摘要
It has been said that “lactulose is a many splendored thing . . . with many other beneficial actions in its bag of tricks.”1 Is the routine use of lactulose as prophylaxis for hepatic encephalopathy following an acute variceal bleed another “trick” to be pulled out of the proverbial bag? The use of lactulose has long been applied in the setting of constipation. In 1966, lactulose was introduced in the treatment of hepatic encephalopathy by Bircher et al.2 Despite a relative lack of high-powered studies conclusively proving its beneficial effect over placebo,3 lactulose is still considered first-line management (along with the identification and correction of precipitating factors).4 The development of overt hepatic encephalopathy is itself a poor prognostic indicator.5, 6 Gastrointestinal bleeding, in particular, an acute variceal bleed, is a common precipitant of hepatic encephalopathy. The exact prognostic significance of hepatic encephalopathy in the context of an acute variceal bleed is unclear; however, following a first episode, the overall transplant-free survival at 1 year is only 42%.6 The pathogenesis of hepatic encephalopathy is complex and imprecisely defined. It is thought to revolve around elevated levels of ammonia, an inflammatory response, and subsequent astrocyte swelling leading to cerebral edema.5 The neuropsychiatric disorder that results is variable, and is by consensus clinically defined using the West Haven criteria developed by Conn et al. in 1977.7, 8 Ammonia is produced as a byproduct of the metabolism of nitrogen-containing compounds, abundant in the bacterial flora of the gastrointestinal tract. In the “normal” system, the liver removes systemic ammonia by converting it to the water soluble urea. In liver disease, however, this function is impaired (due to either hepatocellular failure or portosystemic shunting) and brain and muscle cells are increasingly involved, converting ammonia to glutamine.5, 9 The treatment of hepatic encephalopathy has thus focused around reducing the production and absorption of ammonia in the gut.5 Precipitants of hepatic encephalopathy are many. In the case of an acute gastrointestinal bleed, increased ammonia levels arise from the high protein load in the gut. The Baveno IV guidelines, and subsequent AASLD (American Association for the Study of Liver Diseases) practice guidelines, for the management of portal hypertension, outline key management issues immediately after an acute bleeding episode, including the recommendation for antibiotic prophylaxis for preventing bacterial infections/spontaneous bacterial peritonitis.10, 11 Furthermore, there is a recommendation that treatment with lactulose is indicated if hepatic encephalopathy eventuates.10, 11 However, while it would seem prudent to use lactulose as prophylaxis in a setting that is well known to precipitate the condition, this is not currently a guideline recommendation. Lactulose, a non-absorbable disaccharide, is not degraded in the upper gastrointestinal tract. Aside from its cathartic effect, lactulose reduces the synthesis and absorption of ammonia by driving the conversion of ammonia to the non-absorbable ammonium via a reduction in the colonic pH.5, 12 A Cochrane Review in 2004 of studies from 1969 to 2003 evaluated the beneficial effect of lactulose in hepatic encephalopathy, and concluded that the evidence was insufficiently sound to support its use. Furthermore, the authors recommended that it should not be utilized as a comparator in future trials.3 However, a study published in 2009 by Sharma et al. concluded that lactulose was in fact effective in the prevention of recurrent hepatic encephalopathy.13 In this open-label, randomized, controlled trial of 125 patients, randomly assigned to either lactulose or non-lactulose treatment, 20% (12/61) and 47% (30/64) developed overt hepatic encephalopathy, respectively (P = 0.001).13 In this study, non-response to lactulose was reported in patients with hyponatremia and very high venous ammonia levels.13 The study by Sharma et al. in this issue of the Journal of Gastroenterology and Hepatology14 examined the effectiveness of lactulose in the prevention of hepatic encephalopathy following an acute variceal bleed. The study involved the randomization of 70 patients to either 30 mL lactulose three to four times per day (ultimately to ensure 2–3 semiformed stools per day) or non-lactulose. Inclusion criteria included admission within 24 h of gastrointestinal bleed, no history of lactulose intake in the preceding 6 weeks, and no significant distracting comorbidity.14 The inclusion of patients with a prior history of encephalopathy rendered it as neither a primary nor secondary prophylaxis study. Bleeding was assessed and managed in accordance with Baveno IV guidelines, and hepatic encephalopathy was assessed as per West Haven criteria by two independent assessors. Approximately one-sixth of patients in each group received terlipressin instead of somatostatin as a vasoactive drug. Nineteen patients developed clinically-overt hepatic encephalopathy; 14% (5/35) in the lactulose group and 40% (14/35) in the non-lactulose group (P = 0.03), giving a relative risk reduction of 66%.14 The resultant increase in length of hospital stay of those who developed encephalopathy was also significant (11 ± 2.2 vs 7 ± 1.8 days, P = 0.001).14 Those in the control group were treated with lactulose therapy once hepatic encephalopathy did develop. Confounding precipitants of hepatic encephalopathy are difficult to separate, although there were no significant biochemical differences between the two groups. Alternative therapies for hepatic encephalopathy include antibiotics, such as neomycin, that reduce ammonia-producing bacteria. These have been used in the long-term treatment of hepatic encephalopathy.4 However, more recently, they have lost favor secondary to their side-effect profiles. Rifaximin is a semisynthetic rifamycin-based antibiotic with gut-specific action as a result of its poor solubility and absorption.15 The negligible plasma levels of rifaximin bode well for avoiding potential drug resistance.15 Solid evidence for the use of rifaximin in the treatment of hepatic encephalopathy is largely a consequence of the randomized, controlled trial conducted by Bass et al.16 In that study, rifaximin was compared to placebo in patients in remission from recurrent hepatic encephalopathy, with breakthrough episodes reported in 22% (31/140) of patients in the rifaximin group and 46% (73/159) in the placebo group (P < 0.001). Of interest, over 90% of patients in both groups concomitantly used lactulose (the common adverse effects of which are bloating, abdominal cramping, and diarrhea). In contrast, rifaximin has not been shown to have a significant side-effect profile,16 certainly when compared to other previously utilized antibiotics, although long-term data are not yet available. Some of the adverse effects of lactulose can be minimized with the use of lactitol (an analog of lactulose);17 however, this is not available worldwide. Overtreatment of lactulose can lead to the more serious side-effects of marked dehydration, hyponatremia, and worsening of hepatic encephalopathy.5 The dosing regime of rifaximin can be either fixed daily or cyclical,15 whereas lactulose does rely on some degree of patient adjustment to ensure approximately two to three semiformed stools daily. Costing considerations are not straightforward. While the direct cost of rifaximin compared to lactulose is marked, approximately $US1120 as opposed to $150 per month in the USA,15 the total annual costing of rifaximin has been reported to be less than lactulose when hospital admissions are taken into account.18 The role of rifaximin in the prevention or treatment of hepatic encephalopathy following an acute variceal bleed remains to be elucidated. Even if no further conclusive studies confirming the beneficial effect of lactulose occur, it is difficult to conceive of it being substituted as first-line therapy for hepatic encephalopathy. Its routine prophylactic use in the setting of an acute variceal bleed (post-initial stabilization and endoscopic control of bleeding) seems intuitive. Perhaps prophylactic lactulose therapy should be considered for incorporation into acute variceal bleed practice guidelines as the “standard of care.”