 |
INTRODUCTION |
Iron deficiency is a nutritional problem all over the world. Iron deficiency anemia is the most frequent form of nutritional anemia, with the highest prevalence occurring in developing countries (DeMayer and Adiels-Tegman 1985
). Infants <2 y of age and preterm newborns, in particular, represent major risk groups because of their low iron stores at birth, high iron requirements for growth, and diets consisting of foods with low iron content and low iron bioavailability (Dallman et al. 1980).
The high prevalence and consequences of iron deficiency justify the search for alternative prophylactic measures. Among the measures to be taken are the prevention of low birth weight and prematurity, the advocation of exclusive breast-feeding up to 6 mo of age, discouraging the use of fresh cow's milk at least up to mo 9 of life, the use of iron-fortified infant formulas when formulas are used, the inclusion of foods that facilitate iron absorption in the diet, the distribution of iron-fortified foods and iron supplementation (AAP 1992). However, some of these recommendations are impeded by political and socioeconomic problems, especially in developing countries, so that iron deficiency remains an unresolved public health problem.
An alternative approach to food enrichment with iron that has possible applications in developing countries is cooking infant diets in iron pots. The iron content of most diets increases significantly when the diet is cooked in an iron utensil (Borigato and Martinez 1992, Brittin and Nossaman 1986). Furthermore, in an experiment conducted in rats, the iron present in food cooked in an iron pot had good bioavailability (Martinez and Vannucchi 1986).
The objective of this study was to evaluate iron nutritional status in preterm infants fed food cooked in iron pots between mo 4 and 12 of life.
| |
SUBJECTS AND METHODS |
Research design.
The infants that would participate in the study were selected at 2 mo postnatal age. The controlled clinical trial was initiated during the beginning of mo 4 postnatal age, with follow-up to 12 mo of age. Pairs of infants were randomly selected and placed into two groups. The study group consisted of 22 infants whose food was cooked in iron pots (IP)4; the control group consisted of 23 infants whose food was cooked in aluminum pots (AP). The AP (Almar) was purchased on the Ribeirão Preto market. The IP, a 2-L pot, was manufactured especially for the study by the Fabbris foundry, Ribeirão Preto, Brazil, using pig iron. A single IP was used by each infant.
One of the authors (E.V.M.B.) evaluated the infants at the outpatient clinic on a monthly basis during the first half of the study and on a bimonthly basis during the second, in addition to also seeing them during unscheduled visits (such as in the case of an illness). At each return visit, a careful history was taken and each infant was given a clinical examination to determine general health and nutritional status. Anthropometric measurements were also made and blood was collected for hematologic evaluation. The dates scheduled for each return were monitored carefully, with a tolerance of 3 d. Growth (weight, length and head circumference) was evaluated every 2 mo. The infants were weighed on a Filizola (São Paulo, Brazil) pediatric scale with a 10-g sensitivity. Length was measured using an anthropometric ruler with a metric scale and 0.5-cm sensitivity. Head circumference was measured by using a Stanley (São Paulo, Brazil) 2-m flexible, nonexpandable steel tape with millimeter divisions. On the occasion of the return visits scheduled during mo 4, 8 and 12, venous blood was collected between 0700 and 1100 h for the determination of hemoglobin, hematocrit, mean corpuscular volume (MCV), free erythrocyte protoporphyrin (FEP), serum iron, total iron-binding capacity (TIBC), transferrin saturation (TS) and serum ferritin.
Illnesses noted during the study were classified according to degree of severity as follows: 1) mild when the mother did not seek medical care, when the infant presented with slight irritability, a cold, or changes in intestinal habits; 2) moderate when the mother sought medical care, with the infant presenting fever, irritation, coughing, acute otitis media, acute diarrhea or refusal of food for more than 24 h; and 3) severe when the mother sought medical care and the infant required hospitalization. When severe illness occurred, the infant was excluded from the study. Morbidity was assessed by voluntary maternal recall, depending on the child's clinical state and individual needs.
