If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
a Contributed equally and share joint 1st authorship
Rajiv Jalan
Correspondence
Corresponding author. Address: Liver Failure Group, Institute for Liver and Digestive Health, UCL Medical School, Royal Free Hospital, Rowland Hill Street, London, NW3 2PF, United Kingdom.
a Contributed equally and share joint 1st authorship
Affiliations
European Foundation for Study of Chronic Liver Failure, EF-Clif, Barcelona, SpainLiver Failure Group, Institute for Liver and Digestive Health, UCL Medical School, Royal Free Hospital, London, United Kingdom
European Foundation for Study of Chronic Liver Failure, EF-Clif, Barcelona, SpainAPHP, Hôpital Beaujon, Service d’Hépatologie, Clichy, FranceInserm, Université de Paris, Centre de Recherche sur L’Inflammation, Paris, France
Traditionally, the complications of cirrhosis, namely variceal bleeding, ascites and hepatic encephalopathy, were thought to result predominantly from circulatory dysfunction and altered organ perfusion arising as a result of portal hypertension. Over the past 20 years, large, international prospective studies have indicated the importance of systemic inflammation and organ immunopathology as additional determinants of organ dysfunction in cirrhosis, which not only manifests in the liver, brain, circulation and the kidneys, but also the immune system, gut, muscles, adrenal glands, reproductive organs, heart and lungs. This review provides an overview of the traditional and emerging concepts around the initiation and maintenance of organ dysfunction in cirrhosis and proposes a new paradigm based upon a better understanding of acute decompensation of cirrhosis. The interaction between the traditional concepts and the emerging perspectives remains a matter of great interest and the basis for future research.
Cirrhosis represents the culmination of decades of liver injury and is thought to represent an irreversible disease. Traditionally, the occurrence of the first major complication, e.g. variceal bleeding, ascites or hepatic encephalopathy, is thought to change the natural history of cirrhosis, marking the transition from a ‘compensated’ to a ‘decompensated’ state
associated with a high risk of mortality over the subsequent 5 years. Following the clinical, prognostic and pathophysiologic characterisation of acute-on-chronic liver failure (ACLF), this classical view of the clinical course of cirrhosis needs to be revisited.
From a clinico-pathophysiological perspective, the complications of cirrhosis have been thought to be directly related to the severity of liver dysfunction and changes in portal haemodynamics that involve portal hypertension, portosystemic shunting, hepatic and extrahepatic organ perfusion.
Recently, the importance of systemic inflammation, particularly in patients who present with acute decompensation of cirrhosis has been highlighted and shown to be independently associated with a high-risk of short-term mortality.
Perhaps, the greatest change in our understanding of the clinical course of cirrhosis comes from evidence demonstrating the importance of extrahepatic organs on short-term outcomes in patients with acute decompensation.
Although there is some debate regarding the diagnostic criteria of ACLF, the EASL-CLIF Consortium criteria are the best studied and most widely validated and will be referred to for the most part in this review [reviewed in
]. These extrahepatic manifestations of cirrhosis have been shown to be associated with evidence of inflammation and cell death within extrahepatic organs, collectively referred to as organ immunopathology.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
There are 3 relevant clinical features that are important to highlight as introductory statements in the current review. The first is ‘acute decompensation’, which frequently signals the transition of cirrhotic patients from a compensated to decompensated state, and subsequently develops in a recurrent form during the entire clinical course. It is classically defined as the acute development of ascites, gastrointestinal haemorrhage or hepatic encephalopathy, or any combination of these, and may run widely different clinical courses. The second is ‘stable decompensated cirrhosis’, which defines patients with decompensated cirrhosis who, while receiving sustained prophylaxis with diuretics, and/or lactulose or rifaximin, and/or non-selective beta-blockers or repeated endoscopic treatment of esophagogastric varices, do not present episodes of AD for a long-time period. The concept of stable decompensated cirrhosis should be differentiated from ‘recompensated cirrhosis’, which is the clinical phase of the disease prior to the resolution of cirrhosis induced by successful treatment of the underlying aetiology.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
In this review, we aim to discuss our evolving understanding of cirrhosis and its complications and to provide an overview of traditional and emerging concepts around the initiation and maintenance of organ dysfunction, as well as proposing a potential new paradigm.
Insights into the clinical course of decompensated cirrhosis: The traditional view
Cirrhosis may result from any type of chronic insult to the liver through inflammation, parenchymal necrosis, fibro/angio-genesis, and progressive vascular changes. Once established, cirrhosis is initially characterised by the absence of symptoms and good/acceptable quality of life until the appearance of one or more of its clinical manifestations. At this point, the disease acquires a rapidly progressive course associated with deteriorating liver function, repeated hospital admissions and poor quality of life.
Since the very early descriptions of the natural history of cirrhosis, the disease has been termed compensated in the absence of symptoms and decompensated in their presence. In the last decades, long-term prospective studies have shown that liver-related mortality in cirrhosis only occurs at decompensation but can occur at first decompensation.
