Acute Liver Failure: Current Practice and Recent Advances




Acute liver failure (ALF), also referred as “fulminant hepatic failure,” is a rare syndrome characterized by rapid onset of severe hepatic dysfunction in the absence of preexistent liver disease. The principal manifestations of ALF include jaundice, hepatic encephalopathy, and coagulopathy, often owing to massive or submassive hepatic necrosis. An estimated 2000 to 3000 cases of ALF occur in the United States every year. The syndrome has diverse etiologies that manifest wide geographic distribution. In North America and Western Europe, acetaminophen and idiosyncratic drug hepatotoxicity are the predominant causes. In contrast, viral hepatitis B and E are the major causes in Asia. Without specialized intensive care management and the availability of liver transplantation, a diagnosis of ALF portends a poor outcome. This article summarizes current practice and recent advances in the management of ALF.


Diagnosis and Initial Management


Early diagnosis of ALF is crucial. Although jaundice and hepatic encephalopathy are the cardinal manifestations of ALF, many patients present with nonspecific symptoms of generalized malaise, nausea, and abdominal discomfort. Thus, concern for severe liver injury is often triggered by abnormal laboratory values. Hepatic encephalopathy develops rapidly in patients with hyperacute failure sometimes preceding jaundice. Careful clinical evaluation and laboratory and imaging studies are required to help exclude cirrhosis and to establish the etiology of the disease. The role of liver biopsy is limited and it is often problematic in view of cerebral and hemodynamic instability, and associated coagulopathy.


Patients with ALF are best managed in an intensive care unit. Once diagnostic criteria are met, transfer to a tertiary facility with a liver transplant program is urgent. Transfer may be considered earlier if a patient with severe acute liver injury is deemed likely to progress to ALF. Worsening coagulopathy and behavioral changes are particularly relevant in this regard. A decision regarding tracheal intubation should be made before transfer. Intubation is particularly recommended for patients with grades III or IV hepatic encephalopathy. Rapid development of hypoglycemia is a concern and should be addressed by a continuous intravenous dextrose infusion and frequent monitoring of blood sugar levels. Encephalopathy in ALF may be associated with agitation and seizures; therefore, a quiet environment is needed to minimize external stimuli.




Cerebral Edema


Cerebral edema and intracranial hypertension (ICH) are among the most serious complications of ALF. As cerebral edema progresses, fatal uncal herniation may ensue. Residual neurologic deficits have been noted among survivors. Cerebral edema should be suspected in patients with progressive hepatic encephalopathy. Although it is rare among those with grades I or II, its incidence is 25% to 35% with grade III and as high as 65% to 75% with grade IV hepatic encephalopathy. A high index of suspicion is needed because patients may not exhibit classic features including headache, vomiting, bradycardia, hypertension, blurred vision, papilledema, brisk reflexes, and decerebrate rigidity.


The pathogenesis of ICH from ALF is not fully understood; however, hyperammonemia is considered to be a key factor. Once ammonia traverses the blood–brain barrier, astrocytes detoxify it by conversion to glutamine through its reaction with glutamic acid, a process facilitated by the enzyme glutamine synthetase. Glutamine has a direct osmotic effect resulting in astrocyte swelling. In addition, it serves as a “Trojan horse” by releasing ammonia inside the cell that causes mitochondrial dysfunction and oxidative stress, which in turn results in cytotoxicity and cellular edema. Furthermore, release of inflammatory cytokines leads to systemic vasodilation and increased intracranial blood volume.


Intracranial Pressure Monitoring


Although leading transplant centers in the United States tend to employ intracranial pressure (ICP) measurement, there are no randomized trials or consensus guidelines to support this practice. ICP is often measured to have an objective parameter to aid in management and prognostication. However, target values are not well-defined and monitor placement is associated with the risk of intracerebral bleeding that may occur in 8% to 10% of cases, although less frequently with an epidural catheter. This risk has been shown to be mitigated by the use of recombinant activated factor VII (rFVIIa). Some centers use ICP monitors routinely, whereas others employ it in patients who are at the highest risk of cerebral edema, including those with grades III to IV hepatic encephalopathy, or those receiving vasopressors. Intracranial monitors are more frequently used among patients listed for liver transplantation. The United States Acute Liver Failure Study Group endorses the use of ICP monitors in patients with a high risk of ICH, including nontransplant candidates who have increased likelihood of spontaneous survival.


