Hemorrhage

Yesterday we talked about a patient presenting with a Hg of 42 and an MCV of 50 with no signs of bleeding. During our discussion, a number of topics related to anemia, hemorrhage and transfusions came up.

We talked briefly about the TRICC trial published in the NEJM. This study looked at 838 ICU patients with a Hg of less than 90 and randomized them to a liberal (transfuse if Hg less than 100) vs. a conservative (transfuse if Hg less than 70) transfusion strategy.

The study showed that at 30 days there was no between arm difference in mortality but there was reduced in-hospital mortality in the restrictive transfusion arm. In sub-group analysis, this arm also demonstrated reduced mortality in younger patients (less than 55 years old) and less acutely ill patients (APACHE-II score less than 20). No difference in outcome was seen between the arms in patients with an ACS.

TRICC Bottom Line:

1) These were ICU patients so be cautious with generalizing to GIM patients, but we generally follow the results from TRICC.

2) Consider a RBC transfusion in patients with a Hg of less than 70, especially if symptomatic. This is not a hard and fast rule - you may not transfuse a patient with a Hg of 65 who is well and asymptomatic, especially if it is chronic.

3) Consider not transfusing RBCs in a patient with a Hg greater than 100. However, it may be reasonable to aim for a higher Hg (i.e. greater than 100) in patients with an acute coronary syndrome.

4) A significant gray zone exists inbetween 70 and 100.

Sources of Blood Loss:

1) When presented with a very low Hg, you need to rule out immediate life theatening causes right away - bleeding and hemolysis.

2) A reasonable initial screen for hemolysis includes an LDH, haptolgobin, blood film and bilirubin. The combination of an elevated LDH and low haptoglobin are 90% specific and 92% sensitive for hemolysis.

3) Looks for obvious sources of bleeding: GI tract (BRBPR, melena, hematemesis), GU tract, trauma (laceration, hemo-thorax/peritoneum, etc.) or Gyne.

4) If no obvious source, consider hemolysis and/or a retroperitoneal bleed.

5) Large hemodynamically significant blood loss will not occur from an intracranial hemorrhage due to the limited space in which there is to bleed. Additionally, in a pulmonary hemorrhage, hypoxia will be a greater threat to life before anemia is.

A few points on microcytic anemias:

1) We talked briefly about differentiating iron deficiency anemia from thalassemia. Thank you to Lauren for pointing out that the RDW may be useful in this case. In thalassemia, the RDW is generally low or normal whereas in iron deficiency anemia it may be elevated as the bone marrow pushes out RBC precursors.

2) In a patient over the age of 50 with iron deficiency anemia, remember to look for occult bleeding from a GI malignancy.

3) A ferritin of less than 10-15 is virtually diagnostic of iron deficiency anemia (99% specific, 59% sensitive). Using a value of 41 increases the sensitivity to 98% and the specificity to 99%. The challenge is that ferritin is an acute phase reactant and is increased in inflammation. One possible approach is to divide the ferritin value by three in these patients, if it's less than 20, this is highly suggestive of co-existing iron deficiency anemia.

4) Don't forget celiac disease as a not uncommon cause of iron deficiency anemia.

5) Many other tests including iron saturation, TIBC, Hg electrophoresis, reticulocyte count, soluble transferrin receptor and blood film examination can help in the assessment of a microcytic anemia.