Therefore, the most important outcomes are decompensation in patients with compensated cirrhosis and death in patients with decompensated cirrhosis. Thus, compensated and decompensated cirrhosis, defined by the absence or presence/history of variceal bleeding, ascites, encephalopathy or jaundice, are usually perceived as distinct clinical states.
In compensated cirrhosis, 2 clinical states have been defined based on the presence or absence of gastro-oesophageal varices. The absence of varices defines state 1 while their presence defines state 2: 5-year death risk is 1.5% and 10%, respectively.
Patients in state 1 may be sub-classified as having mild portal hypertension (hepatic venous pressure gradient [HVPG] >5 mmHg and <10 mmHg) with minimal or no risk of clinical events and clinically significant portal hypertension (CSPH) with HVPG ≥10 mmHg,
Portal hypertensive bleeding in cirrhosis: risk stratification, diagnosis and management: 2016 practice guidance by the American Association for the Study of the Liver.
This sub-classification has clinical relevance since in mild portal hypertension, hyperdynamic circulation is not yet established and no response to non-selective beta-blockers (NSBBs) is detected,
Portal hypertensive bleeding in cirrhosis: risk stratification, diagnosis and management: 2016 practice guidance by the American Association for the Study of the Liver.
B-blockers to prevent decompensation of cirrhosis in patients with clinically significant portal hypertension (PREDESCI): a randomised, double-blind, placebo-controlled, multicentre trial.
Importantly, liver stiffness measurement ≥20–25 KPa alone or in combination with low platelet count and increased spleen size may non-invasively identify CSPH (specificity 0.90)
in compensated cirrhosis without varices, thus enabling the sub-classification of state 1 in clinical practice.
Given the associated risk of further clinical events and death, decompensation is a critical point in the clinical course of cirrhosis. In decompensated cirrhosis, clinical states with increasing risk of death have been defined by the type and number of decompensating events (Fig. 1). When decompensation presents with variceal bleeding alone, state 3, 5-year mortality is 20%; for patients decompensating with any single non-bleeding event (mostly ascites), state 4, 5-year mortality is 30%; after any second decompensating event or when decompensation occurs with any 2 or more decompensating events at once, 5-year mortality is 88%, state 5.
reported that the occurrence of infection or renal failure in decompensated cirrhosis was associated with 1-year mortality of 63%. It is now clear that bacterial infections play a relevant role throughout the course of cirrhosis by precipitating or aggravating decompensation and that any organ dysfunction beyond the liver is associated with a very high risk of imminent mortality,
Therefore, it is the recurrence of infections, the appearance of extrahepatic organ dysfunction, ACLF, refractory ascites, persistent encephalopathy or jaundice that may define this late decompensation state; although heterogeneous, this group of patients share a very high 1-year risk of death, ranging from 60% to 80%,
The multistate model describing the clinical outcomes of cirrhosis has been well validated.
Fig. 1Clinical trajectory of decompensated cirrhosis across different clinical states with increasing risk of death based on the traditional multistate model. Clinical states are defined according to the type of decompensation and increasing mortality. Decompensation may be precipitated by acute or non-acute events. ACLF may occur at any disease state. The relative incidence of acute and non-acute decompensation is not yet known. State 3 is defined by the occurrence of variceal bleeding alone, state 4 by any single non-bleeding event, state 5 by any 2 or more events and the late decompensate state by any event with organ failures either with or without ACLF. 5-year mortality across states from 3 to 5 is in the order of, respectively: 20%, 30%, 88%. With late decompensation mortality ranges between 60% and 80% at 1 year. The arrows at the bottom indicate the intensity of some major mechanisms of disease progression, respectively: hyper/hypodynamic circulation, Bacterial translocation and risk of infections, systemic inflammation, organ failures. ACLF, acute-on-chronic liver failure.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
showed that most patients (60.7%) presented ≥1 organ failures (OFs, severe impairment in organ function) or organ dysfunctions (moderate impairment of organ function), according to the CLIF-Consortium organ failure (CLIF-C OF) diagnostic criteria. The CANONIC data showed that among the 22% of patients hospitalised with organ failure, 64.9% had a single organ failure, 24.4% had 2 organ failures, and 10.6% had 3 organ failures or more.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Likewise, in patients with AD who develop ACLF after hospital admission, the most common OF was renal failure (56%), followed by liver, coagulation, cerebral, circulatory, and respiratory failures (44%, 28%, 24%, 17%, and 9%, respectively).
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
where the most common precipitating events are bacterial infection and active alcoholism, kidney failure is a common OF. However, in China, where the most common precipitating event is the reactivation of HBV infection, liver and coagulation failures are the most common OFs.
of portal hypertension led to its pharmacological treatment and, more recently, aetiologic treatments have shown that fibrosis may also be substantially reduced,
The outcome of patients recovering from (acute) decompensation is not yet clearly known and hence, there is no consensus on how to define recompensation as follow-up time is limited to 1-year at present. It would require a symptom-free time-interval since the (only) previous decompensation and the ability to maintain this state without treatment. Such a situation might be expected after aetiologic cure, which could result in a progressive reduction of fibrosis and portal hypertension, and even reversion of cirrhosis ;
however, this is unlikely with ongoing exposure to the aetiological agent. In fact, stable recompensation has been reported after effective antiviral treatment in patients with HBV-
Treatment with direct-acting antivirals improves the clinical outcome in patients with HCV-related decompensated cirrhosis: results from an Italian real-life cohort (Liver Network Activity-LINA cohort).
