Objectives

1. To recognize the constellation of critical illness that complicates the care of patients with cirrhosis.
2. To identify the unique challenges of patients with cirrhosis and GI bleeding present.
3. To recognize the common pathogens that cause infections in patients with cirrhosis admitted to the ICU with sepsis related to peritonitis.
4. To review the different strategies for treating patients with cirrhosis presenting with hepatic encephalopathy (HE).

Abbreviations

FFP = fresh frozen plasma; HE = hepatic encephalopathy; SBP = spontaneous bacterial peritonitis; TIPS = transjugular intrahepatic portosystemic shunt

In 2000, chronic liver disease and cirrhosis were responsible for 353,000 hospital discharges and 26,552 deaths in the United States. Cirrhosis is the 12th most common cause of death and the single most common cause of death from nonmalignant digestive diseases in the United States.¹ Patients with cirrhosis and decompensated disease present a variety of unique challenges when they are admitted to the ICU. Mortality rates range from 37 to 98% in such patients when they are admitted to the ICU.² The mortality rate may be as high as 100% in patients with multisystem organ failure who require mechanical ventilation and in patients with septic shock admitted to the ICU.^3,4 This PCCU lesson will focus on three types of complications in patients with cirrhosis who are admitted to the ICU:

1. The patient with cirrhosis and bleeding from esophageal varices.
2. The patient with cirrhosis and altered mental status consistent with hepatic encephalopathy (HE).
3. The patient with cirrhosis and bacterial peritonitis.

These patients can present with different clinical manifestations and have different prognoses, based on the constellation of findings present upon arrival in the ICU.   Each of these types of patients' problems will be reviewed. Although there are other important manifestations of acute and chronic liver failure relevant to ICU management, such as hepatorenal syndrome, infection other than peritonitis, and other end organ failure, these issues will not be addressed in this lesson. In addition, the underlying cause of chronic liver disease may be an important factor in management; disease due to chronic alcohol toxicity, for example, may be associated with cardiomyopathy, while primary biliary cirrhosis may be associated with malabsorption, hypercoagulability, and higher bilirubin levels. Refer to recent reviews on the broader topic of end stage liver disease.^5-8

Management of the Patient With Cirrhosis and Acute GI (Variceal) Bleeding

GI bleeding occurs in one third of patients with cirrhosis and is responsible for > 25% of the deaths in these patients.⁹ Approximately 80 to 90% of upper GI bleeding in patients with cirrhosis is a result of esophageal and gastric varices.¹⁰ Variceal hemorrhage as a subtype of GI bleeding has a mortality in the range of 30 to 50%.¹¹ Up to 30% of initial bleeding episodes are fatal, and approximately 70% of the patients who survive the initial episode have recurrent bleeding. The current approach to the patient with cirrhosis and active upper GI bleeding includes pharmacologic and endoscopic therapeutic maneuvers (most commonly sclerotherapy and band ligation) and the administration of blood products. In some patients, a transjugular intrahepatic shunt insertion is necessary. Although less relevant to acute ICU management and, thus, not discussed further here, there are multiple approaches to primary and secondary prevention of variceal bleeding that come into play in later follow-up.¹²

Pharmacologic agents used to control acute bleeding belong to one of two groups: agents that reduce the collateral portal venous flow (vasoconstrictors) and those that reduce the intrahepatic vascular resistance (vasodilators).¹² Several studies have demonstrated mortality benefits in patients with cirrhosis receiving pharmacologic therapy, aimed at the management of variceal hemorrhage prior to endoscopy.^10,13,14 Most commonly, a vasoconstrictor is given in combination with a vasodilator. A double-blind, placebo-controlled trial of early administration of somatostatin (a naturally occurring peptide that reduces portal pressure) vs placebo, initiated prior to sclerotherapy, was associated with a decreased mortality, volume of blood product support needed, active bleeding, and a greater ease of sclerotherapy performance.¹³ The effectiveness and safety of a vasopressin analogue (terlipressin) combined with glyceryl trinitrate (vasodilatation) administered by the emergency medical team has also been studied. The treatment group demonstrated superior bleeding control and significantly reduced mortality at days 15 and 42, without serious side effects compared to the placebo group.¹⁰ In another study, octreotide, a synthetic analogue of somatostatin, was given in combination with sclerotherapy for 5 days and compared with sclerotherapy alone in a double-blind prospective trial. Patients who received octreotide (25 m g/h) had less episodes of rebleeding and less units of blood transfused.¹⁵ Based on these studies, the early institution of vasopressin analogues for treatment of suspected variceal bleeding in patients with cirrhosis should occur even before a definitive diagnosis is determined.

