Reducing the load: The evolution and management of obesity and nonalcoholic steatohepatitis before liver transplantation


  • Kymberly D. Watt

    Corresponding author
    1. Division of Gastroenterology and Hepatology, Mayo Clinic and Foundation, Rochester, MN
    • Division of Gastroenterology and Hepatology, Mayo Clinic and Foundation, 200 First Street Southwest, Rochester, MN 55905
    Search for more papers by this author
    • Telephone: 507-266-1586; FAX: 507-266-1856

  • Potential conflict of interest: Nothing to report.


Key Points

1. Obesity is common in patients with end-stage liver disease of any cause.

2. Obesity is associated with comorbidities that can affect liver transplant candidacy, such as metabolic syndrome (diabetes, hyperlipidemia, and hypertension), cardiovascular disease, pulmonary disorders (related to obstructive sleep apnea), renal dysfunction, and malignancies.

3. Nonalcoholic steatohepatitis is the only indication for transplantation that is increasing in frequency, and it may soon become the leading indication.

4. There is no set body mass index above which liver transplantation is contraindicated. Transplant candidacy and outcomes depend on the cumulative comorbidities of the individual patient.

5. Weight loss is an important component of metabolic syndrome management and is recommended before liver transplantation.

6. Bariatric surgery may be considered in carefully selected patients with well-compensated cirrhosis but is not recommended in patients with decompensated cirrhosis. Liver Transpl 18:S-S, 2012. © 2012 AASLD.


AIH, autoimmune hepatitis; ALD, alcoholic liver disease; BMI, body mass index; CAD, coronary artery disease; CC, cryptogenic cirrhosis; HBV, hepatitis B virus; HCV, hepatitis C virus; NASH, nonalcoholic steatohepatitis; PBC, primary biliary cirrhosis; PSC, primary sclerosing cholangitis; SRTR, Scientific Registry of Transplant Recipients.

The World Health Organization categorizes obesity according to body mass index (BMI) ranges (overweight, 25-29.9 kg/m2; class I obesity, 30-34.9 kg/m2; class II obesity, 35-39.9 kg/m2; and class III obesity, ≥40 kg/m2) and has affirmed that the worldwide prevalence of obesity has doubled since 1980.1 Between 1986 and 2000, the prevalence of class III obesity quadrupled, and the prevalence of BMIs ≥ 50 kg/m2 quintupled.2 The 2008/2010 World Health Organization statistics are staggering, with more than 1.5 billion adults determined to be overweight and more than 200 million men and nearly 300 million women found to be obese. It is even more worrisome that 43 million children less than 5 years old were found to be overweight. Overall, more than 1 in 10 of the world's adults and 20% to 30% of all people in North American and European countries are obese.

Obesity is a serious health concern because it is a major risk factor for many diseases, including diabetes, hypertension, hyperlipidemia, cardiovascular diseases, chronic liver disease and cirrhosis, musculoskeletal disorders, renal disease, and malignancies.


The risk of developing fatty liver disease is 5 times greater for obese individuals versus nonobese individuals.3 Fatty liver can be demonstrated on ultrasound in 20% to 32% of the general population4, 5 and more specifically in 10% to 15% of nonobese individuals and in up to 76% of obese individuals.3 In patients with biopsy-confirmed nonalcoholic steatohepatitis (NASH), obesity is independently associated with fibrosis progression.6 NASH is the only disease etiology that is increasing in frequency as the primary indication for liver transplantation in the United States, and it is currently the third most common indication for transplantation7 (Fig. 1). This may become even more dramatic as the eradication rates for hepatitis C virus (HCV) increase. Posttransplant survival rates for patients with NASH appear similar to those for patients with other diseases.7–10 However, a group of high-risk NASH patients who were older than 60 years and had a BMI > 30 kg/m2, pretransplant diabetes, and hypertension were found to be at particularly high risk for worse outcomes. These select patients experienced an immediate mortality rate of 25% and a 1-year mortality rate of 50% in comparison with Model for End-Stage Liver Disease score–matched controls.9

Figure 1.

Frequencies of specific indications for liver transplantation among adults in the United States. Reprinted with permission from Gastroenterology.7 Copyright 2011, AGA Institute.

