Acute liver failure
How can I be sure that the patient has acute liver failure?
- A tabular or chart listing of features and signs and symptoms
- How can I confirm the diagnosis?
- What other diseases, conditions, or complications should I look for in patients with acute liver failure?
- What is the right therapy for the patient with acute liver failure?
- What is the most effective initial therapy?
- Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.
- A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies
- Listing of these, including any guidelines for monitoring side effects.
- How should I monitor the patient with acute liver failure?
- What's the evidence?
How can I be sure that the patient has acute liver failure?
In patients with acute liver disease, the combination of encephalopathy and laboratory evidence of a coagulopathy is the basis for the diagnosis of acute liver failure. The diagnosis must be made within 26 weeks of the first symptoms of the liver disease and there should be no preexisting diagnosis of liver disease. Viral hepatitis (A, B, and E), acetaminophen, drug reactions, and no identifiable cause (seronegative) represent the four most common etiological subgroups of acute liver failure.
The encephalopathy of acute liver failure can range from fairly subtle impairment of mental capacity to an unresponsive state. This spectrum is graded on a scale of 1 to 4 and, simplistically, grade 1 is drowsiness, grade 2 is confusion, grade 3 is responding only to major stimuli (often very agitated), and grade 4 is unresponsiveness. The fetor and tremor that are typical features of encephalopathy in patients with chronic liver disease are not characteristic of this condition. Most patients with acute liver failure will be jaundiced at the time of presentation, the possible exceptions being very rapidly evolving liver injury (e.g., after an acetaminophen overdose).
The liver is normal in size or mildly enlarged at the time of presentation. Subsequently, the liver may reduce in size. An enlarged liver is very atypical and should raise the possibility of Budd-Chiari syndrome or malignant infiltration. Ascites and other manifestations of portal hypertension are not characteristic findings in acute liver failure, with the exception of the acute presentation of Wilson's disease, and in some patients with a more protracted clinical course who are often referred to as having subacute liver failure.
A tabular or chart listing of features and signs and symptoms
Jaundice: Usually present
Small liver: Particularly subacute
Normal size liver: Most cases at outset
Large liver: Budd-Chiari syndrome, fatty liver of pregnancy
Ascites: Some patients with subacute liver failure, Wilson's disease
Less common features
Pyrexia: Hepatitis A
Hyperpyrexia (>40 degrees): Heat shock, some cases of ecstasy-induced liver injury
Chronic liver disease and cirrhosis. There is considerable overlap between the clinical features of acute liver failure and liver failure resulting from cirrhosis. However, it is unusual for the initial presentation of cirrhosis to be as severe as acute liver failure, with the exception of Wilson's disease. Patients with significant portal hypertension at the time of presentation should be investigated for possible underlying chronic liver disease.
Acute alcoholic hepatitis. This may be confused for acute liver failure if the history of alcohol consumption is suppressed. It should always be considered in patients with hepatomegaly and when the gamma-glutamyl transferase is more than 5 times the upper limit of normal. A markedly elevated white cell count that is disproportionate to an apparent sepsis profile may be another clue to this diagnosis.
Malignant infiltration. This is a trap for the unwary. Again, hepatomegaly is the clue that additional investigation is required. Lymphoma is the most common causative malignancy. Breast carcinoma is another cancer that can cause malignant infiltration.
How can I confirm the diagnosis?
The initial investigations are directed at two objectives: 1) assessing the severity of the disease and 2) establishing the etiology.
Assessing severity of disease
The basic tests assessing severity include blood count, measures of coagulation (prothrombin time; international normalized ratio [INR]; or specific coagulation factors, e.g., factor V), serum bilirubin, liver enzymes, serum creatinine, and arterial blood gases.
The single most important laboratory test is the coagulation parameter and this should be repeated on a daily basis. The pattern of coagulation test results gives valuable prognostic information. Some management protocols mandate the use of prophylactic fresh frozen plasma, which reduces the value of serial measurements of coagulation in this regard. The coagulation results must be interpreted in the clinical context because of a very wide range seen in acute liver failure. Extreme derangements may be seen after an acetaminophen overdose, whereas in subacute liver failure, the abnormalities observed may seem relatively trivial.