Cerebral Blood Flow Measurement


ALF is associated with the loss of cerebral blood flow autoregulation. Depending on the cerebral perfusion pressure (CPP), either cerebral hyperemia or hypoxia could develop. Measurement of cerebral blood flow (CBF), in addition to ICP monitoring, may therefore help to manage such patients. Still, CBF-oriented therapies are difficult to implement because the metabolic demands of the brain vary depending on the patient’s underlying level of inflammation or sedation, and there is no evidence that CBF monitoring influences outcome. Several tools are available for the indirect measurement of CBF, including the jugular bulb catheter, transcranial Doppler, and xenon-enhanced computed tomography (CT).


Jugular bulb


A jugular bulb catheter may be utilized to assess cerebrovascular autoregulation. This process involves passage of a fine-bore catheter into the internal jugular vein until the tip reaches the jugular bulb, thereby allowing for sampling of blood that drains exclusively from intracranial circulation. Cerebral oxygen uptake is then determined by calculating the arteriojugular oxygen content difference in paired blood samples. During assessment of cerebral autoregulation, it is assumed that cerebral metabolic rate of oxygen remains constant so that the arteriojugular oxygen content difference solely reflects CBF. In reality, the cerebral metabolic rate of oxygen may vary, and that limits the utility of the jugular bulb in the assessment of CBF autoregulation. Jugular bulb placement may also be used to measure brain cytokine production indicative of an inflammatory response that correlates with the severity of ICP.


Transcranial Doppler ultrasonography


Transcranial Doppler ultrasound is a noninvasive method to measure blood flow velocity in the basal intracranial cerebral arteries, thereby indirectly determining CBF via the linear relationship between flow and velocity. Blood flow velocity and CBF measured by transcranial Doppler has been shown to correlate with xenon-enhanced CT determination of CBF. Among patients with ALF, monitoring with transcranial Doppler provides an assessment of cerebral hemodynamics and could be used to predict dynamic changes in CBF and ICP.


Xenon-enhanced CT


CBF measurement with xenon, an inert anesthetic gas, was first described 40 years ago. After the advent of CT, xenon-enhanced scanning was developed to determine regional blood flow. Xenon CT provides both anatomic and functional information that could help to manage patients with ALF. However, there are disadvantages to this modality. For instance, the reproducibility of findings is poor as comparisons of the same region from one study to another are difficult. Furthermore, CBF assessment is restricted to mainly cortical and subcortical regions within the territory of the middle cerebral artery. There is no consensus regarding the use of xenon-enhanced CT in patients with ALF and its application therefore remains limited.


Electroencephalography


Seizure activity is not uncommon among patients with ALF ; however, it is often masked by the use of sedatives and paralytic agents. Seizures may acutely elevate ICP, and may also increase cerebral oxygen consumption resulting in ischemia and worsening cerebral edema. Monitoring by continuous electroencephalography would detect subclinical seizure activity that could be treated with anti-epileptic agents. Studies examining the effect of prophylactic anticonvulsants have remained inconclusive.




Cerebral Edema


Cerebral edema and intracranial hypertension (ICH) are among the most serious complications of ALF. As cerebral edema progresses, fatal uncal herniation may ensue. Residual neurologic deficits have been noted among survivors. Cerebral edema should be suspected in patients with progressive hepatic encephalopathy. Although it is rare among those with grades I or II, its incidence is 25% to 35% with grade III and as high as 65% to 75% with grade IV hepatic encephalopathy. A high index of suspicion is needed because patients may not exhibit classic features including headache, vomiting, bradycardia, hypertension, blurred vision, papilledema, brisk reflexes, and decerebrate rigidity.


The pathogenesis of ICH from ALF is not fully understood; however, hyperammonemia is considered to be a key factor. Once ammonia traverses the blood–brain barrier, astrocytes detoxify it by conversion to glutamine through its reaction with glutamic acid, a process facilitated by the enzyme glutamine synthetase. Glutamine has a direct osmotic effect resulting in astrocyte swelling. In addition, it serves as a “Trojan horse” by releasing ammonia inside the cell that causes mitochondrial dysfunction and oxidative stress, which in turn results in cytotoxicity and cellular edema. Furthermore, release of inflammatory cytokines leads to systemic vasodilation and increased intracranial blood volume.


