Visual Abstract

INTRODUCTION
Hepatic encephalopathy (HE) is one of the most consequential complications of cirrhosis, affecting patients, caregivers, and health systems through recurrent hospitalizations, impaired cognition, falls, and reduced quality of life. 1 It is now the most common first decompensating event in cirrhosis and is the leading cause of readmission among patients hospitalized for complications of cirrhosis, with 90‐day readmission rates of 21%–53%. 2 , 3
HE encompasses a spectrum of neuropsychiatric abnormalities ranging from subclinical cognitive deficits (covert HE) to overt disorientation, asterixis, and coma. The 2026 American College of Gastroenterology (ACG) guideline presents 24 recommendations spanning covert HE/minimal HE (CHE/MHE), inpatient overt HE (OHE), recurrence prevention, nutrition and frailty, TIPS‐related HE, and transplant considerations. 1
For hospitalists, the most practice‐changing issues are diagnostic evaluation of altered mental status in patients with cirrhosis, selective use of neuroimaging, first‐line and adjunctive inpatient therapy, prophylaxis and strategies for preventing readmission, and nutritional management (Table 1).
Table 1.
Guideline summary.
| Guideline title | ACG Clinical Guideline: Hepatic Encephalopathy |
| Release date | March 3, 2026 |
| Prior version(s) | 2014 AASLD/EASL practice guideline; 2022 EASL clinical practice guideline |
| Developer | American College of Gastroenterology |
| Funding source | American College of Gastroenterology; no commercial support |
| Target population | Adults with cirrhosis across the spectrum of covert, minimal, and overt hepatic encephalopathy |
KEY RECOMMENDATIONS FOR THE HOSPITALIST
Recommendation 1: Do not use serum ammonia levels to diagnose overt HE or guide inpatient treatment decisions.
The guideline recommends against routine serum ammonia testing to diagnose HE or guide treatment decisions, because ammonia levels do not reliably correlate with HE severity and are affected by sample handling, processing, and dietary protein intake. In a patient with known cirrhosis and clinical features consistent with prior episodes of overt HE, ammonia testing is not required to confirm the diagnosis; HE remains a clinical diagnosis based on history, examination, exclusion of alternative causes of altered mental status, and evaluation for precipitants.
When obtained, a normal ammonia value should prompt evaluation for alternative causes of altered mentation.
Recommendation 2: Do not obtain routine brain imaging in patients with cirrhosis and confusion without new‐onset focal neurologic deficits.
Routine brain imaging is not recommended to diagnose HE. CT or MRI is indicated to evaluate alternative conditions such as intracranial hemorrhage or stroke in patients with new focal findings, seizures, or lack of response to primary HE treatment.
Recommendation 3: Treat hospitalized acute overt HE with lactulose and early adjunctive rifaximin; consider PEG 3350 as an alternative strategy.
Lactulose is recommended as first‐line therapy for OHE and is typically titrated to 2–4 soft bowel movements daily. The guideline suggests adding rifaximin to lactulose for acute episodes, supported by a randomized trial showing higher rates of complete HE reversal, shorter hospital stays, and reduced mortality compared with lactulose alone. Careful titration is essential because lactulose overuse can cause dehydration, hypernatremia, and hypokalemia, which may themselves worsen or precipitate HE.
Patients with grade 3–4 HE who cannot safely swallow may require nasogastric administration or enemas. PEG 3350 is suggested as an alternative cathartic strategy, particularly for patients intolerant of lactulose or when rapid catharsis is desired, based on one RCT demonstrating faster HE resolution. 1 , 4 , 5
Recommendation 4: Conduct a thorough precipitant evaluation and reassess patients who fail to improve within 48–72 h.
Evaluate for common precipitating factors such as infection, gastrointestinal bleeding, electrolyte disorders, constipation, and dehydration. Because infection is a common precipitant of HE, the guideline recommends obtaining blood, urine, and ascitic fluid cultures in all patients, with stool and sputum studies when clinically indicated. Medication review is essential: opioids, benzodiazepines, gabapentinoids, and unnecessary proton pump inhibitors increase HE risk and should be deprescribed when feasible.
Improvement in mental status is expected within 24–48 h. Lack of improvement after 48–72 h despite adequate bowel movements and precipitant management should prompt reassessment for alternative diagnoses, undertreated precipitants, or spontaneous portosystemic shunts. Hepatology and interventional radiology consultations should be obtained if imaging identifies a potentially embolizable shunt.
Recommendation 5: Prescribe lactulose for secondary prophylaxis after the first episode of OHE, and add rifaximin for recurrent episodes.
After an initial episode of OHE, the guideline recommends lactulose titrated to 2–3 soft bowel movements daily as first‐line outpatient therapy for prevention of HE recurrence. For patients on lactulose maintenance who experience recurrent episodes, the addition of rifaximin is strongly recommended.
Recommendation 6: Avoid protein restriction; target 1.2–1.5 g/kg/day protein intake and consider BCAA supplementation.
The guideline recommends a protein intake of 1.2–1.5 g/kg/day and explicitly discourages protein restriction, which increases muscle breakdown without reducing HE duration. Branched‐chain amino acid (BCAA) supplementation is recommended when dietary intake is insufficient, and a late‐evening snack is suggested to reduce the accelerated starvation seen in cirrhosis.
Recommendation 7: Counsel patients with recent OHE episodes against driving.
