Question from EM - how do you differentiate liver disease vs hemolysis labs?

[EMgordo]

Hello IM friends, I am in EM and have a question that I think better fits your expertise. I've had several chronic alcoholic patients come in with a lab pattern of anemia, low platelets, and elevated bilirubin. I know this can be expected for some cirrhotics, but it is also the pattern that might prompt consideration of hemolysis/MAHA. What do you do in these situations to differentiate between the two?

I'm sure that sending a full hemolysis panel and smear would help guide us, but we often don't get those results in time while in the ED.
I've thought that the direct or indirect predominance might help, but then I've heard that the cirrhotic pattern of bilirubin can be predominantly conjugated or unconjugated. Is that true?
I've also heard that cirrhotics can have some low levels of hemolysis as part of its disease process, so is it a matter of seeing the degree of abnormalities on the hemolysis labs?

I've tried to look this up but have not had much look. I appreciate any wisdom you are willing to share

---

Members don't see this ad.

 

[whoknows2012]

Unfortunately many of the findings can be present in both with caveats and variabilities depending on severity of cirrhosis/liver disease and the type of maha/tma (vs let’s say pure hemolytic anemia).

Hemolysis causes an indirect hyperbili not direct. In liver disease and cirrhosis you can have an indirect bilirubinemia but the direct is often higher

Reticulocytosis is more likely to be present in maha/tma vs liver disease. You’re also more often to find hemolysis cases presenting with macrocytic anemia as opposed to normocytic anemias encountered in liver disease (though this can obviously also be confounded by alcohol itself which can cause macrocytosis)

Sure there are other indirect clues-in significant maha/tma there may be a significant elevation in LDH and cytopenias will vary depending on process. Also in cirrhosis with hypersplenism as you probably already know you can have pancytopenia though of course it’s most classically associated with thrombocytopenia, Leukopenia and anemia do occur in more severe circumstances.

Unfortunately the best way to differentiate is good history review of labs and looking at the smear. If the smear does not show significant schistocytes or microspherocytes (or other indirect evidence of hemolysis or maha/tma) I’d consider it less likely to be from that

 

[DrMetal]

Alcohol also suppresses bone marrow, so you can have a pancytopenia, in and of itself, hence the pattern of anemia and thrombocytopenia. Sometimes the liver disease is already present, sometimes not (the liver disease and pancytopenia are not necessarily correlated, but onset at the same time).

https://pubs.niaaa.nih.gov/publications/arh21-1/42.pdf

 

[thumbz]

I may be off base here, but I think a haptoglobin would be helpful? Caveat being it's produced by the liver, but one study suggested even in cirrhosis 70% of patients have a normal haptoglobin. So if you check it and it's normal, less likely hemolysis where as if it's low it would make me much more likely to grab the smear.

Interestingly in obstructive jaundice haptoglobin seems to be elevated which would actually be helpful in pointing towards the increased bili not being hemolytic in nature.

 

[EMgordo]

Thank you all for the replies. I am going to try to summarize and let me know if any of this is off-base.

Next time I have a patient like this, with ETOH hx, no prior labs, and presents with anemia, thrombocytopenia, and elevated bilirubin, I can do the following:
-send labs to assess liver function and hepatobiliary injury like INR, albumin, LFTs, alk phos +/- GGT, direct and indirect bilirubin, imaging RUQ US vs CTAP, etc
-send some simple labs for an initial screen for hemolysis like haptoglobin, LDH.

Things that would suggest the lab derangements are more likely from hemolysis/MAHA/TMA would be:
-bilirubin is overwhelmingly indirect/unconjugated (not a perfect test, as sometimes cirrhosis can be majority indirect too, but that is less frequent)
-low haptoglobin (not perfect, as liver produces haptoglobin, but most of the time liver can still produce haptoglobin even in cirrhosis)
-elevated reticulocytes
-elevated MCV (not perfect)
--> if these are present, send the full hemolysis panel like peripheral smear etc.

