During pregnancy, the iron requirement generally exceeds the amount provided through the diet,1,2 and there is a need to accurately determine iron status. Conventional methods for assessment of iron status during pregnancy are either not sensitive enough, eg, red cell indices, or are altered by gestation independently of iron status, eg, hemoglobin and markers of iron transport. The main shortcoming of serum ferritin as a marker is that a low value does not distinguish between depleted iron stores—a common condition in late gestation—and iron deficiency.2,3 The soluble transferrin receptor (sTfR) in serum seems to be a new, promising way to detect iron deficiency.
The cell-surface transferrin receptor (TfR) transports iron into the cell. The amount of the truncated form in serum, sTfR, reflects both the receptor density on cells (tissue iron deficiency) and the numbers of cells with receptors (erythropoietic activity).4,5 Consequently, sTfR correlates inversely with the amount of iron available for tissues, and directly with the rate of erythropoiesis. Thus, sTfR is useful for detection of iron deficiency that has not yet developed into anemia6 and may distinguish individuals with iron-deficiency anemia from those with anemia caused by chronic disease.7–9 Combining sTfR with serum ferritin to a ratio increases the accuracy with which iron deficiency can be diagnosed.6,10–14
Previous studies on sTfR during pregnancy have either focused on late pregnancy or used a cross-sectional study design.15–17 Recently, we showed that sTfR is a specific and sensitive index of iron deficiency during pregnancy.18 The aim of this follow-up study was to use sTfR for evaluation of the impact of pregnancy and lactation on iron status and erythropoiesis. We measured sTfR and the ratio sTfR:serum ferritin (sTfR/Fer) longitudinally during puerperium, lactation, and postlactation and compared data to the results previously obtained during pregnancy. We also measured sTfR in cord blood.
MATERIALS AND METHODS
Women in early gestation attending any of the three antenatal care units in Solna, Stockholm County, Sweden, and who understood written information in Swedish were invited to participate and to be followed for almost 2 years in an environmental exposure assessment study.18–20 In total, 254 of the 618 pregnant women registered at the antenatal care units during 15 months (1994–1996) agreed to participate and were eligible. Of these, 17 had miscarriage, two had induced abortion, and five delivered before second blood sampling.
We collected venous blood samples twice during pregnancy, at gestational weeks 11 and 36,18 and three times postpartum (during the first week, in the lactation period, and postlactation), and measured ferritin and sTfR in serum. The average time (and range) of sampling was 3 days (1–8 days), 3 months (1.3–5.8 months), and 15 months (12–19 months) postpartum. In addition, cord serum was collected at delivery (on average at gestational week 40, range 34–42) and analyzed for sTfR in a subsample of the neonates.
All sampling except that postlactation was carried out during routine maternity care. The participation rate postlactation was lower (n = 67) because of problems in locating the women and difficulties in getting them to come to the hospital for sampling. Seven of these 67 women were pregnant again. To be able to make a longitudinal comparison of the iron status markers throughout the study, we compared maternal age, number of pregnancies, parity, smoking, bleeding at delivery, length of gestation, birth weight, birth length, time of sampling, serum ferritin and sTfR between the group who was lost to follow-up and those who continued the study. We regarded the women continuing the study as representative of the recruited group, because those who were lost to follow-up at various stages did not differ significantly with respect to the analyzed variables from those who continued the study.
As is customary at the antenatal care units, the women were advised to take iron supplements, mainly in the form of a low dose (18 mg Fe whereof 2.4 mg as heme iron) daily from gestational week 20. No individual data on iron supplementation were collected during pregnancy, but the subjects who remained in the study (n = 67) completed a questionnaire postlactation. According to the questionnaire, 82% of the women reported taking iron supplements during pregnancy, and 12% during lactation, mainly in the low-dose form (18 mg). The average length of lactation was, according to the questionnaire, 6 months (5th–95th percentiles, 2.3–12 months).
