Morbidity and mortality in cirrhotic patients undergoing anesthesia and surgery Avishai Zisera and David J. Plevakb

Several studies have demonstrated increased morbidity and mortality in patients with cirrhosis undergoing anesthesia and surgery. Cirrhosis is a chronic liver disease, which may affect all body systems. The severity of the disease, assessed by the Child±Pugh classification, has a substantial effect on patient outcome. The extent of surgery and co-morbid conditions also have a major impact. In the past few years, changes have been made in the diagnosis, preoperative preparation, surgical and anesthetic management and perioperative care of patients with liver disease. The aim of this review is to examine whether these changes have resulted in improved perioperative outcomes. Curr Opin Anaesthesiol 14:707±711.

#

2001 Lippincott Williams & Wilkins.

a Department of Anesthesiology, Rambam Medical Center and the Technion Faculty of Medicine, Haifa, Israel; and bDepartment of Anesthesiology, Mayo Clinic and Mayo Graduate School of Medicine, Rochester, Minnesota, USA

Correspondence to Avishai Ziser, Department of Anesthesiology, Rambam Medical Center, PO Box 9602, Haifa 31096, Israel Tel: +972 53 782 105; fax: +972 4 629 0036; e-mail: [email protected] Current Opinion in Anaesthesiology 2001, 14:707±711 Abbreviations ARF GIB HCC LC OC

acute renal failure gastrointestinal bleeding hepatocellular carcinoma laparoscopic cholecystectomy open cholecystectomy

2001 Lippincott Williams & Wilkins 0952-7907 #

Introduction

A cirrhotic liver can affect all body systems, and result in shortened life expectancy [1]. However, selected asymptomatic cirrhotic patients may not have reduced life expectancy [1]. Liver and renal failure, postoperative bleeding and sepsis are common causes of poor perioperative outcome. Previous reports have demonstrated increased perioperative morbidity and mortality in cirrhotic patients [2]. High mortality rates in cirrhotic patients admitted for liver resection occurred despite careful preoperative assessment and patient selection [3]. In this review, we shall examine the most recent reports, and try to assess whether newer and more advanced surgical techniques, anesthetic drugs and perioperative care can provide better outcomes for patients with cirrhosis who undergo anesthesia and surgery.

The effect of cirrhosis on body systems

The liver is a vital organ. Its main functions include glucose and fat metabolism, protein synthesis (including albumin, globulins and coagulation factors), bilirubin formation and excretion, drug and hormone metabolism and detoxi®cation. Liver damage in cirrhotic patients may cause several hemodynamic, hematological, metabolic and other abnormalities. Hemodynamic and pulmonary alterations

A hyperdynamic state with increased cardiac output, decreased systemic vascular resistance and increased intravascular volume is typical in cirrhosis [4]. Tissue perfusion, however, may be reduced because of arteriovenous shunting. The cardiovascular response to endogenous and exogenous sympathomimetic amines is diminished. Intrapulmonary shunting, increased extravascular water, elevated diaphragm by ascites and pleural effusion, may all induce a mismatch of the ratio of ventilation to blood ¯ow in the lung, hypoxemia and hypoventilation. Hypoxic pulmonary vasoconstriction is blunted and the oxygen dissociation curve is shifted to the right [5]. Renal blood ¯ow tends to decrease, and renal failure may develop in patients with portal hypertension and ascites (the hepatorenal syndrome). Prerenal azotemia, altered vasoregulatory mechanisms and hepatic toxins are responsible for the hepatorenal syndrome [6]. Hematological alterations

Increased plasma volume, gastrointestinal bleeding (GIB), malnutrition, bone marrow depression, hemolysis 707

708 Anaesthesia and medical disease

and hypersplenism result in anemia and thrombocytopenia. The prothrombin time is elevated as a result of the depressed synthesis of coagulation factors II, VII, IX and X. Fibrinogen synthesis is decreased [7]. Metabolic changes and altered host-defense mechanisms

