Impact of Zinc Supplementation in Malnourished Children
with Acute Watery Diarrhoea
by P. Dutta,a U. Mitra,a A. Datta,a S. K. Niyogi,a S. Dutta,a B. Manna,a M. Basak,b T. S. Mahapatra,b and
S. K. Bhattacharyaa
a
Divisions of Clinical Medicine, Microbiology and Biochemistry National Institute of Cholera and Enteric Diseases,
Calcutta, India
b
Department of Pediatric Medicine, Dr. B. C. Roy Memorial Hospital for Children, Calcutta, India
Summary
A double-blind, randomized, controlled clinical trial was conducted on 80 malnourished children
with acute dehydrating diarrhoea to evaluate the ef®cacy of oral supplementation of zinc as an
adjunct therapy to oral rehydration solution (ORS). After decoding it was observed that 44 children
received zinc sulphate (177 mg/kg/day in three divided doses equivalent to 40 mg elemental zinc) in a
syrup form and 36 children received only syrup placebo. Clinical parameters and microbiological
®ndings of stool samples were comparable in the two groups at the time of enrolment. All the
children (100 per cent) in the zinc supplemented group and 32 (89 per cent) children in the
placebo group recovered within 5 days of hospitalization (p = 0.04). The zinc supplemented group
had a signi®cantly shorter duration of diarrhoea (70.4 6 10.0 vs. 103.4 6 17.1 h; p = 0.0001), passed
less liquid stool (1.5 6 0.7 vs. 2.4 6 0.7 kg; p = 0.0001), consumed less oral rehydration solution
(2.5 6 1.0 vs. 3.6 6 0.8 litre; p = 0.0001) and other liquids (867.0 6 466.1 vs. 1354.7 6 675.6 ml;
p = 0.0001) as compared to the placebo group. Our ®ndings suggest that zinc supplementation as
an adjunct therapy to ORS has bene®cial effects on the clinical course of dehydrating acute
diarrhoea.
Introduction
For the last three decades, all age groups including
infants and young children with dehydrating acute
diarrhoea of any aetiology, have been treated with standard
oral rehydration salts (ORS) solution in accordance with
the recommendations of the World Health Organization
(WHO) and the United Nation Children's Fund
(UNICEF).1 Because of its proven ef®cacy, safety and
low cost, standard ORS has been widely used, and along
with continued feeding has helped to bring about a decline
in the case fatality rate of acute dehydrating diarrhoea in
children of developing countries.1 However, the major
drawback of standard ORS is that it does not reduce the
Acknowledgements
The authors acknowledge Dr Dilip Mahalanabis, Director,
Society of Applied Studies, Calcutta for helping us to develop
the protocol of the study. We thank Mr R. K. Sinha, Managing
Director, Greenco Biologicals (Pvt) Ltd., Calcutta for preparing
the zinc syrup and placebo used in the study. We also thank Mr
Shyamal Kumar Das for typing the manuscript.
Correspondence: Dr. P. Dutta, Deputy Director & Head,
Division of Clinical Medicine, National Institute of Cholera
and Enteric Diseases, P-33, C.I.T. Road Scheme XM,
Beliaghata, Calcutta ± 700 010, India. Tel. 91-33-350-5533/
Fax. 91-33-350-5066. E-mail <[email protected]>.
Journal of Tropical Pediatrics
Vol. 46
October 2000
stool volume, diarrhoeal duration and frequency.2 This
raises a practical problem of its acceptance, since the
major concern of the mother is the duration, frequency
and volume of diarrhoea. Hence, there is a persistent
need to overcome these problems.
Malnutrition is a predominant host factor associated
with diarrhoea in developing countries.3 Zinc de®ciency
is usually associated with malnutrition and diarrhoea in
malnourished children leads to further zinc depletion.4,5
Recent clinical studies have shown that supplementation
of zinc has a signi®cant role in reducing the duration and
frequency of acute dehydrating diarrhoea among
severely malnourished children who have pre-existing
zinc de®ciency.6±8 However, the studies have failed to
document any reduction in stool volume, which is also a
major concern of the mother, and of treatment providers.
Furthermore, the bene®cial effect of zinc supplementation in mild to moderately malnourished children, who
constitute about 40 per cent of our child population
(compared to 5.7 per cent severely malnourished) in
India, is not known.
We therefore conducted a double-blind, randomized,
controlled clinical trial of the supplementation of zinc as
an adjunct therapy to ORS, to evaluate the impact of zinc
supplementation on diarrhoeal duration and volume and
intake of ORS in different grades of malnourished
children suffering from acute dehydrating diarrhoea.
q Oxford University Press 2000
259
P. DUTTA ET AL.
Patients and Methods
The children included in the study were admitted to the
Dr B. C. Roy Memorial Hospital for Children, Calcutta,
India between June 1997 and May 1998. Male children
were chosen (for ease of collection of stools and urine
separately) aged between 3 and 24 months and having
less than 80 per cent Harvard Standard weight for age.
These children had a history of watery diarrhoea (more
than four times within last 24 h) for 72 h or less and had
clinical signs and symptoms (e.g. sunken eyes, reduced
skin elasticity, rapid pulse, dry mouth and thirst) of
`some' dehydration.1
Criteria of exclusion were: (a) history of treatment
with antibiotics; (b) other systemic infections, e.g.
septicaemia, meningitis, pneumonia, urinary tract infection, otitis media; (c) chronic underlying diseases, e.g.
tuberculosis, liver diseases; (d) need for intensive care,
i.e., life-support system, blood transfusion or total
parenteral nutrition; and (e) exclusively breastfed.
Children who ful®lled the inclusion and exclusion
criteria were randomized into two treatment groups
according to a random number table. Patients were
allocated a speci®c-numbered bottle of zinc syrup
(177 mg/day in three divided doses, 40 mg elemental
zinc/day) or only syrup. The person who made the
randomization was not associated with the study. The
serial code numbers were kept in a sealed envelope with
one of the senior of®cers who identi®ed the groups after
completion of the study. The bottles of zinc syrup and
placebo were identical in size, shape and colour. Taste,
consistency and colour of zinc syrup and placebo were
also identical. These were prepared by Greenco
Biologicals Private Limited, Calcutta, India according
to the speci®cation that 5 ml zinc syrup should contain
59 mg of zinc sulphate.
This project was approved by the Scienti®c Advisory
Committee and Ethical Committee of our Institute.
Informed written consent was obtained from the parents
of the children explaining the details of the study
procedure. After selection, a complete history was taken
from the parents and a thorough physical examination
was done and ®ndings were recorded in a pre-designed
proforma. Stool samples were collected in sterile
MacCartney's bottles on admission for detection of
various established enteropathogens using standard
methods.9 Children were weighed unclothed on admission using weighing scales with a sensitivity of 20 g;
length and mid-arm circumference were also recorded.
Nutritional status was assessed as weight-for-age and
classi®ed into different grades according to the classi®cation of Indian Academy of Pediatrics (IAP) which is
the only recommended classi®cation for Indian children.10 The degree of dehydration was assessed by WHO
criteria.1 All the children received standard oral
rehydration salts (ORS) solution (mmol/l: sodium, 90;
potassium, 20; citrate, 10; chloride, 80; glucose, 111) at
the rate of 75±100 ml/kg body weight for the ®rst 4±6 h
of admission for correction of initial dehydration.1 If
260
initial dehydration was not corrected, the same solution
was repeated for another 4±6 h. When all the signs and
symptoms of dehydration disappeared ORS solution
was given as maintenance therapy in amounts matching
stool volume and loss in vomitus. However, more ¯uid
was given if the child wanted it and if there were
clinical indications. If any patient developed severe
dehydration during the follow-up period, he received
intravenous infusion of Ringer's lactate according to
WHO guidelines.1
Immediately after rehydration, feeding was resumed
in both groups. Breastfeeding was allowed as wanted in
breastfed children. Non-breastfed children received halfstrength milk for the ®rst 24 h, and the strength was
gradually increased until discharge. Older children were
offered the standard hospital diet of rice, lentil, ®sh
(cereal/vegetable diet) appropriate for their age.1
All intake and output were measured and recorded
every 8 h until diarrhoea stopped, withdrawal from the
study, or up to day 5 if the child did not ful®l the criteria
of recovery. Recovery was de®ned as the passage of
normal stool or no stool for the last 18 h. Stool losses
were measured on pre-weighed disposable diapers, urine
was separated from stool by use of urine collection bags.
Vomitus was weighed on pre-weighed gauze pads. Body
weight was recorded after rehydration and every 24 h
(between 10.00 and 10.30 am), thereafter until discharge
or until day 5. The following daily records were
maintained: (i) number of stools; (ii) number of
vomits; (iii) stool output (g); (iv) body weight (g);
(v) intake of ORS; (vi) intake of other ¯uids, such as
liquid diet, milk and water. At the time of discharge, all
the children were advised to continue the assigned bottle
of syrup until it was ®nished. Mothers were advised to
give at least one extra meal or liquid food per day during
the recovery period and to attend hospital for follow-up
on days 15 and 30 after hospitalization. The following
nutritional assessment was recorded on days 15 and 30 of
the follow-up period: (i) weight, (ii) height, (iii) mid-arm
circumference.
Before analysing the trial results, a check list of
treatment assignment was collected for decoding of
experimental category of sample children. Data was
entered and analysed using dBase IV and SPSS.
Comparability of the study and control groups, according
to the patient characteristics, was determined by using
the chi-squared test. The means of outcome variables of
the two groups were compared by applying Student's
t-test. The differences in proportion of cured patients in
the two groups was also examined by the chi-squared
test.
Results
A total of 80 male children aged between 3 and 24
months suffering from dehydrating acute watery diarrhoea
were included in the study. After decoding it was observed
that 44 children were in the zinc supplementation group
and 36 were in the placebo group.
Journal of Tropical Pediatrics
Vol. 46
October 2000
P. DUTTA ET AL.
TABLE 1
Initial characteristics of the treatment groups
Zinc supplemented
group (n = 44)
Features
Placebo group (n = 36)
Mean age 6 SD (months)
Mean body weight 6 SD (kg)
Mean height 6 SD (cm)
Mean MAC 6 SD (cm)
10.4 6 5.4
5.5 6 1.6
65.5 6 8.4
10.3 6 1.3
11.0 6 4.9
5.8 6 1.5
67.5 6 6.9
10.5 6 1.0
Nutritional status (no. weight-for-age % median)
$ 80
70 < 80
60 < 70
< 60
Mean duration of diarrhoea before admission 6 SD (hours)
Frequency (per 24 h) of diarrhoea before admission
Degree of dehydration
±
6(13)
10(23)
28(64)
33.4 6 11.5
13.8 6 3.8
Some
±
6(17)
11(30)
19(53)
38.3 6 10.3
13.3 6 3.9
Some
Enteropathogens (no. %)
Single pathogen
Enteropathogenic E. coli
Enteroaggregative E. coli
Salmonella typhimurium
Shigella ¯exneri
Shigella sonnei
V. cholerae O1
Clostridium dif®cile
Rotavirus
V. cholerae non-O1 non-O139
15(34)
1(2.3)
3(6.8)
2(4.6)
1(2.3)
1(2.3)
1(2.3)
8(18.0)
2(4.6)
9(25)
1(2.8)
3(8.3)
0
1(2.8)
0
1(2.8)
5(13.9)
3(8.3)
Mixed pathogens
EPEC + S. typhimurium
EPEC + Rotavirus
EPEC + S. ¯exneri
Rotavirus + S. ¯exneri
Rotavirus + S. typhimurium
No pathogen
1(2.3)
3(6.8)
1(2.3)
1(2.3)
1(2.3)
3(6.8)
1(2.8)
4(11.1)
1(2.8)
1(2.8)
2(5.5)
4(11.1)
Both the groups were comparable with regards to
average values of the following variables: age, preadmission duration of diarrhoea, status of clinical
dehydration, nutritional status, height, mid-arm circumference, and microbiological ®ndings of stool sampls
(Table 1).
Forty-four (100 per cent) children in the zinc
supplemented group and 33 (89 per cent) children in
the placebo group recovered within 5 days of hospitalization. The differences in cure rate in the two treatment
groups was signi®cant (p = 0.04). The zinc supplemented
group had signi®cantly shorter duration of diarrhoea
(70.4 6 10.0 vs. 103.4 6 17.1 h; p = 0.0001), passed less
liquid stool (1.5 6 0.7 vs. 2.4 6 0.7 kg; p = 0.0001),
consumed less oral rehydration solution (2.5 6 1.0 vs.
3.6 6 0.8 litre; p = 0.0001) and other liquids (867.0 6
466.1 vs. 1354.7 6 675.6 ml; p = 0.0001) as compared to
the placebo group (Table 2).
However, no signi®cant differences were observed
with regard to gain in weight, mid-arm circumference
and height on recovery or on day 30 of hospitalization
among the children who received the zinc or the placebo
for up to 14 days (Table 2).
Journal of Tropical Pediatrics
Vol. 46
October 2000
In subgroup analysis of different nutritional status, it
was observed that the duration of diarrhoea, stool output,
consumption of ORS and other ¯uids were also
signi®cantly less in the zinc supplemented group as
compared to the placebo group.
Discussion
This double-blind controlled, clinical trial shows signi®cant reduction in the duration of diarrhoea, stool
volume and reduction in consumption of total ORS and
other ¯uids (liquid food and plain water) among the zinc
supplemented malnourished children suffering from
acute dehydrating diarrhoea. The bene®cial effect of
zinc supplementation might be due to the role of zinc in
electrolyte transport,11 early epithelial regeneration,12,13
and improved synthesis of digestive enzymes,14,15
causing a reduction in osmotic diarrhoea and an
improvement in immunity,16 which limits the bacterial
overgrowth and early clearance of intestinal pathogens.
Zinc supplementation causing a reduction in the
duration of diarrhoea in malnourished children, has
also been reported by other investigators.6±8 However,
261
P. DUTTA ET AL.
TABLE 2
Outcome variables
No. (%) of patients recovered
Mean recovery 6 SD (h)
Consumption of total ORS (litres)
Total stool output (kg)
Total liquid (liquid food + watery; ml)
Percentage of weight gain on recovery (% admission wt.) 6 SD
Percentage weight gain on 30th day (% recovery wt.) 6 SD
Percentage gain in mid-arm circumference on 30th day
(% on recovery MAC) 6 SD
Percentage gain in height on 30th day (% on recovery height) 6 SD
a
b
Placebo group
(n = 36)
p value
44(100)
70.4 6 10.0
2.5 6 1.0
1.5 6 0.7
867.0 6 466.1
3.9 6 4.1
2.6 6 3.3a
5.2 6 3.4a
32(89)
103.4 6 17.1
3.6 6 0.8
2.4 6 0.7
1354.7 6 675.6
3.2 6 2.9
2.9 6 3.7b
3.4 6 2.3b
0.04
0.0001
0.0001
0.0001
0.0001
0.41
0.88
0.08
1.1 6 0.9a
0.6 6 0.5b
0.06
Follow-up of 18 patients.
Follow-up of 16 patients.
the present ®ndings of less stool output and less
consumption of ORS in addition to reduced duration of
diarrhoea adds to the favourable impact of zinc
supplementation in acute watery diarrhoea. Decreased
stool output and less consumption of ORS are
considered today as the prerequisite criteria for genuine
anti-diarrhoeal impact.
In lieu of 20 mg elemental zinc used in other
studies,6±8 40 mg elemental zinc was used in this study
which was four times the recommended, daily allowance
(RDA) of the United States Food and Nutrition Board for
non-diarrhoeal healthy children. Such a high dose was
used for prompt normalization of zinc status in
malnourished diarrhoeal children who were also losing
a signi®cant amount of body zinc through liquid
stools.17,18
The major drawback of this study is that, zinc status in
serum or other body tissues of the study population was
not measured. Zinc was supplemented on an empirical
basis with a view that the malnourished study population
might have a zinc de®ciency as there is low zinc content
in the traditional Indian diets of children of poor
socioeconomic status. Also, the bioavailability of zinc
from their diet is low due to the high ®bre and phytate
content.19,20 Moreover, the loss of indigenous zinc is
high in acute diarrhoea which contributes further to zinc
depletion.17,18
Recent studies showed that zinc supplementation
offered substantial clinical bene®t in acute diarrhoea in
severely malnourished children with con®rmed zinc
de®ciency.6±8 However, this study was aimed to observe
the same bene®cial effect among the mild to moderately
malnourished diarrhoeal children and the desired effect
has been documented in those children who constituted
40 per cent of our study population.
References
1. World Health Organization. Programme for control
diarrhoeal diseases: a manual for treatment of diarrhoea.
WHO/CDD/SER/80.2 Rev 2. WHO, Geneva, 1990.
262
Zinc supplemented
group (n = 44)
2. Mahalanabis D, Marson M. Development of an improved
formulation of oral rehydration salts (ORS) with antidiarrhoeal and nutritional prospective: a super ORS. In:
Holmgrane J, Lindbarg A, Mollby R (eds), Development
of Vaccines and Drugs against Diarrhoea. Studentlitteratur,
Sweden, 1986; 240±56.
3. Synder JD, Marson MH. Magnitude of global problem of
acute diarrhoeal diseasesÐa review of active surveillance
data. Bull WHO 1982; 60: 605±13.
4. Khanum S, Alam AN, Anowar J, Ali AM, Rahaman MM.
Ef®cacy of zinc supplementation on the dietary intake and
weight gain of Bangladeshi children recovering from
protein energy malnutrition. Eur J Clin Nutr 1988; 42:
709±14.
5. Hambridge KM. Zinc and diarrhoea. Acta Pediatr 1992;
Suppl 381: 82±6.
6. Sachdev HPS, Mittal NK, Mittal SK, et al. A controlled
trial on utility of oral zinc supplementation in acute
dehydrating diarrhoea in infants. J Pediatr Gastroenterol
Nutr 1988; 7: 877±81.
7. Behran RH, Tomkin AM, Roy SK. Zinc supplementation
during diarrhoea, a forti®cation against malnutrition?
Lancet 1990; 336: 422±43.
8. Roy SK, Behran RH, Haider R, et al. Impact of zinc
supplementation on intestinal permeability in Bangladeshi
children with acute and persistent diarrhoea syndrome. J
Pediatr Gastroenterology Nutr 1992; 15: 289±96.
9. World Health Organization. Manual for laboratory investigation of acute enteric infection. Programme for control
of diarrhoeal diseases. CDD/83.3. WHO, Geneva, 1983.
10. Nutritional Subcommittee of Indian Academy of Pediatrics. Report of the Convener. Indian J Pediatr 1972; 9: 360.
11. Marson JR, Lewis JC. The effect of severe zinc de®ciency
on intestinal permeability: an ultra-structural study. Pediatr
Res 1985; 19: 968±73.
12. Hambridge KM, Casey CE, Krebs NF. Zinc. In: Mertz W
(ed.), Trace Elements in Human and Animal Nutrition, 5th
edn. Academic Press, Florida, 1986; 2: 1±137.
13. Bettger WJ, O'Dell BL. A critical physiological role of
zinc in the structure and function of biomembrane. Life Sci
1981; 28: 1425±38.
14. Park JHY, Grandjean CJ, Antonson DL, Vanderhoof JA.
Effects of short tern isolated zinc de®ciency on intestinal
growth and activities of brush border enzymes in weaning
rates. Pediatr Res 1985; 12: 1333±36.
Journal of Tropical Pediatrics
Vol. 46
October 2000
P. DUTTA ET AL.
15. Zarling EJ, Mobarham S, Donahuc PE. Does zinc
de®ciency affect intestinal protein content or disaccharide
activity? J Lab Clin Med 1985; 106: 708±11.
16. Chandra PK (ed.). Trace elements and immune response.
In: Trace Elements in Nutrition in Children±11. Nestle
Nutrition Workshop Series 1991, 23: 201±14.
17. Castillo-Duran C. Zinc and copper wastage during acute
diarrhoea. Nutr Ref 1990; 48: 19±22.
Journal of Tropical Pediatrics
Vol. 46
October 2000
18. Castillo-Duran C, Vail P, Uany R. Trace mineral balance
during acute diarrhoea in infants. J Pediatr 1988; 113: 452±57.
19. Gopalan C, Sastri BVR, Balasubramanian SC. Nutritive
vale of Indian foods. Report from National Institute of
Nutrition, ICMR, Hyderabad, India, 1994.
20. Gopalan C. Combating undernutrition. Basic issues and
practical approaches. National Foundation of India, Special
Publication Series 3, 1989.
263