Acute decompensation of chronic end-stage liver disease

Chronic liver disease is defined as any sort of hepatic dysfunction which has persisted for longer than 6 months. There are several key differences in the presentation of an acute decompensation of chronic disease, as opposed to the presentation of an acutely sick liver which was previously well.

Causes of decompensation

Question 4 from the first paper of 2018 asked for six reasons for why chronic liver disease might suddently decompensate. Kim et al (2012) give a list which probably varies slightly depending on geographical region:

  • Sepsis
  • Spontaneous bacterial peritonitis
  • Reactivation of a hepatitis virus (eg. hep B, C)
  • Alcohol
  • Drug-induced toxicity
  • Gastrointestinal haemorrhage
  • Surgery
  • Portal vein thrombosis
  • Ischaemic insult of any sort (eg. shock)
  • Autoimmune hepatitis
  • Wilson disease

Major differences between acute and chronic liver failure

This had come up in Question 18 from the first paper of 2020, where the college wanted a tabulated answer comparing these conditions on the basis of their "history, examination, biochemical findings, haematological findings, imaging and biopsy". 

Differences between Acute and Chronic Liver Failure
Domain Acute fulminant hepatic failure Decompensated chronic liver disease
  • Abrupt onset
  • Clear antecedent cause, eg. toxin ingestion like paracetamol, or trauma, or ischaemia
  • Gradual onset
  • Associated virus (eg. Hep C)
  • Associated toxin exposure (eg. alcohol)
  • Acute lung injury
  • Systemic inflammatory response, vasodilated state
  • SvO2 is high because of hyperdynamic circulation
  • Decreased level of consciousness due to cerebral oedema
  • Euvolaemic and without ascites
  • Well-nourished
  • No evidence of portosystemic shunting (thus, no varices)
  • SvO2 is an unreliable marker of tissue perfusion, because of all the shunts.
  • Confusion and delirium due to hepatic encephalopathy
  • Fluid-overloaded, with ascites
  • Portal hypertension is present (thus, varices)
  • Malnourished
  • Spider naevi
  • High ammonia level
  • Normal urea
  • Extremely high LFTs
  • Acute renal failure
  • Ammonia may be normal 
  • Low urea (unless bleeding)
  • Normal-ish LFTs
  • Initially, normal coags
  • Chronically low levels of clotting factors
  • US/CT: enlarged liver, hypodense, heterogenous echogenicity
  • US/CT: small liver, irregular contours, increased density, hyperechoic. Dilated portal vein, evidence of portal hypertension
  • "Acute necrosis" is what it usually says (i.e. it is often uninformative)
  • Discriminates between different causes
  • Diagnostic for cirrhosis
  • Useful for prognosis

A few other differences exist between ICU management of acute and acute-on-chronic liver failure.

In brief summary:

  • Encephalopathy does not lead to intracranial hypertension. Management of an acute exacerbation of chronic hepatic encephalopathy is really all about supporting their airway while the lactulose has a chance to work.
  • SvO2 is an unreliable marker of tissue perfusion, because of all the shunts.
  • Lactate is an unreliable marker of tissue perfusion, because of the crap liver.
  • Albumin may be helpful in sepsis, in contrast to other scenarios where it probably is not.
  • Spontaneous bacterial peritonitis belongs on any list of differentials
  • Hepatorenal syndrome and hepatopulmonary syndrome need to be remembered as causes of renal and respiratory failure
  • Variceal bleeding leads to sepsis, and sepsis leads to variceal bleeding.

Idiosyncrasies of managing sepsis in chronic liver failure

Well, they are already having a hyperdynamic circulation, but apart from that, several other differences exist.

There is increased susceptibility to sepsis

One of the reasons for such severe systemic consequences of sepsis is the normal role of the liver in clearing LPS from the gut. Lipopolysaccharide would normally be gobbled up by Kupffer cells, but in cirrhosis they are all dead, and unclean portal blood escapes into the systemic circulation via portasystemic shunts. Systemic exposure to LPS (and to translocated bacteria) seems to be the main reason there is such an extensive activation of nitric oxide synthase, and explains the hyperdynamic circulation of the cirrhosis patient.

On top of all that, there tends to be a "sepsis-like immune paralysis" where numerically normal macrophage and neutrophil populations somehow fail to opsonise and phagocytose the invading enemy. This must be to some extent a consequence of the complement deficiency (as the liver is normally responsible for the syntesis of complement).

SvO2 is an unreliable marker of tissue perfusion

Because of massive shunts, there may be whole lumps of tissue bypassed by the circulation entirely, thus generating a falsely elevated SvO2.

