Clinical Chemistry AACC Online Job Center
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 43: 2408-2412, 1997;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gambino, R.
Right arrow Articles by Gunter, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gambino, R.
Right arrow Articles by Gunter, E.
Related Collections
Right arrow Laboratory Management
Right arrow General Clinical Chemistry
Right arrow Proteomics and Protein Markers
Right arrow Hematology
(Clinical Chemistry. 1997;43:2408-2412.)
© 1997 American Association for Clinical Chemistry, Inc.

Articles

The relation between chemically measured total iron-binding capacity concentrations and immunologically measured transferrin concentrations in human serum

Raymond Gambino1,a, Edouard Desvarieux1, Michael Orth2, Holly Matan3, Tomy Ackattupathil1, Elenore Lijoi1, Cathy Wimmer1, John Bower3 and Elaine Gunter4

1 Quest Diagnostics Inc., 1 Malcolm Ave., Teterboro, NJ 07608-1070;
2 Wallingford, CT 08492; and
3 Horsham, PA 19044.

4 Centers for Disease Control and Prevention, Atlanta, GA 30341.
a Author for correspondence. Fax 201-393-5903; e-mail Doclab{at}aol.com


   Abstract
Top
 Abstract
Introduction
 Design of Experiments
 Results
 Discussion
 References
 
We sought to determine if serum total iron-binding capacity (TIBC) is equivalent to serum transferrin (TRF) so that a low-cost colorimetric chemical assay for unsaturated iron-binding capacity (UIBC) could be substituted for a high-cost immunologic assay for TRF. Our study design included independent and blinded measurements of UIBC, serum iron, and TRF concentrations in human serum samples. Data from five independent correlation studies carried out at three different Quest Diagnostics laboratories were combined into one data set containing 570 paired results for TIBC and TRF. r2 was 0.941 when three outliers were eliminated from the 570-sample data set. Scatter about the regression line was fully accounted for by the CVs for the TIBC and TRF assays. When each test is measured precisely and without bias, the ratio of TIBC (µmol/L) to TRF (g/L) in SI units is close to the theoretically expected value of 25.0.


   Introduction
Top
 Abstract
 Introduction
Design of Experiments
 Results
 Discussion
 References
 
Theoretically, 2 mol of iron (molecular mass 55.84 Da) bind to 1 mol of serum transferrin (TRF) (80 000 Da).1 Chemical theory, therefore, predicts that 25.0 µmol of iron will saturate the binding sites of 1 g of TRF (1)(2). If the binding of iron to TRF is uncomplicated, then a high degree of correlation between the two measurements is likely and the expected ratio of total iron-binding capacity (TIBC) in µmol/L to TRF in g/L should be 25.0 (i.e., TIBC µmol/L = 25.0 x TRF g/L). Tsung et al. report that TIBC and TRF are indeed equivalent (3).

The relation between the two measurements, however, could vary a great deal if iron binds to other proteins to any significant degree, or if the binding of iron to TRF is altered as iron concentrations change, or if either assay has poor precision or bias, or both.

Several articles suggest that, contrary to theory, the relation between TIBC and TRF is not fixed, especially when results are outside the reference range (4)(5). The reported mean ratio between TIBC and TRF has ranged from a low of 11.3 to a high of 26.1 (6)(7).

We undertook this multilaboratory study of the relation between TIBC and TRF to investigate these discrepancies.


   Design of Experiments
Top
 Abstract
 Introduction
 Design of Experiments
Results
 Discussion
 References
 
analytical methods
TIBC was calculated from the sum of measured unsaturated iron-binding capacity (UIBC) and measured serum iron (i.e., TIBC = UIBC + serum iron). At all three laboratories, UIBC and serum iron were measured colorimetrically on either Hitachi 736–50 (Boehringher Mannheim Corp.) or Olympus AU5200 (Olympus America) high-throughput chemistry analyzers with reagents and aqueous or bovine-based calibrators from Diagnostic Chemicals Limited (DCL). The DCL assay includes Ferene® to form an iron–ferritin complex with maximum absorption at 593 nm and a molar absorptivity of 35 500.

The UIBC assay is essentially identical on each of these high-throughput chemistry analyzers. A precisely known amount of iron in solution is added to diluted serum. The added iron binds to previously unsaturated binding sites on TRF. After a short time—<5 min—the amount of iron that did not bind to TRF is measured colorimetrically. UIBC is defined as the difference in the color produced by the original amount of added iron (highest color) and the color produced by the residual iron in solution that did not bind to TRF (less color).

