Clinical Chemistry Siemens Point of Care - Urinalysis
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 45: 2191-2199, 1999;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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 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 (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Looker, A. C.
Right arrow Articles by Gordeuk, V. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Looker, A. C.
Right arrow Articles by Gordeuk, V. R.
Related Collections
Right arrow General Clinical Chemistry
Right arrow Hematology
(Clinical Chemistry. 1999;45:2191-2199.)
© 1999 American Association for Clinical Chemistry, Inc.

Articles

Increased Serum Transferrin Saturation Is Associated with Lower Serum Transferrin Receptor Concentration

Anne C. Looker1,a, Mark Loyevsky2 and Victor R. Gordeuk2

1 National Center for Health Statistics, Centers for Disease Control and Prevention, Hyattsville, MD 20782.

2 Department of Medicine, The George Washington University Medical Center, Washington, DC 20037.
a Address correspondence to this author at: Room 900, National Center for Health Statistics, 6525 Belcrest Rd., Hyattsville, MD 20782. Fax 301-436-3436.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Serum transferrin receptor (sTfR) concentrations are increased in iron deficiency. We wished to examine whether they are decreased in the presence of potential iron-loading conditions, as reflected by increased transferrin saturation (TS) on a single occasion.

Methods: We compared sTfR concentrations between 570 controls with normal iron status and 189 cases with increased serum TS on a single occasion; these latter individuals may be potential cases of iron overload. Cases and controls were selected from adults who had been examined in the third National Health and Nutrition Examination Survey (1988–1994) and for whom excess sera were available to perform sTfR measurements after the survey’s completion. Increased TS was defined as >60% for men and >55% for women; normal iron status was defined as having no evidence of iron deficiency, iron overload, or inflammation indicated by serum ferritin, TS, erythrocyte protoporphyrin, and C-reactive protein.

Results: Mean sTfR and mean log sTfR:ferritin were ~10% and 24% lower, respectively, in cases than in controls (P <0.002). Cases were significantly more likely to have an sTfR value <2.9 mg/L, the lower limit of the reference interval, than were controls (odds ratio = 1.8; 95% confidence interval, 1.04–2.37).

Conclusion: Our results support previous studies that suggested that sTfR may be useful for assessing high iron status in populations.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The serum transferrin receptor (sTfR)1 concentration reflects the number of tissue cell transferrin receptors (TfRs), which in turn reflect tissue needs for iron. Developing erythrocytes in the bone marrow have the greatest need for iron, and therefore sTfR concentrations tend to correlate with the amount of these erythroid precursors (1). As such, the sTfR concentration could be expected to vary according to both degree of erythropoiesis and amount of body iron stores (2). The concentration of TfRs is increased in the presence of iron deficiency anemia (3)(4)(5)(6), but studies have differed in regard to whether sTfR concentrations are decreased in those with iron-loading conditions. Two studies (7)(8) have found sTfR concentrations within the reference interval in patients with hereditary hemochromatosis; one of these studies also found concentrations within the reference interval among African blacks with dietary iron overload (8). Results from the first study are somewhat difficult to interpret, given that the hemochromatosis patients were under treatment to reduce iron stores when the receptor values were measured (7). Three other studies reported low sTfR values in individuals with high iron status or hemochromatosis (9)(10)(11), and an additional study reported a significant negative correlation between receptor concentrations and ferritin in thalassemia patients who had been cured via bone marrow transplantation (12).

The present study was designed to assess whether sTfR concentrations are lower in individuals with increased serum transferrin saturation (TS), based on a single measurement, than in individuals who have normal iron status, using a sample derived from the third National Health and Nutrition Examination Survey (NHANES III, 1988–1994). NHANES III was a large, population-based health survey (13) which included a component to assess iron status (14). However, this component did not include sTfR measurements. We therefore initiated a special study after the main survey was completed to analyze sTfR in excess sera available from NHANES III for a subsample of 759 men and women who had increased TS on a single occasion or normal iron status. Individuals with increased TS values were selected because they may potentially have an iron-loading condition such as hereditary hemochromatosis, African iron overload, or one of the various forms of secondary iron overload.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
subjects
NHANES III is the most recent of a series of surveys conducted periodically by the National Center for Health Statistics (NCHS), CDC, to assess the health and nutrition of a large representative sample of the noninstitutionalized, civilian US population. Data were collected via household interviews and direct standardized physical examinations conducted in specially equipped mobile examination centers (13). NHANES III was designed to provide reliable estimates for three racial or ethnic groups: non-Hispanic whites, non-Hispanic blacks, and Mexican Americans. Race and ethnicity were self-reported by the participants. All procedures were approved by the NCHS Institutional Review Board, and written informed consent was obtained from all subjects.

Blood was collected as part of the physical examination conducted in the mobile examination centers or in the homes of participants who were unable to travel to the mobile centers. At the time of blood collection, respondents were asked to report the last time they consumed any food or beverage other than plain water. Responses to this question were used in the present study to calculate actual fasting time in hours.

