Copyright 2003 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2003
To measure iron absorption in children from meals containing apple juice or orange juice so as to determine if iron absorption will be greater with orange juice because of its higher ascorbic acid content than apple juice, a noncitrus fruit juice that US children reportedly prefer.
On 2 successive days, children consumed identical meals that included apple juice on one day and orange juice on the other, in random order. The meals were labeled with iron-57 on one day and iron-58 on the other. Iron absorption was measured from red blood cell incorporation of the iron stable isotopes 14 days later.
Nutrition research institute in a major metropolitan medical center.
A total of 25 healthy children, 3 to 6 years of age, were recruited, of whom 21 (11 male and 10 female) completed the study.
Identical meals served with orange juice and apple juice were given on consecutive days, in a balanced randomized design.
Main Outcome Measures
Iron absorption measured by established stable isotope methods.
Median iron absorption from the meal ingested with apple juice was 7.17% (mean ± SD, 9.48% ± 9.68%). Median iron absorption from the meal ingested with orange juice was 7.78% (9.80% ± 6.66%; P = .44). Iron absorption from the meal that included apple juice was significantly correlated with serum ferritin concentration (P = .02); iron absorption from the meal that included orange juice tended to correlate with serum transferrin receptor concentration (P = .051).
As children absorb iron well from a meal that includes either orange or apple juice, a preference for apple juice does not pose a concern with regard to the prospect of iron-deficiency anemia, which remains a significant health problem in the United States.
IRON-DEFICIENCY anemia remains the most prevalent nutritional deficiency in the world1 and affects more than 2 billion people globally.2 In the United States, it is estimated that 8% of children aged 1 to 5 years are iron deficient.3 Iron-deficiency anemia has been associated with lower scores on tests of mental and motor functioning in toddlers4 and with worse school performance in adolescents.5 In humans, iron absorption can be greatly influenced by the presence of enhancers and inhibitors of iron absorption in the rest of the diet. Ascorbic acid can significantly increase iron absorption,6- 9 while phytate and calcium may inhibit it.9- 12
As children mature, their intake of foods and fruit juices becomes more diversified. A recent US Department of Agriculture study found that children are more likely to consume apple juice than orange juice.13 Apple juice contains far less ascorbic acid than orange juice, so this raises the question of whether such a preference adversely affects iron absorption and iron nutrition. Previous studies in children have suggested that ascorbic acid can increase iron absorption from a school breakfast in Peruvian children14 and from a chocolate-flavored milk drink in Jamaican children.15 However, apple and orange juices are complex beverages that differ in many ways. Substitution of one for the other does not, therefore, change solely ascorbic acid intake. Although one study in adults has suggested that iron absorption is lower when apple juice, vs orange juice, accompanies a meal,16 there are few data available on this important nutrition issue in children.
In this study, we compared the effect of apple juice, vs that of orange juice, on iron absorption in children consuming a meal. As apple juice naturally contains less ascorbic acid, we hypothesized that iron absorption would be higher with orange juice.
Twenty-five children, 3 to 6 years of age, were recruited from the greater Houston, Tex, area by public advertisement. A total of 21 children completed the study (11 male and 10 female). Children were considered eligible if they were 3 to 6 years of age, between the 5th and 95th weight-for-height percentiles, had no underlying medical problems, took no medications or vitamin supplements, and would drink both apple juice and orange juice. The Baylor College of Medicine Institutional Review Board approved the protocol, and informed written consent was obtained from the parent or guardian of each child before enrollment.
