Nutritional Methodology
Folate Deficiency Does Not Alter the Usefulness of the Serum Transferrin Receptor Concentration as an Index for the Detection of Iron Deficiency in Mexican Women during Early Lactation1,2 Deborah L. O’Connor,3 Marie E. Latulippe,* Cristina Campos,† Claide Merlos,† Salvador Villalpando,** and Mary Frances Picciano* ** Department of Nutritional Sciences, University of Toronto and The Hospital for Sick Children, Toronto, Canada M5G 1X8; *Nutrition Department, Pennsylvania State University, University Park, PA 16802; †Instituto Nacional de Salud Publica, Cuernavaca, Mexico; and **Office of Dietary Supplements, National Institutes of Health, Bethesda, MD 20892-7517 ABSTRACT The purpose of this study was to investigate the usefulness of soluble serum transferrin receptor (TfR) concentration, TfR index and log TfR:ferritin ratio (TfR outcomes) in detecting tissue Fe deficiency among breast-feeding Otomi women residing in Capulhuac, Mexico (2800 m above sea level) and to determine whether folate deficiency modifies the interpretation of these data. Lactating women (n ⫽ 68) provided blood samples at 22 ⫾ 13 d (mean ⫾ SD) postpartum. Using the 3-index Fe assessment model with and without Hb, 2 women (3%) had Fe-deficient erythropoiesis, 24 (36%) Fe deficiency anemia, and 19 (29%) indeterminate Fe status; 29 (43%) and 5 (7.5%) women had plasma and erythrocyte folate concentrations below normative cutoff values, respectively. Mean values for TfR outcomes were higher among women classified as Fe deficient than those who were Fe sufficient, but did not differ with low or normal blood folate concentrations. Similarly, TfR outcomes did not differ among women with normocytic or macrocytic erythrocytes. Receiver-operating characteristic (ROC) curves generated for TfR outcomes yielded areas under the curve from 0.62 to 0.68, indicating that each of these measures, on its own, is a poor predictor of tissue Fe deficiency in lactating women. In conclusion, low blood folate concentrations or the presence of macrocytosis in Otomi women from Capulhuac, Mexico (moderate altitude) did not influence the utility of TfR outcomes for the detection of Fe deficiency during early lactation. Further, on their own, TfR, TfR index, and TfR:ferritin ratio were poor predictors of tissue Fe deficiency for any given individual. J. Nutr. 135: 144 –149, 2005. KEY WORDS:
●
iron
●
folate
●
lactation
●
transferrin receptor
Although prevalence figures for iron (Fe) deficiency with or without anemia vary widely, it is clear that a disproportionate number of reproducing women globally show evidence of Fe deficiency compared with nonpregnant, nonlactating women (⬃50 – 60% vs. 20 –30%) (1–3). The elevated Fe requirement and high incidence of Fe deficiency anemia during pregnancy, particularly in the third trimester, is well characterized (4,5). Specifically, rapid expansion of maternal blood volume and red cell mass during pregnancy and placental and fetal growth leave many women, even in developed countries, with low Fe stores and tissue Fe deficiency at parturition (6 –12). The consequences of poor maternal Fe status at parturition include decreased work capacity, decreased ability to care for children,
maternal morbidity and mortality, and poor maternal and infant outcomes in the event of a subsequent pregnancy (13). A battery of conventional tests [e.g., hemoglobin (Hb),4 mean cell volume (MCV), transferrin (Tf) saturation, ferritin] and a 3-index Fe assessment model can be used to assess an individual woman’s Fe status or predict the incidence of Fe deficiency among lactating women in a population (14). However, conventional tests of Fe status are frequently affected by underlying infection and inflammatory disease independent of Fe status; thus, interpretation of results can be difficult. The soluble serum transferrin receptor (TfR) assay is reported to be a reliable laboratory indicator of Fe status because it can distinguish Fe deficiency in individuals with inflammatory disease and exhibits small day-to-day variation (4,15,16). The concentration of TfR on the cell surface and in serum reflects erythropoietic activity, intracellular Fe requirement, and Fe status. That is, circulating concentrations of TfR increase
1 Presented in abstract form at Experimental Biology ’00, April, 2000, San Diego, CA [Latulippe, M. E., Rosas, M. G., Villalpando, S. & Picciano, M. F. (2000) Utility of TfR and TfR-ferritin index for assessment of iron deficiency is not complicated by folate deficiency in lactating women. FASEB J. 14: A508 (abs.)]. 2 Supported by the Consejo Nacional de Ciencia y Tecnologia of Mexico; the Natural Sciences and Engineering Research Council of Canada and Pennsylvania State University provided financial support. 3 To whom correspondence should be addressed. E-mail: deborah_l.o’
[email protected].
4 Abbreviations used: Hb, hemoglobin; Hct, hematocrit; ⫺Fe, iron deficient; ⫹Fe, iron sufficient; MCV, mean cell volume; ROC, receiver-operating characteristic; TfR, transferrin receptor.