A large battery of biochemical measurements was performed on blood collected from NHANES III participants. To allow for out-of-range results that would need to be repeated, more specimen volume was sent for measurement than was usually needed for the various assays performed in the main survey. Because most individuals did not have out-of-range results, excess sera was available for many adult respondents in NHANES III. A surplus serum bank was established to store these excess specimens at -70 °C to allow for their use in special ad hoc projects. Proposals to obtain these excess sera samples must be approved for scientific merit by a review committee convened by NCHS and by the NCHS Institutional Review Board for human subjects concerns. All excess sera specimens have been through at least two freeze-thaw cycles.

excess sera selection criteria
Excess sera were requested for two groups of adults >=20 years of age in the present study: individuals with a single increased serum TS, and individuals with normal iron status. The following criteria were used to select the samples:

High TS.
High TS was defined as >60% for men and >55% for women. These particular cutoffs were chosen because they had been used in a previous study of high TS values conducted by the CDC (15). Because serum ferritin (SF) values can be increased in inflammatory conditions (16) as well as in individuals with high body iron stores, we did not use SF as a criterion to select cases. In addition, a SF concentration within the reference interval does not necessarily preclude the possibility of hereditary hemochromatosis; it is also possible that affected individuals will not have increased SF concentrations because they have not yet developed substantial hepatic iron overload (17).

Normal iron status.
Normal iron status was defined as having no evidence of iron deficiency, iron overload, or inflammation. To exclude those with iron deficiency, we excluded individuals with abnormal values for two or more of the following: SF, TS, or free erythrocyte protoporphyrin (FEP). Cutoff values indicative of iron deficiency for these variables have been published previously (18). We used a multivariable approach to define iron deficiency rather than excluding individuals with a single abnormal value because of the well-known overlap between normal and abnormal values in iron status indicators when assessing iron deficiency (19). We also excluded individuals with high values for either TS (as defined previously) or SF (>400 µg/L for men, >200 µg/L for women 20–49 years of age, and >300 µg/L for women >=50 years of age) (16). Finally, individuals with C-reactive protein values >10 mg/L were deleted to remove individuals with possible inflammation (19).

We selected all individuals in the excess sera pool who had increased TS on a single occasion to serve as cases. Three controls were selected at random from among those individuals in the excess sera pool who had normal iron status and who matched cases on single year of age, sex, and race or ethnicity. There were 26 adults in NHANES III who had high TS values on a single occasion but did not have excess sera. When compared with the actual high TS cases in our study, these 26 potential cases differed significantly only on 4 of the 14 selected demographic and biochemical characteristics considered in the present study: age, hematocrit, alanine aminotransferase (ALT), and aspartate aminotransferase (AST). Potential cases without excess sera were older and had higher values of the three serum analytes than did cases with excess sera. Further analyses of the 26 potential cases without excess sera revealed that the significant difference in ALT and AST was attributable entirely to one individual with AST and ALT values that far exceeded those of the other potential cases (e.g., were outliers); when this potential case was deleted, ALT and AST values no longer differed between cases with and without excess sera. There were 793 potential normal iron status controls who matched cases on age, sex, and race or ethnicity but did not have excess sera. These potential controls differed from the potential controls who did have excess sera for only 2 of the 11 biochemical variables included in this study: hematocrit (higher in potential controls without excess sera) and lactate dehydrogenase (lower in potential controls without excess sera). Furthermore, for each high TS case, there was an mean of 24 potential controls with excess sera (range, 2–55) available to be randomly selected as one of the three matched controls; thus it is unlikely that lack of excess sera introduced serious bias into our random selection of matched controls.

laboratory analyses
In the present study, we kept the excess sera samples frozen at -70 °C after their receipt from the storage bank until assayed for sTfR. We did not perform sTfR assays on hemolyzed samples. We measured sTfR using the Ramco TfR test kit (Ramco Laboratories). This assay is an enzyme immunoassay based on the double antibody sandwich method. We diluted plasma or serum samples in buffer and pipetted them into microwells precoated with polyclonal antibody to TfR. We added horseradish peroxidase-conjugated murine monoclonal antibody specific for TfR to the wells and incubated for 2 h at room temperature. We removed any unbound TfR and excess horseradish peroxidase conjugate from the wells by washing. We then added enzyme substrate to the wells. After the addition of an acid "stop" solution, we measured the product of the enzymatic reaction in a microplate reader (Multiskan MS) at 450 nm. The absorbance of the resulting solution is directly proportional to the concentration of the TfR in the unknown samples and in the calibrators. We generated a calibration curve by plotting the absorbance values vs the concentration of the TfR calibrators provided in the kit. We determined the concentration of the TfR in the sample by comparing the sample’s absorbance reading with the calibration curve graph with the aid of Genesis Lite software (Life Sciences International).