Intracranial Pressure Monitoring


Although leading transplant centers in the United States tend to employ intracranial pressure (ICP) measurement, there are no randomized trials or consensus guidelines to support this practice. ICP is often measured to have an objective parameter to aid in management and prognostication. However, target values are not well-defined and monitor placement is associated with the risk of intracerebral bleeding that may occur in 8% to 10% of cases, although less frequently with an epidural catheter. This risk has been shown to be mitigated by the use of recombinant activated factor VII (rFVIIa). Some centers use ICP monitors routinely, whereas others employ it in patients who are at the highest risk of cerebral edema, including those with grades III to IV hepatic encephalopathy, or those receiving vasopressors. Intracranial monitors are more frequently used among patients listed for liver transplantation. The United States Acute Liver Failure Study Group endorses the use of ICP monitors in patients with a high risk of ICH, including nontransplant candidates who have increased likelihood of spontaneous survival.


Cerebral Blood Flow Measurement


ALF is associated with the loss of cerebral blood flow autoregulation. Depending on the cerebral perfusion pressure (CPP), either cerebral hyperemia or hypoxia could develop. Measurement of cerebral blood flow (CBF), in addition to ICP monitoring, may therefore help to manage such patients. Still, CBF-oriented therapies are difficult to implement because the metabolic demands of the brain vary depending on the patient’s underlying level of inflammation or sedation, and there is no evidence that CBF monitoring influences outcome. Several tools are available for the indirect measurement of CBF, including the jugular bulb catheter, transcranial Doppler, and xenon-enhanced computed tomography (CT).


Jugular bulb


A jugular bulb catheter may be utilized to assess cerebrovascular autoregulation. This process involves passage of a fine-bore catheter into the internal jugular vein until the tip reaches the jugular bulb, thereby allowing for sampling of blood that drains exclusively from intracranial circulation. Cerebral oxygen uptake is then determined by calculating the arteriojugular oxygen content difference in paired blood samples. During assessment of cerebral autoregulation, it is assumed that cerebral metabolic rate of oxygen remains constant so that the arteriojugular oxygen content difference solely reflects CBF. In reality, the cerebral metabolic rate of oxygen may vary, and that limits the utility of the jugular bulb in the assessment of CBF autoregulation. Jugular bulb placement may also be used to measure brain cytokine production indicative of an inflammatory response that correlates with the severity of ICP.


Transcranial Doppler ultrasonography


Transcranial Doppler ultrasound is a noninvasive method to measure blood flow velocity in the basal intracranial cerebral arteries, thereby indirectly determining CBF via the linear relationship between flow and velocity. Blood flow velocity and CBF measured by transcranial Doppler has been shown to correlate with xenon-enhanced CT determination of CBF. Among patients with ALF, monitoring with transcranial Doppler provides an assessment of cerebral hemodynamics and could be used to predict dynamic changes in CBF and ICP.


Xenon-enhanced CT


CBF measurement with xenon, an inert anesthetic gas, was first described 40 years ago. After the advent of CT, xenon-enhanced scanning was developed to determine regional blood flow. Xenon CT provides both anatomic and functional information that could help to manage patients with ALF. However, there are disadvantages to this modality. For instance, the reproducibility of findings is poor as comparisons of the same region from one study to another are difficult. Furthermore, CBF assessment is restricted to mainly cortical and subcortical regions within the territory of the middle cerebral artery. There is no consensus regarding the use of xenon-enhanced CT in patients with ALF and its application therefore remains limited.


Electroencephalography


Seizure activity is not uncommon among patients with ALF ; however, it is often masked by the use of sedatives and paralytic agents. Seizures may acutely elevate ICP, and may also increase cerebral oxygen consumption resulting in ischemia and worsening cerebral edema. Monitoring by continuous electroencephalography would detect subclinical seizure activity that could be treated with anti-epileptic agents. Studies examining the effect of prophylactic anticonvulsants have remained inconclusive.




