Pro: Noninvasive Imaging Has Replaced Biopsy as the Gold Standard in the Evaluation of Nonalcoholic Fatty Liver Disease

Watch a video presentation of this article

Watch the interview with the author

Abbreviations

MRE

magnetic resonance elastography

NAFLD

nonalcoholic fatty liver disease

NASH

nonalcoholic steatohepatitis

TE

transient elastography

When defining the term gold standard, the most appropriate definition is “a benchmark test that is the best available under reasonable conditions.”1, 2 When considering this definition, it becomes clear that noninvasive imaging has replaced liver biopsy as the gold standard for evaluation of fibrosis in nonalcoholic fatty liver disease (NAFLD).

The NAFLD epidemic is estimated to affect almost a quarter of the global population, with many high‐risk features being especially prevalent in the United States.3 However, the population most at risk for poor outcomes is those with nonalcoholic steatohepatitis (NASH), because these are the patients at greatest risk for progression to cirrhosis.3, 4 In the United States, the number of individuals with NASH is estimated to be more than 11 million, and our ability to delineate which patients with NAFLD have NASH remains poor at present. The long‐standing use of liver enzymes to differentiate these different entities has been debunked and proved inadequate,4, 5, 6, 7 and despite extensive work in the research realm, no other straightforward serological tests or panels have shown results convincing enough to usurp the previous gold standard.6, 7 The outdated dogma of liver biopsy being the gold standard for identifying those at risk among these millions of patients would mean extrapolating to this population the known severe complication rate of 1.1%8 and mortality rate of 0.3%.9 If we are to appropriately identify the NAFLD population most at risk for progression to cirrhosis by means of biopsy, it corresponds to daunting figures of morbidity and death. When considering the earlier definition of gold standard, it becomes clear that liver biopsy, for this purpose and in this population, cannot be considered “under reasonable conditions” when a noninvasive test has been proved noninferior.

Transient elastography (TE) has been a clinically useful as well that has the means to replace liver biopsy as the gold standard. TE has high accuracy when identifying patients with F3‐F4 fibrosis who are at greater risk for worse clinical outcomes.10 The most remarkable advantage of TE is that the procedure is noninvasive, without any of the complications associated with liver biopsy. In addition, its cost is one‐fourth of that of liver biopsy, and it can be done in 5 minutes in the outpatient setting without any associated pain.11 This also allows this test to be done more frequently, and to be followed over time in patients with higher risk features.

Nevertheless, liver biopsies continue to be done for confirmation in patients who are at risk for fibrosis/cirrhosis. The most useful factor that we acquire from the biopsy and the resultant NAFLD activity score is the degree of fibrosis. Although fibrosis is a result of ongoing inflammation, the grade of inflammation on biopsy has not been shown to have an effect on clinical outcomes. Thus, stage of fibrosis remains the only significant information obtained from histology.12 When considering this, it becomes clearer that, if we are able to accurately assess the degree of fibrosis without invasive measures in patients with NAFLD, we will identify the population at risk for adverse clinical outcomes without the aforementioned complications of biopsy.

Some may use an alternative definition for gold standard, referring to the most accurate test without restrictions, and will suggest that liver biopsy is more accurate than TE. However, liver biopsy samples 1/50,000th of the entire liver, and thus is susceptible to inadequate biopsy size, which can decrease accuracy of staging by 25%.13 Studies have shown that paired biopsies, from two different areas of the liver, can differ in histological fibrosis staging up to 40% of the time.14 Also, studies have demonstrated significant subjectivity in fibrosis staging, with different pathologists disagreeing on staging up to 50% of the time.13, 15, 16 These inherent flaws in liver biopsy as a means for staging fibrosis in NAFLD have come to light only recently and further demonstrate the need for alternative methods (Table 1).

Table 1.

Pros and Cons of Methods of Measurement of Liver Fibrosis

	Pros	Cons
Liver biopsy	‐Current Gold Standard ‐Can determine underlying etiology of Fibrosis	‐Pain ‐Bleeding ‐Mortality Risk ‐Small sample size ‐Variability ‐Reader dependent accuracy ‐Expensive
VC transient elastography	‐No complications ‐1/4 of the cost ‐< 5 minutes completion time ‐Easily repeated	‐Operator dependent accuracy ‐Limited by body habitus/ ascites ‐Altered by infiltrative processes ‐Inexpensive
MR elastography	‐No complications ‐Non‐inferior to liver biopsy	‐Reader dependent accuracy ‐Long process ‐Not standardized ‐Limited space

Magnetic resonance elastography (MRE) is a test that is not yet widely available, but which has an accuracy of fibrosis staging on par with liver biopsy. In identifying patients with F3‐F4 disease, the area under the receiver operating characteristic curve for MRE is at the very least noninferior to liver biopsy at 0.98.16, 17, 18, 19, 20 Although limited head‐to‐head studies are available, at least one has shown equivalence in detecting and differentiating between significant fibrosis, severe fibrosis, and cirrhosis.19 In addition, it has been shown that the rate of agreement on fibrosis stage between different radiologists reading MRE is greater than that of separate pathologists assessing biopsy specimens.17, 18, 19, 20, 21 (Table 1)

It is clear that the NAFLD epidemic will continue to expand in the coming years, and that liver biopsy is not an ideal methodology for evaluation of these patients. Noninvasive modalities can be used more frequently to follow at‐risk patients over time, as well as be instituted for screening evaluations in the absence of the morbidity that unfortunately comes with liver biopsy. Lastly, MRE has emerging data to support its noninferiority to liver biopsy in terms of accuracy in fibrosis staging and, combined with the dramatic risk profile differences, should be considered a superior test. Although the liver biopsy complication rates appear low on an individual basis, expanding those rates to the massive and growing population in need of assessment for liver fibrosis makes it evident that the risk is inexplicable when faced with noninvasive and noninferior alternatives. In our opinion and for the reasons stated earlier, these noninvasive imaging modalities are the new gold standard for assessment of liver fibrosis in patients with NAFLD.