We found comparable serum hepcidin levels in patients with ACD and IDA. The 5 to 95% reference interval in our control population was comparable to the data delivered by the manufacturer. It is noteworthy that none of our control subjects had clinical or laboratory signs of inflammation or iron deficiency.
Our results are, however, not in accordance with four recent studies [6
]. Ganz et al. found that the serum hepcidin levels were undetectable in 18 of the 19 patients with iron deficiency (serum ferritin <10μ
g/L) but significantly elevated in patients with inflammation defined as a CRP >10
mg/L. There is no information regarding the ferritin concentration in patients with inflammation in this study [6
]. In another study, Theurl et al. showed that patients with ACD had significantly higher serum hepcidin levels than patients with IDA, ACD/IDA, and controls [8
]. Due to the definition of subpopulations in the study by Theurl et al., ferritin was also significantly different between the subgroups of ACD with and without iron deficiency [8
]. Thomas et al. studied a 155 anaemic patients with latent iron deficiency and reported that serum hepcidin can differentiate patients with IDA from ACD and ACD/IDA. Serum hepcidin correlated with ferritin and ferritin index [9
]. On the other hand, urinary hepcidin levels were not correlated with proinflammatory markers in a large sample of an older general population [7
There are several limitations to our study. Our study population is by its nature a heterogeneous group of vulnerable geriatric patients. Aging is associated with increased inflammation and a mild rise in inflammatory markers such as IL-6, but additional laboratory analyses were not performed. A higher CRP level is common in hospitalized older patients with IDA [3
], and 3 out of the 11 patients with IDA had a CRP >10
mg/L. We are also aware that the number of patients is limited, but we doubt whether a larger group could significantly alter our results. Although there are no standard criteria for the diagnosis of ACD and IDA, we chose to define our study groups according to well-established laboratory criteria. Finally, several mass spectrometry and immunochemical methods have been developed for hepcidin quantification in serum and urine. Currently, there is no reference method for hepcidin measurements. A generally accepted reference interval is not yet available at this moment, nor a validated cutoff for the diagnosis of IDA or ACD. In the study by Kroot et al., several quantitative methods were compared [10
]. The analytical variation of the methods was comparable and was low for all methods [10
]. However, the hepcidin concentrations in urine and serum differed considerably between all methods. These differences are suggested to be caused by (i) the use of different calibrators, (ii) possible hepcidin aggregation in the sample or the standard solution, (iii) measuring only free, only bound, or both fractions of hepcidin, or (iv) by interference of hepcidin isoforms [10
]. The commercial DRG ELISA assay used in this study was not included in the comparison study of Kroot et al. [10
In conclusion, this is the first study that investigates the possible role of serum hepcidin for the diagnosis of different types of anemia in a geriatric hospitalized population using a commercially available DRG ELISA kit. According to other recent studies, we would expect higher hepcidin levels in ACD and lower levels in IDA patients as compared to our control group. The reason for this is not clear, although an assay-related problem that could explain these inadequate results is possible. However, at this moment it is premature to draw firm conclusions, and further investigation is needed.