Glycerin Enemas and Suppositories inPremature Infants: A Meta-analysisMichael H. Livingston, MDa,b, Anna C. Shawyer, MD, MSc, FRCSCc, Peter L. Rosenbaum, MD, FRCPCd,e,Connie Williams, MD, MSc, FRCPCe, Sarah A. Jones, MD, PhD, FRCSCf, J. Mark Walton, MD, FRCSCa,g

abstractBACKGROUND AND OBJECTIVE: Premature infants are often given glycerin enemas or suppositories tofacilitate meconium evacuation and transition to enteral feeding. The purpose of this studywas to assess the available evidence for this treatment strategy.

METHODS:We conducted a systematic search of Medline, Embase, Central, and trial registries forrandomized controlled trials of premature infants treated with glycerin enemas orsuppositories. Data were extracted in duplicate and meta-analyzed using a random effectsmodel.

RESULTS: We identified 185 premature infants treated prophylactically with glycerin enemas inone trial (n = 81) and suppositories in two other trials (n = 104). All infants were less than32 weeks gestation and had no congenital malformations. Treatment was associated withearlier initiation of stooling in one trial (2 vs 4 days, P = .02) and a trend towardsearlier meconium evacuation in another (6.5 vs 9 days, P = .11). Meta-analysis demonstratedno effect on transition to enteral feeding (0.7 days faster, P = .43) or mortality (P = 0.50).There were no reports of rectal bleeding or perforation but there was a trend towards increasedrisk of necrotizing enterocolitis with glycerin enemas or suppositories (risk ratio = 2.72, P = .13).These three trials are underpowered and affected by one or more major methodological issues.As a result, the quality of evidence is low to very low. Three other trials are underway.

CONCLUSIONS: The evidence for the use glycerin enemas or suppositories in premature infants ininconclusive. Meta-analyzed data suggest that treatment may be associated with increasedrisk of necrotizing enterocolitis. Careful monitoring of ongoing trials is required.

aMcMaster Pediatric Surgery Research Collaborative, bClinician Investigator Program, dCanChild Center for Childhood Disability Research, eDepartment of Pediatrics, and gDivision ofPediatric Surgery, McMaster University, Hamilton, Ontario, Canada; Divisions of fPediatric Surgery, Western University, London, Ontario, Canada; and cDivision of Pediatric Surgery, AlbertaChildren’s Hospital, Calgary, Alberta, Canada

Dr Livingston designed the study, reviewed records, analyzed and interpreted data, and drafted the initial manuscript; Dr Shawyer reviewed records, extracted andinterpreted data, and reviewed and revised the manuscript; and Drs Rosenbaum, Williams, Jones, and Walton helped conceptualize the study, interpreted data, andreviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

www.pediatrics.org/cgi/doi/10.1542/peds.2015-0143

DOI: 10.1542/peds.2015-0143

Accepted for publication Mar 18, 2015

Address correspondence to J. Mark Walton, MD, FRCSC, McMaster Children’s Hospital, 1200 Main St West, Room 4E3, Hamilton, ON, Canada, L8N 3Z5. E-mail: waltonj@mcmaster.ca

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

Copyright © 2015 by the American Academy of Pediatrics

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

FUNDING: Dr Livingston is supported by the Clinician Investigator Program at McMaster University, funded by the Ontario Ministry of Health and Long-Term Care; andDr Rosenbaum is a Canada Research Chair in Childhood Disability Research, Dissemination and Mentoring. These agencies had no role in selecting the topic ordeveloping the content of this review.

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

PEDIATRICS Volume 135, number 6, June 2015 REVIEW ARTICLE by guest on January 5, 2020www.aappublications.org/newsDownloaded from

Feeding and nutrition are significantchallenges for premature infants inthe NICU.1–8 These patients oftenreceive glycerin enemas orsuppositories to stimulate thepassage of meconium and improvefeeding tolerance.9 This practice isbased on the observation that manypremature infants experiencesignificant delays in the passage ofmeconium, which is more viscousthan normal stool.10,11 Delays inmeconium evacuation seem to beassociated with a delay in thetransition to enteral feeding.12 Ifmeconium evacuation could beexpedited by using glycerin enemasor suppositories, this approach mightlead to faster transition to enteralfeeding, decreased reliance onparenteral nutrition, and betterclinical outcomes.13 Unfortunately,there is little evidence to support thispractice and to guide treatmentdecisions.9,14

The objective of the present articlewas to assess the level of evidenceregarding the use of glycerin enemasand suppositories in prematureinfants by updating the systematicreview on this topic. We consideredthe results from randomizedcontrolled trials of glycerin enemas orsuppositories, and data underwentmeta-analysis whenever possible. Thegray literature and trial registrieswere also searched to identify trialsthat are underway or have not yetbeen published.

METHODS

Search Strategy

A systematic search was conducted ofMedline, Embase, and the CochraneCentral Register of Controlled Trialsfor randomized controlled trials ofinfants treated with glycerin enemasor suppositories. An experiencedmedical librarian developed queriesfor each database to identify studiesthat mentioned “premature infants”and “glycerin laxatives.” Theseconcepts were expanded to ensurethat no studies were missed due to

variation in syntax and nomenclature.Appendices 1 through 3 present thesearch strategies and results.

We also performed manual searchesof conference proceedings,15 thesesand dissertations,16–18 and trialregistries.19–22 We included allcitations up until July 2014. Nolanguage limits were placed on thesearch.

Study Selection

Title and abstract screening wascompleted independently and induplicate by the first 2 authors(M.H.L. and A.C.S.). The defaultapproach for any disagreements wasto automatically include the record.Inclusion criteria were: (1)participants who were prematureinfants ,32 weeks’ gestation and/orwho had a birth weight ,1500 g; (2)interventions that were glycerinenemas or suppositories usedprophylactically or as rescue therapyfor jaundice or feeding intolerance;and (3) studies that were randomizedcontrolled trials. Glycerin enemas andglycerin suppositories were selectedas important subgroups a priori. Thereview protocol is available from theauthors upon request.

Data Extraction

Studies selected for full-text reviewunderwent data extractionindependently and in duplicate by thefirst 2 authors (M.H.L. and A.C.S.). Toensure consistency, we developedstandardized data collection formsbefore use. Participant data includedgestational age, birth weight, gender,presence of congenital anomalies,type of feeding (breastfeeding versusformula), age at start of enteralfeeding, and calendar years ofrecruitment. Interventioncharacteristics included treatmenttype (glycerin enemas versussuppositories), type of placebo (nointervention versus sham procedure),dose, and treatment duration.

