Nonalcoholic steatohepatitis and nonalcoholic fatty liver disease

How can I be sure that the patient has nonalcoholic fatty liver disease or nonalcoholic steatohepatitis?

Nonalcoholic fatty liver disease (NAFLD) can be diagnosed by liver biopsy or imaging studies in patients who do not abuse alcohol (<2 drinks/day is unlikely to contribute to fatty liver, > 5 drinks/day has a good chance of playing a role, 3-5 drinks/day is a gray zone – alcohol may or may not be playing a role).

Nonalcoholic steatohepatitis (NASH) can only be diagnosed by liver biopsy. Noninvasive markers can increase or decrease the likelihood that a patient has NASH but cannot establish the diagnosis. Normal ALT (alanine aminotransferase) and AST (aspartate aminotransferase) levels do not exclude NASH.

Few diseases need to be ruled out to establish a diagnosis of NAFLD or NASH. Wilson’s disease always should be considered, especially if the patient’s age is less than 35 years (diagnosed with a low serum ceruloplasmin and high 24-hour urine copper). Other diseases can coexist with NAFLD or NASH and thus do not need to be excluded to establish a diagnosis of NAFLD or NASH.

A tabular or chart listing of features and signs and symptoms

Signs and symptoms

NAFLD can cause fatigue and right upper quadrant abdominal pain. The presence of such symptoms does not help distinguish patients with just steatosis from patients with NASH. The pain occurs in about a third of patients and is variable in character and severity. Some patients have only a mild aching fullness sensation, whereas others complain of recurrent or steady sharp right upper quadrant abdominal pains that can keep them awake at night. The pain may be caused by distention of the liver capsule, although this cannot be proven. Irritable bowel syndrome and cholelithiasis are other considerations for such pain.

There are no signs or symptoms that specifically indicate NASH.

Physical exam

Hepatomegaly is the only physical exam finding that might point to NAFLD, although NAFLD can be difficult to detect in the obese patient and is nonspecific when it is found.

Common coexisting findings that reflect the underlying insulin-resistant state include centripetal obesity, hirsutism of polycystic ovarian syndrome (PCOS), and acanthosis nigricans (pigmentation typically over the elbows, knees, and knuckles).

How can I confirm the diagnosis?


– Elevated aminotransferases (ALT and AST) are often found in patients with the spectrum of NAFLD, from simple steatosis to NASH.

– The ALT and AST can be within the laboratory specified normal range in patients with NASH. This has been repeatedly demonstrated in patients undergoing bariatric surgery. They have the full spectrum of NASH with normal liver enzymes.

– Aminotransferase levels often drop into the normal range as NASH progresses to cirrhosis.

– The ALT is typically higher (1.5- to 2.5-fold higher than the AST in NAFLD), a feature that helps distinguish NAFLD from alcoholic liver disease. In alcoholic liver disease, the AST is higher than the ALT. However, it is important to note that as NASH progresses to cirrhosis, the AST can become higher than the ALT. Thus, if cirrhosis is suspected, the ALT/AST ratio is no longer helpful in distinguishing NASH from alcoholic liver disease.


Ultrasound. A bright liver appearance by ultrasound is typical of severe steatosis, but cirrhosis can also cause a bright appearance of the liver parenchyma. The brightness of cirrhosis is typically described as courser in texture, but cirrhosis can be mistaken for steatosis and vice versa. Milder degrees of steatosis (even with NASH) can be difficult to detect by ultrasound. Therefore, a negative ultrasound does not rule out NAFLD or NASH. The increased echogenicity can cause problems in properly identifying other echodense structures such as bile ducts and hemangiomas, which are usually identified by their being more echodense than the surrounding liver parenchyma.

CT. A low density liver by CT imaging is nearly pathognomonic of NAFLD, although rare lipid storage disorders may mimic this appearance. CT in NAFLD shows a low density liver compared to the spleen or paraspinous muscles. Regions with sparing can sometimes be seen adjacent to the gallbladder or in focal areas of the hilum. Focal areas of fat can sometimes be seen as well and are typically geographic in shape and do not displace the vascular structures.

MRI. Magnetic resonance imaging (MRI) has the greatest sensitivity for NALFD, but it cannot distinguish simple steatosis from NASH.

Liver biopsy

The presence of hepatocyte lipid droplets of varying size, a mixed lobular inflammatory infiltrate (lymphocytes and PMNs), hepatocyte ballooning, small or indistinct Mallory-Denk bodies, and perisinusoidal fibrosis are characteristic of NASH in the absence of alcohol abuse.

A predominance of portal inflammation raises the question of other contributing factors (e.g., hepatitis C), but portal inflammation can be out of proportion to lobular inflammation in pediatric NASH. Prominent Mallory-Denk bodies can suggest alcohol abuse as the cause of the steatohepatitis.

