Indomethacin for symptomatic patent ductus arteriosus in preterm infants

Abstract

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

To determine the effectiveness and safety of indomethacin (given by any route) compared to placebo or no treatment in reducing mortality and morbidity in preterm infants with a symptomatic patent ductus arteriosus (PDA).

Background

Description of the condition

Fetal circulation relies on the placenta, as well as a patent ductus arteriosus (PDA) (Matthew 1998). The ductus arteriosus connects the main pulmonary artery to the proximal descending aorta, allowing the vast majority of right ventricular output to bypass the pulmonary circulation (Clyman 2000). Shortly after birth, with the initiation of breathing and separation of the low‐resistance placenta, the functional closure of the ductus arteriosus begins. Physiological mechanisms for closure include increased oxygen tension, and decreased circulating prostaglandin (PGE₂) and prostacyclin (PGI₂). This generally occurs within 24 to 72 hours of birth in the term infant (Clyman 2000). However, the ductus arteriosus frequently fails to close in the preterm infant, with an inverse relationship between gestational age and ductal patency (Clyman 2000; Hammerman 1995). Seventy percent of infants born before 28 weeks' gestational age have historically received medical or surgical closure of the PDA (Clyman 2000). Infants with respiratory distress syndrome (RDS) (Thibeault 1975), and perinatal asphyxia (Cotton 1981), are more likely to have a significant PDA present, as are infants who receive high volumes of fluid early in their life (Bell 1980). The likelihood of spontaneous closure of a PDA in infants less than or equal to 1500 grams remains high, reaching 85% in one cohort prior to discharge (Semberova 2017).

The failure of the ductus arteriosus to close in preterm infants is partially related to altered physiological mechanisms, including increased ductal sensitivity to the vasodilatory effects of prostaglandins and nitric oxide (Dice 2007). Additionally, there is decreased tone and muscle fibers in the ductus of preterm infants compared to the term infant (Hammerman 1995). RDS can further exacerbate these mechanisms, increasing the likelihood of ductal patency (Hammerman 1995).

Although several clinical findings may be present, none are entirely specific to a hemodynamically‐significant PDA. While often associated with a continuous, "machine‐like" murmur best heard in the left infraclavicular region, a murmur can be purely systolic or absent, particularly in the first few days of life (Ellison 1983). In cases of hemodynamically‐significant PDA with an absent murmur, PDA often presents with respiratory distress or cardiac enlargement (Hammerman 1986). Other common findings include a hyperdynamic precordial impulse, tachycardia, bounding pulses, widened pulse pressure, worsening respiratory status, or apnea (Ellison 1983). In the late 1970s, it was recognized that preterm infants without signs or symptoms of PDA had lower risk for RDS, bronchopulmonary dysplasia (BPD), and death when compared to similar infants with signs or symptoms of PDA (Brown 1979). Both surgical and pharmacological interventions were shown to be possible and potentially beneficial for preterm infants with a significant left to right ductal shunt (Cotton 1978; Friedman 1976; Heymann 1976; Kitterman 1972). While systolic blood pressure may be sustained despite a left to right ductal shunt, the drop in diastolic blood pressure and localized vasoconstriction is believed to contribute to the clinical consequences of PDA (Clyman 2000). The PDA has been associated with complications that include increased rate of BPD (Brown 1979), necrotizing enterocolitis (NEC) (Dollberg 2005), decreased middle cerebral artery blood flow (Weir 1999), intraventricular hemorrhage (IVH) (Ballabh 2010), worsening RDS (Jacob 1980), and death (Dice 2007). However, the precise causal link between these associations has not been demonstrated (Benitz 2010).

Description of the intervention

For hemodynamically‐significant PDAs that do not spontaneously close, a medical or surgical intervention may be chosen to achieve ductal closure. Procedural interventions include surgical ligation or transcatheter occlusion. Pharmacological agents include cyclooxygenase inhibitors, such as ibuprofen or indomethacin, and paracetamol, which is a derivative of acetanilide with weak anti‐inflammatory properties.