The other biochemical measures used in the present study were assayed as part of the main survey. Because details of the assay methods for these measures have been published elsewhere (20), they will only be summarized briefly here. Hemoglobin, hematocrit, and red cell distribution width were measured in the mobile examination centers, using a Coulter S-Plus Jr electronic counter (Coulter Electronics) (14)(20). Assays for the remaining iron status indicators used in the present study were performed by the Division of Environmental Health Laboratory Sciences, National Center for Environmental Health, CDC, Atlanta GA. TS was calculated by dividing serum iron by total iron binding capacity. Serum iron and total iron binding capacity were measured colorimetrically with an ALPKEM RFA analyzer (Alpkem), and 10 g/L thiourea was added to complex Cu2+ to prevent copper interference (20). Serum iron and total iron binding capacity were not measured on hemolyzed samples. FEP was measured via fluorescence extraction, and SF was measured with the Bio-Rad Quantimmume IRMA kit (Bio-Rad Laboratories).

Because both SF and sTfR data were available in the present study, it was also possible to examine the ratio of these two variables. In specific, we calculated the log of the sTfR:ferritin ratio by first converting sTfR from mg/L to µg/L and then calculating the log of the ratio of sTfR (µg/L):SF (µg/L) for each individual. Cook et al. (3) found that this ratio had a precise, linear relationship with body iron stores, and thus it has been proposed as a useful index of body iron.

C-reactive protein was measured by latex-enhanced nephelometry at the Immunology Division, University of Washington, Seattle WA (20). Serum analytes related to liver function (ALT, AST, lactate dehydrogenase, and total bilirubin) were measured with a Hitachi Model 737 multichannel Analyzer (Boehringer Mannheim Diagnostics) at the White Sands Research Center, Alamagordo NM (20).

statistics
We performed all analyses using SAS (21). Because the sample was not a random or representative subset of the larger NHANES III sample, we did not use sampling weights in the analyses; therefore, results cannot be considered representative of the US population. We used statistical methods to account for the use of individually matched cases and controls in all statistical comparisons. This is necessary to ensure that matching does not introduce bias into the exposure distribution of the controls (22). Specifically, to assess whether sTfR values differed by iron status group, we (a) calculated the mean of the differences in sTfR and log sTfR:ferritin ratio for each matched case-control set and tested whether this mean was zero; (b) computed the odds ratio of having a sTfR value below the lower limit of the reference interval as stated by the assay kit manufacturer (e.g., <2.9 mg/L) by iron status group using the Mantel-Haenzel procedure for matched cases and controls (23); and (c) used analysis of variance to compare means between iron status groups while adjusting for age, sex, and race or ethnicity (the matching characteristics). The last approach using analysis of variance is valid for individually matched data as long as the number of categories in the analysis is not unusually large relative to the number of cases (22).

We also performed secondary analyses to assess whether further subdivision of the normal and the increased TS groups was associated with a greater degree of discrimination between sTfR and log sTfR:ferritin ratio values. Because our definition of iron deficiency required abnormal values for two or more of three iron status indicators (SF, TS, and FEP), some individuals with low values for one of these three variables were included in the normal iron status group. These individuals were unlikely to have significant iron deficiency (19), but they may represent a "low normal" iron status state. In addition, we did not use high SF when selecting cases who may potentially have an iron-loading condition in our main analysis because high SF alone is not specific for increased iron stores (16) and because requiring both increased TS and high SF yielded fewer cases (n = 47). To assess whether use of more detailed iron status groups affected results, we divided the normal iron status group into two subgroups: (a) normal iron status, one abnormal value (e.g., those with a value indicating iron deficiency for one of the three variables used to define normal iron status); and (b) normal iron status, zero abnormal values. We used two approaches to subdivide the increased TS group: (a) those with high TS on a single occasion only vs those with both a single high TS and high SF; and (b) those with TS values ranging from 55% to 64.9% vs those with TS values >64.9%. We used the latter approach primarily to provide a second method to examine the relationship of log sTfR:ferritin ratio in the detailed iron status groups because values for log sTfR:ferritin ratio might be artificially reduced in the group that was formed on the basis of both high SF and high TS. The value of 64.9% was the median TS value in the increased TS group overall (range, 55–98.5%).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The distribution of the single high TS cases by age, sex, and race or ethnicity is shown in Table 1 . Most of the male cases were either young Mexican-American men (29%) or non-Hispanic white men of either age group (22% <50 years of age and 21% >=50 years of age). In contrast, almost one-half (45%) of the female cases were young non-Hispanic whites, with a smaller, and roughly equal, percentage (~16–19%) being older white women and younger black or Mexican-American women.


View this table:
[in this window]
[in a new window]
  		Table 1. Demographic characteristics of high TS cases among NHANES III excess sera sTfR samples (n = 189).

Selected characteristics of the cases with increased TS on a single occasion and controls with normal iron status are compared in Table 2 . Data on characteristics used for sample selection and matching are provided to confirm the successful implementation of the selection process. Differences in these characteristics between iron status groups cannot be validly tested for statistical significance because values for these variables were set by definition. The distribution by age, sex, and race or ethnicity in our sample reflects the pattern for these variables among those with a single high TS in the excess sera pool, i.e., most of those with a single high TS were <50 years of age, male, and non-Hispanic white. Because individual matching of these characteristics between cases and controls was used, this pattern must, by definition, occur among the normal iron status controls as well and does not necessarily reflect the actual pattern of these characteristics in the total population with normal iron status.


