Introduction to the Proceedings of the International Research on Infant Supplementation (Iris) Initiative

Specially formulated foods for treating children with moderate acute malnutrition in low- and middle-income countries

BACKGROUND

Moderate acute malnutrition, also called moderate wasting, affects around 10% of children under five years of age in low- and middle-income countries. There are different approaches to addressing malnutrition with prepared foods in these settings; for example, providing lipid-based nutrient supplements or blended foods, either a full daily dose or in a low dose as a complement to the usual diet. There is no definitive consensus on the most effective way to treat children with moderate acute malnutrition.

OBJECTIVES

To evaluate the safety and effectiveness of different types of specially formulated foods for children with moderate acute malnutrition in low- and middle-income countries, and to assess whether foods complying or not complying with specific nutritional compositions, such as the WHO technical specifications, are safe and effective.

SEARCH METHODS

In October 2012, we searched CENTRAL, MEDLINE, LILACS, CINAHL, BIBLIOMAP, POPLINE, ZETOC, ICTRP, mRCT, and ClinicalTrials.gov. In August 2012, we searched Embase. We also searched the reference lists of relevant papers and contacted nutrition-related organisations and researchers in this field.    

SELECTION CRITERIA

We planned to included any relevant randomised controlled trials (RCTs), controlled clinical trials (CCTs), controlled before-and-after studies (CBAs), and interrupted time series (ITS) that evaluated specially formulated foods for the treatment of moderate acute malnutrition in children aged between six months and five years in low- and middle-income countries.

DATA COLLECTION AND ANALYSIS

Two authors assessed trial eligibility and risk of bias, and extracted and analysed the data. We summarised dichotomous outcomes using risk ratios (RR) and continuous outcomes using mean differences (MD) with 95% confidence intervals (CI). Where appropriate, we combined data in meta-analyses using the random-effects model and assessed heterogeneity. The quality of evidence was assessed using GRADE methods.

MAIN RESULTS

Eight randomised controlled trials, enrolling 10,037 children, met our inclusion criteria. Seven of the trials were conducted in Africa. In general, the included studies were at a low risk of bias. There may have been a risk of performance bias as trial participants were aware which intervention group they were in, but we did not consider this likely to have biased the outcome measurement. We were unable to assess the risk of reporting bias in half of the trials and two trials were at high risk of attrition bias. Any specially formulated food versus standard care - the provision of food increased the recovery rate by 29% (RR 1.29, 95% CI 1.20 to 1.38; 2152 children, two trials; moderate quality evidence), decreased the number dropping out by 70% (RR 0.30, 95% CI 0.22 to 0.39; 1974 children, one trial; moderate quality evidence), and improved weight-for-height (MD 0.20 z-score, 95% CI 0.03 to 0.37; 1546 children, two trials; moderate quality evidence). The reduction in mortality did not reach statistical significance (RR 0.44; 95% CI 0.14 to 1.36; 1974 children, one trial; low quality evidence). Lipid-based nutrient supplements versus any blended foods (dry food mixtures, without high lipid content), at full doses - there was no significant difference in mortality (RR 0.93, 95% CI 0.54 to 1.62; 6367 children, five trials; moderate quality evidence), progression to severe malnutrition (RR 0.88, 95% CI 0.72 to 1.07; 4537 children, three trials; high quality evidence), or the number of dropouts from the nutritional programme (RR 1.14, 95% CI 0.62 to 2.11; 5107 children, four trials; moderate quality evidence). However, lipid-based nutrient supplements significantly increased the number of children recovered (RR 1.10, 95% CI 1.04 to 1.16; 6367 children, five trials; moderate quality evidence), and decreased the number of non-recovering children (RR 0.53, 95% CI 0.40 to 0.69; 4537 children, three trials; high quality evidence). LNS also improved weight gain, weight-for-height, and mid-upper arm circumference, although for these outcomes, the improvement was modest (moderate quality evidence). One trial observed more children with vomiting in the lipid-based nutrient supplements group compared to those receiving blended food (RR 1.43, 95% CI 1.11 to 1.85; 2712 children, one trial; low quality evidence). Foods at complementary doses - no firm conclusion could be drawn on the comparisons between LNS at complementary dose and blended foods at complementary or full dose (low quality evidence). Lipid-based nutrient supplements versus specific types of blended foods - a recently developed enriched blended food (CSB++) resulted in similar outcomes to LNS (4758 children, three trials; moderate to high quality evidence). Different types of blended foods - in one trial, CSB++ did not show any significant benefit over locally made blended food, for example, Misola, in number who recovered, number who died, or weight gain (moderate to high quality evidence). Improved adequacy of home diet - no study evaluated the impact of improving adequacy of local diet, such as local foods prepared at home according to a given recipe or of home processing of local foods (soaking, germination, malting, fermentation) in order to increase their nutritional content.

AUTHORS' CONCLUSIONS

In conclusion, there is moderate to high quality evidence that both lipid-based nutrient supplements and blended foods are effective in treating children with MAM. Although lipid-based nutrient supplements (LNS) led to a clinically significant benefit in the number of children recovered in comparison with blended foods, LNS did not reduce mortality, the risk of default or progression to SAM. It also induced more vomiting. Blended foods such as CSB++ may be equally effective and cheaper than LNS. Most of the research so far has focused on industrialised foods, and on short-term outcomes of MAM. There are no studies evaluating interventions to improve the quality of the home diet, an approach that should be evaluated in settings where food is available, and nutritional education and habits are the main determinants of malnutrition. There are no studies from Asia, where moderate acute malnutrition is most prevalent.

Critical appraisal of the management of severe malnutrition: 2. Dietary management

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.

The International Research on Infant Supplementation initiative

The International Research on Infant Supplementation trials emerged as an attempt to develop a credible evidence base to guide policy decision making in this area. The fact that infant diets in developing country settings are more often deficient in multiple micronutrients rather than in single nutrients and/or energy was increasingly recognized during the 1990s. The concept of a "foodlet," a crushable hybrid between a tablet and a food that is water dispersible, was developed at the first IRIS meeting in Rio de Janeiro in 1999. Trials were performed in 4 countries: Indonesia, Peru, South Africa, and Vietnam. Four different formulations of foodlets were tested: a daily placebo supplement, a daily iron supplement, a daily multiple micronutrient supplement, and a weekly multiple micronutrient supplement with a placebo on the other 6 d. The multiple micronutrient foodlets contained adequate daily intakes of 13 micronutrients considered most likely to be inadequate in infant diets. The foodlets were produced in Peru and distributed to all 4 of the country study sites. The methods used in the efficacy trials were standardized across countries, and the trials were carried out in infants 6-11 mo of age for 6 mo. Anthropometry was measured monthly and morbidity checked daily. Blood samples were collected at baseline and at the end of the trial. The data and the biological samples collected in each country were analyzed centrally. The results of this remarkable international collaboration are encouraging and, hopefully, will spur the development of programs in this area of nutrition interventions.

Daily iron alone but not in combination with multimicronutrients increases plasma ferritin concentrations in indonesian infants with inflammation

Iron deficiency is a public health problem in infancy. We assessed the efficacy of iron supplements in infants with inflammation on iron status and subsequent inflammation. This was a prospective, nested, case-control study of 6- to 12-mo-old infants participating in the International Research on Infant Supplementation study, Indonesia. Cases (n = 46) were selected on the basis of their inflammation status at baseline, C-reactive protein (>5 mg/L) or alpha-1 acid glycoprotein (>1 g/L); there were 44 controls without inflammation. Infants received 10 mg/d of elemental iron alone or in combination with multimicronutrients, or placebo. Blood samples were collected at baseline and at 6 mo for determinations of plasma ferritin, zinc, copper, retinol, beta-carotene, alpha-tocopherol, and inflammation status. Data on breast-feeding and acute respiratory infections (ARI) were collected daily. At baseline, 33% of infants had iron deficiency, and those with inflammation had lower retinol, beta-carotene, higher concentrations of copper and higher rates of ARI compared with controls. After 6 mo, compared with infants given placebo, ferritin concentration increased significantly in infants administered iron alone independently of inflammation status at baseline or at the end of the study. In those given multimicronutrients with iron, ferritin increased significantly in infants who did not have inflammation at baseline or at the end of the study compared with those given placebo. Consequently, iron alone resolved iron deficiency, whereas multimicronutrients reduced the deterioration of iron stores compared with placebo (chi(2), P < 0.05), without enhancing inflammation. Iron alone is recommended in populations in which iron deficiency is a public health problem despite the presence of inflammation in infants who are still breast-feeding.