MANUSCRIPT CITATION:

Bassler D, Plavka R, Shinwell ES, et al. NEUROSIS Trial Group. Early Inhaled Budesonide for the Prevention of Bronchopulmonary Dysplasia. N Engl J Med 2015;373:1497-506 DOI: 10.1056/NEJMoa1501917 PMID:26465983

REVIEWED BY:

Ravi Mangal Patel MD, MSc
Assistant Professor of Pediatrics, Division of Neonatology
Emory University School of Medicine, Atlanta, GA

TYPE OF INVESTIGATION:

Prevention

QUESTION:

(P) In infants with a gestational age of 23+0 to 27+6 weeks, (I) does inhaled budesonide initiated within 24 hours after birth and continued until 32+0 weeks postmenstrual age or discontinuation of positive pressure support with supplemental oxygen (C) compared to placebo (O) reduce the incidence of either death or bronchopulmonary dysplasia (BPD)?

METHODS:

• Design: Randomized clinical trial
• Allocation: Computer-generated randomization scheme with fixed block size of 8 and allocation ratio of 1:1. Randomization was stratified by gestational age (23-25 vs. 26-27 weeks of completed gestation). Whether siblings were assigned to the same treatment group was not specified.
• Blinding: Treatment group using coded inhalers that were manufactured to be similar between placebo and budesonide arms.
• Follow-up period: Patients were followed to the first discharge home. Longer-term follow-up, including assessment of neurodevelopmental outcomes, at 18 to 22 months of corrected age is planned.
• Setting: 40 centers in nine European countries
• Patients: Infants with a gestational age of 23+0 weeks to 27+6 weeks who required positive pressure respiratory support from invasive or non-invasive ventilation or CPAP and were within 12 hours of birth were eligible. Infants who were to receive only palliative care, had dysmorphic features or congenital anomalies likely to affect life expectancy or development, had known or strongly suspected cyanotic heart disease, or were from a multiple birth pregnancy other than the second infant in birth order were excluded.
• Intervention: Randomization to either experimental group (budesonide 200 microgram per puff) or control (placebo contained only hydrofluoralkane propellant). Two puffs of study drug were administered every 12 hours in the first 14 days of life (daily dose of 800 micrograms) and then one puff was given every 12 hours from day 15 until the last dose of drug administered (daily dose of 400 micrograms). Study drug was administered through the ventilator circuit for those infants with an endotracheal tube or was given by face mask with spacer (AeroChamber mini) for infants without an endotracheal tube. Study drug was continued until 32+0 weeks postmenstrual age or when positive pressure respiratory support with supplemental oxygen was discontinued, whichever was earlier. Open-label inhaled glucocorticoids and systemic glucocorticoid treatment were discouraged. Doses of study drug could be withheld or decreased at the discretion of the treating attending physician.
• Outcomes:
  • Primary outcome: The primary outcome was a composite of death or BPD at 36 week of postmenstrual age.
    • BPD was defined as the need for positive pressure support, supplemental oxygen at FiO2 > 0.30, or inability to maintain an oxygen saturation >90% during a trial of weaning from supplemental oxygen to room air (oxygen-reduction test).
  • Secondary outcomes: Secondary outcomes were prespecified and included individual components of the composite primary outcome as well as a number of other outcomes including: duration of positive-pressure respiratory support or supplemental oxygen, ventriculomegaly with or without intraventricular hemorrhage, patent ductus arteriosus requiring drug treatment or surgery, intestinal perforation or necrotizing enterocolitis, retinopathy of prematurity stage 2 or greater, culture-proven infections, growth determined by increases in weight and head circumference from birth to 28d, need for reintubation, oral candidiasis requiring treatment, hyperglycemia requiring insulin treatment or hypertension requiring treatment.
  • The reported primary and secondary outcomes were consistent with the planned outcome assessment based on the trial protocol dated 8/21/09, with the exception of the planned secondary outcome assessment of neurodevelopmental disability at 18-22 months of corrected age (trial protocol available as supplementary material at NEJM.org)
• Analysis and Sample Size: Based on the assumption of an incidence of 50% in the primary outcome (death or BPD) in the control arm, the authors calculated that 808 infants would need to be enrolled to have 80% power, with a two-sided alpha level of 0.05, to detect a 20% lower risk in the budesonide group (equivalent to a risk difference of 10%). The sample size was inflated to 850 to account for anticipated loss to follow-up. The authors accounted for the stratifying variable of gestational age in their analysis of the primary outcome using a Mantel-Haenszel chi-square test. The primary analysis was intention-to-treat. A secondary analysis was conducted with adjustment for various covariates, although the rationale behind this was not specified. Sample size estimation and analysis plan are consistent with the study protocol dated 8/21/09.
• Data and safety monitoring and stopping rules: The authors planned four safety analyses to be completed once 100, 300, 500 and 700 infants reached status (36wk postmenstrual age). The authors also planned one formal interim analysis for efficacy using the Peto-Haybittle approach. This approach uses a very conservative threshold (P=0.001) as a stopping rule. This approach allows the final analysis to be performed at the initially determined level of significance (i.e. alpha of 0.05) without having to account for alpha-spending when analyzing the final outcomes. Upon the last safety review, the data and safety monitoring committee recommend study drug be withheld for a borderline significant between-group difference in the rate of death, although study drugs for all enrolled patients had already been discontinued per the trial protocol.
• Patient follow-up: Of 2233 screened infants, 863 were randomized (441 assigned to budesonide group and 422 assigned to placebo). Of the 863 randomized infants, 856 received the study drug and were analyzed for the primary outcome in an intention to treat analysis.