View this table:
[in this window]
[in a new window]
  		Table 2. Selected characteristics of iron status groups: NHANES III excess sera sTfR samples (n = 759).

Comparison of variables that were not used for matching or sample selection revealed that adjusted means for the hematology variables were similar in cases and controls, with the exception of hemoglobin (Table 2Up ). Of note are the significantly higher values of enzymes related to liver function (ALT, AST, and total bilirubin) and of alcohol intake among the cases. Cases also fasted longer than did controls.

The association between TfR or the ratio of receptor to SF and iron status group is shown in Table 3 . The cases had significantly lower mean receptor and log sTfR:ferritin ratio values, after adjusting for age, sex, and race or ethnicity, than did the normal iron status controls. Mean adjusted receptor and log sTfR:ferritin ratio values were ~10% and 24% lower in those with a single high TS than in the normal iron status controls, respectively. The cases were significantly more likely to have a sTfR value below the reference interval for sTfR stated by the kit manufacturer (e.g., <2.9 mg/L) than were the normal iron status controls (odds ratio = 1.8; 95% confidence interval, 1.04–2.37; Table 3 ).


View this table:
[in this window]
[in a new window]
  		Table 3. Association between sTfR concentration and iron status group.

The mean of the difference in sTfR between individually matched case-control pairs was -0.48 mg/L, which was significantly different from zero (P <0.0025). Differences in sTfR and log sTfR:ferritin ratio between cases and controls remained statistically significant after adjustment for liver enzymes, alcohol intake, or fasting time and were of the same magnitude seen for unadjusted results (0.43–0.71 mg/L and 0.59–0.75 for sTfR and log sTfR:ferritin mass ratio, respectively). Figs. 1 and 2 illustrate the difference in the sTfR or sTfR:ferritin ratio distributions between the cases with a high TS on a single occasion and normal iron status controls. For both measurements, values of the cases are shifted toward lower values. However, there is also significant overlap in values between cases and controls.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 1. Cumulative distribution of sTfR by iron status group.

(——–), normal iron status (controls); (—{blacksquare}—), high TS (cases).



View larger version (13K):
[in this window]
[in a new window]
  		Figure 2. Cumulative distribution of log sTfR:ferritin mass ratio by iron status group.

(——–), normal iron status (controls); (—{blacksquare}—), high TS (cases).

Results of the secondary analyses using more detailed iron status groups are shown in Table 4 . A gradation in sTfR values appears when the data are analyzed in this manner. The trend is most evident when log sTfR:ferritin ratio is used, with significantly higher ratios in the two normal iron status subgroups compared with the two high TS subgroups using either approach to subdivide the latter group. A similar trend is also evident, but to a lesser degree, when mean sTfR alone is examined by either approach: mean sTfR in at least one of the normal iron status subgroups was significantly higher than means in the two high iron status subgroups. Mean sTfR values in the two high TS groups did not differ, however. There was also a significant overall trend toward higher prevalences of sTfR <2.9 mg/L with increasing iron status; this prevalence differed by 11–16% (P <0.01 or <0.0006, respectively) between the highest and lowest iron status groups. It is important to note that results based on the four detailed iron status groups should be interpreted with some caution because further analyses revealed significant differences between cases and controls in age and sex when these more detailed groups were used instead of the two broader iron status groups. We have statistically adjusted for these differences in Table 4 ; this adjustment can likely reduce, but cannot completely rule out, possible biases attributable to lack of matching.


View this table:
[in this window]
[in a new window]
  		Table 4. Association between sTfR concentration and detailed iron status group.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
sTfR and log sTfR:ferritin ratio values were reduced in subjects with a single increased TS in the present study. Subjects with high TS values on one occasion had sTfR values that were ~10% lower than subjects with normal iron status, and they were 1.8 times more likely to have a sTfR value below the reference interval. These differences remained after adjustment for differences in liver enzyme values, alcohol intake, or fasting time between cases and controls. Results were also similar when a more stringent case definition was used, e.g., high TS and SF. A tendency toward successively lower mean values in subjects with potentially higher iron status relative to normal iron status appeared when results were analyzed by more detailed iron status groups, especially for the sTfR:ferritin ratio. Although subjects with iron-overloading conditions could not be conclusively identified in our study, our results are consistent with other studies (9)(10)(11)(12) that have suggested that sTfR may be useful for assessing high iron status, at least in populations.

