Short-Term Zinc Supplementation with Dispersible Tablets or Zinc Sulfate

Solution Yields Similar Positive Effects on Plasma Zinc Concentration of

Young Children in Burkina Faso: A Randomized Controlled Trial
K. Ryan Wessells, BS1, Zinewend
e P. Ou
edraogo, MD2, Noel Rouamba, MD2, Sonja Y. Hess, PhD1,
Jean-Bosco Ou
edraogo, MD, PhD2, and Kenneth H. Brown, MD1
Objective To assess zinc absorption from dispersible tablets by investigating the effects of short-term zinc
supplementation, provided either as zinc (Zn) sulfate dispersible tablets or solution, on changes in plasma Zn concentration in young children.
Study design We conducted a randomized, partially-masked, placebo-controlled trial in 451 children 6 to 23
months of age in Burkina Faso, randomly assigned to receive a dispersible tablet containing 5 mg Zn, a Zn solution
containing 5 mg Zn/5 mL, or a placebo solution, daily for 3 weeks. The main outcome measure was change in
plasma zinc concentration after supplementation compared with baseline.
Results The mean plus or minus SD change in plasma Zn concentration (mg/dL) was significantly greater in both
Zn supplemented groups (tablets: 16.9
13.1mg/dL, liquid: 16.6
14.2 mg/dL), compared with the placebo group
(0.2
10.9 mg/dL; P < .001, ANOVA). In both Zn supplemented groups, but not in the placebo group, change in
plasma Zn concentration was progressively less with increasing age in months (0.79 mg/dL/mo and 1.15
mg/dL/mo, respectively; P < .001); this effect did not differ in the Zn supplemented groups (P = .18).
Conclusions Short-term supplementation results in a large increase in plasma Zn concentration, regardless of
whether the additional Zn is provided as a dispersible tablet or solution. (J Pediatr 2012;160:129-35).

inc (Zn) deficiency is estimated to account for approximately 4% of child deaths worldwide, and the International Zinc
Nutrition Consultative Group (IZiNCG) estimates that one-third of the worlds population lives in countries with an
elevated risk of Zn deficiency.1,2 Infants and young children in developing countries are particularly vulnerable, because
of their relatively high requirements for rapid growth, inadequate intake of bioavailable Zn from complementary foods, and
high rates of infection.2,3 Preventive Zn supplementation has been shown to reduce morbidity from diarrhea and pneumonia,
lower all-cause mortality rates, and increase linear growth and weight gain in infants and young children in populations at risk
of Zn deficiency.4
In earlier studies, Zn supplements have generally been provided to young children in the form of flavored syrups. However,
other types of supplements, including dispersible tablets (eg, ZinCfant, Nutriset SAS, Malauney, France), have been developed
to simplify transport and storage logistics for large-scale Zn supplementation programs. Dispersible Zn tablets were used in two
large-scale studies in young children in Tanzania and Nepal, which found no effect on morbidity and a less than expected effect
on plasma Zn concentration.5,6 These latter results are in contrast to the findings of two meta-analyses, which have concluded
that Zn supplementation produces a consistent, moderately large increase in mean serum Zn concentration across a wide range
of daily doses (2.9-21.4 mg/day) and duration of supplementation (2 weeks-14 months).4,7
These differing sets of results raise concerns that the Zn may be less well absorbed from the dispersible tablet. Therefore, this
study was designed to assess Zn absorption from the dispersible tablets by investigating the effects of short-term Zn supplementation, provided either as a Zn sulfate-containing dispersible tablet (ZnTab) or solution (ZnLiq), on changes in plasma Zn
concentration in young children in Burkina Faso.

AGP
CRP
Cu
IZiNCG
LAZ
WAZ
WLZ
Zn
ZnLiq
ZnTab

a-1-acid glycoprotein
C-reactive protein
Copper
International Zinc Nutrition Consultative Group
Length-for-age z-score
Weight-for-age z-score
Weight-for-length z-score
Zinc
Zinc sulfate liquid supplement
Zinc sulfate dispersible tablet

From the 1Department of Nutrition, University of

cherche en
California, Davis, CA; and 2Institut de Re

, Ouagadougou, Burkina Faso
Sciences de la Sante
Funded by a grant from Nutriset, SAS, Malauney, France,
which had no influence on the study design, collection,
analysis, interpretation of data, writing of the report, and
the decision to submit the article for publication. The
authors declare no conflicts of interest.
Registered at ClinicalTrials.gov: NCT00944853.
0022-3476/$ - see front matter. Copyright 2012 Mosby Inc.
All rights reserved. 10.1016/j.jpeds.2011.06.051

