Iron deficiency screening in the first three years of life: A three-decade-long retrospective case study

Abstract

A three-decade-long retrospective study of iron status in a cohort of 1250 children aged 8–36 months was carried out at the Pediatric Department at the Second University Naples. Iron status was evaluated with independent variables such as family income, weight for height, introduction of cow's milk (CM), iron supplementation and weaning. Iron deficiency (ID) is prevalent in children with low income, early introduction of CM, delayed weaning, over-weight, and in those not receiving iron supplementation (P < 0.05). The first decade (1980–1990) was marked by low family income, while early introduction of CM characterized the first two decades (1980–1990, 1990–2000) (P < 0.05). ID depends on a variety of social and dietary factors. Hematological tests should be performed early to identify children at risk for ID.

Keywords:

• ID
• Cow's milk
• Income
• Weaning
• Growth

Introduction

Iron deficiency affects more than two billion people worldwide, particularly children and young females.^1,2

It represents a public health problem in both developing and industrialized countries.^1,2

The first 3 years of life are an important time in child development. Furthermore, childhood iron deficiency is also associated with behavioral and cognitive delays and increased morbidity rates due to impaired immunity.^3,4

The long-term effects and high prevalence of iron deficiency signal the need for early prevention starting in the first months of life.

Correct diet is important to avoid iron deficiency. Dietary risk factors include prolonged exclusive breastfeeding beyond 6 months without supplementation of iron fortified foods or vitamins with iron, early introduction of cow's milk (CM) before 12 months associated with an excessive consumption.^5–8

It has been reported that in children belonging to low-income families, the prevalence of iron deficiency is higher.^5–8

Furthermore, in Europe the prevalence of iron depleted stores (iron deficiency-ID) may vary from 2 to 48% and iron deficiency anemia (IDA) from 2 to 4%. In Italy the prevalence of ID has been estimated to be at about 24–26%.^9,10

Moreover, the epidemiological evidence of ID could be underestimated because ID without Hb reduction occurs in the early stages and represents the most widely observed form.

Positive effects achieved by introducing iron fortified baby foods, particularly in children from populations with more anemia as well as in children from low-income groups led to recommend its use worldwide. It has recently been pointed out that iron fortified foods should not be used without first assessing the real benefits, because of possible risks in iron sufficient infants and children.

Adverse effects on growth and morbidity have been reported.^6,11

The aim of this study is to examine the influence of some independent variables such as family income, weight for height status, early CM introduction, weaning, iron supplementation, age, and gender on iron status. The analysis was conducted with retrospective clinical records from outpatients at the age of 8–36 months followed at the Pediatric Department – Second University of Naples – Italy in the last 30 years (1980–2010). The prevalence of these variables in the three decades was also evaluated.

Materials and methods

The medical records of 1600 outpatients from 1980 to 2010 were reviewed in this retrospective cohort study at the Pediatric Department of the Second University – Naples, Italy. Of these, 1250 children of both sexes (age range 8–36 months) were selected by the following eligibility criteria: gestational age ≥37 weeks, birth weight ≥2500 g, availability of dietary history since the first months of life, some hematological parameters (Hb, MCV, serum ferritin, and iron-binding capacity), no associated malabsorption or blood loss, hemoglobinopathies, or other chronic diseases. In children with family history of thalassemia or other hemoglobinopathies or with reduced Hb and MCV values but normal serum ferritin, Hb electrophoresis was performed.

Patients diagnosed with chronic immunologic or infective diseases and high C-reactive protein (CRP) values were excluded from the study because of possible interaction with iron metabolism, falsifying the results.

Children were segregated into three groups based on decades: 1980/1990 (350 children), 1990/2000 (520 children), and 2000/2010 (380 children).

Some independent variables influencing iron status were identified.

Family income was evaluated as low (annual income lower than €15 000) or medium-high according to household food security.

Anthropometric measurements (weight, length, and height) were performed. Weight for height was evaluated in the first 2 years of life, while body mass index (BMI) was considered over 2 years using age-specific and gender-specific percentiles.

Age and gender were also evaluated.

Introduction of CM before the first year and the start of weaning were analyzed.

The intake of oral iron supplementation with ferrous sulfate or vitamins with iron preparations and in the last decade iron fortified foods was also evaluated.

On the basis of hematological parameters ID or IDA were diagnosed. Hb and MCV values were compared with the percentile values for age and sex. Serum ferritin values ≥15 ng/ml and iron-binding capacity ≥15% were considered normal. We defined IDA as Hb and MCV values lower than the third percentile for age and sex, serum ferritin <15 ng/ml, and iron-binding capacity <15%. ID was defined as serum ferritin level <15 ng/ml.

Comparison of independent variables among the three groups was performed.

Statistical analysis was evaluated by χ² test. Significance was set at P ≤ 0.05.

Odds ratio (OR) with confidence interval (CI) 95% was evaluated for the association between each independent variable and iron deficiency (ID-IDA). Values >1 were considered a positive association with the disease.

Results

In our cohort study, we compared the prevalence of iron deficiency with some independent variables in children aged 8–36 months observed in the last three decades (1980–1990; 1990–2000; 2000–2010) and respectively segregated in three groups (I, II, III). Regardless the decades we observed a higher prevalence of iron deficiency in children from deprived families with low food security than in children from families with medium-high income (P < 0.05). Furthermore, children who had CM introduced before the 12th month and weaned after the 6th month of life showed significantly more iron deficiency than other examined children (P < 0.05) (Tables 1, 2 and 3).

Table 1. Iron deficiency prevalence in children aged 8–36 months related to independent variables in the first decade

Table 2. Iron deficiency prevalence in children aged 8–36 months related to independent variables in the second decade

Table 3. Iron deficiency prevalence in children aged 8–36 months related to independent variables in the third decade

Overweight children also had a higher prevalence of iron deficiency than normal or lower weight ones (P < 0.05) (Tables 1, 2 and 3).

Prevalence of iron deficiency was also significantly higher in children not receiving iron supplementation (P < 0.05). On the contrary, iron status there appeared to be no significant difference among children of various ages or gender (P < 0.05) (Tables 1, 2 and 3).

Prevalence analysis of independent variables: family income, weight for height, introduction of CM, iron supplementation and weaning in the three decades is reported in Table 4. We found that family income was significantly lower in children in the first decade (1980–1990) than other decades (1990–2000, 2000–2010) (P < 0.05). Furthermore, in our cohort introducing CM before the first year of life was more prevalent in the decades 1980–1990 and 1990–2000 than in the third decade (2000–2010) (P < 0.05) (Table 4).

Table 4. Prevalence of variables over three decades

The evaluation of OR shows a positive association among low income (OR: 4; 95% CI: 1.16–0.04), age of CM introduction <12 months (OR: 6.8; 95% CI: 1.55–0.11), age of weaning >6 months (OR:2.5; 95% CI: 0.694–0.106), lack of iron supplementation (OR:17; 95% CI: 1.63–0.83), over-weight (OR 5.5; 95% CI: 0.85–0.55), and ID.

A negative association was found among other variables such as age, gender, and ID.

Discussion

Our retrospective study shows the importance of dietary history to identify children at risk for IDA and iron deficiency without anemia (ID).

Moreover, Hb screening in children between the ages of 9 months and 1 year has been recommended by many nutritional programs, but this practice has recently been questioned because normal values of Hb may not identify ID without anemia with possible consequences on health and neurodevelopment in early childhood.^12–16

Both ID and IDA are attributable to the imbalance between iron needs and available iron with consequent deficiency of iron stores mobilization and alterations of Hb concentration, MCV, MCH, ferritin, and transferrin saturation.

The children are considered anemic when Hb concentration falls 2sds below the mean value for age and gender. Furthermore, infants at 12 months of age with 11.0 g/dl of Hb are defined as anemic by WHO criteria. Moreover, the iron status of a child cannot be defined by a single measurement and the use of Hb alone to characterize ID is neither specific, nor sensitive.¹⁴ Furthermore, there is a marked overlap in Hb concentration in both people with iron sufficiency and ID. The diagnosis of both ID and IDA requires that Hb concentration is combined with the other markers of iron status. Only the Hb can be studied in the follow-up of IDA to measure the response to treatment.¹⁴

Preventing ID in childhood should include not only the screening but also the identification of children at high risk for ID by counseling practices. The correct record of nutritional habits since early infancy and social income is important to prevent the risk of ID.¹⁷

The sources of iron during the first months of life include human milk, formula feeds, or CM feeds. The iron in human milk is highest in early transitional milk, but decreases during lactation and reaches levels of about 0.3 mg/l after the fifth month with an absorption of about 50%. Cow and human milk feeds contain a similar amount of iron, but the iron bioavailability of CM is less than the human milk with an absorption of about 10%. Iron fortified formula as well as iron fortified weaning foods increase the iron absorption, while the absorption from infant formula without iron addition is about 10%.¹⁸

The iron status is also influenced by other nutrients. It is positively associated with vitamin C and negatively with calcium intake. It is well known that CM has a 10-fold greater calcium content than breast milk (1.3–1.4 g/l) as well as more protein (32–33 g/l) with different types of proteins, influencing the bioavailability of iron.¹⁸

Child food insecurity and ID are also associated with low social income with adverse health consequences.^7,19

Our findings confirm that dietary history can identify children at high risk for ID. The markers of iron status were in agreement with dietary questionnaire.

An analysis of our findings reveals the relationship between family income and development of ID. Furthermore, in the first decade characterized by lower socioeconomic status we found earlier introduction of CM, often beginning in the first months of life. There appears to be a clear correlation between early intake of CM and ID. In our cohort, there is a higher prevalence of CM before the twelfth month of life in the first decade than in other decades with better family income.

Iron supplementation represents another variable influencing iron status significantly related to family socioeconomic conditions.

We also saw that weaning after the sixth month of life negatively influences the iron status.

Our findings show no significant differences in the start of weaning over three decades. In our region the weaning starts early, often before the sixth month of life, regardless of family income.

The relationship between over-weight and ID or IDA is also evident in our cohort without significant differences over three decades.

Our findings are consistent with those reported in previous studies.^7,17

Medical and diet history are available to identify children at risk for ID or IDA at a young age, excluding those at very low risk from the hematological screening.¹⁷

Our retrospective study of clinical records from children aged 8–36 months underlines the relationship between some variables and iron status.

The socioeconomic and cultural progress over the last 30 years influencing dietary habits has improved the health of children. Specifically adding formula milk and cereal to the diet has played an important role in reducing iron deficiency.

In Italy, dietary habits and family income are very different from region to region with a gap between northern and southern regions. The latter have lower income. This finding is very important because each region should promote screening using clinical records to prevent iron deficiency in younger children.

Conclusions

The prevalence of ID and IDA depends on a variety of social, dietary factors, and health promotion activities in the first years of life. Medical and dietary records represent a screening tool to early identify children at risk for ID or IDA who need hematological tests, while those at low risk could be excluded from laboratory tests.

Our findings are in agreement with other reports about ID or IDA in young children. We think that these findings could be important in the face of the austerity affecting all countries and particularly the southern regions of Italy.