Circulating TfRs are derived by proteolytic cleavage from TfRs expressed on the cell surface (24). Intracellular iron influences the posttranscriptional regulation of expression of ferritin, the TfR, and several other genes important in mammalian iron metabolism (25)(26). Regulation of expression occurs by means of an interaction between the iron regulatory protein, a molecule that senses changes in the chelatable intracellular iron pool (27)(28)(29), and iron-responsive elements located on untranslated regions of ferritin and TfR mRNAs (30)(31). When intracellular iron is ample, iron regulatory protein has aconitase activity and does not bind to the iron-responsive elements, which increases ferritin mRNA translation and TfR mRNA degradation. Conversely, in iron deprivation, iron regulatory protein loses aconitase activity and binds to iron-responsive elements, causing a repression in ferritin mRNA translation and increased TfR mRNA stability (32). Thus, at the cellular level, TfR expression is decreased when iron supply is ample and increased when iron supplies are reduced (28)(33)(34). In keeping with these observations in vitro, patients with secondary iron overload and with hereditary hemochromatosis have decreased TfR expression in hepatocytes and other cells (35)(36)(37). It is possible that in individuals with an iron-loading condition, the decrement in cellular TfRs in the liver might be large enough to be reflected in reduced values of sTfR.

The combination of sTfR and ferritin has been suggested for evaluating iron status in the range between normally replete stores and iron deficiency (3)(5). Specifically, Cook et al. (3) have proposed that ferritin can provide information regarding the amount of iron stores in replete individuals, whereas sTfR provides information regarding iron-deficient erythropoiesis after iron stores are depleted but before other significant alterations in iron metabolism would be detectable. Our results suggest sTfR and the sTfR:ferritin ratio may also reflect high iron status, as defined by a single increased TS measurement, relative to normal iron status. In addition, the results from the secondary analyses using the detailed iron status groups suggest that the combination of sTfR and ferritin may be superior to sTfR alone for evaluating potentially high iron status because a gradient toward lower means with increasing iron status was more apparent for the ratio than for sTfR alone. This could be because sTfR and SF are affected by different confounding variables: TfR may decrease with iron overload but also with suppressed erythropoiesis, whereas SF may increase with iron overload but also with inflammation and hepatocellular dysfunction. It is conceivable that the ratio of TfR to ferritin is a good way to identify those subjects who have alterations in these measurements that are indeed attributable to iron overload vs other processes.

The majority of the cases in our sample were under 50 years of age, which may explain why most did not have an increased SF concentration, e.g., these individuals could be possible cases of hemochromatosis who have not yet developed significant iron overload (17). The majority (~52%) were non-Hispanic whites, in whom we would expect a comparable prevalence of homozygotes for hereditary hemochromatosis as observed in other studies of whites in the US (38). Approximately 18% of the cases were non-Hispanic blacks—classic hereditary hemochromatosis is believed less common in African Americans than in non-Hispanic whites in the US (39), but a form of genetically linked primary iron overload that is distinct from hereditary hemochromatosis may exist in Africans and African Americans (40)(41). Little is known about the prevalence of hereditary hemochromatosis or other iron-loading conditions in Mexican Americans. It previously had been thought uncommon in individuals from Mexico: the frequency of HLA-A3, which is associated with a greater population frequency of known hemochromatosis, is estimated as 6% in Mexicans compared with 18% for African Americans and 24–26% in European and Australian Caucasians (39)(42). Estimates of genotypic frequencies for mutations in HFE in Mexicans suggest an allele frequency of ~7% for H63D (compared with frequencies of ~13% for the total European population) and 0% for C282Y (compared with frequencies of 3.8% for the total European population) (43). However, this estimate is based on an analysis of chromosomes from only 54 Mexican individuals. Other data support the possibility that hemochromatosis does occur in Mexican Americans, although the exact frequency is not clear. Data on the contribution of ancestral Spanish genes and the frequency of hemochromatosis mutations in the Spanish population provide one line of support. The contribution of putative ancestral populations to the current gene pool of Mexican Americans has been estimated as ~60–68% Spanish, 29–31% Native American, and 3–8% West African (44)(45). Estimates of the frequency for the H36D and C282Y gene mutations in contemporary Spanish populations appear comparable to the frequencies of the total European population (44)(46)(47). In addition, results from a screening program implemented in a health maintenance organization with a large Hispanic population in California found that the prevalence of iron overload in the Hispanics was 5 per 1000 persons, which was comparable to the rate seen in whites (15).

The overlap in the sTfR and sTfR:ferritin ratio distributions observed between the normal iron status controls and the cases with a single increased TS values suggests that the ability of these measurements to diagnose high iron status in individuals may be low. However, the general health survey from which our sample was derived was not designed to conclusively identify individuals with iron overload or hemochromatosis; therefore, our study was not optimally designed to assess the clinical utility of sTfR in individual patients.

Our study was better able to address whether there is a possible role for sTfR and the sTfR:ferritin ratio in assessing iron overload in large epidemiological studies, where definitive measures of iron overload or hemochromatosis, such as liver biopsy or repeated TS measurements, are not feasible. Our results suggest that sTfR or the sTfR:ferritin ratio may have some utility in this regard because sTfR and sTfR:ferritin values tended to be lower in the cases with possible high iron status than in controls with normal iron status. However, the observed overlap in sTfR or sTfR:ferritin ratio distributions also reduces their utility in identifying high iron status in large studies if they are used as single tests. A similar situation exists for many iron status indicators that are used to assess iron deficiency, e.g., considerable overlap in distributions has been observed between groups with normal iron status and those with iron deficiency (48). To address this problem, an expert group recommended the use of multivariable models to define iron deficiency in large epidemiological studies where more definitive tests of iron deficiency, such as bone marrow staining or response to iron therapy, are not possible (19). A similar multivariable approach to assess high iron status in epidemiological studies may be worth considering, given the impractical nature of the definitive tests and the lack of specificity of tests such as high SF or a single high TS. Our results suggest sTfR or the sTfR:ferritin ratio may be candidates for such a model, but additional studies that are better able to conclusively identify high iron status are needed to further evaluate this possibility.

