The Quarterly Consultis a quarterly publication highlighting pediatric clinical expertise. If you have an item of interest to include in the Quarterly Consult, contact Dr. Steve Dassel via Children's Physician Relations at 206-987-5765.

November 2010: Managing Anemia

I like anemia. Everything is neatly compartmentalized under little red blood cells, normal red blood cells and big red blood cells. I talked to Dana Matthews, MD, about it. Dana is an associate professor of hematology and oncology at the University of Washington and is director of Clinical Hematology and medical director of Anticoagulation Services at Seattle Children's Hospital.

Question:Dana, a primary care doctor calls you about microcytic anemia. What is the first question you ask about the patient?

Answer:Age of the child, of course, and diet. If the diet is obviously iron-deficient, i.e., lots of cow's milk, but little else, no further studies are necessary unless the patient obviously has complications such as significant GI blood loss. Otherwise, start iron, 3 mg-4 mg per kg daily as a b.i.d., watch for compliance and treat for 4 to 8 weeks. Get a complete blood count (CBC) at the end of that time, and check the mean cell volume (MCV) to be sure it is improving. Once the hematocrit has normalized, treat for another 2 to 3 months to replenish iron stores.

Question:What is a full CBC?

Answer:Hematocrit and hemoglobin, MCV, Mean corpuscular hemoglobin (MCH), mean cell hemoglobin concentration (MCHC), red blood cell count (RBC), red cell distribution width (RDW), white blood count and platelet count.

Question:What is an RDW?

Answer:Red cell distribution of width. It is the electronic cell counter equivalent of "anisocytosis," or variable cell size. This tends to be elevated in iron deficiency because the red cells made most recently, as iron deficiency has become more severe, are smaller than those made 3 to 4 months ago. In thalassemia trait, all the cells are small, so the RDW is usually normal.

Question:Can you get a retic count earlier in therapy to see if you are on the right track?

Answer:I don't find retic counts to be very helpful in iron deficiency. Unless the iron deficiency is very severe, you may not be able to detect a marked increase in retic counts with initiation of therapy. It is much more helpful to follow hematocrits. In a patient with severe anemia, but who is adequately compensated and so does not need red cell transfusion, I usually recheck a hematocrit in a few days to be sure it is going the right direction.

Question:When do you check again in the patient who has already demonstrated that they developed iron deficiency anemia once?

Answer:Usually 3 months after stopping iron, though depending on how it is going with the family, child, compliance and diet, I sometimes wait up to 6 months.

Question:What about education of this patient?

Answer:Dietary education is by far the most important. I use this as an opportunity to really push getting toddlers "off the bottle" - cutting back on cow's milk, usually to no more than 24 ounces per 24 hours. I try to stress that milk causes problems for a couple of reasons. First of all, children fill up on it and then are much less interested in other foods that would actually contain iron. Second, it impairs the absorption of iron. Ideally, oral iron should be given with vitamin C-containing items such as vitamin C-supplemented juice, or with meat proteins such as chicken, beef, fish or pork. Either of those will markedly increase the percentage of a given dose of iron that is actually absorbed.

Question:What do you do with treatment failure?

Answer:Think about sources of blood loss, e.g., inflammatory bowel disease, or an absorption problem, e.g., celiac disease. Consider an underlying inflammatory process that could block iron utilization. Check compliance. Finally, there are rare problems like the recently described iron-refractory iron deficiency anemia. We would be glad to help you with any of these, especially that last one.

Question:Please go over iron studies for us.

Answer:Ferritin is very helpful if it is low. Inflammation, either from a recent acute illness or a chronic process, can increase ferritin to falsely "normal" or even high. Serum iron is, again, helpful if it is low, but is exquisitely sensitive to iron treatment so just one dose can normalize values. Total iron binding capacity may be elevated in iron deficiency or depressed in chronic disease. If I am really trying to figure out if a child is iron deficient, I will do a ferritin, iron/TIBC and sometimes a zinc protoporphyrin.

Question:How about microcytosis when the diet is iron-replete?

Answer:The CBC can be helpful. If the RBC number is normal or high, think thalassemia trait. Especially consider thal trait if there is a high ethnic likelihood, i.e., if the patient is African-American or of Southeast Asian descent. This would also be a good time to go back and check the newborn screen for the presence of Bart's hemoglobin, as that is a clue to alpha thalassemia trait. If the RBC is low, especially if the RDW is elevated, consider that iron deficiency may be present but the history inaccurate, in terms of diet and/or compliance with oral iron. Other possibilities include the same problems that can cause failure to respond to iron treatment.