On the morning of the study, fasted subjects were admitted to the Metabolic Research Unit of the Children's Nutrition Research Center in Houston. Height and weight were measured by standard clinical methods. They were given a meal of toast with jelly or butter, a choice of a noncitrus fruit, and 60 mL of low-pulp, non–calcium-fortified orange juice (Tropicana Products, Inc, Bradenton, Fla; ascorbic acid content, 39 mg/100 mL) or apple juice (Tree Top, Inc, Selah, Wash; ascorbic acid content, 1 mg/100 mL), to which 5 mg of aqueous ferrous sulfate enriched in either iron-57 (57Fe) or iron-58 (58Fe) had been added 12 to 18 hours earlier. Once this portion was consumed, a further 60 mL of the same juice was used to rinse the glass, and the children were encouraged to consume as much of this as possible. The children were then discharged home and required to fast for an additional 2 hours. The following day, the children were given a meal identical to the one consumed the previous day, but with the alternative juice labeled with the other isotope. The order in which the 2 juices were given was randomized, as was the isotope used to label each juice. The contents of the study meals were weighed at the start and end of the meal. Nutrient intake from the meals was calculated by means of the Minnesota Nutritional Data System (version 2.91, Food Database 12A, Nutrition Database 27; University of Minnesota, Minneapolis). Except for the test meals given on the first 2 study days, no other dietary intervention took place.
The children returned to the Metabolic Research Unit 14 days later, when 5 mL of blood was drawn for isotope ratios, a complete blood cell count, and measurement of ferritin and transferrin receptor levels. Iron absorption was measured by a well-described and validated method that used stable isotopes.17,18 Stable isotopes are naturally occurring, nonradioactive isotopes that can be used safely in any patient population.17
Iron-57 (95.82 atom percent) and 58Fe (93.13 atom percent) were purchased (Trace Sciences International Corp, Richmond Hill, Ontario) and converted to aqueous solutions of ferrous sulfate, as described previously.18 The fruit juices were labeled with 5 mg of the 57Fe-enriched tracer or 1.5 mg of the 58Fe-enriched tracer. In the latter case, 3.5 mg of nonenriched ferrous sulfate was added to ensure that the amount of ferrous sulfate added to the 2 juices was the same. The isotopes were drawn up into preweighed syringes and added to the juices 18 to 24 hours before use. The exact amount of isotope given was measured by the change in weight of the syringe.
Blood samples were collected by venipuncture into an EDTA-anticoagulated tube and a plain tube (no anticoagulant). A portion of the EDTA-anticoagulated sample was used to measure a complete blood cell count. The remainder was separated by centrifugation and prepared for iron isotope ratio analysis as previously described.18 Iron isotope ratios were measured by thermal ionization magnetic sector mass spectrometry (Finnigan MAT 261; Thermo Finnigan, Bremen, Germany). Data were expressed as 57Fe/iron-56 (56Fe) and 58Fe/56Fe ratios and corrected for differences in fractionation by means of the ratio of the 2 nonadministered isotopes, iron-54 (54Fe) and 56Fe. Three blocks of 10 scans each were made until the desired degree of precision was obtained (typically relative SD <0.2%). This method is similar to that used for reference analysis and is capable of high precision and accuracy for iron isotope ratio measurement.17
Serum was separated from the plain tube by centrifugation and stored at −20°C pending analysis. Serum ferritin concentration was measured with a solid-phase, 2-site fluoroimmunometric assay (DELFIA method; PerkinElmer Inc, Boston, Mass), and soluble serum transferrin levels were measured with an enzyme-linked immunosorbent assay (Quantikine; R&D Systems, Minneapolis, Minn).
Iron isotope ratios were converted to tracer-tracee ratios (TTR) as described previously.19,20 Iron incorporation into red blood cells was calculated from the following equations18,21:
57Fe Incorporation = 100% × [(57Fe TTR × Fecirc)/Dose of 57Fe-Enriched Tracer];
58Fe Incorporation = 100% × [(58Fe TTR × Fecirc)/Dose of 58Fe-Enriched Tracer],
where Fecirc is the total amount of circulation hemoglobin iron, and is determined by the following formula19:
Fecirc (in milligrams) = Blood Volume × Hemoglobin Concentration × Body Weight × 3.47,
where Blood Volume is 65 mL/kg.
Iron incorporation was converted to fractional iron absorption on the basis of the assumption that 80% of absorbed iron was incorporated into red blood cells within 14 days; therefore,
Iron Absorption = Iron Incorporation/0.8.
