Critical care hepatology: A narrative review of current concepts and management

Abstract

Patients with advanced liver disease requiring intensive care were long considered to have limited benefit from aggressive treatment, given the high associated mortality. Over recent decades, outcomes have improved with advances in liver transplantation, critical care, and a better understanding of multiorgan failure. Acute liver failure and acute-on-chronic liver failure are now recognized as complex multisystem syndromes in which survival is determined not only by hepatic dysfunction but also by extrahepatic organ failure, systemic inflammation, and infection. Critical care hepatology has therefore evolved as a clinical discipline integrating hepatology, intensive care, and transplant medicine. Modern management requires adaptation of conventional intensive care strategies to the distinct pathophysiological features of liver failure syndromes, together with coordinated multidisciplinary care, often delivered within specialized liver intensive care units. Despite substantial progress, important challenges remain, including limitations of current prognostic models, ongoing debate surrounding transplant allocation policies, and complex decisions regarding transplant candidacy and futility. A structured approach to critical care hepatology may improve care integration and outcomes in critically ill patients with liver disease.

Key Words: Critical care hepatology; Acute-on-chronic liver failure; Liver cirrhosis; Acute liver failure; Liver transplantation

Core Tip: Critical care hepatology has emerged as a distinct multidisciplinary field integrating hepatology, intensive care, and transplant medicine. Modern management of acute liver failure and acute-on-chronic liver failure extends beyond supportive care and requires individualized strategies for circulatory dysfunction, kidney injury, infection, neurological complications, and rebalanced hemostasis. This review summarizes current concepts in organ support, antimicrobial stewardship, extracorporeal therapies, and liver transplantation, while highlighting the growing role of specialized liver-focused intensive care models in improving outcomes for critically ill patients with liver disease.

INTRODUCTION

Patients with liver failure were traditionally considered to have limited benefit from intensive care, and intensive care unit (ICU) admission was often deferred based on perceived futility[1]. This reflected limited therapeutic options and an incomplete understanding of multiorgan failure, with mortality in acute liver failure (ALF) remaining extremely high before modern ICU management.

Over subsequent decades, this perspective has changed. The development of specialized liver units, beginning with Roger Williams’ Liver Unit at King’s College Hospital in 1966, improved the organization of care for severe liver disease[2]. Liver transplantation, developed in earlier decades and established during the 1980s, with 1-year survival exceeding 85% by the early 2000s, shifted care toward bridging patients to definitive treatment[3,4]. Advances in intensive care further improved the management of circulatory failure, renal dysfunction, and neurological complications, reducing mortality[5]. In ALF, the incidence of intracranial hypertension declined from 76% in the mid-1980s to around 20% by the mid-2000s, with associated mortality decreasing from 95% to 55%[6]. Organ allocation systems, including model for end-stage liver disease and pediatric end-stage liver disease, further refined prioritization based on short-term mortality risk[7].

These developments led to the recognition of critical care hepatology as a distinct subspecialty. The United States Acute Liver Failure Study Group, established in 1997, enabled multicenter research[8], while Asian Pacific Association for the Study of the Liver (2009) and the European Association for the Study of the Liver (EASL)-Chronic Liver Failure Consortium (2013) formalized acute-on-chronic liver failure (ACLF) as a distinct, high-mortality syndrome[9,10]. By the 2020s, disease-specific Medical Intensive Liver Units had emerged, reflecting the need for specialized multidisciplinary care[11].

ALF and ACLF are major causes of hospitalization associated with high short-term mortality in the absence of transplantation. ALF is relatively rare, with an incidence of 1.1-1.6 per 100000 person-years in Western countries, whereas ACLF affects 7%-10% of hospitalized patients with cirrhosis and carries 28-day mortality exceeding 30%-40% and 90-day mortality approaching 50%-55%. The burden varies geographically, with drug-induced liver injury predominating in the West and viral hepatitis and alcohol-related disease more common in Asia, while infection and alcohol remain key precipitants globally. These data emphasize the growing burden of liver failure in both specialized and general intensive care settings[12].

This narrative review is based on a focused appraisal of the literature, including key clinical studies, guidelines, and consensus statements identified through PubMed searches and reference list screening, with emphasis on clinically relevant evidence in critical care hepatology. With increasing numbers of patients with ALF, ACLF, and advanced cirrhosis requiring ICU care, critical care hepatology has become an integral component of modern practice[13,14]. Prognosis is determined not only by liver dysfunction but also by extrahepatic organ failure, infection, and systemic inflammation, as illustrated in Figure 1[15-19]. This narrative review addresses clinical management, the evolving role of liver transplantation in critical illness, and the organizational models that define modern critical care hepatology.

Figure 1 Systemic consequences of liver dysfunction and the development of extrahepatic organ failure although liver injury is the defining feature of acute liver failure and acute-on-chronic liver failure, patient outcomes are largely determined by the development of extrahepatic organ dysfunction. The figure highlights the major pathophysiological pathways linking hepatic failure to renal, neurological, cardiopulmonary, coagulation, metabolic, and immune abnormalities, emphasizing the systemic nature of critical illness in patients with severe liver disease and the need for multidisciplinary management. RAAS: Renin-angiotensin-aldosterone system; HRS: Hepatorenal syndrome; TNF: Tumor necrosis factor; IL: Interleukin; SIRS: Systemic inflammatory response syndrome; LPS: Lipopolysaccharide; DIC: Disseminated intravascular coagulation; ALF: Acute liver failure; ARDS: Acute respiratory distress syndrome; HRS-AKI: Hepatorenal syndrome-acute kidney injury.

NARRATIVE REVIEW METHODOLOGY

This narrative review was developed in accordance with the Scale for Assessment of Narrative Review Articles framework. A focused literature search was performed using PubMed/MEDLINE as the primary database, supplemented by targeted searches of Google Scholar and guideline repositories. English-language publications from 1990 through February 2026 were reviewed, with emphasis placed on clinical practice guidelines, consensus statements, landmark trials, multicenter cohort studies, and clinically relevant evidence related to ALF, ACLF, liver transplantation, organ support strategies, and intensive care management in hepatology. Additional references were identified through manual review of bibliographies from key articles and society guidance documents, including publications from EASL, AASLD, Asian Pacific Association for the Study of the Liver, and ADQI. Priority was given to contemporary evidence with direct applicability to critically ill patients with liver disease.

CLINICAL MANAGEMENT: BEYOND SUPPORTIVE CARE

The management of critically ill patients with liver disease has evolved from predominantly supportive care to an approach that targets the underlying pathophysiologic mechanisms of organ failure. Advances in the understanding of systemic inflammation, circulatory dysfunction, immune dysregulation, and metabolic disturbances in ALF and ACLF have reshaped ICU management[5,20]. Contemporary care emphasizes early recognition of organ dysfunction, prompt treatment of precipitating factors such as infection or bleeding, and prevention of secondary organ injury[21]. Management increasingly relies on coordinated multidisciplinary care involving hepatologists, intensivists, transplant specialists, nephrologists, and infectious disease physicians[21]. In parallel, advances in bedside monitoring and organ support technologies have improved the precision of hemodynamic, renal, and coagulation management. The following sections focus on circulatory support, renal replacement therapy (RRT), infection management, neuroprotection, and coagulation, with particular attention to the challenges specific to patients with acute or chronic liver failure. Indications for intensive care admission in patients with acute and chronic liver failure are summarized in Table 1.

Table 1 Indications for intensive care admission in critically ill patients with liver-driven disease.