ICH


Normal ICP in a supine healthy adult ranges from 7 to 15 mmHg. The goal of therapy in ALF is to maintain ICP below 20 and 25 mmHg and to preserve CPP, calculated as mean arterial pressure (MAP) – ICP, at 50 to 60 mmHg. Among trauma patients, CPP of 70 mmHg has been noted to be beneficial. However, such goals are not well-defined and are based on the experiences of individual centers. CPP indicates pressure gradient across the cerebral vascular bed. The rationale of increasing the CPP is to enhance CBF to regions of the brain with critically low flow when autoregulation has failed or CPP is below the lower limit of autoregulation.


Initial maneuvers should include elevation of the head of the bed to 30 degrees, maintenance of a neutral neck position and minimizing painful stimuli to reduce patient agitation. Additional modalities include hyperventilation, which rapidly but transiently restores autoregulation of CBF by lowering the partial pressure of carbon dioxide in arterial blood (P a CO 2 ), hypothermia, and medical treatment including mannitol and thiopental. Limited evidence also supports the use of hypertonic saline, propofol sedation, and indomethacin.


Hyperventilation


Hyperventilation provides a rapid but transient improvement in ICP, with restoration of cerebral autoregulation within several minutes. The goal of hyperventilation is to induce the hypocapnia that causes cerebral vasoconstriction, which in turn reduces CBF leading to a decrease in ICP. In brain-injured patients, both moderate (P a CO 2 , 31–35 mmHg) and forced (P a CO 2 ≤30 mmHg) hyperventilation have been used to treat ICH. Although hyperventilation effectively reduces ICP, there is concern that the resultant vasoconstriction could worsen cerebral hypoxia and even cause ischemia. Careful monitoring of cerebral perfusion and oxygenation is therefore essential. There is no role for prophylactic hyperventilation in patients with ALF. Hyperventilation is likely best used as short-term rescue therapy.


Mannitol


Intravenous mannitol is a potent osmotic agent that does not cross the blood–brain barrier. It reduces ICP by osmotically drawing water from the brain parenchyma into the intravascular space. Although widely accepted as first-line medical therapy for ICH in ALF, there are few published studies to support its use. Mannitol is typically administered as a 20% solution in intravenous boluses of 0.25 to 1 g/kg that are repeated every 4 to 6 hours depending on ICP response. Serum osmolality should be measured before each dose, because a serum osmolality greater than 320 mOsm is associated with greater risk of renal tubular toxicity. However, the evidence for this threshold remains obscure. A retrospective study of 605 patients treated with mannitol showed increased mortality only with severe hypernatremia that equated to serum osmolality of 335 to 340 mOsm. Mannitol may thus be used safely, even if higher serum osmolalities are reached, as long as adequate hydration is maintained. Additionally, mannitol fails to normalize ICP once a level of greater than 60 mmHg is reached.


Hypothermia


Hypothermia can transiently reduce ICP in patients with cerebral edema who are refractory to medical therapy. It could thus be used as a bridge to recovery or to liver transplantation. Hypothermia reduces ICP by restoring CBF autoregulation and reactivity to CO 2 . The efficacy of moderate hypothermia, as defined by a reduction in core body temperature to 32°C to 33°C, was first noted in brain-injured patients. In subsequent studies among patients with ALF, hypothermia seemed to hasten neurologic recovery. However, the clinical evidence regarding induced hypothermia is limited to small case series. A meta-analysis of 5 case series demonstrated that ICP monitoring combined with hypothermia led to improved ICP, CPP, and CBF in 4 of the 5 studies. Hypothermia could be induced by external cooling blankets, intravascular cooling devices, and body suits with monitoring of core body temperatures via a rectal or intravascular thermometer. A paralytic agent or a deep sedative may be needed to prevent reflexive shivering. Moderate hypothermia needs to be used cautiously in view of the potential risks including cardiac arrhythmias, coagulopathy, hypotension, and impaired liver regeneration. Mild hypothermia (core temperature of 34°C–35°C) may also be beneficial with a lesser likelihood of adverse effects. Randomized, controlled trials are needed to confirm the benefits of hypothermia before it is applied routinely.