Outcomes of interest were selecteda priori. These included (in order ofdecreasing importance): mortality,

necrotizing entercolitis (NEC), rectalperforation, rectal bleeding, feedingintolerance, jaundice, transition toenteral feeding, and meconiumevacuation. These outcomes wereselected because they are clinicallyrelevant and/or likely to be affectedby the use of glycerin enemas orsuppositories. Mortality, NEC, andrectal perforation were specifieda priori as critical outcomes.

Statistical Analysis

Outcome data were pooled wheneverpossible in Review Manager version5.2.23 We decided a priori to usea random effects model becausemoderate heterogeneity wasanticipated between studies due tovariation in the participants and typesof interventions. Summary statisticswere reported as risk ratios (RRs) fordichotomous outcomes and meandifferences for continuous outcomes.We also reported 95% confidenceintervals (CIs) and P values.

Heterogeneity was reportedquantitatively by using the I2 statistic:0% to 40% might not be important(ie, low heterogeneity), 30% to 60%may represent moderateheterogeneity, 50% to 90% mayrepresent substantial heterogeneity,and 75% to 100% representsconsiderable heterogeneity.24 Theimportance of the observed value(eg, “low” versus “moderate”)depends on: (1) magnitude anddirection of effects; and (2) strengthof evidence for heterogeneity (eg,P value from the x2 test).

For the purposes of meta-analysis,SDs were estimated whenever theywere not reported explicitly by usingthe following approach: range/4(when n = 15–70) and range/6 (whenn .70).25 Given the low number ofstudies meeting our inclusion criteria,we decided not to produce a funnelplot for any of the pooled outcomes.24

Risk of Bias Assessment forIndividual Trials

Risk of bias for individual trials wasassessed independently and in

1094 LIVINGSTON et al by guest on January 5, 2020www.aappublications.org/newsDownloaded from

duplicate by using the CochraneCollaboration’s tool for assessing riskof bias.24 This instrument consists of6 domains that classify risk of bias aslow, unclear, or high. Differencesbetween reviewers were resolvedthrough discussion and consensus.

Quality of Evidence Across Studies

The quality of evidence for eachoutcome was reported by using theGrading of RecommendationsAssessment, Development andEvaluation (GRADE) system26 withGRADEpro version 3.6.27 This systemtakes into account findings frommultiple studies and grades thequality of evidence for each outcomeas high (4 of 4 points), moderate (3 of4), low (2 of 4), or very low (1 of 4).By default, the quality of evidenceis high for results from randomizedcontrolled trials and is low for resultsfrom observational studies. Ratingscan be downgraded due to risk ofbias, inconsistency, indirectness,imprecision, or publication bias andupgraded due to the presence ofa large effect (RR ,0.5 or .2), ifattempts to control for confoundingwould change the observed effect,and the presence of a dose–responsegradient.

RESULTS

Search Results

Our systematic search of Medline,Embase, and the Cochrane CentralRegister of Controlled Trials yielded68 titles and abstracts, which wasreduced to 45 records afterduplicates were removed. Titles andabstracts were screenedindependently and in duplicate.

Six records were selected for full-textreview. Two of these were excludedbecause they were not randomizedcontrolled trials.13,28 Another studywas excluded because theintervention comprised glycerinenemas as well as oral probiotics3 times a day for 7 days.29 Becauseprobiotics have been shown todecrease the risk of NEC and all-cause

mortality in preterm infants,combined treatment likely hasdifferent effects than using glycerinsuppositories alone.30 Our selectionreview resulted in 3 randomizedcontrolled trials of glycerinenemas (1 trial) or suppositories(2 trials).14,31,32 An overview of thestudy selection procedure is shownin Fig 1.

Participants

The 3 randomized controlled trialsincluded 185 premature infants witha gestational age ,32 weeks and/orbirth weight ,1500 g. One trialincluded infants between 28 and32 weeks’ gestational age only.32 All3 trials specifically excluded infantswith major congenital malformationsor structural gastrointestinalanomalies. One of the glycerinsuppository trials also excludedpremature infants with hypoxic-ischemic encephalopathy stage $2.31

The other trial excluded those withany signs of hemodynamicinstability.32

Interventions

Study interventions consisted ofprophylactic glycerin enemas orsuppositories administered once dailyfor several days. Each trialestablished slightly different dosing,treatment start date, and durationcharacteristics (Table 1). In the trialof glycerin enemas, study treatmentsstarted at 12 hours of life andcontinued until meconium evacuationwas complete (defined as 2 normalstools free of meconium staining).14

No maximum treatment duration wasreported. In the 2 trials of glycerinsuppositories, study interventionsstarted at 24 (or 48) hours of life andcontinued for a total of 10 (or 13)days of treatment, regardless ofstooling or meconiumevacuation.31,32

FIGURE 1Search results and study selection procedures according to the Preferred Reporting Items forSystematic Reviews and Meta-Analyses guidelines.49

PEDIATRICS Volume 135, number 6, June 2015 1095 by guest on January 5, 2020www.aappublications.org/newsDownloaded from

Participants in the control group in 2of the trials were assigned to nointervention.14,31 In Shinde et al,32

participants in the control group weretreated with a sham procedure, whichinvolved opening the diaper andclosing it again (K. Nandkishor, MD,MSc, personal communication, 2015).

Mortality

Mortality rates between treatmentgroups were similar in all 3 trials andranged from 5% to 17% (N. Haiden,MD, MSc, personal communication,2015).31,32 Mortality data in theglycerin enema trial were confirmedafter communication with theprincipal investigator (N. Haiden,personal communication, MD, MSc,2015). Meta-analysis of data from all3 trials revealed no differencebetween treatment groups (RR: 1.34[95% CI: 0.58–3.11]; I2 = 0%, P = .50)(Fig 2).

Necrotizing Enterocolitis

The rates of NEC between treatmentgroups were similar in all 3 trials andranged from 5% to 9%.14,31,32 Thediagnosis of NEC was based on Bellstage33 $1 in Shinde et al32 and $2in the other 2 trials (N. Haiden,personal communication, MD, MSc,2015).31 In Shinde et al, the rate ofNEC was reported only for infantswho survived or were not transferredto another hospital before reachingfull enteral feedings.32 As a result,there were 12 cases of NEC among179 premature infants (7% overall);10% (9 of 93) of these occurred afterthe use of glycerin enemas orsuppositories and 4% (3 of 86)occurred in the control group. Meta-analysis approached statisticalsignificance for increased risk of NECwith the use of glycerin enemas orsuppositories (RR: 2.72 [95% CI:0.76–9.81]; I2 = 0%, P = .13) (Fig 3).

Rectal Perforation and Bleeding

There were no cases of rectalperforation in any of the 3 trials.Absence of rectal perforation wasreported explicitly in 2 of theTA

BLE1

Characteristicsof

Random

ized

ControlledTrialsof

Prem

atureInfantsTreatedWith

Glycerin

Enem

asor

Suppositories

Trial

Participants

NYearsRecruited

Country

Intervention

Protocol

Haiden

etal,142007

Inclusioncriteria:

812000–2001

aAustria

1.Daily

glycerinenem

a(n

=42);10

mL/kg

(0.8gof

glycerinin

10mLof

norm

alsaline)

Notregistered

a

•Gestationalage24–32

wk

•Birthweight#1500

gExclusioncriteria:

2.No

intervention(n

=39)

•Major

congenitalanom

alies

•Gastrointestinal

anom

alies

Khadret

al,312011

Inclusioncriteria:

542006–2008

UnitedKingdom

1.Daily

glycerin

suppository

(n=29)

InternationalStandard

Random

ized

ControlledTrialNumber4706576434

•Gestationalage24–32

wk

250mg(24–28

wk)

Exclusioncriteria:

500mg(28–32

wk)

•Major

congenitalanom

alies

2.No

intervention(n

=25)

•Gastrointestinal

anom

alies

•Hypoxic-ischem

icencephalopathy

stage$2

Shinde

etal,322014

Inclusioncriteria:

502010–2011

India

1.Daily

glycerin

suppository

(n=25)

Notregistered

b

•Gestationalage28–32

wk

1000

mg(28–32

wk)

•Birthweight1000–1500

g2.Sham

procedure(n

=25)

Exclusioncriteria:

Open

andclosingdiaper

b

•Major

congenitalanom

alies

•Gastrointestinal

anom

alies

•Hemodynam

icinstability

aN.

Haiden,p

ersonalcommunication,

MD,

MSc,2015.

bK.Nandkishor,personalcommunication,

MD,

MSc,2015.

1096 LIVINGSTON et al by guest on January 5, 2020www.aappublications.org/newsDownloaded from

trials14,31 and was confirmed in theother trial after communication withits principal investigator (N. Haiden,personal communication, MD, MSc,2015).

There were no cases of rectalbleeding. Absence of rectal bleedingwas reported explicitly in 1 of theglycerin suppository trials31 and wasconfirmed in the other 2 studies

(K. Nandkishor, MD, MSc, personalcommunication, 2015; N. Haiden, MD,MSc, personal communication, 2015).

Feeding Intolerance

There were no differences betweentreatment groups in terms of feedingintolerance. This outcome wasreported in the 2 glycerin suppositorytrials but was variably defined.7,31,32

In Khadr et al,31 there were nodifferences between groups in terms ofincidence of abdominal distension .2cm, number of bilious residuals, numberof feedings withheld, number offeedings reduced or not increased, andpercentage of gastric residualscompared with total enteral feedingvolume. These outcomes were allmeasured within the first 10 days of life.

FIGURE 2Risk of mortality in premature infants treated with glycerin enemas or suppositories versus no treatment. M-H, Mantel-Haenszel.

FIGURE 3Risk of NEC in premature infants treated with glycerin enemas or suppositories versus no treatment. M-H, Mantel-Haenszel.

PEDIATRICS Volume 135, number 6, June 2015 1097 by guest on January 5, 2020www.aappublications.org/newsDownloaded from

In Shinde et al,32 there was nodifference between treatment groupsin terms of the number ofparticipants who had feedingswithheld: 7 of 21 versus 4 of 21 (RR:1.75 [95% CI: 0.60–5.10]; P value notreported). Differences in how feedingintolerance was reported in these 2trials resulted in data that could notundergo meta-analysis.

Jaundice

None of the trials reported outcomesrelated to jaundice, such as or serumbilirubin level or need forphototherapy.

Transition to Enteral Feeding

There were no differences betweentreatment groups in transition toenteral feeding in any of the 3 trials.Each trial defined full enteral feedingdifferently: in Haiden et al,14150mL/kg per day; in Shinde et al,32

maintenance of 180 mL/kg per dayfor 24 hours; and in Khadr et al,31

tolerance of full enteral feedings(prespecified volume not defined)and discontinuation of parenteralnutrition for .48 hours withoutfeedings being reduced or withheld.There were also differences infeeding regimens, with feedinginitiated on median day 1 (range:1–1.8 days),31 median day 2 (range:

0–9 days),14 or mean day 5.5 (SD:2.5).32

The 2 glycerin suppository trials alsoreported the type of feeding used. Inthe trial by Khadr et al,31 whichincluded premature infants 24 to32 weeks’ gestation, expressedbreast milk was used exclusively in10 (19%) of 54 infants. In Shindeet al,32 which included prematureinfants 28 to 32 weeks’ gestationonly, 37 (74%) of 50 receivedexpressed breast milk exclusively.Communication with the principalinvestigator of the glycerin enematrial indicated that the use ofexpressed breast milk when thestudy was conducted was ∼70% to80% (N. Haiden, MD, MSc, personalcommunication, 2015).

For the purposes of meta-analysis,transition to enteral feeding wascalculated as the difference (innumber of days) between the meanor median start of enteral feeding andfull enteral feeding. Across all 3 trials,there was no statistically significantdifference between treatment groups(0.7 day faster with glycerin enemasor suppositories; 95% CI of 2.4days faster to 1.0 day slower, P = .49)(Fig 4). Heterogeneity was substantial(I2 = 51%), but there was nostatistically significant subgroup

difference between glycerin enemasand suppositories (P = .89).

