STRUCTURED ABSTRACT

Question: What is the optimal approach to diagnose, manage, and prevent hepatic encephalopathy (HE) in patients with cirrhosis?

Design: A clinical practice guideline developed under the ACG Practice Parameters Committee. The chosen expert writing group conducted a systematic literature review in PubMed, Embase, and the Cochrane Library, applying the GRADE framework to assess evidence quality and formulate recommendations based on a PICO-based approach.

Setting: Multidisciplinary guideline development intended for use across all clinical settings (outpatient clinics, community hospitals, and transplant centers) where adults with cirrhosis and HE are evaluated or managed.

Patients: Adults with cirrhosis presenting with any stage of hepatic encephalopathy, from covert (CHE/MHE) to overt (OHE) (West Haven Grades 2–4), including those undergoing TIPS, as well as liver transplant candidates.

Definitions:  Minimal Hepatic Encephalopathy (MHE) is the subclinical form of HE that can only be identified through specialized neuropsychological testing; it is not detectable on standard clinical examination.

Covert Hepatic Encephalopathy (CME) is a broader term that combines MHE with West Haven grade 1 HE, distinguishing these subclinical clinical forms from overt HE (grades 2–4). Since both MHE and grade 1 HE are clinically difficult to differentiate, MHE and CHE are generally used interchangeably in practice.

Overt Hepatic Encephalopathy (OHE) corresponds to West Haven grades 2–4; forms of HE that can be diagnosed on standard clinical examination without specialized neuropsychological testing.

Interventions: Key recommendations include

• Single-test strategy preferred for CHE/MHE diagnosis; serum ammonia alone is insufficient.
• Lactulose as first-line therapy for OHE (strong recommendation); rifaximin added for recurrence prevention.
• Protein intake of 1.2–1.5 g/kg/day; branched chain amino acid (BCAA) supplementation if dietary needs unmet; late-evening snack to reduce frailty.
• Initiate rifaximin 14 days before elective transjugular intrahepatic portosystemic shunt (TIPS) and continue for ≥6 months to prevent post-TIPS HE.
• Consider living donor liver transplantation evaluation for patients with multiple HE episodes and MELD <15

Outcomes: The primary outcomes assessed were: HE recurrence, quality of life (QoL), HE-related hospitalizations and readmissions, cognitive function, caregiver burden, mortality, and liver transplant access and outcomes.

Data Analysis: The GRADE methodology was used, rating evidence certainty as high, moderate, low, or very low based on risk of bias, heterogeneity, and directness of evidence. Many recommendations rest on low or very low certainty evidence given the paucity of high-quality randomized controlled trials (RCTs) in this population. Strong recommendations were issued when benefits clearly outweighed harms. Key concepts not amenable to GRADE were provided as expert consensus statements.

Funding: Developed with ACG support. No external or industry funding was utilized.

Results: The guideline provides 24 formal recommendations and 29 key concepts, summarized in Table 1. Highlights are presented below.

Diagnosis of CHE/MHE. CHE is identified in 20-80% of patients with cirrhosis but is tested in fewer than 15% of centers. Given the low reproducibility of its clinical diagnosis, CHE should be defined by excluding OHE in the appropriate clinical context. For OHE exclusion, a single-test strategy is preferred over a 2-test combination, given that dual testing lowers sensitivity without improving predictive accuracy. Serum ammonia alone is insufficient for diagnosis. Priority populations for testing include those with hypoalbuminemia, decompensated cirrhosis, portosystemic shunts, cognitive complaints, falls, or occupational impairment. Alternative diagnoses (including obstructive sleep apnea, mild cognitive impairment, and mood disorders) should be excluded, as over 50% of referred patients with cognitive complaints do not have MHE.

Inpatient Management of OHE. The decision to admit a patient for OHE should be individualized based on key clinical domains (Figure 1). Clinical severity should be graded using West Haven criteria with Glasgow Coma Scale for high-grade HE.  A thorough evaluation for bleeding, infection, sedating medications, and metabolic abnormalities is necessary to identify possible precipitating factors. Serum ammonia should not guide treatment; importantly, a normal ammonia level should prompt consideration of alternative diagnoses in addition to OHE. Brain imaging is not recommended routinely in cirrhosis with confusion unless new focal deficits or seizures are present. Lactulose is the cornerstone of treatment, with a target of 2-3 soft bowel movements daily; PEG 3350 is an acceptable alternative for patients who are intolerant to lactulose. Adding rifaximin to lactulose for acute OHE provides an incremental benefit. If there is marginal improvement in mental status after 48-72 hours of adequate therapy and reversal of potential precipitating factors, evaluation for alternative causes or portosystemic shunts should be considered.

Prevention of Recurrence. After a first episode of OHE, lactulose is preferred first-line therapy. The Bristol Stool Scale should supplement bowel movement frequency for titration. Rifaximin is recommended as add-on therapy for breakthrough episodes on lactulose and suggested for primary outpatient prophylaxis. Zinc supplementation is suggested for those with documented deficiency on dual therapy. Spontaneous portosystemic shunts (SPSS) should be investigated in refractory cases and considered for embolization in appropriate candidates (MELD <15). A thorough medication review should identify and minimize opiates, benzodiazepines, gabapentin, sleep medications, and proton pump inhibitors, as these increase risk for developing HE. Digital tools (Patient Buddy App, electronic reminders) can support adherence and reduce readmissions. Oral and written advice to avoid driving should be given to patients with recent OHE (less than 3 months), and all patients and caregivers should receive HE educational materials; caregivers should also be screened for burnout.

