Vitamin D, a-fat soluble vitamin, represents vitamin D2 (ergocalciferol) and vitamin D3 (cholecalciferol). Vitamin D3 is found in some natural foods including fish, fish oil and egg yolk, or may be fortified in certain foods such as milk, margarine, butter and infant formula. On the other hand, vitamin D can also be made in skin through solar UV-B (wavelengths of 290–315 nm) irradiation of 7-dehydrocholesterol(Reference Holick, Chen and Lu1). The classical function of vitamin D is to maintain Ca and phosphate homeostasis, thereby promoting bone mineralization(Reference DeLuca2). In addition, vitamin D plays a role in cellular growth and regulation in many organs(Reference Bikle3).
Vitamin D deficiency causes rickets in children and osteomalacia in adults. It also precipitates and exacerbates osteopenia, osteoporosis and fractures in adults.
Long-term vitamin D insufficiency is associated with increased risks of several chronic diseases including cancers, hypertension, heart diseases, diabetes and multiple sclerosis(Reference Kulie, Groff and Redmer4, Reference Holick5). A high prevalence of vitamin D deficiency has been reported in infants, children and adolescents from different countries around the world(Reference Gordon, Feldman and Sinclair6–Reference Ward, Gaboury and Ladhani9). Vitamin D deficiency might be attributed to low sunlight exposure, dark skin and breast-feeding without vitamin D supplementation(Reference Pettifor10).
A high degree of association between Fe-deficiency anaemia and vitamin D deficiency in Asian children has been reported, with Fe-deficiency anaemia found to be a significant risk factor for low vitamin D levels in children(Reference Lawson and Thomas11, Reference Grindulis, Scott and Belton12). In another study, thirteen out of seventeen infants with Fe-deficiency anaemia had serum 25-hydroxyvitamin D (25(OH)D) concentration below the normal range despite the fact that these infants received 10 μg vitamin D/d from the age of 1 month. Infants’ Hb level, serum Fe as well as 25(OH)D concentrations were increased after they were treated with intramuscular iron dextran(Reference Heldenberg, Tenenbaum and Weisman13). Studies on patients with chronic kidney disease have also supported this association(Reference Albitar, Genin and Fen-Chong14, Reference Patel, Gutiérrez and Andress15).
Several mechanisms for such an association have been proposed. Vitamin D may influence Hb through a direct effect on erythropoiesis where it has a synergistic action with erythropoietin; it also increases the storage and retention of Fe and reduces pro-inflammatory cytokines(Reference Albitar, Genin and Fen-Chong14–Reference Meguro, Tomita and Katsuki18). Thus, vitamin D deficiency reduces the ability of red blood cells to become active. On the other hand, it is known that Fe deficiency impairs fat and vitamin A intestinal absorption. Therefore, it is suggested that absorption of vitamin D may also be impaired(Reference Grindulis, Scott and Belton12). Accordingly, it is still controversial which deficiency causes the other; however it is important to be aware of such association especially when proposing treatment.
There are numerous natural sources of vitamin D for nurslings. These include primarily the stores they developed parentally (for newborns), the vitamin D they produce with exposure of their skin to sunlight, and to lesser extent from human milk(Reference Makin, Seamark and Trafford19, Reference Ala-Houhala20).
Although human breast milk has only a small amount of Fe, anaemia is uncommon in the breast-fed baby. Healthy, full-term babies have sufficient Fe stores in their bodies to last for at least the first 6 months(Reference Ziegler, Nelson and Jeter21, Reference Dewey and Chaparro22). The Fe in breast milk is better absorbed than that from other sources; vitamin C and high lactose levels in breast milk aid in Fe absorption. Fe in mothers’ milk is absorbed five times more than that from cows’ milk or Fe-fortified formula(Reference Ziegler, Nelson and Jeter21–Reference Hicks, Zavaleta and Chen24).
Vitamin D deficiency and insufficiency are common among children who live in northern Jordan (28·0 % and 28·4 %, respectively)(Reference Abdul-Razzak, Ajlony and Khoursheed25). A significant association between infant feeding practices and vitamin D status was found. Infants who were exclusively breast-fed had higher risk for vitamin D deficiency and insufficiency than those who were bottle-fed(Reference Abdul-Razzak, Ajlony and Khoursheed25). On the other hand, WHO estimates that an average of 52·7 % of Jordanian infants (aged 0·50–0·99 years) and 40·76 % of toddlers (aged 1·00–2·99 years) are anaemic(26).
