Surfactant therapy via thin catheter in preterm infants with or at risk of respiratory distress syndrome

Abstract

Background

Non‐invasive respiratory support is increasingly used for the management of respiratory dysfunction in preterm infants. This approach runs the risk of under‐treating those with respiratory distress syndrome (RDS), for whom surfactant administration is of paramount importance. Several techniques of minimally invasive surfactant therapy have been described. This review focuses on surfactant administration to spontaneously breathing infants via a thin catheter briefly inserted into the trachea.

Objectives

Primary objectives

In non‐intubated preterm infants with established RDS or at risk of developing RDS to compare surfactant administration via thin catheter with:

1. intubation and surfactant administration through an endotracheal tube (ETT); or

2. continuation of non‐invasive respiratory support without surfactant administration or intubation.

Secondary objective

1. To compare different methods of surfactant administration via thin catheter

Planned subgroup analyses included gestational age, timing of intervention, and use of sedating pre‐medication during the intervention.

Search methods

We used the standard search strategy of Cochrane Neonatal to search the Cochrane Central Register of Controlled Trials (CENTRAL), in the Cochrane Library; Ovid MEDLINE(R) and Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Daily and Versions(R); and the Cumulative Index to Nursing and Allied Health Literature (CINAHL), on 30 September 2020. We also searched clinical trials databases and the reference lists of retrieved articles for randomised controlled trials (RCTs) and quasi‐randomised trials.

Selection criteria

We included randomised trials comparing surfactant administration via thin catheter (S‐TC) with (1) surfactant administration through an ETT (S‐ETT), or (2) continuation of non‐invasive respiratory support without surfactant administration or intubation. We also included trials comparing different methods/strategies of surfactant administration via thin catheter. We included preterm infants (at < 37 weeks' gestation) with or at risk of RDS.

Data collection and analysis

Review authors independently assessed study quality and risk of bias and extracted data. Authors of all studies were contacted regarding study design and/or missing or unpublished data. We used the GRADE approach to assess the certainty of evidence.

Main results

We included 16 studies (18 publications; 2164 neonates) in this review. These studies compared surfactant administration via thin catheter with surfactant administration through an ETT with early extubation (Intubate, Surfactant, Extubate technique ‐ InSurE) (12 studies) or with delayed extubation (2 studies), or with continuation of continuous positive airway pressure (CPAP) and rescue surfactant administration at pre‐specified criteria (1 study), or compared different strategies of surfactant administration via thin catheter (1 study). Two trials reported neurosensory outcomes of of surviving participants at two years of age. Eight studies were of moderate certainty with low risk of bias, and eight studies were of lower certainty with unclear risk of bias.

S‐TC versus S‐ETT in preterm infants with or at risk of RDS

Meta‐analyses of 14 studies in which S‐TC was compared with S‐ETT as a control demonstrated a significant decrease in risk of the composite outcome of death or bronchopulmonary dysplasia (BPD) at 36 weeks' postmenstrual age (risk ratio (RR) 0.59, 95% confidence interval (CI) 0.48 to 0.73; risk difference (RD) ‐0.11, 95% CI ‐0.15 to ‐0.07; number needed to treat for an additional beneficial outcome (NNTB) 9, 95% CI 7 to 16; 10 studies; 1324 infants; moderate‐certainty evidence); the need for intubation within 72 hours (RR 0.63, 95% CI 0.54 to 0.74; RD ‐0.14, 95% CI ‐0.18 to ‐0.09; NNTB 8, 95% CI; 6 to 12; 12 studies, 1422 infants; moderate‐certainty evidence); severe intraventricular haemorrhage (RR 0.63, 95% CI 0.42 to 0.96; RD ‐0.04, 95% CI ‐0.08 to ‐0.00; NNTB 22, 95% CI 12 to 193; 5 studies, 857 infants; low‐certainty evidence); death during first hospitalisation (RR 0.63, 95% CI 0.47 to 0.84; RD ‐0.02, 95% CI ‐0.10 to 0.06; NNTB 20, 95% CI 12 to 58; 11 studies, 1424 infants; low‐certainty evidence); and BPD among survivors (RR 0.57, 95% CI 0.45 to 0.74; RD ‐0.08, 95% CI ‐0.11 to ‐0.04; NNTB 13, 95% CI 9 to 24; 11 studies, 1567 infants; moderate‐certainty evidence). There was no significant difference in risk of air leak requiring drainage (RR 0.58, 95% CI 0.33 to 1.02; RD ‐0.03, 95% CI ‐0.05 to 0.00; 6 studies, 1036 infants; low‐certainty evidence). None of the studies reported on the outcome of death or survival with neurosensory disability.

Only one trial compared surfactant delivery via thin catheter with continuation of CPAP, and one trial compared different strategies of surfactant delivery via thin catheter, precluding meta‐analysis.

Authors' conclusions

Administration of surfactant via thin catheter compared with administration via an ETT is associated with reduced risk of death or BPD, less intubation in the first 72 hours, and reduced incidence of major complications and in‐hospital mortality. This procedure had a similar rate of adverse effects as surfactant administration through an ETT. Data suggest that treatment with surfactant via thin catheter may be preferable to surfactant therapy by ETT. Further well‐designed studies of adequate size and power, as well as ongoing studies, will help confirm and refine these findings, clarify whether surfactant therapy via thin tracheal catheter provides benefits over continuation of non‐invasive respiratory support without surfactant, address uncertainties within important subgroups, and clarify the role of sedation.

Plain language summary

Surfactant therapy via thin catheter in preterm infants with or at risk of respiratory distress syndrome

Review question

Is giving surfactant via a minimally invasive technique involving placement of a thin catheter in the trachea of a spontaneously breathing infant effective and safe?

Background

Respiratory distress syndrome (RDS) is an important cause of disease and death in preterm infants. It is commonly treated with a medication called surfactant, which is given by a tube (called an endotracheal tube, or ETT). The ETT is placed in the windpipe (trachea). However, more infants with RDS are now being treated from the onset with non‐invasive respiratory support (through a mask) without use of an ETT. This means that the usual means of administering surfactant is not available. In such infants, surfactant therapy requires placement of an ETT, with or without the intent to remove it soon after the procedure. Surfactant improves clinical outcomes, but insertion of the ETT and mechanical ventilation (assisted breathing) can cause lung injury. This can contribute to development of a chronic lung disease known as bronchopulmonary dysplasia (BPD) and other problems. Alternatives to ETT insertion have been developed. The most popular method is the use of a thin catheter (tube) that is briefly inserted into the windpipe.

Study characteristics

We searched the electronic databases and found 16 randomised trials (18 publications) that met our selection criteria. These trials involved delivery of surfactant via a thin catheter. Evidence is up‐to‐date as of 30 September 2020.

Key results

Surfactant delivery via a thin catheter to spontaneously breathing preterm infants compared with surfactant administration through an ETT was associated with a decrease in the following: risk of death or BPD, need for assisted breathing in the first 72 hours of life, severe brain bleeding, death during first hospitalisation, and BPD among survivors. We are uncertain as to whether the intervention has an important effect on air leak requiring drainage because the results are imprecise. None of the studies reported on the outcome of death or survival with disability. The procedure had rates of adverse effects similar to surfactant administration through an ETT. These data suggest that treatment with surfactant via a thin catheter is preferable to surfactant therapy through an ETT. Further well‐designed studies of adequate size and power, as well as ongoing studies, are required to confirm and refine these findings, and to clarify whether surfactant therapy via a thin catheter provides benefits over continuation of non‐invasive respiratory support without surfactant.

Certainty of evidence

Most of the studies had important methodological weaknesses. We used the GRADE approach to assess the certainty of evidence. We downgraded the evidence to 'moderate to low'. More good quality studies are urgently needed to address uncertainties within important subgroups.

Summary of findings

Summary of findings 1. Surfactant administration via thin catheter (S‐TC) vs surfactant administration through an endotracheal tube (S‐ETT) in preterm infants with or at risk of respiratory distress syndrome.

Surfactant administration via thin catheter (S‐TC) vs surfactant administration through an endotracheal tube (S‐ETT) in preterm infants with or at risk of respiratory distress syndrome
Patient or population: preterm infants with or at risk of respiratory distress syndrome Setting: neonatal intensive care units. Countries: Germany, Turkey, Canada, China, India, Iran, and Pakistan Intervention: surfactant administration through thin catheter (S‐TC) Comparison: surfactant administration through endotracheal tube (S‐ETT)
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	№. of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with surfactant administration through endotracheal tube (S‐ETT)	Risk with surfactant administration through thin catheter (S‐TC)
Death or bronchopulmonary dysplasia (BPD) at 36 weeks' postmenstrual age	Study population		RR 0.59 (0.48 to 0.73)	1324 (10 RCTs)	⊕⊕⊕⊝ MODERATEa
	26 per 100	16 per 100 (13 to 19)
Need for intubation within the first 72 hours	Study population		RR 0.63 (0.54 to 0.74)	1422 (12 RCTs)	⊕⊕⊕⊝ MODERATEa
	36 per 100	23 per 100 (20 to 27)
Air leak requiring drainage	Study population		RR 0.58 (0.33 to 1.02)	1036 (6 RCTs)	⊕⊕⊝⊝ LOWa,b
	6 per 100	3 per 100 (2 to 6)
Severe intraventricular haemorrhage (grade III or IV)	Study population		RR 0.63 (0.42 to 0.96)	857 (5 RCTs)	⊕⊕⊝⊝ LOWa,b
	12 per 100	7 per 100 (5 to 11)
Death during first hospitalisation (all causes)	Study population		RR 0.63 (0.47 to 0.84)	1424 (11 RCTs)	⊕⊕⊝⊝ LOWa,b
	13 per 100	8 per 100 (6 to 11)
Bronchopulmonary dysplasia (BPD) among survivors at 36 weeks' postmenstrual age	Study population		RR 0.57 (0.45 to 0.74)	1567 (11 RCTs)	⊕⊕⊕⊝ MODERATEa
	18 per 100	10 per 100 (8 to 13)
Death or survival with neurosensory disability ‐ not reported	‐	‐	‐	‐	‐	None of the studies reported this outcome
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: 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. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

^aDowngraded by one level for serious study limitations (high risk of bias due to uncertainty about methods used to generate random sequence, conceal allocation, and mask outcome assessments) in many trials.

^bDowngraded by one level for serious imprecision of effect estimate (inadequate optimal effect size and/or 95% CI around estimate consistent with substantial harm or benefit).

Background

Description of the condition

Respiratory distress syndrome (RDS) and its complications are major contributors to morbidity and mortality in preterm infants. Recognition that surfactant deficiency is an important cause of RDS (as reported in Avery 1959) ultimately led to the development of surfactant replacement therapy for RDS (Jobe 1993). Administration of exogenous surfactant is known to reduce mortality and risk of air leak, and has become a mainstay of therapy for preterm infants with RDS (Suresh 2005; Sweet 2016).

In recent years, non‐invasive respiratory support has become more popular for the management of respiratory dysfunction in preterm infants (Berger 2013; Soll 2013). Three large randomised controlled trials (RCTs) found that applying nasal continuous positive airway pressure (CPAP) from birth is at least as effective as intubation and ventilation among infants at < 30 weeks' gestation (Dunn 2011; Finer 2010; Morley 2008). Applying CPAP from the outset in an unselected population of preterm infants does, however, run the risk of under‐treating those with RDS, for whom CPAP may fail to provide adequate respiratory support. Absence of an endotracheal tube means the usual conduit for exogenous surfactant administration is unavailable. The risks and consequences of CPAP failure under these circumstances are now being appreciated (Ammari 2005; Dargaville 2013b; Dargaville 2016). Such infants, once intubated, receive surfactant at a later than ideal time and have increased risk of adverse outcomes compared to like‐gestation infants managed by CPAP alone (Ammari 2005; Dargaville 2013b; Dargaville 2016).

One approach to resolving the CPAP‐surfactant dilemma has been to briefly insert an endotracheal tube (ETT) to administer surfactant to infants on CPAP, followed by rapid extubation back to CPAP (InSurE; Intubate, Surfactant, Extubate procedure) (Stevens 2007; Verder 1994; Victorin 1990). This technique provides benefits over continuation of CPAP, but most often it requires sedating pre‐medication, and extubation may be delayed due to respiratory suppression.

Numerous investigators have sought an alternative solution to the problem of administering surfactant to infants on non‐invasive respiratory support. Several techniques of minimally invasive surfactant therapy have been described and are the topic of other Cochrane Reviews, including surfactant administration by aerosolisation (Abdel‐Latif 2012), by pharyngeal deposition (Abdel‐Latif 2011a), and through a laryngeal mask (Abdel‐Latif 2011b). The topic of this review, which has not previously been systematically reviewed, is administration of surfactant via a thin catheter briefly inserted into the trachea (Dargaville 2011; Kribs 2007).

Description of the intervention

Surfactant administration via thin catheter (S‐TC) encompasses any method in which a thin catheter, expected to be narrower than a standard endotracheal tube (ETT), is passed through the vocal cords to allow surfactant instillation. The most commonly used methods are:

1. flexible thin catheter and Magill's forceps (Cologne method), as described by Kribs and colleagues (Kribs 2007);

2. flexible thin feeding tube without Magill's forceps (take care method), as described by Kanmaz and colleagues (Kanmaz 2013);

3. semi‐rigid thin catheter (Hobart method), as described by Dargaville and colleagues (Dargaville 2011); and

4. modifications of the above methods.

Variation may be encountered in (1) the pre‐medication used, (2) the means of laryngoscopy used, including videolaryngoscopy, (3) the type of catheter, (4) the method used to guide the catheter through the vocal cords, (5) the approach to surfactant delivery (bolus versus infusion, rapid versus slow), (6) the surfactant preparation, (7) the surfactant dose, and (8) the approach to respiratory management before, during, and after the technique, including the type of non‐invasive respiratory support used. It is expected that infants are spontaneously breathing, and therefore positive‐pressure inflations are not required for surfactant dispersal. Unlike an ETT, a thin catheter is unsuitable for delivery of positive‐pressure inflations.

Several different acronyms may be used for the above methods, including:

1. MIST (minimally invasive surfactant therapy);

2. LISA (less invasive surfactant administration);

3. SurE (surfactant without endotracheal tube);

4. MISA (minimally invasive surfactant administration); and

5. NISA (non‐invasive surfactant administration).

For this review, we elected to not use any of these in preference to others, instead using a term capturing the essence of the method: surfactant administration via thin catheter (S‐TC).

How the intervention might work

For infants with RDS managed by non‐invasive respiratory support, administering surfactant directly into the trachea using a minimally invasive approach has the potential to overcome surfactant deficiency and replenish the endogenous surfactant pool. Progressive respiratory deterioration culminating in CPAP failure may thus be avoided, and along with it, the known associated adverse outcomes. Non‐randomised studies have demonstrated that surfactant administration via tracheal catheterisation is feasible (Kribs 2007; Kribs 2008; Kribs 2009; Kribs 2010; Dargaville 2011; Dargaville 2013a), and it appears to be safe (Aguar 2014; Porth 2011), and that a reduction in the need for subsequent ventilation or supplemental oxygen, or both, may be achievable. These short‐term clinical benefits have the potential to lead to improvement in longer‐term outcomes.

Why it is important to do this review

Surfactant administration via thin catheter is a promising, feasible therapy that is being adopted in many sites around the world (Bhayat 2020; Heiring 2017; Klotz 2017; Jeffreys 2019; Roberts 2020). Therefore it is important to determine whether this treatment is safe and effective. This technique has not been the topic of a previous Cochrane Review.

Objectives

Primary objectives

In non‐intubated preterm infants with established RDS or at risk of developing RDS to compare surfactant administration via thin catheter with:

1. intubation and surfactant administration through an endotracheal tube (ETT); or

2. continuation of non‐invasive respiratory support without surfactant administration.

Secondary objective

1. To compare different methods of surfactant administration via thin catheter

Planned subgroup analyses included gestational age, timing of intervention, and use of sedating pre‐medication during the intervention.

Methods

Criteria for considering studies for this review

Types of studies

We included parallel interventional trials, randomised or quasi‐randomised, regardless of the unit of allocation (individual or cluster).

Types of participants

We included preterm infants (< 37 weeks' gestation) with or at risk of RDS.

Types of interventions

We included the following methods of surfactant administration via thin catheter.

1. A flexible catheter and Magill's forceps (Kribs 2007).

2. A flexible catheter without Magill's forceps (Kanmaz 2013).

3. A semi‐rigid catheter without Magill's forceps (Dargaville 2011).

4. Variations or modifications of the above methods, including use of videolaryngoscopy for catheter placement.

We included studies that compared different tracheal catheterisation techniques (e.g. semi‐rigid versus flexible catheter, sedation versus no sedation). We included trials using any surfactant formulation, including animal‐derived and synthetic surfactants (with or without surfactant protein activity).

Types of comparisons

In accordance with the objectives of this review, we categorised trials by the form of intervention used in the comparator (control) group, as below. Given the fundamental difference in the three therapeutic approaches for control infants (see later), we analysed data from trials within these categories separately and did not pool data together in a meta‐analysis.

The three comparisons are discussed below.

Comparison of surfactant administration via thin catheter (S‐TC) with surfactant administration via ETT (S‐ETT)

In this category, infants in the comparison (control) group were intubated and received surfactant by ETT. We further divided these trials into two groups.

S‐TC versus surfactant administration via ETT with the intent to rapidly extubate (InSurE)

In these trials, for controls, there was the intent to extubate soon after surfactant delivery, as in Haberman 2002 (i.e. the INtubate‐SURfactant‐Extubate (InSurE) procedure; Reininger 2005; Victorin 1990).

S‐TC versus surfactant administration via ETT with delayed extubation

In these trials, control infants remained intubated after surfactant delivery, with delayed extubation after a period of mechanical ventilation.

Comparison of S‐TC with continuation of non‐invasive respiratory support

In this category, management in the comparison (control) group consisted of continuation of non‐invasive respiratory support (CPAP, high‐flow (HF), variations thereof) without surfactant administration, unless pre‐specified failure criteria were met.

Comparison of different methods of surfactant delivery via thin catheter

In this category, a thin catheter was used for surfactant delivery to all participants, with comparison of different methods, including different approaches to the use of sedation.

Types of outcome measures

Primary outcomes

The following were recognised as critical outcomes for this review.

1. Death or bronchopulmonary dysplasia (BPD): the composite outcome of death or BPD, defined as the need for oxygen or respiratory support at 36 weeks' postmenstrual age (PMA) (Shennan 1988).

2. Need for intubation within the first 72 hours of life.

3. Air leak requiring drainage (during first hospitalisation).

4. Severe intraventricular haemorrhage (IVH), including grades III and IV (Papile 1978).

5. Death during first hospitalisation (all causes).

6. BPD (clinical definition) among survivors to 36 weeks' PMA.

7. Death or survival with neurosensory disability, with the latter measured beyond one year PMA and defined as any of (1) cerebral palsy by clinical examination or other means; (2) developmental delay more than two standard deviations below the population mean on standardised testing; (3) blindness (visual acuity < 6/60); or (4) deafness (hearing impairment requiring amplification).

Secondary outcomes

Measures of safety of the surfactant administration procedure

1. Catheter/ETT placement unsuccessful at first attempt (during trial‐related intervention)

2. Bradycardia (heart rate < 100 beats per minute (bpm)) during the intervention

3. Hypoxaemia (oxygen saturation < 80%) during the intervention

And for studies comparing thin catheter methods

1. Need for positive‐pressure ventilation during the intervention

2. Need for immediate intubation (within 15 minutes of the intervention)

Metrics of respiratory support

1. Need for intubation within the first 72 hours, or not intubated but reached failure criteria

2. Need for intubation at any time

3. Need for intratracheal surfactant therapy post intervention

4. Duration of mechanical ventilation via ETT (days; among survivors)

5. Duration of any respiratory support (mechanical ventilation, CPAP, heart failure (HF)) (days; among survivors)

6. Duration of oxygen therapy (days; among survivors)

7. Postnatal systemic corticosteroid therapy for BPD mitigation

Outcomes during first hospitalisation

1. BPD (physiological definition), evaluated when necessary by a room‐air challenge at 36 weeks' PMA for infants with borderline oxygen requirements (Walsh 2004)

2. IVH, any grade (Papile 1978)

3. Cystic periventricular leukomalacia (PVL)

4. Patent ductus arteriosus (PDA) requiring medical therapy

5. Necrotising enterocolitis (NEC): modified Bell stage 2 or greater (Bell 1978; Walsh 1988)

6. Spontaneous intestinal perforation

7. Retinopathy of prematurity (ROP), stage 3 or greater

8. Duration of hospitalisation (days; among survivors)

Postdischarge outcomes

1. Oxygen therapy at home

2. Number of hospital re‐admissions with respiratory illness in the first two years

3. Parent‐reported wheeze in the first two years

4. Bronchodilator use in the first two years

5. Neurosensory disability (defined per primary outcome above), among survivors

Search methods for identification of studies

We used the criteria and standard methods of Cochrane and Cochrane Neonatal (see the Cochrane Neonatal search strategy for specialised register).

Electronic searches

We conducted a comprehensive search on 30 September 2020. This search included the Cochrane Central Register of Controlled Trials (CENTRAL; 2020, Issue 9), in the Cochrane Library; Ovid MEDLINE(R) and Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, Daily and Versions(R) (1946 to 30 September 2020); and the Cumulative Index to Nursing and Allied Health Literature (CINAHL; 1946 to September Week 2 2020).

We have included the search strategies for each database in Appendix 1. We did not apply language or date restrictions.

