Iron during first phase of malnutrition treatment
Iron during first phase of malnutrition treatment Ariane Curdy 14.09.2000
Re: Iron during first phase of malnutrition treatment André Briend 14.09.2000

Date: 14.09.2000 1326

Subject: Iron during first phase of malnutrition treatment

From: Ariane CurdyatICRC_GVA


Dear All

It is usually agreed on that no iron should be given during the first phase of malnutrition treatment.

At times, a two weeks-ban for iron is given.

But all commercialized HEM formulas (both 75, and 100) do contain iron, varying between less than 0.4 mg to less than 1 mg per liter. Couldn't find any cut-off points of what level was considered as acceptable. Does anybody has an answer on that?

Thanks .... and keep smiling. Ariane

Date : 14.09.2000

From: "André Briend" <>

Subject: Iron during first phase of malnutrition treatment


Dear all,

To answer Ariane's question, it helps to go back to the origin of the current "ban" on iron during phase 1 of treatment of malnutrition.

As far as I understand, all this started by several observations

a) There is an association between plasma ferritin concentration / iron status and mortality from PEM.

See Ramdath et Golden, Non hematological aspects of iron deficiency. Nutrition Research Review, 1989; 2 29-50 (quoted in JC Waterlow's book p 140).

Ferritin levels are usually considered to be a reflexion of iron body stores, and these data suggests that higher body reserves increase mortality (maybe through increased oxydative stress or making iron avaialble for bacterial growht). Yet, ferritin is also a marker of infection, and the interpretation of these resutls is not so straightforward.

See also McFarlane et al. Immunity, transferin and survival in Kwashiorkor. Brit Med J 1970; 4 268-70.

b) Anaemia during severe malnutrition is usually not related to iron deficiency.

Apparently, severely malnourished children do have stores of unused iron in the bone marrow, which suggests that their anaemia is not related to iron deficiency.

See Fondu et al. Protein energy malnutritoin and anemia in Kivu. Am J Clin Nutr 1978; 31 46-56.

c) Giving iron supplement apparently increases mortality.

Iron supplements were routinely given until a few years ago (see the WHO protocol published in the 80's), but enthusiasm was refrained following a paper from Nigeria showing that mortality decreased when iron tablets were missing and they could not give iron supplement.

See Smith F, Taiwo O, Golden MHN. Plant protein rehabiliation diets and iron supplementation of the protein energy malnourished child. Eur J Clin Nutr 1989, 43 763-768.

Interestingly, in this paper (which resulted in a change of protocols in NGO's when results were known) there is no mention of the dose of iron given. (I understood from Mike that the dose was 'nutritional'. One may assume that it was not very different from 10 mg/day, but I would surprised it would be much less given the dosage of usual iron tablets, maybe more). You can also calculate from the table in the paper that the difference in mortality when children received iron was not statistically significant

Given the weight of evidence, none in favour of iron, doubts about potential risks, NGO's decided to "ban" iron SUPPLEMENTS in phase 1 of tt of severe manutrition.

Yet, this does not mean we should try to give a totally iron free diet. Actually, this would be very difficult and prevent the use of DSM in formulas. There is some iron in cow's milk, (the calf need some) and this explains that all F100 F75 in the market have some. UN specifications for indusrtrial F100 preparation accept that and the request usually made to F100 producers is NOT to add iron. The UN limits of iron in F100 formula were adjusted for iron content in DSM, but it was realised afterwards that iron content of DSM may vary (according to season) and the UN initial limits may be slightly too tight.

Please note that levels found by Ariane are much less than the dose presumably given as supplement in the Nigeria study.

Other remark iron in milk is also known to be poorly absorbed, and in any case much less than iron given between meals. Iron tablets are more toxic than iron in food. Even micro-encapsulaton of iron dramatically reduces toxicity in acute animal models.

See Boccio et al. Bioavailability, absorption mechanism and toxicity of microencapsulated iron (I) sulfate. Biol Trace El Res 1998; 62 65-73.

In conclusion, the pendulum should not go too far, and until there is more evidence, be happy provided iron is not added to the F100 formula.

I hope this helps.

Best regards,


André Briend, MD, PhD
5 rue du Vertbois, 75003 Paris, France
tel 33-1-53 01 80 36, fax 33-1-53 01 80 05