The next most important information comes from the pattern of liver function tests. The serum transaminases reflect both the acuity as well as the extent of liver injury. The highest transaminase levels (>5,000IU/l) are seen after acetaminophen overdose, in ischemic hepatitis, and with toxemia of pregnancy. Intermediate levels (1000-5000 IU/l) are seen at early stages of viral or autoimmune hepatitis. In many cases, the peak in serum transaminases is not documented and the levels are considerably lower at the time of presentation. The serum bilirubin mainly correlates with the duration of the disease in its upward trajectory and with increasing likelihood of recovery when in a downward trajectory.
After acetaminophen overdose, serum creatinine and arterial pH are very important in assessing prognosis. Serum creatinine is also important in all other etiologies.
Assessment of the severity of the disease is generally done by using one of a number of composite models that rely heavily on combinations of the laboratory tests outlined earlier (above). Some models also include clinical features, such as severity of encephalopathy or presence of cerebral edema. Additional information may become available from liver biopsy or imaging. When the liver injury is homogeneous, the extent of hepatocyte necrosis reflects the severity of the injury; however, in many patients (particularly those in whom the severity of the illness is most difficult to assess), the pathological pattern is heterogenous and difficult to evaluate with a standard biopsy. Both absolute and serial estimates of liver volume by imaging are valuable in assessing severity of disease, particularly in subacute liver failure.
Serology for hepatitis A, B, and E and autoantibodies with serum immunoglobulins should be performed in all patients. Acetaminophen levels should be measured in all patients with a history of drug use and in all cases with a clinical history of being unwell for up to 4 days. Specific investigations pertinent to a given clinical situation should also be performed at this stage.
Confirmatory tests are often not required when a clear-cut diagnosis is made on initial investigation. Hepatitis B is the significant exception because a number of different patterns of serology are seen with de novo infection, seroconversion-related injury, or increased viral replication. IgM anticore antibody establishes a diagnosis of acute infection and, in a minority of these patients, HBsAg will be negative at the time of presentation. HBV DNA levels are very high in cases related to enhanced viral replication. Seroconversion-related illness is characterized by absence of IgM anticore antibody and low HBV DNA levels. Acute infection with hepatitis D is the explanation for acute liver failure in some patients with hepatitis B.
In cases of doubt, additional information may be forthcoming from liver histology or radiological investigation. In the most common causes of acute liver failure, the findings on liver biopsy are nonspecific or confirm information obtained from alternative sources. Diagnoses of malignant infiltration or acute alcoholic hepatitis may be established in patients with clinical features of acute liver failure who have enlarged livers. Similarly, radiological imaging might confirm what is expected (e.g., pregnancy-related) or be diagnostic (e.g., Budd-Chiari syndrome
If there is a history of acetaminophen overdose, measure blood levels. Otherwise, check viral serology, serum immunoglobulins and autoantibodies.
When there is no evidence of viral or autoimmune hepatitis review drug exposure (including herbal remedies) over the preceding 6 months and assess likelihood of culpability.
Determine if the clinical background gives clues to the cause (e.g., Is the patient in the third trimester of pregnancy?).
Consider Wilson's disease in adolescents and young adults with hemolytic anemia or severe jaundice or ascites and screen for Kayser-Fleischer rings.
If the liver is enlarged, proceed to imaging and possible liver biopsy to investigate the possibility of Budd-Chiari syndrome or malignant infiltration (particularly, lymphoma and breast carcinoma).
When all are negative, make a diagnosis of seronegative hepatitis or acute liver failure of indeterminate etiology (30%-40% of patients will be in this category).
Investigation of etiology of acute liver failure
Hepatitis A (HAV): IgM anti-HAV 95% positive initially, 100% on repeat testing
+ D(HBV, HDV): Full profile IgM anti-core positive – acute infection
Elevated HBV DNA: Reactivation that may be spontaneous or as a consequence of immunosuppression or chemotherapy
Hepatitis E (HEV): Anti-HEV IgM antibody – diagnostic
Acetaminophen Drug levels in blood may be negative on third or subsequent days after overdose.