Maternal compliance with the continuous use of the cooking utensil and with the daily administration of medications in the two groups was evaluated by surprise monthly home visits by E.V.M.B. When cooked diets were found, samples were collected. The results of analysis have been reported elsewhere (Borigato and Martinez 1992).
Population sample.
From a total of 63 infants selected at 2 mo postnatal age, 45 fulfilled the requirements (absence of severe illness or blood transfusions, for example) to enter and complete the study, which began at 4 mo postnatal age, with follow-up to 12 mo postnatal age. These infants came from low socioeconomic backgrounds.
Gestational age was 30-36 wk, as determined by the date of last menstruation and confirmed by the method of Dubowitz et al. (1970). Birth weight was <2500 g, and appropriate weight for gestational age was determined according to the curves proposed by Battaglia and Lubchenco (1967). There were no congenital malformations, malnutrition or perinatal complications, and no blood transfusions. The infants were born and followed up at the University Hospital, Faculty of Medicine of Ribeirão Preto, State of São Paulo, Brazil.
Written consent to participate in the study was obtained from each mother after both the objective of the study and the type of observations in which they would be involved with their infants were explained to them. The research protocol was approved by the Department of Pediatrics and the Ethics Committee of the University Hospital.
Dietary advice was offered, in accordance with the routine of the service. Breast-feeding was encouraged whenever possible. Breast-fed infants started to receive fruit juices at the end of mo 3, fruit at the end of mo 4, cereals by the second half of mo 5, vegetables by the end of mo 5, meat and liver by the end of mo 6 and legumes by the second half of mo 7. Artificially fed infants started to receive fruit juices by the second half of mo 2, fruit by the second half of mo 4, cereals by mo 4, vegetables by the beginning of mo 5, meat and liver by the end of mo 6 and legumes by the beginning of mo 7 (Woiski 1988). The cereals (oatmeal or corn meal), vegetables, meat or viscera, and legumes, as well as milk when the infants were artificially fed, were always prepared in the pot indicated in the study. No iron-fortified cereals or formulas were used. The infants received the diet in the form of a pap, i.e., the solid part mixed with broth. This schedule has now been modified and we are currently following the recommendations of the American Academy of Pediatrics (AAP 1992).
The mothers of all infants were instructed to start giving their children a daily vitamin supplement consisting of 50 mg vitamin C and 3 mg vitamin E starting on d 7 of life and continuing up to 7.5 mo. Ferrous sulfate, providing 2 mg elemental iron/(kg·d), was recommended from d 15 to 12 mo of age.
Blood analyses.
Hemoglobin, hematocrit and MCV levels were determined by using a model CC-510 (Celm, Barveri, São Paulo, Brazil) automatic cell counter, a model DA-500 automatic diluter and a model HB-520 CELM hemoglobinometer. FEP was measured by the method of Piomelli (1973). Serum iron concentration and TIBC were determined by the method of Ramsay adapted to a micromethod (Ramsay 1957a and 1957b). Transferrin saturation was calculated by the serum iron/TIBC ratio. Serum ferritin was determined quantitatively by an immunoenzymatic method (Ferrizyme, Abbott Laboratories, Chicago, IL).
View this table:
[in this window]
[in a new window]
|
Table 1.
General characteristics at birth of the infants studied, sociodemographic characteristics of the study
population and type of cooking utensil used
|
|
The reference values employed for the classification of the hematologic parameters of preterm infants at 4 mo were as follows: hemoglobin >100 g/L, MCV > 74 fL, serum iron > 7 µmol/L, TIBC < 97 µmol/L, TS > 7% and serum ferritin > 11 µg/L (Halliday et al. 1984, Heese et al. 1990). The following cut-off levels were considered at 12 mo: hemoglobin > 110 g/L, hematocrit > 0.32, MCV > 70 fL, FEP < 0.53 µmol/L whole blood, serum iron > 5 µmol/L, TIBC < 72 µmol/L, TS > 10%, and serum ferritin > 10 µg/L (Dallman and Reeves 1984).
The iron nutritional status of the infants was evaluated on the basis of combined laboratory data and classified as follows: 1) normal when hemoglobin, serum ferritin, and 2 or 3 criteria for MCV, FEP, TS levels were within normal limits; 2) iron deficient when hemoglobin was normal, serum ferritin low, and 0 or 1 of the criteria for MCV, FEP, TS was abnormal; 3) iron deficiency anemia when hemoglobin was low, serum ferritin low, and 0 or 1 of the criteria for MCV, FEP, TS was abnormal; and 4) other type of anemia when hemoglobin was low and serum ferritin was normal.
Statistical analysis.
Student's t test for data with normal distribution, the Mann-Whitney U test for data not normally distributed and the chi-square test for two independent samples were used to compare the data between the two groups and to analyze the change in blood concentrations between 4 and 12 mo. Fisher's exact test was used when the chi-square test was not feasible. The level of significance was set at P < 0.05 (Siegel 1975).
| |
RESULTS |
The general characteristics of the children in the two groups were similar and are presented in Table 1. There were no significant differences in sex, gestational age, percentage of appropriate for gestational age, weight, length, head circumference at birth or sociodemographic characteristics.
During the longitudinal follow-up, 18 losses occurred in the initial group of infants selected at 2 mo (28.6%; 18 of 63) for the following reasons: 7 were due to severe disease, 2 from the IP group (1 with prolonged diarrhea and the other with bronchopneumonia) and 5 from the AP group (3 with prolonged diarrhea, 1 with infectious hepatitis, and 1 with gastroesophageal reflux); 2 losses were due to the fact that the family moved to another town (both in AP group); 7 did not return to the clinic (2 in IP group and 5 in AP group); and 2 were due to the fact that the mothers stopped using the iron pot after 4 and 5 mo of use.
Of the 45 infants who completed the longitudinal follow-up, 10 from the IP group (10 of 22, 45.4%) and 9 from the AP group (9 of 23, 39.1%) presented mild-to-moderate illnesses consisting of infection of the upper airways, acute otitis media or acute infectious diarrhea.
The groups did not differ significantly in duration of breast-feeding (P > 0.05), which was 6 mo or more for 36.4% of the infants (8 of 22) in the IP group and 21.7% (5 of 23) in the AP group (Table 2).
On the occasion of the home visits, it was observed that the iron pot was used in the daily routine, whereas the administration of medication (ferrous sulfate and multivitamins) was not regular, reflecting the low maternal compliance.
The two groups did not differ significantly in weight, length or head circumference at 4 or 12 mo (Table 3).
View this table:
[in this window]
[in a new window]
|
Table 3.
Anthropometric variables measured at 4 and 12 mo of age in preterm infants fed food cooked in iron or aluminum pots1
|
|
Blood concentrations of hemoglobin, MCV, FEP, serum iron, TIBC, TS and serum ferritin at 4 mo of age did not differ significantly between groups, whereas at 12 mo of age they were significantly improved in IP infants, except for serum iron, TIBC and TS (Table 4).
Between 4 and 12 mo, the change in several hematologic indices differed significantly between groups (Table 5). In the IP group, mean hemoglobin (P = 0.01) and transferrin saturation (P = 0.05) increased, whereas they decreased in the AP group. The IP group experienced a smaller decrease in mean corpuscular volume (P = 0.05) and serum iron (P = 0.04) than the AP group.
View this table:
[in this window]
[in a new window]
|
Table 5.
Change in blood variables from 4 to 12 mo of age in preterm infants fed food cooked in iron or aluminum pots1
|
|
No significant differences in blood concentrations of hemoglobin, MCV, FEP, serum iron, TIBC, TS and serum ferritin were observed at any of the ages tested between infants with appropriate for gestational age weight and infants with small for gestational age weight, within or between groups (data not shown).
Iron nutritional status at 12 mo of age, evaluated on the basis of combined hemoglobin, serum ferritin, MCV, FEP and TS data, was significantly better in the IP group than in the AP group (P = 0.03) (Table 6).
View this table:
[in this window]
[in a new window]
|
Table 6.
Classification of iron status at the age of 12 months in preterm infants fed food cooked in iron or aluminum pots
|
|
| |
DISCUSSION |
The diets offered to infants are frequently inadequate in terms of satisfying the iron requirements for growth (Pizarro et al. 1991). Despite the rapid accumulation of new information about iron deficiency, the diversity of feeding regimens and the economic situations prevailing in many parts of the world represent special problems that require a variety of solutions. Fortification of milk and cereals and oral iron administration have prevented iron deficiency in term and preterm newborn infants in developed countries (Miller et al. 1985). In developing countries, the establishment of programs of food fortification for the target population is often impeded by political and socioeconomic problems. Appropriate strategies directed at the prevention of iron deficiency anemia are necessary in these countries, where the incidence of the disorder is so high as to reach epidemic proportions among infants (Florentino and Guirriec 1984, Macphail and Bothwell 1992). The low maternal compliance with oral iron administration and the precarious living conditions of the population may lead to failure of standard programs for the prevention and treatment of iron deficiency.
The supply of some trace elements can be increased by the use of different forms of food processing and by the type of cooking utensils employed. These provide an approach to diet fortification and possibly effective prevention of certain deficiencies. Analysis of the iron content of an infant diet cooked in an iron pot demonstrated a significant increase in dietary iron concentration (Borigato and Martinez 1992). However, the nutritional and toxicologic importance of these casual metal sources in the diet should be carefully analyzed (Reilly 1987). In a study comparing iron and aluminum utensils in terms of increased dietary content of Fe, Zn, Cu and Pb, the detection of toxic levels of any of the metals was not reported. There was a higher Fe concentration in food cooked in the iron utensil and no important differences were reported for Pb, Zn or Cu. In some diets, Pb levels were higher in foods cooked in aluminum utensils (Reilly 1985).
Aluminum utensils have been widely used since 1960, and have the following advantages compared with iron utensils: they are good heat conductors, do not alter food color, do not rust, they are lightweight and cost less. However, aluminum utensils not only supply less iron but also release aluminum, which is suspected of being harmful to health (Karlik et al. 1980). Although this was not evaluated, in theory it would be safer to use an iron pot to cook food because it releases less Al and Pb (Reilly 1985).
During mo 4 of life or earlier, the iron reserves derived from the mother are depleted in infants and, unless available supplemental iron is supplied, late anemia of prematurity develops (Friel et al. 1990). A study evaluating low birth weight infants receiving 2 mg Fe/(kg·d) from d 15 of life and comparing them with a group receiving 4 mg Fe/(kg·d) after 4 mo of age showed that the lower iron dose was insufficient to prevent iron deficiency (Lundström and Siimes 1980). In that study, the use of an iron pot was started at 4 mo of age as a complement to the supplementary iron dose recommended by the American Academy of Pediatrics Committee on Nutrition (AAP 1976). Diversification of the diet offered to the infants was also started during this period.
In this study, mothers were instructed to administer 2 mg Fe/(kg·d) as ferrous sulfate starting at 15 d and extending to the end of the first year of life for both infant groups, in addition to the use of an iron pot for cooking the diet for the IP group and of an aluminum pot for the AP group. On the occasion of the home visits it was observed that the iron pot was used in the daily routine, whereas the administration of medication (ferrous sulfate and multivitamins) was not regular, reflecting the low maternal compliance; however, no quantitative documentation has been obtained at this point. The reason reported by the mothers for not offering the medication on a daily basis was that their household dutes caused them to forget. The mothers participating in this study were of a low socioeconomic level, ranging in age from 20 to 39 y; most of them were multiparous housewives with minimal schooling, residing in the periphery of the city. Daily administration of medication to apparently healthy children for a period of several months is difficult to sustain even for mothers of relatively high socioeconomic status (Palti et al. 1987). The greater compliance in the use of the iron pot represented a practical advantage of this method of iron enrichment of the diet.
Growth pattern affects iron needs. Low birth weight infants have lower iron reserves and rapidly growing infants require more iron. The growth patterns observed in this study were similar to those reported by Gorten and Cross (1964). The two groups of children studied here were of similar size at birth and presented adequate growth during the first year of life, with no significant difference between them at 4 and 12 mo of age. This indicates that growth rates did not affect the different iron nutritional status indices detected in the two groups.
The type of feeding is one of the variables affecting the iron status of infants. In this study, 64.4% (29 of 45) of the infants received artificial feeding at 4 mo, a proportion that increased to 71.1% (32 of 45) at 6 mo. Despite the subtle shift toward more breast-feeding in the IP group, there was no significant difference between groups. All infants received boiled cow's milk after being weaned from the breast. No iron-fortified food or formula was used by the infants because of the families' economic restrictions. In poor Brazilian communities, despite recommendations to the contrary (AAP 1992), boiled cow's milk is the only possible option for artificial feeding of infants after they are weaned from the breast.
The longitudinal design of the study reduces the ability to follow up all infants. On the other hand, follow-up provides the opportunity of evaluating clinical problems that occur during treatment. Cooking food in iron pots was not associated with a difference in morbidity between groups. Most of the infants that became ill during the period of observation (42.2%, 19 of 45) presented signs and symptoms of diseases commonly occurring in infants and classified as mild to moderate, i.e., infection of the upper airways, acute otitis media or acute infectious diarrhea. Despite the good doctor-patient relationship created during the study, there is the possibility that the method used for morbidity surveillance would not be sensitive to mild illnesses. The two groups also did not differ in terms of the infants excluded because of diseases that were classified as serious and that would interfere with nutritional iron status (11.1%, 7 of 63). The losses in longitudinal follow-up observed in this study (28.6%, 18 of 63) were similar to those reported by Heese et al. (1990) in a study on the same type of population receiving iron supplementation by the oral or parenteral route. Friel et al. (1990) reported a 39.5% drop-out rate at the end of 12 mo of observation of low birth weight infants receiving milk formula fortified with iron.
Analysis of the results obtained during mo 4, when the study was started, did not show a significant difference in blood variables between the two groups studied in terms of levels observed or their clinical importance. At that time, the iron stores of the infants were already depleted. At 12 mo of age, hemoglobin, hematocrit, MCV and serum ferritin were higher and FEP lower in the IP group than in the AP group. The change in blood concentrations from 4 to 12 mo showed a similar trend. Iron nutritional status, evaluated on the basis of a combination of examinations, was significantly better in the IP group.
At the end of the study, we noted that supplementation with medicinal iron in both groups and iron fortification by the use of iron pots in the IP group did not adequately increase the iron stores. Also, even though FT levels were significantly higher in the IP group than in the AP group, the exogenous iron offered was not sufficient to replenish depleted iron stores during the first year of life.
A combination of various factors that adversely affect iron reserves was present in the population studied here: low iron reserves at birth, very rapid growth rate and a diet consisting of foods with low intrinsic iron content.
Evaluation of these results demonstrates that the increase in iron in food cooked in iron pots was insufficient to satisfy the high iron requirements of preterm infants during the first year of life. The better nutritional iron status of the IP infants indicates that further studies should be conducted on foods cooked in iron pots containing factors that facilitate non-heme iron absorption or lower levels of inhibitory factors. Although cooking in iron pots alone was not sufficient to satisfy the high requirements of preterm infants, under the presents conditions, they exerted a positive effect on iron balance and might be considered a useful adjunct to programs to prevent iron deficiency in populations with high rates of this condition.
| |
ACKNOWLEDGMENT |
We thank Isabel Machado de Souza for technical assistance.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
SARAH, Brasília, 70330-150, DF, Brazil and 
Abstract
Introduction
110 g/L) was observed in 36.4% (8 of 22) of infants in the group fed food cooked in iron pots and in 73.9% (17 of 23) of the infants fed food cooked in aluminum pots (P = 0.03). These results indicate that the iron added to food cooked in iron pots is bioavailable. However, this increased iron availability was insufficient to satisfy the high iron requirements of this group of preterm infants.
Abstract
