A Randomized Controlled Trial of Zinc as AdjuvantTherapy for Severe Pneumonia in Young Children

WHAT’S KNOWN ON THIS SUBJECT: Pneumonia is stilla significant problem in young children from developing countrieswhere zinc deficiency is prevalent. Although zinc supplementationreduces the risk of childhood pneumonia, the effect of adjunctzinc on severe pneumonia is unclear with conflicting results.

WHAT THIS STUDY ADDS: The overall effect, if any, of zinc asadjuvant therapy for World Health Organization–defined severepneumonia in young children is small.

abstractBACKGROUND AND OBJECTIVE: Diarrhea and pneumonia are the lead-ing causes of illness and death in children ,5 years of age. Zincsupplementation is effective for treatment of acute diarrhea and canprevent pneumonia. In this trial, we measured the efficacy of zincwhen given to children hospitalized and treated with antibiotics forsevere pneumonia.

METHODS: We enrolled 610 children aged 2 to 35 months who pre-sented with severe pneumonia defined by the World Health Organiza-tion as cough and/or difficult breathing combined with lower chestindrawing. All children received standard antibiotic treatment andwere randomized to receive zinc (10 mg in 2- to 11-month-olds and20 mg in older children) or placebo daily for up to 14 days. Theprimary outcome was time to cessation of severe pneumonia.

RESULTS: Zinc recipients recovered marginally faster, but this differ-ence was not statistically significant (hazard ratio = 1.10, 95% CI 0.94–1.30). Similarly, the risk of treatment failure was slightly but not sig-nificantly lower in those who received zinc (risk ratio = 0.88 95% CI0.71–1.10).

CONCLUSIONS: Adjunct treatment with zinc reduced the time to ces-sation of severe pneumonia and the risk of treatment failure only mar-ginally, if at all, in hospitalized children. Pediatrics 2012;129:701–708

AUTHORS: Sudha Basnet, MD,a Prakash S. Shrestha, MD,a

Arun Sharma, MD,a Maria Mathisen, PhD,b Renu Prasai, DCH,c

Nita Bhandari, PhD,d Ramesh K. Adhikari, MD,a HalvorSommerfelt, PhD,b,e Palle Valentiner-Branth, PhD,f Tor A.Strand, PhD,b,e,g and members of the Zinc SeverePneumonia Study GroupaChild Health Department, Institute of Medicine, TribhuvanUniversity, Kathmandu, Nepal; bCentre for InternationalHealth, University of Bergen, Bergen, Norway; cKanti Children’sHospital, Kathmandu, Nepal; dSociety for Applied Studies, NewDelhi, India; eDivision of Infectious Disease Control, NorwegianInstitute of Public Health, Oslo, Norway; fDivision of Epidemiology,Department of Epidemiology, Statens Serum Institut, Copenhagen,Denmark; and gMedical Microbiology, Innlandet Hospital TrustLillehammer, Norway

KEY WORDSzinc, pneumonia, young children, therapeutic, treatment failure

ABBREVIATIONSCR—chest radiographLCI—lower chest indrawingSpO2—oxygen saturationWHO—World Health Organization

Drs Basnet, Shrestha, Sharma, Adhikari, Sommerfelt, and Strandmade primary contributions to the design, conduct, analysis,interpretation, and writing of this manuscript; Drs Prasai andMathisen contributed to the field conduct, training andstandardization, collection of biological specimen, and qualitycontrol of the trial; Dr Mathisen was also responsible for thevirus analyses; Drs Bhandari and Valentiner-Branth contributedto the development and design of the trial and to theinterpretation of the study results; Dr Strand had full access toall the data in the study and took the final decision to submitthis report for publication.

This trial has been registered at www.clinicaltrials.gov(identifier NCT00252304).

www.pediatrics.org/cgi/doi/10.1542/peds.2010-3091

doi:10.1542/peds.2010-3091

Accepted for publication Nov 28, 2011

(Continued on last page)

