Hepatic Encephalopathy: Grades, Causes, Treatment and What Families Should Watch For
Hepatic encephalopathy is brain dysfunction caused by a failing liver, and in most cases it is reversible once the trigger is found. Overt episodes occur in 30–40% of people with cirrhosis at some point; the milder covert form is commoner still and is missed on routine examination. Treatment means correcting the precipitant — infection, bleeding, constipation, dehydration — plus lactulose, with rifaximin added to prevent recurrence. Protein is never restricted.
— Dr Chetan Kalal, DM Hepatology, Gleneagles Hospital Mumbai
Key takeaways
- Hepatic encephalopathy (HE) is a spectrum, from subtle slowing on psychometric tests (covert HE) to coma (grade IV). The 2014 AASLD/EASL guideline estimates that 30–40% of people with cirrhosis will have an overt episode at some stage, and that covert HE affects 20–80%.
- Nearly every episode has a trigger. Infection, gastrointestinal bleeding, diuretic overuse and dehydration, electrolyte disturbance, constipation and sedatives account for most. Find it, fix it, and the brain usually recovers.
- Diagnosis is clinical. A raised ammonia level alone adds nothing to the diagnosis, grading or prognosis of HE in chronic liver disease (AASLD/EASL 2014, GRADE II-3, A, 1). Acute liver failure is a different disease, and ammonia carries different weight there.
- Lactulose is first-line treatment and first-line prevention; rifaximin added to lactulose cut recurrence from 45.9% to 22.1% over six months in the landmark 2010 trial. The Delhi trial that established lactulose for secondary prophylaxis came from Prof Sarin’s group.
- Do not restrict protein. EASL’s 2019 nutrition guideline says so in four words. Patients with HE need 1.2–1.5 g/kg/day, split into small meals with a late-evening snack — starvation worsens encephalopathy.
- Covert HE impairs driving; patients cannot judge this themselves. Families spot HE before doctors do — sleep reversal, a changed signature, salt added twice. A second episode within six months, or any episode with poor liver function, is the point to ask about transplant evaluation.
A composite from the clinic. A 58-year-old man from Dombivli with alcohol-related cirrhosis, abstinent two years, is brought in by his wife because “he is not himself”. He sleeps through the afternoon and paces the flat at 3 a.m. His signature on a cheque last week was a scrawl she did not recognise. He salted his dal, forgot, and salted it again. He had stopped his lactulose a fortnight earlier — the loose motions were embarrassing on the train — and had not passed stool in four days. He would pass a casual conversation. His wife is right and the conversation is wrong: this is early hepatic encephalopathy, precipitated by constipation, and it is entirely reversible.
Hepatic encephalopathy is the complication of cirrhosis that families notice first and doctors sometimes confirm last. It can look like dementia or a stroke. Most of the time it is neither permanent nor mysterious — a reversible brain dysfunction caused by a liver that can no longer clear the gut’s toxins, tipped over by something specific that can be found and corrected. This guide covers the forms and grades, the triggers, diagnosis (and the ammonia misunderstanding), treatment, the protein myth, driving, and when encephalopathy becomes a reason to discuss liver transplantation. It is anchored to the AASLD/EASL 2014 and EASL 2022 guidelines and the EASL 2019 nutrition guideline, with Indian trial data. Educational, not a prescription.
What hepatic encephalopathy is: covert, overt and the West Haven grades
The liver sits between the gut and the brain. Ammonia and other nitrogen compounds produced by gut bacteria from dietary protein pass through it first, and a healthy liver clears them. In cirrhosis there are fewer working liver cells and blood is diverted around the liver through portosystemic shunts, so the toxins reach the brain. The result is a disorder of attention, sleep, personality and eventually consciousness.
Guidelines split this spectrum into two halves. Covert HE includes minimal HE — abnormal only on psychometric or neurophysiological testing — and West Haven grade I, where changes exist but a doctor meeting the patient for the first time will usually miss them. Overt HE begins at grade II, when the patient is disoriented for time or has asterixis (the flapping tremor of the outstretched hands), and runs to grade IV coma. Overt HE is present in 10–14% of people when cirrhosis is diagnosed and occurs in 30–40% at some point in the disease, with repeated bouts the norm in survivors; the quoted range for covert HE is 20–80%. Indian data sit in the middle. Two prospective series from G B Pant Hospital, New Delhi, found minimal HE in 55% of 110 and 41% of 200 cirrhotic outpatients, climbing with Child-Pugh class to 76% in Child C.
