INTRODUCTION — Clinical trials have shown that postnatal systemic corticosteroid therapy improves short-term lung function and pulmonary outcome of infants with established bronchopulmonary dysplasia (BPD) and reduces the risk of BPD in high-risk preterm infants. However, systemic corticosteroid administration (primarily dexamethasone) is associated with serious adverse effects (eg, increased risk of cerebral palsy [CP]). As a result, the potential benefits of routine administration of postnatal corticosteroids appear to be outweighed by its downsides. Nevertheless, low-dose systemic corticosteroid therapy may be beneficial in a select subset of patients (ie, those with evolving or established BPD who remain ventilator-dependent at two to four weeks after birth). (See "Bronchopulmonary dysplasia: Definition, pathogenesis, and clinical features", section on 'Epidemiology'.)
The use of postnatal corticosteroid therapy in preterm infants to prevent BPD will be reviewed here. Antenatal corticosteroid therapy, a broader discussion of strategies to prevent BPD, and management of established BPD are discussed separately:
●(See "Bronchopulmonary dysplasia: Prevention".)
●(See "Bronchopulmonary dysplasia: Management".)
DEFINITIONS
Bronchopulmonary dysplasia (BPD) — Clinically, BPD is defined by a requirement of oxygen supplementation either at 28 days postnatal age or 36 weeks postmenstrual age (PMA) [1-3]. More nuanced definitions that account for both the supplemental oxygen requirement and mode of ventilatory support are summarized in the table (table 1) and are discussed in greater detail separately. (See "Bronchopulmonary dysplasia: Definition, pathogenesis, and clinical features", section on 'Definitions'.)
The definition of BPD has evolved over time because of changes in the population at risk (ie, more surviving neonates at earlier gestational ages [GAs]) and advances in neonatal care (ie, surfactant, antenatal glucocorticoid therapy, early use of noninvasive respiratory support), which have altered the clinical course of BPD. When evaluating the literature, it is important to recognize that the definition may vary across studies [4,5].
Neurodevelopmental impairment (NDI) — NDI is a broad term that generally includes cognitive, motor, sensory, behavioral, and/or psychologic impairments. Severe NDI is commonly defined in research studies as the presence of any of the following:
●Cognitive delay (ie, ≥2 standard deviations (SD) below the mean on standardized assessments)
●Moderate to severe cerebral palsy (CP)
●Severe bilateral hearing loss
●Severe visual impairment
NDI in preterm infants is discussed in greater detail separately. (See "Long-term neurodevelopmental impairment in infants born preterm: Epidemiology and risk factors".)
SYSTEMIC CORTICOSTEROIDS
Rationale — Because inflammation is thought to play a key role in the pathogenesis of BPD, the use of corticosteroids to suppress inflammation has been of an area of active clinical research [6]. (See "Bronchopulmonary dysplasia: Definition, pathogenesis, and clinical features", section on 'Pathology' and "Bronchopulmonary dysplasia: Definition, pathogenesis, and clinical features", section on 'Inflammation'.)
In addition, relative adrenal insufficiency has been described in extremely preterm (EPT) infants (gestational age [GA] <28 weeks) due to impaired synthesis or release of adrenocortical hormones and this may contribute to hemodynamic and respiratory instability in the early postnatal course [7]. It has been suggested that administration of low-dose corticosteroid therapy as a replacement for this relative deficiency may have a beneficial effect. (See "Neonatal shock: Management", section on 'Suspected adrenal insufficiency'.)
Our approach — Our approach in the use of postnatal systemic corticosteroid therapy to prevent BPD is based on review of the available literature and is consistent with the recommendations of restrictive use of systemic corticosteroid therapy to prevent BPD in the preterm infant by both the American Academy of Pediatrics and the Canadian Paediatric Society (AAP/CPS) [8,9] (see 'General efficacy' below):
●We recommend against the routine use of postnatal systemic corticosteroids to prevent BPD for all EPT infants at risk for BPD. Although prophylactic systemic corticosteroids reduce the risk of BPD, this approach would unnecessarily expose the majority of EPT infants, who would not develop BPD, to the known short-term and long-term adverse effects of corticosteroids. (See 'General efficacy' below and 'Adverse effects' below.)
●We suggest using systemic corticosteroids selectively in EPT infants who remain ventilator-dependent at a postnatal age of two to four weeks and in whom attempts to wean from the ventilator have failed and/or who require oxygen supplementation >50 percent [10-12]. These infants are at high risk for developing BPD.
In this setting, we administer low-dose dexamethasone according to the protocol used in DART (Dexamethasone: A Randomized Trial) (see 'Corticosteroid dose' below). Parents/caregivers should be informed of the risks and benefits and should participate in decision-making.
●We also reserve the use of systemic corticosteroid for "rescue therapy" in select older infants (≥36 weeks postmenstrual age [PMA]) with established BPD who require sustained ventilatory and oxygen support or are experiencing an episode of acute pulmonary deterioration. Such infants should first be evaluated for other causes of lung disease, including ventilator-associated pneumonia, progressive pulmonary hypertension, or the development of bronchomalacia. Use of corticosteroids in this setting is discussed separately. (See "Bronchopulmonary dysplasia: Management", section on 'Corticosteroids'.)
General efficacy — The efficacy of postnatal systemic corticosteroid therapy, either routinely administered during the first week after birth or used selectively in older infants with ongoing mechanical ventilator-dependence, for preventing development of BPD is supported by clinical trials and meta-analyses [13-17]. In general, the available evidence suggests that the benefits of routine postnatal systemic corticosteroid therapy are outweighed by the adverse effects. (See 'Adverse effects' below.).
●Early use (within first 7 to 10 days) – In a meta-analysis of 26 trials (4167 neonates), early corticosteroid therapy administered within the first 7 to 10 days after birth decreased the incidence of BPD at 36 weeks PMA compared with placebo (25 versus 31 percent; relative risk [RR] 0.80, 95% CI 0.73-0.88) [14]. Mortality was similar in both groups. As discussed below, the beneficial effect in reducing rates of BPD was seen only in trials of dexamethasone, whereas trials involving hydrocortisone did not detect a significant reduction in BPD. (See 'Dexamethasone' below and 'Hydrocortisone' below.)
However, early corticosteroid therapy, particularly with dexamethasone, is associated with increased risk of cerebral palsy (CP) in long-term follow-up. Corticosteroid therapy is also associated with an increased risk of gastrointestinal perforation, as discussed below. (See 'Adverse effects' below.)
●Later use in infants with evolving or established BPD – In a meta-analysis of 21 trials involving 1428 infants >7 days old with ongoing ventilator dependence, late administration of corticosteroid therapy reduced mortality compared with control (18 versus 23 percent; RR 0.81, 95% CI 0.66-0.99), and reduced the incidence of BPD at 36 weeks PMA (53 versus 59 percent, RR 0.89, 95% CI 0.80-0.99; 14 trials, 988 neonates) [16]. In subgroup analyses, a significant reduction in BPD was detected in trials using dexamethasone but not in trials using hydrocortisone. (See 'Dexamethasone' below and 'Hydrocortisone' below.)
Adverse effects
Short-term — In the previously mentioned meta-analyses, postnatal systemic corticosteroids were associated with the following short-term adverse effects [14,16]:
●Hyperglycemia
●Hypertension
●Gastrointestinal bleeding
●Gastrointestinal perforation
●Hypertrophic cardiomyopathy
Two trials were stopped early because of increased rates of intestinal perforation in the active treatment groups compared with placebo; one of these trials involved hydrocortisone [18], the other dexamethasone [19]. In a meta-analysis of 16 trials, early corticosteroid treatment was associated with a higher incidence of intestinal perforation compared with placebo (7 versus 4 percent; RR 1.84, 95% CI 1.36-2.49) [14]. The risk was similar regardless of agent (dexamethasone or hydrocortisone). A separate meta-analysis examining late corticosteroid treatment was inconclusive as to the risk of intestinal perforation as there were few events (16 total events out of 552 neonates) [16]. The risk of intestinal perforation is heightened when corticosteroids are administered concomitantly with indomethacin.
Long-term — Follow-up studies have raised concerns that postnatal systemic corticosteroid therapy contributes to neurodevelopmental impairment (NDI), especially CP [20-23]. In a meta-analysis of trials involving early administration of corticosteroids (within first seven days after birth), long-term data showed early corticosteroid treatment was associated with increased risk CP compared with control (11 versus 7 percent; RR 1.42, 95% CI 1.06-1.91) [14]. However, as discussed below, the increased risk of CP was seen only in trials involving dexamethasone, not in trials involving hydrocortisone. (See 'Dexamethasone' below and 'Hydrocortisone' below.)
Specific corticosteroid agents
Dexamethasone — Dexamethasone is the corticosteroid agent most commonly used in the prevention and treatment of BPD. Randomized clinical trials have shown that it reduces the risk of BPD [14-16,24]. However, postnatal dexamethasone therapy is associated with serious adverse effects, including CP, especially when high doses are used within the first week after birth [14,15,24]. As a result, we continue to recommend against routine early dexamethasone therapy [25]. However, the benefit-risk ratio depends on the underlying risk of BPD [21,26]. In high-risk patients (ie, older infants with evolving or established BPD who remain ventilator-dependent after two to four weeks), the benefits of low-dose dexamethasone likely outweigh the risk of adverse effects [16,27]. (See 'Our approach' above.)
●Early use (within first 7 to 10 days) – Randomized trials and meta-analyses evaluating dexamethasone given early (within the first 7 to 10 days after birth) have reported the following findings [14,15,19,24]:
•No apparent effect on mortality – In a meta-analysis of 20 trials (n=2940 neonates), mortality rates were similar in infants randomized to early dexamethasone treatment compared with control (24 percent each; RR 1.02, 95% CI 0.90-1.16) [14].
•Lower incidence of BPD at 36 weeks PMA – A meta-analysis of 17 trials (n=2791 neonates) early dexamethasone therapy reduced rates of BPD at 36 weeks PMA compared with control (19 versus 27 percent; RR 0.72, 95% CI 0.63-0.82) [14].
•Increased incidence of NDI – In a meta-analysis of seven trials (n=921 neonates), rates of CP were higher among infants who received early dexamethasone treatment compared with control (16 versus 9 percent; RR 1.77, 95% CI 1.21-2.58) [14]. Rates of major neurosensory disability (ie, hearing or vision impairment) were also higher in the dexamethasone group (22 versus 16 percent; RR 1.37, 95% CI 1.03-1.83). Other neurodevelopmental outcomes (eg, standardized cognitive and motor assessments) were not consistently reported in the different trials.
●Later use (after first 7 to 10 days) – Randomized trials and meta-analyses evaluating use of dexamethasone in neonates who remain ventilator-dependent beyond the first 7 to 10 days after birth have reported the following findings [13,15,16,22,28]:
•Uncertain effect on mortality – A meta-analysis of 19 randomized trials (n=993 neonates) did not detect a difference in mortality between infants treated with late dexamethasone compared with control (16 versus 19 percent; RR 0.85, 95% CI 0.66-1.11) [16].
•Reduced incidence of BPD at 36 weeks PMA – In a meta-analysis of 12 randomized trials (n=553 neonates), late dexamethasone therapy reduced the incidence of BPD at 36 weeks PMA compared with control (50 versus 66 percent; RR 0.76, 95% CI 0.66-0.87) [16].
•Earlier extubation – In a meta-analysis of 16 randomized trials (n=783 neonates), infants treated with late dexamethasone were less likely to remain intubated after seven days of treatment (55 versus 83 percent; RR 0.66, 95% CI 0.6-0.73) [16]. Similar findings were reported in an observational study in which approximately 80 percent of treated infants were successfully extubated within 14 days of starting dexamethasone [29].
•Uncertain effect on NDI – A meta-analysis of 17 randomized trials (n=1290 neonates), did not detect a significant difference in rates of CP in later childhood among neonates treated with late dexamethasone compared with control (14 versus 12 percent; RR 1.12, 95% CI 0.79-1.60) [16].
Hydrocortisone — When a decision is made to treat with corticosteroids, we generally prefer low-dose dexamethasone over hydrocortisone. Clinical trials directly comparing the two agents are lacking. Although the available trials evaluating hydrocortisone in this setting suggest that the risk of NDI may be lower with this agent, its efficacy for reducing BPD among neonates with ongoing ventilator dependence is less certain. Some of the uncertainty in this area stems from the considerable variability between trials regarding the timing and dosing of hydrocortisone.
●Early use (within first 7 to 10 days) – Randomized trials and meta-analyses evaluating hydrocortisone given early (within the first 7 to 10 days after birth) at a low dose (1 to 2 mg/kg per day) have reported the following findings [14,30-35]:
•Reduced mortality – In a meta-analysis of 11 randomized trials (n=1433 neonates) early treatment with hydrocortisone reduced in-hospital mortality compared with control (18 versus 22 percent; RR 0.80, 95% CI 0.65-0.99) [14].
•No apparent effect on rates of BPD at 36 weeks PMA – A meta-analysis of nine randomized trials (n=1376 neonates) did not detect a significant difference in rates of BPD at 36 weeks PMA among neonates treated with early hydrocortisone compared with control (35 versus 38 percent; RR 0.92, 95% CI 0.81-1.06) [14].
•No apparent effect on rates of NDI – In a meta-analysis of six randomized trials (n=1052 neonates), rates of cerebral palsy (CP) in later childhood were similar in both groups (6 percent each; RR 1.05, 95% CI 0.66-1.66), as were rates of neurosensory disability (ie, hearing or vision impairment) (15 versus 17 percent; RR 0.86, 95% CI 0.64-1.14) [14]. Other neurodevelopmental outcomes (eg, standardized cognitive and motor assessments) were not consistently reported in the different trials. In the largest trial, which included 523 EPT infants, neurodevelopmental outcomes at a median corrected age of 22 months were similar in both groups (rates of moderate to severe NDI were 7 versus 11 percent; rates of mild NDI were 20 versus 18 percent) [35]. In a subgroup analysis limited to infants born at 24 to 25 weeks GA, the rate of NDI was lower in infants randomized to hydrocortisone compared with placebo, but the exploratory nature of this subgroup analysis and the small number of events preclude drawing firm conclusions [36]. Magnetic resonance imaging (MRI) at term equivalent age performed in 300 of the 412 survivors demonstrated a higher score for white matter abnormality for the hydrocortisone versus control groups [37]. However, after adjusting for GA and other risk factors, MRI scores were similar between the two groups. Observational studies have also generally reported similar neurodevelopmental outcomes in preterm infants treated with or without hydrocortisone [38-41]; although some studies have reported worse neurodevelopmental outcomes among hydrocortisone-treated infants [33].
●Later use (after first 7 to 10 days) – Randomized trials and meta-analyses evaluating use of hydrocortisone at higher doses (3 to 5 mg/kg per day) in neonates who remain ventilator-dependent beyond the first 7 to 10 days after birth have reported the following findings [16,42-45]:
•Uncertain effect on mortality – In a meta-analysis of two randomized trials (n=435 neonates) mortality was lower among neonates treated with late hydrocortisone compared with control; however the finding was not statistically significant (23 versus 30 percent; RR 0.74, 95% CI 0.54-1.02) [16]. In a subsequent randomized trial published after the meta-analysis, in-hospital mortality was similar in neonates who received late hydrocortisone treatment (at 14 to 28 days) compared with placebo (9 versus 10 percent; RR 0.86, 95% CI 0.56-1.31) [43].
•No apparent effect on rates of BPD at 36 weeks PMA – In a meta-analysis of two randomized trials (n=435 neonates), rates of BPD at 36 weeks PMA were similar in both groups (57 versus 52 percent; RR 1.10, 95% CI 0.92-1.31) [16]. Similar findings were reported in a subsequent randomized trial published after the meta-analysis, in which rates of BPD at 36 weeks PMA were comparable in both groups (83 versus 86 percent; RR 0.96, 95% CI 0.91-1.02) [43].
•Earlier extubation – In the two trials that reported extubation rates, infants treated with late hydrocortisone were more likely to be extubated within one to two weeks after starting treatment [42,43]. For example, in a multicenter trial involving 800 infants <30 weeks GA who were ventilator-dependent at 14 to 28 days postnatal age, more infants in the hydrocortisone group were successfully extubated during the 10-day treatment period compared with placebo (45 versus 34 percent; RR 1.54, 95% CI 1.23-1.93) [43].
●No apparent effect on rates of NDI – A meta-analysis of two randomized trials (n=435 neonates), did not detect a significant difference in rates of CP in later childhood among neonates treated with late hydrocortisone compared with control (28 versus 34 percent; RR 0.82, 95% CI 0.62-1.08) [16]. Similarly, in a subsequent randomized trial published after the meta-analysis, rates of moderate to severe NDI among survivors were comparable in both groups (58 versus 57 percent; RR 1.03, 95% CI 0.90-1.17) [43].
Corticosteroid dose — The risk of adverse effects of corticosteroid therapy appears to increase with increasing cumulative dose [25,46,47]. An optimal dose that provides maximal benefit while minimizing adverse effects has not been established. Decisions regarding criteria for starting postnatal corticosteroid therapy, preferred agent, and dosing remain center dependent. In our centers, we selectively use low-dose dexamethasone only in patients who are at high risk of developing BPD (ie, ventilator- or oxygen-dependent after two to four weeks postnatal age). (See 'Our approach' above.)
We administer low-dose dexamethasone according to the protocol used in DART (Dexamethasone: A Randomized Trial), which provides intravenous dexamethasone (cumulative dose 0.89 mg/kg) as follows [13]:
●0.075 mg/kg per dose 12 hourly for three days, then,
●0.05 mg/kg per dose 12 hourly for three days, then,
●0.025 mg/kg per dose 12 hourly for two days, then,
●0.01 mg/kg per dose 12 hourly for two days, then discontinue
Several cohort studies have reported an increasing risk of neurologic impairment with higher doses of corticosteroid therapy or longer duration of therapy:
●A multicenter prospective cohort study reported that for infants exposed to postnatal corticosteroid therapy (dexamethasone in 92 percent of cases) an increase of 1 mg/kg in dexamethasone dose was associated with a two-point reduction in the Bayley mental development index (MDI) score, and a 40 percent increase in risk for disabling CP at 18 to 22 months corrected age [46].
●In the EPICure study (a large English prospective cohort study of preterm infants born <26 weeks gestation), increasing duration of postnatal steroid treatment was associated with poor motor outcomes. [47].
In a network meta-analysis that included 14 trials that used different corticosteroid regimens (ie, different agents [dexamethasone or hydrocortisone], timing of administration [early versus late], and/or cumulative daily dose [low, moderate, or high]), dexamethasone given between 8 to 14 days after birth at moderate dose was the regimen associated with lowest risk of BPD or morality at a PMA of 36 weeks compared with the other 13 regimens [48]. However, the indirect nature of these comparisons limit the certainty of this finding.
Trends in practice over time — The use of postnatal systemic corticosteroid therapy in preterm neonates has declined considerably since the late 1990s and early 2000s when concerns first surfaced about long-term adverse effects. There was further decline after 2010 when the AAP published guidelines recommending against postnatal corticosteroid therapy due to these concerns.
Several population-based studies examining the impact of diminished corticosteroid use on the incidence and severity of BPD have reported conflicting results. A study of three large neonatal network databases in North America reported that a reduction in steroid use starting in 1999 was not associated with significant changes in mortality or the incidence and severity of BPD [49]. In contrast, several studies have reported an increase in the incidence of BPD with varying effects on mortality. One retrospective study of extremely low birth weight (ELBW) infants (BW <1000 g) reported that decreased use of postnatal corticosteroid therapy was associated with an increase in the incidence of BPD without a change in mortality [50]. There was also no change in the rate of major NDI. In another population-based study found a reduction in postnatal corticosteroid use was associated with a decrease in mortality and an increased incidence of BPD for preterm infants (GA 24 to 32 weeks) [51]. These findings support the notion that prophylactic corticosteroids are effective in reducing the incidence of BPD. (See 'General efficacy' above.)
Unanswered questions — As discussed in the previous sections, we recommend against routine use of postnatal corticosteroid therapy since the available evidence suggests that the potential benefits do not outweigh the known complications of this intervention [8,9]. However, it remains uncertain whether there is a clinical setting in which postnatal corticosteroid therapy would be beneficial despite the risk of adverse effects.
Well-powered randomized controlled trials with adequate follow-up are still needed to answer the following questions [52]:
●Can we identify a subpopulation of preterm infants for whom the benefits of postnatal corticosteroids clearly outweigh the risks?
●If corticosteroid therapy is used, what is the optimal regimen (ie, agent, timing, dosing)? (See 'Specific corticosteroid agents' above and 'Corticosteroid dose' above.)
●Do short-term improvements in pulmonary function associated with corticosteroid administration convey additional long-term pulmonary benefits (eg, prevention of long-term severe pulmonary complications of BPD, including severe pulmonary hypertension and cor pulmonale)? (See "Complications and long-term pulmonary outcomes of bronchopulmonary dysplasia".)
OTHER ROUTES OF ADMINISTRATION
Inhaled corticosteroids — It has been proposed that inhaled corticosteroids might be an effective and safer alternative for preventing BPD in at-risk EPT infants rather than using systemic corticosteroids, since the latter is associated with long-term neurodevelopmental impairment (NDI), as discussed above (see 'Adverse effects' above). However, postnatal administration of inhaled corticosteroids has not consistently been shown to reduce the risk of BPD, and there are concerns about a possible increased mortality risk.
Our approach — Our approach is as follows:
●Routine use for prevention of BPD – Based upon available data (described below), we suggest not routinely using inhaled corticosteroids to prevent BPD.
●For treatment of severe BPD – Despite the limited data on the efficacy of inhaled corticosteroids, we do selectively use inhaled corticosteroids in older infants if they have severe BPD, are dependent upon substantial ventilator and oxygen support, and have signs of severe airway obstruction or reactive airway disease. This is discussed separately. (See "Bronchopulmonary dysplasia: Management", section on 'Corticosteroids'.)
Evidence — Based on the available clinical trial data, postnatal administration of inhaled corticosteroids does not appear to meaningfully reduce the risk of BPD, and some studies suggest a possible increased mortality risk [53-55].
In a meta-analysis of six trials involving mechanically ventilated preterm neonates randomized to early (in the first two weeks after birth) inhaled corticosteroid therapy or placebo, rates of BPD at 28 days of age or 36 weeks postmenstrual age (PMA) were similar in both groups (15 percent in each; relative risk [RR] 0.97, 95% CI 0.62-1.52) [53]. Mortality was also similar at 36 weeks PMA (14 and 15 percent, respectively). However, inhaled corticosteroids reduced the incidence of BPD among survivors (24 versus 31 percent; RR 0.76, 95% CI 0.63-0.93). Adverse events were similar between the two groups.
The largest trial in the above meta-analysis was the NEUROSIS trial (Neonatal EUROpean Study of Inhaled Steroids). A follow-up study of the NEUROSIS trial reported that of the 813 infants for whom vital status was known at two years, there were more deaths in the inhaled corticosteroid group compared with placebo (20 versus 15 percent; RR 1.37 95% CI 1.01-1.86) [54]. Rates of NDI among survivors were similar in both groups (48 versus 51 percent; adjusted RR 0.93, 95% CI 0.80-1.09).
There are fewer data on use of late (>7 days after birth) inhaled corticosteroid therapy in preterm neonates with ongoing need for mechanical ventilation or oxygen beyond the first week after birth. In a meta-analysis five placebo-controlled trials (n=79), infants who received late inhaled corticosteroid therapy were less likely to remain intubated after seven days of treatment (74 versus 95 percent; RR 0.80, 95% CI 0.66-0.98) [55]. However, overall duration of mechanical ventilation was similar in both groups. No firm conclusions could be reached regarding the effect of inhaled corticosteroids on mortality, rates of BPD at 36 weeks PMA, or serious adverse events because these outcomes were inconsistently reported and/or there were few events. Another important limitation of these data is the considerable heterogeneity between trials. In addition, most of the trials were performed in the late 1990s to early 2000s and the applicability to modern-day practice is unclear.
Intratracheal corticosteroids — Because of the concerns for adverse effects associated with systemic steroids (eg, increased risk of cerebral palsy), postnatal intratracheal delivery of corticosteroids mixed with surfactant was proposed as an alternative. However, we suggest not using this approach as it remains uncertain whether intratracheal corticosteroid therapy is effective and safe in extremely preterm (EPT) neonates at risk for BPD.
Two randomized controlled trials from a single center examined early administration of intratracheal budesonide at 0.25 mg/kg mixed in surfactant compared with surfactant alone in infants <1500 g [56-60]. In a meta-analysis that pooled the results of both trials plus three other smaller trials (n=549 infants), rates of BPD were lower in the budesonide group compared with control (22 versus 42 percent; RR 0.58, 95% CI 0.41-0.82) [59]. Short-term mortality was also lower in the budesonide group (13 versus 20 percent; RR 0.64, 95% CI 0.41-0.99). A separate meta-analysis assessing long-term outcomes at age 18 to 36 months did not detect a difference in rates of neurodevelopmental impairment (31 versus 39 percent; RR 0.78, 95% CI 0.55-1.11) or long-term mortality (18 versus 24 percent; RR 0.76, 95% CI 0.52-1.11). [61].
The results of these trials need to be replicated in larger multicenter trials before use of intratracheal budesonide can be recommended as a routine.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Bronchopulmonary dysplasia".)
SUMMARY AND RECOMMENDATIONS
●Rationale for postnatal corticosteroid therapy ‒ Because inflammation is thought to play a key role in the pathogenesis of bronchopulmonary dysplasia (BPD), the use of corticosteroids to suppress inflammation has been proposed to prevent BPD in high-risk preterm infants, particularly extremely preterm (EPT) infants (gestational age [GA] <28 weeks . (See 'Rationale' above.)
●Systemic corticosteroid therapy ‒ The potential benefits of postnatal corticosteroid therapy must be weighed against the adverse effects, which include hyperglycemia, hypertension, gastrointestinal perforation, and risk of neurodevelopmental impairment (eg, cerebral palsy [CP]). The optimal approach to identifying patients who are likely to benefit from postnatal corticosteroid therapy, selecting a corticosteroid agent, timing of treatment, and dosing remain uncertain. Practice varies between centers. Our general approach is as follows (see 'Systemic corticosteroids' above):
•For most EPT infants, we recommend against routine use of systemic corticosteroid therapy (Grade 1B). Although postnatal systemic corticosteroid therapy reduces the risk of BPD, it is associated with serious adverse effects, particularly increased risk of CP. Thus, the potential benefits of routine postnatal corticosteroid therapy in this population appear to be outweighed by its downsides. (See 'Our approach' above and 'General efficacy' above and 'Adverse effects' above.)
•For EPT infants who remain ventilator- and oxygen-dependent at two to four weeks postnatally, we suggest low-dose systemic corticosteroid therapy (Grade 2B). In this setting, we suggest low-dose dexamethasone rather than hydrocortisone (Grade 2C). In our center, we use a low-dose 10-day regimen. However, the optimal regimen remains uncertain. (See 'Dexamethasone' above and 'Corticosteroid dose' above.)
●Inhaled and intratracheal corticosteroid therapy ‒ Because of the concerns for adverse effects associated with systemic steroids, inhaled corticosteroids and intratracheal administration (delivered mixed with surfactant) have been proposed as alternatives. (See 'Other routes of administration' above.)
•We suggest not routinely using these therapies for prevention of BPD (Grade 2C). Postnatal inhaled corticosteroid therapy has not consistently been shown to reduce the risk of BPD, and there are concerns about a possible increased mortality risk. Data on intratracheal corticosteroid administration are limited and the safety and efficacy of this approach remain uncertain. (See 'Inhaled corticosteroids' above and 'Intratracheal corticosteroids' above.)
•Despite the limited data, inhaled corticosteroid therapy is a reasonable treatment option for older infants with established severe BPD, particularly if they are dependent upon substantial ventilator and oxygen support and have signs of severe airway obstruction or reactive airway disease. This is discussed separately. (See "Bronchopulmonary dysplasia: Management", section on 'Corticosteroids'.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges James Adams, Jr., MD, who contributed to an earlier version of this topic review.