TRF was measured immunologically on either the Behring Nephelometer II (Behring Diagnostics) or the Beckman Array® 360-CE (Beckman Instruments), or both, with respective calibrators from Behring or Beckman. The calibrators were standardized to the IFCC's preparation for plasma proteins Clinical Reference Material lot 470, which was certified by the Bureau of Reference of the European Community and designated the Reference Preparation for Proteins in Human Serum by the College of American Pathologists. This new calibrator yields TRF concentrations that are 13% lower than those obtained with the older calibrator. Table 1 specifies the nephelometer used and in which laboratory. The two nephelometers differ in how they measure the protein concentrations. The Beckman Array monitors the rate at which scattered light changes. The Behring nephelometer measures the amount of light scattered at a defined end point.


View this table:
[in this window]
[in a new window]
  		Table 1. Selection criteria for 570 patient samples included in five independent studies of paired TRF and TIBC assays and the instruments used for the assays.

precision studies
Two independent precision studies were conducted: one in March 1996 in the Wallingford laboratory, and one in October 1996 in the Teterboro laboratory. In each study, human serum pools were assayed in duplicate once daily for 20 days. The Wallingford study used three pools: one with a low TIBC concentration (~30 µmol/L), one with a normal TIBC concentration (~50 µmol/L), and one with a high TIBC concentration (~70 µmol/L). The Teterboro study used two pools: one with a saturation of TRF <15%, and one with a saturation of TRF >50%. All pools were generated from discarded samples of serum that had been separated from cells by centrifugation of a barrier tube soon after blood had been collected. Once assembled, each pool was aliquoted into multiple 5-mL plastic vials and stored frozen (-30 °C) until used.

bias studies
Bias was assessed in two independent experiments (April 1996 and January 1997). In each experiment, three frozen reference pools (9114, 9115, and 9116) were obtained from the CDC. The TIBC and serum iron concentrations of each pool were previously defined by CDC with a reference AutoAnalyzer method (8) based on the procedures of Giovanniello et al. (9) and of Ramsey (10). We measured the TRF concentration of each pool immunologically on Behring and Beckman equipment at the Teterboro and Wallingford laboratories, respectively. The pools were also assayed colorimetrically on Olympus equipment at the Teterboro and Wallingford laboratories.

patient sample selection
We studied 570 serum samples from 491 patients. Study samples were selected from serum samples submitted for analysis of iron and TIBC concentrations to each of the three laboratories (Teterboro, Horsham, and Wallingford) in the Quest Diagnostics laboratory network (formerly Corning Clinical Labs.). Samples were specially selected to include a wide range of TIBC and TRF concentrations as well as high and low percent saturation of TRF with iron. Three independent studies of patient samples were conducted at the Teterboro laboratory over 13 months; one study was conducted at the Horsham laboratory over 2 weeks, and one study was conducted at the Wallingford laboratory where 72 samples were assayed for TRF by the Beckman immunologic assay; 52 of the 72 were also assayed for TRF by the Behring immunologic assay. Patient samples from the Horsham and Wallingford laboratories were also analyzed for TIBC and TRF at the Teterboro laboratory.

Table 1Up lists the selection criteria used in each of the five studies, and Table 2 lists the range of values for TRF and TIBC, percent saturation of TRF, and the ratios of TIBC to TRF encompassed by the five studies. The mean ratio of TIBC to TRF for 567 patient samples (570 - 3 outliers) was 22.3—not the ideal ratio of 25.0. The TIBC/TRF ratio of the 567 samples ranged from a low of 17.6 to a high of 29. This wide range can be attributed to the inherent imprecision of Behring's immunoassay for TRF, and to the unavailability of standardized human-based calibrators for UIBC. The TIBC/TRF ratios of the three outliers were 15.2, 13.8, and 10.0; the TIBC concentrations were 43.7, 36.5, and 41.0 µmol/L; and the TRF concentrations were 2.87, 2.64, and 4.09 g/L, respectively.