PEDIATRICS Volume 129, Number 4, April 2012 701

ARTICLE

by guest on April 19, 2021www.aappublications.org/newsDownloaded from

The estimated number of deaths glob-ally in children aged ,5 years was8.8 million in 2008.1 Pneumonia anddiarrhea combined are responsible for∼45% of these deaths.2 It is estimatedthat zinc deficiency in association withdiarrhea, pneumonia, and malaria con-tributes to 4.4% of deaths and 3.8% oflost disability adjusted life years amongchildren aged 6 to 59 months in Africa,Latin America, and Asia.3 In a meta-analysis of clinical trials evaluating thepreventive role of zinc, daily supple-mentation led to 14% and 8% reductionsin the risk of diarrhea and pneumonia,respectively.4 The World Health Orga-nization (WHO) now recommends zincfor the treatment of children withdiarrhea5 because there is sufficientevidence demonstrating that supple-mentation reduces the severity andduration of the episode.6 The benefitof zinc in the treatment of pneumoniais, however, unclear. Although zinc asadjuvant therapy for hospitalized chil-dren with pneumonia was found to bebeneficial in 1 clinical trial in Bangla-desh,7 other trials in India8 and Aus-tralia9 found no effect. In developingcountries, the estimated incidence ofclinical pneumonia in children aged,5 years is 0.29 episodes per childyear.10 Despite a declining trend, theincidence of pneumonia in Nepal is0.13 episodes per child per year withsevere pneumonia accounting for 1.2%of the all cases of acute respiratoryinfections.11 We have previously shownthat zinc deficiency is common in chil-dren12 and in women of reproductiveage13 in the Kathmandu valley. In a largecommunity-based trial in Bhaktapur,Nepal, zinc administration did notreduce time till recovery or risk oftreatment failure in children with pneu-monia.14 Whether zinc administrationhas a beneficial effect when given tochildren hospitalized with severe pneu-monia needs to be clarified. We there-fore undertook a clinical trial to assessthe efficacy of zinc as adjuvant therapy

to standard antibiotic treatment in re-ducing the time to cessation and therisk of treatment failure of a severepneumonia episode.

METHODS

This was a double-blind, randomized,placebo-controlled trial in young chil-dren designed to measure the impactof daily zinc administration for up to14 days on time to cessation of severepneumonia. An important secondaryoutcome was the risk of treatment fail-ure. Clearances were obtained from theethical board of the Institute of Medi-cine, Tribhuvan University, and NepalHealth Research Council, Kathmandu.

Sample Size Calculation

The study was based on the hypothesisthat zinc sulfate given as adjuvanttherapy to standard antibiotic treat-ment would result in a reduction of themedian duration of hospital stay by1 day. In a similar study in Bangladesh,7

the median time to resolution fromsevere pneumonia was 3 days in thezinc and 4 days in the placebo group.With a minimal detectable hazard ratioof 1.30 for time to cessation of severepneumonia, we estimated a total sam-ple size of 500 children. Calculationswere done with 80% power and an a

error of 5%. Assuming up to 20% loss tofollow-up, we enrolled 610 children.

Enrollment, Baseline ClinicalWorkup, and Initiation ofAntibiotic Therapy

Children aged 2 to 35 months pre-senting to the Kanti Children’s hospitalemergency or outpatient departmentswith complaints of cough lasting ,14days and/or difficult breathing of ,72hours duration with lower chest in-drawing (LCI) were screened for en-rollment by trained physicians. TheIntegrated Management of ChildhoodIllnesses algorithm was used to definesevere pneumonia.15 Eligible children

were initially assessed for hypoxemiaby using a pulse oximeter (NellcorPuritan Bennett NPB-40, Pleasanton,CA) with a pediatric sensor (NellcorPedichek D-YSPD) and presence ofwheezing. Oxygen saturation (SpO2)was recorded twice after stabilizationof the reading for 1 minute. The higherof the 2 readings was used. For chil-dren with SpO2 of ,90%, oxygen wasprovided before further evaluation.Children with wheezing were given upto 3 doses of nebulized Salbutamol 15minutes apart, reassessed, and ex-cluded if LCI disappeared.16 A history ofthe child’s illness was taken and phys-ical examination carried out by usinga standardized form. Children wereweighed by using an Electronic Scale890 (SECA, Hamburg, Germany) thatmeasures to the nearest 100 g. Heightwas measured by using a standardwooden height measuring board, andrecumbent length in children ,2 yearswas measured by using an infan-tometer, both to the nearest 0.1 cm.Stunting defined as length-for-age #2z score and wasting as weight-for-length #2 z score were calculated byusing the 2006 WHO Child GrowthStandards.17 We measured hemoglo-bin concentrations by using Hemocue(Ångelholm, Sweden). A chest radio-graph (CR) taken in all children wasinterpreted by a radiologist blinded toclinical data by using the WHO stan-dardized tool for interpretation of CRsand findings classified as endpoint con-solidation, other infiltrates, or normal.18

Children with recurrent wheezing (de-fined as .3 episodes over the past6 months and on treatment with bron-chodilators), disappearance of LCI afternebulized salbutamol, severe wasting,19

severe anemia (hemoglobin ,7 g/dL),heart disease, documented tuberculo-sis, or concomitant diarrhea with de-hydration and those with severe illnessrequiring special care or surgical inter-vention were excluded. Informed con-sent was obtained for eligible children.

702 BASNET et al by guest on April 19, 2021www.aappublications.org/newsDownloaded from

Literate parents signed the consentform after they had read a writtenstatement in the local Nepali language.For parents who were unable to reador write, verbal informed consent wasobtained in the presence of a witness.After obtaining consent, blood wascollected for investigations, and thefirst dose of intravenous antibioticswas administered. This was followedby collection of nasopharyngeal aspi-rate for identification of 7 respiratoryviruses. Details of the sampling tech-nique and the results of these analyseshave been described in a separatepublication20

Randomization, Intervention,and Blinding

The intervention was dispersible tab-lets containing 10 mg of elemental zincsulfate or placebo, manufactured andpacked by Nutriset (Malaunay, France)in a blister pack with 15 tablets. Tabletsof both groups were identical in pack-aging, appearance, and inactive in-gredients with no indication of groupidentity and content. For each child inthe study, there were 3 blister packswith zinc or placebo tablets labeledwith a serial number tomatch the childstudy identification number. The ran-domization list linking the treatmentgroups to these identification numberswas generated by and kept with a per-son not otherwise involved in the study.Children were allocated to either ofthe 2 intervention groups by randomi-zation in blocks of 16 in a 1:1 ratio.Randomization was stratified on age,12months or$12 completed monthsand on wheezing status before nebuli-zation. Study physicians selected blis-ter packs with the lowest number froma box specifying the stratum to whicheach child belonged. Children ,12monthswere given 1 tablet and children$12 months were given 2 tablets dis-solved in 5 mL of clean water or breastmilk. The first dose was dispensed bythe study physician and subsequently

by trained study assistants who werenot otherwise involved in patient care.The zinc or placebo dispersions weregiven as a single daily dose until dis-charge or for amaximum of 14 days. Allchildren were observed for vomiting.For children who vomited within thefirst 15 minutes, a repeat dose wasgiven. Children who vomited the secondtime were given the required amount in2 divided doses over a 24-hour periodfrom the next day onward.

Outcome Definitions

The primary outcome, time to cessationof severe pneumonia, was defined asthe period starting from enrollmentto the beginning of a 24-hour consec-utive period of absence of LCI, hypoxia,and any danger signs. We used WHOguidelines to define hypoxia (SpO2 of,90%) and danger signs such as in-ability to breastfeed or drink, vomit-ing everything consumed, convulsions,lethargy, or unconsciousness.19 Thesecondary outcome, treatment failure,was defined as a requirement forchange in antibiotics, development ofcomplications, such as empyema orpneumothorax requiring surgical in-tervention, or admission to the ICUfor ventilator or inotropic support.

Cointerventions

Enrolled children were admitted tothe hospital and monitored by studyphysicians at ∼8 hourly intervals un-til discharge. Benzyl penicillin (50 000U/kg intravenously every 6 hours) andgentamicin (7.5 mg/kg intravenouslyonce daily) were given until clinicalimprovement, defined as absence ofdanger signs, of hypoxia for 24 con-secutive hours and of LCI for a 48-hourperiod. Patients were then dischargedwith advice to continue oral amoxicillinto complete treatment of a total dura-tion of 10 days.

Antibiotics were changed to cefotax-ime in children with failure to improve,

defined as persistence of LCI or of anydanger signs present at enrollmentdespite 48 hours of treatment or ap-pearance of new danger signs or hyp-oxia with deterioration of patient’sclinical status anytime after initiationof treatment. A decision to changeantibiotics was made only after con-sultation with senior pediatricians in-volved in the study. For children unableto eat/drink or breastfeed, intravenousfluids based on daily requirementswere initiated.19 Humidified oxygenwas given to children with documentedhypoxia. During each physician visit,oxygen saturation was documentedafter a washout period of 5 minutesand oxygen discontinued when theywere no longer hypoxic. The absenceof hypoxia was confirmed after a sec-ond reading taken 30 minutes later.

Study physicians were trained to as-sess and manage children with severepneumonia, and theirperformancewasmonitored each day by experiencedpediatricians, who were also the inves-tigators for this study. Standardizationexercises were conducted before thestart of the trial. Each physician wasassigned to record axillary tempera-ture, count respiratory rate, observefor LCI, and listen for wheezing andcrepitations in at least 10 children.Their findings were matched againstthose of an experienced pediatrician un-til the desired agreement was reached.

Data Management and Analysis

The completed forms with patient datawere collected by the study assistantson a daily basis. All formswere checkedmanually by 1 of the clinical supervi-sors before data entry, which was donewithin 48 hours. The data were doubleentered into a database (Visual FoxPro6.0, Microsoft Corp, Redmond, WA) withinbuilt logic, range, and consistencychecks. Statistical analyses were under-taken by using Stata, version 10 (StataCorp, College Station, TX). Data cleaning,

ARTICLE

PEDIATRICS Volume 129, Number 4, April 2012 703 by guest on April 19, 2021www.aappublications.org/newsDownloaded from

definition of outcome variables, exclu-sion of cases as well as programmingof scripts in the statistical packageswere done before the analysis fileswere merged with the randomizationlists. We used Cox proportional hazardsregression models to compare thetime to cessation of severe pneumoniabetween the treatment groups, theeffect estimates expressed as hazardratios. Treatment failure, risk of pro-longed illness and vomiting after thefirst dose of the intervention was com-pared by using generalized linear re-gression models with log link functionsand binomial distributions, yielding re-lative risks. We coded the outcomesand interventions so that hazard ratios.1 and relative risks ,1 would repre-sent a beneficial effect of zinc. Differ-ences were considered significant whena two-sided P value was,.05.

RESULTS

From January 8, 2006, to June 30, 2008,we screened 2199 children meeting in-clusion criteria. There were 1589 (72%)who were not eligible for randomiza-tion, of whom 1282 (58%) fulfilled thepredefined exclusion criteria (Fig 1),227 (10%) did not consent, and 80 (4%)had been previously enrolled. After en-rollment and randomization of the re-maining 610 children, we discoveredthat 11 with heart disease 1 with coughduration.14 days had been included.These trial deviates were evenly dis-tributed between the study arms (Fig 1)and were excluded from the analysis.Of the remaining 598 children, 299were randomized to receive zinc and299 to receive placebo. In the zincgroup, 199 of the 245 (81%) infants and45 of the 54 (83%) older children hadwheezing. In the placebo group, 208 ofthe 248 (84%) infants compared with40 of the 51 (78%) older children pre-sented with wheezing. Eleven childrenwere lost to follow-up in the zinc armand 7 in the placebo arm (Fig 1). The

remaining 580 (288 in the zinc and 292in placebo arms) stayed in the studyuntil recovery from severe pneumonia.

Viruses were isolated from nasopha-ryngeal aspirates in 29% with detailsdescribed elsewhere.20 Among 533CRs available for interpretation, 520films were of adequate quality. End-point consolidation was identified in126 (24%), 196 (38%) had other infil-trates, and 198 (38%) were normal.Most baseline characteristics wereevenly distributed between groups(Tables 1 and 2).

Analyses were done by intention totreat. The time until cessation of se-vere pneumonia was slightly shorteramong the zinc recipients, with a hazardratio of 1.10, 95% confidence interval0.94 to 1.30; P = .22 (Table 3). We ex-plored whether the effect of zinc wasdifferent in subgroups on the basis ofage, gender, presence of fever, hypoxia,

wheezing, crepitations, virus isolatedin nasopharyngeal aspirate, endpointconsolidation on CR, wasting, and stunt-ing. In the subgroup consisting of chil-dren with endpoint consolidation, zincrecipients recovered significantly fasterthan those in the placebo group (Fig 2).The effect of zinc, however, was notsignificantly different between thosewith and without radiographic pneu-monia, that is, the interaction was notstatistically significant. We also com-pared the proportion of children withsevere pneumonia at 72, 96, or 120hours after admission between thestudy groups. These comparisons werealso in favor of zinc, but none reachedstatistical significance (Table 3). Therisk of treatment failure was loweramong the zinc recipients; however,this was also not statistically signifi-cant (risk ratio: 0.88; 95% confidenceinterval: 0.71–1.10). Adjusting for po-tential confounders altered the results

FIGURE 1Trial profile of a randomized, placebo-controlled trial of zinc as adjunct therapy for severe pneumonia inchildren aged 2 to 35 months of age in Kanti Children’s Hospital, Kathmandu, Nepal.

704 BASNET et al by guest on April 19, 2021www.aappublications.org/newsDownloaded from

only marginally. The proportion of chil-dren who vomited after the first doseof supplement was higher (14%) inthe zinc than in the placebo group (9%;P = .052).

DISCUSSION

This study on zinc as adjunct therapyinchildrenwithseverepneumoniashowsa modest but not statistically signif-icant effect of daily zinc administration

in reducing time to cessation of severepneumonia defined as a 24-hour con-secutive period of absence of LCI, hyp-oxia, and any other danger sign.

In a study undertaken in Bangladesh,7

children who received zinc recoveredfaster, and fewer had treatment fail-ure and duration of severe pneumonialasting .72, 96, or 120 hours. The re-sults from our trial were in the samedirection but smaller. Another trial inSouth India8 failed to show any bene-ficial effect of zinc on duration of illnessin young children with severe pneu-monia. In another study from Kolkatta,India,21 although zinc was efficacious inboys, the overall effect as well as theinteraction between gender and zincadministration was not statisticallysignificant. All these studies, like ourstudy, were double-blind randomizedcontrolled trials assessing the efficacyof zinc in children with severe pneu-monia. Inherent differences in the pop-ulations studied and differences in theillness characteristics including pre-enrollment duration and definition ofrecovery would explain the discrepancybetween studies.

In the current trial, the proportion ofchildren with wheezing was 82% com-pared with 37% in the Bangladeshi7

and 62.5% in the South Indian trial.8

Because children with reactive airwaydisease would meet criteria for inclu-sion in our study, we excluded childrenwith a history of recurrent wheezingand enrolled others only if LCI per-sisted after salbutamol administration.Brooks et al reported that in childrenwithout wheezing, administration of zincresulted in earlier resolution of clinicalsigns.7 The effect of zinc was not modi-fied by wheezing status in our subgroupanalysis (Fig 2), a finding similar to thatreported from South India.8 However,because there were only 106 childrenwithout wheezing, we had insufficientpower to detect an effect of zinc in thissubgroup.