Clinical context	Definition	Indications for ICU level care
Acute decompensation of cirrhosis	Acute worsening of cirrhosis (ascites, HE, bleeding, infection) without established extrahepatic organ failure, primarily driven by portal hypertension	Impaired consciousness requiring airway protection (HE grade III-IV or active bleeding); severe infection requiring hemodynamic monitoring; acute variceal bleeding requiring endoscopic or radiologic intervention (including rescue TIPS), particularly with instability; early signs of deterioration with risk of progression to ACLF
ACLF	Acute decompensation with ≥ 1 organ failure and high short-term mortality	Requirement for advanced organ support (vasopressors, renal replacement therapy, invasive ventilation); failure of ward-based management; ACLF grade ≥ 2; evolving multiorgan dysfunction; persistent hyperlactatemia or worsening shock despite resuscitation
ALF	Acute liver injury with coagulopathy and HE in a previously non-cirrhotic liver	Any degree of encephalopathy due to risk of cerebral edema; significant coagulopathy (e.g., INR > 2); rapidly evolving biochemical failure; need for early transplant assessment and neurocritical monitoring
Non-hepatic or peri-procedural indications in liver disease	Patients with liver disease admitted for non-hepatic illness or post-intervention monitoring	Post-procedural observation (e.g., TIPS, interventional radiology); postoperative care after non-hepatic surgery; acute cardiovascular or neurological events; need for intensified monitoring due to limited physiological reserve in cirrhosis

Summary of key clinical scenarios and indications for intensive care admission in patients with acute and chronic liver failure, reflecting the need for timely escalation of organ support and transplant evaluation. ICU: Intensive care unit; HE: Hepatic encephalopathy; TIPS: Transjugular intrahepatic portosystemic shunt; ACLF: Acute-on-chronic liver failure; ALF: Acute liver failure; INR: International normalized ratio.

Hemodynamic management and circulatory dysfunction

Circulatory dysfunction in ACLF: Circulatory dysfunction in ACLF reflects an interplay of systemic vasodilation, cardiac dysfunction, and impaired vascular responsiveness[10,22]. The hyperdynamic circulation, characterized by increased cardiac output, reduced systemic vascular resistance, and relative central hypovolemia, becomes maladaptive during acute decompensation. Superimposed systemic inflammation, driven by infection, alcohol-related injury, or other precipitants, further amplifies vasodilatation and impairs vascular responsiveness to both endogenous and exogenous vasoconstrictors[20,23]. The net result is a distinct circulatory failure phenotype with overlapping characteristics of distributive and cardiogenic shock, requiring tailored hemodynamic assessment and management.

Cirrhotic cardiomyopathy (CCM), characterized by impaired contractile reserve and diastolic dysfunction, further limits the cardiac response to acute stress and predisposes to hepatorenal syndrome-acute kidney injury (HRS-AKI). CCM results from chronic exposure to the hyperdynamic circulatory state, with downregulation of β-adrenergic receptors, impaired calcium channel function, and overactivity of negative inotropic mediators, including nitric oxide, all contributing to progressive myocardial dysfunction and a blunted response to vasopressors. CCM is common and clinically relevant, identified in 34.7% of patients overall and 75.4% of those with HRS-AKI[5]. It independently predicted mortality at 90 days (adjusted hazard ratio = 8.9) and 1 year (adjusted hazard ratio = 1.7), and was associated with increased need for RRT[24].

Point-of-care ultrasound (POCUS) has become an integral tool in the hemodynamic assessment of critically ill patients with ACLF, where clinical examination alone is often unreliable and static hemodynamic parameters can be misleading. A structured POCUS evaluation incorporating assessment of left and right ventricular function, inferior vena cava respiratory variations, and lung ultrasound to detect pulmonary congestion can help differentiate among volume depletion, distributive circulatory failure, and cardiac dysfunction at the bedside. A more advanced hemodynamic ultrasound approach includes venous excess ultrasound (VexUS), which integrates Doppler assessment of the hepatic, portal, and intrarenal veins to quantify systemic venous congestion and may improve volume phenotyping beyond what is possible with conventional physical examination[25]. In patients with cirrhosis and AKI, this approach may help distinguish true hypovolemia from congestive nephropathy, potentially reducing iatrogenic fluid overload associated with empiric albumin administration[25,26]. VexUS has also been proposed as a noninvasive adjunct in congestive hepatopathy and hepatic pressure mapping, although altered hepatic architecture and tense ascites may limit waveform interpretation in advanced cirrhosis[27]. Further studies are needed to validate liver-specific VexUS protocols and determine their impact on outcomes in hepatocardiorenal syndrome[28]. In a prospective cohort of patients with cirrhosis with AKI, 79.6% were hypovolemic at ICU admission, highlighting frequent under-recognition of volume depletion despite advanced liver disease[24]. POCUS-guided reassessment refined AKI classification and supports the growing role of multiorgan ultrasound in individualized hemodynamic evaluation.

Invasive hemodynamic monitoring should be reserved for selected patients in whom clinical examination, POCUS, and VexUS remain insufficient to guide resuscitation, particularly refractory shock, mixed distributive-cardiogenic physiology, severe CCM, persistent hypoperfusion, or transplant evaluation[29,30]. Transpulmonary thermodilution with PiCCO can quantify cardiac output, preload, extravascular lung water, and pulmonary vascular permeability, which may be useful in cirrhotic patients with septic shock at high risk of fluid overload[30]. In ALF, invasive intracranial pressure monitoring has declined in use and should be individualized, as recent data suggest limited routine utility, though it may be useful in specialized centers[31,32]. Evidence remains limited, and monitoring should support, not replace, repeated physiological reassessment[29,31].

Volume expansion is a key initial step in managing circulatory dysfunction in ACLF[33]. Trials of liberal or prophylactic albumin strategies, including ATTIRE and INFECIR, did not demonstrate a mortality benefit and were associated with increased adverse events[34,35]. Consequently, albumin use in cirrhosis has been refined to well-defined indications, namely initial resuscitation and specific conditions such as spontaneous bacterial peritonitis, large-volume paracentesis, and AKI. When crystalloids are required, balanced solutions such as Ringer’s lactate or Plasma-Lyte are preferred over normal saline due to the risk of hyperchloremic acidosis and adverse renal effects[36].

Norepinephrine is recommended as first-line vasopressor therapy restoring mean arterial pressure through α-adrenergic vasoconstriction[29,33]. Higher doses are often required due to the marked systemic vasodilation characteristic of advanced cirrhosis and ACLF, and titration should target a mean arterial pressure of at least 65 mmHg, with higher targets individualized according to renal and cerebral perfusion[29]. Vasopressin acts on V1 receptors to produce splanchnic and systemic vasoconstriction independently of adrenergic pathways, and it is an appropriate second-line agent when norepinephrine requirements are escalating, though its use requires careful monitoring given the risk of mesenteric and hepatic ischemia at higher doses[29]. Terlipressin, a synthetic vasopressin analog with greater V1 selectivity and a longer duration of action, is widely used in HRS, where it has established efficacy in improving renal function and HRS reversal compared with placebo, although its survival benefit remains inconsistent across trials[37,38]. Angiotensin II, which targets the renin-angiotensin system, has been reported in selected cases of cirrhosis with septic shock, but evidence remains limited, and its role is not yet established in routine practice[39,40].

Inotropic agents are not routinely indicated in liver failure, where circulatory dysfunction is predominantly vasodilatory. Their use is reserved for patients with documented systolic dysfunction confirmed on echocardiography, mostly for patients with persisting signs of tissue hypoperfusion despite adequate fluid loading and stabilization of hemodynamics with vasopressor support[41]. Dobutamine is the most commonly used inotrope when required, although β-adrenergic receptor downregulation in CCM may attenuate its effect[42]. When used, it should be combined with a vasopressor given the risk of worsening hypotension through its vasodilatory β2 effects.

Circulatory dysfunction in ALF: In ALF, circulatory dysfunction presents as a vasodilatory state with low systemic vascular resistance and high cardiac output, driven by inflammatory mediators and nitric oxide[43]. It is a consequence of massive acute hepatocyte necrosis and the resulting systemic inflammatory response. Unlike ACLF, where circulatory dysfunction evolves over a background of chronic hyperdynamic physiology with established neurohormonal compensation, ALF produces acute and rapidly progressive vascular collapse in a previously normal circulation, often without time for compensatory adaptation[44].

Fluid resuscitation must be carefully titrated to restore perfusion while avoiding exacerbation of cerebral edema - a complication that is not a typical feature of other shock states. EASL Clinical Practice Guidelines on ALF identify intracranial hypertension as a major determinant of fluid strategy in ALF and caution against generic resuscitation approaches not adapted to this specific risk[43]. When volume expansion is required, isotonic crystalloids are generally used and albumin may be considered as an adjunct, though ALF-specific evidence remains limited, and its role is largely extrapolated from cirrhosis and general critical care practice.

