A Practical Guide to Abnormal Liver Tests

David A. Johnson, MD


September 21, 2010

This feature requires the newest version of Flash. You can download it here.

Abnormal Liver Tests: Introduction and Clinical Scenario

Hello. I'm Dr. David Johnson, Professor of Medicine and Chief of Gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia. Welcome back to another session of GI Common Concerns -- Computer Consult.

Today I thought we would talk about abnormal liver-related enzymes, abnormal liver tests -- not liver function tests -- liver tests. These [abnormal liver tests] are a very common problem for us in day-to-day practice, and I am sure you see them every day. Let us start with a clinical scenario, and then we can build our discussion around an individual case.

This is a case of a 36-year-old woman who comes to you because she had an insurance physical and, when they did her liver tests, they found that the alanine transaminase (ALT) was elevated, about 1.5 times normal. Her aspartate transaminase (AST) was essentially normal.

What would you do with this patient? What would be your starting discussion? How would you work up this patient in a cost-efficient way? I know we are pragmatic in everything we do, but [in terms of being] cost efficient, what would be your approach to this patient?

History and Physical Examination

First, you take a great history. That goes without saying. In particular, here are things that I want you to remember: any risk factors for hepatitis, any type of family history of liver disease, or any occupational exposures, in particular any use of drugs or any type of alcohol consumption that would result in a disproportionate risk for liver enzyme elevation. This all needs to be noted.

Talk about what medications they are taking, because we find that this is a very common abnormality that is picked up just because of coincident use of a medication, whether it is prescription or over the counter. Expand on the over-the-counter medications. Ask about acetaminophen and nonsteroidal anti-inflammatory drugs. Also ask about any herbal therapies, any particular teas, or any unusual things that they may be taking in excess. You want to know that because it can be very, very subtle, but is clearly implicated as it relates to liver test abnormalities.

[In regard to our case], you essentially have a normal, healthy adult with nothing [unusual] in the history.

Let us talk about the physical. In this case, her body mass index (BMI) is elevated. She probably has a BMI of around 34, so she meets the criteria for obese. It certainly makes you think about things like hepatic steatosis, which is very likely in a 36-year-old woman with a little bit of obesity.

Evaluation of Common Liver Tests

Let us talk about how we evaluate her liver tests [to] better direct our workup.

Her transaminases are up minimally in the AST. The ALT is up. We are always taught that there is a disproportion in patients that have alcohol [injury]. The AST goes up, and the ALT lags behind. There is a 2:1 ratio that we are all taught and is somewhat strongly suggestive of alcohol injury. Why is that? It is because of where the enzymes live in the cell. For alcohol injury, it is a subcellular enzyme. It does not necessarily require the whole cell to be killed to have this released; but, the ALT is a cytosolic component to the enzyme, so you really have to kill the whole cell to get the ALT released. The injury from alcohol does not necessarily have to kill the whole cell, so the AST goes up, and the ALT lags behind, in particular, unless the whole cell is killed. That can happen with alcohol, too. Think about the ratio when you look at this as it relates to considering alcohol. The ratio in this case is not helpful.

This patient's alkaline phosphatase is normal and her bilirubin is normal, so we are not thinking about a cholestatic-type injury. I tried to look at the pattern of abnormalities and say, this just a transaminase elevation that makes me think about infiltrated disease and inflammatory diseases in the liver.

Is it a cholestatic picture? If so, think about bile duct-type diseases, like primary biliary cirrhosis and sclerosing cholangitis, or obstructive-type diseases, such as common duct stone, neoplasm in the bile duct or in the pancreas, something at the hilum, or extrinsic compression from metastatic disease. Those things all can make you have much more of a cholestatic injury. In between, you have this mixed pattern.

We will come back to this patient again, who has minimal elevation of her transaminases, normal alkaline phosphatase, and normal bilirubin.

What else do I look at in the liver panel?

I always look at the immunoglobulin fraction. This is helpful because it may pinpoint you very quickly to a patient with autoimmune hepatitis. The immunoglobulin fraction is something you can easily calculate; it is simple subtraction. You start with the total protein and then subtract the albumin. It should be less than 4. If it is less than 4, autoimmune hepatitis is less likely. Now, can that happen? Sure. More likely than not, if it is over 4, you are starting to think very quickly about autoimmune hepatitis. That is what I look at when I look at the profile.

