Cholera in severely malnourished patients
Cholera in severely malnoursihed patients André Briend 15.04.97
Cholera in severely malnourished patients Michael H.N. Golden 15.04.97
cholera in severe pem André Briend 15.04.97
Cholera in severely malnourished patients Michael H.N. Golden 15.04.97
cholera and iv fluids in severely malnourished Michael H.N. Golden 16.04.97
Re: Cholera and PEM George Fuchs 16.04.97
Cholera, IV and malnutrition André Briend 18.04.97
Cholera, IVF, and PEM George Fuchs 22.04.97

From André Briend

Tue Apr 15 08:29 BST 1997

Subject: Cholera in severely malnoursihed patients



There are cholera cases in Burundi in TFC. The question rasied from ACF is how to rehydrate these patients ?

I feel embarassed to say more that only a good clinician can strike the balance between the risk of fluid overload and of collapse by insufficient rehydration in a malnourished patient.

I wonder what to do for K.

Resomal has not enough Na to be used in cholera, and WHO ORS has not enough K to be used in severely malnourished patients.

What about adding 1 Therapeutic CMV dose to 1L WHO ORS ? This would lead to an easy to prepare ORS with 90 mEq /L Na and 50 mEq K + Mg which would not harm. Will this be too much K ? I suspect that if we add K requirements for cholera and severe malnutrition at the early stage, this may be all right;

Any comment welcome,


Dr. André Briend

INSERM U 290, Hopital Saint Lazare

107 rue du Faubourg Saint Denis, 75 475 Paris Cedex 10, France

tel 33-1-45 23 24 07, tel (direct) 33-1-48 00 56 04, fax 33-1-47 70 28 35

Subject: Cholera in severely malnourished patients

Date: Tue, 15 Apr 1997 13:33:08 +0100 (BST)

From: "Michael H.N. Golden" <>


Dear André

I am very worried about your suggestion of adding the mineral/vitamin mix to WHO-ORS for those with cholera and severe malnutrition. The osmolality will be to high (precisely the situation we need to avoid in these patients in particular) and the amount of potassium (not concentration) that would be given is far to high for any person in whom we are trying to replace catastrophic losses and will have pre-renal failure.

Resomal is not designed for this purpose.

If there is cholera-like diarrhoea (stool losses of over 10% body weight per day) then I would treat the children in the classical way with WHO-ORS without any addatives. It is precisely this situation that WHO-ORS is designed to treat and there is very extensive experience with its use. The urgency is to restore the extracellular/intravascular fluid compartment and everything else takes second priority.

(When the purging rate is much above this (>20% body weight per day for example) then there is even a good case for giving a solution that has higher sodium and lower potassium than the WHO-ORS! - but this is uncommon and usually requires IV fluid to manage these patients - particularly if they are children or are so exhausted that they cannot drink sufficiently)

For the marasmic children there is a good case for using standard ORS when the diagnosis of dehydration due to diarrhoea in unequivocal at any rate - it has been well tested!

For oedematous children there is no information on what fluid is best when they have catastropic diarrhoea - when in doubt use the standard management!

For Lower rates of stool losses than 10% body weight per day then I would use resomal in the malnourished children.

The issues are:

1) with watery diarrhoea there is a loss of both sodium and potassium.

2) as the purging rate increases the potassium concentration in the stool drops and the sodium concentration rises. The classical paper of Darrow is quite clear on this - The WHO ORS was DESIGNED initially for cholera because the electrolyte solution approximates what is lost with severe purging.

3) already malnourished children have a deficit in whole body potassium and magnesium before they get diarrhoea, and this will affect them adversly.

4) the malnourished children are very prone to heart failure due to volume overload. Particularly, the children with kwashiorkor.

These are the arguements that have used to increase the potassium and decrease the sodium content to give ReSoMal (Rehydration Solution for the Malnourished) and to add magnesium and zinc to the mixture. This has been used very successfully in severe malnutrition.

HOWEVER, with severe watery diarrhoea like cholera, where there is no equivocation about the amount of fluid that is being lost, correction of the dehydration definitely takes precedence over the other considerations. AND I WOULD USE WHO-ORS in this specific situation (cholera-like-diarrhoea).