Oral and written advice to avoid driving should be provided to patients and caregivers after recent overt HE, particularly within 3 months of an episode. Impaired driving is an underrecognized functional consequence of HE, and discharge counseling is directly relevant to hospitalist care transitions.
Recommendation 8: Consider zinc supplementation in patients with persistent HE and low serum zinc.
Zinc supplementation may be added in patients with persistent HE despite lactulose and rifaximin therapy who have low blood zinc levels. Zinc plays a cofactor role in nitrogen metabolism, and the intervention is low‐cost, low‐risk, and may improve cognitive function.
Recommendation 9: Initiate rifaximin prophylaxis 14 days before elective TIPS placement.
For patients with decompensated cirrhosis undergoing elective TIPS insertion, the guideline recommends initiating rifaximin therapy 14 days before the procedure and continuing for at least 6 months to decrease the risk of recurrent or de novo OHE.
CRITIQUE
This guideline has several notable strengths. The writing group includes hepatologists, gastroenterologists, and GRADE methodology experts from nine academic institutions, and the GRADE framework was applied to each recommendation. The addition of social infrastructure—defined in the guideline as family and social support, financial resources for medication adherence, and institutional support for follow‐up care—as a fifth axis in the HE classification framework, alongside type, grade, time course, and precipitants, is a meaningful conceptual advance. This is particularly relevant for hospitalists because readmissions for HE are often driven by social and care‐transition barriers, including medication access, caregiver support, and follow‐up reliability, rather than by biomedical factors alone.
However, several limitations should be noted. Many recommendations are supported by low or very low certainty of evidence. The panel did not include hospitalists, emergency physicians, or critical care physicians, even though these clinicians frequently manage acute OHE in practice. Certain recommendations regarding TIPS management and shunt embolization for refractory cases assume subspecialty access that may not be available at community or rural hospitals. The cost of rifaximin also remains a significant barrier.
Although the source guideline presents lactulose‐based therapy and adjunctive rifaximin as separate formal recommendations, they are combined here because hospitalists typically make these treatment decisions simultaneously at the time of admission for acute OHE. This framing better reflects bedside inpatient practice while preserving the distinction that lactulose remains first‐line and rifaximin is adjunctive.
An area of potential ambiguity is the recurrence‐prevention recommendations. The guideline clearly supports lactulose after a first OHE episode and rifaximin addition for recurrence despite lactulose, but one recommendation could be interpreted as endorsing rifaximin more broadly without clearly specifying whether it is add‐on therapy, an alternative for lactulose intolerance, or use before breakthrough recurrence. Notably, the guideline's narrative discussion acknowledges the superior efficacy of combination therapy, yet the formal recommendations do not clearly define whether upfront combination therapy should have a role after a first OHE episode. For hospitalists, a more explicit stepwise outpatient algorithm would improve bedside implementation and discharge planning.
AREAS IN NEED OF FUTURE STUDY
Randomized trials evaluating fecal microbiota transplant as a strategy for secondary prevention of HE recurrence and readmission.
Comparative effectiveness of electronic clinical decision support tools, best practice alerts, and patient‐facing health technology interventions (e.g., the Patient Buddy App) in reducing HE‐related readmissions.
Development and validation of reproducible inpatient HE diagnostic criteria that could improve transplant prioritization and organ allocation.
CONFLICT OF INTEREST STATEMENT
The author declares no conflicts of interest.
DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. Full guidelines in The American Journal of Gastroenterology: https://doi.org/10.14309/ajg.0000000000003899. Supplementary Digital Content: http://links.lww.com/AJG/D846. Patient/caregiver HE educational materials: https://cirrhosiscare.ca/confusion/.
REFERENCES
- 1. Bajaj JS, Jakab SS, Jesudian AB, et al. ACG clinical guideline: hepatic encephalopathy. Off J Am Coll Gastroenterol ACG. 2026;121(3):588‐618. 10.14309/ajg.0000000000003899 [DOI] [PubMed] [Google Scholar]
- 2. Bajaj JS, Reddy KR, Tandon P, et al. The 3‐month readmission rate remains unacceptably high in a large North American cohort of patients with cirrhosis. Hepatology. 2016;64(1):200. 10.1002/hep.28414 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Tapper EB, Halbert B, Mellinger J. Rates of and reasons for hospital readmissions in patients with cirrhosis: a multistate population‐based cohort study. Clin Gastroenterol Hepatol. 2016;14(8):1181‐1188.e2. 10.1016/j.cgh.2016.04.009 [DOI] [PubMed] [Google Scholar]
- 4. Rahimi RS, Singal AG, Cuthbert JA, Rockey DC. Lactulose vs polyethylene glycol 3350‐electrolyte solution for treatment of overt hepatic encephalopathy: the HELP randomized clinical trial. JAMA Intern Med. 2014;174(11):1727‐1733. 10.1001/jamainternmed.2014.4746 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Sharma BC, Sharma P, Lunia MK, Srivastava S, Goyal R, Sarin SK. A randomized, double‐blind, controlled trial comparing rifaximin plus lactulose with lactulose alone in treatment of overt hepatic encephalopathy. Off J Am Coll Gastroenterol ACG. 2013;108(9):1458. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. Full guidelines in The American Journal of Gastroenterology: https://doi.org/10.14309/ajg.0000000000003899. Supplementary Digital Content: http://links.lww.com/AJG/D846. Patient/caregiver HE educational materials: https://cirrhosiscare.ca/confusion/.