Time course matters too. An episodic bout is a single event; recurrent HE means bouts six months or less apart; persistent HE means the patient never returns fully to baseline. Each step changes the treatment conversation.
The West Haven criteria are the standard clinical grading scale. They were written for clinicians. The table keeps the guideline’s descriptions and adds the two columns families actually need.
| West Haven grade | Guideline description (AASLD/EASL 2014) | What the family sees at home | What to do |
|---|---|---|---|
| Minimal (covert) | Abnormal psychometric or neurophysiological tests without clinical evidence of mental change. | Nothing obvious. Perhaps slower at work, more careless with money or driving, “not as sharp”. Often noticed only in retrospect. | Mention it at the next hepatology visit. Ask about psychometric testing, especially if the patient drives or works with machinery. |
| Grade I (covert) | Trivial lack of awareness; euphoria or anxiety; shortened attention span; impaired addition or subtraction; altered sleep rhythm. Oriented in time and space. | Sleeps in the day, awake at night. Handwriting or signature deteriorates. Loses the thread of conversation. Irritable, or oddly cheerful. Repeats tasks (salting food twice). | Check for a trigger — constipation, missed lactulose, new sedative, fever, reduced fluid intake. Contact the treating team within a day or two. Do not let the patient drive. |
| Grade II (overt) | Lethargy or apathy; disorientation for time; obvious personality change; inappropriate behaviour; dyspraxia; asterixis. | Does not know the day or date. Hands flap when held out. Behaves out of character — rude, childlike, undressing. Slow and sleepy but rousable. | Same-day medical review. Most patients at this grade need hospital assessment to find and treat the precipitant. |
| Grade III (overt) | Somnolence to semi-stupor; responsive to stimuli; confused; gross disorientation; bizarre behaviour. Disoriented also for place. | Barely wakeable, speaks nonsense, does not know where he is or who you are. Cannot swallow medicines safely. | Emergency. Hospital now. Do not give oral medicines or food to a drowsy patient — aspiration risk. |
| Grade IV (overt) | Coma. Does not respond even to painful stimuli. | Unresponsive. | Emergency. Call an ambulance. Airway protection and intensive care are needed. |
Grade I is unreliable between observers — mild slowing is easy to overlook in a ten-minute consultation. The guideline’s own operative description of grade I is decay “with respect to his or her standard … to the caregivers”. The spouse is a better detector than the clinician, which is why we ask families to describe, not just to report.
What triggers an episode
The most useful sentence in the 2014 guideline: precipitating factors can be identified in nearly all bouts of episodic HE in cirrhosis and should be actively sought and treated (GRADE II-2, A, 1). Encephalopathy in a stable cirrhotic almost never arrives from nowhere. Something pushed.
The guideline ranks the triggers by frequency. For an episodic bout: infection, gastrointestinal bleeding, diuretic overdose, electrolyte disturbance, constipation, and a residual group where nothing is found. For recurrent HE, electrolyte disorder and infection lead. Add sedatives and sleeping tablets (benzodiazepines above all), opioid painkillers, dehydration from vomiting or diarrhoea, a recent TIPS procedure, kidney injury, and — in India — the unprescribed herbal or ayurvedic preparation.
Each has a story a family can recognise. Infection — urinary, chest, or spontaneous bacterial peritonitis in ascitic fluid — may produce confusion before it produces fever. Bleeding from oesophageal varices delivers a protein load into the gut and drops the blood pressure; black stool with confusion is a double emergency. Diuretics cause encephalopathy when the dose outruns the fluid — the patient “doing well” on a hot week with less water and the same tablets. Low sodium or potassium lowers the threshold. Constipation is the trigger patients create by stopping lactulose. Sedatives are the trigger doctors create, usually prescribed for the insomnia that was itself an early sign.
How it is diagnosed, and why ammonia is not the test
Overt HE is diagnosed clinically, in a patient with liver insufficiency or shunting, after excluding the other causes of a clouded mind (AASLD/EASL 2014, GRADE II-2, B, 1). Hypoglycaemia, alcohol withdrawal, Wernicke’s encephalopathy, a subdural haematoma after a fall, sepsis, hyponatraemia, stroke — several coexist with cirrhosis and several are more dangerous untreated than HE itself. A new confusional state earns a glucose, electrolytes, a septic screen and often a brain scan before anyone writes “HE” in the notes.
