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

This is not a single experiment but a **clinical practice guideline** — a formal, graded synthesis of the world literature on hepatic encephalopathy (HE) up to 2014. It covers:

How to define, classify, and diagnose HE (including its covert/minimal forms)

Which treatments reduce HE episodes (e.g. lactulose, rifaximin, dietary protein management, zinc supplementation)

How to prevent recurrence of overt HE bouts

How to manage HE in the context of liver transplantation

What outcome measures and diagnostic tools are most valid

Comparators discussed include: lactulose vs. placebo, rifaximin vs. placebo or lactulose, dietary protein restriction vs. normal protein intake, branched-chain amino acids, zinc, probiotics, and portosystemic shunt modification.

This guideline synthesises data from **adults with chronic liver disease (CLD)**, primarily cirrhosis, across a wide range of disease severity. It explicitly excludes:

Acute liver failure (covered in separate guidelines)

Paediatric populations

Studies published more than ~30 years before 2014 (due to changed management of cirrhosis complications)

No single sample size applies — the guideline draws from dozens of randomised trials and observational studies, many of which are noted to have **small patient numbers and inconsistent methodology**, which is itself a key finding of the document.

The guideline discusses and grades multiple measurement approaches:

**West Haven Criteria (WHC):** The most widely used clinical grading scale for overt HE, rating severity from Grade 0 (no impairment) to Grade IV (coma). Criticised for subjectivity and poor reliability at the lower grades.

**Minimal/Covert HE detection:** Psychometric tools including the Psychometric Hepatic Encephalopathy Score (PHES), Critical Flicker Frequency (CFF), and the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). No single test is universally endorsed.

**Number of HE episodes / time to first episode:** Used in most prevention trials as the primary outcome.

**Health-related quality of life (HRQoL):** Flagged as critically important but inconsistently measured across trials.

**Survival:** Noted as poor once overt HE develops; used in some trials as a secondary outcome.

**Ammonia levels:** Measured in blood but explicitly noted to be an **unreliable diagnostic criterion** — elevated ammonia does not confirm HE, and normal ammonia does not rule it out.

Evidence quality is graded using the **GRADE system**: Level I–III (source), quality A/B/C (high/moderate/low), and strength 1/2 (strong/weak).

This is a **systematic practice guideline**, not a primary RCT.

**Design:**

Formal literature review by an international expert panel (AASLD + EASL)

Evidence graded using GRADE methodology

Recommendations classified by evidence quality (A/B/C) and recommendation strength (1 = strong, 2 = weak)

**Strengths of this design:**

Synthesises evidence across many trials rather than relying on a single study

GRADE methodology explicitly signals where evidence is weak, helping clinicians and researchers identify gaps

Produced by two major independent professional bodies, reducing single-institution bias

**What this design cannot prove:**

It cannot establish new causal evidence — it can only grade what already exists

Where high-quality RCTs are absent (which is frequent in this field), the guideline relies on expert consensus, which is inherently subject to opinion bias

The guideline cannot resolve contradictions between individual studies; it can only flag that contradiction exists

**Major methodological weaknesses acknowledged:**

The underlying evidence base is described as having **limited high-quality studies**. Many trials are small, use inconsistent diagnostic criteria, and are of short duration (most less than 6 months)

No universally accepted diagnostic standard for minimal/covert HE at time of publication

Heterogeneity of patient populations across studies makes pooling results unreliable

Industry funding of individual underlying trials is not comprehensively disclosed within the guideline text

**Diagnosis and classification:**

HE should be classified by: (1) underlying disease type, (2) severity (covert vs. overt, WHC grade), (3) time course (episodic, recurrent, persistent), and (4) presence or absence of precipitating factors

Covert (minimal) HE is present in an estimated **30–84% of cirrhotic patients** depending on the tests used and population studied

Overt HE develops in approximately **30–40% of cirrhotic patients** at some point during their disease course

Ammonia measurement alone is **not sufficient** for diagnosis (low/moderate quality evidence)

**Treatment — Acute/Overt HE:**

**Lactulose** is recommended as first-line treatment for overt HE (strong recommendation, moderate-quality evidence)

**Rifaximin** is an effective alternative or add-on to lactulose (strong recommendation, moderate-quality evidence)

**Dietary protein restriction is not recommended**: normal protein intake (~1.2–1.5 g/kg/day) is advised; severe restriction is harmful

**Zinc supplementation** may be beneficial but evidence is limited (weak recommendation, low-quality evidence)

**Branched-chain amino acids (BCAAs)** can be used as an alternative or supplement in patients intolerant of standard protein sources (weak recommendation, moderate-quality evidence)

**Probiotics** show some evidence of benefit but data are insufficient for a strong recommendation

**Prevention of recurrence:**

**Rifaximin 550 mg twice daily** was shown in one pivotal RCT to reduce risk of HE recurrence vs. placebo over 6 months: HE recurrence occurred in **22.1% of rifaximin patients vs. 45.9% of placebo patients** (hazard ratio 0.42, 95% CI 0.28–0.64; p<0.001), with the majority of patients also on lactulose

**Lactulose** is also recommended for secondary prevention (strong recommendation, moderate-quality evidence)

**Portosystemic shunt occlusion** may be considered in selected patients where shunting is driving recurrent HE (weak recommendation, low-quality evidence)

**Liver transplantation:**

Transplantation is the only intervention that treats the underlying cause; HE alone is not a formal indication but recurrent/severe HE substantially affects transplant listing decisions

The rifaximin recurrence prevention finding is the most numerically robust result in the guideline: the absolute risk reduction was approximately **24 percentage points** over 6 months (45.9% → 22.1%), meaning roughly 1 in 4 patients was spared an HE episode compared to placebo. This is a clinically large effect.

Lactulose vs. placebo for acute HE: various trials show reduction in overt episode duration and improved cognitive test scores, but effect sizes are inconsistently reported across studies.

Covert HE treatment (with lactulose or rifaximin) shows improvements in psychometric test scores, but the real-world clinical significance of these improvements is not well quantified.

Overall, the guideline is candid that **most effect sizes in the HE literature are uncertain** due to small trials and heterogeneous populations.

**Evidence base is weak overall:** The authors explicitly state that high-quality evidence is scarce, and many recommendations rest on moderate or low-quality evidence

**No universally accepted diagnostic standard** for covert HE makes it impossible to compare results across studies

**Short trial durations:** Most studies ran for fewer than 6 months, making long-term efficacy and safety conclusions unreliable

**Small sample sizes** in most underlying trials reduce statistical power and increase risk of false-positive findings

**Self-report and caregiver-report bias:** HE symptoms (personality change, sleep disturbance, cognitive slowing) are partially subjective

**Population generalisability:** Many trials