Vous êtes sur la page 1sur 8

Efficacy of zinc given as an adjunct in the treatment of severe and very

severe pneumonia in hospitalized children 224 mo of age: a


randomized, double-blind, placebo-controlled trial13
Nitya Wadhwa, Aruna Chandran, Satinder Aneja, Rakesh Lodha, Sushil K Kabra, Mona K Chaturvedi, Jitender Sodhi,
Sean P Fitzwater, Jagdish Chandra, Bimbadhar Rath, Udaypal S Kainth, Savita Saini, Robert E Black, Mathuram Santosham,
and Shinjini Bhatnagar

INTRODUCTION

Pneumonia is a leading cause of death in young children


and causes w18% of all deaths in children ,5 y of age (1). In
India, which has the single highest burden of pneumonia
globally, an estimated 370,000 children die of pneumonia each
year. Community-based standardized case management has been
a hallmark of strategies to reduce pneumonia mortality; it is
estimated that such an approach may reduce the mortality rate
by #70% (2).
Zinc deficiency has been well characterized in Indian children
and other low- and middle-income countries (35). Zinc plays an

important role in the immune system; it is essential for the


normal function and development of innate and adaptive immune systems (6). In clinical trials, zinc supplementation has
provided a cost-effective approach to decreasing morbidity,
mortality, and the economic burden associated with childhood
diseases in low- and middle-income countries. A recent metaanalysis showed a 19% reduction in pneumonia morbidity and
13% reduction in incidence of diarrhea in children who received
preventative zinc supplementation (7). Zinc is currently recommended by the WHO as an adjunct therapy for treating diarrhea (8, 9).
However, zincs therapeutic effect on respiratory infections is
less clear. Studies that examined the clinical effects of zinc for
treating pneumonia in children have shown conflicting results,
with some studies that showed a beneficial effect on the duration
of recovery and severity (1012), whereas other studies suggest
that zinc has no treatment benefit (13, 14). Additional information is needed to determine whether zinc has a role in the
treatment of children hospitalized for pneumonia before policy
recommendations can be made. In addition, the earlier studies
did not evaluate the effect of zinc in a subgroup of children with
very severe pneumonia. It is possible that zinc supplementation
only has an effect on very severe pneumonia but not on milder

From the Centre for Diarrheal Diseases and Nutrition Research, Department of Pediatrics (NW, MKC, JS, SS, and SB) and the Department of
Pediatrics (RL and SKK), All India Institute of Medical Sciences, New
Delhi, India; the Pediatric Biology Centre, Translational Health Science
and Technology Institute, Gurgaon, Haryana, India (NW and SB); the Departments of International Health (AC, SPF, REB, and MS) and Pediatrics
(AC and MS) and the Center for American Indian Health (MS), Johns
Hopkins Bloomberg School of Public Health, Baltimore, MD; the Department of Pediatrics, Lady Hardinge Medical College and Kalawati Saran
Childrens Hospital, New Delhi, India (SA, JC, and BR); and the Department
of Pediatrics, Deen Dayal Upadhyay Hospital, New Delhi, India (USK).
2
Cofunded by the WHO and Johns Hopkins Bloomberg School of Public
Health (to SB).
3
Address correspondence to S Bhatnagar, Pediatric Biology Centre, Translational Health Science and Technology Institute (an autonomous institute of
Department of Biotechnology, Government of India), Gurgaon, Haryana
122016, India. E-mail: shinjini.bhatnagar@thsti.res.in.
Received October 25, 2012. Accepted for publication March 25, 2013.
First published online May 1, 2013; doi: 10.3945/ajcn.112.052951.

Am J Clin Nutr 2013;97:138794. Printed in USA. ! 2013 American Society for Nutrition

1387

Downloaded from ajcn.nutrition.org by guest on December 7, 2016

ABSTRACT
Background: Pneumonia is a leading cause of death; in India, an
estimated 370,000 children die of pneumonia each year. Zinc has
multiple influences on the immune response to infections. Zinc
supplementation has been shown to prevent diarrhea and pneumonia
in children. However, zincs therapeutic effect on respiratory infections is less clear.
Objective: We evaluated the role of zinc as an adjunct to antibiotics
in the treatment of children hospitalized for severe or very severe
pneumonia.
Design: In this randomized, double-blind, placebo-controlled trial,
we enrolled 550 children aged 224 mo with severe or very severe
pneumonia. Within each hospital and pneumonia-severity stratum,
children were randomly assigned to receive zinc (20 mg elemental
zinc/d) or a placebo in addition to antibiotics and supportive care.
Results: The time to recovery from severe or very severe pneumonia
was similar in both groups (HR: 0.98; 95% CI: 0.82, 1.17). In the
stratified analysis, zinc was shown to be efficacious in reducing the
time to recovery in children with very severe pneumonia (HR: 1.52;
95% CI: 1.03, 2.23); however, the effect was no longer statistically
significant after adjustment for differences in severely underweight
children in the 2 groups.
Conclusions: This study showed no overall benefit of the addition
of zinc to antibiotics in reducing the time to recovery from pneumonia but showed a possible benefit of zinc supplementation in a subgroup of children with very severe pneumonia. Additional research is
needed in specific subgroups such as children with very severe pneumonia. This trial was registered at http://www.controlled-trials.
com as ISRCTN48954234.
Am J Clin Nutr 2013;97:1387
94.