Idiosyncratic drug reactions: Eosinophil count (minority); most diagnoses based on temporal relationship
Autoimmune hepatitis: Autoantibodies, high titers of liver specific, elevated IgGs
Immunoglobulins (IgGs)L: Low titers occur in other etiologies
Pregnancy-related including: Fatty liver uric acid: ultrasound, usually in first pregnancy; HELLP syndrome: platelet count; toxemia transaminases: relevant obstetric background
Wilson's disease, ceruloplasmin: Urinary copper may be elevated in other etiologies.
Ischemic hepatitis: Transaminases
Heat stroke: Transaminases, rhabdomyolysis may be a prominent feature.
What other diseases, conditions, or complications should I look for in patients with acute liver failure?
Risk factors for acute liver failure
In the majority of patients, there are no identifiable risk factors for developing acute liver failure other than any that might have triggered the liver injury in the first instance (e.g., taking acetaminophen, acquiring hepatitis B). However, within specific etiologies of acute liver failure, there is some evidence of differential risk. Acute liver failure is more likely to develop after a paracetamol overdose in malnourished patients and in those with liver enzyme induction secondary to chronic alcohol consumption or antiepileptic drugs. Patients with hepatitis B or drug-induced liver injury were found to be more likely to develop liver failure if they used acetaminophen. Chemotherapy and immunosuppression increase the risk of liver failure in chronic hepatitis B carriers. The risk of developing liver failure with hepatitis A increases after 40 years of age. There is also a wide spectrum of risk of developing drug-induced liver failure.
Complications of liver failure
Acute liver failure can be associated with complications affecting almost every system in the body. This is reflected in the fact that the most common modality of death in this condition is multi-organ failure.
Cerebral edema is the most dramatic specific complication of acute liver failure. Young patients and those with rapidly progressive disease are most at risk. The clinical syndrome of cerebral edema is seen only in patients with advanced encephalopathy. The incidence of cerebral edema has fallen to around 20% to 30% of patients at risk for reasons that are unclear.
Renal failure follows two patterns. Early renal failure is seen particularly in acetaminophen-induced disease and mushroom poisoning. In these settings, it appears to reflect primary renal injury. In other etiologies, renal failure tends to develop later.
The systemic inflammatory response syndrome (SIRS) or culture-proven sepsis occurs very regularly. The diagnosis of SIRS is based on three parameters – temperature, white cell count, and pulse rate – that overlap with changes seen with infection. The incidence of bacterial infection can be manipulated by antibiotic strategies but still occurs in around 30% of cases. Invasive fungal infection occurs in about 10% of cases and carries a very poor prognosis.
There is a poor correlation between the risk of bleeding and the severity of derangements of laboratory parameters of coagulation. This might be due to the fact that the balance between coagulation and thrombolytic mechanisms is maintained despite the marked reduction in synthesis of related products by the failing liver. The highest risk of bleeding is seen in patients with thrombocytopenia or disseminated intravascular coagulation. Common sites of bleeding include puncture sites and the gastrointestinal tract.
Circulatory failure or shock is seen at an early stage in acetaminophen-induced disease and with mushroom poisoning. This may respond to resuscitation with fluids. Circulatory failure triggered by sepsis may also respond to treatment with fluids and antibiotics. However, many patients develop circulatory failure with decreasing inotrope responsiveness as part of disease progression to multi-organ failure and death.
Respiratory failure follows a similar pattern to circulatory failure. Some cases are secondary to a specific complication (e.g., infection or fluid overload) and may respond to treatment directed at the underlying cause. In other cases, it is a component of progressive multi-organ failure.
Cerebral edema: Incidence declining to 20-30% of cases with grade 4 encephalopathy
Renal failure: 75% of acetaminophen and 30% of other etiologies
Sepsis and systemic: Overall 80%, including bacterial and fungal sepsis
Inflammatory response syndrome
Bleeding: Clinically significant hemorrhage,10-20%
Circulatory failure: Common at some point in illness and refractory in 20-30%
Respiratory failure: Contributes to multi-organ failure, which is now the commonest cause of death and is closely linked to circulatory failure.
