Menu ENN Search
Language: English Français

MUAC and oedema

This question was posted the Assessment and Surveillance forum area and has 18 replies. You can also reply via email – be sure to leave the subject unchanged.

» Post a reply

Susana Moreno Romero

Nutrition and food security specialist

Normal user

24 Apr 2012, 11:52

Hi there, I have read several times in this forum that MUAC is NOT affected by oedema or pregnancy. I can not find any study about it. Could anyone give me some references please? Thank you

Mark Myatt

Frequent user

24 Apr 2012, 19:26

Difficult to show that MUAC is not "affected by oedema" as oedema often presents alongside low MUAC. One reason for using MUAC is that oedema is not a well recognised sign in (e.g.) PHC clinics (i.e. MUAC picks up oedema as well as cases of simple wasting).

MUAC may be affected by pregnancy (others are more expert than I am) but it is less affect than weight (and hence BMI). There are very many studies showing strong associations between maternal MUAC and poor birth outcomes.

What exactly do you want references on? There is a lot of literature there. Try "MUAC pregnancy", "MUAC birthweight", "MUAC birth outcome", &c. in PubMed. That will bring up quite a few references.


Normal user

25 Apr 2012, 06:02

It depends on the degree of oedema. MUAC may not be much affected by odema + and ++ but its definately affected by oedema +++. In such cases of oedema +++ MUAC may be useful in monotoring after oedema has subsided.


Frequent user

25 Apr 2012, 06:58

Dear Anonymous,

MUAC has been shown to be less affected than weight-for-height by hydration status in children with diarrhoea. See:

Mwangome MK, Fegan G, Prentice AM, Berkley JA. Are diagnostic criteria for acute malnutrition affected by hydration status in hospitalized children? A repeated measures study. Nutr J. 2011 Sep 13;10:92. Available at:

So it seems likely that MUAC is less affected by oedema as well, but to my knowledge this has never been formally tested. A study measuring MUAC repeatedly is a group of oedematous children during treatment would be useful in this respect.

I hope this helps,


Susana Moreno Romero

Nutrition and food security specialist

Normal user

25 Apr 2012, 08:48

Really helpful,

So, according to this, we could say that MUAC is LESS affected by oedema or pregnancy than weight or weight relative measures (W/H, W/A or BMI), but a relative impact cannot be ruled out.

Many thanks to all.


Mark Myatt

Frequent user

25 Apr 2012, 09:32

I am not sure what you mean by "relative impact".

I am not even sure that there is an issue here at all.

We use MUAC in pregnancy and have case-defining thresholds that are calibrated by outcome. Whether pregnancy affects MUAC or not is IRRELEVANT because the thresholds have been established from analysis of many cohorts of pregnant women in many settings. The selected thresholds vary a little but that is a reflection of the relative weights given to sensitivity and specificity in different settings (usually informed by resource availability) rather to variability in underlying risk.

In young children we use MUAC and oedema as independent admission criteria. If (e.g.) a child has oedema then they are admitted regardless of their level of MUAC. If (e.g.) child has MUAC < 115 mm they are admitted regardless of whether they have oedema. In the case of monitoring using MUAC, studies are underway in settings in which a high proportion of patients are admitted by oedema. Preliminary analysis shows MUAC gain with oedema loss. Discharge is always on loss of oedema and MUAC above a discharge threshold. This also applies to adults (threshold differ). Here also, the thresholds are calibrated by outcome in many cohorts in many settings.

The same (i.e. MUAC and oedema as independent case-definitions) applies to survey applications.

I am interested in your motivation ... please explain what you think the issue is. Why is it important for you to state "a relative impact cannot be ruled out". It seems to me to be confusing nitpicking? I cannot see what practical difference such a statement could would make other than to confuse practitioners and impair programming to the detriment of beneficiary populations. I fear that I may be missing something important here. PLEASE EXPLAIN.


BTW ... partly in response to Florence ... It would have to be massive generalised oedema for it to affect MUAC. Such a child (or adult) would be so sick that the MUAC would be a side issue at admission.

Mija ververs

Normal user

25 Apr 2012, 10:20

Dear Anonymous
My answer is exclusively on MUAC and pregnancy, not on oedema.
I studied for MSF-CH what anthropometric indicator to use in the field for pregnant women (to identify risks to adverse birth and maternal outcomes). What I saw in the literature is that maternal MUAC seems to be relatively stable during pregnancy in women from developing countries and is considered to be useful for screening of maternal fat and lean tissue stores in pregnant and non-pregnant women.

