Measuring iron and zinc bioavailability in humans

Transcription

1 Janet C. King, Carmen M. Donangelo, Leslie R. Woodhouse, Sarah D. Mertz, David M. Shames, Fernando E. Viteri, Z. Cheng, and Ross M. Welch Abstract and zinc deficiencies are common in populations dependent on cereal-based diets because of the poor bioavailability of these minerals in those foods. Selective breeding of high-mineral grains can improve the total intake of iron and zinc. However, the additional iron and zinc from those grains may not be available for absorption because of the high phytate content of cereals. and zinc bioavailability needs to be measured before the high-mineral crops are promoted. or zinc bioavailability can be measured from the response of a physiological variable, assessment of body retention, tissue or blood uptake, changes in pool size, or rates of absorption. bioavailability is preferentially measured from erythrocyte uptake of oral radioactive or stable iron tracers; zinc bioavailability is measured from the rate of absorption of an oral isotopic tracer compared with an intravenous tracer. The oral label, which is required for studies of both iron and zinc, may be intrinsically added to the plant during growth or extrinsically added before feeding. and zinc bioavailability from intrinsically and extrinsically labelled normal and high-mineral common bean varieties was tested in young women with low iron stores. The absorption of intrinsic and extrinsic labels of iron and zinc did not differ. The bioavailability of iron and zinc from both varieties was low, about 1.5% and 13%, respectively. Methods to improve the bioavailability of iron and zinc from plant foods need to be developed. Janet King, Carmen Donangelo, Leslie Woodhouse, Sarah Mertz, and David Shames are affiliated with the US Department of Agriculture-Agricultural Research Service (USDA- ARS) Western Human Nutrition Research Center at the University of California, Davis, California, USA. Fernando Viteri is affiliated with the University of California, Berkeley, California. Z. Cheng and Ross Welch are affiliated with the USDA-ARS Plant, Soil, and Nutrition Laboratory at Cornell University in Ithaca, New York, USA. Mention of the names of firms and commercial products does not imply endorsement by the United Nations University. Introduction and zinc deficiencies are common not only in developing countries, where the diet is essentially composed of cereals and root crops, but also in some industrialized countries. It is estimated that about half of the world s population consumes insufficient iron and zinc. -deficiency anaemia can impair mental and psychomotor development in children, increase morbidity and mortality of mothers and children at childbirth, decrease work performance, and decrease resistance to infection [1]. deficiency in humans impairs growth, sexual maturity, and the immune defense system [1]. Although body zinc homeostasis is maintained over a wide range of zinc intakes by increasing or decreasing both intestinal zinc absorption and endogenous intestinal excretion, low zinc intakes combined with poor bioavailability and increased needs during growth or reproduction ultimately result in zinc deficiency [2]. absorption from different foods varies considerably. There are two types of iron in food: nonhaem iron, which is present in both plant foods and animal tissues, and haem iron from haemoglobin and myoglobin in animal products. Haem iron represents 30% to 70% of the total iron in lean meat and is always well absorbed. Non-haem iron enters a common nonhaem iron pool in the gastric juice, and the amount of iron absorbed depends to a large extent on the presence of enhancing and inhibitory substances in the meal and the iron status of the individual. Approximately 50% of the variation in non-haem iron absorption can be explained by intake of animal tissue, phytic acid, ascorbic acid, and iron status [3]. Haem iron is absorbed by a different mechanism from non-haem iron. The composition of the meal has little influence on haem iron absorption because it is taken up intact by pinocytosis. [4]. Although haem iron represents only 10% to 15% of dietary iron intake in populations with high meat intakes, it may contribute more than 40% of the total absorbed iron. The absorption of non-haem iron varies widely 434 Food and Nutrition Bulletin, vol. 21, no , The United Nations University.