Now ammonia, asked about at almost every consultation. Recommendation 9 of the 2014 guideline reads: “Increased blood ammonia alone does not add any diagnostic, staging, or prognostic value for HE in patients with chronic liver disease. A normal value calls for diagnostic reevaluation” (GRADE II-3, A, 1). A high ammonia does not make the diagnosis, does not tell you the grade, and serial levels do not guide treatment in cirrhosis. The one useful direction is the reverse: an encephalopathic patient with a genuinely normal ammonia needs another explanation. Families arrive with ammonia reports charted like blood sugars. The patient’s orientation, sleep and handwriting tell us more.
Acute liver failure — a previously normal liver collapsing over days — is a different disease. There, very high arterial ammonia is linked to cerebral oedema and carries prognostic weight, though even in that setting my colleagues and I argued in a 2018 letter to Hepatology that a venous ammonia should not drive decisions on its own. The cirrhosis rule should not be exported to acute liver failure, and the acute liver failure anxiety should not be imported into the cirrhosis clinic.
Covert HE is diagnosed by psychometric or neurophysiological testing in experienced hands (GRADE II-2, B, 1) — the Psychometric Hepatic Encephalopathy Score, number and figure connection tests, critical flicker frequency, the inhibitory control test. Routine testing of everyone is not required; the guideline recommends it where the result would change something — quality of life, employment or public safety. A professional driver with cirrhosis belongs in that group.
Treatment: the trigger, lactulose, rifaximin — and the protein myth
The guideline’s four-pronged approach to an overt episode (GRADE II-2, A, 1): care for the patient with altered consciousness, which in grade III–IV means airway protection and intensive care; exclude other causes of confusion; identify and correct the precipitant; start empirical treatment. The order is deliberate. A patient whose encephalopathy came from peritonitis recovers because the peritonitis was treated, not because of the laxative.
Lactulose is the first choice (GRADE II-1, B, 1). A non-absorbable sugar, it acidifies the colon, traps ammonia and speeds its exit. The dose is adjusted by the treating team to produce regular soft stools, and patients get this wrong in both directions: too little, out of embarrassment or taste, and the protection is lost; too much, in the belief that more is safer, and the resulting dehydration and salt loss themselves precipitate encephalopathy. The guideline is blunt that overuse is a real danger. Lactulose is not a drug to self-titrate.
Secondary prophylaxis after recovery is recommended at the highest grade (GRADE I, A, 1), and the trial that established lactulose in this role is Indian. Sharma, Sharma, Agrawal and Sarin at G B Pant Hospital, New Delhi, randomised 140 patients who had recovered from an episode to lactulose or placebo; over a median of 14 months, overt HE recurred in 19.6% against 46.8%. A second trial from the same group, in 235 patients, showed that lactulose and a probiotic both reduced recurrence compared with no therapy.
Rifaximin, a minimally absorbed antibiotic that reduces the ammonia-producing gut flora, is recommended as an add-on to lactulose to prevent recurrence (GRADE I, A, 1). The evidence is the 2010 New England Journal of Medicine trial by Bass and colleagues: 299 patients in remission from recurrent HE, over 90% already on lactulose, randomised to rifaximin or placebo for six months. Breakthrough HE occurred in 22.1% versus 45.9% — a 58% reduction in risk — and HE-related hospitalisation fell from 22.6% to 13.6%. In practice it is added after a second episode on lactulose, or after a severe first one. It is costlier than lactulose in India; whether it earns that cost in a given patient is the hepatologist’s call, not the pharmacy counter’s.
Prophylaxis can occasionally stop (GRADE III, C, 2) when the precipitant has been durably controlled — a treated infection, banded varices — or when liver function has clearly improved with sustained abstinence or antiviral therapy. The reverse is far commoner: patients stop on their own, and relapse.
Now the protein myth, which does more harm in India than any drug error I see. For decades patients with encephalopathy were told to cut protein; dal, milk, eggs and paneer were struck off; and the patient, already wasted, lost more muscle. The 2014 guideline calls chronic protein restriction detrimental, notes that roughly 75% of patients with HE already have moderate-to-severe protein-calorie malnutrition, and recommends 1.2–1.5 g/kg/day of protein and 35–40 kcal/kg/day of energy (GRADE I, A, 1). The EASL 2019 nutrition guideline says it in four words: “Avoid protein restriction in patients with HE.” Normal-to-high protein intake does not precipitate encephalopathy and may improve mental state.
