In Reply

We agree with Dr. Ostendorf that the lower reference limit of ferritin that is normally used is probably already in the range of an initial stage of iron deficiency, at least if iron deficiency is defined by the fact that intestinal iron absorption has already been upregulated (by reducing hepcidin levels).

We can imagine that the ferritin limits for iron deficiency will be redefined in the future, once standardized hepcidin values are available.

Dr. Ostendorf is right that the lower limit of the normal range for reticulocyte hemoglobin (CHr) is usually given as 28 pg. We gave the lower limit value (1), which also allows an iron deficiency to be diagnosed with great certainty in hemodialysis patients. Fishbane et al. (2) report that patients with a CHr <26 pg show an increase in reticulocytes when they receive intravenous iron dextran. The iron deficiency determined in this way could be diagnosed with a lower CHr limit of 26 pg, with a sensitivity of 100% and a specificity of 80%. Similarly, Mittman et al. (3) also detect iron deficiency with the slightly higher limit of 28 pg, but with reduced sensitivity and specificity. In our review article (1), we should probably have named the limit value that has the greatest general acceptance.

We would support a proposal to use the term “mean reticulocyte hemoglobin” (MRH, in analogy to MCH) to refer to the hemoglobin content of the reticulocytes, in a company-independent manner. We also see the advantage of using MRH as an iron deficiency parameter with a fast response time.

In his contribution to the discussion, Dr. Steinmetz rightly points out that it is problematic to interpret reduced transferrin saturation (TSAT) in oncological patients as being indicative of iron deficiency. Due to the inflammatory activity that is often present in cancer, TSAT can be reduced without having a deficiency in total body iron stores. Therefore, strictly speaking, the limit value of < 20% specified by the WHO for the diagnosis of iron deficiency should only be used for patients who do not have cancer or chronic inflammation.

We agree with Dr. Steinmetz that the soluble transferrin receptor (sTfR) and the reticulocyte hemoglobin content (CHr) are well-suited parameters for assessing the iron supply of erythropoiesis. We have also stated this in our text.

We thank Dr. Alexopoulos for the valuable addition to the problem of interpreting laboratory results in patients with chronic kidney disease. Because of the strict limitations on the scope of this wide-ranging topic, we could not go into more detail about the connections he presented.