Brief Communication: Sensitivity, Specificity, and Predictive Value of Serum Soluble Transferrin Receptor at Different Stages of Iron Deficiency

Available online at www.annclinlabsci.org Annals of Clinical & Laboratory Science, vol. 35, no. 4, 2005 435 Brief Communication: Sensitivity, Specificity, and Predictive Value of Serum Soluble Transferrin Receptor at Different Stages of Iron Deficiency Jong Weon Choi Department of Laboratory Medicine, College of Medicine, Inha University, Incheon, South Korea Abstract. This study investigated the efficiency of serum soluble transferrin receptor (stfr) for assessing body iron status at different stages of iron deficiency. Among 72 patients with advanced iron-deficiency anemia (IDA), the sensitivity and specificity of stfr (at a diagnostic cutoff of 3.24 mg/l) were 70.8% and 90.6%, respectively, with a positive predictive value of 85.0%. Sensitivities of stfr in patients at the earliest stage of iron deficiency (n = 41) and the intermediate stage of iron-deficient erythropoiesis (n = 15) were 21.9% and 26.7%, respectively, at the same cutoff value of stfr. Serum ferritin concentrations averaged 6.7 ± 1.9 µg/l in IDA patients with stfr <3.24 mg/l, which were significantly above the values in IDA patients with stfr 3.24 mg/l (4.8 ± 1.2 µg/l, p <0.05). In healthy controls, blood reticulocyte counts were significantly higher in subjects with stfr 3.24 mg/l than in those with stfr <3.24 mg/l (0.045 ± 0.013 (10 12 /L) vs 0.034 ± 0.011 (10 12 /L), p <0.05]. In conclusion, stfr level is not a sensitive indicator for the early or intermediate stages of iron deficiency, although stfr assay can be a useful aid in the diagnosis of advanced IDA. Serum stfr concentration has significant relationships with blood reticulocyte counts in healthy subjects and with serum ferritin levels in IDA patients. Keywords: serum soluble transferrin receptor, iron-deficiency anemia, serum ferritin, blood reticulocytes Introduction Iron is essential to most living organisms and participates in a variety of vital processes, such as oxygen transport and enzymatic reactions. Body iron stores are affected by dietary intake and by the physiological need of iron for erythropoiesis [1]. Subclinical iron deficiency is common in adolescents, women of childbearing age, and elderly persons [2,3]. Correct diagnosis of iron deficiency is important for successful treatment of patients because it may be the first clinical sign of serious illness [4]. Measurement of serum ferritin is a laboratory test for diagnosing iron deficiency, and a ferritin value <12 µg/l is generally acknowledged to be a specific indicator of iron deficiency [5]. Address correspondence to Jong Weon Choi, M.D., Ph.D, Department of Laboratory Medicine, Inha University Hospital, 7-206, 3-ga, Shinheung-dong, Jung-gu, Incheon, 400-711, South Korea; tel 82 32 890 2503; fax: 82 32 890 2529; e-mail: jwchoi@inha.ac.kr. However, since ferritin is an acute phase reactant, diagnosis of iron deficiency in hospitalized patients can be difficult, and patients may have increased ferritin values even when they are iron-deficient. Serum soluble transferrin receptor (stfr) concentration increases with iron deficiency but is not influenced by acute or chronic inflammatory diseases [6]. Accordingly, stfr assay has been suggested as a sensitive test for iron deficiency. Investigators showed that subjects with subclinical iron deficits could be readily identified using stfr levels and a stfr-ferritin index [7,8]. Clinical utility of stfr to diagnose iron-deficiency anemia (IDA) and to differentiate IDA from the anemia of chronic disease was extensively studied [9-11]. However, few studies have closely examined the ability of stfr to identify the early or intermediate stages of iron deficiency. In the present study, we investigated the use of stfr to detect iron deficits in patients at different stages of iron deficiency, from the earliest 0091-7370/05/0400-0435. $1.25. 2005 by the Association of Clinical Scientists, Inc. 10 Choi 435-439.indd 435 10/11/05 8:23:37 AM

436 Annals of Clinical & Laboratory Science, vol. 35, no. 4, 2005 stage of iron depletion to advanced IDA. We also investigated the clinical features of subjects who showed disparate values for stfr concentrations and body iron status. Materials and Methods The study was approved by the Committee of Ethics of the Inha University Hospital, and informed consent was obtained from all subjects. Two hundred twenty-four adolescents (59 males and 165 females) with a median age of 18 yr (range 17-19 yr) were investigated by measurements of blood hemogram, serum iron profiles, blood reticulocyte counts, and serum stfr concentrations. The subjects were all South Korean volunteers without diversity of racial composition. Subjects were excluded from this study if they had recent infections (n = 5) or a previous history of vitamin or iron supplementation (n = 3). The subjects were assigned to 4 groups on the basis of their body iron status: (i) the earliest stage (iron depletion, n = 41), (ii) intermediate stage (iron-deficient erythropoiesis, n = 15), (iii) advanced stage (overt IDA, n = 72), and (iv) healthy controls (n = 96). Non-anemic subjects with normal serum iron levels (>50 µg/dl), but with decreased serum ferritin concentration (<12 µg/l), were classified as the earliest stage of iron depletion. Intermediate stage of iron-deficient erythropoiesis was defined as a reduction in body iron beyond the point of depleted iron stores: serum ferritin <12 µg/l and serum iron <50 µg/dl without overt anemia. Subjects showing a decreased serum ferritin concentration, decreased serum iron level, and decreased blood hemoglobin level (<12 g/dl) were considered to have overt IDA (advanced stage). The diagnostic cutoff value of stfr for iron deficiency was 3.24 mg/l, based on the reference intervals of stfr in healthy adolescents [12]. After the subjects had fasted >12 hr, venous blood was drawn in iron-free evacuated tubes. Serum stfr concentrations were measured by an immunoenzymometric method (IDeATM stfr assay, Orion Diagnostica, Espoo, Finland). Complete blood cell counts and red cell indices were measured with EDTA-anticoagulated blood using an electronic counter (SE 9000; Sysmex, Kobe, Japan). Blood reticulocytes and their subpopulations were analyzed by flow cytometry (R-3000; Sysmex). Corrected reticulocyte counts were calculated, based on a normal hematocrit of 45%, and a reticulocyte maturity index was determined from the proportion of reticulocyte subpopulations, as described previously [13]. Serum iron and total iron-binding capacity were assayed spectrophotometrically with a chemical analyzer (Hitachi 7600; Hitachi, Tokyo, Japan) and serum ferritin was measured by a chemiluminescence technique (ACS 180; Bayer Diagnostics, Tarrytown, NY, USA). Data analysis was performed using a non-parametric test (Wilcoxon rank sum test). All p values <0.05 were considered statistically significant. Results and Discussion In this study, we evaluated the usefulness of stfr assays for the assessment of iron status in welldefined populations, who were categorized into several stages of iron deficiency by using conventional laboratory tests such as serum ferritin, serum iron, and blood hemoglobin levels. The present study indicates that stfr is not a sensitive indicator for the subjects in early or intermediate stages of iron deficiency, although stfr has relatively high sensitivity and specificity in the advanced stage of IDA. Incidences of subjects with elevated stfr concentrations at different stages of iron deficiency are listed in Table 1. Among 72 IDA patients, the sensitivity and specificity of stfr (at a diagnostic cutoff of 3.24 mg/l) were 70.8% and 90.6%, respectively, with a positive predictive value of 85.0%. Serum stfr concentrations were 3.24 mg/l in 9 of 41 subjects at an early stage of iron deficiency and in 4 of 15 subjects with iron-deficient erythropoiesis, demonstrating sensitivities of 21.9% and 26.7%, respectively. Our data for stfr concentrations are in partial agreement with the results of Gimferrer et al [14] who found that an increase in stfr levels does not exceed the diagnostic value of a low serum ferritin for evaluating iron deficiency. In contrast, Skikne et al [15] reported that stfr measurement is particularly valuable in identifying a mild degree of iron deficiency of recent onset. Suominen et al [7] reported a significant contribution of stfr to the detection of clinically silent iron deficiency. In our study, diagnostic sensitivity of stfr was low, especially in the ferritin alone-depleted group, which corresponds to an early stage of iron deficiency. These observations suggest that the ability of stfr to screen for iron deficiency remains controversial. These discrepancies may reflect differences in the diagnostic cutoff values for stfr; in our study a slightly higher cutoff point was applied than in the other studies. The inconsistencies may also reflect the substantial differences in clinical characteristics of subject populations among the studies, such as age (adolescents vs adults), nutritional status, and body iron status. 10 Choi 435-439.indd 436 10/11/05 8:23:37 AM

Soluble transferrin receptor levels at stages of Fe deficiency 437 Table 1. Incidence of elevated serum stfr levels in patients at different stages of iron deficiency and in healthy control subjects. Stages of iron deficiency Total No. of cases (n, %) stfr <3.24 mg/l stfr 3.24 mg/l Early stage (iron depletion) 41 32 (78.1%) 9 (21.9%) Intermediate stage (iron-deficient erythropoiesis) 15 11 (73.3%) 4 (26.7%) Advanced stage (overt iron-deficiency anemia) 72 21 (29.2%) 51 (70.8%) Healthy controls 96 87 (90.6%) 9 (9.4%) Table 2. Assays of iron parameters, hematologic parameters, and reticulocyte parameters in patients with IDA. Parameters Total Serum stfr concentrations in IDA patients <3.24 mg/l 3.24 mg/l No. of cases 72 21 51 Age (yr) 17.8 ± 1.7 (18) 17.6 ± 1.2 (18) 17.9 ± 1.9 (18) stfr levels (mg/l) 4.89 ± 2.65 (4.61) 2.10 ± 0.89 (2.21) 6.04 ± 2.26 (5.01) a Iron parameters Serum iron (µg/dl) 28.7 ± 12.4 (32) 34.4 ± 10.6 (36) 26.4 ± 12.1 (30) Serum TIBC (µg/dl) 449.2 ± 46.5 (447) 437.7 ± 51.3 (448) 453.9 ± 44.0 (447) Serum ferritin (µg/l) 5.4 ± 1.6 (5.1) 6.7 ± 1.9 (6.1) 4.8 ± 1.2 (4.7) a Hematologic parameters Blood erythrocytes (10 12 /L) 4.32 ± 0.34 (4.28) 4.15 ± 0.29 (4.21) 4.39 ± 0.36 (4.42) Blood hemoglobin (g/dl) 10.5 ± 0.9 (10.8) 11.1 ± 1.0 (11.2) 10.4 ± 0.9 (10.5) Mean corpuscular volume (fl) 79.6 ± 6.8 (78.7) 82.8 ± 5.2 (83.8) 78.5 ± 6.4 (78.9) Reticulocyte parameters Corrected reticulocytes (%) 0.67 ± 0.20 (0.65) 0.65 ± 0.21 (0.63) 0.67 ± 0.19 (0.65) Reticulocytes (10 12 /L) 0.032 ± 0.012 (0.030) 0.029 ± 0.011 (0.032) 0.034 ± 0.012 (0.031) Reticulocyte maturity index (%) 2.11 ± 1.79 (1.92) 1.47 ± 1.33 (1.35) 2.38 ± 1.89 (2.17) Data are expressed as mean ± SD (median). stfr = soluble transferrin receptor; TIBC = total iron-binding capacity. a Statistically significant (p <0.05) vs subjects with stfr <3.24 mg/l, computed by Wilcoxon rank sum test. Table 3. Assays of iron parameters, hematologic parameters, and reticulocyte parameters in healthy subjects. Parameters Total stfr concentrations in healthy subjects <3.24 mg/l 3.24 mg/l No. of cases 96 87 9 Age (yr) 17.4 ± 1.9 (18) 17.3 ± 2.1 (17) 17.5 ± 1.5 (18) stfr levels (mg/l) 1.71± 1.06 (1.35) 1.47 ± 0.81 (1.29) 3.98 ± 0.54 (3.87) a Iron parameters Serum iron (µg/dl) 120.6 ± 49.7 (109) 123.5 ± 50.3 (114) 92.7 ± 37.9 (86) Serum TIBC (µg/dl) 373.6 ± 52.3 (374) 375.1 ± 53.4 (377) 360.2 ± 40.6 (359) Serum ferritin (µg/l) 26.8 ± 13.9 (21.9) 27.4 ± 14.1 (22.1) 21.3 ± 10.4 (15.2) Hematologic parameters Blood erythrocytes (10 12 /L) 4.76 ± 0.35 (4.74) 4.75 ± 0.36 (4.73) 4.88 ± 0.29 (4.87) Blood hemoglobin (g/dl) 13.9 ± 1.1 (13.8) 13.9 ± 1.0 (13.9) 13.8 ± 0.7 (13.6) Mean corpuscular volume (fl) 86.9 ± 3.2 (87.1) 87.2 ± 2.9 (87.3) 84.5 ± 4.8 (84.6) Reticulocyte parameters Corrected reticulocytes (%) 0.80 ± 0.26 (0.76) 0.79 ± 0.24 (0.74) 0.93 ± 0.32 (0.85) Reticulocytes (10 12 /L) 0.036 ± 0.012 (0.035) 0.034 ± 0.011 (0.032) 0.045 ± 0.013 (0.044) a Reticulocyte maturity index (%) 1.05 ± 0.81 (1.03) 1.07 ± 0.75 (0.91) 0.91 ± 1.26 (0.84) Data are expressed as mean ± SD (median). Abbreviations: see Table 2. a Statistically significant (p < 0.05) vs subjects with stfr < 3.24 mg/l, computed by Wilcoxon rank sum test. 10 Choi 435-439.indd 437 10/11/05 8:23:38 AM

438 Annals of Clinical & Laboratory Science, vol. 35, no. 4, 2005 Mast et al [16] used a diagnostic cutoff level of stfr for iron deficiency in adults of 2.8 mg/l. Skikne [17] emphasized that the reference interval for stfr concentrations depends upon the subjects age and varies with different assay systems. In a previous study, we reported that healthy adolescents, with no evidence of iron deficiency or IDA, had a reference interval of 1.18-3.23 mg/l [12]. On the basis of these results, a stfr concentration of 3.24 mg/l was used in the present study as the diagnostic cutoff value for assessing iron deficiency. If the diagnostic limit for stfr were lowered to 2.8 mg/l in the present study, several subjects (7 IDA patients, 3 subjects at an early stage of iron deficiency, and 7 healthy subjects) would be included in the positive group; the diagnostic sensitivity for stfr would increase from 70.8% to 80.6% in IDA patients and from 21.9% to 29.3% in the early stage of iron deficiency, but the specificity for stfr would decrease from 90.6% to 83.3%. These findings suggest that application of a diagnostic cutoff point of 2.8 mg/l would not significantly improve the sensitivity of stfr assays to detect an early stage of iron deficiency. Serum ferritin undergoes a characteristic sequence of changes as body iron stores decrease. The progressive stages of iron deficiency during a period of negative iron balance include iron depletion, iron-deficient erythropoiesis, and frank IDA. The stage of iron deficiency that is observed in individuals with absent iron stores but who have not yet developed overt anemia is referred to as iron-deficient erythropoiesis or iron deficiency without anemia [15]. Baynes [18] demonstrated that the only indicator of early iron-deficient erythropoiesis is the compensatorily elevated stfr concentration. Suominen et al [7] found that the presence of iron-deficient erythropoiesis can be disclosed by measurement of stfr concentrations. In the present study, sensitivities of stfr levels for iron-deficient erythropoiesis were slightly higher than those in early stage iron deficiency but were still lower compared to those in advanced IDA. These results imply that the diagnostic efficacy of stfr for iron-deficient erythropoiesis is not good enough to be comparable to that for overt IDA. In the present study, 21 of 72 IDA patients showed decreased stfr levels <3.24 mg/l, and 9 exhibited stfr levels that were lower than the mean value (1.71 mg/l) of stfr in healthy subjects. Among 96 healthy controls, 9 displayed an increase in stfr levels 3.24 mg/l. The clinical characteristics of the adolescents with disproportionate results for serum stfr levels and iron parameters were investigated. As shown in Table 2, there were no significant differences in reticulocyte parameters between IDA patients with stfr 3.24 mg/l and those with <3.24 mg/l, nor between the patients with stfr 1.71 mg/l and those with stfr <1.71 mg/l (data not shown). However, serum ferritin concentrations were significantly higher in IDA patients with stfr <3.24 mg/l than in those with stfr 3.24 mg/l (6.7 ± 1.9 µg/l vs 4.8 ± 1.2 µg/l, p <0.05). In healthy adolescents, reticulocytes averaged 0.045 ± 0.013 (10 12 /L) in the subjects with stfr 3.24 mg/ L, which were significantly above the values in those with stfr <3.24 mg/l (0.034 ± 0.011 (10 12 /L), p <0.05), although no significant differences were noted in iron parameters between the 2 groups (Table 3). It thus appears that in healthy subjects, blood reticulocytes counts have some relationship to increased serum stfr concentrations, but in IDA patients the influence of serum ferritin levels on stfr levels is more important than that of blood reticulocyte counts. Serum stfr originates mostly from erythroblasts and to a lesser extent from reticulocytes. In a healthy adult, approximately 80% of stfr is in erythroid precursors in bone marrow [19]. Erythroid precursors and circulating reticulocytes shed their transferrin receptor into circulating blood during the maturation sequence. In our study, diminished serum stfr concentrations, which were observed in some IDA patients without significant reduction in reticulocyte parameters, presumably reflect a decline in erythroid precursors in bone marrow. However, because our study measured only the stfr concentrations and reticulocyte subpopulations in circulating blood, it does not provide direct evidence for a relationship between intramedullary erythroblasts and serum stfr concentrations. In conclusion, the diagnostic performance of stfr was lower in the ferritin alone-depleted group and in subjects with iron-deficient erythropoiesis than in IDA patients. Serum stfr measurement 10 Choi 435-439.indd 438 10/11/05 8:23:39 AM

Soluble transferrin receptor levels at stages of Fe deficiency 439 does not offer a particular advantage over ferritin determination for assessing the early or intermediate stages of iron deficiency, although stfr can be a useful predictor for the diagnosis of frank IDA. References 1. Punnonen K, Irjala K, Rajamaki A. Serum transferrin receptor and its ratio to serum ferritin in the diagnosis of iron deficiency. Blood 1997;89:1052-1057. 2. Holyoake TL, Stott DJ, McKay PJ, Hendry A, MacDonald JB, Lucie NP. Use of plasma ferritin concentration to diagnose iron deficiency in elderly patients. J Clin Pathol 1993;46:857-860. 3. Bruner AB, Joffe A, Duggan AK, Casella JF, Brandt J. Randomised study of cognitive effects of iron supplementation in non-anaemic iron-deficient adolescent girls. Lancet 1996;348:992-996. 4. Rockey DC, Cello JP. Evaluation of the gastrointestinal tract in patients with iron-deficiency anemia. NEJM 1993;329:1691-1695. 5. Ali MA, Luxton AW, Walker WH. Serum ferritin concentration and bone marrow iron stores: a prospective study. Can Med Assoc J 1978;118:945-946. 6. Pettersson T, Kivivuori SM, Siimes MA. Is serum transferrin receptor useful for detecting iron-deficiency in anaemic patients with chronic inflammatory diseases? Br J Rheumatol 1994;33:740-744. 7. Suominen P, Punnonen K, Rajamaki A, Irjala K. Serum transferrin receptor and transferrin receptor-ferritin index identify healthy subjects with subclinical iron deficits. Blood 1998;8:2934-2939. 8. Punnonen K, Irjala K, Rajamaki A. Serum transferrin receptor, ferritin and TfR-F index in identification of latent iron deficiency. Eur J Haematol 1998;60:135-137. 9. Flowers CH, Skikne BS, Covell AM, Cook JD. The clinical measurement of serum transferrin receptor. J Lab Clin Med 1989;114:368-377. 10. 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;119:385-390. 11. Rimon E, Levy S, Sapir A, Gelzer G, Peled R, Ergas D, Sthoeger ZM. Diagnosis of iron deficiency anemia in the elderly by transferrin receptor-ferritin index. Arch Intern Med 2002;162:445-449. 12. Choi JW, Pai SH, Im MW, Kim SK. Change in transferrin receptor concentrations with age. Clin Chem 1999;45:1562-1563. 13. Choi JW, Pai SH. Reticulocyte subpopulations and reticulocyte maturity index (RMI) rise as body iron status falls. Am J Hematol 2001;67:130-135. 14. Gimferrer E, Ubeda J, Royo MT, Marigo GJ, Marco N, Fernandez N, Oliver A, Padros R, Gich I. Serum transferrin receptor levels in different stages of iron deficiency. Blood 1997;90:1332-1334. 15. Skikne BS, Flowers CH, Cook JD. Serum transferrin receptor: a quantitative measure of tissue iron deficiency. Blood 1990;75:1870-1876. 16. Mast AE, Blinder MA, Gronowski AM, Chumley C, Scott MG. Clinical utility of the soluble transferrin receptor and comparison with serum ferritin in several populations. Clin Chem 1998;44:45-51. 17. Skikne BS. Circulating transferrin receptor assay: coming of age. Clin Chem 1998;44:7-9. 18. Baynes RD. Refining the assessment of body iron status. Am J Clin Nutr 1996;64:793-794. 19. Huebers HA, Finch CA. The physiology of transferrin and transferrin receptors. Physiol Rev 1987;67:520-582. 10 Choi 435-439.indd 439 10/11/05 8:23:39 AM