TABLE 1 Baseline Demographic Characteristics of Children aged 2 to 35 Months in a Clinical TrialEvaluating Efficacy of Zinc as Adjuvant Therapy for Severe Pneumonia

Characteristic Zinc Group Placebo Group

n Value n Value

Mean age in months (SD) 299 7.8 (6.0) 299 7.1 (5.6)Age groups (months)2–6, n (%) 299 167 (55.9) 299 185 (61.7)7–11, n (%) 299 78 (26.1) 299 63 (21.1)12–23, n (%) 299 45 (15.1) 299 45 (15.0)24–35, n (%) 299 9 (3.0) 299 6 (2.0)

Male subjects, n (%) 299 177 (59.2) 299 190 (63.6)Mean age of mother, y (SD) 292 24.7 (4.2) 294 24.1 (4.1)Illiteratea mother (%) 292 79 (27.1) 294 75 (25.5)Illiteratea father (%) 293 20 (6.8) 295 18 (6.1)Unemployedb mother (%) 290 205 (70.7) 294 213 (72.5)Unemployedb father (%) 289 10 (3.5) 291 9 (3.1)Child still breastfeeding 299 283 (94.6) 299 288 (96.3)a No schooling with inability to read part or whole of a sentence.b No work/housework.

TABLE 2 Baseline Clinical Characteristics of Children Aged 2 to 35 Months Enrolled in a ClinicalTrial on the Efficacy of Zinc as Adjuvant Therapy for Severe Pneumonia

Characteristic Zinc Placebo

n Value n Value

Mean duration of cough in days (SD) 299 4.0 (2.0) 299 4.2 (2.2)Mean duration of difficulty breathing in hours (SD) 299 30.2 (20.2) 299 29.6 (18.3)Mean duration of fever in days (SD) 264 3.5 (2.0) 279 3.5 (2.1)Wt for length/height #2 z scorea (%) 296 86 (29.1) 298 71 (23.8)Length/height for age #2 z scorea (%) 299 28 (9.4) 299 22 (7.4)Mean hemoglobin in g/dL (SD) 299 11.0 (1.4) 299 10.9 (1.4)Mean axillary temperature in °C (SD) 299 37.7 (0.8) 299 37.7 (0.9)Febrileb (%) 299 46 (15.4) 299 46 (15.4)Mean respiratory rate in breaths per minute (SD)2–11 mo 245 64 (12) 248 65 (12)12–35 mo 54 60 (13) 51 63 (10)

Mean oxygen saturation (SD) 299 86 (8) 299 87 (9)Oxygen saturation (%),90% 299 186 (62.2) 299 187 (62.5),93% 299 255 (85.3) 299 249 (83.3)

Wheezing (%) 299 244 (81.6) 299 248 (82.9)Crepitations (%) 299 273 (91.3) 299 276 (92.3)Endpoint consolidation on CR 258 60 (23.3) 262 66 (25.2)Symptoms of severe pneumoniaChild unable to drink/breastfeed (%) 299 30 (10.0) 299 24 (8.0)History of convulsions (%) 299 1 (0.3) 299 2 (0.7)Vomiting everything child eats (%) 294 9 (3.1) 292 5 (1.7)Child unconscious/lethargic (%) 299 25 (8.4) 299 30 (10.0)

Signs of severe pneumoniaNasal flaring (%) 298 116 (38.9) 299 116 (38.8)Grunting (%) 299 60 (20.1) 299 71 (23.8)Head nodding (%) 299 65 (21.7) 299 73 (24.4)Central cyanosis (%) 299 1 (0.3) 299 2 (0.7)

a Derived by using the recent WHO Child Growth Standards.17b Defined as axillary temperature of .38.5°C.

ARTICLE

PEDIATRICS Volume 129, Number 4, April 2012 705 by guest on April 19, 2021www.aappublications.org/newsDownloaded from

This study enrolled 610 children andto our knowledge is the largest trialconducted to date on zinc given dur-ing severe pneumonia. The study was

conducted in an area where zinc de-ficiency is common12,13 and in a countrywhere pneumonia in young childrenis a significant health problem.11 The

study site is the only government hos-pital and referral center for childrenliving in as well as outside Kathmandu.We had study physicians who werewell trained and dedicated only to thistrial. Daily doses of supplement wereprovided by study assistants and notthe physicians recruited for the study.Limiting duration of difficulty breath-ing to #72 hours ensured that we en-rolled children with an acute episodeof pneumonia. We used objective out-comes for defining resolution, disap-pearance of LCI and SpO2$90%, whichare replicable findings. In a review as-sessing the precision of clinical signsin the diagnosis of pneumonia, Margoliset al concluded that there is betteragreement among observers for aclinical sign that can be observed, suchas respiratory effort retractions (k =0.48), than for an auscultatory sign,such as presence of adventitious sounds(k = 0.3).22

The high proportion of children withwheezing is a limitation of this study, afinding likely due to use of WHO criteriain defining severe pneumonia. This def-inition has high specificity for severelower respiratory tract infection butdoes not define the etiology.23 Severepneumonia includes a wide spectrumof causes and predisposing factors thatmay respond differently to zinc adminis-tration. Furthermore, this heterogeneitycan also result in poor specificity ofthe outcomes, which again may dilute

TABLE 3 Primary and Secondary Outcomes in a Randomized, Placebo-Controlled Trial on Oral Zinc as Adjunct Therapy for Severe Pneumonia inChildren 2 to 35 Months of Age

Zinc Group Placebo Group Hazard Ratioa (95% CI) P

n Value (IQR) n Value (IQR)

Median time to cessation of severe pneumonia in hours 288 49 (33, 77) 292 49 (29, 91) 1.10 (0.94 – 1.30) .22Proportion with duration of severe pneumonia in hours Risk Ratiob (95% CI).72 295 83 (28) 297 104 (35) 0.80 (0.63–1.02) .07.96 294 56 (19) 296 64 (22) 0.88 (0.64–1.21) .44.120 293 31 (11) 294 46 (16) 0.67 (0.44–1.03) .07

Proportion with treatment failure 296 98 (33) 298 111 (37) 0.88 (0.71–1.10) .29Proportion with vomiting after supplementc 299 41 (14) 299 26 (9) 1.57 (0.99–2.50) .05

IQR, interquartile range.a Hazard ratio for time until cessation of severe pneumonia was estimated by using Cox proportional hazards models. A value of .1 indicates beneficial effect of zinc.b From generalized linear models with a log link function and binomial distribution.c After the first dose.

FIGURE 2The efficacy of oral zinc given during an episode of severe pneumonia in Nepalese children aged 2 to35 months in selected subgroups. Hazard ratios for time until cessation of severe pneumonia wereestimated by using Cox proportional hazards models. A value .1 indicates beneficial effect of zinc.Weight for length/height#2 z score by using WHO child growth standards. Length/height for age#2z score by using WHO child growth standards.

706 BASNET et al by guest on April 19, 2021www.aappublications.org/newsDownloaded from

a measureable effect of zinc. Using CRsto improve the diagnosis enabled us toidentify only 24% with radiographicpneumonia. It is also noteworthy thatthere was significant beneficial effect ofzinc in this subgroup. Future studies areneeded exploring the role of zinc in se-vere pneumonia in which children withwheezing are excluded and CR, micro-biologic, and inflammatory markers areused in an attempt to arrive at a morespecific diagnosis.24

CONCLUSIONS

This trial yielded a small but not sta-tistically significant efficacy estimatefor zinc in the resolution of severe pneu-monia in hospitalized 2- to 35-month-oldchildren. All study participants receivedoptimized antibiotic and other therapies,

an aspect that needs to be taken intoaccount when the results from this trialas well as those of other completed,ongoing, and planned studies are in-terpreted and summarized to deter-mine the role of zinc in the treatmentof severe pneumonia.

ACKNOWLEDGMENTSThe Zinc Severe Pneumonia Study Groupis:RameswarManShrestha,MD;UdayRajUpadhaya, DCH; ChandeshwarMahaseth,MRCP; Rojen Sundar Shrestha, DCH;Puja Shrestha, MBBS; Geetika KC, MBBS;YagyaRatnaShakya,MBBS;UjmaShrestha,MBBS.

We thank all the children and their fam-ilies who took part in the study. We areindebted to field assistants MaheshKumar Thapa and Ram Krishna Khatri

and the study physicians for their contri-bution to thestudyaswellasotherstaff ofthe Child Health Research Project. Wethank the director of the Kanti ChildrenHospital and staff of the Emergency, Ob-servation, Acute Respiratory Illness/OralRehydration Therapy, Medical, Payingand Special Cabin wards for their invalu-able support. We also thank the headof the Department of Child Health, Pro-fessor Pushpa Raj Sharma, and otherfaculty members for their support, theDepartment of Microbiology, TribhuvanUniversity TeachingHospital, Kathmandu,for providing the laboratory facilities andDr Dhiraj Man Shrestha for interpreta-tion of chest radiographs. We thank Sol-frid Vikøren for excellent administrativesupport and Hans Steinsland for gen-erating the randomization list and main-taining it until the end of the study.

REFERENCES

1. Black RE, Cousens S, Johnson HL, et al;Child Health Epidemiology Reference Groupof WHO and UNICEF. Global, regional, andnational causes of child mortality in 2008:a systematic analysis. Lancet. 2010;375(9730):1969–1987

2. Rudan I, El Arifeen S, Black RE, Campbell H.Childhood pneumonia and diarrhoea: set-ting our priorities right. Lancet Infect Dis.2007;7(1):56–61

3. Fischer Walker CL, Ezzati M, Black RE. Globaland regional child mortality and burden ofdisease attributable to zinc deficiency. EurJ Clin Nutr. 2009;63(5):591–597

4. Aggarwal R, Sentz J, Miller MA. Role of zincadministration in prevention of childhood di-arrhea and respiratory illnesses: a meta-analysis. Pediatrics. 2007;119(6):1120–1130

5. World Health Organization; United NationsChildren’s Fund; Johns Hopkins BloombergSchool of Public Health; USAID. Implement-ing the New Recommendations on theClinical Management of Diarrhea: Guide-lines for Policy Makers and ProgrammeManagers. Geneva, Switzerland: World HealthOrganization; 2006

6. Fontaine O. Effect of zinc supplementationon clinical course of acute diarrhea. J HealthPopul Nutr. 2001;19(4):338–346

7. Brooks WA, Yunus M, Santosham M, et al.Zinc for severe pneumonia in very young

children: double-blind placebo-controlledtrial. Lancet. 2004;363(9422):1683–1688

8. Bose A, Coles CL, Gunavathi JH, et al. Efficacyof zinc in the treatment of severe pneumo-nia in hospitalized children ,2 y old. Am JClin Nutr. 2006;83(5):1089–1096, quiz 1207

9. Chang AB, Torzillo PJ, Boyce NC, et al. Zincand vitamin A supplementation in Indi-genous Australian children hospitalisedwith lower respiratory tract infection: a rand-omised controlled trial. Med J Aust. 2006;184(3):107–112

10. Rudan I, Boschi-Pinto C, Biloglav Z, MulhollandK, Campbell H. Epidemiology and etiologyof childhood pneumonia. Bull World HealthOrgan. 2008;86(5):408–416

11. Ghimire M, Pradhan YV, Maskey MH.Community-based interventions for diarrhealdiseases and acute respiratory infectionsin Nepal. Bull World Health Organ. 2010;88:216–221

12. Strand TA, Adhikari RK, Chandyo RK, SharmaPR, Sommerfelt H. Predictors of plasma zincconcentrations in children with acute di-arrhea. Am J Clin Nutr. 2004;79(3):451–456

13. Chandyo RK, Strand TA, Mathisen M, et al.Zinc deficiency is common among healthywomen of reproductive age in Bhaktapur,Nepal. J Nutr. 2009;139(3):594–597

14. Valentiner-Branth P, Shrestha PS, ChandyoRK, et al. A randomized controlled trial of

the effect of zinc as adjuvant therapy inchildren 2–35 mo of age with severe ornonsevere pneumonia in Bhaktapur, Nepal.Am J Clin Nutr. 2010;91:1667–1674

15. World Health Organization; United NationsChildren’s Fund. Model IMCI Handbook:Integrated Management of Childhood Illness.WHO/FCH/CAH/00.12. Geneva, Switzerland:Division of Child and Adolescent Health andDevelopment, World Health Organization; 2005

16. Hazir T, Qazi S, Nisar YB, et al. Assessmentand management of children aged 1–59months presenting with wheeze, fast breath-ing, and/or lower LCI; results of a multicentredescriptive study in Pakistan. Arch Dis Child.2004;89:1049–1054.

17. World Health Organization Multicentre GrowthReference Study Group. WHO Child GrowthStandards: Length/Height-for-Age, Weight-for-Age, Weight-for-Length, Weight-for-Heightand Body Mass Index-for-Age: Methods andDevelopment. Geneva, Switzerland: WorldHealth Organization; 2006

18. World Health Organization PneumoniaVaccine Trial Investigator’s Group. Stan-dardization of Interpretation of ChestRadiographs for the Diagnosis of Pneumo-nia in Children. WHO/V&B/ 01.35. Geneva,Switzerland: World Health Organization; 2001

19. World Health Organization. Pocket Book ofHospital Care for Children: Guidelines for

ARTICLE

PEDIATRICS Volume 129, Number 4, April 2012 707 by guest on April 19, 2021www.aappublications.org/newsDownloaded from

the Management of Common Illnesses WithLimited Resources. Geneva, Switzerland:World Health Organization; 2005

20. Mathisen M, Basnet S, Sharma A, et al.RNA viruses in young Nepalese childrenhospitalized with severe pneumonia. PediatrInfect Dis J. 2011;30:1032–1036

21. Mahalanabis D, Lahiri M, Paul D, et al. Ran-domized, double-blind, placebo-controlled

clinical trial of the efficacy of treatmentwith zinc or vitamin A in infants and youngchildren with severe acute lower respiratoryinfection. Am J Clin Nutr. 2004;79(3):430–436