0022-3166/05 $8.00 © 2005 American Society for Nutritional Sciences. Manuscript received 23 July 2004. Initial review completed 19 August 2004. Revision accepted 20 October 2004. 144
TRANSFERRIN RECEPTOR CONCENTRATION AND LACTATION
when erythropoiesis is stimulated by hemolysis or ineffective erythropoiesis. Similarly, in uncomplicated tissue Fe deficiency, the synthesis of TfR is upregulated. To date, however, only 2 published studies have assessed the utility of TfR for detecting maternal tissue Fe deficiency during lactation (6,17). In addition to Fe deficiency, reproducing women also are often at risk for folate deficiency. It is estimated that up to one third of reproducing women globally have some degree of folate undernutrition (18). Reproducing women in countries that have not adopted an elevated folic acid fortification program or who are not consuming folic acid– containing supplements develop a certain degree of folate depletion (e.g., increase in serum homocysteine) during pregnancy and lactation (19,20). Folate deficiency can cause ineffective erythropoiesis, resulting in an abnormal increase in immature RBC. For this reason, the possibility exists that folate deficiency may confound the interpretation of values for TfR. Although Fe status usually improves postpartum, folate is preferentially utilized by the mammary gland and secreted into milk at the expense of maternal folate reserves (19). Preferential folate partitioning to the mammary gland was first documented by Metz et al. (21) in a study of mothers with megaloblastic anemia and low milk folate concentrations. These investigators reported that folic acid supplementation increased milk folate concentration without changing either serum folate concentration or reticulocyte count (21). We were able to locate only 1 study that systematically investigated the effect of folate deficiency on the usefulness of TfR concentration for the detection of tissue Fe deficiency and none that examined this relation during lactation (22). The purpose of this study was to investigate the usefulness of serum TfR, TfR index (TfR/log ferritin) (23), and log TfR:ferritin ratio [log(TfR ⫻ 1000 /ferritin)] (24) to detect tissue Fe deficiency and whether these measures were affected by folate deficiency among Otomi women residing in Capulhuac, Mexico during early lactation. Previously, we reported a high prevalence of Fe and folate deficiency among lactating women in this community (7). SUBJECTS AND METHODS Subjects. Indigenous lactating Otomi women (n ⫽ 68) residing in the rural farming Indian community of San Mateo Capulhuac, Mexico, located 2800 m above sea level (moderate altitude), furnished blood samples as well as dietary and demographic information at 22 ⫾ 13 d (mean ⫾ SD) postpartum. Detailed description of this community and the dietary habits of its residents were published elsewhere (7,25). Briefly, Otomi women subsist on a maize-predominant diet; the amount of meat included in the diet is minor. Protein from animal sources (mostly eggs) comprises only 30% of all protein consumed. Dark leafy vegetables and legumes are frequently eaten after prolonged boiling; hence, the folate content of these usually good sources of folate is low (26). The calculated dietary intake of phytic acid is very high due to consumption of corn (65% of their energy intake) and other phytate-containing vegetables. In the majority of cases, exclusive breast-feeding extends for ⱖ4 mo. Women willing to breast-feed their infant exclusively were recruited from the local Medical Clinic and enrolled if the following eligibility requirements were met: 1) age: 17–37 y; 2) free of chronic diseases; 3) currently taking no medication; 4) not consuming alcoholic beverages; 5) free of pregnancy complications; 6) gave birth to a term infant of appropriate length and weight; and 7) willingness to comply with experimental protocol. The participation of human subjects was approved by the Institutional Review Boards of both the Instituto Mexicano del Seguro Social (IMSS), Mexico City, Mexico and the Pennsylvania State University, State College, PA. Anthropometric and blood sample collection. Maternal weight and height were determined in the clinic using standardized proce-
145
dures (27). Weights were determined using electronic scales with a precision of ⫾50 g (Tanita W8, Tanita). Heights were determined using a clinical stadiometer (Holtain). After an overnight fast, blood samples were collected into tubes containing EDTA or trace element–free tubes without an anticoagulant (Vacutainer, Becton-Dickinson). A portion (100 L) of whole blood was diluted in 10 volumes of 0.1 mol/L potassium phosphate buffer containing 0.05 mol/L Na-ascorbate. The remaining whole blood was separated by centrifugation (850 ⫻ g for 20 min at 5°C); sodium ascorbate (0.02 mol/L) was dissolved into aliquots of plasma and frozen at ⫺70°C for later determination of folate status. Serum samples for determination of Fe status were immediately frozen at ⫺20°C until analysis. Biochemical analyses. A complete blood count analysis, including measurement of hematocrit (Hct), Hb, and MCV, was performed using fresh blood samples and an electronic particle counter analyzer (ACT8 Coulter Counter, Beckman Coulter) The accuracy of the hemoglobin determinations was ⫾2 g/L, with an interassay CV of 3%. C-reactive protein was determined by immunoprecipitation (SANOFI, Pasteur Diagnostic) with an interassay CV