University of California, Berkeley, California, USA.

SUMMARY.

The absorption of a commercial brand of small-particle reduced iron was evaluated in 10 normal subjects. For each subject, the hemoglobin incorporation method was used to measure the true absorption of 60 mg of iron from either ferrous sulfate or ferric ammonium citrate. The iron tolerance test (ITT) was also studied for these two compounds and for reduced iron. This procedure consisted of measuring the area under the curve of plasma iron elevations at specified times for 6 hours, or the peak plasma iron, corrected by the plasma iron disappearance rate obtained from measuring plasma iron at specified times for 4 hours after the slow intravenous injection of 0.4 mg of iron as ferric citrate. Only the ITT was used to measure the absorption of 60 mg of reduced iron. Reference dose iron ascorbate absorption was measured in each subject. The absorption of ferric ammonium citrate and reduced iron was expressed as percent of dose and also as absorption percent of that of ferrous sulfate. Mean % geometric "true absorptions" were 39.0 for reference dose, 10.4 for FeSO4 and 2.4 for ferric ammonium citrate. The later was 23% that of FeSO4. By ITT the mean geometric % absorptions were 7.9, 3.7 and 3.2 for FeSO4, ferric ammonium citrate and reduced iron respectively, or 47 and 41% of that of FeSO4. We propose that the true absorption of the commercial brand of reduced iron tested was 20% that of FeSO4 based on the relation between the ITT results of reduced iron and the ITT and true absorption values of ferric ammonium citrate in relation to FeSO4. The use of this method for measuring absorption of unlabeled iron compounds is discussed.

Key words: Reduced iron, relative absorption, iron absorption methodology, unlabeled iron compounds.

Absorción de hierro reducido comercial no-marcado y de partícula pequeña. Un método para predecir la absorción de compuestos no-marcados de hierro en humanos. La absorción de una forma comercial de hierro reducido de partícula pequeña se evaluó en 10 sujetos normales. La incorporación de hierro radioactivo en la hemoglobina se utilizó en cada sujeto para medir la "absorción verdadera" de 60 mg de hierro proveniente de sulfato ferroso o de citrato férrico amoniacal. La prueba de tolerancia de hierro (ITT) se usó también en cada sujeto para evaluar la absorción de hierro de estos dos compuestos y del hierro reducido no-marcado. Esta prueba consiste en medir el área bajo la curva de hierro plasmático obtenida por mediciones periódicas por 6 horas subsiguientes a la administración oral de los compuestos de hierro o por el valor pico de hierro plasmático corregidos por la constante de desaparición plasmática del hierro obtenida por mediciones seriadas de hierro plasmático por 4 horas después de la administración endovenosa lenta de 0.4 mg de hierro como citrato férrico. Para medir la absorción de 60 mg de hierro reducido se utilizó unicamente la prueba ITT. Se midió también la absorción de una dosis oral de referencia ( 3 mg de ascorbato de hierro) en cada sujeto, por incorporación en la hemoglobina. La absorción del citrato férrico amoniacal y la del hierro reducido se expresaron como porcentaje de la dosis administrada y como porcentaje de la absorción de la misma dosis de sulfato ferroso. El promedio geométrico de la "absorción verdadera" de la dosis de referencia fue de 39.0%, la del sulfato ferroso fue de 10.4% y la del citrato férrico amoniacal fue de 2.4%. Esta última fue 23% la absorción del sulfato ferroso. Por la prueba ITT los promedios geométricos de las absorciones fueron 7.9, 3.7 y 3.2% respectivamente para sulfato ferroso, citrato férrico amoniacal y hierro reducido, o sea 47 y 41% de la absorción del hierro de sulfato ferroso por esta prueba. Proponemos que la "absorción verdadera" de la fuente comercial de hierro reducido estudiada fue de 20 % la absorción del hierro del sulfato ferroso, basados en la relación entre los resultados de la prueba ITT para el hierro reducido y las del ITT y la "absorción verdadera" para el citrato férrico amoniacal, en relación a ambos resultados para el sulfato ferroso. Se discute el uso de este método para medir la absorción de compuestos de hierro no-marcados.

Palabras clave: Hierro reducido, absorción relativa, metodología de absorción de hierro, compuestos de hierro no-marcados.

Recibido: 09-02-2001 Aceptado: 10-07-2001

INTRODUCTION

Food fortification with iron has increased significantly in the last three decades. The utilization of reduced iron for that purpose in the United States has increased from 72 tons in 1970 to 10,124 tons in 1987. Reduced iron constituted 13.2% and 96.7% of the total amount of iron used in fortification, respectively (1). The use of reduced iron as a fortificant of cereals has continued to increase, but the total amounts are not available (personal communication P. Whittaker and L. A. Mejia, 2000). The absorption of fortification iron is of special interest in order to determine what can be expected in terms of iron nutrition in populations consuming different amounts of iron fortified products using various compounds.