129

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Methods
The study was designed as a randomized, partially masked,
placebo-controlled intervention trial. The study was conducted from September to December 2009, in 31 rural communities within the catchment area of the governmental
health clinic located in Toussiana, Burkina Faso. Infants
and young children were identified with a door-to-door census and invited to attend screening examinations at the health
clinic; their participation in the intervention trial was requested when they met these inclusion criteria: age 6 to 23
months, currently breastfeeding, hemoglobin level $60 g/L,
and no fever or diarrhea (>3 liquid or semi-liquid stools in
a 24-hour period) reported in the past week. Those children
who were currently consuming vitamin or mineral supplements or Zn-fortified infant formulas or who demonstrated
bipedal edema or other serious medical conditions, were excluded. In the case of twins, only one of each pair was enrolled in the trial; however, both received identical
supplements. Children ineligible because of an aforementioned acute illness were excluded temporarily and rescreened after recovery.
Consent materials were presented, both orally and in written form, in the presence of a neutral witness. Informed consent, documented with either a written signature or
a fingerprint, was obtained from the parents of each child before his or her enrollment in the study. The study protocol
was approved by the institutional review boards of the Centre
Muraz in Bobo-Dioulasso, Burkina Faso, and the University
of California, Davis, and registered as a clinical trial (www.
ClinicalTrials.gov; NCT00944853).
Eligible study participants were randomly assigned to 1 of
3 treatment groups by using an independently generated
block randomization scheme, with a varied block length
of 3 or 6.8 The daily supplementation regimens were: (1)
dispersible tablet, containing 5 mg Zn as zinc sulfate
(ZnTab); (2) liquid Zn supplement, containing 5 mg Zn
as zinc sulfate per 5 mL (ZnLiq); or (3) liquid placebo supplement. ZnTabs were provided by Nutriset SAS (Malauney,
France). ZnLiq supplements, prepared by study personnel,
were formulated by compounding Zn sulfate heptahydrate
with cherry syrup (Humco, Texarkana, Texas). Zn and placebo syrups were indistinguishable in appearance, flavor,
and packaging. Zn concentrations of all supplements were
confirmed with atomic absorption spectrophotometry at
the Institute for Research in the Health Sciences (Bobo-Dioulasso, Burkina Faso) and the University of Otago (Dunedin, New Zealand) before distribution. Blister packs or
bottles were independently coded with subject identification
numbers. Supplements were administered daily for 21 days
by study fieldworkers who visited the childrens homes each
morning. Tablets were dissolved in approximately 5 mL of
water, and liquid supplements was measured with a 5-mL
spoon. Caregivers were instructed to feed their children
only water or breast milk for 30 minutes before and after
supplement consumption.
130

Vol. 160, No. 1

Data on childrens morbidity were collected daily during
the fieldworkers home visit, with a systematic, symptombased recall. Elicited information included general health status, appetite, stool number and consistency, and presence of
vomiting, nasal discharge, cough, respiratory difficulty, rapid
respiration, and fever. Axillary temperature was measured
whenever fever was reported. During the supplementation
period, medical consultations and treatment for malaria
and diarrhea were provided gratis, according to previously
developed diagnostic and treatment algorithms, and this information was recorded. Information on household assets,
parental education and occupation, and housing quality
was obtained through a structured interview and direct observation of housing characteristics during the second week
of the intervention period.
Childrens recumbent length and weight were measured
at baseline, following the protocols established by the
Food and Nutrition Technical Assistance Project.9
Weight-for-age (WAZ), length-for-age (LAZ), and weightfor-length (WLZ) z-scores were calculated according to
the World Health Organization growth standards.10 Venous
blood samples were collected at baseline and after supplementation for measurement of plasma Zn and copper
(Cu) concentrations and levels of C-reactive protein
(CRP), a-1-acid glycoprotein (AGP), and hemoglobin
(baseline only). When a child was acutely ill with diarrhea
or fever within 48 hours of the final scheduled blood drawing, sample collection was postponed and supplementation
was continued for as long as 1 week more until the child
was symptom-free for at least 48 hours. When it was not
possible to collect the final blood sample in these conditions, the child was excluded from the study. Blood samples
were collected in the morning, from 1 to 2 hours after the
last breastfeeding episode. For each blood-sampling event, 5
mL of blood was drawn from an antecubital or dorsal metacarpal vein according to procedures recommended by
IZiNCG.2,11 The blood was collected in evacuated, trace
element-free polyethylene tubes containing lithium heparin
(Sarstedt AG & Co, Numbrecht, Germany), and immediately stored at 4 C until separation. Blood samples were
transported from the field site to the Institute for Research
in the Health Sciences laboratory, and plasma was separated
from heparinized blood by centrifuging at 1520  g for 10
minutes. Plasma was separated within 8 hours of collection,12 and the plasma aliquots were stored at 20 C. Standard protocols and trace element-free materials were used
to avoid Zn contamination of samples.11
At baseline, hemoglobin concentration was analyzed immediately after blood collection with a HemoCue 201+ photometer (HemoCue AB, Angelholm, Sweden). Plasma Zn
and Cu concentrations were measured with inductively coupled plasma optical emission spectrophotometry (Vista; Varian Inc, Walnut Creek, California) at the Childrens Hospital
of Oakland Research Institute. Plasma samples were digested
overnight at 60 C in OmniTrace 70% HNO3 (VWR International, West Chester, Pennsylvania), which was then diluted
to a final concentration of 5.5% HNO3 and centrifuged at
Wessells et al

ORIGINAL ARTICLES

January 2012
3000  g for 10 minutes before analysis.13 Each batch of samples run on the inductively coupled plasma optical emission
spectrophotometry was analyzed with these reference materials, which were prepared in a manner identical to that of
the clinical samples: Seronorm Trace Elements Serum L-1
and L-2 (lot 090310 and lot NO0371z, respectively; Accurate
Chemical and Scientific Corp, Westbury, New York), and internal pooled human plasma and animal serum controls. Zn
and Cu concentrations of all analyzed reference materials
were found to be within the acceptable ranges according to
the manufacturers specifications. Inter-run co-efficients of
variation for reference materials ranged between 2.5% and
4.5%. Samples were analyzed in duplicate, and all samples
from a particular subject were analyzed within the same analytic run. If plasma Zn concentrations differed by greater
than 10% between duplicates, all samples for that subject
were rerun. Duplicates where plasma Cu differed by greater
than 10% were excluded from analyses (n = 6). Plasma
CRP and AGP concentrations were measured with radial immunodiffusion (The Binding Site Limited, Birmingham,
United Kingdom and Kent Laboratories, Bellingham, Washington, respectively). All plasma samples were coded and analyzed subsequently in the laboratory without knowledge of
the subjects study group.
The main outcome variable was change in plasma Zn
concentration. To detect treatment-related differences having an effect size of 0.40 SD, by using 3-way group-wise
comparisons, a sample size of 150 participants per group
was necessary (a = 0.05, b = 0.20, 20% estimated attrition).
All analyses were conducted on an intention-to-treat basis.
Group-wise differences at baseline were compared by using
ANOVA for continuous variables and c2 tests for categorical variables. Changes in plasma Zn and Cu concentrations
after the intervention were assessed by analysis of
co-variance, with supplementation group assignment as
the main effect and controlling for co-variates (age, initial
plasma Zn or Cu concentration, initial anthropometric
measurements, methodological factors related to blood collection, acute phase protein concentrations, and morbidity
prevalence) as appropriate. Group means were compared
post-hoc using least-square means with the Tukey-Kramer
adjustment. The co-variates that remained significant in
the analysis of co-variance model were adjusted for in calculating within-group partial correlation co-efficients.
Plasma Zn and Cu concentrations, adjusted for the presence
of inflammation (elevated acute phase proteins categorically
defined as CRP $10 mg/L, AGP $1 g/L, or both) were obtained with the co-efficients obtained from the main effects
model at baseline. Between-group differences in morbidity
prevalence were analyzed with the non-parametric Kruskal-Wallis test. All statistical analyses were completed with
SAS System software for Windows release 9.2 (SAS Institute, Cary, North Carolina). Data are presented as means
plus or minus SE for analysis of co-variance, and means
plus or minus SD for all other analyses, unless otherwise
noted. Treatment groups remained masked until all statistical analyses were completed.

Results
Of the 556 children screened for eligibility, 451 participants
(81%) were enrolled, and 431 participants (96% of those
enrolled) completed the study (Figure 1; available at www.
jpeds.com). There were no significant group-wise differences
in attrition, nor significant differences in baseline characteristics
between children who completed the study and children who
did not. Treatment groups did not differ at baseline in the
childrens sex, age, socioeconomic indicators, anthropometrics,
and biochemical indicators of nutritional status (Table I).
Reported compliance with supplement intake was 99.9% of
possible doses, and 98.0% of these were consumed under the
observation of a study fieldworker. There were no groupwise differences in supplement consumption.
Baseline plasma Zn and Cu concentrations were normally
distributed and did not differ among groups. The overall
mean plasma Zn concentration at baseline was 62.9
11.6
mg/dL. The prevalence of low plasma Zn concentrations at
baseline, with a cutoff point of 65 mg/dL, as recommended
by IZiNCG, was 62.7%.2 Infants aged 6 to 11 months had
a significantly higher plasma Zn concentration than children
aged 12 to 23 months (65.4
12.5 mg /dL versus 60.9
10.4
mg /dL, respectively; P < .0001), and boys had significantly
higher plasma Zn concentrations than girls (64.1
12.7 mg
/dL versus 61.6
10.2 mg/dL, respectively; P = .02). The
mean plasma Cu concentration at baseline was 195.8
43.8
mg/dL, and the prevalence of plasma Cu concentrations elevated higher than the reference range (64-156 mg/dL)14 at
baseline was 81.6%. There was a high prevalence of elevated
acute phase proteins in this population; children with elevated CRP concentration, AGP concentration, or both at
baseline had significantly lower mean plasma Zn concentrations and significantly higher mean plasma Cu concentrations (Table II).
Mean plasma Zn concentrations in children who received
either the ZnTab or the ZnLiq increased significantly
(16.9
13.1 mg/dL and 16.6
14.2 mg/dL, respectively) compared with children who received the placebo supplement
(0.2
10.9 mg/dL; P < .0001, ANOVA); the plasma Zn responses did not differ for the two forms of Zn supplement
(P = .98; Table III). Several child characteristics and study
methodological factors were significantly associated with
the change in plasma Zn concentrations, independent of
study group. Specifically, initial plasma Zn concentration,
childs age and LAZ, time of day (final), CRP concentration
$10 mg/L (final), and AGP concentration $1g/L (final) were
negatively related to change in plasma Zn concentration.
Elapsed time since last breastfeeding (final), CRP
concentration $10 mg/L (baseline), and the occurrence of
a medical consultation during the intervention period were
positively related to change in plasma Zn concentration.
When included in subsequent analyses, however, the
combined effect of these co-variates on group-wise mean
change in plasma Zn concentration was minimal (<0.5 mg/
dL). Children with higher initial plasma Zn concentrations