Pancreatitis: 5% across all etiologies
Bone marrow failure: Aplastic anemia in 1-2% of seronegative hepatitis and hemophagocytic syndrome in 5%
What is the right therapy for the patient with acute liver failure?
There are four components to therapy.
1. Commence specific therapies for the underlying cause when these exist. Some of these interventions are time critical and need to be started at the first point of contact (e.g., N-acetylcysteine in acetaminophen overdose).
2. Institute supportive medical care ranging from aggressive resuscitation on presentation to intensive care support as the disease progresses. The resuscitation phase will happen at the first point of contact with medical services but after stabilization, most patients should be referred to specialist centers that also offer the opportunity for liver transplantation.,
3. Identify patients who might benefit from liver transplantation.
4. Consider liver support devices or strategies.
N-acetylcysteine: All etiologies
Intensive-care-based monitoring and treatment
Liver transplantation: Whole graft or auxiliary
Biological liver support devices (e.g., porcine hepatocyte or human tumor cell line)
Nonbiological liver support devices (e.g., albumin dialysis systems, plasmapheresis)
What is the most effective initial therapy?
Patients with hypotension or hypovolemia require immediate fluid resuscitation. This is commonly seen in acetaminophen-induced liver failure and may be associated with a severe metabolic acidosis. The degree of response to resuscitative measures is one of the key prognostic indicators in this setting.
Hypoglycemia occurs frequently in acute liver failure and may be a relatively early event. The possibility of hypoglycemia should be considered in all patients with impaired conscious levels.
Airway protection should be instituted in encephalopathic patients prior to transfer to specialist centers.
N-acetylcysteine should be given to all patients presenting after an overdose of acetaminophen. During the first 16 to 24 hours after ingestion of acetaminophen, it acts to replete glutathione, which is essential to minimize the formation of the hepatotoxic metabolite of acetaminophen. However, later administration also improves outcome despite the development of liver injury. N-acetylcysteine has also been shown to be of benefit in other etiologies of acute liver failure.
Antiviral agents active against hepatitis B may be beneficial, particularly with faster response profiles (e.g., lamivudine or tenofovir). These are used intuitively in patients known to have elevated HBV DNA levels but good data on efficacy are not available. A beneficial role in new infections is less likely on theoretical grounds.
Acyclovir is recommended in any patient suspected as having this unusual cause of acute liver failure. The presence of mucocutaneous herpetic lesions should trigger consideration of this treatment.
Although corticosteroids are the first-line treatment for autoimmune hepatitis, they are often ineffective once acute liver failure has developed. Furthermore, potential side effects (e.g., infection) may prejudice successful liver transplantation. The risk/benefit profile needs to be assessed in individual patients. Patients with fairly early disease, or those for whom liver transplantation may not be an option, may benefit from corticosteroids.
Penicillamine can rescue patients with acute presentations of Wilson's disease who have a coagulopathy but are not yet encephalopathic. Once encephalopathy develops, liver transplantation is the only effective therapy.
Early antibiotic therapy with penicillin can be effective in patients with mushroom poisoning.
Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.
Use in all acetaminophen cases and continue until coagulation parameters improve in early use or until clinical recovery in more severe disease. Also consider in other etiologies, especially those with more rapidly evolving disease. Patients with subacute liver failure are unlikely to respond. Early therapy in acetaminophen patients may prevent acute liver failure, whereas later therapy may reduce complication rates and improve survival.
Hepatitis B: One strategy would be to use antivirals in all patients with hepatitis B. An alternative approach is to direct therapy to those shown to have elevated HBV DNA levels, although this might delay institution of therapy. The use of antivirals is not evidence-based. Institution of antivirals may be of benefit to patients undergoing liver transplantation by reducing the risk of recurrent infection.
Herpes simplex: Acyclovir therapy may improve outcome.
Corticosteroids: Limited role on a risk/benefit evaluation in autoimmune hepatitis. No role in other etiologies, including drug-induced acute liver failure.
Penicillamine. The only established role for penicillamine in the acute presentation of Wilson's disease is prevention of progression to encephalopathy. Treatment should not delay liver transplantation once encephalopathy develops. Therapy is long-term in patients surviving without liver transplantation.
