Introduction

Management of HE has traditionally been with non-absorbable disaccharides for lowering the production and absorption of ammonia (Blei and Cordoba 2001). Antibiotics have been mainly used as alternatives to the non-absorbable disaccharides in improving HE symptoms (Bajaj 2010). This review will highlight the role of antibiotics in the acute treatment and prevention of HE obtained through a detailed literature, abstract and guidelines search for studies that responded to the MESH terms “antibiotics and hepatic encephalopathy”. There will also be a brief description of the role of antibiotics in treating infections as a precipitant of overt HE, on minimal or covert HE, and evaluation of economic considerations in antibiotic therapy.

Rationale

The prevention of production and absorption of gut-derived neurotoxins along with reduction in endotoxemia and inflammation underlie the rationale for antibiotic use in HE. A major component of these toxins is ammonia, while, other gut-derived products, oxindole, phenols, mercaptans, and short-chain fatty acids, have also been implicated (Riggio et al. 2010). It has also been shown that cirrhotic patients have significantly altered stool microbiome, which can also impact cognition (Bajaj et al. 2012a). Lowering systemic ammonia and reduction of inflammation and endotoxemia can be achieved through antimicrobials (Kalambokis et al. 2012; Jalan 2010; Sanyal et al. 2010). Based on this gut-specific action, most antibiotics used have been gut non-absorbable in nature. Neomycin, vancomycin, paromomycin, metronidazole, and recently rifaximin, have been shown to be effective in acute and chronic therapy of HE (Poh and Chang 2012).

Systemic antibiotic use for infections that can precipitate overt HE

Bacterial infections are significant precipitants of overt HE, especially spontaneous bacterial peritonitis, and in situations where bacterial translocation is increased such as an upper GI bleeding episode (Gustot et al. 2009). The recommended antibiotics in this situation depend on the infection and local microbial susceptibilities but are broad-spectrum, systemic agents that are aimed at the infection rather than HE itself (Bajaj 2010). In contrast, antibiotics specifically for HE are gut-specific and are not effective for treating potential infections during the acute episode.

Antibiotic use in covert and overt HE

These are antibiotics specifically directed towards HE; most studies pertaining to antibiotic use have concentrated on the acute overt HE episode or prevention of overt HE recurrence (Table 1). Only rifaximin has been studied for covert/minimal HE and therefore it will be discussed in the section pertaining to rifaximin.

Table 1 Summary of overt HE acute, long-term and recurrence trials with antibiotics
Full size table

Aminoglycosides (neomycin and ribostamycin)

Neomycin and ribostamycin are the aminoglycosides that have been studied. Only one study comparing ribostamycin to lactulose in 15 patients has been performed, that showed equivalence; rest have used neomycin. Neomycin has activity against most gram-negative aerobes, except pseudomonas, and staphylococcal species. It inhibits bacterial protein synthesis via binding to the bacterial 30S ribosomal subunit and in some reports, has also been shown to inhibit intestinal glutaminase. There is limited data on the effectiveness of neomycin, since its use predated our current concept of evidence-based or randomized trials and it was the standard to which lactulose was initially compared to when it was first studied. A randomized, double blind, controlled trial comparing neomycin versus placebo in acute HE demonstrated no significant difference in symptomatic improvement (Strauss et al. 1992). In addition, other randomized controlled studies comparing neomycin to lactulose found no significant difference between the two agents (Orlandi et al. 1981; Atterbury et al. 1978; Conn et al. 1977). Dosing of neomycin is usually 1,000 mg every 6 h for up to 6 days in an acute episode of overt HE and 1–2 g daily for chronic use.

Though neomycin is FDA approved for HE, its continued clinical use is not recommended given its extensive side-effect profile. Common adverse events include intestinal malabsorption, nephrotoxicity, and ototoxicity. These side effects are common in chronic use because systemic absorption is much higher in hepatic and renal failure, as compared to roughly 4 % for subjects without this organ dysfunction, increasing the systemic exposure (Fig. 1). It is due to these side effects and lack of demonstrated clinical benefit that neomycin use has fallen out of favor for HE treatment.

Fig. 1
figure 1

Pharmacokinetics of rifaximin compared to neomycin and rifampin (U.S. Food and Drug Administration 2010). The figure shows that the systemic exposure to rifaximin compared to neomycin and rifampin in healthy individuals. The comparative exposure to rifaximin increases with worsening liver disease severity

Full size image

Metronidazole, vancomycin and paromomycin

There have been limited studies on the efficacy of metronidazole and vancomycin in the management of HE. In a small study by Morgan et al., 11 mild to moderate HE patients and seven chronically affected HE patients were treated for 1 week with 250 mg oral dose of metronidazole twice daily with similar efficacy as neomycin (Morgan et al. 1982). However given its prolonged rate of elimination in HE patients and increased risk for irreversible peripheral neurotoxicity metronidazole is not recommend for the management of an acute episode or for chronic management of HE (Loft et al. 1987).

Oral vancomycin, on the other hand, may be safer for the management for an acute HE episode, and has been studied in a limited group of HE patients who were resistant to lactulose (Tarao et al. 1990). Nevertheless, the limited studies, high expense, and with increased prevalence of vancomycin-resistant enterococci and other bacterial resistance, preclude its routine use (Fernández et al. 2012; Bajaj et al. 2012b).

