Key PointsQuestion
Which is the most appropriate postnatal corticosteroid regimen for preventing mortality or bronchopulmonary dysplasia (BPD) in preterm neonates at 36 weeks’ postmenstrual age?
Findings
In this systematic review and network meta-analysis of 62 studies with 5559 neonates, a moderately early-initiated, medium cumulative dose of systemic dexamethasone had the highest relative risk reduction of mortality or BPD among 14 regimens, with a low confidence in the evidence.
Meaning
Results of this study suggested that a moderately early-initiated, medium cumulative dose of systemic dexamethasone appeared to be the best regimen for preventing mortality or BPD at 36 weeks’ postmenstrual age.
Importance
The safety of postnatal corticosteroids used for prevention of bronchopulmonary dysplasia (BPD) in preterm neonates is a controversial matter, and a risk-benefit balance needs to be struck.
Objective
To evaluate 14 corticosteroid regimens used to prevent BPD: moderately early-initiated, low cumulative dose of systemic dexamethasone (MoLdDX); moderately early-initiated, medium cumulative dose of systemic dexamethasone (MoMdDX); moderately early-initiated, high cumulative dose of systemic dexamethasone (MoHdDX); late-initiated, low cumulative dose of systemic dexamethasone (LaLdDX); late-initiated, medium cumulative dose of systemic dexamethasone (LaMdDX); late-initiated, high cumulative dose of systemic dexamethasone (LaHdDX); early-initiated systemic hydrocortisone (EHC); late-initiated systemic hydrocortisone (LHC); early-initiated inhaled budesonide (EIBUD); early-initiated inhaled beclomethasone (EIBEC); early-initiated inhaled fluticasone (EIFLUT); late-initiated inhaled budesonide (LIBUD); late-initiated inhaled beclomethasone (LIBEC); and intratracheal budesonide (ITBUD).
Data Sources
PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), Embase, World Health Organization’s International Clinical Trials Registry Platform (ICTRP), and CINAHL were searched from inception through August 25, 2020.
Study Selection
In this systematic review and network meta-analysis, the randomized clinical trials selected included preterm neonates with a gestational age of 32 weeks or younger and for whom a corticosteroid regimen was initiated within 4 weeks of postnatal age. Peer-reviewed articles and abstracts in all languages were included.
Data Extraction and Synthesis
Two independent authors extracted data in duplicate. Network meta-analysis used a bayesian model.
Main Outcomes and Measures
Primary combined outcome was BPD, defined as oxygen requirement at 36 weeks’ postmenstrual age (PMA), or mortality at 36 weeks’ PMA. The secondary outcomes included 15 safety outcomes.
Results
A total of 62 studies involving 5559 neonates (mean [SD] gestational age, 26 [1] weeks) were included. Several regimens were associated with a decreased risk of BPD or mortality, including EHC (risk ratio [RR], 0.82; 95% credible interval [CrI], 0.68-0.97); EIFLUT (RR, 0.75; 95% CrI, 0.55-0.98); LaHdDX (RR, 0.70; 95% CrI, 0.54-0.87); MoHdDX (RR, 0.64; 95% CrI, 0.48-0.82); ITBUD (RR, 0.73; 95% CrI, 0.57-0.91); and MoMdDX (RR, 0.61; 95% CrI, 0.45-0.79). Surface under the cumulative ranking curve (SUCRA) value ranking showed that MoMdDX (SUCRA, 0.91), MoHdDX (SUCRA, 0.86), and LaHdDX (SUCRA, 0.76) were the 3 most beneficial interventions. ITBUD (RR, 4.36; 95% CrI, 1.04-12.90); LaHdDX (RR, 11.91; 95% CrI, 1.64-44.49); LaLdDX (RR, 6.33; 95% CrI, 1.62-18.56); MoHdDX (RR, 4.96; 95% CrI, 1.14-14.75); and MoMdDX (RR, 3.16; 95% CrI, 1.35-6.82) were associated with more successful extubation from invasive mechanical ventilation. EHC was associated with a higher risk of gastrointestinal perforation (RR, 2.77; 95% CrI, 1.09-9.32). MoMdDX showed a higher risk of hypertension (RR, 3.96; 95% CrI, 1.10-30.91). MoHdDX had a higher risk of hypertrophic cardiomyopathy (RR, 5.94; 95% CrI, 1.95-18.11).
Conclusions and Relevance
This study suggested that MoMdDX may be the most appropriate postnatal corticosteroid regimen for preventing BPD or mortality at a PMA of 36 weeks, albeit with a risk of hypertension. The quality of evidence was low.
Bronchopulmonary dysplasia (BPD) remains one of the most common, complex, and intriguing diseases among preterm neonates. Bronchopulmonary dysplasia is associated with an increased risk of mortality as well as short- and long-term morbidity, especially in extremely low-gestational-age neonates (ELGANs).1-3 Although the pathogenesis of BPD has yet to be fully deciphered, an imbalance between ongoing pulmonary inflammation and simultaneous repair seems to play a major role.4,5 In ELGANs, postnatal corticosteroids (PNCs) have been used for both the prevention and treatment of BPD. The anti-inflammatory action of PNCs on the pulmonary system has been postulated to improve gas exchange and lung mechanics, thus facilitating weaning from invasive mechanical ventilation.6 However, the use of PNCs in ELGANs remains controversial because of fear of adverse neurodevelopmental effects.7
Multiple studies have indicated that BPD is associated with risk of poor long-term neurological outcomes, such as cerebral palsy, as well as cognitive, visual, speech, and learning impairments and behavioral problems.1 Doyle et al,2 in their follow-up to a longitudinal study of extremely preterm infants born during the past 3 decades, found that the rates of BPD did not decrease despite an increase in the use of noninvasive respiratory support in the latter years. Doyle et al2 postulated that less use of PNCs, even in infants who were at the highest risk of BPD, might be a contributing factor. As such, clinicians often have to negotiate between the risks of adverse outcomes associated with long-term ventilation and the potential adverse outcomes of PNCs.
Field studies report a varying prevalence of PNC use (3%-50%).8,9 Appreciating the chasm, different academic bodies provide advice on the use of both systemic and inhaled PNCs for preterm neonates who are at risk of BPD.10-12 As a common denominator, the recommendations unanimously warn against the early use of systemic dexamethasone at fewer than 8 days of life. However, the policy statements for other regimens of systemic dexamethasone or different corticosteroids, including hydrocortisone and inhaled corticosteroids, differ widely among advisory bodies.10-12
An abundance of systematic reviews and meta-analyses that compare different PNCs using pairwise comparisons are available.13-18 However, evaluating such a complex intervention in a network meta-analysis would ease interpretation and enable the ranking of PNCs according to improved outcomes and safety.19 To our knowledge, only a single network meta-analysis has covered this subject.20 To guide safe clinical practice, we performed a comprehensive systematic review and network meta-analysis to scrutinize the existing body of evidence as well as to evaluate and compare treatments that have not been compared against each other in randomized clinical trials (RCTs).
The study protocol was registered with PROSPERO (CRD42020205685).21 We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.22
Interventions and Studies
This systematic review and network meta-analysis evaluated 14 interventions for BPD: (1) moderately early-initiated (8-14 days), low cumulative dose (<2 mg/kg) of systemic dexamethasone (MoLdDX); (2) moderately early-initiated, medium cumulative dose (2-4 mg/kg) of systemic dexamethasone (MoMdDX); (3) moderately early-initiated, high cumulative dose (>4 mg/kg) of systemic dexamethasone (MoHdDX); (4) late-initiated (15-27 days), low cumulative dose of systemic dexamethasone (LaLdDX); (5) late-initiated, medium cumulative dose of systemic dexamethasone (LaMdDX); (6) late-initiated, high cumulative dose of systemic dexamethasone (LaHdDX); (7) early-initiated (<8 days) systemic hydrocortisone (EHC); (8) late-initiated (≥8 days) systemic hydrocortisone (LHC); (9) early-initiated (<8 days) inhaled budesonide (EIBUD); (10) early-initiated inhaled beclomethasone (EIBEC); (11) early-initiated inhaled fluticasone (EIFLUT); (12) late-initiated (≥8 days) inhaled budesonide (LIBUD); (13) late-initiated inhaled beclomethasone (LIBEC); and (14) intratracheal budesonide (ITBUD; using surfactant as a vehicle). The cumulative dose and duration of systemic dexamethasone were categorized on the basis of previously published systematic reviews.14
We included studies of preterm neonates with a gestational age of 32 weeks or younger that compared any of the previously mentioned interventions in an RCT. Published and peer-reviewed full-text articles and abstracts in all languages were included. The characteristics of the included studies are provided in Table 1.23-83 Studies that evaluated early-initiated dexamethasone (<8 days of postnatal age) or dexamethasone in established cases of BPD (≥28 days) were excluded.
Primary and Secondary Outcomes
The primary combined outcome was mortality or BPD, which was defined as oxygen dependency, at a postmenstrual age (PMA) of 36 weeks. The secondary outcomes were incidences of BPD at a PMA of 36 weeks, BPD at 28 days of age, successful extubation from invasive mechanical ventilation, mortality before discharge, an intraventricular hemorrhage grade of III or IV,84 periventricular leukomalacia, hyperglycemia, hypertension (defined as systolic or diastolic blood pressure >2 SDs above the mean for neonates’ gestational and postnatal age), necrotizing enterocolitis (stage ≥II per Bell et al85), gastrointestinal perforation, retinopathy of prematurity stage 3 or higher,86 blood culture–proven sepsis, hypertrophic cardiomyopathy, cerebral palsy or neurodevelopmental impairment (NDI) at 18 to 24 months (cerebral palsy or mental development index <70 per Bayley Scales of Infant Development II), or visual or hearing impairment.
Literature Search, Study Selection, and Risk-of-Bias Assessment
We searched the online databases MEDLINE (via PubMed), Cochrane Central Register of Controlled Trials (CENTRAL), Embase, World Health Organization’s International Clinical Trials Registry Platform (ICTRP), and CINAHL (via CENTRAL) from inception through August 25, 2020. The citations from previously published systematic reviews on PNCs for prevention and treatment of BPD were also searched. The search strategy used is shown in eTable 1 in the Supplement.
Two of us (T.B. and J.A.) independently screened titles, abstracts, and full-text articles. Full-text articles of relevant RCTs were selected and evaluated for inclusion. In case of differences of opinion between the 2 reviewers, we consulted a third author (V.V.R.).
The risk of bias for the included studies was assessed by 2 of us (T.B. and V.V.R.) using the Cochrane Collaboration risk-of-bias tool.87 The domains of selection bias, performance bias, detection bias, attrition bias, reporting bias, and other bias were evaluated. Disagreements between the 2 assessors were resolved by consulting a third reviewer (C.C.R.).
The quality of evidence for the primary and secondary outcomes was assessed using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) Working Group approach for network meta-analysis.88 The direct evidence from pairwise comparisons was first evaluated on risk of bias, inconsistency (heterogeneity), imprecision, indirectness, and publication bias. The quality of indirect evidence was evaluated from the lowest quality of direct evidence (of pairwise comparisons) from the first-order loops. When 2 or more first-order loops were detected, the highest quality of evidence among them was used. If no first-order loops were present, the quality of indirect evidence was inferred from the lowest quality of direct evidence from the higher-order loops. The highest quality of evidence between direct and indirect evidence was adjudged to be the quality of the evidence for the final network estimate. The quality of evidence was ranked as follows: high (high confidence in the evidence), moderate (moderate confidence), low (some confidence), and very low (little confidence) (Table 2).
Network meta-analysis using bayesian and frequentist approaches was performed using R, version 3.6.2; netmeta; BUGSnet; and gemtc packages (R Foundation for Statistical Computing).89,90 The bayesian network meta-analysis was performed with Markov chain Monte Carlo simulation using vague priors. We used generalized linear models with 4 chains, burn-in of 50 000 iterations followed by 100 000 iterations, and 10 000 adaptations.90 Network geometry was evaluated with network plots. Model fit was assessed with leverage plots, total residual deviance, and deviance information criterion. Gelman-Rubin as well as trace and density plots were inspected for model convergence.91 Node-splitting and inconsistency model method in the bayesian approach as well as net-splitting in the frequentist approach were used to look for any inconsistency between the direct and indirect evidence.92 A P < .05 was considered as statistically significant. The sensitivity of the bayesian results was assessed by comparing them with the results of the frequentist approach.
Pairwise comparisons were depicted using forest plots and the I2 statistic to detect heterogeneity. Estimates were reported as risk ratios (RRs, with 95% credible intervals [CrIs] or 95% CIs). Forest plots, league plots, and matrix plots were used to illustrate the network estimates of various comparisons. Ranking of the interventions was performed with the surface under the cumulative ranking curve (SUCRA) plots.93 Caution is warranted when interpreting the SUCRA. The SUCRA values can vary across outcomes for the same regimen, and the differences might be attributed to chance alone; the values should be interpreted along with the quality of evidence and do not capture the magnitude of differences in outcomes between 2 regimens.93
In addition, we performed the following sensitivity analyses: (1) excluding RCTs that enrolled neonates on noninvasive respiratory support; (2) subdividing dexamethasone therapy according to the total duration of the treatment course into short course (<8 days), medium course (8-14 days) and long course (>14 days); (3) excluding RCTs with antenatal corticosteroid coverage of less than 70%; (4) excluding RCTs with a high risk of bias; (5) excluding RCTs that evaluated intratracheal budesonide; and (6) combining the different types of inhaled corticosteroids.
Of the 2211 studies screened, 62 studies (involving 5559 neonates) were included in the final analysis23-83 (Table 1 and eFigure 1 in the Supplement). The mean (SD) gestational age of the enrolled neonates was 26 (1) weeks (eFigure 2 in the Supplement). We contacted 19 authors for additional information regarding their studies, and 2 responded. The included studies consisted of two 3-group RCTs,24,63 and 41 had enrolled ventilated neonates,23-29,31,34,35,38-42,44,46,48,51,55-57,59,60,62-66,68-71,74-76,78-81,83 but only 18 had included neonates with an antenatal corticosteroid coverage of more than 70%.25,26,28,29,31,34,42-44,46,51,59,60,62,65,66,75,79 Excluded studies are listed in eTable 2 in the Supplement.
Thirteen studies were judged to have a high risk of bias.25,32,33,37,50,54,55,58,65,68,75,81,82 Three studies were classified as having variable risk of bias,24,28,69 and 45 studies had a low risk of bias.23,26,27,29-31,34-36,38-49,51-53,56,57,59-64,66,67,70-74,76-80,83 The RCTs with a high risk of bias predominantly had concerns regarding random sequence generation and allocation concealment. The risk-of-bias summary and graph are shown in eFigure 3 in the Supplement.
A total of 45 studies with 5236 neonates and 2768 events (53% event rate) were included for the assessment of the primary outcome (Figure 1 and eTable 3 in the Supplement).23-29,31,34,35,38-44,46,59,49-51,55-57,60,62-66,68-71,74-76,78-81,83 When compared with placebo, EHC (RR, 0.82; 95% CrI, 0.68-0.97); EIFLUT (RR, 0.75; 95% CrI, 0.55-0.98); LaHdDX (RR, 0.70; 95% CrI, 0.54-0.87); MoHdDX (GRADE: moderate) (RR, 0.64; 95% CrI, 0.48-0.82); ITBUD (RR, 0.73; 95% CrI, 0.57-0.91); and MoMdDX (GRADE: low) (RR, 0.61; 95% CrI, 0.45-0.79) decreased the risk of the primary outcome, mortality or BPD at 36 weeks’ PMA (Figure 1). Furthermore, MoMdDX was better than LaMdDX (GRADE: low) (RR, 0.67; 95% CrI, 0.45-0.94), and MoHdDX was better than LaMdDX (GRADE: low) (RR, 0.70; 95% CrI, 0.48-0.99) (Figure 2).
SUCRA values ranked MoMdDX (SUCRA, 0.91); MoHdDX (SUCRA, 0.86); and LaHdDX (SUCRA, 0.76) as the 3 most beneficial interventions. The least beneficial intervention was LHC (SUCRA, 0.17) (Figure 1). Direct evidence from the pairwise comparisons and the split between direct and indirect evidence are shown in eFigures 4 and 5 in the Supplement. The quality of evidence for the various comparisons for the primary outcome is described in Table 2.
The network characteristics for all 15 secondary outcomes are provided in eTable 3 in the Supplement. The network plots; SUCRA plots; forest plots depicting the network estimate compared with placebo, split between direct and indirect evidence, and direct pairwise comparisons; and the league plots or matrix plots are shown in eFigures 6 to 56 in the Supplement. The quality-of-evidence assessment is outlined in eTable 4 in the Supplement. Because of the sparseness of some of the networks, node-splitting to assess inconsistency was not possible for some of the 15 outcomes (intraventricular hemorrhage, periventricular leukomalacia, necrotizing enterocolitis, hypertrophic cardiomyopathy, and cerebral palsy at 18-24 months).
EIBUD was associated with decreased incidence of BPD at a PMA of 36 weeks (GRADE: moderate) (RR, 0.71; 95% CrI, 0.51-0.94). Furthermore, EHC showed a pattern of lesser incidence of BPD at 36 weeks’ PMA but did not reach statistical significance (RR, 0.86; 95% CrI, 0.71-1.02).
When compared with placebo, MoMdDX (GRADE: low) (RR, 0.43; 95% CrI, 0.18-0.82) and EHC (GRADE: moderate) (RR, 0.69; 95% CrI, 0.44-0.98) were associated with reduced mortality before discharge. EIBUD did not show any significant difference in the incidence of mortality compared with placebo (RR, 0.89; 95% CrI, 0.46-1.36).
The following regimens were associated with more extubations: ITBUD (RR, 4.36; 95% CrI, 1.04-12.90); LaHdDX (RR, 11.91; 95% CrI, 1.64-44.49); LaLdDX (RR, 6.33; 95% CrI, 1.62-18.56); MoHdDX (RR, 4.96; 95% CrI, 1.14-14.75); and MoMdDX (GRADE: very low to low) (RR, 3.16; 95% CrI, 1.35-6.82). According to SUCRA ranking, LaHdDX was the best intervention.
None of the interventions were associated with either a decreased or an increased risk of NDI. In addition, LaHdDX showed a lower incidence of NDI at 18 to 24 months compared with LaMdDX (GRADE: low) (RR, 0.31; 95% CrI, 0.03-0.90).
Compared with placebo, EHC was associated with a higher risk of gastrointestinal perforation (GRADE: moderate) (RR, 2.77; 95% CrI, 1.09-9.32). The network was sparse, and inconsistency assessment was not possible. The direct pairwise analysis revealed a higher risk of hypertrophic cardiomyopathy with MoHdDX compared with placebo (GRADE: low) (RR, 5.94; 95% CrI, 1.95-18.11).
Compared with placebo, MoMdDX was associated with higher risk of hypertension (GRADE: low) (RR, 3.96; 95% CrI, 1.10-30.91). Inconsistency was detected in the network meta-analysis.
Moderately early-initiated, medium cumulative dose, short course of systemic dexamethasone (RR, 0.54; 95% CrI, 0.30-0.86); moderately early-initiated, high cumulative dose, long course of systemic dexamethasone (RR, 0.65; 95% CrI, 0.42-0.95); and late-initiated, high cumulative dose, long course of systemic dexamethasone (RR, 0.71; 95% CrI, 0.52-0.94) were associated with decreased risk of the primary outcome, mortality or BPD at PMA of 36 weeks (Figure 3). On the basis of SUCRA values, a moderately early-initiated, medium cumulative dose, short course of systemic dexamethasone was found to be the best intervention (SUCRA, 0.82). The network characteristics for all of the sensitivity analyses are shown in eTable 3 in the Supplement. Inconsistency assessment by splitting direct and indirect evidence as well as the league plot depicting the network estimates for the different comparisons are given in eFigures 57 and 58 in the Supplement.
In excluding RCTs that enrolled neonates on noninvasive respiratory support, we found that the results for the primary outcome were similar to those of the primary analysis. When RCTs with antenatal corticosteroid coverage of less than 70% were excluded, only ITBUD showed a significant decrease in the incidence of the primary outcome (RR, 0.64; 95% CrI, 0.44-0.89) (eFigures 59-61 in the Supplement).
The results for the primary outcome were similar to those of the primary analysis when RCTs that evaluated ITBUD were excluded. When trials with a high risk of bias were excluded, the network estimates for both EHC (RR, 0.75; 95% CrI, 0.53-1.01) and EIFLUT (RR, 0.84; 95% CrI, 0.67-1.02) compared with placebo did not show any significant decrease in the mortality or BPD at a PMA of 36 weeks (eFigures 62-64 in the Supplement).
By combining the different types of inhaled corticosteroids, we found that early-initiated inhaled corticosteroid was associated with a decreased risk of the primary outcome measure when compared with placebo (RR, 0.81; 95% CrI, 0.69-0.94). All of the other results for different comparisons were similar to those of the primary analysis (eFigures 65-67 in the Supplement).
This systematic review and network meta-analysis included 62 RCTs that evaluated the outcomes and safety of different regimens of PNCs administered through various routes (systemic, inhaled, and intratracheal). The primary outcome measure was the combined outcome of BPD or mortality at a PMA of 36 weeks.
These results indicate that moderately early initiation (8-14 days of postnatal age) of systemic dexamethasone with a medium cumulative dose (2-4 mg/kg) appeared to be the most successful intervention in preventing mortality or BPD. In contrast, Zeng et al20 in their network meta-analysis found the early initiation (<8 days) of systemic dexamethasone in a high dose (>3 mg/kg/d) to be the most successful in preventing BPD at 36 weeks. Substantial differences exist between the present network meta-analysis and that of Zeng et al.20 First, we excluded studies that had evaluated early systemic dexamethasone because this therapy is unanimously touted as an unsafe intervention by different academic bodies and by authors of several systematic reviews.7,8,10,94 Second, the interventions that we evaluated were different with the addition of LHC and ITBUD, as well as splitting inhaled corticosteroids (EIBUD, EIBEC, EIFLUT, LIBEC, and LIBUD) and subdividing systemic dexamethasone into different regimens based on a previously published Cochrane review.14 Third, although Zeng et al20 had evaluated safety by assessing cerebral palsy alone, we assessed 11 short- and long-term adverse effects. Fourth, we included 25 trials28-34,36,37,40,41,43,49,52-54,57,58,62,72,75,78,80,82,83 that were not evaluated by Zeng et al20 because of the different inclusion criteria.
The 2020 Canadian Paediatric Society recommendation advises initiating systemic dexamethasone at a dose of 0.15 to 0.2 mg/kg/d for a total duration of 7 to10 days in preterm neonates who require continued invasive mechanical ventilation beyond the first week of life.12 In keeping with this recommendation, our findings also suggested that moderately early initiation (8-14 days) of dexamethasone is better than late initiation (>14 days). However, the cumulative dose suggested by the Canadian Paediatric Society is anywhere between 1 and 2 mg/kg, whereas findings in the present study favored a dose of 2 to 4 mg/kg. The 2020 Canadian Paediatric Society recommendation relied on evidence from pairwise meta-analyses. In addition, a guidelines panel takes into consideration multiple other parameters when moving from evidence to recommendation, such as clinician and parental preference, adverse effects profile, and cost-benefit ratio.95
In this study, the quality of evidence for MoMdDX or MoLdDX compared with placebo was low, implying that the true effect (or the reality, as defined by the GRADE group) might be substantially different from the estimated outcome. Risk of bias and imprecision were the reasons for downgrading the quality of evidence for MoMdDX vs placebo. The sensitivity analysis that evaluated the duration of dexamethasone course indicated that a moderately early-initiated, medium dose, and short course (<8 days) of systemic dexamethasone might be the best intervention. A short course has also been advocated in the 2019 update of the European Consensus Guidelines on the Management of Respiratory Distress Syndrome.11
Of the different regimens with late initiation, only LaHdDX was associated with a significant decrease in risk of the primary outcome. Although LaLdDX appeared to be a good intervention to facilitate extubation, it did not translate into better clinical outcomes (GRADE: low). Thus, late initiation might warrant a relatively high dose for a better outcome than moderately early initiation. In their prediction model of the various risk factors of BPD, Laughon et al96 concluded that, with advancing age, the risk of BPD of all severity increased with continued mechanical ventilation. This risk might be secondary to the ongoing pulmonary inflammation from a ventilator-induced lung injury and hence requiring a higher dose of systemic dexamethasone in the third rather than the second week of life. This analysis of long-term outcomes indicated that none of the interventions were associated with an increased risk of NDI when compared with placebo (GRADE: low). We also found that LaHdDX had a lower incidence of NDI compared with LaMdDX. In a sparse network, the evidence was derived from a single study published 3 decades ago, and the overall quality of evidence was judged to be low.63
This study showed that, although LHC did not have any benefits for short-term outcomes, EHC had a decreased risk of the combined outcome of BPD or mortality at 36 weeks that was attributable to a decreased risk of mortality. These findings are in agreement with the 2010 American Academy of Paediatrics and 2020 Canadian Paediatric Society policy statements.10,12 In addition, EHC was associated with higher rates of gastrointestinal perforation. It was suggested that this increased risk of perforation was associated with concomitant indomethacin use.12 The present analysis also indicated that EIFLUT might be associated with a decreased risk of mortality or BPD. Moderate-quality evidence points toward a decreased risk of BPD at a PMA of 36 weeks associated with EIBUD. However, EIBUD was not found to have a lower risk of the competing outcome of BPD or mortality at a PMA of 36 weeks. The possibility of this finding being associated with an increase in the risk of mortality could not be ruled out.
In the present analysis, ITBUD appeared to be beneficial, a finding that is in sync with other pairwise meta-analyses.15,16 However, intratracheal budesonide is used very early in postnatal life during the window of surfactant administration, when a higher proportion of the neonates are successfully extubated. Further studies are warranted to investigate this intervention’s benefits in neonates who remain ventilated beyond the first week of life.
This study has several limitations. Most of the included RCTs had used open-label dexamethasone. Analysis of some of the postnatal systemic dexamethasone regimens (pulse, individualized, or based on daily dose) could not be performed. We could not conduct a meta-regression analysis similar to that of Doyle et al97 by assessing neonates according to their baseline risk of BPD. Some of the networks assessing the safety outcomes were sparse, with low to very low quality of evidence.
Although it presents a risk of hypertension, a moderately early-initiated (8-14 days), medium cumulative dose (2-4 mg/kg), short course (<8 days) of systemic dexamethasone might be the most appropriate PNC regimen for preventing the risk of BPD or mortality at PMA of 36 weeks, with a low quality of evidence. In view of this low confidence in the evidence, the successful outcome and safety of this regimen need to be confirmed by an adequately powered multicentric RCT.
Neither a moderately early-initiated nor late-initiated low dose of systemic dexamethasone (<2 mg/kg) appeared successful in decreasing the risk of BPD or mortality at a PMA of 36 weeks, with a low quality of evidence.
EHC might confer a survival advantage, with a moderate quality of evidence. EIFLUT might be a better alternative to EIBUD, especially in lieu of the possibility of increased risk of mortality with EIBUD, with a moderate quality of evidence. Although ITBUD appeared to be a good intervention with a low quality of evidence, further research is warranted in neonates who require continued mechanical ventilation beyond the first week of life.
Accepted for Publication: October 27, 2020.
Published Online: March 15, 2021. doi:10.1001/jamapediatrics.2020.6826
Corresponding Author: Charles C. Roehr, MD, PhD, National Perinatal Epidemiology Unit, Nuffield Department of Population Health Medical Sciences Division, University of Oxford, Old Road Campus, Richard Doll Building, Oxford OX3 7LF, United Kingdom (charles.roehr@ouh.nhs.uk).
Author Contributions: Dr Ramaswamy had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Ramaswamy, Bandyopadhyay, Roehr, Nangia.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Ramaswamy, Bandyopadhyay, Bandiya, Roehr.
Critical revision of the manuscript for important intellectual content: Ramaswamy, Bandyopadhyay, Nanda, Ahmed, Garg, Roehr, Nangia.
Statistical analysis: Ramaswamy, Bandyopadhyay.
Administrative, technical, or material support: Nanda, Bandiya, Ahmed, Roehr, Nangia.
Supervision: Ramaswamy, Bandyopadhyay, Roehr, Nangia.
Conflict of Interest Disclosures: None reported.
2.Doyle
LW, Carse
E, Adams
AM, Ranganathan
S, Opie
G, Cheong
JLY; Victorian Infant Collaborative Study Group. Ventilation in extremely preterm infants and respiratory function at 8 years.
N Engl J Med. 2017;377(4):329-337. doi:
10.1056/NEJMoa1700827
PubMedGoogle ScholarCrossref 3.Schmidt
B, Asztalos
EV, Roberts
RS, Robertson
CM, Sauve
RS, Whitfield
MF; Trial of Indomethacin Prophylaxis in Preterms (TIPP) Investigators. Impact of bronchopulmonary dysplasia, brain injury, and severe retinopathy on the outcome of extremely low-birth-weight infants at 18 months: results from the Trial of Indomethacin Prophylaxis in Preterms.
JAMA. 2003;289(9):1124-1129. doi:
10.1001/jama.289.9.1124
PubMedGoogle ScholarCrossref 8.Nuytten
A, Behal
H, Duhamel
A,
et al; EPICE (Effective Perinatal Intensive Care in Europe) Research Group. Correction: evidence-based neonatal unit practices and determinants of postnatal corticosteroid-use in preterm births below 30 weeks GA in Europe: a population-based cohort study.
PLoS One. 2017;12(2):e0172408. doi:
10.1371/journal.pone.0172408
PubMedGoogle Scholar 9.Gortner
L, Misselwitz
B, Milligan
D,
et al; Members of the MOSAIC Research Group. Rates of bronchopulmonary dysplasia in very preterm neonates in Europe: results from the MOSAIC cohort.
Neonatology. 2011;99(2):112-117. doi:
10.1159/000313024
PubMedGoogle ScholarCrossref 10.Watterberg
KL; American Academy of Pediatrics, Committee on Fetus and Newborn. Policy statement—postnatal corticosteroids to prevent or treat bronchopulmonary dysplasia.
Pediatrics. 2010;126(4):800-808. doi:
10.1542/peds.2010-1534
PubMedGoogle ScholarCrossref 15.Delara
M, Chauhan
BF, Le
ML, Abou-Setta
AM, Zarychanski
R, 'tJong
GW. Efficacy and safety of pulmonary application of corticosteroids in preterm infants with respiratory distress syndrome: a systematic review and meta-analysis.
Arch Dis Child Fetal Neonatal Ed. 2019;104(2):F137-F144. doi:
10.1136/archdischild-2017-314046
PubMedGoogle ScholarCrossref 16.Zhong
YY, Li
JC, Liu
YL,
et al; Intratracheal Administration of Corticosteroid and Pulmonary Surfactant for Preventing Bronchopulmonary Dysplasia in Preterm Infants With Neonatal Respiratory Distress Syndrome. Early intratracheal administration of corticosteroid and pulmonary surfactant for preventing bronchopulmonary dysplasia in preterm infants with neonatal respiratory distress syndrome: a meta-analysis.
Curr Med Sci. 2019;39(3):493-499. doi:
10.1007/s11596-019-2064-9
PubMedGoogle ScholarCrossref 17.Shah
SS, Ohlsson
A, Halliday
HL, Shah
VS. Inhaled versus systemic corticosteroids for preventing bronchopulmonary dysplasia in ventilated very low birth weight preterm neonates.
Cochrane Database Syst Rev. 2017;10(10):CD002058. doi:
10.1002/14651858.CD002058.pub3PubMedGoogle Scholar 22.Hutton
B, Salanti
G, Caldwell
DM,
et al. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations.
Ann Intern Med. 2015;162(11):777-784. doi:
10.7326/M14-2385
PubMedGoogle ScholarCrossref 24.Halliday
HL, Patterson
CC, Halahakoon
CW; European Multicenter Steroid Study Group. A multicenter, randomized Open Study of Early Corticosteroid Treatment (OSECT) in preterm infants with respiratory illness: comparison of early and late treatment and of dexamethasone and inhaled budesonide.
Pediatrics. 2001;107(2):232-240. doi:
10.1542/peds.107.2.232
PubMedGoogle ScholarCrossref 27.Jangaard
KA, Stinson
DA, Allen
AC, Vincer
MJ. Early prophylactic inhaled beclomethasone in infants less than 1250 g for the prevention of chronic lung disease.
Paediatr Child Health. 2002;7(1):13-19. doi:
10.1093/pch/7.1.13
PubMedGoogle ScholarCrossref 28.Lin
YJ, Lin
HC, Lin
CH, Su
BH, Yeh
TF. Early endotracheal instillation of budesonide (B) for prevention of CLD in preterm infant with RDS-a double blind clinical trial.
Ped Res. 2001;21:278A.
Google Scholar 29.Yeh
TF, Lin
HC, Chang
CH,
et al. Early intratracheal instillation of budesonide using surfactant as a vehicle to prevent chronic lung disease in preterm infants: a pilot study.
Pediatrics. 2008;121(5):e1310-e1318. doi:
10.1542/peds.2007-1973
PubMedGoogle ScholarCrossref 32.Ke
H, Li
ZK, Yu
XP, Guo
JZ. Efficacy of different preparations of budesonide combined with pulmonary surfactant in the treatment of neonatal respiratory distress syndrome: a comparative analysis [in Chinese].
Zhongguo Dang Dai Er Ke Za Zhi. 2016;18(5):400-404.
PubMedGoogle Scholar 33.Pan
J, Chen
MW, Ni
WQ,
et al. Clinical efficacy of pulmonary surfactant combined with budesonide for preventing bronchopulmonary dysplasia in very low birth weight infants [in Chinese].
Zhongguo Dang Dai Er Ke Za Zhi. 2017;19(2):137-141. doi:
10.7499/j.issn.1008-8830.2017.02.002PubMedGoogle Scholar 34.Merz
U, Kusenbach
G, Häusler
M, Peschgens
T, Hörnchen
H. Inhaled budesonide in ventilator-dependent preterm infants: a randomized, double-blind pilot study.
Biol Neonate. 1999;75(1):46-53. doi:
10.1159/000014076
PubMedGoogle ScholarCrossref 36.Bassler
D, Shinwell
ES, Hallman
M,
et al; Neonatal European Study of Inhaled Steroids Trial Group. Long-term effects of inhaled budesonide for bronchopulmonary dysplasia.
N Engl J Med. 2018;378(2):148-157. doi:
10.1056/NEJMoa1708831
PubMedGoogle ScholarCrossref 38.Cao
YY, Yao
G, Wang
Y,
et al. Aerosol inhalation of budesonide and pulmonary surfactant to prevent bronchopulmonary dysplasia.
J Pediatr Pharmacy. 2018;24(03):25-29.
Google Scholar 39.Fok
TF, Lam
K, Dolovich
M,
et al. Randomised controlled study of early use of inhaled corticosteroid in preterm infants with respiratory distress syndrome.
Arch Dis Child Fetal Neonatal Ed. 1999;80(3):F203-F208. doi:
10.1136/fn.80.3.F203
PubMedGoogle ScholarCrossref 40.Yong
WSC, Carney
S, Pearse
RG, Gibson
AT. The effect of inhaled fluticasone propionate (FP) on premature babies at risk for developing chronic lung disease of prematurity.
Arch Dis Child Fetal Neonatal Ed. 1999;80:G64.
Google ScholarCrossref 43.Efird
MM, Heerens
AT, Gordon
PV, Bose
CL, Young
DA. A randomized-controlled trial of prophylactic hydrocortisone supplementation for the prevention of hypotension in extremely low birth weight infants.
J Perinatol. 2005;25(2):119-124. doi:
10.1038/sj.jp.7211193
PubMedGoogle ScholarCrossref 45.Watterberg
KL, Shaffer
ML, Mishefske
MJ,
et al. Growth and neurodevelopmental outcomes after early low-dose hydrocortisone treatment in extremely low birth weight infants.
Pediatrics. 2007;120(1):40-48. doi:
10.1542/peds.2006-3158
PubMedGoogle ScholarCrossref 46.Peltoniemi
O, Kari
MA, Heinonen
K,
et al. Pretreatment cortisol values may predict responses to hydrocortisone administration for the prevention of bronchopulmonary dysplasia in high-risk infants.
J Pediatr. 2005;146(5):632-637. doi:
10.1016/j.jpeds.2004.12.040
PubMedGoogle ScholarCrossref 49.Ng
PC, Lee
CH, Bnur
FL,
et al. A double-blind, randomized, controlled study of a “stress dose” of hydrocortisone for rescue treatment of refractory hypotension in preterm infants.
Pediatrics. 2006;117(2):367-375. doi:
10.1542/peds.2005-0869
PubMedGoogle ScholarCrossref 51.Baud
O, Maury
L, Lebail
F,
et al; PREMILOC Trial Study Group. Effect of early low-dose hydrocortisone on survival without bronchopulmonary dysplasia in extremely preterm infants (PREMILOC): a double-blind, placebo-controlled, multicentre, randomised trial.
Lancet. 2016;387(10030):1827-1836. doi:
10.1016/S0140-6736(16)00202-6
PubMedGoogle ScholarCrossref 52.Baud
O, Trousson
C, Biran
V, Leroy
E, Mohamed
D, Alberti
C; PREMILOC Trial Group. Two-year neurodevelopmental outcomes of extremely preterm infants treated with early hydrocortisone: treatment effect according to gestational age at birth.
Arch Dis Child Fetal Neonatal Ed. 2019;104(1):F30-F35. doi:
10.1136/archdischild-2017-313756
PubMedGoogle ScholarCrossref 57.Rozycki
HJ, Byron
PR, Elliott
GR, Carroll
T, Gutcher
GR. Randomized controlled trial of three different doses of aerosol beclomethasone versus systemic dexamethasone to promote extubation in ventilated premature infants.
Pediatr Pulmonol. 2003;35(5):375-383. doi:
10.1002/ppul.10269
PubMedGoogle ScholarCrossref 62.Onland
W, Cools
F, Kroon
A,
et al; STOP-BPD Study Group. Effect of hydrocortisone therapy initiated 7 to 14 days after birth on mortality or bronchopulmonary dysplasia among very preterm infants receiving mechanical ventilation: a randomized clinical trial.
JAMA. 2019;321(4):354-363. doi:
10.1001/jama.2018.21443
PubMedGoogle ScholarCrossref 64.Kothadia
JM, O’Shea
TM, Roberts
D, Auringer
ST, Weaver
RG
III, Dillard
RG. Randomized placebo-controlled trial of a 42-day tapering course of dexamethasone to reduce the duration of ventilator dependency in very low birth weight infants.
Pediatrics. 1999;104(1, pt 1):22-27. doi:
10.1542/peds.104.1.22
PubMedGoogle ScholarCrossref 65.Malloy
CA, Hilal
K, Rizvi
Z, Weiss
M, Muraskas
JK. A prospective, randomized, double-masked trial comparing low dose to conventional dose dexamethasone in neonatal chronic lung disease.
Internet J Pediat Neonatol. 2005;5(1):10473.
Google Scholar 66.Doyle
LW, Davis
PG, Morley
CJ, McPhee
A, Carlin
JB; DART Study Investigators. Low-dose dexamethasone facilitates extubation among chronically ventilator-dependent infants: a multicenter, international, randomized, controlled trial.
Pediatrics. 2006;117(1):75-83. doi:
10.1542/peds.2004-2843
PubMedGoogle ScholarCrossref 67.Doyle
LW, Davis
PG, Morley
CJ, McPhee
A, Carlin
JB; DART Study Investigators. Outcome at 2 years of age of infants from the DART study: a multicenter, international, randomized, controlled trial of low-dose dexamethasone.
Pediatrics. 2007;119(4):716-721. doi:
10.1542/peds.2006-2806
PubMedGoogle ScholarCrossref 69.Ramanathan
R, Siassi
B, Sardesai
S, deLemos
RA. Comparison of two dosage regimens of dexamethasone for early treatment of chronic lung disease in very low birth weight (VLBW).
Pediatric Research. 1994;34:250A.
Google Scholar 71.Durand
M, Sardesai
S, McEvoy
C. Effects of early dexamethasone therapy on pulmonary mechanics and chronic lung disease in very low birth weight infants: a randomized, controlled trial.
Pediatrics. 1995;95(4):584-590.
PubMedGoogle Scholar 72.Scott
SM, Backstrom
C, Bessman
S. Effect of five days of dexamethasone therapy on ventilator dependence and adrenocorticotropic hormone-stimulated cortisol concentrations.
J Perinatol. 1997;17(1):24-28.
PubMedGoogle Scholar 74.Armstrong
DL, Penrice
J, Bloomfield
FH, Knight
DB, Dezoete
JA, Harding
JE. Follow up of a randomised trial of two different courses of dexamethasone for preterm babies at risk of chronic lung disease.
Arch Dis Child Fetal Neonatal Ed. 2002;86(2):F102-F107. doi:
10.1136/fn.86.2.F102
PubMedGoogle ScholarCrossref 76.Romagnoli
C, Zecca
E, Vento
G, De Carolis
MP, Papacci
P, Tortorolo
G. Early postnatal dexamethasone for the prevention of chronic lung disease in high-risk preterm infants.
Intensive Care Med. 1999;25(7):717-721. doi:
10.1007/s001340050935
PubMedGoogle ScholarCrossref 77.Romagnoli
C, Zecca
E, Luciano
R, Torrioli
G, Tortorolo
G. A three year follow up of preterm infants after moderately early treatment with dexamethasone.
Arch Dis Child Fetal Neonatal Ed. 2002;87(1):F55-F58. doi:
10.1136/fn.87.1.F55
PubMedGoogle ScholarCrossref 78.Durand
M, Mendoza
ME, Tantivit
P, Kugelman
A, McEvoy
C. A randomized trial of moderately early low-dose dexamethasone therapy in very low birth weight infants: dynamic pulmonary mechanics, oxygenation, and ventilation.
Pediatrics. 2002;109(2):262-268. doi:
10.1542/peds.109.2.262
PubMedGoogle ScholarCrossref 80.McEvoy
C, Bowling
S, Williamson
K, McGaw
P, Durand
M. Randomized, double-blinded trial of low-dose dexamethasone: II. functional residual capacity and pulmonary outcome in very low birth weight infants at risk for bronchopulmonary dysplasia.
Pediatr Pulmonol. 2004;38(1):55-63. doi:
10.1002/ppul.20037
PubMedGoogle ScholarCrossref 83.Marr
BL, Mettelman
BB, Bode
MM, Gross
SJ. Randomized trial of 42-day compared with 9-day courses of dexamethasone for the treatment of evolving bronchopulmonary dysplasia in extremely preterm infants.
J Pediatr. 2019;211:20-26.e1. doi:
10.1016/j.jpeds.2019.04.047PubMedGoogle ScholarCrossref 88.Puhan
MA, Schünemann
HJ, Murad
MH,
et al; GRADE Working Group. A GRADE Working Group approach for rating the quality of treatment effect estimates from network meta-analysis.
BMJ. 2014;349:g5630. doi:
10.1136/bmj.g5630
PubMedGoogle ScholarCrossref 91.Brooks
S, Gelman
A. General methods for monitoring convergence of iterative simulations.
J Comput Graph Stat. 1998;7(4):434-455.
Google Scholar 94.Doyle
LW, Cheong
JL, Ehrenkranz
RA, Halliday
HL. Early (< 8 days) systemic postnatal corticosteroids for prevention of bronchopulmonary dysplasia in preterm infants.
Cochrane Database Syst Rev. 2017;10(10):CD001146. doi:
10.1002/14651858.CD001146.pub5
PubMedGoogle Scholar 96.Laughon
MM, Langer
JC, Bose
CL,
et al; Eunice Kennedy Shriver National Institute of Child Health and Human Development Neonatal Research Network. Prediction of bronchopulmonary dysplasia by postnatal age in extremely premature infants.
Am J Respir Crit Care Med. 2011;183(12):1715-1722. doi:
10.1164/rccm.201101-0055OC
PubMedGoogle ScholarCrossref 97.Doyle
LW, Halliday
HL, Ehrenkranz
RA, Davis
PG, Sinclair
JC. An update on the impact of postnatal systemic corticosteroids on mortality and cerebral palsy in preterm infants: effect modification by risk of bronchopulmonary dysplasia.
J Pediatr. 2014;165(6):1258-1260. doi:
10.1016/j.jpeds.2014.07.049
PubMedGoogle ScholarCrossref