Penicillin. Initiation of therapy early after ingestion of Amanita phalloides may improve outcome.
A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies
Management of complications
Cerebral edema is managed in the first instance with boluses of mannitol (1g/kg), and these can be repeated as required while effective in lowering intracranial pressure. Second-line measures include hypertonic saline and hypothermia. Third-line measures include sedation (barbiturates, propofol) and intravenous indomethacin.
Progression beyond treatment with mannitol in the management algorithm indicates that survival is unlikely unless liver transplantation is performed. The ultimate act of desperation in patients listed for liver transplantation is a hepatectomy, which stabilizes the condition for up to 24 hours.
The initial management of circulatory failure is volume repletion to restore physiological filling pressures. Persistent hypotension is managed by vasopressors with norepinephrine being the first-choice drug, followed by vasopressin when necessary. Adrenal insufficiency should be excluded in refractory patients.
Practice is split on the routine use of prophylactic fresh frozen plasma. Although it is intuitive to correct the (often profound) coagulopathy, there is no evidence base to support this practice. Targeted prophylaxis prior to invasive procedure is consistently practiced. An additional argument against prophylactic fresh frozen plasma administration is the loss of a valuable prognostic indicator. Active bleeding is managed in the standard way by repletion of an appropriate combination of blood products.
Established renal failure is best managed by continuous renal support systems rather than intermittent hemodialysis in the early stages to reduce hemodynamic instability. Patients with acetaminophen may need extended hemodialysis (often up to 6 weeks) after recovering liver function. Lesser levels of acute kidney injury are managed by addressing the targets discussed under circulatory failure. The threshold for instituting renal support therapy tends to be lower than in isolated renal failure.
Infection is managed by standard antimicrobial regimens delivering broad-spectrum cover in the first instance. The choice of specific antibiotics will depend on the antibiotic policy of individual institutions. Prophylactic antibiotic therapy has been shown to reduce the incidence of culture-proven infection but at the cost of increasing antibiotic resistance and not improving survival. Systemic antifungal therapy should be instituted in patients at risk rather than after the diagnosis of definitive infections because of the high mortality associated with the latter approach. Patients at risk include those with fevers unresponsive to antibiotics, high white cell counts, ITU stays of more than 5 days, and renal failure.
Between 30% to 50% of patients with acute liver failure undergo liver transplantation with overall 1-year survival rates at 80% or higher in many centers. One of the most critical decisions to be made in the management of a patient with acute liver failure is whether the patient should be wait-listed for a transplant. There are two broad approaches to selection for listing: 1) consider all patients potential candidates for liver transplantation and make a decision once an organ is available, and 2) direct liver transplantation at those considered to have the worst prognosis, with patients being identified using one of the available prognostic models.
Most organ allocation systems prioritize patients with acute liver failure, and more than 70% of wait-listed patients receive transplants within 2 to 3 days. The most frequently performed operation is total replacement of the liver from a deceased donor. A few centers perform auxiliary transplants, leaving part of the native liver in situ, in the hope that it may regenerate. About 70% of carefully selected patients having this procedure do achieve this goal, allowing them to revert to a nontransplant state 2 to 4 years later. Another small minority of patients receive a transplant from a living donor.
Liver support devices
The current role for liver support devices appears to be as a bridge to transplantation. None of the devices assessed in clinical trials demonstrated improved survival without recourse to liver transplantation. However, these trials did usually identify subgroups of patients that appeared to benefit, such as acetaminophen patients in an albumin dialysis (MARS) study and patients with definable etiologies in a porcine hepatocyte device study. These devices are not currently considered to be standard care.
Listing of these, including any guidelines for monitoring side effects.
Management of complications
Mannitol: Monitor serum osmolality
Hypertonic saline: Aim for serum sodium in 145-155 mmol/L range
Propofol or barbiturates
Hypothermia: Aim for body temperature of 32-34 degrees
Norepinephrine, vasopressor: Avoid excessive vasoconstriction.
Fresh frozen plasma: Monitor for fluid overload.
Renal support systems: Use continuous rather than intermittent systems.
Liver transplantation is used in about 40% of patients considered to have the poorest prognosis and results in survival rates of 70% to 80%.