Therefore, accurate predictors of the risk of further decompensation are needed to define recompensation. The inflammatory pattern may be one such predictor, as recently suggested by the PREDICT study,
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
but long-term prospective validation is needed. For now, patients recovering from decompensation should still be included in the decompensated population and be considered to have lower life expectancy compared to patients who have never experienced any decompensation, unless they have successfully undergone aetiological treatment. However, even following successful aetiologic cure, the occurrence of oesophageal varices, decompensation and hepatocellular carcinoma has been reported. In patients with HCV-related cirrhosis, the risk of de novo/additional acute decompensation was still about 7% two years after the achievement of sustained virological response and was associated with baseline HVPG ≥16 mmHg and previous ascites.
Clinical outcome and hemodynamic changes following HCV eradication with oral antiviral therapy in patients with clinically significant portal hypertension.
Expanding consensus in portal hypertension. Report of the Baveno VI consensus workshop: stratifying risk and individualizing care for portal hypertension.
Recent studies describing outcomes in patients with acute decompensation provide new insights into the trajectory of cirrhosis.
Current understanding of the pathophysiologic basis of decompensated cirrhosis, AD and its limitations
Cirrhosis as a systemic disease
The current understanding of the pathophysiological basis of decompensated cirrhosis originated from investigations by Pavlov in Saint Petersburg and Starling in London at the end to the 19th century,
who described hyperammonaemia and portal hypertension associated with exaggerated production of hepatic lymph as the main mechanisms of encephalopathy and ascites, respectively. Three modern concepts developed over these theories are of major interest for this review. The first was the functional component of portal hypertension, which is related to under-expression of vasodilators in the hepatic microcirculation, leading to increased intrahepatic vascular resistance, and overexpression of vasodilators in the splanchnic circulation, leading to overflow of blood into the portal venous system and the hyperdynamic systemic circulation of cirrhosis.
The second was the identification of new mechanisms of hyperammonaemia, of ammonia entry into the brain, and on the deleterious effects of ammonia on neuronal function, which led to better understanding of brain dysfunction in cirrhosis.
Finally, the third important modern concept was the peripheral arterial vasodilation hypothesis, which reformulated the traditional pathophysiology of ascites and hepatorenal syndrome into a more complex sequence of events, proposing splanchnic arterial vasodilation and left ventricular dysfunction as the initial mechanism of effective arterial hypovolemia; the homeostatic activation of the renin-angiotensin-aldosterone system, sympathetic nervous system and antidiuretic hormone the intermediate processes; and renal fluid retention the final consequence.
These new concepts have heavily influenced the treatment of decompensated cirrhosis within recent decades.
Soon after this proposal, however, investigators began to understand the limitations of this pathophysiological paradigm of decompensated cirrhosis. Among the major arguments against this new view of the pathophysiology of AD and ACLF, the new concept of cirrhosis as a systemic disease stands out. A systemic disease is a condition that affects the whole body, and there were clear observations suggesting that this definition fits well for patients with decompensated cirrhosis. Typical examples of extrahepatic manifestations of cirrhosis are renal dysfunction,
Characteristics, risk factors, and mortality of cirrhotic patients hospitalized for hepatic encephalopathy with and without acute-on-chronic liver failure (ACLF).
Such widespread organ/system dysfunction fits badly into the paradigm derived from the peripheral arterial vasodilation hypothesis (with other independent mechanisms proposed for each organ dysfunction). Secondly, this paradigm did not offer a reasonable explanation for the extremely high incidence of bacterial infections at admission and during early follow-up (roughly 60%) in patients hospitalised with AD.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Such a high risk of bacterial infections was proposed to be related to severe impairment of the immune system, which was also not satisfactorily explained in the context of the other complications. Finally, the clinical observation that ascites, encephalopathy, gastrointestinal haemorrhage and infections frequently develop simultaneously in combinations of 2, 3 or even 4 complications in about 50% of patients hospitalised with AD
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
is more compatible with a common pathophysiological mechanism than with specific mechanisms for each complication.
Background of systemic inflammation in decompensated cirrhosis
Systemic inflammation is a condition in which there is inflammation throughout the body. It may be acute or chronic; mild, moderate or severe; cause or consequence of various pathological processes. It is characterised by activation of the innate immune system, increased circulating levels of inflammatory mediators and, in severe cases, proliferation of neutrophils, monocytes and dendritic cells. Systemic inflammation culminates in marked neuroendocrine and metabolic changes that aim to conserve metabolic energy and allocate more nutrients to the activated immune system. It is a major pathophysiological process in many clinical conditions including infections, obesity, atherosclerosis, metabolic syndrome, diabetes, non-alcoholic steatohepatitis, chronic pulmonary obstructive disease, depression and neurodegenerative diseases, osteoporosis, autoimmune diseases and cancer, and it is considered the most significant cause of death in the world today.