of 4.9% at a level of 2 mg/L. Erythrocyte and plasma folate concentrations were determined by microbiological assay using cryoprotected Lactobacillus casei (ATCC #7469, American Type Tissue Culture Collection) as the test organism (28). Before the microbiological assay, whole blood lysates were incubated for 25 min at 37°C to convert folates to their microbiological assayable forms (e.g., short-chain folylpolyglutamates). Erythrocyte folate concentration was determined using the analyzed wholeblood folate concentration minus the plasma folate concentration corrected for Hct. The accuracy and reproducibility of these assays were assessed using infant formula with a certified value (1.29 ⫾ 0.28 mg folic acid/kg; Standard reference material #1846, National Institute of Standards and Technology). Analysis in our laboratory yielded a folate concentration of 1.33 ⫾ 0.05 mg folate/kg. The interassay CV was 5.4 and 12.3% for plasma and erythrocyte folate, respectively. Serum Fe (Fe) and total Fe binding capacity were determined by the colorimetric method of Fielding (29), modified for determination in microliter sample volumes. Fetal bovine serum with a certified value [32 ⫾ 2 mol/L (176 ⫾ 10 g/dL), Lot #7000C, Atlanta Biologicals] was used to verify the accuracy and reproducibility of this method. Analysis in our laboratory yielded a Fe concentration of 30 ⫾ 0.7 mol/L. Plasma ferritin was determined by RIA (Diagnostic Products) and verified by a human recombinant ferritin standard (19 ⫾ 1.5 g/L). Results from our laboratory yielded 20 ⫾ 1.1 g/L with an interassay CV of 3.6%. Serum transferrin receptor was determined by enzyme immunoassay (RAMCO Laboratories) using both high- and lowquality control serum samples. The interassay CV was 4.6% at a concentration of 13.5 g/L. Maternal Fe and folate status assessment models. We used a 3-index Fe assessment model and normative cutoff values for nonpregnant women described by Cook et al. (14), including ferritin (ⱕ12 g/L), transferrin saturation (ⱕ16%), and MCV (ⱕ80 fL). Using receiver-operating characteristic (ROC) curve analysis and blood samples from lactating women in this community, a sample specific Hb cutoff value for anemia of 133 g/L was derived as reported previously (30,31). This Hb cutoff value was used and is in agreement with the WHO (32) recommendation derived after adjusting for altitude (33–35). According to the 3-index model, subjects with ⱖ2 of these measures (ferritin, transferrin saturation, and MCV) positive for Fe deficiency were considered Fe deficient (⫺Fe). Based on this 3-index model and Hb levels, subjects were classified as follows: 1) Iron sufficient: ⫹Fe and adequate Hb; 2) Iron deficient erythropoiesis: ⫺Fe with or without adequate Hb; 3) Indeterminate iron status: one index of Fe status (ferritin, transferrin saturation, or MCV) indicative of Fe depletion with or without low Hb. The following criteria were applied in the assessment of folate status: 1) Normal blood folate individuals had erythrocyte folate ⬎ 360 nmol/L (27), plasma folate ⬎ 10 nmol/L (36), and MCV ⬍ 94 fL (14) with or without adequate Hb. This plasma folate cutoff point is based on plasma homocysteine as an indicator of folate undernu-
O’CONNOR ET AL.
146
trition (36). 2) Low blood folate individuals had a erythrocyte or plasma folate concentrations below these cutoff values. Statistical analysis. All data were analyzed using SAS for Windows, version 8.01, at the 5% level of statistical significance. Statistical comparisons of the mean TfR concentration, TfR index, and the log TfR:ferritin ratios by Hb concentration (ⱕ133 g/L or ⬎133 g/L), MCV category (ⱕ80, ⬎80 –⬍94, ⱖ94 fL) and Fe status [Fe deficient (with or without anemia), Fe indeterminate, Fe sufficient] were assessed by ANOVA and Least Significant Difference statistics as appropriate. A number of potentially significant covariates were investigated including days postpartum and energy intake, but they were not prognostic of TfR outcomes and hence were not included in the model. In a subgroup analysis, comparisons of mean TfR concentration, TfR index, and the log TfR:ferritin ratios by Hb concentration and Fe status among women with either low or normal blood folate levels were preformed. Transferrin receptor concentrations, TfR indices, and log TfR:ferritin ratios were not normally distributed and were log transformed before performing statistical analysis. Values in the text are means ⫾ SD unless otherwise noted. To determine appropriate cutoff values for TfR concentration, TfR index, and log TfR:ferritin ratio for diagnosing Fe deficiency (with or without anemia) in lactating women at moderate altitude, ROC curves and tables were generated (30). Descriptive statistics including the sensitivity (true positive), specificity (true negative), positive predictive value [true positive divided by (true positive ⫹ false positive)], negative predictive value[true negative divided by (true negative ⫹ false negative)], and positive likelihood ratio [sensitivity divided by (1 ⫺ specificity)] of diagnosing Fe deficiency at various proposed cutoff vales using the aforementioned indicators were generated. Women defined as being Fe deficient (with or without anemia) according to the 3-index Fe assessment model were considered indicative of true Fe deficiency and used as the “gold standard” against which TfR concentration, this classification was TfR index, and log TfR:ferritin ratio were evaluated.
RESULTS The weight and height of women in the sample were 53 ⫾ 7.6 kg (range 38 –77) and 1.5 ⫾ 0.1 m (range 1.3–1.7), respectively. All but 9 of the subjects were stunted (height ⬍ 152.7 cm) and none were wasted (BMI ⬍ 17.8 kg/m2); 12% were overweight (BMI ⬎ 26.5 kg/m2) (37). Mean parity was 3, and 24% of the women were primiparous (n ⫽ 17). Almost 90% of women had ⬍3 y of education. Evidence of infection and/or inflammation was apparent in 6 (8.8%) subjects from measures of C-reactive protein ⬎ 11 mg/L. A considerable number of women had blood biochemical indicators outside the normal range during early lactation (Table 1). Approximately 62 and 58% of women exhibited
anemia based exclusively on Hb (ⱕ133g/L) (31–34) and Hct (41.0%) (31,35) cutoff values, respectively. The prevalence of low serum Fe (ⱕ11 mol/L), transferrin saturation (ⱕ16%), and ferritin (ⱕ12 g/L) all approached 50%. Using the 3-index Fe assessment model with and without Hb, 2 women (3%) were classified as having Fe deficient erythropoiesis, 24 (36%) Fe deficiency anemia, and 19 (29%) indeterminate Fe status. Exclusion of women who had elevated C-reactive protein concentrations did not appreciably alter this distribution. Low plasma folate concentrations were present in 29 women (⬃43%); 13 women (⬃19%) had low blood folate concentrations in conjunction with Hb values ⱕ 133 mg/L. An additional 3 subjects (⬃5%) had low blood folate concentrations in conjunction with both elevated MCV (⬎ 94 fL) and low Hb values; 13 women (20%) exhibited signs of both Fe and folate deficiencies, and 4 women (⬃6%) had anemia of unknown origin. In the overall sample, serum TfR concentration, TfR index, and log TfR:ferritin ratios were significantly higher among lactating Otomi women with a hemoglobin concentration ⱕ 133 g/L compared with those with a hemoglobin concentration ⬎ 133 g/L (Table 2). Similarly, TfR outcomes were higher among women who were classified as Fe deficient according to the 3-index Fe assessment model compared with those categorized as Fe sufficient (Fig. 1). The mean TfR concentration and Tfr index did not differ in women classified either as having indeterminate vs. Fe-deficient status. In contrast, the log TfR:ferritin ratio of women classified as having indeterminate Fe status differed from that of women categorized as either Fe deficient or Fe sufficient. Serum TfR concentration, TfR index, and log TfR:ferritin ratio were higher among women with microcytosis (MCV ⱕ 80 fL, n ⫽ 13) compared with those with normocytosis (n ⫽ 47) or macrocytosis (MCV ⱖ 94 fL, n ⫽ 8). TfR outcomes did not differ between women who were normocytic or had macrocytosis. Blood folate concentrations did not affect TfR outcomes whether all women were included in the analysis or statistical comparisons were made among women with similar Hb concentration or Fe status classification (Table 3). A cutoff value for TfR concentration of 7.75 mg/L appeared to produce the best balance of sensitivity (69.2%) and specificity (72.5%) for diagnosis of tissue Fe deficiency (Table 4). Similarly, cutoff values for TfR index and log TfR: ferritin ratio of 3.5 and 6.0, respectively, produced the best compro-
TABLE 1 Iron and folate measures in blood samples collected from lactating Otomi women 22 ⫾ 13 d postpartum1,2 Mean
Median
Cutoff values3
Outside normal range %
Hemoglobin, g/L Hematocrit, % Mean cell volume, fL Serum iron, mol/L Total iron binding capacity, mol/L Transferrin saturation, % Ferritin, g/L Plasma folate, nmol/L Erythrocyte folate, nmol/L
127 ⫾ 17 40.0 ⫾ 4.8 84.8 ⫾ 7.8 12.2 ⫾ 5.9 65.0 ⫾ 13.6 19.7 ⫾ 11.3 16.2 ⫾ 16.5 12.6 ⫾ 6.0 707.8 ⫾ 302.3
130 (118, 140) 40.0 (37.5, 43.5) 83.4 (80.8, 89.2) 10.9 (7.5, 15.0) 62.6 (55.4, 75.5) 16.5 (11.6, 24.7) 11.2 (4.9, 21.6) 10.9 (8.5, 14.4) 654.6 (509.1, 851.1)
1 n ⫽ 68. 2 Includes women with elevated C-reactive protein concentration (⬎11 mg/L). 3 Cut-off values were derived from References 14, 27, 31, 32, and 36.
ⱕ133 ⬍41 ⱕ80 or ⱖ94 ⱕ11
61.8 57.7 30.9 53.7
ⱕ16 ⱕ12 ⱕ10 ⱕ360
47.0 52.9 42.6 7.5
TRANSFERRIN RECEPTOR CONCENTRATION AND LACTATION
147
TABLE 2 TfR concentrations, TfR indices, and log TfR:ferritin ratios in lactating Otomi women by Hb and MCV1 Hemoglobin (g/L)
TfR, mg/L TfR index log TfR:ferritin ratio
MCV (fL)
All subjects n ⫽ 67–68
ⱕ133 n ⫽ 41–42
⬎133 n ⫽ 26
ⱕ80 fL n ⫽ 13
⬎80–⬍94 n ⫽ 46–47
ⱖ94 n⫽8
8.6 ⫾ 6.7 5.3 ⫾ 6.9 6.5 ⫾ 1.5
10.6 ⫾ 7.5a 7.1 ⫾ 8.2a 7.0 ⫾ 1.2a
5.4 ⫾ 3.4b 2.4 ⫾ 1.9b 5.6 ⫾ 1.3b
16.5 ⫾ 9.8a 13.8 ⫾ 10.9a 8.0 ⫾ 1.0a
6.7 ⫾ 4.2b 3.3 ⫾ 3.4b 6.1 ⫾ 1.4b
7.2 ⫾ 2.0b 2.8 ⫾ 0.8b 6.1 ⫾ 0.7b
1 Values are unadjusted means ⫾ SD. Differences in mean transferrin concentration, TfR index, or log TfR:ferritin ratios were assessed on log transformed data. Means without a common superscript letter differ within each subcategory, P ⬍ 0.05.
mise between sensitivity (73.1 and 92.3%, respectively) and specificity (82.5 and 62.5%, respectively). DISCUSSION Results from this study indicate that the use of TfR concentration, TfR index, and log TfR:ferritin ratio, on their own, do not enhance our ability to detect tissue Fe deficiency among Otomi women with a relatively low prevalence of infection (⬃9%) during early lactation. Further, data show that suboptimal folate status does not influence interpretation of TfR values for the detection of tissue Fe deficiency among lactating women despite a theoretical basis to indicate otherwise. Although TfR concentration, TfR index, and log TfR: ferritin ratio were higher among Fe-deficient than Fe-sufficient lactating Otomi women, there was considerable overlap between these 2 groups for these outcomes (Table 2 and Fig. 1). In addition, the ROC curves and tables generated for TfR concentrations, TfR indexes, and log TfR:ferritin ratios yielded areas under the curve or an accuracy of 0.62, 0.67, and 0.68, respectively, indicating that these measures, on their
own, are poor prognostic tests for determining Fe deficiency among Otomi women during early lactation. Our results are consistent with those of others who reported a great deal of overlap in TfR concentrations among pregnant and lactating Malawian and Zairian women with and without anemia (17,38). In contrast, others report that plasma TfR was found to be a specific marker of Fe deficiency during pregnancy and lactation among American and Swedish women (6,13,39). In other human studies either with men or nonpregnant nonlactating women, TfR outcome measures did not provide substantially more information on Fe status than serum ferritin (24,40,41). Similarly, we were unable to demonstrate that TfR provided substantially more information than serum ferritin concentration (sensitivity 96.2%, specificity 75.0% and accuracy of 0.68 with a cutoff value ⱕ12 g/L) for distinguishing Fe-sufficient from Fe-deficient lactating women. We acknowlTABLE 3 TfR concentrations, TfR indices, and log TfR:ferritin ratios by blood folate measurements in lactating Otomi women residing in Capulhuac, Mexico (elev. 2800 m)1
TfR, mg/L Hemoglobin ⱕ133 Hemoglobin ⬎133 Iron deficient3 Iron sufficient4 TfR index Hemoglobin ⱕ133 Hemoglobin ⬎133 Iron deficient Iron sufficient log TfR:ferritin ratio Hemoglobin ⱕ133 Hemoglobin ⬎133 Iron deficient Iron sufficient FIGURE 1 TfR concentration, TfR index, and log TfR:ferritin ratio of lactating Otomi women residing in Capulhuac, Mexico (elev. 2800 m) by iron status. Using a battery of conventional tests (e.g., Hb, MCV, transferrin saturation, and ferritin) and a 3-index iron assessment model, women were classified as iron deficient (n ⫽ 25–26), iron indeterminate (n ⫽ 19) or iron sufficient (n ⫽ 21). Values are presented as the unadjusted means ⫾ SD (n). Differences in mean transferrin concentration, TfR index, or log TfR:ferritin ratios were assessed on logtransformed data by ANOVA and Least Significant Difference tests. Bars within each category without a common superscript letter differ, P ⬍ 0.05.
g/L g/L
g/L g/L
g/L g/L
Low blood folate2
n
Normal blood folate2
n
9.7 ⫾ 12.7 ⫾ 5.7 ⫾ 11.6 ⫾ 4.6 ⫾
8.4 9.6 4.1 8.1 3.1
30 17 13 12 9
7.8 ⫾ 5.0 9.1 ⫾ 5.4 5.1 ⫾ 2.6 11.4 ⫾ 6.2 5.4 ⫾ 2.5
38 25 13 14 12
8.8 ⫾ 10.6 2.5 ⫾ 2.1 10.5 ⫾ 12.1 1.4 ⫾ 1.0
16 13 11 9
6.1 ⫾ 6.1 2.2 ⫾ 1.8 9.0 ⫾ 7.0 1.7 ⫾ 1.0
25 13 14 12
17 13 12 9
6.9 ⫾ 1.225 5.5 ⫾ 1.3 7.5 ⫾ 1.0 5.2 ⫾ 0.9
13 14 12
7.3 ⫾ 5.7 ⫾ 7.8 ⫾ 4.9 ⫾
1.3 1.4 1.2 1.0
1 Values are unadjusted means ⫾ SD. Differences in mean transferrin concentration, TfR index, or log TfR:ferritin ratios were assessed on log transformed data. The 2 groups did not differ. 2 Individuals with either an erythrocyte folate concentration ⱕ360 nmol/L or a plasma folate concentration ⱕ10 nmol/L were classified as having low blood folate; otherwise they were classified as having normal blood folate. 3 Includes subjects with a low hemoglobin (Hb ⱕ133 g/L) and ⱖ2 indices positive for iron deficiency of serum ferritin (ⱕ12 g/L) transferritin saturation (ⱕ16%) or MCV (ⱕ80 fL). 4 Includes subjects with no values below the normative cutoff for the following measures: serum ferritin, transferrin saturation and MCV.
O’CONNOR ET AL.
148
TABLE 4 Sensitivity, specificity, positive and negative predictive values, and positive likelihood ratio for diagnosing iron deficiency using various cut-off values for TfR, TfR index, and TfR:ferritin ratio Variable
Cut-off
Sensitivity
Specificity
Positive predictive value
Negative predictive value
Positive likelihood ratio
2.75 3.75 4.75 5.75 6.75 7.75 8.75 9.75
96.2 88.5 88.5 73.1 69.2 69.2 57.7 53.8
12.5 22.5 47.5 55.0 65.0 72.5 82.5 87.5
52.3 53.3 62.8 61.9 66.4 71.6 76.7 81.2
76.5 66.1 80.5 67.1 67.9 70.2 66.1 65.5
1.1 1.1 1.7 1.6 2.0 2.5 3.3 4.3
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
92.3 84.6 80.7 76.9 73.1 61.5 61.5 53.8
35.0 45.0 60.0 67.5 82.5 85.0 90.0 90.0
58.7 60.6 66.9 70.3 80.7 80.4 86.0 84.3
82.0 74.5 75.7 74.5 75.4 68.8 70.1 66.1
1.4 1.5 2.0 2.4 4.2 4.1 6.2 5.4
4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
100.0 100.0 100.0 96.2 92.3 84.6 73.1 53.8
5.0 20.0 27.5 35.0 62.5 75.0 80.0 92.5
51.3 55.6 58.0 59.7 71.1 77.2 78.5 87.8
100.0 100.0 100.0 90.1 89.0 83.0 74.8 66.7
1.1 1.3 1.4 1.5 2.5 3.4 3.7 7.2
TfR, mg/L
TfR index
TfR:ferritin ratio
edge, however, that we may have come to a different conclusion if the prevalence of infection had been greater in our sample (4,15). Blood folate concentrations did not affect TfR outcomes whether all women were included in the analysis or statistical comparisons were made among women with similar hemoglobin concentration (ⱕ133 g/L or ⬎133 g/L) or Fe status classification (Fe deficient, Fe sufficient, Table 3). We could find no other studies with lactating women in which the relation between folate status and TfR concentration was examined. Consistent with the low median dietary folate intakes (86 g/d) of lactating women in this community reported previously, the prevalence of low blood folate values was widespread in our sample (7). Folate deficiency was infrequently associated with megaloblastosis in our study; however, megaloblastosis in many instances may have been masked by a concomitant Fe deficiency. de Azevedo Paiva et al. (22) are the only other investigators we are aware of who systematically examined the relation between folate status and TfR concentration in a study of pregnant women. The serum and erythrocyte folate concentrations for lactating women reported in our study are similar to those of the pregnant women in that study. No significant correlation was found in either study between serum/plasma folate and serum TfR concentrations. However, de Azevedo Pavia et al. (22) found a positive correlation (r ⫽ 0.40) between erythrocyte folate and TfR concentrations. This is puzzling because a negative correlation would be expected if folate deficiency affects erythropoiesis. Although the number of Otomi women with macrocytosis in our sample was low, we did not find differences in mean TfR outcomes between women with an elevated MCV (n ⫽ 8) and those with normocytosis (n ⫽ 47, Table 3). Similarly, Carmel et al. (42) did not find that macrocytosis secondary to vitamin
B-12 deficiency in the absence of anemia elevated TfR concentrations. However, Carmel et al. (42) reported in their study of 33 patients with vitamin B-12 deficiency that TfR concentrations were elevated in 6 patients with classic megaloblastic anemia and that the rise correlated strongly with the severity of anemia. All subjects in that study were Fe sufficient. In our study, only 3 subjects (⬃5%) had low blood folate concentrations in conjunction with both elevated MCV (⬎94 fL) and a low Hb concentration. This small sample precludes formal testing of the hypothesis that macrocytosis in conjunction with anemia results in an elevated TfR concentration Although we did not examine the vitamin B-12 concentration in the blood of lactating women in this study, we previously reported that their median dietary intake of vitamin B-12 (1.5 g/d) was low compared with recommended levels (2.8 g/d) (7). Similarly, others report low vitamin B-12 intakes among rural Mexican lactating women (43). Given the low incidence of megaloblastic anemia in the sample and the finding of Carmel et al. (42), we suspect that vitamin B-12 similarly has little effect on TfR concentrations and hence the utility of this index in assessing tissue Fe deficiency. The mean TfR concentrations reported for lactating Otomi women in the present study are similar to published values for lactating Zairian women but higher than in a sample of lactating Swedish women (6,17). Differences likely reflect the level of tissue Fe deficiency among samples and perhaps the moderate altitude at which women in our study resided. Increased erythropoiesis and red cell mass and a resultant increase in TfR concentration is anticipated at higher altitudes in response to reduced partial pressure of oxygen (44). In conclusion, on their own, serum TfR concentration, TfR index, and TfR:ferritin ratio appear to be poor predictors of tissue Fe deficiency for an individual woman during early lactation in Capulhuac, Mexico (moderate altitude); however
TRANSFERRIN RECEPTOR CONCENTRATION AND LACTATION
as a group, lactating women with Fe deficiency have higher values for these indices than do Fe-sufficient women. The utility of these measures for the detection of Fe deficiency does not appear to be influenced by low blood folate concentrations or the presence of megaloblastosis. LITERATURE CITED 1. Stolzfus, R. J. (2001) Defining iron-deficiency anemia in public health terms: a time for reflection. J. Nutr. 565S–567S. 2. Yip, R. (1998) Iron deficiency. Bull. WHO 76 (suppl. 2): 121–123. 3. DeMaeyer, E. & Adiels-Tegman, M. (1985) The prevalence of anaemia in the world. World Health Stat. Q. 38: 302–316. 4. Beard, J. L., Dawson, H. & Pinero, D. J. (1996) Iron metabolism: a comprehensive review. Nutr. Rev. 54: 295–317. 5. National Research Council, Food and Nutrition Board and the Institute of Medicine (2001) Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium and Zinc. National Academy Press, Washington, DC. 6. Akesson, A., Bjellerup, P., Berglund, M., Bremme, K. & Vahter, M. (2002) Soluble transferrin receptor: longitudinal assessment from pregnancy to postlactation. Obstet. Gynecol. 99: 260 –266. 7. Villalpando, S., Latulippe, M. E., Rosas, G., Irurita, M. J., Picciano, M. F. & O’Connor, D. L. (2003) Milk folate but not milk iron concentrations may be inadequate for some infants in a rural farming community in San Mateo, Capulhuac, Mexico. Am. J. Clin. Nutr. 78: 782–789. 8. Picciano, M. F. (2003) Pregnancy and lactation: physiological adjustments, nutritional requirements and the role of dietary supplements. J. Nutr. 133: 1997S–2002S. 9. Ettyang, G. A., van Marken Lichtenbelt, W. D., Oloo, A. & Saris, W. H. (2003) Serum retinol, iron status and body composition of lactating women in Nandi, Kenya. Ann. Nutr. Metab. 47: 276 –283. 10. Haidar, J., Muroki, N. M., Omwega, A. M. & Ayana, G. (2003) Malnutrition and iron deficiency in lactating women in urban slum communities from Addis Ababa, Ethiopia. East Afr. Med. J. 80: 191–194. 11. Dijkhuizen, M. A., Wieringa, F. T., West, C. E., Muherdiyantiningsih & Muhilal (2001) Concurrent micronutrient deficiencies in lactating mothers and their infants in Indonesia. Am. J. Clin. Nutr. 73: 786 –791. 12. Takimoto, H., Yoshiike, N., Katagiri, A., Ishida, H. & Abe, S. (2003) Nutritional status of pregnant and lactating women in Japan: a comparison with non-pregnant/non-lactating controls in the National Nutrition Survey. J. Obstet. Gynaecol. Res. 29: 96 –103. 13. Carriaga, M. T., Skikne, B. S., Finley, B., Cutler, B. & Cook, J. D. (1991) Serum transferrin receptor for the detection of iron deficiency in pregnancy. Am. J. Clin. Nutr. 54: 1077–1081. 14. Cook, J. D. & Skikne, B. S. (1989) Iron deficiency: definition and diagnosis. J. Intern. Med. 226: 349 –355. 15. Beguin, Y. (2003) Soluble transferrin receptor for the evaluation of erythropoiesis and iron status. Clin. Chim. Acta 329: 9 –22. 16. Cooper, M. J. & Zlotkin, S. H. (1996) Day-to-day variation of transferrin receptor and ferritin in healthy men and women. Am. J. Clin. Nutr. 64: 738 –742. 17. Kuvibidila, S., Yu, L. C., Ode, D. L., Warrier, R. P. & Mbele, V. (1994) Assessment of iron status of Zairean women of childbearing age by serum transferrin receptor. Am. J. Clin. Nutr. 60: 603– 609. 18. Sauberlich, H. (1995) Folate status of US population groups. In: Folate in Health and Disease (Bailey, L., ed.), pp. 171–194. Marcel Decker, New York, NY. 19. O’Connor, D. L. (1994) Folate status during pregnancy and lactation. In: Nutrient Regulation During Pregnancy, Lactation, and Infant Growth. (Allen, L., King, J. & Lo¨nnerdal, B., eds.), pp. 157–172. Plenum Press, New York, NY. 20. Qvist, I., Abdulla, M., Jagerstad, M. & Svensson, S. (1986) Iron, zinc and folate status during pregnancy and two months after delivery. Acta Obstet. Gynecol. Scand. 65: 15–22. 21. Metz, J. (1970) Folate deficiency conditioned by lactation. Am. J. Clin. Nutr. 23: 843– 847.
149
22. de Azevedo Paiva, A., Rondo, P. H., Guerra-Shinohara, E. M. & Silva, C. S. (2003) The influence of iron, vitamin B(12), and folate levels on soluble transferrin receptor concentration in pregnant women. Clin. Chim. Acta 334: 197–203. 23. Punnonen, K., Irjala, K. & Rajamaki, A. (1997) Serum transferrin receptor and its ratio to serum ferritin in the diagnosis of iron deficiency. Blood 89: 1052–1057. 24. Malope, B. I., MacPhail, A. P., Alberts, M. & Hiss, D. C. (2001) The ratio of serum transferrin receptor and serum ferritin in the diagnosis of iron status. Br. J. Haematol. 115: 84 – 89. 25. Villalpando, S. F., Butte, N. F., Wong, W. W., Flores-Huerta, S., Hernandez-Beltran, M. J., Smith, E. O. & Garza, C. (1992) Lactation performance of rural Mesoamerindians. Eur. J. Clin. Nutr. 46: 337–348. 26. Bindra, G. S. & Gibson, R. S. (1987) Vitamin B12 and folate status of East Indian immigrants living in Canada. Nutr. Res. 7: 365–374. 27. Gibson, R. S. (1990) Principles of Nutritional Assessment. Oxford University Press, New York, NY. 28. Tamura, T. (1990) Microbiological assay of folate. In: Folic Acid Metabolism in Health and Disease (Picciano, M. F., Gregory, J. & Stokstad, R.E.L., eds.), pp. 121–137. Wiley-Liss and Sons, New York, NY. 29. Fielding, J. (1980) Serum and iron binding capacity. In: Iron (Cook, J., ed.). Churchill Livingstone, New York, NY. 30. Campbell, M. & Machin, D. (1993) Medical Statistics: A Common Sense Approach. John Wiley and Sons, Chichester, UK. 31. Latulippe, M. E., Rosas, M. G., Villalpando, S. & Picciano, M. F. (2000) Utility of TfR and TfR-ferritin index for assessment of iron deficiency is not complicated by folate deficiency in lactating women. FASEB J. 14: A508 (abs.). 32. WHO/UNICEF/UNU (2001) Iron deficiency anemia, assessment, prevention and control: a guide for programme managers. WHO/NHD/01.3. World Health Organization, Geneva, Switzerland. 33. Yip, R. (1993) Altitude and hemoglobin elevation: implications for anemia screening and health risk of polycythemia [abs.]. In: Hypoxia and the Brain: Proceedings of the 8th International Hypoxia Symposium. Charles Houston, Burlington, VT. 34. Dirren, H., Logman, M. H., Barclay, D. V. & Freire, W. B. (1994) Altitude correction for hemoglobin. Eur. J. Clin. Nutr. 48: 625– 632. 35. Centers for Disease Control (1989) CDC criteria for anemia in children and childbearing-aged women. Morb. Mortal. Wkly. Rep. 38: 400 – 404. 36. Brouwer, D. A., Welten, H. T., Reijngoud, D. J., van Doormaal, J. J. & Muskiet, F. A. (1998) Plasma folic acid cutoff value, derived from its relationship with homocyst(e)ine. Clin. Chem. 44: 1545–1550. 37. Kuczmarski, R. J., Ogden, C. L., Grummer-Strawn, L. M., Flegal, K. M., Guo, S. S., Wei, R., Mei, Z., Curtin, L. R., Roche, A. F. & Johnson, C. L. (2000) CDC growth charts: United States. Adv. Data 314: 1–27. 38. Semba, R. D., Kumwenda, N., Hoover, D. R., Taha, T. E., Mtimavalye, L., Broadhead, R., Eisinger, W., Miotti, P. G. & Chiphangwi, J. D. (2000) Assessment of iron status using plasma transferrin receptor in pregnant women with and without human immunodeficiency virus infection in Malawi. Eur. J. Clin. Nutr. 54: 872– 877. 39. Akesson, A., Bjellerup, P., Berglund, M., Bremme, K. & Vahter, M. (1998) Serum transferrin receptor: a specific marker of iron deficiency in pregnancy. Am. J. Clin. Nutr. 68: 1241–1246. 40. Pettersson, T., Kivivuori, S. M. & Siimes, M. A. (1994) Is serum transferrin receptor useful for detecting iron-deficiency in anaemic patients with chronic inflammatory diseases? Br. J. Rheumatol. 33: 740 –744. 41. Mast, A. E., Blinder, M. A., Gronowski, A. M., Chumley, C. & Scott, M. G. (1998) Clinical utility of the soluble transferrin receptor and comparison with serum ferritin in several populations. Clin. Chem. 44: 45–51. 42. Carmel, R. & Skikne, B. S. (1992) Serum transferrin receptor in the megaloblastic anemia of cobalamin deficiency. Eur. J. Haematol. 49: 246 –250. 43. Black, A. K., Allen, L. H., Pelto, G. H., de Mata, M. P. & Chavez, A. (1994) Iron, vitamin B-12 and folate status in Mexico: associated factors in men and women and during pregnancy and lactation. J. Nutr. 124: 1179 –1188. 44. Stray-Gundersen, J., Chapman, R. F. & Levine, B. D. (2001) “Living high-training low” altitude training improves sea level performance in male and female elite runners. J. Appl. Physiol. 91: 1113–1120.