All synthetic functions of the liver are depressed in cirrhosis. Hypoalbuminemia is common and wound healing is delayed. Pseudocholinesterase production is reduced. Glucose utilization is impaired and increased plasma levels of growth hormone and glucagon induce glucose intolerance. Decreased elimination of ammonia and other neurotoxins may cause encephalopathy. This may be aggravated by the absorption of red blood cell breakdown products from the gastrointestinal tract as a result of upper GIB from esophageal and gastric varices. The decreased production of globulins is associated with an impaired immune defense system and an elevated risk of postoperative infections [8]. Gonadal dysplasia is common in both sexes. Altered drug metabolism

The liver is a major organ for the metabolism and elimination of many drugs. Drug metabolism in cirrhosis is impaired because of a substantial decrease in the number of functioning hepatocytes and altered hepatic blood supply. The volume of distribution of medications is increased. The half life of many medications is increased and elimination is slower, therefore the dose of several medications needs to be adjusted in cirrhotic patients. The free portion of highly protein-bound medications, increased by hypoalbuminemia, may counteract the effect of an increased volume of distribution [9]. Ascites, renal failure and hepatic encephalopathy

Several organs and systems are affected by cirrhosis. Ascites is common and it is a poor prognostic sign. Diuretic therapy for ascites may induce hyponatremia, hypokalemia, hypovolemia, prerenal azotemia, hepatorenal syndrome and encephalopathy. Various levels of hepatic encephalopathy might be present, from subtle behavioral changes to frank hepatic coma. Liver dysfunction and the diversion of portal venous blood ¯ow away from the liver are the main reasons for encephalopathy [10]. Patients with cirrhosis often have mild to moderate jaundice, weakness, reduced appetite, nausea, vomiting and abdominal pain. The effects of anesthesia and surgery on liver function

Cirrhosis may produce only minor clinical manifestations, and liver function tests may be close to normal. Hepatic function reserve, however, may be very limited, and portal venous pressure increased. The result is a decrease in portal blood ¯ow and multiple venous

collaterals. Local hepatic autoregulation of arterial blood ¯ow is impaired, making the cirrhotic liver less tolerant of hemodynamic changes. These alterations render the cirrhotic liver more vulnerable to hemorrhage, hypovolemia, hypotension and hypoxemia. Most inhalational anesthetic agents reduce hepatic blood ¯ow and can impair the liver's response to these insults. When these events occur during anesthesia and surgery, hepatocyte damage and liver failure can result [11].

Surgical procedures

Several publications have documented recent studies regarding the morbidity and mortality of cirrhotic patients undergoing various surgical procedures. Liver resection

Liver resection is now associated with decreased morbidity and mortality as a result of better diagnostic surveillance, surgical techniques, anesthetic management and anesthetic drugs, and improved perioperative care [12 .]. Tumors, which were considered unresectable in the past, are now amenable to surgical removal. Resection is utilized in primary liver tumors, or metastases (usually originating from colorectal cancer). The most common primary liver malignancy is hepatocellular carcinoma (HCC), and 50±85% of these patients have associated cirrhosis [13 .,14,15]. Some 10±20% of cirrhotic patients of any cause will develop HCC [13 .]. Close surveillance of cirrhotic patients, including repeated ultrasound and computerized tomography examinations of the liver, have increased the frequency of diagnosis of an early stage and resectable HCC [16 .]. Liver regeneration, which normally occurs within a few weeks after resection, is severely impaired in patients with cirrhosis [17]. Liver failure from extensive resection, in a condition of functional impairment, is the main cause of morbidity and mortality [13 .,14,18,19]. Several recent studies have reported their results in liver resection for patients with cirrhosis [12 .,14,15,16 .,18, 20 .±22 .,23,24 .,25 .,26,27 .]. The reported mortality rate was between zero and 8.7% [12 .,16 .,18,20 .,26], the postoperative complication rate was between 20 and 41.2%, and hospital stay was between 10.4 and 14.1 days [12 .,16 .,26]. In such studies, mortality and morbidity were associated with an increased prothrombin time [12 .], longer surgical time [12 .], the presence of active hepatitis [20 .], reduced indocyanine green clearance [20 .,24 .], male sex [21 .], Child's classi®cation B or C [24 .], the presence of malignant tumor [18], tumor size and the extent of tumor invasion [14,15,21 .,23,24 .]. As well as being a perioperative risk factor [18,23,24 .,25 .], cirrhosis was found to be associated with late recurrence of HCC [27 .]. Recent investigations have reported reduced mortality rates, fewer transfusion requirements and shorter post-

