WHO/UNICEF/WFP/UNHCR informal consultation on moderate malnutrition management in U5’s

Summary of meeting^a

The World Health Organisation (WHO) convened a meeting in Geneva (September 30th - October 3rd, 2008) with the overall aim of answering the question, 'What diets should be recommended to feed moderately malnourished children?'^b. The general objectives of the meeting were:

1. to identify areas of consensus on the nutrient needs and dietary management of moderate malnutrition (MM) in children that can be translated into evidence-based global guidelines and
2. to identify knowledge gaps that should be addressed by research, both in the area of dietary management and the modalities for providing that diet.

Examples of different RUTF formulations developed by Prof Jeya Henry, Oxford Brookes University 2005

Moderate malnutrition (or moderate wasting) includes all children with a weight-forheight between -3 and -2 z-scores^c. Moderate stunting is defined by a height-for-age between -3 and -2 z-scores. Most of these children will be moderately underweight (weight-for-age between -3 and -2 z-scores).

Four background papers were commissioned by the WHO in advance of the meeting and circulated among participants. Presentations of these and other papers were followed by discussions and working group sessions to develop consensus statements and identify areas for research on the improved dietary management of MM.

Nutrient requirements of moderately malnourished children

A background paper (prepared by Prof Mike Golden, Emeritus Professor, Aberdeen University) provided tentative recommendations for diets suitable for feeding MM children. Nigel Rollins's (WHO) presentation on managing the needs of HIV-infected children emphasised how little is known about the relationship between HIV and MM in infected children and how there is currently no basis for recommending different nutritional management, apart from increased energy intake, compared to non-HIV infected children. A presentation by Mark Manary (St. Louis Children's Hospital) on recent attempts to supplement the diet of MM children to prevent kwashiorkor in Malawi, highlighted the lack of an evidence base to make specific recommendations for the dietary management of MM children in areas of high kwashiorkor prevalence.

Consensus statements

• The nutritional requirements of moderately malnourished children probably fall somewhere between the nutritional requirements for healthy children and those of children with severe acute malnutrition during the catch up growth phase.
• The nutrient intakes of moderately malnourished children need to be adequate to allow wasted children to synthesise the lean tissue deficits and to allow stunted children to achieve both accelerated linear growth and associated accrual of lean tissue.
• Diets with a nutrient density equivalent to F100 and a low anti-nutrient content, provided at an energy intake to support the desired rate of weight gain, are adequate to promote height and weight gain. Such diets may also be effective at restoring functional outcomes, including physiological and immunological function towards normal, in moderately wasted children.
• There is evidence that growth deficits can be treated (i.e. that catch up growth for height can occur) in children far beyond two years of age, and even in adolescents, provided that a high quality diet is sustained. Though there is no evidence of similar recovery of other deficits associated with stunting, such as cognitive deficits. However, the prevention of stunting should always be directed at the 'window of opportunity' from conception to the first 24 months of life, when most growth faltering occurs and impacts on health and brain development are greatest.
• Proteins used to feed moderately malnourished children should have a PDCAAS of at least 70%. Giving lower amounts of proteins with higher PDCAAS^d may be advantageous.
• The diets of children recovering from moderate wasting should provide at least 30% of their energy as fat. A higher percentage of energy derived from fat (35 to 45 %) might have advantages provided the density of nutrients is adequate.
• It is recommended that diets for moderately malnourished children contain at least 4.5% of their total energy content from n-6 polyunsaturated fatty acids (PUFA) and 0.5% of their total energy content from n-3 PUFA. It is advised that the ratio of linoleic/_-linolenic acid remains in the range of 5-15. A ratio within the range of 5- 9, however, may be preferable.
• When large quantities of nutrients known to have an effect on acid-base metabolism are added to foods, their potential effect on the acid-base balance of the body after being absorbed and metabolised should be estimated. Their overall effect should remain neutral.
• Energy needs of moderately malnourished HIV-infected children are increased by 20- 30% compared to non HIV-infected children who are growing well. There is no evidence for an increased protein requirement in relation to energy, i.e. 10-15% of the total energy intake is sufficient.

Research needs

It is unclear whether a diet adequate for treating a moderately wasted child will also be adequate to treat a stunted child. The length of time required for catch up growth is also not known. Further studies are needed to clarify the effect of the diet on the timing of linear growth in relation to weight gain.

Research is also needed on safe upper limits of different nutrients at different ages, as well as the requirements and importance of specific and often 'forgotten' nutrients like potassium, sulfur, phosphorus and selenium.

More field friendly techniques (like blood spot technology) for assessing deficiency of certain Type I nutrients are needed.