Why does the myth hurt? Muscle is the body’s second ammonia-clearing organ; lose it and the brain is more exposed, not less. Starvation — including the overnight fast — breaks down the body’s own protein and releases ammonia from within. So: small frequent meals, protein at breakfast and in a late-evening snack, vegetable and dairy protein where tolerated, and branched-chain amino acid supplements for those who cannot reach the target otherwise. The full argument is in my post on nutrition in cirrhosis.
Living with hepatic encephalopathy: driving, safety and the family’s job
The hardest conversation in the HE clinic is about the car. Patients with minimal HE have no insight into their impairment — Bajaj and colleagues showed in 2008 that they rated their own driving as normal while family observers and a driving simulator disagreed. The same group then checked 167 patients with cirrhosis against official transport-department records: those with minimal HE on the inhibitory control test had a 17% crash rate in the preceding year against 3% in those without, a relative risk of 5.8, and more crashes and violations on prospective follow-up. In a 2011 randomised trial, rifaximin improved simulator performance in minimal HE — 76% of treated patients improved versus 31% on placebo — proof the deficit is real, and treatable.
The 2014 guideline says doctors “cannot evade the responsibility” of counselling patients with HE about driving, and that the safest advice is often to stop until formally cleared. Mine is simpler. After an episode of overt HE, do not drive until your hepatologist says otherwise. If you have cirrhosis and your family thinks you drive badly, they are probably right — ask to be tested. Mumbai traffic forgives nothing.
Beyond driving, the family’s job is detection and prevention. Keep a baseline — a signature on paper every month, a note of the usual sleep pattern. Watch for day-night reversal, changed handwriting, slowness, inappropriate laughter, a drop in performance at work. Keep the bowels moving and the lactulose going, and call before stopping it. Hide the sleeping tablets. Know the warning signs of worsening liver disease and the three that mean hospital now: black or bloody stool or vomit, a drowsy patient who cannot be fully woken, and fever with confusion. And understand that the patient at grade II is not being difficult. The stubbornness about the medicines is the disease.
When encephalopathy means talking about transplant
Overt HE is a decompensating event. The 2014 guideline places it alongside variceal bleeding and ascites as the marker that cirrhosis has crossed into its decompensated phase, where the time horizon shortens. Its recommendation is that recurrent intractable overt HE together with liver failure is an indication for liver transplantation (GRADE I). HE by itself, with preserved liver function, is not usually an indication; such a patient should first be checked for a large portosystemic shunt that might be closed, and for a reversible cause of liver injury.
The MELD score, which governs transplant priority, contains no term for encephalopathy. A patient with recurrent HE and a MELD of 14 can have a worse life and a higher risk of a fatal admission than one with a MELD of 20 and a clear head. And the guideline notes that repeated bouts of overt HE may leave cumulative cognitive deficits that do not fully reverse, even after transplantation. Recurrent HE is a reason to bring the conversation forward, not to wait for the number to catch up.
A second overt episode within six months, an episode that needed intensive care, encephalopathy alongside refractory ascites or a bleed, or a life that has shrunk around the condition — any of these should prompt formal evaluation at a transplant centre. Living-donor transplantation, the usual route in India, means the timing can be planned rather than waited for. Across more than 1,000 liver transplants I have been part of — around 300 of them complex — with 92% survival, the patients I remember most are the ones referred after the third admission, when a year earlier the family had been told it was “just confusion”. My transplant preparation checklist explains what an evaluation involves.
Frequently asked questions
Is hepatic encephalopathy reversible?
Usually, yes. An episode precipitated by infection, bleeding, constipation, dehydration or a sedative will generally reverse once the trigger is corrected and treatment with lactulose is started. Repeated severe episodes may leave some residual cognitive deficit, which is one reason to prevent recurrence and to consider transplant evaluation early in recurrent disease.
What is the difference between covert and overt hepatic encephalopathy?
Covert HE includes minimal HE, detectable only on psychometric or neurophysiological tests, and West Haven grade I, where changes in sleep, attention and personality are present but subtle. Overt HE begins at grade II, when the patient is disoriented for time or has asterixis, and extends to grade IV coma. Covert HE is far commoner and affects driving and work performance.
Does a high ammonia level mean hepatic encephalopathy?
Not on its own. The AASLD/EASL 2014 guideline states that a raised blood ammonia alone adds no diagnostic, staging or prognostic value in chronic liver disease. The diagnosis is clinical. A normal ammonia in a confused patient should prompt a search for another cause. Acute liver failure is a different setting, where very high ammonia is linked to brain swelling.