Fatty liver disease can also be superimposed upon patients with other chronic liver diseases, such as viral hepatitis and alcohol-related disease. In some cases, such as patients with HCV11 and even patients with an alpha-1-antitrypsin deficiency,12 this can result in more severe and more rapidly progressive disease. Obesity has also been associated with reduced rates of response to antiviral therapy in HCV-infected patients.13, 14 Obesity and fatty liver are notable in up to 30% of Asian patients with hepatitis B virus (HBV), but they have no apparent impact on disease severity and may be associated with lower HBV viral loads.15, 16

In alcohol-related liver disease, the combination of increased BMI and alcohol use may be additive: the combination of obesity and more than 2 drinks of alcohol per day has been associated with elevated liver transaminases, which are assumed to be related to fatty liver disease.17 In nonobese patients with heavy alcohol exposure, the risk of steatosis is increased 3-fold in comparison with nondrinkers, whereas the risk is 4.6-fold for obese nondrinkers and almost 6-fold for obese heavy drinkers.3

Obese patients with compensated cirrhosis of any etiology are at increased risk of decompensation over a period of 2 to 5 years in comparison with normal-weight or overweight patients, with a 6% increase for each BMI point increase.18 The cumulative probabilities of clinical decompensation at 2 and 5 years were found to be 0% and 13%, respectively, for normal-weight patients, 14% and 28%, respectively, for overweight patients, and 21% and 37%, respectively, for obese patients18 (Fig. 2). In that study, obese patients were less likely to experience a reduction in the hepatic venous pressure gradient than overweight and normal-weight patients (who did experience a reduction), even though 50% of the obese patients received beta-blocker therapy.18

Figure 2.

Probability of first clinical decompensation of cirrhosis according to BMI. Reprinted with permission from Hepatology.18 Copyright 2011, American Association for the Study of Liver Diseases.

Obese patients are also at higher risk for hepatocellular carcinoma and have an 83% increased risk of primary liver cancer (mostly hepatocellular cancer) in comparison with the general population.19 This risk is highest in patients with cirrhosis and HCV (but not HBV).19 Steatosis in patients with HCV cirrhosis has been identified as an additional risk factor for hepatocellular carcinoma, and the risk increases as the grade of steatosis within the liver increases.20


Transplant Evaluation of the Obese Patient

With obesity come additional comorbidities that affect outcomes both before and after liver transplantation. A thorough pretransplant evaluation is required. The well-established association with metabolic syndrome and diabetes generally leads to a rigorous workup for covert cardiovascular disease in the pretransplant evaluation. Metabolic syndrome (particularly hypertension and hyperlipidemia) and cardiovascular disease can be masked by the effects of end-stage liver disease, so the use of standard cardiovascular risk assessment tools can be falsely reassuring. Noninvasive testing for coronary artery disease (CAD)—particularly nuclear stress imaging—in the pre–liver transplant population is suboptimal because end-stage liver disease results in low systemic vascular resistance and hypotension, and many patients take beta-blockers.21–23 It is generally felt that normal dobutamine stress echocardiography in low-risk patients effectively rules out significant CAD, but the definition of low risk is generally a patient with no cardiac risk factors. Coronary calcification studies may be helpful in determining higher risk CAD patients because standard cardiac risk factors and risk scores correlate with increased calcification scores, but data on perioperative or posttransplant outcomes are not available.24 The noncontrast computed tomography scan can be accessed by all providers and certainly requires further investigation. Biomarkers such as serum troponin and homocysteine levels (but not C-reactive protein levels) have been identified as additional risk markers for covert CAD in the end-stage liver disease population.25, 26 Obesity alone is not typically included in cardiac risk models (eg, Framingham risk scores), but it has been independently associated with an elevated cardiovascular risk.27 Thus, until a better risk model exists for the obese patient undergoing the evaluation for transplantation with at least 1 cardiac risk factor28–30 (Table 1), it would be reasonable to pursue dobutamine stress echocardiography with or without serum biomarkers (troponin and homocysteine) and, on the basis of these results and clinical judgment, coronary angiography. Unfortunately, no clear consensus exists for the management of CAD in the pre–liver transplant patient once it is identified.31 This is discussed further elsewhere in this syllabus.32 It must also be remembered that these patients are at risk not only for CAD but also for cardiomyopathy, cerebrovascular disease, and pulmonary consequences of obstructive sleep apnea (including pulmonary hypertension, right-sided heart failure, and restrictive lung disease). Thus, electrocardiography, chest X-rays, transthoracic echocardiography, carotid artery Doppler studies, and pulmonary function testing are warranted.

