Model for end-stage liver disease (MELD) exception for hepatic encephalopathy


Hepatic encephalopathy (HE), a common complication of cirrhosis and end-stage liver disease, has been shown by univariate and multivariate analyses to be a predictor of death in patients with liver failure.1, 2 The Model for End-Stage Liver Disease (MELD) scoring system does not prioritize patients with HE.3 The original multivariate analysis leading to the institution of the MELD did not demonstrate that HE increased the model's ability to predict 3-month pretransplant mortality.4 Recently, however, data have begun to emerge that point to HE as a predictive factor for short-term mortality and that suggest that the inclusion of HE would improve the predictive ability of the MELD score. An analysis of a model by the Scientific Registry of Transplant Recipients that included HE was strikingly better than MELD at predicting short-term mortality.2

Said and colleagues5 recently identified HE as an independent predictor of mortality, and others agree that HE provides additional prognostic information.2, 6, 7 Despite this, HE remains a subjective diagnosis, and an easily quantifiable and verifiable method for scoring patients is lacking.

Recurrent or poorly responsive HE often has an underlying pathophysiology that may be controlled in the setting of otherwise relatively good liver function, such as large spontaneous portosystemic shunts, a proportion of which can be closed; recurrent dehydration, which can be managed by rehydration; sepsis, which can be treated by identifying the underlying cause; and dietary noncompliance, which can be corrected. Cases that are submitted to Regional Review Boards for additional priority because of failure to respond to maximal medical therapy should include clear documentation of patient medication compliance with neomycin, lactulose, sodium benzoate or Buphenyl, rifaximin, and metronidazole.

Other conditions for which Regional Review Boards might consider granting additional priority include grade IV hepatic coma requiring hospitalization in an intensive care unit and endotracheal intubation for mechanical ventilation. Cerebral edema associated with HE in the patient with cirrhosis is quickly lethal and may justify an exception if the intracranial pressure is high.8, 9 Another example is profound HE associated with a large portosystemic shunt that cannot be occluded for technical reasons. Chronic intermittent sepsis in a patient with a low MELD score who has, for example, low-grade prostatitis or breakthrough severe HE requiring hospitalization and airway protection might also justify a MELD exception. To facilitate the development of an accurate method for fair prioritization of liver transplant candidates with HE, we suggest that an estimate of the time spent in (duration of) grade III/IV HE be included in the model.

We propose that the West Haven criteria10 for grading mental status be used to assess for the presence of HE (Table 1). Other, more sophisticated measures need to be studied for their potential use as objective criteria of HE over time. Inclusion of duration of severe HE may identify difficult-to-manage and frequently recurrent HE. The Scientific Registry of Transplant Recipients analysis established that a one-time (cross-sectional) assessment of HE predicts mortality independent of the MELD score.2 We propose that the strength of this prediction will be greatly enhanced by factoring in duration of severe HE.

Table 1. West Haven Criteria of Altered Mental Status10
0• Normal—no clinical signs or symptoms
1• Trivial lack of awareness
 • Euphoria or anxiety
 • Shortened attention span
 • Impaired performance of addition
2• Lethargy or apathy
 • Minimal disorientation for time or place
 • Inappropriate behavior
 • Subtle personality change
 • Impaired performance of subtraction
3• Somnolence to semistupor, but responsive to verbal stimuli
 • Confusion
 • Gross disorientation
4• Unresponsive to verbal or noxious stimuli


Currently there is only one situation in which the patient with HE due to chronic liver disease should be automatically given priority: cerebral edema and intracranial hypertension as described by Donovan and colleagues.8 Aside from this rare situation, there is currently no justification for automatically and systematically increasing priority for candidates with symptoms of HE. The debate over whether intracranial pressure measurements are an absolute requirement for the above MELD exception has not been resolved.


HE, hepatic encephalopathy; MELD, Model for End-Stage Liver Disease.


Encephalopathy should be defined by using nomenclature from the consensus statement in the final report of the working party at the 11th World Congresses of Gastroenterology11, 12 as follows:

  • Type A—HE associated with acute liver failure.

  • Type B—HE associated with portosystemic bypass (noncirrhotic).

  • Type C—HE associated with chronic liver disease/cirrhosis with the following subclassifications:

    • Episodic HE—single or recurrent.

    • Persistent HE—mild or severe.

    • Minimal HE—formerly called subclinical.

Data on HE should be prospectively collected via UNet, with the West Haven criteria10 used to estimate severity. Additional data are needed to begin to more accurately define HE as a predictor of mortality. It is evident, however, that the burden of data collection cannot be borne by UNet and that a new system needs to be devised that maximizes the accuracy of data collection and yet is practical. Optimally, in addition to severity of HE over time, the following information should be prospectively collected:

  • Time spent in duration of stage III/IV HE.

  • Number of hospitalization admissions for HE.

  • Number of days in an intensive care unit due to HE.

  • Presence of a transjugular intrahepatic portosystemic shunt complicated by worsening HE.

  • Number of episodes of severe HE necessitating endotracheal intubation.

  • Duration of grade III/IV HE per week or month (can serve as a surrogate for the other parameters).


One of the major virtues of the MELD scoring system is that it eliminates the subjective assessment of factors that contribute to higher-priority testing. Currently, the concern is that the severity of HE may be overestimated once again if it contributes to the priority for liver transplantation over and above the MELD score. If duration of severe HE is a strong predictor of early mortality, it may be valid for use in MELD exceptions, but data collection must be rigorous. A very strong case can be made that duration of severe HE should be estimated by non–liver transplant staff (i.e., critical care nurses). Also, the issue of how to assess severe HE in the patient with an endotracheal tube who is mechanically ventilated and frequently sedated has not been resolved. We must not reward inappropriate endotracheal intubation. Does severe HE always warrant intubation to protect the airway in a comatose patient? Before data collection and analysis are undertaken to answer this question, the care of patients with HE needs to be standardized.


At this time, because of the lack of a quantifiable, verifiable, and reproducible method of documenting HE, we propose that intractable or complicated HE continue to be addressed by the Regional Review Boards and that additional priority be assigned on a case-by-case basis after the above-described data have been submitted. Additional priority should not be not automatically granted at this time. There are no data that permit extrapolation of this recommendation to children.