Morbidity and mortality in cirrhotic patients Ziser and Plevak 709

operative hospital stay when compared with older studies [12 .,18]. Despite recent advances, hepatic resection for HCC is mainly justi®ed in patients with adequate liver reserve, either no cirrhosis or Child±Pugh class A cirrhosis, and well-de®ned tumors [22 .]. For patients with advanced cirrhosis and portal hypertension, liver transplantation or non-surgical modes of therapy may be the more plausible alternatives [22 .]. Laparoscopic cholecystectomy

Patients with cirrhosis have a higher rate of gall stone formation, compared with patients without cirrhosis [28]. This ®nding is attributed to accelerated pigment stone formation. Earlier studies demonstrated high morbidity and mortality for laparoscopic cholecystectomy (LC) in patients with cirrhosis [29]. At that time, LC was felt to be contraindicated in these patients [30]. However, three recently published studies reported experience with LC in a total of 107 patients [31 .,32,33 .] (Table 1). The Child±Pugh classi®cation was A, B, and C in 81.3, 16.8 and 1.9%, respectively. Postoperative complications were recorded in 10.3% of patients, without mortality. The conversion rate to open cholecystectomy (OC) was 8.4%. Although the complication rate is higher when compared with LC in patients without cirrhosis [31 .], it is lower than in OC in patients with cirrhosis [33 .]. Other advantages of LC compared with OC in cirrhotic patients include shorter surgery time, reduced hospital stay, less need for transfusion, and a reduced risk for healthcare workers from innoculation with blood-borne pathogens. Despite the apparent safety of LC in cirrhotic patients, the procedure is certainly not without risk, especially in those patients with poorly compensated (Child±Pugh C) cirrhosis. Trauma

In a recent retrospective study of 27 500 trauma patients, Wahlstrom et al. [34 .] demonstrated a 0.89% incidence of cirrhosis. The 17 cirrhotic patients who underwent

Table 1. Recent publications concerning morbidity and mortality in cirrhotic patients undergoing laparoscopic cholecystectomy Reference no. Number of patients Study years Mean age (years) Child's class (%) A B C Length of surgery (min) Hospital stay (days) Blood transfusion (units) Complications (%) Conversion to OC (%) OC, Open cholecystectomy.

[31 .] 48 1993±1998 58.3 79.2 20.8 104 6.5 0.16 10.4 8.3

[32] 33 1990±1997 60 81.8 12.1 6.1 64 2.8 0 0 6

[33 .] 26 1990±1997 59.1 85 15 117 2.4 0 19 12

emergency laparotomy had 44% mortality and 71% morbidity. That represented a fourfold increase in mortality, compared with a matched control group. Increased mortality in trauma patients with cirrhosis has been reported in several previous studies, and is a widely accepted maxim for surgical teaching rounds [35]. Unfortunately, because of the emergent nature of these cases, the diagnosis of cirrhosis is usually made at the time of visual inspection of the liver during laparotomy. Therefore, it is very dif®cult to make well-informed decisions regarding surgical candidacy in these patients. It may also be dif®cult to optimize perioperative conditions in an effort to minimize the risk to the liver and other organs. Miscellaneous surgical procedures

