Risk stratification and preoperative evaluation of the patient with known or suspected liver disease

Watch a video presentation of this article

Watch the interview with the author

Abbreviations

ASA

American Society of Anesthesiologists

CTP

Child‐Turcotte‐Pugh

INR

international normalized ratio

MELD

Model for End‐Stage Liver Disease

NAFLD

nonalcoholic fatty liver disease

NASH

nonalcoholic steatohepatitis

OR

operating room

TIPS

transjugular intrahepatic portosystemic shunt

Patients with liver disease have a unique pathophysiology that results in the need for a specialized evaluation before undergoing any surgical procedure. This article reviews strategies to assess surgical risk and optimize these patients.

Timing of Surgery

The first consideration for the timing of surgery is whether the surgery is emergent or elective. Emergent cases should proceed without delay, but with the understanding that those with decompensated cirrhosis have poorer surgical outcomes than those with normal liver function.1 Elective cases are contraindicated in several clinical scenarios listed in Table 1 and include patients with acute liver failure, acute alcoholic hepatitis, and acute viral hepatitis.2 Patients for whom there are no absolute contraindications should undergo individualized risk stratification based on the cause of their underlying liver disease, the degree of liver dysfunction, often indicated by the presence or absence of portal hypertension, and the type of surgical procedure.

Table 1.

Contraindications for Elective Procedures in Patients With Liver Disease

Acute liver failure

Acute viral hepatitis

Acute alcoholic hepatitis

Acute renal failure

Severe coagulopathy (despite correction)

Hypoxemia

Cardiomyopathy

ASA class V

Assessment of Liver Disease Severity

A key distinction that helps prognosticate postoperative outcomes is the presence or absence of cirrhosis. Cirrhosis is a state of decreased systemic vascular resistance, and blood flow to the liver is often reduced, especially in the presence of portal hypertension. Surgical procedures are often a state of constant hemodynamic shifts because of the vasoactive properties of anaesthesia and intraoperative blood loss. Cirrhotic patients are highly susceptible to these shifts, which result in hypoxemic injury to the liver and can lead to hepatic decompensation. Patients with chronic liver disease, but no cirrhosis do not have this susceptible hemodynamic profile and should be risk‐stratified similarly to the general population.3 Those who have cirrhosis should undergo further testing to identify the presence of clinically significant portal hypertension, because this increases their risk for hypoxemic tissue injury. Indicators of significant portal hypertension include thrombocytopenia (platelets <150), the presence of splenomegaly, varices, or ascites, and a hepatic venous pressure gradient greater than 10.

Prediction Models of Disease Severity

Once the diagnosis of cirrhosis is established, further risk stratification depends largely on the degree of hepatocellular dysfunction. The two most commonly used scores to stratify liver disease severity are the Child‐Turcotte‐Pugh (CTP) and the Model for End‐Stage Liver Disease (MELD). Both scores have been applied to preoperative risk stratification, with higher MELD and CTP scores correlating with higher 30‐day mortality. The CTP score emphasizes the sequelae of portal hypertension have been validated in several small retrospective studies with mortality rates as high as 30% to 31% for Child class B and 76% to 82% for Child class C cirrhotic patients.4, 5 A large retrospective study of 772 individuals who underwent digestive, orthopedic, or cardiovascular surgeries demonstrated that the MELD score was an independent predictor of postoperative mortality. For every MELD score point increase above 8, there was a 14% increase in mortality in the first 30 to 90 days. Median survival based on the MELD score is shown in Table 2.6 There is still a debate whether the CTP or MELD score is more accurate at predicting postoperative mortality. Although the CTP is criticized for its subjectivity and heterogeneity within a class, a patient with severe portal hypertension may not have a MELD score that reflects the severity of their illness. Age older than 70 years and an American Society of Anesthesiologists (ASA) physical status class of 4 or 5 have also been shown to be independent predictors of mortality and have been incorporated into prediction models such as the Mayo MELD.

Table 2.

Median Survival After Surgery by Model for End‐Stage Liver Disease Category

MELD	0‐7	8‐11	12‐15	16‐20	21‐25	>26
Median survival	4.8 years	3.4 years	1.6 years	64 days	23 days	14 days

Adapted with permission from Gastroenterology.6 Copyright 2007, AGA Institute.

Surgical Considerations

Biliary Surgery

Cirrhotic patients have a susceptibility to development of gallstones. There is a risk for postoperative hepatic decompensation that occurs in 7.7% of laparoscopic and 18.1% of open cholecystectomy cases.7 Studies have shown that patients with MELD scores up to 11 to 13, Child class A and Child class B cirrhosis without portal hypertension may undergo intra‐abdominal procedures such as cholecystectomy for symptomatic gallstones. Mortality rate with these parameters is about 0% to 6%, with the laparoscopic approach preferred over open cholecystectomy.8

Hernias

Hernias are relatively common in cirrhotic patients because of the increased intra‐abdominal pressure from ascites. Umbilical hernias should be prophylactically repaired before the development of ascites, but if ascites is already present, it should be aggressively controlled before repair because it can contribute to wound dehiscence and poor healing. Repair has been successfully performed even in Child class C cirrhotic patients; however, it should not be considered in patients with high MELD scores or hypoalbuminemia because of increased infectious complications.

