| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Clinical Pathology, College of Medicine, the Catholic University of Korea, Uijongbu St. Mary’s Hospital, 65-1 Kumoh-Dong, Uijongbu-City, Kyunggi-Do 480-130, Korea;
2
Department of Clinical Pathology, College of Medicine, the Catholic University of Korea, Kangnam St. Mary’s Hospital, 505 Banpo-dong Seocho-ku, Seoul 137-040, Korea
aauthor for correspondence: fax 82-2-592-4190, e-mail ejoh{at}catholic.ac.kr
The soluble transferrin receptor (sTfR) has been introduced as a promising new diagnostic tool for differentiating between iron deficiency anemia (IDA) and anemia of chronic disease (ACD) (1)(2)(3). The circulating sTfR concentration is proportional to cellular expression of the membrane-associated TfR and increases with increased cellular iron needs and cellular proliferation (4). Furthermore, because serum ferritin reflects the storage iron compartment and sTfR reflects the functional iron compartment, the sTfR/log ferritin index (sTfR-F index), based on these two values, has been suggested as a good estimate of body iron compared with the sTfR/ferritin ratio (5). However, whether they could be useful in evaluating the iron deficiency in various malignancies has not been reported. In addition, some data have demonstrated that sTfR offers little advantage over conventional laboratory indicators of iron status (6) and might not assess the iron status of patients with ACD. A potential explanation for these differences may be the ACD patient population studied. Because several studies of patients with solid malignancies have reported that the erythropoietin concentrations are inappropriate for the degree of anemia (7) and chemotherapy-induced bone marrow (BM) suppression may also decrease sTfR concentration, the sTfR concentrations in the ACD population may depend on the proportion of patients with malignancy in the study group.
In the present study, we assessed the diagnostic performance of sTfR, ferritin, and sTfR-F index for detecting iron depletion in several groups of patients (IDA, chronic inflammation or infection, and nonhematologic malignancy) according to the guidelines (8). The diagnostic classification of all patients was based on an examination of the BM iron stain as the gold standard for iron depletion.
The patient population consisted of 120 (58 men and 62 women; age range, 21–85 years; mean, 54 years) anemic adult patients who underwent a BM examination for anemia study and 81 nonanemic controls. The 120 anemic patients were divided into five populations on the basis of the BM examination and clinical data: IDA (n = 31), which included patients who had no stainable iron in the BM because of an uncomplicated iron deficiency (i.e., simple blood loss); I-IDA (n = 15), which included patients who had chronic inflammatory disease (chronic infection, rheumatoid arthritis, liver cirrhosis, or chronic renal failure) accompanying a C-reactive protein (CRP) concentration >4 mg/L (mean ± SD, 45.2 ± 60.5 mg/L) and no iron in the BM; I-ACD (n = 23), which included patients with chronic inflammatory disease (CRP, 32.4 ± 29.1 mg/L) and stainable iron in the BM; M-IDA (n = 26), which included patients with nonhematologic malignancy and no iron stores; and M-ACD (n = 25), which included patients with nonhematologic malignancy and iron stores. The nonanemic control groups included 24 healthy blood donors (Control), 32 patients with a chronic inflammatory disease (CRP, 35.4 ± 43.2 mg/L; I-Control), and 25 patients with a nonhematologic malignancy (M-Control).
Anemia was defined as hemoglobin <140 g/L in men and <120 g/L in women. All blood samples were obtained before blood transfusion. Patients who had hematologic malignancies, hemolytic anemia, a defined deficiency of vitamin B12 or folic acid, or marked hypocellularity of the BM as well as patients on oral iron therapy were excluded because these conditions can influence sTfR concentrations irrespective of iron status (9). For patients with malignancies, chemotherapy- and radiotherapy-naive patients were selected to exclude the effect of hypocellularity on sTfR irrespective of iron status. The patients with malignancies had gastrointestinal malignancies or other solid tumors, including Hodgkin and non-Hodgkin lymphoma, lung cancer, or ovarian cancer.
To detect iron in the BM, both aspiration smear specimens and biopsies from the iliac crest were stained with Prussian blue. In each sample, positive and negative controls were performed, and a minimum of three particles of BM were examined. Routine complete blood cell counts and red cell indices were measured with a SE9000 electronic counter (Sysmex Co.). For serum sTfR assays, we used an automated immunoturbidimetric method (IDeA sTfR-IT; Orion Diagnostica) (10) on a 7600 analyzer (Hitachi). The intraassay CVs were 1.5% and 2.0% for two serum samples with 10 replicates, and the interassay CVs were 2.5% and 3.2% for two serum samples in 10 assays during 2 weeks. The serum ferritin and CRP concentrations were measured with a nephelometer (Behringwerke AG), and the sTfR-F index (sTfR/log ferritin) was calculated. ROC curve analysis was performed with SPSS 10.1, MedCalc, and AccuROC for Window.
