37

Original Article

Zinc Supplementation and Serum Zinc During Diarrhea

Abdullah H. Baqui, Robert E. Black, Christa L. Fischer Walker, Shams Arifeen, Khalequz Zaman,
Mohammad Yunus, Mohammad A. Wahed and Laura E. Caulfield

Johns Hopkins Bloomberg School of Public Health, Department of International Health, Baltimore, Maryland
(AHB, REB, CFW, LEC) and International Centre for Diarrheal Diseases Research, Bangladesh (ICDDR,B), Dhaka,
Bangladesh (AHB, SA, KZ, MY, MAW)

ABSTRACT
Objective. Zinc deficiency is very common in developing countries and is more pronounced during an episode of diarrhea.
Supplementation with zinc improves diarrhea and might correct zinc deficiency in both the short and longer term.

Method. We conducted a nested study within a cluster randomized treatment trial. Fifty children with diarrhea living in the zinc
treated clusters, 50 children with diarrhea living in control clusters, and 50 healthy children living in the control clusters were
enrolled. We assessed serum zinc at the start of the diarrhea episode, which was 1-3 days after supplementation began in zinc
treated children, and again one week after the diarrhea ended and supplementation ceased. Baseline characteristics and serum
zinc concentration were assessed.

Results. Serum zinc was low in 44% of healthy children at the first blood draw. Compared to healthy controls, serum zinc was
3.1 mmol/L higher among children with diarrhea who were supplemented with zinc at first blood draw and 1.3mmol/L higher
3 weeks later.

Conclusion. Zinc supplementation enhances serum zinc concentration when given as a treatment for diarrhea and helps
children maintain a more adequate zinc status during the convalescent period. [Indian J Pediatr 2006; 73 (6) : 493-497]
E-mail : abaqui@jhsph.edu

Key words : Zinc; Diarrhea; Supplementation

Zinc deficiency is highly prevalent in children in
developing countries and is one of the major risks factors
for poor child health accounting for over 800 000 child
deaths per year, including 10% of all diarrhea deaths.1, 2
Dietary zinc insufficiency is exacerbated during a diarrhea
episode by net zinc loss in the stool. 3 Zinc
supplementation during diarrhea reduces the duration
and severity of the episode and the risk of subsequent
diarrhea and ALRI morbidity.4-6 It is now recommended
that all children under 5 receive 20 mg supplemental zinc
for 10-14 days as part of clinical treatment of diarrhea.7
How zinc exerts these therapeutic and preventive
effects is not fully understood. Zinc is an essential mineral
for human cell growth, differentiation, and DNA
synthesis 8 , and it is known to improve water and
electrolyte absorption,9, accelerate the regeneration of the
gut epithelium, 10 increase the activity of brush border
enzymes11 and enhance overall immune function. 12 A
better understanding of the mechanism of protection may
be possible if more is known about the effect of short-term
supplementation on zinc status, as described by serum
zinc. Studies to date have assessed serum zinc prior to
supplementation during diarrhea and then again on day
14 in both zinc-supplemented and control children
generally showing higher zinc concentration at the end of
supplementation.13-15 In a study nested within a cluster
randomised trial of zinc therapy for diarrhea we
examined the serum zinc concentrations of children with
diarrhea in the 1-3 days after the initial supplement was
given as well as the lasting effect one week after a 14-day
course of zinc supplementation to determine if the higher
serum zinc concentrations continued after
supplementation had been stopped.
MATERIAL AND METHODS

Correspondence and Reprint requests : Dr. Abdullah H. Baqui Johns

We have previously reported the details of the study
Hopkins Bloomberg School of Public Health 615 North Wolfe St. Rm
E8138 Baltimore, MD 21205 Phone : 410-955-3850 Fax: 410-614-1419 design and impact of two weeks of zinc therapy for
(attn: A.Baqui) diarrhea on acute lower respiratory infections (ALRI) and

Indian Journal of Pediatrics, Volume 73—June, 2006 493


Abdullah H. Baqui et al
diarrhea morbidity and mortality in a community-based
cluster randomised trial 4 in the Matlab field area of
ICDDR,B: Centre for Health and Population Research.
Thirty service areas (clusters) around Matlab ICDDR,B
Hospital, each with about 200 children 3-59 months old,
were randomly allocated to intervention or comparison
areas. Each cluster was served by one community health
worker (CHW). From November 1998 to October 2000, all
children 3-59 months old were included.
CHWs and Bari Mothers (BMs - community
volunteers) distributed oral rehydration solution (ORS)
packets and adviced on feeding to parents who sought
treatment for a child with diarrhea in both intervention
and control areas. In the intervention clusters, CHWs and
BMs also distributed zinc syrup and instructed the
mothers on how to give the syrup to her child. The syrup
was a 20 mg dose of zinc acetate prepared in 5 ml syrup
to be given once a day for 14 days regardless of the
duration of diarrhea. Both ORS and zinc were provided
free. In the comparison communities, CHWs and BMs
distributed ORS and provided advice on feeding only.
To assess the differences in serum zinc concentrations
among children with diarrhea supplemented with zinc,
children with diarrhea who were not supplemented with
zinc, and children without infection, blood specimens
were collected in 3 groups of children: (i) children with
acute diarrhea in the zinc treatment group after 1-3 days
of zinc supplementation, and on day 21 (1 week after
30 Clusters
15 Intervention Clusters with 3974 partially
or fully complete follow-up resulting in 5866
observable child years
50 Children
with diarrhoea
13 loss to 8 loss to
follow-up/ follow-up/
bad bad blood
blood work work
37 with 2 successful
blood draws
Fig. 1. Trial Profile
494
38
completing 14 days of treatment); (ii) children with acute
diarrhea in the control area on day 1-3 of diarrhea and 21
days after the first blood draw; (iii) healthy children in the
control areas, defined as those not having diarrhea for at
least 2 weeks prior to the blood draw. The sample size
was calculated to see an 11% difference (80% power and
95% confidence) in mean serum zinc after 14 days of
supplementation between zinc-supplemented children
and non-zinc-supplemented children with diarrhea. 16
Fifty children with acute diarrhea from the villages
receiving zinc supplementation were selected over the
study year. As each child from one of these villages was
enrolled, an age (within 1 month) and gender-matched
child with acute diarrhea from a control village and an
age and gender-matched healthy child also from a control
village were selected. The children from the control areas
were chosen within one week of the selection of the child
in the treatment group to also account for seasonal
influences. The larger cluster-randomised trial and the
selection of subjects for this nested study are described in
Fig. 1.
All selected children were brought to the Matlab
hospital for blood drawing. Approximately 3 ml of blood
was collected via venipuncture using plastic tubes
carefully washed to make them zinc free. The blood
samples were centrifuged immediately, frozen at –20 °C,
and transported on ice to the ICDDRB laboratory in
Dhaka where serum zinc concentrations were measured
15 Control Clusters with 4096 partially or
fully complete follow-up resulting in 6015
observable child years
50 Children with 50 Healthy Controls
diarrhoea with 1 successful
blood draw
42 with 2
successful
blood draws
Indian Journal of Pediatrics, Volume 73—June, 2006
39

Zinc Supplementation and Serum Zinc During Diarrhea

using atomic absorption spectrophotometry (AAS, Perkin TABLE 1. Baseline Characteristics of Children
Elmer model 3100). A certified trace element control
Selected Children With Diarrhoea Healthy
serum was used daily to ensure accuracy and precision Characteristics Controls
(UTAK Lab Inc., California, USA). All lab analyses were Zinc No Zinc (n=50)
blinded to diarrhea and intervention status of the children (n=37) (n=42)
sampled.
Male (%) 51.4 61.9 58.0

Of the 50 children per group, 45 children in the zinc Age in months (at 1st

treatment group and 43 children with acute diarrhea in blood collection) 28.2 (13.3) 27.6 (13.5) 27.2 (13.7)

the control group had a second blood draw (Fig 1). Of the [Mean (SD)]

238 blood draws, 11 of them were either contaminated, Birth Order [median 2 (1, 8) 3 (1, 7) 3 (1, 7)

insufficient or grossly haemolyzed. Samples from the (range)]

Mother’s education (%)

index diarrhea children and diarrhea controls which did

None 54.0 37.5 23.7

not have matching first and second blood draws were 1-5 years 29.7 40.0 61.7

excluded. Thus, 37 index diarrhea children, 42 diarrhea > 6 years 16.2 22.5 14.8

controls, and 50 healthy controls were included in the Household size 5 (2, 11) 5 (2, 16) 5 (2, 15)

analysis. [median (range)]

The study was approved by the ethical review ** There were no statistically significant differences between groups
committees of ICDDR,B and the Johns Hopkins
Bloomberg School of Public Health. We obtained verbal TABLE 2. Mean Serum Concentration (µ µ mol/L) and Number of

consent from village leaders and parents for this nested Children with Low Serum Zinc Concentration*

study.
Children with Diarrhoea Healthy
Statistical analyses were done with STATA 8.0 Controls
statistical software. Mean differences in baseline Zinc (n=37) No Zinc (n=42) (n=50)
characteristics between zinc-supplemented and control
children were analyzed by analysis of variance. First Blood Draw 13.1 ± 4.3a,b 9.4 ± 1.9 10.0 ± 1.6
Mean + SD 8 (21.6) 29 (69.1) 22 (44.0)
Differences in proportions were assessed by chi-squared Number deficient (%)
analysis. The mean serum zinc concentration for each Second Blood Draw
blood draw in each group of children was calculated, as Mean + SD 11.3 ± 2.0a 10.5 ± 1.8
well as the proportion of children in each group with Number deficient (%) 10 (27.0) 15 (35.7)
concentrations less than 9.9 µmol/L, a recommended cut­
(a) Compared to healthy controls by linear regression analysis
point for zinc deficiency in young children.1 Comparisons adjusted for sex (p <0.05)
of mean serum zinc at each time point were assessed by (b) Compared to children with diarrhea not receiving zinc at first
multiple linear regression controlling for the sex of the blood draw by linear regression analysis adjusted for sex (p =
child. Although there were no differences in mean serum 0.000)
zinc by sex, we adjusted for sex to reduce the variance in *Low serum zinc defines as <9.9 µmol/L [1].
the serum zinc measures. A p value of <0.05 was
considered statistically significant. For comparison of zinc had a significantly higher mean serum zinc
mean serum zinc concentrations after the diarrhea concentrations at the first blood draw (13.1 µmol/L) and
episode, the duration of the diarrhea was added to the 1 week after supplementation ended (11.3 µmol/L) than
regression. healthy controls after controlling for sex (Table 2).
Duration of diarrhea did not influence serum zinc
concentration of the second blood draw when added into
RESULTS the regression analysis. Zinc-supplemented children had
an overall lower duration of diarrhea (4.5 + 3.6 days)
There were no differences in characteristics of the study compared to children with diarrhea who did not receive
children by group at baseline (Table 1). All children with zinc (7.0 + 5.7 days).
diarrhea from the zinc intervention and control villages
had their first blood draw within 1-4 days of starting DISCUSSION
diarrhea (1-3 days after starting zinc in the intervention
villages).
Healthy controls had a mean serum zinc concentration This study investigated the immediate and sustained
of 10.0 µmol/L (+ 1.7), the suggested lower threshold for biochemical response of zinc supplementation on serum
zinc deficiency; 44% had a serum zinc concentration lower zinc concentration within the context of a randomised
than 9.9 µmol/L. Children with diarrhea who were not controlled trial of zinc therapy for diarrhea. Given the
supplemented with zinc had a mean serum zinc community-based design, it was not possible to collect a
concentration similar to healthy controls (Table 2). blood sample from children with diarrhea prior to zinc
Children with diarrhea who were supplemented with treatment; however, given random assignment of villages,

Indian Journal of Pediatrics, Volume 73—June, 2006 495


Abdullah H. Baqui et al
it is reasonable to make post-treatment comparisons
regarding changes in zinc status with a 14-day regimen of
zinc supplementation.
The results indicate that zinc deficiency is common in
otherwise healthy children in Bangladesh; 44%, based on
the 2.5 th percentile cut off in population studies; 1 this
indicates poor individual zinc status of these children and
reflects the high prevalence of zinc deficiency among
children under 5 in Bangladesh. Children with diarrhea
who were not treated with zinc had lower serum zinc
concentrations (69% were below that threshold),
significantly more than the healthy children (p<0.05)
which could either be due to more zinc deficient children
being at greater risk of diarrhea. 17 or shifts in zinc from
blood to the liver during infections.18 These results are
congruent with previous studies in Bangladesh of
children presenting with diarrhea.16, 19
After 1-3 days of supplementation, children treated
with zinc had significantly higher serum zinc
concentrations than either unsupplemented children with
diarrhea or healthy controls. Although compliance was
not specifically measured in this subset, children living in
the zinc clusters on average consumed 7 tablets per
treatment.4 In absolute terms, the difference was 3.1-3.7
µmol/L, and when considered in relation to sample
variability, was about 1 SD. One week after finishing the
14-day zinc regimen, the serum zinc concentrations were
lower than during supplementation, but were still
significantly higher than those of healthy controls (1.3
µmol/L or 0.77 SD). They were higher than those of
control children with diarrhea (0.8 µmol/L or 0.38 SD),
although the difference was not statistically significant.
This may have been due to the small sample size. The
effects on zinc status persisted beyond the delivery of the
supplement. These results are largely consistent with the
study by Raquib et al19 in which Bangladeshi children
with shigella dysentery were randomized to receive 20
mg of zinc in a multivitamin mix or a multivitamin mix
alone for 14 days in addition to antibiotic treatment, and
were evaluated at baseline, 7 days, and 30 days (16 days
after supplementation). In that study, serum zinc
concentrations were 0.76 µmol/L higher at 7 days, and
1.22 µmol/L higher than the control group after 30 days.
As was observed here, zinc rose over time in the zinc and
control groups, but the serum zinc concentrations
continued to rise even 16 days after supplementation.
One week after supplementation, the serum zinc
concentration of our study children showed evidence of
decline from those during supplementation, but in the
study by Raquib et al 19 serum zinc concentrations
continued to rise and remained higher than baseline even
16 days after supplementation. Thus the persistence of
higher zinc concentrations following supplementation in
young children may not be universal; mediating factors
for this are yet to be identified.
Zinc supplementation has been shown to continue to
benefit the child after supplementation has ended. Short­
496
40
course treatment trials have shown a decrease in both
incidence and prevalence of diarrhea and acute lower
respiratory infections in the months following the
treatment of one episode of diarrhea. 6 It may be that a
persisting biochemical response to zinc therapy as shown
here is partly responsible for these effects. Concerns
remain about the utility of plasma or serum zinc
concentration as an indicator of individual zinc status.
Strand et al20 found that serum zinc was influenced by the
presence of other clinical symptoms or illnesses in
children with diarrhea. The children bled in this study did
not have additional illnesses. In a recent study plasma
zinc correlated well with changes in dietary zinc intake
and kinetic indicators of zinc status in adult men with no
underlying infection.21 On a population level, serum zinc
is a useful indicator of overall zinc status and shifts in zinc
nutriture. This study showed that zinc treatment for
childhood diarrhea replaced zinc stores immediately after
supplementation and kept zinc status elevated in the
weeks following illness. Further studies are needed to
define the relation between zinc concentration and whole
body zinc status during illness in order to understand the
mechanisms by which short regimens of zinc are effective
as therapeutic agents for childhood diarrhea and other
common infections, and continue to have benefits after
the supplementation period.
Acknowledgements
The study has been supported by the Johns Hopkins Family Health
and Child Survival Co-operative Agreement and ICDDR,B Co­
operative Agreement (#HRN-A-OO-06-90005-00), with funding
from the US Agency for International Development.
REFERENCES
1. Brown, KH et al. Assessment of the risk of zinc deficiency in
populations. Food and Nutrition Bulletin 2004; 25(1): S130-S162.
2. WHO, World Health Report 2002: Reducing risks, promoting
healthy life. 2002, World Health Organization: Geneva.
3. Castillo-Duran C, Vial P. and R. Uauy. Trace mineral balance
during acute diarrhea in infants. J Pediatr 1988; 113(3): 452-457.
4. Baqui, AH et al. Effect of zinc supplementation started during
diarrhoea on morbidity and mortality in Bangladeshi children:
community randomised trial. BMJ 2002; 325(7372): 1059.
5. Bhutta ZA et al. Therapeutic effects of oral zinc in acute and
persistent diarrhea in children in developing countries: pooled
analysis of randomized controlled trials. Am J Clin Nutr 2000;
72(6): 1516-1522.
6. Bhutta ZA et al. Prevention of diarrhea and pneumonia by zinc
supplementation in children in developing countries: pooled
analysis of randomized controlled trials. Zinc Investigators’
Collaborative Group. J Pediatr 1999; 135(6): 689-697.
7. WHO and UNICEF. WHO-UNICEF Joint statement on the
clinical management of acute diarrhea. in World Health Assembly.
2004. Geneva.
8. Sandstead HH, Zinc deficiency. A public health problem? Am
J Dis Child 1991; 145(8): 853-859.
9. Ghishan FK. Transport of electrolytes, water, and glucose in
zinc deficiency. J Pediatr Gastroenterol Nutr 1984; 3(4): 608-612.
10. Roy SK, et al. Impact of zinc supplementation on intestinal
Indian Journal of Pediatrics, Volume 73—June, 2006
41
Zinc Supplementation and Serum Zinc During Diarrhea
permeability in Bangladeshi children with acute diarrhoea and
persistent diarrhoea syndrome. J Pediatr Gastroenterol Nutr
1992; 15(3): 289-296.
11. Jones PE and TJ Peters, Oral zinc supplements in non-
responsive coeliac syndrome: effect on jejunal morphology,
enterocyte production, and brush border disaccharidase
activities. Gut 1981; 22(3): 194-198.
12. Ibs, K.H. and L. Rink, Zinc-altered immune function. J Nutr
2003; 133(5 Suppl 1): 1452S-6S.
13. Bhatnagar S et al. Zinc with oral rehydration therapy reduces
stool output and duration of diarrhea in hospitalized children:
a randomized controlled trial. J Pediatr Gastroenterol Nutr 2004;
38(1): 34-40.
14. Bhutta ZA, S.Q. Nizami, and Z. Isani, Zinc supplementation in
malnourished children with persistent diarrhea in Pakistan.
Pediatrics 1999; 103(4): e42.
15. Roy SK et al. Impact of zinc supplementation on persistent
diarrhoea in malnourished Bangladeshi children. Acta Paediatr
1998; 87(12): p. 1235-1239.
Indian Journal of Pediatrics, Volume 73—June, 2006
16. Roy SK et al. Randomised controlled trial of zinc
supplementation in malnourished Bangladeshi children with
acute diarrhoea. Arch Dis Child 1997; 77(3): 196-200.
17. Bhandari N et al. Substantial reduction in severe diarrheal
morbidity by daily zinc supplementation in young north
Indian children. Pediatrics 2002; 109(6): e86.
18. Brown, K.H., Effect of infections on plasma zinc concentration
and implications for zinc status assessment in low-income
countries. Am J Clin Nutr 1998; 68(2 Suppl): 425S-429S.
19. Raqib R et al. Effect of zinc supplementation on immune and
inflammatory responses in pediatric patients with shigellosis.
Am J Clin Nutr 2004; 79(3): 444-450.
20. Strand TA et al. Predictors of plasma zinc concentrations in
children with acute diarrhea. Am J Clin Nutr 2004; 79(3): 451­
456.
21. Lowe, NM et al. Kinetic parameters and plasma zinc concen­
tration correlate well with net loss and gain of zinc from men.
J Nutr 2004; 134(9): 2178-2181.
497