Bariatric Surgery

With increasing numbers of nonalcoholic steatohepatitis (NASH) patients, bariatric surgery has come to the forefront as a potential treatment because it has been shown to decrease steatosis, inflammation, and fibrosis as patients lose weight. Patients without advanced fibrosis are not at increased risk for postoperative complications, and multiple studies have shown that it can be performed safely in compensated cirrhotic patients. A national study of patients undergoing bariatric surgery from 1998 to 2007 showed a mortality rate of 0.9% for compensated cirrhotic patients.9 In a questionnaire of bariatric surgeons worldwide who proceeded with bariatric surgery after finding cirrhosis incidentally during the procedure, the pooled perioperative mortality rate was approximately 3.2%.10

Cardiac Surgery

Cardiac surgeries such as coronary artery bypass grafting and valve replacement have increased postoperative mortality in cirrhotic patients because of hemodynamic shifts of cardiopulmonary bypass. A study of cirrhotic patients undergoing cardiopulmonary bypass showed that those with a CTP score greater than 8 had a postoperative 90‐day mortality rate of 4.6%, similar to that of propensity matched control subjects. Those with a CTP score greater than 8 had a 90‐day mortality rate of 70%.11 With the advent of percutaneous valve replacement, patients with chronic liver disease may have safer options.

Transjugular Intrahepatic Portosystemic Shunt

Given the impact of portal hypertension on postsurgical mortality, a few studies have evaluated the use of transjugular intrahepatic portosystemic shunt (TIPS) to reduce portal hypertension preoperatively to improve survival. The largest study, consisting of 25 patients with a median MELD score of 15 (28% Child class C), showed a perioperative mortality risk rate of 12% with prophylactic TIPS placement. Thus, placement 4 to 6 weeks before a procedure may allow patients to undergo surgery that would typically be contraindicated.12

Disease‐Specific Considerations

Nonalcoholic Fatty Liver Disease

Fatty liver disease has long been associated with the metabolic syndrome composed of hyperlipidemia, obesity, diabetes, and elevated body mass index. Although these features all increase cardiovascular risk in patients without liver disease, nonalcoholic fatty liver disease (NAFLD) itself is thought to also increase cardiovascular risk. Although the mechanisms are still unclear, patients with increased endothelial dysfunction induced by the proinflammatory state of NAFLD have higher numbers of cardiovascular events.13 Patients with NASH undergoing liver transplantation have a higher risk for postoperative cardiovascular events; however, their long‐term postoperative mortality is similar to cirrhotic patients undergoing transplant for other indications.14 There are no formal recommendations for preoperative testing in NASH patients, but given the increased cardiovascular risk, they should receive cardiac clearance before major surgeries and be carefully monitored postoperatively.

Wilson's Disease

Patients using d‐penicillamine should have the dose reduced before elective surgeries because it can interfere with wound healing.

Autoimmune Liver Disease

Steroids are commonly used in patients with autoimmune conditions. Patients receiving steroids may require stress‐dose steroids surrounding the perioperative period to prevent complications of hypotension.

Preoperative Laboratory Evaluation

All liver patients should be evaluated with a complete metabolic panel, complete blood count, prothrombin time, international normalized ratio (INR) and activated partial thromboplastin time, and fibrinogen preoperatively because these tests may be abnormal in patients with liver disease and influence surgical outcomes.

Thrombocytopenia

Due to hypersplenism and thrombopoietin deficiency, liver patients are prone to thrombocytopenia. A platelet transfusion for a goal greater than 50,000 in those undergoing moderate‐risk surgery and greater than 100,000 in high‐risk surgery has been traditionally recommended.15

Coagulopathy

Patients with advanced liver disease may have an elevated INR because of the inability of the liver to synthesize appropriate clotting factors. However, cirrhotic individuals also have a deficit in anticlotting factors, making them prone to both bleeding and thrombosis. Because the INR measures only a piece of the thrombin pathway, it is believed to be an inaccurate estimate of a patient's true bleeding risk. Tests such as thromboelastography and rotational thromboelastometry measure clot formation, strength, and dissolution and have been used in transplant surgeries to better predict disorders in coagulopathy and reduce the need for transfusion in cirrhotic patients.16 Correction of coagulopathy can be accomplished with 10 mg intravenous vitamin K for 3 days. Cryoprecipitate may be administered in patients with a fibrinogen level less than 100. Fresh‐frozen plasma may also be used perioperatively as well.

Renal Dysfunction

Splanchnic dilatation seen in cirrhotic patients can result in decreased renal perfusion and renal insufficiency. Patients with renal dysfunction of preoperative laboratory results will be more sensitive to intraoperative fluid loss and should be monitored closely.

Infectious Risk

Preoperative antibiotics should be administered per general surgery guidelines. There are no recommendations for extended courses of antibiotics in patients with chronic liver disease.

Summary

Surgical risk for patients with liver disease increases in those with advanced portal hypertension and acute liver injury. Although a general approach is shown in Figure 1, risk stratification should be individualized. Improved prognostication with tools such as the MELD score and preoperative optimization can help guide a discussion of the risks and benefits to help patients make educated decisions about invasive procedures.

Figure 1.

Recommended approach for preoperative risk assessment. Abbreviation: OR, operating room.