The endoscopic therapies are crucial in the acute management of esophageal variceal bleeding in patients with cirrhosis. Endoscopy is useful not only to define the cause of the bleeding (thereby providing prognostic information), but also to serve as definitive therapy. In addition, alternative sources of bleeding, such as gastric varices and peptic ulcer disease, may be identified. The available endoscopic techniques used to control bleeding include sclerotherapy, band ligation, and use of cyanoacrylate glue in patients with gastric varices. Several randomized trials have shown that band ligation is equivalent or superior in achieving hemostasis, when compared to sclerotherapy in the setting of variceal bleeding, and is associated with a lower rate of complications.¹² A meta-analysis by Laine and Cook,¹⁶ that pooled data from seven relevant randomized trials, compared ligation to sclerotherapy for bleeding esophageal varices. Ligation therapy compared with sclerotherapy reduced the rebleeding rate, the mortality rate, and the rate of death due to bleeding. It was calculated that four patients would require ligation rather than sclerotherapy to avert one rebleeding episode, and 10 would require ligation instead of sclerotherapy to prevent one death.

A combination of pharmacologic and endoscopic therapy is beneficial in the setting of esophageal variceal bleeding. Two major randomized, controlled trials have demonstrated a mortality improvement in subjects receiving sclerotherapy with a vasopressin analogue as opposed to sclerotherapy alone.^14,15 A third study involving band ligation, with or without octreotide, showed that the use of combination therapy was associated with a decreased risk of rebleeding, as well as need for balloon tamponade and a lower 30-day mortality, with the latter demonstrating a trend towards statistical significance.¹⁷ Thus, pharmacologic therapy should be instituted prior to endoscopic therapy and when endoscopy is performed, if possible, band ligation is favored over sclerotherapy.

While esophageal varices are the most common type of variceal hemorrhage requiring ICU admission, gastric varices may be the source of bleeding in 10 to 36% of patients with variceal hemorrhage.¹⁸ Bleeding from isolated gastric varices may have a very different course than bleeding manifested by bleeding esophageal varices, particularly because of the high rate of rebleeding. Whereas such varices may initially also respond to the pharmacologic therapies, transjugular intrahepatic portosystemic shunt (TIPS) is arguably the therapy of choice.¹⁹ Additional therapies, such as the use of cyanoacrylate glue for the treatment of gastric varices, have been found to be safe and cost effective,²⁰ yet not fully developed in large clinical trials.

Many centers are capable of providing TIPS for management of patients with severe uncontrolled bleeding after failure of endoscopic therapy as a single modality or, as is more likely the case in current practice, failure of combination therapy.²¹ If TIPS is successfully performed, it almost invariably stops the bleeding in the subset of patients with refractory acute variceal hemorrhage.¹² TIPS is done by passing a needle from the hepatic vein into a major branch of the portal vein through the liver parenchyma and deploying an expandable stent. Major acute complications include hemoperitoneum, cardiopulmonary arrest, heart failure, acute renal failure, and bacteremia. Potential for early restenosis is greater when the procedure is performed in emergency situations.²² One of the most common, and perhaps difficult to manage, situations is the chronic complication of worsening HE following TIPS.²³

The management of variceal hemorrhage may be complicated by the fact that patients with cirrhosis often have coagulopathies due to a reduction in the levels of clotting factors. Endoscopic and pharmacologic management should occur in conjunction with optimal blood product and fluid management with specific attention to coagulopathy. The liver is the site of production of all the clotting factors, except the von Willebrand factor. Liver disease results in coagulopathy for multiple reasons. Thrombocytopenia (platelet count <150,000 x10³/mL) may occur in the majority of patients with cirrhosis²⁴ and may be significantly contributed to by active ethanol-related bone marrow suppression in the subset of patients actively abusing alcohol. Thrombocytopenia is also associated with splenomegaly due to portal hypertension, impaired synthesis, low levels of thrombopoietin,²⁵ elevated levels of antiplatelet immunoglobulins,²⁶ and other mechanisms, including interactions with other coagulation factors.

