doi:10.1111/j.1440-1754.2006.00932.x

REVIEW ARTICLE

Critical appraisal of the management of severe malnutrition: 2. Dietary management David R Brewster Fiji School of Medicine, Suva, Fiji

Abstract: In the dietary management of severe acute malnutrition in children, there is evidence to support the WHO Manual’s protocol of cautious feeding of a low energy and protein formula with small frequent feeds in the initial phase of treatment, particularly in kwashiorkor. However, this initial milk diet (WHO F-75) might benefit from increasing the sulphur amino acid, phosphorus and potassium content and reducing the lactose content, but further studies are needed. Careful tube-feeding results in faster initial recovery and weight gain, but has a significant risk of aspiration in poorly supervised settings. Ready-to-use therapeutic food is an important recent advance in the dietary management of malnutrition in ambulatory settings, allowing more effective prevention programmes and earlier discharge from hospital where community follow-up is available. It should be included in future protocols. There is very good evidence on the use of micronutrients such as zinc, and preliminary evidence suggests that smaller doses of daily vitamin A are preferable to a single large dose on admission for severe malnutrition. Key words:

diet; high-energy milk; malnutrition; micronutrients; RUTF.

The basic principles of hospital treatment of severe malnutrition, particularly for kwashiorkor, were worked out in specialised research units in Jamaica (Tropical Metabolism Research Unit, TMRU), Guatemala (Institute of Nutrition of Central America and Panama), Cape Town (JD Hansen and colleagues) and Kampala (UK Medical Research Council) more than 25 years ago. Briefly, it became clear that high-protein diets, salt supplements (for hyponatraemia) and fluid intakes >150 mL/kg/day in the initial phase resulted in high mortality rates, whereas potassium supplements, antibiotics and careful feeding regimes (commencing with low volumes of dilute milk and grading up in volume and density towards a high-energy milk formula for rapid catch-up growth) greatly reduced mortality. On high-energy milk diets (5.8 kJ/mL), mean rates of weight gain without complications or infections were 8.4 g/kg/day

Key Points 1 There is some evidence to support increasing the sulphur amino acid, phosphorus and potassium content of the initial milk diet to treat severe malnutrition, and also to reduce the lactose and osmolality. 2 Ready-to-use therapeutic food warrants inclusion into management protocols for severe malnutrition. 3 Smaller repeat doses of vitamin A may be better than an initial large dose in severe malnutrition.

Correspondence: Professor David Brewster, Fiji School of Medicine, Private Mail Bag, Suva, Fiji Islands. Fax: +67 9 3305 781; email: [email protected] Accepted for publication 22 May 2006.

with an energy cost of growth of 13.9 kJ/g. Following these early studies, TMRU1 and World Health Organization (WHO)2 developed protocols for the management of severe malnutrition.

What Is the Best Diet for Severe Malnutrition? In this section, we address issues such as whether higher-energy diets with additional sucrose and vegetable oil result in higher weight gain compared with milk alone (Table 1), and whether lowlactose and low-osmolality diets improve outcomes in malnutrition. In many developing countries, the low-energy density of weaning foods is a major contributor to growth faltering and malnutrition. Although weaning diets may be adequate when children are healthy, they fail to allow rapid catch-up growth after diarrhoea and other infections.3 Breast milk is an important source of fat and essential fatty acids up until the end of the second year of life because weaning foods tend to be low in fat, leading to a difficult transition after weaning with inadequate fat intake during recovery from infections.

Feeding schedules There is good evidence for the effectiveness of nasogastric tubefeeding in anorexic children in the hospital context in terms of rapid catch-up growth and shortening hospital length of stays. A Malawi trial by the author compared hospital treatment with and without nasogastric tube-feeding at two sites, with no other differences in protocol, and found more rapid weight gain (8.2 vs. 4.5 g/kg/day) in the nasogastric tube-fed children with kwashiorkor.4 However, the rate of weight gain during rehabilitation varies greatly between marasmus and kwashiorkor, as the latter will lose weight initially from loss of oedema, so usually only commence to gain weight in the second week of rehabilitation. Tube-feeding in a hospital setting is

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

575

Dietary management

Table 1

DR Brewster

Comparison of milk diets

Diet

Lactose (g/L)

Protein (g/L)

Energy kJ (kcal)/L

Sodium/ potassium (mmol/L)

F-75 F-100 Human milk Cow milk Goat milk

13 42 74 48 48

9 29 10 34 34

3150 (750) 4200 (1000) 2911 (693) 2814 (670) 3087 (735)

6/36 19/59 6/15 21/38 20/50

important for overcoming anorexia and for rapid weight gain, but has dangers of overfeeding resulting in cardiac failure or aspiration pneumonia so demands good nursing care. Feeding children 2– 4 hourly is difficult for mothers or nursing staff without tube-feeding. Even after anorexia and infection have resolved, weight gains of 10– 15 g/kg/day are often not achieved in hospitalised children without tube-feeding. There may be a trade-off between these dangers of tube-feeding, and the danger of anorexia with hypoglycaemia, slow improvement and even dehydration from poor intake. In a hospital setting, both tube-feeding and interosseous infusions are important medical interventions that can greatly benefit malnourished children when used appropriately. The WHO Manual recommends 4-hourly feeds night and day during initial treatment (p. 15) and during rehabilitation, and often mentions 2–3 hourly feeds in other contexts. Although this may well be optimal, it is not feasible in small hospitals with limited nursing care at night. It is very difficult to get nursing staff to prepare and give more than about five feeds per day in most settings, especially during the rehabilitation phase. During the initial phase of treatment, it is often possible to give 3- or 4-hourly feeds during working hours with possibly one middle-of-the-night feed, particularly for very sick children. Although continuous nasogastric tube-feeding (especially overnight) may result in more rapid weight gain, it demands close nursing supervision to prevent overfilling of the stomach with risk of aspiration. Bolus feeding is preferable. In a different context, a Cochrane review concluded that the clinical benefits and risks of continuous versus intermittent nasogastric tube milk feeding cannot be reliably discerned from randomised trials to date.5

