovery transplantation. Various modalities, extracorporeal devices bioartificial systems, have investigated, efficacy subject debate.2 ACLF, on other hand, represents distinct syndrome chronic disease, most commonly cirrhosis. Hernaez colleagues summarized comprehensive review definition, diagnosis epidemiology ACLF.3 ACLF develops setting decompensation may also develop stage decompensation, so-called non-acute decompensation.4 For precipitating event, bacterial infection, alcoholic gastrointestinal bleeding, triggers systemic inflammatory response exacerbates injury. But necessary interventions 2) induce precipitate Praktiknjo et al.5 dysregulated immune leads widespread tissue damage organ dysfunction, extra-hepatic failures, driving extreme short-term 30% 50%.6 Alcoholic specific trigger still unaddressed condition.7 Similarly hepatitis,7 complex multifactorial, involving interactions between system, gut microbiota inflammation. progress made, we expect technological (omics) digital (artificial intelligence) improve understanding, potentially design approaches.8 Especially using large cohorts omics data sets will facilitate construction algorithms predict prognosis ACLF. Existing future reviewed Valainathan al.9 Current focuses treating factors, providing care and, some cases, considering life-saving intervention.10 role inflammation dysregulation demonstrated over last this exploration novel targets, interleukin-22 Hwang al.11 Strategies aimed at modulating response, corticosteroids, immunomodulatory agents anti-inflammatory cytokines, shown promise preclinical studies early-phase trials.12 approaches multimodal, indeed, one shape strategy based molecular mechanism supporting clear rationale 3), nicely Morrison al.13 addition pharmacological interventions, increasing use potential support, toxins mediators, regeneration. While early promising results, larger randomized controlled trials needed establish ACLF.14, conclusion, morbidity mortality. need further research better understand conditions more targeted therapies holds improving outcomes validate safety practice. Jonel Trebicka was supported German Research Foundation (DFG) project ID 403224013—SFB 1382 (A09), Federal Ministry Education (BMBF) DEEP-HCC Hessian Higher Education, Arts (HMWK) ENABLE ACLF-I cluster projects. MICROB-PREDICT (project 825694), DECISION 847949), GALAXY 668031), LIVERHOPE 731875) IHMCSA 964590) projects received funding European Union's Horizon 2020 innovation program. manuscript reflects authors' views, Commission not responsible any made information it contains. funders had influence study design, collection analysis, decision publish preparation manuscript. speaking and/or consulting fees Versantis, Gore, Boehringer-Ingelheim, Falk, Grifols, Genfit CSL Behring. article does contain primary patient nor did involve human animal subjects. As such, ethical approval required production article. data, materials sources. authors writing assistance disclose. editorial present new synthesis existing findings. Therefore, sets, code submission. All included publicly available literature references cited within text.

Language: Английский

Citations

0
Clinical and Pathophysiological Characteristics of Non-acute Decompensation of Cirrhosis DOI
Nipun Verma, Parminder Kaur, Pratibha Garg

et al.

Journal of Hepatology, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Language: Английский

Citations

0
Liver stiffness as a dynamic predictor of decompensation and mortality in alcohol-related liver disease DOI Creative Commons
Majeed M. A. Ali

Deleted Journal, Journal Year: 2024, Volume and Issue: 1(1)

Published: Nov. 5, 2024

Alcohol-related liver disease (ALD) is a significant global health concern, contributing substantially to liver-related morbidity and mortality. Liver stiffness measurement (LSM) has emerged as non-invasive, reliable method for assessing fibrosis cirrhosis in ALD. This review examines the role of LSM predicting monitoring risk decompensation mortality ALD patients. We conducted comprehensive analysis recent literature on techniques, including transient elastography, magnetic resonance acoustic radiation force impulse imaging, shear wave elastography. The focused pathophysiology ALD, predictive value hepatic mortality, its use progression treatment response. demonstrates strong power with values above 20–25 kPa generally associated increased risk. Serial measurements provide valuable insights into effects alcohol abstinence or relapse. Integration other biomarkers prognostic models enhances stratification clinical decision-making. However, challenges remain standardizing techniques establishing ALD-specific cutoff values. represents advance management, offering dynamic repeatable measure health. Its ability predict outcomes monitor response makes it tool guiding care. Future research should focus standardization, multimodal approaches combining biomarkers, application artificial intelligence data further improve utility management.

Language: Английский

Citations

3
Definition and prognosis of acute-on-chronic liver failure DOI

FE Uschner,

Jonel Trebicka

Deleted Journal, Journal Year: 2024, Volume and Issue: 19(4), P. 322 - 330

Published: May 22, 2024

Das akut-auf-chronische Leberversagen (ACLF) bezeichnet das Endstadium chronischer Lebererkrankungen und ist durch eine hohe Mortalität gekennzeichnet. In dieser Übersichtsarbeit sollen Pathophysiologie, Definition Prognoseabschätzung des ACLF erläutert werden. Literaturrecherche, Analyse Diskussion bestehender Grundlagenarbeiten zum Thema ACLF. entsteht in Folge einer schweren systemischen Inflammation, die verschiedene Auslöser (Infektionen, akute Alkoholhepatitis, Blutungen) induziert wird einem Circulus vitiosus den Krankheitsprozess unterhält vorantreibt. Neue Prognosescores, bestehend aus Lebersynthesemarkern elastographischen Verfahren, können dabei helfen, Risikopatienten zu identifizieren. Krankheitsdefinierend für dekompensierte Leberzirrhose mit neu aufgetretenem (Mehr‑)Organversagen. Die möglichen Organversagen werden anhand abgewandelten Chronic-liver-failure(CLIF)-sequential-organ-failure-assessment(SOFA)-Scores festgelegt (Leber, Gerinnung, Niere, Enzephalopathie, Atmung Kreislauf) gehen zusammen Patientenalter Markern der Inflammation CLIF-consortium(C)-ACLF-Score ein. Nierenversagen hepatische Enzephalopathie zeigen besondere Relevanz sind schlechten Prognose assoziiert. Der CLIF-C-ACLF-Score dafür geeignet, Patienten identifizieren deren 30-Tage-Mortalität abzuschätzen. ein lebensbedrohliches Krankheitsbild. Prognosescores Zukunft dazu beitragen, frühzeitige Diagnose eines sowie individuellen entscheidend Therapie.

Citations

0
Safe and successful gut-restricted adsorbent strategy against cirrhosis and acute-on-chronic liver failure DOI
Van der Merwe, Maite G. Fernández‐Barrena

Gut, Journal Year: 2024, Volume and Issue: unknown, P. gutjnl - 332457

Published: Oct. 26, 2024

Language: Английский

Citations

0
Predicting 90-day mortality in patients with HBV-ACLF using machine learning tools DOI Creative Commons
Juan Liu, Wentao Zhu, Ting Deng

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 4, 2024

Abstract Background Acute chronic liver failure (ACLF) is characterized by a systemic inflammatory response, mainly associated with hepatitis B virus (HBV) in the Asia-Pacific region, and has high mortality rate. We aimed to develop stable feasible prognostic prediction model based on machine learning (ML) tools predict 90-day patients virus-associated acute-on-chronic (HBV-ACLF). Method Clinical data from 573 HBV-ACLF across two hospitals were retrospectively collected. Prognostic models of constructed using support vector (SVM), decision tree (DT), random forest (RF), K nearest neighbour (KNN), least absolute shrinkage selection operator (LASSO), logistic regression (LR). Model performance metrics included accuracy, area under (AUC) receiver operating characteristic curve, sensitivity, specificity, positive predictive value (PPV), negative (NPV). Results In training cohort, RF demonstrated significantly higher AUC, PPV, NPV than LASSO, LR, SVM, DT, KNN models. However, AUC validation cohort was 0.728, decline PPV 0.688, 0.545, 0.655, respectively. LASSO had lowest at 0.739, while sensitivity 0.694. testing cohorts, SVM DT exhibited both 0.581. Although LR performed less effectively it outperformed cohorts. Conclusions summary, demonstrates efficacy greater stability, making more practical for clinical treatment decision-making.

Language: Английский

Citations

0