Determination of serum ferritin is described elsewhere.18 The sTfR was measured with kits from Ramco Laboratories, Inc. (Houston, TX).21 The total interassay coefficients of variation for two different control concentrations were 8.9% and 10.3% (38 duplicate measurements at each concentration). In addition, we stored and analyzed one authentic serum sample together with the samples in each kit (coefficient of variation = 7.8%; 19 duplicate measurements). There was no systematic change over time. Values of sTfR less than 2.9 mg/L were considered indicative of decreased erythropoiesis and values greater than 8.3 were indicative of tissue iron deficiency according to the manufacturer (Ramco Laboratories, Inc.). A value of sTfR/Fer greater than 500 was considered to be indicative of depleted iron stores, and greater than 2000 as indicative of iron deficiency.6,14,22
Statistical analyses were made using SPSS Inc. (Chicago, IL) and SAS Institute Inc. (Cary, NC) software programs. The repeated measures analyses of variance on logarithmically transformed serum ferritin, sTfR, and sTfR/Fer (time as the within factor with five levels), where the missing values were assumed to be at random, were carried out using procedure “Mixed” in SAS (handling missing values).23 To compensate for multiple tests in the comparison over time (minimizing the type 1 error), the obtained P value was multiplied by seven (equal to the total number of comparisons made). Parity (grouped according to zero, one, or two to three previous children) and iron treatment (grouped as nonsupplemented or supplemented) were analyzed as the between-factor effect in the repeated measures analysis of variance. The interaction between iron treatment and time was also investigated. Between-subject heterogeneity was incorporated in the model if necessary. We used Mann-Whitney test to detect differences between two independent groups, and Wilcoxon signed rank test for two dependent groups. Correlation analysis was performed by Spearman rank correlation analysis. Boxplots depict 25th, 50th, and 75th percentiles and whiskers minimum and maximum, excluding outliers. The reference interval was determined based on the 2.5th and 97.5th percentiles.24 This study was carried out in accordance with the Helsinki declaration with the approval of the ethics committee at Karolinska Institutet.
There was a loss to follow-up in the present study, especially postlactation. However, in all the tests performed, the women who continued the study were found to be representative of the recruited group, indicating no discernable selection bias. We were therefore able to perform analysis of variance for repeated measures in a program designed to handle missing values.
Figure 1 shows the longitudinal variations in serum ferritin, sTfR, and sTfR/Fer from early gestation to postlactation. The overall results of the repeated measures analyses showed statistically significant differences over time for all three markers (P < .001). Tests for differences between sampling occasions showed that sTfR and sTfR/Fer were lower in early gestation compared with all later sampling occasions (P < .001). The two markers decreased significantly from late gestation to the puerperium (P < .001). From the puerperium, sTfR, but not sTfR/Fer, increased significantly to the lactation period (P = .015). There were no changes in sTfR or sTfR/Fer from lactation to postlactation. The serum ferritin values showed similar statistically significant changes as did sTfR/Fer but in the opposite direction (P ≤ .002). When parity was analyzed as the between factor, the overall differences were toward a deterioration of iron status with increasing parity for all three markers (P ≤.012).
The sTfR concentrations below the reference interval of 2.9–8.3 mg/L (indicating decreased erythropoiesis) were present in 5% of the women in the puerperium (Figure 1). The percentages of sTfR greater than 8.3 mg/L (indicating tissue iron deficiency) were 8%, 7%, and 10% at puerperium, lactation, and postlactation, respectively. The percentages of sTfR/Fer greater than 500 units (indicating depleted iron stores) were 21%, 13%, and 20% at puerperium, lactation, and postlactation, respectively.
To study alterations in erythropoiesis, we controlled for variations in iron status by selecting the women who had serum ferritin greater than 20 μg/L both at gestational week 11 and postlactation. In these women, the median sTfR concentration in early gestation (4.0 mg/L) was significantly lower than postlactation (4.3 mg/L, P = .002, n = 25). If the same comparison was made between gestational week 11 and puerperium, no difference was observed (P = .44, n = 35). Similar results were obtained for the women with serum ferritin greater than 30 μg/L.