Hypertonic Saline


Uncontrolled studies of hypertonic saline in patients with traumatic brain injury showed favorable effect on systemic hemodynamics and ICP. Hypertonic saline and dextran solution was noted to be superior to mannitol infusion in another study. However, a recently published, large, randomized trial failed to show a beneficial effect of hypertonic saline on neurologic outcome or survival. The experience in patients with ALF has remained limited. A study of 30 patients with ALF and grades III or IV encephalopathy showed a reduction in the incidence and severity of ICH with hypertonic saline infusion in comparison with the standard of care. Hypertonic saline mitigates ICH through both osmotic and nonosmotic effects. The blood–brain barrier prevents the free flow of water and solutes into and out of the cerebral cells. A high level of circulating sodium creates a gradient that favors the movement of water from the brain tissue into the circulation. It also reduces endothelial swelling that improves CBF through the microcirculation. Hypertonic saline inhibits neutrophil activation and release of proinflammatory cytokines that may help control the systemic inflammatory response syndrome (SIRS) that is often a component of ALF.


Hypertonic saline is administered as a continuous infusion or as boluses through a central venous catheter to prevent phlebitis. In addition, potassium and magnesium supplementation may be necessary to avoid the risk of cardiac arrhythmia. Other possible adverse effects include pulmonary edema, central pontine myelinolysis, and hyperchloremic metabolic acidosis.


Barbiturates


Thiopental and pentobarbital are centrally acting hypnotics that reduce brain oxygen utilization. They are considered second-line therapy for severe ICH. Their administration requires continuous electroencephalographic monitoring for dose titration. Because barbiturate infusion may cause systemic hypotension, blood pressure support is often needed to maintain an adequate CPP. Other adverse effects include cardiac arrhythmia and immune suppression that increases the risk of infections and ileus.




Coagulopathy


The liver plays a central role in hemostasis as it synthesizes almost all coagulation factors; in addition, it produces many inhibitors of coagulation and proteins involved in fibrinolysis. Coagulopathy [International Normalized Ratio (INR) > 1.5] is therefore considered a key feature of ALF. There are a number of factors involved in the bleeding and clotting diatheses seen in ALF including platelet dysfunction (quantitative and qualitative) and deficiency of clotting factors (II, V, VII, IX, and X) and possibly that of vitamin K. There are reduced levels of fibrinolytic proteins except plasminogen activator inhibitor-1 that is greatly increased, resulting in suppression of fibrinolysis. Anticoagulant factors such as antithrombin III and proteins C and S are also reduced that further helps to rebalance the system. Hemostatic changes in ALF thus incorporate a coagulopathy (measured as prolonged prothrombin and partial thromboplastin time) as well as a tendency to develop thrombotic events such as disseminated intravascular coagulation.


In a patient with ALF, correction of the INR to a level that is considered normal is not required for prophylactic purposes and it is likely unobtainable. Any improvement in INR from fresh frozen plasma (FFP) is transient and correction also obscures prothrombin time as a prognostic marker. However, an attempt at improving coagulation parameters is recommended in situations such as clinically significant bleeding, before performing an invasive procedure with high bleeding risk, or before insertion of an ICP monitoring device. There are insufficient data available to support a specific platelet count or INR, although general guidelines include an INR of less than 1.5 and platelet count of more than 50,000/mL. Cryoprecipitate is recommended in patients who have significant hypofibrinogenemia (<100 mg/dL). Administration of vitamin K may be helpful and should be considered in all patients with ALF. However, because of the potential for poor absorption of oral vitamin K, especially in those with bowel wall edema or cholestatic jaundice, intravenous vitamin K is recommended. For patients who do not respond to FFP or who cannot receive FFP owing to volume overload or risk of adverse reactions, rFVIIa may be given. It is advisable to replenish other coagulation factors with FFP and fibrinogen with cryoprecipitate, before rFVIIa is administered. The effect of rFVIIa persists for approximately 2 to 6 hours. Because rFVIIa increases the risk of thrombosis, it should be avoided in patients with prior myocardial infarction, ischemic stroke, or pulmonary embolism, as well as in those with ALF from pregnancy, acute portal vein thrombosis, or Budd–Chiari syndrome.