The approach to vasopressor and inotrope therapy differs between ALF and ACLF. Vasopressors are frequently required, in ALF with norepinephrine as first-line therapy, consistent with practice in ACLF. Vasopressin may be added as a norepinephrine-sparing agent in refractory cases, though its use in ALF is empirical and carries theoretical risk of worsening hepatic microcirculatory perfusion in the already failing liver[43]. Terlipressin, a cornerstone of HRS-AKI management in ACLF, has no established role in ALF[45]. When cardiac dysfunction is confirmed by echocardiography, dobutamine is the preferred inotrope in ALF, and may retain greater efficacy than in ACLF given that β-adrenergic receptors have not been subject to the chronic downregulation. Hemodynamic instability contributes to early mortality and complicates transplantation, with postreperfusion syndrome occurring in approximately 41% of emergency liver transplants[46]. Key evidence guiding vasoactive therapy in acute and ACLF is shown in Table 2.

Table 2 Key evidence guiding vasoactive therapy in acute and acute-on-chronic liver failure.

Ref.	Population (n)	Design	Intervention	Comparator	Primary outcome	Key findings	Clinical relevance
Cavallin et al[94], 2015	Cirrhosis, HRS (n = 49)	RCT	Terlipressin 3-12 mg/day CTI + albumin	Midodrine + octreotide + albumin	Renal recovery	Terlipressin superior: Partial/complete response 70% vs 29% (P = 0.01); complete response 56% vs 5% (P < 0.001)	First head-to-head terlipressin vs midodrine/octreotide; terlipressin clearly favoured
Boyer et al[95], 2016	Cirrhosis, HRS-1 (n = 196)	RCT	Terlipressin 1 mg every 6 hours + albumin	Placebo + albumin	CHRSR	CHRSR 19.6% vs 13.1% (P = 0.22; NS); creatinine reduction significantly greater (P < 0.001); survival similar	Phase 3; missed primary endpoint; HRS reversal predicted 90-day survival; more ischaemic AEs
Piano et al[96], 2018	298 patients with cirrhosis and type 1 HRS, stratified by ACLF grade	Retrospective multicenter cohort study	Terlipressin + albumin	Comparison according to ACLF grade (ACLF-1, ACLF-2, ACLF-3) rather than a separate treatment arm	Response to treatment (serum creatinine < 1.5 mg/dL at end of therapy) and 90-day mortality	Treatment response declined with increasing ACLF severity: 60% in ACLF-1, 48% in ACLF-2, and 29% in ACLF-3 (P < 0.001). ACLF grade and baseline creatinine were independently associated with treatment response. ACLF grade independently predicted 90-day mortality irrespective of HRS reversal	Key study establishing ACLF grade as determinant of terlipressin response; EASL-CLIF criteria
Arora et al[97], 2020	ACLF (n = 120)	RCT	Terlipressin 2-12 mg/day CTI + albumin	Noradrenaline 0.5-3 mg/hour + albumin	HRS-AKI reversal day 7; 28-day survival	Terlipressin superior: HRS reversal 40% vs 17% (P = 0.004); survival 48% vs 20% (P = 0.001); less RRT 57% vs 80% (P = 0.006)	APASL ACLF criteria; open-label; landmark ACLF-specific RCT
Wong et al[98], 2021	Cirrhosis, HRS-1 (n = 300)	RCT	Terlipressin + albumin	Placebo + albumin	HRS reversal	Verified HRS reversal 32% vs 17% (P = 0.006); no significant survival benefit; respiratory failure more frequent with terlipressin	Largest placebo-controlled terlipressin trial; pivotal study underlying FDA approval
Wong et al[99], 2022	ACLF (n = 299)	Post-hoc	Terlipressin + albumin	Placebo + albumin	Respiratory failure by ACLF grade; 90-day survival	RF with terlipressin 30% in ACLF grade 3 vs 9.4% grade 1-2 (P = 0.0002); 90-day survival lower in grade 3 terlipressin arm. Use with caution in ACLF grade 3	Basis for FDA black box warning; terlipressin contraindicated in ACLF grade 3 with hypoxia
Jindal et al[100], 2024	ACLF (n = 60)	RCT	Early terlipressin (after 12 hours volume expansion)	Standard terlipressin (after 48 hours albumin challenge)	28-day mortality; AKI reversal	Early terlipressin: Lower mortality (17% vs 43%, P = 0.03); greater AKI reversal; regression of ACLF stage	First study on timing; supports early AKI intervention before 48 hours window in ACLF
Gupta et al[101], 2025	ACLF + septic shock (n = 70)	RCT	Terlipressin 2.6 mcg/kg/minute	Noradrenaline 0.1 mcg/kg/minute	MAP > 65 mmHg at 6 hours	Noradrenaline superior: MAP target achieved 74% vs 14% (P < 0.001); 3- and 7-day mortality higher with terlipressin	Terlipressin INFERIOR in septic shock context; noradrenaline first-line for ACLF septic shock

This table summarizes key studies on vasoactive therapy in acute-on-chronic and acute liver failure. Evidence is largely derived from cirrhosis and acute-on-chronic liver failure, where terlipressin improves hepatorenal syndrome reversal without consistent survival benefit, with reduced efficacy in advanced acute-on-chronic liver failure. While terlipressin may outperform norepinephrine for renal recovery in selected patients, norepinephrine remains first-line in septic shock. Data in acute liver failure are limited and suggest differing cerebral hemodynamic effects, supporting cautious use. HRS: Hepatorenal syndrome; RCT: Randomised controlled trial; CTI: Continuous terlipressin infusion; NS: Not significant; AEs: Adverse events; ACLF: Acute-on-chronic liver failure; EASL-CLIF: European Association for the Study of the Liver-Chronic Liver Failure; APASL: Asian Pacific Association for the Study of the Liver; FDA: Food and Drug Administration; RF: Respiratory failure; AKI: Acute kidney injury; MAP: Mean arterial pressure; CHRSR: Confirmed hepatorenal syndrome reversal; RRT: Renal replacement therapy.

Kidney failure

Kidney failure in ACLF: AKI occurs in up to 50% of hospitalized patients with cirrhosis and is the most common extrahepatic organ failure in ACLF (29%-75% by EASL-Chronic Liver Failure criteria)[22,33]. The differential diagnosis is broad and includes hypovolemia, acute tubular necrosis, prerenal azotemia, nephrotoxic exposure, and structural kidney disease, each with distinct therapeutic implications.

HRS-AKI: HRS-AKI is a functional renal failure driven by splanchnic vasodilation, reduced effective arterial blood volume, and compensatory renal vasoconstriction[24]. Diagnosis requires exclusion of shock, nephrotoxins, and structural kidney disease, and lack of response to diuretic withdrawal and albumin expansion[22]. The updated International Club of Ascites definition incorporates dynamic changes in serum creatinine, rather than absolute threshold values alone, enabling earlier recognition[22].

Treatment is based on albumin and vasoconstrictor therapy with the goal of reversing the hemodynamic disturbance. Terlipressin has the strongest evidence, improving HRS reversal rates compared with placebo, although survival benefit is less consistent, and adverse events, including ischemia and respiratory failure, remain relevant[37,38]. Norepinephrine shows comparable efficacy in smaller studies and is a practical alternative in the ICU setting[29]. Midodrine-octreotide has limited efficacy compared with intravenous vasoconstrictors and is not preferred when other options are available[38]. Treatment response varies by AKI phenotype, emphasizing the importance of accurate clinical classification[47].

Kidney failure in ALF: In ALF, renal dysfunction results from systemic vasodilation and reduced effective arterial blood volume despite increased cardiac output[20]. Activation of vasoconstrictor systems, excessive nitric oxide, and systemic inflammation further impair renal perfusion[20,23]. In addition, endotoxins, bile acids, and other tubular toxins released from the necrotic liver may contribute directly to structural kidney injury, producing a mixed functional and structural renal failure phenotype that is less amenable to vasoconstrictor-based treatment strategies than classic HRS-AKI in cirrhosis[20].