Initial diet The WHO F-75 milk formula (75 kcal/100 mL) used in the ‘initiation of cure’ phase has a low-protein content (4.8% of energy as protein), which does not meet the phosphorus requirements of 2 mmol/kg/ day (60 mg) recommended by the WHO Manual (Appendix 5), providing only about 1.16 mmol/kg/day (35 mg) with an intake of 135 mL/kg. The rationale for using such a very-low-protein diet initially is unclear. Based upon Whitehead’s work in Uganda, the recommended protein intake in the 1977 TMRU book (p. 110) and 1981 WHO Manual (p. 21) (which were both after the so-called ‘protein fiasco’6) was 3.0–3.3 g/kg/day.7,8 However, in Waterlow’s 1992 book, this dropped to 0.7 g/kg/day for the ‘initiation of cure’ with intakes of 120 mL/kg/day (p. 173), increasing in the rehabilitation phase to 5.75 g/kg/day on intakes of 150 mL/kg/day of milk-oil formula.9 The evidence for this change was not explained by Waterlow, but there 576

was clearly a clinical need for an ‘initiation of cure’ diet for severely ill children, especially those with kwashiorkor, who would not tolerate high-energy milk. The proposed diet was intended as an improvement on half-strength cow milk ± oral rehydration solution, which was in common use. More recently, Golden has justified this initial low-protein diet in terms of the degree of hepatic derangement and presence of abnormal amino acid metabolites in the urine of kwashiorkor cases.10 He also cites the study of high-protein diets in adults by Collins et al.,11 but this study (see later) compared a diet providing 16.4% of energy as protein with one providing 8.5%, not the 4.8% of F-75. The alleged toxic metabolic effects of amino acids in sick malnourished children hardly justify the extent of protein restriction in F-75, which may only be appropriate for a small minority of severe kwashiorkor cases. The disadvantage of the low protein content of F-75 is that it has a low sulphur amino acid content, and cysteine is needed for glutathione synthesis. It has been shown recently that cysteine supplementation has favourable consequences on oedema clearance in kwashiorkor patients.12 This study, however, was carried out with children receiving a diet based on a high whey infant formula with a higher cysteine content than F-75. In addition, the low protein content of a milk-based diet like F-75 is also low in phosphorus, and low phosphorus status is associated with an increased risk of dying in severe malnutrition.13,14 So there is room to improve F-75 by increasing sulphur amino acids, phosphorus and potassium content. Although the evidence is not robust, reducing the lactose content of F-75 would arguably save many more lives than this low-protein content. There is a need for good studies comparing F-75 with a comparable lower-lactose, higher-protein and lower-osmolality formula in hospitalised children with severe malnutrition, including cases of both kwashiorkor and marasmus. Although of only marginal relevance, a UK study in an intensive care setting found no need for extra energy intake in critically ill infants and children with sepsis.15 According to the study, these children are not hypermetabolic and do not require higher energy intake than stable non-septic patients. Overfeeding these patients may have detrimental effects on liver metabolism and respiratory dynamics. This is consistent with accepted practice in the developing world of ‘initiation of cure’, where high-energy milk is delayed until the child’s clinical condition is stable.

High-energy milk diets Once a child with severe malnutrition is clinically stable (e.g. resolution of oedema, diarrhoea and sepsis), a high-energy milk formula is introduced to promote rapid catch-up growth. These formulas are not tolerated in the initial phase of treatment (especially in kwashiorkor) for the reasons given above and also because of its high lactose content (42 g/L for F-100). High-energy milk diets such as the WHO F-100 and milk-oil formula also have high osmolarity, whereas lactose-free commercial formulas use maltodextrins or glucose polymers with medium chain triglycerides to lower the osmolality and improve tolerance. Many hospitals and feeding centres in the developing world do not have access to F-75 and F-100 diets, so have to mix their own formulas of milk-oil-sugar and water, and very few have access to expensive lactose-free commercial formulas such as O-Lac (Mead-Johnson) or DeLact (Nutricia). A study from Peru reported the effects of increased energy density of foods in children recovering from malnutrition by varying the

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

DR Brewster

energy density of diets from 1.67 to 6.28 kJ/g by means of non-fat additives.16 Although there were some compensatory decreases in the quantity of food consumed, the overall energy intake was more than twice as great with the highest energy density formulation than with the lowest, confirming the importance of energy density in achieving high-energy intakes in malnourished children. Arora et al. carried out a randomised controlled trial in 39 malnourished Indian children comparing low- versus high-fat mixed diets (22% vs. 48% of energy as fat) with a fixed volume of 185 mL/kg/day such that the high-fat group were receiving 50% higher energy intake.17 The outcomes were the coefficients of absorption of energy, nitrogen and carbohydrate, which were similar in the two groups. Median fat absorption was 7% higher (96.3% vs. 89.3%, P < 0.01) in the high-fat group of children, who retained 50% more energy compared with the low-fat group. A Guinea-Bissau study showed that children with persistent diarrhoea can be managed with home-based dietary treatment with a single low-lactose, cereal and milk-based diet consisting of locally available foods, micronutrient supplementation and antibiotic treatment when indicated.18 Although no significant effect on diarrhoeal morbidity was shown, the intervention did have a beneficial effect on both ponderal and linear growth. There were 141 episodes of persistent diarrhoea in 110 children during the study. During the relatively short intervention period (about 17 days), weight gain in the treatment group exceeded that of the control group by 61.5 g/ week (49.2–73.8), whereas there was no significant difference in linear growth. At a median follow-up period of 6.6 months after the intervention, there was greater weight gain and length gain of 12.5 g/week (7.7–17.3) and 0.65 cm/year (0.11–1.19), respectively, in the treatment than control group. Thus, therapeutic feeding and micronutrient supplementation had an immediate and sustained beneficial effect on growth in children with persistent diarrhoea. An Indian community study of 184 preschool children compared the acceptability of three different foods, namely two sweet ready mixes (wheat-soy-sugar with and without amylase, energy density 1638 kJ/100 g) and therapeutic food (1974 kJ/100 g).19 Limitations of the study were lack of random allocation and most subjects were not malnourished. The key outcome (mean consumption of the diet) was 46.6, 52.9 and 68.1 g/day by dietary group, respectively, for the 50 malnourished children, which was significantly higher for the more energy-dense therapeutic food, but only when all 184 children were included (P < 0.001). Collins et al. compared two diets in 573 severely malnourished Somali adults during the 1992 famine, namely a high-protein diet of milk, soy, flour, oil, sugar, rice, beans and biscuit (158 g protein/day, 16.4% of energy) versus a low-protein diet of milk, oil, sugar formula (MOF) with rice and banana (82 g protein/day, 8.5% of energy).11 Two other key differences in the diets were 137 g lactose and 70 g of vegetable oil in the high-protein diet compared with 95 g of lactose and 150 g of oil per day in the low-protein diet. Although diets were not allocated randomly, the case-fatality rates for marasmic versus oedematous cases for the high-protein diet was 22% (64/291) versus 52% (14/27) whereas for the low-protein diet it was 13% (11/86) and 25% (14/56), respectively. Anecdotally, the oedematous cases had improved appetite and faster loss of oedema on the low-protein diet, but still had problems with diarrhoeal symptoms of probable lactose intolerance requiring dilution of the milk-oil formula. Although the authors downplay the importance of lactose intolerance (because the high-protein diet did

Dietary management

not result in more diarrhoea despite its higher lactose content), there were only 27 oedematous cases on the high-protein diet due to biased allocation, so this may not be a valid observation for this group. This methodological flaw also limits the strength of the findings of this study. These findings are not surprising to paediatricians with experience in treating kwashiorkor, as the danger of highprotein diets is well documented.9,20–28 Collins and Sadler also carried out a retrospective study of emergency relief outpatient management of 170 severely malnourished children 6–120 months of age (median 36 months) during the Ethiopian famine.29 The overall median outcomes (interquartile range) were: case-fatality 4% (but another 11% defaulted or were transferred), length of stay 42 days (28–56) and weight gain 3.2 g/kg/day (1.9–5.6). For marasmus, marasmic kwashiorkor and kwashiorkor, the lengths of stay were 42, 56 and 42 days and the weight gain 4.8, 4.0 and 2.7 g/kg/day, respectively. It is important to note that these results relate to emergency relief of large numbers of malnourished children due to food insecurity, and cannot necessarily be extrapolated to hospitals and nutrition rehabilitation centres in the developing world, which is the focus of this review. In a study of adults during the Somali famine, Collins and Myatt showed that models based on clinical signs predicted death better than body mass index.30 Of the eight clinical signs, severe oedema (odds ratio 2.45; 1.41–4.27), apparent dehydration (2.73; 1.60–4.66) and inability to stand (2.96; 1.40–6.26) were independently associated with mortality. The most useful clinical model was that based on the presence of any one of these three signs, with a sensitivity of 77% and a specificity of 59%. Although counting relevant clinical signs is an easy and effective prognostic tool in severe adult and adolescent malnutrition during famine, it is not sensitive enough for use as a screening tool. We carried out a randomised controlled trial of three milk formulas in Australian Aboriginal children with diarrhoea and malnutrition, which showed significantly better outcomes for the lowest osmolality lactose-free formula (165 mOsmol/kg).31 Although this particular commercial formula (DeLact) is very expensive for use in the developing world, the study is an indication that malnourished children with severe enteropathy benefit from lactose-free low osmolality milk feeds. The WHO formulas F-75 and F-100 have osmolalities of 333 and 419 mOsmol/L, and lactose contents of 13 and 42 g/L, respectively, so lowering these could well improve outcomes in severe malnutrition, particularly in the context of associated HIV infection, cryptosporidiosis and other causes of persistent diarrhoea. However, the increased cost implications of even selective use of commercial formulas is an important constraint. Similarly, a recent Jamaican study used isotope probes to study fat digestion in 24 children during each of the three phases (9 days each) of recovery from severe malnutrition on a coconut and maize oil diet as the lipid source.32 The results indicated that over half of the children with severe malnutrition had a significant degree of fat and bile salt malabsorption. Although the clinical significance of this is unclear, it was related to the underlying enteropathy, which improved with treatment from phase 1 to phases 2 and 3. Bacterial overgrowth was not a major factor contributing to malabsorption.33

Ready-to-use therapeutic food (RUTF) Ready-to-use therapeutic food was developed by the French company Nutriset as an alternative to the F-100 formula (100 kcal or

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

577

Dietary management

DR Brewster

420 kJ/100 mL) or milk-oil formula (567 kJ/100 mL) used in hospitals and nutrition rehabilitation centres for rapid catch-up growth after initiation of cure. Part of the skim milk was replaced by lactoserum and groundnut paste, giving it a flavour of peanut butter, an energy density of 2281 kJ/100 g and, most importantly, it does not support bacterial growth.34 RUTF was designed to be consumed without the addition of water, hence eliminating the risk of bacterial proliferation in case of accidental contamination, but it is also energy-dense without any clinically significant osmolarity problems. This is true both for the RUTF produced by Nutriset, which contains mainly dextrin maltose as source of carbohydrate, and also for a cheaper locally made version containing sugar, and this is due to the physical structure of RUTF slowing down the dissolution of water-soluble molecules. In the hospital setting, a major limitation of RUTF is the inability to easily administer it by nasogastric tube. An ideal RUTF formulation should have the following key attributes: low cost, good nutritional quality (i.e. protein, energy and micronutrient content), long shelf life, resistant to bacterial growth, highly palatability, consistency suitable for feeding infants/children and require no additional processing prior to feeding.35 It may also be feasible to produce low-cost RUTF with local foods without the need for peanuts and milk powder. The recent advent of RUTF that can be stored without refrigeration yet without the high risk of contamination of liquid milk has made it easier for the rapid catch-up growth phase to occur out of hospital, at lower cost and without the risk of nosocomial infections. In an unblinded randomised trial, RUTF was compared with the F100 formula (high-energy milk) in 60 severely malnourished Senegalese children.36 The key outcomes (RUTF versus F-100) were energy intake (808 vs. 573 kJ/kg/day), weight gain (15.6 vs. 10.1 g/ kg/day) and duration of rehabilitation (13.4 vs. 17.3 days, respectively), which were all highly significant differences favouring RUTF (P < 0.001). Although the unblinded methodology and loss of 10 of 70 children to follow-up are concerns, this study provides some evidence that RUTF is superior to F-100 for the rapid catch-up growth phase of severe malnutrition, presumably related to its greater energy density (2281 vs. 414 kJ/100 g). An additional benefit of RUTF is protection against bacterial contamination for outpatient or home treatment as well as the feasibility of local production of RUTF in a tropical setting. A recent Malawian study of 1178 malnourished children compared home-based RUTF with standard inpatient management, and documented better outcomes with RUTF.37 These included lower case-fatality or relapse rates (8.7% vs. 16.7%), better weight gain (3.5 vs. 2.0 g/kg/day) and lower cross-infection rates. Another Malawian study compared RUTF with two home-based diets (corn-soy and RUTF supplement) in the home-based treatment of 282 HIVseronegative malnourished children, followed fortnightly for 6 months after about 2 weeks of hospital treatment.38 The RUTF group had more rapid weight gain (5.2 vs. 3.1 g/kg/day) and recovery of wasting (WHZ > 0 in 35 vs. 56 days) than the home-based diets. The main limitations of the study were the lack of formal randomisation and 16% loss to follow-up, which were related to the difficulty of carrying our research in this setting. The better outcomes with RUTF were again related to its fivefold higher energy density than corn-soy. Yet another study from Malawi compared commercially produced RUTF (Nutriset, France; cost US$3/kg + $2 shipping) with a locally produced RUTF (US$2/kg) in 260 severely malnourished 578

children discharged from hospital.39 Outcomes were similar between the two diets, including weight gain (commercial vs. local) of 5.2 versus 4.8 g/kg/day. On this evidence, a locally produced RUTF can be as successful as the commercial product, which has important implications in the developing country setting. We conclude that RUTF is an important advance in the management of severe malnutrition, so needs to be incorporated into treatment protocols. There are likely to be a new generation of diets for young children that present as spreads (FOODlet).40–42 However, such foods still suffer from the well-known limitations of food supplement handouts, namely high cost, sharing with other family members (‘leakage’), compensatory reductions in home feeds (‘substitution’) and poor compliance (‘wastage’). This problem was documented in a periodic 24-h dietary recall study that found index children received only 30% and 43%, respectively, of the provided supplement.43 Supervised ambulatory feeding (directly observed therapy) can improve compliance, but it is often not feasible. Yet the advantages of home feeding in terms of reduction of cross-infection, increased programme coverage and reduced social disruption largely compensate for the problems of leakage, substitution and wastage of food.

What Micronutrients Are Beneficial in Severe Malnutrition? There have been a host of recent micronutrient treatment trials, both hospital and community studies, especially for zinc and vitamin A and their interactions. This is the issue concerning severe malnutrition management with the highest quality evidence available and probably the least concern about external generalisability. Gibson et al. point out that many of the complementary foods in developing countries fail to meet requirements for micronutrients such as iron, zinc and calcium.44 Although strategies such as germination, fermentation and soaking can reduce the phytate content (which inhibits micronutrient absorption), fortification of plant-based complementary foods is considered the best strategy for meeting micronutrient requirements. Phytate reduction strategies have only had a modest effect on micronutrient status in Malawian children.45–47 Nevertheless, an intervention to improve the micronutrient adequacy of diets did lead to an increase in indices of lean body mass and reductions in the incidence of anaemia and common infections in stunted children in rural Malawi.48

Zinc There have been four key randomised trials involving zinc from Bangladesh. A community study documented significantly greater weight gains (3.15 vs. 2.66 kg, P < 0.04) in the zinc (5 mg/day from 4 to 24 weeks of age) than in the placebo group for 43 infants who were zinc-deficient at baseline, but not for the other infants.49 Zincdeficient infants also showed a reduced risk of acute lower respiratory infection after zinc supplementation (relative risk 0.30; 0.10–0.92) but not among the zinc-sufficient infants. On the other hand, a second trial followed 653 children aged 12–35 months for 6 months with four intervention groups: 20 mg zinc daily for 14 days, 200 000 IU of vitamin A on day 14, zinc plus vitamin A and placebo.50 They found no significant differences in gains in weight and length among the four groups, so concluded that combined short-term

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

DR Brewster

zinc supplementation and a single dose of vitamin A had no significant effects on weight and length increments in children over a 6-month period. A third micronutrient study assessed the impact of zinc supplementation (20 mg daily for 14 days) during acute diarrhoea on subsequent growth and morbidity in 65 malnourished children aged 3–24 months.51 Zinc-supplemented children showed significantly greater cumulative length gain (18.9 mm vs. 14.5 mm, P < 0.03) but not weight gain than controls. Zinc-supplemented and stunted children (n = 18) experienced significantly fewer episodes of diarrhoea (0.07 vs. 0.6, P < 0.05) and respiratory illness (1.0 vs. 2.4, P < 0.01) compared with controls. Zinc-supplemented underweight children (n = 38) also had fewer episodes of diarrhoea (0.4 vs. 1.0, P < 0.04) and shorter duration of diarrhoeal episodes (1.0 vs. 3.0 days, P < 0.04) compared with controls. The authors concluded that a short course of zinc supplementation for malnourished children with diarrhoea reduces growth-faltering and diarrhoeal and respiratory morbidity over the subsequent 2 months. Finally, a fourth trial measured the effects of zinc supplementation on catch-up growth in 141 children 6 months−3 years of age with severe protein-energy malnutrition, with particular reference to linear growth.52 Subjects received zinc 1.5 mg/kg for 15 days, or 6.0 mg/kg for 15 days, or 6.0 mg/kg for 30 days, and were followed for a total of 90 days. Not only did the higher zinc dose (6.0 mg/kg) not lead to improved anthropometric indices, but also mortality was significantly greater in these children compared with the low zinc dose (risk ratio 4.53; 1.1–18.8). Similar results were found in a still unpublished Malawi study.53 From these studies, it is clear that highdose zinc (>6 mg/kg/day for 2–4 weeks) in the rapid catch-up growth phase of nutritional rehabilitation has a potentially detrimental effect on delayed cutaneous hypersensitivity, weight gain, incidence of infections and mortality. Zinc supplementation (10 mg total dose as zinc sulphate daily for 3 months) was evaluated in the management of severe malnutrition in a trial of 300 children, aged 6–60 months in Lesotho.54 The case-fatality rate was significantly lower in the zinc group (4.7%) than in controls (16.7%). The prevalence of morbidity was also significantly higher in the control group at 1, 2 and 3 months follow-up. In the zinc group, 58% of the children had a weight-forage >80% of the median (note that the article seems to confuse centile and per cent of the median) 3 months after discharge compared with 28% in controls, with significant reduction in diarrhoeal disease, respiratory morbidity, episodes of clinical anaemia, skin infections and fever as well as vomiting in the malnourished children. These trends were also evident on biochemical profiles.55 However, unlike the Bangladeshi studies, there are methodological concerns about this trial, and the magnitude of the benefit from zinc seems disproportionate. The main difficulty in many settings is the availability of micronutrients such as zinc, especially in a suitable form for administration to children. It is important to acknowledge these sorts of constraints for poor countries, so donor agencies can provide funding or information about sources of micronutrients for poor countries. In conclusion, malnourished children should receive zinc supplementation, and a total daily dose of 20 mg is safe and beneficial whereas doses ≥6 mg/kg/day for long periods may be detrimental. This is consistent with the WHO Manual and zinc sulphate 10 mg is now included on the WHO list of essential drugs.56

Dietary management

Vitamin A The WHO Manual recommends vitamin A administration on admission in standard age-related doses (50 000 IU, 100 000 IU or 200 000 IU) and repeat doses on days 2 and 14 (or later) with clinical deficiency. Initially, randomised studies in the developing world showed a definite reduction in mortality with vitamin A supplementation. However, more recent studies with combined micronutrients have shown interactions, and some have found higher rates of respiratory infection in the Vitamin A supplemented group. There is evidence from Senegalese randomised studies on vitamin A dosage in severe malnutrition showing that daily low doses of vitamin A improved mortality better than a single high dose as recommended in the Manual.57,58 The most recent study abstract59 compared the effect of daily low doses of vitamin A (5000 IU/day) with a single high dose (200 000 IU) on the case-fatality rate in hospitalised malnourished children. Both groups were similar at baseline in terms of age and nutritional status with low serum retinol concentrations in more than 75% of the study sample. Mortality was 9.7% (59 deaths) in the low-dose group and 11.1% (67 deaths) in the high-dose group, which was not statistically significant. However, in children with oedema on admission, mortality was significantly lower in the low daily dose group (adjusted odds ratio 0.21; 0.05– 0.99). The proportion of children without acute respiratory infection during hospitalisation was significantly higher in the low daily dose group (83.5 vs. 77.8, P = 0.043) but there was no difference in diarrhoea between groups. Vitamin A reduced the severity of acute shigellosis in children in Bangladesh where vitamin A deficiency and dysentery are a major public health problems.60 One of the problems with studies is that retinol values are much higher in subjects with normal protein levels than in those with raised acute phase proteins (16% higher for α1antichymotrypsin, 18% for α1-acid-glycoprotein, 25% for C-reactive protein and 32% for serum amyloid A).61 Thus the reduction in plasma retinol for individuals with infection is estimated at 13% (incubation), 24% (early convalescent) and 11% (late convalescent). Thus, surveys to estimate vitamin A deficiency need to include measurements of acute-phase proteins so plasma retinol concentrations can be adjusted. A recent updated meta-analysis of the effect of vitamin A supplementation on childhood morbidity from respiratory tract infections and diarrhoea (1966–2000) identified nine randomised controlled trials.62 The combined results indicated that vitamin A supplementation had no consistent overall protective effect on the incidence of diarrhoea (relative risk (RR) 1.00; 0.94–1.07) and that it slightly increased the incidence of respiratory tract infections (RR 1.08; 1.05–1.11). The reviewers concluded that high-dose vitamin A supplements are not recommended on a routine basis for all preschool children and should be offered only to individuals or populations at risk of vitamin A deficiency. A recent abstract presented at the International Vitamin A Consultative Group Meeting63 studied the interaction between vitamin A and zinc supplementation in a trial of 260 Brazilian children with vitamin A deficiency (serum retinol levels <0.70 µmol/L). The two treatment groups were: vitamin A 200 000 IU on day 1 + zinc sulphate 25 mg daily for 30 days, compared with vitamin A 200 000 IU on day 1 + placebo daily. Vitamin A supplementation alone significantly increased serum retinol levels but zinc + vitamin A did not.

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

579

Dietary management

DR Brewster

Thus, there is preliminary evidence suggesting a need to modify the current single high-dose regime to a lower daily dose (5000 IU) for children with severe malnutrition. The possibility of the single high dose on admission increasing the risk of respiratory infection and leading to poor absorption are concerns. Further information is available on the web from the International Vitamin A Consultative Group (http://ivacg.ilsi.org/).63

Iron The standard practice in developing countries, as recommended in the WHO Manual, is not to commence iron therapy on admission in severe malnutrition, but to delay it until the rehabilitation phase. A Nigerian study reported an excess mortality when iron was given to severely malnourished children.64 However, the dose is not mentioned and the difference in mortality reported in the ironsupplemented group was not significant. Similar methodological issues have been raised about other studies on iron and infection in well-nourished children. A recent systematic review of infection with iron supplementation found no evidence of a detrimental effect.65 This does not of course mean that it is wise to give iron to critically ill children. There is an apparent excess of iron that may predispose to bacterial infections and free radical-mediated injury in children with kwashiorkor.66–68 Incidentally, although the free radical theory of kwashiorkor has been widely accepted, a recent trial of antioxidant supplementation for 20 weeks in eight Malawian villages (2332 children) failed to prevent kwashiorkor, which occurred in 3.3% of the antioxidant group compared with 1.9% of controls, casting doubt on the hypothesis that free radicals are an important aetiological factor.69 With regards to iron status, kwashiorkor may result in inappropriate down-regulation of hepcidin leading to increased intestinal iron absorption and increased iron release from the reticuloendothelial system. Hepcidin is a small peptide hormone produced by the liver induced by inflammation (acute-phase reactant), which controls iron homeostasis. It normally acts to limit intestinal absorption and macrophage release of iron (presumably to deny iron to the invading organism causing the inflammatory response) and is regulated by iron availability to red cells.70,71 The early onset and high prevalence of iron deficiency in Aboriginal children from remote communities is related to the high burden of infections, presumably causing chronic up-regulation of hepcidin.72,73 The administration of oral iron is problematic, but new studies on less frequent administration and newer formulations administered as microencapsulated iron fumarate ‘sprinkles’ may help to improve compliance.74–77 Although the delay in commencing iron in severe malnutrition seems a sensible policy, the evidence to support it is lacking.

Conclusions 1 RUTF: Evidence for the use of RUTF for the rapid catch-up growth phase of severe malnutrition is convincing. It should be included in treatment protocols, including instructions for producing RUTF from local foods. More research is needed about RUTF from local produce that does not contain milk or peanuts and takes the form of spreads (FOODlet). The key advantages of RUTF is its high-energy density, storage at room temperature, protec580

tion against bacterial contamination and lowered risk of nosocomial infections by enabling earlier hospital discharge. 2 F-75 milk: Consideration should be given to increasing the sulphur amino acid, phosphorus and potassium content of this formula for the initiation of cure in severely affected cases of malnutrition, especially for kwashiorkor. The current recommendation of restricting intakes to 130 mL/kg/day during initial treatment is essential for kwashiorkor. 3 Availability of WHO F-75 and F-100 formulas: The poor availability of these milk formulas (including micronutrients) at low cost, especially for small countries, means that milk-oil-sugar formulas will continue to need to be mixed by health facilities treating child malnutrition. 4 Micronutrients: Although there has been a considerable number of publications on micronutrients, they have not resulted in much need to change severe malnutrition protocols. Preliminary evidence suggests that smaller doses of daily vitamin A are preferable to a single large dose, and that children with kwashiorkor may need to be given more potassium than recommended in the Manual.

References 1 Picou D, Alleyne GAO, Brooke O et al. Malnutrition and Gastroenteritis in Children: A Manual for Hospital Treatment and Management. Kingston, Jamaica: Tropical Metabolism Research Unit, 1978. 2 World Health Organization. Management of Severe Malnutrition: A Manual for Physicians and Other Senior Health Workers. Geneva: WHO, 1999. 3 Prentice AM, Paul AA. Fat and energy needs of children in developing countries. Am. J. Clin. Nutr. 2000; 72 (Suppl.): 1253S–65S. 4 Brewster DR, Manary MJ, Graham SM. Case management of kwashiorkor: an intervention project at 7 Nutritional Rehabilitation Centres in Malawi. Eur. J. Clin. Nutr. 1997; 51: 139–47. 5 Premji S, Chessell L. Continuous nasogastric milk feeding versus intermittent bolus milk feeding for premature infants less than 1500 grams. Cochrane Database Syst. Rev. 2003;1: CD001819. 6 McLaren DS. The great protein fiasco. Lancet 1974; 2: 93–6. 7 Alleyne GAO, Hay RW, Picou DI, Stanfield JP, Whitehead RG. ProteinEnergy Malnutrition. London: Edward Arnold, 1977. 8 Anonymous. The Treatment and Management of Severe Protein-Energy Malnutrition. Geneva: WHO, 1981. 9 Waterlow JC. Protein-Energy Malnutrition. London: Edward Arnold, 1992. 10 Golden MH. The development of concepts of malnutrition. J. Nutr. 2002; 132: 2117S–22S. 11 Collins S, Myatt M, Golden BE. Dietary treatment of severe malnutrition in adults. Am. J. Clin. Nutr. 1998; 68: 193–9. 12 Badaloo A, Reid M, Forrester T, Heird WC, Jahoor F. Cysteine supplementation improves the erythrocyte glutathione synthesis rate in children with severe edematous malnutrition. Am. J. Clin. Nutr. 2002; 76: 646–52. 13 Freiman I, Pettifor JM, Moodley GM. Serum phosphorus in protein energy malnutrition. J. Pediatr. Gastroenterol. Nutr. 1982; 1: 547–50. 14 Manary MJ, Hart CA, Whyte MP. Severe hypophosphatemia in children with kwashiorkor is associated with increased mortality. J. Pediatr. 1998; 133: 789–91. 15 Turi RA, Petros AJ, Eaton S et al. Energy metabolism of infants and children with systemic inflammatory response syndrome and sepsis. Ann. Surg. 2001; 233: 581–7. 16 Brown KH, Sanchez Grinan M, Perez F, Peerson JM, Ganoza L, Stern JS. Effects of dietary energy density and feeding frequency on total daily

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

DR Brewster

17

18

19 20 21

22 23

24 25

26 27 28 29

30

31

32

33

34

35

36

37

38

energy intakes of recovering malnourished children. Am. J. Clin. Nutr. 1995; 62: 13–18. Arora NK, Anand NK, Bhan MK et al. Nutrient absorption from a fatenriched diet in young malnourished children: a randomized controlled trial. Acta Paediatr. 1998; 87: 143–8. Valentiner-Branth P, Steinsland H, Santos G et al. Community-based controlled trial of dietary management of children with persistent diarrhea: sustained beneficial effect on ponderal and linear growth. Am. J. Clin. Nutr. 2001; 73: 968–74. Rao DH, Sarma KV, Kumar S, Reddy CG, Roa NP. Acceptability trials with ready to eat foods in a rural area. Indian Pediatr. 1992; 29: 1513–18. Pretorius PJ, Hansen JD, Davel JGA, Brock JF. Skimmed milk and kwashiorkor. S. Afr. Med. J. 1956; 33: 447–50. Hansen JDL, Schendel HE, Wilkins JA, Brock JF. Nitrogen metabolism in children with kwashiorkor receiving milk and vegetable diets. Pediatrics 1960; 25: 258–82. Holt LE, Snyderman SE, Norton PM, Roitman E, Finch J. The plasma aminogram in kwashiorkor. Lancet 1963; ii: 1342–8. Pereira SM, Isaac T, Begum A, Dumm ME, Webb JK. Moderate protein, high calorie diets in treatment of kwashiorkor. Indian Pediatr. 1965; 2: 76–84. McLaren DS. A fresh look at protein-calorie malnutrition. Lancet 1966; 2: 485–8. Gopalan C. Kwashiorkor and marasmus: evolution and distinguishing features. In: McCance RA, Widdowson EM, eds. Calorie Deficiencies and Protein Deficiencies. London: J.&A. Churchill, 1968; 49–58. Waterlow JC. Classification and definition of protein-calorie malnutrition. Br. Med. J. 1972; 3: 566–9. Golden MH, Golden BE, Jackson AA. Albumin and nutritional oedema. Lancet 1980; 1: 114–16. Golden MH. Oedematous malnutrition. Br. Med. Bull. 1998; 54: 433–44. Collins S, Sadler K. Outpatient care for severely malnourished children in emergency relief programmes: a retrospective cohort study. Lancet 2002; 360: 1824–30. Collins S, Myatt M. Short-term prognosis in severe adult and adolescent malnutrition during famine: use of a simple prognostic model based on counting clinical signs. JAMA 2000; 284: 621–6. Kukuruzovic RH, Brewster DR. Milk formulas in acute gastroenteritis and malnutrition: a randomised trial. J. Paediatr. Child Health 2002; 38: 571–7. Murphy JL, Badaloo AV, Chambers B, Forrester TE, Wootton SA, Jackson AA. Maldigestion and malabsorption of dietary lipid during severe childhood malnutrition. Arch. Dis. Child. 2002; 87: 522–5. Murphy JL, Robinson EN, Forrester TE, Wootton SA, Jackson AA. Gastrointestinal handling and metabolic disposal of 13C-labelled tripalmitin during rehabilitation from childhood malnutrition. Br. J. Nutr. 2001; 85: 705–13. Briend A, Lacsala R, Prudhon C, Mounier B, Grellety Y, Golden MH. Ready-to-use therapeutic food for treatment of marasmus. Lancet 1999; 353: 1767–8. Anonymous. CTC Workshop, Dublin, October 2003. 2004. Available from: http://www.fantaproject.org/ctc/workshop2003.shtml [accessed 23 March 2006]. Diop El HI, Dossou NI, Ndour MM, Briend A, Wade S. Comparison of the efficacy of a solid ready-to-use food and a liquid, milk-based diet for the rehabilitation of severely malnourished children: a randomized trial. Am. J. Clin. Nutr. 2003; 78: 302–7. Ciliberto MA, Sandige H, Ndekha MJ et al. Comparison of home-based therapy with ready-to-use therapeutic food with standard therapy in the treatment of malnourished Malawian children: a controlled, clinical effectiveness trial. Am. J. Clin. Nutr. 2005; 81: 864–70. Manary MJ, Ndekha MJ, Ashorn P, Maleta K, Briend A. Home based therapy for severe malnutrition with ready-to-use food. Arch. Dis. Child 2004; 89 (6): 557–61.

Dietary management

39 Sandige H, Ndekha MJ, Briend A, Ashorn P, Manary M. Home-based treatment of malnourished Malawian children with locally produced or imported ready-to-use food. J. Pediatr. Gastroenterol. Nutr. 2004; 39: 141–6. 40 Gross R, Dwivedi A, Solomons NW. Introduction to the proceedings of the International Research on Infant Supplementation (IRIS) Initiative. Food Nutr. Bull. 2003; 24 (Suppl. 3): S3–6. 41 Lopez DR, Gross R. IRIS. III: proposal of a multicenter efficacy study using a high-energy, micronutrient-dense spread. Food Nutr. Bull. 2003; 24 (3 Suppl.): S39–41. 42 Briend A, Solomons NW. The evolving applications of spreads as a FOODlet for improving the diets of infants and young children. Food Nutr. Bull. 2003; 24 (3 Suppl.): S34–8. 43 Maleta K, Kuittinen J, Duggan MB et al. Supplementary feeding of underweight, stunted Malawian children with a ready-to-use food. J. Pediatr. Gastroenterol. Nutr. 2004; 38: 152–8. 44 Gibson RS, Ferguson EL, Lehrfeld J. Complementary foods for infant feeding in developing countries: their nutrient adequacy and improvement. Eur. J. Clin. Nutr. 1998; 52: 764–70. 45 Manary MJ, Krebs NF, Gibson RS, Broadhead RL, Hambidge KM. Community-based dietary phytate reduction and its effect on iron status in Malawian children. Ann. Trop. Paediatr. 2002; 22: 133–6. 46 Manary MJ, Hotz C, Krebs NF et al. Zinc homeostasis in Malawian children consuming a high-phytate, maize-based diet. Am. J. Clin. Nutr. 2002; 75: 1057–61. 47 Gibson RS, Yeudall F, Drost N, Mtitimuni BM, Cullinan TR. Experiences of a community-based dietary intervention to enhance micronutrient adequacy of diets low in animal source foods and high in phytate: a case study in rural Malawian children. J. Nutr. 2003; 133 (11 Suppl. 2): 3992S–9S. 48 Yeudall F, Gibson RS, Kayira C, Umar E. Efficacy of a multi-micronutrient dietary intervention based on haemoglobin, hair zinc concentrations, and selected functional outcomes in rural Malawian children. Eur. J. Clin. Nutr. 2002; 56: 1176–85. 49 Osendarp SJ, Santosham M, Black RE, Wahed MA, van Raaij JM, Fuchs GJ. Effect of zinc supplementation between 1 and 6 mo of life on growth and morbidity of Bangladeshi infants in urban slums. Am. J. Clin. Nutr. 2002; 76: 1401–8. 50 Rahman MM, Tofail F, Wahed MA, Fuchs GJ, Baqui AH, Alvarez JO. Shortterm supplementation with zinc and vitamin A has no significant effect on the growth of undernourished Bangladeshi children. Am. J. Clin. Nutr. 2002; 75: 87–91. 51 Roy SK, Tomkins AM, Haider R et al. Impact of zinc supplementation on subsequent growth and morbidity in Bangladeshi children with acute diarrhoea. Eur. J. Clin. Nutr. 1999; 53: 529–34. 52 Doherty CP, Sarkar MA, Shakur MS, Ling SC, Elton RA, Cutting WA. Zinc and rehabilitation from severe protein-energy malnutrition: higher-dose regimens are associated with increased mortality. Am. J. Clin. Nutr. 1998; 68: 742–8. 53 MacDonald C. The Effect of Zinc Fortification of a Therapeutic Complementary Food on the Growth and Immune Function of Severely Malnourished Children in Malawi. Guelph, Canada: University of Guelph, 1999. 54 Makonnen B, Venter A, Joubert G. A randomized controlled study of the impact of dietary zinc supplementation in the management of children with protein-energy malnutrition in Lesotho. I: mortality and morbidity. J. Trop. Pediatr. 2003; 49: 340–52. 55 Makonnen B, Venter A, Joubert G. A randomized controlled study of the impact of dietary zinc supplementation in the management of children with protein-energy malnutrition in Lesotho. II: special investigations. J. Trop. Pediatr. 2003; 49: 353–60. 56 Anonymous. WHO List of Essential Drugs. Geneva: WHO. Available from: http://www.who.int/medicines/services/essmedicines_def/en/ [accessed 26 March 2006].

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)

581

Dietary management

DR Brewster

57 Donnen P, Dramaix M, Brasseur D, Bitwe R, Vertongen F, Hennart P. Randomized placebo-controlled clinical trial of the effect of a single high dose or daily low doses of vitamin A on the morbidity of hospitalized, malnourished children. Am. J. Clin. Nutr. 1998; 68: 1254–60. 58 Sylla A, N’diaye O, Diouf S et al. [Effect of systematic vitamin A supplementation on persistent diarrhea in Senegal: single high dose versus small daily doses]. Arch. Pediatr. 2002; 9: 975–6. 59 Donnen P, Sylla A, Dramaix M, Sall G, Kuakuvi N, Hennart P. Daily Low Doses of Vitamin A Compared with Single High Dose Improves Survival of Malnourished Children in Senegal. 2004. Poster T15:[56]. Available from: http://ivacg.ilsi.org/file/IVACGfinal.pdf [accessed 23 March 2006]. 60 Hossain S, Biswas R, Kabir I et al. Single dose vitamin A treatment in acute shigellosis in Bangladesh children: randomised double blind controlled trial. Br. Med. J. 1998; 316: 422–6. 61 Thurnham DI, McCabe GP, Northrop-Clewes CA, Nestel P. Effects of subclinical infection on plasma retinol concentrations and assessment of prevalence of vitamin A deficiency: meta-analysis. Lancet 2003; 362: 2052–8. 62 Grotto I, Mimouni M, Gdalevich M, Mimouni D. Vitamin A supplementation and childhood morbidity from diarrhea and respiratory infections: a meta-analysis. J. Pediatr. 2003; 142: 297–304. 63 Anonymous. Improving the Vitamin A Status of Populations. Report of the XXI International Vitamin A Consultative Meeting. 2004. Available from: http://ivacg.ilsi.org/file/IVACGfinal.pdf [accessed 26 March 2006]. 64 Smith IF, Taiwo O, Golden MH. Plant protein rehabilitation diets and iron supplementation of the protein-energy malnourished child. Eur. J. Clin. Nutr. 1989; 43: 763–8. 65 Gera T, Sachdev HP. Effect of iron supplementation on incidence of infectious illness in children: systematic review. Br. Med. J. 2002; 325: 1142. 66 Dempster WS, Sive AA, Rosseau S, Malan H, Heese HV. Misplaced iron in kwashiorkor. Eur. J. Clin. Nutr. 1995; 49: 208–10. 67 Sive AA, Subotzky EF, Malan H, Dempster WS, Heese HD. Red blood cell antioxidant enzyme concentrations in kwashiorkor and marasmus. Ann. Trop. Paediatr. 1993; 13: 33–8.

582

68 Sive AA, Dempster WS, Rosseau S, Kelly M, Malan H, Heese HD. Bone marrow and chelatable iron in patients with protein energy malnutrition. S. Afr. Med. J. 1996; 86: 1410–13. 69 Ciliberto H, Ciliberto M, Briend A, Ashorn P, Bier D, Manary M. Antioxidant supplementation for the prevention of kwashiorkor in Malawian children: randomised, double blind, placebo controlled trial. Br. Med. J. 2005; 330: 1109. 70 Ganz T. Hepcidin, a key regulator of iron metabolism and mediator of anemia of inflammation. Blood 2003; 102: 783–8. 71 Roy CN, Weinstein DA, Andrews NC. 2002 E. Mead Johnson Award for Research in Pediatrics Lecture: the molecular biology of the anemia of chronic disease: a hypothesis. Pediatr. Res. 2003; 53: 507–12. 72 Ritchie BK, McNeil Y, Brewster DR. Soluble transferrin receptor in aboriginal children with a high prevalence of iron deficiency and infection. Trop. Med. Int. Health 2004; 9: 96–105. 73 Kruske SG, Ruben AR, Brewster DR. An iron treatment trial in an Aboriginal community: improving non-adherence. J. Paediatr. Child Health 1999; 35: 153–8. 74 Zlotkin S, Antwi KY, Schauer C, Yeung G. Use of microencapsulated iron(II) fumarate sprinkles to prevent recurrence of anaemia in infants and young children at high risk. Bull. World Health Organ. 2003; 81: 108–15. 75 Zlotkin S, Arthur P, Schauer C, Antwi KY, Yeung G, Piekarz A. Homefortification with iron and zinc sprinkles or iron sprinkles alone successfully treats anemia in infants and young children. J. Nutr. 2003; 133: 1075–80. 76 Ermis B, Demirel F, Demircan N, Gurel A. Effects of three different iron supplementations in term healthy infants after 5 months of life. J. Trop. Pediatr. 2002; 48: 280–4. 77 Haidar J, Omwega AM, Muroki NM, Ayana G. Daily versus weekly iron supplementation and prevention of iron deficiency anaemia in lactating women. East Afr. Med. J. 2003; 80: 11–16.

Journal of Paediatrics and Child Health 42 (2006) 575–582 © 2006 The Author Journal compilation © 2006 Paediatrics and Child Health Division (Royal Australasian College of Physicians)