Systemic inflammation is a topic of growing interest in cirrhosis. The first studies were published in the 1970s and since then a large body of work has been published.
Tumor necrosis factor and interleukin-6 in spontaneous bacterial peritonitis in cirrhosis: relationship with the development of renal impairment and mortality.
Clinical hepatologists’ interest in systemic inflammation is logical considering the predisposition of patients with decompensated cirrhosis to bacterial translocation due to quantitative and qualitative changes in the microbiota, increased permeability of the mucosa and impaired function of the submucosal immune system. The intestinal flora is, therefore, a formidable source of systemic inflammation in these patients.
In the second part of the article we summarise current data suggesting that the clinical course of decompensated cirrhosis occurs in a context of severe chronic systemic inflammation associated with transient episodes of acute inflammatory bursts, during which patients may develop AD or ACLF.
Patients with acute decompensation of cirrhosis have different risks of short-term mortality depending upon whether they develop ACLF.
The systemic inflammation hypothesis: A new perspective on the clinical course of acute decompensation of cirrhosis
As indicated, the current paradigm of AD and organ failure in cirrhosis does not satisfactorily explain the complexity of AD, particularly the simultaneous development of ≥2 major complications (ascites, encephalopathy, gastrointestinal haemorrhage or infections) together with widespread impairment in the function of most extrahepatic organs, including the kidney, brain, lungs, heart, thyroid, adrenal glands, immune system, circulation and coagulation. For such a complex systemic syndrome, this paradigm only offers individual mechanisms specific for each complication or organ dysfunction. Systemic inflammation, in contrast, is a systemic syndrome that may lead to severe impairment of all these organs/systems. In addition, it may increase portal hypertension and liver failure, impair cardiocirculatory and renal function, increase the permeability of the blood brain barrier, and impair neuronal function, exerting a synergetic effect with hyperammonaemia.
This section summarises the clinical course of AD, stratified into 6 phenotypes with different prognoses, and a new pathophysiological paradigm of AD and ACLF proposed by the EASL-CLIF consortium, in which systemic inflammation plays a predominant role.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
(Box 1); it does not exclude the traditional specific mechanisms of ascites, encephalopathy or variceal haemorrhage (i.e. portal hypertension, circulatory and renal dysfunction and hyperammonaemia) but proposes that they act synergistically with systemic inflammation in the development of these complications and in the widespread impairment of extrahepatic organ function.
Patients with ‘no ACLF’ can be divided into those with a high risk of developing ACLF (pre-ACLF), portal hypertensive complications (unstable decompensated cirrhosis) and a group that are at low risk of developing complications (stable decompensated cirrhosis) in 3-months after acute decompensation.
The proposals of the systemic inflammation hypothesis.
1.
AD is a clinical entity with common pathophysiological background for all complications and organ failures.
2.
In the majority of patients, systemic inflammation is a major driver of progression from compensated to decompensated cirrhosis, the recurrence of AD during the clinical course of the disease, and the development of single or multiple organ failure.
3.
Once the first episode of AD develops, systemic inflammation follows a chronic course, with transient periods of aggravation due to proinflammatory precipitants or bursts of bacterial translocation resulting in repeated episodes of AD.
4.
The clinical course of AD largely depends on the evolution of systemic inflammation.
5.
AD-ACLF is the extreme expression of systemic inflammation.
6.
Systemic inflammation perturbs peripheral (non-immune) organ function and causes organ failures mainly through severe metabolic dysregulation leading to mitochondrial dysfunction and impaired energy production. Other mechanisms include direct tissue damage immunopathology or the synergistic effects of organ-specific mechanisms such as hyperammonaemia in encephalopathy.
Due to the complexity of AD and the ACLF syndrome, the large number of patients included in the PREDICT and CANONIC studies, and the need for a strategy to correlate the severity of systemic inflammation with patients’ clinical course and mortality, the first aim of both studies was to develop and propose new stratification criteria for patients with ACLF (AD-ACLF group) and with AD without ACLF (AD-No ACLF group) (Fig. 2). AD was diagnosed as the development of ascites, encephalopathy or gastrointestinal haemorrhage or any combination of these complications. The ACLF syndrome was defined as an episode of AD associated with single or multiple organ failure and high risk of short-term patient mortality (>15% in 28 days). Patients with AD-ACLF were stratified into 3 grades (ACLF 1, 2, and 3) according to the number of organ failures at admission. Six organ/system failures (liver, kidney, brain, coagulation, circulation, and respiration) were considered for this stratification, and the CLIF-C OF score was specifically designed to diagnose organ/system failures.