The impaired synthesis of coagulant factors and impaired clearance of their inhibitors add to the hypocoagulability state. The prothrombin time is elevated in association with worsening liver disease, and, because of its strong association with liver disease, prothrombin time is an integral component of several prognostic scoring measures.^27,28 Vitamin K levels can also be low in association with malnutrition²⁹ and impaired absorption due to impaired levels of biliary salt excretion. Other abnormalities, such as dysfibrinogenemia, resulting in elevated thrombin time and hyperfibrinolysis, are also common.^30,31

Management of coagulopathy is indicated when the patient is actively bleeding. Fresh frozen plasma (FFP) should be infused when the INR is >2 in an actively bleeding, coagulopathic patient with cirrhosis.³² FFP is typically dosed at 10 to 20 mL/kg. (Given that one unit of FFP is usually 250 mL, in a 70-kg adult, this equates to 1400 mL of FFP, based on 20 mL/kg. Thus, the patient should receive 6 units, rounded up from 5.6 units.) In patients with hypofibrinogenemia, as manifested by continued bleeding or elevated prothrombin time and activated partial thromboplastin time despite repletion of factors with FFP, administration of cryoprecipitate is indicated. While FFP does have relatively normal levels of fibrinogen, the volume needed to replenish a patient with fibrinogenopenia is inadequate. One unit per 10 kg will increase fibrinogen by 50 mg/dL.³³ Platelet transfusions are typically indicated in the context of bleeding and platelet count < 40,000 to 50,000 ´ 10 3 /mL.³³ Vitamin K replacement with 10 mg IV infusion for 3 days is typically performed. DDAVP, or desmopressin, stimulates the release of stored factor VIII and the von Willebrand factor from the endothelium; yet, it has not been proven to be useful in bleeding due to varices.³⁴ Large volume-packed RBCs or blood products may lead to a relative deficiency of endogenous procoagulant factors, acid base disturbances, potential for lung injury, and other complications.

In summary, for patients with active variceal bleeding, ligation is the preferred treatment based on its efficacy and low rate of complications. Patients for whom endoscopic therapy or the required expertise is not immediately available, somatostatin or octreotide should be used while awaiting endoscopic or more definitive treatments.³⁵

Management of the Patient With Cirrhosis and Altered Mental Status

HE is a general term that includes the entire range of (reversible) neuropsychiatric disturbances manifested in patients with liver failure. The diffuseness of this description has led to a need to better define the syndrome for precision in clinical trials.³⁶ Because there is no specific set of findings associated with HE, it is an exclusionary diagnosis. Upon recognizing the syndrome, in addition to usual supportive care, the common potential mimics and causes of HE should be reviewed (Table 1).⁵

The development of an acute confusional state that can progress to coma is termed acute encephalopathy . Acute encephalopathy can complicate the course of acute liver failure and cirrhosis.⁸ Acute HE can be precipitated by a wide range of hemodynamic, metabolic, or infectious complications of disease progression or decompensation or as the result of a specific treatment, such as TIPS.

The pathogenesis of acute HE is unknown; however, it is accepted that nitrogenous substances, derived from the gut, adversely affect brain function.⁸ The exact mechanisms underlying acute HE are unclear, but possible explanations include alteration of neurotransmitters, like glutamatergic, serotoninergic, g -aminobutyric acid, catecholamine pathways, and other alterations of cerebral metabolism.⁸ Ammonia metabolism is implicated (elevated ammonia levels correlate with severe encephalopathy), but a causative association has not been established.³⁷

Grading of HE is based on clinical findings, but, in the ICU, it may be manifested as obtundation and coma necessitating airway protection.⁸ The management of HE is aimed at the identification and treatment of the precipitating factors and providing supportive care, including adequate nutrition and reduction of the load of nitrogenous products in the gut. The reduction of the nitrogenous load is achieved by the induction of catharsis and administration of disaccharides (lactulose) and/or antibiotics. The administration of lactulose to induce an increase in the amount of stools to remove ammoniagenic substrates has been shown to improve HE.³⁸ The daily dose of lactulose should be titrated to result in two to four soft acidic (pH <6) stools per day.⁵ Lactulose may be administered orally or by retention enema. Antibiotics, such as neomycin and metronidazole, have been used with some additive benefit to lactulose but may also be considered a therapeutic alternative.⁸ Other adjunctive therapies, such as zinc, ornithine aspartate, and protein restriction may also be helpful, but the evidence to support their routine use is limited. The use of flumazenil   (an antagonist of benzodiazepines) has been demonstrated to improve the EEG abnormalities found in patients with HE^39,40 and may be used if the patient may have received benzodiazepines acutely (in chronic users of benzodiazepines, flumazenil may induce withdrawal). A recent meta-analysis suggests that flumazenil is associated with an improvement in encephalopathy compared with placebo. However, the benefit is usually short-term and probably confined to those patients with a good prognosis.⁴¹ Despite this meta-analysis, the potential complications of flumazenil use make this therapy not part of standard practice.