The results for the iron status markers and diagnostic accuracy with optimal cutoffs defined by the ROC curves analysis are summarized in Table 1
and Fig. 1
, according to the study groups. The areas under the curves (AUCs) for distinguishing between the IDA (n = 24) and ACD (n = 48) were 0.995 for sTfR-F index, 0.987 for ferritin, and 0.936 for sTfR. The sTfR-F index (cutoff of >1.36), ferritin (cutoff of 
35 µg/L), and sTfR (cutoff of >1.8 mg/L) had sensitivities of 100%, 94%, and 97%, respectively, and specificities of 98%, 98%, and 88%, but there was no significant difference among the AUCs for these tests (P >0.05).
|
|
In patients with a chronic inflammatory disease, both ferritin 153 µg/L and sTfR-F index >0.75 were favorable markers for discriminating between I-IDA and I-ACD (AUCferritin = 0.870; AUCsTfR-F index = 0.865; AUCsTfR = 0.704), but sTfR-F index did not improve the diagnostic efficiency for detecting iron depletion compared with ferritin alone (P >0.05). The poor diagnostic accuracy of sTfR may be attributable to an adequate iron supply for erythropoiesis provided by circulating iron (absorbed from the intestine or recycled from senescent red cells) even without iron stores (11), or functional iron deficiency even with adequate iron stores (12).
In the patients with a nonhematologic malignancy, there was no good indicator for detecting iron depletion (AUCsTfR = 0.543, AUCsTfR-F index = 0.613, and AUCferritin = 0.777). The low diagnostic accuracy of sTfR in this population suggests that anemia caused by a malignancy can be associated with a decrease in the total BM erythropoietic activity, a decrease in TfR cellular expression, disproportionate expression of cellular TfR compared with sTfR, or the suppression of iron mobilization by the malignancy. The pathogenesis of the anemia caused by cancer remains unclear, but it may involve a combination of the shortened survival of erythrocytes in circulation with a failure of the BM to increase the red cell population in compensation under the influence of several cytokines (13). This study could not confirm the pathogenesis of the anemia caused by cancer, but it supports the report by Dowlati et al. (14) in that anemia of cancer is mainly attributable to an impaired erythroid marrow response to erythropoietin stimulation.
Serum ferritin has been widely used to define iron depletion, and the results of our study also confirmed it. However, careful diagnostic classification of the patients as well as knowledge of all the factors that may cause changes in serum ferritin is required (5). Serum ferritin concentrations 30, 50, and 100 µg/L have been cited as evidence for iron deficiency in anemic patients with coexisting chronic diseases such as inflammation, infection, and malignancy (15)(16)(17). However, in our study, these cutoffs might be too low, and serum ferritin with different cutoff values (153 µg/L in patients with inflammation or infection and 257 µg/L in patients with a nonhematologic malignancy) was a favorable iron status marker in each population. Furthermore, considering the patients with a malignancy, although it had low diagnostic value, serum ferritin (cutoff of 257 µg/L) was the best indicator of iron status in patients with a nonhematologic malignancy, providing a sensitivity of 77% and a specificity of 76%. Caution is urged in extrapolating the serum ferritin results of this study to other patients with hypocellularity or BM involvement of a malignancy. This study was limited to patients with a malignancy who had normocellular BM and no evidence of BM involvement.
Because serum ferritin varies with iron stores, whereas sTfR is assumed to reflect the degree of the tissue iron supply (1)(18), the sTfR-F index has been suggested as a good estimate of body iron (3)(5). However, in this study, the sTfR-F index did not improve diagnostic accuracy compared with serum ferritin alone in the populations studied, possibly because of the low specificity of the sTfR.
In conclusion, these results are in agreement with some reports (15)(19) showing that sTfR is not superior to ferritin for detecting iron depletion. Furthermore, in well-defined patients with a nonhematologic malignancy, sTfR did not reflect the iron status because of its unknown mechanism. Therefore, evaluation of iron status in patients with chronic disease requires different serum ferritin cutoffs according to diagnostic classification, and the sTfR-F index adds information on patients with chronic inflammation or infection.
Acknowledgments
This study was supported by grants from the Samkwang Medical Foundation. We thank Seong Hee Kim and Hyeon Im Lee for excellent technical assistance.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. S Shord, J.M. Hamilton Jr, and S. Cuellar Parenteral iron with erythropoiesis-stimulating agents for chemotherapy-induced anemia Journal of Oncology Pharmacy Practice, March 1, 2008; 14(1): 5 - 22. [Abstract] [PDF]
|
|
|
|
 |
|
 A. S. W. Mburu, D. I. Thurnham, D. L. Mwaniki, E. M. Muniu, F. Alumasa, and A. de Wagt The Influence and Benefits of Controlling for Inflammation on Plasma Ferritin and Hemoglobin Responses following a Multi-Micronutrient Supplement in Apparently Healthy, HIV+ Kenyan Adults J. Nutr., March 1, 2008; 138(3): 613 - 619. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A Northrop-Clewes Interpreting indicators of iron status during an acute phase response - lessons from malaria and human immunodeficiency virus Ann Clin Biochem, January 1, 2008; 45(1): 18 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Kupka, G. I. Msamanga, F. Mugusi, P. Petraro, D. J. Hunter, and W. W. Fawzi Iron Status Is an Important Cause of Anemia in HIV-Infected Tanzanian Women but Is Not Related to Accelerated HIV Disease Progression J. Nutr., October 1, 2007; 137(10): 2317 - 2323. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, control;
, IDA; 
, patient population with infection or inflammatory disease;
, patient population with nonhematologic malignancy. Shown are interquartile ranges (boxes), medians (horizontal lines inside boxes), outliers (circles), and extremes (asterisks). The cutoff values for sTfR (>1.8 mg/L; A), serum ferritin [