Our study had several limitations. Because NHANES III participants only provided blood on a single occasion, individuals who may potentially have an iron-loading condition had to be defined on the basis of a single TS measurement. TS values exhibit considerable within-person variability (39); therefore, many individuals with an increased TS on one occasion may have a normal value upon retesting (38)(49). Current screening algorithms for hemochromatosis require increased TS values on two occasions to suspect presence of the disease (39). Increased TS values could also reflect liver disease or alcohol abuse. Other factors, such as fasting time, may affect TS values as well. However, we observed similar results in the present study regardless of case definition (e.g., a single increased TS only vs concurrently increased TS and SF). The difference in sTfR and log sTfR:ferritin ratio also remained significant after statistically controlling for differences in liver enzymes, alcohol intake, or fasting time between cases and controls. These analyses suggest, but cannot definitively establish, that the presence of false positives for high TS does not invalidate our findings. A final potential limitation relates to use of excess sera specimens that had been thawed and refrozen at least twice. We are unaware of any published data regarding the effect of freezing and thawing on sTfR measures. Any effect would be expected to impact samples from cases and controls equally, so that comparisons between these groups would still be valid.

In summary, interest in assessing high iron status in populations has grown in recent years, both in terms of detecting hemochromatosis (39) and evaluating whether high iron stores in otherwise healthy individuals are associated with increased risk of chronic diseases (50)(51)(52). Our results suggest that additional studies are warranted to further explore the potential role of sTfR in this assessment.