2 from less than 1% to more than 50%, depending on the dietary components and the iron status of the individual. It is usually about 1% to 20%. The main inhibitory substances are phytic acid from cereals and legumes and polyphenol compounds from tea and coffee [5]. The main enhancers are ascorbic acid from fruits and vegetables and partially digested peptides from muscle tissues. In the United States and other developed countries, meat, seafood, and dairy products are the primary sources of zinc, providing about 70% of the total intake. In developing countries, where the intake of animal products or tissues is low, cereals and legume seeds provide most of the dietary zinc [6]. Phytic acid is a strong inhibitor of zinc absorption. Animal proteins counteract the inhibiting effect of phytic acid on zinc absorption [7]. The total amount of zinc in the diet also influences zinc absorption. At low zinc levels with no inhibitory agents, zinc absorption can be as high as 50%; high intakes cause a lower fractional absorption. The reduction is not directly proportional to the increase in content, however, so a higher amount is always absorbed at a higher intake [7]. Measurement of iron and zinc bioavailability Bioavailability is the fraction of the ingested nutrient that is utilized for normal physiological functions or storage [8]. One of the major determinants of bioavailability is the proportion that is absorbed from the gastrointestinal tract. However, tissue utilization (or lack thereof) of the absorbed nutrient also influences 435 bioavailability. This is particularly true for some vitamins and minerals, such as selenium. Gastrointestinal absorption is the primary determinant of iron and zinc bioavailability. When evaluating the bioavailability of iron and zinc from a food item in humans, it is important to remember that a number of factors unrelated to the characteristics of the food influence the proportion absorbed. Those factors include the previous intake of the nutrient, the body status of the nutrient, gut transit time, and gastrointestinal function. Methods of measuring bioavailability Five general approaches are available for assessing the bioavailability of iron and zinc in humans (table 1). The first is the response of a physiological variable or reversal of a deficiency symptom, such as low haemoglobin for iron deficiency. This method is limited, however, to studies in individuals with iron depletion, and the response, or change in haemoglobin concentration, is likely to be influenced by the degree of depletion. Measurement of whole-body iron or zinc retention is a second approach. Mass balance can quantitate the whole-body retention of the minerals from a controlled diet containing the food or foods of interest. This method is flawed for some nutrients, such as zinc, where the amount absorbed is not tightly regulated and the excess absorbed is secreted back into the gastrointestinal tract. Thus, faecal zinc losses are a mixture of unabsorbed dietary zinc and absorbed endogenous zinc. This secretion into the gut leads to an underestimate of bioavailability. Administrating an oral zinc TABLE 1. Techniques for measuring iron and zinc bioavailability in humans Measurement Response of a physiological variable Assessment of body retention Tissue or blood uptake Changes in body pool sizes Rates of absorption Technique Change in haemoglobin concentration in iron-deficient individuals Not used; no sensitive indicator of changes in zinc status available mass balance; retention of a single oral dose of an isotopic tracer; whole-body retention of radioiron mass balance; whole-body retention of radiozinc incorporation into erythrocytes Not used; no single tissue available for assessing zinc incorporation Plasma iron response test Plasma zinc response test; changes in exchangeable zinc pool sizes Compartmental modeling of iron absorption; dual isotopic tracer studies Compartmental modeling of zinc absorption; dual isotopic tracer studies; deconvolution analysis

3 436 J. C. King et al. isotope and measuring faecal losses only partly corrects the problem, because the isotope absorption usually exceeds the need, and excesses are also secreted back into the gut [9]. Since iron absorption is regulated tightly, measuring the excretion of a single oral dose of iron in the stool is a valid estimate of iron bioavailability [10]. Whole-body isotope retention may also be estimated from the retention of a radioactive isotopic tracer of iron or zinc, i.e., 55 Fe, 59 Fe, or 65 Zn, which is given orally with the test food. The initial whole-body activity is measured after consumption of the test food, before any unabsorbed isotope has been excreted, and two weeks after tracer administration, when the unabsorbed tracer has cleared the gut [11 13]. The percentage