A variety of iron products are used for food fortification purposes, including ferrous sulfate, ferrous lactate, ferrous gluconate, ferric ammonium citrate, ferrous fumarate, ferrous succinate, ferric saccharate, ferric orthophosphate, ferric ammonium orthophosphate, ferric pyrophosphate, EDTA iron, hemoglobin, and elemental iron (electrolytic and carbonyl iron).

The total amount of these iron sources in fortification programs is much less than that of reduced iron (2).

In addition, there is the concept of preventive supplementation of vulnerable groups, and inexpensive and well-tolerated sources of iron could be used for this purpose. Viteri, Ali and Tujague (3) have demonstrated that iron reserves can be progressively increased in childbearing-age women by the administration of 30 weekly iron doses of 60 mg as ferrous sulfate consumed in an interval of 7 months (a total of 30 doses). Initially, 16% had low hemoglobin levels, but by three months of weekly supplementation no woman had hemoglobin (Hb) levels below 125 g/L. The general idea of preventive supplementation is the safe, long-term intake of a weekly iron-folate (and possibly vitamin A and zinc) supplement by vulnerable groups who can not benefit sufficiently from adequate diets, fortified or not (4).

Studies on the absorption of reduced iron have rendered very dissimilar results in relation to the absorption of ferrous sulfate. These results range from 13 up to 148%. The variability in the absorption of different reduced iron products and of other iron compounds was shown by Callender over 30 years ago (5, 6). The variability of results is related to the particle size, surface area, and porosity of reduced iron, which depends on the manufacturing process (7).

The absorption of commercial reduced iron has been difficult because of the inexistence of commercial reduced iron in an isotopic form. Bjorn-Rasmussen and Hallberg (7) as well as Roe and Fairweather-Tait (8) have produced labeled reduced iron in their laboratories trying to mimic the commercial iron sources. The first authors, however, were clear in pointing out the difficulties in predicting the behavior of commercial reduced iron produced in the laboratory on a micro-scale. Moreover, the majority of the studies on reduced iron absorption have been made adding this type of iron to different foods making it difficult to establish its true availability in relation to that of ferrous sulfate (please see Table 6).

The purpose of the present study was to compare the absorption of an unlabeled commercial source of reduced iron administered alone (without food) at doses compatible with supplementation (60 mg) with those of two other sources of iron whose absorption rates are known: ferrous sulfate and ferric ammonium citrate. Grebe et al. (9) have shown that ferric ammonium citrate is absorbed 22% in relation to ferrous sulfate when both were administered without food. We used the "iron tolerance test" (ITT) to compare the iron absorption from the three compounds and the double radioisotopic absorption of ferrous sulfate and ferric ammonium citrate by hemoglobin incorporation of radiolabeled iron. The absorption of a radiolabeled reference dose of iron ascorbate was also measured in order to adjust the absorptions of the compounds as is customary (10). For comparison purposes, the absorption of ferrous sulfate by either method was considered 100% and that of the other compounds as a proportion of this absorption.

The source of the commercial reduced iron was a reputable company from which we purchased it without the company being aware of the purpose of such purchase to insure the representativeness of the product.

The ITT measured by integrating the area under the serum iron elevation after an oral dose of the compound, unadjusted or adjusted by the iron disappearance constant after an intravenous iron dose, has been used to measure iron absorption before the development of isotopic methods to determine the "true absorption" or "retention of iron". The true absorption has been measured by the incorporation of radioisotopic iron to circulating hemoglobin; retention has been measured by whole body counting (11). Many of the results from the ITT are considered imprecise given that the individual values obtained correlate poorly with the other two methods (12). However, the mean and ranges of absorption in groups of subjects obtained by the ITT are acceptable when compared to the other methods if adjustments for disappearance rates are used and the absorptions are reported as percent of the ferrous sulfate standard (11, 13-15).

On the other hand, Heinrich (13) and Heinrich and Fischer (14) validated the results of the tolerance test by comparing them to those of whole body counting as a reference method. They, as well as Ekenved (11) were able to obtain correlation coefficients in the order of 0.88 between whole body radioactivity and integrated areas under the serum iron curve or with the peak increments, adjusted or not by disappearance rates, when different iron doses were given to the same subjects.

The utilization of both the ITT and the radioiron incorporation into Hb of a high and a low bioavailability iron compounds (ferrous sulfate and ferric ammonium citrate, respectively) and the ITT for estimating the absorption of non-radiolabeled reduced iron in relation to the previous compounds, constitutes a new approach to estimating the "true absorption" of reduced iron.

