Introduction

Nutrition in childhood has a significant impact on lifelong health. Obesity and iron deficiency (ID) are two of the most common nutritional disorders worldwide. ID has been associated with decreased exercise capacity, impaired cognitive function, developmental delays, and behavioral disturbances.^1,2 Obese children are at risk of increased morbidity starting already in childhood or adolescence.³ Interestingly, despite their excessive dietary and caloric intake, obese children and adolescents may be at risk of ID because they tend to consume unbalanced meals, particularly rich in carbohydrates and fat. Both obesity and ID are also more prevalent in lower socioeconomic groups, who consume low-cost fast foods, that are low in essential nutrients and rich in fats, sugars and preservatives.

Screening for iron-deficiency anemia (IDA) by a blood count with indices, is recommended only in infants, because of the critical role of iron in brain development, and in menstruating adolescents.⁴ However, we were impressed by the number of children and adolescents, referred to our obesity clinics, who had evidence of microcytic anemia on complete blood count screen. To determine the prevalence of ID in this patient population, we measured serum iron levels as an early sensitive screen. Low serum iron together with low percent iron saturation precede microcytosis and hypochromia, which in turn leads to IDA. Our findings suggest that obese children and adolescents are commonly iron deficient.

Methods

The study sample consisted of 321 children attending two university-affiliated endocrine clinics in Israel between 1999 and 2001. The mean age was 11.33.6 y (range 3–19 y), and female-to-male ratio was 1.8 : 1. A total of 138 children were overweight (n=33) or obese (n=152) as determined by body mass index (BMI) for age and gender according to the NIH data:⁵ overweight—BMI above 85th percentile, but below 97th percentile; obese—BMI above 97th percentile (1.88 standard deviation score (SDS)). The remaining 136 children were of normal weight (BMI below 85th percentile, <1.036 SDS) attending the endocrine clinics because of precocious or delayed puberty, familial short stature, or hirsutism. Celiac disease was excluded in all these cases.

Iron levels were measured in the morning after an overnight fast using the FerroZine method without depolarization (Roche Diagnostics GmbH, Mannheim, Germany). ID was defined as iron level <8 mol/l (45 mcg/dl), and IDA was defined as ID with hemoglobin level below 2 SDS for age and gender.

Statistics

Data were analyzed with BMDP software.⁶ The following statistical analyses were applied: one- and two-way ANOVA, one-way analysis of covariance, Pearson correlation, stepwise logistic regression; P-values <0.05 were considered statistically significant.

Results

Age (means.d.), gender distribution, and iron levels of the normal-weight (group 1), overweight (group 2), and obese subjects (group 3) are shown in Table 1. Mean serum iron level in the obese children was significantly lower (10.6 mol/l, 59.4 mcg/dl) than that of the overweight children (13.3 mol/l, 74.2 mcg/dl) (P<0.01), and even more significantly lower than that of the normal-weight children (15.8 mol/l, 88.4 mcg/dl) (P<0.001). Adjustment of the findings for age and gender further strengthened these differences.

Table 1 - Iron levels according to BMI SDS corrected for age and gender.

Full table

Iron levels below 8 mol/l (45 mcg/dl) were noted in 38.8% of the obese subjects, 12.1% of the overweight subjects, and only 4.4% of the normal-weight group (P<0.001). The percentage of males and females with iron levels below this cutoff point was identical (21.2%). When the subjects were reclassified into three groups by age (5–10 y, 10–15 y, and >15 y), no differences were noted in the percentage of iron-deficient subjects among groups (21.5, 20.1 and 23.5%, respectively).

Iron levels showed a highly significant negative correlation with BMI SDS (r=-0.44, P<0.001) (Figure 1). For each increase in BMI SDS, there was a concomitant decrease in iron level.

Figure 1.

Body mass index SDS vs. iron levels

Full figure and legend (41K)

On logistic regression analysis with iron <8 mol/l (45 mcg/dl) as the dependent variable, and gender, age and BMI SDS as the independent variables, only BMI SDS was a predictor of ID, with an odds ratio of 1.75 (95% CI 1.48–2.06). Receiver operating characteristics (ROC) plot yielded an area under the polygon of 0.79, indicating that BMI alone was sufficient to predict ID.

Pearson correlations showed that iron levels were not gender-dependent. Age was also not associated with an increased risk of ID in either males or females. On ANOVA, the interaction of all three variables (BMI SDS, age, and gender), only BMI SDS had a significant effect on iron level (P<0.001).

IDA was noted in 26.6% of the subjects with ID: 6.7% were in normal-weight group compared with 58.3% in obese group (P<0.001). As expected, there was a significant positive correlation between iron levels and both delta hemoglobin and delta MCV (r=0.37 and 0.45, respectively, with P<0.001 for both).

Discussion

Obesity is increasing worldwide, reaching epidemic proportions.⁷ Poor nutrition is one of the main reasons. It is also one of the main reasons for ID. Recently, the National Health and Nutrition Examination Survey III⁸ demonstrated that ID without anemia is associated with lower standardized math scores in school-aged children and adolescents. ID without anemia in young women impairs adaptation to aerobic exercise.⁹

In the present study, 4.4% of children in the normal-weight group had low iron levels. This rate is consistent with the estimated prevalence of 3% (1.5–8.7%) of ID in children and adolescents in the USA.¹⁰ However, the high rate of ID in our obese children was surprising, and the precise reason remains unclear. To the best of our knowledge, this is the first report of ID in obese children.

The total amount of iron in the body is determined by its intake, absorption, loss, and storage.¹⁰ In children, risk factors for ID include diet poor in iron, diet poor in enhancers of iron absorption such as vitamin C, and diet rich in inhibitors of iron absorption such as polyphenols (in certain vegetables), tannins (in tea), phytates (in bran), and calcium (in dairy products).¹¹ Dietary intake was not documented in our study and needs further investigation. Moreover, even though the obese patients in the present study were new to our clinic and had not yet started a weight-loss diet, some of them may have tried different unbalanced diets on their own before referral, and had therefore been consuming low iron levels for some time. Even short-term very low-energy diets can cause detectable changes in iron stores.¹²

Iron balance is regulated mainly in the gastrointestinal tract through absorption. As obese children gain weight easily, there is no reason to assume they have a problem with absorption. Furthermore, studies in ob/ob mice have shown that they absorb and retain approximately twice as much iron as wild-type lean mice.¹³ Periods of rapid growth, such as in premature babies, neonates and adolescents,¹⁴ are known to be associated with ID. Similarly, it is possible that obese children have increased iron needs because of their increased growth and body surface area.

Iron loss may be a reason for ID in menstruating adolescents. However, in our study, serum iron levels were low also in males and in prepubertal children. Therefore, iron loss because of menstruation alone is not a sufficient explanation for the observed findings.

ID and IDA are associated with decreased physical endurance and maximal exercise capacity.¹⁵ The extent to which these disorders might affect daily activities that do not involve maximal exercise capacity is unknown. Nevertheless, we may assume that ID may further contribute to the poor exercise capacity of obese children, who already have difficulties performing even mild physical tasks.

The main limitation of our study is the use of serum iron level, which was selected because it is readily available in many biochemical tests. However, infections and inflammations can decrease serum iron concentration, and there may be day-to-day variations within individuals. Other parameters such as transferrin, transferrin receptor, and ferritin should be studied as well in the future.

In summary, our findings show that obese children and adolescents have a higher than normal rate of ID. We believe that causes may be poor diet content, repeated short-term restrictive diets, and increased iron needs. Because of the potentially harmful effects of ID, obese children should be routinely screened and treated as necessary.

References

1. Saloojee H, Pettifor JM. Iron deficiency and impaired child development. Br Med J 2001; 15: 1377–1378.
2. Bhatia D, Seshadri S. Growth performance in anemia and following iron supplementation. Indian Pediatr 1993; 30: 195–200.
3. Barlow ES, Dietz WH. Obesity evaluation and treatment: expert committee recommendations, Vol 102. The Maternal and Child Health Bureau, Health Resources and Services Administration and the Department of Health and Human Services. Pediatrics, 1998. p E29.
4. American Academy of Pediatrics, Committee on Practice and Ambulatory Medicine. Recommendations for preventive pediatric health care. Pediatrics 1995; 96: 373–374.
5. http://www.cdc.gov/nchs/about/major/nhanes/growthcharts/charts.htm
6. Dixon WJ (Chief ed). BMDP statistical software. University of California Press: Berkeley, CA; 1993.
7. Popkin B. Obesity is a worldwide phenomenon. Nutr Rev 1998; 56: 106–114. | PubMed | ISI | ChemPort |
8. Halterman JS, Kaczorowski MJ, Aligne AC, Auigner P, Szilagyij PG. Iron deficiency and cognitive achievement among school-aged children and adolescents in the United States. Pediatrics 2001; 107: 1381–1386. | Article | PubMed |
9. Hinton PS, Giordano C, Brownlie T, Haas JD. Iron supplementation improves endurance after training in iron-depleted, non-anemic women. J Appl Physiol 2000; 888: 1103–1111.
10. Andrews NC. Disorders of iron metabolism. N Engl J Med 1999; 341: 1986–1995. | Article | PubMed | ISI | ChemPort |
11. Conrad ME, Umbreit JN. A concise review: iron absorption—the mucin–mobilferrin–integrin pathway. A competitive pathway for metal absorption. Am J Hematol 1993; 42: 67.
12. Beguin Y, Grek V, Weber G, Sautois B, Paquot N, Pereira M, Scheen A, Lefebvre P, Fillet G. Acute functional iron deficiency in obese subjects during a very-low-energy all-protein diet. Am J Clin Nutr 1997; 66: 75–79. | PubMed | ISI | ChemPort |
13. Failla ML, Kennedy ML, Chen ML. Iron metabolism in genetically obese (ob/ob) mice. J Nutr 1988; 118: 46–51. | PubMed | ISI | ChemPort |
14. Gibson RS, MacDonald AC, Smit-Vanderkooy PD. Serum ferritin and dietary iron parameters in a sample of Canadian preschool children. J Can Dietetic Assoc 1988; 49: 23.
15. Haas JD, Brownlie T. Iron deficiency and reduced work capacity: a critical review of the research to determine a causal relationship. J Nutr 2001; 131: 676S–688S. | PubMed |