absorbed is the portion of initial wholebody activity that remains after two weeks with correction for physical decay and for background activity. The third approach involves measuring tissue or blood uptake or retention of an isotopic tracer. This approach, the incorporation of radioiron into erythrocytes, is the preferred method for measuring iron absorption [14]. The protocol entails feeding a single meal labelled with a radioactive iron tracer, withholding food (but not water) for several hours thereafter, analysing the radioactivity in the blood two weeks after the meal, and expressing the fraction absorbed as a ratio to absorption from a reference dose given with water to minimize potential confounding effects of intersubject variability caused by differences in iron status. The method requires an estimate of total blood volume and an assumption that 80% or more of the absorbed iron isotope is incorporated into erythrocytes within 14 days. This approach has not been used to measure zinc bioavailability because of the lack of a suitable tissue for measuring retention. Assessment of changes in body pool sizes is a fourth approach used to measure the bioavailability of minerals. Evaluating the plasma concentration curve over several hours following an oral dosing of the mineral under study is an example of this approach. This technique has the drawbacks that a large amount of the nutrient must be given to produce a reproducible plasma response, and it is uncertain that the plasma concentration reflects the concentration at the tissue site of action [8]. Nevertheless, it has been used to screen the bioavailability of zinc or iron from different foods and to study mineral mineral interactions [15 19]. Since the method only requires measurement of plasma zinc or iron concentration in several blood samples, it is relatively easy and inexpensive. Radioactive and stable isotopic tracers may be used in whole-body kinetic models for measuring changes in zinc exchangeable pool sizes [20, 21]. Although it appears that the size of the exchangeable zinc pool measured over two days varies with dietary zinc intake, further studies are needed to determine if the size of the pool is sensitive to intakes of diets differing in zinc bioavailability. The fifth approach, measuring the rate of iron or zinc gastrointestinal absorption, is the most direct measure of the bioavailability of the mineral. Isotopic tracers provide several methods for measuring the rate of iron or zinc absorption: compartmental analysis, deconvolution curves, or dual isotopic tracer studies. All of these methods require administering two isotopes, one orally with the test food and the second intravenously. In the deconvolution and dual isotopic tracer methods, zinc absorption is estimated from the ratio of the oral to the intravenous isotope concentrations in plasma and/or urine over the following three to five days. Lowe and co-workers recently compared the estimates of zinc absorption in a group of women when all of these methods were done simultaneously [9]. The deconvolution and the dual isotopic tracer methods compared favourably with the compartmental model estimate of zinc absorption. The dual isotopic tracer method requires only a single blood or urine sample, however, whereas the deconvolution method requires multiple blood and urine samples. Thus, the dual isotopic tracer method is the preferred method for measuring zinc absorption in humans (table 2). The intravenous infusion and the expense of administering and measuring several stable isotopes are drawbacks to this method. Administration of oral isotopic tracer The preferred methods for measuring both iron and zinc bioavailability require oral administration of isotopic tracers. Oral tracers may be administered in two ways: intrinsically or extrinsically. An intrinsic tracer is one that has been incorporated into the food during growth of the plant or animal; an extrinsic tracer is added to the food in small amounts as inorganic salts several hours before feeding. When an extrinsic tracer is used, it is assumed that the label equilibrates with all pools of iron or zinc in the food and is absorbed to the same extent and at the same rate as the intrinsic element in the food. An extrinsic tracer is the preferred method for administering the oral tracer, because it is less expensive and easier to prepare than an intrinsic label. Numerous validity studies comparing extrinsic with intrinsic tracers of iron or zinc have been done [10, 22 26]. The studies suggest that the validity of extrinsic tracers for measuring iron or zinc bioavailability varies with the test food. Complete equilibration between extrinsic and intrinsic tags in the gastrointestinal tract is dependent on gut transit time and on the mineral-binding ligands present in the food. Thus, to be certain that an extrinsic tag is a valid tracer for bioavailability studies, the two types of tags should be compared before an extrinsic tag is used. In general, extrinsic labelling