Current data are inconclusive regarding the comparative efficacy of surgery or medical management as the initial treatment of a PDA in a preterm infant (Malviya 2013). In clinical practice, surgical ligation or occlusion is frequently used following failure of medical management. Surgical ligation of the ductus is associated with reduced mortality, but surviving infants are at increased risk of neurodevelopmental impairment. However, there is a lack of studies addressing survival bias and confounding by indication (Weisz 2014).

Cyclooxygenase inhibitors (including indomethacin and ibuprofen) have been used to treat preterm infants with symptomatic PDA, following demonstration in both observation and randomized controlled trials (RCTs) that indomethacin increases the PDA closure rate (Friedman 1976; Gersony 1983; Heymann 1976). While numerous studies have indicated that prophylactic closure of the ductus with indomethacin reduces the incidence of severe IVH, hemodynamically significant PDA, and surgical ligation, complications of indomethacin therapy have been noted, including increased risk of bleeding, and transient renal insufficiency (Davis 1990; Fowlie 2010; Gersony 1983; Stavel 2017; Yeh 1981). Several studies report NEC as an adverse outcome of indomethacin treatment (Fujii 2002; Grosfeld 1996). However, this association was not found in a systemic systematic review of prophylactic intravenous indomethacin (Fowlie 2010). Systematic reviews have indicated ibuprofen may offer similar efficacy to indomethacin, with lower rates of transient renal insufficiency and NEC, though the optimal dosing, duration, and timing of both therapies are uncertain (Noori 2009; Ohlsson 2015).

Regarding cyclooxygenase inhibitors, several approaches have been taken for PDA treatment, including prophylactic medical therapy within the first 24 hours of life for at risk infants (Fowlie 2010; Ohlsson 2015), treatment following diagnosis of PDA, or symptomatic treatment, based on clinical and/or echocardiographic criteria to define hemodynamic significance (Clyman 1996).

This Cochrane Review will evaluate symptomatic treatment of PDA with indomethacin in the preterm infant. A previously published review has evaluated the role of ibuprofen for treatment of PDA (Ohlsson 2015).

How the intervention might work

Prostaglandins play a key role in maintaining ductal muscle relaxation (Clyman 1977). Indomethacin is a non‐selective cyclooxygenase inhibitor, which prevents the enzymatic process leading to the production of prostaglandins. Inhibition of prostaglandin production results in arteriolar vasoconstriction, which aids ductal closure. Prophylactic treatment aims to achieve ductal closure prior to development of significant left to right shunting and the clinical consequences of hemodynamic instability. PDA closure rates with prophylactic use of ibuprofen and indomethacin approach 58% and 57% respectively (Fowlie 2010; Ohlsson 2015). However, there are potential side effects to exposure to these drugs, and unclear benefits (Benitz 2010). Although there are associated benefits with indomethacin use, treatment with indomethacin may reduce perfusion to certain organs. This includes reduced cerebral blood flow, which may oppose neurodevelopmental outcomes associated with decreased IVH (Edwards 1990), decreased intestinal perfusion potentially increasing rate of NEC (Coombs 1990), and decreased renal perfusion (Cifuentes 1979). Indomethacin may also lead to bleeding via inhibition of platelet function (Friedman 1976). Rather than medically treating all preterm infants at risk of PDA or with trivial symptoms of PDA, targeting treatment to only those infants with hemodynamically significant PDA (by clinical and/or echocardiographic/Doppler criteria) may improve the risk/benefit ratio and limit the known adverse exposure to those with the greatest potential to benefit from therapy.