Let us talk about a pragmatic approach to what I would do for this elevation. I would start with lab tests to supplement a good history and physical.

If I am thinking this patient even has remote exposure for hepatitis, I would order a hepatitis profile. If we are talking about acute hepatitis, [the profile would include] hepatitis B surface antigen and hepatitis B core antibody, immunoglobulin M (IgM). If you just ordered hepatitis B core antibody, IgM, this may be the only thing that is elevated in acute hepatitis B. The acute hepatitis profile would include hepatitis A IgM antibody and a hepatitis C [virus antibody] test. Certainly those things would be very easy to get and not unreasonable to do in the initial evaluation.

For an autoimmune hepatitis workup, would I order an antinuclear antibody test (ANA) in a 36-year-old [woman]? I would, just because you can certainly see the globulins here are not 100% suggestive, but you may miss the diagnosis. I would start with an ANA in a woman in whom I am considering autoimmune hepatitis. It is still a differential diagnosis.

Things like primary biliary cirrhosis, I would not consider in the absence of alkaline phosphatase elevation, so in a 36-year-old [woman], is where you start to see the patients [with primary biliary cirrhosis] start to encroach on the diagnosis and that may be something to consider, if the alkaline phosphatase and bilirubin are up, not necessarily just her ALT and minimal elevation of AST.

One thing that you should consider in patients is celiac disease. We are seeing an incredible number of people with celiac disease, and their only manifestation, as it was in this case, potentially, was that they just had their blood drawn and their liver tests are up. Celiac disease can cause what looks like an autoimmune-type hepatitis pattern. It is something that we are seeing more and more. It has responded nicely to control of celiac disease. I would consider a celiac profile, maybe not on the first pass of this patient, but, if there is persistent elevation, I would consider it on the second pass.

Elevation of transaminases in the face of hemochromatosis [is possible]. Hemochromatosis is more common in men and less common in women. One percent of the US population carries the allele for hemochromatosis. It is certainly easy to check with a ferritin and serum iron study, so that would be an early test in this patient as well.

I would start with transaminase elevation. I would start with viral hepatitis, acute and chronic panel. I think it is reasonable to do a ferritin. It is very reasonable to do an ANA in this patient. It is very reasonable to consider a celiac profile.

Finally, you should consider Wilson's disease. It is something that we used to tell people, you never screen them over 40 [years]. We do see that Wilson's disease will typically present earlier in life with liver disease, such as] in this 36-year-old [woman]. In patients younger than 25-30 years, Wilson's disease may present with hemolysis and liver test abnormalities. Wilson's disease can cause fulminant hepatic failure. Older patients, that is, older than 35 years or so, may present more with neurologic sequelae. We have seen Wilson's disease reported now in patients in their seventh decade of life. A ceruloplasmin is an easy one to get, and I never fault the residents for ordering a ceruloplasmin even on patient like this.

I think it is a pragmatic way to start transaminase elevation. Think about viral hepatitis, think about iron storage disease, think about your autoimmune disease, think about Wilson's disease, and think about celiac disease.

I would also consider a hepatic ultrasound on virtually every one of these patients. I think it is very easy to do and very inexpensive. You may get some changes compatible with hepatic steatosis. These are simple tests to do. Make sure you are not looking at masses and such, and I think that is really easy.

One other thing that I did not mention was lactate dehydrogenase (LDH) elevation. I almost never use the LDH as an indicator of liver disease, with a couple exceptions. You have a patient with an LDH elevation and may be a little bit jaundice. Think about things that create infectious hepatitis. In particular I see this with mononucleosis. So a 36-year-old [woman] can get mononucleosis, not necessarily from Epstein-Barr virus, as is common, but more in that age range CMV. So mononucleosis can be a presentation. The LDH would be up in that patient, more so than you would see in other diseases, where the LDH typically does not go up.

Using a pragmatic approach, in an otherwise healthy, mildly obese patient, start with a very simple workup. Again you can pursue further evaluation.

When would you refer this patient for a liver biopsy?