View this table:
[in this window]
[in a new window]
  		Table 2. Characteristics of the 567 patient samples used in five independent studies.1


   Results
Top
 Abstract
 Introduction
 Design of Experiments
 Results
Discussion
 References
 
precision
In two independent studies, the CV of the colorimetric assay for TIBC ranged from 2.1% to 4.1% (two experiments, five pools), the CV of the Behring end-point nephelometric immunoassay for TRF ranged from 4.4% to 6.0% (two experiments, five pools), and the CV of the Beckman kinetic nephelometric immunoassay for TRF ranged from 2.1% to 2.9% (one experiment, three pools).

bias
When CDC's target values for TIBC were muliplied by 25.0 we obtained a theoretical TRF concentration that was in close agreement with the immunologically measured TRF concentration as shown in Table 3 . For the three pools, the ratio of CDC's target concentration for TIBC to our measured TRF concentration ranged from a low of 23.6 to a high of 24.7. This is close to the theoretically expected ratio of 25.0, and confirms the validity of using the concentration of TRF to check the bias of the TIBC assay. Similar results were obtained in two separate studies that were 10 months apart. Each study was carried out in two laboratories. Table 3 also demonstrates that all laboratories that follow current Hitachi and Olympus recommendations for measuring UIBC are underestimating TIBC by ~8–14% when compared with expected values based on CDC reference pools. This a generic problem with the manufacturers' methods and with currently available calibrators from manufacturers. The individual data points in these bias experiments also reconfirmed the higher imprecision of the Behring immunoassay for TRF when compared with the Beckman immunoassay.


View this table:
[in this window]
[in a new window]
  		Table 3. Data from two separate studies in each of two laboratories comparing target and measured values (mean of 5) for reference pools from the CDC.

patient correlations
Figure 1 is a scatter plot of 567 comparisons of TIBC by colorimetric assay and TRF by Behring nephelometric immunoassay, representing the combined results of five independent experiments in three laboratories. Three outliers were excluded. The correlation between TIBC measured colorimetrically and TRF measured immunologically over a wide range of values is high: r2 = 0.941.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 1. Scatter plot comparing TRF measured immunologically on the Behring nephelometer and TIBC measured colorimetrically on either Hitachi or Olympus high-volume chemistry analyzers.

n = 567 (three outliers excluded). TIBC µmol/L = 6.90 + 19.41 x TRF g/L. r2 = 0.941.

We estimated the contributions of the variances of the TIBC and TRF assays to the total variance of the scatter plot as follows:

The SE of predicted y from x = Sy|x = 3.57; the mean for y = 57.8 µmol/L







Our estimate for the residual variance is essentially zero. This supports our hypothesis that the observed scatter is secondary to imprecision inherent in the assay and not to any biological variability in the binding of iron to TRF.

Figure 2 is a scatter plot of the results of 72 comparisons of TIBC by colorimetric assay and TRF by Beckman nephelometric immunoassay. The correlation is improved when TRF is assayed with the more precise Beckman assay: r2 = 0.968.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 2. Scatter plot comparing TRF measured immunologically on the Beckman nephelometer and TIBC measured colorimetrically on the Olympus high-volume chemistry analyzer.

n = 72. TIBC µmol/L = 3.45 + 20.06 x TRF g/L. r2 = 0.968.


   Discussion
Top
 Abstract
 Introduction
 Design of Experiments
 Results
 Discussion
References
 
Our results support the comparability of TIBC measured colorimetrically and TRF measured immunologically over a wide range of values. We found no evidence of any unusual patient- or disease-related effect on the bias of either the TIBC or the TRF assay. That does not mean that it might not happen, but if it does it is a rare event.

There is wide scatter in literature-reported ratios of TIBC to TRF, ranging from a low of 11.3 (5) to a high of 34.4 (4). Only two (1)(11) of six other studies (1)(3)(7)(11)(12)(13) reported ratios close to 25.0 (i.e., 24.4 (1) and 23.3 (11)), and then only when results in the older study (1) were corrected by us to be equivalent to those obtained with the new IFCC TRF standard. Most significant in our analysis of these disparate ratios is the high proportion of observations that differ from ideal when radial immunodiffusion was used as the immunologic method for TRF. Manual radial immunodiffusion methods are not as precise as automated nephelometric methods (1).

Indeed, our results and our review of published articles suggest that any reported deviation from the theoretically expected ratio of 25.0 between TIBC and TRF indicates a problem with either or both assays. Moreover, unless precise and unbiased TIBC and TRF methods are used, identifying a genuine occurrence of "deviant" TRF or TIBC will be impossible.

The results we obtained for the CDC pools with our colorimetric UIBC assay on Hitachi and Olympus equipment suggest a manufacturer-related error in the calibration of the UIBC channel, but not of the serum iron channel.