Things that would suggest the lab derangements are more likely from hepatobiliary disease would be:
-bilirubin is predominantly direct, which would not be seen in hemolysis.
-haptoglobin is elevated (not perfect)
-dilated CBD or other imaging findings suggesting cirrhosis etc
-if prior labs show this is chronic, it would be more likely cirrhosis cause than MAHA.

 

[MD46]

When in doubt consult Hematology, wouldnt wana miss a TTP/HUS patient. Some places will have hematopathologist on call to look at smears for shistocytes etc

 

[Drrrrrr. Celty]

I may be off base here, but I think a haptoglobin would be helpful? Caveat being it's produced by the liver, but one study suggested even in cirrhosis 70% of patients have a normal haptoglobin. So if you check it and it's normal, less likely hemolysis where as if it's low it would make me much more likely to grab the smear.

Interestingly in obstructive jaundice haptoglobin seems to be elevated which would actually be helpful in pointing towards the increased bili not being hemolytic in nature.

A cirrhotic with an INR >2 probably is going to have low haptoglobin.

 

[Drrrrrr. Celty]

When in doubt consult Hematology, wouldnt wana miss a TTP/HUS patient. Some places will have hematopathologist on call to look at smears for shistocytes etc

It's not hard to order a DIC panel and interpret it prior to consulting a hematologist.

 

[Drrrrrr. Celty]

Thank you all for the replies. I am going to try to summarize and let me know if any of this is off-base.

Next time I have a patient like this, with ETOH hx, no prior labs, and presents with anemia, thrombocytopenia, and elevated bilirubin, I can do the following:
-send labs to assess liver function and hepatobiliary injury like INR, albumin, LFTs, alk phos +/- GGT, direct and indirect bilirubin, imaging RUQ US vs CTAP, etc
-send some simple labs for an initial screen for hemolysis like haptoglobin, LDH.

Things that would suggest the lab derangements are more likely from hemolysis/MAHA/TMA would be:
-bilirubin is overwhelmingly indirect/unconjugated (not a perfect test, as sometimes cirrhosis can be majority indirect too, but that is less frequent)
-low haptoglobin (not perfect, as liver produces haptoglobin, but most of the time liver can still produce haptoglobin even in cirrhosis)
-elevated reticulocytes
-elevated MCV (not perfect)
--> if these are present, send the full hemolysis panel like peripheral smear etc.

Things that would suggest the lab derangements are more likely from hepatobiliary disease would be:
-bilirubin is predominantly direct, which would not be seen in hemolysis.
-haptoglobin is elevated (not perfect)
-dilated CBD or other imaging findings suggesting cirrhosis etc
-if prior labs show this is chronic, it would be more likely cirrhosis cause than MAHA.

INR is the benchmark for synthetic liver function. You can't really characterize a liver injury or chronicity without one.

Low platelets should be confirmed with a citrated platelet to rule out assay issue.
Low platelet confirmed should reflex to a DIC panel ( INR, Schistocyte screen, PT/PTT).
If DIC panel wnl. It's likely neither a MAHA or DIC. You have time. If PTT is the only abnormality consider autoimmune non-MAHA like lupus ( Consult Heme)
Get U/S abdomen and look at liver and spleen. Hepatosplenomegaly? That's portal hypertension, chronic or acute likely in 99% of cases.
No ultrasound evidence or confirmation then get a hematologist. Because now you're getting into more expensive tests and consideration for things like Cold or warm hemolysis or PNH.

Haptoglobin is a good test for healthy individuals. Malnutrition and poor synthetic function make it less useful.

Either way as an EM person you probably will be able to feel out what needs an admission with a lab abnormality like that. Ex. a 20 yo with no alcohol hx with the above labs 100% needs it. Someone who is a chronic alcoholic and stable won't. Someone who is a chronic alcoholic and looks like they have a new INR of 9 and a Bili of 10 needs it.