Changes in Serum Lipid Concentrations during Iron Depletion and after Iron Supplementation

Annals of Clinical & Laboratory Science, vol. 31, no. 2, 2001 151 Changes in Serum Lipid Concentrations during Iron Depletion and after Iron Supplementation Jong Weon Choi, 1 Soon Ki Kim, 2 and Soo Hwan Pai 1 1 Department of Clinical Pathology 2 Department of Pediatrics College of Medicine, Inha University, and Inha University Hospital, Inchon, South Korea, Abstract. To investigate the effects of body iron depletion and iron supplementation on serum lipid concentrations, hematologic indices, iron markers, and serum lipid profiles were measured in 427 girls, age 14-19 yr. There were no significant differences in serum lipid concentrations between subjects with moderate iron deficiency anemia (blood Hb <12.0 g/dl) and healthy controls. However, serum total cholesterol concentration (mean ± SD, 148 ± 16 mg/dl) in severely anemic subjects with blood Hb <8.0 g/dl was significantly lower than in subjects with blood Hb 14.0 g/dl (170 ± 17 mg/dl) (p <0.01). Moreover, serum triglyceride concentration in subjects with blood Hb >14.0 g/dl was 2-fold higher than in the severely anemic subjects. Mean values of serum total cholesterol and triglyceride (149 ± 17 mg/dl and 58 ± 22 mg/dl) in girls with severe anemia were significantly elevated after iron supplementation (164 ± 17 mg/dl and 98 ± 26 mg/dl) (p <0.01, respectively). In the severely anemic subjects, blood Hb concentration was correlated with serum total cholesterol (r = 0.49, p <0.01) and triglyceride concentrations (r = 0.51, p <0.01). These findings indicate that severe iron deficiency anemia in girls is attended by decreased concentrations of serum total cholesterol and triglyceride, and that these reduced serum lipid levels return to normal following iron supplementation. (received 7 November 2000, accepted 30 November 2000) Keywords: iron deficiency, iron supplementation, serum cholesterol, serum triglyceride, serum lipid profiles Introduction Elevated cholesterol levels increase the risk of arteriosclerosis and coronary heart disease [1]. As abnormal lipid and lipoprotein levels are related to atherogenesis, many attempts have been made to evaluate serum lipid and lipoprotein profiles. During adolescence, considerable changes in endogenous sex hormone levels and in serum lipoprotein metabolism have been observed [2]. Iron is an essential metal involved in a wide spectrum of physiological functions in the body, such as oxygen transport and enzymatic reactions. Nevertheless, excess iron can be harmful because it promotes the generation of free radicals, which lead to tissue damage [1,3]. Iron deficiency is commonly Address correspondence to Soo Hwan Pai, M.D., Department of Clinical Pathology, Inha University Hospital, 7-206, 3-ga, Shinheung-dong, Jung-gu, Inchon, 400-103, South Korea; tel 82 32 890 2502; fax 82 32 890 2529; e-mail shpaimd@inha.ac.kr. encountered in young infants and adolescent girls. The progressive stages of iron deficiency during a period of negative iron balance include iron depletion (stage I), iron deficient erythropoiesis (stage II), and iron deficiency anemia (stage III). During the iron depletion phase, tissue iron stores are exhausted, but neither anemia nor decreased serum iron concentration is evident. In the iron deficient stage, erythropoiesis, serum iron, and serum ferritin levels are decreased, but anemia and hypochromia are still not apparent [4]. In an epidemiological study, high body iron stores were related to an increased risk of coronary heart disease [5]. Subsequent studies have shown that low serum iron-binding capacity and high serum iron concentrations are risk factors for myocardial infarction [6,7]. Although associations have been found between dietary iron intake and serum lipid and lipoprotein concentrations in animal models [8,9], such relationships have not been investigated extensively in humans, and the available data are inconsistent. Knowledge about the effect of iron deficiency on serum lipid 0091-7370/01/0200/0149 $1.50; 2001 by the Association of Clinical Scientists, Inc.

152 Annals of Clinical & Laboratory Science profiles is limited. Therefore, in the present study, we investigated the changes in serum lipid concentrations that occur during progressive stages of iron depletion and after iron supplementation. Materials and Methods We measured complete blood cell counts (CBC), iron markers, and serum lipid profiles in 427 girls, age 14 19 yr, selected from 1,704 middle- or high-school students who had been screened for iron deficiency anemia. We excluded 76 subjects from this study: 21 were excluded because they had recently had infections, 19 had histories of iron or vitamin supplementation, 8 had undergone surgical operations, and 28 had inconsistent assay results for serum iron and serum ferritin concentrations. The subjects were all South Korean students from middle-class families; there were no significant differences in racial composition or socioeconomic status among the groups. This study was explained to and approved by the subjects parents and by the academic administrators at each educational center, and only volunteers were included in the study population. The study was approved by the Committee of Ethics of the Inha University Hospital, and informed consent was obtained from all subjects. The subjects were divided into 4 groups: iron depletion phase (n = 102), iron deficient erythropoiesis (n = 73), iron deficiency anemia (n = 115), and healthy controls (n = 137). Non-anemic subjects with normal serum iron level (>50 µg/dl) but decreased ferritin concentration (<12 µg/l) were classified as being in the iron depletion phase. Iron deficient erythropoiesis was defined as the presence of serum ferritin concentration <12 µg/l and serum iron level <50 µg/ dl without overt anemia. When subjects with decreased serum ferritin concentration and decreased serum iron level also had decreased blood hemoglobin level (<12.0 g/dl), they were considered to have iron deficiency anemia. We assigned subjects with blood hemoglobin <8.0 g/dl to the severely anemic group and compared their serum lipid levels with those of subjects with hemoglobin >14.0 g/dl. To investigate the effect of iron supplementation on serum lipid concentrations, 52 girls with hemoglobin <8.0 g/dl were supplemented with one 256 mg tablet of ferrous sulfate (80 mg of elemental iron) per day for 3 mo; however, vitamin B12 or folate was not supplemented. The compliance for iron supplementation was checked by interviews throughout the period of treatment; blood samples were obtained from 41 of these girls for assays of serum lipid profiles after 3 mo of iron supplementation. Venous blood (7 ml) was drawn into iron-free, evacuated, serum separator tubes after 12 h of fasting. CBC was measured with EDTA-anti-coagulated blood within 3 h after collection. Routine CBC and red cell indexes were determined with an electronic counter (SE 9000; Sysmex, Kobe, Japan). Serum iron, total iron-binding capacity (TIBC), and all lipid profiles were assayed with an automatic chemical analyzer (Hitachi 747; Hitachi, Tokyo, Japan). Serum ferritin was measured by the chemiluminescence method (ACS 180; Chiron, MA, USA) within 4 h after collection. Serum concentrations of triglyceride, total cholesterol, and low- and high-density lipoproteincholesterol were analyzed by an enzymatic colorimetric method using triglyceride GPO-PAP reagents (Roche Diagnostics GmbH, Mannheim, Germany), SICDIA L T-CHO reagents (Eiken Chemical Industries, Tokyo, Japan), Cholestest-LDL reagents (Daiichi Chemicals, Tokyo, Japan), and Cholestest-HDL reagents (Daiichi Chemicals), respectively [10 13]. Data analysis was performed using the SAS statistical computing software package (version 6.12 for Windows; SAS Institute, Inc., Cary, NC, USA). The Mann-Whitney U test was used to calculate the statistical significance of differences. The correlation between serum lipid concentrations and blood hemoglobin levels was assessed by Pearson correlation coefficients. All p values <0.01 were considered statistically significant. Results The clinical features and laboratory data of the subjects in this study are summarized in Table 1. Of the 427 girls, high-school students comprised 53.8% of the total. In the 197 middle-school students, 48 (24%) were anemic (hemoglobin <12.0 g/dl) and 127 (64%) had ferritin levels <12 µg/l, while in the 230 highschool students, 67 (29%) were anemic and 163 (71%) had hypoferritinemia. There were no significant

Serum lipid concentrations in severely iron-deficient girls 153 Table 1. Clinical parameters and laboratory data c of the girls included in this study. Parameters Middle- High- p-value school school students students No. of subjects 197 230 nt Age (yr, range) 14-16 17-19 nt Anemia a (n) 48 (24%) 67 (29%) ns Hypoferritinemia b (n) 127 (64%) 163 (71%) ns Blood hemoglobin (g/dl) 13.0 ± 3.9 12.6 ± 3.6 ns Serum iron (µg/dl) 110 ± 59 91 ± 59 ns Transferrin saturation (%) 32 ± 16 23 ± 17 ns Serum ferritin(µg/l) 16 ± 11 16 ± 11 ns a Blood Hb 12 g/dl b serum ferritin 12 µg/l c mean ± SD nt = not tested ns = not significant differences in the blood hemoglobin, serum iron, and serum ferritin concentrations between the middle- and high-school students. Figure 1 demonstrates the changes in serum lipid concentrations with changing blood hemoglobin (Hb) values. Total cholesterol and triglyceride concentrations increased progressively with increasing Hb levels. There were no significant changes in serum HDL- and LDLcholesterol levels with increasing Hb levels. Table 2 summarizes the differences in the mean values of serum lipids during the various stages of iron depletion and after iron supplementation. Total cholesterol and triglyceride levels in healthy controls were slightly higher than in subjects with moderate iron deficiency anemia (Hb 12.0 g/dl), but there were no significant differences between these groups. To compare the differences in serum lipid concentrations between the severely anemic group and subjects with high hemoglobin levels, we divided them into two groups: subjects with blood Hb <8.0 g/dl, and those with blood Hb >14.0 g/dl. As shown in Table 2, the total cholesterol level in the severe anemic group (148 ± 16 mg/dl) was significantly lower than in the other group (170 ± 17 mg/dl), ie, subjects with hemoglobin > 14.0 g/dl (p < 0.01). The triglyceride level in subjects with Hb >14.0 g/dl was 2-fold higher than in subjects with Hb < 8.0 g/dl. Fig. 1. Relationships of serum lipids to blood hemoglobin (Hb) levels in 427 girls. Total cholesterol (TC, closed triangles) and triglyceride (TG, closed circles) increase with Hb levels; there were no significant changes of HDLcholesterol (open circles) or LDL-cholesterol (open triangles) with increasing Hb levels. Table 2 also demonstrates the change in serum lipid concentrations in the girls with severe anemia after iron supplementation. Total cholesterol and triglyceride concentrations (164 ± 17 mg/dl and 98 ± 26 mg/dl, respectively) were significantly higher after iron supplementation, compared with the values (149 ± 17 mg/dl and 58 ± 22 mg/dl) before supplementation. The correlations between hemoglobin levels and lipid profiles in severely anemic girls are shown in Figs. 2 and 3. Blood Hb levels were correlated significantly with serum total cholesterol (r = 0.49, p <0.01) and triglyceride levels (r = 0.51, p < 0.01). No significant correlations were observed between serum HDL- or LDL-cholesterol concentrations and blood Hb values. Discussion In this study, we investigated the changes in the serum concentrations of triglyceride, total cholesterol, HDLcholesterol and LDL-cholesterol during iron deficiency

154 Annals of Clinical & Laboratory Science Table 2. Changes in serum lipid concentrations with iron deficiency and iron supplementation in apparently healthy girls. Population No. of Lipid concentrations (mean ± SD; [median]) parameters cases Total cholesterol Triglyceride HDL-cholesterol LDL-cholesterol mg/dl mg/dl mg/dl mg/dl Body iron status Iron depletion 102 159 ± 20 [158] 85 ± 32 [85] 54 ± 11 [54] 87 ± 19 [88] Iron deficient erythropoiesis 73 156 ± 19 [157] 82 ± 31 [78] 55 ± 10 [56] 86 ± 18 [86] Iron deficient anemia 115 154 ± 19 [152] 70 ± 29 [69] 57 ± 9 [55] 83 ± 17 [84] Healthy controls 137 161 ± 20 [162] 96 ± 32 [94] 54 ± 10 [54] 89 ± 19 [90] Hemoglobin levels Hb <8.0 g/dl 52 148 ± 16 [146] a 56 ± 22 [56] a 58 ± 9 [58] 81 ± 16 [80] Hb 14.0 g/dl 54 170 ± 17 [170] a 115 ± 30 [115] a 51 ± 8 [52] 93 ± 19 [94] Iron treatment Before treatment 41 149 ± 17 [147] b 58 ± 22 [58] b 57 ± 9 [57] 79 ± 16 [79] After treatment 41 164 ± 17 [165] b 98 ± 26 [97] b 53 ± 9 [52] 85 ± 16 [85] a p value = <0.01 for subjects with Hb <8 g/dl versus those with Hb 14 g/dl, calculated by Mann-Whitney U test. b p value = <0.01 for subjects before iron treatment versus those after treatment, calculated by Mann-Whitney U test. Fig. 2. Scatter plot that shows the correlation between blood hemoglobin concentration (X-axis) and serum total cholesterol concentration (Y-axis) in 52 girls with hemoglobin concentration <8.0 g/dl. The equation of the correlation line is Y = 10.72X + 74.65; the correlation coefficient (r) = 0.49 (p <0.01). Fig. 3. Scatter plot that shows the correlation between blood hemoglobin concentration (X-axis) and serum triglyceride concentration (Y-axis) in 52 girls with hemoglobin concentration <8.0 g/dl. The equation of the correlation line is Y = 13.32X - 35.48; the correlation coefficient (r) = 0.51 (p <0.01).

Serum lipid concentrations a nd iron-deficiency anemia 155 and after iron supplementation. Our data are in general agreement with the results of Au and Schilling [14], who found that the degree of anemia was associated with lower serum cholesterol levels in sex-linked anemic mice. The changes in serum lipid concentrations during iron depletion have varied in different animal studies. Guthrie et al [15] and Amine et al [16] found that hyperlipidemia developed when anemia was produced by an iron-deficient diet. Sherman et al [17] reported that plasma lipid levels were elevated in rat pups from anemic dams. Bristow-Craig et al [8] showed that higher dietary iron levels were associated with higher serum cholesterol levels in rats. These observations suggest that changes in serum lipid parameters depend on the age, sex, and strain of the animal model and the composition of the diets used in the experiments. El-Hazmi et al [18] reported that serum cholesterol levels were significantly lower in patients with sickle cell anemia compared to matched controls with normal hemoglobin levels. In our study, the serum total cholesterol concentration in subjects with moderate iron deficiency anemia did not differ significantly from healthy controls, although it tended to decrease as iron depletion progressed. However, our data showed that serum concentrations of total cholesterol and triglyceride were significantly lower in girls with severe anemia than in subjects with high hemoglobin levels. Our results are in accord with the previous studies in showing that severe iron deficiency affects lipid metabolism. On the basis of our results, we believe that the decreases in serum lipid concentrations are related to the severity of iron deficiency anemia. Animal studies have indicated that lipid status may be influenced by dietary iron intake [19,20]. In our study, after iron supplementation, the initially reduced concentrations of total cholesterol and triglyceride returned to levels comparable to the healthy controls. These results are consistent with the reports of investigators [9,19] who demonstrated that dietary iron supplementation increases plasma lipids in rats. Obviously, it is difficult to compare such data from animal models with the present observations in young human females. Ohira et al [19] found that total cholesterol concentrations were elevated following an increase in hemoglobin levels by transfusion and iron treatment. They also reported that the concentration of red blood cells might affect cholesterol synthesis or its mobilization from tissue to plasma. In contrast to these findings, Stangl and Kirchgessner [20] reported hypertriglyceridemia in rats receiving a very low level of dietary iron. On the other hand, Ece et al [21] found that iron deficiency itself has no direct effect on the lipid and lipoprotein profile. They suggested that an iron-deficient diet may be deficient in energy and protein, and that a hypocaloric diet could cause hypolipidemia. In the present study, only an iron tablet was given as a supplement, and no hypercaloric diet or vitamins were provided. Therefore, it appears that iron supplementation can elevate serum total cholesterol and triglyceride levels in girls with severe anemia caused by iron deficiency. Relationships between hypocholesterolemia and anemia have been reported, and the effects of anemia on the development of atherosclerosis have been studied [22 24]. Dabbagh et al [23] demonstrated that iron overload causes a significant increase in plasma total and HDL-cholesterol levels in rats. However, another of their studies [24] did not support the hypothesis that elevation of iron stores increases the risk of coronary artery disease. Studies have shown a positive correlation between serum cholesterol level and hematocrit in healthy humans [25] and in anemic patients during iron supplementation [26]. El-Hazmi et al [18] reported a significant positive correlation between serum cholesterol and blood hemoglobin levels in patients with sickle cell anemia. In our study, blood hemoglobin levels correlated significantly with serum cholesterol concentrations, but only in girls with severe iron deficiency anemia. Our data are in accord with the report of Ohira et al [19], who showed that in subjects with hemoglobin < 9.0 g/dl, there was a significant relationship between hemoglobin and cholesterol levels; this relationship was not seen in subjects with hemoglobin > 9.0 g/dl. They found that triglyceride levels were independent of blood hemoglobin levels. In contrast, in our study, blood hemoglobin levels correlated significantly with serum concentrations of triglyceride in girls with severe anemia. In our opinion, this discrepancy may be attributed to differences in the subjects. Our study only involved girls, age 14 19 yr, who showed anemia with hemoglobin < 8.0 g/dl; this experimental design avoids

156 Annals of Clinical & Laboratory Science the influences of gender, age, and occupations on serum lipid levels. In conclusion, our results suggest that severe iron deficiency anemia in girls is attended by decreased serum concentrations of total cholesterol and triglyceride, and that these serum lipid abnormalities return to normal with iron supplementation. References 1. Lee TS, Shiao MS, Pan CC, Chau LY. Iron-deficient diet reduces atherosclerotic lesions in apoe-deficient mice. Circulation 1999;99:1222-1229. 2. Stozicky F, Slaby P, Volenikova L. Longitudinal study of serum cholesterol, apolipoproteins and sex hormones during puberty. Acta Paediatr Scand 1991;80:1139-1144. 3. Boldt DH. New perspective on iron: an introduction. Am J Med Sci. 1999;318:207-212. 4. 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. 5. Salonen JT, Nyyssonen K, Karpela H, Tuomilekto J, Seppanen R, Salonen R. High stored iron levels are associated with excess risk of myocardial infarction in Eastern Finnish men. Circulation 1992;86:803-811. 6. Magnusson MK. Sigfusson N, Sigvaldason H, Johanneson GM, Magnusson S, Thorgeirsson G. Low iron binding capacity as a risk factor for myocardial infarction. Circulation 1994;89:102-108. 7. Morrison HI, Semenciw RM, Mao Y, Wigle DT. Serum iron and risk of fatal acute myocardial infarction. Epidemiology 1994;5:243-246. 8. Bristow-Craig HE, Strain JJ, Welch RW. Iron status, blood lipids and endogenous antioxidants in response to dietary iron level in male and female rats. Int J Vit Nutr Res 1994;64:324-329. 9. Cunnane SC, McAdoo KR. Iron intake influences essential fatty acid and lipid composition of rat plasma and erythrocytes. J Nutr 1987;117:1514-1519. 10. Allain CC, Poon LS, Chan CS, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem 1974;20:470-475. 11. Friedwald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502. 12. Olson RE. A critique of the report of the National Institute of Health Expert Panel on detection, evaluation, and treatment of high blood cholesterol. Arch Intern Med 1989;149:1501-1503. 13. Gordon T, Castelli WP, Hjortland MC, Kanel WB, Dawber TR. High density lipoprotein as a protective factor against coronary heart disease. Am J Med 1977; 62:707-714. 14. Au YP, Schilling RF. Relationship between anemia and cholesterol metabolism in sex-linked anemic (gene symbol, sla) mouse. Biochim Biophys Acta 1986;883: 242-246. 15. Guthrie HA, Froozani M, Sherman AR, Barron GP. Hyperlipidemia in offspring of iron-deficient rats. J Nutr 1974;104:1273-1278. 16. Amine EK, Desilets EJ, Hegsted DM. Effect of dietary fats on lipogenesis in iron deficiency anemic chicks and rats. J Nutr 1976;106:405-411. 17. herman AR, Guthrie HA, Wolinsky I, Zulak IM. Iron deficiency hyperlipidemia in 18-day-old rat pups: effects of milk lipids, lipoprotein lipase and triglyceride synthesis. J Nutr 1978;108:152-162. 18. El-Hazmi MAF, Jabbar FA, Warsy AS. Cholesterol and triglyceride level in patients with sickle cell anaemia. Scan J Clin Lab Invest 1987;47:351-354. 19. Ohira Y, Edgerton VR, Gardner GW, Senewiratne B. Serum lipid levels in iron deficiency anemia and effects of various treatments. J Nutr Sci Vitaminol 1980;26: 375-379. 20. Stangl GI, Kirchgessner M. Different degrees of moderate iron deficiency modulate lipid metabolism of rats. Lipids 1998;33:889-895. 21. Ece A, Yigitoglu MR, Vurgun N, Guven H, Iscan A. Serum lipid and lipoprotein profile in children with iron deficiency anemia. Pediatr Int 1999;41:168-173. 22. Bari MA, Rahman MA. Effect of iron deficiency anemia on the development of atherosclerosis in chicks. Atherosclerosis 1975;21:401-408. 23. Dabbagh AJ, Mannion T, Lynch SM, Frei B. The effect of iron overload on rat plasma and liver oxidant status in vivo. Biochem J 1994;300:799-803. 24. Dabbagh AJ, Shwaery GT, Keaney JF Jr, Frei B. Effect of iron overload and iron deficiency on atherosclerosis in the hypercholesterolemic rabbit. Arterioscler Thromb Vasc Biol 1997;17:2638-2645. 25. Bottiger LE, Carlson LA. Relation between serum cholesterol and triglyceride concentration and haemoglobin values in non-anaemic healthy persons. Br Med J 1972;3:731-733. 26. Westerman MP. Hypocholesterolemia and anaemia. Br J Haematol 1975;31:87-94,