Paromomycin has been compared with rifaximin for the treatment of acute episodes of HE in three studies. Only one trial showed improvement in cognitive testing that was greater in rifaximin; rest showed equivalence (Marco et al. 1984; Testa et al. 1985; Parini et al. 1992). Metronidazole, paromomycin and vancomycin are not FDA approved for treatment of overt HE.

Rifaximin

Rifaximin is gut-specific antimicrobial agent for the management of HE that is FDA-approved for prevention of overt HE recurrence. It has a broad spectrum of activity against both gram positive and gram negative organisms, and specifically against anaerobic enteric bacteria (Jiang and DuPont 2005). It binds to the b-subunit of the bacterial DNA-dependent RNA polymerase and disrupts RNA synthesis. However, unlike its derivative, rifamycin, less than 1 % is absorbed systemically after oral administration, resulting in greater concentration in the gastrointestinal tract. This systemic exposure increases with worsening liver disease severity (Fig. 1) (U.S. Food and Drug Administration 2010). It also has minimal effects on normal gut flora, though increased doses were shown to initially decrease GI flora such as enterococcus, Escherichia coli, Lactobacillus spp., Bacteroides spp., Bifidobacterium spp. and Clostridium perfringens all of which returned to initial values after a wash-out period (Brigidi et al. 2012).

Rifaximin has minimal side effects – headache, flatulence, abdominal pain, constipation, nausea, and vomiting – and no reported drug interactions make it relatively safe. It has been demonstrated to be superior to lactulose and other antimicrobials in numerous trials in patients with mild to moderate severe HE (Festi et al. 1993; Bucci and Palmieri 1993; Massa et al. 1993; Fera et al. 1993; Mas et al. 2003; Leevy and Phillips 2007; Paik et al. 2005; Alcorn 2008; Bajaj and Riggio 2010) (Table 1).

In a randomized, double-blind, placebo controlled prospective trial, Bass et al. compared rifaximin, dosed at 550 mg twice daily, to placebo over 6 months in patients with two prior overt HE episodes and showed a reduction in risk of developing HE (Bass et al. 2010). In addition the risk of hospitalization was significantly reduced in the rifaximin group. This study showed reduction in venous ammonia and improvement in health-related quality of life in patients randomized to rifaximin (Sanyal et al. 2010; Sanyal et al. 2011). Additionally a recent study by Neff et al, showed that rifaximin used for greater than 6 months to be an effective agent for HE, particularly in patients with a model for end-stage liver disease (MELD) score less than 20 (Neff et al. 2012).

Rifaximin is one of the few antimicrobial agents that has been tested in patients with covert/minimal HE. The RIME trial randomized minimal HE patients into rifaximin and placebo and found that rifaximin therapy was associated with a significant improvement in cognitive performance and health-related quality of life compared to placebo (Sidhu et al. 2011). Another study in minimal HE extended these findings on to the real-world outcomes of driving, in which patients randomized to rifaximin not only improved their cognition, but also significantly bettered their driving simulator performance and reduced systemic inflammation, but only psycho-social aspects of quality of life improved (Bajaj et al. 2011).

Economic analysis for HE treatment with antibiotics

The economic analysis of HE is important given the daunting and ever-increasing costs of health care. Therefore a pragmatic approach includes the cost analysis of the drug as well as the savings if subsequent negative outcomes, i.e. hospitalizations are prevented (Prakash and Mullen 2010; Shawcross et al. 2011; Stepanova et al. 2012; Poordad 2007). As with any management for a medical condition, the risks and benefits have to be considered. Both non-absorbable disaccharides and neomycin have dose limiting side-effects and ambiguous effectiveness. While, on the other hand, rifaximin, has a more tolerable side-effect profile as well as being more efficacious, but is more expensive.

A comprehensive decision analysis published by Huang et al. (2007) addressed this dilemma through a cost effectiveness of 6 different strategies in the management of HE. The six arms were: (1) no HE treatment; (2) lactulose monotherapy (3) lactitol monotherapy; (4) neomycin monotherapy; (5) rifaximin monotherapy; and (6) up front lactulose with crossover to rifaximin if there was a poor response or intolerance to lactulose (Huang et al. 2007). Through using decision-analysis software, the study concluded that the “do nothing” arm was the least efficacious and rifaximin salvage was most efficacious. However it was also noted that rifaximin monotherapy was not cost effective and that lactulose monotherapy and rifaximin salvage therapy was less expensive and more effective than alternate therapies. Conversely, studies have shown that since rifaximin may be associated with a lower rate of hospitalizations, this could result in overall cost-saving since the costs of hospitalizations far outweigh the costs of the therapy (Neff et al. 2007; Mantry and Munsaf 2010). In minimal or covert HE however, where hospitalizations are few and the therapy has to continue for a longer period potentially, rifaximin was not cost-saving for the prevention of motor vehicle accidents compared to lactulose from a societal perspective (Bajaj et al. 2012c).

Conclusions

Antimicrobial agents form a substantial component of the armamentarium against HE. Although several agents have been used, the antimicrobial agent with the most published experience is rifaximin. The use of rifaximin has evolved over several decades and is one of the most widely used antibiotics for overt HE. However further studies into appropriate place of antimicrobials as first or second line therapeutic strategies for HE are needed.