Recent publications have documented morbidity and mortality in cirrhotic patients under a variety of surgical circumstances (Table 2). In cirrhotic patients undergoing esophagectomy for cancer, the postoperative complication rate was 83.3%, and blood loss was 78% higher in comparison with non-cirrhotic patients (1287 versus 723 ml, P50.05) [36 .]. However, if cirrhotic patients survive hospitalization, long-term survival was not signi®cantly different from those without cirrhosis. Cirrhotic patients undergoing minor surgical procedures were also found to be at higher risk. In a Danish survey of 59 501 patients undergoing appendectomy [37], the odds ratio for 30 day mortality risk in cirrhotic patients was 8 [3±20, 95% con®dence interval (CI)] when compared with non-cirrhotic patients. Another Danish study [38] demonstrated an adjusted odds ratio of 3.0 (0.4±22.9, 95% CI) for 30 days postoperative mortality in cirrhotic patients undergoing transurethral resection of the prostate. In patients with hepatic hydrothorax, thoracoscopy and talc poudrage was associated with a 47.6% success rate, a morbidity and mortality rate of 57.1 and 27.8%, respectively, and a mean postoperative hospital stay of 15 days [39 .]. By altering the surgical technique, De la Pena et al. [40] demonstrated that umbilical hernia can be surgically repaired in cirrhotic patients without morbidity or mortality. Cirrhosis and the intensive care unit

Cirrhosis was found to be one of the most important risk factors for the development of acute renal failure (ARF) in a prospective multicenter intensive care unit European study [41 .]. The odds ratio for cirrhotic patients to develop ARF was 2.18 (1.2±4.1, 95% CI, P50.01). In addition, the risk of developing multiple organ failure in ARF was 9.06 higher if liver failure was present [41 .]. Another prospective observational study [42 .] examined the risk factors for poor outcome in patients with acute GIB. Cirrhosis was one of the four independent predictors of poor outcome (de®ned as recurrent GIB, emergency surgery, or death) in these patients. In a

710 Anaesthesia and medical disease Table 2. Recent publications concerning morbidity and mortality in cirrhotic patients undergoing various surgical procedures Surgical procedure Reference no. Number of patients Mean age (years) Child's class (%) A B C Study period 30 Day mortality (%) Odds ratio for mortality Morbidity (%) Postoperative hospital stay (days)

Esophagectomy

Appendectomy

TURP

Thoracoscopy talc poudrage

Umbilical heniorrhaphy

[36 .] 18 64.4 61.1 38.9

[37] 69 N/A N/A

[38] 30 67.3 N/A

[39 .] 18 57.6 N/A

[40] 15 57 N/A

1981±1999 16.7 N/A 83.3 N/A

1977±1993 9 8 N/A N/A

1977±1993 6.7 3 N/A N/A

1985±1999 27.8 N/A 57.1 15.1

1994±1997 0 N/A 0 3±6

N/A, not available; TURP, transurethral resection of prostate.

multiple logistic regression analysis of the relationship of selected predictor with poor outcome, the odds ratio for poor outcome in cirrhotic patients was 2.27 (1.2±4.3, 95% CI, P=0.01).

Minimizing the risk

Careful preoperative assessment and close perioperative monitoring are of utmost importance in cirrhotic patients. After assessing the Child±Pugh classi®cation, the anesthesiologist should evaluate the clinical condition requiring surgical intervention. If time permits, hepatological evaluation and the diagnosis and treatment of the underlying liver disease should be carried out. A review of organ systems should be accomplished, speci®cally looking for encephalopathy, ascites, portal hypertension with variceal bleeding and renal insuf®ciency. Laboratory tests, including complete blood count, coagulation studies, liver and renal function tests, liver enzymes, plasma proteins, arterial blood gas, electrolytes and glucose, are usually helpful. Certain associated medical problems (coagulation, volume status, electrolytes, malnutrition) can frequently be improved before elective surgery. Invasive procedures should be performed after treating any coagulopathy with vitamin K, fresh frozen plasma or other coagulation products. Most induction drugs are well tolerated by cirrhotic patients, as recovery from their effect is mainly dependent on redistribution and not on hepatic metabolism. Volatile anesthetic agents may adversely affect the jeopardized hepatic circulation. When volatile agents are employed, iso¯urane is least disruptive of normal physiological mechanisms of liver blood ¯ow preservation [43]. The duration of action of muscle relaxants requiring liver metabolism can be prolonged in patients with hepatic disease. These medications should be titrated using a nerve stimulator. Because the elimination of atracurium and cis-atracurium is independent of liver function, many clinicians prefer their use in cirrhotic patients. Pseudocholinesterase levels tend to be below normal in cirrhosis, but are seldom low enough to increase the duration of action of