   Footnotes
 
1 Nonstandard abbreviations: STfR, serum transferrin receptor; TS, transferrin saturation; NHANES III, third National Health and Nutrition Examination Survey; NCHS, National Center for Health Statistics; SF, serum ferritin; FEP, free erythrocyte protoporphyrin; ALT, alanine aminotransferase; and AST, aspartate aminotransferase.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Beguin Y, Clemons G, Pootrakul P, Fillet G. Quantitative assessment of erythropoiesis and functional classification of anemia based on measurements of transferrin receptor and erythropoietin. Blood 1993;81:1067-1076. [Abstract/Free Full Text]
  2. Baynes RD. Refining the assessment of body iron stores. Am J Clin Nutr 1996;64:793-794. [Free Full Text]
  3. Cook JD, Skikne B, Baynes R. The use of serum transferrin receptor for the assessment of iron status. Hallberg L Asp N-G eds. Iron nutrition in health and disease 1996:49-58 John Libbey & Co London. .
  4. Kohgo Y, Niitsu Y, Nishisato T, Kato J, Kondo H, Sasaki K, Urushizaki I. Quantitation and characterisation of serum transferrin receptor in patients with anemias and polycythemias. Jpn J Med 1988;27:64-70. [Medline] [Order article via Infotrieve]
  5. Skikne BS, Flowers CH, Cook JD. Serum transferrin receptors: quantitative measure of tissue iron deficiency. Blood 1990;75:1870-1876. [Abstract/Free Full Text]
  6. Flowers CH, Skikne BS, Covell AM, Cook JD. Clinical measurement of serum transferrin receptors. J Lab Clin Med 1989;114:368-377. [Web of Science][Medline] [Order article via Infotrieve]
  7. Huebers HA, Beguin Y, Pootrakul P, Einspahr D, Finch CA. Intact transferrin receptors in human plasma and their relation to erythropoiesis. Blood 1990;75:102-107. [Abstract/Free Full Text]
  8. Baynes RD, Cook JD. Bothwell T, Friedman BM, Meyer TE. Serum transferrin receptors in hereditary hemochromatosis and African siderosis. Am J Hematol 1994;45:288-292. [Web of Science][Medline] [Order article via Infotrieve]
  9. Thorstensen K, Romslo I. Measurement of serum transferrin receptors in screening for hemochromatosis [Letter]. Clin Chem 1992;38:1510.[Free Full Text]
  10. Ledue TB, Craig WY. Serum concentrations of transferrin receptor in hereditary hemochromatosis [Letter]. Clin Chem 1995;41:1053-1054. [Free Full Text]
  11. Khumalo H, Gomo ZAR, Moyo VM, Gordeuk VR, Saungweme T, Roualt T, Gangaidzo IT. Serum transferrin receptors are decreased in the presence of iron overload. Clin Chem 1998;44:40-44. [Abstract/Free Full Text]
  12. Centis F, Delfini C, Agostinelli F, Barbanti I, Annibali M, Lucarelli G. Correlation between soluble transferrin receptor and serum ferritin levels following bone marrow transplantation for thalassemia. Eur J Haematol 1995;54:329-333. [Web of Science][Medline] [Order article via Infotrieve]
  13. National Center for Health Statistics. Plan and operation of the third National Health and Nutrition Examination Survey, 1988–94. Vital Health Stat 1(32). DHHS Publication No. PHS 94-1308. Hyattsville, MD: NCHS, 1994..
  14. Looker AC, Gunter EW, Johnson CL. Methods to assess iron status in various NHANES surveys. Nutr Rev 1995;53:246-254. [Web of Science][Medline] [Order article via Infotrieve]
  15. CDC. Iron overload disorders among Hispanics–San Diego California, 1995. Morbid Mortal Wkly Rep 1996;45:991–3..
  16. Custer EM, Finch CA, Sobel RE, Zettner A. Population norms for serum ferritin. J Lab Clin Med 1995;126:88-94. [Web of Science][Medline] [Order article via Infotrieve]
  17. Edwards CQ, Kushner JP. Screening for hemochromatosis. N Engl J Med 1993;328:1616-1620. [Free Full Text]
  18. Looker AC, Dallman PR, Carroll MD, Gunter EW, Johnson CL. Prevalence of iron deficiency in the United States. JAMA 1997;227:973-976.
  19. Expert Scientific Working Group. Summary of a report on assessment of the iron nutritional status of the United States. Am J Clin Nutr 1985;42:1318–30..
  20. Gunter EW, Lewis BL, Koncikowski SM. Laboratory methods used for the third National Health and Nutrition Examination Survey (NHANES III), 1988–1994 [CD-ROM; Available from National Technical Information Service (NTIS), Springfield, VA] 1996 Centers for Disease Control and Prevention Hyattsville, MD. .
  21. SAS Institute, Inc. SAS user’s guide, Ver. 6, 4th ed. Cary, NC: SAS Institute, 1989:943pp..
  22. Rothman KJ. Modern epidemiology 1986:236-283 Little, Brown & Co. Boston, MA. .
  23. Kahn HA, Sempos CT. Statistical methods in epidemiology 1989:123-129 Oxford University Press New York. .
  24. Turkewitz AP, Amatruda JF, Borhani D, Harrison SC, Schwatz AL. A high yield purification of the human transferrin receptor and properties of its major extracellular fragment. J Biol Chem 1988;263:8318-8325. [Abstract/Free Full Text]
  25. Klausner RD, Rouault TA, Harford JB. Regulating the fate of mRNA: the control of cellular iron metabolism. Cell 1993;72:19-28. [Web of Science][Medline] [Order article via Infotrieve]
  26. Hentze MW, Kuhn LC. Molecular control of vertebrate iron metabolism: mRNA-based regulatory circuits operated by iron, nitric oxide, and oxidative stress. Proc Natl Acad Sci U S A 1996;93:8175-8182. [Abstract/Free Full Text]
  27. Haile DJ, Rouault TA, Tang CK, Chin J, Harford JB, Klausner RD. Reciprocal control of RNA binding and aconitase activity in the regulation of iron responsive element binding protein: role of iron sulfur cluster. Proc Natl Acad Sci U S A 1992;89:7536-7540. [Abstract/Free Full Text]
  28. Rouault T, Rao K, Harford J, Mattia E, Klausner RD. Hemin, chelatable iron, and the regulation of transferrin receptor biosynthesis. J Biol Chem 1985;260:14862-14866. [Abstract/Free Full Text]
  29. Rouault TA, Hentze MW, Caughman SW, Harford BJ, Klausner RD. Binding of a cytosolic protein to the iron responsive element of human ferritin messenger RNA. Science 1988;241:1207-1210. [Abstract/Free Full Text]