We searched clinical trial registries for ongoing or recently completed trials (ISRCTN Registry). We searched the World Health Organization’s International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en/) and the US National Library of Medicine’s ClinicalTrials.gov (clinicaltrials.gov) via Cochrane CENTRAL.

Searching other resources

We communicated with expert informants and searched bibliographies of reviews and trials for references to other trials. We also searched previous reviews including cross‐references, abstracts, and conference and symposia proceedings (as above) from 1990 to 30 September 2020. For unpublished trials, we contacted the contact investigator to request information. We considered unpublished studies and studies reported only as abstracts as eligible for review only if final trial data were reported (i.e. data from an interim analysis were not included). We contacted the corresponding authors of identified trials for additional information when needed. We searched clinical trial registries for ongoing and recently completed trials (as above). We searched the reference lists of any articles selected for inclusion in this review to identify additional relevant articles.

Data collection and analysis

We used the standard methods of Cochrane and Cochrane Neonatal. Two review authors independently conducted searches, assessed study eligibility, and extracted study results and risk of bias. We resolved discrepancies by discussion and consensus.

Selection of studies

Two review authors (MEA and PAD) independently reviewed the titles and abstracts of potentially relevant studies against inclusion and exclusion criteria. We (MEA and PAD) independently assessed the titles and abstracts of studies identified by the search strategy for eligibility for inclusion in this review. We obtained full‐text versions of studies for closer examination of eligibility, or when inadequate information was provided in the abstract.

Data extraction and management

We (MEA and PAD) independently extracted data from full‐text articles using a specifically designed spreadsheet to manage the information. We (MEA and PAD) resolved discrepancies through discussion and consensus, or, if required, by consultation with a third review author (PGD). We (MEA and PAD) entered data into Review Manager software (Review Manager 2020), and we checked it for accuracy.

Assessment of risk of bias in included studies

Two review authors (MEA and PAD) independently assessed the risk of bias (low, high, or unclear) of all included trials using the Cochrane ‘Risk of bias’ tool for the following domains (Higgins 2011).

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 resolved any disagreements by discussion or by consultation with a third assessor. See Appendix 2 for a more detailed description of risk of bias for each domain.

Measures of treatment effect

We analysed the results of included studies using the statistical package Review Manager 5 software (Review Manager 2020). We used the standard methods of Cochrane Neonatal. We used a fixed‐effect model for meta‐analysis. In assessing treatment effects for dichotomous data or categorical data, we reported the risk ratio (RR) or the risk difference (RD), respectively, along with the 95% confidence interval (CI). If the RD was statistically significant, we calculated the number needed to treat for an additional beneficial outcome (NNTB) and the number needed to treat for an additional harmful outcome (NNTH) (1/RD). For outcomes measured on a continuous scale, we reported the mean difference (MD), along with the 95% CI.

Unit of analysis issues

We combined cluster‐randomised and individually randomised trials in a single meta‐analysis using the generic inverse variance method.

Dealing with missing data

In the case of missing data, we described the number of participants with missing data in the Results section and in the Characteristics of included studies table. When possible, we performed an intention‐to‐treat (ITT) meta‐analysis using reconstructed denominators. We discussed the implications of data missing from the review as appropriate.

Assessment of heterogeneity

We used Review Manager 5 to assess the heterogeneity of treatment effects between trials (Review Manager 2020). We used two formal statistical approaches to assess the presence of statistical heterogeneity.

1. The Chi² test for homogeneity: because this test has low power when the number of studies included in the meta‐analysis is small, we set the level of significance at 10% probability (P < 0.1) (Higgins 2019).

2. The I² statistic: the I² statistic describes the percentage of total variation across studies due to heterogeneity rather than to sampling error, and is thus a measure of the validity of data pooling for meta‐analysis. We graded the degree of heterogeneity as follows: ≤ 24%, no heterogeneity; 25% to 49%, low heterogeneity; 50% to 74%, moderate heterogeneity; and ≥ 75%, high heterogeneity.

When we noted evidence of apparent or statistical heterogeneity, we assessed the source of heterogeneity by using sensitivity and subgroup analyses to look for evidence of bias or methodological differences between trials.

Assessment of reporting biases

We made attempts to obtain the study protocols of all included studies and to compare outcomes reported in the protocol versus those reported in the findings for each of the included studies. If reporting bias was suspected (see Assessment of reporting biases), we made attempts to contact the study authors to ask them to provide further information. When this was not possible, and when missing data were thought to introduce serious bias, we examined the impact of including/excluding such studies in the overall assessment of results by performing a sensitivity analysis.

We investigated non‐reporting (including publication) bias by visually assessing funnel plot asymmetry, and by using Egger's test in meta‐analyses if data from at least 10 trials contributing events were available (Egger 1997).

Data synthesis

We performed meta‐analyses using the standard methods of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019). We used a fixed‐effect model. When studies were statistically heterogenous, we examined study characteristics including study design and quality. When appropriate, we performed sensitivity analysis including only trials with higher methodological rigour.

We did not pool trials that included different comparison groups (see Types of interventions).

Subgroup analysis and investigation of heterogeneity

We identified several factors that could influence the safety and efficacy of interventions examined in this review and therefore planned sub‐group analyses based on:

1. gestational age (≤ 28 weeks (extremely preterm), 29 to 32 weeks (very preterm), 33 to 36 weeks (preterm));

2. timing of surfactant administration (i.e. prophylaxis versus rescue). Here we defined prophylaxis trials as those in which surfactant treatment was administered soon after birth to infants at risk of RDS, and rescue trials as those that used treatment with surfactant selectively in infants demonstrating features of RDS; and

3. use of sedation and analgesia pre‐medication in the tracheal catheterisation group (i.e. sedation and analgesia used versus withheld).

Sensitivity analysis

We explored methodological heterogeneity through the use of sensitivity analysis. We assessed studies as having low risk of bias if sequence generation and allocation concealment were adequate, and if losses were less than 10% with ITT analysis.

Summary of findings and assessment of the certainty of the evidence

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

1. Death or bronchopulmonary dysplasia (BPD) at 36 weeks' PMA.

2. Need for intubation within the first 72 hours of life.

3. Air leak requiring drainage.

4. Severe IVH (grade III or IV).

5. Death during first hospitalisation (all causes).

6. BPD among survivors to 36 weeks' PMA.

7. Death or survival with neurosensory disability.

Two review authors (MEA and PAD) independently assessed the certainty of evidence for each of the outcomes above. We considered evidence from RCTs as high certainty but downgraded the evidence by one level for serious (or by 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 used the GRADEpro GDT Guideline Development Tool to create Table 1 to report the certainty of evidence.

The GRADE approach results in an assessment of the certainty of a body of evidence as belonging to one of four grades.

1. High certainty: further research is very unlikely to change our confidence in the estimate of effect.

2. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

3. Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.

4. Very low certainty: we are very uncertain about the estimate.

Results

Description of studies

See the Characteristics of included studies and Characteristics of excluded studies tables.

Results of the search

We identified 6133 records from the initial search of PubMed, CENTRAL, MEDLINE, and CINAHL. We performed additional searches of clinicaltrials.gov and other registries and identified 202 further records that appeared to be relevant. Additional searches of reference lists and other Internet resources yielded 14 additional relevant articles. After we removed duplicates, there were 4849 records. Among these, 28 articles remained relevant after inspection of titles or abstract, or both. We evaluated the abstracts or full‐texts of articles and excluded 12 records. The diagram of the flow of studies from the initial search to the meta‐analysis is shown in Figure 1. A description of all included studies is displayed under Characteristics of included studies, and excluded studies with reasons for exclusion are given in the Characteristics of excluded studies table.

1.

Study flow diagram.

We included in our meta‐analysis 16 primary studies (18 publications including 16 primary studies and two reports of neurosensory outcomes in two primary trials among surviving participants at two years of age). Studies included in this review include those that examined effects of administration of surfactant via thin catheter on clinical outcomes among infants with or at risk of RDS. We divided these studies into groups based on treatment strategy (Figure 2).

2.

Primary and follow‐up studies included in the review categorised by comparison group.

Trials comparing S‐TC with S‐ETT

S‐TC versus InSurE

We identified 12 studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Halim 2019; Han 2020; Jena 2019; Kanmaz 2013; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Yang 2020).

S‐TC versus surfactant administration via ETT with delayed extubation

We identified two studies (Kribs 2015; Olivier 2017), along with a further report in Mehler 2020 detailing two‐year neurosensory outcomes for infants recruited in the NINSAPP trial (Kribs 2015).

Trials comparing S‐TC with continuation of non‐invasive respiratory support

We identified a single study (AMV trial; Göpel 2011). In this study, the comparison group (control) continued on CPAP respiratory support without surfactant administration unless certain failure criteria were met. Herting 2020 is a report of two‐year neurosensory outcomes for infants recruited in the AMV trial.

Trials comparing different methods or strategies of surfactant delivery via thin catheter

We identified a single study (Dekker 2019). This study compared two methods of performing MIST: one with sedation and one without sedation during the MIST procedure.

Included studies

Of the 16 included studies (18 publications), seven were multi‐centre (Göpel 2011; Han 2020; Jena 2019; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Olivier 2017), and nine involved a single centre (Bao 2015; Boskabadi 2019; Choupani 2018; Dekker 2019; Gupta 2020; Halim 2019; Kanmaz 2013; Mosayebi 2017; Yang 2020). Herting 2020 and Mehler 2020 are reports of two‐year neurosensory follow‐up of infants recruited in Göpel 2011 and Kribs 2015, respectively.

In total, we recruited 2164 preterm infants. The number of infants included in each trial and their gestational age are provided in Table 2. The description of all studies is summarised under Characteristics of included studies.

1. Number of infants recruited and gestational age ranges for included trials.

Trial	Multicentre study	Country	Total number of infants recruited	Gestational age (weeks)			Age at surfactant administration	Sedation
				Eligiblity criteria	Intervention group	Control group
Bao 2015	No	China	90	28 to 32	29.1 ± 1.5	29.3 ± 1.6	Within 2 hours of birth	None
Boskabadi 2019	No	Iran	40	< 32	29.1 ± 2.6	28.2 ± 2.1	Not specified	Not specified
Choupani 2018	No	Iran	104	28 to 37	32.9 ± 2.6	33.1 ± 2.3	Within 1 hour of birth	Not specified
Dekker 2019	No	Netherlands	78	26 to 37	29 + 0 (27 + 5 to 32 + 0)	29 + 0 (28 + 0 to 31 + 0)	Within first 24 hours of life	Propofol
Gupta 2020	No	India	58	28 to 34	30.07 ± 1.51	29.90 ± 1.67	Within 6 hours of birth	None
Göpel 2011 (follow‐up reported in Herting 2020)	Yes (n = 12)	Germany	220	26 to 28	27.6 ± 0.8	27.5 ± 0.8	Within 12 hours of birth	Sedation and analgesia were used at the discretion of attending neonatologist
Halim 2019	No	Pakistan	100	≤ 34	32 to 34 weeks = 26 (52%) 30 to 31 + 6 weeks = 11 (22%) 28 to 29 + 6 weeks = 8 (16%) < 28 weeks = 5 (10%)	32 to 34 weeks 24 (48%) 30 to 31 + 6 weeks = 14 (28.6%) 28 to 29 + 6 weeks = 6 (12.2%) < 28 weeks = 5 (10.2%)	Within 12 hours of birth	None
Han 2020	Yes (n = 8)	China	344	25 + 0 to 31 + 6	30.6 ± 1.6	30.8 ± 1.3	Within 6 hours of birth	None
Jena 2019	Yes (n = 3)	India	350	≤ 34	31.0 (29.0 to 33.0)	31.0 (29.0 to 33.0)	Within 6 hours of birth	None
Kanmaz 2013	No	Turkey	200	< 32	28 ± 2	28.3 ± 2	Not specified	None
Kribs 2015 (follow‐up reported in Mehler 2020)	Yes (n = 13)	Germany	211	23 to 26	25.3 ± 1.1	25.2 ± 0.91	10 to 120 minutes of age	None
Mirnia 2013a	Yes (n = 3)	Iran	136	27 to 32	29.6 ± 1.7	29.6 ± 1.7	Not specified	None
Mohammadizadeh 2015	Yes (n = 2)	Iran	38	≤ 34	30 ± 2	31 ± 2	Within 1 hour of birth	Not specified
Mosayebi 2017	No	Iran	53	28 to 34	32.6 ± 1.1	31.9 ± 1.5	Not specified	None
Olivier 2017	Yes (n = 3)	Canada	45	32 to 36	34 ± 1.4	33.8 ± 1.5	Within first 24 hours of life	Fentanyl
Yang 2020	No	China	97	32 + 0 to 36 + 6	33.7 ± 1.0	34.1 ± 1.3	Within 12 hours of birth	Not specified
TOTAL			2164

Data reported as mean ± SD; median (interquartile range) or number (%).

Excluded studies

We excluded two randomised studies (Characteristics of excluded studies).

1. One single‐centre study ‐ Mirnia 2013b ‐ that was reported as part of another included multi‐centre randomised trial (Mirnia 2013a).

2. One randomised study comparing different ventilation strategies within the minimally invasive surfactant therapy approach (Oncel 2016).

Ongoing studies

We identified 11 ongoing studies (see Characteristics of ongoing studies).

1. Still recruiting (ChiCTR1900020970; NCT04016246; NCT04445571).

2. Finished recruiting but not analysed yet (ACTRN12611000916943).

3. Recruitment not yet started (ACTRN12611000917932; NCT01848262; NCT04073173).

4. Terminated or suspended (NCT01615016; NCT02772081).

5. Non‐randomised observational studies (NCT03989960; UMIN000021785).

Risk of bias in included studies

The risk of bias for studies included in this review based on the review authors' judgements is summarised in Figure 3 and Figure 4 and is discussed below.

3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

4.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Randomisation was performed and was reported adequately in 10 studies (Bao 2015; Choupani 2018; Dekker 2019; Göpel 2011; Gupta 2020; Halim 2019; Jena 2019; Kanmaz 2013; Kribs 2015; Olivier 2017).

We considered randomisation as high risk in Choupani 2018, as it involved initial convenience sampling followed by random allocation. Five studies did not report adequately on the method of randomisation (Boskabadi 2019; Han 2020; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017). One study used a quasi‐randomised method (alternate allocation) (Yang 2020).

Allocation concealment was adequately reported and was appropriately performed in 11 studies (Bao 2015; Dekker 2019; Göpel 2011; Gupta 2020; Han 2020; Jena 2019; Kanmaz 2013, Kribs 2015; Mohammadizadeh 2015; Olivier 2017; Yang 2020). Five studies did not report on the method of allocation concealment (Boskabadi 2019; Choupani 2018; Halim 2019; Mirnia 2013a; Mosayebi 2017).

Blinding

Blinding of participants and personnel (performance bias) was achieved in one study (Yang 2020), and blinding of outcome assessment (to prevent detection bias) was achieved in two studies (Dekker 2019; Yang 2020). In all other studies, blinding was not performed or was unclear.

Incomplete outcome data

There was complete follow‐up of all enrolled participants with minimal risk of attrition bias, with the exception of two studies that did not report the outcome of 11% to 13% of randomised infants (Dekker 2019; Han 2020).

Selective reporting

There was no reporting bias in four studies (Göpel 2011; Gupta 2020; Kanmaz 2013; Kribs 2015). We obtained the study protocol from the authors of five studies (Bao 2015; Han 2020; Kanmaz 2013; Mosayebi 2017; Olivier 2017).

The study protocol was not available for 10 studies (Boskabadi 2019; Choupani 2018; Halim 2019; Jena 2019; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Yang 2020; and Herting 2020 and Mehler 2020, which were reports of two‐year neurosensory follow‐up of infants recruited in the Göpel 2011 and Kribs 2015 trials, respectively).

Ten studies were not registered with an international trial registry (Boskabadi 2019; Choupani 2018; Halim 2019; Jena 2019; Mirnia 2013a; Mohammadizadeh 2015; Olivier 2017; Yang 2020; and Herting 2020 and Mehler 2020, which were reports of two‐year neurosensory follow‐up of infants recruited in the Göpel 2011 and Kribs 2015 trials, respectively). Four studies were registered with an international trials registry either retrospectively or after the start of patient recruitment (Bao 2015; Han 2020; Kanmaz 2013; Mosayebi 2017).

Other potential sources of bias

Seven trials were multi‐centre trials with no treatment standardisation between centres (e.g. caffeine use, surfactant dose, type of catheter used), leading to variability between centres (Göpel 2011; Han 2020; Jena 2019; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Olivier 2017). However, multi‐variate logistical regression analysis was implemented in some of these studies and showed no significant centre effect.

Nine trials were single‐centre studies (Bao 2015; Boskabadi 2019; Choupani 2018; Dekker 2019; Gupta 2020; Halim 2019; Kanmaz 2013; Mosayebi 2017; Yang 2020). In one trial, some infants who might have been eligible could not be enrolled because of concern for standardisation of the intervention (Take Care) (Kanmaz 2013). Seven studies were not reported according to CONSORT guidelines, hence it is difficult to judge their quality (e.g. randomisation, allocation concealment, blinding) (Boskabadi 2019; Choupani 2018; Han 2020; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Yang 2020).

Effects of interventions

See: Table 1

See Table 1 for trials comparing S‐TC with S‐ETT in preterm infants with or at risk of RDS; see Data and analyses.

We have reported information in this section under three trial categories with different comparator groups.

1. A. Trials comparing S‐TC with S‐ETT.

   1. S‐TC versus InSurE.

   2. S‐TC versus surfactant via ETT with delayed extubation.

   3. S‐TC versus S‐ETT (all studies).

2. B. Trials comparing S‐TC with continuation of non‐invasive respiratory support.

3. C. Trials comparing different methods or strategies of thin catheter surfactant delivery.

We reported the analyses relevant to each comparison under the following three subtitles.

1. Overall analysis (primary and secondary outcomes).

2. Sub‐group analyses.

3. Sensitivity analysis.

A. Trials comparing S‐TC with S‐ETT

1. S‐TC compared with S‐ETT ‐ overall analysis

Primary outcomes

1.1 Death or BPD

See Analysis 1.1; Figure 5; Figure 6.

1.1. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 1: Death or BPD

5.

Forest plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.1 Death or BPD.

6.

Funnel plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.1 Death or BPD.

S‐TC versus InSurE

This outcome was reported by nine studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Jena 2019; Kanmaz 2013; Mirnia 2013a; Mohammadizadeh 2015; Yang 2020). One study showed a significant difference in the risk of this outcome (Jena 2019). The meta‐analysis of treatment trials showed significant differences in the risk of this outcome between S‐TC and InSurE (typical risk ratio (RR) 0.52, 95% confidence interval (CI) 0.40 to 0.68; typical risk difference (RD) ‐0.11, 95% CI ‐0.16 to ‐0.07; number needed to treat for an additional beneficial outcome (NNTB) 9, 95% CI 6 to 15; 9 studies, 1113 infants). Heterogeneity among the studies was low (I² = 2%).

S‐TC versus surfactant via ETT with delayed extubation

This outcome was reported by one study (Kribs 2015). This study showed no significant differences in the risk of this outcome (RR 0.79, 95% CI 0.55 to 1.13; RD ‐0.09, 95% CI ‐0.22 to 0.04; 1 study, 211 infants).

S‐TC versus S‐ETT (all studies)

This outcome was reported by 10 studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Yang 2020). The meta‐analysis of trials showed a significant decrease in the risk of this outcome with S‐TC compared to S‐ETT (typical RR 0.59, 95% CI 0.48 to 0.73; typical RD ‐0.11, 95% CI ‐0.15 to ‐0.07; NNTB 9, 95% CI 7 to 16); 10 studies, 1324 infants). Heterogeneity among the studies was low (I² = 19%). There was no statistically significant evidence of funnel plot asymmetry consistent with trials favouring controls missing from the meta‐analysis (Egger test for bias, P = 0.467).

1.2 Need for intubation within the first 72 hours

See Analysis 1.2; Figure 7; Figure 8.

1.2. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 2: Need for intubation within the first 72 hours

7.

Forest plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.2 Need for intubation within the first 72 hours.

8.

Funnel plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.2 Need for intubation within the first 72 hours.

S‐TC versus InSurE

This outcome was reported by 10 studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Jena 2019; Kanmaz 2013; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Yang 2020). Two studies showed a significant difference in the risk of this outcome (Jena 2019; Kanmaz 2013). The meta‐analysis of treatment trials showed a significant decrease in the risk of this outcome with S‐TC compared to InSurE (typical RR 0.61, 95% CI 0.50 to 0.75; typical RD ‐0.12, 95% CI ‐0.17 to ‐0.07; NNTB 8, 95% CI 6 to 14; 10 studies, 1166 infants). There was no heterogeneity among these studies (I² = 0%).

S‐TC versus surfactant via ETT with delayed extubation

This outcome was reported by two studies (Kribs 2015; Olivier 2017). These studies showed a significant difference in the risk of this outcome. The meta‐analysis of treatment trials showed a significant decrease in the risk of this outcome with S‐TC compared to surfactant via ETT with delayed extubation (typical RR 0.68, 95% CI 0.0.53 to 0.86; typical RD ‐0.21, 95% CI ‐0.32 to ‐0.09; NNTB 5, 95% CI 3 to 12; 2 studies, 256 infants). Heterogeneity among the studies was high (I² = 85%).

S‐TC versus S‐ETT (all studies)

This outcome was reported by 12 studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Olivier 2017; Yang 2020). The meta‐analysis of trials showed a significant decrease in the risk of this outcome with S‐TC compared to S‐ETT (typical RR 0.63, 95% CI 0.54 to 0.74; typical RD ‐0.14, 95% CI ‐0.18 to ‐0.09; NNTB 8, 95% CI; 6 to 12; 12 studies, 1422 infants). Heterogeneity among the studies was low (I² = 31%). There was no statistically significant evidence of funnel plot asymmetry consistent with trials favouring controls missing from the meta‐analysis (Egger test for bias, P = 0.322).