Liver support devices
Biological and nonbiological devices have been used in acute liver failure but without strong evidence that these improve survival except by bridging patients to liver transplantation. At present, these devices should only be used in clinical trials.
Assess likely prognosis of patient and reassess regularly.
Consider for liver transplantation and wait-list, per local protocol.
Institute monitoring of clinical condition.
Treat complications as these arise.
Reassess prognosis at time of organ becoming available for transplantation.
How should I monitor the patient with acute liver failure?
Encephalopathy is monitored by clinical evaluation, with a limited role for blood ammonia levels.
Patients are monitored for early clinical signs of cerebral edema (e.g., arterial hypertension). Some centers use transcranial ultrasound to assess cerebral blood flow. Opinion is divided on the use of direct intracranial pressure measurements. Protagonists argue that it is the most accurate and timely method to monitor the complication and the response to therapeutic interventions. Antagonists argue that it is a risky procedure that has not been shown to improve outcome.
Renal function is monitored by measuring urine output and serum creatinine levels. Note that urea levels may understate the degree of renal impairment. Regular electrolyte monitoring is also required.
Hemodynamic monitoring should be as sophisticated as possible.
Infection should be monitored on a daily basis with repeated cultures of all blood, urine, sputum/bronchial secretions, etc. A role for serologic markers of fungal infection has not been definitively established in acute liver failure.
Monitoring progress of disease
This is a combination of repeated clinical evaluation and serial laboratory tests. The coagulation parameters are the best indicators of recovery of liver function.
Full hemodynamic monitoring
Doppler of intracranial vessels: Increasingly used but is not the widespread practice.
Intracranial pressure monitoring: Controversial; when used, needs to continue until resolution of cerebral edema or 24 hours after successful liver transplantation
Repeated sepsis screens
Daily laboratory tests
What's the evidence?
Overview of acute liver failure
Bernal, W, Auzinger, G, Dhawan, A, Wendon, J. "Acute liver failure". Lancet . vol. 376. 2010. pp. 190-201.(This is a recent and comprehensive overview of acute liver failure.)
Larson, AM, Polson, J, Fontana, RJ. "Acetaminophen-induced acute liver failure: results of a United States multicenter prospective study". Hepatology. vol. 42. 2005. pp. 1364-72.(This article gives a good insight into the U.S. perspective of acetaminophen-induced acute liver failure. Note that the UK perspective is different on a number of issues.)
Management of acute liver failure
Hu, J, Zhang, Q, Ren, X. " Efficacy and safety of acetylcysteine in ‘non-acetaminophen’ acute liver failure: a meta-analysis of prospective clinical trials". Clin Res Hepatol Gastroenterol. vol. 39. 2015 Oct. pp. 594-9.(A meta-analysis evaluating the use of NAC in non-acetaminophen causes of ALF showing that NAC is safe in this setting and associated with improved survival with native liver.)
Siddiqui, MS, Stravitz, RT. "Intensive care unit management of patients with acute liver failure". Clin Liver Dis. vol. 18. 2014 Nov. pp. 957-78.(A review summarizing the current ICU management practices of patients with acute liver failure.)
Stravitz, RT, Kramer, DJ. "Management of acute liver failure". Nat Rev Gastroenterol Hepatol . vol. 6. 2009. pp. 542-53.(This is a very detailed review of specific components of the management of acute liver failure.)
Auzinger, G, Wendon, J. "Intensive care management of acute liver failure". Curr Opin Crit Care . vol. 14. 2008. pp. 179-88.(This review is from a very experienced center and complements the Stravitz and Kramer review.)
Liver transplantation for acute liver failure
Bernal, W, Cross, TJ, Auzinger, G. "Outcome after wait-listing for emergency liver transplantation in acute liver failure: a single centre experience". J Hepatol. vol. 50. 2009. pp. 306-13.(This work gives insight into outcomes from the point of listing for emergency liver transplantation.)
Liver support devices
Phua, J, Lee, KH. "Liver support devices". Curr Opin Crit Care . vol. 14. 2008. pp. 208-15.(This work is a concise review of liver support devices.)
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.