We evaluated the results of the three markers in relation to iron supplementation as recorded in the post-lactation questionnaire (Figure 2). There was a statistically significant difference over time in sTfR (P = .010), but not in serum ferritin (P = .071) or sTfR/Fer (P = .084) between the supplemented and nonsupplemented group. There was no interaction between time and supplementation for sTfR (P = .87).
Iron status indices during pregnancy were not associated with gestational age at term. Women who miscarried did not differ significantly in serum ferritin (median 41 μg/L) or sTfR (3.8 mg/L) compared with the other women (33 μg/L and 4.1 mg/L, respectively, P = .11 and .35).
The average blood loss at delivery was 400 mL (range 100–3300 mL). Iron status (all markers) during the lactation period was significantly correlated to the blood loss (eg, sTfR/Fer and mL blood loss: correlation coefficient [rs] = 0.35, P < .001). Postlactation (in both pregnant and nonpregnant women), both serum ferritin and sTfR/Fer, but not sTfR, remained significantly correlated to the blood loss (rs = 0.26, P = .042, and rs = 0.31, P = .022, respectively). There were no correlations between iron status on the one hand, and start of menses postpartum or length of breast feeding on the other hand.
The median sTfR in cord was 9.5 mg/L (range 6.3–18.4 mg/L). On average, the concentration was 1.6 times higher than maternal sTfR at gestational week 36 (P < .001, n = 24) and 2.2 times higher than maternal sTfR during the puerperium (P = .002, n = 13). The correlation coefficients between sTfR in cord blood and maternal blood at gestational week 11 and 36 were 0.45 (P = .025, n = 25) and 0.40 (P = .054, n = 24), respectively. Cord blood sTfR was not correlated to maternal serum ferritin, gestational age at birth, weight, length, or head circumference at birth. There was no statistically significant difference in sTfR between girls (9.2 mg/L) and boys (9.9 mg/L, P = .67). Associations between cord sTfR and other birth outcomes were not possible to calculate because of the limited number of subjects.
This longitudinal study revealed marked changes in tissue iron status and erythropoiesis in relation to pregnancy and lactation. Iron status did not return to first-trimester levels postpartum. Although iron status may be improved in early gestation compared with before pregnancy because of the absence of menstruation, we were able to prove that pregnancy led to a deterioration of iron status. Each successive pregnancy, as well as loss of blood at parturition, resulted in a lower iron status postpartum. Postlactation, more than 70% of the women had reduced stores of iron (serum ferritin less than 30 μg/L), and 10% had tissue iron deficiency (sTfR greater than 8.3 mg/L).
Several studies have shown that iron supplementation during pregnancy is required to restore serum ferritin concentration to first-trimester levels during the lactation period.25–27 As a part of standard routine at the antenatal care units, all the women in the present study were advised to take iron orally in a low dose from midpregnancy. According to the questionnaire completed by 67 women postlactation, about 80% of the pregnant women did so. Apparently, the amount was insufficient to replenish the iron stores after pregnancy. Although this study was not designed to evaluate the effect of iron supplementation, and we have no information on actual compliance, the supplementation data (Figure 2) further support that the supplementation in this study had very little effect on iron status. The observed statistically significant differences in sTfR between the supplemented and nonsupplemented groups may have been there already before the start of supplementation, as there was no significant interaction between treatment and time. We, therefore, conclude that to prevent this negative effect of childbearing on iron status, postpartum supplementation must be considered in women who bleed excessively at parturition and in those who, for any reason, choose to take a low dose of iron—or none at all—during pregnancy.
