JOURNAL OF TROPICAL PEDIATRICS, VOL. 58, NO.

1, 2012

Effect of Zinc Supplementation on Growth in Very Low Birth

Weight Infants
by T. V. Ram Kumar, and Siddarth Ramji
Neonatal Division, Department of Pediatrics, Maulana Azad Medical College, New Delhi 110002, India

Correspondence: T. V. Ram Kumar, Neonatal Division, Department of Pediatrics, Maulana Azad Medical College,
New Delhi 110002, India. Tel: (91)09650784403; E-mail: <starringram@gmail.com>.
Summary

Downloaded from https://academic.oup.com/tropej/article/58/1/50/1686554 by guest on 27 December 2023

This was a randomized blinded placebo controlled trial undertaken to study the role of zinc supplemen-
tation on growth, primarily the linear growth velocity in very low birth weight (VLBW) infants at
3 months corrected age (CA). Out of 134 neonates with birth weight <1500 g, 101 babies were eligible.
Due to lack of consent 10 were excluded. The remaining 91 neonates who were comparable for sex,
gestational age, birth weight, APGAR and age at enrollment were randomized to receive either 1 ml of
zinc sulfate (10 mg elemental zinc) (n ¼ 46) or 1 ml placebo (n ¼ 45) from enrollment to 60 days. The
infants in the zinc group had significantly higher linear growth velocity (0.98  0.12 cm week1) com-
pared to a placebo group (0.67  0.15 cm week1) (p < 0.001) at 3 months CA. We concluded that zinc
supplementation at 10 mg day1 for 60 days in VLBW infants improves their linear growth during
infancy.

Key words: Low birth weight, zinc, growth.

Introduction study was designed to test the hypothesis that zinc

Zinc is one of the most important trace elements for
the body. Zinc plays a major role in various aspects
of physiology, immunity and skeletal growth. Its
many roles include participation in basic metabolic
functions such as cellular respiration, synthesis of
many proteins and enzymes, DNA and RNA repli-
cation, carbohydrate metabolism, cell division and
growth, pituitary and adrenal gland functions, integ-
rity of biological membranes and bone metabolism
[1]. The problem of zinc deficiency in low birth
weight (LBW) infants has been a subject of investi-
gation for over a decade. During the first weeks of
life, there is a progressive decrease in the serum levels
of zinc, accentuated in the small-for-gestational age
(SGA) group [2]. Supplementation studies in young
term, AGA infants have not shown benefits of zinc
supplements, as the zinc content of breast milk is
adequate to meet the needs for first several months
of life. However, in SGA and LBW infants, there
occurs postnatal zinc deficiency due to several fac-
tors, including higher zinc nutritional requirements
to support catch-up growth inadequate intake, im-
paired zinc absorption, increased endogenous losses
associated with organ immaturity, especially the
gastro intestinal tract, increased losses associated
with infections and impaired zinc status at birth
[3, 4]. The exchangeable zinc pool is smaller in
SGA infants. They have smaller liver size, the mag-
nitude of difference being 35–40% [5]. The present
supplement at twice RDA for 60 days would improve
the growth of VLBW (VLBW) neonates at 3 months
corrected gestational age.
Methods
This study was a double-blinded randomized control
trial, conducted in the Neonatal Division,
Department of Pediatrics, Maulana Azad Medical
College and associated Lok Nayak Hospital, New
Delhi, India. The study was approved by the ethical
committee of the institution. Neonates with birth
weight <1500 g on enteral feed of at least
30 ml kg1 day1, whose parents were residents of
Delhi, underwent screening for eligibility. Children
with gross congenital malformations were excluded.
To detect a difference in linear growth velocity of
0.1 cm week1 with SD of 0.07 between the interven-
tion and placebo groups at an alpha error of 5% and
power of 80%, it was estimated that 20 subjects
would be required in each trial arm. Allowing for
an attrition rate of 20%, it was decided to enroll 25
subjects in each group. On obtaining informed writ-
ten consent from parents, the subjects were rando-
mized using computer-generated random number
sequence to intervention and placebo groups (blinded
as A or B). Allocation concealment was done by pla-
cing them in opaque sealed envelopes. The neonates
in one arm received 1 ml of zinc sulfate (10 mg ml1
of elemental zinc in reconstituted solution) i.e. twice