If this patient had more of a progression, I would certainly consider a liver biopsy. We are looking for a hepatic injury now, beyond just simple fat, but nonalcoholic fatty liver disease with an inflammatory response. This would be something that would warrant a very aggressive approach, including controlling the diet, looking at weight reduction, and controlling of lipids, because these people can get steatohepatitis that can result in cirrhosis and more consequences of end-stage liver disease later in life.

By the Numbers: Liver Disease and Patient Age

What else do I do as far as looking at these people? I think about different ages.

If a young patient comes in (the patient in their 20s, or sub 20s), you think about things like Wilson's disease and autoimmune hepatitis. That would be something that you really could not afford to miss.

When [patients] get into their 20s and 30s, you start to think about things like hemochromatosis, Wilson's disease continues, and also viral hepatitis as they may experiment more in their abuse habits, that is something that certainly comes in to play.

For patients in the 30s, 40s, and early 50s, you start to worry about things like primary biliary cirrhosis, especially in women, do not miss that diagnosis. The tip here is that these patients will typically have pruritus when they come in. It is very common to see pruritus, even though they are not even quite jaundice. Interestingly, intrahepatic cholestasis, meaning not obstructive biliary disease, is very likely to cause pruritus early as opposed to a patient who comes in and is profoundly jaundice from a malignancy or something. Those patients get pruritus (itching) later in the stages of their disease, as opposed to [patients with] primary biliary cirrhosis who may have [pruritus] very early in their presentation.

So for patients in the 30s, 40s, and 50s, I am also thinking about autoimmune hepatitis. Alpha-1 antitrypsin deficiency is one we had not mentioned yet, but it is simple enough to look for. It typically does not present as liver disease until in the fourth, maybe fourth to fifth, decade, so it is later in presentation. A simple test here is a serum protein electrophoresis, looking at the Alpha-1 level, in particular. If you see any abnormalities, then go straight to an Alpha-1 level, it is a lot more expensive to order that test up front, so I do not typically do that. I order a serum protein electrophoresis.

For patients in the 50s, 60s, and 70s, you start to always think about bad things, such as infiltrative diseases and metastatic diseases. There is second peak for autoimmune hepatitis in the 50s and 60s so it can come back; do not forget autoimmune hepatitis. Think again about primary biliary cirrhosis. Think about other things, particularly drug-induced abnormalities, as these patients are exposed to more drugs as they age. One thing that we start to see in the older patient population too is, as they may have a component of congestive failure, we may see some elevation there; this more or less creates an ischemic pattern of injury, and, in that patient, I may also look at hepatic portal venous flow and do hepatic duplex, looking at their right heart influence on the hepatic flow.

Take-Away Points

There are some simple things to do. A liver biopsy is the gold standard; but again, it is not without significant potential risk. I do not jump to a liver biopsy. I think you can figure out a lot with simple things and without a liver biopsy. Be pragmatic.

Going back to our patient, what have we found?

We found this patient to have increased epigenicity on hepatic ultrasound. I was very comfortable with only 1 significant abnormality here. In following this patient cautiously, I encourage weight reduction, and I checked her lipid profile. Those are the 2 factors that I think really played out for hepatic steatosis. If you are a diabetic patient, then you are getting more into a metabolic syndrome, and strict diabetic control obviously goes without saying. Those factors would all mitigate toward the better outcome for hepatic steatosis.

I hope I have given you some pragmatic points as you look at your next patient with liver-related enzyme abnormalities.

To summarize these points:

  • Think about the pattern of injury;

  • Look at the age of the patient;

  • Think about the history. Does it focus you on particular areas to work up as far as it relates to viral hepatitis or infectious causes?;

  • Take a good drug history, including over-the-counter medications; and

  • Ask about occupational exposures.

A lot of these tests go with a lot of cost, so be pragmatic in your approach and hopefully your patients will be well served.

We see these [liver-related enzyme abnormalities] a lot. When you start to do surveys, 8% of the population have abnormal liver enzymes, so if you have not seen it, expect it soon. Hopefully I have given you some guidance to make your next interaction a very favorable one and to give you a little bit more insight into a very common clinical problem.

I am Dr. David Johnson. I look forward to chatting with you again soon. Thanks for listening.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: