| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
Central Laboratory
1 Department of Clinical Chemistry
2 Department of Hematology, University
Hospital of Turku, Kiinamyllynkatu 4-8, 20520 Turku, Finland
a Author for correspondence; Fax 358-02-2613920;
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animal and clinical studies indicate that serum concentrations of the soluble TfR reflect the total mass of tissue receptor (13). The serum-soluble transferrin receptor (sTfR) is a truncated form of the previously described intact receptor, lacking the cytoplasmic and transmembrane domains (14). It exhibits a molecular mass of 85 kDa and forms a complex with transferrin in serum. The main sources of the sTfR are the erythropoietic cells of the bone marrow and the circulating reticulocytes that eventually shed their receptors during maturation (15). Other tissues that contain considerable amounts of TfR, such as the liver (16) and the placenta (17), contribute only slightly to the sTfR pool (7). Thus, the concentrations of sTfR correlate directly with erythropoietic activity and inversely with the amount of iron available for erythropoiesis, providing a quantitative measure of functional iron status (18)(19).
Confirmation of the absence of iron stores in a bone marrow aspirate sample is still considered the gold standard for the diagnosis of iron deficiency, yet demand for a less strenuous and noninvasive method is evident. In rheumatoid arthritis and other inflammatory diseases, episodes of anemia occur frequently, and the importance of differential diagnostics is pronounced. The quantitation of sTfR has been shown to differentiate effectively between iron-deplete and iron-replete anemic states, irrespective of the presence of acute or chronic inflammatory reactions (18)(20)(21)(22)(23)(24)(25). This finding emphasizes the clinical usefulness of the quantitation of the circulating receptor, as ferritin is a known acute-phase reactant (26)(27), and concentrations of serum iron and transferrin or total iron binding capacity are known to decrease in the presence of an acute-phase reaction (28).
Although the main focus is toward the detection of iron deficiency in anemic patients, other conditions affect the amounts of sTfR as well. Disorders such as autoimmune hemolytic anemia, sickle cell anemia, hereditary spherocytosis, ß-thalassemia, and polycythemia vera have been reported to increase the number of circulating receptors (29), whereas idiopathic hemochromatosis has been associated with decreased values (30).
Binding assays with radiolabeled transferrin (31)(32) and ELISA-based assays (33)(34)(35) have thus far been used to study the sTfR quantitatively. Both polyclonal (34) and monoclonal (31)(33)(35)(36) antibodies have been directed against the receptor.
The objective of this study was to assess the analytical and functional quality of a newly introduced IDeATM TfR immunoenzymometric assay (IEMA) by Orion Diagnostica. Also, the purpose was to show that the concentrations of sTfR measured by this new kit can be used effectively in the differential diagnosis of iron-deficiency anemia (IDA). We also compared this new method with two other commercially available reagents: ClinigenTM (R&D Systems) and TfR enzyme immunoassay (Ramco Laboratories).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The amount of bound enzyme-labeled antibody that eventually remains after incubations and washing is directly proportional to the concentration of sTfR in the sample. The actual concentrations in unknown samples are deduced by means of a calibration curve on the basis of 6 calibrator specimens of known concentrations of TfR, which are analyzed in parallel with the samples. The calibrator values span the range of 0.7–11 mg/L. No predilution of the samples is routinely required, yet the instruction manual advises the user to dilute and reanalyze patient samples having initial concentrations above that of the highest calibrator. In these situations the samples can be diluted 1/2 and 1/10 with zero-calibrator serum, a sufficient amount of which is provided in the kit. In this study, however, predilution was performed on patient samples yielding concentrations >7 mg/L, which we discovered to mark the upper limit of the true linear range. The total time required to perform the analysis, including all the necessary preparations and using all of the 96 microtitration wells, is ~5 h. All of the procedures in the course of performing the assay should be carried out within the temperature range of 20–25 °C.
The analyses performed on the Clinigen and TfR enzyme immunoassay methods were carried out according to the instructions provided by the manufacturers and with the same equipment as described above. Patient samples with concentrations above the highest calibrators of the respective assays were diluted and reanalyzed according to the instruction manuals.
