Drug-induced removal of nitrogen derivatives in urine: A new concept whose time has come


  • Juan Córdoba M.D.,

    Corresponding author
    1. Liver Unit Hospital Vall Hebron Barcelona, Spain
    • Address reprint requests to: Juan Córdoba, M.D., Liver Unit, Hospital Vall Hebron, Paseo Vall Hebron 119, Barcelona 08035, Spain. E-mail: jcordoba@vhebron.net; fax: +34 93 2746068.

    Search for more papers by this author
  • Meritxell Ventura-Cots M.D.

    1. Departament de Medicina, Universitat Autònoma de Barcelona, CIBERehd Instituto de Salud Carlos III, Madrid, Spain
    Search for more papers by this author

  • See Article on Page 1073

  • Potential conflict of interest: Dr. Cordoba received grants from Ocera.


glycerol phenylbutyrate


hepatic encephalopathy


Model for End-Stage Liver Disease







Hepatic encephalopathy (HE) corresponds to a confusional syndrome that can progress to coma and is caused by liver failure and/or portal-systemic shunting.[1] HE is frequent among hospitalized patients with cirrhosis; their admission into the hospital causes an important burden of care and carries important economic costs.[2] In spite of major progress in the treatment of complications of cirrhosis, therapy of HE episodes has remained unchanged for many years. Most recommendations have been focused on the general management of patients, with the correction of precipitating factors being the basis of treatment.[3] Therapies for HE are mostly based on empirical observations without a good pathophysiological understanding.[4] Recent theories have been focused on the accumulation of water in the astrocyte (the cell that supports neuronal function) and the effects of activated neuroinflammation.[5] However, they have not led to new therapeutic developments.

For centuries, ammonia has been considered an important factor in the pathogenesis of HE.[6] Ammonia precipitates a cascade of disturbances in multiple neurotransmitter systems[7] and may led to neurodegeneration.[8] However, the role of ammonia is controversial. Though several precipitating factors increase plasma ammonia (e.g., constipation, gastrointestinal bleeding, infection, and so on), there is no good association between plasma levels and outcome. This may be explained by the coexistence of other factors that enhance ammonia injury (nitrosative stress, cytokines, Toll-like receptor polymorphisms, hyponatremia, dehydration, and so on). In addition, blood ammonia may not be an adequate biomarker of ammonia exposure to the brain, which may depend more on time of exposure than on peak values. Nevertheless, an ultimate element to judge the role of ammonia would be demonstrating the benefits of drugs that eliminate ammonia or related metabolites.

Glycerol phenylbutyrate (GPB) is a new drug for HE that acts by providing an alternate pathway for ammonia removal in the form of urinary phenylacetylglutamine (PAG). This drug was designed for hyperammonemia secondary to urea-cycle disorders, where there is a large experience with phenylbutyrate (PB) and related compounds.[9] These compounds were reserved for inborn errors of metabolism and did not reach wider use in HE. The current issue of Hepatology publishes the first study evaluating the effects of GPB in patients with cirrhosis.[10] The strategy consists of stimulating the elimination of waste nitrogen with orally administered GPB. The study shows that GPB improves outcome among patients with cirrhosis with highly recurrent HE. The new drug, which improves the organoleptic properties of its predecessor (sodium PB) and avoids the risk of sodium overload, was well tolerated and had a good safety profile.

The study will become a hallmark because it initiates a new therapeutic approach to HE. As previously demonstrated for lactulose[11] and rifaximin (RIF),[12] GPB is effective in the prevention of recurrence in a group of patients with cirrhosis with a relatively preserved liver function. The study reports that GPB prevents recurrence of HE, even in patients that receive lactulose. This is an interesting observation, because GPB has a different mechanism of action, compared to currently used drugs. The efficacy of nonabsorbable disaccharides (lactulose and lactitol) and antibiotics (RIF and neomycin) depends on interaction with gut flora.[13] Both types of drugs appear to have several effects on the intestinal flora that may enhance the ammonia-lowering effect of GPB. A remarkable effect of GPB is that it can promote pathways for excreting nitrogen metabolites in urine that result in a decreased frequency of HE recurrence. Thus, the combination with gut-focused therapies may have added efficacy, as demonstrated for lactulose, but could not be demonstrated in the limited number of patients that received GPB plus RIF; combined treatment was administered to those with a more severe clinical presentation. Future studies should include patients with cirrhosis over a large range of severity and examine the value of combining several drugs. Ornithine phenylacetate, a new product that also enhances the disposal of nitrogenous molecules in urine, is under clinical development in decompensated cirrhosis and may become an intravenous treatment for episodic HE.[14]1

Figure 1.

Abbreviations: NH3, ammonia; GLN, glutamine; GPB, glycerol phenylbutyrate; PAG, Phenylacetylglutamine; GLU, glutamate; GNASE, glutamine synthetase.

The efficacy of currently used drugs for treating the episode of HE has been derived from preventive studies among highly recurrent HE patients.[10, 12] The effect of therapy was investigated in patients with relatively preserved liver function (according to Model for End-Stage Liver Disease [MELD] or Child-Pugh scores); the majority of them showed no precipitating factor. These results suggest that chronic preventive therapy should be reserved to recurrent HE and not generalized to all patients. Once an episode of HE has been resolved, chronic therapy should be tailored to the patient. Lactulose or lactitol is recommended after the first episode, in combination with RIF after the second. Therapy after an episode related to a precipitating factor may not be needed if the precipitating factor is under control. Cirrhosis is not a stable condition; several factors, such as alcoholic liver injury, bacterial infection, liver injury, or dehydration, change over time. Patients with low MELD and dominating shunts can benefit from their occlusion.[15]

HE has been usually considered a “multifactorial” disorder without understanding the role of the involved factors. Response to a treatment specifically addressed to remove ammonia in urine as related metabolites (PAG),provides additional value to the ammonia hypothesis.[6] The results of the study extend the importance of ammonia to the whole metabolic process of nitrogenous metabolites, which are closely interrelated. The renewed interest in ammonia provides justification for the name of the society that deals with HE (International Society for Hepatic Encephalopathy and Nitrogen Metabolism). The lesson that we can learn from our pediatric colleagues is that a therapeutic strategy, commonly used in nitrogen disorders, may be beneficial for patients with cirrhosis and HE. The results of the current study are very encouraging and provide a theoretical framework for better understanding the role of ammonia in HE. Next, it would be important to corroborate its preventive efficacy in different groups of patients and confirming the effects in episodic HE.

  • Juan Cordoba, M.D.1,2Meritxell Ventura-Cots, M.D.1

  • 1Liver Unit, Hospital Vall Hebron, Paseo Vall Hebron 119, Barcelona 08035 Departament de Medicina, Universitat Autònoma de Barcelona

  • 2CIBERehd, Instituto de Salud Carlos III, Madrid Spain