Management of the Patient With Cirrhosis and Infection

Spontaneous bacterial peritonitis (SBP) is the prototypical infection of decompensated cirrhosis. The diagnosis of SBP is based on clinical suspicion and the finding of >250 polymorphonuclear leukocyte/mm³ in the peritoneal fluid. Seventy percent of patients with SBP are culture-positive (predominantly enteric Gram-negative organisms). The pathogenesis of SBP is associated with the immunoregulatory dysfunction present in patients with cirrhosis. Theories of gut translocation, mucosal failure, decreased opsonization, dilution of ascites factors, impaired Kupffer cell activity in the reticuloendothelial system, and other mechanisms have explained the high infection rates with enteric bacteria. Therapies directed at these organisms have changed over the years, with third-generation cephalosporins now favored for patients that require ICU management. Studies comparing the use of ampicillin and tobramycin vs third-generation cephalosporins showed better resolution and 10% fewer complications of superinfection and nephrotoxicity with the use of third-generation cephalosporins.⁴² Interestingly, results from a recent Cochrane systematic review concluded that there was no clear evidence for the antibiotic treatment of patients with cirrhosis and SBP due to the lack of placebo-controlled trials and different antibiotic regimens precluding meta-analysis. Until large, well-conducted trials provide adequate evidence of improved morbidity and mortality, treatment should be based on clinical judgment.⁴³ It is our clinical experience to treat these patients.

Since the era of antibiotic prophylaxis, resistant organisms have become an increasingly important issue.⁴⁴ Nonetheless, quinolone antibiotics remain the mainstay of primary and secondary prophylaxis for patients with GI bleeding,⁴⁵ in addition to patients admitted with low protein content ascitic fluid.⁴⁶

Albumin is the other adjunctive therapy indicated for use in the treatment of SBP. A recent, prospective, randomized, controlled study assessing the mortality impact of the addition of the albumin to cefotaxime for the treatment of SBP was impressive for demonstrating short- and long-term mortality benefits that appeared to be related to the development of renal failure in this setting.⁴⁷ The study did not include patients in shock; however, specific mention of the need for volume resuscitation, or that it was adequate, was not made either. The cost of albumin therapy is high and, therefore, still remains a matter for debate.⁴⁸

Conclusions

Patients with cirrhosis who require ICU care suffer a broad range of abnormalities as a result of the various manifestations of liver failure. The effects of portal hypertension and variceal hemorrhage need to be addressed immediately; pharmacologic therapy should even be started before endoscopy can be performed. Band ligation appears to be better than sclerotherapy, in terms of rebleeding risk and mortality with few adverse events. The management of blood and blood product support is a critical component of the management of the bleeding patient with cirrhosis. TIPS is a salvage therapy indicated for uncontrollable bleeding. HE is a clinical syndrome without a clear causal pathway and, therefore, should be considered to be a diagnosis of exclusion. The exercise of reviewing the various causes of altered mental status in the patient with cirrhosis can uncover several abnormalities requiring other specific interventions. The treatment of HE is based predominantly on the ammoniagenic model and traditionally employs lactulose to increase excretion of toxins. The risk of infection in patients with cirrhosis is high. All severely ill patients with cirrhosis should be thoroughly investigated at admission for possible sources of infection and at any point thereafter, if warranted by the clinical status. The ascites fluid should be sampled, particularly after procedures that may introduce infection. Therapy for SBP is based on clinical experience, favoring third-generation cephalosporins. Prophylaxis of SBP may alter the sensitivity of the common organisms, even as Gram-positive infections are occurring in greater frequency.