succinylcholine and mivacurium appreciably. Sedatives and opioids should be titrated, because exaggerated effects can occur in patients with hepatic encephalopathy. In addition, these medications can accumulate from altered hepatic metabolism. A detailed description of the anesthetic management and drug dose adjustments for cirrhotic patients can be found in anesthesia textbooks, or in the excellent review by Patel [44].

Conclusion

Recent studies con®rm past investigations, demonstrating high morbidity and mortality rates, and poor longterm survival in cirrhotic patients undergoing anesthesia and surgery. However, there is evidence that patients with compensated cirrhosis (Child±Pugh class A) can undergo elective surgery with an acceptable rate of morbidity and mortality. In all cirrhotic patients who must undergo anesthesia and surgery, preoperative evaluation and perioperative care are of utmost importance. The extent of surgical intervention should be limited (or, if possible, avoided) in those instances when patients have decompensated cirrhosis (Child±Pugh class B and C). Despite recent advances, poor outcomes are still likely when cirrhotic patients require major surgical interventions, emergency surgery, care after trauma, and treatment in which renal failure is a complicating factor.

References and recommended reading

Papers of particular interest, published within the annual period of review, have been highlighted as: . of special interest .. of outstanding interest 1

Uddenfeldt P, Danielsson AÊ. Primary biliary cirrhosis: survival of a cohort followed for 10 years. J Intern Med 2000; 248:292±298.

2

Ziser A, Plevak DJ, Wiesner RH, et al. Morbidity and mortality in cirrhotic patients undergoing anesthesia and surgery. Anesthesiology 1999; 90:42± 53.

Morbidity and mortality in cirrhotic patients Ziser and Plevak 711 3

Gozzeti G, Mazziotti A. Hepatic resection for tumors in cirrhotic livers. Int Surg 1987; 72:82±86.

4

Kowalski HJ, Abelman WH. The cardiac output at rest in Laennec's cirrhosis. J Clin Invest 1953; 32:1025±1033.

5

Rodriguez-Roisin R, Roca J, Agusti AG, et al. Gas exchange and pulmonary vascular reactivity in patients with liver cirrhosis. Am Rev Respir Dis 1987; 135:1085±1092.

6

Epstein M, Berk DP, Hollenberg NK, et al. Renal failure in the patient with cirrhosis. The role of active vasoconstriction. Am J Med 1970; 49:175±185.

7

Kang Y. Coagulation and liver transplantation. Transplant Proc 1993; 25:2001±2005.

8

Teloh HA. Serum proteins in hepatic disease. Ann Clin Lab Sci 1978; 8:127± 129.

9

Roberts RK, Desmond PV, Schenker S. Drug prescribing in hepatobiliary disease. Drugs 1979; 17:198±212.

10 Basile AS, Jones EA. Ammonia and GABA-ergic neurotransmission: interrelated factors in the pathogenesis of hepatic encephalopathy. Hepatology 1997; 25:1303±1305. 11 Galindo A. Hepatic circulation and hepatic function during anaesthesia and surgery. II. The effect of various anaesthetic agents. Can Anaesth Soc J 1965; 12:337±348. 12 Wu CC, Yeh DC, Lin MC, et al. Improving operative safety for cirrhotic liver . resection. Br J Surg 2001; 88:210±215. A large study on 329 cirrhotic patients, comparing two periods (1989±1994 versus 1995±1999) and risk factors for postoperative morbidity. 13 Moser MAJ, Kneteman NM, Minuk GY. Review: research toward safer . resection of the cirrhotic liver. HPB Surg 2000; 11:285±297. A comprehensive review of the problems associated with the cirrhotic liver, and the various options for surgical and medical management to improve outcome. 14 Nagasue N, Ono T, Yamanoi A, et al. Prognostic factors and survival after hepatic resection for hepatocellular carcinoma without cirrhosis. Br J Surg 2001; 88:515±522. 15 Abdel-Wahab A, El-Enein AA, Abou-Zeid M, et al. Hepatocellular carcinoma in Mansoura-Egypt: experience of 385 patients at a single center. HepatoGastroenterol 2000; 47:663±668. 16 Chiappa A, Zbar AP, Audisio RA, et al. Factors affecting survival and long-term . outcome in the cirrhotic patient undergoing hepatic resection for hepatocellular carcinoma. Eur J Surg Oncol 2000; 26:389±392. This study describes the uni- and multivariate factors for long-term survival in cirrhotic patients undergoing liver resection. 17 Nagasue N, Yukaya H, Ogawa Y, et al. Human liver regeneration after hepatic resection. A study of normal livers and livers with chronic hepatitis and cirrhosis. Ann Surg 1987; 206:30±39. 18 Belghiti J, Hiramatsu K, Benoist S, et al. Seven hundred forty seven hepatectomies in the 1990s: an update to evaluate the actual risk of liver resection. J Am Coll Surg 2000; 191:38±46. 19 Bolder U, Brune A, Schmidt S, et al. Preoperative assessment of mortality risk in hepatic resection by clinical variables: a multivariate analysis. Liver Transpl Surg 1999; 5:227±237. 20 Eguchi H, Umeshita K, Sakon M, et al. Presence of active hepatitis associated . with liver cirrhosis is a risk factor for mortality caused by posthepatectomy liver failure. Dig Dis Sci 2000; 45:1382±1388. A detailed study about the risk of mortality in cirrhotic versus chronic hepatitis patients with or without active hepatitis, undergoing hepatectomy. 21 Lee CC, Chau GY, Lui WY, et al. Better post resectional survival in female . cirrhotic patients with hepatocellular carcinoma. Hepato-Gastroenterol 2000; 47:446±449. An elegant study demonstrating the differences in disease markers between men and women with cirrhosis, and its effect on outcome. 22 Rust C, Gores GJ. Locoregional management of hepatocellular carcinoma. Clin . Liver Dis 2001; 5:161±173. A review about the surgical and non-surgical treatments for HCC. 23 Hanazaki K, Kajikawa S, Shimozawa N, et al. Hepatic resection for hepatocellular carcinoma in the elderly. J Am Coll Surg 2001; 192:38±46. 24 Hanazaki K, Kajikawa S, Shimozawa N, et al. Survival and recurrence after . hepatic resection of 386 consecutive patients with hepatocellular carcinoma. J Am Coll Surg 2000; 191:381±388. A retrospective review of the risk factors for mortality, morbidity and recurrence of HCC in a large cohort of patients.

25 Shimada M, Rikimaru T, Sugimachi K, et al. The importance of hepatic resection . for hepatocellular carcinoma originating from nonfibrotic liver. J Am Coll Surg 2000; 191:531±537. A comparison between cirrhotic and non-cirrhotic patients looking into long-term survival and the factors affecting it. 26 Mazziotti A, Maeda A, Ercolani G, et al. Isolated resection of segment 8 for liver tumors. Arch Surg 2000; 135:1224±1229. 27 Poon RTP, Fan ST, Ng IOL, et al. Different risk factors and prognosis for early . and late intrahepatic recurrence after resection of hepatocellular carcinoma. Cancer 2000; 89:500±507. A prospective study demonstrating the effect of cirrhosis on the late recurrence of HCC. 28 Conte D, Fraquelli M, Formari F, et al. Close relationship between cirrhosis and gall stones: cross-sectional and longitudinal survey. Ann Intern Med 1999; 159:49±52. 29 Bloch RS, Allaben RD, Walt AJ. Cholecystectomy in patients with cirrhosis. Arch Surg 1985; 120:669±672. 30 Cuschieri A, Dubois F, Mouiel J, et al. The European experience with laparoscopic cholecystectomy. Am J Surg 1991; 161:385±387. 31 Fernandes NF, Schwesinger WH, Hilsenbeck SG, et al. Laparoscopic . cholecystectomy and cirrhosis: a case±control study of outcomes. Liver Transpl 2000; 6:340±344. The most recent study on laparoscopic cholecystectomy and cirrhosis. A detailed retrospective review and a comparison with non-cirrhotic controls. 32 Morino M, Cavuoti G, Miglietta C, et al. Laparoscopic cholecystectomy in cirrhosis: contraindication or privileged indication? Surg Laparosc Endosc Percutan Tech 2000; 10:360±363. 33 Poggio JL, Rowland CM, Gores GJ, et al. A comparison of laparoscopic and . open cholecystectomy in patients with compensated cirrhosis and symptomatic gallstone disease. Surgery 2000; 127:405±411. This study demonstrates the advantage of LC versus OC in cirrhosis. 34 Wahlstrom K, Ney AL, Jacobson, S, et al. Trauma in cirrhosis: surgical and . hospital sequelae in patients requiring abdominal exploration. Am Surg 2000; 66:1071±1076. A large retrospective study comparing the outcome of abdominal exploration in cirrhotic versus non-cirrhotic trauma patients. 35 Maull KI, Turnage B. Trauma in the cirrhotic patient. South Med J 2001; 94:205±207. 36 Tachibana M, Kotoh T, Kinugasa S, et al. Esophageal cancer with cirrhosis of . the liver: results of esophagectomy in 18 consecutive patients. Ann Surg Oncol 2000; 7:758±763. A detailed description of perioperative variables and risk factors in a rare group of cirrhotic patients with esophageal malignancy. 37 Poulsen LO, Thulstrup AM, Sorensen HT, Vilstrup H. Appendectomy and perioperative mortality in patients with liver cirrhosis. Br J Surg 2000; 87:1664±1665. 38 Nielsen SS, Thulstrup AM, Lund L, et al. Postoperative mortality in patients with liver cirrhosis undergoing transurethral resection of the prostate: a Danish nationwide cohort study. BJU Int 2001; 87:183±186. 39 De Campos JRM, Filho LOA, Werebe EDC, et al. Thoracoscopy and talc . poudrage in the management of hepatic hydrothorax. Chest 2000; 118:13±17. The surgical management of hydrothorax in cirrhosis, its indications, limitations and complications are thoroughly discussed in this paper. 40 De la Pena CG, Fakih F, Marquez R, et al. Umbilical herniorrhaphy in cirrhotic patients: a safe approach. Eur J Surg 2000; 166:415±416. 41 De Mendonca A, Vincent JL, Suter PM, et al. Acute renal failure in the ICU: risk . factors and outcome evaluated by the SOFA score. Intensive Care Med 2000; 26:915±921. A large prospective multicenter study on the risk factors for acute renal failure in the ICU. 42 Afessa B. Triage of patients with acute gastrointestinal bleeding for intensive . care unit admission based on risk factors for poor outcome. J Clin Gastroenterol 2000; 30:281±285. In this prospective study the authors discuss the triage options of patients, according to the risk factors for complications. 43 Gelman S, Fowler KC, Smith LR. Liver circulation and function during isoflurane and halothane anesthesia. Anesthesiology 1984; 61:726±730. 44 Patel T. Surgery in the patient with liver disease. Mayo Clin Proc 1999; 74:593±599.