Completion and Initiation ofMeconium Evacuation

The trial of glycerin enemasreported a nonsignificant trendtoward earlier completion ofmeconium evacuation with activetreatment (median: 6.5 vs 9 days;P = .11).14 The other 2 trials did notreport this outcome.31,32

Initiation of meconium evacuationwas not affected by the use ofglycerin enemas starting at 12 hoursof life (median: 1 vs 1 day; P = .68),14

but a statistically significant effectwas observed for the use of glycerinsuppositories starting at 24 hours oflife (median: 2 vs 4 days; P = .016).31

In the glycerin suppository trial,treatment was also associated withlower frequency of delayed initiationof meconium evacuation (24% vs64%; P = .003).

Meta-analysis of data for initiation ofmeconium evacuation demonstratedno significant treatment effect (1 dayfaster with treatment, 95% CI of 3.0days faster to 0.9 day slower; I2 = 89%,P = .30) (Fig 5). The CIs from the2 trials did not overlap, and the testfor subgroup differences wassignificant (P = .002).

FIGURE 4Transition to enteral feeding (mean difference in days to full enteral feeding) in premature infants treated with glycerin enemas or suppositories versusno treatment. IV, inverse variance.

1098 LIVINGSTON et al by guest on January 5, 2020www.aappublications.org/newsDownloaded from

Other Outcomes

There were no differences betweentreatment groups in terms ofintraventricular hemorrhage,retinopathy of prematurity, patentductus arteriosus, sepsis based onpositive culture results, or oxygenrequirements (Table 2) (N. Haiden,personal communication, MD, MSc,2015).31 There were no differencesreported between groups for weightgain or length of hospital stay in anyof the 3 trials.14,31,32

Risk of Bias of Individual Studies

All 3 trials were at high risk of biasin $2 of 6 domains (Table 3).Sequence generation was created byusing random number software in 2of the trials (K. Nandkishor, MD,MSc, personal communication, 2015;N. Haiden, MD, MSc, personalcommunication, 2015) and shufflingof sealed envelopes in Khadr et al.31

Allocation concealment wasmaintained in all 3 studies by usingopaque envelopes (K. Nandkishor,MD, MSc, personal communication,2015; N. Haiden, MD, MSc, personalcommunication, 2015).32 Two of the3 trials were open-label studies withno blinding.14,31 Shinde et al32

relied on a research nurse toadminister study interventions, andall other clinicians, study personnel,

and outcome assessors wereblinded to treatment allocation(K. Nandkishor, personalcommunication, 2015).

Incomplete outcomes were discussedin all 3 trials. In the most recentglycerin suppository trial, 3participants in each group (.10% ofthe total sample size) weretransferred to another hospital beforecomplete outcomes could beobtained.32 A sensitivity analysis wasperformed for this situation byimputing missing data for mortalityand NEC by using the approach of“best- and worst-case scenario.”24

This method did not affect the meta-analysis for mortality. For NEC,however, the worst-case scenario(ie, 3 of 3 infants in the treatmentgroup developing NEC versus 0 of 3in the control group) would havemade our meta-analysis statisticallysignificant (RR: 3.68 [95% CI:1.07–12.65]; P = .04). As a result, weconcluded that the lack of completeoutcome data for this trial resulted ina high risk of bias.

The protocol for Khadr et al31 wasregistered and available online,34 andthere was no evidence of selectivereporting. The other 2 trials were notregistered, and the risk of bias due toselective reporting was thereforeunclear (K. Nandkishor, MD, MSc,

personal communication, 2015;N. Haiden, MD, MSc, personalcommunication, 2015).

All 3 trials were at high risk of biasdue to low power. The glycerin enematrial was powered to detect a “30%difference” in days to completemeconium evacuation.14 The glycerinsuppository trials were powered todetect a reduction in days to fullenteral feeding of 3.63 and 3 days,respectively.31,32 These effect sizesare substantially larger than themagnitude of the true effect size (ifit actually exists). Powering thesetrials for a more moderate effect sizewould have required substantiallylarger sample sizes.

The glycerin enema trial was also athigh risk of bias due to the numberof protocol violations: 15 of 42infants in the intervention groupmissed a scheduled glycerin enema,and 8 of 39 in the control groupreceived at least 1 enema despitebeing assigned to no treatment.14

Outcomes were reported on the basisof intention-to-treat and per-protocol analyses, but there were nosignificant differences between theseapproaches. The other 2 trials alsoexamined the results on the basis ofan intention-to-treat analysis(K. Nandkishor, MD, MSc, personalcommunication, 2015).31

FIGURE 5Mean difference in days to initiation of meconium evacuation in premature infants treated with glycerin enemas or suppositories versus no treatment.IV, inverse variance.

PEDIATRICS Volume 135, number 6, June 2015 1099 by guest on January 5, 2020www.aappublications.org/newsDownloaded from

Quality of Evidence

The quality of evidence across all 3trials for mortality, NEC, transition toenteral feeding, and completion ofmeconium evacuation was low (2 of 4points). Using the GRADE approach,the quality of evidence wasdowngraded for all outcomes due torisk of bias (–1 point) (Table 3).35,36

Ratings were also downgraded forimprecision by 1 point for lownumber events (for mortality andNEC) and/or 1 point for CIs thatcrossed 1 (for NEC, transition toenteral feeding, and completion ofmeconium evacuation).37 The qualityof evidence for NEC was upgradeddue to the presence of a large effect(+1 point).38

The quality of evidence for rectalbleeding, rectal perforation, andinitiation of meconium evacuationwas very low (1 of 4 points)(Table 4). This outcome was due torisk of bias (–1 point) andimprecision due to low number ofevents (–1 point) and/or CIs thatcrossed 1 (–1 point).36,37 The qualityof evidence for initiation ofmeconium evacuation was alsodowngraded due to inconsistencybecause the CIs from the 2 trials didnot overlap.39 The other possibleexplanation for this lack of overlap isthat glycerin enemas and glycerinsuppositories may have trulydifferent effects on initiation ofmeconium evacuation.

Ongoing Trials

Our systematic search of the grayliterature and trial registries resultedin 77 records. Three of these recordswere protocols for ongoingrandomized controlled trials ofpremature infants treatedprophylactically with glycerinsuppositories. The first trial recruited79 premature infants from 30 to35 weeks’ gestation who requiredphototherapy for physiologichyperbilirubinemia.40 The primaryoutcome was total duration ofphototherapy; secondary outcomesincluded duration of initialphototherapy, need to restartphototherapy, peak serum bilirubin,mean serum bilirubin, rate of declineof serum bilirubin, and use of stoolsofteners after completion ofphototherapy. Recruitment wascompleted in 2013, and the resultsare pending.

We also identified a 2-center trial ofglycerin suppositories from SaudiArabia with plans to recruit 220premature infants with a birth weight,1250 g.41 The primary outcome isdays to full enteral feeding;recruitment began in 2013. InJanuary 2015, our center startedrecruitment for a pilot, randomizedcontrolled trial of 30 prematureinfants of 24 to 32 weeks’ gestationand/or birth weight 500 to 1500 g.42

The purpose of this study is to furtherassess the feasibility and safety ofglycerin suppositories beforeembarking on a larger, multicentertrial.

DISCUSSION

The quality of evidence for the use ofglycerin enemas and suppositories inpremature infants is low to very low.Three single-center, randomizedcontrolled trials have been conductedand published to date.14,31,32 Thesestudies are underpowered and at riskfor bias due to a variety ofmethodologic issues, includingproblems with blinding, outcomeassessment, incomplete follow-up,

TABLE 2 Outcomes Reported Among Included Trials

Outcomes Reported Haiden et al,14 2007 Khadr et al,31 2011 Shinde et al,32 2014

Relevant outcomesa

Mortality Yesb Yes YesNEC Yes Yes YesRectal perforation Yes Yes Yesb

Rectal bleeding Yesb Yes Yesb

Feeding intolerance Yes YesTransition to enteral feeding Yes Yesc Yesc

Completion of meconium evacuation Yesc,d

Initiation of meconium evacuation Yes Yesc,e

Other outcomesIntraventricular hemorrhage Yesb YesRetinopathy of prematurity YesPatent ductus arteriosus Yesb YesSepsis based on positive culture results YesOxygen requirements YesLength of stay Yes Yes YesWeight gain Yes Yes Yes

a Clinically relevant and/or likely to be affected by use of glycerin enemas or suppositories.b Confirmed after communication with principal investigator (K. Nandkishor, MD, MSc, personal communication, 2015;N. Haiden, MD, MSc, personal communication, 2015).c Primary outcome.d Trend toward faster completion of meconium evacuation with glycerin enemas (median: 6.5 vs 9 days; P = .11).e Faster initiation of meconium evacuation with glycerin suppositories (median: 2 vs 4 days; P = .016).

TABLE 3 Risk of Bias of Individual Studies According to the Cochrane Collaboration Risk of BiasTool24

Domain Risk of Bias

Haiden et al,14 2007 Khadr et al,31 2011 Shinde et al,32 2014

Random sequence generation Low Low LowAllocation concealment Low Low LowBlinding of patients and personnel High High LowBlinding of outcome assessment High High LowIncomplete outcomes addressed Low Low HighNo selective reporting Unclear Low UnclearNo other bias Higha,b Higha Higha

a Low statistical power.b Frequent protocol violations.

1100 LIVINGSTON et al by guest on January 5, 2020www.aappublications.org/newsDownloaded from

TABLE4

GRADEEvidence

Profile33

Quality

Assessment

No.ofParticipants

Effect

Quality

Importance

No.ofStudies(n

=participants)

Design

Risk

ofBias

Inconsistency

Indirectness

Imprecision

Other

Considerations

Glycerin

Laxatives

NoTreatm

ent

Relative

(95%

CI)

Absolute

Mortality

3(n

=179)

Random

ized

trials

Seriousa

None

None

Seriousb

None

12/93(13%

)8/86

(9%)

RR:1.34

(0.58–3.11)

32moreper1000

(39fewer

to196more)

ÅÅΟΟ

Low

Critical

NEC 3(n

=179)

Random

ized

trials

Seriousa

None

None

Very

seriousb

,cLargeeffectd

9/93

(10%

)3/86

(4%)

RR2.72:

(0.76–9.81)

60moreper1000

(8fewer

to307more)

ÅÅΟΟ

Low

Critical

Rectal

perforation

3(n

=185)

Random

ized

trials

Seriousa

None

None

Very

seriousb

,cNone

0/96

(0%)

0/89

(0%)

——

ÅΟΟΟ

Very

low

Critical

Rectal

bleeding

3(n

=185)

Random

ized

trials

Seriousa

None

None

Very

seriousb

,cNone

0/96

(0%)

0/89

(0%)

——

ÅΟΟΟ

Very

low

Important

Jaundice

0(n

=0)

——

——

——

——

——

—Important

Transitionto

enteral

feeding

3(n

=177)

Random

ized

trials

Seriousa

None

None

Seriousb

None

9285

—0.7dfaster

(2.4faster

to1.0

slow

er)

ÅÅΟΟ

Low

Important

Completionof

meconiumevacuation

1(n

=81)

Random

ized

trials

Seriousa

None

None

Seriousb

None

4239

—2.5dfaster

(P=.11,CI

not

reported)

ÅÅΟΟ

Low

Important

Initiationof

meconium

evacuation

2(n

=135)

Random

ized

trials

Seriousa

Seriouse

None

Seriousb

None

7164

—1dfaster

(3.0faster

to0.9

slow

er)

ÅΟΟΟ

Very

low

Important

aSeeTable3.

b95%

CIsforrelativerisk

include13

5or

P.

.05.

cSm

allnumberof

events

(,300).35

dRR

.2.36

eNo

overlapbetweenCIsfrom

different

studies.37

PEDIATRICS Volume 135, number 6, June 2015 1101 by guest on January 5, 2020www.aappublications.org/newsDownloaded from

and/or the possibility of selectivereporting.24

The published data from these 3 trialssuggest that the use of glycerinenemas and suppositories has noeffect on transition to enteral feedingor mortality. One trial reported earlierinitiation of meconium evacuationwith glycerin suppositories31 andanother reported a trend towardearlier completion of meconiumevacuation with glycerin enemas.14

Although there were no reports ofrectal bleeding or perforation in these3 trials, our meta-analysis founda trend toward increased risk of NECwith the use of glycerin enemas orsuppositories compared with notreatment. The total number of eventsin each group was small (9 of 92 vs3 of 85), and although the risk ratiowas 2.72, the accuracy of thisestimate remains uncertain.

Given the methodologic limitations ofrandomized trials published to date,we believe that the evidence for theuse of glycerin enemas orsuppositories in premature infants isinconclusive. Meta-analysis of thedata suggests that treatment is notassociated with any consistentbenefits and may result in harm.These conclusions are similar to thoseof previous systematic reviews butinclude an additional message ofcaution given the concerning butnonsignificant trends found by ourmeta-analysis.

A systematic review published in2011 assessed whether glycerinenemas and suppositories decreasefeeding intolerance in prematureinfants.9 The authors considered theresults from the trial on glycerinenemas by Haiden et al14 anda historical cohort study of 83 verylow birth weight infants from a singlehospital in South Korea.13 In thecohort study, full enteral feeding wasachieved within 16 days during thetreatment period and 22.9 days forthe control period (P , .001).Significant improvements were alsoreported for days to passage of first

meconium, duration of centralcatheter usage, sepsis after 7 days oflife, and positive results of catheterculture. The authors of the systematicreview reported that the historicalcohort study was of fair quality andreported positive results, whereas therandomized controlled trial byHaiden et al was of good quality butreported negative results. Theauthors of the review concluded: “Theevidence regarding the effectivenessof glycerin [enemas or suppositories]for improving feeding intolerance invery low birth weight infants isinconclusive.”9

In 2014, an unpublished CochraneReview was conducted thatconsidered the results from the trialsby Haiden et al and Shindeet al.14,31,43 The authors concludedthat the available evidence does notsupport the use of glycerin enemas orsuppositories in clinical practice andthat further studies are needed.43

Previous trials and systematicreviews have not assessed the effectof glycerin enemas or suppositorieson jaundice in premature infants.Results are pending from a recentlycompleted trial of 79 prematureinfants treated with glycerinsuppositories to reduce jaundice.40

Three other randomized controlledtrials have also been conducted inhealthy term infants.44–46 Asystematic review of these 3 studieswas published in 2011 and concludedthat early meconium evacuationwith the use of glycerin suppositorieswas not associated with any clinicalbenefit in healthy term infants.47

Our search strategy identified 3 trialson the use of glycerin suppositories inpremature infants that are underwayor have not yet been published. Oneof these was recently completed, andthe results are pending.11,40 This trialdid not specify NEC as a primary orsecondary outcome a priori, butcommunication with the principalinvestigator confirmed that none ofthe 79 infants was diagnosed withNEC (C.T. D’Angio, MD, personal

communication, 2015). This outcomewas likely because the trial enrolledlate preterm (30 to 35 weeks’gestation) rather than early preterm(,32 weeks’ gestation) infants, whoare at higher risk for developing NEC.

The other 2 ongoing trials of glycerinsuppositories in premature infantshave started recruiting participantsand are ongoing. These trials shouldbe closely monitored for increasedrisk of NEC. The nature of theassociation between glycerinmedications and NEC will becomeclearer as additional data becomeavailable. One possibility is that theapparent relationship is nothing morethan a spurious correlation that willdisappear with the inclusion of moredata. The other explanation is thatthere is a real effect that has not yetbecome statistically significant. Ourcenter has elected to proceed witha trial as a pilot study limited to 30premature infants.42,48 This trial andothers like it should be discontinuedif it becomes clear that glycerinsuppositories are associated withincreased harm.

CONCLUSIONS

The quality of evidence for the use ofglycerin enemas and suppositories inpremature infants is low to very low.Previous randomized controlled trialsare underpowered and at risk forbias due to a variety of methodologicissues. As a result, the evidence forthis treatment is inconclusive.Meta-analysis of the data suggeststhat glycerin enemas andsuppositories have no consistenteffect on meconium evacuation,transition to full enteral feeding, ormortality. Previous trials haveassessed the effect of thesemedications on jaundice in terminfants but not in premature infants.No cases of rectal bleeding orperforation were reported in any ofthe trials published to date, but thesedata indicate a nonsignificant trendtoward increased risk of NEC whenglycerin enemas or suppositories are

1102 LIVINGSTON et al by guest on January 5, 2020www.aappublications.org/newsDownloaded from

used on a daily basis. We recommendthat ongoing trials be monitoredcarefully because treatment may beassociated with increased harm.

ACKNOWLEDGMENTS

We thank Alla Iansavichene, BSc,MLIS, for helping us to develop andexecute the search strategy. We alsoacknowledge Nathan Evaniew, MD,and Moin Khan, MD, for their reviewand constructive feedback.

REFERENCES

1. Su BH. Optimizing nutrition in preterminfants. Pediatr Neonatol. 2014;55(1):5–13

2. Belfort MB, Rifas-Shiman SL, Sullivan T,et al. Infant growth before and afterterm: effects on neurodevelopment inpreterm infants. Pediatrics. 2011;128(4).Available at: www.pediatrics.org/cgi/content/full/128/4/e899

3. Sharma R, Hudak ML. A clinicalperspective of necrotizing enterocolitis:past, present, and future. Clin Perinatol.2013;40(1):27–51

4. Raval MV, Hall NJ, Pierro A, Moss RL.Evidence-based prevention and surgicaltreatment of necrotizing enterocolitis-a review of randomized controlled trials.Semin Pediatr Surg. 2013;22(2):117–121

5. Gardner VA, Walton JM, Chessell L. A casestudy utilizing an enteral refeedingtechnique in a premature infant withshort bowel syndrome. Adv NeonatalCare. 2003;3(6):258–268, quiz 269–271

6. Al-Harbi K, Walton JM, Gardner V,Chessell L, Fitzgerald PG. Mucous fistularefeeding in neonates with short bowelsyndrome. J Pediatr Surg. 1999;34(7):1100–1103

7. Lucchini R, Bizzarri B, Giampietro S, DeCurtis M. Feeding intolerance in preterminfants. How to understand the warningsigns. J Matern Fetal Neonatal Med.2011;24(suppl 1):72–74

8. Morris BH, Miller-Loncar CL, Landry SH,Smith KE, Swank PR, Denson SE. Feeding,medical factors, and developmentaloutcome in premature infants. ClinPediatr (Phila). 1999;38(8):451–457

9. Shah V, Chirinian N, Lee S; EPIQ EvidenceReview Group. Does the use of glycerinlaxatives decrease feeding intolerance in

preterm infants? Paediatr Child Health(Oxford). 2011;16(9):e68–e70

10. Loening-Baucke V, Kimura K. Failure topass meconium: diagnosing neonatalintestinal obstruction. Am FamPhysician. 1999;60(7):2043–2050

11. Meetze WH, Franz AR, Linder W, PohlandtF. Meconium passage in very-lowbirthweight infants. J Parenter EnteralNutr. 1993;17:537–540

12. Mihatsch WA, Franz AR, Lindner W,Pohlandt F. Meconium passage inextremely low birthweight infants and itsrelation to very early enteral nutrition.Acta Paediatr. 2001;90(4):409–411

13. Shim SY, Kim HS, Kim DH, et al. Inductionof early meconium evacuation promotesfeeding tolerance in very low birthweight infants. Neonatology. 2007;92(1):67–72

14. Haiden N, Jilma B, Gerhold B, et al. Smallvolume enemas do not acceleratemeconium evacuation in very low birthweight infants. J Pediatr GastroenterolNutr. 2007;44(2):270–273

15. Conference Proceedings Citation Index.New York, New York: Thomson Reuters.Accessed August 1, 2014

16. Index to Theses. Cambridge, UnitedKingdom: ProqQuest. Accessed August 1,2014

17. ProQuest Dissertations & ThesesDatabase. Ann Arbor, Michigan:ProqQuest. Accessed August 1, 2014

18. OpenGrey. Vandoeuvre-lès-Nancy Cedex,France: Institut de l’InformationScientifique et Technique. Available at:www.opengrey.eu/. Accessed August 1,2014

19. ClinicalTrials.gov. Available at: http://clinicaltrials.gov/. Accessed August 1,2014

20. International Clinical Trials RegistryPlatform. Available at: http://apps.who.int/trialsearch/. Accessed August 1, 2014

21. European Union Clinical Trials Register.Available at: https://www.clinicaltrialsregister.eu/. AccessedAugust 1, 2014

22. International Standard RandomisedControlled Trial Number Registry.Available at: www.isrctn.org/. AccessedAugust 1, 2014

23. Review Manager. (RevMan) Version 5.2.Copenhagen, Denmark: The Nordic

Cochrane Centre, The CochraneCollaboration; 2012

24. Higgins JPT, Green S. CochraneHandbook for Systematic Reviews ofInterventions Version 5.1.0. The CochraneCollaboration, 2011. Available at: www.cochrane-handbook.org. AccessedJanuary 7, 2015

25. Hozo SP, Djulbegovic B, Hozo I. Estimatingthe mean and variance from the median,range, and the size of a sample. BMCMed Res Methodol. 2005;5:13

26. Guyatt GH, Oxman AD, Schünemann HJ,Tugwell P, Knottnerus A. GRADEguidelines: a new series of articles in theJournal of Clinical Epidemiology. J ClinEpidemiol. 2011;64(4):380–382

27. Grading of RecommendationsAssessment. Development andEvaluation (GRADE) profiler (GRADEpro)Version 3.6. GRADE Working Group; 2011.Available at: www.gradeworkinggroup.org/. Accessed April 3, 2015

28. Zenk KE, Koeppel RM, Liem LA.Comparative efficacy of glycerin enemasand suppository chips in neonates. ClinPharm. 1993;12(11):846–848

29. Wang G, Li H, Huang Y, Huang C, Tan H.Effect of early intervention with glycerineenema and golden diplococci on feedingintolerance in very low birth weightinfants. J Applied Clinical Pediatrics.2008;14:1110–1111, 1135

30. AlFaleh K, Anabrees J. Probiotics forprevention of necrotizing enterocolitis inpreterm infants. Cochrane DatabaseSyst Rev. 2014;4:CD005496

31. Khadr SN, Ibhanesebhor SE, Rennix C,et al. Randomized controlled trial:impact of glycerin suppositories on timeto full feeds in preterm infants.Neonatology. 2011;100(2):169–176

32. Shinde S, Kabra NS, Sharma SR, AvasthiBS, Ahmed J. Glycerin suppository forpromoting feeding tolerance inpreterm very low birthweightneonates: a randomized controlledtrial. Indian Pediatr. 2014;51(5):367–370

33. Bell MJ, Ternberg JL, Feigin RD, et al.Neonatal necrotizing enterocolitis.Therapeutic decisions based uponclinical staging. Ann Surg. 1978;187(1):1–7

34. Ibhanesebhor S. Tolerance of milk feedsin preterm, very low birth weight babies:

PEDIATRICS Volume 135, number 6, June 2015 1103 by guest on January 5, 2020www.aappublications.org/newsDownloaded from

the use of glycerine suppositories.International Standard RandomisedControlled Trial Number Registry.Available at: www.isrctn.com/ISRCTN47065764. Accessed August 1,2014

35. Guyatt G, Oxman AD, Akl EA, et al; GRADEWorking Group. GRADE guidelines: 1.Introduction-GRADE evidence profiles andsummary of findings tables. J ClinEpidemiol. 2011;64(4):383–394

36. Guyatt GH, Oxman AD, Vist G, et al; GRADEWorking Group. GRADE guidelines: 4.Rating the quality of evidence—studylimitations (risk of bias). J ClinEpidemiol. 2011;64(4):407–415

37. Guyatt GH, Oxman AD, Kunz R, et al;GRADE Working Group. GRADE guidelines6. Rating the quality of evidence—imprecision. J Clin Epidemiol. 2011;64(12):1283–1293

38. Guyatt GH, Oxman AD, Sultan S, et al;GRADE Working Group. GRADE guidelines:9. Rating up the quality of evidence. JClin Epidemiol. 2011;64(12):1311–1316

39. Guyatt GH, Oxman AD, Kunz R, et al;GRADE Working Group. GRADEguidelines: 7. Rating the quality of

evidence—inconsistency. J ClinEpidemiol. 2011;64(12):1294–1302

40. D’Angio C. Glycerin suppositories toreduce jaundice in premature infants.NCT01746511. Available at: https://www.clinicaltrials.gov/ct2/show/NCT01746511.Accessed January 7, 2015

41. Khadawardi EM. Efficacy of prophylacticglycerin suppositories for feedingintolerance in very low birth weightpreterm infants: a randomized trial.NCT01799629. Available at: https://www.clinicaltrials.gov/ct2/show/NCT01799629.Accessed January 7, 2015

42. Walton JM. Glycerin suppositories usedprophylactically in premature infants(SUPP): a pilot randomized placebo-controlled trial. NCT02153606. Availableat: https://clinicaltrials.gov/ct2/show/NCT02153606. Accessed January 7, 2015

43. Anabrees J, Shah VS, Al Osaimi A, AlFalehK. Glycerin laxatives for prevention ortreatment of feeding intolerance in verylow birth weight infants. CochraneDatabase Syst Rev. 2014;(3):CD010464