However, recent research (2008 and 2009), demonstrated that MUAC can change during pregnancy (and lactation), albeit this was studied in young to very young women (so perhaps not what you are looking for, as these changes have possibly more to do with adolescence, i.e. redistribution of fat, than pregnancy….). MUAC changed during pregnancy and lactation (amongst women <25 yr):
- it increased in pregnancy
- it decreased post-partum (6-12 weeks after)
I did not find any studies that demonstrated changes in MUAC of adult pregnancies above 25 years.
As you asked for references, See also:
- Maternal mid-arm circumference and other anthropometric measures of adiposity in relation to infant birth size among Zimbabwean women. Ogbana C, Woelk GB, NING Y, et al. Acta Obstetricia et Gynecologica. 2007; 86: 26_32.
- The post-partum mid-upper arm circumference of adolescents is reduced by pregnancy in rural Nepal. Katz J, Khatry SK, LeClerq SC, et al. Matern Child Nutr. 2010 Jul 1;6(3):287-95.
- Pregnancy and lactation hinder growth and nutritional status of adolescent girls in rural Bangladesh. Rah JH, Christian P, Shamim AA et al. J Nutr 2008;138, 1505–1511.

But I agree with Mark, do a search on the by him provided key words and a lot will come up.

PS Because of all this MUAC is such a great anthropometric indicator as you don't need to know anything of the pregnant women' s gestational age!!!!

regards, Mija

Mark Myatt

Frequent user

25 Apr 2012, 11:30


Thanks for that.

Can you confirm that MUAC thresholds for predicting adverse negative and maternal outcomes are stable from setting to setting? Is the variation due to resource availability / health service priority or to differences in underlying risks, or a mixture of both? What strikes me is that the recommended thresholds are similar from setting to setting (e.g. a centimetre one way or the other). If it not too much work for you, can you present (in this forum) a short summary table of MUAC thresholds in current use.

I think that the data on "teenage mums" is very interesting. Am I right in thinking that these are very young mothers (i.e. 14 or 15 years old rather than just < 25 years old)? I don't think this invalidates the use of MUAC in pregnancy as (correct me if I am wrong) the risk of low MUAC in these younger mothers is little different from in older mothers but the younger mothers are more at risk of low MUAC and adverse outcomes.

The practical difference that this data makes is that we might consider "teenage mum" as a risk category and provide nutritional support regardless of MUAC.

BTW ... shouldn't we be seeing increased MUAC (i.e. increased muscle mass) with increasing age. If this is correct then these mothers may be growth faltering due to the pregnancy. Perhaps this is complicated by changing fat distribution but we might expect "puppy fat" to be "lost" mostly in the pre-pubertal / pubertal growth spurt. I'm straying outside my expertise here.

I must stress again that, for our purposes (i.e. prevalence surveys, screening, and admission), "normal" changes in MUAC during pregnancy are irrelevant because we have established thresholds from cohorts of pregnant women.

Mija ververs

Normal user

25 Apr 2012, 12:06

Dear Mark and colleagues
The discussion is now redirected more into MUAC and pregnancy in general. And…what indicator to use.
This is what I found:
PLW are often included in nutritional programmes, most frequently supplementary feeding programmes (SFP). However, there often lacks any rationale on who to include when and how. Similarly, many programmes lack objectives on what the nutritional aims are for those PLW and products are provided based on availability rather than biological needs.

Over the last 5 years there is an increasing number of national nutrition protocols that use to some extent anthropometric indicators for inclusion of PLW. As per Mark’s request, here an overview. The cut-off for inclusion varies from MUAC <185 mm (Zimbabwe 2008), <210 mm (Madagascar 2007, Malawi 2007, Burkina Faso, DRC 2008, Guinea 2005, Mali 2007, Senegal 2008, Niger 2006, Burundi 2002), <220 mm (Mozambique 2008, UNHCR 2007), <225 mm (Zambia 2009). MUAC cut-off point of 230 mm has also been used (Indonesia 1996, Sri Lanka 2006).
Some protocols enroll PLW on the base of gestational age (mostly only in 3rd trimester) or if they have an infant <6 months (malnourished or not).
The majority of PLW are included in SFP because they are pregnant or lactating, not because of specific needs that are spelled out.
So….. we are doing something with MUAC in pregnancy, but it is not based thoroughly on extensive research (at least not in global terms: in emergencies we tend all to do the same for the same population groups, and cut-offs lower due to lack of inputs such as food/nutritional products). In simple terms, we need to know which threshold of MUAC (or another indicator) has proven to be harmful for the infant or mother healthwise.
Unfortunately, I am still working with MSF on this and results will come out, hopefully, in 2012.
BUT, I can already tell you that all the studies I looked at, settings are different (chronic food insecurity, ‘normal’ situation, impoverished settings, etc). So based on these heterogenous settings, I see now certain common thresholds of MUAC below which adverse outcomes increase. Overall, the thresholds in Asian populations seem to be slightly lower than in African settings (difference genetic make-up?....or something else, who knows…).
That’s all for now that I can say. I realise I have not answered all your questions. Eventually we hope, with MSF-CH, to demonstrate which MUAC thresholds provide which kind of risk, and the issue will be addressed whether we should differ per continent (Asia, Africa), per country, etc. It is clear whatever threshold for MUAC we will find, it should not be lowered because there is an emergency or not enough resources. That would certainly not make any medical sense.
And yes Mark, the teenage pregnancies were, if I recall well, below 16 years.
Thanks, Mija