22. Margolis P, Gadomski A. The rational clinicalexamination. Does this infant have pneu-monia? JAMA. 1998;279(4):308–313

23. Puumalainen T, Quiambao B, Abucejo-Ladesma E, et al; ARIVAC Research Con-

sortium. Clinical case review: a method toimprove identification of true clinical andradiographic pneumonia in children meet-ing the World Health Organization defini-tion for pneumonia. BMC Infect Dis. 2008;8:95

24. Das RR. Differential effects of zinc in severepneumonia in children. Indian J Pediatr.2011;78:1159–1160

THE LASTING IMPACT OF A GOOD TEACHER: How important is a good teacher?That question has been the subject of intense debate over the past several years.Although most people agree that a good primary school teacher is incrediblyvaluable, how to define a good teacher is quite difficult, and defining a poorteacher is even more challenging. One controversial way to assess the quality ofa teacher is through use of value-added ratings. Most simply, value-added ratingscompare the test scores of students for a teacher in one year to test scores ofprevious years and are also compared to students of other teachers. Previousresearch has shown that the positive effect of a single teacher on the subsequentscores of students fades within a few years. As reported in The New York Times(Education: January 6, 2012), a large study that looked at 2.5 million students formore than 20 years, concluded that good teachers have a lasting impact not juston student test scores but on many other measures including adult income. Theresearchers used value-added ratings to characterize elementary and primaryschool teachers as good, average, or poor. Students of teachers characterized asgood had lower teenage-pregnancy rates, were more likely to attend college, andhad greater adult earnings. The impact of a single teacher on a single student isquite small. For example, having a single good teacher rather than all averageteachers between fourth and eighth grade may result in a net increase in adultlifetime income of only $4,600. However, if the results of an entire class are ag-gregated over time, the total effect becomes quite large. Importantly, the dif-ference between poor and average was just as large and strong as the differencebetween good and average. The implications are that given the challenge offinding, hiring, and training good teachers, weeding out poor teachers may bemore valuable. Of course, the use of value-added ratings is quite controversial.Some school officials argue that teachers may be unfairly singled out for dis-missal or that if high value-added ratings become critical to job security,teachers may begin teaching to the test. Regardless, the data shows that havingeither good or poor primary or elementary school teachers has a lasting impacton student lives. This is just one way to identify those teachers.

Noted by WVR, MD

(Continued from first page)

Address correspondence to Tor A. Strand, PhD, Medical Microbiology, Innlandet Hospital Trust, Anders Sandvigsgate 17, 2629 Lillehammer, Norway. E-mail: tors@me.com

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2012 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

FUNDING: Supported by a grant from the European Commission (EU-INCO-DC contract INCO-FP6-003740); by funding from the Meltzer Foundation in Bergen, Norway;by the Danish Council of Developmental Research (project 91128); and by the Research Council of Norway (projects 151054 and 172226). The sponsors of the studyhad no role in study design, data collection, data analysis, data interpretation, or writing of the report.

708 BASNET et al by guest on April 19, 2021www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2010-3091 originally published online March 5, 2012; 2012;129;701Pediatrics 

Strand and members of the Zinc Severe Pneumonia Study GroupBhandari, Ramesh K. Adhikari, Halvor Sommerfelt, Palle Valentiner-Branth, Tor A. Sudha Basnet, Prakash S. Shrestha, Arun Sharma, Maria Mathisen, Renu Prasai, Nita

Pneumonia in Young ChildrenA Randomized Controlled Trial of Zinc as Adjuvant Therapy for Severe

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/129/4/701including high resolution figures, can be found at:

Referenceshttp://pediatrics.aappublications.org/content/129/4/701#BIBLThis article cites 15 articles, 6 of which you can access for free at:

Subspecialty Collections

http://www.aappublications.org/cgi/collection/pharmacology_subPharmacologyfollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

http://www.aappublications.org/site/misc/Permissions.xhtmlin its entirety can be found online at: Information about reproducing this article in parts (figures, tables) or

Reprintshttp://www.aappublications.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

by guest on April 19, 2021www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2010-3091 originally published online March 5, 2012; 2012;129;701Pediatrics 

Strand and members of the Zinc Severe Pneumonia Study GroupBhandari, Ramesh K. Adhikari, Halvor Sommerfelt, Palle Valentiner-Branth, Tor A. Sudha Basnet, Prakash S. Shrestha, Arun Sharma, Maria Mathisen, Renu Prasai, Nita

Pneumonia in Young ChildrenA Randomized Controlled Trial of Zinc as Adjuvant Therapy for Severe

http://pediatrics.aappublications.org/content/129/4/701located on the World Wide Web at:

The online version of this article, along with updated information and services, is

by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397. the American Academy of Pediatrics, 345 Park Avenue, Itasca, Illinois, 60143. Copyright © 2012has been published continuously since 1948. Pediatrics is owned, published, and trademarked by Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on April 19, 2021www.aappublications.org/newsDownloaded from