In our previous study,18 we showed that in women who had depleted iron stores both in early and late gestation, sTfR increased by 74%. Generally, this change reflects the overall effect of both increased erythropoiesis and the development of tissue iron deficiency. In women who maintained their iron stores during pregnancy, sTfR still increased by 25%. This increase in sTfR is likely to be the result of increased erythropoiesis. We also suggested that sTfR was lowered in early gestation.18 In the present work, we were able to confirm this, as sTfR was 7–14% lower in early gestation than post-lactation in the group of women who had adequate iron stores on both occasions. Apparently, this issue is not well investigated, and it has often been stated that there is a continuous increase in erythropoiesis already from early pregnancy.28,29 Our results speak against this, as do some other studies showing lower sTfR and blunted erythropoietin production in early gestation compared with late gestation.30,31 In addition, serum erythropoietin has been shown to be the major predictor of sTfR in pregnancy.30 One way to confirm a decreased erythropoiesis in early pregnancy is to study women longitudinally starting before conception. One can speculate that such decreased erythropoiesis helps ensure the necessary increase in plasma volume initiated early in gestation, which in turn may provide the stimulus for a later accelerated red-cell production. Most likely, sTfR in early gestation is affected by this decrease in erythropoiesis, and we propose 2.4–6.7 mg/L as a more appropriate reference interval in early pregnancy (compared with 2.9–8.3 mg/L for nonpregnant women and men according to the manufacturer of the kit used).
All three markers changed toward an improvement of iron status from gestational week 36 to the first week of the puerperium. This could be explained by a return of iron from the expanded erythrocyte mass. However, it is not known whether this return can be detectable just 3 days after delivery, as the time course of this release is not clear.32 On the other hand, a decreased erythropoiesis, lowering the sTfR concentration just after delivery, cannot be excluded. This is supported by the fact that there was a statistically significant increase in sTfR from puerperium to lactation, but no correspondent decrease in serum ferritin, findings similar to those reported by others.33 In addition, similarly low concentrations of sTfR were obtained in the puerperium as at gestational week 11, when those with depleted iron stores were excluded. Furthermore, sTfR results below the reference interval were only present at these two occasions. Although increased erythropoiesis in early puerperium has been suggested,2,30,34 recent data obtained using modern techniques speak against that conclusion.32 Thus, the low sTfR concentration during the puerperium in the present study would rather suggest a decreased erythropoiesis around 3 days after delivery. We did not recalculate the reference interval for the puerperium, as the sample size was not large enough.
The higher concentration of sTfR in the neonate than in the mother agrees with other reports, which indicate that the concentration gradually declines from birth to childhood to adolescence.35–39 Possible explanations for higher sTfR in neonates include different regulation of TfR production, different shedding of TfR from maturing fetal erythrocytes, or contribution of receptors from the placenta (rich in TfR). Our results and those of others36 argue against the possibility that cord sTfR levels reflect the rapid cell proliferation and tissue growth, as no association between cord sTfR and gestational age at birth or birth weight was found. Also, a higher age-related rate of erythropoiesis was excluded as explanation in a study of 1–2-year-old children.37 The question remains, however, whether the higher concentration of sTfR in neonates still reflects the iron status of the newborn. Previous reports concerning sTfR in neonates have been contradictory. One study of anemic mothers and their newborns considers sTfR to identify those with iron-deficient erythropoiesis.40 Other studies conclude that sTfR levels are independent of iron metabolism36,38 or reflect fetal erythropoiesis.38 We did not evaluate iron status by means of other indices in the newborn, but found that sTfR in cord blood was weakly correlated to maternal sTfR but not to maternal serum ferritin. However, fetal iron status is relatively independent of maternal iron stores, as the fetus accumulates iron normally despite mild-to-moderate maternal iron deficiency.41 This may explain the disagreement between the above-mentioned studies. Further studies are warranted to clarify the implication of sTfR findings in neonates.
1. Milman N, Bergholt T, Byg KE, Eriksen L, Graudal N. Iron status and iron balance during pregnancy. A critical reappraisal of iron supplementation. Acta Obstet Gynecol Scand 1999;78:749–57.
2. Letsky EA. The haematological system. In: Chamberlain G, Broughton-Pipkin F, eds. Clinical physiology in obstetrics. 3rd ed. Oxford: Blackwell Science, 1998:71–110.
