Cirrhosis as a Perioperative Risk Factor (see Chapter 56 )

Cirrhosis is a syndrome of end-stage liver disease pathologically characterized by severe fibrosis and nodular regeneration of the liver parenchyma. Alcohol abuse remains the most common cause of cirrhosis. Other etiologies include chronic hepatitis, primary biliary cirrhosis, hemochromatosis, Wilson's disease, and idiopathic cirrhosis. The Child-Pugh scoring system is the most frequently used tool to predict perioperative risk in cirrhosis patients undergoing abdominal surgery, exclusive of portosystemic shunt procedures ( Table 55-1 ).^[107] This scoring system assigns points based on levels of serum albumin and bilirubin, the INR, the degree of ascites, and the presence and grade of encephalopathy and stratifies risk in order of severity as class A, B, or C. A number of studies have used this system to predict perioperative outcome in cirrhosis patients undergoing a variety of surgeries.

Garrison and coworkers^[108] retrospectively evaluated outcomes in 100 patients with histologically proven cirrhosis who were undergoing surgery predominantly for biliary tract procedures (cholecystectomy, choledochotomy), as well as for gastroduodenal repair, colon and small bowel resection, and open liver biopsy. An overall operative mortality of 30% and an additional perioperative morbidity of 30% were noted, with sepsis-mediated multiorgan system failure being the major cause of death (87%). When stratified to Child-Pugh classes A, B, or C, mortality was 10%, 31%, and 76%, respectively. Excluding the Child-Pugh classification, the authors also performed a multivariate analysis of other perioperative variables and concluded that preoperative prolongation of the

2218

Rice and colleagues,^[114] however, described the lack of predictability of the Child-Pugh classification in patients with chronic liver failure undergoing not only abdominal surgery but also coronary bypass grafting, orthopedic procedures, and other peripheral surgeries as well. In this retrospective analysis of 40 patients, perioperative prolongation of the INR and clinical evidence of encephalopathy were the only variables most associated with increased mortality (10 and 35 times that of normal individuals, respectively).

Ziser and coworkers,^[115] in the largest of all these retrospective studies, established an 11.6% 30-day mortality rate in 773 patients with cirrhosis who were undergoing a variety of surgical procedures involving local, regional, or general anesthesia. Multivariate analysis identified male gender, Child-Pugh class C, ascites, azotemia, perioperative infection, higher American Society of Anesthesiologists (ASA) physical classification, a diagnosis of cryptogenic cirrhosis, and surgery on the respiratory system as risk factors independently associated with mortality. Further analysis identified additional risk factors for perioperative complications ( Fig. 55-6 ).

Despite the lack of prospective studies, cirrhosis is still thought to be a major risk factor for patients undergoing nonhepatic surgery, with elective surgery considered contraindicated in Child-Pugh class C patients.^[116] In addition, it is probably prudent to avoid elective surgery in cirrhotic patients with a prolonged INR, hypoalbuminemia, and evidence of preoperative infection or encephalopathy. Alternatives to extensive abdominal procedures, such as cholecystostomy or laparoscopic cholecystectomy versus open cholecystectomy, should be considered for the treatment of cholelithiasis.^{[117] [118]} Preoperative decompression of portal hypertension by transjugular intrahepatic portosystemic shunting (TIPS) may also improve postoperative outcomes in certain patients.^[119] Finally, elective abdominal surgery should probably be avoided if a diagnosis of cirrhosis is made fortuitously at the time of surgery.^[102]