1388

WADHWA ET AL

forms of pneumonia. We conducted a randomized, double-blind,


controlled trial to evaluate the efficacy of daily orally administered zinc given in addition to standard antimicrobial therapy in
the treatment of severe and very severe pneumonia.
SUBJECTS AND METHODS

Study design
This randomized, double-blind, placebo-controlled trial was
undertaken from February 2007 to March 2010 at the following
3 tertiary hospitals in New Delhi, India: the Kalawati Saran
Childrens Hospital, the Deen Dayal Upadhyay Hospital, and the
All India Institute of Medical Sciences. The study protocol
was approved by the respective institutional ethics committees
as well as the Ethics Review Committee of WHO, the Institutional
Review Board of the Johns Hopkins Bloomberg School of
Public Health, and the Health Ministry Screening Committee of
the Ministry of Health And Family Welfare, Government of
India.

We screened children aged 2 (60 d)24 mo who presented to


the emergency department of study hospitals with a cough for
,30 d or difficult breathing for #7 d for clinical symptoms and
signs of severe or very severe pneumonia. Definitions of severe
pneumonia and very severe pneumonia were adapted from the
WHO criteria (15). Children were categorized as having severe
pneumonia if they had fast breathing ($50 breaths/min in
children ,12 mo of age or $40 breaths/min in children $12 mo
of age), chest indrawing, and crepitations on chest auscultation.
Children who had signs of pneumonia (fast breathing and
crepitations with or without chest indrawing) and cyanosis or
any general danger sign (lethargy, inability to drink, or convulsions) were categorized as having very severe pneumonia.
Children were excluded if they presented with a need for
mechanical ventilation or inotropic medications, had major
congenital anomalies, inborn errors of metabolism, chronic
disorders such as renal failure, preexisting seizure disorders,
surgical or other conditions that interfered with oral feeding,
HIV infection, were born to mothers with documented HIV infection, had active measles, severe malnutrition that required
separate medical attention, or had any other serious underlying
medical condition. Children who had received intravenous
antimicrobials .48 h for the current illness or zinc supplementation in the past 3 mo were also excluded. Children who
presented with clinical signs of wheezing were treated with 3
nebulizations of salbutamol at 20-min intervals. At the end of
this therapy, if the signs of severe or very severe pneumonia
persisted irrespective of wheeze, the child was eligible for enrollment. Study physicians obtained written (or a thumb print
from individuals who were illiterate) informed consent from
the parent or guardian of eligible patients in the presence of a
witness.
Random assignment, masking, and intervention
Immediately after informed consent was obtained, children
were stratified by pneumonia severity (severe or very severe
pneumonia, as previously defined). A scientist, who was other-

Clinical monitoring and antibiotic therapy


Study physicians examined patients for signs of severe and
very severe pneumonia and other relevant clinical features every
6 h (or more often if clinically indicated) until recovery or day 14.
The respiratory rate count was repeated if it was $50 breaths/min
in children ,12 mo of age or $40 breaths/min in children $12
mo of age and recorded. If the discrepancy between the 2 counts
was .5 breaths/min, a third count was done, and the average of
the 2 closest counts was recorded as the respiratory rate. Study
staff recorded nude weights at random assignment and every
24 h until recovery or day 14. The site supervisor, who was a
pediatrician, supervised the monitoring and was responsible for
overseeing at least one daily assessment done by each study
physician.
Children were treated according to a standardized protocol
adapted from the Indian Academy of Pediatrics guidelines (16) in
accordance with the standard of care that was being provided at
the hospitals. Oxygen saturation was measured by using a pulse
oximeter at admission and every 6 h thereafter. Oxygen was
administered to children with a saturation ,90%. Breastfeeding
was continued in children who were able to feed. Children with
severe pneumonia who came directly to our study sites were
treated with recommended doses of intravenous ampicillin and
an aminoglycoside (intravenous gentamicin or amikacin).
Children with very severe pneumonia were given a thirdgeneration cephalosporin. Children who had been initiated on
antibiotics before presenting to study sites were given a thirdgeneration cephalosporin if diagnosed with severe pneumonia
and vancomycin along with ceftriaxone-sulbactam or piperacillintazobactam if diagnosed with very severe pneumonia. Antistaphylococcal antibiotics were administered if there was clinical
suspicion of staphylococcal infection. Children who had not recovered by the end of 14 d were no longer monitored by the study
staff but were managed by hospital physicians.