3. Cook JD. Defining optimal body iron. Proc Nutr Soc 1999;58:489–95.
4. Feelders RA, Kuiper-Kramer EP, van Eijk HG. Structure, function and clinical significance of transferrin receptors. Clin Chem Lab Med 1999;37:1–10.
5. R'zik S, Beguin Y. Serum soluble transferrin receptor concentration is an accurate estimate of the mass of tissue receptors. Exp Hematol 2001;29:677–85.
6. Skikne BS, Flowers CH, Cook JD. Serum transferrin receptor: A quantitative measure of tissue iron deficiency. Blood 1990;75:1870–6.
7. Suominen P, Mottonen T, Rajamäki A, Irjala K. Single values of serum transferrin receptor and transferrin receptor ferritin index can be used to detect true and functional iron deficiency in rheumatoid arthritis patients with anemia. Arthritis Rheum 2000;43:1016–20.
8. Ferguson BJ, Skikne BS, Simpson KM, Baynes RD, Cook JD. Serum transferrin receptor distinguishes the anemia of chronic disease from iron deficiency anemia. J Lab Clin Med 1992;119:385–90.
9. Punnonen K, Irjala K, Rajamäki A. Iron-deficiency anemia is associated with high concentrations of transferrin receptor in serum. Clin Chem 1994;40:774–6.
10. Means RT Jr, Allen J, Sears DA, Schuster SJ. Serum soluble transferrin receptor and the prediction of marrow aspirate iron results in a heterogeneous group of patients. Clin Lab Haematol 1999;21:161–7.
11. Punnonen K, Irjala K, Rajamäki A. Serum transferrin receptor and its ratio to serum ferritin in the diagnosis of iron deficiency. Blood 1997;89:1052–7.
12. Suominen P, Punnonen K, Rajamäki A, Irjala K. Serum transferrin receptor and transferrin receptor-ferritin index identify healthy subjects with subclinical iron deficits. Blood 1998;92:2934–9.
13. Cermak J, Brabec V. Transferrin receptor-ferritin index: A useful parameter in differential diagnosis of iron deficiency and hyperplastic erythropoiesis. Eur J Haematol 1998;61:210–2.
14. van den Broek NR, Letsky EA, White SA, Shenkin A. Iron status in pregnant women: Which measurements are valid? Br J Haematol 1998;103:817–24.
15. Carriaga MT, Skikne BS, Finley B, Cutler B, Cook JD. Serum transferrin receptor for the detection of iron deficiency in pregnancy. Am J Clin Nutr 1991;54:1077–81.
16. Choi JW, Im MW, Pai SH. Serum transferrin receptor concentrations during normal pregnancy. Clin Chem 2000;46:725–7.
17. Rusia U, Flowers C, Madan N, Agarwal N, Sood SK, Sikka M. Serum transferrin receptors in detection of iron deficiency in pregnancy. Ann Hematol 1999;78:358–63.
18. Åkesson A, Bjellerup P, Berglund M, Bremme K, Vahter M. Serum transferrin receptor: A specific marker of iron deficiency in pregnancy. Am J Clin Nutr 1998;68:1241–6.
19. Osman K, Åkesson A, Berglund M, Bremme K, Schütz A, Ask K, et al. Toxic and essential elements in placentas of Swedish women. Clin Biochem 2000;33:131–8.
20. Åkesson A, Berglund M, Schütz A, Bjellerup P, Bremme K, Vahter M. Cadmium exposure in pregnancy and lactation in relation to iron status. Am J Public Health. In press.
21. Åkesson A, Bjellerup P, Vahter M. Evaluation of kits for measurement of the soluble transferrin receptor. Scand J Clin Lab Invest 1999;59:77–81.
