Plasma free iron: a possible cause of oedema in kwashiorkor.

BACKGROUND: Oedema is a sine qua non for the diagnosis of kwashiorkor yet the mechanisms leading to oedema remain ill defined. AIMS: To relate the plasma concentration of radical promoting 'free' iron to the degree of oedema in patients with kwashiorkor. SETTING: University teaching hospital. PATIENTS: Fifteen children with kwashiorkor, nine of whom had severe and six of whom had a moderate degree of oedema. METHODS: Plasma 'free' iron was measured as bleomycin detectable iron (BDI) and related to severity of oedema and plasma albumin concentration. RESULTS: BDI was significantly higher in the patients with severe oedema (20.5 v 6.75 mumol/l) whereas the albumin concentrations were similar (16 v 17 g/l). BDI was no longer present in any patients 30 days after admission. CONCLUSIONS: 'Free' circulating iron may contribute to the oedema of kwashiorkor, and its sequestration could hasten recovery and decrease morbidity and mortality.

Bone marrow and chelatable iron in patients with protein energy malnutrition.

OBJECTIVES: To examine the iron status of malnourished children by comparing bone marrow iron deposits in children with protein energy malnutrition with those in well-nourished controls, and measuring chelatable urinary iron excretion in children with kwashiorkor. DESIGN: Bone marrow iron was assessed histologically in postmortem specimens from children with kwashiorkor or marasmus, and from controls. Twenty-four-hour urinary iron was measured in children with severe kwashiorkor, half of whom received 10 mg/kg of intramuscular desferrioxamine (DFO) on admission. SETTING: Red Cross War Memorial Children's Hospital, Cape Town. SUBJECTS: Thirteen children with kwashiorkor, 6 with marasmus and 16 well-nourished children underwent bone marrow examination. Urinary iron excretion was assayed in 17 children with kwashiorkor. RESULTS: Stainable iron was present in the bone marrow of half the children with kwashiorkor but in only 1 child in each of the other groups. The median iron excretion was 945.5 micrograms/24 hours in the DFO group compared with 28.5 micrograms/24 hours in the non-DFO group. CONCLUSIONS: There is an apparent excess of iron which may predispose to bacterial infections and free radical-mediated injury in children with kwashiorkor.

Misplaced iron in kwashiorkor.

OBJECTIVES: To determine the presence of radical promoting iron (non-protein-bound or loosely bound or free iron) in the plasma of children with kwashiorkor. DESIGN: The bleomycin assay was employed for the quantitation of free or loosely bound iron. SETTING: The Red Cross War Memorial Children's Hospital, Cape Town, Tertiary Care. SUBJECTS: Fifty children on admission with kwashiorkor: six with marasmus and twelve healthy well-nourished controls. RESULTS: Non-protein-bound iron was detected in the plasma of 58% of children with kwashiorkor but was absent in marasmic and healthy well-nourished children. CONCLUSIONS: The presence of radical promoting iron supports the hypothesis that a free radical injury probably plays a role in the pathogenesis of kwashiorkor and its removal may improve mortality.

Red blood cell antioxidant enzyme concentrations in kwashiorkor and marasmus.

Kwashiorkor may occur when an imbalance between pro- and antioxidants in malnourished children results in an excess of free radicals. The concentrations of the antioxidant enzymes catalase (CAT), superoxide dismutase (SOD), reduced glutathione (GSH) and glutathione peroxidase (GPX) were measured in erythrocytes of 22 children with kwashiorkor on admission to hospital and repeated on days 5, 10 and 30 of recovery. The concentrations were compared with those in 22 children with marasmus and in 20 children who were normally nourished but had infective illness necessitating their hospitalization. CAT and SOD were similar in all groups and did not change during recovery. GSH and GPX were significantly lower in kwashiorkor than in the other groups. Concentrations of thiobarbituric acid-reactive substances (TBARS), a marker of lipid peroxidation, were significantly elevated in children with kwashiorkor. During clinical recovery, GSH but not GPX concentrations rose despite an increase in plasma selenium levels and decreased concentrations of TBARS. These findings suggest that the antioxidant status of children with kwashiorkor differs from that of well nourished and marasmic children. Whether these differences are the cause of the consequence of the clinical picture is unresolved.

Plasma zinc, copper, selenium, ferritin and whole blood manganese concentrations in children with kwashiorkor in the acute stage and during refeeding.

Plasma zinc, copper, selenium, ferritin and whole blood manganese concentrations were measured in 22 children with kwashiorkor on admission to hospital and on days 5, 10 and 30 of refeeding. Twenty similarly aged, healthy, well nourished children served as controls. The mean (SEM) zinc, copper and selenium concentrations of 7.5 (0.93), 10.8 (0.64) and 0.29 (0.02) mumol/l, respectively, in the children with kwashiorkor on admission were all significantly lower than the values of 13.7 (0.66), 25.6 (1.72) and 0.72 (0.04) mumol/l in the controls. In contrast, the erythrocyte manganese level of 1.67 (0.09) micrograms/gHb and the median ferritin concentration of 293 micrograms/dl were significantly higher than in the controls. After 30 days there was full clinical recovery with significant weight gain and a return of the plasma albumin, caeruloplasmin, copper and ferritin to normal. However, manganese remained elevated and zinc and selenium concentrations remained significantly low. Our results suggest that nutritional rehabilitation of children with kwashiorkor is incomplete by 30 days and cannot be judged purely by a return of the plasma proteins to normal. Addition of selected trace elements to the diet may hasten full recovery.

Iron fortification of infant milk formula: the effect on iron status and immune function.

We conducted a randomized double-blind trial of a cow's milk infant formula with increased iron fortification in order to confirm its safety and to measure its effects on iron status and immune function. A group of full-term, well nourished and healthy infants was followed from the age of 3 months to 1 year. A control group of 74 infants was given a commercially available infant formula containing 8.3 mg Fe/100g. The test group of 75 infants received a similar formula with 40 mg Fe/100 g. The formula with the extra iron proved to be safe and, when compared with the control group, the children in the test group had significantly improved iron status as reflected by the proportion of children classed as normal (25 of 61 cf. 44 of 65; p less than 0.003), and by the mean values of the haemoglobin concentration (11.5 cf. 11.9 g/dl; p = 0.04), red cell distribution width (15.5% cf. 14.4%; p = 0.0005), red cell zinc protoporphyrin (3.4 cf. 4.0 micrograms/g Hb; p = 0.04) and ferritin (29 cf. 17.3 micrograms/l; p = 0.004). The extra iron fortification depressed zinc concentration in plasma (90.6 cf. 83.5 micrograms/l; p = 0.05). There was no significant difference between the two groups for laboratory measures of immune function or for incidence of infection. No adverse effects such as infection could be attributed to the increased iron. We conclude that iron fortification of cow's milk infant formula may be safely increased to 40 mg/100 g (i.e. by a factor of 4.8 over the common concentration of 8.3 mg/100 g), but that this has less than the expected effect on iron status. Further studies are required to define (a) the long-term role of facilitators of iron absorption such as ascorbic acid, (b) the interaction of iron with absorption of divalent trace elements such as zinc, and (c) the effect of iron status on immune function and susceptibility to infection.

Prevention of iron deficiency in preterm neonates during infancy.

The preterm infant inevitably develops iron deficiency unless supplementary iron is given. Oral iron supplementation is preferred in ideal social circumstances but, where compliance with such therapy is uncertain, intramuscular iron dextran may be a more effective treatment. A study was conducted to compare the effectiveness of two methods of preventing iron deficiency of prematurity. One group of healthy premature infants was given oral iron 2 mg/kg/d until the age of 6 months. The second similar group was given 100 mg as intramuscular iron dextran (Imferon; Fisons) between the ages of 6 and 8 weeks. Both kinds of supplementary iron appeared to have benefited the majority of infants in this trial.

Reference concentrations of serum selenium and manganese in healthy nulliparas.

Reference serum selenium and manganese concentrations were established for healthy nulliparas aged 18-23 years resident in Cape Town. Measurements were determined for selenium in 100 female students who had been taking low-dosage triphasic contraceptive medication for a minimum of 3 months and in 100 female students who were not on contraceptive therapy. Manganese concentrations were determined for 25 female students from each group. The mean serum selenium concentrations were 0.988 +/- 0.189 micrograms/l (78 +/- 15 micrograms/dl) and 0.925 +/- 0.177 mumol/l (73 +/- 14 micrograms/l) respectively for females taking and not taking oral contraceptives. The corresponding concentrations for manganese were 21.84 +/- 9.82 nmol/l (1.20 +/- 0.54 micrograms/l) and 21.66 +/- 7.64 nmol/l (1.19 +/- 0.42 micrograms/l) respectively. The differences in selenium were statistically significant (P = 0.0231) but not for manganese (P = 0.910).

Reference values for serum copper, ceruloplasmin and zinc and haematological indices for healthy nulliparous females.

Reference serum copper, ceruloplasmin and zinc values were established for 100 healthy white nulliparous students aged 18-23 years resident in Cape Town who had been taking low-dosage triphasic contraceptives for a minimum period of 3 months, and in 100 female students not taking contraceptives. The mean serum copper values were 26.5 +/- 4.2 mumol/l and 16.9 +/- 2.7 mumol/l for those taking and not taking oral contraceptives respectively; corresponding values for ceruloplasmin were 181 +/- 43.9 IU/ml and 110 +/- 22.7 IU/ml respectively. Both differences were statistically significant. Serum zinc values for those on contraceptives were 14.1 +/- 2.1 mumol/l and for the others 14.7 +/- 2.0 mumol/l. There were no differences in the haematological parameters except for a significantly higher mean corpuscular volume in females taking oral contraceptives. Of possible clinical significance in this student population are prevalence rates of 2.2% for anaemia (haemoglobin value less than 11.5 g/dl), 7% for iron deficiency (serum ferritin less than 12 micrograms/l) and 6.6% for iron depletion (serum ferritin 12-20 micrograms/l).

Ferritin levels in human milk.

Ferritin was detected and quantitated in breast milk from 24 mothers of healthy fullterm infants 3 days, 7 days, 6 weeks and 3 months after delivery. Highest concentrations were found at day 3 and demonstrated a marked decline at day 7, with negligible levels at 6 weeks and 3 months. The values in breast milk were compared with mother's serum ferritin concentration at delivery and a significant but weak correlation (r = 0.475, P less than 0.05) was found between the paired values of maternal serum ferritin at delivery and breast milk ferritin at day 3. The possible biological significance of ferritin in breast milk is discussed.

Serum zinc and copper levels in healthy middle and higher socio-economic class infants in Cape Town. A longitudinal study during the 1st year of life.

Serum zinc and copper levels were measured using a flame atomic absorption technique in 55 healthy infants and their middle and upper socio-economic class mothers at delivery, and also longitudinally in the infants during the first 12 months of life. Mean cord serum zinc levels (99.50 +/- 27.63 micrograms/100 ml) were 1.5 times higher than maternal serum levels at birth (66.3 +/- 21.0 micrograms/100 ml). The mean serum zinc levels showed no significant change (P greater than 0.05) at each of the study periods. However, the mean serum zinc levels at birth were significantly higher (P less than 0.05), than those after 6 weeks. The infants' levels remained within the normal adult range at all the study periods. The mean cord serum copper levels (34.46 +/- 12.60 micrograms/100 ml) were about one-sixth of the mean maternal serum copper levels at delivery (217.7 +/- 64.39 micrograms/100 ml). The mean serum copper levels increased significantly (P less than 0.05) from birth to 12 months of age. Mean serum copper levels similar to those of normal adults were reached at 12-24 weeks.

Serum zinc and copper levels in healthy South African women at term and during the first year after childbirth.

Serum zinc and copper levels were measured longitudinally in 55 healthy middle and upper socio-economic group white mothers living in Cape Town from the birth of the baby to 12 months after parturition. Mean maternal serum zinc levels showed a significant rise (P less than 0.05) from delivery (66.34 +/- 21.07 micrograms/100 ml) to 12 weeks (87.88 +/- 15.93 micrograms/100 ml), but no further changes were detected at 24, 36 and 52 weeks after parturition. The mean maternal serum copper levels decreased from 217.73 +/- 64.34 micrograms/100 ml at delivery to 141.65 +/- 45.60 micrograms/100 ml at 12 weeks (P less than 0.05); they remained constant at all the other sampling periods. No differences (P greater than 0.05) were noted at all the different sampling periods between mean serum zinc levels of primiparous and multiparous mothers, but mean serum copper levels were significantly higher (P less than 0.05) in the primiparas. The mean serum zinc and copper levels of healthy white South African mothers at delivery and 12 months after parturition correspond with those for Northern American mothers. Normal non-pregnant adult serum zinc and copper levels were attained within 12 weeks of delivery.

Blood lead levels in preschool children in Cape Town.

Blood lead levels were assessed in 293 children aged between 4 and 6 years attending preschool centres in metropolitan Cape Town in order to establish the degree of lead absorption. Anthropometric data, blood count, zinc protoporphyrin and blood lead level were obtained for each child. A questionnaire was used to determine socio-economic status, dietary habits and history of pica. Thirteen children, or 4,4% of those sampled, had blood levels of greater than or equal to 30 micrograms/dl. The majority of these children lived in close proximity to one another in a socially deprived inner urban environment. Environmental sampling for lead was carried out in the homes of children with the highest blood levels as well as in the homes of a matched control group with low levels living in the same area. The only difference was a significantly higher incidence of pica in the children with high levels.

Zinc and copper levels in the breast-milk of Cape Town mothers.

Zinc and copper levels were measured longitudinally in the breast-milk of 55 healthy South African women from the middle and upper socio-economic classes. The mean foremilk zinc levels showed a significant fall (P less than 0,05) from 652,9 +/- 240,7 micrograms/dl at 3 days after delivery to 413,3 +/- 143,0 micrograms/dl at 7 days and 228,1 +/- 136,1 micrograms/dl at 6 weeks. Lower foremilk zinc levels were obtained at 36 weeks (64,3 +/- 50,0 micrograms/dl). In contrast, foremilk copper levels were low at 3 days (57,0 +/- 74,8 micrograms/dl), with a gradual decline to very low levels (28,0 +/- 29,7 micrograms/dl) at 36 weeks. No differences were detected between the fore- and hindmilk zinc and copper levels at any stage of lactation studied (P greater than 0,05).

Serum zinc and copper levels in the 1st year of life.

Zinc and copper concentrations in the serum of 240 healthy infants aged 1-12 months were measured by the atomic absorption spectrophotometric method. The study was designed so that on completion there would be 20 infants in each of 12 groups according to age by month. The mean serum zinc levels remained fairly constant in all the age groups except for a significant fall from levels at 1 month (77,11 +/- 12,85 micrograms/dl) to those at 2 months of age (66,60 +/- 15,18 micrograms/dl) (P less than 0,025). Another significant decline in serum zinc levels was present between 4 and 5 months of age (P less than 0,05), which was followed by a significant increase between 5 and 6 months of age (P less than 0,05). The infants' mean serum zinc levels were lower than those of normal adults at all ages (P less than 0,05). The mean serum copper levels were low at 1 month of age (71,17 +/- 17,30 micrograms/dl), and increased significantly between 1 and 2 months (P less than 0,001) and 3 and 4 months of age (P less than 0,02). Mean serum copper levels similar to those of adults were reached at 4 months of age. These levels remained fairly constant from 5 months of age onwards, the only significant decline being noted at between 7 and 8 months of age (P less than 0,05). The present study substantiates previous findings that a specific pattern of change in serum copper concentrations which differs from the trend in serum zinc levels occurs in infants after birth. To detect zinc and copper depletion it is important to interpret serum concentrations in relation to normal values for age.

Serum-IgE levels in rural Namibian infants.

Serum IgE levels were examined in 237 infants ages 2 wk to 12 mo in a remote rural area of Namibia. There was a wide range of values (0.5 to 884 IU/ml). The highest value in the first month of life was 295 IU/ml. Median values for the age groups 2 wk to 3 mo, 3 to 6 mo, and 6 to 12 mo were higher than those reported from Western countries. Values in general increased with age, but only 4% of the variation is explicable on this basis. There were differences between median values for male and female infants, but they did not reach statistical significance at the 5% level. Levels between Baster and Nama infants, the main ethnic groups in the area, did not differ. The median IgE levels in breast-fed infants, although high, tended to be lower than those in weaned infants. The differences were not statistically significant (p greater than 0.05). Investigations did not suggest that parasitic infestations or atopy were of significant importance. It was not possible to identify the factor(s) responsible for the high IgE levels. They must have been operative from very early life. However, the basic immunologic mechanisms involved are presumably genetically determined and similar to those responsible for the corresponding high IgG, IgA, and IgM levels reported in infants from developing communities.

PIP: Serum IgE levels were exminaed in 237 infants ages 2 weeks-12 months in a remote rural area of Namibia. There was a wide range of values (0.5-884 IU/ml). The highest value in the 1st month of life was 295 IU/ml. Median values for the age groups 2 weeks-3 months, 3-6 months, and 6-12 months were higher than those reported from the Western countries. Values in general increased with age, but only 4% of the variation is explicable on this basis. There were differences between median values for female and male infants, but they did not reach statistical significance at the 5% level. Levels between Baster and Nama infants, the main ethnic groups in the area, did not differ. The median IgE levels in breastfed infants, although high, tended to be lower than those in weaned infants. The differences were not statistically significant (P0.05). Investigations did not suggest that parasitic infestations or atopy were of any significance. It was not possible to identify those factors resposnible for the high IgE levels. Perhaps they were operative from a point early in life. However, the basic immunologic mechanisms involved are presumably genetically determined and similar to those responsible for the corresponding high IgG, IgA, and IgM levels reported in infants from developing communities.

The prevalence of iron deficiency in apparently healthy Cape coloured infants.

In order to obtain a more accurate picture of the prevalences of anaemia, iron deficiency anaemia, haematological iron deficiency and diminished iron stores, 240 Cape Coloured infants from the lower socio-economic groups were studied. Anaemia was diagnosed in 42 infants (17,5%), iron deficiency anaemia in 81 (34%), and haematological iron deficiency in 28%; 64 (27%) showed evidence of diminished iron stores. The findings indicate that iron deficiency was a common problem in the infants studied, and the same probably also applies to the community at large.

Free erythrocyte protoporphyrin levels in the first year of life.

Free erythrocyte protoporphyrin (FEP) levels were measured in 161 infants judged not to be iron-deficient. In addition to high mean FEP levels in cord blood, a significant number of infants had FEP levels in excess of 120 micrograms/dl red blood cells during the first 6 months of life. These levels are comparable to those found in iron-deficient adults. It is concluded that the clinical significance of an FEP level in excess of 120 micrograms/dl and its value as a marker of early iron deficiency during the first 6 months of life require further study. FEP values in excess of 120 micrograms/dl were not observed in infants aged 7-12 months.