Article Text
PDF
Abstract
Background There are no evidence-based recommendations for surfactant use in late preterm (LPT) and term infants with respiratory distress syndrome (RDS).
Objective To investigate the safety and efficacy of surfactant in LPT and term infants with RDS.
Methods Systematic review, meta-analysis and evidence grading.
Interventions Surfactant therapy versus standard of care.
Main outcome measures Mortality and requirement for invasive mechanical ventilation (IMV).
Results Of the 7970 titles and abstracts screened, 17 studies (16 observational studies and 1 randomised controlled trial (RCT)) were included. Of the LPT and term neonates with RDS, 46% (95% CI 40% to 51%) were treated with surfactant. We found moderate certainty of evidence (CoE) from observational studies evaluating infants supported with non-invasive respiratory support (NRS) or IMV that surfactant use may be associated with a decreased risk of mortality (OR 0.45, 95% CI 0.32 to 0.64). Very low CoE from observational trials in which surfactant was administered at FiO2 >0.30–0.40 to infants on Continuous Positive Airway Pressure (CPAP) indicated that surfactant did not decrease the risk of IMV (OR 1.20, 95% CI 0.40 to 3.56). Very low to low CoE from the RCT and observational trials showed that surfactant use was associated with a significant decrease in risk of air leak, persistent pulmonary hypertension of the newborn (PPHN), duration of IMV, NRS and hospital stay.
Conclusions Current evidence base on surfactant therapy in LPT and term infants with RDS indicates a potentially decreased risk of mortality, air leak, PPHN and duration of respiratory support. In view of the low to very low CoE and widely varying thresholds for deciding on surfactant replacement in the included studies, further trials are needed.
- neonatology
- respiratory medicine
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
What is already known on this topic?
8%–10% of neonates of gestational age ≥340/7 weeks are diagnosed with respiratory distress syndrome (RDS).
Uncertainty exists on the risks and benefits of surfactant use in these infants.
What this study adds?
Of the late preterm and term neonates who are diagnosed with RDS, 46% -are treated with surfactant.
Surfactant therapy in neonates of gestational age ≥340/7 weeks may be associated with decreased risk of mortality and short-term respiratory morbidities.
There is wide variation in the thresholds used for deciding on surfactant therapy in these neonates, indicating the need for future trials.
Introduction
The management of respiratory distress syndrome (RDS) has evolved considerably over the past decades and exogenous surfactant replacement is one of its particular cornerstones.1 National and international guidelines recommend early surfactant therapy for very preterm neonates (less than 32 weeks’ gestation) at defined cut-off levels of fraction of inspired oxygen (FiO2).2 3 While the incidence of RDS in the more mature late preterm (LPT) and term neonates is only 8%–10%, the absolute number affected by RDS is significant.4–6
Most of the recommendations for surfactant use in infants with RDS are derived from evidence gained from those born <32 weeks’ gestation.3 However, extrapolating the treatment criteria and also the expected good outcomes from the surfactant-deficient group of very low birthweight infants (VLBWI) to the more mature groups of LPT and term infants with RDS is contentious. To date, off-label use of surfactant in LPT and term infants is common despite a paucity of evidence for these groups.7 Likewise, the safety profile and cost-effectiveness of surfactant replacement therapy are also almost exclusively researched in VLBWI, making risk–benefit assessments in the LPT and term infants difficult to adjudge.8 9 This systematic review and meta-analysis investigates the safety and efficacy of surfactant use in LPT and term infants with RDS to guide clinical practice decisions and highlight gaps in knowledge.
Methods
The review protocol was registered with PROSPERO (International Prospective Register of Systematic Reviews; CRD42021255292),10 and our report is written in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidance.11
Literature search
We searched five databases, Medline, Embase, Web of Science, Cochrane Central Register of Controlled Trials (CENTRAL) and Cumulative Index to Nursing and Allied Health Literature (CINAHL), from inception to 5 March 2021. The reference list of the included studies was also searched. Both English and non-English literature were reviewed. Randomised controlled trials (RCTs), observational studies and conference abstracts were eligible for inclusion. The search strategy is provided in online supplemental file 1.
Supplemental material
Inclusion criteria
Population: Neonates born at ≥34 weeks’ gestation and diagnosed with RDS. Neonates with differing severity of RDS including those on non-invasive respiratory support (NRS) or invasive mechanical ventilation (IMV) were eligible for inclusion.
Intervention: Surfactant which could have been administered by any modality.
Comparator: Standard management comprising NRS and/or IMV with or without placebo.
Primary outcomes: Mortality before discharge and requirement for IMV.
Secondary outcomes: Proportion of LPT or term infants with RDS who were treated with surfactant, air leak, pulmonary haemorrhage, persistent pulmonary hypertension of the newborn (PPHN), ventilator-associated pneumonia (VAP), duration of IMV/NRS/oxygen support/hospital stay, arterial blood gas parameters (pH, a:AO2 (ratio of arterial to alveolar partial pressure of oxygen), PaO2 (arterial partial pressure of oxygen), PaCO2 (arterial partial pressure of carbon oxide), PaO2/FiO2, oxygenation index (OI)) and FiO2 before and after surfactant therapy at different time points.
Data extraction and synthesis
Two authors (TA, TB) extracted the data independently in duplicates. A random-effects meta-analysis was done. Heterogeneity was evaluated based on I2, τ2 and Cochran Q. Effect estimates were expressed as Odds Ratio (OR) with 95% Confidence Interval (CI) or mean difference (MD) with 95% CI.
Risk of bias
The risk of bias in RCTs was evaluated using Cochrane Risk of Bias Tool V.2.0.12 Risk Of Bias In Non-randomized Studies - of Interventions (ROBINS-I) tool and modified Quality in Prognostic Studies (QUIPS) scale were used to evaluate risk of bias in observational studies and studies included in proportion-based meta-analysis, respectively.13 14 Certainty of evidence (CoE) for all the outcomes was evaluated as per the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) working group recommendations.15 Risk of bias and CoE assessments were done by two authors independently (TA, TB) and disagreements were resolved by discussing with a third author (VVR).
Results
Of the 7970 titles and abstracts screened and 165 full texts evaluated, 17 studies were included in the final meta-analysis.16–32 Of the 17 studies, 1 was a three-armed RCT including 175 infants27 and the rest were observational studies. While clinical outcomes were assessed by four observational studies enrolling 1775 infants,17 22 25 32 changes in blood gas indices and FiO2 after surfactant therapy were assessed in nine studies enrolling 1707 neonates. The proportion of infants with RDS treated with surfactant was assessed in nine studies including 45 080 LPT and term infants.17 18 20 21 24 25 28 31 32 The PRISMA flow is depicted in figure 1.
Literature search flow. WOS: Web of Science, CENTRAL: Cochrane Central Register of Controlled Trials, CINAHL: Cumulative Index to Nursing and Allied Health Literature.
Characteristics of included studies
Of the 16 observational studies, only 2 had a prospective design.22 28 While eight studies had evaluated LPT and term infants with RDS exclusively, others had analysed them as subgroups. The baseline characteristics of included neonates were variable across studies. Of the three studies clearly stating the FiO2 threshold used for surfactant administration, two had used a threshold of >0.3022 29 and one >0.60–0.70.23 Four of the 16 observational studies had enrolled infants with severe RDS requiring IMV.19 20 24 32 The only three-armed RCT had enrolled infants who required IMV with an OI of 10–20.27 The efficacy of two different dose regimens of calf-derived surfactant was evaluated in this trial. The characteristics of the included studies are listed in online supplemental table 1.
Primary outcomes
Mortality
Low CoE from the RCT enrolling LPT and term infants with severe RDS requiring IMV with an OI of 10–20 did not show any difference in the risk of mortality between the groups (OR 0.51, 95% CI 0.23 to 1.13; I2=0%, τ2=0%).27 Moderate CoE from three observational studies which had enrolled LPT and term infants with RDS and who were managed with NRS or IMV indicated that surfactant use might be associated with a decreased risk of mortality (OR 0.45, 95% CI 0.32 to 0.64) (figure 2A). Although the CoE from the observational studies was downrated by one level for risk of bias, it was uprated by one level each for a large magnitude of the effect and plausible confounding for the critical outcome of mortality. ‘Plausible confounding’ was adjudged based on one study which had the highest weightage in the meta-analysis, in which surfactant was used in a sicker subgroup of neonates with RDS and was shown to have decreased mortality despite this sicker group being at higher risk of death.25 The adjusted OR was not reported in the included studies and hence the effect estimate for this outcome was derived from raw data.
Forest plot depicting the effect estimates for the outcomes: (A) mortality from observational studies and (B) proportion of late preterm and term infants with respiratory distress syndrome treated with surfactant. IV, inverse variance
Requirement for IMV
Very low CoE from two observational trials which had administered surfactant at an FiO2 threshold of >0.30–0.40 had indicated that surfactant administration in LPT infants might not decrease the risk of requirement for IMV (OR 1.20, 95% CI 0.40 to 3.56; I2=0%, τ2=0%).17 22
Secondary outcomes
Proportion of LPT and term infants with RDS treated with surfactant
Of the LPT and term neonates who were diagnosed with RDS, 46% (95% CI 40% to 51%) were treated with surfactant. There was significant heterogeneity between the studies on this outcome (figure 2B).
Air leak, PPHN, VAP and pulmonary haemorrhage
Low CoE from the RCT showed that surfactant use was associated with a significant decrease in risk of air leak (OR 0.28, 95% CI 0.13 to 0.60; I2=0%, τ2=0%) and PPHN (OR 0.19, 95% CI 0.08 to 0.45; I2=0%, τ2=0%).27 Very low to low CoE showed that the risk of VAP (OR 0.54, 95% CI 0.20 to 1.47; I2=61%, τ2=0.32) and pulmonary haemorrhage (OR 0.84, 95% CI 0.25 to 2.86; I2=0%, τ2=0) was comparable between the groups.27 No differences in risk of air leak (OR 1.01, 95% CI 0.32 to 3.15), PPHN (OR 3.12, 95% CI 0.83 to 11.79) and pulmonary haemorrhage (OR 1.82, 95% CI 0.59 to 5.59) were seen in the observational study in which surfactant was administered while stabilised on continuous positive airway pressure (CPAP) (CoE: very low).32
Duration of IMV, NRS, oxygen support and hospital stay
Low CoE from the RCT enrolling neonates with severe RDS suggested that surfactant decreased the duration of IMV (MD −56.46 hours, 95% CI −100.56 to –12.37; I2=97%, τ2=983.13), NRS (MD −46.46 hours, 95% CI −53.26 to –39.67; I2=0%, τ2=0%) and hospital stay (MD −2.03 days, 95% CI −3.99 to –0.07; I2=76%, τ2=1.51).27
Very low CoE from the observational study indicated that surfactant administration at a lower FiO2 cut-off of >0.30 on CPAP decreased the duration of NRS (MD −23.60 hours, 95% CI −42.63 to −4.27) and oxygen support (MD −19.50 hours, 95% CI −31.76 to −7.24), but not the duration of hospital stay (MD −1.70 days, 95% CI −3.8 to 0.4).22
Surrogate outcomes of FiO2, indices of oxygenation and ventilation
Significant between-study heterogeneity was detected in the meta-analysis of these outcomes. These were explained on subgroup analyses based on the type of respiratory support at enrolment and the time points at which the outcomes were measured. The meta-analysis indicated that there was a significant decrease in the FiO2 requirement at 3–12 hours after surfactant therapy only in studies that had enrolled a homogenous population of infants who were either on NRS or IMV (online supplemental figure 1A). Also, a significant increase in PaO2 and a:AO2 and a decrease in PaCO2 were demonstrated post surfactant (online supplemental figures 1B and 2A,B). There was significant heterogeneity for the outcome of OI response after surfactant which could not be explained by subgroup or sensitivity analysis (I2=92%) (online supplemental figure 2C).
Risk of bias
The RCT was evaluated as having a high risk of overall bias due to ‘some concerns’ in four domains. Among them, randomisation process was rated as having ‘some concerns’ due to inadequate allocation concealment, for which no information was available and due to baseline characteristics not suggesting any imbalance among the three groups (online supplemental figure 3A). Concerning the observational studies included in the proportion-based meta-analysis, three18 24 28 had high and three17 20 21 had low risk of bias, while two31 32 were evaluated as having a moderate risk of bias (online supplemental figure 3B). For the cohort studies, ROBINS-I assessment indicated that eight studies had a moderate risk of bias16 17 19 20 22 26 30 32 and five were adjudged as having serious risk of bias.18 23 25 29 32 The domain ‘selective reporting’ could not be assessed for any of the included studies as none of them had registered protocols available for evaluation (online supplemental figure 4). The CoE for all the outcomes is given in online supplemental table 2.
Discussion
This systematic review and meta-analysis on the efficacy of surfactant in LPT and term infants with RDS included 17 studies. To the best of our knowledge, this is the only systematic review and meta-analysis that has exclusively evaluated the efficacy of surfactant in LPT and term infants with RDS.
Only a few studies addressed our research question and most were observational studies. The proportion-based meta-analysis revealed that 46% (40%–51%) of the LPT and term infants with RDS were treated with surfactant. The percentage appears high; however, we suspect that the lack of specific recommendations for surfactant use in these relatively mature infants on the one hand and possibly the enrolment of sicker infants are the reasons. From these studies, we assessed with moderate CoE that in LPT and term infants supported by NRS or IMV, surfactant therapy may be associated with a decreased risk of mortality. These findings match the outcomes of the largest cohort study to date by Jackson et al,21 who evaluated surfactant use in relatively mature preterm neonates with RDS. Their results indicated that surfactant use was associated with decreased odds of mortality in neonates >30 weeks’ gestation or 2000 g birth weight (adjusted OR 0.6, 95% CI 0.4 to 1.0).21 We could not pool this study in our meta-analysis as 24% of the enrolled infants were very preterm or moderately preterm infants. Due to significant differences in the baseline sickness of neonates and the threshold for surfactant administration in the included studies, it is difficult to infer with reasonable certainty which profile of LPT and term neonates with RDS would have a survival advantage with surfactant therapy.22 25 32
The evidence for using a lower FiO2 threshold of 0.30 for surfactant administration in very preterm infants, as suggested by the European consensus on RDS management (2019), is based on evidence from observational studies and an appreciation that early targeted surfactant treatment prevents IMV.3 33 Results from pooled analysis of two observational trials that had evaluated LPT neonates stabilised on CPAP with an FiO2 of 0.30–0.40 indicate that surfactant therapy might not decrease the risk of IMV.17 22 However, the CoE was very low, precluding any reasonable conclusions. To date, Verder et al 34 published the only RCT that had evaluated different thresholds of FiO2 for surfactant administration in preterm neonates of less than 30 weeks’ gestation with RDS. However, this trial was performed before widespread use of antenatal corticosteroid therapy for threatened preterm labour. Therefore, it is difficult to ascertain whether the results from this RCT, which suggest administering surfactant at FiO2 thresholds between 0.37 and 0.55, would still be applicable to present-day preterm infants with RDS. One large multicentric RCT which aims to study different thresholds of surfactant administration in LPT and early term infants with RDS stabilised on NRS is currently ongoing.35 Presently, in the absence of evidence from trials comparing surfactant therapy at different FiO2 cut-offs and for different modes of respiratory support, it remains to be speculated as to which are the ideal thresholds for surfactant therapy in LPT and term infants.
One of the important limitations of the existing literature is the wide variation in definitions used to differentiate RDS from other causes of respiratory distress such as transient tachypnea of newborn (TTNB), neonatal acute respiratory distress syndrome (ARDS), congenital pneumonia, etc, in this subgroup of neonates. Future studies may use imaging modalities such as lung ultrasound or lamellar body counts to clearly delineate RDS from other conditions such as TTNB.36–38 Neonatal ARDS has been increasingly recognised as a separate entity resulting in respiratory failure in neonates in the recent years. Due to absence of a uniform definition for neonatal ARDS prior to ‘the Montreux definition of neonatal ARDS’, it is plausible that many of the included studies had not differentiated this condition from primary RDS. Further research is warranted with respect to neonatal ARDS.36
This systematic review has several limitations. Except for the critical outcome of mortality, the CoE for all the other outcomes was very low to low. Also, parts of the outcomes reported by the included studies were surrogate markers of respiratory physiology which are not considered as critical clinical outcomes while contemplating clinical decisions. Further, sensitivity analyses that were planned a priori based on antenatal corticosteroid usage, type/dosage of surfactant used and the modality by which it was administered could not be performed due to paucity of data. Finally, subgroup analyses to explore the between-study heterogeneity for the various surrogate outcomes were done post-hoc, which is prone to selective reporting bias.
Conclusion
We conclude that a significant proportion of LPT and term infants with RDS are being treated with surfactant. Surfactant therapy in these infants might be associated with a decreased risk of mortality, air leak, PPHN and duration of respiratory support. In view of the widely heterogeneous group of infants with RDS being evaluated by the included studies, it is unclear as to what is the optimal window of surfactant administration. We suggest future large multicentric RCTs comparing different thresholds and different forms of respiratory support for surfactant administration in LPT and term infants with RDS. These studies may include additional clinical parameters and diagnostic modalities to differentiate RDS from other causes of respiratory distress in these neonates.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors CCR, EB and VVR conceptualised the systematic review. TB, TA and VVR did acquisition, analysis and interpretation of data. CCR and EB provided further intellectual input and revised the first draft. All authors approved the final version submitted for publication and agree to be accountable for all aspects of the work. VVR is the guarantor for this manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Linked Articles
- Fantoms