Short-Term Zinc Supplementation with Dispersible Tablets or Zinc Sulfate Solution Yields Similar Positive Effects
on Plasma Zinc Concentration of Young Children in Burkina Faso: A Randomized Controlled Trial

131

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Vol. 160, No. 1

Table I. Initial characteristics of study participants and their households

Subjects, n
Male, n (%)
Age, monthsz
Socioeconomic indicators
Paternal education, n (%) completing primary school
Material education, n (%) completing primary school
Source of water for household, n (%)
Surface
Well
Hand pump
Electricity, n (%)
Latrine facilities, n (%)
Anthropometric measurements
Length, cm
Weight, kg
LAZ
WAZ
WLZ
Biochemical assessments
Hb concentration (g/L){
Hb <110 g/L, n (%){
AGP $1 g/L, n (%)**
CRP $10 mg/L, n (%)**

ZnTab

ZnLiq

Placebo

P value*

150
86 (57.3)
13.4
5.1

150
73 (48.7)
13.9
5.1

151
70 (46.4)
13.7
5.4

.13
.76x

46 (31.7)
24 (16.6)

38 (25.9)
14 (9.5)

35 (23.2)
21 (13.9)

.51
.20
.39

18 (12.4)
75 (51.7)
52 (35.9)
33 (22.8)
98 (67.6)

18 (12.2)
81 (55.1)
48 (32.7)
31 (21.1)
97 (66.0)

16 (10.6)
95 (62.9)
40 (26.5)
23 (15.2)
98 (64.9)

.23
.89

72.6
5.8
8.3
1.4
1.4
1.4
1.3
1.1
0.8
1.1

72.6
5.6
8.4
1.5
1.5
1.2
1.2
1.1
0.6
0.9

72.3
5.8
8.2
1.5
1.5
1.2
1.4
1.1
0.8
1.0

.83x
.43x
.53x
.46x
.17x

89
16
132 (89.2)
113 (75.3)
57 (38.0)

91
16
130 (89.0)
111 (76.0)
54 (37.0)

88
14
137 (94.5)
106 (70.7)
44 (29.3)

.30x
.17
.52
.23

Hb, hemoglobin.
*P values are group-wise.
c2 test.
zMean
SD, all such values.
xANOVA.
{ZnLiq, n = 148; ZnTab, n = 146; placebo, n = 147.
**ZnTab, n = 150; ZnLiq, n = 146; placebo, n = 150.

had a smaller change in plasma Zn concentration, and vice

versa (0.36 mg/dL less increase in plasma Zn concentration
for each mg/dL greater initial plasma Zn concentration).
This occurred regardless of treatment group, probably
indicating regression to the mean (Figure 2; available at
www.jpeds.com). There were significant group-wise
differences in the effect of age on change in plasma Zn
concentration (P < .0001). In both the ZnTab and ZnLiq
groups, but not the placebo group, change in plasma Zn
concentration declined progressively with increasing age
in months (b = 0.79 mg/dL/month and 1.15 mg/dL/
month, respectively; Figure 3, available at www.jpeds.com);
these age-dependent effects on change in plasma Zn
concentration did not differ in the two Zn supplemented
groups (P = .18). Mean plasma Cu concentrations in
children who received either ZnTab or ZnLiq decreased
significantly (17.7
2.6 mg/dL and 19.3
2.6 mg/dL,
respectively) compared with children who received the
placebo supplement (8.4
2.5 mg/dL; P < .0001) when

controlling for baseline Cu copper concentration. The

plasma Cu response did not differ for the two forms of Zn
supplements (P = .90). The prevalence of reported fever
(6.6%), diarrhea (1.4%), and other signs of infection were
low during the period of observation and did not differ
significantly among treatment groups. The change in
plasma Zn and Cu concentrations did not differ by illness
prevalence, independent of study group, although the small
number of observations limits the statistical power of this
analysis. There were no group-wise differences in the
presence of elevated CRP or AGP concentrations after
supplementation. Acute phase protein (CRP and AGP)
concentrations were significantly associated with change in
plasma Zn and Cu concentrations (Table IV).

Discussion
The results of this study indicate that both ZnTab and ZnLiq
substantially and significantly increase mean plasma Zn

Table II. Effect of CRP and AGP on baseline plasma Zn and Cu concentrations*

Subjects, n
Plasma Zn concentration, mg/dLz
Plasma Cu concentration, mg/dL

CRP 10 mg/L
AGP <1 g/L

CRP 10 mg/L
AGP 1 g/L

CRP <10 mg/L

AGP 1 g/L

Neither
elevated

P value

10
62.5
9.8ab
197.0
34.4a-c

145
60.1
11.8a
219.9
44.3a

185
61.9
10.5a
193.6
38.5b

105
66.8
12.3b
166.0
32.1c

<.001x
<.001x

*Values in a row with different superscript letters were significantly different, P < .0001.
P values are group-wise.
zMean
SD, all such values.
xANOVA.

132

Wessells et al

ORIGINAL ARTICLES

January 2012

Table III. Plasma zinc concentrations of study participants by treatment group and stage of study*
Subjects, n
Initial plasma Zn concentration, mg/dLz
Initial plasma Zn <65mg/dL, n (%){
Unadjusted
Adjusted
Change in plasma Zn concentration, mg/dLzz
Final plasma Zn <65mg/dL, n (%)
Unadjusted
Adjusted

ZnTab

ZnLiq

Placebo

P value

149
63.6
12.4

146
62.1
11.4

150
63.0
11.0

.49x

88 (59.1)
71 (47.7)
16.9
13.1a

99 (67.8)
74 (50.7)
16.6
14.2a

92 (61.3)
73 (48.7)
0.2
10.9b

.27**
.87**
<.0001x

23 (16.2)a
15 (10.6)a

30 (21.9)a
21 (15.3)a

91 (62.3)b
59 (40.4)b

<.0001x
<.0001x

*Values in a row with different superscript letters were significantly different, P < .0001.
P values are group-wise.
zMean
SD, all such values.
xANOVA; means in Table III were not adjusted for co-variates.
{n (%), all such values.
**c2 test.
Predicted n (%) of children with initial plasma Zn concentrations <65 mg/dL when values are adjusted for CRP $10 mg/L and AGP $1 g/L.
zzZnLiq, n = 137; ZnTab, n = 142; placebo, n = 146.

concentrations in Burkinabe children 6 to 23 months of age,

after a 21-day supplementation period. The randomized,
partially masked clinical trial design, direct observation of
supplementation consumption, and highly standardized
and rigorous collection and processing of plasma Zn samples
lend strength to these findings. The overall effect size of supplementation in this study was 1.31 SD (95% CI, 1.09-1.53).
A recent meta-analysis of preventive Zn supplementation trials in pre-pubertal children found that plasma Zn concentration consistently was increased after Zn supplementation,
with a moderately large overall effect size of 0.60 SD (95%
CI, 0.44-0.77).4 Although the results of this study are in
agreement with the overall findings of the meta-analysis,
the observed effect size was much larger in this trial, possibly
because of the high rates of compliance with supplement intake and the instructions to consume the supplements apart
from meals. The results of this study are consistent with the
findings from a recent study conducted by Taneja et al, which
found a 26.9 mg/dL (95% CI, 19.6-34.2) difference in plasma
Zn concentrations between low-birth-weight infants supplemented with ZnTab compared with a placebo for 12
months.15 However, these results differ from those of other
recent studies that used commercially prepared ZnTab as opposed to the ZnLiq used in most earlier studies. Studies conducted by Sazawal et al5 and Mazariegos et al16 found either
less effect than expected or no effect of Zn supplementation
on plasma Zn concentrations. Although Sazawal et al and
Mazariegos et al concluded that Zn was bioavailable from
these supplements because of the trend toward increased

plasma Zn concentrations in the supplemented groups,5,16

concerns were raised that the bioavailability of Zn from
ZnTab was lower than that of ZnLiq, especially considering
the lack of effect of Zn supplementation on linear growth16
and morbidity6 in these studies. The previously cited metaanalysis found significant heterogeneity of results, but the
source of this heterogeneity could not be identified.4 In particular, the results were unrelated to dose, duration of supplementation, or combined supplementation with other
micronutrients. The lack of expected response in plasma
Zn concentration in the aforementioned trials that used
ZnTab may be caused by other confounding factors not related to the physical form of the supplement, such as compliance with supplement intake or lack of standardization of
blood collection methods.
We found that plasma Zn concentrations were significantly higher in infants 6 to 11 months of age than children
12 to 23 months of age, before supplementation. Although
the older children had significantly higher AGP concentrations, an indicator of inflammation and possible sequestration of Zn in the liver, the age-related differences in plasma
Zn concentration remained significant even after controlling
for AGP concentration. There are currently no reference data
for plasma Zn concentrations for children <3 years of age. A
review by Hess et al recommended a single reference value for
all children >1 year of age, but noted that a different value
may be necessary for infants.17 A longitudinal study of
healthy Danish infants found a significant decrease in serum
Zn concentrations between 6 and 9 months of age18;

Table IV. Effect of CRP and AGP concentrations on change in plasma Zn and Cu concentrations*,

Subjects, n
Change in plasma Zn concentration, mg/dLz
Change in plasma Cu concentration, mg/L

Elevated APP,
baseline only

Elevated APP,
baseline and final

Elevated APP,
final only

Elevated APP,
neither

82
13.9
1.3a
34.9
3.1a

244
10.2
0.8ab
10.4
1.8b

48
7.6
1.7b
7.2
4.0c

51
14.4
1.7b
27.1
4.0a

APP, acute phase protein.

*Elevated APP defined as CRP concentration $10 mg/L, AGP concentration $1 g/L, or both.
Analysis of co-variance model, controlling for initial plasma Zn or Cu concentration and group. Values in a row with different superscript letters were significantly different, P < .05.
zMean
SE, all such values.

Short-Term Zinc Supplementation with Dispersible Tablets or Zinc Sulfate Solution Yields Similar Positive Effects
on Plasma Zinc Concentration of Young Children in Burkina Faso: A Randomized Controlled Trial

133

THE JOURNAL OF PEDIATRICS

www.jpeds.com

however, it is unknown whether Zn concentrations would

have been even lower in older children, as in this study. In
contrast to the results of this study, a study in healthy Belgian
children found that infants <12 months of age had significantly lower mean serum Zn concentrations than children
12 to 23 months old.19 We are unable to determine whether
the age-related differences in baseline plasma Zn concentration observed in this study are caused by physiological differences, which would also be present in well-nourished
populations with low rates of infection, or whether they are
caused by an increasing risk of Zn deficiency in older children
resulting from inadequate dietary intake from complementary foods and increased Zn requirements.
Change in plasma Zn concentration in the Zn supplemented groups was significantly associated with age, with
younger children having a larger change in plasma Zn concentration than older children. This effect may be doserelated (mg Zn/kg body weight), because all children received
5 mg of Zn per day regardless of age and body weight. Because of the design of this study, we are unable to distinguish
between the effect of age and weight, so additional studies will
be necessary to further elucidate these relationships. Alternatively, the observed age-related responses could reflect differences in infants and young children in absorption efficiency
or the development of regulatory mechanisms of Zn homeostasis. However, results from dual stable isotope studies in
infants 4 months of age suggest early maturation of mechanisms that regulate absorption and conserve endogenous
Zn, with efficiencies comparable with adults after adjusting
for intestinal length.20 Thus, maturation of these functions
does not seem to explain the observed differences. Finally,
the observed age-related responses could reflect an increase
in the prevalence of intestinal mucosal injury with increasing
age.21
This study was conducted in a population with a high
prevalence of infection and inflammation, as evidenced by
the large number of children with elevated plasma levels
of acute phase proteins. Elevated CRP and AGP concentrations were significantly associated with the change in
plasma Zn concentration, as has been observed previously.22-24 Our results do not support the recent suggestion
by Mburu et al that inflammation actually blocks the uptake
of Zn, rather than transiently reducing plasma Zn concentration irrespective of Zn status.25 Instead, children with elevated acute phase proteins at both points in this study had
similar changes in plasma Zn concentration as children with
non-elevated acute phase proteins at both points. The larger
change in plasma Zn concentrations observed when acute
phase proteins were elevated only at baseline and the
smaller change when the acute phase proteins were elevated
only at the final sampling of blood probably reflects redistribution of Zn in different body compartments, partially
confounding measurement of Zn status by change in
plasma Zn concentration.26
Decreases in mean plasma Cu concentrations from baseline were significantly greater in the Zn supplemented groups
compared with the placebo group, although all plasma Cu
134

Vol. 160, No. 1

values remained within or higher than the reference range.
Although the recommended upper level of Zn intake in
adults is based on the effect of excess Zn on Cu status,27
the decreases in plasma Cu concentrations associated with
Zn supplementation in this study are inconsistent with results of most Zn supplementation studies in young children
that reported on biomarkers of Cu status.28-35 However,
plasma Cu concentration is an insensitive indicator of Cu status, and because of the role of ceruloplasmin as an acute
phase reactant, its concentrations during inflammation and
infection are often 2 to 3 times greater than reference
values.36 Post-supplementation decreases in plasma Cu concentration in the Zn supplemented groups may be indicative
of slightly reduced concentrations of acute phase proteins
that the semiquantitative CRP and AGP radial immunodiffusion assays were not sensitive enough to discern.
The results of this study support the possibility of using
dispersible tablet supplements in large-scale Zn intervention
programs, which could reduce costs associated with transportation and storage. Further studies should be conducted
to determine reference values for plasma Zn concentration
in infants and young children and to elucidate the impact
of childrens age and weight on changes in plasma Zn concentration after Zn supplementation. n
We greatly appreciate the contributions of the entire Burkinab
e ZincTab study team, including the community fieldworkers, phlebotomists,
laboratory assistants, nurses, data entry and office support staff, and
the Laboratoire Phytofla. We thank David Killilea for assistance with
the zinc analyses in the laboratory of Janet King at the Childrens Hospital of Oakland Research Institute, Karl Bailey (University of Otago,
Dunedin, New Zealand) for laboratory training in Burkina Faso, and
Janet Peerson (University of California, Davis) for assistance with the
statistical analyses. Finally, we sincerely appreciate the children, their
parents, and local community whose support and participation made
this study possible.
Submitted for publication Feb 25, 2011; last revision received May 6, 2011;
accepted Jun 28, 2011.

References
1. Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al.
Maternal and child undernutrition: global and regional exposures and
health consequences. Lancet 2008;371:243-60.
2. Brown K, Rivera J, Bhutta Z, Gibson R, King J, L
onnerdal B, et al. International Zinc Nutrition Consultative Group (IZiNCG) technical document #1. Assessment of the risk of zinc deficiency in populations and
options for its control. Food Nutr Bull 2004;25:S99-203.
3. Kimmons JE, Dewey KG, Haque E, Chakraborty J, Osendarp SJ,
Brown KH. Low nutrient intakes among infants in rural Bangladesh
are attributable to low intake and micronutrient density of complementary foods. J Nutr 2005;135:444-51.
4. Brown KH, Peerson JM, Baker SK, Hess SY. Preventive zinc supplementation among infants, preschoolers, and older prepubertal children.
Food Nutr Bull 2009;30:S12-40.
5. Sazawal S, Black RE, Ramsan M, Chwaya HM, Dutta A, Dhingra U, et al.
Effect of zinc supplementation on mortality in children aged 1-48
months: a community-based randomised placebo-controlled trial.
Lancet 2007;369:927-34.
6. Tielsch JM, Khatry SK, Stoltzfus RJ, Katz J, LeClerq SC, Adhikari R, et al.
Effect of daily zinc supplementation on child mortality in southern

Wessells et al

January 2012

7.