Mark Myatt

Frequent user

25 Apr 2012, 13:52


Very good. Thanks. This is (IMO) important work. I look forward to seeing the results. I am sure that we all do.

Susana Moreno Romero

Nutrition and food security specialist

Normal user

25 Apr 2012, 16:19

Thanks Mija, very interesting.

Well Mark, I never thought that to clarify one point on these issues could be “confusing nitpicking” or have such negative impacts on programming. Sorry, but my opinion is exactly the opposite. I find the discussion and on-going research very interesting.

I don’t see any problem to clarify that pregnancy or oedema could have an impact on MUAC:
Because oedema could have, oedema and MUAC are independent admission criteria or case definitions.

Because pregnancy could have and/or biological importance of MUAC during pregnancy or lactation is different, thresholds for pregnant women are different (please correct me if I am wrong).

I don’t understand. Why to say that MUAC is not influenced by these factors if it could be? I find it an unnecessary and excessive simplification that was confusing and not very useful.
Thanks anyway.

Mark Myatt

Frequent user

25 Apr 2012, 17:06

There are many people opposed to the use of MUAC or are sceptical about the use of MUAC. They will take statements such as the ones you are making here as a reason not to use MUAC. Failure to use MUAC does limit program activities, community involvement, and coverage. This is to the detriment of beneficiary populations.

"Why to say that MUAC is not influenced by these factors if it could be? I find it an unnecessary and excessive simplification that was confusing and not very useful" is a very misleading statement because it is context free.

In the context of this forum (i.e. the assessment of individual and population need in emergency settings), the MUAC thresholds we use in pregnant women are developed from cohort studies of pregnant women. What you see as important and interesting is irrelevant detail in that it can make no practical difference in how we should behave.

In the context of this forum, oedema and MUAC are independent criteria. What you see as important and interesting is irrelevant detail in that it can make no practical difference to how we should behave.

I think you are confusing "excessive simplification" with an absence of irrelevant detail.

As I wrote above "I may be missing something important here". Am I? Please let me know by giving me an example of how what you find important and interesting might affect our behaviour. Prove me wrong. That would be interesting.

Chantal Autotte Bouchard


Normal user

25 Apr 2012, 20:54


I would like to reply to the following comment made by Mark - “Discharge is always on loss of oedema and MUAC, above a discharge threshold”.

My question is the following, using the following example:

A child is admitted according to their MUAC (with or without Oedemas (here we have approximately 16 % of oedemas)).

Mark said “Discharge is always on loss of oedema and MUAC above a discharge threshold”.

My question is “Above a threshold, but what threshold?”

Is the threshold determined in relation to the criteria of admission of that child or with another criterion?

Here we notice that the children stay in the program for a long time as if they are admitted by their MUAC, it will take them a long time to reach exit criteria MUAC (normally if they are admitted by their MUAC they need to reach exit criteria with the same measure, i.e. MUAC). They will more quickly reach exit criteria for the weight / height in the standards, but the MUAC exit criteria takes much longer.

In the national protocol the discharge criteria always needs to be the same as the admission criteria. If the measure is MUAC the child will stay in the program a long time. If we could use different criteria for Entry i.e. MUAC and for Exit criteria i.e. Weight / Height our length of stay could be shorter?

Thank you C.