Management is primarily supportive, focusing on hemodynamic stabilization, avoidance of nephrotoxins, and treatment of precipitating factors. Continuous RRT (CRRT) is preferred in hemodynamically unstable patients, offering more precise control of fluids and solutes than intermittent hemodialysis and avoiding the hemodynamic instability associated with rapid fluid shifts that may exacerbate cerebral edema and intracranial hypertension[48]. An additional benefit is its capacity to reduce ammonia levels in ALF.

RRT and SLKT considerations: RRT should not be withheld solely due to liver disease severity, as renal recovery is possible after liver transplantation[49,50]. Liver transplantation is the definitive treatment for HRS-AKI, reversing the underlying hemodynamic disturbance[20]. Simultaneous liver-kidney transplantation is reserved for patients with chronic kidney disease or prolonged, non-recovering AKI unlikely to improve after liver transplantation alone[20,51].

Augmented renal clearance: Augmented renal clearance (ARC), defined as creatinine clearance > 130 mL/minute/1.73 m², is increasingly recognized in critically ill patients and may be present in patients with cirrhosis despite apparent renal dysfunction[52]. ARC reflects a hyperdynamic circulatory state with increased renal blood flow and enhanced drug elimination. Its presence may lead to subtherapeutic antibiotic exposure, particularly for renally cleared agents, with potential risk of treatment failure[52]. Recognition of ARC is therefore important, and dose adjustment guided by therapeutic drug monitoring should be considered when available[53].

Infection management and antimicrobial stewardship

Infections in ACLF: Bacterial infections are both a precipitant and a consequence of ACLF, occurring in 48% of cases globally (38% in Southern Europe, 75% in the Indian subcontinent)[22,54]. In prospective studies of patients developing ACLF, bacterial infection has been identified as the precipitating event in approximately 44% of cases[10]. Half of initially uninfected ACLF patients develop infections within 4 weeks[54].

The pathophysiology reflects cirrhosis-associated immune dysfunction syndrome, characterized by systemic inflammation and immunoparalysis affecting innate and adaptive immunity, gut barrier function, and microbial translocation[55,56]. Most patients with decompensated cirrhosis do not mount a febrile response, requiring a high index of suspicion[19]. Clinical and laboratory surrogates must therefore prompt infectious workup, including rising CRP, unexplained worsening of encephalopathy, new or deteriorating AKI, hemodynamic instability, and rising lactate, even in the absence of fever[55].

Multidrug-resistant organisms (MDRO) now account for over 30% of infections worldwide, with marked geographic variation: > 70% in India, < 20% in the United States, and approximately 35% in Europe[18,54,57,58]. Extensively drug-resistant bacteria range from 2% to 28%, with the highest rates reported in India[54,59]. Resistance patterns are heterogeneous and dynamic across regions and institutions[54].

Delay in antimicrobial therapy is strongly associated with mortality. Each hour of delay increases the risk of death by a factor of 1.86[55]. In spontaneous bacterial peritonitis, delay in treatment nearly triples in-hospital mortality, increasing absolute risk by 3.3% per hour[55]. Failure of empirical antibiotic therapy is an independent predictor of ACLF development[54].

Empiric therapy should be guided by the site of infection, healthcare exposure (community-acquired, healthcare-associated, or nosocomial), MDRO colonization status, and local resistance patterns[33]. Broad-spectrum antibiotics should be initiated immediately according to local epidemiology[22,60]. Early de-escalation within 24-72 hours is recommended based on microbiological data, clinical response, and susceptibility testing[22,60]. Altered antibiotic pharmacokinetics in liver failure, driven by ascites-related expansion of volume of distribution, hypoalbuminemia with reduced protein binding, impaired hepatic metabolism, and ARC, frequently result in subtherapeutic drug concentrations. Therapeutic drug monitoring should therefore be considered, particularly for β-lactams, vancomycin, and aminoglycosides in critically ill patients with cirrhosis[61].

Spontaneous bacterial peritonitis (ascitic fluid neutrophils ≥ 250/mm³) is the most common infection in ACLF[62]. Standard therapy includes appropriate antibiotics and intravenous albumin (1.5 g/kg on day 1 and 1 g/kg on day 3) to reduce renal impairment and mortality[60]. In nosocomial or healthcare-associated spontaneous bacterial peritonitis, where third-generation cephalosporin failure rates reach 40%-60%, carbapenem-based regimens, particularly meropenem combined with daptomycin, have demonstrated improved resolution and are associated with better 90-day survival[63].

Fungal infections occur in 10%-15% of ACLF patients, predominantly nosocomial, and in 3%-6% of patients with decompensated cirrhosis overall[60,64]. Invasive candidiasis accounts for 70%-90% of cases and invasive aspergillosis for 10%-20%, with case fatality rates of 45%-60% and higher, respectively. Coexisting bacterial and fungal infections further worsen outcomes. Invasive fungal infections are a contraindication to liver transplantation and a frequent cause of delisting[65]. Empirical antifungal therapy may be considered in ACLF patients with nosocomial septic shock and additional risk factors[66].

Several aspects of antimicrobial management in this setting lack a clear evidence-based standard. The factors that justify broad empiric coverage, namely nosocomial onset, recent antibiotic exposure, and a high model for end-stage liver disease score, are also the principal drivers of resistance selection, and the point at which therapy can be safely narrowed in this population has not been defined[34,57,58,60]. De-escalation is further complicated by the absence of a reliable means to distinguish colonization from invasive infection, particularly in fungi; an isolate of uncertain significance is therefore often treated empirically, yet autopsy data show that invasive disease is still frequently missed during life[60,64,66]. Antimicrobial dosing is a third unresolved issue, as therapeutic drug-monitoring targets extrapolated from non-cirrhotic patients perform poorly when expanded volume of distribution, hypoalbuminemia, and variable clearance are taken into account[21,61].

Infections in ALF: Bacterial infections complicate ALF through similar mechanisms of immune dysfunction, but with more rapid progression. Early recognition and prompt initiation of broad-spectrum antimicrobial therapy remain essential, together with attention to MDRO risk and source control. Fungal infections, particularly invasive candidiasis and aspergillosis, are associated with poor outcomes and may preclude emergency liver transplantation[43].

Hepatic encephalopathy and neuroprotection

Hepatic encephalopathy in ACLF: Severe hepatic encephalopathy (HE, Grade 3-4 per West Haven criteria) occurs in 20%-30% of patients with ACLF[22,33]. Management prioritizes identification and treatment of precipitants, including infection, gastrointestinal bleeding, constipation, electrolyte disturbances, AKI, and central nervous system-depressant medications.

Lactulose remains first-line therapy[22,33]. In Grade 3-4 HE, lactulose enema (300 mL lactulose in 700 mL water) may be used[33]. Polyethylene glycol is an effective alternative, particularly in critically ill patients at risk of ileus or abdominal distention[22,33,67]. Rifaximin is used for recurrent HE, although its role as add-on therapy in acute ACLF remains uncertain[33].

Sedation should be minimized. When required, short-acting agents such as propofol or dexmedetomidine are preferred[33]. Dexmedetomidine may reduce ventilation duration and benzodiazepine requirements in alcohol withdrawal. Opioids should be avoided or used cautiously.

Neurological complications in ALF: Cerebral edema and intracranial hypertension, once associated with mortality exceeding 80% in ALF during the 1980s, now account for approximately 20% mortality with modern ICU care[4,68]. CRRT reduces serum ammonia and is associated with improved outcomes, providing an adjunctive neuroprotective strategy[48]. Plasma exchange may be used as bridging therapy to reduce ammonia and other neurotoxic substances[5]. Intracranial pressure monitoring has declined due to bleeding risk and lack of clear outcome benefit. Management is focused on early recognition of neurological deterioration, avoidance of factors that increase intracranial pressure, and timely escalation to organ support and transplantation.