44. Weisman LE, Merenstein GB, Digirol M,Collins J, Frank G, Hudgins C. The effect ofearly meconium evacuation on early-onset

hyperbilirubinemia. Am J Dis Child. 1983;137(7):666–668

45. Chen JY, Ling UP, Chen JH. Earlymeconium evacuation: effect on neonatalhyperbilirubinemia. Am J Perinatol. 1995;12(4):232–234

46. Bader D, Yanir Y, Kugelman A, Wilhelm-Kafil M, Riskin A. Induction of earlymeconium evacuation: is it effective inreducing the level of neonatalhyperbilirubinemia? Am J Perinatol.2005;22(6):329–333

47. Srinivasjois R, Sharma A, Shah P, Kava M.Effect of induction of meconiumevacuation using per rectal laxatives onneonatal hyperbilirubinemia in terminfants: a systematic review ofrandomized controlled trials. Indian JMed Sci. 2011;65(7):278–285

48. Thabane L, Ma J, Chu R, et al. A tutorialon pilot studies: the what, why and how.BMC Med Res Methodol. 2010;10:1

49. Moher D, Liberati A, Tetzlaff J, AltmanDG; PRISMA Group. Preferred reportingitems for systematic reviews andmeta-analyses: the PRISMA statement.J Clin Epidemiol. 2009;62(10):1006–1012

1104 LIVINGSTON et al by guest on January 5, 2020www.aappublications.org/newsDownloaded from

APPENDIX 1 Medline Search Strategy and Results

No. Searches Results

1 exp glycerol/ 34 3792 (glycerol$ or glycerin$ or glyrol or microglycerin or

ophthalgan or osmoglyn or rovi or vilardell).tw.42 071

3 or/1-2 62 9794 suppositories/ 36865 (suppository or suppositories or suppositorium$).tw. 39006 or/4-5 52767 enema/ 64138 (enema$ or clysma$ or clyster$ or enteroclysis).tw. 93659 or/7-8 11 99510 3 and (6 or 9) 1511 exp Infant/ 942 47012 ((premature$ or pre-mature$ or prematuritas or preterm$1

or pre-term$1 or (low adj$ wt$) or low-birth weight$ orunderweight$) adj5 (infant$ or neonate$ or neonatal$ orbaby or babies or birth or child or children or childbirthor newborn$)).mp.

99 818

13 (infant$ or neonate$ or neonatal$ or baby or babies ornewborn$).tw.

544867

14 or/11-13 1 164 29715 10 and 14 20

APPENDIX 2 Embase Search Strategy and Results

No. Searches Results

1 glycerol/ 34 9232 (glycerol$ or glycerin$ or glyrol or microglycerin or

ophthalgan or osmoglyn or rovi or vilardell).tw.50 661

3 or/1-2 61 9424 suppository/ 51685 (suppository or suppositories or suppositorium$).tw. 59796 or/4-5 75967 exp enema/ 84988 (enema$ or clysma$ or clyster$ or enteroclysis).tw. 14 2269 or/7-8 17 34110 3 and (6 or 9) 35311 prematurity/ or exp low birth wt/ or newborn/ or infant/ 954 35512 ((premature$ or pre-mature$ or prematuritas or preterm$1

or pre-term$1 or (low adj3 wt$) or low-birth weight$ orunderweight$) adj5 (infant$ or neonate$ or neonatal$ orbaby or babies or birth or child or children or childbirthor newborn$)).mp.

115 604

13 (infant$ or neonate$ or neonatal$ or baby or babies ornewborn$).tw.

711 577

14 or/11-13 1 221 29815 10 and 14 38

PEDIATRICS Volume 135, number 6, June 2015 1105 by guest on January 5, 2020www.aappublications.org/newsDownloaded from

APPENDIX 3 Cochrane Central Register of Controlled Trials Search Strategy and Results

No. Searches Results

1 exp glycerol/ 22322 (glycerol$ or glycerin$ or glyrol or microglycerin or

ophthalgan or osmoglyn or rovi or vilardell).tw.1083

3 or/1-2 28704 suppositories/ 5375 (suppository or suppositories or suppositorium$).tw. 11196 or/4-5 12317 enema/ 4448 (enema$ or clysma$ or clyster$ or enteroclysis).tw. 8549 or/7-8 95910 3 and (6 or 9) 3711 exp Infant/ 23 20812 ((premature$ or pre-mature$ or prematuritas or preterm$1

or pre-term$1 or (low adj3 wt$) or low-birth weight$ orunderweight$) adj5 (infant$ or neonate$ or neonatal$ orbaby or babies or birth or child or children or childbirthor newborn$)).mp.

7946

13 (infant$ or neonate$ or neonatal$ or baby or babies ornewborn$).tw.

24 480

14 or/11-13 35 99115 10 and 14 10

1106 LIVINGSTON et al by guest on January 5, 2020www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2015-0143 originally published online May 18, 2015; 2015;135;1093Pediatrics 

Sarah A. Jones and J. Mark WaltonMichael H. Livingston, Anna C. Shawyer, Peter L. Rosenbaum, Connie Williams,Glycerin Enemas and Suppositories in Premature Infants: A Meta-analysis

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/135/6/1093including high resolution figures, can be found at:

Referenceshttp://pediatrics.aappublications.org/content/135/6/1093#BIBLThis article cites 32 articles, 0 of which you can access for free at:

Subspecialty Collections

http://www.aappublications.org/cgi/collection/neonatology_subNeonatologysubhttp://www.aappublications.org/cgi/collection/fetus:newborn_infant_Fetus/Newborn Infantine_subhttp://www.aappublications.org/cgi/collection/evidence-based_medicEvidence-Based Medicinefollowing 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 January 5, 2020www.aappublications.org/newsDownloaded from

DOI: 10.1542/peds.2015-0143 originally published online May 18, 2015; 2015;135;1093Pediatrics 

Sarah A. Jones and J. Mark WaltonMichael H. Livingston, Anna C. Shawyer, Peter L. Rosenbaum, Connie Williams,Glycerin Enemas and Suppositories in Premature Infants: A Meta-analysis

http://pediatrics.aappublications.org/content/135/6/1093located on the World Wide Web at:

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

ISSN: 1073-0397. 60007. Copyright © 2015 by the American Academy of Pediatrics. All rights reserved. Print the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has 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 January 5, 2020www.aappublications.org/newsDownloaded from