Downloaded from ajcn.nutrition.org by guest on December 7, 2016

Patient eligibility and assessment

wise not involved in the study, generated allocation sequences


with STATA software (version 9.0; StataCorp) so that, within
each hospital and pneumonia severity, patients were randomly
assigned equally in permuted blocks of 6 to receive zinc or a
placebo. Dispersible tablets that were identical in color, taste, and
appearance and packed in identical looking blister strips with or
without zinc sulfate (10 mg elemental zinc; Nutriset) were
provided by the WHO. Each strip was labeled with a unique serial
number according to the randomization list. A zip-lock bag that
held 3 such blister strips coded with the same unique serial
number was assigned for each subject. Information regarding
random assignment was not available to any of the investigators
until the data had been obtained, entered, and locked. Participants
and investigators were masked to treatment allocation. We
maintained masking during data analysis by coding the treatment
allocation with 2 letters.
After random assignment, study physicians gave each child
one tablet of zinc (10 mg elemental zinc) or a placebo dissolved in
3 mL distilled water every 12 h until recovery (ie, the end of a
24-h period without any clinical signs of severe or very severe
pneumonia) or the completion of 14 d, whichever was earlier.
Administration was repeated #3 times in children who vomited
within 15 min of dose administration. The day of randomization
and initial dose administration was designated as day 1.

ZINC FOR PNEUMONIA TREATMENT

Clinical outcomes
The primary outcome was the time to recovery, which was
calculated as the time from random assignment until recovery.
Recovery was defined as the end of a 24-h period without any
clinical signs of severe or very severe pneumonia (as previously
defined).
The secondary outcome was treatment failure. Treatment failure
was a composite outcome, which was defined as a need to change
antimicrobial therapy in #7 d of random assignment or a need for
intensive care (mechanical ventilation or vasoactive drug infusion
or both) or withholding of the intervention for worsening in
clinical condition or death at any time within 14 d of random assignment, or a failure to recover by day 14. The decision to change
antimicrobials was made by 2 senior pediatricians (site investigators)
on the basis of a priori guidelines stated in the study protocol
(ie, persistence of chest indrawing or any initially identified danger
sign $48 h of random assignment or worsening of clinical features
that defined severe or very severe pneumonia at any time after
random assignment).

1389

met every 3 mo to ensure that the study was done according to


the protocol.
Sample-size estimation

Laboratory procedures

Data management and statistical analysis

Study physicians obtained blood specimens at random assignment, 48 h, and recovery. Hemoglobin, total and differential
leukocyte counts, a blood culture, serum zinc, and C-reactive
protein (CRP) were determined at random assignment at each
clinical site. Chest X-rays were done in the emergency room.
Blood cultures were processed in the microbiology laboratory at
the All India Institute of Medical Sciences by using the fully
automated microbiology growth and detection system, BACTEC
(model PEDS PLUS/F; Becton Dickinson), and antibiotic sensitivity was determined using the Kirby-Bauer disc diffusion
technique (17). Serum for the zinc concentration was obtained at
48 h and recovery, whereas serum was taken at recovery for the
CRP concentration. The serum was stored at 2208C. The laboratory at the All India Institute of Medical Sciences measured
serum concentrations of zinc with a flame furnace atomic absorption spectrophotometer (GBC Avanta) as per standard procedures (18) and CRP concentrations by using a commercial
ELISA (Biocheck).

Site supervisors checked all completed case-report forms


twice before they were sent for interactive double data with
Microsoft Access software (version 2007; Microsoft Corp) with
in-built logic, range, and consistency checks. With STATA
software (version 11.0; StataCorp), all analyses were done on an
intention-to-treat basis. We used time-to-event analyses for the
primary outcome that censored children in whom recovery could
not be documented because of withdrawal of consent for continuation in the study. Subjects who did not recover by day 14
(336 h) were censored at that time. Children who died during the
study period of 14 d were censored at 337 h (ie, 1 h longer than the
maximum time of observation). Irrespective of whether treatment
failure had occurred, a Cox proportional hazards regression
model compared the time until recovery between study groups;
an HR .1 indicated a beneficial effect of zinc. We calculated the
proportion of treatment failures in the 2 trial arms and the corresponding RR and risk difference. We compared the change in
serum zinc concentration from baseline to recovery between the
2 intervention groups by using Students t test. Serum zinc at
recovery was compared between the 2 groups by using the
Mann-Whitney U test.
We performed a predefined subgroup analysis on severity
groups by using time-to-event analyses to determine whether the
baseline severity of pneumonia modified the effect of treatment
with zinc. Interactions were assessed in Cox-proportional hazard
models to determine whether the severity of pneumonia modified
the effects of treatment with zinc.

Quality assurance and monitoring of adverse serious events


Before study initiation, study procedures were standardized to
consistently identify primary and secondary outcomes. Site supervisors (trained pediatricians) received standard training with
print and video material on patient assessment and outcomes
according to a standard protocol to ensure uniformity across
study sites. Regular standardization exercises for clinical outcomes
and laboratory procedures were conducted among the research
team members to minimize intraobserver and interobserver
variability within and across the 3 hospitals. Particular emphasis was given to counting the respiratory rate, identifying
chest indrawing, and auscultating the chest for crepitations. An
independent data safety monitoring board was constituted before
initiation of enrollment; the board was comprised of a biostatistician and 3 clinicians experienced in research who were
not affiliated with study hospitals. The board reviewed each
serious adverse event (death or need for intensive care) and

RESULTS

Enrolled patients
A total of 10,066 children between the ages 2 and 24 mo were
identified with difficult breathing for #7 d or cough for ,30 d or
both. Of these children, 3164 subjects had signs of severe or
very severe pneumonia. On additional evaluation, 2284 children
were not eligible for the study for reasons described in Figure 1.
The remaining 880 children were approached for enrollment

Downloaded from ajcn.nutrition.org by guest on December 7, 2016

On the basis of unpublished data from one of the participating


hospitals, the total time (6SD) taken to recover from severe
pneumonia or very severe pneumonia in children ,2 y old and
receiving standard care was estimated at 96 6 48 h and 144 6
60 h, respectively. With the use of this assumption, a total
sample size of 492 children [292 children with severe pneumonia (146 children in each group) and 200 children with very
severe pneumonia (100 children in each group)] was calculated
to detect a 20% reduction in the total time taken to recover from
severe or very severe pneumonia by administering zinc. With
assumption of a 10% loss to follow-up, this sample size provided a power of 90% and 95% confidence. Because of the
difficulty of enrolling children in the very severe pneumonia
category, the total enrollment limit was increased from 492 to
550 to recruit children in the very severe pneumonia stratum.

1390

WADHWA ET AL

(Figure 1), of which consent was refused for 330 children. Of


the 550 enrolled patients, 274 and 276 children were assigned
to receive zinc and a placebo, respectively. Overall, 4.9% of
children (zinc: 12 subjects; placebo: 15 subjects) were lost to
follow-up, 1.5% of children (zinc: 4 subjects; placebo: 4 subjects) died during the course of the study, and 2.18% of children
(zinc: 6 subjects; placebo: 6 subjects) did not recover by day 14.
There was one protocol deviate in the placebo group who inadvertently received one dose that was assigned for another
patient and also received antibiotics that were different from
what was defined in the protocol.
The 2 groups had comparable baseline characteristics (Table 1);
82.4% of children were ,12 mo of age, and 67.3% of children
were boys. Nearly one-quarter (22.9%) of children had a weightfor-age z score less than 23, and a large number (75%) of children
were anemic (hemoglobin concentration ,11 g/dL). A total of
430 enrolled children (78.2%) were diagnosed with severe
pneumonia, whereas 120 enrolled children (21.8%) were categorized as having very severe pneumonia. The median duration of
symptoms before admission into the study was 3.2 d (IQR: 3, 5 d).
Illness severity, as determined by clinical characteristics and
laboratory markers of inflammation, was similar in the 2 groups.
Median (IQR) concentration of CRP was 9.41 mg/L (2.32, 24.51
mg/L) in the zinc group and 8.46 mg/L (2.17, 19.64 mg/L) in the
placebo group. The baseline serum zinc concentration was
similar in both zinc and placebo groups at 9.3 and 9.2 mmol/L,
respectively. More than 50% of the patients had serum zinc
concentrations less than 9.2 mmol/L (zinc: 55.4%; placebo:
55.4%).

Clinical outcomes
The time until recovery from severe or very severe pneumonia
was similar in both groups [median (IQR): zinc, 78.5 h (59, 122 h);
placebo, 77.0 h (59, 117 h); HR: 0.98; 95% CI: 0.82, 1.17)]
(Table 2, Figure 2). Treatment failure could be assessed in 525
children (Table 2). Of the 27 children who were lost to followup, 2 children (given placebo) had the outcome of treatment
failure before they withdrew consent for continuation in the study.
Risk of treatment failure was similar in the 2 study groups
(Table 2). We did a post hoc analysis and showed no significant difference between the proportion of children who recovered by day 2 or 5 (P = 0.63 and P = 0.97, respectively).
The case fatality did not differ between the 2 intervention groups
(Table 2). Risk of vomiting the first dose of the study intervention
within 15 min of receiving it was similar in the 2 groups [zinc: 9
children (3.28%); placebo: 7 children (2.54%); RR: 1.3; 95% CI:
0.49, 3.43)].
The mean (6SD) change in serum zinc concentrations from
enrollment to recovery was higher in the children given zinc
(4.30 6 5.6 mmol/L) than in children who received the placebo
(1.06 6 3.9 mmol/L), with a difference of 3.24 mmol/L (95% CI:
2.4, 4.1 mmol/L; P , 0.0001). The effect of zinc on laboratory
markers of inflammation (CRP) did not differ between the 2
groups.
In the stratified analysis, zinc was shown to be efficacious in
reducing the time to recovery in children with very severe
pneumonia [median (IQR): zinc, 80 h (63, 114 h); placebo, 106 h
[66, 177 h)] (Table 3) but not in children with severe pneumonia
[median (IQR): zinc, 78 h (57, 122 h); placebo, 73 h (54, 106 h)]

Downloaded from ajcn.nutrition.org by guest on December 7, 2016

FIGURE 1. Enrollment and random assignment profile.

1391

ZINC FOR PNEUMONIA TREATMENT


TABLE 1
Baseline characteristics of the enrolled children in the zinc and placebo groups
Zinc (n = 274)

Aged $2 to ,12 mo [n (%)]


Age (mo)
Sex (F) [n (%)]
Breastfed [n (%)]
Severely underweight [n (%)]2
Mothers education more than primary [n (%)]
Antimicrobial therapy before admission y [n (%)]
Hours of symptoms before admission
Enrollment category [n (%)]
Severe pneumonia
Very severe pneumonia
Clinical characteristics
Respiratory rate
Chest indrawing [n (%)]
Wheezing persisted after nebulization [n (%)]
Fever [n (%)]4
Laboratory findings
Oxygen saturation ,90% [n (%)]
Hemoglobin ,11 g/dL [n (%)]
Leukocytosis [n (%)]
CRP5 (mg/L)
Positive blood culture [n (%)]
Serum zinc concentrations (mmol/L)6
,9.2 mmol/L [n (%)]

221
5.5
98
235
58
81
74
96

Placebo (n = 276)

(80.7)
(3, 10)1
(35.8)
(85.8)
(21.2)
(29.6)
(27.0)
(72, 120)

232
5
82
240
68
68
63
72

215 (78.5)
59 (21.5)

(84.1)
(3, 10)
(29.7)
(87.0)
(24.6)
(24.6)
(22.8)
(72, 120)

215 (77.9)
61 (22.1)

61.8
271
150
93

6 8.53
(98.9)
(54.7)
(33.9)

62.0
274
151
105

6 7.4
(99.3)
(54.7)
(38.0)

11
207
34
9.41
16
9.3
149

(4.0)
(75.5)
(12.4)
(2.32, 24.51)
(5.8)
6 3.9
(55.4)

8
205
32
8.46
20
9.2
149

(2.9)
(74.3)
(11.6)
(2.17, 19.64)
(7.2)
6 3.6
(55.4)

Median; IQR in parentheses (all such values).


Weight-for-age z score less than 23.
3
Mean 6 SD (all such values).
4
Fever axillary temperature $37.58C
5
CRP, C-reactive protein.
6
Zinc concentrations were available for 269 patients in the zinc group and 269 patients in the placebo group.
2

(P-interaction = 0.010). In the very severe pneumonia group, the


reduction in the time to recovery was also evident in the hazards
model (HR: 1.52; 95% CI: 1.03, 2.23) (Figure 3). When adjusted for prespecified potential confounders such as the weightfor-age z score less than 23 (severely underweight) status, age,
breastfeeding status, baseline serum CRP, and serum zinc, the
effect of zinc on the time to recovery in subjects with very se-

vere pneumonia (HR: 1.54; 95% CI: 1.02, 2.33) continued to be


significant. Of note, the proportion of children with a weight-forage z score less than 23 (severely underweight) in the very
severe pneumonia group was lower in zinc recipients than in the
placebo group (15.3% compared with 32.8%, respectively).
Therefore, we also did an exploratory time-to-event analysis
in which we adjusted for only severely underweight; in this

TABLE 2
Effect of oral zinc on clinical outcomes in children with severe or very severe pneumonia
Outcomes

Zinc
1

Time to recovery (h)


Recovered by 2 d [n (%)]
Recovered by 5 d [n (%)]
Treatment failure [n (%)]4
Any failure5
Death
Need for intensive care
Not recovered by 14 d
Change of antibiotics

Placebo
2

78.5 (59, 122)


33 (12.0)
194 (70.8)
37
4
0
2
31

(14.1)
(1.5)
(0)
(0.8)
(11.8)

Values
2

77.0 (58, 117)


37 (13.4)
195 (70.7)
28
4
1
1
22

(10.6)
(1.5)
(0.4)
(0.4)
(8.4)

0.98 (0.82, 1.17)3

1.3 (0.8, 2.1)6


1.0 (0.3, 4.0)6

2.0 (0.2, 22.0)6


1.4 (0.8, 2.4)6

1
Number of patients in whom the primary outcome could be assessed was 274 in the zinc group and 276 in the placebo
group.
2
Median; IQR in parentheses
3
HR; 95% CI in parentheses. Time-to-event analyses by using a Cox proportional hazards regression model.
4
Number of patients in whom treatment failure could be assessed was 262 in the zinc group and 263 in the placebo
group.
5
Absolute risk reduction was 3.5% (95% CI: 22.2%, 9.1%).
6
RR; 95% CI in parentheses.

Downloaded from ajcn.nutrition.org by guest on December 7, 2016

Characteristics

1392

WADHWA ET AL

FIGURE 2. Kaplan-Meier survival estimates for the time to recovery in


patients with severe or very severe pneumonia in the zinc (n = 274)
compared with placebo (n = 276) groups. Median (IQR) time to recovery:
zinc, 78.5 h (59, 122 h); placebo, 77.0 h (58, 117 h). HR (95% CI): 0.98
(0.82, 1.17); P = 0.838 (time-to-event analyses by using a Cox proportional
hazards regression model).

DISCUSSION

This study showed that there was no overall benefit of zinc


supplementation in reducing the time to recovery in children with
pneumonia. Risk of treatment failure was similar in zinc and
placebo groups. We observed a significant effect of zinc supplementation in reducing the time to recovery in children with
very severe pneumonia; however, this effect was attenuated when
we controlled for severely underweight. Nevertheless, when we
adjusted the analysis for effect of zinc supplementation on time to
recovery controlling for severely underweight, the positive effect
of zinc supplementation continued, which suggested that our
sample size may not have been large enough to document this
effect.
Overall results of our study were in line with the opinion
expressed in a recent review of zinc for treatment of pneumonia in
children (19). Studies that examined the treatment of pneumonia
with zinc have shown mixed results. Two studies, one in Iran
and one in Bangladesh, have shown treatment benefits of zinc
(10, 11), which has not been reported from other studies done in
different regions of the Indian subcontinent such as Kolkata,
Vellore, Chandigarh, and Nepal (13, 14, 2022). Some beneficial effects have been seen in certain subgroups; the trial based
in Kolkata showed significant effect in male subjects (12),
whereas the trial in Vellore showed a seasonal benefit of zinc
treatment (14).
Several studies have inferred that the benefits of treating
pneumonia with zinc are seen in bacterial but not viral infections.
In Bangladesh, a greater effect of treatment with zinc was seen in
children who presented without wheeze, which the authors
suggested may have been due to a limited therapeutic benefit in
children who presented with viral pneumonia (11). Similarly,
Bose et al (14) hypothesized that the seasonal effect seen in
Vellore may have been due to benefits seen in pneumonia of
bacterial origin, which are thought to be more common in the hot
season. However, to our knowledge, no studies have assessed

TABLE 3
Effect of oral zinc on clinical outcomes in children with very severe
pneumonia
Outcomes

Zinc
1

Time to recovery (h)


Recovered by 2 d [n (%)]
Recovered by 5 d [n (%)]
Treatment failure [n (%)]4
Any failure
Death
Need for intensive care
Not recovered by 14 d
Change of antibiotics

Placebo
2

Values

80 (63, 114) 106 (66, 177) 1.52 (1.03, 2.23)3


4 (6.8)
5 (8.2)

43 (72.9)
31 (50.8)

5
1
0
0
4

(8.8)
(1.8)
(0)
(0)
(7.0)

8
2
1
0
5

(14.0)
(3.5)
(1.8)
(0)
(8.8)

0.63 (0.2, 1.8)5


0.50 (0.05, 5.4)5

0.80 (0.2, 2.8)5

1
Number of patients in whom the primary outcome could be assessed
was 59 in the zinc group and 61 in the placebo group.
2
Median; IQR in parentheses.
3
HR; 95% CI in parentheses. Time-to-event analyses by using a Cox
proportional hazards regression model.
4
Number of patients in whom treatment failure could be assessed was
57 in the zinc group and 57 in the placebo group.
5
RR; 95% CI in parentheses.

Downloaded from ajcn.nutrition.org by guest on December 7, 2016

analysis, the effect continued in the same direction but was not
significant (HR: 1.35; 95% CI: 0.92, 2.00). In the very severe
pneumonia group, there was also a trend toward lower treatment
failure in children who received zinc.

pneumonia cause directly by using microbiological techniques.


The prevalence of wheezing in the earlier studies has ranged from
37.4% to 62.5% (11, 14, 21, 22), whereas we observed wheezing
in 54.7% if subjects. Some of these studies have reported a higher
prevalence of hypoxemia at enrollment (11, 14, 22), which may
have been due to differences in the oxygen-saturation cutoff used,
altitude of the study site, and proportion of children with
wheezing. In the current study, no difference in treatment outcomes were seen when patients were separated into categories
based on sex, hot and cool or rainy and dry seasons, or the
presence of wheezing (data not shown).
The significant benefits seen with zinc in treating very severe
pneumonia could be important in the background of other recent
studies that have shown beneficial effects of zinc therapy in
children with more-severe infection. A trial conducted in our
setting that looked at the treatment benefit of zinc in young
infants admitted with probable serious bacterial infection showed
a 40% reduction in the treatment failure rate in zinc recipients
(23). In addition, a study from Nepal reported that the subgroup of
children with endpoint consolidation seen in chest radiographs
had a significantly faster recovery when they received zinc compared with a placebo (22). A recent study from Uganda showed
a significant reduction in case fatality with zinc supplementation in
the subgroup of HIV-infected children with severe pneumonia (24).
The case fatality for very severe pneumonia can range from 10%
to 50% even with hospitalization (25, 26). Therefore, effective
adjunct-treatment options could be important (27). Although our
findings could have been due to a higher rate of severe underweight
in children with very severe pneumonia, these effects of zinc may
be significant in extremely ill children and, therefore, need to be
evaluated further.
It has been hypothesized that baseline zinc deficiency may
alter the outcome of zinc treatment. In Bangladesh, where a clear
benefit to treatment with zinc was seen, the baseline serum zinc
concentration was 10.1 mmol/L, whereas in Vellore and Kolkata,
which showed effects only in subgroup analyses, baseline concentrations were 11.0 and 9.6 mmol/L, respectively (11, 12, 14).
In the Chandigarh study, which saw no effect, children had

ZINC FOR PNEUMONIA TREATMENT

FIGURE 3. Kaplan-Meier survival estimates for the time to recovery in


patients with very severe pneumonia in the zinc (n = 59) compared with
placebo (n = 61) groups. Median (IQR) time to recovery: zinc, 80 h (63, 114 h);
placebo, 106 h (66, 177 h). HR (95% CI): 1.52 (1.03, 2.23); P = 0.034 (time-toevent analyses by using a Cox proportional hazards regression model).

We thank the medical and nursing staff of the 3 hospitals for their contribution as well as the patients and their parents. We also thank Mohini Kumari,
Ritu Chawla, Vinod Kumar Sharma, Ashok Kumar Dutta, Arvind Bagga, and
Madhulika Kabra for providing clinical support to patients enrolled in the
study at the Deen Dayal Upadhyay Hospital, Kalawati Saran Childrens Hospital, and All India Institute of Medical Sciences, respectively. We are thankful

to Arti Kapil for supervising blood cultures and Uma Chandra Mouli Natchu for
helping us with data analysis and interpretation of data. We acknowledge the late
Anil Kumar Sharma for assisting in quality assurance and supervision of the
research staff, Mukesh Juyal for secretarial support, Shilpa Chopra for
data-management support, and Dharmendra Sharma for data-management
support and analysis of data. We are thankful to Olivier Fontaine, medical
officer, Child and Adolescent Health at the WHO, for providing the intervention
(zinc and placebo tablets from Nutriset) and helping with the design and periodic
reviews of the conduct of the study.
The authors responsibilities were as followsSB, AC, REB, and MS:
designed the research; NW, SA, RL, SKK, MKC, JS, JC, BR, USK, SS, and
SB: conducted the research; NW, RL, SPF, and SB: analyzed data; NW, AC,
SA, RL, and SB: wrote the manuscript; NW and SB: had primary responsibility for the final content of the manuscript; and all authors: read and
approved the final manuscript. None of the authors had a conflict of interest.

REFERENCES
1. Black RE, Cousens S, Johnson HL, Lawn JE, Rudan I, Bassani DG, Jha
P, Campbell H, Walker CF, Cibulskis R, et al. Global, regional, and
national causes of child mortality in 2008: a systematic analysis.
Lancet 2010;375:196987.
2. Theodoratou E, Al-Jilaihawi S, Woodward F, Ferguson J, Jhass A,
Balliet M, Kolcic I, Sadruddin S, Duke T, Rudan I, et al. The effect of
case management on childhood pneumonia mortality in developing
countries. Int J Epidemiol 2010;39(suppl1):i15571.
3. Brundtland GH. Nutrition and infection: malnutrition and mortality in
public health. Nutr Rev 2000;58:S14; discussion S6373.
4. Kapil U, Jain K. Magnitude of zinc deficiency amongst under five
children in India. Indian J Pediatr 2011;78:106972.
5. Dhingra U, Hiremath G, Menon VP, Dhingra P, Sarkar A, Sazawal S.
Zinc deficiency: descriptive epidemiology and morbidity among preschool children in peri-urban population in Delhi, India. J Health Popul
Nutr 2009;27:6329.
6. Shankar AH, Prasad AS. Zinc and immune function: the biological basis of
altered resistance to infection. Am J Clin Nutr 1998;68(2 suppl):447S63S.
7. Yakoob MY, Theodoratou E, Jabeen A, Imdad A, Eisele TP, Ferguson J,
Jhass A, Rudan I, Campbell H, Black RE, et al. Preventive zinc supplementation in developing countries: impact on mortality and morbidity due to
diarrhea, pneumonia and malaria. BMC Public Health 2011;11(suppl 3):S23.
8. WHO. Clinical management of acute diarrhoea. In: WHO/UNICEF, ed.
WHO/UNICEF joint statement. New York; 2004. Available from:
http://whqlibdoc.who.int/hq/2004/WHO_FCH_CAH_04.7.pdf (cited
15 October 2012).
9. Santosham M, Chandran A, Fitzwater S, Fischer-Walker C, Baqui AH,
Black R. Progress and barriers for the control of diarrhoeal disease.
Lancet 2010;376:637.
10. Valavi E, Hakimzadeh M, Shamsizadeh A, Aminzadeh M, Alghasi A.
The efficacy of zinc supplementation on outcome of children with
severe pneumonia. A randomized double-blind placebo-controlled
clinical trial. Indian J Pediatr 2011;78:107984.
11. Brooks WA, Yunus M, Santosham M, Wahed MA, Nahar K, Yeasmin
S, Black RE. Zinc for severe pneumonia in very young children:
double-blind placebo-controlled trial. Lancet 2004;363:16838.
12. Mahalanabis D, Lahiri M, Paul D, Gupta S, Gupta A, Wahed MA,
Khaled MA. Randomized, double-blind, placebo-controlled clinical
trial of the efficacy of treatment with zinc or vitamin A in infants and
young children with severe acute lower respiratory infection. Am J Clin
Nutr 2004;79:4306.
13. Ganguly A, Chakraborty S, Datta K, Hazra A, Datta S, Chakraborty J. A
randomized controlled trial of oral zinc in acute pneumonia in children
aged between 2 months to 5 years. Indian J Pediatr 2011;78:108590.
14. Bose A, Coles CL, Gunavathi, John H, Moses P, Raghupathy P, Kirubakaran
C, Black RE, Brooks WA, et al. Efficacy of zinc in the treatment of severe
pneumonia in hospitalized children ,2 y old. Am J Clin Nutr 2006;83:
108996, quiz 1207.
15. WHO. Pocket book of hospital care for children. Geneva, Switzerland:
World Health Organization Press, 2005. Available from: http://whqlibdoc.
who.int/publications/2005/9241546700.pdf (cited 15 October 2012).
16. Agarwal R, Singh V, Yewale V. RTI FACTS: IAP consensus guidelines
on rational management of respiratory tract infections in children.
Mumbai, India: Indian Academy of Pediatrics, 2006.