Lactate is an unreliable marker of tissue perfusion

Because the liver-diseased patient lacks the major clearance mechanism for lactate, it would be silly to expect their lactate to fall satisfyingly in response to resuscitation. A falsely elevated lactate may exist in a well-resuscitated patient.

Albumin may be helpful

Small weirdly-designed studies seem to support the use of albumin as a resuscitation fluid in a septic patient with chronic liver disease. Sure, its better than hetastarch, but then probably toilet water is better than hetastarch. There may also be some benefit in preventing hepatorenal syndrome and renal failure.

Spontaneous bacterial peritonitis

This pathology attracted college atttention in Question 4 from the first paper of 2018, and has therefore been given a chapter all of its own. Key features to remember when you draw some ascitic fluid for culture:

  • lots of fluid is required - the bacteria are usually present in a very low concentration
  • A single organism is most likely SBP, whereas a zoo of organisms is most likely a product of "secondary peritonitis", like a perforated bowel.

Hepatorenal syndrome

This topic is discussed elsewhere. In short summary, it is renal failure without any better reason for renal failure, in the patient with severe liver disease.The management of confirmed hepatorenal syndrome would consist of:

Hepatopulmonary syndrome

This uncommon syndrome features the dilatation of pulmonary vessels, which leads to an inappropriate increase in perfusion to relatively underventilated regions. Again, nitric oxide synthase is to blame. There are diagnostic criteria:

Diagnostic criteria for hepatopulmonary syndrome:

  • PaO2 under 75mmHg on room air, OR
  • A-a gradient of over 20mmHg
  • Pulmonary vessel dilatation in the presence of cirrhosis

Apparently this is present in 20% of cirrhosis patients. It does tend to resolve after transplantation, but this may take up to 1 year.

Portopulmonary hypertension

This is pulmonary hypertension with evidence of increased pulmonary resistance (in contrast to the "normal" pulmonary hypertension of the cirrhosis patient, which is associated with increased pulmonary pressure due to increased flow though dilated unresisting arteries).

Diagnostic criteria for portopulmonary hypertension

  • Portal hypertension
  • Mean PA pressure over 25mmHg
  • PAWP over 15mmHg
  • Pulmonary vascular resistance over 120 dynes/s/cm-5

This is also present in 20% of cirrhosis patients. Management here is different; the goal is to reduce the pulmonary pressure. Bosantan sildenafil and epoprostenol have been used.

The presence of portopulmonary hypertension is a major issue; perioperative mortality due to right heart decompensation is 100% if the MPAP is over 50mmHg. All sorts of advanced hemodynamic monitoring tools should be deployed to keep these patients alive; milrinone and intravenous prostacycline have been used by various authors.

Bleeding from the varices

This topic is discussed elsewhere. In the chronic liver disease patient, there are few additional points to remember:

  • Sepsis is more likely to develop if you have uncontrolled gastrointestinal bleeding; and acute gastrointestinal bleeding is more likely to develop if the patient already has sepsis. Empirical antibiotics are recommended in variceal bleeding- and they clearly decrease the risk of re-bleeding.


Chapter 44   (pp. 501) Liver  failure by Christopher  Willars  and  Julia  Wendon

Chapter 101  (pp. 1040)  Liver  transplantation by Anish  Gupta,  Simon  Cottam  and  Julia  Wendon

Gustot, Thierry, et al. "Severe sepsis in cirrhosis." Hepatology 50.6 (2009): 2022-2033.

Fernández, Javier, et al. "A randomized unblinded pilot study comparing albumin versus hydroxyethyl starch in spontaneous bacterial peritonitis."Hepatology 42.3 (2005): 627-634.

Sort, Pau, et al. "Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis." New England Journal of Medicine 341.6 (1999): 403-409.

Hou, Ming‐Chih, et al. "Antibiotic prophylaxis after endoscopic therapy prevents rebleeding in acute variceal hemorrhage: a randomized trial." Hepatology 39.3 (2004): 746-753.

Lange, Paul A., and James K. Stoller. "The hepatopulmonary syndrome."Annals of Internal Medicine 122.7 (1995): 521-529.

Krowka, Michael J., et al. "Pulmonary hemodynamics and perioperative cardiopulmonary‐related mortality in patients with portopulmonary hypertension undergoing liver transplantation." Liver Transplantation 6.4 (2000): 443-450.

Kim, Tae Yeob, and Dong Joon Kim. "Acute-on-chronic liver failure." Clinical and molecular hepatology 19.4 (2013): 349.

Bernal, William, et al. "Acute liver failure." The Lancet 376.9736 (2010): 190-201.