In the PREDICT study it was not possible to stratify patients with AD-No ACLF into subgroups with different prognosis at the time of hospital admission. Therefore, they were stratified according to clinical course during the 3-month follow-up period. Three subgroups of patients with different prognoses were identified: the pre-ACLF subgroup, including patients developing ACLF within 3 months after admission; the unstable decompensated cirrhosis (UDC) subgroup, including patients dying in hospital or requiring re-hospitalisation for reasons other than ACLF during the 3-month follow-up period; and the stable decompensated cirrhosis (SDC) subgroup, including patients who were discharged alive and did not require re-hospitalisations during the 3-month follow-up period. These 6 subgroups showed marked differences in cumulative 1-year mortality (Fig. 3A).Fig. 3B shows the cumulative development of ACLF and mortality during the 3-month follow-period in the pre-ACLF group.
Fig. 2Clinical trajectory of cirrhosis based on the new understanding of acute decompensation of cirrhosis. Patients with compensated cirrhosis have different risks of developing an episode of acute decompensation on the basis of whether they have clinical features of cirrhosis such as mild to moderate ascites, grade 1 hepatic encephalopathy or clinically significant portal hypertension. Following acute decompensation, the patients may either have ACLF or no ACLF. No-ACLF phenotype: patients with AD without ACLF which may follow 3 clinical courses: (A) stable decompensated cirrhosis sub-phenotype: patients not requiring further hospital readmission during a 90-day follow-up period; (B) unstable decompensated cirrhosis sub-phenotype: patients requiring ≥1 hospital readmissions unrelated with ACLF development during a 90-day follow-up period; (C) pre-ACLF sub-phenotype: patients with AD no-ACLF developing ACLF during a 90-day follow-up period. The second is the ACLF phenotype, which is defined by single renal failure or single non-renal organ (liver, brain, coagulation, circulation, respiration) failure if associated with renal dysfunction and/or brain dysfunction (chronic-liver failure organ failure score). Its severity and the patient’s risk of mortality is defined by the number of organ failures. ACLF, acute-on-chronic liver failure; AD, acute decompensation; CSPH, Clinically significant portal hypertension. ∗Stable Decomp Cirrhosis: 100% Survival over 3-months; ∗∗Unstable Decomp Cirrhosis: 70% Survival over 3-months; ∗∗∗pre ACLF: 50% Survival over 3-months
Fig. 3Outcome of acute decompensation and role of inflammation. (A) Cumulative incidence of mortality curves of the 6 subgroups of patients with AD. Patients with SDC, UDC and pre-ACLF were included in the PREDICT study. Patients with ACLF-1, ACLF-2 and ACLF-3 were included in the CANONIC study. (B) Cumulative percentage of patients with pre-ACLF developing ACLF or dying during the 3-month follow-up period per week. (C) Severity of systemic inflammation at admission, estimated by the plasma concentration of CRP, in patients with ACLF-1, ACLF-2 and ACLF-3 included in the CANONIC study (No-ACLF: patients without ACLF at admission). (D) Severity of systemic inflammation, estimated by the plasma concentration of IL-6 in a control group of patients with compensated cirrhosis and in patients with SDC, UDC and pre-ACLF included in PREDICT study. (E) Plasma concentration of CRP at hospital admission and at the last visit during the 3-month follow-up period in patients with SDC, UDC and pre-ACLF included in the PREDICT study. (F) Plasma concentration of IL-6 at hospital admission and at the last follow-up visit in patients with improving or worsening ACLF included in the CANONIC study. ACLF, acute-on-chronic liver failure; AD, acute decompensation; CRP, C-reactive protein; FU, follow-up; IL-6, interleukin-6; SDC, stable decompensated cirrhosis; UDC, unstable decompensated cirrhosis.
Thus, the CANONIC and PREDICT studies focus on a set of patients at higher risk of acute decompensation and therefore mortality. Additionally, Tonon et al. suggest the existence of a new entity referred to as ‘acute kidney disease’ that affects about 30% of patients with cirrhosis during routine clinical follow-up. About 78% of these patients develop this syndrome without any episode of acute kidney injury, with a 5-year survival rate of about 12%.
Tonon M, Rosi S, Gambino CG, Piano S, Calvino V, Romano A. et al. The natural history of Acute Kidney Disease (AKD) in patients with cirrhosis. J Hepatol. (in press).
Patients with compensated cirrhosis show no evidence of systemic inflammation, as estimated by the frequency of inflammatory cells in blood and the plasma concentration of markers of systemic inflammation and inflammatory cytokines. In contrast, systemic inflammation is severe in all subgroups of AD-No ACLF and increased further in patients with AD-ACLF
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
(Fig. 3C,D). Systemic inflammation is higher in patients with pre-ACLF than in patients with SDC and UDC. In patients with AD-ACLF, the severity of systemic inflammation increased in parallel with the number of OFs
(Fig. 3E). Therefore: i) the evolution of compensated to decompensated cirrhosis represents the transition from a status of unremarkable inflammation to one of severe systemic inflammation; ii) the severity of systemic inflammation at hospital admission correlates closely with severity of AD and short-term mortality during follow-up.