Tamsin Walters

en-net moderator

Forum moderator

26 Apr 2012, 10:36

From Nikki Blackwell:

in my opinion adolescent pregnancy is a reason for nutritional support regardless of the muac

- we know that contrary to the case of the pregnant adult woman nutrition is preferentially directed to the foetus; this is not the case with pregnant adolescent girls where there are competing needs - the growth of the girl vs her foetus. furthermore, research shows that adolescent girls given nutritional supplementation increase their pelvic size (decreased risk of obstructed labour) more than the birth weight of the baby increases (the concern being that supps to these girls would lead to more obstructed labour because the babies would be
too big)

being pregnant and adolescent should be a de facto reason to receive nutritional supps in food insecure situations (niger; most of the sahel countries)

Nikki Blackwell


Pr Nikki Blackwell
Professeur associé de Soins Critiques (Université de Queensland, Australie)
Directrice Médicale ALIMA
Coordinatrice de recherche

Tamsin Walters

en-net moderator

Forum moderator

26 Apr 2012, 14:10

Dear all,

Please could I remind discussants to be respectful of each other's thoughts and suggestions on en-net. Personal or derogatory comments are not appropriate. The forum promotes open discussion and sharing of ideas and dilemmas.

The Terms and Conditions of use can be found here:

Many thanks

Tamsin Walters
Forum Moderator

Mark Myatt

Frequent user

27 Apr 2012, 17:33


We might be better moving this to one of the treatment forums. They are people more expert than I on those forums.

Discharge thresholds vary from program to program. I have seen two main thresholds in use. These are > 115 mm and > 125 mm. I am involved in a study looking at safe discharge levels and we are using > 125 mm. We are closing recruitment on this study soon. We have seen some cases of relapse to MAM (i.e. MUAC < 125 mm but > 115 mm) but no cases of relapse to SAM (i.e. oedema or MUAC < 115 mm) in three month follow-up. It looks like > 125 mm is a safe discharge criteria when discharging to the community. I am not sure about > 115 mm. Reports from programs are of little or no relapse but follow-up may be patchy. If pushed, I would say > 115 mm is probably safe for discharge to SFP or in a good community-based program where follow-up and re-admission is likely.

I am mostly working with MUAC-only programs. These admit of MUAC (or oedema) and discharge on proportional weight gain. This is not a great method as it means that the most wasted get the least treatment. Some programs have begun discharging on MUAC (as the the study above and in some MSF programs ... there may be many more I don't know about) and this solves the problem of the least needy getting the most treatment and has reduced lengths of stay considerably.

In the study above we have been seeing lengths of stay of 8 weeks (median). In a study in Bangladesh with very high coverage we have seen lengths of stays of just over 4 weeks (median) due to early treatment.

I do not recommend the use of W/H for any purpose (I have many good reasons for this which you can find on these forums). A child with a MUAC < 115 mm is at considerable risk of mortality (15% - 20% die within a year) and should (IMO) not be discharged. MUAC thresholds are based on mortality / relapse risk (i.e. children with MUAC < 115 mm are at several times the baseline mortality risk, children at MUAC > 125 mm are at about baseline mortality risk) rather than some arbitrary statistical measure ascertained from examining a population of kids living in ideal conditions (as with W/H).

The correct ways to get your lengths of stay down are to promote compliance in centres (e.g. give the full CTC protocol) and in beneficiaries (e.g. reduce sharing of RUTF) and to organise the community aspects of a program to promote early treatment seeking. In one program in Bangladesh we had > 75% of admission at or just below admission criteria and had a length of stay of 30 days (median).

I hope this is of some help. As I say ... might be better to move this to the SAM treatment forum.


I agree about rude and disrespectful people. I'm still waiting for a reply (asked twice).

See my response to Chantal (above). Should we moved some of this thread?


I agree.

Chantal Autotte Bouchard


Normal user

2 May 2012, 16:31

Hello Mark,

Thank you for this answer.
I do not think that finally my question was move, as I took a lot of time.

In brief, the answer is very interesting, and I am going to re-bend over it more in detail and to see the most sensible choices has to make with the means which we have (we already have the psychosocial follow-up to limit the sharing among others but not only, and the community mobilization) but especially with the means which the government could have after our presence and see if discussion for the revision of the protocol (not planned) for the moment could concern it and see how we could work together.

I am going to find and to bend over the studies concerning specifically it (I believe that MSF already published on the subject). thank you

Mark Myatt

Frequent user

2 May 2012, 16:45

I believe that MSF-CH has a paper in-press about this issue based on their experiences in Gedaref, Sudan. I think they found that length of stay and weight gains in program were negatively associated with MUAC (i.e. lower MUAC = longer stays and greater proportional weight gain) when MUAC is used for discharge. You should contact then directly for details.

Chantal Autotte Bouchard


Normal user

2 May 2012, 19:15

Thank you Mark, I ask to my desk to contact them about that just after I sent my post this morning.


Back to top

» Post a reply