Dr Tapper discusses how diagnostic testing with vibration-controlled transient elastography (VCTE) and magnetic resonance elastrography are often better choices for diagnosing and staging liver disease than traditional liver biospy. He recently presented on this topic at the American Association for the Study of Liver Diseases annual meeting.

Elliott Tapper, MD, is the director of the cirrhosis program and assistant professor at the University of Michigan.

TRANSCRIPT:

Hi, I'm Elliott Tapper. I am the director of the Cirrhosis Program at the University of Michigan in Ann Arbor where I'm a transplant hepatologist and assistant professor.

I had the privilege of speaking at the liver meeting this year about a biopsy-free world at the postgraduate course. I wanted to share with you a few of the highlights and take-home points.

We all use biopsy from time to time, but I used my 15 to 20 minutes to take a step back and ask about its role within the context of decision making for contemporary patients. In order to build up the arguments that I left the viewers with, I wanted to frame this discussion in terms of the expected yield and the risks and benefits of biopsy in the first place.

A liver biopsy really only samples about 1/50,000 of the liver. It can cause pain. It has complications, although most people sail right through it. It has costs, your time, the nurse's time, the time for people to take a day off of work and rope someone into driving to and fro.

You better have some benefit to balance out those risks. The benefits are broken down into 2 main categories, diagnosis and staging or prognosis.

What I reviewed is that in today's patients where we have access to a plethora of blood tests and the prevalence of certain diseases are increasing while others are decreasing, the role of biopsy for diagnosis is really small.

Most patients who are coming to us happen to have fatty liver disease. At the point that you've excluded those diagnosable causes using serology, the only diagnosis which is more common on a liver biopsy turns out to be fatty liver disease.

Later in my talk, I actually reviewed some of the data for the diagnostic performance of biopsy for some of the more important diagnoses that we are looking for on a biopsy. It may shock people to learn that the performance for diagnosis in autoimmune hepatitis is far from ideal.

It turns out that even though we often use biopsy to make a diagnosis of autoimmune hepatitis, it's a far from certain. The features that are classic for this diagnosis are missing on a huge chunk of liver biopsies, which means that every time we get that pathologist report, there are wide confidence intervals around their estimates and diagnoses.

It turns out that those wide confidence intervals are even more important when it comes to staging because it turns out that patients' livers are highly heterogeneous. If you stab one part of the liver, it could turn out that the other side had a variable amount of fibrosis.

If you give the same biopsy specimen to two of the world's best pathologists, the chance that they agree with each other on the overall staging of fibrosis is far from awesome. Interclass coefficient in the people who gave us the Metavir staging system was .78.

What this means is that there are uncertainties in liver biopsy. We hope that we can try to select patients using noninvasive tests. I reviewed for the most part 2 main concepts. One is that we have serologic tests which allow us to select patients for further evaluation.

I highlighted the FIB-4 because it's free and because it performs about as well as the other serological tests, expensive or not. This leverages the concept that the lower the platelet count, the higher the AST, the bigger the chance that your patient has advanced fibrosis or cirrhosis.

I advocated using this test because for the average patient walking into your clinic, they're actually at relatively low risk of advanced fibrosis. If you apply the FIB-4 to that person, you can either find a person who's at really low risk or a person who becomes at moderate risk.

At that point, you need another test to move the goal post towards clinical certainty about their advanced fibrosis. You could do a liver biopsy. It has problems. We advocated for elastographic tests. Principal one that we use is Fibroscan.

This is a very reliable test, particularly in people with a BMI [body mass index] less than 40. If your BMI is greater than 40, we talked about the role of magnetic resonance elastography [MRE], which is going to perform better in people with very high BMIs.

Now MRE gives you far more beautiful pictures, and it might be more accurate at earlier stages of fibrosis, but there's no difference when you're comparing Fibroscan to MRE for the diagnosis of advanced fibrosis. That's really what we were interested in.

The pitfalls of diagnosis using elastographic tests are 1, if your patient has a really high ALT, they're liable to have a super stiff liver. Something specific to MRE is that if your patient's liver is loaded with iron, iron will confuse the MRI.

We concluded with the concept that all of our tests are going to have some uncertainty and we just have to work through the pre-test probabilities to help us select the best test for our patient, aiming to define that low-risk group first and then selecting our next step after that, which means that the best assessment for fibrosis is a sequential set of tests.

Thanks for your attention.