Review ID

Reason for exclusion

Abou El Senoun 2014

Secondary neonatal outcomes included:

• Disability at time of childhood follow‐up (as defined by authors).

• Serious disability (as defined by authors) after two years.

No outcome data for these outcomes.

Bricker 2015

Primary outcomes included:

• Neurodevelopment at age two.

No outcome data for this outcome.

Buchanan 2010

Secondary neonatal outcomes included:

• Disability at time of childhood follow‐up.

No outcome data for this outcome.

Chapman 2014

No outcomes focused on development or disability at follow‐up.

Crowley 2016

Secondary infant outcomes included:

• Neurodevelopmental delay at 12 months and 24 months.

No outcome data for this outcome.

Dare 2006

foetal, neonatal, and infant outcomes included:

• Disability at time of childhood follow‐up.

No outcome data for this outcome

East 2014

Primary outcomes included:

• Long‐term neurodevelopmental outcome.

No outcome data for this outcome.

Gomi 2015

No outcomes focused on development or disability at follow‐up.

Han 2013

Primary outcomes included:

• Any neurological disability at follow‐up.

No outcome data for this outcome.

Hofmeyr 2015

Primary outcomes included:

• Perinatal or infant death (excluding fatal anomalies) or disability in childhood.

Secondary long‐term infant outcomes included:

• Disability in childhood, as defined by trial authors.

No outcome data for these outcomes.

Hopkins 2002

No outcomes focused on development or disability at follow‐up.

Kenyon 2013

Secondary outcome included:

• Long‐term health outcomes (as defined by trial authors) after at least two years.

Outcome data only reported for 'Serious childhood disability at seven years'.

Khunpradit 2011

Secondary outcomes include:

• Neonatal neurodevelopment.

No outcome data for this outcome.

Kiiza 2015

Protocol.

Secondary baby outcomes will include:

• Long‐term neurodevelopmental outcome.

Lewin 2010

No outcomes focused on development or disability at follow‐up.

Lutomski 2015

No outcomes focused on development or disability at follow‐up.

Mackeen 2014

No outcomes focused on development or disability at follow‐up.

Neilson 2003

No outcomes focused on development or disability at follow‐up.

Olsen 2012

No outcomes focused on development or disability at follow‐up.

Othman 2007

No outcomes focused on development or disability at follow‐up.

Sangkomkamhang 2015

No outcomes focused on development or disability at follow‐up.

Say 1996

No outcomes focused on development or disability at follow‐up.

Say 1996a

No outcomes focused on development or disability at follow‐up.

Say 1996b

No outcomes focused on development or disability at follow‐up.

Say 1996c

No outcomes focused on development or disability at follow‐up.

Say 2001

No outcomes focused on development or disability at follow‐up.

Say 2003

No outcomes focused on development or disability at follow‐up.

Say 2003a

No outcomes focused on development or disability at follow‐up.

Say 2003b

No outcomes focused on development or disability at follow‐up.

Siegfried 2011

No outcomes focused on development or disability at follow‐up.

Siriwachirachai 2014

No outcomes focused on development or disability at follow‐up.

Stan 2013

Secondary outcomes included:

• Long‐term sequelae: neurologic impairment and chronic lung disease.

No outcome data for this outcome.

Thomas 2007

Outcomes included:

• Adverse neonatal outcomes in terms of longer‐term neurological outcomes.

No outcome data for this outcome.

Review ID

Date of search; date assessed as up‐to‐date

No. included trials; countries and years of publication

No. participants in included trials

Inclusion criteria for "Types of participants"

Relevant comparison interventions (no. trials and participants)

Overview outcomes for which data were reported (pre‐specified unless stated otherwise)

Abalos 2014

30 April 2013

49 RCTs

34 RCTs in industrialised countries (Australia, France, Hong Kong, Ireland, Israel, Italy, Sweden, UK and USA)

15 RCTs in low‐ or middle‐income countries (Argentina, Brazil, Caribbean Islands, India, South Africa, Sudan and Venezuela)

RCTs published in:

1960s: 1

1970s: 2

1980s: 22

1990s: 17

2000s: 5

2010s; 2

4723 women and their babies

women with mild to moderate hypertension during pregnancy, regardless of whether or not they had proteinuria, previous antihypertensive treatment, or whether the pregnancy was singleton or multiple

any antihypertensive drug versus no drugs

(29 RCTs, 3350 women)

cerebral palsy reported as a secondary outcome

primary outcome was impaired long‐term growth and development in infancy and childhood

Alfirevic 2013

Search: 31 December 2012

Up‐to‐date: 31 January 2013

13 RCTs (1 qRCT)

No. RCTs in:

Australia: 2

Denmark: 1

Greece: 1

India: 1

Ireland: 2

Pakistan: 1

Sweden: 1

UK: 1

USA: 3

RCTs published in:

1970s: 4

1980s: 6

1990s: 2

2000s: 1

37,715 women and their babies

pregnant women in labour and their babies

continuous cardiotocography versus intermittent auscultation

(12 RCTs, 33,681 women)

cerebral palsy reported as a primary review outcome

Churchill 2013

Search: 28 February 2013

Up‐to‐date:

10 July 2013

4 RCTs

No. RCTs in:

Egypt: 1

Europe: 1

South Africa: 1

USA: 1

RCTs published in:

1990s: 2

2000s: 2

425 women and their babies

women with severe pre‐eclampsia, up to and including 34 weeks' gestation

interventionist care versus expectant (delayed delivery) care

(4 RCTs, 425 women)

cerebral palsy reported as a secondary review outcome

pre‐specified outcome was 'measures of long‐term growth and development, such as important impairment and cerebral palsy'

Crowther 2010

Search: 20 December 2010

Up‐to‐date:

15 February 2011

8 RCTs Countries of trials not reported

RCTs published in:

1980s: 4

1990s: 3

2000s: 1

879 women and their babies

women at risk of imminent very preterm birth

vitamin K versus control

(8 RCTs, 879 women)

cerebral palsy reported as a primary review outcome

pre‐specified outcome was long‐term neurodevelopment

Crowther 2010a

Search: 20 December 2010

Up‐to‐date: 9 January 2011

9 RCTs

Countries of trials not reported

RCTs published in:

1980s: 4

1990s: 5

1752 women and their babies

women at risk of imminent very preterm birth (before 34 weeks' gestation)

phenobarbital versus control

(9 RCTs, 1752 women)

cerebral palsy, motor dysfunction (other neuromotor impairment) reported as primary review outcomes

pre‐specified outcome was 'long‐term neurodevelopment'

Crowther 2014

31 January 2014

37 RCTs (4 qRCTs)

No. RCTs in: China: 3

Iran: 5

Italy: 1

Mexico: 1

Thailand: 1

Turkey: 1

USA: 25

RCTs published in:

1980s: 7

1990s: 18

2000s: 10

2010s: 2

3571 women and their babies

women considered to be in preterm labour given magnesium sulphate to reduce their risk of preterm birth

magnesium sulphate versus placebo, no treatment, or other tocolytic agent

(37 RCTs, 3571 women)

cerebral palsy reported as a secondary review outcome

primary review outcome was a composite outcome including cerebral palsy, listed as 'serious infant outcome' ("...death or chronic lung disease...grade three or four intraventricular haemorrhage or periventricular leukomalacia, major neurosensory disability (legal blindness, sensorineural deafness requiring hearing aids, moderate or severe cerebral palsy, or developmental delay or intellectual impairment...))"

Crowther 2015

20 January 2015

10 RCTs

No. RCTs in:

Australia and New Zealand: 1

Canada: 1

Finland: 1

India: 1

USA: 5

20 countries: 1

RCTs published in:

2000s: 9

2010s: 1

4733 women and their babies

women considered to be at risk of preterm birth who had already received a single course of prenatal corticosteroid seven or more days previously

repeat doses of prenatal corticosteroids versus placebo or no treatment

(10 RCTs, 4733 women)

cerebral palsy reported as a secondary review outcome;

composite primary outcomes that included cerebral palsy

• survival free of any disability (however defined by authors)

• survival free of major disability (however defined by authors)

• major neurosensory disability (not pre‐specified)

• disability at childhood follow‐up (however defined by authors)

• composite serious outcome (however defined by authors)

Dodd 2013

14 January 2013

36 RCTs

No. RCTs in:

Albania: 1

Brazil: 1

Denmark and Austria: 1

Egypt: 3

Finland: 1

France: 3

India: 2

Iran: 5

Italy: 1

Netherlands: 1

Spain: 1

Turkey: 1

UK: 1

USA: 11

international: 3

RCTs published in:

1970s: 2

1980s: 2

2000s: 13

2010s: 19

8523 women and their babies

pregnant women considered to be at increased risk of preterm birth:

• Past history of spontaneous preterm birth

• Multiple pregnancy

• Ultrasound identified short cervical length

• fetal fibronectin testing

• Following acute presentation with symptoms or signs of threatened preterm labour

• Other reason considered to be at increased risk of preterm birth

progesterone versus placebo or no treatment

(11 RCTs, 1899)

cerebral palsy reported as a secondary review outcome

motor dysfunction, pre‐specified as motor impairment, reported as secondary review outcome

Doyle 2009

Search: 31 August 2008

Up‐to‐date: 5 November 2008

5 RCTs

No. RCTs in: Australia and New Zealand: 1

France: 1

USA: 2

international; 1 (predominately in developing countries)

RCTs published in:

2000s: 5

5560 women and their 6145 babies

women considered to be at risk of preterm birth

magnesium sulphate versus placebo

(5 RCTs, 6145 babies)

cerebral palsy; cerebral palsy or death; severity of cerebral palsy (mild, moderate, moderate to severe, severe cerebral palsy); composite outcomes including cerebral palsy (any neurologic impairment; major neurological disability; death or any neurologic impairment; death or major neurological disability); motor dysfunction (substantial gross motor dysfunction; death or substantial gross motor dysfunction) reported as primary outcomes

pre‐specified outcomes were: neurological impairments (developmental delay or intellectual impairment (developmental quotient or intelligence quotient less than one standard deviation (SD) below the mean), cerebral palsy (abnormality of tone with motor dysfunction), blindness (corrected visual acuity worse than 6/60 in the better eye), or deafness (hearing loss requiring amplification or worse)); neurological disabilities (abnormal neurological function caused by any of the preceding impairments) at follow‐up later in childhood; substantial gross motor dysfunction (motor dysfunction such that the child was not walking at age two years or later, or the inability to grasp and release a small block with both hands); major neurological disability (legal blindness, sensorineural deafness requiring hearing aids, moderate or severe cerebral palsy, developmental delay or intellectual impairment (developmental quotient or intelligence quotient less than two SD below the mean)); paediatric mortality combined with cerebral palsy, substantial gross motor dysfunction, neurological impairment, or major neurological disability (these combined outcomes recognise the competing risks of death or survival with neurological problems)

Duley 2010

Search: 4 June 2010

Up‐to‐date: 1 September 2010

15 RCTs

No. RCTs in: Denmark: 1

India; 1

Malaysia: 1

Mexico: 2

South Africa: 2

Taiwan: 1

USA: 5

international: 2 (1 in 33 countries, with 85% recruitment in low‐ and middle‐income countries; 1 in 8 countries)

RCTs published in:

1990s: 10

2000s: 5

15,570 women and their babies

any women with pre‐eclampsia, regardless of whether: before or after delivery, a singleton or multiple pregnancy, or whether an anticonvulsant had been given before trial entry

magnesium sulphate versus placebo or no anticonvulsant

(6 RCTs, 11,444 women)

severe cerebral palsy; other composite outcomes including cerebral palsy (neurosensory disability; death or neurosensory disability) were all reported as secondary review outcomes

pre‐specified outcomes were 'long‐term growth and development: blindness, deafness, seizures, poor growth, neurodevelopmental delay and cerebral palsy'

Flenady 2013

Search: 31 August 2013

Up‐to‐date: 3 October 2013

14 RCTs

No. RCTs in:

Canada: 1

Chile: 1

Denmark: 1

Germany: 1

Iran: 1

South Africa: 1

Uruguay: 1

USA: 6

international: 1

RCTs published in:

1980s: 1

1990s: 10

2000s: 3

7837 women and their babies

women thought to be in preterm labour with intact membranes, between 20 and 36 completed weeks of gestation

any antibiotics versus no antibiotics

(14 RCTs, 7837 women)

cerebral palsy reported as a primary review outcome

pre‐specified outcome was 'major long‐term infant neurosensory impairment'

Magee 2003

Search:

4 July 2012 (results added to Studies awaiting classification)

Up‐to‐date: 30 January 2004

29 RCTs

No. RCTs in:

Argentina: 1

Australia: 2

Brazil: 1

England: 5

Fr Caribbean: 1

France: 3

Hong Kong: 1

India: 1

Israel: 4

Scotland: 4

Sweden: 3

USA: 2

Venezuela: 1

RCTs published in:

1970s: 1

1980s: 17

1990s: 11

2548 women and their babies

women with mild to moderate hypertension during pregnancy, however defined

beta‐blockers versus placebo or no beta‐blocker

(13 RCTs, 1480 women)

Cerebral palsy reported as a review outcome, rather than primary or secondary

pre‐specified outcome was 'measures of long‐term health and development such as cerebral palsy'

Neilson 2014

31 December 2013

28 RCTs (20 RCTs contributed data)

No. RCTs in:

Australia: 1

Canada: 1

Europe: 6

Iran: 1

Italy: 1

Japan: 1

Sweden: 1

USA: 10

not reported: 6

RCTs published in:

1960s: 1

1970s: 5

1980s: 18

1990s: 3

2010s: 1

2715 women and their babies in 20 RCTs

pregnant women assessed as being in spontaneous preterm labour and considered suitable for tocolytic agents

betamimetics versus placebo

(12 RCTs, 1367 women)

Cerebral palsy reported as a primary review outcome

pre‐specified outcome was 'abnormal long‐term neurodevelopmental status at more than 12 months corrected age (moderate to severe developmental delay, cerebral palsy, sensory impairment, for example, blind and deaf, or a combination)'

Roberts 2006

Search: 30 April 2010 (added the search to Studies awaiting classification)

Up‐to‐date:

15 May 2006

21 RCTs

No. RCTs in:

Brazil: 1

Canada: 1

Finland: 2

Jordan: 1 RCT

Netherlands: 1

New Zealand: 1

South Africa: 1

Spain: 1

Tunisia: 1

UK: 1

USA: 10

RCTs published in:

1970s: 3

1980s: 8

1990s: 8

2000s: 2

over 3999 women (data available for 3885 women and their babies)

women with a singleton or multiple pregnancy, expected to deliver preterm as a result of either spontaneous preterm labour, preterm pre‐labour rupture of the membranes, or elective preterm delivery

antenatal corticosteroids versus placebo or no treatment

(21 RCTs, 3885 women)

cerebral palsy reported as a secondary review outcome

other composite outcome including cerebral palsy (neurodevelopmental delay) reported as a primary review outcome

pre‐specified primary outcome was 'neurodevelopmental disability' at follow‐up (blindness, deafness, moderate/severe cerebral palsy (however defined by authors), or development delay or intellectual impairment (defined as developmental quotient or intelligence quotient less than 2 SD below population mean))'

Stock 2016

30 April 2016

1 RCT

1 RCT in Belgium, Cyprus, Czech Republic, Germany, Greece, Hungary, Italy, Netherlands, Poland, Portugal, Saudi Arabia, Slovenia, United Kingdom

RCT published in 2000s

548 women and their babies

pregnant women at less than 36 weeks' gestation in whom there was clinical suspicion of fetal compromise as defined by trialists

Immediate delivery versus deferred delivery

(1 RCT, 548 women)

cerebral palsy reported as a secondary outcome

other composite outcomes including cerebral palsy (death or disability at or after two years of age) reported as a primary review outcome

'neurodevelopmental impairment at or after two years of age' reported as a secondary outcome

'death or severe disability in childhood' reported but not pre‐specified

Review ID

Summary of trial limitations (risk of bias)

Abalos 2014

Sequence generation: 12 RCTs low risk; 35 RCTs unclear risk; 2 RCTs high risk

Allocation concealment: 17 RCTs low risk; 32 RCTs unclear risk

Blinding (participants and personnel): 10 RCTs low risk; 1 RCT unclear risk; 38 RCTs high risk

Blinding (outcome assessors): 10 RCTs low risk; 2 RCTs unclear risk; 37 RCTs high risk

Incomplete outcome data: 45 RCTs low risk; 4 RCTs high risk

Selective reporting: 9 RCTs low risk; 40 RCTs unclear risk

Other: 23 RCTs low risk; 24 RCTs unclear risk; 2 RCTs high risk

Overall: "Overall, the quality of the studies included in this review is moderate to poor"

Alfirevic 2013

Sequence generation: 3 RCTs low risk; 8 RCTs unclear risk; 2 RCTs high risk

Allocation concealment: 3 RCTs low risk; 6 RCTs unclear risk; 4 RCTs high risk

Blinding (participants and personnel): 13 RCTs high risk

Blinding (outcome assessors): 12 RCTs unclear risk; 1 RCT high risk

Incomplete outcome data: 8 RCTs low risk; 3 RCTs unclear risk; 2 RCTs high risk

Selective reporting: 13 RCTs high risk

Other: 13 RCTs low risk

Overall: Only 2 RCTs were judged to be of high quality.

"The overall quality of the evidence can best be described as low to moderate"

Churchill 2013

Sequence generation: 3 RCTs low risk; 1 RCT unclear risk

Allocation concealment: 3 RCTs low risk; 1 RCT unclear risk

Blinding (participants and personnel): 1 RCT low risk; 3 RCTs unclear risk

Blinding (outcome assessors): 1 RCT low risk; 3 RCTs unclear risk

Incomplete outcome data: 1 RCT low risk; 2 RCTs unclear risk; 1 RCT high risk

Selective reporting: 4 RCTs low risk

Other: 1 RCT low risk; 2 RCTs unclear; 1 RCT high risk

Overall: "Overall, two trials were judged to have a low risk of bias, one was unclear and one a high risk of bias"

Crowther 2010

Sequence generation: 2 RCTs low risk; 4 RCTs unclear risk; 2 RCTs high risk

Allocation concealment: 7 RCTs unclear risk; 1 RCT high risk

Blinding: 2 RCTs low risk; 1 RCT unclear risk; 5 RCTs high risk

Incomplete outcome data: 3 RCTs low risk; 4 RCTs unclear risk; 1 RCT high risk

Selective reporting: 6 RCTs low risk; 2 RCTs unclear risk

Other: 5 RCTs low risk; 2 RCTs unclear risk; 1 RCT high risk

Overall: "The trials were of variable quality."

Crowther 2010a

Sequence generation: 1 RCT low risk; 4 RCTs unclear risk; 4 RCTs high risk

Allocation concealment: 7 RCTs unclear risk; 1 RCT high risk; 1 RCT: not reported

Blinding: 4 RCTs low risk; 1 RCT unclear risk; 4 RCTs high risk

Incomplete outcome data: 1 RCT low risk; 5 RCTs unclear risk; 3 RCTs high risk

Selective reporting: 9 RCTs low risk

Other: 6 RCTs low risk; 1 RCT unclear risk; 2 RCTs high risk

Overall: "Poor‐quality trials contribute excessively to the weight in the overall analysis due to the higher rate of adverse outcomes in those trials"

Crowther 2014

Sequence generation: 15 RCTs low risk; 18 RCTs unclear risk; 4 RCTs high risk

Allocation concealment: 6 RCTs low risk; 27 RCTs unclear risk; 4 RCTs high risk

Blinding (participants and personnel): 4 RCTs low risk; 7 RCTs unclear risk; 26 RCTs high risk

Blinding (outcome assessors): 1 RCT low risk; 35 RCTs unclear risk; 1 RCT high risk

Incomplete outcome data: 20 RCTs low risk; 15 RCTs unclear risk; 2 RCTs high risk

Selective reporting: 11 RCTs low risk; 19 RCTs unclear risk; 7 RCTs high risk

Other: 17 RCTs low risk; 20 RCTs unclear risk

Overall: "Overall, we judged the included trials to be of moderate to high risk of bias"

Crowther 2015

Sequence generation: 8 RCTs low risk; 2 RCTs unclear risk

Allocation concealment: 10 RCTs low risk

Blinding (participants and personnel): 9 RCTs low risk; 1 RCT high risk

Blinding (outcome assessors): 4 RCTs low risk; 6 RCTs unclear risk

Incomplete outcome data: 7 RCTs low risk; 3 RCTs unclear risk

Selective reporting: 9 RCTs low risk; 1 RCT unclear risk

Other: 7 RCTs low risk; 3 RCTs high risk

Overall: "Overall, the included trials were assessed as having a low to moderate risk of bias"

Dodd 2013

Sequence generation: 23 RCTs low risk; 13 RCTs unclear risk

Allocation concealment: 23 RCTs low risk; 13 RCTs unclear risk

Blinding (participants and personnel): 24 RCTs low risk; 7 RCTs unclear risk; 4 RCTs high risk

Blinding (outcome assessors): 15 RCTs low risk; 17 RCTs unclear risk; 4 RCTs high risk

Incomplete outcome data: 31 RCTs low risk; 5 RCTs unclear risk

Selective reporting: 25 RCTs low risk; 10 RCTs unclear risk; 1 RCT high risk

Other: 21 RCTs low risk; 15 RCTs unclear risk

Overall: "The overall quality of the included trials varied from good to fair"

Doyle 2009

Sequence generation: 4 RCTs low risk; 1 RCT unclear risk

Allocation concealment: 4 RCTs low risk; 1 RCT unclear risk

Blinding: 3 RCTs low risk; 2 RCT unclear risk

Incomplete outcome data: 2 RCTs low risk; 3 RCT unclear risk

Selective reporting: 4 RCTs low risk; 1 RCT unclear risk

Overall: "Overall, the methodological quality of the trials was relatively good, with a low risk of bias. However, the quality was better, and the risk of bias lower, in some studies compared with others"

Duley 2010

Sequence generation: 6 RCTs low risk; 9 RCTs unclear risk

Allocation concealment: 5 RCTs low risk; 9 RCTs unclear risk; 1 RCT high risk

Blinding: 4 RCTs low risk; 3 RCTs unclear risk; 8 RCTs high risk

Incomplete outcome data: 7 RCTs low risk; 3 RCTs unclear risk; 5 RCTs high risk

Overall: "The quality of the studies included in this review ranged from excellent to poor. However, most of the poor quality studies were small. The large study comparing magnesium sulphate with placebo was of high quality"

Flenady 2013

Sequence generation: 7 RCTs low risk; 7 RCTs unclear risk

Allocation concealment: 9 RCTs low risk; 5 RCTs unclear risk

Blinding (participants and personnel): 12 RCTs low risk; 2 RCTs high risk

Blinding (outcome assessors): 12 RCTs low risk; 2 RCTs high risk

Incomplete outcome data: 13 RCTs low risk; 1 RCT unclear risk (long‐term: 1 RCT low risk; 13 RCTs unclear risk)

Selective reporting: 12 RCTs low risk; 2 RCTs unclear risk

Other: 13 RCTs low risk; 1 RCT unclear risk

Overall: "Overall the quality of the included trials was good"

Magee 2003

Allocation concealment: adequate in 5 RCT

Double blinding (of physicians and patients) for outcome assessment: 7 RCTs

For maternal and pregnancy outcomes, follow up of greater than 90%: 20 RCTs

Overall: "The quality of these trials was poor"

Neilson 2014

Sequence generation: 12 RCTs low risk; 16 RCTs unclear risk

Allocation concealment: 7 RCTs low risk; 21 RCTs unclear risk

Blinding (participants and personnel): 15 RCTs low risk; 1 RCTs unclear risk; 12 RCTs high risk

Blinding (outcome assessors): 9 RCTs low risk; 8 RCTs unclear risk; 11 RCTs high risk

Incomplete outcome data: 16 RCTs low risk; 10 RCTs unclear risk; 2 RCTs high risk

Selective reporting: 1 RCT low risk; 26 RCTs unclear risk; 1 RCT high risk

Other: 8 RCTs low risk; 19 RCTs unclear risk; 1 RCT high risk

Overall: not detailed

Roberts 2006

Allocation concealment: 8 RCTs: A (adequate); 12 RCTs: B (unclear); 1 RCT: C (inadequate)

Overall: not detailed

Stock 2016

Sequence generation: 1 RCT low risk

Allocation concealment: 1 RCT low risk

Blinding (participants and personnel): 1 RCT high risk

Blinding (outcome assessors): 1 RCT low risk

Incomplete outcome data: 1 RCT low risk (high for childhood outcomes)

Selective reporting: 1 RCT low risk

Other: 1 RCT high risk

Overall: "large study of good quality"

Review ID

AMSTAR criteria

Total score

A priori design

Duplicate selection and extraction

Comprehensive search

Grey literature considered

Included and excluded studies lists

Characteristics of included studies

Quality assessed and documented

Quality considered for conclusions

Methods for combining studies appropriate

Publication bias considered or assessed

Conflicts stated

Abalos 2014

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Alfirevic 2013

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Churchill 2013

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Crowther 2010

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Crowther 2010a

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Crowther 2014

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

11/11

HIGH QUALITY

Crowther 2015

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

11/11

HIGH QUALITY

Dodd 2013

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Doyle 2009

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

✗

9/11

HIGH QUALITY

Duley 2010

✓

✓

✓

✓

✓

✓

✓

✓

?

✗

✗

8/11

HIGH QUALITY

Flenady 2013

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Magee 2003

✓

✓

✓

✓

✓

✓

✓

✓

?

✗

✗

8/11

HIGH QUALITY

Neilson 2014

✓

✓

✓

✓

✓

✓

✓

✓

✓

✓

✗

10/11

HIGH QUALITY

Roberts 2006

✓

✓

✓

✓

✓

✓

✓

✓

?

✗

✓

9/11

HIGH QUALITY

Stock 2016

✓

✓

✓

✓

✓

✓

✓

✓

N/A

✓

✗

9/10

HIGH QUALITY

Review ID

ROBIS domains

Overall risk of bias

Study eligibility criteria

Identification and selection of studies

Data collection and study appraisal

Synthesis and findings

Abalos 2014

Low risk

Low risk

Low risk

Low risk

LOW RISK

Alfirevic 2013

Low risk

Low risk

Low risk

Low risk

LOW RISK

Churchill 2013

Low risk

Low risk

Low risk

Low risk

LOW RISK

Crowther 2010

Low risk

Low risk

Low risk

Low risk

LOW RISK

Crowther 2010a

Low risk

Low risk

Low risk

Low risk

LOW RISK

Crowther 2014

Low risk

Low risk

Low risk

Low risk

LOW RISK

Crowther 2015

Low risk

Low risk

Low risk

Low risk

LOW RISK

Dodd 2013

Low risk

Low risk

Low risk

Low risk

LOW RISK

Doyle 2009

Low risk

Low risk

Low risk

Low risk

LOW RISK

Duley 2010

Low risk

Low risk

Low risk

Unclear risk

LOW RISK

Flenady 2013

Low risk

Low risk

Low risk

Low risk

LOW RISK

Magee 2003

Low risk

Low risk

Low risk

Unclear risk

LOW RISK

Neilson 2014

Low risk

Low risk

Low risk

Low risk

LOW RISK

Roberts 2006

Low risk

Low risk

Low risk

Unclear risk

LOW RISK

Stock 2016

Low risk

Low risk

Low risk

Low risk

LOW RISK

Intervention and comparison

Outcome

Assumed risk with comparator

Corresponding risk with intervention

Relative effect (95% CI)

Number of participants (trials)

Quality of the evidence (GRADE)

Comments

Interventions for the treatment of mild to moderate hypertension

Any antihypertensive drug versus placebo for mild to moderate hypertension during pregnancy (Abalos 2014)

and

Oral beta‐blockers versus placebo for mild to moderate hypertension during pregnancy (Magee 2003)

Cerebral palsy at 1 year (definition not clear; one child with 'spastic quadriparesis with a severe pseudo‐bulbar palsy') assessed by clinical evaluation

(assessment method taken from RCT manuscript as not detailed in reviews)

18 per 1000

(1/55)

6 per 1000 (0 to 146)

RR 0.33 (95% CI 0.01 to 8.01)

110 (1 RCT)

VERY LOW

study limitations (‐1): 1 RCT with unclear sequence generation, allocation concealment, and selective reporting

Imprecision (‐2): Wide confidence intervals crossing line of no effect; 1 small RCT with few events

Interventions for the treatment of pre‐eclampsia

Interventionist care versus expectant (delayed delivery) care for severe pre‐eclampsia (Churchill 2013)

Cerebral palsy at 2 years (definition not clear) assessed by family practitioner or paediatrician using short questionnaire

(assessment method taken from RCT manuscript as not detailed in review)

8 per 1000

(1/121)

50 per 1000 (6 to 398)

RR 6.01 (95% CI 0.75 to 48.14)

262 (1 RCT)

LOW

imprecision (‐2): wide confidence intervals crossing line of no effect; 1 small RCT with few events

Magnesium sulphate versus placebo for women with pre‐eclampsia (Duley 2010)

Severe cerebral palsy at 18 months defined as 'not walking or unlikely to walk unaided by 24 months; children screened with Ages and Stages Questionnaires; screen‐positive and a sample of screen‐negative children had clinical and neurodevelopmental assessments (using Bayley Scales of Infant Development; Griffiths Tests, or other); if this was not possible, clinical history and examination (using the Health Status Questionnaire)

(definition and assessment method taken from RCT manuscript as not detailed in review)

6 per 1000

(9/1464)

2 per 1000 (1 to 8)

RR 0.34 (95% CI 0.09 to 1.26)

2895 (1 RCT)

LOW

study limitations (‐1): 1 RCT with unclear risk of attrition bias for this outcome (2895 of 6922 children included in original RCT )

imprecision (‐1): wide confidence intervals crossing line of no effect

Interventions for the diagnosis and prevention of fetal compromise in labour

Continuous cardiotocography (CTG) versus intermittent auscultation (IA) for fetal assessment during labour (Alfirevic 2013)

Cerebral palsy at 18 months to 4 years

• 1 RCT: defined as 'non‐progressive disorder of movement or posture due to a defect in or damage to the developing brain'; a developmental paediatrician performed neurological examinations at 18 months

• 1 RCT: definition not clear; children with abnormal neurological signs in the neonatal period underwent a general physical and detailed neurological examination by a paediatrician at 4 years; other cases identified from records of specialist remedial clinics

(definitions and assessment methods taken from RCT manuscripts as not detailed in review)

3 per 1000

(17/6643)

4 per 1000 (2 to 9)

average RR 1.75 (95% 0.84 to 3.63)

13,252 (2 RCTs)

LOW

quality of evidence (GRADE), taken from published review:

"study limitations (‐1): most of the pooled effect provided by studies "B" or "C" without a substantial proportion (i.e. < 40% (actual 0% weight) from studies "C"

imprecision (‐1): 95% confidence interval around the pooled or best estimate of effect includes both 1) no effect and 2) appreciable benefit favouring IA"

Interventions for the prevention of preterm birth

Prenatal administration of progesterone versus placebo for preventing preterm birth in women with a previous history spontaneous preterm birth (singletons) (Dodd 2013)

Cerebral palsy at 4 years, definition not clear; assessed by general physical examination by paediatrician or nurse practitioner, or from chart abstraction

(age and assessment method taken from RCT manuscript as not detailed in review)

12 per 1000

(1/82)

2 per 1000 (0 to 42)

RR 0.14 (95% 0.01 to 3.48)

274 (1 RCT)

LOW

imprecision (‐2): wide confidence intervals crossing line of no effect; 1 small RCT with few events

Any prophylactic antibiotics versus no antibiotics for inhibiting preterm labour with intact membranes (Flenady 2013)

Cerebral palsy at 7 years, definition not clear; measured using proxy information provided by parents through a postal questionnaire (or by telephone in a small number) using validated tools

16 per 1000

(12/770)

28 per 1000 (15 to 52)

RR 1.82 (95% CI 0.99 to 3.34)

3173 (1 RCT)

MODERATE

imprecision (‐1): wide confidence interval crossing the line of no effect

Betamimetics versus placebo for inhibiting preterm labour (Neilson 2014)

Cerebral palsy at 18 months, definition not clear; assessed by paediatrician examination

(assessment method taken from RCT manuscript as not detailed in review)

41 per 1000

(5/121)

8 per 1000 (1 to 67)

RR 0.19 (95% CI 0.02 to 1.63)

246 (1 RCT)

LOW

study limitations (‐1): 1 RCT at unclear risk of attrition bias for this outcome

imprecision (‐1): wide confidence intervals crossing line of no effect; 1 small RCT

Magnesium sulphate versus other tocolytic agents for preventing preterm birth in threatened preterm labour (Crowther 2014)

Cerebral palsy at 18 months, definition not clear; assessed by neurodevelopment examinations at 4, 8, 12, and 18 months, with diagnoses made or verified by developmental paediatrician after 18 month examination

(assessment method taken from RCT manuscript as not detailed in review)

59 per 1000

(3/51)

8 per 1000 (1 to 148)

RR 0.13 (95% CI 0.01 to 2.51)

106 (1 RCT)

VERY LOW

study limitations (‐1): 1 RCT with unclear risk of selection, attrition, and reporting bias, and high risk of performance bias

imprecision (‐2): wide confidence intervals crossing line of no effect; 1 small RCT with few events

Interventions prior to preterm birth for fetalmaturation or neuroprotection

Phenobarbital versus placebo or no treatment prior to preterm birth for preventing neonatal periventricular haemorrhage (Crowther 2010a)

Cerebral palsy at 18 months to 3 years

• 1 RCT: defined as presence of hypertonicity, hyperreflexia, and dystonic or spastic movement quality in the affected extremity (including diplegia, hemiplegia, triplegia, or quadriplegia); assessed by certified examiners trained to perform neurologic examinations at 18 months

• 1 RCT: described as cerebral palsy associated with motor delay (diplegia, monoplegia, quadriplegia, and hemiplegia); assessed by trained nurse practitioner using detailed physical and neurologic examination at 3 years

(definitions and assessment methods taken from RCT manuscript as not detailed in review)

91 per 1000

(23/252)

61 per 1000 (37 to 117)

RR 0.71 (95% CI 0.40 to 1.28)

517 (2 RCTs)

LOW

study limitations (‐1): 2 RCTs with limitations: 1 with high risk of selection bias, bias due to lack of blinding and attrition bias; 2 with unclear risk of selection bias

imprecision (‐1): wide confidence intervals crossing line of no effect

Vitamin K versus placebo prior to preterm birth for preventing neonatal periventricular haemorrhage (Crowther 2010)

and

Phenobarbital versus placebo prior to preterm birth for preventing neonatal periventricular haemorrhage (Crowther 2010a)

Cerebral palsy at 7 years, definition not clear; method of assessment not clear

79 per 1000

(12/151)

61 per 1000 (26 to 140)

RR 0.77 (95% CI 0.33 to 1.76)

299 (1 RCT)

VERY LOW

study limitations (‐1): 1 RCT with unclear risk of selection bias, and high risk of attrition bias

indirectness (‐1): dual intervention of vitamin K and phenobarbital

imprecision (‐1): wide confidence intervals crossing line of no effect

Magnesium sulphate versus placebo for women at risk of preterm birth for neuroprotection of the fetus (Doyle 2009)

Cerebral palsy at 18 months to 2 years

• 1 RCT: defined as abnormalities of tone and loss of motor function; assessed by developmental paediatrician at 2 years

• 1 RCT: defined severe cerebral palsy as not walking or unlikely to walk unaided by 2 years; children screened with Ages and Stages Questionnaires; screen‐positive and a sample of screen‐negative children had clinical and neurodevelopmental assessments at 18 months (using Bayley Scales of Infant Development; Griffiths Tests; or other); if this was not possible, clinical history and examination (using the Health Status Questionnaire) was used

• 1 RCT: used the European Cerebral Palsy Network definition; paediatricians evaluated motor function at 2 years; if examination was not possible, parent telephone interview was used

• 1 RCT: definition not provided; assessed by neurodevelopment examinations at 4, 8, 12, and 18 months, with diagnoses made or verified by developmental paediatrician after the 18 month examination

• 1 RCT: defined as presenting with 2 or more of the following three features: a delay of 30% or more in gross motor developmental milestones; abnormality in muscle tone, 4+ or absent deep tendon reflexes, or movement abnormality; or persistence of primitive reflexes or absence of protective reflexes; assessed by paediatrician or paediatric neurologist at 2 years

50 per 1000

(154/3093)

34 per 1000 (27 to 43)

RR 0.68 (95% CI 0.54 to 0.87)

6145 (5 RCTs)

HIGH

Not downgraded

Antenatal corticosteroids versus placebo for accelerating fetal lung maturation for women at risk of preterm birth (Roberts 2006)

Cerebral palsy at 2 to 6 years

• 1 RCT: defined as any of hemiparesis, diplegia, tetra‐, quadriplegia; assessed using detailed physical and neurologic examination (including of motor co‐ordination) at 3 years

• 1 RCT: defined as pathological muscle tonus, pre‐existing primitive reflexes and delay in motor coordination; assessed using neurological examination by trial author (including evaluation of gross and fine motor) at 2 years

• 3 RCTs: definition not clear

(definitions and assessment methods taken, where possible, from RCT manuscripts as not detailed in review)

68 per 1000

(28/414)

41 per 1000 (23 to 70)

RR 0.60 (95% CI 0.34 to 1.03)

904 (5 RCTs)

LOW

study limitations (‐1): 2 RCTs with unclear and 1 RCT with inadequate allocation concealment

imprecision (‐1): wide confidence intervals crossing line of no effect

Repeat doses of corticosteroids versus single course for women at risk of preterm birth (Crowther 2015)

Cerebral palsy at 18 months to 3 years

• 1 RCT: defined as abnormalities of muscle tone as well as loss of motor function; assessed by developmental paediatrician using neurological examinations at 2 years

• 1 RCT: defined as non‐progressive motor impairment characterised by abnormal muscle tone and decreased range of movements; assessed (by neonatologists, general paediatricians, developmental paediatricians, and trained nurses) using a standardised neurological assessment at 18 months to 2 years

• 1 RCT: definition as described by Rosenbaum 2007; assessed by paediatric neurologist or paediatrician using specific neurological examination at 2 years

• 1 RCT: defined as severe delay in gross motor milestones, failure to walk by 17 months of corrected age; abnormality of tone or reflexes; and aberration of primitive reflexes or postural reactions; assessed by paediatricians or paediatric neurologists using neurologic examination between 2 and 3 years

(definitions and assessment methods taken from RCT manuscripts as not all detailed in review)

27 per 1000

(52/1891)

28 per 1000 (20 to 41)

RR 1.03 (95% CI 0.71 to 1.50)

3800 (4 RCTs)

MODERATE

imprecision (‐1): wide confidence intervals crossing line of no effect

Interventions for the management of preterm fetalcompromise

Immediate versus deferred delivery of the preterm baby with suspected fetal compromise (Stock 2016)

Cerebral palsy at or after 2 years

definition not clear; assessed by family practitioner or paediatrician using short questionnaire

(assessment method taken from RCT manuscript as not detailed in review)

8 per 1000

(2/251)

47 per 1000 (11 to 207)

RR 5.88 ( CI 95% 1.33 to 26.02)

507 (1 RCT)

MODERATE

study limitations (‐1): 1 RCT at high risk of performance bias and other bias (did not account for non‐independence of data for twin pregnancies)

Intervention and comparison

Outcome

Subgroup or sensitivity analysis

Assumed risk with comparator

Corresponding risk with intervention

Relative effect (95% CI)

Number of participants (trials)

Test for subgroup differences

Interventions for the diagnosis and prevention of fetalcompromise in labour

Continuous cardiotocography (CTG) versus intermittent auscultation (IA) for fetal assessment during labour (Alfirevic 2013)

Cerebral palsy at 18 months to 4 years

Pregnancy risk status

High

77 per 1000

195 per 1000 (85 to 451)

RR 2.54 (95% CI 1.10 to 5.86)

173 (1 RCT)

Chi² = 1.52, df = 1 (P = 0.22), I² = 34%

Mixed or not specified

2 per 1000

2 per 1000 (1 to 4)

RR 1.20 (95% CI 0.52 to 2.79)

13079 (1 RCT)

Onset of labour

Not specified

3 per 1000

4 per 1000 (2 to 8)

RR 1.74 (95% CI 0.97 to 3.11)

13,252 (2 RCTs)

Not applicable

Gestational age

Preterm labour

77 per 1000

195 per 1000 (85 to 451)

RR 2.54 (95% CI 1.10 to 5.86)

173 (1 RCT)

Chi² = 1.52, df = 1 (P = 0.22), I² = 34%

Both or gestation not specified

2 per 1000

2 per 1000 (1 to 4)

RR 1.20 (95% CI 0.52 to 2.79)

13,079 (1 RCT)

Number of babies

Singleton

77 per 1000

195 per 1000 (85 to 451)

RR 2.54 (95% CI 1.10 to 5.86)

173 (1 RCT)

Chi² = 1.52, df = 1 (P = 0.22), I² =34%

Both or not specified

2 per 1000

2 per 1000 (1 to 4)

RR 1.20 (95% CI 0.52 to 2.79)

13,079 (1 RCT)

Access to fetal blood sampling

Yes

3 per 1000

4 per 1000 (2 to 8)

RR 1.74 (95% CI 0.97 to 3.11)

13,252 (2 RCTs)

Not applicable

Parity

Both or not specified

3 per 1000

4 per 1000 (2 to 8)

RR 1.74 (95% CI 0.97 to 3.11)

13,252 (2 RCTs)

Not applicable

Quality

High

2 per 1000

2 per 1000 (1 to 4)

RR 1.20 (95% CI 0.52 to 2.79)

13,079 (1 RCT)

Chi² = 1.52, df = 1 (P = 0.22), I² = 34%

Unclear

77 per 1000

195 per 1000 (85 to 451)

RR 2.54 (95% CI 1.10 to 5.86)

173 (1 RCT)

Interventions for the prevention of preterm birth

Prenatal administration of progesterone versus placebo for preventing preterm birth in women with a previous history spontaneous preterm birth (singletons) (Dodd 2013)

Cerebral palsy at 4 years

Route of administration

Intramuscular

12 per 1000

2 per 1000 (0 to 42)

RR 0.14 (95% CI 0.01 to 3.48)

274 (1 RCT)

Not applicable

Prophylactic antibiotics versus no antibiotics for inhibiting preterm labour with intact membranes (Flenady 2013)

Cerebral palsy at 7 years

Type of antibiotic

Beta‐lactam antibiotics alone

16 per 1000

19 per 1000 (6 to 57)

Average RR 1.22 (95% CI 0.41 to 3.63)

1049 (1 RCT)

Chi² = 1.41, df = 2 (P = 0.49), I² = 0.0%

Macrolide antibiotics alone

16 per 1000

22 per 1000 (7 to 65)

Average RR 1.42 (95% CI 0.48 to 4.15)

1073 (1 RCT)

Macrolide and beta‐lactam antibiotics

16 per 1000

44 per 1000 (16 to 123)

Average RR 2.83 (95% CI 1.02 to 7.88)

1052 (1 RCT)

Any macrolide versus no macrolide for inhibiting preterm labour with intact membranes (Flenady 2013)

Any macrolide versus no macrolide antibiotics

17 per 1000

33 per 1000 (21 to 52)

RR 1.90 (95% CI 1.20 to 3.01)

3173 (1 RCT)

Not applicable

Any beta‐lactam versus no beta‐lactam for inhibiting preterm labour with intact membranes (Flenady 2013)

Any beta‐lactam versus no beta‐lactam antibiotics

19 per 1000

32 per 1000 (20 to 49)

RR 1.67 (95% CI 1.06 to 2.61)

3173 (1 RCT)

Not applicable

Interventions prior to preterm birth for fetal maturation or neuroprotection

Phenobarbital versus placebo prior to preterm birth for preventing neonatal periventricular haemorrhage (Crowther 2010a)

Cerebral palsy at 18 months to 3 years

Excluding trials with non‐concealment at randomisation (C quality)

91 per 1000

61 per 1000 (37 to 117)

RR 0.71 (95% CI 0.40 to 1.28)

517 (2 RCTs)

Not applicable

Vitamin K versus placebo prior to preterm birth for preventing neonatal periventricular haemorrhage (Crowther 2010)

and

Phenobarbital versus placebo prior to preterm birth for preventing neonatal periventricular haemorrhage (Crowther 2010a)

Cerebral palsy at 7 years

Excluding trials with inadequate concealment of allocation of treatment

79 per 1000

61 per 1000 (26 to 140)

RR 0.77 (95% CI 0.33 to 1.76)

299 (1 RCT)

Not applicable

Magnesium sulphate versus placebo for women at risk of preterm birth for neuroprotection of the fetus (Doyle 2009)

Cerebral palsy between 18 and 2 years

neuroprotective intent

Neuroprotective

65 per 1000

46 per 1000 (36 to 59)

RR 0.71 (95% CI 0.55 to 0.91)

4446 (4 RCTs)

Chi² = 1.69, df = 2 (P = 0.43), I² = 0%

(Performed by overview authors)

Maternal neuroprotective (pre‐eclampsia)

6 per 1000

3 per 1000 (1 to 13)

RR 0.40 (95% CI 0.08 to 2.05)

1593 (1 RCT)

Tocolytic

59 per 1000

7 per 1000 (1 to 148)

RR 0.13 (95% CI 0.01 to 2.51)

106 (1 RCT)

Single or multiple pregnancy

Single

28 per 1000

26 per 1000 (16 to 42)

RR 0.92 (95% CI 0.57 to 1.49)

2321 (2 RCTs)

Chi² = 1.28, df = 1 (P = 0.26), I² = 22.1%

(Performed by overview authors)

Multiple

53 per 1000

28 per 1000 (11 to 67)

RR 0.52 (95% CI 0.21 to 1.25)

527 (2 RCTs)

Gestational age

Less than 34 weeks at randomisation

56 per 1000

39 per 1000 (30 to 50)

RR 0.69 (95% CI 0.54 to 0.88)

5357 (5 RCTs)

Not applicable (subgroups not exclusive)

Less than 30 weeks at randomisation

56 per 1000

48 per 1000 (31 to 73)

RR 0.86 (95% CI 0.56 to 1.31)

1537 (2 RCTs)

Loading dose

4 g (any or no maintenance)

43 per 1000

34 per 1000 (24 to 47)

RR 0.79 (95% CI 0.56 to 1.10)

3595 (4 RCTs)

Chi² = 1.33, df = 1 (P = 0.25), I² = 24.6%

(Performed by overview authors)

6 g (any or no maintenance)

59 per 1000

35 per 1000 (24 to 50)

RR 0.59 (95% CI 0.40 to 0.85)

2444 (1 RCT)

Maintenance dose

No maintenance (any loading)

82 per 1000

113 per 1000 (15 to 879)

RR 1.37 (95% CI 0.18 to 10.70)

747 (2 RCTs)

Chi² = 0.44, df = 1 (P = 0.51), I² = 0%

(Performed by overview authors)

Any maintenance (any loading)

45 per 1000

31 per 1000 (23 to 41)

RR 0.68 (95% CI

0.51 to 0.91)

5292 (3 RCTs)

Loading and maintenance dose

Loading dose (4 g) and no maintenance

82 per 1000

113 per 1000 (15 to 879)

average RR 1.37 (0.18, 10.70)

747 (2 RCTs)

Chi² = 2.94, df = 3 (P = 0.40), I² = 0%

(Performed by overview authors)

Loading dose (4 g) and lower‐dose maintenance (1 g/hour)

33 per 1000

27 per 1000 (18 to 41)

average RR 0.81 (95% CI 0.54 to 1.23)

2848 (2 RCTs)

Loading dose (4 g) and higher‐dose maintenance (2 to 3 g/hour)

59 per 1000

8 per 1000 (1 to 148)

RR 0.13 (95% CI 0.01 to 2.51)

106 (1 RCT)

Loading dose (6 g) and higher‐dose maintenance (2 to 3 g/hour)

59 per 1000

35 per 1000 (24 to 50)

RR 0.59 (95% CI 0.40 to 0.85)

2444 (1 RCT)

Retreatment permitted

Yes

59 per 1000

35 per 1000 (24 to 50)

RR 0.59 (95% CI 0.40 to 0.85)

2444 (1 RCT)

Chi² = 1.26, df = 2 (P = 0.53), I² = 0%

(Performed by overview authors)

No

44 per 1000

33 per 1000 (24 to 46)

RR 0.76 (95% CI 0.55 to 1.06)

3536 (3 RCTs)

Unclear

38 per 1000

35 per 1000 (8 to 170)

RR 0.94 (95% CI 0.20 to 4.53)

165 (1 RCT)

High antenatal corticosteroids

66 per 1000

44 per 1000 (35 to 56)

RR 0.67 (95% CI 0.53 to 0.86)

4493 (4 RCTs)

Not applicable

Studies with lowest risk of bias only

62 per 1000

42 per 1000 (32 to 56)

RR 0.68 (95% CI 0.52 to 0.91)

3699 (2 RCTs)

Not applicable

Antenatal corticosteroids versus placebo for accelerating fetal lung maturation for women at risk of preterm birth (Roberts 2006)

Cerebral palsy at 2 to 6 years

In babies born from pregnancies complicated by hypertension syndromes

59 per 1000

16 per 1000 (2 to 177)

RR 0.28 (95% CI 0.03 to 3.01)

94 (1 RCT)

Not applicable

Main decade of recruitment

In babies from trials conducted in 1970s

28 per 1000

27 per 1000 (7 to 97)

RR 0.95 (95% CI 0.26 to 3.45)

322 (2 RCTs)

Chi² = 1.05, df = 2 (P = 0.59), I² = 0%

(Performed by overview authors)

In babies from trials conducted in 1980s

73 per 1000

45 per 1000 (20 to 100)

RR 0.62 (95% CI 0.28 to 1.38)

406 (1 RCT)

In babies from trials conducted in 1990s

136 per 1000

57 per 1000 (22 to 149)

RR 0.42 (95% CI 0.16 to 1.09)

176 (2 RCTs)

Intervention and comparison

Outcome

Assumed risk with comparator

Corresponding risk with intervention

Relative effect (95% CI)

Number of participants (trials)

Quality of the evidence (GRADE)

Comments

Interventions prior to preterm birth for fetalmaturation or neuroprotection

Magnesium sulphate versus placebo for women at risk of preterm birth for neuroprotection of the fetus (Doyle 2009)

Death or cerebral palsy between 18 months and 2 years (as above under cerebral palsy)

188 per 1000

(583/3093)

177 per 1000 (147 to 211)

Average RR 0.94 (95% CI 0.78 to 1.12)

6145 (5 RCTs)

HIGH

not downgraded

Intervention and comparison

Outcome

Assumed risk with comparator

Corresponding risk with intervention

Relative effect (95% CI)

Number of participants (trials)

Quality of the evidence (GRADE)

Comments

Interventions for the treatment of pre‐eclampsia

Magnesium sulphate versus placebo for women with pre‐eclampsia (Duley 2010)

Severe cerebral palsy at 18 months (definition: not walking or unlikely to walk unaided by 24 months)

(definition taken from RCT manuscript as not detailed in review)

6 per 1000

(9/1464)

2 per 1000 (1 to 8)

RR 0.34 (95% CI 0.09 to 1.26)

2895 (1 RCT)

LOW

study limitations (‐1): 1 RCT with unclear risk of attrition bias for this outcome (2895 of 6922 children in original RCT included)

imprecision (‐1): wide confidence intervals crossing line of no effect

Interventions prior to preterm birth for fetalmaturation or neuroprotection

Magnesium sulphate (neuroprotective intent) versus placebo for women at risk of preterm birth for neuroprotection of the fetus (Doyle 2009)

Mild cerebral palsy at 2 years

• 1 RCT: definition: walking at 2 years; assessed by developmental paediatrician at 2 years

• 1 RCT: definition: Gross Motor Function Classification System of level 1; assessed by paediatrician or paediatric neurologist at 2 years

• 1 RCT: definition not clear; paediatricians evaluated motor function at 2 years; if examination was not possible, parent telephone interview was used

33 per 1000

(74/2218)

25 per 1000 (17 to 35)

RR 0.74 (95% CI 0.52 to 1.04)

4387 (3 RCTs)

MODERATE

imprecision (‐1): wide confidence intervals crossing line of no effect

Moderate cerebral palsy at 2 years

• 1 RCT: definition: not walking at 2 years but likely to do so; assessed by developmental paediatrician at 2 years

• 1 RCT: definition not clear; paediatricians evaluated motor function at 2 years; if examination was not possible, parent telephone interview was used

22 per 1000

(21/962)

14 per 1000 (7 to 28)

RR 0.66 (95% CI 0.34 to 1.28)

1943 (2 RCTs)

MODERATE

imprecision (‐1): wide confidence intervals crossing line of no effect

Moderate to severe cerebral palsy at 2 years

• 1 RCT: definition: not walking at 2 years but likely to do so (moderate); not likely to walk (severe); assessed by developmental paediatrician at 2 years

• 1 RCT: definition: Gross Motor Function Classification System level of 2 or 3 (moderate), or level 4 or 5 (severe); assessed by paediatrician or paediatric neurologist at 2 years

• 1 RCT: definition not clear; paediatricians evaluated motor function at 2 years; if examination was not possible, parent telephone interview was used

32 per 1000

(72/2218)

21 per 1000 (14 to 30)

RR 0.64 (95% CI 0.44 to 0.92)

4387 (3 RCTs)

HIGH

not downgraded

Severe cerebral palsy at 2 years

• 1 RCT: definition: not likely to walk; assessed by developmental paediatrician at 2 years

• 1 RCT: definition not clear; paediatricians evaluated motor function a 2 years; if examination was not possible, parent telephone interview was used

14 per 1000

(13/962)

11 per 1000 (5 to 25)

RR 0.82 (95% CI 0.37 to 1.82)

1943 (2 RCTs)

MODERATE

imprecision (‐1): wide confidence intervals crossing line of no effect

Intervention and comparison

Outcome

Assumed risk with comparator

Corresponding risk with intervention

Relative effect (95% CI)

Number of participants (trials)

Quality of the evidence (GRADE)

Comments

Interventions for the treatment of pre‐eclampsia

Magnesium sulphate versus placebo for women with pre‐eclampsia (Duley 2010)

Neurosensory disability at 18 months (definition: functional blindness (binocular visual acuity < 6/60), deafness (severe enough to need a hearing aid), severe cerebral palsy, or DQ < 2 SD below the mean)

10 per 1000

(17/1648)

8 per 1000 (4 to 16)

RR 0.77 (95% CI 0.38 to 1.58)

3283 (1 RCT)

LOW

study limitations (‐1): 1 RCT with unclear risk of attrition bias for this outcome (3283 of 6922 children in original RCT included)

imprecision (‐1): wide confidence intervals crossing line of no effect

Death or neurosensory disability at 18 months (definition as above for 'neursensory disability at 18 months')

141 per 1000

(233/1648)

150 per 1000 (127 to 177)

RR 1.06 (95% CI 0.90 to 1.25)

3283 (1 RCT)

LOW

study limitations (‐1): 1 RCT with unclear risk of attrition bias for this outcome (3283 of 6922 children in original RCT included)

imprecision (‐1): wide confidence intervals crossing line of no effect

Interventions for the prevention of preterm birth

Magnesium sulphate versus other tocolytic agents for preventing preterm birth in threatened preterm labour (Crowther 2014)

Serious infant outcome (definition: total perinatal and infant mortality; IVH 3/4 or PVL; cerebral palsy at 18 months; assessment method as above under 'cerebral palsy')

(1 RCT included cerebral palsy in composite outcome)

59 per 1000

(3/51)

145 per 1000 (41 to 518)

RR 2.47 (95% CI 0.69 to 8.81)

106 (1 RCT)

VERY LOW

study limitations (‐1): 1 RCT with unclear risk of selection, attrition, and reporting bias and high risk of performance bias

imprecision (‐2): wide confidence intervals crossing line of no effect; 1 small RCT with few events

Interventions prior to preterm birth for fetalmaturation or neuroprotection

Magnesium sulphate versus placebo for women at risk of preterm birth for neuroprotection of the fetus (Doyle 2009)

Any neurologic impairment at 18 months or 2 years (definition: any of cerebral palsy, blindness, deafness, or developmental delay or intellectual impairment (DQ or IQ less than 1 SD below the mean))

141 per 1000

(200/1421)

142 per 1000 (121 to 167)

RR 1.01 (95% CI 0.86 to 1.19)

2848 (2 RCTs)

HIGH

not downgraded

Major neurological disability at 18 months or 2 years (definition: any of moderate or severe cerebral palsy, blindness, deafness, or an MDI less than 70)

64 per 1000

(91/1421)

69 per 1000 (53 to 90)

RR 1.07 (95% CI 0.82 to 1.40)

2848 (2 RCTs)

MODERATE

imprecision (‐1): wide confidence intervals crossing the line of no effect

Death or any neurologic impairment at 18 months or 2 years (definition as above for 'any neurological impairment')

348 per 1000

(495/1421)

348 per 1000 (317 to 387)

RR 1.00 (95% CI 0.91 to 1.11)

2848 (2 RCTs)

HIGH

not downgraded

Death or major neurological disability at 18 months or 2 years (definition as above for 'major neurological disability')

272 per 1000

(386/1421)

277 per 1000 (244 to 312)

RR 1.02 (95% CI 0.90 to 1.15)

2848 (2 RCTs)

HIGH

not downgraded

Antenatal corticosteroids versus placebo for accelerating fetal lung maturation for women at risk of preterm birth (Roberts 2006)

Neurodevelopmental delay at 2 years (definition: severe disability: tetraplegic cerebral palsy and/or mental retardation (Bayley's mental index < 70))

(definition taken from RCT manuscript as not detailed in review)

94 per 1000

(3/32)

60 per 1000 (13 to 279)

RR 0.64 (95% CI 0.14 to 2.98)

82 (1 RCT)

VERY LOW

study limitations (‐1): 1 RCT with unclear allocation concealment

imprecision (‐2): wide confidence intervals crossing line of no effect; 1 small RCT with few events

Repeat doses of corticosteroids versus single course for women at risk of preterm birth (Crowther 2015)

Survival free of any disability 18 months to 2 years, defined as:

• 1 RCT: definition: survival free of severe, moderate or mild neurosensory disability: severe neurosensory disability defined as severe cerebral palsy (child considered permanently non‐ambulant), severe developmental delay (MDI score, > 3 SD below the mean), or blindness; moderate disability defined as moderate cerebral palsy (child non‐ambulant at 2 years but likely to walk), moderate developmental delay (MDI score, > 2 SD to 3 SD below the mean), or deafness; mild disability defined as either mild cerebral palsy (walking at 2 years) or mild developmental delay (MDI score, > 1 SD to 2 SD below the mean)

• 1 RCT: definition: survival free of neurological impairment (cerebral palsy or cognitive delay (2 SD below the normative value))

(definitions taken from RCT manuscripts as not clearly detailed in review)

773 per 1000

(1215/1571)

781 per 1000 (750 to 812)

RR 1.01 (95% CI 0.97 to 1.05)

3155 (2 RCTs)

HIGH

not downgraded

Survival free of major neurosensory disability at 2 to 3 years, defined as:

• 1 RCT: definition: survival being ambulant by 2 years of age and not having blindness, deafness, a developmental index score of more than 2 SD below the mean, or cerebral palsy

• 1 RCT: definition: survival without severe neurodevelopmental impairment (cerebral palsy, MDI < 70, DQ < 70, deafness, or blindness)

(definitions taken from RCT manuscripts as not clearly detailed in review)

847 per 1000

(572/675)

856 per 1000 (780 to 941)

Average RR 1.01 (95% CI 0.92 to 1.11)

1317 (2 RCTs)

LOW

study limitations (‐1): 1 RCT with unclear risk of selection bias, attrition bias and high risk of other bias

inconsistency (‐1): substantial heterogeneity (I² = 88%)

Major neurosensory disability at 2 to 3 years, defined as:

• 1 RCT: definition: severe or moderate neurosensory disability: severe neurosensory disability defined as severe cerebral palsy (child considered permanently non‐ambulant), severe developmental delay (MDI score > 3 SD below the mean), or blindness; moderate disability defined as moderate cerebral palsy (child non‐ambulant at 2 years but likely to walk), moderate developmental delay (MDI score > 2 SD to 3 SD below the mean), or deafness

• 1 RCT: definition: severe neurodevelopmental impairment (cerebral palsy, MDI < 70, DQ < 70, deafness, or blindness)

(definitions taken from RCT manuscripts as not clearly detailed in review)

110 per 1000 (71/643)

119 per 1000 (34 to 415)

Average RR 1.08 (95% CI 0.31 to 3.76)

1256 (2 RCTs)

LOW

study limitations (‐1): 1 RCT with unclear risk of selection bias, attrition bias, and high risk of other bias

imprecision (‐1): wide confidence intervals crossing line of no effect

Disability at 2 years, defined as:

• 1 RCT: definition: severe, moderate, or mild neurosensory disability: severe neurosensory disability defined as severe cerebral palsy (child considered permanently non‐ambulant), severe developmental delay (MDI score > 3 SD below the mean), or blindness; moderate disability defined as moderate cerebral palsy (child non‐ambulant at 2 years but likely to walk), moderate developmental delay (MDI score > 2 SD to 3 SD below the mean), or deafness; mild disability was defined as either mild cerebral palsy (walking at 2 years) or mild developmental delay (MDI score > 1 SD to 2 SD below the mean)

(definition taken from RCT manuscript as not clearly detailed in review)

361 per 1000 (182/504)

354 per 1000 (300 to 419)

RR 0.98 (95% CI 0.83 to 1.16)

999 (1 RCT)

HIGH

not downgraded

Composite serious outcome at 18 months to 2 years, defined as:

• 1 RCT: definition: death or any neurosensory disability: severe, moderate or mild neurosensory disability: severe neurosensory disability defined as severe cerebral palsy (child considered permanently non‐ambulant), severe developmental delay (MDI score > 3 SD below the mean), or blindness; moderate disability defined as moderate cerebral palsy (child non‐ambulant at 2 years but likely to walk), moderate developmental delay (MDI score > 2 SD to 3 SD below the mean), or deafness; mild disability defined as either mild cerebral palsy (walking at 2 years) or mild developmental delay (MDI score > 1 SD to 2 SD below the mean)

• 1 RCT: definition: death or neurologic impairment (cerebral palsy or cognitive delay (cognitive delay was defined as 2 SD below the normative value))

(definitions taken from RCT manuscripts as not clearly detailed in review)

227 per 1000

(356/1571)

224 per 1000 (197 to 254)

RR 0.99 (95% CI 0.87 to 1.12)

3164 (2 RCTs)

HIGH

not downgraded

Interventions for the management of preterm fetalcompromise

Immediate versus deferred delivery of the preterm baby with suspected fetal compromise (Stock 2016)

Death or disability at or after 2 years (definition: cerebral palsy, little or no vision, requirement for hearing aid, or Griffiths DQ of 70 or less)

155 per 1000

(44/283)

190 per 1000 (132 to 272)

RR 1.22 (95% CI 0.85 to 1.75)

573 (1 RCT)

LOW

study limitations (‐1): 1 RCT at high risk of performance bias and other bias (did not account for non‐independence of data for twin pregnancies)

imprecision (‐1): wide intervals crossing line of no effect

Neurodevelopmental impairment at or after 2 years (definition: cerebral palsy, little or no vision, requirement for hearing aid, or Griffiths DQ of 70 or less)

48 per 1000

(12/251)

82 per 1000 (41 to 163)

RR 1.72 (95% CI 0.86 to 3.41)

507 (1 RCT)

LOW

study limitations (‐1): 1 RCT at high risk of performance bias and other bias (did not account for non‐independence of data for twin pregnancies)

imprecision (‐1): wide intervals crossing line of no effect

Death or severe disability at 6 to 13 years (definition: classified severe blindness, severe deafness, cerebral palsy, or Kaufman Mental Processing Component < 70 points)

(definition taken from RCT manuscript as not detailed in review)

168 per 1000

(25/149)

138 per 1000 (81 to 235)

RR 0.82 (95% CI 0.48 to 1.40)

302 (1 RCT)

LOW

study limitations (‐1): 1 RCT at high risk of performance, attrition and other bias

imprecision (‐1): wide intervals crossing line of no effect

Intervention and comparison

Outcome

Assumed risk with comparator

Corresponding risk with intervention

Relative effect (95% CI)

Number of participants (trials)

Quality of the evidence (GRADE)

Comments

Interventions for the prevention of preterm birth

Prenatal administration of progesterone versus placebo for preventing preterm birth in women with a previous history spontaneous preterm birth (singletons) (Dodd 2013)

Motor impairment at 4 years (definition: overall activity problems or co‐ordination problems)

(age and definition taken from RCT manuscript as not detailed in review)

24 per 1000

(2/82)

16 per 1000 (3 to 92)

RR 0.64 (95% CI 0.11 to 3.76)

274 (1 RCT)

LOW

imprecision (‐2): wide confidence intervals crossing line of no effect; one small RCT with few events

Interventions prior to preterm birth for fetal maturation or neuroprotection

Phenobarbital versus control prior to preterm birth for preventing neonatal periventricular haemorrhage (Crowther 2010a)

Other neuromotor impairment at 3 years (definition: tonal abnormalities with no delay in ambulation or other motor milestones)

(definition taken from RCT manuscript as not detailed in review)

73 per 1000

(4/55)

49 per 1000 (9 to 254)

RR 0.67 (95% CI 0.13 to 3.49)

96 (1 RCT)

VERY LOW

study limitations (‐2): 1 RCT with high risk of selection bias, bias due to lack of blinding, and attrition bias

imprecision (‐2): wide confidence intervals crossing line of no effect; 1 small RCT with few events

Magnesium sulphate versus no placebo for women at risk of preterm birth for neuroprotection of the fetus (Doyle 2009)

Substantial gross motor dysfunction between 18 months and 2 years (definition: motor dysfunction such that the child was not walking at age 2 years or later, or the inability to grasp and release a small block with both hands)

31 per 1000

(94/3013)

19 per 1000 (14 to 27)

RR 0.61 (95% CI 0.44 to 0.85)

5980 (4 RCTs)

HIGH

not downgraded

Death or substantial gross motor dysfunction between 18 months and 2 years (definition as above for 'substantial gross motor dysfunction')

174 per 1000

(523/3013)

160 per 1000 (130 to 194)

Average RR 0.92 (95% CI 0.75 to 1.12)

5980 (4 RCTs)

MODERATE

inconsistency (‐1): substantial heterogeneity (I² = 65%)