The present study investigated the existence of a correlation between vitamin D status and anaemia in healthy infants and toddlers living in northern Jordan. The findings of the study are intended to improve treatment protocols for both conditions, and also to attain best understanding of the ideal feeding pattern for children in an attempt to prevent such nutrition-related problems, because inadequate vitamin D status and Fe-deficiency anaemia are considered as major public health problems in Jordan(Reference Faqih, Kakish and Izzat27).
Materials and methods
Study participants
The final sample consisted of 203 infants and toddlers between the age of 6 and 36 months who were seen between October 2008 and January 2009 for primary care at Princess Rahma Teaching Hospital. This hospital is the main public paediatrics hospital in the city of Irbid, which is located about 85 km north of the capital Amman. The study was approved by the Institutional Review Board at Jordan University of Science and Technology. All parents approved their infants’ and toddlers’ participation in the study and signed a consent form. Infants with disease that affects the level of vitamin D, such as gastrointestinal, liver or renal diseases, or who consumed vitamin D or Fe supplements were excluded from the study.
One of the parents completed a self-guided questionnaire which included questions on the type of feeding, the duration of breast-feeding, use of supplements and whether the infant has chronic diseases. A detailed description of field data collection can be found elsewhere(Reference Abdul-Razzak, Ajlony and Khoursheed25).
Laboratory measurements
About 5 ml of venous blood was collected in a heparinzed test-tube from each infant/toddler who participated in the study. Plasma was assayed for 25(OH)D using the enzyme immunoassay method (Immunodiagnostic Systems, UK).
Plasma vitamin D levels were defined as follows: vitamin D deficiency, <50 nmol/l (<20 ng/ml); vitamin D insufficiency, <75 nmol/l (<30 ng/ml); and vitamin D sufficiency, ≥75 nmol/l (≥30 ng/ml). The reference range was provided by the manufacturer of the assay as previously published(Reference Robert, Susan and Cecilia28). Anaemia was defined as Hb < 11 g/100 ml blood using WHO criteria for age 0·50–4·99 years(29). The participants were divided into two groups according to age: infants aged 6–12 months and toddlers aged >12–36 months.
Statistical analyses
Data were analysed using the SPSS statistical software package version 16·0 (SPSS Inc., Chicago, IL, USA). The χ 2 test was performed to test for difference in anaemia prevalence between participants who had vitamin D deficiency, insufficiency and sufficiency, as well as by feeding pattern. One-way ANOVA followed by the Duncan post hoc test was performed to compare the means of continuous variables according to vitamin D status. Findings with P < 0·05 were considered to be statistically significant.
Results
Characteristics of the study population
The majority of the study population were males (60·6 %); more than half of them were infants (54·2 %). The majority of the study population had vitamin D sufficiency (normal level), followed by vitamin D deficiency then vitamin D insufficiency. The prevalence of anaemia among the study population was 40·4 %. No statistically significant association between vitamin D status and Hb level was found (Table 1).
Characteristics of the study population: healthy infants and toddlers aged 6–36 months, Irbid, Jordan, October 2008–January 2009
ANOVA was performed to test the differences between variables; P < 0·05 was considered as statistically significant.
*Anaemia was defined as Hb < 11 g/100 ml blood.
Hb level among the study population (total) and stratified by age group
Mean Hb level was 10·81 (sd 1·52) g/100 ml blood for infants and 11·47 (sd 0·98) g/100 ml blood for toddlers. About half of the infants were anaemic, while the vast majority of toddlers were not anaemic; most of the anaemic infants had sufficient vitamin D levels while most of the anaemic toddlers had sufficient and deficient vitamin D levels (Table 2). No statistically significant association between vitamin D status and Hb level was found in either age category (Table 3).
The prevalence of anaemia stratified by vitamin D status among healthy infants (6–12 months) and toddlers (>12–36 months), Irbid, Jordan, October 2008–January 2009
The χ 2 test was performed to test the differences between variables; P < 0·05 was considered as statistically significant.
*Anaemia was defined as Hb < 11 g/100 ml blood.
The association of vitamin D status and Hb level stratified by age group among healthy infants (6–12 months) and toddlers (>12–36 months), Irbid, Jordan, October 2008–January 2009
ANOVA was performed to test the differences between variables; P < 0·05 was considered as statistically significant.
Hb level stratified by type of feeding
Although the average Hb level in the present study tended to be higher in bottle-fed than in breast-fed participants, this increment in Hb level was not statistically significant (Table 4). Similarly, the prevalence of anaemia was higher in breast-fed participants but the difference was also not statistically significant (Table 4).
Anaemia prevalence stratified by type of feeding among healthy infants (6–12 months) and toddlers (>12–36 months), Irbid, Jordan, October 2008–January 2009
The χ 2 test was performed to test the differences between variables; P < 0·05 was considered as statistically significant.
*Anaemia was defined as Hb < 11 g/100 ml blood.
Discussion
The study findings indicate the presence of high anaemia prevalence among our Jordanian infants and toddlers, particularly infants, but there was no statistically significant association between vitamin D status and Hb level. The prevalence of vitamin D deficiency was lower than that of anaemia (29·6 % v. 40·4 %).
Our findings are in agreement with Özsoylu and Aytekin, who could not find a relationship between anaemia and rickets due to vitamin D deficiency(Reference Özsoylu and Aytekin30), but disagree with Yoon et al.(Reference Yoon, Park and Seo31) and Grindulis et al.(Reference Grindulis, Scott and Belton12) who found a significant association of Fe-deficiency anaemia and low vitamin D concentration among children <2 years of age. However no mechanism for this association has been proposed. An association between vitamin D deficiency and a greater risk of anaemia and lower mean Hb was demonstrated in adults(Reference Sim, Lac and Liu16).
In contrast to Yoon et al.(Reference Yoon, Park and Seo31) and Grindulis et al.(Reference Grindulis, Scott and Belton12), Heldenberg et al.(Reference Heldenberg, Tenenbaum and Weisman13) found that severe Fe-deficiency anaemia in infants aged 1 to 12 months decreased plasma vitamin D concentrations; a finding which was attributed to reduced vitamin D intestinal absorption caused by Fe deficiency. However in a study in rats, Katsumata et al.(Reference Katsumata, Katsumata-Tsuboi and Uehara32) found that severe Fe-deficiency anaemia can affect both bone formation and bone resorption by affecting the synthesis and metabolism of Fe-dependent enzymes, prolyl and lysyl hydroxylases in collagen synthesis and renal 25-hydroxyvitamin D 1-hydroxylase, resulting in the synthesis of abnormal collagen and reduced plasma 1,25-dihydroxycholecalciferol concentration, thus reduced bone mineral density.
In our study we found no correlation between anaemia and early feeding practices; however infants who were exclusively breast-fed had a higher risk for anaemia. This finding is in agreement with Yoon et al.(Reference Yoon, Park and Seo31) who found most of the anaemic children <2 years of age were breast-fed.
The prevalence of anaemia among our Jordanian infants and toddlers aged 6–36 months was found to be 40·4 %; this is lower than the previously published worldwide anaemia prevalence, which covered 48·8 % of the global population of children of pre-school age, of 47·4 (95 % CI 45·7, 49·1) %(Reference McLean, Cogswell and Egli33). In our study the anaemia prevalence in infants 1 year and younger, and among the remaining children older than 1 year, was 51·8 % and 26·9 %, respectively. Previously published WHO results showed relatively similar prevalence of anaemia among Jordanian infants (2003 v. 2008) but higher prevalence of anaemia among Jordanian toddlers(26).
The decrease in anaemia prevalence after the age of 1 year may be attributed to several factors. First is the introduction of solid foods and the decreased dependence on milk as the main diet. This is supported by an observational cohort study which demonstrated that infants weaned at <6 months of life had, regardless of milk feeding, higher Hb and ferritin levels than those weaned at >6 months(Reference Capozzi, Russo and Bertocco34). Second is the high prevalence of anaemia among Jordanian pregnant women who live in the north(Reference Albsoul-Younes, Al-Ramahi and Al-Safi35). Significantly high incidence of anaemia was found among infants born to anaemic mothers in Jordan(Reference Kilbride, Baker and Parapia36). This explains the lower prevalence and risk of anaemia in bottle-fed infants compared with those exclusively breast-fed and breast-/bottle-fed. This finding is supported by many previous studies which have demonstrated that breast-feeding is not by itself a causal factor for anaemia(Reference Ziegler, Nelson and Jeter21, Reference Saarinen, Siimes and Dallman23, Reference Hicks, Zavaleta and Chen24). Therefore it would be important to explore the relationship between anaemia during pregnancy and the development of anaemia among Jordanian infants.
Acknowledgements
This research was supported by a grant (168/2008) from the Deanship of Research, Jordan University of Science and Technology. The authors have no conflicts of interest to declare. K.K.A.-R. designed and supervised the work, applied for funding, wrote part of the manuscript and edited the manuscript; A.M.K. conducted the work at Princess Rahma Teaching Hospital; S.M.A. wrote part of the manuscript; B.A.O. performed data analysis; and M.-J.A.A. supervised the work at Princess Rahma Teaching Hospital.
