Very hot dry environments
Very hot dry environments Mike Golden 01.07.98
Re: very hot dry environments Ted Greiner 03.07.98
water needs in malnutrition - very hot dry environments Mike Golden 03.07.98

Date: Wed, 01 Jul 1998 16:51:30 +0100

From: Michael Golden <>

Subject: very hot dry environments



I have recently been in Northern Mali and Tchad where the temperature was

between 44 and 47C and a very low relative humidity.

(this is the hottest time of year). All the severely malnourished children

had good skin - I think that bad skin in that environment would be lethal

because of the rate of water loss. I was worried because I saw a number of

children with clinical hypernatraemic dehydration. [skin with the

consistency of dough, loss of consciousness, convulsions]. Clearly, at

sustained temperatures above body heat the only way heat can be dissipated

is by evaporation (I drank 8-10 litres a day and urinated 3 times in 5 days

in Mali); they will gain heat by radiation and conduction. Children with

a low weight/body surface area ratio must be very susceptible to pure water

deficiency, which if treated with electrolyte solutions will lead to

hypernatraemia - perhaps if sufficient is given to maintain a good urine

flow rate and their renal concentrating ability is not as compromised as it

is elsewhere (like the skin) this problem could be avoided - but this may

take very large volumes which is not recommended.

I do not know of literature of water balance/requirements in young children

exposed to sustained temperatures over body heat and a relative humidity

<15%. The literature on premature infants does not include these

conditions. However, rough calculations from Ultman JS (Computational Model

for insensible water loss from the newborn - Pediatrics 79:760-765,1987)

indicate that the water loss can be very high. Have conventional ORS

solutions been tested in these extreme temperatures?

It is not easy to measure a child's temperature under these conditions as

the thermometers are above body temperature. Are cheap robust user-friendly

thermocouple thermometers available from any supplier?

All metal items potentially burn the children such as metal tapes on

measuring boards etc because of the high specific heat of metal.

The signs of shock change so that "cool peripheries" become "warm

peripheries" and if blood flow to the skin is reduced then hyperthermia can

result. Indeed, failure to have dilated skin vessels (scalp) was a very

poor prognostic sign in these conditions.

The children should be cooled with wet cloths.

It is likely that the management of severe malnutrition under these

environmental circumstances should not be the same as in "normal" tropical

temperatures and the diets and other provisions of most manuals varied to

take into account the problems of water and heat exchange. All the ACF

centres now have maximum-minimum thermometers in the phases to monitor the

environmental temperature.

Have those who have worked in the Sahel and other hot desert areas seen

similar problems. Parenthetically, in the winter in the desert the

temperature can fall at night to very low levels and we have seen

hypothermia under these conditions.

I wonder what the sweat sodium is in the young children and newborn infants

in these areas and whether they have a greater renal concentrating ability

than in other areas - perhaps loss of renal concentrating ability would be

a lethal change similar to my observations with the skin.


Prof. Michael H.N.Golden

Date: Fri, 03 Jul 1998 13:13:40 +0100

From: Ted Greiner <>

Subject: Re: Ngonut: very hot dry environments


Dear Dr. Golden,

You may find the following two references of interest. In the first one,

Almroth1 reviews the relevant literature on renal concentrating capacity in

infants (some of which would be unethical to do today) and compares this

with the osmolarity of the urine of exclusively breast-fed infants under

Jamaica's warm but somewhat humid conditions. In the second one,2 she and

Bidinger examine the outcome under conditions nearly as extreme as those

you mention. (Not much urine was produced, but it was dilute. Recall that,

unlike everyone else, exclusively breast-fed babies have no excess protein

or minerals to excrete--the main reason we need to drink water or risk

electrolyte disequilibrium and death--despite our bodies still being made

up large of water at death. Your question of the sodium content of sweat is

interesting, but urinary osmolality is the "bottom line" and therefore

presumably sufficient. Nonetheless, there may be times during illness when

these infants do need additional water.) A study in Pakistan arrived at

similar findings under almost equally intense environmental conditions.

They also tested the infants' renal concentrating capacity and found it to

be adequate3. (This may be another example in which the fact that some few

individuals are lacking in this capacity has been generalized so that

physicians often assume that virtually all young infants have low renal

concentrating capacity. Again, prematures are of course another issue.)

Almroth and Bidinger also document that temperature and humidity vary

enormously in 24-hour cycles in such a setting, with evening and early

morning values being much easier to tolerate. Thus one strategy is to

remain indoors during the worst period, particularly if homes have thick

walls of mud or stone, as the cooling of the night passes through to the

inside during the day and vice versa. Did you notice many infants being

exposed to outdoor conditions when they were worst at mid-day?

On the issue of ORS, I think it is particularly relevant to keep in mind,

as Almroth and Latham pointed out4, that all relevant studies have found

that many mothers make ORS or SSS too strong. Thus it is irrational to

promote these solutions for PREVENTING dehydration in normal circumstances.

Better are any foods or fluids already available in the household, since

these are unlikely to be excessively high in sodium or cause osmotic

diarrhea through being too high in sugar. Indeed they maintain hydration

through the actions of amino acids, peptides and carbohydrates which

simulate intestinal fluid absorption just like ORS does. Indeed, in some

places it would appear that promotion of ORS as the "water of life" leads

to a version of it called "water with sugar and salt" being used regularly

even when the infant does not have diarrhea. Anything that threatens

exclusive breastfeeding may have a triple negative effect, replacing breast

milk, increasing the renal solute load, and introducing bacterial and other

contaminants to the infant. Moreover, well-meaning medical intervention may

have harmed traditional practices supportive of giving extra food to

infants with diarrhea5.

In case you want to correspond directly with Dr. Almroth on any of this

research, her email address is included as a cc above. She now works with

Population Council's regional office in Delhi.

Best regards,


Ted Greiner



1. Almroth SG. Water requirements of breast-fed infants in a hot climate.

American Journal of Clinical Nutrition 1978;31:1154-1157.

2. Almroth A, Bidinger PD. No need for water supplementation for

exclusively breast-fed infants under hot and arid conditions. Transactions

of the Royal Society of Tropical Medicine and Hygiene 1990;84:602-4.

3. Ashraf RN, Jalil F, Aperia A, Lindblad BS. Additional water is not

needed for healthy breast-fed babies in a hot climate. Acta Paediatrica

Scandinavica 1993;82(12):1007-11.

4. Almroth S, Latham MC. Rational home management of diarrhoea. Lancet


5. Almroth S, Mohale M, Latham MC. Grandma ahead of her time: traditional

ways of diarrhoea management in Lesotho. Journal of Diarrhoeal Disease

Research 1997;15(3):167-172.


Ted Greiner, PhD, Nutritionist

Section for International Child Health, Department of Women's and Children's Health, Entrance 11

Uppsala University, 75185 Uppsala, Sweden

Date: Fri, 03 Jul 1998 13:17:39 +0100

From: Michael Golden <>

Subject: Ngonut: water needs in malnutrition - very hot dry environments


Dear Ted,

Thank you very much for your thoughtful message. Indeed, I was in Jamaica

when one of the studies was done in Port Antonio, and remember it well. I

fully agree totally that in the all tropical countries where studies have

been done NOTHING is required, apart from breast milk, for normal infants.


The environmental conditions in the Sahara are very extreme. Indeed, the

difference between 40 and 45 degrees is very substantial as water loss

rises exponentially this temperature. And the very low relative humidity

is important.

The conditions in the studies that you site are not as extreme as these:

(Almroth,'78 = 24-28C, humidity 62-90%: Almroth '90 = 35-40C, humidity

10-35%: Ashraf '93 = 27-40, humidity 24-77).


However, I was not addressing the normal breast fed infant - but rather

severely malnourished children, most of whom are over 6 months of age but

less than 70% weight for height and severely stunted (so that their surface

area/weight is very large). Here the formulae that are used elsewhere may

indeed be inappropriate because of the high renal solute load and attempts

to make the diets "energy-dense". Even in the Jamaican climate we had to

stop giving any child with fever or a rapid respiratory rate and all small

children the standard diet that was advocated at that time (135 kcal/100ml)

because of the high renal solute load/low water intake. More dilute

formulae are now used - but in conditions such as the Sahel they may still

be to concentrated.


There have been studies of renal concentrating ability in severe

malnutrition, and the maximum is about 600 mOsm/l in most children in

Jamaica and sometimes much less - but as I said in my message the children

adapted to the Sahalian environment may have retained renal function in a

way that was not seen in Jamaica. Alternatively, those who become

malnourished that do not retain this ability may not be available for study

as they will have passed on - and we could make incorrect assumptions when

we only look at a survivor cohort.


With the current formulae I have calculated the maximum insensible water

loss tolerable for various volumes of intake and maximum renal

concentrating ability:


For example the tables give the insensible water loss in ml/kg/d:


Intake 100 kcal/kg/d:

Urine osmolality = 600 800 1000

F100 45 57 66

F100-diluted* 77 91 98

F75 104 112 117

F135** 30 41 48


Intake 150 kcal/kg/d:

Urine osmolality = 600 800 1000

F100 66 86 98

F100-diluted* 116 136 >140

F75*** >140 >140 >140

F135** 45 61 71


*diluted to give 75 kcal/100 ml

** no longer recommended

*** F75 is only given at 100 kcal/kg/d included here for comparison only.


In health full term infants not exposed to extreme environments the

insensible water loss is between 30 and 70 ml/kg/d.


>From the above calculations it is clear that F100 has sufficient water for

most children with a surface area/weight similar or less disadvantageous

than a healthy full-term (say 3.5kg) - water may become the limiting

nutrient with this diet in smaller children who have fever, tachopnoea,

mild diarrhoea or under extreme environmental conditions.


For small children (say less than 4 kg) the dilution of F100 to 75kcal/100

ml should ensure sufficient water when the child gets a fever, tachopnoea

or mild increases of stool water when there is excessive evaporative loss.


With F75 there should not be any difficulty except under the most extreme

of conditions in small infants.


These figures show that F135 is dangerous and its use should only be

contemplated for large children who do not have a fever, respiratory

infection or diarrhoea, take more than 100 kcal/kg/d, are not in a very hot

or dry environment and can be closely monitored so that the diet can be

changed if they develop intercurrent infection.


Clearly though the conditions that prevail in the Sahara are quite unusual

- we need to know at what weight children should be given the more dilute

feed as a routine when the temperature is sustained above 40C with very low



Prof. Michael H.N.Golden