22. Baynes RD. Refining the assessment of body iron status. Am J Clin Nutr 1996;64:793–4.
23. Littell RC, Milliken GA, Stroup W, Wolfinger RD. SAS system for mixed models. Clark, NC: SAS Institute, Inc., 1996.
24. Solberg HE. Establishment and use of reference values. In: Burtis CA, Ashwood ER, eds. Tietz textbook of clinical chemistry. 3rd ed. Philadelphia: WB Saunders, 1999: 336–56.
25. Taylor DJ, Mallen C, McDougall N, Lind T. Effect of iron supplementation on serum ferritin levels during and after pregnancy. Br J Obstet Gynaecol 1982;89:1011–7.
26. Williams MD, Wheby MS. Anemia in pregnancy. Med Clin North Am 1992;76:631–47.
27. Milman N, Agger AO, Nielsen OJ. Iron supplementation during pregnancy. Effect on iron status markers, serum erythropoietin and human placental lactogen. A placebo controlled study in 207 Danish women. Dan Med Bull 1991;38:471–6.
28. Duvekot JJ, Peeters LLH. Very early changes in cardiovascular physiology. In: Chamberlain G, Broughton-Pipkin F, eds. Clinical physiology in obstetrics. 3rd ed. Oxford: Blackwell Science, 1998:3–32.
29. Cavill I. Iron and erythropoiesis in normal subjects and in pregnancy. J Perinat Med 1995;23:47–50.
30. Beguin Y, Lipscei G, Thoumsin H, Fillet G. Blunted erythropoietin production and decreased erythropoiesis in early pregnancy. Blood 1991;78:89–93.
31. Bianco I, Mastropietro F, D'Asero C, Graziani B, Piergrossi P, Mezzabotta M, et al. Serum levels of erythropoietin and soluble transferrin receptor in the course of pregnancy in non-beta-thalassemic and beta-thalassemic women. Haematologica 2000;85:902–7.
32. Richter C, Weitzel H, Huch A, Huch R. Failure to demonstrate postpartum functional iron deficiency using quantitative red blood cell analysis. Eur J Obstet Gynecol Reprod Biol 1999;84:69–74.
33. Mara M, Zivny J, Eretova V, Kvasnicka J, Kuzel D, Umlaufova A, et al. Changes in markers of anemia and iron metabolism and how they are influenced by antianemics in postpartum period. Acta Obstet Gynecol Scand 2001;80:142–8.
34. Richter C, Huch A, Huch R. Erythropoiesis in the post-partum period. J Perinat Med 1995;23:51–9.
35. Olivares M, Walter T, Cook JD, Hertrampf E, Pizarro F. Usefulness of serum transferrin receptor and serum ferritin in diagnosis of iron deficiency in infancy. Am J Clin Nutr 2000;72:1191–5.
36. Kuiper-Kramer EP, Baerts W, Bakker R, van Eyck J, van Raan J, van Eijk HG. Evaluation of the iron status of the newborn by soluble transferrin receptors in serum. Clin Chem Lab Med 1998;36:17–21.
37. Virtanen MA, Viinikka LU, Virtanen MK, Svahn JC, Anttila RM, Krusius T, et al. Higher concentrations of serum transferrin receptor in children than in adults. Am J Clin Nutr 1999;69:256–60.
38. Carpani G, Buscaglia M, Ghisoni L, Pizzotti D, Vozzo N, Bellotti M, et al. Soluble transferrin receptor in the study of fetal erythropoietic activity. Am J Hematol 1996;52:192–6.
39. Choi JW, Pai SH, Im MW, Kim SK. Change in transferrin receptor concentrations with age. Clin Chem 1999;45:1562–3.
40. Rusia U, Flowers C, Madan N, Agarwal N, Sood SK, Sikka M. Serum transferrin receptor levels in the evaluation of iron deficiency in the neonate. Acta Paediatr Jap 1996;38:455–9.
41. Petry CD, Wobken JD, McKay H, Eaton MA, Seybold VS, Johnson DE, et al. Placental transferrin receptor in diabetic pregnancies with increased fetal iron demand. Am J Physiol 1994;267:E507–E514.