Liver biopsy is the only means of detecting early fibrosis.

Liver biopsy is also useful in detecting unsuspected contributors to chronic liver disease, such as iron overload, alpha-1 antitrypsin abnormalities, granulomatous disease, or occult alcohol abuse.

Diagnostic algorithm

Patients with NASH typically come to medical attention because of elevated aminotransferases or evidence of fatty liver on imaging studies. When elevated aminotransferases are discovered and the use of a medication and alcohol abuse are not suspected to be the cause, a complete serologic workup for chronic liver disease should be undertaken. If medications or alcohol are suspected, a period of avoidance of the suspected agent(s) and rechecking the liver enzymes in 4 to 8 weeks is reasonable.

There is no simple answer regarding which patient to biopsy. The decision is made on a case-by-case basis taking into consideration the likelihood of finding advanced disease and of knowing that the diagnosis will change the patient’s behavior. Some patients are not motivated to make necessary lifestyle modifications until a diagnosis of NASH is established, whereas others are motivated to make changes with the threat of needing a biopsy if they fail to lose weight and increase exercise over a defined time period.

Blood tests to evaluate elevated ALT and/or AST levels

Viral hepatitis. Anti-HBs, anti-HBc total (not IgM), HBsAg, anti-HCV (also consider anti-HAV total [not IgM] to recommend vaccination if negative)

Note: There is never any utility in measuring acute antibody response (IgM) to hepatitis A or B when evaluating chronic liver disease.

Autoimmune hepatitis. ANA, anti-smooth muscle antibody (ASMA)

Antimitochondrial antibody (AMA), if alkaline phosphatase is elevated

Metabolic liver disease. Alpha-1 antitrypsin level and phenotype, iron level and transferrin saturation, ceruloplasmin (follow up with a 24-hour urine copper if the ceruloplasmin is low or borderline low, especially if age <35 years).

Factors for and against liver biopsy
Features that favor obtaining a liver biopsy when NASH is suspected

– Clinical evidence of cirrhosis (e.g., low platelets, varices, cutaneous spider angiomas)

– Risks for advanced disease: obesity, diabetes, hypertension, age > 50, AST > ALT

– Patient is motivated to know the diagnosis.

– ALT >30 U/L in men or >19 U/L in women

Features that favor not doing a biopsy

Concomitant diseases

– Usual contraindications: severe comorbid diseases (e.g., advanced cancer, severe CHF)

– Bleeding disorder (Transjugular liver biopsy is an option in this situation.)

– Truly normal ALT (<30 U/L for men, <19 U/L for women; disregard lab-specified upper limit for ALT)

– Disinterested patient

Pathognomonic findings

– Clinical: There are no pathognomonic symptoms or laboratory findings for NAFLD or NASH.

– Imaging: Bright liver by ultrasound, low density liver by CT, and MRI studies showing increased liver fat fraction indicate NAFLD; no imaging study can identify NASH.

– By definition, only liver biopsy findings are pathognomonic for NASH.

Other diagnostic considerations

Patients with NASH typically come to medical attention because of elevated aminotransferases or evidence of fatty liver on imaging studies. When elevated aminotransferases are discovered and the use of a medication and alcohol abuse are not suspected to be the cause, a complete serologic workup for chronic liver disease should be undertaken. (See list of blood test above.) If medications or alcohol are suspected, a period of avoidance of the suspected agent(s) with rechecking the liver enzymes in 4 to 8 weeks is reasonable.

In the setting of elevated aminotransferases, all forms of chronic liver disease and drug-induced liver injury must be considered as alternatives or concomitant processes to NAFLD.

Identifying another cause of liver disease does not exclude NAFLD and NASH as coexisting disease processes such as viral hepatitis, primary biliary cirrhosis, autoimmune hepatitis, alpha-1 antitrypsin abnormalities, and hemochromatosis that have been shown to coexist with NASH. Even alcoholic liver disease probably coexists with NASH in some patients with risk factors for both. Obesity is known to be associated with more advanced alcoholic liver disease and this may in fact represent the synergistic injury from both pathophysiological processes.

Hepatitis C is often associated with NAFLD. This appears to be due to coexisting risks (e.g., type 2 diabetes, obesity) in hepatitis C patients rather than the virus itself. The exception is genotype 3 hepatitis C infection, which can cause NAFLD, and the NAFLD improves with successful eradication of the virus.

In the setting of imaging evidence of NAFLD, NASH may be present in 10% to 30% of patients.

What other diseases, conditions, or complications should I look for in patients with nonalcoholic fatty liver disease and nonalcoholic steatohepatitis?

Major risk factors

– Obesity

– Type 2 diabetes

– Hypertension

– Hyperlipidemia

– Sedentary lifestyle

– Poor dietary habits (sugar-sweetened beverage consumption, regular fast food consumption)

Concomitant diseases

– Cardiovascular disease

– Diabetes

– Hyperlipidemia

– Obstructive sleep apnea

– Polycystic ovarian syndrome (PCOS)

– Hepatitis C genotype 3


– Progression to cirrhosis

– Hepatocellular carcinoma (also seen rarely in noncirrhotics)

– Inappropriate cholecystectomy

What is the right therapy for the patient with nonalcoholic fatty liver disease or nonalcoholic steatohepatitis?

Treatment options

Weight loss. Losing 10% of body weight results in histological improvement of NASH. Gradual weight loss (no more than 1 lb per week) achieved through choosing healthy foods in normal quantities is optimal. More rapid weight loss, missing meals, especially breakfast, and fasting in general should be discouraged. Because relatively modest weight reduction results in improved liver histology, this suggests that obesity per se may not be the issue but rather it is the dietary and exercise habits that perpetuate the obesity that lead to NASH.

Healthy food choices in the right amount. Using the word “diet” when discussing what patients should eat implies deprivation, invoking the powerful and counterproductive psychology of denial. A more productive approach is to focus on healthy eating habits while discouraging excessive consumption of foods known to cause obesity or promote hepatic synthesis of fatty acids (i.e., simple carbohydrates, especially fructose).

Fast foods are often high in calories and should be avoided. Single meals in excess of 1500 calories and 100 g of fat (a stick of butter is 92 g) are easily consumed when eating fast foods. Eating several such meals daily will inevitably cause weight gain in most people, especially those who are sedentary.

Consumption of excessive carbohydrates in soft drinks, juices, and other beverages drives hepatic de novo lipogenesis (synthesis of fatty acids), and high quantities of fructose especially promote the formation of fatty acids in the liver. Thus, beverages containing large amounts of sugar are to be discouraged. For some patients, eliminating such beverages alone results in weight loss. Patients should be reminded that a 12-ounce can of commonly consumed sugar-sweetened soda and sweetened tea typically contains about 10 sugar cubes of sugar.

Exercise. Many patients with NAFLD and NASH are remarkably sedentary. Regular aerobic exercise has been shown to be associated with reduced liver fat content. Because of the risk of concomitant cardiovascular disease in patients with NAFLD, screening through a careful history and testing as appropriate should be undertaken before patients engage in strenuous exercise regimens. Walking is a good start for many patients but progressing to regular aerobic activities, as tolerated, is best.

Vitamin E. Natural vitamin E at a dose of 800 IU daily was found to cause resolution of NASH in about half of adults treated in the PIVENS trial (Sanyal et al. 2010) with a similar result in children in the TONIC trial (Lavine et al. 2011). Ways to identify which patients might benefit from vitamin E have not been established. The patients treated in the PIVENS and TONIC trials were nondiabetic and noncirrhotic and thus the role of vitamin E in treating patients with diabetes or cirrhosis is unknown.

Pioglitazone. Pioglitazone is not FDA-approved for the treatment of NASH. The thiazolidinediones (TZDs, or “glitazones”) improve insulin sensitivity and reduce inappropriate lipolysis in adipose tissue, which may decrease the amount of fat the liver must take up from the circulation. Because adipose triglyceride lipolysis is the major source of free fatty acids delivered to the liver, regulation of adipose lipolysis with drugs such as pioglitazone is a logical way to treat NASH. However, treatment of patients resulted in resolution of NASH in only half of the patients in the PIVENS trial. Its use was associated with weight gain, which can be especially problematic when treating obese patients with NASH. Note that the drug has a black box warning: not to use it in patients with significant congestive heart failure.

Metformin. Improving hepatic insulin sensitivity with metformin has not been shown to be an effective treatment option for NASH.

Bariatric surgery. Laparoscopic banding and restrictive or malabsorptive surgical procedures to treat obesity have been shown to improve NASH. The precipitation or worsening of NASH observed after jejunoileal bypass procedures performed decades ago has not been seen with modern procedures that do not leave a blind loop of small bowel. The blind loop was a source of small bowel bacterial overgrowth that probably played a key role in the liver disease following that now abandoned surgical technique. Bariatric surgery may be a reasonable option for the properly motivated and educated patient with NASH in the setting of obesity.

Other. Many patients seek alternative, readily available therapies, such as fish oil, herbal remedies, and “liver cleansing” regimens. Fish oil (omega-3 fatty acids) preparations are currently in clinical trials for NASH. Of the potentially beneficial herbal remedies, milk thistle (silymarin) is in clinical trials. Patients must be questioned about herbal remedy use because some agents, such as kava kava used for weight loss, can cause significant liver injury. The liver is remarkably capable of cleansing itself of xenobiotics and there is no evidence that the liver is cleansed of anything by frequently touted liver cleansing regimens, diets, and remedies.