Coagulation management: Rebalanced hemostasis

Hemostasis in ACLF: The traditional view of cirrhosis as a pure bleeding diathesis has been replaced by the concept of “rebalanced hemostasis” - simultaneous reductions in liver-synthesized procoagulant factors (II, V, VII, IX, X, XI) and anticoagulant proteins (protein C, protein S, antithrombin) that produce a fragile equilibrium prone to both bleeding and thrombosis[69,70]. Standard coagulation tests [prothrombin time/international normalized ratio (INR)] reflect only procoagulant factor deficiencies and fail to capture compensatory anticoagulant loss, thereby overestimating bleeding risk[69,71].

Thrombocytopenia occurs in approximately 70% of cirrhosis patients (vs 6% in pre-cirrhosis), yet many maintain preserved thrombin generation despite low platelet counts[71]. Bleeding events in cirrhosis and ACLF (reported incidence 7.35%-67% across studies, reflecting heterogeneity in definitions and populations) predominantly arise from portal hypertension and mechanical causes rather than primary hemostatic failure[72]. Conversely, thrombotic complications occur in 4.7%-20% of ACLF patients, including portal vein thrombosis (which shows ~40% spontaneous recanalization and ~60%-70% recanalization with anticoagulation), venous thromboembolism, and arterial events[71,72].

Prophylactic correction of coagulopathy before procedures based solely on INR or platelet thresholds is no longer recommended, as these tests do not predict bleeding risk and large-volume transfusions may worsen portal hypertension[69,71]. Guidelines favor individualized assessment using global hemostatic assays and targeted, restrictive transfusion strategies[71].

Hemostasis in ALF: ALF produces marked INR prolongation (median INR ~2.7 in large cohorts) yet clinically significant bleeding occurs in only 7.4%-18% of patients, as thrombin generation and platelet function are often preserved despite laboratory abnormalities[71,72]. This rebalanced state is more fragile than in stable cirrhosis but does not equate to uniform hypocoagulability. Thrombotic events occur in 6%-21% of ALF patients[72]. Management principles mirror those in ACLF: Avoid routine INR-based correction, prioritize cause-specific therapy for portal hypertension-related bleeding, and use global assays to guide targeted hemostatic interventions when necessary[71].

Viscoelastic testing: Viscoelastic tests (VETs) (thromboelastography and rotational thromboelastometry) provide real-time, whole-blood assessment of clot initiation, kinetics, strength, and fibrinolysis, capturing interactions among platelets, fibrinogen, coagulation factors, and fibrinolytic pathways[70,73]. In cirrhosis and ACLF, VETs frequently demonstrate normal or near-normal global hemostasis despite prolonged INR and thrombocytopenia, better reflecting the rebalanced state[73,74].

VET-guided algorithms are best validated in liver transplantation, where adoption has coincided with dramatic reductions in transfusion requirements, from historical averages of ~29 red blood cell units, 37 fresh frozen plasma units, and 32 platelet units per transplant to contemporary medians of 0.5 units, 0.3 units, and 0.2 units, respectively[71]. In non-transplant settings, VET guidance reduces unnecessary prophylactic transfusions and enables targeted component therapy (fibrinogen concentrate, prothrombin complex concentrates)[70,73]. However, high-quality randomized data demonstrating reductions in mortality or bleeding in procedural settings remain limited, and ongoing trials (e.g., the RECIPE trial evaluating rotational thromboelastometry-guided prophylaxis for invasive procedures) aim to provide definitive evidence of outcomes[73].

Extracorporeal liver support

Extracorporeal liver support systems are used as bridging therapies to transplantation or recovery. Plasma exchange has the strongest evidence, improving short-term, but not overall, survival in ALF and supporting its use as standard care in specialized centers[75]. Albumin dialysis systems (molecular adsorbent recirculating system, Prometheus) improve HE and biochemical parameters but have not shown a consistent survival benefit. Their use is therefore limited to selected patients, primarily as a bridge to transplantation in specialized centers. Outside of plasma exchange, CRRT remains the only extracorporeal support routinely used in ICU practice for patients with liver failure[22,43]. Bioartificial liver devices have not demonstrated a survival benefit and remain investigational. Emerging approaches, including organoid-based and cell-derived therapies, are under evaluation but are not yet clinically applicable[76].

Liver transplantation: Redefining transplantability

The concept of “too sick to transplant” has undergone fundamental revision. Survival rates for transplanted ACLF patients now approach those of non-ACLF liver transplant recipients[12]. 1-year post-transplant survival now exceeds 80% even in the most severely ill patients[77,78].

Early assessment for liver transplantation should be proposed for all patients with severe ACLF (ACLF-2 or -3)[22]. Delaying transplantation for severe ACLF increases waitlist and post-transplant mortality[22]. In a multicenter study of 73 patients with ACLF-3, 1-year survival following liver transplantation was 77%[77]. Long-term follow-up nine years later demonstrated sustained survival benefit[78]. A systematic review demonstrated that liver transplantation improves prognosis across all grades of ACLF[79]. In the United States, 1-year survival increased from 81% in 2005-2008 to 87% in 2017-2020[80].

Recent changes in allocation policy have addressed this. The United Kingdom introduced an ACLF-3-specific tier in 2021: 81% underwent transplant at a median of 2 days, with 81% post-transplant survival[81]. In a cohort of 270 ACLF patients listed for transplantation, living donor liver transplantation enabled transplantation within a median of 3 days and achieved a transplant rate of 94.5%, compared with 53.8% for deceased donor transplantation, resulting in superior intention-to-treat survival at 3 months (92.9% vs 67.1%) while maintaining comparable 5-year post-transplant survival (~74%)[82].

A multidisciplinary panel of 35 international experts developed consensus criteria for when a critically ill cirrhotic patient is too sick to transplant[83]. These criteria emphasize that transplantability should not be determined solely by the severity of liver disease or the number of organ failures, but rather by a dynamic assessment of reversibility, illness trajectory, adequacy of infection control, frailty, and the intensity and duration of organ support[76].

These observations have fundamentally reshaped the concept of transplantability in ACLF. Contemporary evidence demonstrates that carefully selected patients with high-grade ACLF, including ACLF-3, can achieve favorable post-transplant outcomes despite multiple organ failures, mechanical ventilation, vasopressor support, or RRT, challenging historical assumptions that such features invariably represent prohibitive risk[12,77-80]. These advances have shifted the clinical question from whether such patients should undergo transplantation to how transplantability should be assessed and futility defined among those with more advanced disease[12].

Importantly, the expert consensus framework defining when a patient is “too sick to transplant” should not be considered synonymous with traditional contraindications to transplantation[83]. Rather, it addresses a different question: Not whether transplantation is possible, but whether it is likely to provide meaningful benefit. Unlike classical contraindications, which generally reflect irreversible conditions unlikely to be corrected by transplantation, such as devastating neurological injury, uncontrolled refractory shock, extensive bowel ischemia, or other irreversible systemic insults, the consensus framework focuses primarily on potentially reversible extrahepatic factors and the overall trajectory of critical illness[12,83,84]. Rather than applying binary exclusion criteria, it promotes dynamic assessment of organ failure reversibility, infection control, frailty, and the evolution of intensity of organ support requirements over time[22,33,83].

Nevertheless, important controversies remain. Although patients with ACLF face extremely high short-term mortality, current allocation systems were primarily designed to prioritize chronic liver disease severity and may not fully capture the rapid clinical deterioration characteristic of ACLF[81]. Consequently, debate persists regarding whether patients with severe ACLF should receive additional priority for transplantation based on their immediate risk of death and potential transplant benefit[12,81]. At the same time, expanding transplantation to increasingly critically ill recipients has generated debate regarding post-transplant utility, long-term functional recovery, and the ethical allocation of scarce donor organs when prognostic uncertainty remains considerable[12,51]. Another unresolved question is whether static severity scores should continue to guide listing decisions, or whether dynamic trajectories of organ failure and response to intensive care better identify patients most likely to benefit from transplantation[85-87]. Future efforts will likely focus on dynamic prognostic models, refinement of futility thresholds, and broader implementations of living donor transplantation strategies to improve timely access to transplantation while maintaining equitable organ allocation[12,82,85].

Liver transplantation remains the definitive treatment for ALF when spontaneous recovery is unlikely, with post-transplant survival now reaching approximately 90%[88], only slightly inferior to transplantation for chronic liver disease. European Liver Transplant Registry data (2007-2022) demonstrate 1-year survival of 78%, 5-year survival of 71%, and 10-year survival of 64% in 5818 ALF transplant recipients[89]. United States multicenter data (1998-2018) report even higher 1-year survival of 91% and 3-year survival of 90%[89], reflecting improvements in critical care management and earlier recognition.

Selection criteria balance the risk of premature transplantation in patients who may recover spontaneously against the risk of delaying transplantation until irreversible complications develop. In contrast to ACLF, where score-based thresholds increasingly inform futility decisions (Table 3), ALF remains less amenable to rigid criteria. Apart from irreversible brain injury, which represents the only absolute contraindication to liver transplantation, decisions in ALF rely on dynamic clinical assessment and multidisciplinary judgment, as reflected in the prognostic models summarized in Table 4. King’s College Criteria remain most widely used due to bedside applicability, demonstrating high specificity (98%-100%) but modest sensitivity (58%)[44]. Dynamic models incorporating serial measurements (ALFED, Bernal Decision Support Tool) and etiology-specific factors improve prediction[90].

Table 3 Prognostic models and futility assessment in acute-on-chronic liver failure.

Score	Components	Outcome predicted	Use/threshold	Type
Established prognostic and futility scores in the critical illness setting
CLIF-C ACLF[102,103]	CLIF-C OF + age + WBC	28- and 90-day mortality in ACLF patients	Score ≥ 64 → 28-day mortality ~80%; used to assess LT candidacy futility	Prognostic + futility
CLIF-C OF[10,102]	Liver (bilirubin), kidney (creatinine), brain (HE grade), coagulation (INR), circulation (MAP), lungs (PaO2/FiO2)	OF burden and short-term mortality	Defines ACLF grade (1-3) per EASL-CLIF criteria; grade 3 = ~75% 28-day mortality	Prognostic
MELD/MELD-Na[104,105]	Creatinine, bilirubin, INR (+ sodium for MELD-Na)	90-day mortality; waitlist priority	MELD > 35 in ACLF associated with futile outcomes without LT	Prognostic + futility
NASCELD-ACLF[15,106]	Number of extrahepatic OFs (kidney, brain, circulation, lungs)	30-day in-hospital mortality in cirrhotic inpatients	≥ 2 OFs = ACLF; 30-day mortality ~50%; ≥ 3 OFs approaches futility (~75%)	Prognostic + futility
APACHE II/III	Age, vitals, GCS, AaDO2, pH, electrolytes, creatinine, Hct, WBC	ICU mortality (general critical care score applied to ACLF)	APACHE II > 20 in liver patients correlates with high ICU mortality; used adjunctively	Prognostic
APACHE IV[107]	Acute physiology variables, age, chronic health status, admission diagnosis	ICU and hospital mortality in critically ill patients, including cirrhosis cohorts	Used as a general ICU severity score; in a cohort of 64 critically ill patients with cirrhosis, median APACHE IV scores were significantly higher among ICU non-survivors (117 vs 79.5) and demonstrated good discrimination for ICU mortality (AUC = 0.87)	Prognostic
SOFA[10,108]	PaO2/FiO2, platelets, bilirubin, MAP/vasopressors, GCS, creatinine	Daily organ dysfunction course in ICU	Rising SOFA score (≥ 2-point increase) denotes clinical deterioration; high scores used in futility discussions	Prognostic + futility
Child-Pugh[108]	Bilirubin, albumin, PT, ascites, encephalopathy	Cirrhosis severity and surgical risk	Child C (10-15) in ACLF context predicts poor short-term prognosis; limited discrimination vs MELD	Prognostic
AARC-ACLF	Bilirubin, lactate, creatinine, PT-INR, HE grade	28-day mortality (Asia-Pacific cohorts)	AARC score 5-6 = 28-day mortality ~47%; ≥ 9 = ~100% (futility threshold used in India/APASL populations)	Prognostic + futility
LT-specific futility and post-transplant mortality scores
TAM score[86]	4 pre-LT binary criteria (1 point each): Age ≥ 53 years, arterial lactate ≥ 4 mmol/L, mechanical ventilation with PaO2/FiO2 ≤ 200 mmHg, WBC ≤ 10 G/L	1-year post-LT mortality in ACLF grade 3 patients (multicenter European cohort, n = 152)	TAM > 2 → 1-year post-LT survival 83% (futility); TAM ≤ 2 → survival 83%. Assessed immediately before LT. Validated in independent multicenter cohort	Futility/LT
CLIF-C ACLF > 64[103]	CLIF-C OF score + age + WBC (same formula as CLIF-C ACLF; threshold specific to LT futility context)	Post-LT mortality and ICU futility in ACLF-3; validated at Royal Free Hospital (n = 202)	Score > 64 at 48 hours ICU → high 28-day mortality (~80%); used as a threshold for LT futility discussions alongside TAM; EASL CPG 2023 cites ≥ 70 for ICU withdrawal in non-LT candidates	Futility/LT
≥ 3 OFs[22]	ACLF grade 3 as defined by CLIF-C OF criteria: Hepatic, renal, brain, coagulation, circulatory, and/or respiratory failure (any ≥ 3 simultaneously)	28-day mortality; probability of benefit from LT	ACLF-3 → 28-day mortality ~75%-79% without LT; > 3 OFs (i.e., 4-6) associated with increasing futility risk; EASL CPG cites ≥ 4 OFs + CLIF-C ACLF > 70 for ICU withdrawal	Futility/LT
Futility consensus framework[83]	Consensus framework by 35 international experts (Delphi method): Severe frailty, persistent fever or < 72 hours appropriate antimicrobials for ongoing sepsis, uncontrolled septic shock, and other contraindications; not a single numeric score but a structured decision framework	Definition of futile LT and criteria to delay or deny LT in critically ill cirrhotic/ACLF patients	Severe frailty OR active uncontrolled sepsis (< 72 hours antibiotics) = defer LT. Framework uses 1-year and 5-year post-LT survival benchmarks to define futility (< 50% at 5 year); meant to complement numeric scores with clinical judgment	Futility/LT
CFS ≥ 7[83,87]	9-point ordinal scale assessing activity, energy, function, and dependence: 1 (very fit) → 9 (terminally ill). Scores ≥ 7 = severely frail/completely dependent	Post-LT mortality; identifies patients too frail to survive transplantation regardless of hepatic disease severity	CFS ≥ 7 → independent predictor of post-LT mortality; endorsed by Weiss 2021 consensus as a criterion to defer or deny LT. Stabilization before LT (Huebener et al[87], 2018) with organ recovery associates with better post-LT survival	Futility/LT
SALT-M score > 2023[85]	Age, BMI, diabetes (pre-LT comorbidity); respiratory failure (PaO2/FiO2 < 200 or mechanical ventilation); circulatory failure (vasopressors); infection history; RRT use; WBC at LT - derived in 521 United States + 120 French ACLF-2/3 patients (MODEL Consortium)	1-year post-LT mortality in ACLF grade 2 or 3; c-statistic 0.72 (derivation) and 0.80 (external validation)	SALT-M > 30 → substantially elevated 1-year post-LT mortality risk (futility threshold cited in subsequent literature); outperforms MELD-Na, CLIF-C ACLF and BAR score for post-LT mortality prediction. Also estimates post-LT ICU length of stay (ACLF-LT-LoS sub-score)	Futility/LT
HALT (Zhuang)	Recipient age, number of organ failures, lactate, DCD graft, cold ischemia time	1-year post-LT mortality in HBV ACLF	Estimated mortality 66%-86% in highest-risk graft-recipient combination; proposed to identify futile transplantation	Futility/LT

This table summarizes established prognostic models and futility criteria used in critically ill patients with acute and chronic liver disease, including both general intensive care unit scores and liver-specific tools. It highlights their key components, predicted outcomes, and clinically relevant thresholds to support risk stratification, transplant decision-making, and identification of patients unlikely to benefit from continued intensive care or liver transplantation. CLIF-C ACLF: Chronic Liver Failure Consortium acute-on-chronic liver failure score; CLIF-C OF: Chronic Liver Failure Consortium organ failure score; WBC: White blood cell count; ACLF: Acute-on-chronic liver failure; LT: Liver transplantation; HE: Hepatic encephalopathy; INR: International normalized ratio; MAP: Mean arterial pressure; OF: Organ failure; EASL-CLIF: European Association for the Study of the Liver-Chronic Liver Failure; MELD: Model for end-stage liver disease; MELD-Na: Model for end-stage liver disease including sodium; NASCELD: North American Consortium for the Study of End-Stage Liver Disease; APACHE: Acute Physiology and Chronic Health Evaluation; GCS: Glasgow Coma Scale; Hct: Hematocrit; ICU: Intensive care unit; AUC: Area under the curve; SOFA: Sequential Organ Failure Assessment; PaO₂/FiO₂: Ratio of arterial oxygen partial pressure to fraction of inspired oxygen; PT: Prothrombin time; AARC: Asian Pacific Association for the Study of the Liver Acute-on-Chronic Liver Failure Research Consortium; APASL: Asian Pacific Association for the Study of the Liver; TAM: Transplantation for acute-on-chronic liver failure-3 model; CPG: Clinical practice guidelines; EASL: European Association for the Study of the Liver; CFS: Clinical Frailty Scale; SALT-M: Survival After Liver Transplantation in acute-on-chronic liver failure model; BMI: Body mass index; BAR: Balance of Risk score; RRT: Renal replacement therapy; LoS: Length of stay; DCD: Donation after circulatory death; HBV: Hepatitis B virus.

Table 4 Prognostic models and liver transplantation listing criteria in acute liver failure.

Score	Components	Outcome predicted	Use/threshold	Type
Established prognostic and LT-listing criteria
KCC[109-111]	Acetaminophen: PH < 7.3, OR all 3 of (INR > 6.5, creatinine > 300 μmol/L, HE grade III-IV); lactate added in 2002 update. Non- acetaminophen: INR > 6.5 alone, OR any 3 of 5 factors (age < 10/> 40, non-A/B/drug etiology, jaundice-to-HE > 7 days, INR > 3.5, bilirubin > 300 μmol/L)	Death without ELT; need for emergency liver transplantation; PPV 70%-100%, specificity ~89%-95% in APAP-ALF	Meeting KCC → list for ELT; endorsed by AASLD and EASL. Sensitivity limited (~59%-69%); sequential use improves specificity. Lactate > 3.5 mmol/L (4 hours) or > 3.0 (12 hours post-resuscitation) incorporated as APAP-KCC criterion	LT listing
MELD/MELD-Na[104,112]	Creatinine, bilirubin, INR (+ sodium). A continuous score; rising values help guide urgency, particularly in ALF	Short-term mortality; ELT urgency. MELD correlates with severity but was not designed for ALF	MELD > 30 → high-urgency ELT listing; dynamic worsening more predictive than single value. Better for non-APAP-ALF (higher sensitivity) than KCC; both used complementarily	Prognostic
ALFSG-PI[113]	HE coma grade, INR, bilirubin, phosphate (≥ 3.7 mg/dL vs < 3.7 mg/dL), log10M30 (caspase-cleaved cytokeratin-18 apoptosis marker) - derived in 500 United States ALF patients; validated in independent 250-patient cohort	21-day transplant-free survival; need for LT or death at study entry. AUROC = 0.822 - superior to KCC (0.654) and MELD (0.704)	Higher index = worse prognosis; covers all ALF etiologies; requires M30 ELISA (limits routine use); 85.6% sensitivity/64.7% specificity for LT/death	Prognostic
CVC[114]	HE grade III-IV + factor V < 20% (age < 30 years) or < 30% (age ≥ 30 years); originally derived in fulminant hepatitis B patients	Survival without LT; primarily validated in viral (HBV) and non-paracetamol ALF in French cohorts	Criteria met → consider ELT listing; particularly used in France and Germany. Factor V assay not universally available. Sensitivity 69%-75%, specificity 50%-56% in mixed etiology (Ichai et al[114], 2015)	LT listing
APACHE II	Acute physiology (12 variables), age, chronic health - 0-71 points; validated in general ICU; applied to ALF as a general severity tool	ICU mortality in ALF patients; adjunctive severity stratification	APACHE II > 15 in ALF correlates with poor outcome; used alongside KCC; AUROC comparable to SOFA and ALFSG-PI in single-center studies	Prognostic
SOFA[115]	PaO2/FiO2, platelets, bilirubin, MAP/vasopressors, GCS, creatinine - 0-4 per organ; total 0-24	Multiorgan failure trajectory; daily ICU reassessment in ALF	Serial SOFA ≥ 15 or rapidly rising score used in futility discussions; SOFA outperforms KCC in multiorgan failure setting; AUROC ~0.84-0.85 in ALF studies (comparable to ALFSG-PI)	Prognostic
Bernal/United Kingdom ALF dynamic model[116]	Day 1: Age, GCS, arterial pH, lactate, creatinine, INR, circulatory failure; day 2: Change in lactate + change in INR - two time-point model specific to APAP-ALF	Death without LT in APAP-induced ALF; dynamic 2-day model improves over single time-point criteria. AUROC day 1: 0.82, day 2: Higher	Improves sensitivity over KCC at early admission; particularly useful in first 48 hours to identify non-survivors who do not yet meet standard KCC; used adjunctively in United Kingdom centers	LT-listing
Serum phosphate[117]	Serum phosphate at 48-96 hours post-paracetamol ingestion (threshold > 1.2 mmol/L); reflects failure of hepatic regeneration (phosphate uptake by regenerating hepatocytes)	Hepatic regeneration failure in APAP-ALF; high specificity for death without LT (used as add-on to KCC)	Phosphate > 1.2 mmol/L at 48-96 hours → failure to regenerate → consider ELT; incorporated as optional KCC criterion; specificity ~89%, PPV ~89% for death	LT-listing
Serum lactate[110]	Arterial lactate at admission and post-resuscitation: > 3.5 mmol/L at 4 hours, or > 3.0 mmol/L at 12 hours after adequate fluid resuscitation, in APAP-ALF	Early mortality indicator in APAP-ALF; tissue hypoxia and mitochondrial dysfunction marker; predicts poor outcome before HE develops	Lactate > 3.5 mmol/L at 4 hours (pre-resuscitation) = KCC criterion. Widely available and rapidly measurable; used for early risk stratification to initiate LT referral before KCC fully met	Prognostic
Emerging and investigational scores in ALF
miRNA-based prognostic model[118,119]	Regeneration-linked miRNA signature (early model) + cell-death miRNA signature (late model), combined with MELD score and vasopressor use - machine-learning derived, ALFSG cohort	21-day transplant-free survival in APAP and non-APAP ALF. Early model AUROC 0.78; late model AUROC 0.83 - both outperformed ALFSG-PI and KCC	Not yet in clinical use; requires standardised miRNA profiling. Represents next generation of biomarker-enriched dynamic ALF models; promising for identifying regeneration potential. Further prospective validation needed	Prognostic
Factor V-based composite scores[120]	Factor V level + bilirubin + vasopressor use + HE coma grade - derived within ALFSG cohorts as a dynamic biomarker model; factor V half-life 12-15 hours makes it sensitive to rapid change	Death or LT at 21 days in ALF; outperformed KCC and MELD but not ALFSG-PI in head-to-head comparison within ALFSG dataset	Factor V assay not universally available; currently investigational. Potential to complement KCC particularly in non-APAP ALF where factor V correlates better with prognosis than INR alone	Prognostic
AI/ML models[120]	Variable depending on model: Multi-parameter clinical + laboratory data (INR, bilirubin, creatinine, HE grade, vasopressors, lactate, ammonia, imaging); some incorporate omics data	Transplant-free survival and LT/death at 21-28 days; aim to be dynamic (daily recalculation) and etiology-agnostic	No AI/ML model currently validated for routine clinical ALF decision-making; several retrospective cohort studies promising (AUROC: 0.85-0.92). Major barriers: Explainability, prospective validation, regulatory approval. Active area of research	Prognostic
ALFED model[120]	Dynamic tracking of 4 variables over 3 days: HE grade > II, serum bilirubin, INR, arterial ammonia - assesses whether each remains above threshold or worsens. Derived in 380 non-APAP ALF patients (India)	Death without LT in non-paracetamol ALF (predominantly hepatitis E and hepatitis B etiology); designed for resource-limited settings	Promising for non-APAP ALF in Asia/developing world where hepatitis E predominates; AUROC superior to KCC in derivation cohort. Requires external validation in diverse global cohorts before routine use	Prognostic
Novel biomarker composites (CPS1, FABP1, Gc-globulin)[118,121,122]	CPS1 (hepatocyte-specific mitochondrial enzyme); FABP1 (hepatocyte damage); Gc-globulin (actin-free; reflects hepatic regeneration capacity) - all added to existing models (ALFSG-PI or KCC)	Death or LT in ALF; each has shown incremental improvement in AUROC when added to ALFSG-PI or KCC in ALFSG biobank studies. FABP1 > 350 ng/mL strongly associated with non-survival	All investigational; none in routine clinical use. FABP1 and CPS1 may be commercially assayable in future. Gc-globulin reflects regeneration rather than injury - concept of “regeneration potential” increasingly recognized as key missing element in current models	Prognostic