Table 1. Cardiac Risk Assessment in the General Population
  1. NOTE: The data for this table were taken from Vasan et al.,28 Yusuf et al.,29 and D'Agostino et al.30

Risks equivalent to a previous history of CAD
• Previous or current carotid artery disease, peripheral vascular disease, or abdominal aortic aneurysm
• Diabetes
• Current smoking
• Chronic renal failure (defined as a creatinine glomerular filtration rate < 60 mL/minute)
Major risk factors
• Age > 45 (males) or 55 years (females)
• Hypertension (systolic blood pressure > 140 mm Hg and diastolic blood pressure > 90 mm Hg)
• Total cholesterol level > 240 mg/dL
• Low-density lipoprotein cholesterol level > 160 mg/dL
• High-density lipoprotein cholesterol level < 40 mg/dL
• Family history [defined as myocardial infarction or death from CAD in a first-degree relative before the age of 50 (males) or 60 years (females)]
• Obesity
Borderline risk factors
• Systolic pressure of 120-139 mm Hg
• Diastolic pressure of 80-89 mm Hg
• Low-density lipoprotein cholesterol level of 100-159 mg/dL (2.6-4.1 mmol/L)
• High-density lipoprotein cholesterol level of 40-59 mg/dL (1.0-1.5 mmol/L)
• Impaired fasting glucose without overt diabetes
• Past history of smoking (highest within 2 years of quitting)
• Nonalcoholic fatty liver disease

Additionally, obesity has recently been established as a risk factor for renal injury and chronic renal failure.33 It has been suggested that obese women (relative risk = 1.92) are at even greater risk than men (relative risk = 1.49) for chronic kidney disease.33 A careful evaluation of measured renal function should be considered because sarcopenia is common in the obese population and creatinine may not adequately represent renal function.

According to the updated list at the National Cancer Institute,34 obesity is associated with increased risks of cancers of the esophagus, breast (postmenopausal), endometrium, colon and rectum, kidneys, pancreas, thyroid, and gallbladder and possibly other types of cancer. Patients should be assessed for the risks of these malignancies before transplantation and should be screened accordingly.

Nutritional Evaluation

In any patient with end-stage liver disease, cachexia, malnutrition, and both lean body mass and fat mass loss are common, and obese patients and NASH patients are no exception.35, 36 Nutritional assessment in end-stage liver disease has been reviewed elsewhere in detail37 and should be performed for all patients, including obese patients. Most obese patients have tried for years to lose weight without sustained success. Lifestyle modifications and caloric restriction can work in highly motivated patients but work best in a highly monitored setting. This presents a challenge for most patients. In patients with NASH, weight loss has resulted in histological improvements.38 In liver transplant patients, help from an experienced dietician is beneficial for avoiding protein-calorie malnutrition and vitamin deficiencies. Patients with cirrhosis are frequently and erroneously placed on low-protein diets, which increase the risk of sarcopenia and worsen their overall nutritional status.

Medical Management of Obesity and NASH Before Transplantation

The effects of pharmacological interventions in both obese patients and NASH patients have been disappointing overall, with small amounts of weight being lost that would have minimal clinical relevance in the transplant setting. These agents have been recently reviewed in detail elsewhere.39 At this point, vitamin E in NASH patients may be helpful in reducing intrahepatic inflammation, but a reducing effect on fibrosis is less clear in 2-year follow-up biopsy samples.40 Unfortunately, vitamin E has been associated with a 17% increased risk of prostate cancer in a large prospective control study, and this further limits its use in men who are 50 years old or older.41 Oral hypoglycemics (including peroxisome proliferator-activated receptor agonists) have shown promise,40 but the heavily publicized adverse event profile has so far dampened the enthusiasm. There are interesting associations between the intestinal microbiome and both obesity and NASH,42 but further investigation into therapeutic measures is needed. Thus, at this time, there is no obvious role for pharmacological interventions in the already cirrhotic NASH patient awaiting liver transplantation.

The medical management of metabolic syndrome before transplantation has been reviewed extensively elsewhere.43 In brief, diabetes is managed with insulin or oral hypoglycemics, but the latter may increase the risk of hypoglycemia in patients with advanced liver disease and should be used only with caution and careful follow-up. The goals of management are similar to the goals in the general population because there are no studies specific to the patient with cirrhosis. These goals include a glycosylated hemoglobin level < 7%, a fasting blood sugar level of 70 to 130 mg/dL (3.9-7.2 mmol/L), and a peak postprandial glucose level < 180 mg/dL (10 mmol/L).44 The blood pressure goals are <140/90 mm Hg for nondiabetic patients and <130/80 mm Hg for diabetic patients (per the Adult Treatment Panel III criteria) with hypertension before transplantation. No data exist on optimal lipid goals for patients with cirrhosis, but the treatment should follow the same guidelines used for the general population until more data are available.

Bariatric surgery in patients with noncirrhotic NASH has been well tolerated and results not only in improved weight loss and improvements in metabolic syndrome but also in improvements in histological NASH.45 Bariatric surgery has been performed in patients with cirrhosis. Previous small retrospective studies in patients incidentally found to have cirrhosis suggested outcomes similar to those for patients without cirrhosis.46 However, a recent, large national database study comparing outcomes for all patients undergoing bariatric surgery with or without underlying cirrhosis found a 2-fold higher risk of perioperative mortality in patients with compensated cirrhosis and a >20-fold higher risk in patients with decompensated cirrhosis in comparison with individuals without cirrhosis.47 It should be noted that 40% of the patients with decompensated cirrhosis had surgery at low-volume centers (<50 procedures per year), and this was also an independent risk factor for perioperative mortality. On further review, it was found that the in-hospital postoperative mortality rate was 41% for patients with decompensated cirrhosis at low-volume centers, whereas at high-volume centers, no deaths were noted.47

A case series of 6 patients with some degree of portal hypertension (with only small esophageal varices and no ascites) who received a gastric sleeve before liver transplantation has been described.48 These patients lost 24% to 40% of their excess body weight, their liver function remained stable, and their metabolic syndrome comorbidities all improved. These patients were likely a highly select cohort, and care must be taken in generalizing these results. Because the risk of performing any surgery in patients with end-stage liver disease is high, a bariatric procedure before transplantation for most patients is not possible. One patient was described as undergoing an intraoperative placement of a gastric band at the time of liver retransplantation, and this resulted in a 45% excess body weight reduction and an improvement of her metabolic disturbances.49 At the Mayo Clinic, 7 patients have undergone intraoperative gastric sleeve surgery at the time of liver transplantation in a pilot series: minimal additional time was added to the transplant procedure, and so far, there has been no morbidity in 6 to 24 months of follow-up for 6 of the 7 patients.50 One patient experienced a gastric leak with a complicating intra-abdominal infection that has since resolved, and he was well with normal liver function and a BMI of 24 kg/m2 6 months after transplantation. Further investigation and experience are needed for the combination of a restrictive bariatric procedure and liver transplantation.


Obesity is common before liver transplantation and increases afterward. Although retrospective studies of large databases suggested worse survival for patients with morbid obesity,51 the BMI values were not corrected for ascites. Pretransplant BMI values, when they were corrected for ascites at the time of transplantation, were not independently predictive of patient or graft survival in other studies.52, 53 Correcting for ascites resulted in a lower obesity category for 12% to 20% of patients.52 Importantly, this study demonstrated a 7% increased risk of mortality associated with each liter of ascites removed, and this suggested that the lower survival rate was related more to ascites and advanced liver disease than body weight. Morbidly obese patients (BMI > 40 kg/m2) may have higher frequencies of morbidities such as prolonged hospitalization and readmission as well as infectious, wound, and cardiovascular complications after transplantation.53 As for pediatric patients, although early posttransplant survival was no different for obese children, long-term survival (12 years) was lower for children who were obese at the time of transplantation. However, these children were appreciably sicker at the time of transplantation.54

The technical aspects of surgery can be very different and depend on the distribution of adipose tissue (Fig. 3). Intra-abdominal adiposity creates a technically more challenging dissection, but no data exist on whether it increases perioperative morbidity or mortality in liver transplant patients. In a recent study of patients undergoing hepatectomy for colon cancer metastasis, intra-abdominal fat (defined as perinephric fat) was associated with more postoperative complications and was independently predictive of poor survival.55 Sarcopenia (or reduced muscle mass), however, can equally affect survival in this population of colon cancer resection patients,56 and the combination of obesity and sarcopenia results in even worse outcomes than either entity alone.57 In liver transplant recipients, sarcopenia has been found to be an important determinant of increased mortality after liver transplantation, but the effects of added obesity have not been studied.58

Figure 3.

Subcutaneous and intra-abdominal adipose tissue distribution.

Transplant recipients gain weight after transplantation, with the majority of the weight gain occurring within the first 6 to 12 months.59, 60 With few exceptions, patient who are overweight or obese before transplantation will remain overweight or obese afterward, and up to 50% of normal-weight patients at the time of transplantation can become overweight or obese after transplantation.60 The delta weight change from the pretransplant period to 1 year after transplantation has been independently associated with an increased risk of posttransplant metabolic syndrome,61 and metabolic syndrome is associated with worse long-term outcomes and posttransplant cardiovascular disease.62


The prevalence of obesity continues to rise worldwide and has many repercussions for the liver transplant community. There is no set BMI that absolutely precludes transplantation, but a thorough evaluation of the technical feasibility and comorbid conditions is needed. The role of weight loss in improving liver transplant candidacy and the timing, risks, and benefits of bariatric procedures in this population require more investigation.