We studied the effect of changes in ferritin, percent saturation of TRF, UIBC, and TIBC on the ratio of TIBC measured colorimetrically to TRF measured on the more precise Beckman nephelometer and found that there is about a 10% difference in the ratio between the lowest and highest values of the variables studied. Table 4 summarizes our findings. The highest ratios were found when ferritin was high, when percent saturation was high, when UIBC was low, and when TIBC was low. There is some selection bias in the ferritin data because only 52 of the 72 samples had a ferritin test, and 27 of the 52 ferritin concentrations were >1000 µg/L. There is a common denominator, however, among all of these variables: the number and availability of binding sites for iron on TRF. When ferritin concentrations are increased, the percent saturation of TRF is usually also high and available binding sites are decreased. When the number of unsaturated binding sites decreases, less time is required for the reaction to approach completion, which results in reporting a higher UIBC concentration, a higher TIBC concentration, and a higher ratio of TIBC to TRF. Conversely, when the number of unsaturated binding sites increases, more time is required for the reaction to approach completion, which is the case when percent saturation and ferritin concentrations are low, and when UIBC and TIBC concentrations are high.


View this table:
[in this window]
[in a new window]
  		Table 4. Effect of changes in serum ferritin, UIBC, TIBC, and percent saturation on the ratio of TIBC to TRF when chemistry analyzer is calibrated routinely and TRF is measured with a Beckman nephelometer.

Our results suggest that TRF-containing protein-based calibrators should be used for the primary calibration of Hitachi and Olympus UIBC. A preliminary study of calibration methods at our Teterboro laboratory confirms the superiority of two-point calibration with human serum. Two-point calibration of the UIBC channel with the low and high reference pools from CDC, rather than with an aqueous zero and a single-point commercially available bovine calibrator, yielded superior results. The mean ratio of TIBC to TRF was 24.6 when we assayed 29 different patient-derived minipools covering a wide range of TIBC concentrations. The spread of values for the ratio was very tight, ranging from a low of 23.5 to a high of 26.3, and the correlation of TIBC to TRF was high: r2 = 0.992. Moreover, two-point calibration of the UIBC channel with human serum eliminated the effect of UIBC concentrations on the ratio.

On the basis of these results, Quest Diagnostics is restandardizing all of its UIBC/TIBC methods to be in agreement with TRF concentrations measured immunologically. In addition, we are working with suppliers of reagents and calibrators to assist them in the development of unbiased secondary nonhuman calibrators for UIBC that are traceable to human-based primary calibrators. The UIBC/TIBC concentrations of the primary calibrators are defined by chemical reference methods and by TRF assays standardized to the IFCC's preparation for plasma proteins.

In summary, TIBC and TRF are equivalent when each is measured precisely and without bias.


   Acknowledgments
 
We gratefully acknowledge the assistance of Frank Calascibetta, Kathy Erickson, Carl Garber, Candis Harper, Elaine Labrecque, and Sheila McCreary.


   Footnotes
 
1 Nonstandard abbreviations: TRF, transferrin; TIBC, total iron-binding capacity; and UIBC, unsaturated iron-binding capacity.


   References
Top
 Abstract
 Introduction
 Design of Experiments
 Results
 Discussion
 References
 

