GI Consult: Nonalcoholic Fatty Liver Disease

Once thought to be associated only with chronic alcohol abuse, fatty liver is now recognized as a very common disease with a wide range of causes.

By Charles Maltz, MD

1. Isn't fatty liver caused by alcoholism? What is nonalcoholic fatty liver disease (NAFLD)? Is it the same as nonalcoholic steatohepatitis (NASH)?

The liver plays a pivotal role in fat metabolism, so it is not surprising that fat deposition (steatosis) may develop under certain conditions. Alcoholism is one possible cause, but fat deposition may also develop from a reaction to medication or from a viral infection. Hepatocellular necrosis and bile stasis may also occur.

Nonalcoholic fatty liver disease, therefore, is simply fat deposition in the liver that is not caused by chronic ingestion of alcohol. The terms NAFLD and NASH are often used interchangeably. However, NASH properly refers to that subset of patients with NAFLD who have evidence of inflammation on liver biopsy. The spectrum of NAFLD is similar to that of alcoholic liver disease. At one extreme is fatty liver occurring without inflammation; at the other is cirrhosis. Nonalcoholic steatohepatitis is an intermediate disease state between these two extremes.

2. What is the difference between macrovesicular and microvesicular fat deposition in the liver?

Macrovesicular fat is deposited in large vacuoles in hepatocytes; such accumulation pushes the nucleus toward the wall of the cell. In contrast, microvesicular fat is deposited throughout the cytoplasm; in this instance, the nucleus retains its central location. Macrovesicular fat deposition typically occurs in NAFLD and alcoholic liver disease, whereas microvesicular fat deposition occurs in Reye syndrome and acute fatty liver of pregnancy and as a possible adverse effect of certain medications, such as valproic acid.

3. What is the underlying pathologic mechanism that causes NAFLD?

The underlying cause is unknown, but it appears to be related to a problem with fat metabolism. The condition is associated with type 2 diabetes mellitus, obesity, total parenteral nutrition involving fat emulsions, rapid weight loss, and hyperlipidemia-all of which are characterized by impaired fat metabolism. One study showed that NAFLD is associated with insulin resistance even in patients with normal results from glucose tolerance tests.

4. It would seem that NAFLD must be an old disease. Why are we hearing so much about it now?

Before diagnostic tests for hepatitis C became available, many cases of NAFLD were categorized as non-A, non-B hepatitis. Now that such tests are commonplace, the spectrum of NAFLD can be more accurately defined. The condition was first described as a specific disease about 20 years ago. Originally thought to be of little clinical significance, it is now believed to be the most common cause of cryptogenic cirrhosis.

5. How common is NAFLD in the United States?

Nonalcoholic fatty liver disease is extremely common, affecting approximately 20% of American adults. Several studies have examined abnormal results of liver function tests in asymptomatic patients. The investigators in those studies found that once chronic alcohol ingestion and viral, pharmacologic, autoimmune, and metabolic causes—such as hemochromatosis—were ruled out, virtually all of the remaining patients proved to have NAFLD.

6. Do patients with NAFLD present with any classic symptoms?

Most patients with NAFLD are asymptomatic. Abnormal liver function test results are usually the first sign of a disorder. Hepatomegaly may be present, but it is usually not detected on physical examination. Occasionally, ultrasound examination performed for other reasons will reveal an enlarged liver.

7. Is there a specific pattern of liver function abnormalities that is characteristic of NAFLD?

The serum transaminase levels are usually not greatly elevated, but in contrast with alcoholic liver disease, the alanine aminotransferase (ALT) level will be higher than the aspartate aminotransferase (AST) level. The ferritin level may also be elevated, sometimes leading to a misdiagnosis of hemochromatosis. Patients with a more virulent form of NAFLD may have already progressed through NASH and fibrosis; the diagnosis therefore will be established only when these patients present with cryptogenic cirrhosis.

8. Does the degree of elevation in aminotransferase levels reflect the severity of liver damage in NAFLD?

No. The degree of elevation in these test results is not helpful in determining who has NAFLD and who has NASH or fibrosis. This information can only be determined by means of liver biopsy.

9. What about ultrasound, computed tomography, and magnetic resonance imaging? How useful are these tests in diagnosing NAFLD?

They may all show images suggesting fatty infiltration of the liver. Ultrasound examination should be performed first because it is the least expensive and does not expose the patient to radiation. However, a definitive diagnosis can only be made on the basis of liver biopsy findings.

10. Should I order a liver biopsy for all patients who have abnormal results of unknown cause on liver function tests-that is, when chronic alcohol ingestion and other known causes of fatty liver disease have been ruled out?

The probable diagnosis in this group of patients is NAFLD. Whether a biopsy is necessary to confirm the diagnosis should be decided after discussing its risks and the possible prognostic and therapeutic benefits. Because NAFLD is a benign condition in most patients, it is prudent to reserve biopsy only for those who are most likely to have fibrosis. A recent study has suggested that old age, obesity, and diabetes mellitus increase the chance of fibrosis being found on examination of liver biopsy specimens.

11. Are potentially hepatotoxic medications contraindicated for patients who are thought to have NAFLD?

Although there is no evidence that the incidence of drug-induced liver damage is increased among patients with NAFLD, liver enzyme levels should always be monitored carefully in any patient with presumed fatty liver who is receiving a drug with the potential for hepatotoxic effects. In addition, if the medication is being prescribed on a long-term basis, obtaining a baseline liver biopsy specimen before initiation of treatment may be helpful.

12. What is the overall prognosis for patients with NAFLD?

Although NAFLD is very common and is the most common cause of cryptogenic cirrhosis, the prognosis for patients with this condition is good, mostly because cryptogenic cirrhosis is rare.

13. Can NAFLD be treated successfully?

Treatment of NAFLD focuses on judicious weight loss, abstention from alcohol, control of diabetes, and treatment of dyslipidemia. Solid evidence supporting the efficacy of this approach is lacking, however. Urosodiol therapy, which has been administered in attempts to resolve fibrosis and reduce inflammation in the liver of patients with NAFLD, has yielded disappointing results.

14. Does fatty liver occur only in humans?

You have probably heard of and perhaps even eaten paté de fois gras. The words fois gras mean "fatty liver." This delicacy is made from the fatty livers of geese that have been force fed. Also, fatty liver in cats-hepatic lipidosis-is a well-documented clinical disorder.