Should a patient with hepatic encephalopathy eat less protein?
No. Protein restriction is harmful. The EASL 2019 nutrition guideline says to avoid protein restriction in HE and recommends 1.2–1.5 g/kg/day, split into small frequent meals with protein at breakfast and a late-evening snack. Muscle helps clear ammonia; losing it makes encephalopathy worse.
Can I drive if I have had hepatic encephalopathy?
After an overt episode, not until your hepatologist clears you. Even minimal HE raised the risk of a road crash more than fivefold in the best study, and affected patients cannot judge their own impairment. If you have cirrhosis and drive for a living, ask for psychometric testing.
When does hepatic encephalopathy mean I need a liver transplant?
A single episode with a clear, correctable trigger does not. Recurrent intractable HE together with failing liver function is a recognised indication. A second episode within six months, an episode needing intensive care, or HE alongside refractory ascites or bleeding should prompt a formal transplant evaluation, because the MELD score does not capture encephalopathy and waiting for the number to rise can mean waiting too long.
Book a consultation with Dr Chetan Kalal at Gleneagles Hospital, Mumbai — call +91 83294 76669 or use the appointment form on this site. International and NRI patients may request a teleconsultation.
References
- Vilstrup H, Amodio P, Bajaj J, et al. Hepatic encephalopathy in chronic liver disease: 2014 Practice Guideline by the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver. Hepatology. 2014;60(2):715–735. PMID 25042402. doi:10.1002/hep.27210
- European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of hepatic encephalopathy. J Hepatol. 2022;77(3):807–824. PMID 35724930. doi:10.1016/j.jhep.2022.06.001
- European Association for the Study of the Liver. EASL Clinical Practice Guidelines on nutrition in chronic liver disease. J Hepatol. 2019;70(1):172–193. PMID 30144956. doi:10.1016/j.jhep.2018.06.024
- Bass NM, Mullen KD, Sanyal A, et al. Rifaximin treatment in hepatic encephalopathy. N Engl J Med. 2010;362(12):1071–1081. PMID 20335583. doi:10.1056/NEJMoa0907893
- Sharma BC, Sharma P, Agrawal A, Sarin SK. Secondary prophylaxis of hepatic encephalopathy: an open-label randomized controlled trial of lactulose versus placebo. Gastroenterology. 2009;137(3):885–891. PMID 19501587. doi:10.1053/j.gastro.2009.05.056
- Agrawal A, Sharma BC, Sharma P, Sarin SK. Secondary prophylaxis of hepatic encephalopathy in cirrhosis: an open-label, randomized controlled trial of lactulose, probiotics, and no therapy. Am J Gastroenterol. 2012;107(7):1043–1050. PMID 22710579. doi:10.1038/ajg.2012.113
- Sharma P, Sharma BC, Sarin SK. Predictors of nonresponse to lactulose for minimal hepatic encephalopathy in patients with cirrhosis. Liver Int. 2009;29(9):1365–1371. PMID 19555401. doi:10.1111/j.1478-3231.2009.02067.x
- Sharma P, Sharma BC. Predictors of minimal hepatic encephalopathy in patients with cirrhosis. Saudi J Gastroenterol. 2010;16(3):181–187. PMID 20616413. doi:10.4103/1319-3767.65189
- Bajaj JS, Saeian K, Schubert CM, et al. Minimal hepatic encephalopathy is associated with motor vehicle crashes: the reality beyond the driving test. Hepatology. 2009;50(4):1175–1183. PMID 19670416. doi:10.1002/hep.23128
- Bajaj JS, Saeian K, Hafeezullah M, Hoffmann RG, Hammeke TA. Patients with minimal hepatic encephalopathy have poor insight into their driving skills. Clin Gastroenterol Hepatol. 2008;6(10):1135–1139. PMID 18928938. doi:10.1016/j.cgh.2008.05.025
- Bajaj JS, Heuman DM, Wade JB, et al. Rifaximin improves driving simulator performance in a randomized trial of patients with minimal hepatic encephalopathy. Gastroenterology. 2011;140(2):478–487.e1. PMID 20849805. doi:10.1053/j.gastro.2010.08.061
- Kalal C, Shukla A, Mohanka R, Shah S. Should venous ammonia be used in decision-making acute liver failure patients? Hepatology. 2018;67(2):800–801. PMID 29108118. doi:10.1002/hep.29641