Infections and SIRS


Patients with ALF are highly susceptible to infections. Bacterial infections have been documented in 80% of the cases, most commonly pneumonia (50%), urinary tract infections (22%), intravenous catheter-induced bacteremia (12%), and spontaneous bacteremia (16%). Gram-negative enteric bacilli and gram-positive cocci are most frequently isolated. Fungal infections occur in about one third, largely with candida species. In addition to their propensity to infections, patients with ALF have accentuated hemodynamic, hormonal, and cytokine responses associated with SIRS and septic shock. Patients who develop SIRS are more likely to have worsening hepatic encephalopathy and poor outcome. ALF patients may therefore benefit from routine microbiologic surveillance for early detection and treatment of infections. Although evidence supporting the routine use of prophylactic antibiotics is limited to small studies that showed no effect on survival, such a strategy may be beneficial in those with worsening encephalopathy and progression to SIRS. Empiric antibiotics are also recommended for patients listed for liver transplantation, because development of infection and sepsis may prompt delisting.




Metabolic Issues and Renal Failure


Metabolic and electrolyte abnormalities are common in ALF. Patients are prone to develop hypoglycemia because hepatocyte necrosis causes glycogen depletion and defective glycogenolysis and gluconeogenesis. Dextrose infusion would correct the abnormality, but hyponatremia should be avoided because it may exacerbate cerebral edema. Serum phosphate, magnesium, and potassium levels are frequently low and need supplementation. Relative adrenal insufficiency develops in one third of the patients, requiring steroid replacement. Owing to the hypercatabolic state of ALF, nutrition is vital and enteral feedings should be initiated early. If enteral feedings are contraindicated, parenteral nutrition may be utilized.


Renal failure develops in 40% to 50% of patients with ALF. A number of factors contribute to the development of renal failure that may present as the hepatorenal syndrome, prerenal azotemia, or acute tubular necrosis. Such factors include toxic agents inducing ALF that are also nephrotoxic (acetaminophen, amanita), other nephrotoxic medications, sepsis, and volume depletion. Management involves withdrawal of nephrotoxic agents, correction of hypovolemia and renal support. The decision to start renal replacement therapy should be based on the level of renal dysfunction, volume overload, and metabolic derangements, such as acidosis and hyperkalemia. Continuous renal replacement therapy is recommended because patients with ALF tolerate intermittent hemodialysis poorly because of hemodynamic instability, fluid shifts and a rise in ICP.




Cardiovascular and Respiratory Support


ALF is characterized by a hyperdynamic circulation with high cardiac output, low arterial pressures, and diminished systemic vascular resistance. This results in reduced oxygen delivery and consumption, increasing the risk of tissue hypoxia. Fluid resuscitation is often required and is usually achieved with colloid infusion. After fluid resuscitation, intravenous dopamine or norepinephrine may be initiated to maintain an adequate MAP and CPP. CBF in ALF correlates with arterial pressure owing to loss of autoregulation. It is, therefore, important to maintain cardiovascular stability with well-titrated doses of pressors to avoid worsening cerebral hyperemia and edema. Terlipressin, a vasopressin synthetic analog, has been evaluated in 2 small studies with conflicting results. One of the studies raised concern that terlipressin and vasopressin may worsen cerebral hyperemia and ICH.


The development of severe acute lung injury is associated with poor prognosis in ALF. Acute lung injury is characterized by impaired gas exchange and bilateral alveolar or interstitial infiltrates in the absence of congestive cardiac failure. Among patients with ALF, acute lung injury increases the risk of circulatory failure and cerebral edema. Endotracheal intubation is needed in patients with respiratory failure, severe hepatic encephalopathy, or agitation, and before placement of an ICP monitor. There are insufficient data to recommend a standard mode of mechanical ventilation in patients with ALF. Low tidal volume ventilation is beneficial; it is associated with reduced mortality and shorter duration of ventilator use. However, it must be appreciated that decrements in tidal volume decrease minute ventilation and increase the partial pressure of carbon dioxide, which may increase ICP. Respiratory rate should therefore be increased to avoid marked hypercapnia. It is also prudent to maintain the lowest level of positive end-expiratory pressure that achieves adequate oxygenation because high levels may decrease hepatic blood flow.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Feb 26, 2017 | Posted by in GASTROENTEROLOGY | Comments Off on Acute Liver Failure: Current Practice and Recent Advances

Full access? Get Clinical Tree

Get Clinical Tree app for offline access