8.
9.

10.

11.

12.

13.

14.
15.

16.

17.

18.

19.
20.
21.

22.

Nepal: a community-based, cluster randomised, placebo-controlled

trial. Lancet 2007;370:1230-9.
Brown KH, Peerson JM, Allen LH. Effect of zinc supplementation on
childrens growth: a meta-analysis of intervention trials. Bibl Nutr Dieta
1998;76-83.
Moses L, Oakford R. Tables of random permutation. Stanford, CA: Stanford University Press; 1963.
Cogill B. Anthropometric indicators measurement guide. In: Food and
nutrition. Washington, DC: Technical Assistance Project, Academy for
Educational Development; 2003.
World Health Organization Multicentre Growth Reference Study
Group. WHO Child Growth Standards based on length/height, weight
and age. Acta Paediatr Suppl 2006;450:76-85.
International Zinc Nutrition Consultative Group (IZiNCG). Assessing
population zinc status with serum zinc concentration. IZiNCG technical
brief no. 2. Davis, CA: University of California; 2007.
Tamura T, Johnston KE, Freeberg LE, Perkins LL, Goldenberg RL. Refrigeration of blood samples prior to separation is essential for the accurate determination of plasma or serum zinc concentrations. Biol Trace
Elem Res 1994;41:165-73.
Killilea DW, Ames BN. Magnesium deficiency accelerates cellular senescence in cultured human fibroblasts. Proc Natl Acad Sci U S A 2008;105:
5768-73.
Turnlund JR. Copper nutriture, bioavailability, and the influence of dietary factors. J Am Diet Assoc 1988;88:303-8.
Taneja S, Bhandari N, Rongsen-Chandola T, Mahalanabis D,
Fontaine O, Bhan MK. Effect of zinc supplementation on morbidity
and growth in hospital-born, low-birth-weight infants. Am J Clin Nutr
2009;90:385-91.
Mazariegos M, Hambidge KM, Westcott JE, Solomons NW, Raboy V,
Das A, et al. Neither a zinc supplement nor phytate-reduced maize
nor their combination enhance growth of 6- to 12-month-old Guatemalan infants. J Nutr 2010;140:1041-8.
Hess SY, Peerson JM, King JC, Brown KH. Use of serum zinc concentration as an indicator of population zinc status. Food Nutr Bull 2007;28:
S403-29.
Michaelsen KF, Samuelson G, Graham TW, Lonnerdal B. Zinc intake,
zinc status and growth in a longitudinal study of healthy Danish infants.
Acta Paediatr 1994;83:1115-21.
Van Biervliet S, Van Biervliet JP, Bernard D, Vercaemst R, Blaton V. Serum zinc in healthy Belgian children. Biol Trace Elem Res 2003;94:33-40.
Hambidge KM, Krebs NF, Westcott JE, Miller LV. Changes in zinc absorption during development. J Pediatr 2006;149:S64-8.
Goto K, Chew F, Torun B, Peerson JM, Brown KH. Epidemiology of altered intestinal permeability to lactulose and mannitol in Guatemalan
infants. J Pediatr Gastroenterol Nutr 1999;28:282-90.
Wieringa FT, Dijkhuizen MA, West CE, Northrop-Clewes CA, Muhilal.
Estimation of the effect of the acute phase response on indicators of micronutrient status in Indonesian infants. J Nutr 2002;132:3061-6.

ORIGINAL ARTICLES
23. Brown KH, Lanata CF, Yuen ML, Peerson JM, Butron B, Lonnerdal B.
Potential magnitude of the misclassification of a populations trace element status due to infection: example from a survey of young Peruvian
children. Am J Clin Nutr 1993;58:549-54.
24. Thurnham DI, Mburu AS, Mwaniki DL, De Wagt A. Micronutrients in
childhood and the influence of subclinical inflammation. Proc Nutr Soc
2005;64:502-9.
25. Mburu AS, Thurnham DI, Mwaniki DL, Muniu EM, Alumasa FM. The
influence of inflammation on plasma zinc concentration in apparently
healthy, HIV+ Kenyan adults and zinc responses after a multimicronutrient supplement. Eur J Clin Nutr 2010;64:510-7.
26. Singh A, Smoak BL, Patterson KY, LeMay LG, Veillon C, Deuster PA.
Biochemical indices of selected trace minerals in men: effect of stress.
Am J Clin Nutr 1991;53:126-31.
27. Institute of Medicine. Zinc. In: Food and Nutrition Board, ed. Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium,
and zinc. Washington, DC: National Academy Press; 2002. p. 442-501.
28. Walravens PA, Krebs NF, Hambidge KM. Linear growth of low income
preschool children receiving a zinc supplement. Am J Clin Nutr 1983;38:
195-201.
29. Sazawal S, Malik P, Jalla S, Krebs N, Bhan MK, Black RE. Zinc supplementation for four months does not affect plasma copper concentration
in infants. Acta Paediatr 2004;93:599-602.
30. Ruz M, Castillo-Duran C, Lara X, Codoceo J, Rebolledo A, Atalah E. A
14-mo zinc-supplementation trial in apparently healthy Chilean preschool children. Am J Clin Nutr 1997;66:1406-13.
31. Brooks WA, Santosham M, Naheed A, Goswami D, Wahed MA, DienerWest M, et al. Effect of weekly zinc supplements on incidence of pneumonia and diarrhoea in children younger than 2 years in an urban, lowincome population in Bangladesh: randomised controlled trial. Lancet
2005;366:999-1004.
32. Baqui AH, Walker CL, Zaman K, Arifeen SE, Chowdhury HR,
Wahed MA, et al. Weekly iron supplementation does not block increases
in serum zinc due to weekly zinc supplementation in Bangladeshi infants. J Nutr 2005;135:2187-91.
33. Rosado JL, Lopez P, Kordas K, Garcia-Vargas G, Ronquillo D, Alatorre J,
et al. Iron and/or zinc supplementation did not reduce blood lead concentrations in children in a randomized, placebo-controlled trial. J Nutr
2006;136:2378-83.
34. Lind T, Lonnerdal B, Stenlund H, Ismail D, Seswandhana R,
Ekstrom EC, et al. A community-based randomized controlled trial of
iron and zinc supplementation in Indonesian infants: interactions between iron and zinc. Am J Clin Nutr 2003;77:883-90.
35. Wuehler SE, Sempertegui F, Brown KH. Dose-response trial of prophylactic zinc supplements, with or without copper, in young Ecuadorian
children at risk of zinc deficiency. Am J Clin Nutr 2008;87:723-33.
36. Mason K. A conspectus of research on the copper metabolism and requirements of man. J Nutr 1979;109:1979-2066.

Short-Term Zinc Supplementation with Dispersible Tablets or Zinc Sulfate Solution Yields Similar Positive Effects
on Plasma Zinc Concentration of Young Children in Burkina Faso: A Randomized Controlled Trial

135

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Enrollment

Vol. 160, No. 1

Assessed for eligibility (n= 556)

Randomized (n=451)

Excluded (n=105)
Age <6 months, >23 months (n=5)
Not breastfed (n=3)
Fever (n=70)
Severe malnutrition, edema (n=2)
Hemoglobin <60g/L (n=5)
Consuming Zn fortified foods (n=1)
Twin of enrolled child (n=7)
Blood draw failure (n=3)
Acute lower respiratory infection (n=8)

Allocation

Allocated to Liquid Zinc (n=150)

Allocated to Zinc Tablet (n=150)

Allocated to Placebo (n=151)

Follow-Up

Discontinued intervention (n=9)

Withdrew consent (n=8)
Illness (n=1)
Blood draw failure (n=4)

Discontinued Intervention (n=7)

Withdrew consent (n=3)
Moved from study area (n=3)
Illness (n=1)
Blood draw failure (n=1)

Discontinued Intervention (n=4)

Withdrew consent (n=3)
Illness (n=1)

Anthropometry (n=150)
Complete biochemistry (n=142)

Anthropometry (n=151)
Complete biochemistry (n=146)

Analysis
Anthropometry (n=150)
Complete biochemistry (n=137)

Figure 1. Consort flow diagram of childrens participation in the screening survey and enrollment in the clinical trial. Children
noted as excluded in the consort flow diagram were either excluded from the study because of their fulfillment of a permanent
exclusion criterion or they did not return for re-screening after temporary exclusion because of an acute illness.

135.e1

Wessells et al

January 2012

ORIGINAL ARTICLES

Figure 2. The magnitude of change in plasma Zn concentration depended on the participants baseline plasma Zn concentration, irrespective of study group, probably indicating regression to the mean. Number of participants by group: ZnTab, n = 142;
ZnLiq, n = 137; placebo, n = 146.

Short-Term Zinc Supplementation with Dispersible Tablets or Zinc Sulfate Solution Yields Similar Positive Effects
on Plasma Zinc Concentration of Young Children in Burkina Faso: A Randomized Controlled Trial

135.e2

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Vol. 160, No. 1

Figure 3. In both the ZnLiq and ZnTab groups, but not the placebo group, change in plasma Zn concentration progressively
declined with increasing age in months (P < .0001, analysis of co-variance); these age-dependent effects on change in plasma Zn
concentration did not differ in the two Zn supplemented groups (P = .18). Within-group partial correlation co-efficients are adjusted for initial plasma Zn concentration, LAZ, CRP $10 mg/L (baseline and final), and AGP $1 g/L (final), time of day (final)
elapsed time since last breastfeeding (final), and the occurrence of a medical consultation. Number of participants by group:
ZnTab, n = 142; ZnLiq, n = 137; placebo, n = 146.

135.e3

Wessells et al