Why it is important to do this review

A Cochrane Review on the prophylactic use of indomethacin for medical treatment of PDA in the preterm infant examined studies of asymptomatic preterm infants with PDA and demonstrated only short‐term benefit, with no benefit or harm on long‐term neurodevelopment (Fowlie 2010). It is possible that there is an increased benefit relative to adverse consequences for treatment of symptomatic preterm infants with PDA when compared to prophylactic use. For this Cochrane Review, we will only include studies of indomethacin for preterm infants with a symptomatic PDA.

A Cochrane Review assessed the duration of indomethacin therapy for the treatment of PDA in preterm infants (Herrera 2007). A prolonged course of indomethacin did not seem to significantly impact clinical outcomes. For this Cochrane Review, we will not assess duration of therapy but will examine differences in dosing.

A non‐Cochrane review by Nehgme and colleagues assessed six small randomized trials of intravenous indomethacin for symptomatic PDA (Nehgme 1992). The meta‐analysis suggested that indomethacin significantly increased rates of PDA closure, without differences of mortality, IVH, NEC, or chronic lung disease (Nehgme 1992). In this Cochrane Review we will provide an updated examination of the literature.

Objectives

To determine the effectiveness and safety of indomethacin (given by any route) compared to placebo or no treatment in reducing mortality and morbidity in preterm infants with a symptomatic patent ductus arteriosus (PDA).

Methods

Criteria for considering studies for this review

Types of studies

We will include all published and unpublished RCTs, quasi‐RCTs, and cluster‐RCTs, and randomized crossover trials. Both superiority trials and non‐inferiority trials are eligible for inclusion.

Types of participants

Preterm infants born at less than 37 weeks' gestation or low birth weight infants (< 2500 grams) treated for symptomatic PDA enrolled within the first 28 days of life.

Symptomatic PDA is defined by clinical, echocardiographic, and Doppler criteria. Clinical criteria include a characteristic heart murmur, hyperdynamic precordial impulse, tachycardia, bounding pulses, widened pulse pressure, or worsening respiratory status (Davis 1995). Echocardiographic and Doppler criteria include a transductal diameter greater than 1.5 mm, left atrial aortic root ratio (LA:Ao) greater than 1.3, a left to right shunt, and “disturbed diastolic flow in the main pulmonary artery with diastolic backflow in the aorta immediately below the ductus arteriosus and forward flow above the ductal insertion” (Lago 2002).

We will exclude infants with other known congenital heart disease or infants with prior treatment with a cyclooxygenase inhibitor (indomethacin, ibuprofen).

Types of interventions

Indomethacin (any dose, any route) versus placebo or no treatment.

Types of outcome measures

Primary outcomes

1. PDA closure within one week of administration of the first dose of indomethacin

2. Bronchopulmonary dysplasia (defined as supplemental oxygen need at 28 days postnatal age (Bancalari 1979), or supplemental oxygen at 36 weeks' postmenstrual age with or without compatible clinical and radiographic findings (Shennan 1988))

3. All‐cause neonatal mortality at 28 days and prior to hospital discharge

Secondary outcomes

PDA‐related outcomes

Proportion of infants receiving rescue medical treatment (repeated cyclooxygenase or paracetamol/acetaminophen dosing, or both)

Proportion of infants requiring surgical ligation or transcatheter occlusion

Other outcomes

1. Pneumothorax

2. Pulmonary hemorrhage

3. Late onset‐sepsis

4. NEC (Bell stage 2 or greater) (Bell 1978)

5. IVH (any grade) (Papile 1978)

6. Severe IVH (grade III to IV)(Papile 1978)

7. Periventricular leukomalacia

8. Retinopathy of prematurity (any stage) (ICROP 2005)

9. Severe retinopathy of prematurity (stage III or greater)

10. Infant mortality (first year of life)

11. Use of inotropic agents

12. Duration of assisted ventilation (days)

13. Duration of oxygen dependence (days to last discontinuation of any supplemental oxygen)

14. Duration of hospital stay (days)

15. Time to full enteral feeds (days)

16. Neurodevelopmental outcome (Bayley Scales of Infant Development score at 18 to 28 months)

17. Cerebral palsy at approximately two years corrected age (as defined by the study authors)

18. Neurodevelopmental outcome at approximately two years corrected age (acceptable range 18 months to 28 months) including: cerebral palsy, delayed neurodevelopment (Bayley Scales of Infant Development Mental Developmental Index < 70), legal blindness (< 20/200 visual acuity), and hearing deficit (aided or < 60 dB on audiometric testing). The composite outcome 'neurodevelopmental impairment' was defined as having any one of the aforementioned deficits.