MATERIALS AND METHODS

Healthy adult volunteers between the ages of 18 and 65 years who responded to a public call in the Berkeley, California community were invited to participate in the study. All the procedures from recruiting to the performance of the study were approved by the Committee of the Protection of Human Subjects (CPHS) as well as by the Radiation Safety Committee of the University of California at Berkeley.

Exclusion criteria included pregnancy or the possibility of becoming pregnant within two years, breast feeding, metrorrhagia, gastrointestinal symptoms, suspicions of intestinal malabsorption, chronic inflammatory conditions, current or previous use of therapeutic iron in the last six months, recent blood donation and probability of not being able to finish the full study.

Twenty volunteers entered the study and only ten completed it. The volunteers arrived at the laboratory for screening procedures, signature of informed consent and for scheduling of activities to be undertaken at the penthouse of Morgan Hall at the university campus. Their health status was evaluated and for menstruating women the characteristics of their cycles and birth control methods. They were instructed to fill out a diet history questionnaire and to maintain their usual diet during the whole study period, not to take any supplements except those administered by us and to keep a record of foods consumed the day prior to each test.

The reduced iron administered was hydrogen-reduced 99.8 purity, produced initially by the Domfer process utilizing high pressure water jets (16). Its absorption was estimated by the ITT. This test consisted of measuring the area under the plasma iron curve obtained by repeated venous blood sampling for 6 hours after the oral ingestion of 60 mg of reduced iron in a gelatin capsule accompanied by 125 ml of water. The procedure was performed after an overnight fast. The area under the plasma iron curve was adjusted by the serum iron disappearance rate obtained by the slow intravenous administration of 0.4 mg of iron citrate in a peripheral vein of one arm. Repeated blood samples were obtained from a catheter inserted in a peripheral vein in the other arm. The identical ITT procedure was also performed for ferrous sulfate and ferric ammonium citrate. However, these last two compounds were also labeled with either 55Fe or 59Fe. Iron absorption from these two compounds was also measured by their radioactive incorporation into hemoglobin two weeks later. The absorption of a reference dose of 3 mg of iron ascorbate (15 mg of FeSO4-7H2O + 25 mg Ascorbic acid) administered as a solution in 250 ml of orange juice, was also measured by its incorporation into hemoglobin. In essence, all subjects underwent the 5 following procedures: an intravenous administration of iron citrate; the oral ingestion of the three compounds being tested (ferrous sulfate, ferric ammonium citrate, and commercial reduced iron), and the oral ingestion of the reference iron ascorbate dose. The sequence of these five events was randomly assigned. In half the subjects, three of these events took place at the start of the study and the other two followed two weeks later, after blood was obtained to measure radiolabeled iron incorporation into Hb. In the other half two events took place initially and the other three two weeks later. Each event was separated from the previous one by two days. When two radioactively labeled compounds were administered in sequence, one was labeled with 55Fe and the other with 59Fe. Only reduced iron was unlabeled (Table 1). The radioactivity doses administered were 0.3-0.4 µCi (11.1 - 14.8 mBq) intravenously and 1.0 µCi (37 mBq) of 59Fe orally, or 0.7 µCi (25.9mBq) intravenously and 2.0 µCi (74 mBq) orally of 55Fe. Serial 5 ml blood samples were obtained 10, 15, 30, 45, 60, 90, 120 and 240 minutes after the end of the intravenous infusion, and at 0, 30, 60, 90, 45, 120, 180, 240, 300 and 360 minutes after the oral ingestion of iron.

The blood sampling catheter was kept open between samples by the infusion of 1 ml of saline with 5 u of heparin/100 ml through a 3-way valve. The contents of the valve and the catheter were aspirated before blood sampling for each point in the ITT until about 0.5 ml of blood was recovered. This aspirate was discarded and the 5 ml study sample was subsequently obtained.

A total of 5 doses of 60 mg of ferrous sulfate were also administered to the subjects in order to replace the iron drawn in the five sessions where 55 ml of blood were drawn. Single doses were to be ingested at home by the subjects at the end of the test day, that is at least two days prior to any test.

The following biochemical determinations were performed in each basal sample (time 0): blood hemoglobin, serum iron, total iron binding capacity (TIBC), and plasma ferritin. Percent saturation of TIBC was calculated. Blood hemoglobin was also measured in every sample in order to correct for possible dilution by adjusting all of the values to the basal hemoglobin level.

Blood hemoglobin was measured by the HemoCue System (HemoCue Inc., Angelholm, Sweden). Plasma ferritin was measured by the Spectro Ferritin MT Kits (Ramco Laboratories, Houston, Texas) standardized against WHO standards. Plasma iron and TIBC were measured by the methods published by INACG (17). Iron absorption by incorporation into hemoglobin was measured by liquid scintillation counting (18)

Schematic presentation of the schedule of a subject in one groupÄ