References

2. 1 Minino AM, Arias E, Kochanek KD, et al. Deaths: final data for 2000. Natl Vital Stat Rep 2002; 50:1-119
3. 2 Foreman MG, Mannino DM, Moss M. Cirrhosis as a risk factor for sepsis and death: analysis of the National Hospital Discharge Survey. Chest 2003; 124:1016-1020
4. 3 Zauner C, Schneeweiss B, Schneider B, et al. Short-term prognosis in critically ill patients with liver cirrhosis: an evaluation of a new scoring system. Eur J Gastroenterol Hepatol 2000; 12:517-522
5. 4 Kress JP, Rubin A, Pohlman AS, et al. Outcomes of critically ill patients denied consideration for liver transplantation. Am J Respir Crit Care Med 2000; 162:418-423
6. 5 Menon KV, Kamath PS. Managing the complications of cirrhosis. Mayo Clin Proc 2000; 75:501-509
7. 6 Borzio M, Salerno F, Piantoni L, et al. Bacterial infection in patients with advanced cirrhosis: a multicentre prospective study. Dig Liver Dis 2001; 33:41-48
8. 7 Blendis L, Wong F. The natural history and management of hepatorenal disorders: from pre-ascites to hepatorenal syndrome. Clin Med 2003; 3:154-159
9. 8 Blei AT, Cordoba J. Hepatic Encephalopathy. Am J Gastroenterol 2001; 96:1968-1976
10. 9 Schlichting P, Christensen E, Fauerholdt L, et al. Main causes of death in cirrhosis. Scand J Gastroenterol 1983; 18:881-888
11. 10 Levacher S, Letoumelin P, Pateron D, et al. Early administration of terlipressin plus glyceryl trinitrate to control active upper gastrointestinal bleeding in cirrhotic patients. Lancet 1995; 346:865-868
12. 11 Imperiale TF, Teran JC, McCullough AJ. A meta-analysis of somatostatin versus vasopressin in the management of acute esophageal variceal hemorrhage. Gastroenterology 1995; 109:1289-1294
13. 12 Sharara AI, Rockey DC. Gastroesophageal variceal hemorrhage. N Engl J Med 2001; 345:669-681
14. 13 Avgerinos A, Nevens F, Raptis S, et al. Early administration of somatostatin and efficacy of sclerotherapy in acute oesophageal variceal bleeds: the European Acute Bleeding Oesophageal Variceal Episodes (ABOVE) randomised trial. Lancet 1997; 350:1495-1499
15. 14 Villanueva C, Ortiz J, Sabat M, et al. Somatostatin alone or combined with emergency sclerotherapy in the treatment of acute esophageal variceal bleeding: a prospective randomized trial. Hepatology 1999; 30:384-389
16. 15 Besson I, Ingrand P, Person B, et al. Sclerotherapy with or without octreotide for acute variceal bleeding. N Engl J Med 1995; 333:555-560
17. 16 Laine L, Cook D. Endoscopic ligation compared with sclerotherapy for treatment of esophageal variceal bleeding. A meta-analysis. Ann Intern Med 1995; 123:280-287
18. 17 Sung JJ, Chung SC, Yung MY, et al. Prospective randomised study of effect of octreotide on rebleeding from oesophageal varices after endoscopic ligation. Lancet 1995; 346:1666-1669
19. 18 Stanley AJ, Hayes PC. Portal hypertension and variceal haemorrhage. Lancet 1997; 350:1235-1239
20. 19 Albillos A, Ruiz del Arbol L. "Salvage" transjugular intrahepatic portosystemic shunt: gastric fundal compared with esophageal variceal bleeding. Gastrointest Endosc 1999; 50:294-295
21. 20 Greenwald BD, Caldwell SH, Hespenheide EE, et al. N-2-butyl-cyanoacrylate for bleeding gastric varices: a United States pilot study and cost analysis. Am J Gastroenterol 2003; 98:1982-1988
22. 21 Banares R, Casado M, Rodriguez-Laiz JM, et al. Urgent transjugular intrahepatic portosystemic shunt for control of acute variceal bleeding. Am J Gastroenterol 1998; 93:75-79
23. 22 Gerbes AL, Gulberg V, Waggershauser T, et al. Transjugular intrahepatic portosystemic shunt (TIPS) for variceal bleeding in portal hypertension: comparison of emergency and elective interventions. Dig Dis Sci 1998; 43:2463-2469
24. 23 Sanyal AJ, Freedman AM, Shiffman ML, et al. Portosystemic encephalopathy after transjugular intrahepatic portosystemic shunt: results of a prospective controlled study. Hepatology 1994; 20:46-55
25. 24 Bashour FN, Teran JC, Mullen KD. Prevalence of peripheral blood cytopenias (hypersplenism) in patients with nonalcoholic chronic liver disease. Am J Gastroenterol 2000; 95:2936-2939
26. 25 Kawasaki T, Takeshita A, Souda K, et al. Serum thrombopoietin levels in patients with chronic hepatitis and liver cirrhosis. Am J Gastroenterol 1999; 94:1918-1922
27. 26 Samuel H, Nardi M, Karpatkin M, et al. Differentiation of autoimmune thrombocytopenia from thrombocytopenia associated with immune complex disease: systemic lupus erythematosus, hepatitis-cirrhosis, and HIV-1 infection by platelet and serum immunological measurements. Br J Haematol 1999; 105:1086-1091
28. 27 Pugh RN, Murray-Lyon IM, Dawson JL, et al. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973; 60:646-649
29. 28 Forman LM, Lucey MR. Predicting the prognosis of chronic liver disease: an evolution from child to MELD. Mayo End-stage Liver Disease. Hepatology 2001; 33:473-475
30. 29 Iber FL, Shamszad M, Miller PA, et al. Vitamin K deficiency in chronic alcoholic males. Alcohol Clin Exp Res 1986; 10:679-681
31. 30 Agarwal S, Joyner KA, Jr., Swaim MW. Ascites fluid as a possible origin for hyperfibrinolysis in advanced liver disease. Am J Gastroenterol 2000; 95:3218-3224
32. 31 Hu KQ, Yu AS, Tiyyagura L, et al. Hyperfibrinolytic activity in hospitalized cirrhotic patients in a referral liver unit. Am J Gastroenterol 2001; 96:1581-1586
33. 32 Everson GT. A hepatologist's perspective on the management of coagulation disorders before liver transplantation. Liver Transpl Surg 1997; 3:646-652
34. 33 Practice guidelines for blood component therapy: A report by the American Society of Anesthesiologists Task Force on Blood Component Therapy. Anesthesiology 1996; 84:732-747
35. 34 Mannucci PM. Desmopressin (DDAVP) in the treatment of bleeding disorders: the first 20 years. Blood 1997; 90:2515-2521
36. 35 Gross M, Schiemann U, Muhlhofer A, et al. Meta-analysis: efficacy of therapeutic regimens in ongoing variceal bleeding. Endoscopy 2001; 33:737-746
37. 36 Ferenci P, Lockwood A, Mullen K, et al. Hepatic encephalopathy--definition, nomenclature, diagnosis, and quantification: final report of the working party at the 11th World Congresses of Gastroenterology, Vienna, 1998. Hepatology 2002; 35:716-721
38. 37 Ong JP, Aggarwal A, Krieger D, et al. Correlation between ammonia levels and the severity of hepatic encephalopathy. Am J Med 2003; 114:188-193
39. 38 Conn HO, Leevy CM, Vlahcevic ZR, et al. Comparison of lactulose and neomycin in the treatment of chronic portal-systemic encephalopathy. A double blind controlled trial. Gastroenterology 1977; 72:573-583
40. 39 Dursun M, Caliskan M, Canoruc F, et al. The efficacy of flumazenil in subclinical to mild hepatic encephalopathic ambulatory patients. A prospective, randomised, double-blind, placebo-controlled study. Swiss Medical Weekly. 2003; 133:118-123
41. 40 Laccetti M, Manes G, Uomo G, et al. Flumazenil in the treatment of acute hepatic encephalopathy in cirrhotic patients: a double blind randomized placebo controlled study. Digestive & Liver Disease. 2000; 32:335-338
42. 41 Als-Nielsen B, Kjaergard LL, Gluud C. Benzodiazepine receptor antagonists for acute and chronic hepatic encephalopathy. Cochrane Database Syst Rev 2001:CD002798
43. 42 Felisart J, Rimola A, Arroyo V, et al. Cefotaxime is more effective than is ampicillin-tobramycin in cirrhotics with severe infections. Hepatology 1985; 5:457-462
44. 43 Soares-Weiser K, Brezis M, Leibovici L. Cochrane Database of Systematic Reviews 2003; 3:3
45. 44 Fernandez J, Navasa M, Gomez J, et al. Bacterial infections in cirrhosis: epidemiological changes with invasive procedures and norfloxacin prophylaxis. Hepatology 2002; 35:140-148
46. 45 ACP Journal Club. November/December 2002; 137:94
47. 46 ACP Journal Club. May/June 1991; 114:79
48. 47 Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis [comment]. N Engl J Med 1999; 341:403-409
49. 48 Brand RE. Intravenous albumin in patients with cirrhosis and spontaneous bacterial peritonitis: is it worth the cost? Am J Gastroenterol 1999; 94:3404