subjects
The patient population consisted of 98 consecutive anemic adult
patients at the University Hospital of Turku who underwent a bone
marrow examination to define the type of anemia and to determine the
iron stores. Anemia was defined as a hemoglobin concentration of <128
g/L in men and <17 g/L in women, which constitute the lower 2.5%
reference limits in our hospital. All blood samples were obtained
before any blood transfusions. Patients on oral iron therapy were
excluded from the study population. Patients with hematological
malignancies were also excluded from this study, as certain
hematological malignancies have been reported to be associated with
increased serum TfR concentrations regardless of the iron status of the
patients (19)(34). Additionally, patients who
had hemolytic anemia or defined deficiency of vitamin B12
or folic acid were excluded from our study population, because these
conditions may be associated with increased sTfR concentrations
irrespective of iron status (37). The patients were
assigned to one of three groups on the basis of the bone marrow
examination and clinical data. Forty-five patients, who fulfilled the
morphologic criteria of iron deficiency and who had no stainable iron
in the bone marrow, were classified as having IDA. Thirty-six patients,
who presented with stainable iron in the bone marrow, were classified
as having anemia of chronic disease (ACD). Of these 36 patients, 23 had
recurrent or chronic infections, whereas the remaining 13 had other
chronic diseases, i.e., nonhematological malignancies and inflammatory
diseases such as rheumatoid arthritis. Another 17 patients, who had no
iron in the bone marrow together with an infectious disease, a chronic
inflammatory disease (rheumatoid arthritis or ulcerative colitis), or a
nonhematological malignancy were placed in a group of combined anemic
states (COMBI). These patient populations are a subfraction of the
patient populations in which sTfR concentrations have been evaluated
recently with the R&D sTfR assay (25). The serum samples
of 119 apparently healthy nonanemic subjects (23 men and 96 women, ages
24–69 years) were obtained and used as controls.
statistical analyses
The reference range was calculated from the healthy population by
the nonparametric method according to IFCC recommendations with
SAS® 6.10 software (Sas Institute). ROC curves were
constituted, and the corresponding areas under the curves
(AUCROC) were derived with the GraphROC for Windows
software package (38)(39). GraphROC was also
used to determine the optimal decision limit of sTfR for iron
deficiency. P values were calculated by a two-tailed
t-test with a Microsoft® Windows for
WorkgroupsTM software package (Microsoft). The correlation
scatter plots and the distribution profiles were calculated and
visualized with Fig.P® software (Fig.P Software). The
protocol of this study was approved by the Joint Committee of Ethics of
the University Hospital of Turku and the University of Turku.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Precision.
Intraassay CVs (n = 20) for four serum samples
(mean concentrations 1.6–5.8 mg/L) were 3.4–6.3%; CVs calculated
from duplicate results in 10 subsequent assays were 4.0–6.1%.
Linearity.
Serum samples were diluted with zero-calibrator
serum. The mean results (Table 1
) were 100.2% of expected (range 84.4–122.2%). The assay was
linear from 0.3–7 mg/L (not shown). When the concentration of sTfR in
the sample exceeds 7 mg/L, the sample can be diluted 1/2 and 1/10 with
zero-calibrator serum and reanalyzed.
|
Recovery.
Recovery of purified TfR added to low serum samples
was 91.9–102.4% (mean 96.6%). The evaluation was carried out at
concentrations within the linear range, close to the upper limit of the
reference range. When the expected concentrations exceeded 7 mg/L,
i.e., the upper limit of the linear range, the observed recoveries
decreased notably.
Drift.
Three serum samples and three controls were pipetted at
intervals of 0–25 min to simulate the delay in the course of
pipetting. No drift was observed.
clinical performance
We tested the IDeA for its capability to detect iron deficiency,
i.e., to differentiate between IDA patients and those with ACD. The
COMBI group was included so we could study whether the sTfR
concentrations differ from the group with a noncomplicated iron
deficiency in the presence of an inflammatory reaction. The mean sTfR
in the control group was 2.1 mg/L, and the 95% interval was 1.3–3.3
mg/L.
In the iron-deficient group (IDA) the mean sTfR concentration was 10.9
mg/L (range 3.4–29.0 mg/L). The mean value in the ACD group was 2.4
mg/L (range 1.2–4.2 mg/L), similar to the control values. The
corresponding values for the COMBI group were 9.8 mg/L and 2.0–34.0
mg/L. Fig. 1
displays the distribution profiles for each group in relation
to the upper limit of the reference range.
|
The capability of the IDeA method to discriminate between these three
groups was studied by means of ROC curves. Subsequently, the
corresponding AUCROC depicting the diagnostic accuracy
were calculated and compared with results obtained with the Ramco and R
& D methods (Table 2
). All three methods could well establish the difference between
the patient groups.
|
Correlation coefficients for all patient data (r) and for
the linear ranges of the respective methods
(rLin) were calculated separately (Fig. 2
). All patient samples with initial concentrations >7 mg/L on
the IDeA method and above the highest calibrators (5 and 40 mg/L,
respectively) on the R & D and Ramco methods were diluted and
reanalyzed. We encountered evidence that the linear ranges of the R & D
and Ramco assays also had upper limits (3 mg/L for the R & D and 16
mg/L for the Ramco) that were substantially lower than the respective
highest calibrators. These upper limits of the linear ranges were used
to calculate the rLin values.