Question:Would a referral to Hematology be in order when we suspect thal trait?

Answer:Yes, the treatment is education, and we have more time in our schedule to provide that than the primary care doctor. We emphasize that thal trait is not really a disease, but rather a "variation." There should be no impact on the child's life or ability to participate in athletic activities. We emphasize that the child needs to be made aware of the diagnosis, and if anyone suggests treating with iron, the healthcare provider should be asked to first document iron deficiency. We also provide genetic counseling for the parents, in terms of the potential for more severe thal syndromes in future children, and for the patients, when they are old enough to reproduce.

Question:OK, great. Now, let's turn to normocytic anemia.

Answer:When the MCV is normal, the most important test is the reticulocyte count to separate destructive from hypoproliferative processes. For an otherwise stable patient with a high retic, hereditary spherocytosis is one of the most common diagnoses. This seems especially true in our area, where so many are of Northern European heritage. There is a new screening test for spherocytosis that is more reproducible than the older test, osmotic fragility. This screening test can also be done on the newborn, or in a child who has been transfused, as it uses flow cytometry.

Question:And if the retic is low?

Answer:A low retic suggests a hypoproliferative anemia. You certainly want to check the white blood cell count, differential, and platelet count, because if those are abnormal, it raises the possibility of a marrow replacement process such as leukemia, or marrow failure such as aplastic anemia. You also want to examine the child particularly carefully for adenopathy and hepatosplenomegaly.

Question:And if the other cell lines are normal?

Answer:Check a creatinine, as occasionally renal failure will present as hypoproliferative anemia because of diminished erythropoietin production. I also screen for inflammatory disease, as anemia of chronic disease can be either normocytic or microcytic. Hypothyroidism is generally macrocytic, but I include a TSH to be sure. An impressive hypoproliferative anemia with otherwise normal blood counts and exam can be TEC - transient erythroblastopenia of childhood.

Question:Now we get to big red blood cells. What is your approach?

Answer:Macrocytic anemias are relatively rare in pediatrics, and these patients generally should be referred. Vitamin B12 or folate deficiency can cause microcytic anemia, though such patients usually have a discernible reason such as malabsorption, short gut, or a very abnormal diet. Your patient may have failure to thrive, weakness and diarrhea. Remember that macrocytosis is normal in patients with trisomy 21.


Answer:We really do not understand it, but their hematopoiesis can be mildly abnormal in several ways.

Question:Anything else we should think of?

Answer:Macrocytosis can be a sign of a stressed or failing marrow. Diamond-Blackfan anemia is macrocytic and presents early in life. Fanconi anemia can present at almost any age but is usually macrocytic. Macrocytosis may also be the initial presentation of aplastic anemia.

Question:What about big red blood cells in the newborn?

Answer:They normally have big red blood cells, so the lab needs to be using age-appropriate normals.

Question:My usual closure: What mistakes are we primary care providers making in diagnosis, treatment or referral of anemic patients?

Answer:I see some overdosing of iron with treatment going on for an extended period of time, like more than 6 months. That is usually not necessary, especially if there have been appropriate dietary changes. Physicians often continue treating because the anemia has not resolved or even improved significantly - but that is usually a time to question the original diagnosis. While patients CAN have an MCV in the normal range and still be iron deficient, that iron deficiency is usually mild and responds quickly to iron treatment. Most patients who have significant anemia but are NOT microcytic are not iron-deficient. We also see over-testing for iron deficiency anemia in patients who have a normal MCV. Another testing mistake is the assumption that because a hemoglobin electrophoresis is normal, a child does not have thalassemia: Alpha thalassemia will have normal electrophoresis, but may be picked up with a full "thalassemia screen" that includes a peripheral blood (BCB stain). I see a fair number of thalassemia tests performed on patients who are not microcytic - if a patient is not microcytic, they don't have thalassemia. Also, you should not test for thalassemia while the patient is iron deficient - if they remain microcytic once you have confirmed they are iron-replete, that's the time to test with a thalassemia screen. And finally, there's the mistake of not getting a retic count with a normocytic anemia.

Question:Thank you, Dana.

If you have an item of interest to include in the Quarterly Consult, contact Dr. Steve Dassel via Children's Physician Relations at 206-987-5765.