The difference between iron absorption from the meal containing apple juice and that from the meal containing orange juice was examined by means of a paired t test. Serum ferritin concentrations were log-transformed (to the base 10) to normalize the distribution. Statistical analysis was carried out with StatView version 5.0.1 for Macintosh (SAS Institute Inc, Cary, NC). Results were considered significant at P = .05. Power calculations were carried out with DSTPLAN version 4.2 (University of Texas M. D. Anderson Cancer Center, Houston). Data for iron absorption are given as both the median and the mean ± SD. The effect of iron status on iron absorption was assessed by simple regression analysis. The proxy markers of iron status used were the hemoglobin concentration, serum ferritin concentration (before and after log transformation), the serum transferrin receptor concentration, and the serum ferritin–transferrin receptor ratio.
We expected an iron absorption of approximately 8% (SD, 4%). Assuming the smallest clinically significant decrease in iron absorption to be 3%, a sample of 20 was required to have an 80% power (β = .20) to detect such a difference at a P<.05 (α = .05). To allow for subject attrition, 25 subjects were recruited.
Twenty-five subjects were recruited, of whom 21 completed the study. An insufficient amount of blood was obtained from 2 subjects on day 14 for analysis. Another 2 subjects did not return for the day 14 visit. Demographic data representing the 21 subjects who completed the study are shown in Table 1. One subject was mildly anemic (hematocrit, 33.4%; hemoglobin level, 11.3 g/dL) and had unexpectedly high iron absorption. Results are expressed, and statistical analyses were carried out, both with and without this subject. The composition of the 2 meals, excluding the juices, was similar. However, when the juices were included, differences in ascorbic acid content and in several other nutrients were seen (Table 2).
Median iron absorption from the meal with apple juice was 7.17% (mean ± SD, 9.48% ± 9.68%). Median iron absorption from the meal with orange juice was 7.78% (9.80% ± 6.66%) and was not statistically different (P = .44 after log transformation; 95% confidence interval for difference, −1.54% to 3.50%). When the single outlier was excluded, results were similar, with a median iron absorption of 6.65% (7.63% ± 4.76%) from the meal with apple juice and 7.55% (9.82% ± 6.84%) from the meal with orange juice (P = .19 after log transformation; confidence interval for difference, −3.92% to 3.29%).
Neither iron absorption from apple juice (P = .50) nor that from orange juice (P = .82) was correlated with hemoglobin concentration. Iron absorption from apple juice was significantly negatively correlated with serum ferritin level (P = .02), logserum ferritin (P = .01), and the ferritin–transferrin receptor ratio (P = .02), but was not correlated with the serum transferrin receptor concentration (P = .99). In contrast, iron absorption from orange juice tended to increase as serum transferrin receptor concentration increased (P = .051) but was not correlated with the serum ferritin concentration (P = .67), logserum ferritin (P = .65), and the ferritin–transferrin receptor ratio (P = .45).
In this study, we used a stable isotope–based method to compare the absorption of iron from a meal accompanied by a non–ascorbic acid–fortified apple juice vs that from a meal accompanied by orange juice containing ascorbic acid. As the orange juice had greater ascorbic acid content than apple juice, we hypothesized that iron absorption would be greater from the meal given with orange juice. Contrary to our expectations, iron absorption was similar in children consuming meals that were accompanied by either apple or orange juice.
These results differ from those of a previous study by Fairweather-Tait et al7 in which a beverage containing ascorbic acid led to a 2-fold increase in iron bioavailability. Similarly, a study by Ballot et al16 showed that juices containing citric acid and ascorbic acid significantly increase iron absorption from a rice meal given to adult multiparous women.16 The different findings in these studies may reflect the age range studied, the amount of ascorbic acid used, or the ratio of ascorbic acid to inhibitors of iron absorption in the meals. In the study by Fairweather-Tait et al, the children were 9 months of age, and 50 mg of ascorbic acid was used. In the Ballot et al study, multiparous women were studied, and 28 to 38 mg of ascorbic acid was used. In our study of 3- to 5-year-old children, total ascorbic acid intake was approximately 35 mg.
The beneficial effect of ascorbic acid on iron absorption may be most apparent in meals with high levels of inhibitors of iron absorption. The largest effect in the study by Fairweather-Tait et al was seen in the meals containing whole-meal bread or beans, both of which have relatively high phytate content. Our meals, which reflected the type of meals eaten by children in the United States, contained relatively low amounts of phytate. Similarly, the effect seen in the study by Davidsson et al15 was seen in chocolate milk; bovine proteins and calcium have been shown to inhibit iron absorption.