This table summarizes established and emerging prognostic models used in acute liver failure, including criteria for liver transplantation listing and short-term outcome prediction. It outlines key variables, predicted outcomes, and clinically relevant thresholds to support early risk stratification and time-sensitive transplant decision-making. LT: Liver transplantation; KCC: King’s College Criteria; OR: Odds ratio; INR: International normalized ratio; HE: Hepatic encephalopathy; ELT: Emergency liver transplantation; PPV: Positive predictive value; APAP: Acetaminophen (paracetamol); ALF: Acute liver failure; MELD: Model for end-stage liver disease; MELD-Na: Model for end-stage liver disease including sodium; ALFSG: Acute Liver Failure Study Group; PI: Prognostic index; AUROC: Area under the receiver operating characteristic curve; ELISA: Enzyme-linked immunosorbent assay; CVC: Clichy-Villejuif criteria; HBV: Hepatitis B virus; APACHE: Acute Physiology and Chronic Health Evaluation; ICU: Intensive care unit; SOFA: Sequential Organ Failure Assessment; PaO₂/FiO₂: Ratio of arterial oxygen partial pressure to fraction of inspired oxygen; MAP: Mean arterial pressure; GCS: Glasgow Coma Scale; miRNA: MicroRNA; AI: Artificial intelligence; ML: Machine learning; CPS1: Carbamoyl phosphate synthetase 1; FABP1: Fatty acid binding protein 1.

Among 624 United States patients listed for transplantation, 64% underwent transplantation, 20% recovered spontaneously, and 16% died waiting, the latter group demonstrating more severe multiorgan failure (vasopressors 65% vs 22%, mechanical ventilation 84% vs 57%, RRT 57% vs 30% compared with those transplanted)[91]. Waitlist mortality is higher for paracetamol-induced ALF despite faster transplantation, though post-transplant survival is comparable across etiologies. Absolute contraindications include irreversible brain injury (cerebral herniation, loss of middle cerebral artery flow, non-reactive pupils)[43]; relative contraindications include unstable vasoplegic shock, uncontrollable acute respiratory distress syndrome, severe hemorrhagic pancreatitis, and extensive bowel ischemia[84]. Negative prognostic factors post-transplant include age > 50 years, high vasopressor support, pH ≤ 7.26, grade IV encephalopathy > 48 hours, body mass index > 30 kg/m², serum creatinine > 2.0 mg/dL, and ≥ 3 organ failures[12]. Living donor liver transplantation achieves comparable 5-year survival (65%-80%) and is a major modality in Asia, whereas > 98% of ALF transplants in the United States and Europe use deceased donors[12,43].

Interdisciplinary models: The medical intensive liver unit

The complexity of managing critically ill patients with liver disease has led to two principal models of care: Dedicated medical intensive liver units and general ICUs with specialized hepatology consultation (Figure 2).

Figure 2 Models of intensive care delivery in patients with liver failure: Structure vs integration two organizational approaches commonly used to deliver care for critically ill patients with liver disease are illustrated: Dedicated liver-focused intensive care units and general intensive care units supported by specialist hepatology consultation. Although these models differ in structure, both aim to provide timely transplant evaluation, standardized management of organ failure, and coordinated multidisciplinary care. Current evidence suggests that outcomes depend less on the physical setting itself than on effective integration between hepatology, intensive care, and transplant services, supported by protocolized pathways and access to specialized expertise. ICU: Intensive care unit; MILU: Medical intensive liver unit; MARS: Molecular adsorbent recirculating system; PEX: Plasma exchange; CAID: Cirrhosis-associated immune dysfunction; LT: Liver transplantation.

Medical intensive liver units, predominantly found in North America, function as closed units where hepatology-trained intensivists provide continuous, liver-specific expertise. This model facilitates early recognition of liver-specific complications, rapid transplant evaluation, and implementation of standardized protocols tailored to cirrhosis and ALF, with close integration between hepatology, critical care, and transplant services. In contrast, general ICUs with hepatology consultation, more common in Europe, use established critical care infrastructure while incorporating liver expertise through consultative input, offering broader accessibility but potentially less continuity of disease-specific management.

Both models aim to address the distinctive pathophysiology of liver failure, including altered pharmacokinetics, rebalanced hemostasis, and HRS, which require specialized clinical knowledge[33]. Emerging evidence suggests that structured, liver-focused care pathways may improve outcomes, analogous to stroke and cardiac units, although the optimal balance between specialization and scalability remains uncertain. The key determinant of quality care may be less the physical location than the degree of integration, protocolization, and timely access to hepatology and transplant expertise.

CONCLUSION

Critical care hepatology has developed as a subspecialty in response to the recognition that liver failure is a multisystem condition requiring coordinated expertise across disciplines. Advances in intensive care, transplantation, and understanding of organ failure have shifted management from supportive care to targeted, physiology-based strategies. Optimal care now depends on close collaboration between hepatologists, intensivists, and transplant teams, with shared responsibility for decision-making, organ support, and assessment of transplant candidacy. This model extends beyond consultation and requires integrated, multidisciplinary management. Several challenges remain. These include improving prognostic accuracy, refining transplant allocation strategies, and balancing protocolized care with individualized clinical decision-making, with emerging priorities outlined in Figure 3. Future progress in critical care hepatology will likely depend not only on therapeutic innovation but also on clinical models that facilitate seamless integration of hepatology, intensive care, and transplant medicine, whether delivered through dedicated liver-focused units or multidisciplinary collaborative pathways[11,14]. Equally important will be the expansion of dedicated education and training opportunities in hepatic critical care, allowing clinicians to develop expertise in unique physiological, prognostic, and transplant-related challenges encountered in this population[14]. The development of collaborative research initiatives and prospective registries may help address important evidence gaps, facilitate evaluation of emerging management strategies, and support continuous improvement across diverse clinical settings[92,93]. At the same time, emerging therapies targeting systemic inflammation, organ support technologies, and regenerative approaches may further expand treatment options. As the burden of liver disease continues to rise, the principles of critical care hepatology will become increasingly relevant across both specialized and general intensive care settings. Continued advances in clinical systems, multidisciplinary care, education, research infrastructure, and therapeutic innovation will be essential to further improve outcomes in critically ill patients with liver disease.

Figure 3 Emerging priorities and research directions in critical care hepatology overview of key areas for future development, including precision medicine, regenerative therapies, immunomodulation, microbiome modulation, advanced organ support, transplant allocation strategies, and global health adaptation. These areas reflect evolving efforts to improve prognostication, therapeutic targeting, and care delivery. ACLF: Acute-on-chronic liver failure.