Research is required to understand better the pathophysiology of how HIV causes undernutrition, how HIV-related undernutrition differs from undernutrition due to other causes and how to distinguish between the different aetiologies.

Fundamental research is needed to obtain a better understanding of the pathophysiology of kwashiorkor. Currently, none of the proposed mechanisms for how kwashiorkor develops are supported by strong evidence that can be translated into preventive programming.

Foods and ingredients suitable for use in moderately malnourished children

A background paper (prepared by Prof Kim Michaelsen and colleagues from the University of Copenhagen, and Prof Tsinuel Girma, from the University of Jimma, Ethiopia) provided an extensive description of foods and ingredients most commonly used to feed MM children. Elaine Ferguson (London School of Hygiene and Tropical Medicine) presented a short paper explaining how linear programming can be used to check the nutritional adequacy (and assess the cost) of diets recommended for MM children.

Consensus statements

• The addition of animal source foods to a plant-based diet promotes the recovery of moderately malnourished children.
• Diets based exclusively on plant foods need to be fortified and processed in such a way to remove anti-nutrients.
• Diets with low anti-nutrient and fibre content are beneficial for promoting the recovery of malnourished children.
• Phytate may seriously limit the efficacy of plant based foods. The possibility of safely reducing its content by the use of phytase and/or food processing should be explored.
• Highly refined cereal flours (those with low extraction rates) have lower levels of antinutrients and dietary fibre than less refined flours. Highly refined flours cost more and have lower vitamin and mineral levels - although these vitamins and minerals are more bioavailable.
• Blended flours prepared with de-hulled legumes are preferable to those prepared with whole legume flour.
• Food processing techniques, including home -based processing techniques such as fermentation and soaking, can improve food quality, specifically nutrient bioavailability. The effect of anti-nutrients in complementary foods based on the family diet can be decreased by various traditional food processing methods such as malting or soaking. The feasibility and efficacy of these processing techniques for the management of moderate malnutrition should be assessed.
• The energy density of semi-solid foods can be increased by reducing the water content or by adding fat or sugar. Adding fat and sugar, however, decreases the nutrient density in relation to energy and is accept able only if the overall density of each and every essential nutrient is sustained at a level that supports normal balanced tissue synthesis.
• The increase in viscosity resulting from the reduced water content can be limited by using amylase or amylase rich flours.
• Foods with a high energy density often have a high renal solute load and may not provide enough water for recovering children.
• Children fed diets with a high solute load in relation to their water content may need additional water during and between meals. Breastfeeding provides large quantities of water in addition to a full range of nutrients. It has a low solute load and should always be encouraged before potable water when energy dense foods are provided.
• As most diets in poor countries have a low level of n-3 (omega-3) fatty acids and an inappropriately high ratio of n-6 fatty acids in relation to the n-3 fatty acids, foods with high n-3 fatty acid content should be promoted. These include soybean and rape seed oil, and fatty fish or its products. This is especially important for non breastfed children.

Research needs

There is uncertainty about the minimum quantity or type of animal source foods that are needed in the diets of MM children. Milk, and potentially eggs, seem to have advantages over meat and fish in terms of growth, but not in terms of improving micronutrient status. It is unclear whether children who are stunted but not wasted may benefit from different proportions of animal v plant protein in their diets, as compared to diets designed to treat wasting.

Local production of foritfied blended food

Research is also needed to assess whether dairy/ whey stimulates linear growth and / or reverses wasting in comparison to plant based foods (e.g. soy) with a high PDCAAS, low levels of anti-nutrients and low fibre content in malnourished children. The extent to which cooking/heat treatment denatures bioactive components of dairy products should also be investigated.

Data are needed on the maximum acceptable levels of intake of the most important antinutrients and of different types of fibres for MM children. There is also a need to establish upper acceptable limits for sodium and iron content of foods for MM children.

Dietary counselling for moderately malnourished children

This background paper (prepared by Prof Ann Ashworth, London School of Hygiene and Tropical Medicine) concluded that mothers of MM children are usually given the same general dietary advice as mothers of well-nourished children. The paper suggests that generic dietary recommendations developed by the WHO and Food and Agricultural Organisation (FAO) for well-nourished children may meet requirements of MM children if the recommendations are made more specific and context appropriate.

To date, there have been few studies of the efficacy of dietary counselling in treating MM. Studies looking at dietary counselling for MM report very different weight gains. Little information is available on other outcomes. Even height gains are rarely reported. Differences in reported weight gain are probably due to differences in initial nutritional status (stunted vs wasted). It was noted that one of the most effective pilot nutrition counselling programmes implemented in Bangladesh provided micronutrient supplements that may have increased its efficacy. Save the Children US presented data showing that large scale positive deviance programmes in Vietnam and other countries have not had a significant impact on reducing MM.

Consensus Statements

• Dietary counselling for the prevention and management of malnutrition in general is often weak or absent and should be strengthened for all caregivers, especially those of children aged less than 24 months.
• Dietary counselling, breastfeeding coun selling and improving feeding practices should always be part of the management of MM. This is essential even when food supplements are given.
• Formative research should always be carried out before formulating dietary recommendations. Only foods and feeding practices that are affordable, feasible and acceptable to families should be recommended.
• Where prior assessment indicates that it is not possible to provide all nutrients needed by the child using the accessible family foods, other approaches, (including the use of fortified foods, food supplements, or micronutrient supplements) should be recommended.

Research needs

One of the key research questions is whether to always to aim to maximise the rate of catch-up in wasted children and what the most appropriate delivery channels for dietary counselling are. Research into the effectiveness of a combination of approaches for addressing MM is also needed, e.g. infection control and nutritional support and the combined and separate impact of food supplements and dietary counselling.

In order to inform this research agenda, researchers need to report weight gain as g/kg/d (as well as % moving between different weight-for-height and height-for-age categories), disaggregate weight gain among wasted and non-wasted children and broaden the number of outcomes (e.g. body composition, height gain, immune function, morbidity). Overall, we need a better understanding of how to provide and deliver effective dietary counselling.

Food supplements used to treat moderate malnutrition in children

This background paper (prepared by Dr Saskia de Pee and Dr Martin Bloem, WFP) reviewed specialised food supplements that are currently used to treat MM children in different contexts. This includes fortified blended foods prepared with cereals and legumes as major ingredients, complementary food supplements providing nutrients and energy missing in the family diet, and micronutrient powders.

Dr de Pee and Dr Bloem reiterated that most supplementary feeding programmes (SFPs) for moderately malnourished children supply fortified blended foods (FBFs), such as corn soy blend (CSB) and wheat soy blend (WSB), in combination with oil and sugar. However, there are a number of shortcomings with FBFs such that they are not optimal for feeding moderately malnourished children and need to be improved and/or replaced by foods that better meet their nutritional needs.

Presentations from WFP, UNICEF and USAID described the various improvements the agencies all plan to make to their fortified blended flour products, e.g. increasing the energy density, adding dairy products, dehulling soybeans, possibly removing cereal germ, changing the proportion of energy from fat, improving the EFA and micronutrient profiles.

Improvements and adaptations to lipidbased nutrient supplements (LNS) and readyto- use foods (RUFs) are also being made by the members of the LNS Research Network (supported by grants from the Bill and Melinda Gates Foundation and with support of the USAID- funded FANTA-2 Project) and Valid International.

Papers on field research from Malawi (Prof Ken Maleta, Blantyre College of Medicine), China (Prof Chen Chunming, International Life Science Institute), Niger and Sierra Leone (Dr Susan Shepherd, MSF-Nutrition Working Group) and Ghana (Prof Kathryn Dewey, University of California, Davis) presented data on the impact and outcomes of using specialised products to treat and prevent MM in different contexts. For example, in Malawi, supplementary feeding of milk/peanut and soy/peanut fortified spreads to treat moderately wasted children resulted in slightly higher recovery rates than feeding with CSB. In Niger, a targeted MSF supplementary feeding programme for moderately wasted children using RUF had a 95% recovery rate. In Sierra Leone, soy peanut fortified spread resulted in higher weight gain and shorter treatment than premix CSB-oil.

Consensus statements

• There is an urgent need to develop clear terminology for the different specialised foods used to treat moderate malnutrition.
• When it is expected that a new food product will have at least equal impact on growth, morbidity and micronutrient status compared to an existing product (often a FBF such as CSB or WSB), then it is permissible to use this product in programmes for feeding moderately malnourished children provided that the product is acceptable to the beneficiaries. In that case, it is important to collect programme data to monitor the impact of this new product on the time needed for recovery of MM children, when the product is used for treatment, or on the occurrence of new cases of malnutrition if it is used for prevention. Concurrently, the efficacy of the new product should also be assessed under carefully controlled circum stances in the same or another area or country, depending on local possibilities. Such efficacy testing should include measures of physiological, immunological, cognitive and body compositional recovery as well as simple weight gain.
• It is very likely that different types of specialised foods and programme formats (e.g. blanket or targeted, dietary coun selling) will be used to treat, or prevent, moderate malnutrition in the future, depending on the context (security, prevalence of malnutrition, general food security conditions, etc). The next WHO meeting on moderate malnutrition, which will focus on programming issues, should endeavour to develop algorithms for determining what programme type and product to use in different situations.

Research needs

Areas of uncertainty still exist with respect to improving FBFs. These include: the impact of dehulling and degerming of soy, maize and wheat, addition of phytase and/or amylase to improve nutrient availability and food acceptability, maximum tolerable fibre content, the minimal quantity of energy provided by fat to ensure adequate energy intake, the amount/proportion of milk required in the formula, the possibility and efficacy of using plant protein isolates, especially soy protein isolates, as a possible substitute for dairy products. There is also a question regarding whether the anti-nutrient content of FBFs can be significantly reduced by encouraging farmers to produce crop types that have naturally lower concentrations of anti-nutrients.

Agencies urgently need to collect impact assessment data from the different products (FBF, RUF, LNS, micronutrient powders) being used to treat and prevent MM in different contexts so that field agencies/governments know which product to use in a given context.

The impact and outcome data for new products need to be comparable across studies and programme evaluation. Information on nonfood context factors should also be collected (e.g. programme incentives). The operational advantages of some products/programme types should be recorded (e.g. blanket distributions may be easier in food insecure areas).

It is essential to collect information on the costs of providing different types of specialised products, complementary interventions, and the means of distribution.

Recommendations of the meeting - Next steps

The meeting made a number of recommendations to move forward and to continue to improve current programmes in the next few years.

Establishment of a process to develop specifications for food categories for moderately malnourished children and validation of new products for prevention and treatment of moderate malnutrition in children.

Dr. Carlos Navarro-Colorado (Emergency Nutrition Network) presented a description of a generic approach to validate the efficacy of new foods for moderate malnutrition. This would need to be based upon clear classification of different types of food supplements required and the nutrient specifications for each category of food supplement. Four stages of validation were proposed:

1. analysis of composition and processing,
2. small scale clinical pilot,
3. field efficacy trial, and
4. post-validation monitoring. It would not be necessary to conduct all four stages for all products.

The design of studies and validation of products will face a number of challenges. These include lack of baseline dietary information, accounting for differences in the quality of programme implementation, the need to broaden and define outcome indicators beyond anthropometry, and accounting for the fact that an unknown proportion of moderately malnourished children will recover spontaneously. Another significant challenge will be how to establish an institutional mechanism and identify a lead agency for ensuring coordinated validation of products.

A working group then examined how to move forward and to set up a process of improving existing food supplements and ensure their efficacy is adequately evaluated.

Consensus statements

• A standing task force should be established and led by WHO with appropriate UN agencies and other technical experts to develop specifications for specialised products, in particular for moderately wasted children, in a first step.
• A separate expert group should be established to examine different endogenous food components that have potential negative effects and develop upper limits for these anti-nutrients and toxins. One of the tasks of this group would be to determine the maximum acceptable level of different types of dietary fibres and other potentially deleterious natural constituents that can be present in food supplements.
• There is a need for an independent standing working group to assist national governments and agencies to determine if newly available products that are put onto the market are appropriate and whether (a) particular type(s) of product testing are required before granting approval for their use among specific target groups.
• The meeting recommended that this set of activities should be initiated within the next six months.

Research Needs

In the discussions, the meeting also identified the need to estimate the level at which recovery from moderate malnutrition occurs in absence of supplementation so that this can be accounted for in trials involving new products. This can be achieved either by examining data from previous studies where some children did not receive any supplement or by taking as the control group in intervention studies a group receiving adequate dietary counselling but no food supplement.

There is also a need to elaborate specific nonanthropometric measures that can be used to compare outcomes and product efficacy. This will involve developing and strengthening field friendly techniques for measuring outcomes such as body composition, immunocompetence, micronutrient status, renal concentrating ability, physical activity level, sodium pump function, intellectual development, etc.

Organisation of a second meeting on improving programmes addressing the management of moderate malnutrition.

The focus of this technical meeting was dietary requirements of MM children, so that programmatic issues were not substantively addressed. The WHO is planning a further technical meeting on programming for MM children.

For more information, contact: Zita Weise- Prinzo, WHO, email: weiseprinzoz@who.int

^aWHO (2008). Proceedings of the WHO UNICEF WFP UNHCR informal consultation on the management of moderate malnutrition in under-5 children

^bAnother WHO consultation is planned to review the evidence on strategies and programmatic approaches to managing moderate malnutrition, which aims to answer questions not addressed in this meeting.

^cWHO. Management of severe malnutrition: a manual for physicians and other senior health workers. Geneva, 1999. Available at: http://www.who.int/nutrition/publications/severemalnutrition/en/manage_severe_malnutrition_eng.pdf

^dPDCAAS (Protein Digestibility Corrected Amino Acid Score) is a method of evaluating the protein quality based on the amino acid requirements of humans