In iron-deficient Mexican women, consumption of 50 mg of ascorbic acid per day as agua de limón (limeade) for 2 weeks significantly increased iron absorption.20 However, this population is very different from the one we studied in age, iron status, and diet. Addition of ascorbic acid to complex foods has been shown to increase iron absorption in Peruvian and Jamaican children,14,15 but there are more differences between apple and orange juices than ascorbic acid content.
One could speculate, therefore, that other ingredients present in apple juice may have a beneficial effect on iron absorption and counteract the effect of lower ascorbic acid levels. The apple juice in our study contained approximately 400 mg of malic acid per 100 mL of juice, whereas the orange juice contained negligible amounts. Some studies have shown malic acid to have a beneficial effect on iron absorption in adults.21 Although the effect of malic acid on nonheme iron absorption in children has not been studied, it is possible that malic acid or some other component of the apple juice counteracted the effect of the differing ascorbic acid concentration. Alternatively, the higher levels of zinc and copper in the orange juice meal may have impaired iron absorption and negated any beneficial effect of ascorbic acid. The point remains that substitution of apple juice for orange juice in the diet does not only result in a change in ascorbic acid intake.
Consumption of apple juice is increasing in the United States. A survey conducted by the US Department of Agriculture in 1994 found a large increase in total beverage consumption in the United States, the greatest of which was seen in noncitrus juices, especially apple juice.13 Although a number of studies in developing countries have shown that iron status improves as the intake of ascorbic acid in the diet increases, our study was on children in a developed country with generally good iron status, and we were not able to demonstrate a negative effect of apple juice consumption on iron absorption in this population.
Current dietary recommendations are that children consume 5 servings of fruit, juices, and vegetables daily. However, actual intakes are well below this level. Apple juice is well accepted by children; therefore, it is reassuring that consumption of apple juice, rather than orange juice, does not appear to have an adverse effect on iron absorption from the diet and can be encouraged as a healthy addition to the diets of preschool-aged children. However, it should be remembered that excessive amounts of apple juice can lead to abdominal pain and diarrhea, as its carbohydrate composition can lead to sugar malabsorption.22
Corresponding author: Ian J. Griffin, MB ChB, US Department of Agriculture/Agricultural Research Service Children's Nutrition Research Center, 1100 Bates St, Houston, TX (e-mail: email@example.com).
Accepted for publication May 22, 2003.
This project was financed in part with federal funds from the US Department of Agriculture/Agricultural Research Service under Cooperative Agreement 58-6250-6-001 and by the State of Florida, Department of Citrus, Lakeland.
We thank Keli Hawthorne, Melissa Knox, and Cynthia Edwards and the staff of the Metabolic Research Unit for their assistance with the study; Lily Liang, MS, and Zhensheng Chen, PhD, for sample analysis; and Leslie Loddeke for editorial assistance.
This work is a publication of the US Department of Agriculture/Agricultural Research Service Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, and Texas Children's Hospital, Houston. Contents of this publication do not necessarily reflect the views or policies of the US Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsement by the US government.
Ascorbic acid enhances iron absorption. However, children reportedly prefer noncitrus juices (eg, apple juice), which contain little vitamin C, over citrus juices (eg, orange juice) with high vitamin C content. Previous studies have not established whether apple juice consumption by children reduces iron absorption, and thus increases the risk of pediatric iron deficiency, compared with consumption of orange juice.
This study demonstrates that young children absorb iron well from a meal that includes either apple juice or orange juice. Thus, our findings imply that a child's preference for apple juice over orange juice with a meal will not have an adverse effect on iron absorption or lead to iron-deficiency anemia, a significant health problem that persists worldwide.
Shah M, Griffin IJ, Lifschitz CH, Abrams SA. Effect of Orange and Apple Juices on Iron Absorption in Children. Arch Pediatr Adolesc Med. 2003;157(12):1232-1236. doi:10.1001/archpedi.157.12.1232