Safety outcomes (harms reported within one week of completing intervention)

1. Intestinal perforation

2. Renal function

   1. Oliguria (< 1 mL/kg/hour)

   2. Serum/plasma creatinine (µmol/L) levels during treatment

   3. Serum/plasma creatinine (µmol/L) after treatment

3. Hemostasis

   1. Mucocutaneous or gastrointestinal bleeding

   2. Platelet count (< 50,000 platelets/µL blood)

4. Pulmonary hypertension (diagnosed by echocardiographic or Doppler criteria)

Search methods for identification of studies

We will use the criteria and standard methods of Cochrane and Cochrane Neonatal (see the Cochrane Neonatal search strategy for specialized register). We will search for errata or retractions from included studies published in full‐text on PubMed (www.ncbi.nlm.nih.gov/pubmed) and report the date this was done.

Electronic searches

We will conduct a comprehensive search including: Cochrane Central Register of Controlled Trials (CENTRAL, current issue) in the Cochrane Library; MEDLINE via PubMed (1996 to current); Embase (1980 to current); and CINAHL (1982 to current) using the following search terms: (Indomethacin[MeSH] OR Indomethacin OR indomethacin OR indocid OR Indocin) AND (Ductus Arteriosus, Patent[MeSH] OR patent ductus arteriosus or PDA), plus database‐specific limiters for RCTs and neonates (see Appendix 1 for the full search strategies for each database). We will not apply language restrictions. We will search clinical trials registries for ongoing or recently completed trials, including ClinicalTrials.gov (clinicaltrials.gov), the World Health Organization International Clinical Trials Registry Platform (www.who.int/ictrp/search/en/), and the ISRCTN Registry (www.isrctn.com/).

Searching other resources

We will assess the reference lists of all identified articles for relevant articles not identified in the primary search.

Data collection and analysis

Selection of studies

Working in pairs, the review authors will independently screen the search results by title and abstract for studies that potentially meet the inclusion criteria. We will obtain the full‐text of any articles that are potentially eligible, and at least two review authors will independently perform full‐text assessments. We will resolve any disagreements by discussion between the review author team. We will record the selection process in sufficient detail to complete a PRISMA flow diagram (Moher 2009), and will list all studies excluded after full‐text assessment in a ‘Characteristics of excluded studies' table.

Data extraction and management

Two review authors (PE and SM) will extract, assess, and code all data for each study, using a form designed specifically for this review. We will replace any standard error of the mean by the corresponding standard deviation. We will resolve any disagreements by discussion. For each study, one review author (PE) will enter the extracted data into Review Manager 5 (RevMan 5) (RevMan 2014); a second review author (SM) will check data entry. If there is a conflict regarding interpretation of data, another author (RS) will adjudicate. All review authors will assess the protocol, analysis, and draft manuscript.

We will collect information regarding the method of randomization, blinding, drug intervention, stratification, and whether the trial was single or multicenter for each included study. We will note the information regarding trial participants including gestational age criteria, birth weight criteria, and other inclusion or exclusion criteria. We will analyze the information on clinical outcomes of the primary and secondary outcomes.

Assessment of risk of bias in included studies

Two review authors will independently assess the risk of bias (low, high, or unclear) of all included trials using the Cochrane ‘Risk of bias’ tool for the following domains (Higgins 2017).

1. Sequence generation (selection bias)

2. Allocation concealment (selection bias)

3. Blinding of participants and personnel (performance bias)

4. Blinding of outcome assessment (detection bias)

5. Incomplete outcome data (attrition bias)

6. Selective reporting (reporting bias)

7. Any other bias

We will resolve any disagreements by discussion or by consulting a third review author. See Appendix 2 for a more detailed description of risk of bias for each domain.

Measures of treatment effect

We will perform statistical analyses using RevMan 5 (RevMan 2014). We will analyze categorical data using the risk ratio (RR), and risk difference (RD). For statistically significant outcomes we will calculate the number needed to treat for an additional beneficial outcome, or number needed to treat for an additional harmful outcome. We will analyze continuous data using weighted mean difference (WMD) and the standardized mean difference (SMD). We will report the 95% confidence interval (CI) on all estimates.

Unit of analysis issues

The unit of analysis will be the participating infant in individually randomized trials. We will only consider an infant once in an analysis. We plan to exclude infants with multiple enrollments unless we obtain data from the report or investigators relating to the first episode of randomization. If we cannot separate data from the first randomization, we will exclude the study as we would not be able to address the unit of analysis issues that arise from multiple enrollments of the same infant.

We intend to conduct intention‐to‐treat analyses.

The participating neonatal unit or section of a neonatal unit will be the unit of analysis in cluster‐randomized trials. We plan to analyze these using an estimate of the intracluster correlation coefficient derived from the trial (if possible) or from another source as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2017).

If we identify both cluster‐RCTs and individually RCTs, we plan to only combine the results from both if there was little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomization unit is considered to be unlikely.

Dealing with missing data

We will attempt to obtain data from the primary author if published data provided inadequate information for the review.

Assessment of heterogeneity

We will estimate the treatment effects of individual trials and examine heterogeneity among trials by inspecting the Forest plots and quantifying the impact of heterogeneity using the I² statistic. We will assess the degree of heterogeneity as:

1. < 25% = no heterogeneity

2. 25% to 49% = low heterogeneity

3. 50% to 75% = moderate heterogeneity

4. > 75% = substantial heterogeneity

If we note statistical heterogeneity is present (I² statistic value of > 50%), we will explore the possible causes (for example, differences in study quality, participants, intervention regimens, or outcome assessments).

Assessment of reporting biases

If we suspect reporting bias, we will contact trial investigators to request missing outcome data. If this is not possible and we consider the missing data to have introduced serious bias, we plan to explore the impact of including such trials in the overall assessment of results in a sensitivity analysis.

Data synthesis

If we identify multiple studies that we consider to be sufficiently similar, we will perform meta‐analysis using RevMan 5 (RevMan 2014). For categorical outcomes we will calculate the typical estimates of RR and RD, each with its 95% CI; and for continuous outcomes we will determine the WMD or a summary estimate for the SMD, each with its 95% CI. We will use a fixed‐effect model for meta‐analysis. If we consider meta‐analysis to be inappropriate, we will analyse and interpret individual trials separately.

Quality of the evidence

We will use the GRADE approach, as outlined in the GRADE Handbook (Schünemann 2013), to assess the quality of the evidence for the following (clinically relevant) outcomes.

Efficacy

1. PDA closure within one week of administration of the first dose of indomethacin

2. Proportion of infants requiring surgical ligation or transcatheter occlusion

3. Bronchopulmonary dysplasia (defined as supplemental oxygen need at 28 days postnatal age (Bancalari 1979) or supplemental oxygen at 36 weeks' postmenstrual age with or without compatible clinical and radiographic findings (Shennan 1988))

4. All‐cause neonatal mortality at 28 days and prior to hospital discharge

5. NEC (Bell stage 2 or greater) (Bell 1978)

6. Neurodevelopmental outcome at approximately two years corrected age (acceptable range 18 months to 28 months) including: cerebral palsy, delayed neurodevelopment (Bayley Scales of Infant Development Mental Developmental Index < 70), legal blindness (< 20/200 visual acuity), and hearing deficit (aided or < 60 dB on audiometric testing). The composite outcome ‘neurodevelopmental impairment' was defined as having any one of the aforementioned deficits.

Safety outcomes (harms) (if significant)

1. Intestinal perforation

2. Oliguria (< 1 mL/kg/hour)

3. Increase in serum/plasma creatinine (µmol/L) levels after treatment

4. Mucocutaneous or gastrointestinal bleeding

5. Pulmonary hypertension (diagnosed by echocardiographic or Doppler criteria)

Two review authors will independently assess the quality of the evidence for each of the outcomes above. We will consider evidence from RCTs as high quality but will downgrade the quality of the evidence by one level for serious (or two levels for very serious) limitations based upon the following: design (risk of bias), consistency across studies, directness of the evidence, precision of estimates, and presence of publication bias. We will use the GRADEpro Guideline Development Tool to create a ‘Summary of findings’ table to report the quality of the evidence (GRADEpro GDT 2015).

The GRADE approach results in an assessment of the quality of a body of evidence in one of four grades.

1. High: we are very confident that the true effect lies close to that of the estimate of the effect.

2. Moderate: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

3. Low: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

4. Very low: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Subgroup analysis and investigation of heterogeneity

Subgroup analyses

• Gestational age at birth (< 28 weeks, 28 to 32 completed weeks, 33 to 36 completed weeks)

• Birth weight (< 1000 grams, 1000 to 1500 grams, 1501 to 2500 grams)

• Chronological age (< 3 days, 3 to 7 days; > 7 days)

• Route of administration (intravenous indomethacin, oral indomethacin)

• Higher versus lower dose indomethacin: low dose (≤ 0.4 mg/kg/dose) versus high (> 0.4 mg/kg/dose)

• Cumulative indomethacin dose: standard or lower cumulative dose (≤ 0.6 mg/kg) versus higher cumulative dose (> 0.6 mg/kg)

• Method of PDA diagnosis (clinical diagnosis, echocardiographic diagnosis, Doppler diagnosis)

• Trial methodology (low bias, high bias)

• Co‐interventions (antenatal steroids, post natal surfactant therapy)

Sensitivity analysis

We plan to conduct sensitivity analyses to determine whether or not findings were affected by including only studies of adequate methodology (low risk of bias), defined as adequate randomization and allocation concealment, blinding of intervention and measurement, and up to and including a 10% loss to follow‐up.

Appendices

Appendix 1. Cochrane Neonatal standard search strategy

PubMed

((infant, newborn[MeSH] OR newborn OR neonate OR neonatal OR premature OR low birth weight OR VLBW OR LBW or infan* or neonat*) AND (randomised controlled trial [pt] OR controlled clinical trial [pt] OR randomised [tiab] OR placebo [tiab] OR drug therapy [sh] OR randomly [tiab] OR trial [tiab] OR groups [tiab]) NOT (animals [mh] NOT humans [mh]))

Embase

((exp infant) OR (infan* OR newborn or neonat* OR premature or very low birth weight or low birth weight or VLBW or LBW).mp AND (human not animal) AND (randomised controlled trial or controlled clinical trial or randomised or placebo or clinical trials as topic or randomly or trial or clinical trial).mp

CINAHL

(infan* OR newborn OR neonat* OR premature OR low birth weight OR VLBW OR LBW) AND (randomised controlled trial OR controlled clinical trial OR randomised OR placebo OR clinical trials as topic OR randomly OR trial OR PT clinical trial)

CRS Web

(infan* or newborn or neonat* or premature or preterm or very low birth weight or low birth weight or VLBW or LBW)

Appendix 2. ‘Risk of bias' tool

We will use the standard methods of Cochrane and Cochrane Neonatal to assess the methodological quality of the included studies. For each included study, we will seek information regarding the method of randomization, blinding, and reporting of all outcomes of all infants enrolled in the study. We will assess each criterion as being at a ‘low', ‘high', or ‘unclear' risk of bias. Two review authors will separately assess each study. We will resolve any disagreement by discussion. We will add this information to the ‘Characteristics of included studies' table. We will evaluate the following issues and enter the findings into the ‘Risk of bias' table.

1. Sequence generation (checking for possible selection bias). Was the allocation sequence adequately generated?

For each included study, we will categorize the method used to generate the allocation sequence as:

• low risk (any truly random process e.g. random number table; computer random number generator);

• high risk (any non‐random process e.g. odd or even date of birth; hospital or clinic record number); or

• unclear risk.

2. Allocation concealment (checking for possible selection bias). Was allocation adequately concealed?

For each included study, we will categorize the method used to conceal the allocation sequence as:

• low risk (e.g. telephone or central randomization; consecutively numbered sealed opaque envelopes);

• high risk (open random allocation; unsealed or non‐opaque envelopes, alternation; date of birth); or

• unclear risk

3. Blinding of participants and personnel (checking for possible performance bias). Was knowledge of the allocated intervention adequately prevented during the study?

For each included study, we will categorize the methods used to blind study participants and personnel from knowledge of which intervention a participant received. We will assess blinding separately for different outcomes or class of outcomes. We will categorize the methods as:

• low risk, high risk, or unclear risk for participants; and

• low risk, high risk, or unclear risk for personnel.

4. Blinding of outcome assessment (checking for possible detection bias). Was knowledge of the allocated intervention adequately prevented at the time of outcome assessment?

For each included study, we will categorize the methods used to blind outcome assessment. Blinding will be assessed separately for different outcomes or class of outcomes. We will categorize the methods as:

• low risk for outcome assessors;

• high risk for outcome assessors; or

• unclear risk for outcome assessors.

5. Incomplete outcome data (checking for possible attrition bias through withdrawals, dropouts, protocol deviations). Were incomplete outcome data adequately addressed?

For each included study and for each outcome, we will describe the completeness of data including attrition and exclusions from the analysis. We will note whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomized participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information is reported or supplied by the trial authors, we will re‐include missing data in the analyses. We will categorize the methods as:

• low risk (< 20% missing data);

• high risk (≥ 20% missing data); or

• unclear risk.

6. Selective reporting bias. Are reports of the study free of suggestion of selective outcome reporting?

For each included study, we will describe how we investigated the possibility of selective outcome reporting bias and what we found. For studies in which study protocols were published in advance, we will compare prespecified outcomes versus outcomes eventually reported in the published results. If the study protocol was not published in advance, we will contact study authors to gain access to the study protocol. We will assess the methods as:

• low risk (where it is clear that all of the study's prespecified outcomes and all expected outcomes of interest to the review have been reported);

• high risk (where not all the study's prespecified outcomes have been reported; one or more reported primary outcomes were not prespecified outcomes of interest and are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported); or

• unclear risk.

7. Other sources of bias. Was the study apparently free of other problems that could put it at a high risk of bias?

For each included study, we will describe any important concerns we had about other possible sources of bias (for example, whether there was a potential source of bias related to the specific study design or whether the trial was stopped early due to some data‐dependent process). We will assess whether each study was free of other problems that could put it at risk of bias as:

• low risk;

• high risk;

• unclear risk.

If needed, we plan to explore the impact of the level of bias through undertaking sensitivity analyses.

Contributions of authors

PE and RS conceived the project and drafted the protocol. PE, DO, JNF, SM and RS reviewed all drafts and approved the final version of the protocol.

Sources of support

Internal sources

• No sources of support supplied

External sources

• Vermont Oxford Network, USA.

  Cochrane Neonatal Reviews are produced with support from Vermont Oxford Network, a worldwide collaboration of health professionals dedicated to providing evidence‐based care of the highest quality for newborn infants and their families.

Declarations of interest

The authors declare no competing financial interests.

New