Downloaded from ajcn.nutrition.org by guest on December 7, 2016

a high baseline serum zinc concentration of 22.9 mmol/L compared with in studies in which treatment benefits were seen (20).
The baseline serum zinc concentration in this study was lower,
at 9.3mmol/L, than in any of the previous studies. The lack of
correlations between baseline zinc concentrations and clinical
outcomes may have been due to limitations of serum or plasma
zinc as a marker of zinc status in the body, particularly during
infections, or possibly because of effects of zinc that occur irrespective of zinc concentrations.
In our study, it is possible that the dose of zinc as adjunctive
therapy for the treatment of pneumonia may have been low; however,
this dose was similar to or more than the dose used in other studies
(1014, 22, 24). Doses .20 mg may be studied in future trials.
This study had some limitations. Although we completed our
overall sample size of 492 children with severe or very severe
pneumonia, in the subgroup of very severe pneumonia, we could
enroll only 120 children compared with the proposed 200 children during the study period. And, despite a well-conducted
stratified random assignment, the 2 groups in the strata of
children with very severe pneumonia were substantially different
in terms of the prevalence of severe underweight, which ultimately affected our results. This effect may not have happened if
we had been able to enroll the proposed number of children with
very severe pneumonia.
Our study provides important information on a possible benefit
of zinc supplementation in a subgroup of children with very severe
pneumonia. We did not show an overall benefit of zinc for pneumonia treatment. To date, a clear benefit of zinc in the treatment
of pneumonia has been seen in 2 of 8 studies that examined the
impact of zinc on pneumonia outcomes. In light of these varying
results and differences seen in various subgroup analyses, a metaanalysis is warranted to determine whether there is any overall
benefit of zinc for the treatment of pneumonia, but more importantly, additional research is needed in specific subgroups such as
children with very severe pneumonia with stratification for severely
underweight.

1393

1394

WADHWA ET AL

17. Clinical and Laboratory Standards Institute. M7eA7 dilution antimicrobial


susceptibility tests for bacteria that grow aerobically. Wayne PA: CLSI;
2007. Available from: http://www.microbiolab-bg.com/CLSI.pdf (cited
15 October 2012).
18. Hambidge KM, King JC, Kern DL, English-Westcott JL, Stall C. Prebreakfast plasma zinc concentrations: the effect of previous meals.
J Trace Elem Electrolytes Health Dis 1990;4:22931.
19. Mathew JL. Zinc supplementation for prevention or treatment of childhood pneumonia: a systematic review of randomized controlled trials.
Indian Pediatr 2010;47:616.
20. Bansal A, Parmar VR, Basu S, Kaur J, Jain S, Saha A, Chawla D. Zinc
supplementation in severe acute lower respiratory tract infection in
children: a triple-blind randomized placebo controlled trial. Indian
J Pediatr 2011;78:337.
21. Valentiner-Branth P, Shrestha PS, Chandyo RK, Mathisen M, Basnet S,
Bhandari N, Adhikari RK, Sommerfelt H, Strand TA. A randomized
controlled trial of the effect of zinc as adjuvant therapy in children 2-35
mo of age with severe or nonsevere pneumonia in Bhaktapur, Nepal.
Am J Clin Nutr 2010;91:166774.
22. Basnet S, Shrestha PS, Sharma A, Mathisen M, Prasai R, Bhandari
N, Adhikari RK, Sommerfelt H, Valentiner-Branth P, Strand TA. A

23.

24.

25.
26.
27.

randomized controlled trial of zinc as adjuvant therapy for severe


pneumonia in young children. Pediatrics 2012;129:7018.
Bhatnagar S, Wadhwa N, Aneja S, Lodha R, Kabra SK, Natchu UCM,
Sommerfelt H, Dutta AK, Chandra J, Rath B, et al. Zinc as adjunct
treatment in infants aged between 7 and 120 days with probable serious
bacterial infection: a randomised, double-blind, placebo-controlled
trial. Lancet 2012;379:20728.
Srinivasan MG, Ndeezi G, Mboijana CK, Kiguli S, Bimenya GS,
Nankabirwa V, Tumwine JK. Zinc adjunct therapy reduces case fatality
in severe childhood pneumonia: a randomized double blind placebocontrolled trial. BMC Med 2012;10:14.
Tiewsoh K, Lodha R, Pandey RM, Broor S, Kalaivani M, Kabra SK.
Factors determining the outcome of children hospitalized with severe
pneumonia. BMC Pediatr 2009;9:15.
Sehgal V, Sethi GR, Sachdev HP, Satyanarayana L. Predictors of
mortality in subjects hospitalized with acute lower respiratory tract
infections. Indian Pediatr 1997;34:2139.
Mathew JL, Patwari AK, Gupta P, Shah D, Gera T, Gogia S, Mohan P,
Panda R, Menon S. Acute respiratory infection and pneumonia in India:
a systematic review of literature for advocacy and action: UNICEF-PHFI
series on newborn and child health, India. Indian Pediatr 2011;48:191218.

Downloaded from ajcn.nutrition.org by guest on December 7, 2016