ß The Author [2011]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com 50
doi:10.1093/tropej/fmr036 Advance Access published on 5 May 2011
T. V. RAM KUMAR AND S. RAMJI
RDA, and babies in other arm received 1 ml placebo
each day for 60 days from enrollment. The drug was
available as a white powder in white opaque plastic
container. When sterile water was added up to the
indicator mark, it provided 15 ml of solution contain-
ing 10 mg ml1 of elemental zinc. The placebo (which
did not have zinc as an active ingredient) was also in
a powder form with similar color, taste and packing.
All bottles were labeled either A or B. Each bottle
once reconstituted was meant for use for 3 days and
the remaining being discarded. The drug/placebo was
given by 1 ml disposable syringe by the investigator
initially and later by the mother under investigator’s
supervision, till she became sufficiently proficient to
administer it correctly by herself.
At baseline, the data on maternal and infant char-
acteristics were recorded. They were then followed
weekly for 3 months corrected age (CA). At each
visit, the weight, length, head circumference and tri-
ceps skin-fold thickness were recorded. At discharge
the mother was given a mother booklet (in Hindi)
Subjects screened for eligibility (n=134)
Subjects eligible (n=101)
Randomized (n=91)
Zinc
(n=46)
Not completed trial=10
• Lost to follow up=2
• Died=8
Completed trial
(n=36)
FIG. 1. Flow of the study.
Journal of Tropical Pediatrics Vol. 58, No. 1
and asked to attend weekly or earlier if there was
any problem. The booklet clearly mentioned the
dates of next visit, the date on which she actually
attended and a picture to aid administration of
drug. It also had a sheet where mother marked a
‘tick’ on administration of drug or on opening of a
new bottle. It contained the mobile number of the
investigator to be contacted if the parents felt that
their baby was ill or have any doubt regarding the
drug. Compliance was checked by entries in mother’s
booklet and measuring residual volume of drug in the
bottle at each visit. For those who did not attend
Downloaded from https://academic.oup.com/tropej/article/58/1/50/1686554 by guest on 27 December 2023
follow up on the stipulated date, the parents were
contacted by telephone or post card. Those who
were not traced by either of these were then traced
by social worker by home visit. If no contact was
made even after home visit, the subject was con-
sidered a permanent loss to follow up.
The data were analyzed by Epi Info software. An
intention to treat strategy was used for analyzing the
data. All continuous parametric variables were
Non-eligible=33
• Non-residents=9
• Malformations=7
• Not reached enteral feed of
30ml/kg/day=17
Not randomized=10
• Consent not given=7
• Consent not taken=3
Placebo
(n=45)
Not completed trial=18
• Lost to follow up=4
• Died=14
Completed trial
(n=27)
51
 T. V. RAM KUMAR AND S. RAMJI

compared between the groups by using t-test and all
non-parametric variables by Mann–Whitney test. All
categorical data were analyzed by chi-square/Fisher’s
exact test. A probability of 5% was considered
significant.
Results
The neonates with birth weight <1500 g were
screened for eligibility (n ¼ 134) as depicted in
Fig. 1. Of these, 33 infants (24.6%) were non-eligible
due to reasons mentioned in the Fig.1. Of the 101
were symmetrically growth retarded (ponderal index
>10th centile of reference standards). The proportion
of preterms (<37 weeks) was comparable in both
groups (zinc: 25%; placebo: 26.6%). Table 3 shows
the changes in growth during the period of the study
between the zinc and placebo groups. Weight, length
and triceps skin-fold thickness at 40 weeks post con-
ception age (PCA) showed non-significant trend for
higher values in the zinc group compared to the
placebo group. At 52 weeks PCA, all the growth par-
ameters were significantly higher in the zinc com-
pared to placebo group. The linear growth velocity

Downloaded from https://academic.oup.com/tropej/article/58/1/50/1686554 by guest on 27 December 2023

neonates who were eligible for the study, 10 could
not be randomized due to lack of consent. The re-
maining 91 neonates were randomized to either zinc
(n ¼ 46) or placebo (n ¼ 45). There were 91 eligible
neonates randomized to zinc (n ¼ 46) or placebo
(n ¼ 45) arm. A total of 63 neonates completed the
study (36 in zinc and 27 in placebo arm). The infants
not completing the study were significantly lower
in birth weight and gestation. The baseline param-
eters of mothers in the two groups were compar-
able except for maternal weight (Table 1). The
maternal weight in the placebo group was significant-
ly more (53.9  9.7 kg) compared to zinc group
(50.1  7.9 kg). The neonates randomized to the
zinc and placebo groups were comparable for
gender, birth weight, gestational age, intrauterine
growth status, Apgar scores and age at enrollment
(Table 2). The enrolled neonates were about 1300 g
and 34 weeks gestation at birth. Over two-thirds of
the enrolled subjects were growth retarded (birth
weight <10th centile for gestation of reference
growth standards) and almost half of these infants
in the zinc group was significantly higher
(0.98  0.12 cm week1) compared to the placebo
group (0.67  0.15 cm week1) by an average of
about 0.3 cm week1.
Discussion
The present study demonstrated that VLBW neo-
nates who were exclusively fed with breast milk and
supplemented with zinc for 60 days had significantly
higher weight, length and head size at 52 weeks PCA
compared to those who had received a placebo.
The results of the present study were similar to
those reported by Castillo-Duran et al. [6] in a ran-
domized controlled trial of zinc supplementation
(3 mg day1) among 68 term SGA neonates who
received mixed feeding of breast milk and enriched
supplemented cow’s milk formula over 6 months.
The zinc supplemented groups were observed to
have significantly higher weight and length at
6 months of age. The weight change in these infants
was, however, reported to be also influenced by sex

TABLE 1
Baseline maternal characteristics

Characteristics Zinc (n ¼ 46) Placebo (n ¼ 45) p-value

Age of mother (years) (mean, SD) 24.5  3.2 25.2  3.4 0.3
Parity(median, inter-quartile range) 2 (1–2) 2 (1–3) 0.3
Socio-economic statusa (%)
I 6 (13.6) 5 (11.1) 0.3
II 15 (34.1) 16 (35.6)
III 12 (27.3) 12 (27.3)
IV 11 (25) 11 (25)
V 0 (0) 0 (0)
Antenatal care (3 visits) (%) 26 (60.5) 34 (75.6) 0.2
Hypertension (%) 8 (18.2) 15 (33.3) 0.16
Diabetes (%) 0 (0) 1 (2.2) 0.5
Weight (kg) (mean, SD) 50.1  7.9 53.9  9.7 0.04
Height (cm) (mean, SD) 152.3  4.1 152.9  3.5 0.5
Mode of delivery (%)
Vaginal 38(86.4) 35 (77.8) 0.4
Caesarean 6(13.6) 10 (22.2)
Hemoglobin (g/dl) (mean, SD) 10.2  1.7 10.1  1.7 0.8

ANC ¼ Antenatal care.

a
Kuppuswamy scale (2003).

52 Journal of Tropical Pediatrics Vol. 58, No. 1

T. V. RAM KUMAR AND S. RAMJI

TABLE 2
Baseline infant characteristics

Characteristics Zinc (n ¼ 46) Placebo (n ¼ 45) p-value

Male gender (%) 23 (52.3) 20 (44.4) 0.6

Gestational age (weeks) (mean, SD) 34.8  2.8 34.6  2.9 0.7
Birth weight (g) (mean, SD) 1310  185 1268  197 0.3
APGAR (median, inter-quartile range)
1 min 9 (9–9)(n ¼ 43) 9 (9–9) (n ¼ 43) 0.1
5 min 9 (9–9)(n ¼ 43) 9 (9–9) (n ¼ 44) 0.1
Ponderal index (mean, SD) 2.1  0.2 2.1  0.3 0.6
IUG status (%)

Downloaded from https://academic.oup.com/tropej/article/58/1/50/1686554 by guest on 27 December 2023

AGA (10th–90th centile) 13 (29.5) 13 (28.9) 0.9
SGA (<10th centile) 31 (70.5) 32 (71.1) 0.8
Symmetrical SGA (B.Wt <10th centile and P.I. >10th centile) 15 (48.4) 13 (40.6) 0.7
Age at enrollment (days) (mean, SD) 3.3  1.6 3.7  1.9 0.3
Age when maximal enteral feed achieved (days) (mean, SD) 6.3  1.5 6.4  1.7 0.7

IUG ¼ Intra uterine growth, AGA ¼ Appropriate for gestational age, P.I. ¼ Ponderal index, B.Wt ¼ Birth weight.

TABLE 3
Changes in growth between zinc and placebo groups

Anthropometrical variable Zinc Placebo p-value

Weight (g) (mean, SD)

At enrollment 1310  185 (n ¼ 46) 1268  197 (n ¼ 45) 0.3
At 40 weeks of PCA 1839  348 (n ¼ 38) 1722.7  377 (n ¼ 35) 0.2
At 52 weeks of PCA 4120.1  635 (n ¼ 36) 3404.4  802 (n ¼ 27) 0.0003
Length (cm) (mean, SD)
At enrollment 39.3  1.9 (n ¼ 46) 38.9  1.9 (n ¼ 45) 0.4
At 40 weeks of PCA 43.8  2.4 (n ¼ 38) 42.8  2.3 (n ¼ 35) 0.07
At 52 weeks of PCA 55.9  2.4 (n ¼ 36) 50.7  3.9 (n ¼ 27) <0.001
Head circumference (cm) (mean, SD)
At enrollment 27.5  1.6 (n ¼ 46) 27.4  1.5 (n ¼ 45) 0.6
At 40 weeks of PCA 30.9  1.6 (n ¼ 38) 30.9  2.2 (n ¼ 35) 0.8
At 52 weeks of PCA 37.5  1.7 (n ¼ 36) 36.1  2.2 (n ¼ 27) 0.008
Triceps skin-fold thickness (mm) (mean, SD)
At enrollment 2.5  0.4 (n ¼ 46) 2.6  0.5 (n ¼ 45) 0.3
At 40 weeks of PCA 3.6  0.8 (n ¼ 38) 3.3  0.7 (n ¼ 35) 0.2
At 52 weeks of PCA 7.5  1.4 (n ¼ 35) 5.8  1.6 (n ¼ 27) <0.001
Linear growth velocity (cm week1) (mean, SD) 0.98  0.12 (n ¼ 36) 0.67  0.15 (n ¼ 27) <0.001

and enriched formula and not zinc alone. The study
had also observed that the effect of zinc on linear
growth was noted as early as 30 days after initiation
of zinc supplementation; an observation also noted in
the present study. Two other studies in VLBW in-
fants by Friel et al. [7] and Gomez et al. [8] reported
significant improvement in linear growth among zinc
supplemented infants, but not for weight and head
circumference. The lack of effect on weight and head
circumference in these studies may have been due to
lower dosage of zinc supplementation compared to
the higher (therapeutic) dose used in the present
study.
In contrast, a community-based randomized
double blind trial by Lira et al. [9] which enrolled
full-term LBW infants between 1500 and 2499 g,
observed only a significant difference in weight gain
(mean birth weight 2337  152 g) among the zinc sup-
plemented group at 26 weeks of age. The authors
postulated that the lack of effect of zinc on linear
growth in their study could be due to the low
number (about 27%) of stunted infants (less than
2 SD of length for age), which was about 27%.
However, this argument appears untenable, since a
controlled randomized trial among Ethiopian infants
observed a significant effect of zinc on linear growth
in both stunted and non-stunted infants. This benefit
was postulated to be due to reduction of anorexia,
acute respiratory infection and diarrhea in zinc sup-
plemented infants and the existence of zinc deficiency

Journal of Tropical Pediatrics Vol. 58, No. 1 53


in these children [10]. In yet another community-
based controlled trial [11] among LBW neonates,
zinc supplementation resulted in higher weight and
length at 1 year, but the effects were noted only after
5 months of age. When comparing zinc supplemen-
tation studies among LBW and VLBW neonates, it
appears that the effects of zinc on growth are seen
earlier in VLBW than in LBW neonates.
In conclusion, zinc supplementation (5–10 mg day1)
among LBW neonates, especially VLBW infants, has
significant benefit on their growth, particularly linear
growth. It is suggested that zinc should routinely be
supplemented in VLBW neonates during early in-
fancy to improve their outcomes in growth.
Funding
Drug and placebo were supplied gratis by M/S
Reddy’s Laboratory, Hyderabad, India.
References
1. Prasad AS. Clinical, biochemical and pharmacological
role of zinc. Ann Rev Pharmacol Toxicol 1979;19:
393–426.
2. Doménech E, Dı́az-Gómez NM, Barroso F,
Cortabarria C. Zinc and perinatal growth. Early Hum
Dev 2001;65:S111–7.
3. Hambidge M. Human zinc deficiency. J Nutr 2000;130:
1344s–9s.
54
T. V. RAM KUMAR AND S. RAMJI
4. Higashi A, Matsuda I, Masumoto T, et al. Serum zinc
and copper concentrations in low birth weight infants
during first three months of life: correlation to birth
weight and different feedings. Tohuku J Exp Med
1985;146:253–63.
5. Krebs N, Westcott J, Rodden D, et al. Exchangeable
zinc pool size at birth is smaller in small-for-gestation
age than in appropriate-for-gestational age preterm in-
fants. Am J Clin Nutr 2006;84:1340–3.
6. Castiio-Duran C, Rodriguez A, Venegas G, et al. Zinc
supplementation and growth of infants born small for
gestational age. J Pediatr 1995;127:206–11.
7. Friel J, Andrews W, Matthew J, et al. Zinc supplemen-
Downloaded from https://academic.oup.com/tropej/article/58/1/50/1686554 by guest on 27 December 2023
tation in very low birth weight infants. J Pediatr
Gastroenterol Nutr 1993;17:97–104.
8. Gomez N, Doménech E, Barroso N, et al. The effect of
zinc supplementation on linear growth, body compos-
ition, and growth factors in preterm infants. Pedaiatr
2003;111:1002–9.
9. Lira P, Ashworth A, Morris S. Effect of zinc supple-
mentation on the morbidity, immune function, and
growth of low birth weight, full-term infants in north-
east Brazil. Am J Clin Nutr 1998;68(Suppl):418s–24s.
10. Gibson RS. Zinc supplementation for infants. Lancet
2000;355:2008–9.
11. Sur D, Gupta D, Mondal S, et al. Impact of zinc sup-
plementation on diarrheal morbidity and growth pat-
tern of low birth weight infants in Kolkata, India.
Pediatr 2003;112:1327–32.
Journal of Tropical Pediatrics Vol. 58, No. 1