4 437 TABLE 2. Preferred methods for measuring iron and zinc bioavailability in humans Measurement Procedure Disadvantages : erythrocyte Label meal or food with radioiron ( 55 Fe or 59 Fe) Exposure to radioiron incorporation of Measure initial ingestion before any faecal loss radioiron by whole-body counting Measure retention 2 weeks later by whole-body counting Assume 80% of absorbed iron is incorporated into erythrocytes Estimate total body volume based on body weight : dual isotopic Label meal or food with one stable isotope of Requires intravenous infusion tracer method zinc ( 67 Zn) Infuse second isotope following ingestion of Expense of stable isotopes test meal/food ( 70 Zn) Measure ratio of oral to intravenous isotope 3 5 days later in plasma or urine for non-haem-iron bioavailability appears to be valid for most foods; milk may be an exception because of the more rapid transit of the liquid food [25]. extrinsic tags, however, do not appear to exchange fully with endogenous zinc in many foods [23]. Extrinsic versus intrinsic measures of iron and zinc bioavailability from common beans Several years ago, an international collaborative project to increase the bioavailable concentrations of iron and zinc from staple plant foods was initiated [27]. Studies of the iron and zinc content of the common bean show that selective breeding strategies could increase the amount of iron by 60% to 80% and the amount of zinc by about 50% [28]. Before these high-iron, highzinc varieties of the common bean are introduced into the agricultural systems of those regions of the world where intake of these minerals is inadequate, the bioavailability of iron and zinc from the legume needs to be determined. Therefore, we evaluated iron and zinc bioavailability from a test meal composed only of the common bean in a group of young women with low iron stores. The results of this study will determine whether the less expensive extrinsic tracers can be used in future field studies evaluating the impact of highiron, high-zinc beans on mineral nutriture. Subjects and methods Twenty-three women, 20 to 26 years of age, participated in the study and were divided into two groups. Group I (n = 12) received a test meal of typical common beans; group II (n = 11) received the high-iron, highzinc beans. Intrinsic and extrinsic iron absorption was compared in both groups; intrinsic and extrinsic zinc absorption was compared only in group II. All of the women were non-smokers, apparently healthy with no recent use of vitamin or mineral supplements, of acceptable weight-for-height, and users of oral contraceptive agents. At entry to the study, all women had haemoglobin concentrations greater than 110 g/l and plasma ferritin concentrations less than 20 g/l. The study was approved by the Radiation Safety Review Committee and the Committee for Protection of Human Subjects of the University of California, Berkeley. The typical common beans provided 0.92 µmol of iron and 0.43 µmol of zinc per gram; the highiron, high-zinc beans provided 1.5 µmol of iron and 0.85 µmol of zinc per gram. The test food, fed at 7:15 a.m., consisted of cooked beans, equivalent to 40 g dry weight, and deionized water. Immediately after consumption of the test food, a third stable isotope of zinc was infused into the antecubital vein. absorption was estimated from the incorporation of radiolabelled iron into erythrocytes 14 days after oral administration of the tracer. absorption was estimated by the dual isotopic tracer method in spot urine samples collected in the mornings on days 3, 4, and 5 following the test meal [9]. The iron and zinc contents of the test meal, plasma, erythrocytes, and urine were determined by atomic absorption spectrophotometry (Thermo-Jarrell Ash 22, Massachusetts, USA). Aliquots of samples were digested with concentrated HNO 3 (Fisher Scientific, Pennsylvania, USA, TM grade) in a microwave digester (MDS 2000, CEM Corp., USA) and diluted with deionized water before measurement. absorption was calculated as the ratio of 59 Fe or 55 Fe in total erythrocytes and the ingested tracer dose, expressed as a percentage. The iron tracer content of blood was determined by liquid scintillation counting after acid digestion and ion-exchange chromatography. The amount of 59 Fe or 55 Fe in the blood was estimated from a calculation of total blood volume [29]. The esti-

5 438 J. C. King et al. mated incorporation of radioiron into the erythrocyte cell mass was based on the initial blood haemoglobin and plasma ferritin concentrations [30]. The ratio of oral ( 70 Zn and 68 Zn) to intravenous ( 67 Zn) isotopic zinc tracers was measured in urine by a Sciex ELAN 500 Inductively Coupled Plasma Mass Spectrometer (Perkin-Elmer, Norwalk, Connecticut, USA). Before measurement, zinc was separated from the other elements in the urine by ion-exchange chromatography. The data were tested for normality and log-transformed for plasma ferritin and iron absorption before further statistical analyses. Differences between intrinsic and extrinsic estimates of iron or zinc absorption were determined by a paired t test. The results were considered to be significantly different for p <.05. Results The composition of the typical beans and the highiron, high-zinc beans is shown in table 3. The highiron, high-zinc bean had 67% more iron and twice as much zinc as the typical bean. The amount of phytate (myo-inositolpentaphosphoric acid plus myoinositolhexaphosphoric acid [IP6 + IP5]) and calcium was similar in both beans. The phytate/iron and phytate/zinc molar ratios were high in both varieties of beans. There were no differences between intrinsic and extrinsic iron absorption from either the typical or the high-iron, high-zinc beans.the intrinsic and extrinsic geometric mean iron absorptions from the typical beans were 1.83% and 1.87%, respectively; the values were 1.03% and 1.01%, respectively, from the highiron, high-zinc beans. The correlation between the intrinsic and extrinsic tracers was The absorption of intrinsically labelled zinc was slightly greater than that of extrinsically labelled zinc: 12.3% versus 13.2%. This difference did not reach statistical significance (p =.17). The correlation between the intrinsic and extrinsic estimates of zinc absorption was not as good as that for iron absorption (0.641). Discussion There was no difference in the absorption of intrinsically and extrinsically labelled iron and zinc in common red beans by young women with low iron stores. The correlation between the intrinsic and extrinsic tags for iron was very high (0.989). The absorption of intrinsically labelled zinc tended to be higher than that of extrinsically labelled zinc, but the difference did not reach statistical significance. The precision of the zinc-absorption method is less than that of the iron-absorption method. That may explain TABLE 3. Composition of beans with normal and high iron and zinc contents Component Normal Fe and Zn High Fe and Zn Phytate 31 µmol/g 33 µmol/g (IP5 + IP6) a Fe 0.9 µmol/g 1.5 µmol/g (50 ppm) (83 ppm) Zn 0.4 µmol/g 0.8 µmol/g (28 ppm) (55 ppm) Ca 34 µmol/g 22 µmol/g (1,380 ppm) (900 ppm) Phytate/Fe molar ratio Phytate/Zn molar ratio a. Phytate is myo-inositolpentaphosphoric acid plus myo-inositolhexaphosphoric acid. why the comparison between the two labels differed more than that of iron. Alternatively, the extrinsic zinc tracer may have readily bound with the excessive amounts of phytate present in the beans, so that slightly less was available for absorption than for endogenous zinc. Since zinc absorption varies widely between and within individuals, the small difference between the intrinsic and extrinsic zinc tags is probably not relevant. Thus, it appears that future field studies of the bioavailability of iron and zinc from common beans can be done by using extrinsic tracers. Implications The test meal of beans only, given as the first meal of the day, was used to measure iron and zinc absorption in this study. This does not reflect the usual conditions for absorption of iron and zinc from a diet, however. There is no evidence that time of day influences absorption from a complete diet [5], but inhibitors such as phytate may have a greater effect on total absorption of iron and zinc when bioavailability is studied from the test food alone rather than from a complete diet. Before introducing high-iron, high-zinc beans in populations with iron and zinc deficiencies, the efficacy of high-iron, high-zinc beans for improving the status of these nutrients must be tested. The bioavailabilities of iron and zinc from the beans were low, about 1.5% and 13%, respectively. Since the subjects had low iron stores and low plasma zinc concentrations initially, higher rates of absorption should have occurred if the elements were available. Possibly, the high phytate content of the beans reduced the solubility of the iron and zinc complexes in the gut and limited the amount that could be absorbed. It has been suggested that iron and zinc bioavailability

6 439 can be improved by supplementing the diet with small amounts of animal tissue or by using food-preparation methods that reduce phytate content. Methods to augment the bioavailability of minerals from cereal crops need to be developed and tested in iron- and zincdeficient subjects who normally consume these foods. References 1. Frossard E, Bucher M, Machler F, Mozafar A, Hurrell R. Potential for increasing the content and bioavailability of Fe, Zn and Ca in plants for human nutrition. J Sci Food Agric 2000;80: King JC, Shames DM, Woodhouse LR. homeostasis in humans. J Nutr 2000;130:1360S 6S. 3. Reddy MB, Hurrell RF, Cook JD. Estimation of nonheme-iron bioavailability from meal composition. Am J Clin Nutr 2000;71: Fairbanks VF. in medicine and nutrition. In: Shils ME, Olson JA, Shike M, eds. Modern nutrition in health and disease. Philadelphia, Pa, USA: Lea and Febiger, 1994: Hurrell RF. Bioavailability of iron. Eur J Clin Nutr 1997;51:S4 S8. 6. King JC, Keen CL.. In: Shils ME, Olson JA, Shike M, eds. Modern nutrition in health and disease. 8th ed. Philadelphia, Pa, USA: Lea and Febiger, 1997: Sandstrom B. Bioavailability of zinc. Eur J Clin Nutr 1997;51:S17 S Jackson MJ. The assessment of bioavailability of micronutrients: introduction. Eur J Clin Nutr 1997;51:S1 S2. 9. Lowe NM, Woodhouse LR, Matel JS, King JC. Comparison of estimates of zinc absorption in humans by using 4 stable isotopic tracer methods and compartmental analysis. Am J Clin Nutr 2000;71: Boza JJ, Fox TE, Eagles J, Wilson PDG, Fairweather-Tait S. The validity of extrinsic stable isotopic labeling for mineral absorption studies in rats. J Nutr 1995;125: Hunt JR, Gallagher SK, Johnson LK, Lykken GI. Highversus low-meat diets: effects on zinc absorption, iron status, and calcium, copper, iron, magnesium, manganese, nitrogen, phosphorus, and zinc balance in postmenopausal women. Am J Clin Nutr 1995;62: Hunt JR, Mullen LM, Lykken GI, Gallagher SK, Nielsen FH. Ascorbic acid: effect on ongoing iron absorption and status in iron-depleted young women. Am J Clin Nutr 1990;51: Hunt JR, Roughead ZK. Nonheme-iron absorption, fecal ferritin excretion, and blood indexes of iron status in women consuming controlled lactovegetarian diets for 8 wk. Am J Clin Nutr 1999;69: Benito P, Miller D. absorption and bioavailability: an updated review. Nutr Res 1998;18: Fitzgerald SL, Gibson RS, de Serrano JQ, Portocarrero L, Vasquez A, de Zepeda E, Lopez-Palacios CY, Thompson LU, Stephen AM, Solomons NW. Trace element intakes and dietary phytate/zn and Ca phytate/zn millimolar ratios of periurban Guatemalan women during the third trimester of pregnancy. Am J Clin Nutr 1993;57: Solomons NW. Competitive interaction of iron and zinc in the diet: consequences for human nutrition. J Nutr 1986;116: Solomons NW, Jacob RA. Studies on the bioavailability of zinc in humans: effects of heme and nonheme iron on the absorption of zinc. Am J Clin Nutr 1981;34: Solomons NW, Jacob RA, Pineda O, Viteri F. Studies on the bioavailability of zinc in man. II. Absorption of zinc from organic and inorganic sources. J Lab Clin Med 1979;94: Solomons NW, Pineda O, Viteri F, Sandstead HH. Studies on the bioavailability of zinc in humans: mechanism of the intestinal interaction of nonheme iron and zinc. J Nutr 1983;113: Fairweather-Tait SJ, Jackson MJ, Fox TE, Wharf SG, Eagles J, Croghan PC. The measurement of exchangeable pools of zinc using the stable isotope 70 Zn. Br J Nutr 1993;70: Miller LV, Hambidge KM, Naake VL, Hong Z, Westcott JL, Fennessey PV. Size of the zinc pools that exchange rapidly with plasma zinc in humans: alternative techniques for measuring and relation to dietary zinc intake. J Nutr 1994;124: Egan CB, Smith FG, Houk RS, Serfass RE. absorption in women: comparison of intrinsic and extrinsic stable-isotope labels. Am J Clin Nutr 1991;53: Fairweather-Tait SJ, Fox TE, Wharf SB, Eagles J, Crews HM, Massey R. Apparent zinc absorption by rats from foods labeled intrinsically and extrinsically with 67 Zn. Br J Nutr 1991;66: Gallaher DD, Johnson PE, Hunt JR, Lykken GI, Marchello MJ. Bioavailability in humans of zinc from beef: intrinsic versus extrinsic labels. Am J Clin Nutr 1988;48: Gislason J, Jones B, Lonnerdal B, Hambraeus L. absorption differs in piglets fed extrinsically and intrinsically 59 Fe-labeled sow s milk. J Nutr 1992;122: Meyer NR, Stuart MA, Weaver CM. Bioavailability of zinc from defatted soy flour, soy hulls and whole eggs as determined by intrinsic and extrinsic labeling techniques. J Nutr 1982;113: Bouis H. Enrichment of food staples through plant breeding: a new strategy for fighting micronutrient malnutrition. Nutr Rev 1999;54: Beebe S, Gonzalez AV, Rengifo J. Research on trace minerals in the common bean. Food Nutr Bull 2000;21: Frenkel EP, McCall MS, Reisch JS, Minton PD. An analysis of methods for the prediction of normal erythrocyte mass. Am J Clin Nutr 1972;58: Viteri FE, Kohaut BA. Improvement of the Eakins and Brown method for measuring 59 Fe and 55 Fe in blood and other iron-containing materials by liquid scintillation counting and sample preparation using microwave digestion and ion-exchange column purification of iron. Anal Biochem 1997;244: