SEARCH

SEARCH BY CITATION

Keywords:

  • bronchopulmonary dysplasia;
  • chronic lung disease;
  • corticosteroids;
  • infants;
  • neonates;
  • preterm

Abstract

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

Background

Bronchopulmonary dysplasia (BPD) is an important complication associated with considerable morbidity in preterm infants. Corticosteroids in various regimens have been tried out to prevent BPD.

Objective

To examine the evidence from Cochrane systematic reviews regarding the effectiveness and associated complications of corticosteroids used to prevent BPD in preterm infants.

Methods

The Cochrane Database of Systematic Reviews was searched to identify reviews of corticosteroids for BPD in preterm infants. Data were extracted by one investigator, and checked by a second investigator for accuracy. Results are presented as risk ratios (RR) with 95% confidence intervals (CI). We considered <8 days as early and >7 days as late administration.

Main results

Six reviews (67 trials and 6535 patients) were included and covered three main comparisons: inhaled corticosteroids versus placebo, inhaled versus systemic corticosteroids and systemic corticosteroids versus placebo. Systemic corticosteroids compared with placebo significantly reduced the incidence of BPD (early: RR 0.79, 95% CI 0.71–0.88; late: RR 0.72, 95% CI 0.63–0.82), and BPD or mortality (early: RR 0.89, 95% CI 0.84–0.95; late: RR 0.72, 95% CI 0.63–0.82) at 36 weeks post-menstrual age. Similar results were observed for these outcomes assessed at 28 days of life at which time there was additionally a reduction in mortality (late: RR 0.49, 95% CI 0.28–0.85). There was a higher incidence of cerebral palsy associated with systemic corticosteroids compared with placebo when initiated early (RR 1.45, 95% CI 1.06–1.98). No differences in neurodisability based on Bayley Mental or Psychomotor Developmental Index scores were observed for inhaled or systemic corticosteroids compared with placebo. Hypertension was significantly increased in association with systemic corticosteroids versus placebo (early: RR 1.85, 95% CI 1.55–2.22; late: RR 2.66, 95% CI 1.58–4.49) as were gastrointestinal (GI) perforations when treatment was initiated early (RR 1.81, 95% CI 1.33–2.48).

Author's Conclusion

Systemic corticosteroids decrease BPD and early mortality in premature infants but have a risk of complications, particularly when initiated in the first week of life. Owing to the wide range of dosing protocols, timing of initiation, doses and duration of therapy in the included reviews, it is difficult to determine, based on the evidence examined in this overview, a protocol that is both safe and effective. Further research should focus on determining the most effective dose and timing of corticosteroid administration beyond the first week of life to maximize benefit in decreasing BPD and mortality while avoiding short- and long-term harms.


Plain language summary

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

Bronchopulmonary dysplasia (BPD) is a debilitating disease of the lung occurring in preterm infants due to lung development outside the uterus and artificial breathing or oxygen support. Its characteristics have changed with advancements in care of extremely premature infants over the last few decades but its occurrence has not significantly decreased. Use of corticosteroids can reduce early death in newborns and can significantly reduce the occurrence of BPD. There are serious side effects associated with corticosteroid use and therefore the risks of therapy must be weighed against the benefits. Inhaled corticosteroids may be considered safe by clinicians; however, their side effects have not been sufficiently evaluated. There are some limitations in the available literature; therefore, there is a need for further well conducted research to determine the best corticosteroid treatment regimen (dose, duration and timing) in order to maximize the benefit of reducing BPD while minimizing long-term harm.

Background

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

Description of the condition

Chronic lung disease (CLD) or BPD is a major cause of mortality and morbidity in preterm infants, and in particular in extremely low birthweight (<1000 g) infants. The terms CLD and BPD, while not technically the same, are often used interchangeably in practice and in the literature (1). At a 2001 National Institute of Child Health and Human Development (NICHD) workshop, participants determined that the best name for the disease is BPD as this term clearly distinguishes it from the many other chronic lung diseases occurring later in life (2). We have used BPD throughout this overview. Since it was first described by Northway in 1967 (3), the definition of the disease has changed a number of times as has the clinical picture of the disease. Northway attributed the lung damage in these infants primarily to the use of aggressive mechanical ventilation and high inspired oxygen concentrations. It is now recognized that pulmonary inflammation associated with antenatal and postnatal infection, exposure to O2 and mechanical ventilation, as well as genetic predisposition are also contributors to the development of BPD (46).

There have been many definitions of BPD. Commonly used definitions include the need for supplemental oxygen at either 28 days of age or 36 weeks post-menstrual age (PMA) (1,2,7). Some definitions require radiographic changes in addition to the need for oxygen (8) and others consider whether an infant is receiving positive pressure (9). Severity criteria, classifying BPD as mild, moderate or severe based on duration and degree of oxygen supplementation, were developed at a workshop held by the National Institutes of Health in 2001 (2); these have since been shown to better predict adverse pulmonary and neurodevelopmental outcomes than other definitions (1).

Changes in neonatal and perinatal medicine, including the widespread use of antenatal corticosteroids and postnatal surfactant, along with less aggressive mechanical ventilation have changed the clinical presentation of BPD. Classic BPD was characterized predominantly by a combination of atelectasis and hyperinflation or cystic changes along with fibrosis, the damage being caused by oxygen toxicity and mechanical ventilation (10). More recently, BPD is found in a more preterm population (11,12) of patients who required less oxygen and ventilation in their early course than previously seen. This new BPD is characterized by arrested alveolar and vascular development of the immature lung. Despite the increased use of continuous positive airway pressure (CPAP) therapy and gentle ventilation strategies, the overall rates of BPD are not decreasing and it remains an important health burden. Nearly 25% of infants with birthweight under 1500 g still develop BPD (1214). Infants with BPD can develop chronic respiratory morbidity and lung function abnormalities (4) and they have higher rates of cognitive, educational and behavioural impairments (15).

Description of the interventions

As inflammation is one of the primary intermediaries of injury in the pathogenesis of BPD, corticosteroids have been used in an attempt to prevent or treat BPD in preterm infants.

Studies of corticosteroids in BPD have used different corticosteroids (16) with multiple modes of administration (including intravenous, inhalational and recently intratracheal with surfactant as a vehicle). Dexamethasone is the most studied steroid for BPD in preterm infants. Hydrocortisone has also been studied for this use.

Beclomethasone, budesonide, flunisolide and fluticasone by inhalation have all been studied for the care of infants with or at risk of BPD. Inhaled corticosteroids have been used and studied with a view to decrease the need for systemic steroids and in the hope of decreasing side effects compared to systemic corticosteroid administration (17).

How the interventions might work

Substantial research has been done to understand and substantiate the role of antenatal and postnatal inflammation in the pathogenesis of BPD (1820). Corticosteroids are strong anti-inflammatory agents, which are thought to treat or prevent BPD by controlling or decreasing pulmonary inflammation. Corticosteroids act on diverse targets through multiple mechanisms to control inflammation (21). Some of the main pathways of action include decreasing recruitment of polymorphonuclear leukocytes, suppression of multiple inflammatory mediators and decreasing vascular permeability (22). Corticosteroids can also have rapid effects on inflammation through nongenomic mechanisms such as the activation of endothelial nitric oxide synthetase (23).

Inhaled corticosteroids (beclomethasone dipropionate, budesonide, fluticasone and mometasone) can be delivered to the lungs and remain in the lungs for some time after intratracheal instillation (24). They may have local anti-inflammatory activity; however, they also enter the systemic circulation by absorption from the lungs and the GI tract after oropharyngeal deposition during inhalation. Thus, inhaled corticosteroids may have both local and systemic anti-inflammatory effects.

Why it is important to do this overview

The role of corticosteroids in the prevention of BPD is controversial (25,26). Based on concerns with respect to their long-term impact on neurodevelopment, firm recommendations were made in 2002 by paediatric societies that steroids not be used outside of well-designed trials and in exceptional circumstances (25,27). These recommendations were more recently somewhat relaxed (26,28). Despite the earlier strong recommendations, and the lack of information regarding optimal dosing and timing, clinicians continue to use systemic corticosteroids to treat preterm infants dependent on ventilators (29) albeit generally less often than previously (30). The large number of randomized controlled trials examining the use of corticosteroids for the prevention of BPD in preterm infants has had widely variant methodologies and has been combined into multiple Cochrane reviews based on timing and/or delivery method of the treatment. It is difficult for clinicians to weigh the risks and benefits of treatment when faced with several reviews for consideration. We aim to provide the currently available evidence of corticosteroids for the prevention of BPD in preterm infants from the Cochrane Database of Systematic Reviews (CDSR) in a comprehensive way, synthesized in one accessible paper to assist the clinician in making an evidence-based plan for care of their patients.

Objectives

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

The purpose of this overview is to critically evaluate the evidence available in the CDSR regarding the efficacy and safety of corticosteroids in preventing BPD in premature infants.

Methods

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

Criteria for considering reviews for inclusion

All Cochrane reviews on corticosteroids for the prevention of BPD in preterm infants were considered for inclusion. Reviews on other treatment modalities of BPD were excluded.

Search methods for the identification of reviews

The CDSR was searched to identify reviews of corticosteroids for BPD in preterm infants. The search term ‘corticosteroids’ was used in the CDSR search engine using the ‘Title, Abstract or Keyword’ search field. The search was performed on March 24, 2013. One author (SA) identified potentially relevant systematic reviews based on the inclusion criteria described above. All authors discussed the reviews identified and made final decisions regarding inclusion. All reviews were considered for inclusion regardless of date of last assessment.

Types of outcome measures

The following outcomes were selected by the authors for inclusion in this overview prior to its beginning:

  • 1.
    Mortality at 28 days postnatal age (PNA)
  • 2.
    Mortality at 36 weeks PMA
  • 3.
    BPD at 28 days PNA
  • 4.
    BPD at 36 weeks PMA
  • 5.
    Death or BPD at 28 PNA
  • 6.
    Death or BPD at 36 weeks PMA
  • 7.
    Cerebral palsy (CP)
  • 8.
    Neurodisability (Bayley II Mental Developmental Index (MDI) < 2 standard deviations (SD) or Psychomotor Developmental Index (PDI) < 2SD)
  • 9.
    Hypertension
  • 10.
    GI perforation
  • 11.
    Hypothalamic–pituitary–adrenal axis suppression

Other outcomes such as duration of assisted ventilation, intestinal bleeding, necrotizing enterocolitis, sepsis, duration of hospital stay and neurological assessment by clinical exam were not evaluated. These outcomes were considered to be more subjective and less clinically meaningful with respect to long-term effects.

Data collection and analysis

One investigator (SA) independently extracted and a second reviewer (MO) verified the following data from included systematic reviews: review characteristics (population, intervention, comparison and outcomes of interest); numerical data (results of each individual review) and methodological quality (risk of bias) of included trials as assessed in each review. A predefined data collection form was used.

We present effect estimates as reported in the reviews. Risk ratios (RR) with 95% confidence intervals (CI) were used to describe dichotomous data. Results of meta-analyses were presented with related I2 values, which indicate statistical heterogeneity across trials. Within one review (31), a fixed-effects model was used in the absence of statistical heterogeneity and otherwise a random-effects model was employed; in the other six reviews (3237), a fixed-effects model was used regardless of reported statistical heterogeneity. To quantify the treatment effect for outcomes that were statistically significant we calculated number needed to harm (NNH) or number needed to treat (NNT) according to the formulae below (38):

  • equation image
  • equation image

Results

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

Description of included reviews

The search of the CDSR identified 310 reviews. After screening the titles and abstracts, eight reviews were considered potentially relevant. One was excluded because it was a protocol (39) and the other seven were included (3137).

One of these seven reviews was a subgroup analysis (33) of studies contained in another review (32); therefore, it was not included in the characteristics of reviews table. The remaining six included reviews (31, 32, 3437) contained 67 trials and 6535 participants. Characteristics of the included reviews are shown in Table 1. Only outcomes included in our overview are listed in the table. Reviews were divided into three groups according to the type of comparison:

  • a
    Inhaled corticosteroids versus placebo or no treatment (two reviews) (31,35)
  • b
    Inhaled corticosteroids versus systemic steroids (two reviews) (36,37)
  • c
    Systemic corticosteroids versus placebo or no treatment (two reviews) (32,34)
Table 1. Characteristic of included reviews
Title of the review (reference); assessed up to dateNo. of trials (patients)PopulationIntervention and comparisonOutcomes included in overview
  1. BSID-II: Bayley Scales of Infant Development—Second Edition; BW: birthweight; BPD: bronchopulmonary dysplasia; CP: cerebral palsy; DXM: dexamethasone; GA: gestational age; GI: gastrointestinal; MDI: mental developmental index; NEC: necrotizing enterocolitis; PDI: Psychomotor Developmental Index; PNA: postnatal age; PMA: post-menstrual age; VLBW: very low birthweight.

Inhaled corticosteroid versus placebo/no treatment
Early administration of inhaled corticosteroids for preventing chronic lung disease in ventilated very low birthweight preterm neonates (35); 29 July, 20117 (492)Ventilated preterm infants with birthweight <1500 g within the first 2 weeks of lifeInhaled corticosteroid therapy versus placebo or no treatmentBPD at 36 weeks PMA; death by 36 weeks PMA; death by or BPD at 36 weeks PMA; CP; death or CP; death before follow up for CP; CP in survivors; MDI among survivors on BSID-II < 2SD of the mean; BPD at 28 days of age; death by 28 days of age; death by or BPD at 28 days of age; MDI; hypertension.
Late (≥7 days) inhalation corticosteroids to reduce bronchopulmonary dysplasia in preterm infants (31); 12 March, 20128 (232)Ventilator or supplemental oxygen dependant preterm infants ≥7 days or <36 weeksInhaled corticosteroids versus placeboMortality at 28 days PNA; mortality at 36 weeks PMA; BPD at 28 days PNA; BPD at 36 weeks PMA; mortality or BPD at 28 days PNA; mortality or BPD at 36 weeks PMA.
Inhaled corticosteroids versus systemic corticosteroids
Inhaled versus systemic corticosteroids for preventing chronic lung disease in ventilated very low birthweight preterm neonates (36); 23 June, 20112 (294)Ventilated VLBW infants (<1500 g) or GA < 32 weeks in first 2 weeks of lifeInhaled corticosteroid therapy versus systemic corticosteroidsBPD at 36 weeks PMA; BPD at 28 days; death by 28 days PNA ; death by 36 weeks PMA; death or BPD by 28 days; death or BPD by 36 weeks PMA; hypertension; GI perforation
Inhaled versus systemic corticosteroids for the treatment of chronic lung disease in ventilated very low birthweight preterm infants (37); 23 June, 20113 (431)Ventilated neonates with BW <1500 g or GA < 32 weeks after 2 weeks of life with evolving BPDInhaled corticosteroids for the treatment of evolving BPD versus systemic corticosteroidsBPD at 36 weeks PMA; BPD at 28 days of age; death by 36 weeks PMA; death by 28 days of age; death or BPD at 36 weeks PMA; death or BPD at 28 days of age; hypertension; GI perforation
Systemic corticosteroid versus placebo/no treatment
Early (<8 days) postnatal corticosteroids for preventing chronic lung disease in preterm infants (34); 9 September, 200828 (3740)Ventilator dependent in the first 7 days of life preterm infants at risk of developing BPDIntravenous or oral corticosteroids versus placebo or no treatmentNeonatal mortality (up to 28 days); BPD at 28 days; BPD at 36 weeks; BPD or death at 28 days; BPD or death at 36 weeks; CP; CP in 1–3 years of life; CP in latest reported age; CP in survivors; CP or death at latest reported age; death before follow up for CP; MDI; Bayley PDI; hypertension; GI perforation
Late (>7 days) postnatal corticosteroids for chronic lung disease in preterm infants (32); 5 November, 200819 (1345)Preterm infants with oxygen or ventilator dependant beyond 7 days of life with or without radiographic changes of BPDDXM in initial dose of 0.5–1 mg/kg/day versus placebo or nothingNeonatal mortality before 28 days; mortality at latest reported age; BPD at 28 days; BPD at 36 weeks; BPD or death at 28 days; BPD or death at 36 weeks; CP; CP in 1–3 years of life; CP in latest reported age; CP in survivors; CP or death; death before follow up for CP; MDI; Bayley PDI; hypertension; GI perforation.

All studies included in these reviews were randomized controlled trials. We considered initiation of intervention <8 days of PNA as early and >7 days of PNA as late administration. Three reviews (3436) were classified as early intervention reviews. The other three reviews (31,32,37) were classified as late administration.

The duration and dosing of corticosteroid, as well as the specific drug used varied among the reviews and included studies. Dexamethasone was administered intravenously or orally in doses ranging from 0.04 to 2 mg/kg/day for 3–42 days. Hydrocortisone was administered intravenously in doses from 0.5 mg/kg/day to 30.00 mg/kg/day for 1–15 days. Some trials used short courses of a consistent dose, while others used prolonged regimens with tapering doses. Open label corticosteroid use was reported in some included studies in both reviews of systemic corticosteroids.

Inhaled corticosteroids studied included beclomethasone dipropionate (0.48–200 µg/kg/day for 7–28 days or 400–800 µg/day until 48 hours post extubation), budesonide (800–1200 µg/day for 3–28 days), fluticasone (500 µg/day for 2 weeks) and flunisolide (148–1500 µg/day for up to 4 weeks). Three of the reviews involving inhaled corticosteroids reported open label systemic corticosteroid use (31,35,37). In some included studies this was at the discretion of the treating physician and in other studies the protocols dictated that patients cross over to systemic corticosteroids under certain conditions.

Search methods used in included reviews

Four of the six reviews (31,3537) searched MEDLINE, EMBASE, CINAHL and the Cochrane Central Register of Controlled Trials (CENTRAL); two reviews (32,34) searched only CENTRAL and MEDLINE. Hand searching paediatric and perinatal journals and examining previous review articles was performed in all reviews; hand searching conference proceedings published in Pediatric Research was completed in four reviews (31,3537). Language restrictions were not applied in the same four reviews, while two reviews (32,34) did not specify if language restrictions were applied.

Primary outcomes in included reviews

Four reviews specified their primary outcome (31,3537) while two reviews (32,34) did not specify primary outcomes of interest. The primary outcomes were BPD at 36 weeks PMA in three reviews (3537) and mortality at 36 weeks PMA and BPD defined as oxygen dependency at 36 weeks PMA in one review (31).

Other outcomes that were commonly reported include BPD, mortality at different time points, BPD/mortality by 28 days of age, BPD/mortality by 36 weeks PMA, failure to extubate, requirement for systemic steroids, infection, hyperglycaemia, hypertension, glycosuria, GI bleeding, echodensities on ultrasound scan of brain, cataracts, intraventricular haemorrhage (IVH), periventricular leukomalacia (PVL), necrotizing enterocolitis (NEC), retinopathy of prematurity (ROP), pituitary–adrenal suppression, patent ductus arteriosus (PDA) and long-term neurodevelopmental outcomes. Only pre-specified outcomes of interest detailed above were extracted from these reviews for inclusion and analysis in the overview.

Subgroup analyses in included reviews

Subgroup analyses were conducted in some of the reviews: one review (34) performed subgroup analyses by the type of corticosteroid used (dexamethasone or hydrocortisone) for those outcomes where there were sufficient numbers of trials to make such subgroup analyses meaningful. Three reviews (3537) performed subgroup analysis of BPD at 36 weeks PMA by all randomized patients and by survivors only. One review (31) performed subgroup analyses by ventilated and non-ventilated patients at trial entry.

Methodological quality of included trials

Different aspects of the Cochrane Risk of Bias tool were used to evaluate the methodological quality of included trials. Four reviews (31,3537) assessed all domains: adequacy of sequence generation, allocation concealment, blinding, incomplete outcome data, selective reporting and other risks of bias. Two reviews (32,34) assessed sequence generation, allocation concealment, blinding and incomplete outcome data. In five reviews (32,3437) blinding of intervention and blinding of outcome assessment were assessed separately; one review (31) did not separate these two aspects of blinding.

Risk of Bias assessments are summarized in Table 2. Sequence generation was evaluated in 14 trials and was unclear in the majority (64%). Allocation concealment was assessed in all 67 trials and was deemed low risk in 85%. Blinding of intervention and blinding of outcome assessment were evaluated in 59 trials and were most commonly assessed as low risk (75% and 64%, respectively). In seven trials, blinding in general was assessed; all of these trials were considered low risk. Incomplete outcome data was deemed to be low risk in the majority of trials (88%). Selective reporting and other sources of bias were evaluated in eight and four trials, respectively, and were considered low risk in 63% and 75%, respectively; none of the trials were considered high risk with respect to selective reporting and other sources of bias.

Table 2. Details of risk of bias assessment in the included trials
Risk of bias dimensionLow risk of bias, n (%)Unclear risk of bias, n (%)High risk of bias, n (%)Number of trials assessed
Sequence generation4 (29)9 (64)1 (7)14
Allocation concealment57 (85)9 (13)1 (2)67
Blinding7 (100)0 (0)0 (0)7
Blinding of intervention44 (75)6 (10)9 (15)59
Blinding of outcome assessment38 (64)13 (22)8 (14)59
Incomplete outcome data59 (88)6 (9)2 (3)67
Selective reporting5 (63)3 (37)0 (0)8
Other sources of bias3 (75)1 (25)0 (0)4

In addition to the assessment of the risk of bias domains, the following limitations were noted in many of the reviews: open label use; cross over to the active intervention under certain conditions; inadequate design for long-term follow up, including insufficient statistical power to detect differences in long-term outcomes; inconsistency in assessing CP (e.g. variable timing of outcome assessment and definitions); and variation in the treatment regimen (i.e. dose, timing, duration).

Effects of interventions

Mortality

All reviews reported mortality at 28 days PNA while four reviews (31,3537) also reported mortality at 36 weeks PMA Table 3. Late systemic corticosteroid treatment compared with placebo was associated with reduced mortality at 28 days (RR 0.49, 95% CI 0.28–0.85). The NNT is 19 as calculated by the formula stated above. No statistical heterogeneity was noted. There was no difference in mortality at 28 days for the inhaled versus systemic corticosteroids or inhaled versus placebo/no treatment comparisons. There was no difference in mortality at 36 weeks PMA for early or late inhaled corticosteroids versus placebo and inhaled versus systemic corticosteroids. One review, which compared early systemic corticosteroid with placebo (34), conducted a subgroup analysis based on the type of corticosteroid (dexamethasone or hydrocortisone) and did not show a significant difference at 28 days PNA within either subgroup (dexamethasone RR 1.06, 95% CI 0.90–1.24, 16 studies and 2603 infants; hydrocortisone RR 0.78, 95% CI 0.50–1.23, 3 studies and 347 infants).

Table 3. Mortality
Timing of start of interventionAt 36 weeks PMAAt 28 days PNA
No. of trials (patients)RR (95% CI)I2 (%)No. of trials (patients)RR (95% CI)I2 (%)
  1. NA: Not applicable.

  2. Bold values indicate where results were statistically significant.

Inhaled corticosteroids versus placebo
  Early5 (429)0.73 (0.44, 1.21)385 (429)0.66 (0.39, 1.14)0
  Late3 (61)3.0 (0.35, 25.78)02 (51)3.0 (0.14, 65.90)0
Inhaled versus systemic corticosteroids
  Early2 (294)0.83 (0.56, 1.23)02 (294)0.80 (0.51, 1.25)0
  Late1 (78)2.69 (0.13, 54.15)NA1 (78)2.69 (0.13, 54.15)NA
Systemic corticosteroids versus placebo/no treatment
  Early19 (2950)1.02 (0.88, 1.19)7
  Late8 (656)0.49 (0.28, 0.85)0
Bronchopulmonary dysplasia

There was significantly less incidence of BPD at 28 days for late systemic corticosteroid treatment compared with placebo (RR 0.87, 95% CI 0.81–0.94). A similar reduction was seen at 36 weeks PMA (RR 0.72, 95% CI 0.61–0.85). NNT were 9 and 6 for reduction at 28 days and 36 weeks, respectively. Both outcomes showed moderate heterogeneity. The authors did not comment on reasons for heterogeneity.

Early systemic corticosteroids versus placebo decreased the incidence of BPD at both 28 days (RR 0.87, 95% CI 0.81–0.93) and 36 weeks PMA (RR 0.79, 95% CI 0.71–0.88). NNT were 16 and 15, respectively. As above, there was moderate heterogeneity; however, authors did not examine or comment on potential reasons. There was also a reduction in BPD incidence at 36 weeks PMA in survivors (RR 0.82, 95% CI 0.74–0.90) with a NNT of 14.

A subgroup analysis of infants treated with early systemic steroids included in one review (34) showed a significant difference for dexamethasone but not for hydrocortisone, at both 28 days PNA (dexamethasone: RR 0.85, 95% CI 0.79–0.92, 16 studies and 2621 infants; hydrocortisone: RR 1.00, 95% CI 0.85–1.18, 1 study and 253 infants) and 36 weeks PMA (dexamethasone: RR 0.70, 95% CI 0.61–0.81; 15 studies and 2484 infants; hydrocortisone: RR 0.96, 95% CI 0.82–1.12; 6 studies and 802 infants).

The other comparisons (inhaled steroids versus placebo or inhaled versus systemic steroids for both early and late administration) did not show a significant reduction in BPD incidence. Table 4 provides details of the BPD outcome.

Table 4. Bronchopulmonary dysplasia
Timing of start of interventionAt 36 weeks PMAAt 28 days PNA
No. of trials (patients)RR (95% CI)I2 (%)No. of trials (patients)RR (95% CI)I2 (%)
  1. NA: Not applicable.

  2. Bold values indicate where results were statistically significant.

Inhaled corticosteroids versus placebo
  Early5 (429)0.97 (0.62, 1.52)115 (429)1.05 (0.84, 1.32)4
  Late1 (30)1.0 (0.59, 1.70)NA1 (30)0.93 (0.72, 1.21)0
Inhaled versus systemic corticosteroids
  Early1 (278)1.45 (0.99, 2.11)NA2 (294)1.21 (0.98, 1.48)71.7
  Late2 (139)1.02 (0.83, 1.25)801 (78)0.97 (0.90, 1.05)NA
Systemic corticosteroids versus placebo/no treatment
  Early21 (3286)0.79 (0.71, 0.88)4117 (2874)0.87 (0.81, 0.93)62
  Late8 (471)0.72 (0.61, 0.85)526 (623)0.87 (0.81, 0.94)53
Death or BPD

Composite outcomes, such as the combined outcome of death or BPD are often used to ensure that a decrease in one outcome is not at the expense of another (i.e. a benefit from decreased death but a harm from resultant increased BPD). Early corticosteroids compared with placebo reduced the incidence of death or BPD at 28 days PNA (RR 0.92, 95% CI 0.88–0.96) and 36 weeks PMA (RR 0.89, 95% CI 0.84–0.95). The NNT was 18 for both outcomes. There was moderate to substantial heterogeneity respectively for these estimates. The review authors (34) did not explore reasons for heterogeneity. Subgroup analyses based on the type of medication showed a significant difference for dexamethasone but not for hydrocortisone at both 28 days (dexamethasone: RR 0.91, 95% CI 0.86–0.96; 14 studies and 2293 infants; hydrocortisone: RR 1.00, 95% CI 0.90–1.12; 1 study and 253 infants) and 36 weeks PMA (dexamethasone: RR 0.87, 95% CI 0.80–0.94; 15 studies and 2484 infants; hydrocortisone: RR 0.95, 95% CI 0.86–1.06; 7 studies and 836 infants).

Late administration of systemic corticosteroids also demonstrated decreased mortality or BPD compared with placebo at both 28 days (RR 0.84, 95% CI 0.78–0.89, NNT 7) and 36 weeks PMA (RR 0.72, 95% CI 0.63–0.82, NNT 5). The heterogeneity was moderate and substantial, respectively, in these estimates; however, reasons were not explored by review authors (32). The results for death or BPD are detailed in Table 5.

Table 5. Death or bronchopulmonary dysplasia
Timing of start of interventionAt 36 weeks PMAAt 28 days PNA
No. of trials (patients)RR (95% CI)I2 (%)No. of trials (patients)RR (95% CI)I2 (%)
  1. NA: Not applicable.

  2. Bold values indicate where results were statistically significant.

Inhaled corticosteroids versus placebo
  Early5 (429)0.86 (0.63, 1.17)05 (429)0.96 (0.80, 1.14)0
  Late1 (30)1.10 (0.74, 1.63)NA1 (30)1.00 (0.85, 1.18)NA
Inhaled versus systemic corticosteroids
  Early1 (278)1.09 (0.88, 1.35)NA2 (294)1.05 (0.93, 1.20)78
  Late1 (78)0.94 (0.83, 1.05)NA1 (78)1.00 (0.94, 1.06)NA
Systemic corticosteroids versus placebo/no treatment
  Early22 (3320)0.89 (0.84, 0.95)7216 (2548)0.92 (0.88, 0.96)43
  Late8 (471)0.72 (0.63, 0.82)625 (563)0.84 (0.78, 0.89)40
Cerebral palsy

Early use of systemic corticosteroids showed an increased incidence of CP compared with placebo/no treatment both overall (RR 1.45, 95% CI 1.06–1.98, NNH 26) and in survivors (RR 1.50, 95% CI 1.13–2.00, NNH 15). For the combined outcome of death or CP, the result was not statistically significant (RR 1.09, 95% CI 0.95–1.25). A subgroup analysis contained in the original review (34) showed that CP and the combined outcome of death or CP was statistically significant with dexamethasone but not hydrocortisone (CP: dexamethasone RR 1.75, 95% CI 1.20–2.55, 7 studies and 921 infants; hydrocortisone RR 0.97, 95% CI 0.55–1.69, 5 studies and 531 infants; death or CP: dexamethasone RR 1.17, 95% CI 1.00–1.37, 7 studies and 921 infants; hydrocortisone RR 0.91, 95% CI 0.70–1.19, 5 studies and 531 infants). No other significant differences were found between early or late systemic corticosteroids versus placebo/no treatment for the other methods or time points used to assess for CP (Table 6). Further no significant difference was found in CP among survivors for early inhaled corticosteroid versus placebo.

Table 6. Cerebral palsy (CP) and related outcomes
Comparison and outcomeNo. of trials (patients)RR (95% CI)I2 (%)
  1. NA: Not applicable.

  2. Bold values indicate where results were statistically significant.

Early systemic corticosteroid versus placebo/no treatment
  CP12 (1452)1.45 (1.06, 1.98)22
  CP in survivors assessed12 (959)1.50 (1.13, 2.00)41
  Death before follow up for CP (at latest reported age)12 (1452)0.96 (0.81, 1.14)0
  Death or CP12 (1452)1.09 (0.95, 1.25)29
Late systemic corticosteroid versus placebo/no treatment
  CP in first 3 years of life11 (777)1.14 (0.79, 1.64)0
  CP at latest reported age12 (756)1.22 (0.84, 1.77)0
  CP in survivors assessed (at 1–3 years of age)11 (559)1.10 (0.77, 1.56)0
  CP in survivors assessed (at latest reported age)12 (519)1.18 (0.82, 1.70)0
  Death before follow up for CP (at 1–3 years of age)11 (777)0.85 (0.64, 1.13)7
  Death before follow up for CP (at latest reported age)12 (756)0.84 (0.62, 1.12)0
  Death or CP at 1–3 years11 (777)0.96 (0.78, 1.17)0
  Death or CP at latest reported age12 (756)0.98 (0.80, 1.21)0
Early inhaled corticosteroid versus placebo
  CP in survivors assessed1 (56)1.33 (0.33, 5.42)NA
Neurodisability

Neurodisability assessed by Mental Developmental Index (MDI) or Psychomotor Developmental Index (PDI) tested on the Bayley Scales of Infant Development II (BSID-II) showed no difference between groups either overall or in survivors tested for early or late systemic corticosteroid treatment versus placebo/no treatment. Similarly, no difference in MDI in survivors was found for inhaled corticosteroids versus placebo. Results are described in Table 7.

Table 7. Neurodisability
Comparison and outcomeNo. of trials (patients)RR (95% CI)I2 (%)
  1. NA: Not applicable.

Early systemic corticosteroid versus placebo/no treatment
  Bayley MDI ≤ 2SD3 (842)0.99 (0.77, 1.27)56
  MDI among survivors tested on BSID-II < 2SD of the mean3 (528)0.98 (0.79, 1.23)61
  Bayley PDI ≤ 2SD3 (842)1.16 (0.86, 1.58)0
  Bayley PDI ≤ 2SD in survivors tested3 (528)1.16 (0.87, 1.54)50
Late systemic corticosteroid versus placebo/no treatment
  Bayley MDI ≤ 2SD1 (118)1.25 (0.45, 3.49)NA
  MDI among survivors tested on BSID-II < 2SD of the mean1 (89)1.04 (0.38, 2.86)NA
  Bayley PDI ≤ 2SD1 (118)0.78 (0.34, 1.80)NA
  Bayley PDI ≤ 2SD in survivors tested1 (90)0.67 (0.30, 1.50)NA
Early inhaled corticosteroid versus placebo
  MDI among survivors tested on BSID-II < 2SD of the mean1 (56)1.25 (0.37, 4.17)NA
Hypertension

Early and late use of systemic corticosteroid therapy versus placebo was associated with increased incidence of hypertension (early RR 1.85, 95% CI 1.55–2.22, NNH 10; late RR 2.66, 95% CI 1.58–4.49, NNH 21). Heterogeneity was moderate (43%) in the early treatment comparison; there was no heterogeneity in the late treatment comparison. Within the early systemic steroid versus placebo comparison the results remained statistically significant for dexamethasone (RR 1.84, 95% CI 1.54–2.21, 10 studies and 1946 infants) but not for hydrocortisone (RR 3.00, 95% CI 0.33–26.92, 1 study and 50 infants). Results for hypertension are detailed in Table 8.

Table 8. Hypertension
Timing of start of interventionNo. of trials (patients)Effect estimate RR (95% CI)I2 (%)
  1. NA: Not applicable.

  2. Bold values indicate where results were statistically significant.

Inhaled corticosteroid versus placebo
  Early3 (116)1.20 (0.36, 3.99)72
  Late1 (27)0.0 (0.0, 0.0)NA
Inhaled corticosteroids versus systemic
  Early1 (278)0.76 (0.44, 1.29)NA
  Late2 (139)0.70 (0.31, 1.58)0
Systemic corticosteroid versus placebo/no treatment
  Early11 (1996)1.85 (1.55, 2.22)43
  Late12 (1096)2.66 (1.58, 4.49)0
GI perforation

GI perforation was reported in the reviews of inhaled versus systemic corticosteroids (36) (early only) and systemic steroids versus placebo (32,34) (Table 9). Early systemic steroids versus placebo showed a significant increase in GI perforation (RR 1.81, 95% CI 1.33–2.48, NNH 29). Results remained significant in subgroup analyses by type of corticosteroid (dexamethasone RR 1.73, 95% CI 1.20–2.51, 9 studies and 1340; hydrocortisone RR 2.02, 95% CI 1.13–3.59, 6 studies and 583 infants).

Table 9. Gastrointestinal perforation
Timing of start of interventionNo. of trials (patients)Effect estimate RR (95% CI)I2 (%)
  1. NA: Not applicable.

  2. Bold values indicate where results were statistically significant.

Inhaled corticosteroids versus systemic
  Early1 (278)0.16 (0.02, 1.29)NA
Systemic corticosteroid versus placebo/no treatment
  Early15 (2523)1.81 (1.33, 2.48)0
  Late2 (95)0.36 (0.02, 8.05)0
Hypothalamic–pituitary–adrenal axis suppression

One study within the review (37) comparing inhaled versus systemic corticosteroids for treatment of BPD found that baseline cortisol levels for the high dose (800 µg/day) inhaled beclomethasone group and the systemic group were lower than for the low dose (400 µg/day) inhaled group; however, the levels after stimulation were similar in all three groups. Comparing inhaled corticosteroids to placebo (35) beclomethasone was found in one study to suppress basal cortisol levels but response to stimulation was normal, whereas fluticasone was found, again in one study, to suppress both basal and post-stimulation levels compared with control.

Discussion

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

The use of corticosteroids in the postnatal period to prevent or treat BPD in high-risk preterm infants has been highly controversial. It is difficult for clinicians to determine whether the short-term benefits of corticosteroids outweigh their potential for adverse effects. This overview presents the available evidence from Cochrane reviews to assist clinicians in weighing the evidence and making a decision for their patient.

Summary of main results

The main findings from the included systematic reviews are synthesized in Table 10. Late administration of systemic corticosteroids versus placebo reduces mortality at 28 days compared to placebo. Early and late systemic corticosteroids reduce BPD at 28 days and 36 weeks and decrease the composite outcome of death or BPD at both time points compared with placebo. Early administration of corticosteroids increases CP both overall and in survivors compared with placebo. Additional short-term effects of early corticosteroids compared with placebo include increased rates of hypertension and GI perforation. The increased risk of hypertension is also true of late administration compared with placebo. There is no evidence, in the included reviews, that inhaled corticosteroids decrease either BPD or mortality and while no increased risk of adverse effects were found, small sample sizes for important outcomes and lack of reporting on other outcomes, particularly on long-term outcomes, do not allow for strong conclusions regarding safety.

Table 10. Summary of main findingsThumbnail image of

Limitations of the overview

One of the challenges in this overview was the lack of consistency in included reviews with respect to patient inclusion criteria and corticosteroid drug, dosing and timing. There was moderate to substantial statistical heterogeneity for a number of outcomes, which may be attributable to variation across studies in some of these clinical variables. Given this, it is not possible, with the evidence available in the included reviews, to draw conclusions about ideal timing of therapy, choice of drug, route or dosing. Additionally, in many trials within the reviews involving inhaled corticosteroids, there was open label use of corticosteroids, which may have resulted in performance and/or detection bias. There was a high degree of variability in the included studies with respect to reporting on significant side effects and long-term outcomes. Variation in definition, timing of diagnosis and neurodevelopmental indicators reported made direct comparisons difficult. For the long-term outcomes of particular importance to clinicians and families, there were often smaller numbers of patients reported, both due to fewer studies reporting on the particular outcome and due to significant loss to follow up in some trials. The timing of diagnosis for CP was often before school age, which may affect the accuracy of the diagnosis. In the case of comparisons involving inhaled corticosteroids, there was only follow-up data from one small trial without sufficient power to show a difference. Included trials of systemic corticosteroid treatment were not largely designed with long-term outcomes in mind and none was powered to detect differences in these important outcomes. Together, all of these limitations do not allow for a broad assessment of safety of inhaled or systemic corticosteroid therapy.

Authors' Conclusions

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

Implications for practice

There are many gaps for clinicians in the current evidence available for the efficacy and safety of corticosteroid use for the prevention and treatment of BPD in preterm infants. Early corticosteroid use was shown to decrease the incidence of BPD and death or BPD but not mortality. Early use was also shown, in a large number of studies and patients, to have an increased risk of CP, the most important long-term outcome studied. No review compared early versus late administration of corticosteroids in order to determine whether there is an advantage to early administration with respect to decreasing BPD incidence or the composite outcome of death or BPD. Given that there is no evidence for added benefit of early compared to late administration, and that there is clear evidence for harm, early administration of systemic corticosteroids for the prevention of BPD in preterm infants cannot be recommended.

Late administration of systemic corticosteroids decreases death at 28 days PNA. By definition, corticosteroids in this case are not preventing death due to BPD, which cannot be diagnosed until at least 28 days PNA, but due to some other cause. This may include preventing early death from respiratory illness which may have evolved into BPD. Our pre-specified outcome of death at 36 weeks was not included in the review of late corticosteroids. The review did report on mortality before discharge and mortality at the latest reported age and found no difference at these time points compared to placebo/no treatment. It is therefore difficult to advocate late administration of corticosteroids on the basis of decreasing mortality. Late administration did decrease the incidence of BPD and death or BPD; however, again we only have short-term data of 28 days or 36 weeks. This potential benefit must be weighed by the clinician against the increased risk of hypertension and the uncertain risk to long-term neurodevelopment. If a clinician decides to administer late corticosteroids, there is insufficient evidence in this overview, based on Cochrane reviews, to suggest the optimal drug, dose or duration of therapy.

With respect to inhaled corticosteroids, no evidence of efficacy for the prevention or treatment of BPD in preterm infants was found. This, coupled with a paucity of evidence concerning the safety of inhaled corticosteroids in preterm infants, leads to the conclusion that inhaled corticosteroids cannot be recommended, based on the evidence in this overview, for use in the treatment or prevention of BPD.

Implications for research

Studies are needed to examine the most appropriate dose, timing and route of administration of corticosteroids to yield maximum benefit in decreasing BPD and death while avoiding short- and long-term harms. Well-designed randomized controlled trials, enrolling patients beyond the first week of life, curtailing open label corticosteroid use and with a focus on long-term outcomes are desperately needed. Many major questions remain to be answered: Which patients are at highest risk of not only BPD but also adverse long-term outcome and would benefit most from treatment? Which systemic corticosteroid is most effective and also safe? What is the optimal timing, dosing and duration of therapy (40)?

Interestingly, it is not even clear whether low or high doses of systemic steroids are better and protocols comparing them head to head are needed. Lower doses are recommended by some (28) as it appears logical that lower doses should have fewer short- and long-term side effects. A systematic review of the effects of cumulative dose of dexamethasone on mortality and pulmonary and neurodevelopmental sequelae in preterm infants raises conflicting evidence (41). They found that higher doses were more effective at decreasing the composite outcome of death or BPD when given beyond the first week. Contrastingly, they found that higher doses given between 7 and 14 days of life decreased the risk of mortality or CP and when administered beyond 3 weeks of life no effect of dose on neurodevelopmental outcome was found. This seemingly counterintuitive protective effect of higher doses in the second week of life may be because these doses are more effective at preventing BPD which itself carries a risk of CP and adverse neurodevelopmental outcome independent of receipt of corticosteroids.

Properly designed trials of inhaled corticosteroids, again focusing on long-term risks and benefits are also needed. Despite the lack of evidence for either their efficacy or safety, there are recommendations that may be considered for use as an alternative to dexamethasone to treat BPD (28). Evidence to support or refute this recommendation is urgently needed.

After the firmly worded positions of paediatric societies in 2002 (25,27) that steroids not be used outside of well-designed trials and in exceptional circumstances, it was extremely difficult for even well-designed trials to succeed (42). Perhaps the more recent relaxation of these recommendations, by the same professional societies, will enable future studies to enrol the patients required to answer these important research questions (26,28).

Contributions of Authors

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References

All authors contributed to preparing the protocol and conducting the overview. JH, SA, MO and LH prepared the manuscript and TML provided a critical review. SA performed data extraction and analysis. MO verified data extraction and helped develop tables. JH and TML provided clinical and methodological input throughout the overview. LH supervised the project. All authors approved the final version and take responsibility for the manuscript.

References

  1. Top of page
  2. Abstract
  3. Plain language summary
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' Conclusions
  10. Contributions of Authors
  11. Declarations of interest
  12. References
  • 1
    Ehrenkranz RA, Walsh MC, Vohr BR, Jobe AH, Wright LL, Fanaroff AA, et al. Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics 2005; 116: 13531360.
  • 2
    Jobe AH, Bancalari E. Bronchopulmonary dysplasia. American Journal of Respiratory and Critical Care Medicine 2001; 163: 17231729. Epub June 13, 2001.
  • 3
    Northway WH, Rosan RC, Porter DY. Pulmonary disease following respirator therapy of hyaline-membrane disease—Bronchopulmonary dysplasia. The New England Journal of Medicine 1967; 276: 357368.
  • 4
    Greenough A. Prenatal factors in the development of chronic lung disease. Seminars in Fetal and Neonatal Medicine 2009; 14: 339344.
  • 5
    Merritt TA, Deming DD, Boynton BR. The 'new' bronchopulmonary dysplasia: Challenges and commentary. Seminars in Fetal and Neonatal Medicine 2009; 14: 345357.
  • 6
    Philip AGS. Chronic lung disease of prematurity: A short history. Seminars in Fetal and Neonatal Medicine 2009; 14: 333338.
  • 7
    Askie LM, Ballard RA, Cutter G, Dani C, Elbourne D, Field D, et al. Inhaled nitric oxide in preterm infants: A systematic review and individual patient data meta-analysis. BMC Pediatrics 2010; 10: 15.
  • 8
    Bancalari E, Abdenour GE, Feller R, Gannon J. Bronchopulmonary dysplasia: Clinical presentation. The Journal of Pediatrics 1979; 95(5 Pt 2): 819823. Epub November 1, 1979.
  • 9
    Walsh MC, Wilson-Costello D, Zadell A, Newman N, Fanaroff A. Safety, reliability, and validity of a physiologic definition of bronchopulmonary dysplasia. Journal of Perinatology : Official Journal of the California Perinatal Association 2003; 23: 451456. Epub September 19, 2003.
  • 10
    Kramer BW, Lievense S, Been JV, Zimmermann LJ. From classic to new bronchopulmonary dysplasia [Van klassieke naar nieuwe bronchopulmonale dysplasie[. Nederlands tijdschrift voor geneeskunde 2010; 154: A1024. Epub May 12, 2010.
  • 11
    Fanaroff AA, Stoll BJ, Wright LL, Carlo WA, Ehrenkranz RA, Stark AR, et al. Trends in neonatal morbidity and mortality for very low birthweight infants. American Journal of Obstetrics and Gynecology 2007; 196: 147 e1147 e8. Epub February 20, 2007.
  • 12
    Stoll BJ, Hansen NI, Bell EF, Shankaran S, Laptook AR, Walsh MC, et al. Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics 2010; 126: 443456. Epub August 25, 2010.
  • 13
    Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in premature-infants—Prediction from oxygen requirement in the neonatal-period. Pediatrics 1988; 82: 527532.
  • 14
    Lemons JA, Bauer CR, Oh W, Korones SB, Papile LA, Stoll BJ, et al. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Research Network, January 1995 through December 1996. Pediatrics 2001; 107: e1.
  • 15
    Doyle LW, Anderson PJ. Long-term outcomes of bronchopulmonary dysplasia. Seminars in Fetal and Neonatal Medicine 2009; 14: 391395.
  • 16
    Doyle LW, Ehrenkranz RA, Halliday HL. Dexamethasone treatment after the first week of life for bronchopulmonary dysplasia in preterm infants: A systematic review. Neonatology 2010; 98: 289296.
  • 17
    Gupta S, Prasanth K, Chen CM, Yeh TF. Postnatal corticosteroids for prevention and treatment of chronic lung disease in the preterm newborn. International Journal of Pediatrics 2012; 2012: 315642. Epub November 19, 2011.
  • 18
    Ryan RM, Ahmed Q, Lakshminrusimha S. Inflammatory mediators in the immunobiology of bronchopulmonary dysplasia. Clinical Reviews in Allergy & Immunology 2008; 34: 174190. Epub March 12, 2008.
  • 19
    Speer CP. Inflammation and bronchopulmonary dysplasia: a continuing story. Seminars in Fetal and Neonatal Medicine 2006; 11: 354362. Epub May 17, 2006.
  • 20
    Kallapur SG, Jobe AH. Contribution of inflammation to lung injury and development. Archives of Disease in Childhood. Fetal and Neonatal Edition 2006; 91: F132F135. Epub February 24, 2006.
  • 21
    Rhen T, Cidlowski JA. Antiinflammatory action of glucocorticoids—New mechanisms for old drugs. The New England Journal of Medicine 2005; 353: 17111723.
  • 22
    Laughon MM, Smith PB, Bose C. Prevention of bronchopulmonary dysplasia. Seminars in Fetal and Neonatal Medicine 2009; 14: 374382.
  • 23
    Hafezi-Moghadam A, Simoncini T, Yang Z, Limbourg FP, Plumier JC, Rebsamen MC, et al. Acute cardiovascular protective effects of corticosteroids are mediated by non-transcriptional activation of endothelial nitric oxide synthase. Nature Medicine 2002; 8: 473479. Epub May 2, 2002.
  • 24
    Jarreau PH, Fayon M, Baud O, Autret-Leca E, Danan M, de Verdelhan A, et al. The use of postnatal corticosteroid therapy in premature infants to prevent or treat bronchopulmonary dysplasia: Current situation and recommendations. Archives de Pediatrie 2010; 17: 14801487.
  • 25
    Blackmon LR, Bell EF, Engle WA, Kanto WP, Martin GI, Miller CA, et al. Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Pediatrics 2002; 109: 330338.
  • 26
    Watterberg KL, Papile LA, Adamkin DH, Baley JE, Bhutani VK, Carlo WA, et al. Policy statement-postnatal corticosteroids to prevent or treat bronchopulmonary dysplasia. Pediatrics 2010; 126: 800808.
  • 27
    Committee on Fetus and Newborn. Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Paediatrics & Child Health 2002; 7: 2046. Epub 1 January, 2002.
  • 28
    Jefferies AL, Soc CP, Soc CP, Comm FN. Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Paediatrics & Child Health 2012; 17: 573.
  • 29
    Shinwell ES, Eventov-Friedman S. Impact of perinatal corticosteroids on neuromotor development and outcome: Review of the literature and new meta-analysis. Seminars in Fetal and Neonatal Medicine 2009; 14: 164170.
  • 30
    Yates HL, Newell SJ. Postnatal intravenous steroids and long-term neurological outcome: Recommendations from meta-analyses. Archives of Disease in Childhood: Fetal & Neonatal 2012; 97: F299F303.
  • 31
    Onland W, Offringa M, van Kaam A. Late (> = 7 days) inhalation corticosteroids to reduce bronchopulmonary dysplasia in preterm infants. Cochrane Database of Systematic Reviews 2012; 4: CD002311.
  • 32
    Halliday HL, Ehrenkranz RA, Doyle LW. Late (>7 days) postnatal corticosteroids for chronic lung disease in preterm infants. Cochrane Database of Systematic Reviews 2009; 1: CD001145.
  • 33
    Halliday HL, Ehrenkranz RA, Doyle LW. Moderately early (7–14 days) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database of Systematic Reviews 2003; 1: CD001144. Epub January 22, 2003.
  • 34
    Halliday HL, Ehrenkranz RA, Doyle LW. Early (<8 days) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database of Systematic Reviews 2010; 1: CD001146. Epub January 22, 2003.
  • 35
    Shah VS, Ohlsson A, Halliday HL, Dunn M. Early administration of inhaled corticosteroids for preventing chronic lung disease in ventilated very low birth weight preterm neonates. Cochrane Database of Systematic Reviews 2012; 5: CD001969.
  • 36
    Shah SS, Ohlsson A, Halliday HL, Shah VS. Inhaled versus systemic corticosteroids for preventing chronic lung disease in ventilated very low birth weight preterm neonates. Cochrane Database of Systematic Reviews 2012; 5: CD002058.
  • 37
    Shah SS, Ohlsson A, Halliday HL, Shah VS. Inhaled versus systemic corticosteroids for the treatment of chronic lung disease in ventilated very low birth weight preterm infants. Cochrane Database of Systematic Reviews 2012; 5: CD002057.
  • 38
    McQuay HJ. Moore. Using numerical results from systematic reviews in clinical practice. Annals of Internal Medicine 1997; 126: 712720.
  • 39
    Han S, Yu Z, Guo X, Dong X, Soll R. Intratracheal instillation of corticosteroids using surfactant as a vehicle for the prevention of chronic lung disease in preterm infants with respiratory distress syndrome. Cochrane Database of Systematic Reviews 2011: protocol.
  • 40
    Thebaud B, Lacaze-Masmonteil T, Watterberg K. Postnatal glucocorticoids in very preterm infants: “The good, the bad, and the ugly”? Pediatrics 2001; 107: 413415.
  • 41
    Onland W, Offringa M, De Jaegere AP, van Kaam AH. Finding the optimal postnatal dexamethasone regimen for preterm infants at risk of bronchopulmonary dysplasia: a systematic review of placebo-controlled trials. Pediatrics 2009; 123: 36777. Epub January 2, 2009.
  • 42
    Doyle LW, Davis PG, Morley CJ, McPhee A, Carlin JB. 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: 716721.