Enteral antibiotics for preventing necrotizing enterocolitis in low birthweight or preterm infants

Bury RG, Tudehope D

Background - Methods - Results - Characteristics of Included Studies - References - Data Tables and Graphs

Dates

Date edited: 15/11/2006
Date of last substantive update: 01/09/2000
Date of last minor update: 31/08/2006
Date next stage expected 01/08/2008
Protocol first published: Issue 2, 1998
Review first published: Issue 2, 1998

Contact reviewer

Dr Richard Graham Bury
Director
Neonatal Services
Royal Hobart Hospital
Liverpool St.
Hobart
Tasmania AUSTRALIA
7000
Telephone 1: 61 3 62228308 extension: 8220
Telephone 2: 61 3 62313781
Facsimile: 61 3 62342264
E-mail: graham.bury@dchs.tas.gov.au

Contribution of reviewers

Both reviewer authors assessed the trials for eligibility, quality, and extracted data independently. The review was written by RGB with the editorial assistance of DT.

Internal sources of support

Mater Mother's Hospital, Brisbane, AUSTRALIA
Royal Hobart Hospital, Tasmania, AUSTRALIA
Centre for Clinical Studies - Women's and Children's Health, Mater Hospital, South Brisbane, Queensland, AUSTRALIA

External sources of support

Department of Health and Ageing, Commonwealth Government, Canberra ACT, AUSTRALIA
Centre for Perinatal Health Services Research, University of Sydney, AUSTRALIA

What's new

This review is an update of the previous review cited as: Bury RG, Tudehope D. Enteral antibiotics for preventing necrotizing enterocolitis in low birthweight or preterm infants, published in The Cochrane Library, Issue 3, 2002 (Bury 2002).

No new trials have been found since the last substantive update in September 2000. The conclusion remains unchanged. Although prophylactic enteral antibiotics reduce the incidence of necrotizing enterocolitis in low birth weight infants, their use for this purpose may increase the development of resistant bacteria.

Dates

Date review re-formatted: 09/09/1999
Date new studies sought but none found: 01/08/2006
Date new studies found but not yet included/excluded: / /
Date new studies found and included/excluded: / /
Date reviewers' conclusions section amended: / /
Date comment/criticism added: / /
Date response to comment/criticisms added: / /

Text of review

Synopsis

Not enough evidence to support administering antibiotics through a feeding tube for low birth weight and new born babies to prevent necrotizing enterocolitis

Necrotizing enterocolitis (NEC) is a serious disease that affects the bowel in the first few weeks of life. The cause is unknown but milk feeding and bacteria may contribute. NEC is more common in preterm babies, possibly because of reduced immunity. Oral antibiotics have been used to prevent NEC but there are concerns about the possible adverse effects of oral antibiotics such as resistance to bacteria. The review of trials found there was not enough evidence to support the use of antibiotics to prevent NEC in preterm and low birth weight babies. More research is needed.

Abstract

Background

Necrotizing enterocolitis (NEC) is the most common gastrointestinal problem of preterm neonates. There have been reports published suggesting that the use of enteral antibiotics may be effective as prophylaxis. This systematic review was undertaken to clarify the issue.

Objectives

To evaluate the benefits and harms of enteral antibiotic prophylaxis for necrotizing enterocolitis in low birth weight and preterm infants.

Search strategy

Searches were made of the Oxford Database of Perinatal Trials, Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 3, 2006), MEDLINE (1966 - July 2006), EMBASE (1988 - July 2006) and CINAHL (1982 - July 2006) previous reviews with cross references, abstracts, conference and symposia proceedings, expert informants and journal hand searching in the fields of neonatal pediatrics and microbiology.

Selection criteria

All randomized or quasi-randomized controlled trials where enteral antibiotics were used as prophylaxis against NEC in LBW (< 2500 g) and/or preterm (< 37 weeks gestation) infants.

Data collection & analysis

The standard method of the Cochrane Collaboration and its Neonatal Review Group was used. The methodological quality of each trial was reviewed by the second author who was blinded to the trial authors and institutions. Each author extracted data separately before comparison and resolution of differences. Relative risk (RR), risk difference (RD), and number needed to treat were used in the analysis.

Main results

Five eligible trials involving 456 infants were included. The administration of prophylactic enteral antibiotics resulted in a statistically significant reduction in NEC [typical RR 0.47 (0.28, 0.78); typical RD -0.10 (-0.16, -0.04); NNT 10 (6, 25)].

There was a statistically significant reduction in NEC-related deaths [typical RR 0.32 (0.10, 0.96); typical RD -0.07 (-0.13, 0.01); NNT 14 (8, 100)].

There was a trend towards a reduction in all deaths which was not significant [typical RR 0.67(0.34, 1.32)].

There were no significant differences in NEC-like enteropathies (one trial only).

One study found a statistically significant increase in the incidence of colonization with resistant bacteria and the summary analysis of three trials gave an increase which was just significant [RR 1.73 (1.00, 2.97); RD 0.07 (0.00, 0.13)].

Reviewers' conclusions

Evidence suggests that oral antibiotics reduce the incidence of NEC in low birth weight infants. However concerns about adverse outcomes persist, particularly related to the development of resistant bacteria. To address this question further, a large trial would be required with a sample size sufficient to examine all the important benefits and harms. Adverse outcomes associated with infection should be evaluated, and microbiological studies looking for the development of resistant bacteria should be undertaken

Background

The pathogenesis of necrotizing enterocolitis (NEC) is multi-factorial and any unifying hypothesis remains unproven (Bauer 1992). Hypoxic-ischemic bowel injury with intestinal mucosal disruption in the presence of enteral feeding and bacterial colonization may contribute to the development of NEC. Infants who are preterm or of low birth weight (LBW) have an increased susceptibility, possibly as a result of a compromised immune system.

There have been reports of the effectiveness of using enteral antibiotics as prophylaxis against NEC in low birth weight and preterm infants. There are concerns about the strength of evidence of the effectiveness and possible harmful effects of the use of oral antibiotics. These harmful effects include the development of bacterial resistance and other neonatal complications.

This review updates an earlier version published in the Cochrane Library (Bury 2000).

Objectives

To evaluate the benefits and harms of enteral antibiotic prophylaxis for NEC in LBW and preterm infants.

Criteria for considering studies for this review

Types of studies

Randomized or quasi-randomized trials where enteral antibiotics were used as prophylaxis against NEC in LBW (< 2500 g) and/or preterm (< 37 weeks gestation) infants were considered for this review.

Types of participants

LBW and/or preterm infants at risk for the development of NEC

Types of interventions

Enteral antibiotics or placebo as prophylaxis against NEC

Types of outcome measures

The primary outcome measure was the incidence of NEC. NEC was defined as: Clinical evidence of abdominal distension and feed intolerance plus radiological evidence of pneumatosis intestinalis with or without portal venous gas and pneumoperitoneum.

Secondary outcome measures include:

Death from NEC
All deaths
NEC-like enteropathies
Colonization with resistant bacteria

Death from NEC was defined as death where the primary cause was attributed to NEC. NEC-like enteropathy was defined as clinical evidence of abdominal distension, feed intolerance and blood in the stools in the absence of any defined X-ray abnormalities.

Search strategy for identification of studies

The standard search strategy for the Cochrane Neonatal review Group was used. This included searches of electronic databases: Oxford Database of Perinatal Trials, Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library, Issue 3, 2006), MEDLINE (1966 - July 2006), EMBASE (1988 - July 2006) and CINAHL (1982 - July 2006) using MeSH terms: enterocolitis necrotising; antibiotics; infant low birth weight; infant newborn; infant premature; controlled clinical trials; randomized-controlled trials; and text terms: necrotizing enterocolitis; enterocolitis; antibiotics; low birth weight infant; newborn infant; premature infant; preterm infant and previous reviews with cross references, abstracts, conference and symposia proceedings, expert informants and journal hand searching in the fields of neonatal pediatrics and microbiology.

Methods of the review

Criteria and methods used to assess the methodological quality of included trials:
The standard method of the Cochrane Collaboration and its Neonatal Review Group was used. The methodological quality of each trial was reviewed by the second author who was blinded to the trial authors and institutions.

Methods used to collect data from the included trials:
Each author extracted data separately before comparison and resolution of differences.

Methods used to synthesize data:
Standard method of the Neonatal Review Group with the use of relative risk (RR), risk difference (RD), and their 95% confidence intervals. The number needed to treat (NNT) was calculated as 1/RD. Subgroup analyses were performed a posteriori, limited to the trials that used random allocation. The I² statistic was used to examine heterogeneity.

Description of studies

Ten potentially relevant studies were identified. Five trials were excluded from the final analysis. One (Brantley 1980) was not a randomized or quasi-randomized trial. Another (McCaffree 1975) was an abstract consisting of an interim report of a trial in progress, but with insufficient data for analysis. In two trials (Millar 1992; Fast 1994) there was no control arm which received no treatment. The last (Faix 1988) was a treatment trial in which the babies already had NEC at entry.

Among the five trials included, involving a total of 456 infants, four enrolled very low birth-weight infants and one (Boyle 1978) enrolled infants of less than 2250 grams and less than 36 weeks gestational age at birth. Two studies (Grylack 1978; Rowley 1978) included other infants with specified risk factors for developing necrotizing enterocolitis, and reported on at least one relevant clinical outcome. Additional information was obtained on NEC-related deaths from Siu (Siu 1998).

All trials used oral antibiotics prophylactically, and the primary outcome, incidence of NEC, was reported in all trials. Two studies used enteral gentamicin (Grylack 1978; Rowley 1978) and two studies used enteral kanamycin (Egan 1976; Boyle 1978). One study used enteral vancomycin (Siu 1998). Infants entered the trial on day one (Grylack 1978), day two (Siu 1998), day three (Boyle 1978), and at a mean of 7.1 days (Egan 1976). The age at entry was not specified in one trial (Rowley 1978).

A full description of each of the studies is included in the Table: Characteristics of Included Studies.

Methodological quality of included studies

In four trials, random allocation to treatment groups was used (Boyle 1978; Grylack 1978; Rowley 1978; Siu 1998). In one (Grylack 1978), randomization was by blind card draw in the hospital pharmacy whilst Siu 1998 used random assignment by computer. The method of randomization was not specified by Boyle 1978; Rowley 1978. One trial (Egan 1976) used quasi-random allocation (odd or even hospital record number).

Blinding of the intervention was achieved in three trials (Boyle 1978; Rowley 1978; Siu 1998) and, probably, in a fourth (Grylack 1978).

None of the trials described exclusions after randomization.

Blinding of the outcome measurement was ensured in all trials except for Rowley 1978, where this is uncertain.

Details of the methodologic quality assessments are given in the Table: Characteristics of Included Studies.

Results

Comparison 01: Enteral aminoglycoside versus control (all studies)
Outcome 01: NEC
The administration of prophylactic enteral antibiotics produced a significant overall reduction in NEC [typical RR 0.47 (0.28, 0.78); typical RD -0.10 (-0.16, -0.04); NNT 10 (6, 25)].
In the subgroup analysis limited to the three trials that used random allocation and assessed this outcome, there was a similar significant reduction in NEC [typical RR 0.52 (0.31,0.88); typical RD -0.09 (-0.16, -0.02); NNT 11 (6, 50)].

Outcome 02: All deaths
Two trials (Egan 1976; Siu 1998) reported all deaths. Although there was a reduction in deaths in the treatment group this was not statistically significant [typical RR 0.67 (0.34, 1.32)].

Outcome 03: NEC-related deaths
Two trials (Egan 1976; Grylack 1978 ) reported NEC-related deaths. Additional information was obtained on NEC-related deaths from Siu 1998. There was a reduction in NEC-related deaths which was statistically significant [typical RR 0.32 (0.10, 0.96); typical RD -0.07 (-0.13, 0.01); NNT 14 (8, 100)].

Outcome 04: NEC-like enteropathies
Only one trial (Grylack 1978) reported on this outcome and there was a trend towards a reduction but this was not statistically significant.

Outcome 05: Colonization with resistant bacteria
Boyle 1978 evaluated the development of antibiotic resistance of gram negative organisms to kanamycin during the study period of 24 days. There was a significant increase in resistance in treated compared with control infants [RR 1.96 (1.10, 3.48)] but no increase in resistance in general nursery cultures. Egan 1976 also assessed antibiotic resistance and found no difference in treated compared to control infants. Combining these studies gives an increase that is just statistically significant [typical RR 1.73 (1.00, 2.97); typical RD 0.07 (0.00, 0.13)]. No evidence of vancomycin resistance was found by Siu (Siu 1998) from routine microbiological surveillance or from specimens obtained for sepsis screening during the study. However a significant change in stool flora occurred with a predominant growth of gram negative organisms and yeasts demonstrated immediately and four weeks after vancomycin prophylaxis.

Systemic absorption of the enterally administered antibiotic was evaluated in one study (Egan 1976). Blood levels of kanamycin were found to be undetectable (< 0.5 micrograms/ml).

Discussion

Only one study is recent (Siu 1998). The other relatively small studies were conducted about 20 years ago when trial methodology and reporting were less rigorous and the clinical setting in neonatal units was different. Indications for enrollment varied between studies. Five trials were included, one of which was quasi-randomized (Egan 1976).

Although all trials reported on NEC outcome there was limited reporting of other outcomes, particularly possible harms of treatment. Only two studies reported all deaths, raising some concern about bias in outcomes dependent on survival. Systemic absorption of the antibiotic was evaluated in one study (Egan 1976).

Only one study (Siu 1998) provided the information on the type of feed given to infants in each arm of the study and in those infants developing NEC. No conclusions could be drawn in view of the small numbers involved.

The development of bacterial resistance was evaluated in three studies (Boyle 1978; Egan 1976, Siu 1998). The results varied between the studies. One (Egan 1976) found no difference in bacterial resistance in the treatment or placebo group, another (Boyle 1978) found an increase in the treatment group, while the third (Siu 1998) found no evidence of bacterial resistance developing in either group. The duration of treatment may have had an effect on the development of bacterial resistance since the trial which found increased resistance (Boyle 1978) used oral kanamycin for 24 days, while the other two used oral kanamycin for 21 days (Egan 1976) and oral vancomycin for 7 days (Siu 1998) respectively.

Other trials concerning prevention of NEC have tested immunoglobulins and various combinations of antibiotics. Fast 1994 compared oral gentamicin with an oral IgG/IgA mixture. Millar 1992 compared parenteral vancomycin and aztreonam compared with parenteral vancomycin and gentamicin.

Several Cochrane reviews examine other interventions for the prevention of NEC in preterm infants; human milk compared with formula for feeding (Henderson 2001a; Henderson 2001b), rapid versus slow rate of advancement of feedings (Kennedy 1998; Kennedy 2000), oral immunoglobulin (Foster 2004) and arginine supplementation (Shah 2004). Trials of probiotics to prevent neonatal infection and NEC appear promising and are being reviewed (AlFaleh 2005).

Currently, there is evidence to suggest that oral antibiotics are prophylactic against NEC in low birth weight infants. However, there are concerns about harms and to address this question further, a large trial conducted in selected babies at increased risk for NEC would be required. For example, to demonstrate a 50% relative risk reduction in a population with an anticipated NEC risk of 10%, more than 500 patients in each arm would be required. Babies now at risk of NEC include many who are of substantially lower birth weight and gestational age than those enrolled in previous trials. It would be important to clearly define the groups at risk and to stratify by gestational age. An evaluation of systemic absorption of enterally administered antibiotics should be included.

Reviewers' conclusions

Implications for practice

There is insufficient evidence to support the use of prophylactic enteral antibiotics as prophylaxis for NEC in clinical practice.

Implications for research

To address this question further a large trial would be required with a sample size sufficient to examine all the important benefits and harms. Extended surveillance would be necessary to determine whether antibiotic resistance had emerged. Adverse outcomes associated with infection, particularly with resistant bacteria, should be evaluated.

Acknowledgements

Dr Siu kindly provided additional data from their study.

Potential conflict of interest

None

Characteristics of included studies

Study	Methods	Participants	Interventions	Outcomes	Notes	Allocation concealment
Boyle 1978	Randomized Single centre trial Blinding of randomization: yes Blinding of intervention:yes Complete follow up: yes Blinding of outcome measurement: yes	99 Infants <36 weeks gestation weighing <2,250 grams who were able to begin enteral or gavage feeds	Kanamycin enteral 15mg/kg/day. Same volume of placebo given 3 times a day for 24 days or until discharge or when NEC occurred	Incidence of NEC Microbiological colonisation rates		A
Egan 1976	Quazi-randomized Single unit Blinding of randomisation: no Blinding of intervention: no Complete follow-up: yes Blinding of outcome measurement: yes	75 infants of Birthweight < 1,500 grams Receiving enteral feeds	Kanamycin enteral (15 mg/kg/day in three doses) for 21 days	Incidence of NEC All deaths NEC-related deaths Blood levels of kanamycin Microbiological colonisation rates		C
Grylack 1978	Randomized Single centre trial Blinding at randomisation: yes Blinding of intervention: can't tell Complete followup: yes Blinding of outcome measurement: yes	42 infants of Birthweight < 1,500 grams Any weight plus umbilical catheter Preterm plus fetal distress, postnatal asphyxia or hypotension	Gentamicin enteral (2.5mg/kg/day 6 hourly for 7 days) v 5% dextrose in water (2.5mg/kg/day 6 hourly for 7 days)	Incidence of NEC NEC-related death NEC-like enteropathies		A
Rowley 1978	Randomized Single centre trial Blinding of randomization: yes Blinding of intervention: yes Complete follow up: yes Blinding of outcome measurement: can`t tell	100 infants of Birthweight <1500 grams Umbilical arterial or venous catheters Premature infants with fetal distress, severe hypoxia or hypotension Apgar scores of <7 at 5 minutes in infants <2000 grams	Gentamycin enteral 2.5mg/kg 6 hourly for one week. Sterile water 6 hourly for one week	Incidence of NEC	This trial was described in a letter to Lancet. No deaths or other outcomes reported.	A
Siu 1998	Double blind randomized trial at one centre. Blinding of randomization: yes Blinding of intervention: yes Completeness of followup: not stated Blinding of outcome assessment: yes	140 preterm infants with birthweight <1500 grams, 48 hrs old. Excluded infants with lethal anomalies, major gastrointestinal anomalies and prior gastrointestinal surgery.	Vancomycin oral 15mg per kg 8 hourly for 7 days or placebo (sterile water with one drop multivitamin per 20 ml).	Incidence of NEC, all deaths, microbial cultures.	Author provided information on NEC related deaths	A

Characteristics of excluded studies

Study	Reason for exclusion
Brantley 1980	Not a randomized or quasi-randomized trial.
Faix 1988	Not a prophylaxis trial. Babies had NEC at entry, and were randomised to ampicillin and gentamicin vs ampicillin, gentamicin and clindamycin.
Fast 1994	No control arm. Interim report of trial in progress. Oral gentamicin vs an oral IgG/IgA mixture.
McCaffree 1975	Abstract only. Interim report of trial in progress. Insufficient data for analysis. Primary author could not be located for further data
Millar 1992	No control arm. In this study comparison was between vancomycin and aztreonam vs vancomycin and gentamicin, all drugs being given parenterally.

References to studies

References to included studies

Boyle 1978 {published data only}

Boyle R, Nelson JS, Stonestreet BS, Peter G, Oh W. Alterations in stool flora resulting from oral kanamycin prophylaxis of necrotizing enterocolitis. Journal of Pediatrics 1978;93:857-61.

Egan 1976 {published data only}

Egan EA, Mantilla G, Nelson RM, Eitzmann DV. A prospective controlled trial of oral kanamycin in the prevention of neonatal necrotizng enterocolitis. Journal of Pediatrics 1976;89:467-70.

Grylack 1978 {published data only}

Grylack LJ, Scanlon JW. Oral gentamicin therapy in the prevention of neonatal necrotizing enterocolitis. A controlled double-blind trial. American Journal of Diseases of Children 1978;132:1192-4.

Rowley 1978 {published data only}

Rowley MP, Dahlenburg GW. Gentamicin in prophylaxis of neonatal necrotizing enterocolitis. Lancet 1978;ii:532.

Siu 1998 {published and unpublished data}

Siu YK, Ng PC, Fung SC, Lee CH, Wong MY, Fok TF, So KW, Cheung KL, Wong W, Cheng AF. Double blind, randomised, placebo controlled study of oral vancomycin in prevention of necrotising enterocolitis in preterm, very low birthweight infants. Archives of Disease in Childhood Fetal Neonatal Ed 1998;79:F105-9.

References to excluded studies

Brantley 1980 {published data only}

Brantley VE, Hiatt IM, Hegyi T. The effectiveness of oral gentamicin in reducing the incidence of necrotizing enterocolitis (NEC) in treated and control infants [abstract]. Pediatric Research 1980;14:592 (Abstract).

Faix 1988 {published data only}

Faix RG, Polley TZ, Grasela TH. A randomized, controlled trial of parenteral clindamycin in neonatal necrotizing enterocolitis. Journal of Pediatrics 1988;112:271-7.

Fast 1994 {published data only}

Fast C, Rosegger H. Necrotizing enterocolitis prophylaxis: oral antibiotics and lyophilized enterobacteria vs oral immunoglobulins. Acta Pediatrica 1994;Suppl 396:86-90.

McCaffree 1975 {published data only}

McCaffree MA, Fletcher AB, Avery GB. Prophylactic oral antibiotics in necrotizing enterocolitis [abstract]. Pediatric Research 1975;9:307 (Abstract).

Millar 1992 {published data only}

Millar MR, MacKay P, Levene M, Langdale V, Martin C. Enterobacteriaceae and neonatal necrotising enterocolitis. Archives of Disease in Childhood 1992;67:53-6.

* indicates the primary reference for the study

Other references

Additional references

AlFaleh 2005

AlFaleh K, Bassler D. Probiotics for prevention of mortality and morbidity in preterm infants (Protocol). The Cochrane Database of Systematic Reviews 2005, Issue 4.

Bauer 1992

Bauer CR. Necrotizing enterocolitis. In: Sinclair JC, Bracken MB, editor(s). Effective Care of the Newborn Infant. Oxford: Oxford University Press, 1992:602-14.

Foster 2004

Foster J, Cole M. Oral immunoglobulin for preventing necrotizing enterocolitis in preterm and low birth-weight neonates. Cochrane Database of Systematic Reviews 2004, Issue 1.

Henderson 2001a

Henderson G, Anthony M, McGuire W. Formula milk versus preterm human milk for feeding preterm or low birth weight infants. Cochrane Database of Systematic Reviews 2001, Issue 3.

Henderson 2001b

Henderson G, Anthony M, McGuire W. Formula milk versus term human milk for feeding preterm or low birth weight infants. Cochrane Database of Systematic Reviews 2001, Issue 4.

Kennedy 1998

Kennedy K, Tyson J. Rapid versus slow rate of advancement of feedings for promoting growth and preventing necrotizing enterocolitis in parenterally fed low-birth-weight infants. Cochrane Database of Systematic Reviews 1998, Issue 4.

Kennedy 2000

Kennedy K, Tyson J. Early versus delayed initiation of progressive enteral feedings for parenterally fed low birth weight or preterm infants. Cochrane Database of Systematic Reviews 2000, Issue 1.

Shah 2004

Shah P, Shah V. Arginine supplementation for prevention of necrotising enterocolitis in preterm infants. Cochrane Database of Systematic Reviews 2004, Issue 4.

Other published versions of this review

Bury 1998

Bury RG, Tudehope D. Enteral antibiotics for preventing necrotizing enterocolitis in low birthweight or preterm infants. Cochrane Library 1998, Issue 2.

Bury 2000

Bury RG, Tudehope D. Enteral antibiotics for preventing necrotizing enterocolitis in low birthweight or preterm infants. Cochrane Library 2000, Issue 4.

Bury 2002

Bury G, Tudehope D. Enteral antibiotics for preventing necrotizing enterocolitis in low birthweight or preterm infants. The Cochrane Library 2002, Issue 3.

Comparisons and data

01 Enteral aminoglycoside versus control (all studies)

01.01 NEC

01.02 All deaths

01.03 NEC-related deaths

01.04 NEC-like enteropathies

01.05 Colonization with resistant bacteria

02 Enteral aminoglycoside versus control (randomized studies)

02.01 NEC

02.02 NEC-related deaths

02.03 NEC-like enteropathies

02.04 Colonization with resistant bacteria

Comparison or outcome
Studies

Participants

Statistical method

Effect size

01 Enteral aminoglycoside versus control (all studies)

01 NEC
5

456

RR (fixed), 95% CI

0.47 [0.28, 0.78]

02 All deaths
2

215

RR (fixed), 95% CI

0.67 [0.34, 1.32]

03 NEC-related deaths
3

257

RR (fixed), 95% CI

0.32 [0.10, 0.96]

04 NEC-like enteropathies
1

42

RR (fixed), 95% CI

0.22 [0.03, 1.73]

05 Colonization with resistant bacteria
3

314

RR (fixed), 95% CI

1.73 [1.00, 2.97]

02 Enteral aminoglycoside versus control (randomized studies)

01 NEC
4

381

RR (fixed), 95% CI

0.52 [0.31, 0.88]

02 NEC-related deaths
2

182

RR (fixed), 95% CI

0.41 [0.12, 1.39]

03 NEC-like enteropathies
1

42

RR (fixed), 95% CI

0.22 [0.03, 1.73]

04 Colonization with resistant bacteria
3

314

RR (fixed), 95% CI

1.73 [1.00, 2.97]

Comparison or outcome	Studies	Participants	Statistical method	Effect size
01 Enteral aminoglycoside versus control (all studies)
01 NEC	5	456	RR (fixed), 95% CI	0.47 [0.28, 0.78]
02 All deaths	2	215	RR (fixed), 95% CI	0.67 [0.34, 1.32]
03 NEC-related deaths	3	257	RR (fixed), 95% CI	0.32 [0.10, 0.96]
04 NEC-like enteropathies	1	42	RR (fixed), 95% CI	0.22 [0.03, 1.73]
05 Colonization with resistant bacteria	3	314	RR (fixed), 95% CI	1.73 [1.00, 2.97]
02 Enteral aminoglycoside versus control (randomized studies)
01 NEC	4	381	RR (fixed), 95% CI	0.52 [0.31, 0.88]
02 NEC-related deaths	2	182	RR (fixed), 95% CI	0.41 [0.12, 1.39]
03 NEC-like enteropathies	1	42	RR (fixed), 95% CI	0.22 [0.03, 1.73]
04 Colonization with resistant bacteria	3	314	RR (fixed), 95% CI	1.73 [1.00, 2.97]

Contact details for co-reviewers

Prof David Tudehope
Director of Neonatology, Professor in Perinatal Paediatrics
Division of Neonatology
Mater Misericordiae Mothers' Hospital
Raymond Terrace
South Brisbane
QLD AUSTRALIA
4101
Telephone 1: +61 7 3840 8441
Telephone 2: +61 7 3840 8111
Facsimile: + 61 7 3840 1949
E-mail: dtudehop@mater.org.au

This review is published as a Cochrane review in The Cochrane Library, Issue 1, 2007 (see http://www.thecochranelibrary.com for information). Cochrane reviews are regularly updated as new evidence emerges and in response to feedback, and The Cochrane Library should be consulted for the most recent version of the Review.