|
The methods were observed to correlate fairly well within their
respective linear ranges, whereas the overall correlation of the Ramco
method with the other two assays was notably impaired (Fig. 2
).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The diagnostic accuracy of the IDeA assay was assessed by analyzing three groups of anemic patients, whose iron status had been confirmed from a bone marrow aspirate, and by comparing their sTfR concentrations with those of apparently healthy volunteers. The concentrations of the ACD subjects did not differ from the reference values (P = 0.85); the values of the IDA and COMBI groups were significantly higher (P <0.00001). ROC curves and corresponding AUCROC were used to further assess the clinical applicability of the assay and to compare its differentiating powers with the Ramco and R & D assays. All three assays could well establish the difference between iron-deficient and iron-replete groups, even in the presence of an inflammatory component (COMBI). A slight overlap was observed between the ACD and COMBI groups.
As shown by the scatter plots, the results obtained by the three
different methods correlate, as long as the sTfR concentrations remain
within the linear ranges of the respective methods (Fig. 2
). The
difference between the correlation coefficients of the overall patient
data (r) and the data restricted to the linear ranges
(rLin) indicates that dissociation in
correlation occurs as the values increase beyond the linear ranges. The
overall correlation between the IDeA and R & D assay remains
satisfactory, and the exclusion of values beyond the linear ranges of
the respective methods results in only a minor improvement in
correlation. The overall results obtained by the Ramco assay, however,
correlate poorly with results given by the other two assays. The
rLin values of all three comparisons were
similar. The instruction manuals supplied with each assay advise the
user to dilute and reanalyze any patient samples that yield
concentrations above the highest calibrators of the particular methods.
Because the analytical performance of the IDeA was studied extensively,
we determined the true upper limit of the linear range and used it as
the cutoff value, above which all patient samples were diluted and
reanalyzed. We discovered, however, that the linear ranges of the other
two methods also had upper limits (3 mg/L for R & D and 16 mg/L for
Ramco) that were substantially lower than the highest calibrators of
the respective methods (data not shown).
In whole, none of the evaluated assays maintained their linearity throughout the range indicated by the calibrators. All three instruction manuals advise the user to dilute and reanalyze samples that initially yield concentrations above the highest calibrators, and in doing so imply that the highest calibrators in fact represent the upper limits of the linear ranges. Our findings contradict these implications. We believe that the difference between the r and rLin values is mainly a result of the false assumption regarding the extent of linearity on the R & D and Ramco assays and could be evened by implementing the upper limits of the linear ranges as cutoff values for reanalysis.
Even though the performance characteristics and clinical usefulness of the three kits are similar, the actual numeric values differ significantly (P <0.000001) and cannot be compared as such. The lack of uniform, international calibrators is the main reason for this. The methods and reagents used in the calibration processes differ, as do the isolation and purification procedures, by which the calibrations have been obtained. Also, the assay conditions for the composition of the diluents and buffers vary. The characteristic affinities and binding sites of the different monoclonal antibodies also introduce a plethora of sources of interference, and the conditions, in which the antibody binds the analyte, have therefore been tuned specifically. The possible inhibitory effect of serum depends on the initial amount of serum pipetted into the wells as well as on the properties of the antibody and poses therefore various effects on the different assays. In the IDeA method intact samples are pipetted directly into the wells, whereas the other two methods require predilution. We are therefore justified in assuming that the matrix effects typically involved in immunochemistry are responsible for some of the inconsistencies in the interpretation of results obtained by the different sTfR assays.
In a recent study the conventional reference range for ferritin was found to be considerably lower than the optimal decision limit for iron deficiency (25). These values have generally been calculated from a population that has been determined to be nonanemic on the basis of hemoglobin values only. This is, however, an inadequate measure to exclude the effect of latent iron deficiency on the reference values of ferritin, because anemia is known to develop as a late manifestation of iron deficiency. During this study we have encountered the same problem concerning the reference values of sTfR, because healthy volunteers cannot be subjected to bone marrow sampling. Thus, the actual iron status of the population considered "normal" could not be verified. The optimal decision limit of the IDeA method for iron deficiency, however, differs only slightly from the upper limit of the reference range (3.6 and 3.3 mg/L, respectively). Therefore, the capability of sTfR to differentiate between iron-deplete and iron-replete patient groups seems to be considerably less compromised by this matter than is serum ferritin.