  30. Hentze MW, Caughman SW, Rouault TA, Barriocanal JG, Dancis A, Harford HB, Klausner RD. Identification of the iron responsive element for the transcriptional regulation of human ferritin mRNA. Science 1987;238:1570-1573. [Abstract/Free Full Text]
  31. Casey JL, Koeller DM, Ramin VC, Klausner RD, Harford JB. Iron regulation of transferrin receptor mRNA requires iron responsive elements and a rapid turnover of determinants in the 3' untranslated region of the mRNA. EMBO J 1989;8:3693-3699. [Web of Science][Medline] [Order article via Infotrieve]
  32. Haile DJ, Rouault TA, Harford JB, Kennedy MC, Blondin GA, Beinert H, Klausner RO. Cellular regulation of the iron responsive element binding protein: disassembly of the cubane iron-sulfur cluster results in high affinity RNA binding. Proc Natl Acad Sci U S A 1992;89:11735-11739. [Abstract/Free Full Text]
  33. Bridges KR, Cudkowicz A. Effect of iron chelators on transferrin receptors in K562 cells. J Biol Chem 1984;259:12970-12977. [Abstract/Free Full Text]
  34. Baynes RD, Shih YJ, Cook JD. Production of soluble transferrin receptor in K562 erythroleukaemia cells. Br J Haematol 1991;78:450-455. [Web of Science][Medline] [Order article via Infotrieve]
  35. Lombard M, Bomford A, Hynes M, Naoumov NV, Roberts S, Crowe J, Williams R. Regulation of hepatic transferrin receptor in hemochromatosis. Hepatology 1989;9:1-5. [Web of Science][Medline] [Order article via Infotrieve]
  36. De Vos R, Sciot R, van Eyken P, Desmet VJ. Immunoelectron microscopic localization of hepatic transferrin receptors in human liver with and without iron overload. Virchows Arch B Cell Pathol Incl Mol Pathol 1988;55:11-17. [Web of Science][Medline] [Order article via Infotrieve]
  37. Ward JH, Kushner JP, Ray FA, Kaplan J. Transferrin receptor function in hereditary hemachromatosis. J Lab Clin Med 1984;103:246-254. [Web of Science][Medline] [Order article via Infotrieve]
  38. Edwards CQ, Griffen LM, Goldgar D, Drummond C, Skolnick MH, Kushner JP. Prevalence of hemochromatosis among 11,065 presumably healthy blood donors. N Engl J Med 1988;318:1355-1362. [Abstract]
  39. Witte DL, Crosby WH, Edwards CQ, Fairbanks VF, Mitros FA. Practice guideline development task force of the College of American Pathologists. Hereditary hemochromatosis. Clin Chim Acta 1996;245:139-200. [Web of Science][Medline] [Order article via Infotrieve]
  40. Gordeuk VR, Mukiibi J, Hasstedt SJ, Samowitz W, Edwards CQ, West G, et al. Iron overload in Africa. Interaction between a gene and dietary iron content. N Engl J Med 1992;326:95-100. [Abstract]
  41. McNamara L, MacPhail AP, Gordeuk VR, Hasstedt SJ, Rouault T. Is there a link between African iron overload and the described mutations of the hereditary haemochromatosis gene?. Br J Haematol 1998;102:1176-1178. [Web of Science][Medline] [Order article via Infotrieve]
  42. Terasaki PI eds. Histocompatibility testing 1980:955 UCLA Tissue Typing Laboratory Los Angeles. .
  43. Merryweather-Clarke AT, Pointon JJ, Shearman JD, Robson KJ. Global prevalence of putative haemochromatosis mutations. J Med Genet 1997;34:275-278. [Abstract/Free Full Text]
  44. Hanis CL, Hewett-Emmett D, Bertin TK, Schull WJ. Origins of U. S. Hispanics. Implications for diabetes. Diabetes Care 1991;14(Suppl 3):618-627.
  45. Long JC, Williams RC, McAuley JE, Medis R, Partel R, Tregellas WM, et al. Genetic variation in Arizona Mexican Americans: estimation and interpretation of admixture proportions. Am J Phys Anthropol 1991;84:141-157. [Web of Science][Medline] [Order article via Infotrieve]
  46. Sanchez M, Bruguera M, Bosch J, Rodes J, Ballesta F, Oliva R. Prevalence of the Cys282Tyr and His63Asp HFE gene mutations in Spanish patients with hereditary hemochromatosis and in controls. J Hepatol 1998;29:725-728. [Web of Science][Medline] [Order article via Infotrieve]
  47. Baiget M, Barcelo MJ, Gimferrer E. Frequency of the HFE C282Y and H63D mutations in distinct ethnic groups living in Spain [Letter]. J Med Genet 1998;35:701.
  48. Cook JD, Finch CA, Smith NJ. Evaluation of the iron status of a population. Blood 1976;48:449-455. [Abstract/Free Full Text]
  49. Olsson KS, Eriksson K, Ritter B, Heedman PA. Screening for iron overload using transferrin saturation. Acta Med Scand 1984;215:105-112. [Web of Science][Medline] [Order article via Infotrieve]
  50. Sempos CT, Looker AC, Gillum RF. Iron and heart disease: the epidemiologic data. Nutr Rev 1996;54:73-84. [Web of Science][Medline] [Order article via Infotrieve]
  51. Stevens RG. Iron and the risk of cancer. Med Oncol Tumor Pharmacother 1990;7:177-181. [Web of Science][Medline] [Order article via Infotrieve]
  52. Tuomainen TP, Nyyssonen K, Salonen R, Tervahauta A, Korpela H, Lakka T, et al. Body iron stores are associated with serum insulin and blood glucose concentrations. Population study in 1,013 eastern Finnish men. Diabetes Care 1997;20:426-428. [Abstract]



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


Home page
 Clin. Chem.Home page
M. R. Langlois, M.-E. Martin, J. R. Boelaert, C. Beaumont, Y. E. Taes, M. L. De Buyzere, D. R. Bernard, H. M. Neels, and J. R. Delanghe
The Haptoglobin 2-2 Phenotype Affects Serum Markers of Iron Status in Healthy Males
Clin. Chem., October 1, 2000; 46(10): 1619 - 1625.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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 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 (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Looker, A. C.
Right arrow Articles by Gordeuk, V. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Looker, A. C.
Right arrow Articles by Gordeuk, V. R.
Related Collections
Right arrow General Clinical Chemistry
Right arrow Hematology

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS