MUAC alone admission to SAM treatment in Bangladesh

Summary of research¹

Location: Bangladesh

What we know: MUAC and weight-for-height z score (WHZ) are common anthropometric criteria to diagnose severe acute malnutrition (SAM). They identify different populations of children with some overlap; MUAC appears to be more sensitive to high mortality risk children and to select younger children. There is less evidence of the application of the criteria in Asian populations.

What this article adds: A prospective cohort study of 158 children in an urban slum area of Bangladesh examined the short term consequences of a MUAC only admission policy to nutrition treatment. Children with uncomplicated SAM diagnosis based on WHZ only (MUAC > 115mm) were not admitted for treatment but followed bi-weekly for 3 months, with referral for medical/nutritional deterioration. Thirteen per cent were lost to follow up, 6% deteriorated, 58% had an improved z score and 23% stayed the same. Children who did best had less medical complications. Higher MUAC cut-off may capture higher risk WHZ children.

The World Health Organisation (WHO) recommends the use of two independent anthropometric criteria to diagnose severe acute malnutrition (SAM) among children aged 6–59 months. These two criteria are mid-upper arm circumference (MUAC) <115 mm and weight-for-height Z-score (WHZ) <3. The use of MUAC has been endorsed by the WHO for the community-based management of acute malnutrition (CMAM)². In practice, MUAC is used on a large scale as a single tool for detecting SAM because it is cheap, and community health workers (CHWs) learn to use it effectively with minimal training. The tool is also well accepted by children, due to the simplicity of its measurement. For these reasons, MUAC is a particularly useful screening tool for nutritional assessments in emergencies and for household screening.

MUAC appears to be more sensitive than WHZ in identifying high risk SAM children and predicting mortality, particularly in the younger age group (6–23 months). However, MUAC and WHZ identify different populations of SAM children with only some degree of overlap. Using only MUAC to diagnose SAM may leave out a proportion of high-risk children who might be severely malnourished as determined by WHZ. Is it justifiable to leave such potentially vulnerable SAM children out of a nutritional rehabilitation programme? In particular, how would such children fare in terms of nutritional evolution, morbidity and mortality?

These questions were examined in the context of a Médecins Sans Frontières (MSF) implemented CMAM programme in Kamrangirchar slum in Dhaka, Bangladesh. Routine analysis of nutrition programme data showed that the largest proportion of SAM children were admitted based on WHZ of <–3 while their MUAC was >115 mm. Researchers undertook a prospective cohort study of this group of children in order to assess their nutritional outcomes and report on their morbidity and mortality.

The site of the study, Kamrangirchar, has a population of 400,000 living in an area of 3.1 km2. MSF started providing healthcare services for children in Kamrangirchar from May 2010. All services are provided free of charge through two primary healthcare centres (PHC). The project targets malnutrition among children <5 years of age through the CMAM approach. The management protocols are in line with recommended WHO guidelines. The study was conducted between June 2011 and February 2012 and the study population included children aged 6–59 months (height/length between 65 and 110 cm) with a WHZ of <–3 and a MUAC >115 mm. In view of the context of prevalent stunting, children aged over 59 months but with a height <110cm were also included.

Study group recruitment

Children included in the study were recruited by assessing both MUAC and WHZ. Those with WHZ <–3 and MUAC >115 mm were not admitted to an ambulatory therapeutic feeding centre (ATFC) but recruited for a 3-month follow-up period. Screening took place both at the PHC and in the community. Children recruited in the PHC were assessed clinically and received treatment for any medical conditions. Active door-to-door screening in the community was done by a team of CHWs. MUAC was measured using a colour-coded measurement tape graduated in millimetres. Height/length was measured using a wooden stadiometer with a precision of 0.1 cm. Weight was measured using a hanging scale accurate to 100 g. During the door-to-door screening, children found with symptoms of fever, cough or diarrhoea were referred to the PHC for further clinical assessment. Children found with MUAC<115mm or oedema, or both, were excluded from the study and admitted to the ATFC as standard practice. Children with severe medical complications at baseline (measles, lower respiratory tract infection, severe anaemia, tuberculosis (TB) and severe diarrhoea) were excluded from recruitment.

Follow up

On recruitment, all carers were briefed on the study procedure. A unique identity card was given to the carer to facilitate the follow-up of the child’s medical conditions at the PHC if required. Follow-up home visits were done every 2 weeks by a team of trained CHWs. During the follow-up visit, weight and MUAC were assessed and history of any symptoms of fever, diarrhoea (three or more loose motions per day) or cough in the past two weeks was taken. Children were referred for admission to the ATFC if their MUAC dropped to <115 mm and/or they had lost ≥10% of their baseline weight. Children were also referred to the PHC for clinical assessment and treatment if they were found with any illness. No child recruited into the study was given any ready-to-use therapeutic food (RUTF) unless they required admission to the nutrition programme.

Sample size

The sample size was calculated based on a hypothesis that if children with WHZ ,<–3 and MUAC >115 mm did not receive nutritional rehabilitation, 30% would lose ≥10% of baseline body weight or would need admission to the nutrition programme as a result of the high morbidity and mortality risk during the prospective follow-up period. To detect the above with 95% confidence and 5% error, the required sample size was calculated to be a minimum of 126 children. To correct for losses-to-follow-up in the context of a generally mobile population, the sample size was increased to 158 children.

Results

A total of 158 children were recruited into the study, with a median age of 41 months; 58% were male. Ninety-six (61%) children were recruited through the PHC facilities and 62 (39%) through door-to-door screening. At recruitment, the mean MUAC was 126 mm (SD = 19.6, range 116–144 mm), the median weight was 9.6 kg (range 5.7–13.7 mm) and the median height/length was 89.6 cm (range 67–109.5 cm). Because of the frequency of stunting among children living in Kamrangirchar, 28 children with a height of >65 cm and <110 cm and aged more than 59 months were included.

In the course of the study, a total of 1,002 follow-up home visits were done (average = 167 per 2 weeks), including 222 visits during which children were unexpectedly absent. Of the 158 children recruited into the study, 21 did not complete the 3-month follow-up period; 15 (9%) children left Kamrangirchar and six (4%) children were lost to follow-up.

Of the remaining 137 children with data on nutritional evolution and outcomes:

• Nine out of 158 children (7%) were admitted to the ATFC; five had their MUAC drop to <115 mm, one had weight loss ≥10% of the baseline weight, two had pulmonary TB and one had severe watery diarrhoea.
• Amongst the five children whose MUAC dropped below 115 mm, baseline MUAC was 116 mm in three children and 118 mm in two children. The drop in MUAC occurred after two weeks in two of the children, after 4 weeks in a further two and after six weeks in one child.
• Of the remaining 128 children who completed three months of follow-up, 91 (71%) improved in nutritional status: 81 (63%) progressed to a WHZ ≥–3 and <–2 and 10 (7%) to WHZ ≥–2. The remaining 37 (29%) children maintained a WHZ of <–3.

There were significantly fewer symptoms of cough among those who improved in nutritional status compared with those who maintained a status quo with WHZ <–3 (p = 0.003). There was no other significant difference in morbidity in the groups when compared, with WHZ <-3 as baseline. There was one death, from pulmonary TB.

Discussion

This study shows that most (93%) children who were severely malnourished in terms of their Z score (WHZ <–3) but judged as less severely malnourished on the basis of their MUAC (>115 mm) improved in their nutritional status or maintained a status quo after three months of follow-up. In this context, it would therefore seem possible to rely solely on MUAC for screening and admission into nutritional rehabilitation programmes. One strength of this study is that it is one of the first prospective studies to assess by WHZ nutritional evolution and outcomes among severely malnourished children in the Asian context. Rigorous attention was paid to follow-up visits at home and to staff training on anthropometric measurements, and the CHW team was experienced.

The study has a number of limitations. First, it followed children for a relatively short period of three months. Second, although there are no data to substantiate this, it is plausible that the bi-weekly home visits might have had a positive influence on malnutrition-related behaviour of carers or on the use of healthcare facilities, which might also have affected the results. Third, the findings of this study are representative of a cohort of SAM children without complications, as children with severe medical complications at the baseline were excluded from recruitment. Fourth, it was not possible to distinguish between new episodes of fever, cough and diarrhoea and persistent symptoms from the previous episode.

It is surprising that, in all, around seven in 10 children with WHZ<–3 gained weight, and moved into the moderate malnutrition category or achieved a normal nutritional status according to their WHZ, in the rather short three month follow-up period. A smaller proportion (27%) maintained their status quo. Both findings are encouraging. It is also reassuring that, despite most children having had one or more common morbidities during follow-up, only nine required admission for nutritional rehabilitation. Three of these nine children were admitted as a result of severe medical conditions that would themselves have merited the child’s admission, irrespective of nutritional status. The one death among the study population was of one of these three children, from pulmonary TB. From a purely nutritional viewpoint, this implies that only six (4%) children were negatively affected by their initial nutritional status, which supports use of MUAC alone to screen children for admission to nutritional programmes.

A recent study in rural Bangladesh also demonstrated the feasibility and effectiveness of the CMAM approach, where CHWs used MUAC for community screening and assessment for admission and discharge. Although 27% of children in the present study maintained their status quo in terms of their nutritional evolution, it is reasonable to think that this group might be at a relatively higher risk of adverse events. Appropriate and well-timed access to healthcare for such children would limit the potential deterioration of their nutritional status. Therefore it might be justified to conduct specific research to assess what increased MUAC cut-off threshold (above 115 mm) would be ideal to maximise inclusion of this subgroup. Experience in an African context in Burkina Faso³ suggests the use of a MUAC cut-off of 118 mm or oedema, or both, as admission criteria for SAM children, may be a useful alternative to WHZ.

¹Ali.E et al (2013). Is mid-upper arm circumference alone sufficient for deciding admission to a nutritional programme for childhood severe acute malnutrition in Bangladesh? Transaction of the Royal Society of Tropical Medicine and Hygiene. Doi10.1093/trstmh/trt018

²World Health Organization. Community-based management of severe acute malnutrition: A joint statement by the WHO, the World Food Programme, the United Nations System Standing Committee on Nutrition and the United Nation’s Children’s Fund. http://www.who.int/

³Goossens S, Bekele Y, Yun O et al (2012). Mid-upper arm circumference based nutrition programming: evidence for a new approach in regions with high burden of acute malnutrition. PLoS One 2012;7:e49320.