Nutrition and Sarcopenia. Protein restriction is explicitly discouraged as it worsens muscle breakdown without reducing HE duration. Protein intake of 1.2-1.5 g/kg/day is recommended, with BCAA supplementation when dietary goals cannot be met. A late-evening snack reduces frailty and HE. All patients with HE should be evaluated for impaired muscle health using available tools such as sarcopenia assessments, frailty indices, or cardiorespiratory fitness measures. Exercise is recommended to lower portal pressure, reduce fall risk, and enhance skeletal muscle ammonia metabolism. Exercise-induced increase in ammonia has not been associated with elevated HE risk and should not discourage exercise.

TIPS and Liver Transplant. Prophylactic rifaximin should begin 14 days before elective TIPS and continue for at least 6 months post-TIPS. For patients with multiple HE episodes and MELD less than 15, living donor liver transplantation evaluation is suggested. Multiple HE episodes lead to cumulative cognitive deficits that may not resolve post-transplant, underscoring the need for early referral. Early transplant evaluation should be considered in patients with high MELD 3.0 scores, severe HE grades III to IV, or frequent episodes. In certain circumstances, persistent encephalopathy in admitted patients is a relative contraindication to transplant.

Figure 1. Clinical framework for triage of patients with suspected hepatic encephalopathy.

 

Domain	Recommendation	Evidence
CHE /MHE
Diagnosis	Single-test strategy preferred over 2-test combination	Conditional, very low certainty
Ammonia	Suggest against serum ammonia alone for CHE/MHE diagnosis	Conditional, very low certainty
Treatment	Lactulose suggested over no treatment for MHE/CHE	Conditional, low certainty
Inpatient Management of OHE
Ammonia monitoring	Routine testing not recommended to guide inpatient HE decisions	Conditional, very low certainty
Brain imaging	Not recommended if new focal neurologic deficits are not present	Conditional, very low certainty
Lactulose	Recommended for OHE treatment to improve outcomes	Strong, moderate certainty
PEG 3350	Alternative to lactulose for acute OHE	Conditional, low certainty
Rifaximin (add-on)	Adding rifaximin to lactulose in acute OHE is suggested	Conditional, low certainty
Prevention of HE Recurrence
Lactulose	First-line outpatient prophylaxis, titrated to 2–3 soft BMs/day	Strong, high certainty
Bristol Stool Scale	Recommended for lactulose titration to reduce readmissions	Conditional, very low certainty
Rifaximin	Suggested for outpatient prophylaxis	Conditional, low certainty
Rifaximin (add-on)	Recommended for patients with OHE breakthrough on lactulose	Strong, high certainty
Zinc (add-on)	Suggested as add-on in those with documented zinc deficiency	Conditional, very low certainty
Digital tools	Health IT interventions (Patient Buddy App, e-reminders) suggested when feasible	Conditional, very low certainty
Shunt embolization	Suggested for refractory HE in patients with adequate hepatic function and no contraindications	Conditional, very low certainty
Nutrition and Sarcopenia
Protein intake	Target 1.2–1.5 g/kg/day in outpatients with HE	Strong, moderate certainty
BCAA	Recommended if protein needs unmet by diet alone	Strong, moderate certainty
Late-evening snack	Suggested to reduce frailty and HE	Conditional, very low certainty
Protein restriction	Discouraged; worsens muscle breakdown without reducing HE duration	Conditional, very low certainty
Exercise	Suggested to reduce fall risk, lower portal pressure, and improve ammonia metabolism	Conditional, low certainty
TIPS and HE
Rifaximin pre-TIPS	Initiate 14 days before elective TIPS; continue ≥6 months	Strong, moderate certainty
Collateral embolization	Suggested at time of TIPS to reduce post-TIPS HE	Conditional, low certainty
Liver Transplant and HE
LDLT evaluation	Suggested for patients with multiple HE episodes and MELD <15	Conditional, low certainty

Table 1. Summary of Guideline Recommendations.

• Serum ammonia, once central to clinical decision-making, is now firmly de-emphasized. In a study of 551 HE patients, 40% had normal ammonia levels; venous ammonia does not correlate with HE severity or prognosis and should not guide treatment decisions in any setting.
• Protein restriction, still practiced in some centers, is explicitly discouraged. Adequate protein intake (1.2-1.5 g/kg/day), BCAA supplementation when dietary goals are unmet, and late-evening snacks, are essential to counter sarcopenia and improve outcomes.
• SPSS affect 46% to 71% of patients with refractory HE and represent a remediable cause often overlooked in clinical practice. Cross-sectional imaging should be performed in all patients failing standard therapy, with embolization considered for shunts 8 mm or greater in diameter in patients with preserved hepatic function.
• Rifaximin therapy pre and post-TIPS now carries a strong recommendation, supported by a dedicated RCT showing significant reduction in post-TIPS HE when rifaximin is initiated pre-TIPS and continued for 6 months. This is an immediately actionable change for hepatologists managing TIPS candidates.
• Liver transplant prioritization for HE remains inadequate under the current MELD 3.0 system. Adding 4 to 5 MELD points more accurately reflects 90-day mortality from OHE⁵; multiple HE episodes cause cumulative, potentially irreversible cognitive deficits that may not resolve even after successful transplantation, making early evaluation essential.⁶

Social Media
Follow the authors on X:
@leandrosierraca – Leandro Sierra
@doctornikkid – Nikki Duong