Other management issues

Statin use. Despite what patients learn from direct-to-consumer advertising on television, there is no medical evidence showing that preexisting liver disease predisposes to statin hepatotoxicity. Studies of large prescription databases have in fact shown that aminotransferase levels improve on average with statin use in patients with elevated levels at the time these are started. Nonetheless, as with any medication, patients should be instructed to contact their care provider immediately if they develop any new symptoms after beginning a statin.

Vaccination for hepatitis A and B. All patients with chronic liver disease of any type should be advised to have hepatitis A and B vaccinations if they have not had prior vaccinations or evidence of prior exposure. Note that documented prior vaccination for hepatitis A is adequate and testing for evidence of antibody is not needed when a patient knows he or she was vaccinated. (The commercially available antibody test is relatively insensitive and can be negative in a third to half of patients with prior vaccination despite protective levels of antibodies).

What is the most effective initial therapy?

Making healthy food choices, eating normal portion sizes (best done by avoiding restaurants), and being physically active on a regular basis are the cornerstones for initial treatment of NAFLD and NASH.

Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.


A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies


Listing of these, including any guidelines for monitoring side effects.


How should I monitor the patient with nonalcoholic fatty liver disease or nonalcoholic steatohepatitis?

Patients with NASH should be seen at least every 6 to 12 months to encourage the adoption or maintenance of lifestyle changes.

Aminotransferase levels are typically checked every 6 to 12 months, but there is no evidence that increases or decreases signal worsening of or improvement in the patient’s liver disease.

The only means of monitoring progression is to repeat the liver biopsy. This can be recommended every 3 to 5 years with the interval depending on intervening events. A dropping platelet count may reflect progressive portal hypertension that may justify an earlier biopsy to guide surveillance for varices and hepatocellular carcinoma.

Complete normalization of ALT levels (<30 U/L for men and <19 U/L for women) after significant weight loss and improved exercise habits may signal improved liver disease. Some patients, especially those with fibrosis, may be motivated to maintain these habits if a biopsy shows this improvement.

What's the evidence?

Bellentani, S, Scaglioni, F, Marino, M. “Epidemiology of non-alcoholic fatty liver disease”. Dig Dis. vol. 28. 2010. pp. 155-61. (This article includes recent epidemiological data.)

Browning, JD. “New imaging techniques for non-alcoholic steatohepatitis”. Clin Liver Dis. vol. 13. 2009. pp. 607-19. (Good review of imaging in NAFLD.)

Brunt, EM. “Pathology of nonalcoholic fatty liver disease”. Nat Rev Gastroenterol Hepatol. vol. 7. 2010. pp. 195-203.

Lavine, JE, Schwimmer, JB, Van Natta, ML. “Effect of vitamin E or metformin for treatment of nonalcoholic fatty liver disease in children and adolescents: the TONIC randomized controlled trial”. JAMA. vol. 305. 2011. pp. 1659-68. (Results of the TONIC trial in children showing that vitamin E improved liver biopsy in some children. The primary endpoint was not reached with vitamin E because the ALT levels improved in both the treatment group and the placebo group. Metformin had no beneficial effect.)

Neuschwander-Tetri, BA. “Lifestyle modification as the primary treatment of NASH”. Clin Liver Dis. vol. 13. 2009. pp. 649-65. (Review of the diet and exercise data for the treatment of NASH.)

Neuschwander-Tetri, BA. “Hepatic lipotoxicity and the pathogenesis of nonalcoholic steatohepatitis: the central role of nontriglyceride fatty acid metabolites”. Hepatology. vol. 52. 2010. pp. 774-88. (Review of the pathogenesis of NASH.)

Sanyal, AJ, Chalasani, N, Kowdley, KV. “Pioglitazone, vitamin E or placebo for nonalcoholic steatohepatitis”. N Engl J Med. vol. 362. 2010. pp. 1675-85. (Results of the PIVENS trial showing that both vitamin E and pioglitazone had a benefit on liver biopsy changes in about half of subjects.)

Sullivan, S. “Implications of diet on nonalcoholic fatty liver disease”. Curr Opin Gastroenterol. vol. 26. 2010. pp. 160-4. (Review of dietary factors in NAFLD.)

Tiniakos, DG, Vos, MB, Brunt, EM. “Nonalcoholic fatty liver disease: pathology and pathogenesis”. Annu Rev Pathol. vol. 5. 2010. pp. 145-71. (Review of the pathology of NASH.)

Williams, CD, Stengel, J, Asike, MI. “Prevalence of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis among a largely middle-aged population utilizing ultrasound and liver biopsy: a prospective study”. Gastroenterology. vol. 140. 2011. pp. 24-131. (NASH was found in 12% of patients seeking routine healthcare or undergoing screening colonoscopy.)