Now, the situation might well be different in the marasmic and the kwashiorkor child, and this is where I have no real information. Scouring the literature there are no good reports of cholera-like diarrhoea in kwashiorkor and its response to management. In kwashiorkor, unlike marasmus, there is an avid retention of sodium, reduction of urinary sodium to very low concentrations and a reduction in the ratio of sodium: potassium in the stool. There have been no measurements made with these children at high purging rates to see if the change in ratio of sodium: potassium is maintained at these high rates. We also do not know what the basal activity of the pathways in the intestine, of these particular patients, that are affected by cholera toxin. There is a specific change in the charge on the complex carbohydrates of the membranes in kwashiorkor so that there might even be changes in the intestinal binding sites for the organism and its toxin. Indeed, I can find no references to kwashiorkor patients getting cholera, although I have no reason to doubt that they can get the illness. As I have NOT seen cholera in children that present with kwashiorkor, I wonder if those that are running TFCs during a cholera outbreak have this experience also? My impression is that the attack rate in marasmus (and in the attendants with the malnourished children) is much higher than in kwashiorkor - perhaps there are existing data from the Goma TFCs that could address this point.

If this is the case then this would reinforce the recomendation to always use resomal in the children with oedematous malnutrition unless there is cholera-like diarrhoea, and we could simplify the "rule" to always using resomal in kwashiorkor.

I would like to see George Fuchs comments on this - he has much more experience with cholera than I have, and may have some information on the stool electolytes on oedematous children with very high purging rates.

Best wishes,


Prof. Michael H.N.Golden

Dept of Medicine and Therapeutics, Univ of Aberdeen, Foresterhill, AB9 2ZD. Scotland, (UK)

INTERNET, Tel +44 (1224) 681 818 ext 52793/53014, Tel(direct) +44 (1224) 663 123 527 93, Fax +44 (1224) 699 884

From André Briend,

Tue Apr 15 14:38 BST 1997

Subject: cholera in severe pem


I take your points, Mike, on the danger of adding therapeutic CMV to WHO ORS. This was an ill thought suggestion... but this forum should be used too to bounce silly ideas, otherwise it has no use...

I am still worried though because :

a) WHO ORS has been designed from studies with ADULT cholera cases. K is on the low side. The possible malnutrition component was not taken into account.

b) Most NGO's in the field use in case of IV Ringer lacate which is low in K. I am afraid K deficiency may be not corrected after initial IV.

I had discussions with MSF/Epicentre (Yvan Huttin) last year on K content of IV solutions used in relief cholera. I advised to seriously consider moving to the Dhaka solution (13 mEq K/L, if I am corerct) which would make the addition of extra K less needed. A paper was written in MSF newsletter, but I did not hear of a move towards a more appropriate solution.

Maybe this is an opportunity to raise this issue again in the NGO forum. Many NGO's do not realise that Ringer was not designed to treat cholera especially in the malnourished.



Dr. André Briend

INSERM U 290, Hopital Saint Lazare

107 rue du Faubourg Saint Denis, 75 475 Paris Cedex 10, France

tel 33-1-45 23 24 07, tel (direct) 33-1-48 00 56 04, fax 33-1-47 70 28 35

Subject: Cholera and kwashiorkor

Date: Tue, 15 Apr 1997 23:41:46 +0100 (BST)

From: "Michael H.N. Golden" <>


I have heard from Yvonne Grellety, who was in touch with the TFC in Maramvya, Burundi today, where the cholera cases have occured. This TFC has between 4 and 500 severely malnourished patients in it - the vast majority have oedematous malnutrition (adults as well as children). None of the cholera cases so far have had oedematous malnutrition - they have been either marasmic children or accompanying persons.

If this preliminary observation holds up when we analyse the attack rate by type of malnutrition, it will reinforce the guideline to use only resomal in oedematous patients with diarrhoea and will also have implications for the way in which the cholera bug causes disease.

Best wishes,


Prof. Michael H.N.Golden

Dept of Medicine and Therapeutics, Univ of Aberdeen, Foresterhill, AB9 2ZD. Scotland, (UK)

INTERNET, Tel +44 (1224) 681 818 ext 52793/53014, Tel(direct) +44 (1224) 663 123 527 93, Fax +44 (1224) 699 884

Subject: cholera and iv fluids in severely malnourished

Date: Wed, 16 Apr 1997 00:42:18 +0100 (BST)

From: "Michael H.N. Golden" <>


André Briend is correct that there is a dearth of work on treatment of diarrhoea in the severely malnourished child - and almost none with the oedematous child with diarrhoea.

The early studies by Champ Alleyne and John Garrow in Jamaica and Hansen in SA showed clearly the major deficit in potassium. However, in all their published papers it takes about 14 days for the deficit to be corrected even when quite high supplements of potassium are given.

Incidentally, in kwashiorkor it also takes about this time before both the red-cell GSH and the NADP/NADPH ratio become normal, and a study by Ann Bernabeau last year on longitudinal growth showed that the kwashiorkor kids, unlike the marasmic children took 2 weeks to start to grow in length. So it seems that this is about the time that it takes to reverse the metabolic abnormality in kwashiorkor. Thus, the amount of potassium that there is in the rehydration fluid over a few hours will have little effect upon replacing the intracellular deficit.

The effect of magnesium is critical. It has been repeatedly shown that there will be no retention of potassium without replenishment of magnesium. This is well known by the those treating Chrone's disease and ulcerative cholitis where magnesium deficiency is usuall the most pressing deficiency. Most ORS solutions do not contain magnesium (except resomal). Zinc is in the same category.

In terms of the use of Ringer lactate, I also agree that it is low in potassium.

There is another problem with the anions. During the 80's when we were developing the mineral mixes in Jamaica, we used magnesium acetate at one stage. However, we found to our surprise that the malnourished children had difficulty in metabolising the acetate, this was supported by studies reported in Am J Physiol of studies in dogs given large doses of acetate.

Acetate require an ATP to be converted into acetoacetate before it can be metabolised. Lactate is also a problem. It should not be used in areas or patients where there is any suspicion of thiamine deficiency (thiamine is needed to metabolise lactate). Recently, the Kelifi Wellcome group report children with malaria and encephalopathy where the blood

lactate was very high (English M. et al, Acidosis in severe childhood malaria, Quarterly J Med 90:263-270, 1997). It would be very foolish to give someone who is unable to metabolise endogenous lactate a rehydration fluid containing lactate.

We also had trouble, clinically, in a small series of severely malnourished children when we gave them relatively high doses of phosphate (because of this the experiment we were doing was abandoned). I have no doubt that citrate is the best anion to use, it is used in the oral rehydration solutions, it is easily metabolised straight into the krebs cycle without requiring any prior metabolic conversion, it readily crosses cell membranes, has been added to blood that is used for transfulsion (as an anticoagulant) for many years, is very effective in the management of acidosis and is not toxic even in very large doses (the older clinicians used to manage renal tubular acidosis with Schole's mix which is a concentrated solution of sodium and potassium citrate). And yet there are no IV solutions in common use with citrate as the anion.

Although we do everything in our power to avoid IV lines in the severely malnourished patient it is sometimes unavoidable. I think that there should be a re-evaluation of the composition of the solutions that we recommend in this situation. With a higher potassium content, possibly some magnesium and zinc added and citrate as the anti-acidosis anion.

I am sure that Benny Torun and George Fuchs will want to comment.

Best wishes,

Prof. Michael H.N.Golden

Date: Wed, 16 Apr 1997 10:43:14 +0600

From: George Fuchs <>

Subject: Re: Cholera and PEM


The blizzard of information on the is impressive.

I and my colleagues here would agree that the Dhaka solution with the 13 mEq KCl/L is preferable to Ringer's in the malnourished patient, adult or child, with malnutrition and cholera. We go a step further in our actual practice and add KCl to the Dhaka solution to bring it up to 20 mEq/L.

We agree that the addition of Mg is also importance although it is difficult to recommend it be routinely added to IVF, particularly in a less-controlled environment due to the potential adverse cardiovascular consequences in the severely malnourished patient who already might be hypotensive. In such a circumstance, I wonder if simply providing oral MgOH (not too much, or it will cause diarrhea) in the form of inexpensive and readily available antacid and apart from ORS might be useful. As to zinc, we believe a certain amount (1-2 x RDA?) is very important. I am aware of a recent, yet to be published trial that was discontinued because of excessive mortality in the "high" dose zinc arm. If this is confirmed in the full analyses, it would indicate that more is not better at least as it pertains to zinc. This would actually not be terribly surprising for a few good reasons.

I was very interested to learn from Michael that you discontinued a trial of phosphate. Can you provide more details? We are very interested in phosphorous-PEM issues and have considered some work in this area.

In the malnourished patient with cholera, we cannot advocate Resomal if for no other reason than the lower sodium concentration, although it might be acceptable in the maintenance phase of hydration. I agree that there is little published experience to provide guidance on the cholera and kwashiorkor vs marasmus question. As far as I am aware, ICDDR,B does not have data on the relative attack rate of cholera in K vs M, and we see so few cases of K that we cannot provide even anecdotal insights. Even so, intuitively I would be reluctant to use Resomal instead of standard WHO ORS in children with K who have cholera.

All the best,



George Fuchs, MD

ICDDR,B; GPO 128, Mohakhali, Dhaka 1000, Bangladesh

Tel: 880 (2) 988-2399; Fax: 880 (2) 883-116, email:

Fri, 18 Apr 1997 17:04:54 +0200

From: (Andre' BRIEND)

Subject: Cholera, IV and malnutrition


I have here with me the MSF position paper (in French but there should be an English version somewhere) on sudden death in cholera patients, discussing the possible role of K deficiency.

In my view, the evidence that Ringer solutions are not really adapted to rehydrate cholera patients especially in case of associated malnutrition is compelling. That ICDDR,B moves to higher K concentration really means something (George, could you kindly elaborate on the rationale for giving even more K now than in the 'classical' Dhaka solution' ? Thanks). Yet, I think that in relief, where clinical skills are scarce, possibilities of supervision limited, to have a well balance fine tuned IV solution can make a difference. Nobody may be there to give additional K (or Mg, or Zn or whatever) when needed, as it is the case in Dhaka hospital, full of cholera experts ready to correct any problem at the beginnning.

MSF position is 'wait and see' on two grounds:

1- WHO still recommends Ringer lactate for cholera treatment

2- Price implications of switching to Dhaka solution is unknwon

Let's comment these points:

1- WHO

To me argument 1 is not relevant. Experience with severe PEM (300 000 severely malnourished patients treated with modern treatment by NGO's even before WHO had time to publish its manual...) show that initiatives and improvement are to be expected from people in the field, especially by NGO's, not from UN agencies. If a move is scientifically sound (and we may sort out this issue from this forum) should we wait for a WHO rubber stamp before moving ahead ?

2- Price argument

Price of ingredients of Dhaka solution compared to lactate must be pretty much the same. If there is any price difference, this may be due to a low demand. This can be lowered if all NGO working on relief agree to a common IV cholera solution to be used in malnourished population and are able to explain to donors that a different solution may save lives in a malnourished population. IV solution presumably represent only a small part of the price of a cholera treatment centre in relief. Is price really a problem ? I am not sure. I would like this examined in this forum too.

I suggest we seize this opportunity to discuss too the problem of Mg supplementation and lactate metabolism raised by Mike and George.

Remark for about Mike's remark.

It does take several weeks to restore K stores in a malnourished patient because you have to restore muscle for that. K deficiency in addition to deficit in muscle mass takes much less time to be corrected. This type of severe K deficiency may be life threatening. We should not wait to correct it.


Dr. André Briend

Date: Tue, 22 Apr 1997 07:13:37 +0600

From: George Fuchs <>

Subject: Cholera, IVF, and PEM


In Response To Andre's Request To Elaborate On The K Concentration Of IVF's Used At ICDDR,B In Malnourished Children:

Our situation and practice is as follows. We have many severely malnourished children (45% to <60% WA) with severe dehydration and otherwise uncomplicated diarrhea who are treated as outpatients, i.e. they are initially rehydrated with Dhaka solution (without additional K) over approximately two hours and switched to standard WHO ORS by mouth.

Severely dehydrated and severely malnourished children of WA of 45% to <60% with any apparent complication or all children of WA <45% are admitted as inpatients. They are treated under our PEM protocol which, in addition to other interventions (e.g., antibiotics, Mg, Zn, etc.), calls for initial rehydration with Dhaka solution (with additional K to bring it up to 20 mEq/L) over approximately two hours and subsequently switched to standard WHO ORS by mouth or nasogastric tube as needed.

A few additional comments are probably worthwhile. It would probably be ideal to admit all children with severe malnutrition (WA<60%), but we see so many severely malnourished children and we simply do not have the space and resources for so many prolonged admissions. We see comparatively few of the second group of PEM children (WA of 45% to <60% with any apparent complication or all children of WA <45%) with severe dehydration.

Those children with moderate or less degrees of dehydration are rehydrated with WHO ORS.

We hope to begin a trial very soon to compare standard WHO ORS with a few other solutions, including one that is lower in Na and higher in K.

We infuse approximately 30 ml/kg body wt for the initial rehydration which, even with a solution of 2 mEq/100 ml, delivers only a small part of a child's daily K requirement (approx. 14-42 mEq/d dietary for a normal child, and at least at the higher end of this range for a child with PEM). We therefore feel it is safe and reasonable, in the absence of renal failure, to provide a Dhaka solution with additional KCl to the severely malnourished children. In fact, some people recommend giving a solution of up to 40 mEq/L KCl which seems reasonable to me, particularly as a maintenance solution.

>From a practical standpoint regarding disaster situations, I agree that it is probably not feasible to have a solution that requires additional K be manually added to a stock solution.

Regarding cost, we purchase the Dhaka solution for 32 Taka (43 Taka = U.S. $1) from the Bangladesh Institute of Public Health virtually at cost. Although we receive Ringer's as a donation and so I don't have a price comparison, it is hard to imagine the cost is substantially less. If true, I agree that the Dhaka solution should be recommended instead of Ringer's in crisis situations, which experience cholera. In this regard, it might be worth emphasizing to relief personnel that the Dhaka solution was developed in studies on adults with cholera and that it, like Ringer's, should not routinely be given particularly as a maintenance solution to infants and children with non-cholera diarrhea, irrespective of nutritional status. In all cases, the message should also be consistently given that patients be switched to ORS as early as possible, within a few hours, even in severely dehydrated patients.