QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lartey, A. Articles by Dewey, K. G. Search for Related Content PubMed PubMed Citation Articles by Lartey, A. Articles by Dewey, K. G.

---

Article

Predictors of Micronutrient Status among Six- to Twelve-Month-Old Breast-Fed Ghanaian Infants¹

Anna Lartey^*,, Alhassan Manu^*, Kenneth H. Brown^* and Kathryn G. Dewey^*2

^* Department of Nutrition and Program in International Nutrition, University of California, Davis, CA 95616-8669, and Department of Nutrition and Food Science, University of Ghana, Legon

²To whom correspondence should be addressed.

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
This study describes the factors associated with hemoglobin and plasma ferritin, zinc and retinol concentrations and erythrocyte riboflavin status among 208 Ghanaian infants who participated in a complementary feeding intervention trial from 6 to 12 mo of age. Anthropometric, morbidity and dietary data were collected regularly from 1 to 12 mo; blood samples were collected at 6 and 12 mo. The prevalence of low micronutrient status at 6 and 12 mo, respectively, was as follows: hemoglobin <100 g/L, 30 and 34%; plasma ferritin <12 µg/L, 17 and 43%; plasma zinc <10.7 µmol/L, 4 and 6%; plasma retinol < 0.7 µmol/L, 26 and 26%; erythrocyte riboflavin <200 umol/L of packed red cells, 14 and 10%. Multiple regression was used to identify factors significantly associated with micronutrient status. From 6 to 12 mo, fever prevalence was associated with a decrease in hemoglobin, but an increase in erythrocyte riboflavin concentrations, and diarrhea prevalence was related to a decrease in plasma retinol. Seasonal differences were evident for most of the indicators of micronutrient status, and elevated C-reactive protein levels (indicative of recent infection) were related to lower hemoglobin, retinol and zinc concentrations but higher ferritin and erythrocyte riboflavin concentrations. Weight at birth or at 1 mo of age was positively related to iron, zinc and vitamin A status, but a more rapid weight gain was associated with depletion of iron stores. Socioeconomic status was related to higher hemoglobin, riboflavin and zinc concentrations. The feeding of a micronutrient-fortified food was positively associated with plasma ferritin and retinol concentrations at 12 mo. These results suggest that prenatal factors, socioeconomic status, dietary intake and morbidity all influence infant micronutrient status, and that fortification of complementary foods is one potential avenue for preventing deficiencies.

KEY WORDS: • anemia • ferritin • zinc • vitamin A • riboflavin • humans

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Deficiencies of micronutrients such as vitamin A, iron and iodine are public health problems in many developing countries. Worldwide, over two billion people are at risk and more than one billion are actually affected (The World Bank 1994 ). Marginal zinc deficiency is also suspected to be widespread, especially in populations heavily dependent on cereal-based diets that are high in phytic acid and contain few animal products (Sandstead 1991 ). The deleterious effects of micronutrient deficiencies on human health are well documented. Vitamin A deficiency is an important cause of blindness in children, and subclinical deficiencies increase the risk of severe morbidity and mortality (Beaton et al. 1993 , Sommer and West 1996 ). Iron deficiency, the most common cause of nutritional anemia, is associated with impaired cognitive development and psychomotor function in infants and children (Beard 1995 , Lozoff et al. 1991 , Soewondo et al. 1989 ). Zinc deficiency results in growth retardation in children (Brown et al. 1998b ) and increases the risk of infection (Black 1998 ). The causes of micronutrient malnutrition are varied and include inadequate dietary intakes, repeated infections and poor bioavailability from foods due to the presence of inhibitors or inadequate intake of dietary enhancers.

The extent to which micronutrient deficiencies begin during infancy, in particular among breastfed infants, has not been well documented. Although breastfeeding is the optimal feeding mode for many reasons and has been shown to be protective against clinical vitamin A deficiency (West et al. 1986 ), human milk contains relatively little iron and zinc, and its vitamin content can be compromised by maternal malnutrition (Brown et al. 1998a ). There is a paucity of data on micronutrient status of breastfed infants in the age range of 6 to 12 mo, which is a vulnerable period because of the low nutrient density of many staple complementary foods and the high nutrient requirements of the growing infant.

The purpose of the analyses presented in this paper was to determine the likelihood of micronutrient deficiencies in breastfed infants in Ghana and to explore their potential causes. The data are from a randomized intervention trial (Lartey et al. 1999 ) that was designed to examine the effect of feeding four improved complementary foods from 6 to 12 mo of age. In this study, infants were randomly assigned to receive Weanimix (W)³ (a cereal-legume blend of 75% corn, 15% soybean and 10% groundnuts) or one of three other locally formulated, centrally processed complementary foods: Weanimix fortified with vitamins and minerals (WM), Weanimix with fish powder added (WF) and koko (fermented corn dough) with fish powder added (KF). The study was unique because of the simultaneous, comprehensive assessment of growth, morbidity, weighed dietary intake and micronutrient status of a large sample of infants who were breastfed throughout the first year of life. The results of the intervention showed no significant differences in growth among the four intervention groups, but iron stores and vitamin A status at 12 mo of age were increased in the group fed WM (Lartey et al. 1999 ). In this paper, we present data on other factors associated with iron, zinc, vitamin A and riboflavin status of the same cohort of infants at both 6 and 12 mo of age.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Study design.

Infants (n = 216) were recruited by 1 mo of age from Maternal and Child Health Centers (MCH) in Techiman, Ghana, between November 1994 and April 1995. Techiman, a district capital, is located about 400 km north of Accra and has a population of about 45,000 people. The main occupations for the inhabitants are farming and trading. Techiman has the biggest market in the country. The dry season, when food is less available, is from December to March, and the wet season is from April to approximately July; the weather is variable from August to November. All mothers of eligible infants who attended the MCH centers during the recruitment period were asked to participate, and nearly all agreed. Infants were eligible if they were breast-fed, had no health complications, weighed 2.5 kg at birth, and the mother did not plan to travel during the period of study. The study was approved by the Human Subjects Review Committee of University of California, Davis and by the Ministry of Health, Ghana.

From 1 to 6 mo of age, baseline data were collected and anthropometric status, infant feeding practices and morbidity were assessed monthly. At 6 mo of age, blood was drawn for the assessment of baseline micronutrient status, and infants who remained in the study (n = 208) were randomly assigned to receive one of the four foods: W, WM, WF or KF. Fish powder was added at 20 g/100 g. The nutrient content of the foods has been published elsewhere (Lartey et al. 1999 ). From 6 to 12 mo of age, anthropometric measurements continued monthly, and morbidity data were collected weekly. Intake of project and nonproject foods was monitored during the intervention period. At 12 mo of age, a second blood sample was taken for reassessment of micronutrient status.

Anthropometry.

Birth weights were recorded mainly from birth certificates. Most of the infants were born in hospitals or maternity centers, where birth weight was measured soon after birth by health-care personnel. After enrollment in the study, anthropometric measurements were taken monthly. All anthropometric measurements throughout the study were taken by two trained assistants whose techniques were standardized according to WHO procedures (WHO 1983 ). Recumbent length was measured to the nearest 0.1 cm with a length board (Perspective Enterprises, Portage, MI) and weight to the nearest 100 g using a digital scale (Perspective Enterprises) that was calibrated weekly. Maternal weight was measured monthly, and maternal height was measured at baseline.

Dietary intake.

In the first 6 mo of life, frequency of breastfeeding and consumption of other foods and fluids were assessed using a monthly food frequency questionnaire. At 6, 7, 8, 10 and 12 mo of age, dietary intake (of both project and nonproject foods) was assessed by 24-h recalls in all subjects and by weighed food records in a randomly selected subsample of 50% infants at each time point. For the latter method, a trained observer weighed all foods and beverages consumed by the infant during a 12-h period, using a scale accurate to the nearest gram. Energy and nutrient intakes were calculated from local food composition tables and other published values (Eyeson et al. 1975 , Ferguson et al. 1993 ) and from analyzed values of fat, iron, zinc and riboflavin in project foods. Average breastfeeding frequency from 6 to 12 mo of age was based on maternal recall every 2 wk.

Biochemical measures.

Blood samples were collected in the morning between 9:00 and 12:00 a.m. by venipuncture from all infants at 6 and 12 mo of age. Due to parental refusal or technical difficulties, it was not possible to obtain blood from all subjects, and in some cases the amount of blood obtained was not sufficient for all the biochemical tests. In addition, for plasma zinc, hemolyzed samples were excluded from analysis.

Blood was collected in heparinized, trace element free vacu-tainer tubes. Hemoglobin concentration was determined using the HEMOCUE B-hemoglobin photometer (Ängelholm, Sweden). Erythrocytes were separated from plasma by centrifugation within 15 min of blood collection under laboratory fluorescent light. The erythrocytes were washed three times in saline solution (0.9 NaCl). Racks with the tubes containing the processed erythrocytes and plasma were wrapped in aluminum foil and were stored at -20°C. Frozen samples were transported to the University of California, Davis, Clinical Nutrition Research Unit for analyses of plasma ferritin (in duplicate by immunoradiometric assay; Diagnostic Products Co., Los Angeles, CA), C-reactive protein (in duplicate by rate nephelometry, Beckman Instruments Inc, Galway, Ireland), plasma iron and zinc (three readings of each sample, by atomic absorption spectrophotometry; Butrimovitz and Prudy 1977 , Fernandez and Kahn 1971 ), plasma retinol by HPLC (Arroyave et al. 1982 ) and erythrocyte riboflavin by HPLC (Floridi et al. 1985 ). Commercial controls were run with each batch for ferritin (CV 8.5–10.7%), C-reactive protein (CV 2.8%) and plasma iron and zinc (CV 4.8 and 4.2%, respectively); pooled samples from the Clinical Nutrition Research Unit were used as controls for retinol (CV 6.4%) and erythrocyte riboflavin (CV 9.8%).

Morbidity.

From 1 to 6 mo of age, infant morbidity information was collected monthly by asking mothers to recall specific symptoms of diarrhea, fever and respiratory illness during the 7 d preceding the monthly home visit. In addition, mothers reported whether the infant had been ill during the 3 wk before the 7-d recall. For each type of illness, morbidity from 1 to 6 mo of age was calculated as a morbidity score, that is, the proportion of the number of months during which illness was reported to the total number of months in the interval. One-week morbidity prevalence during the 1- to 6-mo age interval correlated highly with morbidity score (r > 0.6). However, the latter was used in the analysis to capture morbidity information for the whole month. Diarrhea score was based on mother’s perception of diarrhea. From 6 to 12 mo of age, symptoms of these illnesses were recorded weekly. Mothers were provided with a daily grid on which to indicate the occurrence of the morbidity symptoms between visits. During this period, diarrhea was defined as three or more liquid or semiliquid stools, fever was based on mother’s report of elevation of infant’s body temperature above normal, and respiratory illness was defined as the presence of purulent nasal discharge or cough. For each illness category, morbidity prevalence was calculated as the proportion of days illness was present to the number of days information was collected for each subject.

Data analysis.

Statistical analyses were performed using PC-SAS Release 6.04 (SAS Institute, Cary, NC). Baseline characteristics of the four groups of infants were compared using analysis of variance. Multiple stepwise linear regression analysis was used to determine factors associated with micronutrient status, specifically, hemoglobin, plasma ferritin, zinc and retinol and erythrocyte riboflavin at 6 and 12 mo of age and the change in these outcomes between 6 and 12 mo of age. Variables included in the regression models are shown in Table 1 . Seven categories of independent variables known to affect micronutrient status were considered: i) infant characteristics, ii) maternal characteristics, iii) morbidity, iv) socioeconomic indicators, v) dietary intake, vi) biochemical indices reflecting recent infection (C-reactive protein) and, for the analyses of changes in status, initial micronutrient status at 6 mo and vii) season (dry season, wet season, and postwet season). Infant birth weight or weight at 1 mo of age was included in the models, but not infant length because it was highly correlated with infant weight and not significantly related to any of the outcome variables. Maternal body mass index and maternal height were used in all models instead of maternal weight becauseof the highly significant correlation between maternal height and weight. Because the distributions of plasma ferritin and retinol concentrations and household income were skewed, natural logarithmic transformations were applied to these variables. Hemoglobin concentration correlated highly with hematocrit (r > 0.8) and therefore only mean cell hemoglobin concentration and hemoglobin were included in the regression analysis. For the models for hemoglobin, ferritin, retinol, and plasma zinc, a dichotomous variable for elevated C-reactive protein level (>8 mg/L, indicative of recent infection) was included in the analysis.

Multiple logistic regression was used to examine predictors of low values for each of the outcome variables at 6 and 12 mo, using the same independent variables described above. The cut-off values used to define low values were: hemoglobin <100 g/L, plasma ferritin <12 ug/L, plasma retinol <0.7 umol/L, erythrocyte riboflavin <200 umol/L packed red cells, and plasma zinc <10.7 umol/L. Because we could not locate any published cut-off values for erythrocyte riboflavin, we chose an arbitrary value approximately one standard deviation below the mean for our sample.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
During the intervention phase (6 to 12 mo of age), 18 infants dropped out for the following reasons: father refused child’s participation (n = 1), infant rejected project food (n = 4) mother did not feed project food (n = 4), mother traveled out of study area (n = 8) and death of infant (n = 1). The characteristics of the infants with respect to birth weight (3.18 ± 0.44 kg), maternal age (27.7 ± 5.9 y) maternal education (6.5 ± 5.2 y), paternal education (7.5 ± 6.0 y), paternal age (35.2 ± 8.7 y), household size (7.3 ± 3.5) and percentage of household income spent on food (37.2 ± 19.4) did not differ significantly among the four intervention groups. Dropouts did not differ significantly from nondropouts in any of these characteristics except for birth weight, which was lower in the dropouts (2.9 ± 0.3 vs. 3.2 ± 0.4 kg, P < 0.002).

During the 1- to 6-mo age interval, diarrhea, fever and respiratory illness scores were 0.24 ± 0.22, 0.20 ± 0.21 and 0.30 ± 0.26, respectively. Morbidity prevalences from 6 to 12 mo of age for diarrhea, fever and respiratory illness were 15.5 ± 14.4, 8.3 ± 7.1 and 16.9 ± 14.5% of days, respectively.

The mean daily age-adjusted intakes of energy, iron, zinc, calcium, vitamin A, riboflavin and animal protein from nonbreast milk sources (including the project foods) at 6–12 mo of age are shown in Table 2 , along with recommended intakes from complementary foods at this age (Brown et al. 1998a ). The average intakes of iron, zinc, calcium and vitamin A were significantly higher for infants fed WM than for those fed the other three foods, due to the fortification of WM with vitamins and minerals. Average intakes of energy, iron, zinc, calcium, vitamin A and riboflavin by infants in the combined W, WF and KF groups were well below the recommended intakes from complementary foods.

At both 6 and 12 mo of age, plasma ferritin was higher in females than in males and was negatively associated with weight gain since 1 mo of age. Elevated C-reactive protein values were consistently associated with higher ferritin concentrations. In addition, plasma ferritin at 12 mo was higher during the postwet season period (August-November) and was positively associated with weight at 1 mo of age and with available iron intake from nonbreast milk foods (which was correlated with assignment to the WM group, r = 0.77, P < 0.0001; when iron intake was excluded from the model, the WM variable entered instead). The change in ferritin between 6 and 12 mo was inversely related to ferritin at 6 mo and to weight gain between 6 and 12 mo. The results of the mulitple logistic regression analyses were similar. At 6 mo, low ferritin was associated with male sex and higher weight gain since 1 mo of age; at 12 mo it was associated with higher weight gain from 1 to 12 mo of age and being in one of the intervention groups other than WM (i.e., W, WF or KF).

Factors associated with plasma retinol concentration and erythrocyte riboflavin status are shown in Table 5 . Birth weight was positively associated with plasma retinol at 6 mo of age. Diarrhea prevalence between 6 and 12 mo was negatively associated with both plasma retinol at 12 mo and the change in plasma retinol between 6 and 12 mo of age. Intake of vitamin A from nonbreast milk foods was positively associated with plasma retinol at 12 mo and the feeding of WM was positively associated with the change in plasma retinol from 6 to 12 mo of age. [Because of the fortification of WM with vitamin A, these two dietary variables (vitamin A intake and assignment to the WM group) were correlated with each other (r = 0.54, P < 0.0001).] Elevated C-reactive protein values were consistently associated with lower plasma vitamin A concentrations. In the multiple logistic regression analyses, low plasma retinol at 6 mo was significantly associated with lower birth weight; at 12 mo low plasma retinol was associated with lower weight at 1 mo of age, male sex, greater prevalence of diarrhea from 6 to 12 mo, and being in an intervention group other than WM (i.e., W, WF or KF).

Factors associated with plasma zinc concentration are shown in Table 6 . At 6 mo of age, plasma zinc was positively related to weight at 1 mo and household income and negatively associated with maternal height and elevated C-reactive protein values. At 12 mo, plasma zinc concentration was higher during the dry season (December-March) and was associated with several dietary variables. In the initial bivariate analyses for plasma zinc at 12 mo of age, there was an inverse relationship with dietary available zinc, contrary to expectations. However, when other dietary variables were included in the multiple regression (intake of calcium, phytate and animal protein), there was a significant inverse relationship between plasma zinc at 12 mo and calcium intake, but not dietary available zinc (even though the latter variable was based on an algorithm that presumably adjusts for the inhibitory effect of calcium). Energy intake from complementary foods was positively associated with plasma zinc at 12 mo. Change in plasma zinc from 6 to 12 mo of age was also associated positively with energy intake and negatively with calcium intake (as well as with weight at 6 mo and elevated C-reactive protein values). Multicollinearity was a problem with the nutrient intake variables due to the very strong correlation between calcium intake and dietary available zinc (r = 0.98 for the total sample; r = 0.76 when excluding the WM group; and r = 0.66 when excluding the WM group and adjusting for energy intake; all with P < 0.001). Neither of these variables was significant when both were in the model, and each was significantly negative when the other was excluded from the model. To determine whether the predictors of change in zinc status differed without the influence of group WM (who consumed the food fortified with micronutrients, including calcium), the model was rerun after taking this group out of the analysis. The results show that calcium intake was still negatively associated with the change in plasma zinc (this was true even with dietary zinc forced into the model); in addition, there was a positive association with intake of animal protein. We also investigated the possibility that nutrients from project foods had a different effect on change in plasma zinc than nutrients from nonproject foods, but this turned out not to be the case. Project foods were the main sources of calcium for all four groups. Foods W, WM, WF and KF provided 34, 97, 85 and 80%, respectively, of the calcium intake from nonbreast milk sources. Corn-based products, milk powder and rice were the main sources of calcium from nonproject foods.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
The factors associated with micronutrient status among breast-fed infants in this population differ depending on the nutrient. In the case of iron status, the association with birth weight has been well documented previously (Dallman et al. 1980 ). Full-term infants are normally endowed with adequate iron stores at birth, which is also reflected in high hemoglobin concentration during the first few months of life. Hemoglobin and iron stores decline as body iron is mobilized in support of the rapid growth during this period (Dallman et al. 1980 ). Our observation of a positive association of hemoglobin at 6 mo of age with birth weight is consistent with results of a recent study in Honduran infants (Dewey et al. 1998 ). In the present study, hemoglobin concentration at 12 mo of age and the change in hemoglobin between 6 and 12 mo of age were negatively associated with fever prevalence. The close association of fever with malaria in this population is a likely reason for this observation. Malaria has been identified as an important cause of anemia in infants and young children in malaria endemic areas (Binka et al. 1994 , Bloland et al. 1993 , Newton et al. 1997 , van Hensbroek et al. 1995 ). A malaria prevalence survey in northern Ghana reported that parasite density and rates of febrile illness were highest in infants 6 to 11 mo of age, and there was a negative correlation between parasite density and hemoglobin concentration (Binka et al. 1994 ). The proposed mechanisms by which malaria induces anemia are through the increased destruction of parasitized cells (Phillips and Pasvol 1992 ) and impaired red blood cell production during chronic malarial infection (Abdalla et al. 1980 ). Elevated C-reactive protein levels, which indicate recent infection or inflammation, were consistently associated with lower hemoglobin concentrations.

Plasma ferritin is an indicator of iron stores. At 12 mo (though not at 6 mo), plasma ferritin was positively associated with weight at 1 mo of age, which is consistent with the association between birth weight and ferritin in other studies (Dewey et al. 1998 , Edmond et al. 1996 ). Female infants had higher ferritin concentrations at 6 and 12 mo of age than male infants. A similar gender effect on iron stores in favor of girls at 8 and 12 mo has been reported previously (Edmond et al. 1996 , Wharf et al. 1997 ). We also found consistent inverse associations between infant weight gain and plasma ferritin concentrations, indicating that infants with higher growth rates were drawing more on their iron stores. This is in agreement with observations from other recent studies (Dewey et al. 1998 , Edmond et al. 1996 , Wharf et al. 1997 ). As expected, iron intake from complementary foods was positively associated with plasma ferritin at 12 mo. The well-documented association between elevated C-reactive protein levels and higher plasma ferritin was also observed in this study.

The observed positive association between plasma retinol at 6 mo of age and birth weight is also consistent with results from other studies (Shah and Rajalakshmi 1984 , Yassai and Malek 1989 ). Our analysis indicated a negative association between diarrhea prevalence and both retinol concentration at 12 mo and the change in plasma retinol from 6 to 12 mo of age. A negative association between gastrointestinal infection and plasma retinol has been reported previously in preschool and school-aged children (Friis et al. 1996 and 1997 ). However, the causal direction of the relationship is not always clear. Marginal vitamin A deficiency has been associated with increased risk of morbidity and mortality, and some (though not all) intervention trials have reported reduced incidence or severity of illness due to vitamin A supplementation in children (Arthur et al. 1992 , Lie et al. 1993 ). On the other hand, plasma retinol concentration decreases in response to infections (Filteau et al. 1993 , Stephensen et al. 1994 ). At least part of this is due to the acute phase response, as was evident in this study from the inverse association between C-reactive protein level and plasma retinol. However, diarrhea prevalence was inversely related to retinol concentrations even when controlling for C-reactive protein level. Diarrheal infections may decrease plasma retinol by decreasing absorption (Mahalanabis et al. 1979 ,Mahalanabis 1991 ), increasing metabolic requirements (Campos et al. 1987 ), increasing the rate of urinary excretion of retinol (Stephensen et al. 1994 ) and/or reducing the synthesis of hepatic retinol-binding proteins (Rosales et al. 1996 ), thus interfering with retinol mobilization from storage to plasma. In our study, it is likely that diarrhea contributed causally to low plasma retinol levels, rather than the converse, as diarrhea incidence and prevalence were not significantly different between infants given the vitamin A-fortified food (WM) and the other three groups without additional dietary vitamin A (Lartey et al. 1999 ). Furthermore, plasma vitamin A levels at 6 mo did not predict future levels of diarrhea after 6 mo of age (data not shown). Our results suggest that inadequate intake of vitamin A from complementary foods contributed to low plasma retinol levels at 12 mo; a greater increase in plasma retinol was observed among infants receiving either the vitamin A fortified food (WM) or larger amounts of vitamin A from other foods. The project food WM provided about 90% of the vitamin A intake from nonbreast milk foods for this group. For the other three groups (W, WF and KF), dietary intake of vitamin A was apparently inadequate, judging by their lower plasma retinol status at 12 mo. In this community, the main sources of vitamin A in the diet are red palm oil products and dark green leafy vegetables, but these foods are generally not fed to infants. It is also possible that breast milk vitamin A concentrations were lower than normal due to maternal malnutrition, as has been found elsewhere in Ghana (Lartey and Oracca-Tetteh, 1990 ), but we did not assess this.

Very few studies have assessed riboflavin status of breast-fed infants (Bates et al. 1982 ). In our study, erythrocyte riboflavin was used as an indicator of riboflavin status. This method directly measures erythrocyte riboflavin content as flavin adenine dinucleotide and flavin mononucleotide, an indicator of long-term riboflavin status (Fidanza et al. 1989 ). At 6 mo of age, erythrocyte riboflavin was positively associated with household income but negatively associated with maternal body mass index. Interestingly, erythrocyte riboflavin at 6 mo was higher among infants with elevated C-reactive protein levels, and the change in riboflavin status between 6 and 12 mo of age was positively associated with prevalence of fever. These positive associations between indices of infection and erythrocyte riboflavin could be due to malaria, as it has been shown that uptake and metabolism of riboflavin are elevated in erythrocytes infected with Plasmodium falciparum (Dutta 1991 ). The lack of response in erythrocyte riboflavin to the feeding of the vitamin and mineral-fortified food (WM) suggests that most of the study infants were not riboflavin-deficient.

Zinc status was assessed using plasma zinc concentration, which has limitations because of its poor sensitivity and specificity to changes in dietary zinc (King 1990 , Michaelsen et al. 1994 ) and the inability to adequately control for postprandial variation (Brown 1998 ) when conducting studies of infants. Currently no other indicators of zinc status have been validated. The significant negative association of plasma zinc at 6 mo of age with maternal height is difficult to interpret. It is possible that a prenatal influence of maternal height indirectly affects infant plasma status at 6 mo through infant postnatal growth (i.e., weight or length velocity from 1 to 6 mo). We examined this relationship using path analysis, but found no evidence of an indirect effect of maternal height on plasma zinc. Plasma zinc was inversely related to C-reactive protein level, which has been observed previously (Brown 1998 ).

Plasma zinc at 12 mo and the change in plasma zinc from 6 to 12 mo of age were both positively related to energy intake from complementary foods and negatively related to calcium intake from foods. The latter result is consistent with an inhibitory effect of dietary calcium on zinc absorption (Sandstrom 1997 , Wood and Zheng 1997 ). We were concerned that the high calcium content of the WM diet may have been responsible for this relationship, but after excluding infants in group WM from the analysis, the negative association of dietary calcium with plasma zinc was even stronger. This is puzzling because the average calcium intake from nonbreast milk foods in the remaining three groups was quite low (120 ± 95 mg). We did not observe a significant difference in the change in plasma zinc among the four intervention groups (Lartey et al. 1999 ), even though mean calcium intake from nonbreast milk sources in the WM group was 707 ± 379 mg. Although earlier studies in animals showed that high calcium intakes can impair zinc absorption (Forbes 1960 ), in humans the results have not been consistent (McKenna et al. 1997 , Wood and Zheng 1997 ) possibly due to differences in the types and amount of calcium used and the duration of supplementation. Further research is needed to investigate the potential effects, both positive and negative, of fortifying complementary foods with calcium or with food sources rich in calcium. Fortification of foods with multiple nutrients simultaneously may have different effects on nutrient absorption than would be observed in supplementation trials using single nutrients, especially if the single-nutrient supplements are consumed between meals.

To summarize, this study has identified several factors influencing the micronutrient status of breast-fed Ghanaian infants. Among the morbidity variables, fever was associated with anemia but positively associated with erythrocyte riboflavin, whereas diarrhea was related to lower vitamin A status. Seasonal differences were evident for most of the indicators of micronutrient status, and elevated C-reactive protein levels (indicative of recent infection) were related to lower hemoglobin, retinol and zinc concentrations but higher ferritin and erythrocyte riboflavin concentrations. Birth weight (or weight at 1 mo of age) was positively related to iron, zinc and vitamin A status, but a more rapid weight gain was associated with depletion of iron stores. Socioeconomic status was significantly related to anemia, riboflavin and zinc status. Intake of micronutrients from complementary foods was positively associated with ferritin and vitamin A status and intake of energy from complementary foods was positively associated with zinc status, but intake of calcium from foods was negatively related to zinc status. These results have important implications for controlling micronutrient deficiencies in developing countries. Improving micronutrient intake through fortification stands out as a viable medium-term strategy for some micronutrients (e.g., iron and vitamin A); however, the effectiveness of this strategy may be hampered by frequent infections and possibly by nutrient-nutrient interactions (e.g., calcium vs. zinc). Thus, to optimize effectiveness, intervention programs should combine improvement in diet quality with strategies to enhance utilization of nutrients, such as reduction in common childhood illnesses.

	ACKNOWLEDGMENTS

 
We are grateful to UNICEF/Ghana; the Noguchi Memorial Institute for Medical Research and the Department of Nutrition and Food Science, University of Ghana; The Holy Family Hospital; The Maternal and Child Health Center-Techiman office of the Ministry of Health; and the Techiman Weaning Food Project for making this study possible. We would like to thank the Techiman research team and the project mothers for their cooperation. We acknowledge the assistance of the Clinical Nutrition Research Unit (Grant # NIH-NIDDK 35747) at UC Davis in analyzing the blood samples, Janet M. Peerson for statistical advice and Elaine Ferguson, University of Otago, Dunedin, New Zealand, for making a nutrient database on Ghanaian foods available. This paper is dedicated to the memory of Alhassan Manu.

	FOOTNOTES

 
¹ Supported by the Nestle Foundation, a Rockefeller African Dissertation Internship Award to A. Manu, and a Fulbright Scholarship from the Fulbright program, USIA, to A. Lartey.

³ Abbreviations used: KF, koko with fish powder; MCH, Maternal and Child Health Center; W, Weanimix; WF, Weanimix with fish powder; WM, Weanimix fortified with vitamins and minerals.

Manuscript received August 26, 1999. Initial review completed October 14, 1999. Revision accepted October 20, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 

1. Abdalla S., Weatherall D. J., Wickramasinghe S. N., Hughes M. The anemia of P falciparum malaria. Br. J. Haem. 1980;46:171-183[Medline]

2. Arroyave G., Chester C., Flores H., Glover J., Mejía L. A., Olson J. A., Simpson K. L., Underwood B. A. In Biochemical Methodology for the Assessment of Vitamin A Status: A Report of the International Vitamin A Consultative Group 1982 Nutrition Foundation Washington, D.C.

3. Arthur P., Kirkwood B., Ross D., Morris S., Gyapong J., Tomkins A., Addy H. Impact of vitamin A supplementation on childhood morbidity in northern Ghana. Lancet 1992;339:361-362[Medline]

4. Bates C. J., Prentice A. M., Paul A. A., Prentice A., Sutcliffe B. A., Whitehead R. G. Riboflavin status in infants born in rural Gambia, and the effect of a weaning food supplement. Trans. R. Soc. Trop. Med. Hyg. 1982;76:253-258[Medline]

5. Beard J. One person’s view of iron deficiency, development and cognitive function. Am. J. Clin. Nutr. 1995;62:709-710[Free Full Text]

6. Beaton, G. H., Martorell, R., Aronson, K. J., Edmonston, B., McCabe, G., Ross, A. C. & Harvey, B. (1993) Effectiveness of vitamin A supplementation in the control of young child morbidity and mortality in developing countries. ACC/SCN State of the Art Series Nutrition Policy Discussion Paper No. 13: Geneva, Switzerland: Administrative Committee on Coordination-Sub Committee on Nutrition (ACC/SCN).

7. Binka F. N., Morris S., Ross D. A., Arthur P., Aryeetey M. E. Patterns of malaria morbidity and mortality in children in Northern Ghana. Trans. Royal Soc. Trop. Med. Hyg. 1994;88:381-385[Medline]

8. Black R. E. Therapeutic and preventive effects of zinc on serious childhood infectious diseases in developing countries. Am. J. Clin. Nutr. 1998;68(suppl):476S-479S[Abstract]

9. Bloland P. B., Lackritz E. M., Kazembe P. N., Were J. B., Steketee R., Campbell C. C. Beyond chloroquine: implications of drug resistance for evaluating malaria therapy efficacy and treatment policy in Africa. J. Infect. Dis. 1993;167:932-937[Medline]

10. Brown, K. H., Dewey, K. G. & Allen, L. H. (1998a) Complementary Feeding of Young Children in Developing Countries: A Review of Current Scientific Knowledge. WHO/NUT/98 1: World Health Organization, Geneva.

11. Brown K. H., Peerson J. M., Allen L. H. Effect of zinc supplementation on children’s growth: a meta-analysis of intervention trials. Bibliotheca Nutritio et Dieta 1998b;8:76-83

12. Brown K. H. Effect of infections on plasma zinc concentration and implications for zinc status assessment in low-income countries. Am. J. Clin. Nutr. 1998;68(suppl):425S-429S[Abstract]

13. Butrimovitz L. G. P., Prudy W. C. The determination of zinc in blood plasma by atomic absorption spectrometry. Anal. Chim. Acta 1977;94:63-73[Medline]

14. Campos F., Flores H., Underwood B. A. Effect of an infection on vitamin A status of children as measured by the relative dose response (RDR). Am. J. Clin. Nutr. 1987;46:91-94[Abstract/Free Full Text]

15. Dallman P. R., Siimes M. A., Stekel A. Iron deficiency in infancy and childhood. Am. J. Clin. Nutr. 1980;33:86-118[Free Full Text]

16. Dewey K. G., Cohen R. J., Landa Rivera L., Brown K. H. Effects of age of introduction of complementary foods on iron status of breast-fed infants in Honduras. Am. J. Clin. Nutr. 1998;67:878-884[Abstract]

17. Dutta P. Enhanced uptake and metabolism of riboflavin in erythrocytes infected with Plasmodium falciparum. J. Protozoology 1991;38:479-483

18. Edmond A. M., Hawkins N., Pennock C., Golding J., ALSPAC Children in Focus Team Haemoglobin and ferritin concentrations in infants at 8 mo of age. Arch. Dis. Child. 1996;74:36-39[Abstract]

19. Eyeson K., Ankrah E. K., Sundararajan A. R., Karinpaa A., Rudzka J. M. Composition of Foods Commonly Used in Ghana 1975 Food Research Institute Council for Scientific and Industrial Research, Accra.

20. Ferguson E. L., Gibson R. S., Opare-Obisaw C., Osei-Opare F., Stephen A. M., Lehrfeld J., Thomson L. The zinc, calcium, copper, manganese, non-starch polysaccharide and phytate content of seventy-eight locally grown and prepared African foods. J. Food. Comp. Anal. 1993;6:87-99

21. Fernandez F. J., Kahn H. L. Clinical methods for atomic absorption spectroscopy. Clin. Chem. Newsl. 1971;3:24-26

22. Fidanza F., Simonetti M. S., Floridi A., Codini M., Fidanza R. Comparison of methods of thiamin and riboflavin nutriture in man. Internat. J. Vit. Nutr. Res. 1989;59:40-47

23. Filteau S. M., Morris S. S., Abbot R. A., Tomkins A. M., Kirkwood B. R., Arthur P., Ross D. A., Gyapong J. O., Raynes J. G. Influence of morbidity on serum retinol of children in a community-based study in northern Ghana. Am. J. Clin. Nutr. 1993;58:192-197[Abstract/Free Full Text]

24. Floridi A., Plamerini C. A., Fini C., Pupita M., Fidanza F. High performance liquid chromatography analysis of flavin adenine dinucleotide in whole blood. Internat. J. Vit. Nutr. Res. 1985;35:187-191

25. Forbes R. Nutritional interactions of zinc and calcium. Fed. Proc. 1960;19:643-647[Medline]

26. Friis H., Mwaniki D., Omondi B., Muniu E., Magnussen P., Greissler W., Thiong’o F., Michaelsen K. F. Serum retinol concentrations and Schistosoma mansoni, intestinal helmiths, and malarial parasitemia: a cross-sectional study in Kenyan preschool and primary school children. Am. J. Clin. Nutr. 1997;66:665-671[Abstract/Free Full Text]

27. Friis H., Ndhlovu P., Kaondera K., Sandstrom B., Michaelsen K. F., Vennervald B. J., Christensen N. Q. Serum concentration of micronutrients in relation to schistosomiasis and indicators of infections: a cross-sectional study among rural Zimbabwean school children. Eur. J. Clin. Nutr. 1996;50:386-391[Medline]

28. King J. Assessment of zinc status. J. Nutr. 1990;120:1474-1479

29. Lartey A., Manu A., Brown K. H., Peerson J. M., Dewey K. G. A randomized community-based trial of the effects of improved, centrally processed complementary foods on growth and micronutrient status of Ghanaian infants from 6 to 12 mo of age. Am. J. Clin. Nutr. 1999;70:391-404[Abstract/Free Full Text]

30. Lartey A., Orraca-Tetteh R. Nutrient intake and milk yield of lactating rural Ghanaian women. Ghana. Med. J. 1990;24:43-48

31. Lie C., Ying C., En-lin W., Brun T., Geissler C. Impact of large-dose vitamin A supplementation on childhood diarrhea, respiratory disease and growth. Eur. J. Clin. Nutr. 1993;47:88-96[Medline]

32. Lozoff B., Jimenez E., Wolfe A. Long-term developmental outcome of infants with iron deficiency. N. Engl. J. Med. 1991;325:687-694[Abstract]

33. Mahalanabis D. Breastfeeding and vitamin A deficiency among children attending a diarrhoea treatment center in Bangladesh: a case-control study. Br. Med. J. 1991;303:493-496

34. Mahalanabis D., Simpson T. W., Chakraborty M. L., Ganguli C. Malabsorption of water-miscible vitamin A in children with giardiasis and ascariasis. Am. J. Clin. Nutr. 1979;32:313-318[Abstract/Free Full Text]

35. McKenna A. A., Illich J. Z., Andon M. B., Wang C., Matkovic V. Zinc balance in adolescent females consuming a low- or high calcium diet. Am. J. Clin. Nutr. 1997;65:1460-1464[Abstract/Free Full Text]

36. Michaelsen K. F., Samuelson G., Graham T. W., Lönnerdal B. Zinc intake, zinc status and growth in a longitudinal study of health Danish infants. Acta Paediatr 1994;83:1115-1121[Medline]

37. Murphy S. P., Beaton G. H., Calloway D. H. Estimated mineral intakes of toddlers: predicted prevalence of inadequacy in village populations in Egypt, Kenya and Mexico. Am. J. Clin. Nutr. 1992;56:565-572[Abstract/Free Full Text]

38. Newton C. R. J. C., Warn P. A., Winstanley P. A., Peshu N., Snow R. W., Pasvol G., Marsh K. Severe anaemia in children living in a malaria endemic area of Kenya. Trop. Med. Internat. Health 1997;2:165-178[Medline]

39. Phillips R. E., Pasvol G. Anemia of Plasmodium falciparum malaria. Bailliere’s Clin Haem 1992;5:313-330

40. Rosales F. J., Ritter J. J., Zolfaghari R., Smith J. E., Ross C. A. Effects of acute inflammation on plasma retinol, retinol binding protein, and its mRNA in the liver and kidneys of vitamin A-sufficient rats. J. Lipid Res. 1996;37:962-971[Abstract]

41. Sandstead H. H. Zinc deficiency–a public health problem?. Am. J. Dis. Child 1991;145:853-859[Abstract]

42. Sandstrom B. Bioavailability of zinc. Eur. J. Clin. Nutr. 1997;51:S17-S19

43. Shah P. S., Rajalakshmi R. Vitamin A status of newborn in relation to gestational age, body weight and maternal nutritional status. Am. J. Clin. Nutr. 1984;40:794-800[Abstract/Free Full Text]

44. Soewondo S., Husaini M., Politt E. Effects of iron deficiency on attention and learning processes in preschool children: Bandung, Indonesia. Am. J. Clin. Nutr. 1989;50:667-674

45. Sommer A., West K. P. Jr Vitamin A Deficiency. Health, Survival, and Vision. 1996 Oxford University Press, Inc. New York, NY.

46. Stephensen C. B., Alvarez J. O., Kohatsu J., Hardmeire R., Kennedy J. I., Gammon R. B. J. Vitamin A is secreted in the urine during acute infection. Am. J. Clin. Nutr. 1994;60:388-392[Abstract/Free Full Text]

47. van Hensbroek M. B., Morris-Jones S., Meisner S., Jaffar S., Bayo L., Dackour R., Phillips C., Greenwood B. M. Iron, but not folic acid, combined with effective antimalarial therapy promotes haematological recovery in African children after acute falciparum malaria. Trans. Royal Soc. Trop. Med. Hyg. 1995;89:672-676[Medline]

48. West K. P., Chirambo M., Katz J., Sommer A., Malawi Survey Group Breastfeeding, weaning patterns and the risk of xerophthalmia in Southern Malawi. Am. J. Clin. Nutr. 1986;44:690-697[Abstract/Free Full Text]

49. Wharf S. G., Fox T. E., Fairweather-Tait S. J., Cook J. D. Factors affecting iron stores in infants 4–18 mo of age. Eur. J. Clin. Nutr. 1997;51:504-509[Medline]

50. Wood R. J., Zheng J. J. High dietary calcium intakes reduce zinc absorption and balance in humans. Am. J. Clin. Nutr. 1997;65:1803-1809[Abstract/Free Full Text]

51. The World Bank Enriching Lives: Overcoming Vitamin and Mineral Malnutrition in Developing Countries 1994 World Bank Washington D.C.

52. World Health Organization Measuring Change in Nutritional Status 1983 WHO. Geneva

53. Yassai M. B., Malek F. Newborns vitamin A in relation to sex and birth weight. J. Trop. Pediatr. 1989;35:247-249[Abstract/Free Full Text]

This article has been cited by other articles:

				K. Kongsbak, M. A. Wahed, H. Friis, and S. H. Thilsted Acute Phase Protein Levels, T. trichiura, and Maternal Education Are Predictors of Serum Zinc in a Cross-Sectional Study in Bangladeshi Children J. Nutr., August 1, 2006; 136(8): 2262 - 2268. [Abstract] [Full Text] [PDF]

				K. G. Dewey, R. J. Cohen, and K. H. Brown Exclusive Breast-Feeding for 6 Months, with Iron Supplementation, Maintains Adequate Micronutrient Status among Term, Low-Birthweight, Breast-Fed Infants in Honduras J. Nutr., May 1, 2004; 134(5): 1091 - 1098. [Abstract] [Full Text] [PDF]

				T. A Strand, R. K Adhikari, R. K Chandyo, P. R Sharma, and H. Sommerfelt Predictors of plasma zinc concentrations in children with acute diarrhea Am. J. Clinical Nutrition, March 1, 2004; 79(3): 451 - 456. [Abstract] [Full Text] [PDF]

				S. Zlotkin, P. Arthur, C. Schauer, K. Y. Antwi, G. Yeung, and A. Piekarz Home-Fortification with Iron and Zinc Sprinkles or Iron Sprinkles Alone Successfully Treats Anemia in Infants and Young Children J. Nutr., April 1, 2003; 133(4): 1075 - 1080. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lartey, A. Articles by Dewey, K. G. Search for Related Content PubMed PubMed Citation Articles by Lartey, A. Articles by Dewey, K. G.