- Top of page
- Research Methods and Procedures
Objective: This study evaluates whether the iron deficiency suggested in children and adolescents with overweight is also present with increasing age.
Research Methods and Procedures: We examined 50 consecutive postmenopausal nondiabetic white women with a BMI ≥30 kg/m2 and 50 non-obese seemingly healthy women as a control group. In addition to the traditional indices of iron status, we measured the soluble transferrin receptor (sTfR) levels, a sensitive and highly quantitative indicator of early iron deficiency not influenced by the acute phase response.
Results: Obese women have higher serum sTfR levels than control subjects [1.38 (range, 0.89 to 2.39) vs. 1.16 mg/dL (range, 0.69 to 2.03 mg/dL); p < 0.001]. However, no difference in ferritin concentration was observed between the groups [70.50 (range, 18 to 219) vs. 69.50 ng/mL (range, 24 to 270 ng/mL); p = not significant]. A positive correlation between BMI and sTfR concentration was detected. On multiple regression analyses, BMI (positively) and ferritin (inversely) were independent predictors accounting for sTfR.
Discussion: These results suggest that a moderate degree of iron deficiency is also present among adult women with obesity. The determination of sTfR is useful in the evaluation of iron status in this condition. Further studies with a greater number of patients are required to investigate the relationship between tissue iron concentrations and obesity.
- Top of page
- Research Methods and Procedures
Iron is a transition metal capable of causing oxidative tissue damage by catalyzing the formation of free radicals (1). Given that cross-sectional studies have shown a link between increased ferritin levels and several metabolic and vascular disorders (2, 3, 4), it could be speculated that iron stores participate in the higher prevalence of the metabolic syndrome in obese patients. Unexpectedly, the Third National Health and Nutrition Examination Survey (5) has shown that overweight children and adolescents are more than two times more likely than those with normal weight to be deficient of iron. Additionally, compared with lean mice, congenitally obese mice showed lower tissue iron concentration (6). Little, if any, information is available on iron stores in adult obese subjects.
Serum ferritin accurately reflects body iron stores in healthy individuals (7). However, serum ferritin is an acute-phase reactant (8), and a chronic, low-grade, inflammatory status has recently been associated with obesity and the etiopathogenesis of type 2 diabetes (9, 10, 11). We have recently communicated that elevated ferritin levels in type 2 diabetes are caused mainly by inflammatory mechanisms rather than by iron overload (12). In the steady state, circulating iron is bound to transferrin and is taken up from the blood by a high-affinity specific transferrin receptor (TfR).1 The synthesis of TfR and the iron storage protein ferritin is regulated reciprocally at the post-transcriptional level according to the cellular iron status (13). Circulating concentrations of TfR [soluble TfR (sTfR)] are proportional to cellular expression of the membrane-associated TfR (14). Therefore, serum sTfR concentration is closely related to cellular iron demands, and, therefore, the higher the ferritin levels the lower the sTfR concentration. Indeed, as sTfR concentration is not influenced by the acute phase response (15), its measurement permits us to determine more accurately the iron stores in adult obese subjects.
The objective of this study was to determine whether overweight nondiabetic postmenopausal women have an increased prevalence of iron deficiency compared with a group of non-overweight controls.
- Top of page
- Research Methods and Procedures
The main clinical features and the iron status of obese women and control groups are presented in Table 1.
Table 1. Main clinical features and laboratory data of subjects included in the study
| ||Obese women (BMI ≥30 kg/m2) n = 50||Non-obese women (BMI <30 kg/m2) n = 50||p|
|BMI (kg/m2)||39.73 ± 7.94||25.00 ± 2.74||<0.001|
|Age (years)||56.92 ± 9.44||57.36 ± 11.06||0.830|
|Ferritin (ng/mL)||70.50 (18–219)||69.50 (24–270)||0.882|
|Iron (μg/dl)||77.52 ± 27.30||83.54 ± 24.77||0.247|
|Transferrin (mg/dl)||265.56 ± 41.55||278.40 ± 49.61||0.159|
|TSI (%)||29.75 ± 10.84||30.80 ± 10.74||0.626|
|Hemoglobin (g/dl)||13.47 ± 0.90||13.29 ± 0.84||0.320|
|Reticulocytes (/1000 hem)||8.3 (2.60–80.00)||9.20 (4.00–56.00)||0.847|
|sTfR (mg/dl)||1.38 (0.89–2.39)||1.16 (0.69–2.03)||<0.001|
|sTfR/log ferritin||0.73 (0.45–1.67)||0.64 (0.36–1.26)||0.009|
|HOMA-IR||3.86 (1.04–12.47)||2.26 (0.44–9.42)||<0.001|
No differences in age, serum ferritin, iron, transferrin, TSI, hemoglobin, and reticulocytes were observed. Obese women had higher serum sTfR levels [1.38 (range: 0.89 to 2.39) vs. 1.16 mg/dL (range: 0.69 to 2.03 mg/dL); p ≤ 0.001] and sTfR/log ferritin ratio [0.73 (range: 0.45 to 1.67) vs. 0.64 (range: 0.36 to 1.26); p = 0.009] than control subjects. As expected, obese women also had higher levels of insulin resistance.
The correlations obtained between BMI and the parameters of iron metabolism are shown in Table 2. A significant positive correlation between BMI and sTfR and sTfR/log ferritin ratio was detected when all subjects were evaluated (n = 100; Figure 1) and persisted when both groups were analyzed separately. A significant positive correlation between BMI and reticulocytes was also present in obese women but disappeared in non-obese subjects. No correlations between sTfR and HOMA-IR or reticulocytes was detected.
Table 2. Correlations of BMI with other laboratory parameters in all non-diabetic postmenopausal subjects, obese women, and non-obese women (control group)
| ||All subjects||Obese women (BMI ≥30 kg/m2)||Non-obese women (BMI <30 kg/m2)|
|Log sTfR-log ferritin||0.352||<0.001||0.271||0.052||0.241||0.092|
Multiple regression analyses showed that BMI (positively) and serum ferritin levels (inversely) were independent predictors accounting for serum sTfR levels (R2 = 0.457, β standardized coefficients = 0.501 [p < 0.001] for BMI; −0.406 [p < 0.001] for ferritin; Table 3).
Table 3. Multiple regression analysis of independent variables associated with sTfR in all subjects
| ||Dependent variable: log sTfR|
|All subjects|| || || || || |
| Log ferritin||−0.159||0.042||−0.406||−3.760||<0.001|
| Iron|| || ||−0.052||−0.464||0.921|
| Age|| || ||−0.014||−0.119||0.911|
| Reticulocytes|| || ||−0.000||0.002||0.874|
| HOMA-IR|| || ||0.105||0.806||0.685|
| Constant||0.223||0.090|| ||2.471||0.018|
|R2 = 0.457|| || || || || |
- Top of page
- Research Methods and Procedures
Non-anemic iron deficiency has been reported in overweight children and adolescents (5, 23, 24, 25). However, the magnitude of this association has not been previously established in adult subjects. In this study, after excluding type 2 diabetic subjects, no differences in serum ferritin levels between obese and non-obese postmenopausal women were detected. Nevertheless, obese subjects showed a significant increase in circulating sTfR concentrations and in the sTfR/log ferritin ratio in comparison with non-obese women. In this regard, it should be noted that most of the biochemical markers for iron metabolism are affected by acute-phase reaction. Thus, a normal or even high concentration of ferritin does not rule out the existence of iron deficiency in patients with inflammatory conditions, such as obesity. Because sTfR circulating levels are not influenced by the acute phase response that occurs in inflammatory processes (15), their measurement is more accurate than serum ferritin to study iron stores in obese patients. Therefore, our results support the idea that non-anemic iron deficiency could be seen in adult obese subjects.
The relationship between iron deficiency and weight status has mainly been investigated in studies with children and adolescents. Among 355 subjects (163 boys and 192 girls), 11 to 19 years of age, Wenzel, Stults, and Mayer (23) observed that serum iron levels of obese boys and girls were significantly lower than those of normal-weight subjects. In a similar way, Seltzer and Mayer (24) studied 321 adolescents, 160 male and 161 female subjects, 11 to 21 years of age, and reported that serum iron levels and TSI were significantly lower in obese male and female adolescents than in lean individuals, without differences in hemoglobin and hematocrit concentrations. More recently, Pinhas-Hamiel et al. (25) examined the prevalence of iron deficiency among 321 children attending two endocrine clinics in Israel. Using only serum iron levels, iron deficiency was noted in 38.8% of obese subjects, 12.1% of overweight subjects, and only 4.4% of the normal-weight group. The possible association between iron deficiency and obesity deserves further study. Kennedy et al. (6) compared the concentration of iron in tissues from genetically obese (ob/ob) mice and their lean littermates. The results showed that chronic obesity was associated not only with a 21% lower plasma iron concentration but also with a significant decrease in liver, bone, and muscle iron concentrations, irrespective of sex and age of the animal.
Few data have been published on the association of ferritin and obesity in adult subjects. In postmenopausal obese women, we detected an independently positive association between both log sTfR and log sTfR/log ferritin ratio and BMI in obese and non-obese individuals. In two previous cross-sectional studies, an association between elevated serum ferritin levels and the presence of abdominal adiposity (4, 26) was reported. However, the results are difficult to interpret because confounding factors, such as type 2 diabetes, alcohol consumption, or hemochromatosis gene mutations, were not excluded from the initial analysis. In addition, the association between serum ferritin and abdominal adiposity observed in postmenopausal women in univariate analyses disappeared when BMI was included in multivariate models (4). Therefore, the high ferritin levels observed in abdominal obesity could better reflect an inflammatory phenomenon as a part of an acute-phase reaction than an increase of the extrahepatic iron stores. In this regard, we recently reported that serum ferritin levels are increased in type 2 diabetic patients in the absence of a reciprocal decrease of sTfR, thus suggesting that elevated ferritin levels in type 2 diabetic patients are caused mainly by inflammatory mechanisms rather than by iron overload (12).
The specific mechanism that could lead to iron deficiency in obesity remains to be elucidated. Polycythemia is common in obese patients (27), and it is important to know that the most important determinant of serum sTfR levels seems to be marrow erythropoietic activity (28). In fact, in univariate analyses, an association between BMI and serum sTfR was observed in obese women, but no correlation between reticulocytes and sTfR was detected. In addition, reticulocytes were not an independent variable related to sTfR in the multivariable analyses. However, virtually all cells, including adipocytes, express TfR on their surface (29). Davids et al. (30) observed that insulin stimulates fat cell iron uptake by a mechanism that may involve the translocation of TfR to the plasma membrane. The effect of insulin increasing TfR has been also observed in a myeloid leukemia cell line, in mouse mammary gland cells, and in rats (31, 32, 33). Therefore, it is tempting to speculate that the higher level in sTfR detected in obese women originated in the stimulation of adipose tissue by hyperinsulinemia. However, the lack of correlations between HOMA-IR and sTfR concentrations observed in obese patients, as well as the fact that HOMA-IR was not an independent variable related to sTfR, argues against this possibility. Diets for weight reduction may have an iron content below the Recommended Daily Allowances (34), but no significant changes in serum iron, ferritin, and TSI were observed in obese children before and after a 13-week treatment with a hypocaloric balanced diet or a 10-week treatment with a protein-sparing modified fast diet (35). Thus, an inadequate diet, with limited intake of iron-rich foods, does not seem to be the only explanation for the iron deficiency in obese people. Physical inactivity could be another factor associated with iron deficiency in obesity. There is an inverse relation between physical activity and weight gain. Myoglobin is a monomeric heme protein found mainly in muscle tissue, where it serves as an intracellular storage site for oxygen. The breakdown of skeletal muscle caused by injury from physical exercise releases myoglobin into the bloodstream (36). Conversely, iron deficiency detected in obesity could be considered a compensative mechanism in which obese individuals try to protect themselves against the damaging effects of oxidative stress catalyzed by iron. It is interesting that a lower prevalence of diabetes has been recorded among frequent blood donors (37), and a decrease in insulin resistance has been documented after iron depletion in type 2 diabetic patients (38). In addition, when provided with an iron-sufficient diet, obese mice absorbed 2 to 2.5 times more radiolabeled iron than did lean mice, but no increase in tissue concentration was shown (39).
In conclusion, we provide evidence that the non-anemic iron deficiency detected in obese children is also present in postmenopausal women without diabetes. Given the increasing numbers of overweight people, if these data are confirmed, screening guidelines for iron deficiency may need to be modified to include subjects with elevated BMI. Because our study is cross-sectional, a causal relationship between weight status and iron deficiency could not be determined. Further studies with a greater number of patients are required to support our results and to investigate the nature and significance of the relationship among serum iron indices, tissue iron concentrations, and obesity.