The IDeA sTfR assay has proved to be an analytically accurate and clinically promising tool for future studies concerning the differential diagnosis of IDA. In this study we were able to confirm that by measuring the serum sTfR concentrations a good differentiation between the iron-deplete and iron-replete patient groups is possible, irrespective of the concurrent degree of inflammation. All three methods performed similarly.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. E Cogswell, A. C Looker, C. M Pfeiffer, J. D Cook, D. A Lacher, J. L Beard, S. R Lynch, and L. M Grummer-Strawn Assessment of iron deficiency in US preschool children and nonpregnant females of childbearing age: National Health and Nutrition Examination Survey 2003-2006 Am. J. Clinical Nutrition, May 1, 2009; 89(5): 1334 - 1342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Vikstedt, P. von Lode, T. Takala, K. Irjala, O. Peltola, K. Pettersson, and P. Suominen Rapid One-Step Immunofluorometric Assay for Measuring Soluble Transferrin Receptor in Whole Blood Clin. Chem., October 1, 2004; 50(10): 1831 - 1833. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Halonen, J. Mattila, P. Suominen, T. Ruuska, M. K. Salo, and A. Makipernaa Iron Overload in Children Who Are Treated for Acute Lymphoblastic Leukemia Estimated by Liver Siderosis and Serum Iron Parameters Pediatrics, January 1, 2003; 111(1): 91 - 96. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. S. Asobayire, P. Adou, L. Davidsson, J. D Cook, and R. F Hurrell Prevalence of iron deficiency with and without concurrent anemia in population groups with high prevalences of malaria and other infections: a study in Cote dIvoire Am. J. Clinical Nutrition, December 1, 2001; 74(6): 776 - 782. [Abstract] [Full Text]
|
|
|
|
|
|
G. Van den Bosch, J. Van den Bossche, C. Wagner, P. De Schouwer, M. Van De Vyvere, and H. Neels Determination of Iron Metabolism-related Reference Values in a Healthy Adult Population Clin. Chem., August 1, 2001; 47(8): 1465 - 1467. [Full Text] [PDF]
|
|
|
|
|
|
P. Suominen, A. Virtanen, M. Lehtonen-Veromaa, O. J. Heinonen, T. T. Salmi, M. Alanen, T. Mottonen, A. Rajamaki, and K. Irjala Regression-based Reference Limits for Serum Transferrin Receptor in Children 6 Months to 16 Years of Age Clin. Chem., May 1, 2001; 47(5): 935 - 937. [Full Text] [PDF]
|
|
|
|
|
|
M. Olivares, T. Walter, J. D Cook, E. Hertrampf, and F. Pizarro Usefulness of serum transferrin receptor and serum ferritin in diagnosis of iron deficiency in infancy Am. J. Clinical Nutrition, November 1, 2000; 72(5): 1191 - 1195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. T. Goodnough, B. Skikne, and C. Brugnara Erythropoietin, iron, and erythropoiesis Blood, August 1, 2000; 96(3): 823 - 833. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. Harthoorn-Lasthuizen, P. vant Sant, J. Lindemans, and M. M.A.C. Langenhuijsen Serum Transferrin Receptor and Erythrocyte Zinc Protoporphyrin in Patients with Anemia, Clin. Chem., May 1, 2000; 46(5): 719 - 722. [Full Text] [PDF]
|
|
|
|
|
|
J. W. Choi, S. H. Pai, M. W. Im, and S. K. Kim Change in Transferrin Receptor Concentrations with Age Clin. Chem., September 1, 1999; 45(9): 1562 - 1563. [Full Text] [PDF]
|
|
|
|
|
|
P. Suominen, K. Punnonen, A. Rajamaki, R. Majuri, V. Hanninen, and K. Irjala Automated Immunoturbidimetric Method for Measuring Serum Transferrin Receptor Clin. Chem., August 1, 1999; 45(8): 1302 - 1305. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Brugnara, D. Zurakowski, J. DiCanzio, T. Boyd, and O. Platt Reticulocyte Hemoglobin Content to Diagnose Iron Deficiency in Children JAMA, June 16, 1999; 281(23): 2225 - 2230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Suominen, K. Punnonen, A. Rajamaki, and K. Irjala Serum Transferrin Receptor and Transferrin Receptor-Ferritin Index Identify Healthy Subjects With Subclinical Iron Deficits Blood, October 15, 1998; 92(8): 2934 - 2939. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. S. Skikne Circulating Transferrin Receptor Assay—Coming of Age Clin. Chem., January 1, 1998; 44(1): 7 - 9. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, ACD; 
, COMBI.