1.3 Air leak requiring drainage (during first hospitalisation)

See Analysis 1.3; Figure 9.

1.3. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 3: Air leak requiring drainage

9.

Forest plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.3 Air leak requiring drainage.

S‐TC versus InSurE

This outcome was reported by four studies (Jena 2019; Kanmaz 2013; Mirnia 2013a; Yang 2020). None of the individual studies showed significant differences in the risk of this outcome. The meta‐analysis of trials showed no significant differences in the risk of this outcome between S‐TC and InSurE (typical RR 0.72, 95% CI 0.35 to 1.48; typical RD ‐0.01, 95% CI ‐0.04 to 0.01; 4 studies, 783 infants). There was no heterogeneity among the studies (I² = 0%).

S‐TC versus surfactant via ETT with delayed extubation

This outcome was reported by two studies (Kribs 2015; Olivier 2017). These two studies showed no significant differences in the risk of this outcome. The meta‐analysis of treatment trials showed no significant differences in the risk of this outcome between S‐TC and surfactant via ETT with delayed extubation (typical RR 0.41, 95% CI 0.16 to 1.05; typical RD ‐0.07, 95% CI CI‐0.13 to 0.00; 2 studies, 253 infants). There was no heterogeneity among the studies (I² = 0%).

S‐TC versus S‐ETT (all studies)

This outcome was reported by six studies (Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Olivier 2017; Yang 2020). The meta‐analysis of trials showed no significant differences in the risk of this outcome between S‐TC and S‐ETT (typical RR 0.58, 95% CI 0.33 to 1.02; typical RD ‐0.03, 95% CI ‐0.05 to 0.00; 6 studies, 1036 infants). There was no heterogeneity among the studies (I² = 0%).

1.4 Severe intraventricular haemorrhage (IVH), including grades III and IV

See Analysis 1.4; Figure 10.

1.4. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 4: Severe IVH

10.

Forest plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.4 Severe IVH.

S‐TC versus InSurE

This outcome was reported by four studies (Bao 2015; Gupta 2020; Han 2020; Kanmaz 2013). None of the individual studies showed a significant difference in the risk of this outcome. The meta‐analysis of treatment trials showed no significant differences in the risk of this outcome between S‐TC and InSurE (typical RR 0.77, 95% CI 0.45 to 1.32; typical RD ‐0.02, 95% CI ‐0.06 to 0.02; 4 studies, 646 infants). There was no heterogeneity among the studies (I² = 0%).

S‐TC versus surfactant via ETT with delayed extubation

This outcome was reported by one study (Kribs 2015). This study showed a significant difference in the risk of this outcome (RR 0.46, 95% CI 0.24 to 0.90; RD ‐0.12, 95% CI ‐0.22 to ‐0.02; NNTB 8, 95% CI 5 to 49; 1 study, 211 infants).

S‐TC versus S‐ETT (all studies)

This outcome was reported by five studies (Bao 2015; Gupta 2020; Han 2020; Kanmaz 2013; Kribs 2015). The meta‐analysis of trials showed a significant decrease in the risk of this outcome with S‐TC compared to S‐ETT (typical RR 0.63, 95% CI 0.42 to 0.96; typical RD ‐0.04, 95% CI ‐0.08 to ‐0.00; NNTB 22, 95% CI 12 to 193; 5 studies, 857 infants). There was no heterogeneity among the studies (I² = 0%).

1.5 Death during first hospitalisation (all causes)

See Analysis 1.5; Figure 11; Figure 12.

1.5. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 5: Death during first hospitalisation

11.

Forest plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.5 Death during first hospitalisation.

12.

Funnel plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.5 Death during first hospitalisation.

S‐TC versus InSurE

This outcome was reported by nine studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Jena 2019; Kanmaz 2013; Mirnia 2013a; Mohammadizadeh 2015; Yang 2020). One study showed a significant difference in the risk of this outcome (Mirnia 2013a). The meta‐analysis of treatment trials showed significant differences in the risk of this outcome between S‐TC and InSurE (typical RR 0.60, 95% CI 0.44 to 0.82; typical RD ‐0.05, 95% CI ‐0.09 to ‐0.02; NNTB 19, 95% CI 11 to 52; 9 studies, 1213 infants). Heterogeneity among the studies was low (I² = 0%).

S‐TC versus surfactant via ETT with delayed extubation

This outcome was reported by one study (Kribs 2015). This study showed no significant difference in the risk of this outcome (RR 0.81, 95% CI 0.37, 1.79; RD ‐0.02, 95% CI ‐0.10, 0.06; 1 study, 211 infants).

S‐TC versus S‐ETT (all studies)

This outcome was reported by 10 studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Yang 2020). The meta‐analysis of trials showed a significant difference in the risk of this outcome between S‐TC and S‐ETT (typical RR 0.63, 95% CI 0.47 to 0.84; typical RD ‐0.02, 95% CI ‐0.10 to 0.06; NNTB 20, 95% CI 12 to 58; 10 studies, 1424 infants). There was no heterogeneity among the studies (I² = 0%). There was no statistically significant evidence of funnel plot asymmetry consistent with trials favouring controls missing from the meta‐analysis (Egger test for bias, P = 0.217).

1.6 BPD (clinical definition) among survivors to 36 weeks' PMA

See Analysis 1.6; Figure 13; Figure 14.

1.6. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 6: BPD (clinical definition); in survivors to 36 weeks' PMA

13.

Forest plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.6 BPD (clinical definition); in survivors to 36 weeks' PMA.

14.

Funnel plot of comparison: 1 Trials comparing S‐TC with S‐ETT ‐ overall analysis, outcome: 1.6 BPD (clinical definition); in survivors to 36 weeks' PMA.

S‐TC versus InSurE

This outcome was reported by 10 studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Han 2020; Jena 2019; Kanmaz 2013; Mirnia 2013a; Mohammadizadeh 2015; Yang 2020). One study showed a significant difference in the risk of this outcome (Jena 2019). The meta‐analysis of treatment trials showed a significant difference in the risk of this outcome between S‐TC and InSurE (typical RR 0.57, 95% CI 0.44 to 0.75; typical RD ‐0.07, 95% CI ‐0.11 to ‐0.04; NNTB 14, 95% CI 9 to 28; 10 studies, 1378 infants). Heterogeneity among the studies was low (I² = 15%).

S‐TC versus surfactant via ETT with delayed extubation

This outcome was reported by one study (Kribs 2015). This study showed no significant difference in the risk of this outcome (RR 0.58, 95% CI 0.32 to 1.05; RD ‐0.11, 95% CI ‐0.22 to 0.01; 1 study, 189 infants).

S‐TC versus S‐ETT (all studies)

This outcome was reported by 11 studies (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Han 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Yang 2020). The meta‐analysis of trials showed a significant difference in the risk of this outcome between S‐TC and S‐ETT (typical RR 0.57, 95% CI 0.45 to 0.74; typical RD ‐0.08, 95% CI ‐0.11 to ‐0.04; NNTB 13, 95% CI 9 to 24; 11 studies, 1567 infants). There was no heterogeneity among the studies (I² = 0%). There was no statistically significant evidence of funnel plot asymmetry consistent with trials favouring controls missing from the meta‐analysis (Egger test for bias, P = 0.373).

1.7 Death or survival with neurosensory disability (beyond one year)

None of the studies reported on this outcome.

Secondary outcomes

Different studies reported different sets of secondary outcomes (Analysis 1.7; Analysis 1.8; Analysis 1.9; Analysis 1.10; Analysis 1.11; Analysis 1.12; Analysis 1.13; Analysis 1.14; Analysis 1.15; Analysis 1.16; Analysis 1.17; Analysis 1.18; Analysis 1.19; Analysis 1.20; Analysis 1.21; Analysis 1.22; Analysis 1.23; Analysis 1.24).

1.7. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 7: Catheter/ETT placement unsuccessful at first attempt

1.8. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 8: Bradycardia (heart rate < 100 bpm) during the intervention

1.9. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 9: Hypoxaemia (oxygen saturation < 80%) during the intervention

1.10. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 10: Need for intubation within the first 72 hours or not intubated but reached failure criteria

1.11. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 11: Need for intubation at any time

1.12. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 12: Need for intratracheal surfactant therapy post intervention

1.13. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 13: Duration of mechanical ventilation (days; in survivors)

1.14. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 14: Duration of any respiratory support (days; in survivors)

1.15. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 15: Duration of oxygen therapy (days; in survivors)

1.16. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 16: Postnatal systemic corticosteroid therapy for BPD mitigation

1.17. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 17: BPD (physiological definition)

1.18. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 18: IVH, any grade

1.19. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 19: Cystic PVL

1.20. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 20: PDA requiring medical therapy

1.21. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 21: NEC, modified Bell stage ≥2

1.22. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 22: ROP stage ≥ 3

1.23. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 23: Duration of hospitalisation (days; in survivors)

1.24. Analysis.

Comparison 1: Trials comparing S‐TC with S‐ETT ‐ overall analysis, Outcome 24: Discharged home with oxygen

S‐TC versus InSurE

Meta‐analyses showed statistically significant differences in favour of S‐TC in the following outcomes.

1. Need for intubation within the first 72 hours, or not intubated but reached failure criteria (typical RR 0.72, 95% CI 0.53 to 0.96; typical RD ‐0.09, 95% CI ‐0.17 to ‐0.01, I² = 0%; NNTB 11. 95% CI 6 to 102; 4 studies, 464 infants; Analysis 1.10).

2. Need for intubation at any time (typical RR 0.70, 95% CI 0.54 to 0.90; typical RD ‐0.15, 95% CI ‐0.25 to ‐0.05, I² = 29%; NNTB 7, 95% CI 4 to 23; 3 studies, 338 infants; Analysis 1.11).

3. NEC: modified Bell stage ≥ 2 (typical RR 0.34, 95% CI 0.14 to 0.81; typical RD ‐0.04, 95% CI ‐0.07 to ‐0.01, I² = 50%; NNTB 26, 95% CI 15 to 99; 686 infants, 3 studies; Analysis 1.21).

S‐TC versus surfactant via ETT with delayed extubation

Meta‐analyses showed statistically significant differences in favour of S‐TC in the following outcomes.

1. Need for intubation within the first 72 hours, or not intubated but reached failure criteria (typical RR 0.68, 95% CI 0.53 to 0.86; typical RD ‐0.21, 95% CI ‐0.32 to ‐0.09, I² = 85%; NNTB 5, 95% CI 3 to 12; 2 studies, 135 infants; Analysis 1.10).

2. Need for intubation at any time (typical RR 0.75, 95% CI 0.68 to 0.84; typical RD ‐0.24, 95% CI ‐0.33 to ‐0.16, I² non‐applicable; NNTB 5, 95% CI 3 to 6; 183 infants, 1 study; Analysis 1.11).

S‐TC versus S‐ETT (all studies)

Meta‐analyses showed statistically significant differences in favour of S‐TC in the following outcomes.

1. Need for intubation within the first 72 hours, or not intubated but reached failure criteria (typical RR 0.70, 95% CI 0.58 to 0.84; typical RD ‐0.13, 95% CI ‐0.20 to ‐0.07, I² = 29%; NNTB 8, 95% CI 5 to 17; 6 studies, 720 infants; Analysis 1.10).

2. Need for intubation at any time (typical RR 0.73, 95% CI 0.64 to 0.83; typical RD ‐0.18, 95% CI‐0.25 to ‐0.11, I² = 8%; NNTB 6, 95% CI 4 to 10; 4 studies, 549 infants; Analysis 1.11).

3. NEC: modified Bell stage ≥ 2 (typical RR 0.34, 95% CI 0.14 to 0.81; typical RD ‐0.04, 95% CI ‐0.07 to ‐0.01, I² = 50%; NNTB 26, 95% CI 15 to 99; 3 studies, 686 infants; Analysis 1.21).

2. S‐TC compared with S‐ETT ‐ subgroup analyses

See Analysis 2.1.

2.1. Analysis.

Comparison 2: Trials comparing S‐TC with S‐ETT ‐ sub‐group analyses, Outcome 1: Death or BPD

2.1 Gestational age (≤ 28 weeks (extremely preterm), 29 to 32 weeks (very preterm), 33 to 36 weeks (preterm))

Table 2 shows the gestation range of infants recruited in each trial. Only one study exclusively recruited infants at ≤ 28 weeks' gestation (Kribs 2015). One trial recruited infants at < 32 weeks and provided stratified analysis for extremely preterm infants for some outcomes (Kanmaz 2013). Two trials included only infants at > 32 weeks (Olivier 2017; Yang 2020). Eleven trials included infants at ≤ 28 weeks but did not provide stratified analysis based on gestation (Bao 2015; Boskabadi 2019; Choupani 2018; Dekker 2019; Gupta 2020; Halim 2019; Han 2020; Jena 2019; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017). We did not perform meta‐analyses, as subgroup data were not available.

2.2 Trials examining surfactant administration in a prophylactic or rescue context

Table 2 shows the time of surfactant administration in each trial. None of the studies provided stratification based on prophylactic (intervention performed within 15 minutes of birth in infants) versus rescue (intervention performed beyond 15 minutes) surfactant administration. We did not perform meta‐analyses, as subgroup data were not available.

2.3 Trials with and without the use of sedating pre‐medication in the tracheal catheterisation group

Table 2 shows the use of sedation, if any, in each trial. In nine studies, sedating pre‐medication was not used before the intervention (Bao 2015; Gupta 2020; Halim 2019; Han 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mosayebi 2017). Only Olivier 2017 used sedating pre‐medication (fentanyl 1 mcg/kg) before S‐TC. Use of pre‐medication was not clear in four studies (Boskabadi 2019; Choupani 2018; Mohammadizadeh 2015; Yang 2020).

Analyses showed significant subgroup effects in favour of S‐TC for death or BPD in trials where no sedation was used in the S‐TC group (typical RR 0.58, 95% CI 0.46 to 0.73; typical RD ‐0.12, 95% CI ‐0.17 to ‐0.07, I² = 50%; NNTB 8, 95% CI 6 to 14; 6 studies, 1045 infants; Analysis 2.1). We did not perform meta‐analyses for the sub‐group of trials in which sedation was mandatory in the S‐TC group, as only one trial was of this type (Olivier 2017).

3. S‐TC compared with S‐ETT ‐ sensitivity analysis

See Analysis 3.1Analysis 3.2Analysis 3.3Analysis 3.4Analysis 3.5 and Analysis 3.6.

3.1. Analysis.

Comparison 3: Trials comparing S‐TC with S‐ETT ‐ sensitivity analysis, Outcome 1: Death or BPD

3.2. Analysis.

Comparison 3: Trials comparing S‐TC with S‐ETT ‐ sensitivity analysis, Outcome 2: Need for intubation within the first 72 hours

3.3. Analysis.

Comparison 3: Trials comparing S‐TC with S‐ETT ‐ sensitivity analysis, Outcome 3: Air leak requiring drainage

3.4. Analysis.

Comparison 3: Trials comparing S‐TC with S‐ETT ‐ sensitivity analysis, Outcome 4: Severe IVH

3.5. Analysis.

Comparison 3: Trials comparing S‐TC with S‐ETT ‐ sensitivity analysis, Outcome 5: Death during first hospitalisation

3.6. Analysis.

Comparison 3: Trials comparing S‐TC with S‐ETT ‐ sensitivity analysis, Outcome 6: BPD (clinical definition); in survivors to 36 weeks' PMA

We performed sensitivity analysis by excluding eight studies of low or unknown quality based on lack of any of the following: adequate randomisation, allocation concealment, less than 10% loss to follow‐up with ITT analysis (Boskabadi 2019; Choupani 2018; Halim 2019; Han 2020; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Yang 2020; see Figure 4). We identified eight trials of moderate quality (Bao 2015; Dekker 2019; Gupta 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Olivier 2017; and Herting 2020, which reported two‐year neurosensory follow‐up of infants recruited in Kribs 2015). The results of this analysis are shown in Analysis 3.1, Analysis 3.2, Analysis 3.3, Analysis 3.4, Analysis 3.5, and Analysis 3.6.

Meta‐analyses showed statistically significant differences in favour of S‐TC in the following outcomes.

1. Death or BPD (typical RR 0.60, 95% CI 0.46 to 0.78; typical RD ‐0.11, 95% CI ‐0.17 to ‐0.06, I² = 62%; NNTB 9, 95% CI 6 to 18; 5 studies, 799 infants; Analysis 3.1).

2. Need for intubation within the first 72 hours (typical RR 0.60, 95% CI 0.51 to 0.72; typical RD ‐0.18, 95% CI ‐0.23 to ‐0.12, I² = 51%; NNTB 6, 95% CI 4 to 9; 6 studies, 954 infants; Analysis 3.2).

3. Severe IVH (typical RR 0.55, 95% CI 0.34 to 0.89; typical RD ‐0.07, 95% CI ‐0.12 to ‐0.01, I² = 0%; NNTB 15, 95% CI 8 to 64; 3 studies, 559 infants; Analysis 3.4).

4. BPD (clinical definition) in survivors to 36 weeks' PMA (typical RR 0.45, 95% CI 0.31 to 0.64; typical RD ‐0.11, 95% CI ‐0.16 to ‐0.06, I² = 34%; NNTB 9, 95% CI 6 to 16; 5 studies, 858 infants; Analysis 3.6).

B. Trials comparing S‐TC with continuation of non‐invasive respiratory support (CPAP)

One study compared S‐TC with continuation of CPAP: the Avoid Mechanical Ventilation (AMV) study (Göpel 2011). Two‐year outcome data for the AMV trial were published in Herting 2020. Given that there was only one study, meta‐analysis was not applicable.

Findings from the AMV trial are summarised in Analysis 4.1Analysis 4.2, Analysis 4.3, Analysis 4.4, Analysis 4.5, Analysis 4.6, Analysis 4.7, Analysis 4.8, Analysis 4.9, Analysis 4.10, Analysis 4.11, Analysis 4.12, and Analysis 4.13.

4.1. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 1: Death or BPD

4.2. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 2: Incidence of air leak requiring drainage

4.3. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 3: Severe IVH

4.4. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 4: Death during first hospitalisation

4.5. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 5: BPD (clinical definition); in survivors to 36 weeks' PMA

4.6. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 6: Catheter/ETT placement unsuccessful at first attempt

4.7. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 7: Bradycardia (heart rate < 100 bpm) during the intervention

4.8. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 8: Need for intubation within the first 72 hours or not intubated but reached failure criteria

4.9. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 9: Need for intubation at any time

4.10. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 10: Postnatal systemic corticosteroid therapy for BPD mitigation

4.11. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 11: Cystic PVL

4.12. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 12: ROP ≥ stage 3

4.13. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 13: Discharged home with oxygen

We were unable to undertake planned subgroup and sensitivity analyses to determine whether findings are affected by including only studies using adequate methods (low risk of bias), as there was only one study (Göpel 2011), and its two‐year neurosensory follow‐up data are reported in Herting 2020.

C. Trials comparing different methods or strategies of thin catheter surfactant delivery

One study of sedation during MIST (PROMISES) reported a comparison of surfactant via thin catheter with and without sedating pre‐medication (propofol) (Dekker 2019). Given that there was only one study, meta‐analysis was not applicable.

Findings from the above study are summarised in Analysis 5.1, Analysis 5.2, Analysis 5.3, Analysis 5.4, Analysis 5.5, Analysis 5.6, Analysis 5.7, Analysis 5.8, and Analysis 5.9.

5.1. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 1: Air leak

5.2. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 2: Severe IVH

5.3. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 3: Need for intubation during the procedure

5.4. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 4: Need for intubation within the first 24 hours

5.5. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 5: Death during first hospitalisation

5.6. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 6: Need for positive‐pressure ventilation during the intervention

5.7. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 7: Duration of the procedure (seconds)

5.8. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 8: Pain score using a validated instrument for measuring discomfort/pain during the procedure (e.g. COMFORTneo score)

5.9. Analysis.

Comparison 5: Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis, Outcome 9: Hypotension requiring treatment

We were unable to undertake planned subgroup and sensitivity analyses to determine whether findings are affected by including only studies using adequate methods (low risk of bias), as there was only one study (Dekker 2019).

Discussion

Summary of main results

Sixteen primary studies ‐ Bao 2015; Boskabadi 2019; Choupani 2018; Dekker 2019; Göpel 2011; Gupta 2020; Halim 2019; Han 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Olivier 2017; Yang 2020 ‐ and two neurosensory follow‐up reports of infants recruited in two of the primary studies ‐ Herting 2020 (primary study: Göpel 2011) and Mehler 2020 (primary study: Kribs 2015), including 2164 patients, met the inclusion criteria for this review (i.e. total 18 publications).

Evidence from 10 studies including 1324 infants and contributing data to the primary outcomes of this review shows that surfactant therapy via thin catheter (S‐TC) compared to surfactant via endotracheal tube (ETT) reduced the incidence of the combined outcome of death or bronchopulmonary dysplasia (BPD) at 36 weeks' postmenstrual age (PMA) (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Yang 2020). Furthermore, S‐TC was associated with a reduced need for mechanical ventilation within the first 72 hours, and at any time, fewer cases of severe intraventricular haemorrhage (IVH), less BPD among survivors at 36 weeks' PMA, and lower mortality before hospital discharge (Bao 2015; Boskabadi 2019; Choupani 2018; Gupta 2020; Halim 2019; Han 2020; Jena 2019; Kanmaz 2013; Kribs 2015; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Olivier 2017; Yang 2020). The procedure was generally safe and well tolerated with comparable incidences of bradycardia, hypoxaemia, and procedural complications when compared to surfactant dosing via ETT. There was no significant difference in the need for more than one attempt to instrument the trachea with a thin catheter compared to an ETT. A higher rate of surfactant reflux was reported with surfactant administration via a thin catheter. Incidences of other hospital outcomes such as patent ductus arteriosus (PDA), any IVH, and retinopathy of prematurity (ROP) were similar.

Sensitivity analysis after exclusion of low‐quality studies showed similar results.

Only one trial compared surfactant therapy via thin catheter with continuation of continuous positive airway pressure (CPAP) (Göpel 2011). This study did not detect a difference in the incidence of death or BPD but did find a reduction in the need for mechanical ventilation in the first 72 hours of life, and at any time. A report of two‐year follow‐up of participants in this study showed that surfactant therapy via thin catheter appeared to be safe, with comparable two‐year outcomes (Herting 2020).

One study examined surfactant therapy via thin catheter with and without sedation (Dekker 2019). This study showed that low‐dose sedation increased comfort during surfactant administration via thin catheter in preterm infants but increased the need for transient non‐invasive ventilation. Sedation remains a controversial issue, given the uncertainty regarding its benefits and risks in this population (Mehler 2013; Zwicker 2016).

Neurosensory follow‐up data were limited. Two of the primary studies ‐ Göpel 2011 and Kribs 2015 ‐ reported neurosensory outcomes data (Herting 2020 and Mehler 2020, respectively).

Findings of this Cochrane Review suggest that in preterm infants with respiratory distress syndrome (RDS), surfactant administration via thin catheter in spontaneously breathing infants on CPAP may be preferable to surfactant dosing via ETT. Limited or no data were available for subgroup analyses, with no prophylaxis trials, limited numbers of infants at ≤ 28 weeks' gestation, and only one study mandating the use of sedating pre‐medication. Findings within these subgroups should thus be interpreted with caution.

Pulmonary benefits seen with tracheal catheterisation techniques may be related to many factors. These techniques not only avoid endotracheal intubation and associated positive‐pressure ventilation (PPV), they also combine surfactant administration with spontaneous breathing. There is potential for synergy arising from the combination of an effective treatment for RDS and avoidance of lung injury associated with PPV (Björklund 1997). Furthermore, these techniques allow continuation of CPAP, thus avoiding damage from temporary loss of functional lung capacity and atelectasis during the process of intubation (Sinclair 2009), while allowing spontaneous breathing to distribute surfactant within the lungs.

Overall completeness and applicability of evidence

Through a comprehensive search strategy, we identified 16 randomised controlled trials (RCTs) that matched our selection criteria in terms of population, intervention, comparison, and outcomes (16 studies in 18 publications). We excluded one trial report ‐ Mirnia 2013b ‐ due to overlap with a more complete report of the same trial, detailing findings at all three study centres (Mirnia 2013a). Eleven studies are ongoing (Characteristics of ongoing studies); the recruitment status of these studies is as follows.

1. Active recruiting (ACTRN12611000916943; ChiCTR1900020970; NCT04016246; NCT04445571).

2. Active not recruiting (ACTRN12611000917932; NCT03989960; NCT04073173; UMIN000021785).

3. Unknown (NCT01848262).

4. Terminated (NCT01615016).

5. Suspended (NCT02772081).

A total of 2164 newborn infants were assessed in all 16 trials (fewer for some outcomes) (Table 2). We believe the studies gathered in this review represent the best available evidence to answer the questions posed in this review. We were unable to undertake most of the subgroup analyses to further determine applicability of study findings to infants with different clinical and demographic characteristics due to insufficient data.

Quality of the evidence

Risk of bias in the studies included in this review varied (Figure 3), with eight trials having an overall low or unknown risk of bias (Boskabadi 2019; Choupani 2018; Halim 2019; Han 2020; Mirnia 2013a; Mohammadizadeh 2015; Mosayebi 2017; Yang 2020; see Figure 4).

The certainty of evidence, assessed according to GRADE, was moderate to low (Table 1). Although we judged the studies to be at varying risks of bias overall, evidence for our three main outcomes is drawn from studies at low risk of bias. We downgraded the quality of evidence to moderate for the main outcomes below mainly due to serious study limitations in many trials.

1. Death or bronchopulmonary dysplasia (BPD) at 36 weeks' postmenstrual age.

2. Need for intubation within the first 72 hours.

3. Bronchopulmonary dysplasia (BPD) among survivors at 36 weeks' postmenstrual age.

We also downgraded the certainty of evidence to low for the main outcomes below mainly due to serious study limitations and serious imprecision in many trials.

1. Air leak requiring drainage.

2. Severe intraventricular haemorrhage (grade III or IV).

3. Death during first hospitalisation (all causes).

We did not perform subgroup analysis, as subgroup data were limited or were not available.

Potential biases in the review process

Only one study exclusively studied extremely preterm infants at ≤ 28 weeks' gestation (Kribs 2015). Other studies included some infants at ≤ 28 weeks, but it was difficult to extract these data. Further studies are needed to address this high‐risk group. Several review outcomes were not available or could not be definitively addressed due to lack of data. Outcomes such as long‐term neurosensory outcomes are important, and data are scarce (Herting 2020; Mehler 2020). Furthermore, data based on factors such as sedation or analgesia are very important and will need to be addressed in future studies (Dekker 2016).

Agreements and disagreements with other studies or reviews

The findings of our systematic review are consistent in part with meta‐analyses published recently (Aldana‐Aguirre 2016; Gupta 2012; Isayama 2016; Lau 2017; Panza 2020; Wu 2017). They differ from the findings of a previously published meta‐narrative review, which did not include more recent studies (More 2014).

Authors' conclusions

Implications for practice.

Sixteen randomised clinical trials (18 publications) of surfactant administration with a thin catheter have been conducted using different thresholds for surfactant replacement and different comparator groups in preterm infants of varying gestational ages. Evidence from the studies included in this review indicates that infants with RDS treated with surfactant via thin catheter appear to be less likely to need mechanical ventilation, less likely to develop BPD, and less likely to develop severe IVH than are infants receiving surfactant via endotracheal tube. It is unclear whether delivery of surfactant via thin catheter is superior to continuation of CPAP without surfactant therapy. Surfactant administration via thin catheter was generally safe and well tolerated. Although the incidence of failure to catheterise the trachea at first attempt was not significantly different from intubation, training of healthcare personnel is recommended.

Implications for research.

Further research is needed to address high‐risk infants born at ≤ 28 weeks' gestation and to examine whether methods of S‐TC are effective when used for surfactant delivery as delivery room prophylaxis. Data on safety, use of sedating pre‐medication, and long term neurosensory outcomes are also needed. Procedural aspects such as catheter type (flexible versus semi‐rigid), use or not of Magill's forceps, and mode of laryngoscopy also require further investigation. Finally, thin catheter surfactant delivery should be compared with other minimally invasive methods, including aerosolisation and laryngeal mask administration.

What's new

Date	Event	Description
11 May 2021	Amended	Minor edit.

History

Protocol first published: Issue 5, 2015
Review first published: Issue 5, 2021

Appendices

Appendix 1. Search methods

The RCT filters have been created using Cochrane's highly sensitive search strategies for identifying randomised trials (Higgins 2019). The neonatal filters were created and tested by the Cochrane Neonatal Information Specialist.

CENTRAL via CRS Web

Date of search: search completed on 30 September 2020
Terms:
ID Search

#1 (pulmonary or lung or tracheal or catheter or less invasive or minimally invasive or non‐invasive or noninvasive) AND CENTRAL:TARGET
#2 MESH DESCRIPTOR Pulmonary Surfactants EXPLODE ALL AND CENTRAL:TARGET
#3 (surfactant* or Beractant or Poractant or Curosurf or Survanta or Exosurf or Lucinactant) AND CENTRAL:TARGET
#4 #3 or #2
#5 MESH DESCRIPTOR Infant, Newborn EXPLODE ALL AND CENTRAL:TARGET
#6 infant or infants or infant's or "infant s" or infantile or infancy or newborn* or "new born" or "new borns" or "newly born" or neonat* or baby* or babies or premature or prematures or prematurity or preterm or preterms or "pre term" or premies or "low birth weight" or "low birthweight" or VLBW or LBW or ELBW or NICU AND CENTRAL:TARGET
#7 #6 OR #5 AND CENTRAL:TARGET
#8 #1 AND #4 AND #7

MEDLINE via Ovid

Date of search: search completed on 30 September 2020
Terms:
1. (pulmonary or lung or tracheal or catheter or less invasive or minimally invasive or non‐invasive or noninvasive).mp.
2. exp Pulmonary Surfactants/
3. (surfactant* or Beractant or Poractant or Curosurf or Survanta or Exosurf or Lucinactant).mp.
4. 2 or 3
5. exp infant, newborn/
6. (newborn* or new born or new borns or newly born or baby* or babies or premature or prematurity or preterm or pre term or low birth weight or low birthweight or VLBW or LBW or infant or infants or 'infant s' or infant's or infantile or infancy or neonat*).ti,ab.
7. 5 or 6
8. randomized controlled trial.pt.
9. controlled clinical trial.pt.
10. randomized.ab.
11. placebo.ab.
12. drug therapy.fs.
13. randomly.ab.
14. trial.ab.
15. groups.ab.
16. or/8‐15
17. exp animals/ not humans.sh.
18. 16 not 17
19. 7 and 18
20. randomi?ed.ti,ab.
21. randomly.ti,ab.
22. trial.ti,ab.
23. groups.ti,ab.
24. ((single or doubl* or tripl* or treb*) and (blind* or mask*)).ti,ab.
25. placebo*.ti,ab.
26. 20 or 21 or 22 or 23 or 24 or 25
27. 6 and 26
28. 19 or 27
29. 1 and 4 and 28

CINAHL via EBSCOhost

Date of search: search completed on 30 September 2020
Terms:
(pulmonary OR lung OR tracheal OR catheter OR (less invasive) OR (minimally invasive) OR non‐invasive OR noninvasive)
AND
(surfactant* OR Beractant OR Poractant OR Curosurf OR Survanta OR Exosurf OR Lucinactant)
AND
(infant or infants or infant’s or infantile or infancy or newborn* or "new born" or "new borns" or "newly born" or neonat* or baby* or babies or premature or prematures or prematurity or preterm or preterms or "pre term" or premies or "low birth weight" or "low birthweight" or VLBW or LBW)
AND
(randomized controlled trial OR controlled clinical trial OR randomized OR randomised OR placebo OR clinical trials as topic OR randomly OR trial OR PT clinical trial)

ISRCTN

Date of search: search completed on 30 September 2020
Terms:
1. Interventions: Surfactant* AND Participant age range: Neonate

Appendix 2. Risk of bias tool

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

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

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

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

3. unclear risk.

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

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

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

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

3. 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 categorised the methods used to blind study participants and personnel from knowledge of which intervention a participant received. Blinding was assessed separately for different outcomes or classes of outcomes. We categorised the methods as:

1. low risk, high risk, or unclear risk for participants; and

2. 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 categorised the methods used to blind outcome assessment. Blinding was assessed separately for different outcomes or classes of outcomes. We categorised the methods as:

1. low risk for outcome assessors;

2. high risk for outcome assessors; or

3. 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 described the completeness of data including attrition and exclusions from the analysis. We noted whether attrition and exclusions were reported, numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion when reported, and whether missing data were balanced across groups or were related to outcomes. When sufficient information was reported or supplied by trial authors, we re‐included missing data in the analyses. We categorised the methods as:

1. low risk (< 20% missing data);

2. high risk (≥ 20% missing data); or

3. unclear risk.

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

For each included study, we described how we investigated the possibility of selective outcome reporting bias and what we found. For studies for which study protocols were published in advance, we compared pre‐specified outcomes versus outcomes eventually reported in the published results. If the study protocol was not published in advance, we contacted study authors to gain access to the study protocol. We assessed the methods as:

1. low risk (when it is clear that all of the study's pre‐specified outcomes and all expected outcomes of interest to the review have been reported);

2. high risk (when not all of the study's pre‐specified outcomes have been reported; one or more reported primary outcomes were not pre‐specified 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

3. unclear risk.

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

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

1. low risk;

2. high risk; or

3. unclear risk.

If needed, we explored the impact of the level of bias through undertaking sensitivity analyses.

Data and analyses

Comparison 1. Trials comparing S‐TC with S‐ETT ‐ overall analysis.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Death or BPD	10	1324	Risk Ratio (M‐H, Fixed, 95% CI)	0.59 [0.48, 0.73]
1.1.1 S‐TC vs INSURE	9	1113	Risk Ratio (M‐H, Fixed, 95% CI)	0.52 [0.40, 0.68]
1.1.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.79 [0.55, 1.13]
1.2 Need for intubation within the first 72 hours	12	1422	Risk Ratio (M‐H, Fixed, 95% CI)	0.63 [0.54, 0.74]
1.2.1 S‐TC vs INSURE	10	1166	Risk Ratio (M‐H, Fixed, 95% CI)	0.61 [0.50, 0.75]
1.2.2 S‐TC vs surfactant via ETT with delayed extubation	2	256	Risk Ratio (M‐H, Fixed, 95% CI)	0.68 [0.53, 0.86]
1.3 Air leak requiring drainage	6	1036	Risk Ratio (M‐H, Fixed, 95% CI)	0.58 [0.33, 1.02]
1.3.1 S‐TC vs INSURE	4	783	Risk Ratio (M‐H, Fixed, 95% CI)	0.72 [0.35, 1.48]
1.3.2 S‐TC vs surfactant via ETT with delayed extubation	2	253	Risk Ratio (M‐H, Fixed, 95% CI)	0.41 [0.16, 1.05]
1.4 Severe IVH	5	857	Risk Ratio (M‐H, Fixed, 95% CI)	0.63 [0.42, 0.96]
1.4.1 S‐TC vs INSURE	4	646	Risk Ratio (M‐H, Fixed, 95% CI)	0.77 [0.45, 1.32]
1.4.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.46 [0.24, 0.90]
1.5 Death during first hospitalisation	11	1424	Risk Ratio (M‐H, Fixed, 95% CI)	0.63 [0.47, 0.84]
1.5.1 S‐TC vs INSURE	10	1213	Risk Ratio (M‐H, Fixed, 95% CI)	0.60 [0.44, 0.82]
1.5.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.81 [0.37, 1.79]
1.6 BPD (clinical definition); in survivors to 36 weeks' PMA	11	1567	Risk Ratio (M‐H, Fixed, 95% CI)	0.57 [0.45, 0.74]
1.6.1 S‐TC vs INSURE	10	1378	Risk Ratio (M‐H, Fixed, 95% CI)	0.57 [0.44, 0.75]
1.6.2 S‐TC vs surfactant via ETT with delayed extubation	1	189	Risk Ratio (M‐H, Fixed, 95% CI)	0.58 [0.32, 1.05]
1.7 Catheter/ETT placement unsuccessful at first attempt	6	776	Risk Ratio (M‐H, Fixed, 95% CI)	1.08 [0.94, 1.26]
1.7.1 S‐TC vs INSURE	5	565	Risk Ratio (M‐H, Fixed, 95% CI)	1.11 [0.96, 1.28]
1.7.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	1.01 [0.65, 1.57]
1.8 Bradycardia (heart rate < 100 bpm) during the intervention	5	650	Risk Ratio (M‐H, Fixed, 95% CI)	1.14 [0.71, 1.83]
1.8.1 S‐TC vs INSURE	4	439	Risk Ratio (M‐H, Fixed, 95% CI)	0.81 [0.48, 1.39]
1.8.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	3.89 [1.13, 13.38]
1.9 Hypoxaemia (oxygen saturation < 80%) during the intervention	4	553	Risk Ratio (M‐H, Fixed, 95% CI)	1.30 [0.99, 1.71]
1.9.1 S‐TC vs INSURE	3	342	Risk Ratio (M‐H, Fixed, 95% CI)	0.68 [0.43, 1.08]
1.9.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	2.16 [1.50, 3.11]
1.10 Need for intubation within the first 72 hours or not intubated but reached failure criteria	6	720	Risk Ratio (M‐H, Fixed, 95% CI)	0.70 [0.58, 0.84]
1.10.1 S‐TC vs INSURE	4	464	Risk Ratio (M‐H, Fixed, 95% CI)	0.72 [0.53, 0.96]
1.10.2 S‐TC vs surfactant via ETT with delayed extubation	2	256	Risk Ratio (M‐H, Fixed, 95% CI)	0.68 [0.53, 0.86]
1.11 Need for intubation at any time	4	549	Risk Ratio (M‐H, Fixed, 95% CI)	0.73 [0.64, 0.83]
1.11.1 S‐TC vs INSURE	3	338	Risk Ratio (M‐H, Fixed, 95% CI)	0.70 [0.54, 0.90]
1.11.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.75 [0.68, 0.84]
1.12 Need for intratracheal surfactant therapy post intervention	7	963	Risk Ratio (M‐H, Fixed, 95% CI)	1.03 [0.78, 1.36]
1.12.1 S‐TC vs INSURE	6	918	Risk Ratio (M‐H, Fixed, 95% CI)	0.98 [0.73, 1.32]
1.12.2 S‐TC vs surfactant via ETT with delayed extubation	1	45	Risk Ratio (M‐H, Fixed, 95% CI)	1.57 [0.63, 3.96]
1.13 Duration of mechanical ventilation (days; in survivors)	4	361	Mean Difference (IV, Fixed, 95% CI)	‐0.03 [‐0.46, 0.41]
1.13.1 S‐TC vs INSURE	4	361	Mean Difference (IV, Fixed, 95% CI)	‐0.03 [‐0.46, 0.41]
1.14 Duration of any respiratory support (days; in survivors)	2	187	Mean Difference (IV, Fixed, 95% CI)	‐0.11 [‐0.25, 0.03]
1.14.1 S‐TC vs INSURE	2	187	Mean Difference (IV, Fixed, 95% CI)	‐0.11 [‐0.25, 0.03]
1.15 Duration of oxygen therapy (days; in survivors)	2	128	Mean Difference (IV, Fixed, 95% CI)	‐2.04 [‐4.51, 0.43]
1.15.1 S‐TC vs INSURE	2	128	Mean Difference (IV, Fixed, 95% CI)	‐2.04 [‐4.51, 0.43]
1.16 Postnatal systemic corticosteroid therapy for BPD mitigation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.86 [0.62, 1.21]
1.16.1 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.86 [0.62, 1.21]
1.17 BPD (physiological definition)	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.78 [0.50, 1.23]
1.17.1 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.78 [0.50, 1.23]
1.18 IVH, any grade	9	1353	Risk Ratio (M‐H, Fixed, 95% CI)	1.03 [0.81, 1.30]
1.18.1 S‐TC vs INSURE	9	1353	Risk Ratio (M‐H, Fixed, 95% CI)	1.03 [0.81, 1.30]
1.19 Cystic PVL	2	301	Risk Ratio (M‐H, Fixed, 95% CI)	0.40 [0.15, 1.11]
1.19.1 S‐TC vs INSURE	1	90	Risk Ratio (M‐H, Fixed, 95% CI)	0.91 [0.06, 14.18]
1.19.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.35 [0.12, 1.07]
1.20 PDA requiring medical therapy	4	484	Risk Ratio (M‐H, Fixed, 95% CI)	1.03 [0.80, 1.33]
1.20.1 S‐TC vs INSURE	4	484	Risk Ratio (M‐H, Fixed, 95% CI)	1.03 [0.80, 1.33]
1.21 NEC, modified Bell stage ≥2	3	686	Risk Ratio (M‐H, Fixed, 95% CI)	0.34 [0.14, 0.81]
1.21.1 S‐TC vs INSURE	3	686	Risk Ratio (M‐H, Fixed, 95% CI)	0.34 [0.14, 0.81]
1.22 ROP stage ≥ 3	5	734	Risk Ratio (M‐H, Fixed, 95% CI)	0.62 [0.30, 1.29]
1.22.1 S‐TC vs INSURE	4	523	Risk Ratio (M‐H, Fixed, 95% CI)	0.85 [0.36, 1.98]
1.22.2 S‐TC vs surfactant via ETT with delayed extubation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.28 [0.06, 1.31]
1.23 Duration of hospitalisation (days; in survivors)	5	590	Mean Difference (IV, Fixed, 95% CI)	‐1.10 [‐3.26, 1.06]
1.23.1 S‐TC vs INSURE	5	590	Mean Difference (IV, Fixed, 95% CI)	‐1.10 [‐3.26, 1.06]
1.24 Discharged home with oxygen	1	171	Risk Ratio (M‐H, Fixed, 95% CI)	0.26 [0.03, 2.27]
1.24.1 S‐TC vs INSURE	1	171	Risk Ratio (M‐H, Fixed, 95% CI)	0.26 [0.03, 2.27]

Comparison 2. Trials comparing S‐TC with S‐ETT ‐ sub‐group analyses.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Death or BPD	6		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
2.1.1 Death or BPD: ≤ 28 weeks' gestation	1	211	Risk Ratio (M‐H, Fixed, 95% CI)	0.79 [0.55, 1.13]
2.1.2 Death or BPD: 29 to 32 weeks' gestation	0	0	Risk Ratio (M‐H, Fixed, 95% CI)	Not estimable
2.1.3 Death or BPD: 33 to 36 weeks' gestation	0	0	Risk Ratio (M‐H, Fixed, 95% CI)	Not estimable
2.1.4 Death or BPD: surfactant prophylaxis trials	0	0	Risk Ratio (M‐H, Fixed, 95% CI)	Not estimable
2.1.5 Death or BPD: surfactant rescue trials	0	0	Risk Ratio (M‐H, Fixed, 95% CI)	Not estimable
2.1.6 Death or BPD: trials using sedation for S‐TC	0	0	Risk Ratio (M‐H, Fixed, 95% CI)	Not estimable
2.1.7 Death or BPD: trials using no sedation for S‐TC	6	1045	Risk Ratio (M‐H, Fixed, 95% CI)	0.58 [0.46, 0.73]

Comparison 3. Trials comparing S‐TC with S‐ETT ‐ sensitivity analysis.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Death or BPD	5	799	Risk Ratio (M‐H, Fixed, 95% CI)	0.60 [0.46, 0.78]
3.2 Need for intubation within the first 72 hours	6	954	Risk Ratio (M‐H, Fixed, 95% CI)	0.60 [0.51, 0.72]
3.3 Air leak requiring drainage	4	803	Risk Ratio (M‐H, Fixed, 95% CI)	0.53 [0.28, 1.02]
3.4 Severe IVH	4	559	Risk Ratio (M‐H, Fixed, 95% CI)	0.55 [0.34, 0.89]
3.5 Death during first hospitalisation	5	909	Risk Ratio (M‐H, Fixed, 95% CI)	0.72 [0.47, 1.09]
3.6 BPD (clinical definition); in survivors to 36 weeks' PMA	5	858	Risk Ratio (M‐H, Fixed, 95% CI)	0.45 [0.31, 0.64]

Comparison 4. Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
4.1 Death or BPD	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	0.92 [0.48, 1.74]
4.2 Incidence of air leak requiring drainage	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	0.52 [0.16, 1.67]
4.3 Severe IVH	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	1.38 [0.50, 3.85]
4.4 Death during first hospitalisation	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	1.45 [0.48, 4.44]
4.5 BPD (clinical definition); in survivors to 36 weeks' PMA	1	210	Risk Ratio (M‐H, Fixed, 95% CI)	0.62 [0.27, 1.41]
4.6 Catheter/ETT placement unsuccessful at first attempt	1	147	Risk Ratio (M‐H, Fixed, 95% CI)	8.80 [0.46, 167.44]
4.7 Bradycardia (heart rate < 100 bpm) during the intervention	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	9.33 [0.51, 171.25]
4.8 Need for intubation within the first 72 hours or not intubated but reached failure criteria	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	0.61 [0.42, 0.88]
4.9 Need for intubation at any time	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	0.46 [0.34, 0.61]
4.10 Postnatal systemic corticosteroid therapy for BPD mitigation	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	0.35 [0.04, 3.27]
4.11 Cystic PVL	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	2.59 [0.51, 13.08]
4.12 ROP ≥ stage 3	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	3.11 [0.33, 29.45]
4.13 Discharged home with oxygen	1	220	Risk Ratio (M‐H, Fixed, 95% CI)	1.04 [0.07, 16.37]
4.14 Cerebral palsy by clinical examination or other means	1	179	Risk Ratio (M‐H, Fixed, 95% CI)	1.98 [0.60, 6.52]

4.14. Analysis.

Comparison 4: Trials comparing S‐TC with continuation of non‐invasive support ‐ overall analysis, Outcome 14: Cerebral palsy by clinical examination or other means

Comparison 5. Trials comparing different methods of surfactant delivery via thin catheter ‐ overall analysis.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
5.1 Air leak	1	78	Risk Ratio (M‐H, Fixed, 95% CI)	2.57 [0.28, 23.65]
5.2 Severe IVH	1	78	Risk Ratio (M‐H, Fixed, 95% CI)	4.30 [0.21, 86.79]
5.3 Need for intubation during the procedure	1	78	Risk Ratio (M‐H, Fixed, 95% CI)	0.21 [0.03, 1.83]
5.4 Need for intubation within the first 24 hours	1	78	Odds Ratio (M‐H, Fixed, 95% CI)	1.56 [0.51, 4.83]
5.5 Death during first hospitalisation	1	78	Risk Ratio (M‐H, Fixed, 95% CI)	0.86 [0.06, 13.22]
5.6 Need for positive‐pressure ventilation during the intervention	1	78	Odds Ratio (M‐H, Fixed, 95% CI)	14.53 [3.79, 55.73]
5.7 Duration of the procedure (seconds)	1	78	Mean Difference (IV, Fixed, 95% CI)	0.00 [‐78.68, 78.68]
5.8 Pain score using a validated instrument for measuring discomfort/pain during the procedure (e.g. COMFORTneo score)	1	78	Mean Difference (IV, Fixed, 95% CI)	‐5.00 [‐6.59, ‐3.41]
5.9 Hypotension requiring treatment	1	78	Odds Ratio (M‐H, Fixed, 95% CI)	6.47 [0.32, 129.55]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Bao 2015.

Study characteristics
Methods	Parallel‐group, randomised controlled trial conducted at a tertiary NICU in China from January 2012 to December 2012
Participants	90 preterm infants with gestational age between 28 and 32 weeks who were stabilised with nasal continuous positive airway pressure (CPAP) and were eligible for surfactant administration within 2 hours after birth. Infants who were intubated in the delivery room were excluded Inclusion criteria Infants born at 28 to 32 weeks' gestational age Infants with RDS and in need of surfactant administration within 2 hours after birth Exclusion criteria Infants who had been previously intubated Infants with a congenital anomaly affecting respiratory function
Interventions	Intervention (less invasive surfactant administration ‐ LISA): infants were assigned to receive surfactant treatment during spontaneous breathing via a thin vascular catheter of 16 G diameter inserted into the trachea via laryngoscopy Control: infants were assigned to receive surfactant therapy via endotracheal tube with a brief period of mechanical ventilation. The endotracheal tube was withdrawn as soon as clinically possible after surfactant instillation, and the baby returned to CPAP. The whole procedure took about 3 minutes
Outcomes	The following data were recorded during surfactant instillation Oxygen saturation, heart rate, and FiO₂, recorded every 30 seconds pO₂ and pCO₂ values from blood gas samples before and 1 hour after surfactant administration Changes in CPAP pressure recorded every 30 minutes in the first 4 hours, and at 12 and 24 hours of life Intubation and mechanical ventilation in the first 72 hours (and thereafter) Further surfactant therapy Mortality Incidence of bronchopulmonary dysplasia (BPD) Patent ductus arteriosus (PDA) requiring medical and/or surgical therapy Intraventricular haemorrhage (IVH), grades III and IV Retinopathy of prematurity (ROP) > stage 2 Necrotising enterocolitis (NEC), Bell stage II or III Duration of respiratory support, including respiratory assistance (mechanical ventilation and/or CPAP) Oxygen therapy Intensive care admission
Notes	This study was conducted in an upper‐middle‐income country (China)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Generated by a computerised random number generator
Allocation concealment (selection bias)	Low risk	Opaque sealed envelopes
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	Open‐label
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	The trial was registered with Chinese Current Controlled Trials ChiCTR‐ICR‐15006001, on 20 February 2015, after patient recruitment was completed
Other bias	Low risk	Nil noted

Boskabadi 2019.

Study characteristics
Methods	Parallel‐group, randomised controlled trial conducted at tertiary NICUs at Ghaem Hospital of Mashhad, Iran, from 2012 to 2015
Participants	40 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age < 32 weeks Birth weight < 1500 grams Diagnosis of RDS based on clinical and radiological findings Managed with nCPAP with pressure of 5 to 8 cm H₂O Needing surfactant administration based on specific criteria (FiO₂ > 0.4 to keep SpO₂ > 85%, pCO₂ > 60 mmHg, pH < 7.2) Exclusion criteria Congenital anomaly Underlying disease (such as congenital infection, asphyxia, and congenital heart disease)
Interventions	Intervention: administration of surfactant via a thin catheter during spontaneous breathing. A 5F nasogastric tube was inserted into the trachea via laryngoscopy Control: InSurE (Intubate, Surfactant, Extubate) procedure
Outcomes	Primary and secondary outcomes not specified separately Outcomes Death up to 28 days of age Need for reparatory support Need for supplemental oxygen Adverse events during the procedure (coughing, desaturation, and reflux) Mechanical ventilation in the first 72 hours Duration of treatment with CPAP Pulmonary haemorrhage BPD Duration of hospital stay
Notes	This study was conducted in an upper‐middle‐income country (Iran). It included relatively mature infants (mean gestational age 30 (SD 2) to 31 (SD 2) weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation was not described. Study authors mentioned that (quote) "the infants who were included in the study were divided into two groups using the random block method"
Allocation concealment (selection bias)	Unclear risk	It is not clear whether or not allocation was concealed
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	It is not clear whether or not personnel and outcome assessors were blinded. Study authors described the study as (quote) "single‐blind"
Blinding of outcome assessment (detection bias) Short‐term outcomes	Unclear risk	It is not clear whether or not personnel and outcome assessors were blinded
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	The protocol of the trial was not available for review. The trial does not seems to be prospectively registered with any trial registry
Other bias	Low risk	Intervention and control groups were similar in gestation and birth weight

Choupani 2018.

Study characteristics
Methods	Parallel‐group, single‐centre randomised controlled trial conducted at Hajar Hospital of Shahrekord NICU in Iran, from 2016 to 2017
Participants	104 preterm infants with gestational age between 28 and 37 weeks who were stabilised with nasal continuous positive airway pressure (CPAP) and were eligible for surfactant administration within 2 hours after birth. Infants were enrolled when they reached FiO₂> 40%, or if they had moderate to severe respiratory distress Inclusion criteria Clinical signs of RDS at the first hour of life Absence of severe abnormalities Absence of chorioamnionitis Lack of the need for intubation and mechanical ventilation at birth Consent for participation in the study from the guardian of the neonate Exclusion criteria Diagnosis of any disease other than RDS, such as early‐onset sepsis, heart disease, or pneumonia Lack of the need for surfactant administration after treatment with nCPAP Need for continuation of mechanical ventilation immediately after surfactant administration Infants who were intubated in the delivery room
Interventions	Intervention (surfactant‐without‐intubation ‐ SWI): infants were assigned to receive surfactant treatment during spontaneous breathing via a thin vascular catheter of 5F diameter inserted into the trachea via laryngoscopy Control: infants were assigned to receive surfactant therapy via InSurE procedure, with a brief period of mechanical ventilation. The endotracheal tube was withdrawn as soon as clinically possible after surfactant instillation, and the baby was returned to CPAP
Outcomes	The following data were recorded during surfactant instillation HR > 190/min HR < 100/min SpO₂ < 80% Severe coughing or choking Frequency of attempts to successfully insert the tracheal tube or the catheter into the trachea Frequency of surfactant administration Duration of the need for oxygen Need for mechanical ventilation in the first 72 hours Duration of mechanical ventilation (if done) Duration of NCPAP Incidence of bronchopulmonary dysplasia Incidence of intraventricular haemorrhage Incidence of pulmonary haemorrhage Duration of hospitalisation Mortality
Notes	This study was conducted in an upper‐middle‐income country (Iran). It included relatively mature infants (mean (SD) gestational age 33.06 (2.3) to 32.9 (2.6) weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Quote: "convenience sampling method was used to select participants from the neonates diagnosed with RDS in the neonatal intensive care unit (NICU) of Hajar hospital in Shahrekord. Participants were then randomly allocated into two groups by random allocation software"
Allocation concealment (selection bias)	Unclear risk	It is not clear whether or not allocation was concealed
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Study authors mentioned (quote): "this double blind clinical trial"; however, no details were mentioned
Blinding of outcome assessment (detection bias) Short‐term outcomes	Unclear risk	Study authors mentioned (quote): "this double blind clinical trial"; however, no details were mentioned
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Protocol of trial was not available for review. The trial was not registered with a trial registry
Other bias	High risk	The study was not reported according to CONSORT guidelines; hence it is difficult to judge its quality

Dekker 2019.

Study characteristics
Methods	Parallel‐group, single‐centre randomised controlled trial conducted at Leiden University Medical Centre in the Netherlands, from January 2015 and July 2017
Participants	88 preterm infants Inclusion criteria Gestational age (GA) between 26 and 37 weeks Needing surfactant therapy for respiratory distress syndrome according to local criteria (FiO₂ > 0.3 and PEEP ≥ 8 cm H₂O) Exclusion criteria Imminent need for intubation because of respiratory insufficiency Infants with pneumothorax or pulmonary haemorrhage
Interventions	Intervention: infants were assigned to receive intravenous propofol (1 mg/kg) for sedation during the MIST procedure (surfactant treatment during spontaneous breathing via a 16G Angiocath) Control: infants were assigned to receive MIST procedure with no sedation
Outcomes	Primary outcome: percentage of infants with a COMFORTneo score < 14 during the procedure Secondary outcomes Occurrence of positive‐pressure ventilation during and right after the procedure Occurrence of intubation needed during the procedure and within 24 hours Number of attempts of endotracheal insertion of angiocatheter Duration of the total procedure (from start inserting laryngoscope until exit angiocatheter) Complications occurring during the procedure: desaturation < 85%, hypotension (mean below gestational age), bradycardia < 80 bpm, nasal haemorrhage Other complications: pneumothorax, pulmonary haemorrhage, resuscitation Heart rate and blood pressure before, during, and 5 minutes after the procedure
Notes	This study was conducted in a high‐income country (The Netherlands)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation sequence was generated by www.randomization.com. Allocation was stratified by GA (26 to 31 + 6 and 32 to 36 + 6 weeks) using variable block sizes (4 to 8)
Allocation concealment (selection bias)	Low risk	Opaque sealed envelopes were used to determine randomisation
Blinding of participants and personnel (performance bias) All outcomes	High risk	Neonatologists performing the procedure were not blinded
Blinding of outcome assessment (detection bias) Short‐term outcomes	Low risk	Researchers who analysed both primary and secondary outcomes were blinded to treatment allocation: 2 independent NICU nurses, blinded to allocation, reviewed recordings and measured level of comfort using the COMFORTneo scale
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Unclear risk	All infants recruited were accounted for; however, total of 10 of the 88 infants (11%) were randomised and received allocated treatment; they were later excluded from final analysis due to lack of parental consent (n = 7) or lack of availability of video recordings (n = 3)
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Low risk	Trial was registered with the Dutch Trial Registry (NTR5010) on 18 December 2014, and the first patient was recruited on 5 January 2015 (https://www.trialregister.nl/trial/4765)
Other bias	Low risk	Nil noted

Göpel 2011.

Study characteristics
Methods	Göpel 2011 Parallel‐group, multi‐centre randomised controlled trial (the Avoiding Mechanical Ventilation (AMV) trial) involving 12 German NICUs, conducted between October 2007 and January 2010 Herting 2020 2‐year follow‐up for AMV trial
Participants	Göpel 2011 220 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Preterm infants with gestational age between 26 + 0 weeks and 28 + 6 weeks, irrespective of respiratory status Birth weight < 1500 grams Age < 12 hours infants who were already intubated in the delivery room were eligible for inclusion Exclusion criteria Lethal malformations Prior surfactant treatment without intubation Herting 2020 The follow‐up study reported 2‐year outcomes for 179 infants from the 206 surviving infants (86.9%) among the original study participants (n = 220) Inclusion and exclusion criteria for the primary trial were as above All infants from the primary trial (n = 220) who were alive at 2 years of age (n = 206) were eligible for follow‐up. Of the total cohort of 220 infants: 12 died before hospital discharge 2 died after discharge and before 2‐year follow‐up 12 parents declined to participate in follow‐up 29 were not available for follow‐up 14 were lost to follow‐up 2‐year follow‐up was completed in 95 infants from the intervention group and in 84 infants from the control group
Interventions	Intervention: infants were assigned to receive rescue surfactant treatment during spontaneous breathing via a thin catheter of 2.5 to 5 F diameter inserted into the trachea via laryngoscopy and Magill's forceps, if they needed a fraction of inspired oxygen > 0·30 Control: infants were assigned to continue on CPAP and to receive rescue intubation and surfactant treatment if needed
Outcomes	Göpel 2011 Primary outcome: need for any mechanical ventilation, or not being ventilated but having pCO₂ > 65 mmHg (8.6 kPa) or FiO₂ > 0·60, or both, for longer than 2 hours between 25 hours and 72 hours of age. Ventilation on Day 1 was not included in the endpoint analysis to allow surfactant treatment in the standard treatment group Secondary outcomes Incidence and duration of any mechanical ventilation during the infant’s time in hospital Duration of oxygen supplementation or CPAP, or both Number of surfactant doses given per infant Bronchopulmonary dysplasia Death or bronchopulmonary dysplasia; death or treatment with supplemental oxygen at discharge FiO₂ and oxygen saturation in the first 3 days after birth Drug therapy given (sedatives and analgesics, inotropic agents, methyl xanthines, diuretics, and dexamethasone) Serious adverse events (e.g. pneumothorax, intraventricular haemorrhage grade III or IV, pulmonary haemorrhage, periventricular leukomalacia, surgical treatment of patent ductus arteriosus, surgical treatment of necrotising enterocolitis or focal intestinal perforation, laser therapy or cryotherapy of retinopathy, death) Herting 2020 2‐Year Bayley II scales Bayley II mental development index (MDI) Bayley II psychomotor development index (PDI) Ability to walk Growth at 2 years (weight and length) Post‐hospital discharge complications Parent‐reported bronchitis Visual and hearing impairment
Notes	This study was conducted in a high‐income country (Germany)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Infants were randomly assigned with Randomization In Treatment Arms (RITA) programme (version 1.2) in a 1:1 ratio with variable block sizes (4 and 6) by an independent statistician
Allocation concealment (selection bias)	Low risk	An independent statistician prepared sequentially numbered, sealed, opaque envelopes stratified by centre and multiple birth status
Blinding of participants and personnel (performance bias) All outcomes	High risk	None of the participants and none of those giving the interventions, assessing outcomes, or analysing the data were masked to treatment
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	None of the participants and none of those giving the interventions, assessing outcomes, or analysing the data were masked to treatment
Blinding of outcome assessment (detection bias) Long‐term outcomes	High risk	None of the participants and none of those giving the interventions, assessing outcomes, or analysing the data were masked to treatment
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Low risk	All infants recruited were accounted for
Selective reporting (reporting bias)	Low risk	Göpel 2011 (low risk) The trial was prospectively registered with a clinical trial registry (ISRCTN05025922) Herting 2020 (unclear risk) The trial protocol was not available for review and was not prospectively registered with any trial registry
Other bias	Low risk	Göpel 2011 (low risk) Pragmatic trial with no treatment standardisation between centres (e.g. caffeine, surfactant dose) leading to variability between centres. However, multi‐variate logistical regression analysis was implemented and showed no significant centre effect Herting 2020 (high risk) The initial design of the AMV study did not include the planning of a follow‐up study. However, this was planned towards the end of the AMV study. An addendum to the ethical approval for additional follow‐up data collection was sought, and parents were contacted and were asked for additional information. A protocol/case record form for follow‐up data was filled in by respective participating centres Follow‐up assessment was done by different examiners who were not all blinded. Follow‐up examinations were carried out at individual institutes under different conditions, all of which may influence the quality of data. Furthermore, collected data are not complete for all items. Of note is that the AMV trial was powered to assess differences in BPD but not to assess neurodevelopment. Subgroup analyses in each GA stratum, especially the 23‐ and 24‐week GA strata, are based on rather small numbers of patients

Gupta 2020.

Study characteristics
Methods	Parallel‐group, single‐centre randomised controlled trial conducted at a level III NICU in a tertiary care hospital in Kolkata, India, from March 2019 to December 2019
Participants	58 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Preterm infants with gestational age between 28 + 0 weeks and 34 + 0 weeks Inborn at the study hospital Age < 6 hours Respiratory distress syndrome requiring CPAP or nasal intermittent positive‐pressure ventilation (NIPPV) CPAP pressure 5 to 6 cm H₂O and FiO₂ > 0.30 NIPPV settings PEEP 5 to 6 cm H₂O, PIP 15 to 18 cm H₂O, and FiO₂ > 0.30 Exclusion criteria Previously intubated, or in imminent need of intubation because of respiratory distress, apnoea, or persistent acidosis Congenital anomaly or condition that might adversely affect breathing Identifiable alternative cause for respiratory distress (e.g. congenital pneumonia, pulmonary hypoplasia) Neonates with APGAR ≤ 4 at 5 minutes after birth
Interventions	Intervention: infants were assigned to receive rescue surfactant treatment during spontaneous breathing via a gastric tube 5 F diameter inserted into the trachea via laryngoscopy and Magill's forceps if they needed fraction of inspired oxygen > 0·30 Control: infants were assigned to receive surfactant therapy via InSurE procedure, with a brief period of mechanical ventilation. The endotracheal tube was withdrawn immediately a few minutes after surfactant instillation, and the baby was returned to NIPPV
Outcomes	Primary outcome: invasive mechanical ventilation within the first 72 hours after birth Secondary outcomes Need for invasive mechanical ventilation at any time and its duration Duration of non‐invasive ventilation Need for repeated doses of surfactant Incidence of patent ductus arteriosus, necrotising enterocolitis, bronchopulmonary dysplasia, intraventricular haemorrhage grade > 2, pneumothorax, and mortality
Notes	This study was conducted in an upper‐middle‐income country (India). It included relatively mature infants (median gestational age 30.07 weeks (intervention) and 29.90 weeks (control)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisation
Allocation concealment (selection bias)	Low risk	Sequentially numbered, sealed, opaque envelopes
Blinding of participants and personnel (performance bias) All outcomes	High risk	There was no blinding in the study, and even outcome assessors were not blinded
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	There was no blinding in the study, and even outcome assessors were not blinded
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were analysed
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Low risk	Registered with clinical trial registry of India (registration number CTRI/2019/03/017992) prospectively on 8 March 2019
Other bias	Low risk	Nil noted

Halim 2019.

Study characteristics
Methods	Parallel‐group, single‐centre randomised controlled trial conducted at a NICU in Islamabad, Pakistan, from April till December 2017
Participants	100 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Preterm infants with gestational age ≤ 34 weeks Inborn at the study hospital Age < 12 hours Respiratory distress syndrome requiring CPAP, with FiO₂ > 0.40 Exclusion criteria Previously intubated Congenital anomaly
Interventions	Intervention: infants were assigned to receive rescue surfactant treatment during spontaneous breathing via a nasogastric tube of 6 F diameter inserted into the trachea via laryngoscopy if they needed fraction of inspired oxygen > 0·40 on CPAP Control: infants were assigned to receive surfactant therapy via InSurE procedure, with a brief period of 15 to 20 minutes of positive‐pressure ventilation via a T‐piece resuscitaire. The endotracheal tube was withdrawn immediately after surfactant instillation, and the baby was returned to CPAP
Outcomes	Primary and secondary outcomes are not clearly defined. Study authors studied multiple demographic and clinical data Age at the time of procedure FiO₂ reduction after the procedure Duration of respiratory support CPAP Invasive mechanical ventilation Duration of invasive mechanical ventilation Duration of hospital stay Pneumothorax PDA Pulmonary haemorrhage Mortality
Notes	This study was conducted in an upper‐middle‐income country (Pakistan). It included relatively mature infants (only 10% of infants were at < 28 weeks' GA)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Serial numbers from 1 to 100 were randomly divided into 2 groups via a Web‐based randomisation tool (www.randomizer.org)
Allocation concealment (selection bias)	Unclear risk	It is not clear whether or not allocation was concealed
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	It is not clear whether or not personnel were blinded
Blinding of outcome assessment (detection bias) Short‐term outcomes	Unclear risk	It is not clear whether or not outcome assessors were blinded
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Protocol of the trial was not available for review. Trial does not seem to be prospectively registered with any trial registry
Other bias	High risk	Control (InSurE) group had slightly less antenatal steroid coverage. This group also was intubated a bit later compared to the intervention (thin catheter) group. No adjustment for these differences was made in the final analysis

Han 2020.

Study characteristics
Methods	Parallel‐group, multi‐centre randomised controlled trial (Minimally Invasive Surfactant Administration (MISA) trial) involving 8 level III Chinese NICUs, conducted between 1 July 2017, and 31 December 2018. It appears that patients were followed up until 30 March 2019; however follow‐up data are not reported in this manuscript
Participants	344 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age < 31 + 6 Diagnosis of RDS based on clinical signs (respiratory rate > 60/min, with retractions, nasal flaring, grunting, or cyanosis; requiring FiO₂ > 0.4 to maintain SpO₂ > 85%) Managed with CPAP Need for surfactant administration within 6 hours after birth Exclusion criteria Infants who were intubated in the delivery room or before surfactant administration Infants with major congenital malformations affecting respiratory function Infants who died or were transferred to other hospitals for surgery or with uncompleted data Infants who were enrolled in other interventional studies For the MISA group, infants who received second or third dose of surfactant through ETT during the first 72 hours
Interventions	Intervention (MISA): infants were assigned to receive surfactant treatment during spontaneous breathing via a gastric catheter of 5 F diameter inserted into the trachea via laryngoscopy and 10‐cm ophthalmic forceps Control (EISA ‐ endotracheal intubation surfactant administration): infants were assigned to receive surfactant therapy via an endotracheal tube, with mechanical ventilation. Investigators were encouraged to wean the infant from positive‐pressure ventilation as soon as possible. Extubation criteria were established as FiO₂ < 0.3 and mean airway pressure (MAP) < 8 cm H₂O
Outcomes	Primary outcome: BPD at 36 weeks’ GA Secondary outcomes Neonatal sepsis including clinical sepsis and blood culture‐confirmed sepsis Pneumonia (including ventilator‐associated pneumonia) PDA based on clinical signs and echocardiographic confirmation White matter injury and IVH based on cranial ultrasound ROP NEC based on clinical sign and X‐ray finding Duration of positive‐pressure ventilation Days on supplemental oxygen Length of NICU stay Body weight on discharge Short‐term safety including the following Transient bradycardia (heart rate < 100/min) SpO₂ < 85% Choking and coughing, laryngeal spasms Failure of surfactant administration Data on serious adverse events (not specified)
Notes	This study was conducted in an upper‐middle‐income country (China). It included relatively mature infants (mean gestational age 30.6 (SD 1.6) to 30.8 (SD 1.3) weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of sequence generation for randomisation was unclear. Allocation ratio was 1:1
Allocation concealment (selection bias)	Low risk	Sequentially numbered, opaque sealed envelopes were used to complete group assignment
Blinding of participants and personnel (performance bias) All outcomes	High risk	Assigned treatment was not blinded, as the mode of respiratory management was apparent to clinicians and nurses in the NICU
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	Assigned treatment was not blinded
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	High risk	Outcomes for 13.4% of randomised infants not reported
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Trial was registered with ClinicalTrials.gov (NCT04077333) on 4 September 2019, after patient recruitment was completed
Other bias	High risk	Multi‐centre study, with variation in proportion of eligible infants enrolled at each centre (12.8% to 56.5%). Practices were not consistent between centres, particularly in relation to extubation criteria and ventilatory practices. Infants who died or were transferred to other hospitals for surgery and those with incomplete data (n = 34) were excluded after randomisation and after completing/starting the intervention. For the intervention arm (MISA group), infants who were given a second surfactant dose by intubation and/or were ventilated during the first 72 hours (n = 12) were excluded from final analysis, and infants given a second dose of surfactant by MISA were included. Data analyses were performed on a per‐protocol basis rather than on an intention‐to‐treat basis

Jena 2019.

Study characteristics
Methods	Parallel‐group, multi‐centre randomised controlled trial involving 3 Indian NICUs, conducted between 2013 and 2017
Participants	Total of 350 preterm infants at ≤ 34 weeks were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Preterm infants with gestational age ≤ 34 weeks Diagnosis of RDS clinically and confirmed by X‐ray of chest Required fraction of inspired oxygen (FiO₂) > 30% on CPAP of 6 cm H₂O to maintain saturation between 90% and 95% in first 6 hours of life Exclusion criteria Infants with major congenital anomalies including cardiac malformation Infants who required surfactant after 6 hours of life Infants who required intubation at birth
Interventions	Intervention (SurE): surfactant treatment during spontaneous breathing via a 16 G Angiocath (Desilet; Vygon) or a 6 FG feeding tube (based on clinician preference) Control (InSurE): infants were intubated with an appropriately sized endotracheal tube, and surfactant was administered as in the SurE group. After extubation, nCPAP was started as per the SurE group No sedation nor pre‐medication was used in both groups. Criteria for subsequent doses of surfactant, intubation, and MV were the same in both groups
Outcomes	Primary outcome: effect of the SurE technique on the need for MV in the first 72 hours of life Secondary outcomes Repeat dose of surfactant therapy Rates of haemodynamically significant PDA Pneumothorax IVH ≥ grade 2 (Papile classification) NEC: modified Bell’s stage ≥ 2) BPD Early‐onset sepsis (EOS) Duration of oxygen Duration of hospital stay Apnoea Bradycardia or desaturation during surfactant administration Mortality before discharge
Notes	This study was conducted in an upper‐middle‐income country (India). It included relatively mature infants (median (IQR) gestational age 31 (29 to 33) weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation was done by computer‐generated random sequence number. Allocation ratio was 1:1
Allocation concealment (selection bias)	Low risk	Allocation concealment was done by using an opaque sealed envelope. Generation of random numbers and assignment was completed by a person not involved in the study
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of the intervention was not possible due to the nature of the treatment
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	Blinding of the intervention was not possible due to the nature of the treatment
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Protocol of the trial was not available for review. Trial was not prospectively registered
Other bias	Low risk	Data analysis was based on intention‐to‐treat analysis using StataCorp 11.1, Houston, TX Researcher estimated a sample size of 150 for each group and managed to recruit 175

Kanmaz 2013.

Study characteristics
Methods	Parallel‐group, single‐centre randomised controlled trial (Take Care Trial) conducted at a tertiary NICU in Turkey between December 2010 and December 2011
Participants	200 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age < 32 weeks Clinical and laboratory signs of RDS FiO₂ ≥ 0.4 in first 2 hours of life Exclusion criteria Infants who were intubated in delivery room
Interventions	Intervention: early (in NICU) administration of surfactant via a thin catheter during spontaneous breathing (Take Care). A 5 F flexible, sterile nasogastric tube was inserted into the trachea via laryngoscopy without Magill's forceps Control: early (in NICU) InSurE procedure
Outcomes	Primary outcomes: the need for intubation and mechanical ventilation in the first 72 hours of life (and thereafter); feasibility of the technique Secondary outcomes Repeated surfactant therapy Duration of respiratory support Rates of intraventricular pneumothorax Patent ductus arteriosus requiring medical or surgical treatment Intraventricular haemorrhage ≥ grade III according to the Papille classification Retinopathy of prematurity > stage 2 Necrotising enterocolitis: Bell's stage ≥ 2 Length of hospitalisation BPD or death
Notes	This study was conducted in an upper‐middle‐income country (Turkey)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated
Allocation concealment (selection bias)	Low risk	Sequentially numbered sealed opaque envelopes
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	Open‐label
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Trial was registered with ClinicalTrials.gov registry (NCT01329432) on 6 April 2011, after patient recruitment began
Other bias	High risk	Single‐centre study. Some infants who might have been eligible for the study could not be enrolled because of concern for standardisation of the intervention (e.g. unavailability of the study team)

Kribs 2015.

Study characteristics
Methods	Kribs 2015 Parallel‐group, multi‐centre randomised controlled trial involving 13 German NICUs (Nonintubated Surfactant Application (NINSAPP) trial) between April 2009 and June 2012 Mehler 2020 2‐year follow‐up for NINSAPP trial
Participants	Kribs 2015 211 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestation between 23 + 0 and 26 + 6 weeks RDS with Silverman score ≥ 5 and/or FiO₂ ≥ 0.3 to maintain oxygen saturation > 83% Postnatal age > 10 minutes and < 2 hours Exclusion criteria Primary cardiopulmonary resuscitation Prenatally diagnosed severe malformation No parental consent Participation in another interventional trial Mehler 2020 All infants from the primary trial (n = 211) who were alive at 2 years of age (n = 182) were eligible for follow‐up. Of the total cohort of 211 infants 24 died before hospital discharge 7 died after discharge and before 2‐year follow‐up 26 could not be contacted or declined to participate in follow‐up 2‐year follow‐up was completed in 78 infants from the intervention group and in 78 infants from the control group
Interventions	Intervention: surfactant via A 4 F end‐hole catheter was inserted into the trachea via laryngoscopy and Magill's forceps while the infant was breathing via nasal CPAP (LISA) Control: surfactant via endotracheal intubation and mechanical ventilation
Outcomes	Kribs 2015 Primary outcomes: survival without physiological BPD at 36 weeks’ GA Secondary outcomes Survival without major complications (these complications included BPD, severe intraventricular haemorrhage, cystic periventricular leukomalacia, and surgery for necrotising enterocolitis, focal intestinal perforation, or retinopathy of prematurity) Pneumothorax Severe intraventricular haemorrhage Cystic periventricular leukomalacia Laser therapy for retinopathy of prematurity Surgery required for necrotising enterocolitis or focal intestinal perforation Persistent ductus arteriosus requiring surgery Treatment failure (need for intubation and ventilation within the first 72 hours of life) Duration of mechanical ventilation CPAP Oxygen supplementation Length of stay Daily weight gain until 36 weeks’ GA Additional safety analyses including incidence of bradycardia (heart rate < 100/min), oxygen desaturation < 80%, coughing, choking, laryngeal spasms during application, and surfactant application failure Mehler 2020 2‐Year Bayley II scales Bayley II MDI Bayley II PDI Ability to walk Growth at 2 years (weight; length, and head circumference) Post‐hospital discharge complications Parent‐reported days of hospitalisation within the first year of life Visual and hearing impairment
Notes	This study was conducted in a high‐income country (Germany)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random allocation was designed in a 1:1 ratio with variable block sizes by an independent statistician
Allocation concealment (selection bias)	Low risk	Serially numbered opaque, sealed envelopes
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of the procedure was not attempted
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	Open‐label
Blinding of outcome assessment (detection bias) Long‐term outcomes	High risk	Open‐label
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Low risk	All infants recruited were accounted for
Selective reporting (reporting bias)	Low risk	Kribs 2015 (low risk) Trial was prospectively registered with a clinical trial registry (NCT00751959) Mehler 2020 (unclear risk) Protocol of trial was not available for review. Trial does not seems to be prospectively registered with any trial registry
Other bias	Low risk	Kribs 2015 (low risk) Pragmatic trial with no treatment standardisation between centres (e.g. caffeine, surfactant dose) leading to variability between centres. However, multi‐variate logistical regression analysis was implemented and showed no significant centre effect Mehler 2020 (unclear risk) The initial design of the NINSAPP study did not include the planning of a follow‐up study. However, this was planned after the end of the NINSAPP study. An addendum to the ethical approval for additional follow‐up data collection was sought, and parents were contacted and were asked for additional information. A protocol/case record form for follow‐up data was filled in by the respective participating centres Follow‐up assessment was done by different examiners who were not all blinded. Follow‐up examinations were carried out at the individual institutes under different conditions, all of which may influence the quality of data. Furthermore, collected data are not complete for all items. Of note is that the NINSAPP trial was powered to assess differences in BPD but not for assessment of neurodevelopment. Subgroup analyses in each GA stratum, especially the 23‐ and 24‐week GA strata, are based on rather small numbers of patients

Mirnia 2013a.

Study characteristics
Methods	Parallel‐group, multi‐centre randomised controlled trial conducted at 3 tertiary NICUs in Iran (Tabriz, Isfahan, and Mashhad) from February 2010 to October 2012
Participants	136 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age between 27 and 32 weeks Exclusion criteria Apgar score < 6 at 5 minutes Congenital malformations including congenital heart disease.
Interventions	Intervention: administration of surfactant via a thin catheter during spontaneous breathing. A 5 F feeding tube is inserted into the trachea via laryngoscopy. Atropine 5 μg/kg was administered before catheterisation Control: InSurE procedure
Outcomes	The following details during surfactant instillation Pulse oximetry saturation, heart rate, and FiO₂, recorded every 30 seconds for about 3 minutes pO₂ and pCO₂ values from blood gas samples before and 1 hour after surfactant administration Pneumothorax NEC: Bell stage II or III ROP > stage 2 Proven and suspected sepsis IVH Incidence of BPD PDA requiring medical and/or surgical therapy Mortality Duration of CPAP Duration of oxygen therapy Length of intensive care admission Arterial blood gas parameters before and 2 hours after surfactant administration Changes in PEEP, heart rate, FiO₂, and SpO₂ at 0, 5, 10, 15, 60, and 120 minutes, then at 24, 48, and 72 hours after surfactant administration
Notes	This study was conducted in an upper‐middle‐income country (Iran). Atropine was given before intubation
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method not reported
Allocation concealment (selection bias)	Unclear risk	Method not reported
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Method not reported
Blinding of outcome assessment (detection bias) Short‐term outcomes	Unclear risk	Method not reported
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Protocol of trial was not available for review. Study was not registered with prospective trial registry
Other bias	Unclear risk	Study was not reported according to CONSORT guidelines; hence it is difficult to judge its quality

Mohammadizadeh 2015.

Study characteristics
Methods	Parallel‐group, randomised controlled trial conducted at 2 tertiary NICUs in Isfahan, Iran, from December 2012 to May 2013
Participants	38 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age ≤ 34 weeks Exclusion criteria History of chorioamnionitis in mother Apgar score ≤ 4 at 5 minutes, threatening congenital anomalies Infants who were intubated in delivery room
Interventions	Intervention: administration of surfactant via a thin catheter during spontaneous breathing. A 4 F flexible, nasogastric tube was inserted into the trachea via laryngoscopy and Magill's forceps Control: InSurE procedure
Outcomes	Primary outcomes: need for mechanical ventilation in the first 72 hours of life; feasibility of the technique Secondary outcomes Duration of mechanical ventilation Duration of treatment with CPAP Duration of oxygen requirement Number of attempts at successful insertion of endotracheal tube or thin catheter into the trachea Number of surfactant doses and the need for more than 1 dose of the drug Rate of mortality Chronic lung disease (BPD), defined as oxygen requirement by the later of 36 weeks' postmenstrual age or 28 days after birth IVH Adverse events during the procedure (including heart rate < 100/min or > 200/min, SpO₂ ≤ 80%, coughing, and choking)
Notes	This study was conducted in an upper‐middle‐income country (Iran). It included relatively mature infants (mean gestational age 30 (SD 2) to 31 (SD 2) weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation was not described
Allocation concealment (selection bias)	Low risk	Consecutively numbered, opaque and sealed envelopes
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Quote: "the intervention was performed by a medical practitioner (level 2 neonatal trainee) who was not involved in randomisation or outcome assessment" It is not clear whether or not other clinicians were blinded
Blinding of outcome assessment (detection bias) Short‐term outcomes	Unclear risk	It is not clear whether or not outcome assessors were blinded
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Protocol of trial was not available for review. Trial does not seems to be prospectively registered with any trial registry
Other bias	Unclear risk	Intervention (thin catheter) group was slightly smaller in birth weight. An adjustment was made for this difference. There was no comment about practices in the 2 collaborating centres (e.g. extubation criteria, adjuvant treatments such as caffeine)

Mosayebi 2017.

Study characteristics
Methods	Parallel‐group, randomised controlled trial conducted at a single tertiary NICU in Tehran (Roointan‐Arash Maternity Hospital), Iran, from April 2013 to February 2014
Participants	53 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age 28 to 34 weeks RDS Infants with RDS were randomised to one of the study arms once they required FiO₂ > 40% to maintain oxygen saturation (SpO₂) in the range of 85% to 92% Exclusion criteria Infants with congenital anomalies Apgar score ≤ 4 at 5 minutes Infants who were intubated in delivery room Lack of parental consent
Interventions	Intervention: administration of surfactant via a thin catheter during spontaneous breathing. A 5 F feeding tube was inserted into the trachea via laryngoscopy and Magill's forceps Control: InSurE procedure: surfactant was administered by passing a feeding tube through the endotracheal tube
Outcomes	Primary outcomes Need for intubation during first 72 hours Pulmonary haemorrhage Pneumothorax Patent ductus arteriosus requiring medical or surgical treatment Intraventricular haemorrhage grade > 2 Secondary outcomes Retinopathy of prematurity > stage 2 Necrotising enterocolitis ≥ stage 2 Sepsis Bronchopulmonary dysplasia (oxygen dependence at 36 weeks' postmenstrual age) Death No sedation nor pre‐medication was used in both groups Treatment was considered a failure if pH < 7.2, FiO₂ > 60%, and PCO₂ > 60 mmHg persisted for longer than 2 hours, or if apnoea occurred, upon which the infant was intubated, and if required, surfactant was administered
Notes	This study was conducted in an upper‐middle‐income country (Iran). It included relatively mature infants (mean gestational age 31.9 (SD 1.5) to 32.6 (SD 1.1) weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No details (quote): "simple randomization was used in the allocation of participants"
Allocation concealment (selection bias)	Unclear risk	Method not reported
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding was not performed at any stage of the study ‐ from intervention to data analysis and interpretation of results
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	Blinding was not performed at any stage of the study ‐ from intervention to data analysis and interpretation of results
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Study was retrospectively registered with the Iranian Registry of clinical trials, IRCT2014080716937N4
Other bias	Unclear risk	Study was not reported according to CONSORT guidelines; hence it is difficult to judge its quality

Olivier 2017.

Study characteristics
Methods	Parallel‐group, randomised controlled trial conducted at 3 Canadian centres from January 2014 to May 2016
Participants	45 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age 32 + 0 to 36 + 6 weeks Requiring nasal continuous positive airway pressure of 6 cmH₂O and 35% FiO₂ to maintain SpO₂ ≥ 90% Age ≤ 24 hours Exclusion criteria Lethal conditions or significant congenital malformations Intubation or pneumothorax before enrolment
Interventions	Intervention: administration of surfactant via a thin catheter during spontaneous breathing. A 5 F flexible, nasogastric tube was inserted into the trachea via laryngoscopy and Magill's forceps. Atropine (20 mcg/kg) and fentanyl (1 mcg/kg) were administered before the procedure. If desaturation or bradycardia occurred, the procedure was temporarily interrupted. Re‐dosing was allowed if FiO₂ rose to or remained above 0.30, as per Canadian recommendations Control: standard management (surfactant was given only after intubation, based on judgement of attending physician). InSurE was not routinely practiced at participating centres
Outcomes	Primary outcomes Need for mechanical ventilation in the first 3 days of the study Occurrence of a pneumothorax requiring chest tube insertion in the first 3 days of the study Surpassing failure criteria defined as ≥ 1 of Development of respiratory acidosis with pH < 7.20 and pCO₂ > 70 mmHg on 2 blood gas analyses Lack of improvement in FiO₂ in the 4 hours following surfactant therapy Secondary outcomes Number of attempts at successful insertion of endotracheal tube or thin catheter into the trachea Number of laryngoscopy attempts Adverse events documented during surfactant delivery via thin catheter Surfactant reflux Moderate (SpO₂ 60% to 80%) or severe (SpO₂< 60%) hypoxic events Adverse events documented during intubation procedures Moderate (SpO₂ 60% to 80%) or severe (SpO₂< 60%) hypoxic events Duration of mechanical ventilation Duration of treatment with CPAP Duration of oxygen therapy Number of surfactant doses and need for more than 1 dose of the drug Duration of NICU hospitalisation (days)
Notes	This study was conducted in a high‐income country (Canada). Infants meeting failure criteria were regarded as having reached the primary outcome in the analysis, even if MV was not initiated. Multi‐variate regression was used to control for age at oxygen introduction and age at surfactant administration among other confounders. All patients received atropine and fentanyl before the procedure
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation sequence was created in a 1:1 ratio in blocks of 4 by an independent statistician
Allocation concealment (selection bias)	Low risk	Participants were randomised immediately after inclusion via sealed opaque envelopes (prepared by a nurse not involved in the study)
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of the procedure was not attempted
Blinding of outcome assessment (detection bias) Short‐term outcomes	High risk	Open‐label
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for. A total of 7 eligible patients were not recruited and 4 patients were randomised before inclusion criteria were fulfilled
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Trial was not registered with any trial registry
Other bias	Unclear risk	Multi‐centre study. The clinical approach to management of patients in the control group was not standardised after randomisation and was left to the judgement of the clinician (e.g. there was no intubation criteria). Surfactant given after intubation via ETT (S‐ETT) on the judgement of attending physician

Yang 2020.

Study characteristics
Methods	Parallel‐group, quasi‐randomised controlled trial conducted at tertiary NICUs at Children’s Hospital of Shanxi, China, from February 2017 to January 2018 Patients were randomised according to 'numbered list': odd number – LISA group, even number ‐ InSurE group (verbal communication with study authors)
Participants	97 preterm infants were included in the trial. Inclusion and exclusion criteria were as follows Inclusion criteria Gestational age from 32 + 0 to 36 + 6 weeks Requiring nCPAP treatment (PEEP > 6 H₂O and FiO₂ > 0.4) within 12 hours of age RDS diagnosis by physical and by chest X‐ray examination Exclusion criteria Tracheal intubation after birth Congenital developmental abnormalities affecting respiratory outcome Lack of consent
Interventions	Intervention: administration of surfactant via a thin catheter during spontaneous breathing. A 6 F nasogastric tube was inserted into the trachea via laryngoscopy Control: InSurE procedure
Outcomes	Primary and secondary outcomes not specified separately Outcomes Adverse events (reflux, asphyxia, bradycardia < 100 beats/min and apnoea) during procedure Changes in FiO₂, SpO₂, and blood pressure every minute during procedure Changes in PaO₂ and PaCO₂ at 1 hour post treatment Subsequent doses of surfactant Mechanical ventilation within 72 hours Pneumothorax BPD NEC ROP > stage 2 IVH > grade III Sepsis Death
Notes	This study was conducted in an upper‐middle‐income country (China). It included relatively mature infants (mean gestational age 33.7 (SD 1.0) to 34.1 (SD 1.3) weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Patients were randmised according to 'numbered list': odd number – LISA group, even number ‐ InSurE group (verbal communication with study authors)
Allocation concealment (selection bias)	Low risk	Opaque sealed envelopes were used
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Data collector (research assistant) and statistician did not participate in patient management. The medical team who administered surfactant did not participate in patient management thereafter
Blinding of outcome assessment (detection bias) Short‐term outcomes	Low risk	Data collector (research assistant) and statistician did not participate in patient management. The medical team who administered surfactant did not participate in patient management thereafter
Blinding of outcome assessment (detection bias) Long‐term outcomes	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Short‐term outcomes	Low risk	All infants recruited were accounted for
Incomplete outcome data (attrition bias) Long‐term outcomes	Unclear risk	Not applicable
Selective reporting (reporting bias)	Unclear risk	Trial was not registered with any trial registry
Other bias	Low risk	Intervention and control groups were similar in gestation, birth weight, and other background variables

AMV: Avoiding Mechanical Ventilation trial; APGAR: appearance, pulse, grimace, activity, and respiration; BPD: bronchopulmonary dysplasia; bpm: beats per minute; CONSORT: Consolidated Standards of Reporting Trials; CPAP: continuous positive airway pressure; EISA: endotracheal intubation surfactant administration; EOS: early‐onset sepsis; ETT: endotracheal tube; GA: gestational age; HR: heart rate; InSurE: Intubate, Surfactant, Extubate; IQR: interquartile range; IVH: intraventricular haemorrhage; LISA: less invasive surfactant administration; MAP: mean airway pressure; MDI: mental development index; MISA: minimally invasive surfactant administration; MIST: minimally invasive surfactant therapy; MV: mechanical ventilation; nCPAP: nasal continuous positive airway pressure; NEC: necrotising enterocolitis; NICU: neonatal intensive care unit; NINSAPP: Nonintubated Surfactant Application trials; NIPPV: nasal intermittent positive‐pressure ventilation; PDA: patent ductus arteriosus; PDI: psychomotor development index; PEEP: positive end‐expiratory pressure; RDS: respiratory distress syndrome; RITA: Randomization In Treatment Arms (RITA) programme; ROP: retinopathy of prematurity; SD: standard deviation; SAINT: Surfactant Administration by Insure or Thin Catheter; S‐ETT: surfactant administration via ETT; SurE: Surfactant, Extubate; SWI: surfactant‐without‐intubation.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Mirnia 2013b	Study participants are reported as part of another Iranian multi‐centre randomised trial (Mirnia 2013a
Oncel 2016	Single‐centre randomised controlled study comparing the effectiveness of nCPAP and NIPPV as initial respiratory support within the MIST approach in preterm infants with respiratory distress syndrome. Both study groups received MIST

MIST: minimally invasive surfactant therapy; nCPAP: nasal continuous positive airway pressure; NIPPV: nasal intermittent positive‐pressure ventilation.

Characteristics of ongoing studies [ordered by study ID]

ACTRN12611000916943.

Study name	OPTIMIST‐A trial: multi‐centre randomised controlled trial of minimally invasive surfactant therapy in preterm infants 25 to 28 weeks' gestation on continuous positive airway pressure
Methods	Multi‐centre parallel masked randomised controlled trial of minimally invasive surfactant therapy in preterm infants
Participants	Preterm infants at 25 to 29 weeks' gestation on continuous positive airway pressure
Interventions	Intervention: minimally invasive surfactant therapy ‐ delivery of exogenous surfactant to the lung via brief catheterisation of the trachea with a narrow bore vascular catheter (16 gauge Angiocath, Product No. 382259, Becton Dickinson, Sandy UT, USA). The catheter will be advanced through the vocal cords under direct vision via a laryngoscope, without the need for Magill’s forceps. Only a single dose will be given within the first 12 hours of life Control: continued treatment with nasal CPAP delivered by prongs or mask. This is the standard control treatment. After randomisation, infants will receive a sham treatment from a treatment team not engaged in clinical care (this will not involve removal of prongs or discontinuation of CPAP but will require setting up equipment, screening baby, repositioning baby, using suctioning equipment, and changing monitoring)
Outcomes	Primary outcome: death or physiological bronchopulmonary dysplasia at 36 weeks' postmenstrual age Secondary outcome: ≥ 1 of the following: mortality; pneumothorax; duration of respiratory support during first hospitalisation; bronchopulmonary dysplasia; duration of bradycardia and hypoxaemia during intervention; discomfort during intervention
Starting date	1 December 2011
Contact information	Prof Peter Dargaville; Department of Paediatrics, Royal Hobart Hospital, Liverpool St., Hobart, Tasmania 7000; Australia; Tel: +61 3 62228308; Email: peter.dargaville@dhhs.tas.gov.au
Notes	Recruitment status of this study is (quote) "active, recruiting" Registered with ANZCTR: ACTRN12611000916943, on 26 August 2011 Registered with ClinicalTrials.gov: NCT02140580, on 16 May 2014

ACTRN12611000917932.

Study name	OPTIMIST‐B trial: multi‐centre randomised controlled trial of minimally invasive surfactant therapy in preterm infants 29 to 32 weeks' gestation on continuous positive airway pressure
Methods	Multi‐centre parallel blinded randomised controlled trial of minimally invasive surfactant therapy in preterm infants
Participants	Preterm infants at 29 to 32 weeks' gestation on continuous positive airway pressure
Interventions	Intervention: minimally invasive surfactant therapy ‐ delivery of exogenous surfactant to the lung via brief catheterisation of the trachea with a narrow bore vascular catheter (16 gauge Angiocath, Product No. 382259, Becton Dickinson, Sandy, UT, USA). The catheter will be advanced through the vocal cords under direct vision via a laryngoscope, without the need for Magill’s forceps. Only a single dose will be given within the first 12 hours of life Control: continued treatment with nasal CPAP delivered by prongs or mask. This is the standard control treatment. After randomisation, infants will receive a sham treatment from a treatment team not engaged in clinical care (this will not involve removal of prongs or discontinuation of CPAP but will require setting up equipment, screening baby, repositioning baby, using suctioning equipment, and changing monitoring)
Outcomes	Primary outcome: duration of respiratory support during first hospitalisation Secondary outcome: ≥ 1 of the following during first hospitalisation: bronchopulmonary dysplasia, grade III or IV intraventricular haemorrhage, periventricular leukomalacia or retinopathy of prematurity > stage 2; pneumothorax and cost of hospitalisation assessed by computation of length of stay and level of acuity
Starting date	Recruitment not commenced
Contact information	Prof Peter Dargaville; Department of Paediatrics, Royal Hobart Hospital, Liverpool St., Hobart, Tasmania 7000; Australia; Tel: +61 3 62228308; Email:peter.dargaville@ths.tas.gov.au
Notes	Recruitment status of this study is (quote) "not recruiting" Registered with ANZCTR: ACTRN12611000917932, on 26 August 2011

ChiCTR1900020970.

Study name	LPPSA: less invasive surfactant administration versus endotracheal surfactant instillation followed by limited peak pressure ventilation in preterm infants with respiratory distress syndrome in China: study protocol for a randomized controlled trial
Methods	Multi‐centre randomised prospective trial; will be conducted at 14 tertiary NICUs in China from January 2019 to December 2020
Participants	Inclusion criteria Born between 250/7 and 316/7 weeks' gestation with birth weight between 600 grams and 1500 grams Exhibit vigorous spontaneous breathing and can be stabilised by non‐invasive respiratory support PEEP at 6 to 8 cm H₂O and in need of fraction of inspired oxygen (FiO₂) ≧ 0.3 < 6 hours of age Informed parental consent has been obtained Exclusion criteria Requiring MV or intubation in the delivery room Presence of major congenital malformation or chromosomal abnormality or inherited disorders of metabolism Neuromuscular disease affecting respiratory function Presence of congenital pneumonia or pulmonary hypoplasia
Interventions	To compare surfactant application via 2 techniques LISA: less invasive surfactant administration by thin catheter under direct vision by laryngoscopy. During the procedure, the infant will be on nCPAP LPPSA: endotracheal surfactant administration followed by low peak pressure
Outcomes	Primary outcomes: short‐term prognosis of respiration in patients treated with bronchoscopic MIST including Mortality, co‐morbidity, and severity of BPD at 36 weeks' corrected gestational age MV requirement in the first 72 hours of life Secondary outcomes Need for and duration of MV (days) during hospitalisation Duration of noninvasive respiratory support (days) Duration of oxygen need Heart rate and oxygen saturation before and after surfactant administration every 30 seconds for 10 minutes Incidence of pneumothorax and massive pulmonary haemorrhage within 48 hours of surfactant administration Severe neonatal disease including and haemodynamically significant patent ductus arteriosus (hsPDA) that needs medical or surgical intervention Duration of hospitalisation Failure rate of operation (including LISA or LPPSA technique) Time required to perform the procedure of surfactant administration Intraventricular haemorrhage (IVH, grade III to IV) Stage II to III necrotising enterocolitis (NEC) Retinopathy of prematurity (ROP; ≧ stage III) Postnatal steroid use
Starting date	January 2019
Contact information	Jiajun Zhu, Women’s Hospital, Zhejiang University, School of Medicine, Hangzhou 310006, China; Lizhong Du, The Children’s Hospital, Zhejiang University, School of Medicine, Hangzhou 310052, China
Notes	Funded by major scientific and technological projects of medicine and health in Zhejiang Province (WKJ‐ZJ‐2032) Recruitment status of this study is (quote) "recruiting" Registered with Chinese Clinical Trial Registry: ChiCTR1900020970, on 23 January 2019

NCT01615016.

Study name	MISurf: MISurF versus InSurE. A comparison of minimally invasive surfactant application techniques in preterm infants
Methods	Feasibility of a masked, prospective randomised controlled trial with 2 intervention arms
Participants	Eligible are all preterm infants born at ≤ 30 weeks' gestation at McMaster Inclusion criteria CPAP 5 to 6 cmH₂O and FiO₂ ≥ 0.35, or CPAP 7 to 8 cmH₂O and FiO₂ ≥ 0.30 < 36 hours of age Worsening clinical signs of RDS such as retractions (clinical judgement of the responsible physician) Exclusion criteria Previous Intubation or in imminent need of invasive mechanical ventilation because of e.g. apnoea, severe bradycardia, or other deterioration not attributed to RDS (e.g. shock) Congenital anomaly or conditions that might adversely affect breathing Pneumothorax before intervention No parental consent
Interventions	To compare surfactant application using 2 techniques MISurf: minimally invasive intratracheal surfactant application without mechanical ventilation by feeding tube device InSurE: surfactant application by InSurE strategy (Intubation ‐ surfactant ‐ extubation sequence)
Outcomes	Primary feasibility outcome: proportion of included infants who were treated according to protocol Secondary feasibility outcomes Recruitment rate Consent rate Proportion of intervention procedures in which masking has not been successful Proportion of interventions when intervention team has not arrived in time, leading to emergency intervention Success rate in antenatal approach for consent Primary clinical outcome: failure rate of the intervention, where failure is defined as Need for invasive ventilation Requiring either FiO₂ > 0.6 or pCO₂ > 65 mmHg and pH < 7.20, or both, for longer than 2 hours after surfactant administration up to 72 hours of life Intubation/requirement for mechanical ventilation within 48 hours after first intervention (same criteria as above) For InSurE: failed extubation within 15 minutes after intubation for surfactant application Serious adverse event (SAE) during immediate intervention leading to intubation (e.g. severe bradycardia/resuscitation, pneumothorax) Secondary clinical outcomes Proportion of infants not requiring the intervention Proportion of the following co‐morbidities until discharge: incidence of grade III and IV IVH, PVL (periventricular leukomalacia), ROP requiring treatment, NEC stage 2 and 3 Total duration of invasive and non‐invasive ventilation (extubation criteria will follow the extubation and weaning guidelines) Duration of oxygen supplementation until discharge Proportion of patients requiring oxygen supplementation at discharge Proportions of surfactant‐related adverse events such as tube blockade, episodes of desaturation, bradycardia, pulmonary haemorrhage, and pneumothorax differ in the 2 groups Total number of surfactant doses required compared in the 2 groups Incidence of CLD. CLD is assessed as per physiological CLD definition with severity score of mild, moderate, and severe Death
Starting date	July 2012
Contact information	Salhab el Helou,McMaster University, Children's Hospital/Hamilton Health Sciences
Notes	Recruitment status of this study is (quote) "terminated (delay due to infrastructure reasons. Funding withdrawn)" Registered with ClinicalTrials.gov: NCT01615016, on 8 June 2012

NCT01848262.

Study name	ECALMIST: ECALMIST versus InSurE in preterm infant < 32 weeks, multi‐centre, multi‐national RCT
Methods	Prospective open‐label randomised clinical trial
Participants	Newborn at < 32 weeks' gestation at birth Inclusion criteria Newborn at < 32 weeks' gestation at birth Postnatal age < 24 hours of life Clinical diagnosis of RDS Spontaneously breathing on CPAP Clinical decision to give surfactant Exclusion criteria Lack of parental consent Need for mechanical ventilation Major congenital malformation
Interventions	Minimally invasive surfactant therapy via a small vascular catheter ‐ ECALMIST (Early CPAP And Large Volume Minimal Invasive Surfactant Therapy) versus InSurE in preterm infants with RDS Procedure: ECALMIST surfactant administration via 5 French gauge, 133‐mm‐length vascular catheter. 5 mL/kg will be drawn up in a 5‐ or 10‐mL syringe. The vascular catheter will be inserted through the vocal cords under direct vision via a standard laryngoscope with appropriate blade for gestational age. The procedure will be done without removal of the CPAP. A bolus of surfactant of 0.25 to 0.5 mL will be administered after the surfactant is observed moving up and down as an indication of accurate intubation of the trachea. The surfactant will be slowly injected by small pulses of 0.25 to 0.5 mL over 20 to 30 seconds, with each bolus 10 seconds apart. At the end of the procedure, the operator will flush the catheter with 0.5 mL of air before removing the catheter. Other name: CPAP and minimal invasive surfactant therapy Procedure: InSurE patients who will receive surfactant via this technique will have their CPAP removed and then will be orally intubated with a standard ETT via standard endotracheal intubation procedures with ETT of appropriate size according to birth weight. Bovine surfactant will be administered through the ETT via the same technique described above for Arm 1. Manual lung inflation with a Jackson‐Rees anaesthesia bag at 20/5‐cm H₂O pressure will be performed during surfactant instillation, after which the patient will be extubated promptly as per the discretion of the neonatal team. Immediately following extubation, CPAP support will be re‐commenced. No pre‐medication such as sedation or atropine will be used during either procedure. Other name: intubate surfactant extubate
Outcomes	Primary outcome: incidence of early ventilation hours (time frame: 3 days) and number of newborn infants needing ventilation in the first 3 days of life Secondary outcomes: apnoea; bradycardia; desaturation; total ventilation hours; incidence of CLD; early ventilation hours; hospital stay.
Starting date	June 2013
Contact information	Yahya Al Ethawi, University of Manitoba.
Notes	Recruitment status of this study is (quote) "unknown". Completion date has passed and status has not been verified in more than 2 years Registered with ClinicalTrials.gov: NCT01848262, on 7 May 2013

NCT02772081.

Study name	LISPAP: a randomized, controlled study in preterm neonates with RDS to compare two procedures for porcine surfactant (Poractant Alfa, CUROSURF®) administration: a less invasive method (LISA) and the conventional administration
Methods	Multi‐centre randomised, open‐label, parallel‐assignment, controlled study ‐ phase 3
Participants	Inclusion criteria Written informed consent obtained by parents/legal representative (according to local regulation) before or after birth Inborn preterm neonates of either sex aged ≥ 30 minutes and < 24 hours, spontaneously breathing and stabilised on non‐invasive ventilation (NIV) Gestational age of 25 + 0 weeks up to 28 + 6 completed weeks, except for the first 30 enrolled neonates, in which gestational age will be restricted to 27 + 0 weeks up to 28 + 6 weeks Clinical course consistent with RDS Fraction of inspired oxygen (FiO₂) ≥ 0.30 to maintain SpO₂ between 88% and 95% Exclusion criteria Need for immediate endotracheal intubation for cardiopulmonary resuscitation or insufficient respiratory drive Use of nasal high‐frequency oscillatory ventilation (nHFOV) before study entry Use of surfactant before study entry and need for intratracheal administration of any other treatment (e.g. nitric oxide) Known genetic or chromosomal disorders, major congenital anomalies (congenital heart disease, myelomeningocoele, etc) Mothers with prolonged rupture of the membranes (> 21 days' duration) Presence of air leaks if identified and known before study entry Evidence of severe birth asphyxia (e.g. Apgar score ≤ 5 at 10 minutes after birth, continued need for resuscitation at 10 minutes after birth, altered neurological state) Neonatal seizures before study entry Any condition that, in the opinion of the Investigator, would place the neonate at undue risk Participation in another clinical trial of any placebo, drug, or biological substance conducted under the provisions of a protocol
Interventions	To compare surfactant application via 2 techniques Experimental ‐ Curosurf LISA: single dose of poractant alfa 200 mg/kg via brief catheterisation of the trachea with a thin catheter (CHF 6440) in neonates with RDS intervention: drug: LISA combination product (Curosurf+catheter CHF6440) Active comparator ‐ Curosurf endotracheal tube: single dose of poractant alfa 200 mg/kg via conventional intubation with endotracheal tube in neonates with RDS intervention: drug: Curosurf through conventional administration (endotracheal tube)
Outcomes	Numbers of neonates with surfactant‐ and procedure‐related adverse events (time frame: from application of the laryngoscope up to removal of the CHF 6440 catheter or the endotracheal tube within 1 hour after instillation of poractant alfa, device occlusion, apnoea, neonatal oxygen desaturation, bradycardia, hypotension requiring treatment, cough, sneezing, choking, laryngospasm, surfactant regurgitation, vomiting)
Starting date	June 2018
Contact information	Gianluca Lista, Buzzi Hospital, Milan
Notes	Sponsored by Chiesi Farmaceutici S.p.A. Recruitment status of this study is (quote) "suspended (sponsor decision)" Registered with ClinicalTrials.gov: NCT02772081, on 12 May 2019

NCT03989960.

Study name	MOLISAN: modified intubation‐surfactant‐extubation (InSurE) technique in preterm neonates with respiratory distress syndrome
Methods	Parallel‐assignment, non‐randomised, double‐blind (participant, outcomes assessor) study
Participants	Inclusion criteria Preterm infants with birth weight < 2500 grams and gestational age < 36 + 6 weeks High‐risk preterm infants with early symptoms of RDS, or infants with clinical diagnosis of RDS Participating hospital obtained consent of the Ethics Committee Parental informed consents were obtained Exclusion criteria Severe congenital malformations Severe cyanotic congenital heart disease (such as transposition of great artery, tetralogy of Fallot, etc.) that affects systemic haemodynamics Congenital hereditary metabolic disease Parental informed consent not obtained
Interventions	To compare surfactant application via 2 techniques Less invasive surfactant administration (LISA) combined with synchronised nasal intermittent positive‐pressure ventilation (SNIPPV) technique (LISA + SNIPPV group): this group receives pulmonary surfactant by way of SNLISA followed by nasal SNIPPV Traditional InSurE: traditional InSurE group receives intubation‐surfactant‐extubation technique and selects CPAP ventilation
Outcomes	Primary outcomes Average duration of mechanical ventilation (time frame: 40 weeks): average duration of mechanical ventilation for each group Duration of oxygen therapy (time frame: 40 weeks): duration of oxygen therapy for each group Incidence of BPD (time frame: 28 days): incidence of BPD in each group Secondary outcomes Pulmonary severity score (PSC) (time frame: first, second, third, seventh, 14th, 28th days): PSC was defined as FiO₂ × support + medications, where FiO₂ is the actual or 'effective' (for nasal cannula) FiO₂; support is 2.5 for a ventilator, 1.5 for nasal continuous positive airway pressure, or 1.0 for nasal cannula or hood oxygen; and medication is 0.20 for systemic steroids for BPD, 0.10 each for regular diuretics or inhaled steroids, and 0.05 each for methylxanthines or intermittent diuretics. Scores can range from 0.21 to 2.95 Incidence of complications (time frame: 40 weeks) (e.g. NEC, cholestasis, retinopathy of prematurity, extrauterine growth retardation) Oxygenation index and ventilation function (PaO₂, a/APO₂, FiO₂, PaCO₂) (time frame: period of oxygen therapy)
Starting date	1 August 2018
Contact information	Dr. Xiaoqing Chen, The First Affiliated Hospital with Nanjing Medical University, Nanjing, China, 210029
Notes	Recruitment status of this study is (quote) "active, not recruiting" Registered with ClinicalTrials.gov: NCT03989960, on 18 June 2019

NCT04016246.

Study name	PROLISA: propofol versus placebo (with rescue with ketamine) before less invasive surfactant administration: study protocol for a multicenter, double‐blind, placebo controlled trial
Methods	Multi‐centre, parallel‐blinded randomised controlled trial of minimally invasive surfactant therapy in preterm infants with vs without sedation with LISA
Participants	Newborn < 32 weeks' gestation at birth Inclusion criteria Presenting RDS in the first 48 hours of life and treated by non‐invasive ventilation requiring surfactant with the following FiO₂ to obtain a pulse oximetry (SpO₂) between 88% and 95% FiO₂ ≥ 30% for duration ≥ 10 minutes for infants born between 28 and 31 weeks' GA FIO₂ ≥ 25% for duration ≥ 10 minutes for infants born at < 28 weeks' GA Available intravenous line (peripheral, umbilical, or central catheter) Covered by French Social Security Signed parental informed consent Exclusion criteria Congenital and/or major malformations, including upper airway malformations FiO₂ > 60% at the time of inclusion Silverman score > 6 Contraindication to the use of propofol Low mean arterial blood pressure at 2 successive measurements (< gestational age expressed in weeks persisting after 1 volume expansion) Use of inotropic medication to maintain normal blood pressure Use of sedative or analgesic drugs (except paracetamol and ibuprofen) in the previous 24 hours Coma and/or seizures at neurological examination
Interventions	Intervention: infants were assigned to receive intravenous propofol (0.5 mg/kg) for sedation during the LISA procedure Control: infants were assigned to receive intravenous 20% medialipide (0.5 mg/kg) as a placebo during the LISA procedure In both groups, when the team is ready, a solution of atropine, caffeine, and oral sugar solution 30% or 24% is given orally via a syringe before the LISA procedure LISA procedure involves surfactant treatment during spontaneous breathing via an aspiration probe (CH6) or the LISAcath® (catheter for oral endotracheal instillation, CHIESI SAS, Bois Colombes, France). The type of probe is left to the choice of each investigator
Outcomes	Primary outcome: the need for MV from the LISA procedure onwards up to 72 hours of life indicated according to standardised criteria Repeated and severe apnoea (defined by American Academy of Pediatrics Guidelines) with bradycardia and/or low oxygen saturation High FiO₂ (according to local practices) justifying intubation for second administration of surfactant and MV Respiratory complications such as pneumothorax, with aggravation of the RDS Any other cause whereby the clinician judges MV to be necessary: the motive for respiratory assistance is indicated in the case report form (CRF). All intubations for MV are recorded Secondary outcomes Faceless Acute Neonatal Pain Scale (FANS) score is collected before, during, and after the LISA procedure in both groups (propofol and placebo) Ketamine administration (number of injections), dose for rescue (if FANS ≥ 6) is also recorded Per procedure events are recorded including the number of laryngoscopies needed to perform LISA, evolution of cardiorespiratory parameters from baseline to and 120 minutes after the first drug injection: heart rate, respiratory rate, pulse oximetry, blood pressure, FiO₂, ventilatory mode, inspiratory and end‐expiratory ventilation pressures, transcutaneous pCO₂, presence of apnoea requiring bag mask ventilation or additional nasal pressure with NIPPV and emergency intubation for severe apnoea after drug injection before the LISA procedure After the procedure: the clinician who successfully performed the LISA rates his/her performance during laryngoscopy (especially facility of exposure of the glottis) according to the Viby‐Mogensen scale. This score based on 5 items (scored from 0 to 4) explores the facility to expose the larynx and the infant’s behaviour Others items related to tolerance of the procedure such as cardiorespiratory and neurological parameters are collected at 24 hours and at 72 hours after the intervention In‐hospital neonatal morbidity and mortality: pneumothorax, necrotising enterocolitis according to Bell stage, proven early‐ and late‐onset sepsis (respectively defined as positive CSF or blood culture before < and after > 72 hours of life), retinopathy of prematurity according to international classification, cystic periventricular leukomalacia or grade III or IV intraventricular haemorrhage according to the Papile classification, surgical treatment of patent ductus arteriosus, duration of cumulated MV, duration of cumulated non‐invasive ventilation, any intubation, death at 36 weeks' GA, and in‐hospital mortality Data at 2 years of age will be collected: standard paediatric examination, age and stage questionnaire (ASQ) completed by parents, Gross Motor Function Classification Score in cases of motor impairment, any visual and/or hearing disabilities detected during the first 2 years of life
Starting date	7 October 2019
Contact information	Marie Chevallier, UMR 5525 ThEMAS, CNRS, TIMC‐IMAG, Grenoble Alps University, Grenoble, France, and Neonatal Intensive Care Unit, Grenoble Alps University Hospital, Grenoble, France. Email: mchevallier3@chu-grenoble.fr
Notes	Recruitment status of this study is (quote) "recruiting" Registered with ClinicalTrials.gov: NCT04016246, on 06 June 2019, N°EUDRACT: 2018–002876‐41

NCT04073173.

Study name	StrAAS: stress assessment in preterm infants with respiratory distress syndrome treated or not with an analgesic drug during traditional or less invasive method of surfactant therapy
Methods	Single‐centre parallel open‐label randomised controlled trial. Planning to recruit 80 participants
Participants	Inclusion criteria Gestational age at birth between 168 and 223 days Respiratory distress syndrome (diagnosed on the basis of clinical and/or radiological grounds) with fraction of inspired oxygen ≥ 0.30 (for infants born at ≤ 26 weeks' gestational age) or ≥ 0.40 (for infants born at > 26 weeks' gestational age) to achieve peripheral oxygen saturation of 90% to 94% within 24 hours of life and good respiratory drive Written informed consent Exclusion criteria Major malformations Late admission (after 24 hours of life) Intubation in the delivery room Severe birth asphyxia Prolonged rupture of membranes Air leaks No informed consent
Interventions	Procedure: INSURE ‐ patients will be intubated by endotracheal tube; exogenous surfactant (Poractant alfa) will be administered, and then they will be extubated Procedure: LISA ‐ surfactant (Poractant alfa) will be directly delivered into the lungs via a fine bore catheter inserted into the trachea, and then patients will be extubated Drug: analgesic, opioid ‐ remifentanil (0.5 to 2 micrograms/kg/dose) Study arms Experimental: LISA‐analgesic ‐ less invasive surfactant administration (LISA) with remifentanil (0.5 to 2 micrograms/kg/dose) as the analgesic drug Experimental: LISA‐no analgesic ‐ less invasive surfactant administration (LISA) without an analgesic drug Experimental: INSURE‐analgesic ‐ INtubation‐SURfactant‐Extubation (INSURE) with remifentanil (0.5 to 2 micrograms/kg/dose) as the analgesic drug Experimental: INSURE‐no analgesic ‐ INSURE without an analgesic drug
Outcomes	Primary outcome: the need for MV from the LISA procedure onwards up to 72 hours of life indicated according to standardised criteria Cortisol concentrations (time frame: at 1, 3, 6, 12, 24 hours after surfactant administration and then daily in the first week at the same time of the day (to avoid circadian variations)) will be assessed in saliva, as salivary cortisol levels. Saliva samples will be collected via an absorbent swab stick, centrifuged at 4000 rpm for 10 minutes, and kept at ‐80°C until assayed (minimum sample volume 25 μL). Enzymeimmunoassay (ELISA kit) will be used. Basal samples will be obtained at hospital admission and right before surfactant Secondary outcomes Galvanic skin responses (time frame: at 1, 3, 6, 12, 24 hours after surfactant administration, then daily in the first week at the same time of day (to avoid circadian variations)). An instrumental stress‐test device measuring galvanic skin conductance will be used (Pain Monitor, Med‐Storm, Norway): 3 electrodes will be attached to the infant's foot (sole and sides of the ankle); skin conductance is measured in micro Siemens (μS) Heart rate (time frame: 6 hours before and after surfactant therapy will be analysed). Cardiac monitoring will assess heart rate. Traces will be saved onto a computer with a sampling frequency of 1 Hertz. Average heart rate, periods of tachycardia (> 160 bpm for ≥ 5 seconds) and bradycardia (< 100 bpm for ≥ 5 seconds) will be recorded. These parameters may be correlated with stress and haemodynamic instability during procedures Brain oxygenation (time frame: from hospital admission to Day 7 of hospital stay). Brain oxygenation will be assessed by near‐infrared spectroscopy (NIRS) Oxygen saturation (SpO₂) (time frame: from hospital admission to Day 7 of hospital stay). High‐precision oxygenation assessment will be attained by high‐frequency (1 Hz) sampling of SpO₂ data from the cardio monitor to a computer, possibly by using multiple pulse oximeters in the same patient Markers of oxidative stress (time frame: at hospital admission and at 6 and 12 hours after surfactant therapy). 8‐isoprostane and nitrites/nitrates will be dosed on urine samples
Starting date	1 November 2020.
Contact information	Virgilio Carnielli; Azienda Ospedaliero‐Universitaria Ospedali Riuniti di Ancona; Tel: +390715962045; Email: v.carnielli@staff.univpm.it Clementina Rondina; Azienda Ospedaliero‐Universitaria Ospedali Riuniti di Ancona; Tel +390715962014; Email clementina.rondina@ospedaliriuniti.marche.it
Notes	Recruitment status of this study is (quote) "active, not recruiting" Registered with ClinicalTrials.gov: NCT04073173, on 29 August 2019

NCT04445571.

Study name	SAINT: Surfactant Administration by Insure or Thin Catheter
Methods	Single‐centre parallel open‐label randomised controlled trial. Planning to recruit 160 participants
Participants	Inclusion criteria Infants born before 32 completed weeks of gestation on CPAP, with clinical and radiological signs of RDS and need for surfactant treatment Exclusion criteria Infants requiring surfactant as part of delivery room resuscitation are not eligible Infants will be excluded from the final analysis if they have a congenital abnormality or condition that might have an adverse effect on breathing or ventilation, including congenital diaphragmatic hernia; tracheo‐oesophageal fistula; or cyanotic heart disease
Interventions	Procedure: INSURE ‐ surfactant administration by Intubation‐surfactant‐extubation to CPAP according to standard protocol including pre‐medication with analgesia and sedation Procedure: LISA ‐ surfactant administration by thin catheter during spontaneous breathing and continued CPAP according to set protocol including pre‐medication with analgesia
Outcomes	Primary outcomes Oxygenation [Time Frame: 24 hours post procedure ]: arterial to alveolar ratio (a/A ratio) Mechanical ventilation [Time Frame: 48 hours post procedure]: need for intubation and mechanical ventilation (MV) Secondary outcomes Duration of ventilatory support [Time Frame: discharge]: duration of MV (hours), CPAP (days), oxygen (days) Complications [Time Frame: discharge]: incidence of air leaks, bronchopulmonary dysplasia, systemic hypotension, retinopathy, necrotising enterocolitis, intraventricular haemorrhage, persistent duct Mortality [Time Frame: discharge]: death or composite outcome death/BPD Length of stay [Time Frame: discharge]: number of days in NICU and total in neonatal care, including home care
Starting date	15 October 2020
Contact information	Kajsa Bohlin, Karolinska University Hospital, Stockholm, Sweden, 14186, Tel 0858580000 ext 81356; Email: kajsa.bohlin@ki.se Mats Blennow,Karolinska University Hospital, Stockholm, Sweden, 14186, Tel 0858580000 ext 81428; Email: mats.blennow@ki.se
Notes	Recruitment status of this study is (quote) "recruiting" Registered with ClinicalTrials.gov: NCT04445571, on 24 June 2020

UMIN000021785.

Study name	Effectiveness of MIST (minimally invasive surfactant therapy) under bronchoscopy in treating neonatal respiratory distress syndrome
Methods	Single‐arm, non‐randomised, open‐label study
Participants	Inclusion criteria: patients fulfilling all of the following criteria will be included Born at Tokyo Women's Medical University Medical Centre East, and admitted to the NICU Diagnosed with respiratory distress syndrome Informed consent obtained from parents Thought to be appropriate for the study by the attending physician Exclusion criteria: patients fulfilling any of the following criteria will be excluded Severe birth asphyxia Multiple malformation syndrome, or other suspected underlying disease Needing vasoactive drugs Tendency to bleed Congenital infection Thought to be inappropriate for the study by the supervising physician
Interventions	To compare surfactant application using 2 techniques MIST (minimally invasive surfactant therapy) via a thin bronchoscope inserted into the trachea without the need for endotracheal intubation or use of a laryngoscope Historical control
Outcomes	Primary outcomes: short‐term prognosis of respiration in patients treated with bronchoscopic MIST including Days under mechanical ventilation Days under supplemental oxygen Incidence of airway disease Need for endotracheal intubation or additional surfactant Secondary outcomes Morbidity of disease often seen in preterm infants (intraventricular haemorrhage, pneumothorax, patent ductus arteriosus, bronchopulmonary dysplasia, sepsis, NEC) Length of hospitalisation, mortality rate
Starting date	January 2016
Contact information	Masanori Wasa, Tokyo Women's Medical University Medical Centre East, 2‐1‐10 Nishiogu, Arakawaku, Tokyo
Notes	Funded by Tokyo Women's Medical University Medical Centre East Recruitment status of this study is (quote) "recruiting" Registered with UMIN‐CTR Clinical Trial (Japan): UMIN000021785, on 5 April 2016

ASQ: age and stage questionnaire; BPD: bronchopulmonary dysplasia; bpm: beats per minute; CLD: chronic lung disease; CPAP: continuous positive airway pressure; CRF: case report form; CSF: cerebrospinal fluid; ECALMIST: Early CPAP And Large Volume Minimal Invasive Surfactant Therapy;ETT: endotracheal tube; FANS: Faceless Acute Neonatal Pain Scale; GA: gestational age; hsPDA: haemodynamically significant patent ductus arteriosus;InSurE: intubate, surfactant, extubate; IVH: intraventricular haemorrhage; LISA: less invasive surfactant administration; LPPSA: A Randomized, Controlled Study in Preterm Neonates With RDS to Compare Two Procedures for PorcineSurfactant (Poractant Alfa, CUROSURF®) Administration: A Less Invasive Method (LISA) and ConventionalAdministration; MIST: minimally invasive surfactant therapy; MISurF: minimally invasive surfactant; MV: mechanical ventilation; nCPAP: nasal continuous positive airway pressure;NEC: necrotising enterocolitis; nHFOV: nasal high‐frequency oscillatory ventilation; NICU: neonatal intensive care unit; NIPPV: nasal intermittent positive‐pressure ventilation; NIRS: near‐infrared spectroscopy; NIV: non‐invasive ventilation; PEEP: positive end‐expiratory pressure; PSC: pulmonary severity score; SAE: serious adverse event; SNIPPV: synchronised nasal intermittent positive‐pressure ventilation technique; StrAAS: Stress Assessment in Preterm Infants With Respiratory Distress Syndrome Treated or Not With an Analgesic Drug During the Traditional or the Less Invasive Method of Surfactant Therapy.

Differences between protocol and review

1. As of July 2019, Cochrane Neonatal no longer searches Embase for its reviews. RCTs and controlled clinical trials (CCTs) from Embase are added to the Cochrane Central Register of Controlled Trials (CENTRAL) via a robust process (see www.cochranelibrary.com/central/central-creation). Cochrane Neonatal has validated its searches to ensure that relevant Embase records are found while CENTRAL is searched.

2. Also starting in July 2019, Cochrane Neonatal no longer searches for RCTs and CCTs from ClinicalTrials.gov nor from the International Clinical Trials Registry Platform (ICTRP), as records from both platforms are added to CENTRAL on a monthly basis (see www.cochranelibrary.com/central/central-creation). Comprehensive search strategies are executed in CENTRAL to retrieve relevant records. The ISRCTN (at http://www.isrctn.com/, formerly Controlled‐trials.com) is searched separately.

3. We modified the title of the Review from ‘Surfactant therapy via brief tracheal catheterisation in preterm infants with or at risk of respiratory distress syndrome’ to ‘Surfactant therapy via thin catheter in preterm infants with or at risk of respiratory distress syndrome’, and changed ‘surfactant via brief tracheal catheterisation’ to ‘surfactant via thin catheter’ in the text.

4. We added the methods and the plan for 'Summary of findings' tables and GRADE recommendations, which were not included in the original protocol (Wheeler 2015).

5. We also added the following outcomes, which were not included in the original protocol.

   1. Catheter/ETT placement unsuccessful at first attempt (during trial‐related intervention).

   2. Incidence of the need for mechanical ventilation within first 72 hours, or not ventilated but reached failure criteria.

   3. Incidence of spontaneous intestinal perforation..

6. We added a definition for neurosensory disability.

7. We added six new outcomes to the primary outcome section (now a total of seven primary outcomes). These added outcomes were listed as secondary outcomes in the protocol (Wheeler 2015).

   1. Need for intubation within the first 72 hours of life.

   2. Air leak requiring drainage (during first hospitalisation).

   3. Severe intraventricular haemorrhage (IVH), including grades III and IV (Papile 1978).

   4. Death during first hospitalisation (all causes).

   5. BPD (clinical definition) among survivors to 36 weeks' PMA.

   6. Death or survival with neurosensory disability, with the latter measured beyond one year PMA and defined as any of (i) cerebral palsy by clinical examination or other means; (ii) developmental delay more than two standard deviations below the population mean on standardised testing; (iii) blindness (visual acuity less than 6/60); or (iv) deafness (hearing impairment requiring amplification).

8. A number of outcomes in the protocol ‐ Wheeler 2015 ‐ were not included in the review, largely based on anticipated infrequency of reporting. These were:

   1. incidence of discontinuation of the intervention;

   2. number of surfactant doses post intervention (need for intratracheal surfactant therapy post intervention remains);

   3. incidence of dosing failure;

   4. incidence of the need for early intubation (within one hour of surfactant administration);

   5. use of diuretic therapy as prophylactic or rescue treatment;

   6. incidence of death in the first 28 days (in this case, a single death outcome was opted for);

   7. incidence of PDA requiring surgical therapy;

   8. major morbidity and death, or major morbidity (here, the concern was the variability of definitions for this outcome);

   9. time to regain birth weight; and

   10. number of hospital re‐admissions in the first year (all).

9. One of the sub‐group analyses proposed in the protocol ‐ Wheeler 2015 ‐ was omitted.

   1. Surfactant type (animal‐derived, synthetic).

Contributions of authors

The protocol ‐ Wheeler 2015 ‐ was developed by KIW and PAD. All review authors provided feedback on the content of the protocol (Wheeler 2015).

The review manuscript was developed in RevMan 5 by MEA. MEA and PAD independently performed electronic database searches, assessed the certainty of evidence, and synthesised the evidence. PGD also provided methodological input. All review authors provided feedback on the content of the draft and the final manuscript.

Sources of support

Internal sources

• No sources of support provided

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

MEA has no interests to declare.

KIW has no interests to declare.

PGD has no interests to declare.

AGDP has no interests to declare.

PAD is the Chief Investigator of the OPTIMIST‐A trial, a multi‐centre RCT of surfactant via tracheal catheterisation in preterm infants on CPAP (ACTRN12611000916943). Chiesi Farmaceutici (Parma, Italy) is providing in‐kind support for this trial by providing surfactant at reduced cost for the OPTIMIST‐A trial. Dr. Dargaville has served as a consultant for Chiesi Farmaceutici and AbbVie Inc. Neither company is involved with the protocol, analysis, manuscript preparation, or publication processes of this review. The Australian National Health and Medical Research Council (NHMRC) has awarded a project grant (#1049114) for conduct of an RCT of minimally invasive surfactant therapy in preterm infants on CPAP, for which PAD is the Chief Investigator.

Edited (no change to conclusions)