Precipitants
Bacterial infections and acute alcoholic hepatitis, the most important pro-inflammatory precipitants, are present either alone, in combination, or associated with other precipitants in approximately 95% patients with AD-No ACLF or AD-ACLF with identifiable precipitants at admission.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Sequential measurement of the plasma levels of interleukin (IL)-6 in patients with AD-No ACLF followed-up for 12 weeks detected frequent and intense peaks of systemic inflammation of variable duration (between days or weeks) in the absence of apparent clinical events in a significant number of patients, indicating that chronic systemic inflammation in patients with decompensated cirrhosis is not a steady process. AD in patients without identifiable precipitants likely results from transient periods of severe bacterial translocation.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Therefore: i) bacterial infections and acute alcoholic hepatitis are the major precipitants of AD-No ACLF and AD-ACLF, with other precipitants present in a small proportion of patients; ii) the number of precipitants correlates directly with the severity of systemic inflammation at admission and 90-day mortality, suggesting that precipitants have an additive effect on systemic inflammation; iii) most patients with AD-No ACLF (70%) but only one-third of patients with AD-ACLF present without identifiable precipitants, suggesting that AD likely develops in the context of acute bursts of bacterial translocation and systemic inflammation in a significant number of patients.
Clinical course
The distinct clinical courses of AD-No ACLF and AD-ACLF (Fig. 2) subgroups correlate with differences in the progression of systemic inflammation.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Patients with SDC follow an excellent clinical course with very low mortality in the context of a marked reduction in the grade of systemic inflammation (Fig. 3E). In contrast, patients with pre-ACLF develop the most severe clinical course among the AD-No ACLF patients in the setting a significant increase of systemic inflammation (Fig. 3E). The UDC subgroup shows relevant differences with the other AD-No ACLF subgroups. They present a complicated course and relatively high mortality despite moderate systemic inflammation at hospital admission and significant improvement during follow-up (Fig. 3E). On the other hand, they show surrogates of severe portal hypertension (gastrointestinal haemorrhage, treatment with transjugular portosystemic shunt, and hypovolemic shock as a cause of death) more frequently than the other 2 groups. Therefore, the clinical course of patients with UDC likely depends on the progression of portal hypertension and not of systemic inflammation. The clinical course of the AD-ACLF subgroups also correlates with the course of systemic inflammation (Fig. 3F). Thus, whereas resolution or improvement of ACLF occur in the setting of a significant decrease in systemic inflammation, worsening of ACLF develops in parallel with aggravation of inflammatory markers. Therefore: i) the clinical course of most patients with AD (SDC, pre-ACLF, ACLF-1, ACLF-2 and ACLF-3) depends on the evolution of systemic inflammation; ii) however, the clinical course of patients with UDC likely depends on a rapid progression of portal hypertension.
Systemic inflammation is the key pathophysiological mechanism underlying the risk of death in patients with acute decompensation of cirrhosis.
Immunosuppression is the likely mechanism behind bacterial infections in AD
Immunosuppression, a mechanism to limit vigorous systemic inflammation, was first proposed to account for unexplained aggravation of the primary infection or the development of secondary infections after the resolution of the initial infection in patients with severe sepsis.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Such an extremely high incidence of infections strongly indicates underlying immunosuppression. Indeed, among the soluble molecules contributing to immunosuppression, the anti-inflammatory IL-10 and quinolinate were markedly increased at hospital admission in patients with AD-No ACLF and even more so in those with AD-ACLF.
Inhibition of glutamine synthetase in monocytes from patients with acute-on-chronic liver failure resuscitates their antibacterial and inflammatory capacity.
These features are already present at the time of hospitalisation, which suggests that immunosuppression coincides with the onset of systemic inflammation and AD development in most patients and that both bacterial infections present at hospital admission and those developed thereafter are likely due to this immune dysfunction.
Systemic inflammation causes AD through acute metabolic dysregulation affecting energy production by the peripheral organs
As indicated, AD develops in the context of acute bursts of systemic inflammation associated with identifiable precipitants or likely secondary to acute episodes of bacterial translocation. AD can develop in patients with compensated cirrhosis and therefore without baseline systemic inflammation or in patients with decompensated cirrhosis and chronic systemic inflammation. It has been suggested that there is a critical threshold of systemic inflammation beyond which AD develops.
Patients with compensated cirrhosis would therefore require a more severe burst of systemic inflammation to develop AD than patients with decompensated cirrhosis. Indeed, patients with ACLF without a prior history of AD develop the syndrome in the context of a higher white cell count and C-reactive protein concentration and have a higher mortality rate at 28 days than patients with ACLF and a prior history of AD.
The metabolomic and lipidomic fingerprints of patients with ACLF, either infected or not infected, identified 3 characteristic metabolic dysregulations observed in patients with severe sepsis or other clinical conditions associated with systemic inflammation and multiorgan failure.
The first is an intense systemic catabolic reaction in response to the effect of pathogen- and damage-associated molecular patterns, cytokines and other inflammatory mediators on the sympathetic nervous system, the hypothalamic-pituitary-adrenal axis and glucagon secretion.