  1. Vernet-Nyseen M, Pari M, Benoit M-O, Plomteux G. Are usual immunochemical methods for the determination of human serum transferrin influenced by the iron-saturation of the protein?. Clin Chim Acta 1984;143:355-360. [ISI][Medline] [Order article via Infotrieve]
  2. Vernet M. Immunochemical assay of transferrin and iron saturation in serum [Letter]. Clin Chem 1993;39:2352-2353. [ISI][Medline] [Order article via Infotrieve]
  3. Tsung SH, Rosenthal WA, Milewski KA. Immunological measurement of transferrin compared with chemical measurement of total iron-binding capacity. Clin Chem 1975;21:1063-1068. [Abstract]
  4. Rajamaki A, Irjala K, Aitio A. Immunochemical determination of serum transferrin. Reference values, correlation with serum total iron-binding capacity and value in diagnosis of iron deficiency anaemia and anaemia of chronic disorders. Scand J Haematol 1979;23:227-231. [ISI][Medline] [Order article via Infotrieve]
  5. van der Heul C, van Eijk HG, Wiltink WF, Leijnse B. The binding of iron to transferrin and to other serum components at different degrees of saturation with iron. Clin Chim Acta 1972;38:347-353. [ISI][Medline] [Order article via Infotrieve]
  6. Lentjes GWQM, Lindeman JHN, van de Bent W, Berger HM. Measured versus calculated latent iron binding capacity in plasma of newborns. Ann Clin Biochem 1995;32:478-481.
  7. Beilby J, Olynyk J, Ching S, Prins A, Swanson N, Reed W, et al. Transferrin index: an alternative method for calculating the iron saturation of transferrin. Clin Chem 1992;38:2078-2081. [Abstract]
  8. Gunter EW, Lewis BL, Koncikowski SM. Laboratory methods used for the Third National Health and Nutrition Examination Survey (NHANES III), 1988–1994. Hyattsville, MD: Centers for Disease Control and Prevention, 1996:VII-B-2–10..
  9. Giovanniello TJ, Bendetto G, Palmer DW, Peters T. Fully and semiautomated methods for the determination of serum iron and total iron-binding capacity. J Lab Clin Med 1968;71:874-883. [ISI][Medline] [Order article via Infotrieve]
  10. Ramsey WNM. The determination of the total iron-binding capacity of serum. Clin Chim Acta 1957;2:221-226.
  11. Gabbe EE, Heinrich HC, Icagic F. Proposal for the standardization of the serum unsaturated iron binding capacity assay, and results in groups of subjects with normal iron stores and with prelatent, latent, and manifest iron deficiency. Clin Chim Acta 1982;119:51-63. [ISI][Medline] [Order article via Infotrieve]
  12. Finlayson J, Fraser CG. Short-time changes in iron, ferritin, total iron-binding capacity, and transferrin in serum after myocardial infarction [Letter]. Clin Chem 1985;31:782-783. [ISI][Medline] [Order article via Infotrieve]
  13. Huebers HA, Eng MJ, Josephson BM, Ekpoom N, Rettmer RL, Labbé RF, et al. Plasma iron and transferrin iron-binding capacity evaluated by colorimetric and immunoprecipitation methods. Clin Chem 1987;33:273-277. [Abstract/Free Full Text]



The following articles in journals at HighWire Press have cited this article:


Home page
 Hum Exp ToxicolHome page
A M Emara, R S El Kelany, and K A Moustafa
Comparative study of the protective effect between deferoxamine and deferiprone on chronic iron overload induced cardiotoxicity in rats
Human and Experimental Toxicology, July 1, 2006; 25(7): 375 - 385.
[Abstract] [PDF]

Home page
 Clin. Chem.Home page
A. Van Campenhout, C. Van Campenhout, A. R. Lagrou, and B. Manuel-y-Keenoy
Effects of in Vitro Glycation on Fe3+ Binding and Fe3+ Isoforms of Transferrin
Clin. Chem., September 1, 2004; 50(9): 1640 - 1649.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
H. Yamanishi, S. Iyama, Y. Yamaguchi, Y. Kanakura, and Y. Iwatani
Total Iron-binding Capacity Calculated from Serum Transferrin Concentration or Serum Iron Concentration and Unsaturated Iron-binding Capacity
Clin. Chem., January 1, 2003; 49(1): 175 - 178.
[Full Text] [PDF]

Home page
 Clin. Chem.Home page
H. Yamanishi, S. Iyama, Y. Yamaguchi, Y. Kanakura, and Y. Iwatani
Modification of Fully Automated Total Iron-binding Capacity (TIBC) Assay in Serum and Comparison with Dimension TIBC Method
Clin. Chem., September 1, 2002; 48(9): 1565 - 1570.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Chem.Home page
G. Siek, J. Lawlor, D. Pelczar, M. Sane, and J. Musto
Direct Serum Total Iron-binding Capacity Assay Suitable for Automated Analyzers
Clin. Chem., January 1, 2002; 48(1): 161 - 166.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
L. T. Goodnough, B. Skikne, and C. Brugnara
Erythropoietin, iron, and erythropoiesis
Blood, August 1, 2000; 96(3): 823 - 833.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (27)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gambino, R.
Right arrow Articles by Gunter, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gambino, R.
Right arrow Articles by Gunter, E.
Related Collections
Right arrow Laboratory Management
Right arrow General Clinical Chemistry
Right arrow Proteomics and Protein Markers
Right arrow Hematology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS