Neonatal interventions for preventing cerebral palsy: an overview of Cochrane systematic reviews

  • Protocol
  • Overview

Authors

  • Emily Shepherd,

    Corresponding author
    1. The University of Adelaide, ARCH: Australian Research Centre for Health of Women and Babies, Robinson Research Institute, Discipline of Obstetrics and Gynaecology, Adelaide, South Australia, Australia
    • Emily Shepherd, ARCH: Australian Research Centre for Health of Women and Babies, Robinson Research Institute, Discipline of Obstetrics and Gynaecology, The University of Adelaide, Adelaide, South Australia, 5006, Australia. emily.shepherd@adelaide.edu.au.

  • Philippa Middleton,

    1. Healthy Mothers, Babies and Children, South Australian Health and Medical Research Institute, Adelaide, South Australia, Australia
  • Maria Makrides,

    1. Healthy Mothers, Babies and Children, South Australian Health and Medical Research Institute, Adelaide, South Australia, Australia
  • Sarah McIntyre,

    1. University of Sydney, Research Institute, Cerebral Palsy Alliance, Sydney, Australia
  • Nadia Badawi,

    1. University of Sydney, Research Institute, Cerebral Palsy Alliance, Sydney, Australia
    2. The Children's Hospital at Westmead, Grace Centre for Newborn Care, Sydney, Australia
  • Caroline A Crowther

    1. The University of Auckland, Liggins Institute, Auckland, New Zealand

Abstract

This is a protocol for a Cochrane Review (Overview). The objectives are as follows:

The primary objective of this overview is to summarise the evidence from Cochrane systematic reviews regarding the effects of neonatal interventions for preventing cerebral palsy (reducing cerebral palsy risk).

A secondary objective of this overview is to summarise any evidence identified from Cochrane systematic reviews regarding the effects of neonatal interventions that increase cerebral palsy.

Background

Description of the condition

Cerebral palsy: definition and prevalence

‘Cerebral palsy’ was originally (and continues to be) defined by clinical description, at a time when there was little knowledge of aetiology or pathology (Morris 2007). Today, many registries and surveillance programs, including in Australia, the United Kingdom and Europe, highlight five key elements of cerebral palsy: it is an ‘umbrella term’; it is permanent but not unchanging; it involves a disorder of movement or posture or both, and of motor function; it is due to a non-progressive interference, lesion or abnormality; the interference, lesion or abnormality arose in the developing or immature brain (Cans 2000; Mutch 1992; Rosenbaum 2007; Smithers-Sheedy 2014). As cerebral palsy is defined by clinical description which may change over time, a longer time span for diagnosis is considered useful to confirm the condition meets criteria for cerebral palsy, and to accurately describe the motor impairment. Thus, final ascertainment for surveillance programmes across the world range from four to 12 years, with many considering data to be 'complete' at or near five years (Smithers-Sheedy 2014). While average age for diagnosis has been around 18 months, recent evidence has suggested that cerebral palsy may be reliably detected as early as three to four months post-term age, using tests such as Prechtl's Qualitative Assessment of General Movements and medical resonance imaging (Bosanquet 2013; Morgan 2016).

Cerebral palsy is the most common physical disability in childhood. In a recent meta-analysis, including 19 studies (with varying ages of ascertainment), the global pooled prevalence was 2.11 per 1000 live births (95% CI 1.98 to 2.25); a cumulative meta-analysis demonstrated stability over the past 10 years (Oskoui 2013). Similar rates have been shown in countries that have used consistent methods of ascertainment for over 20 years (such as Australia, Sweden and England), with most published estimates in the region of 2 per 1000 (Blair 2006). In low- and middle-income countries, prevalence estimates have tended to be in a similar range or higher (Blair 2006; Cans 2000). However, there is now emerging evidence, including from Australia and Europe, that the overall rates and severity of the condition are starting to decline for the first time (Reid 2015; Sellier 2015).

Cerebral palsy: causes and risk factors

For approximately 6% of individuals with cerebral palsy, their brain injury was acquired during an event more than 28 days after birth (ACPR Group 2013). For the remaining 94% of individuals, their brain injury occurred during pregnancy, birth, or the first 28 days of life (ACPR Group 2013). Preterm birth is one of the principal risk factors for cerebral palsy and associated neurosensory disabilities (Himpens 2008; Oskoui 2013), with over 40% of individuals with cerebral palsy born preterm (ACPR Group 2013). More than half of all individuals with cerebral palsy are, however, born at term (ACPR Group 2013).

Studies on antenatal, intrapartum and neonatal risk factors for cerebral palsy are abundant. Though a great number of risk factors have been identified, their commonality is that separately, or in combination, they influence potentially preventable pathways to brain injury. Risk factors commonly reported include: i) factors prior to conception, e.g. low or advanced maternal age, high parity, nulliparity, a short or long inter-pregnancy interval, a history of stillbirth, multiple miscarriages, neonatal death or preterm birth, family history of cerebral palsy and other genetic predispositions, low socioeconomic status, and pre-existing maternal conditions (such as epilepsy or intellectual disability); ii) factors in early pregnancy, e.g. male sex, multiple gestation, congenital malformations or birth defects, and infections (such as TORCH complex (toxoplasmosis (parasite), other infections, rubella, cytomegalovirus, herpes simplex virus)); iii) factors during pregnancy, e.g. maternal disease (such as thyroid disorders), pregnancy complications (such as pre-eclampsia, placenta praevia and placental abruption), intrauterine infection or inflammation and chorioamnionitis, intrauterine growth restriction, and other precursors to preterm birth; and iv) factors around the time of birth and neonatal period, e.g. acute intrapartum hypoxic events and neonatal encephalopathy, neonatal brain injury (such as intraventricular haemorrhage, periventricular leukomalacia and hydrocephalus), strokes or seizures, cardiovascular disorders (such as patent ductus arteriosus and hypotension), respiratory disorders and associated prolonged ventilation (such as for respiratory distress syndrome or bronchopulmonary dysplasia), infection (such as sepsis and necrotizing enterocolitis), metabolic or endocrine disorders (such as hypoglycaemia and hypothyroidism), neonatal jaundice, along with inborn errors of metabolism, particular syndromes or chromosomal abnormalities (Badawi 2005; Dixon 2002; Drougia 2007; Jacobsson 2004; McIntyre 2011; McIntyre 2013; Murphy 1997; Nelson 2008; Tran 2005; Walstab 2004).

Research has shown that contrary to previous beliefs, birth asphyxia is a relatively rare cause of cerebral palsy (Blair 1988; Ellenberg 2013). A growing body of evidence suggests that genetic abnormalities contribute in some cases (MacLennan 2015; Moreno-De-Luca 2012; O’Callaghan 2009; Oskoui 2015). Common risk factors in the post-neonatal period (some of which also contribute in the neonatal period) include: infection (such as meningitis/encephalitis, or severe infection and subsequent severe dehydration), head injury (such as from traffic accidents, other traumatic injury, or non-accidental injury), vascular episodes (such as post cardiac or brain surgery), and other events (such as near drowning or near sudden infant death) (Cans 2004; Germany 2013).

Cerebral palsy: consequences

Cerebral palsy is the leading cause of physical disability for children, and is a condition with life-long impact. Most individuals will survive to adulthood, with some studies suggesting life expectancy can then be similar to that of the general population (Colver 2012). For known cases of antenatally or neonatally acquired cerebral palsy, the 20-year survival rate has been estimated at 90%. However, strong associations between increasing motor impairment, severe intellectual impairment, number of severe impairments, and early mortality have been shown (Blair 2001; Hemming 2005; Reid 2012). Frequently used definitions for cerebral palsy acknowledge common co-occurring impairments, diseases and functional limitations (Rosenbaum 2007). A recent systematic review estimated that among children with cerebral palsy, "1 in 2 had an intellectual disability… 1 in 4 could not talk; 1 in 4 had epilepsy; 1 in 4 had a behavior disorder… 1 in 10 were blind… and 1 in 25 were deaf" (Novak 2012).

Economic studies have estimated lifetime costs of cerebral palsy, including health care, social care, and productivity costs as EUR 860,000 for men and EUR 800,000 for women in Denmark (in 2000) (Kruse 2009), and USD 921,000 for individuals in the United States (in 2003) (CDC 2004). In Australia, the financial cost of cerebral palsy was an estimated AUD 1.47 billion (in 2007); and the value of lost well-being a further AUD 2.4 billion (Access Economics 2008).

The impacts of cerebral palsy are considerable (Davis 2010). Accordingly, the identification of primary preventive measures has been identified as a key priority, by individuals with cerebral palsy, their families, clinicians and researchers (McIntyre 2010).

Description of the interventions

Neonatal approaches to prevention of cerebral palsy

Research efforts aimed at the prevention of cerebral palsy have increasingly focused on understanding the causes of cerebral palsy. As it is now widely recognised that causes differ by, for example, gestational age (e.g. for preterm and term-born children), and also by clinical subtype of cerebral palsy, it is reasonable to consider that successful primary preventive interventions will also vary according to different aetiologies or causal factors.

In this overview, therefore, we will include a broad range of neonatal interventions (with varying primary aims or indications) which may mediate cerebral palsy risk, including (but not limited to):

  • interventions for neonates following birth asphyxia or with evidence of encephalopathy (e.g. cooling; erythropoietin; darbepoetin; allopurinol; melatonin; magnesium sulphate; anticonvulsants; xenon; naloxone; dopamine; fluid restriction; acupuncture; umbilical cord stem cells);

  • interventions for neonates with neurologic disorders, such as for intracranial haemorrhage or post-haemorrhagic hydrocephalus (e.g. heparin; antithrombin; phenobarbital; diuretic therapy; erythropoietin; repeated lumbar or ventricular punctures); or for seizures (anticonvulsants);

  • interventions for neonates requiring resuscitation (e.g. air or oxygen for positive pressure ventilation; lower or higher oxygen concentrations titrated to target oxygen saturations; face mask, laryngeal mask airway, nasal airway or endotracheal intubation; positive end-expiratory pressure; respiratory function monitoring);

  • interventions for neonates with cardiovascular disorders, such as for hypotension (e.g. corticosteroids; inotropes; early volume expansion; adrenaline; dopamine; dobutamine), or for patent ductus arteriosus (e.g. ibuprofen; indomethacin; fluid restriction; surgical ligation);

  • interventions for neonates with respiratory disorders, such as for apnoea of prematurity (e.g. kinaesthetic stimulation; methylxanthines (caffeine)); for respiratory distress syndrome (e.g. early or delayed, prophylactic or selective, protein-containing or protein-free, animal-derived or synthetic pulmonary surfactant; thyroid hormones; continuous distending pressure); or for bronchopulmonary dysplasia (chronic lung disease) (e.g. early or late, inhaled or systemic, postnatal corticosteroids);

  • interventions for gastrointestinal tract disorders, such as for necrotizing enterocolitis (e.g. lactoferrin; probiotics; antibiotics; immunoglobulin; peritoneal drainage; laparotomy);

  • interventions for neonates with infection, such as for general infection control (e.g. chlorhexidine skin or cord care; patient isolation for infection; gowning by attendants and visitors in newborn nurseries); for fungal and protozoal infections (e.g. prophylactic antifungal agents; or antifungal therapy for invasive fungal infection); for viral infections (e.g. antiviral agents for treatment of herpes simplex virus or cytomegalovirus infection); or for bacterial infections (e.g. intravenous immunoglobulin for prevention of infection, or for suspected or proven infection; antibiotics for suspected early or late onset sepsis; intraventricular antibiotics for meningitis; prophylactic antibiotics for ventilated newborns);

  • interventions for neonates with metabolic or endocrine disorders, such as for disorders of carbohydrate metabolism (e.g. oral dextrose gel for hypoglycaemia; insulin for hyperglycemia); or for thyroid disorders (postnatal thyroid hormones);

  • interventions for neonates with jaundice and liver disorders (e.g. phototherapy);

  • interventions focused on nutrition or metabolism for high-risk neonates (i.e. preterm or low birthweight neonates, or both) including enteral nutrition interventions (e.g. high protein intake; donor breast milk; nutrient-enriched formula; multi-nutrient fortification of human breast milk; responsive or scheduled feeding); parenteral nutrition interventions (e.g. early or late, high or low amino acid administration); or vitamin or mineral supplementation (e.g. glutamine; arginine; iodine; vitamin E);

  • interventions for neurodevelopmental care or physical environment management (or both) for neonates, e.g. developmental care to reduce stressors in the neonatal nursery; kangaroo mother care; massage; co-bedding in the neonatal nursery; early developmental programmes post-discharge to prevent motor and cognitive impairments;

  • interventions for all neonates at birth, such as newborn screening for inborn errors of metabolism.

We will not consider interventions in the antenatal or intrapartum period (such as magnesium sulphate for fetal neuroprotection (Doyle 2009)), as these interventions will be assessed in a separate overview (Shepherd 2016, under review).

How the intervention might work

Advances in research into several factors that modify the risk of cerebral palsy suggest many opportunities for prevention, with the main neonatal strategies focusing on protection of the immature brain through administration of neuroprotective agents or therapies.

For many individuals born at or near term who develop cerebral palsy, their neonatal course was seemingly unremarkable, with the exception of those following perinatal asphyxia and with neonatal encephalopathy (brain injury which may be due to cerebral hypoxia and ischaemia prior to birth) (Badawi 2005; O’Shea 2008). For these neonates, therapeutic hypothermia, applied selectively to the head (as a ‘cooling cap’) or to the whole body, is one such intervention able to mediate cerebral palsy risk (O’Shea 2008). Beyond cooling, there are a range of other interventions (including to be used as adjuvant therapy with cooling) which may contribute to cerebral palsy prevention, either through protecting against secondary cell death and brain damage following hypoxic-ischaemic insult (Robertson 2012), or through treating the underlying cause(s) of encephalopathy (such as infection or metabolic derangement).

For preterm and very low birthweight neonates, or other groups of neonates (such as those with hypoglycaemia) who are at increased risk of brain injury, there are many pharmacological and non-pharmacological interventions in the neonatal period that may mediate cerebral palsy risk (O’Shea 2008). While these interventions differ in their primary aims (such as maintaining adequate ventilation (e.g. through the treatment of apnoea of prematurity with caffeine); maintaining normal metabolic status (e.g. through the treatment of neonatal hypoglycaemia with dextrose gel); or controlling neonatal seizures (e.g. through use of anticonvulsants)), each may contribute to cerebral palsy prevention through reducing the likelihood or severity of brain injury, and thus of long-term neurodevelopmental sequelae.

Why it is important to do this overview

A multitude of individual studies and Cochrane systematic reviews assessing a broad range of neonatal interventions (with varying primary aims or indications) recognise the potential for the intervention of interest to influence cerebral palsy risk. With the acknowledgement that there are many and varied risk factors for cerebral palsy, and that causes of cerebral palsy differ, there is a need to systematically consider all potentially relevant interventions for their ability to contribute to reducing cerebral palsy risk. With new data suggesting possible declining rates and severity of cerebral palsy, it is important to understand the different interventions which may, together, be contributing to these observations.

To our knowledge, to date no ‘overview’ has brought together the evidence around neonatal interventions for cerebral palsy prevention from Cochrane systematic reviews into one coherent document to be used by researchers, funding bodies, policy makers, clinicians and consumers to aid decision making and evidence implementation.

While the objective of this overview is to summarise the evidence from Cochrane systematic reviews regarding the effects of neonatal interventions for preventing cerebral palsy, it is also important to consider whether such interventions may, instead, actually contribute to increasing cerebral palsy risk.

Is an overview the right approach?

We have followed the Editorial Decision Tree proposed by Cochrane's Comparing Multiple Interventions Methods Group to establish whether our review would better fit an overview format or an intervention review format.

Specifically:

  • we will be reviewing systematic reviews, instead of individual trials;

  • we will not compare multiple interventions with the intention of drawing inferences about the comparative effectiveness of the interventions;

  • we intend to present a map of the evidence from systematic reviews but with no attempt to rank the interventions.

Based on this, the Editorial Decision Tree recommends an overview as the right format for the review.

Objectives

The primary objective of this overview is to summarise the evidence from Cochrane systematic reviews regarding the effects of neonatal interventions for preventing cerebral palsy (reducing cerebral palsy risk).

A secondary objective of this overview is to summarise any evidence identified from Cochrane systematic reviews regarding the effects of neonatal interventions that increase cerebral palsy.

Methods

Criteria for considering reviews for inclusion

In this overview of systematic reviews, we will include only published Cochrane systematic reviews of neonatal interventions, as long as cerebral palsy is reported as a primary or secondary review outcome. Cochrane protocols and titles will be identified for future inclusion, and classified as 'Ongoing reviews' (in an Appendix).

We will make note of the publication and search dates of the reviews; however, we will not attempt to update the individual systematic reviews.

Participants:

We will consider reviews that include:

  • neonates with perinatal asphyxia or with evidence of neonatal encephalopathy;

  • neonates born preterm or low or very low birthweight (or both preterm and low/very low birthweight).

We will also include reviews that include other groups of 'at risk' neonates (for example, neonates with hypoglycaemia) so long as the intervention assessed in the Cochrane systematic review is recognised by the review authors as having the potential to influence cerebral palsy risk — cerebral palsy must be pre-specified as a primary or secondary outcome in the review.

Interventions:

We will consider all types of interventions used in the neonatal period compared with placebo, no treatment, or an alternative intervention.

We will include both pharmacological and non-pharmacological interventions (see Description of the interventions for further description of possible interventions).

Outcomes of interest:

Primary:
  • Cerebral palsy (regardless of criteria used for diagnosis by review authors or trialists, and regardless of age at diagnosis; however we will clearly report any variation).

Secondary:
  • Cerebral palsy or death (regardless of criteria used for diagnosis by review authors or trialists, and regardless of age at diagnosis; however we will clearly report any variation).

  • Other composite outcomes that include cerebral palsy as a component (regardless of criteria used for diagnosis by review authors or trialists, and regardless of age at diagnosis; however we will clearly report any variation).

  • Severity of cerebral palsy (e.g. according to: Gross Motor Function Classification System (GMFCS); Manual Ability Classification System (MACS); Communication Function Classification System (CFCS)).

  • Type of cerebral palsy (e.g. according to topography (diplegia; hemiplegia; quadriplegia; monoplegia; triplegia) or motor type (spastic; dyskinetic; ataxic)).

  • Motor dysfunction (regardless of criteria used for diagnosis by review authors or trialists, and regardless of age at diagnosis; however we will clearly report any variation).

To be included, a review must pre-specify our overview’s primary outcome — cerebral palsy (or a composite outcome that includes cerebral palsy*) as a primary or secondary systematic review outcome — and have reported data for this outcome from at least one of the included trials in the review.

Reviews which pre-specify cerebral palsy as a primary or secondary systematic review outcome, but have no reported data from included trials on this outcome, will be included as 'Reviews awaiting further classification' and will be re-considered in future updates of the overview.

* Wherever possible, we will extract data related to the cerebral palsy from any composite outcomes that include cerebral palsy. Where it is not possible to extract only the cerebral palsy data from such composite outcomes, we will report the composite outcome data; however we will report it separately to the data for our primary outcome (i.e. as a secondary outcome).

Search methods for identification of reviews

We will search the Cochrane Database of Systematic Reviews using the term ‘cerebral palsy’. The search term will be used to search ‘all text’, and not limited to ‘title, abstract, or keywords’. We will not apply any language or date restrictions. No other databases will be searched. Citations retrieved through the search will be managed using Covidence (Covidence 2015).

Data collection and analysis

The methodology for data collection and synthesis will be based on Chapter 22 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Where appropriate, the overview will be prepared using Covidence (Covidence 2015) and Review Manager 5 software (RevMan 2014).

Selection of reviews

Two reviewers will independently assess for inclusion all potential systematic reviews we identify that assess the effects of neonatal interventions and report on cerebral palsy. We will resolve any disagreement through discussion or, if required, we will consult a third member of the overview team.

Data extraction and management

Two overview authors will independently extract data from the reviews using a pre-defined data extraction form. We will resolve discrepancies through discussion or, if needed, through discussion with a third overview author. Where information regarding review outcomes is unclear or missing, we will contact the individual review authors, or access the published papers of the individual studies for further details.

Information will be extracted on the following.

  • Review characteristics:

    • Review title and authors.

    • Date that the review was last assessed as up to date.

    • Number of included trials, and number of participants (neonates) in the trials and their characteristics (e.g. countries where the trials were conducted and inclusion criteria in the trials).

    • Quality of the included trials (as reported by the review authors; see 'Quality of included studies within reviews' below under Assessment of methodological quality of included reviews).

    • Interventions and comparisons relevant to this overview.

    • All pre-specified outcomes relevant to this overview (their definitions, and whether they were primary or secondary outcomes in the included reviews).

    • Any other characteristics required to assess and report on review quality (see 'Quality of included reviews' under Assessment of methodological quality of included reviews).

  • Statistical summaries*:

    • The summary intervention effects (including the pooled effects (e.g. risk ratios (RRs), odds ratios (ORs) or mean differences (MDs) as reported in the individual reviews), 95% confidence intervals (CIs), and numbers of studies and participants contributing data to each pooled effect) from comparisons and for outcomes relevant to this overview.

    • Information required to assess and report on the quality of the evidence for the intervention effects extracted above (see 'Quality of evidence in included reviews' under Assessment of methodological quality of included reviews).

* Where reviews were not able to perform meta-analyses and therefore did not report statistical summaries, we plan to extract from those reviews the narrative text relating to the results for our overview outcomes.

Assessment of methodological quality of included reviews

Quality of included reviews

We will assess the methodological quality of each systematic review using the AMSTAR (A Measurement Tool to Assess Systematic Reviews) instrument (Shea 2009). AMSTAR evaluates the methods used in a review against 11 distinct criteria and assesses the degree to which review methods are unbiased. Each item on AMSTAR is rated as 'yes' (clearly done), 'no' (clearly not done), 'cannot answer', or 'not applicable'. These criteria, and the way they will assess review quality, are as follows:

  1. Was an ‘a priori’ design provided?

  2. Was there duplicate study selection and data extraction?

  3. Was a comprehensive literature search performed?

  4. Was the status of publication used as an inclusion criterion?

  5. Was a list of studies (included and excluded) provided?

  6. Were the characteristics of the included studies provided?

  7. Was the scientific quality of the included studies assessed and documented?

  8. Was the scientific quality of the included studies used appropriately in formulating conclusions?

  9. Were the methods used to combine the findings of studies appropriate?

  10. Was the likelihood of publication bias assessed?

  11. Was the conflict of interest stated?

For all items except item 4, a rating of 'yes' is considered adequate. For item 4, a rating of ‘no’ is considered adequate. A review that adequately meets all of the 11 criteria is considered to be a review of the highest quality (Shea 2009). For this overview, we will consider reviews that achieve scores of between 8 to 11 as high quality; scores of 4 to 7 as medium quality; and scores of 0 to 3 as low quality.

To further assess the risk of bias of the systematic reviews, we will additionally use the new ROBIS (Risk Of Bias In Systematic reviews) tool (Whiting 2015). The tool considers risk of bias across four key domains:

  • Study eligibility criteria.

  • Identification and selection of studies.

  • Data collection and study appraisal.

  • Synthesis and findings.

A series of questions within each of the domains elicit information about possible limitations of the systematic review, leading to a judgement about the concerns within that domain (Low, High, or Unclear). The risk of bias of the review as a whole is then considered, with signalling questions and information to support overall judgement of risk of bias (Low, High, or Unclear) (Whiting 2015).

Two overview authors will independently assess the quality of the included reviews using AMSTAR and ROBIS, and another overview author will verify this assessment. We will resolve differences through discussion or, if needed, through discussion with a third overview author.

We will also note and report for each review the publication and search date.

Quality of included studies within reviews

We will not reassess the quality of included studies within reviews but instead will report study quality according to the review authors' assessment. We will collect this information during the data extraction process.

Quality of evidence in included reviews

The quality of the evidence for our primary outcome (cerebral palsy) and secondary review outcomes will be assessed/reported using the GRADE approach, as outlined in the GRADE handbook. We will report the quality of evidence as assessed by the systematic review authors (who were in the best position to assess quality given their familiarity with the study-level data), such as by using GRADEPro ‘Summary of findings’ tables from the reviews if provided (or if necessary we will construct such tables using GRADEpro Guideline Development Tool)). The GRADE system assesses the following features for the evidence found for important outcomes:

  • Study limitations (risk of bias): internal validity of the evidence.

  • Inconsistency: heterogeneity or variability in the estimates of effect across studies.

  • Indirectness: degree of differences between population, intervention, comparator, for the intervention and outcome of interest.

  • Imprecision (random error): extent to which confidence in the effect estimate is adequate to support a particular decision.

  • Publication bias: degree of selective publication of studies.

The GRADE system rates the quality of the evidence as:

  • High (further research is very unlikely to change confidence in the estimate of effect).

  • Moderate (further research is likely to have an important impact on confidence in the estimate of effect and may change the estimate).

  • Low (further research is very likely to have an important impact on confidence in the estimate of effect and is likely to change the estimate).

  • Very low (any estimate of effect is very uncertain).

Data synthesis

A narrative description of the characteristics of the included Cochrane reviews will be undertaken. We will organise the findings by group of neonates, wherever possible as: interventions for neonates with perinatal asphyxia or with evidence of neonatal encephalopathy; interventions for neonates born preterm; interventions for neonates born low or very low birthweight; interventions for neonates born preterm and low or very low birthweight; interventions for other specific groups of 'at risk' neonates.

We will summarise the main results of the included reviews by categorising their findings in the following framework (as has been used within previous Cochrane and non-Cochrane overviews, such as Farquhar 2015 and Lassi 2015):

  • Effective interventions: indicating that the review found high-quality evidence of effectiveness for an intervention.

  • Promising interventions (more evidence needed): indicating that the review found moderate-quality evidence of effectiveness for an intervention, but more evidence is needed.

  • Ineffective interventions: indicating that the review found high-quality evidence of lack of effectiveness for an intervention.

  • Probably ineffective interventions (more evidence needed): indicating that the review found moderate-quality evidence suggesting lack of effectiveness for an intervention, but more evidence is needed.

  • No conclusions possible: indicating that the review found low- or very low-quality evidence, or insufficient evidence to comment on the effectiveness of an intervention.

The choice of category will be based on the quality of the evidence for the primary overview outcome (cerebral palsy). Separate assessments may be made for different comparisons (e.g. where one intervention is compared with both placebo (or no treatment) and to an alternative intervention). This approach to summarising the evidence was based on an earlier Cochrane overview (Jones 2012), which categorised interventions as "What works," "What may work", and "Insufficient evidence to make a judgement".

Acknowledgements

We thank the Cochrane Neonatal Editorial Base for their support. As part of the pre-publication editorial process, this protocol has been commented on by four editors.

Contributions of authors

Emily Shepherd drafted the first version of the protocol for this review, with Sarah McIntyre, Nadia Badawi, Maria Makrides, Philippa Middleton and Caroline Crowther making comments and contributing to the final draft.

Declarations of interest

None known.

Sources of support

Internal sources

  • ARCH: Australian Research Centre for Health of Women and Babies, Robinson Research Institute, The University of Adelaide, Australia.

External sources

  • National Health and Medical Research Council, Australia.

    Funding for the Pregnancy and Childbirth Australian and New Zealand Satellite

  • Cerebral Palsy Alliance Research Foundation, Australia.

    Project Grant: PG0914 – Interventions during the antenatal and neonatal period to prevent cerebral palsy: an overview of Cochrane systematic reviews (Shepherd E, Middleton P, Crowther CA)

  • Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, USA.

    Editorial support of the Cochrane Neonatal Review Group has been funded with Federal funds from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, USA, under Contract No. HHSN275201100016C

Ancillary