This leads to intense glycogenolysis, lipolysis and proteolysis and increased circulating levels of glucose, fatty acids and amino acids which are required to fuel the inflammatory reaction and the function of peripheral organs.
The second metabolic characteristic is the prioritisation of glucose metabolism to the immune system.
Systemic inflammation is an energetically expensive process due to the synthesis of a myriad of pro- and anti-inflammatory molecules and acute-phase proteins, activation and proliferation of immune cells and phagocytosis. Energetic metabolism (ATP synthesis) in activated immune cells largely depends on glucose metabolism via the cytosolic aerobic glycolysis pathway and not on mitochondrial oxidation of acetyl-CoA through the Krebs cycle and oxidative phosphorylation pathway (Fig. 4A). There are a couple of reasons for this change. The first is that energy production by aerobic glycolysis, although energetically inefficient, is extremely rapid, a feature of critical importance to confront an acute process requiring high energy consumption. The second is that metabolites derived from aerobic glycolysis are channelled through specific pathways that increase the nucleotide and RNA synthesis (i.e. the pentose phosphate pathway) required for the inflammatory reaction.
Fig. 4Evidence of severely deranged metabolism in patients with AD of cirrhosis. (A) Major abnormalities of cell metabolism in AD. In the cytosol there is activation of glycolysis, pentose phosphate and glucuronic pathways, and ATP and lactate synthesis. This occurs predominantly in the immune cells. In the mitochondria there is downregulation of carnitine-acylcarnitine translocase, and inhibition of β-oxidation, leading to fatty acid hypometabolism by the Krebs cycle and impaired oxidative phosphorylation and ATP synthesis. There is also increased mitochondrial oxidative stress and protein damage, which further impairs mitochondrial function. This occurs in the cells of peripheral organs. The net effect of the whole process is an increased production of energy by the immune system and hypometabolism and decrease energy production by peripheral organs. These changes explain the widespread dysfunction of peripheral organs in non-severe cases (AD-No ACLF clinical phenotype) and multiorgan failure (AD-ACLF clinical phenotype) in cases with extreme mitochondrial dysfunction. Upregulated mechanisms are represented in green and downregulated mechanisms are represented in red. (Modified from Arroyo V et al., N Engl J Med 220; 382:2137-2145). (B) Cleveland Plots. The right plot shows the whole set of annotated metabolites ranked according to their fold changes in patients with AD-ACLF vs. HI (the highest fold changes on the top, the lowest at the bottom). Fold changes of AD-No ACLF vs. HI and CC vs. HI are also shown. Left inset, zooming of the 50 top metabolites in the 3 comparisons. The dotted vertical line represents values in HI. Metabolites from carbohydrates, fatty acids and amino acids are identified by colours (Reproduced with permissions from Moreau R, et al. J Hepatol 2020; 72:688-701). ACLF, acute-on-chronic liver failure; AD, acute decompensation; CC, compensated cirrhosis; HI, healthy individual.
Energy production by peripheral (non-immune) organs must therefore rely on fatty acid and amino acid catabolism in the mitochondria. However, severe systemic inflammation adversely affects the mitochondrial metabolic pathways. Indeed, the entry of fatty acids into the mitochondria is severely impaired in AD and particularly in patients with AD-ACLF
Tumor necrosis factor and interleukin-6 in spontaneous bacterial peritonitis in cirrhosis: relationship with the development of renal impairment and mortality.
Finally, systemic inflammation leads to increased mitochondrial generation of nitric oxide, carbon monoxide and other reactive molecules that damage mitochondrial DNA and proteins of the electron transport chain, and cause generalised mitochondrial dysfunction.
This likely explains why cytosolic amino acid catabolism is markedly increased in AD-No ACLF, while mitochondrial amino acid catabolism is impaired, as reflected by the lack of increase in ketone body production by the liver.
Therefore, the expression of the metabolic dysregulation in AD is characterised by increased circulating levels of metabolites derived from glycolysis, increased concentrations of carnitines, reflecting impaired transport of cytosolic fatty acids into the mitochondria, and increased concentrations of metabolites derived from the cytosolic catabolism of amino acids
Tumor necrosis factor and interleukin-6 in spontaneous bacterial peritonitis in cirrhosis: relationship with the development of renal impairment and mortality.
(Fig. 4B). Interestingly, whereas the metabolome fingerprint of patients with compensated cirrhosis is almost normal, it is significantly altered in patients with AD-No ACLF, and severely altered in patients with AD-ACLF (Fig. 4B).
In summary, while glycolysis and energy production by immune cells are markedly activated in patients with AD, severe metabolic dysregulation with hypometabolism in peripheral organs develops as a consequence of systemic inflammation. Since the homeostasis of cell function is energy-dependent, the systemic inflammation hypothesis proposes that metabolic dysregulation is the cause of impairment in the function of peripheral organs in AD-No ACLF and of single or multiple organ failure in AD-ACLF.
Metabolic dysregulation may operate by itself or in combination with organ specific mechanism (e.g. hyperammonaemia for brain failure or effective arterial hypovolemia for ascites and renal failure). However, it is the predominant mechanism underlying the widespread impairment of organ function in AD.
At present, how patients with acute decompensation map onto the multistate prognostic models of cirrhosis is not known.
Conclusions and future perspectives
The traditional multistate models of cirrhosis have been validated in several studies and are currently widely used in clinical practice but mainly focus of the natural history of patients that are relatively stable. Recent studies have focused on improving our understanding of the outcomes of patients with acute decompensation, adding to the traditional models. Understanding the revised trajectory of cirrhosis will help to stratify patients into clinical and pathophysiological groups. The associated scoring systems, the CLIF-C OF score to diagnose ACLF,
and the CLIF-C ACLF score to provide prognostic information on patients with ACLF, are likely to help stratify patients for admission to intensive care, liver transplantation, and very importantly clinical trials, while also helping to define futility of ongoing intensive care.
For instance, patients with AD no-ACLF could have 1 of 3 distinct clinical courses with widely differing mortalities.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
The recognition that these clinical courses are associated with relatively distinct pathophysiological states could help guide drug development for specific patient populations. Taking into account the data presented here, it is possible that the failure to find an efficacy signal in large trials of steroids for alcoholic hepatitis (STOPAH study),
was because of a lack of stratification, i.e. the inclusion of patients with widely varying mortality rates. Several clinical trials of novel therapeutics based upon the new understanding of the trajectory of cirrhosis are underway, such as APACHE and DIALIVE, which target patients with ACLF grades 2-3; PRECIOSA, which targets AD no-ACLF patients with UDC; and a trial testing TAK-242, a toll-like 4 receptor antagonist, in patients with pre-ACLF and ACLF 1-2. Additionally, recent studies have indicated that patients with ACLF grade 3 can achieve excellent survival rates after transplantation and are disadvantaged by the current organ allocation system.
This has led to a re-think about how organs should be allocated leading to a pilot programme being introduced in the UK where patients with severe ACLF will be prioritised. We believe that this new understanding will help to reduce mortality in patients with cirrhosis.
Although one of the aims of this paper was to try and build links between the traditional hypothesis and the new paradigm described here, we were not able to do so. Future studies should address this aim.
The authors received no financial support to produce this manuscript.
Authors’ contributions
RJ, GD PA, VA: study concept and design, analysis and interpretation of data, drafting of the manuscript. RM, JT: Critical revision of the manuscript.
Conflict of interest
Dr. Jalan has research collaborations with Yaqrit and Takeda. Dr. Jalan is the inventor of OPA, which has been patented by UCL and licensed to Mallinckrodt Pharma. He is also the founder of Yaqrit limited, a spin out company from University College London. The remaining authors disclose no conflicts.
Please refer to the accompanying ICMJE disclosure forms for further details.
Supplementary data
The following is/are the supplementary data related to this article:
This article is published as part of a supplement entitled New Concepts and Perspectives in Decompensated Cirrhosis. Publication of the supplement was supported financially by CSL Behring. The sponsor had no involvement in content development, the decision to submit the manuscript or in the acceptance of the manuscript for publication.
Trebicka J, Fernández J, Papp M, Caraceni P, Laleman W, Gambino C, et al. The Predict study uncovers three clinical courses in acutely decompensated cirrhosis with distinct pathophysiology. J Hepatol 2020;73:842-854.
Portal hypertensive bleeding in cirrhosis: risk stratification, diagnosis and management: 2016 practice guidance by the American Association for the Study of the Liver.
B-blockers to prevent decompensation of cirrhosis in patients with clinically significant portal hypertension (PREDESCI): a randomised, double-blind, placebo-controlled, multicentre trial.
Treatment with direct-acting antivirals improves the clinical outcome in patients with HCV-related decompensated cirrhosis: results from an Italian real-life cohort (Liver Network Activity-LINA cohort).
Clinical outcome and hemodynamic changes following HCV eradication with oral antiviral therapy in patients with clinically significant portal hypertension.
Expanding consensus in portal hypertension. Report of the Baveno VI consensus workshop: stratifying risk and individualizing care for portal hypertension.
Characteristics, risk factors, and mortality of cirrhotic patients hospitalized for hepatic encephalopathy with and without acute-on-chronic liver failure (ACLF).
Tumor necrosis factor and interleukin-6 in spontaneous bacterial peritonitis in cirrhosis: relationship with the development of renal impairment and mortality.
Tonon M, Rosi S, Gambino CG, Piano S, Calvino V, Romano A. et al. The natural history of Acute Kidney Disease (AKD) in patients with cirrhosis. J Hepatol. (in press).
Inhibition of glutamine synthetase in monocytes from patients with acute-on-chronic liver failure resuscitates their antibacterial and inflammatory capacity.
00029-5&id=gr1.jpg){kind=link}
00029-5&id=gr2.jpg){kind=link}
00029-5&id=gr3.jpg){kind=link}
00029-5&id=gr4.jpg){kind=link}