MAIN RESULTS:

Budesonide treatment, compared to placebo, did not significantly decrease the risk of death or BPD (primary outcome). Budesonide treatment resulted in a significantly decreased incidence of survival with BPD. No significant difference was seen in the effect of budesonide treatment on death, with the upper limit of the confidence interval including a relative risk up to 1.69. In covariate adjusted analysis, budesonide treatment, compared to placebo, was associated with a lower risk of death or BPD (odds ratio 0.71; 95% CI 0.53-0.97) as well as survival with BPD (OR 0.61; 95% CI 0.44-0.85) but not death (OR 1.39; 95% CI 0.89-2.18). Among secondary outcomes, surgical treatment for a PDA was lower among infants in the budesonide group compared to placebo group (RR 0.55; 95% CI 0.36-0.83). In addition, the risk of reintubation was lower among infants in the budesonide group, compared to placebo (RR 0.58; 95% CI 0.35-0.96). Duration of supplementation oxygen was shorter among budesonide treated infants, compared to placebo (31.6 vs 33.1 median days), consistent with the effect on the incidence of BPD. There were no significant differences in other measured secondary outcomes, including retinopathy of prematurity, brain injury, necrotizing enterocolitis, culture-proven infection, adverse treatment effects, duration of hospitalization or growth parameters.

CONCLUSION:

The authors conclude that among extremely preterm infants, the incidence of BPD was lower among infants who received early inhaled budesonide compared to those receiving placebo, but the benefit on BPD may have been gained at the expense of increased mortality.

Visit Acta to access a pdf copy of this EBNEO commentary!

COMMENTARY:

The trial found no difference in the risk of death or survival with BPD among infants treated with early prophylactic inhaled budesonide, compared to placebo treated infants. The trial did find a fairly large decrease in the incidence of survival with BPD among budesonide treated infants (risk difference of approximately 10%, corresponding to a number needed to treat (NNT) of 10 (95% CI 6 to 29) to prevent one outcome of BPD among surviving infants). However, the other component of the composite outcome, death, had a relative risk point estimate in the opposite direction leading to an overall negative result of the primary study outcome.

How are we to interpret these findings? As has been suggested by Freemantle et al., results from trials reporting composite outcomes should focus on the result of the composite outcome itself rather than the individual components.¹ The effects of experimental treatment on individual components of the composite outcome, in this case death and survival with BPD, should be considered as secondary outcomes. One major issue in interpreting composite outcomes arises when the individual components of a composite outcome diverge, which has been noted in other important trials in neonatology, including the aggressive vs. conservative phototherapy² and SUPPORT trials.³ As authors have written with regards to composite outcomes in adult cardiovascular studies,⁴ “Higher event rates and larger treatment effects associated with less important components may result in misleading impressions of the impact of treatment.” Given the unequal weight many families would give to the components of the primary outcome, with many considering death to be a more important outcome than survival with BPD, caution is appropriate in interpreting the results of this study. However, the possibility that the difference in mortality is due to chance alone should also be acknowledged. Importantly, short-term assessment of lung function in the hospital, measured by the outcome of BPD, can vary significantly from longer-term, patient-centered outcomes, such as assessment of lung health (e.g. wheezing, emergency room or physician visits for breathing problems)⁵ as well as more sophisticated measures of lung function into adolescence.⁶ Reassuringly, the authors have planned for longer-term follow-up, allowing for the potential opportunity for additional assessments of pulmonary health to be undertaken.

Questions also remain regarding the effects of early inhaled budesonide on long-term neurodevelopmental outcomes. Of note, the secondary outcome of brain injury, while not statistically different, was higher among budesonide infants (RR 1.25; 95% CI 0.94-1.65), highlighting the importance of the longer-term follow-up at 18-22 months planned by the study investigators.

Although there was a decrease in the incidence of BPD among survivors with early inhaled budesonide treatment, a concurrent increase in death cannot be ruled out. Based on these findings, early preventive inhaled budesonide cannot be currently recommended in clinical practice and additional studies are needed.

REFERENCES

1. Freemantle N, Calvert M, Wood J, Eastaugh J, Griffin C. Composite outcomes in randomized trials: greater precision but with greater uncertainty? JAMA 2003;289:2554-9.
2. Morris BH, Oh W, Tyson JE, et al. Aggressive vs. conservative phototherapy for infants with extremely low birth weight. N Engl J Med 2008;359:1885-96.
3. Network SSGotEKSNNR, Carlo WA, Finer NN, et al. Target ranges of oxygen saturation in extremely preterm infants. N Engl J Med 2010;362:1959-69.
4. Ferreira-Gonzalez I, Busse JW, Heels-Ansdell D, et al. Problems with use of composite end points in cardiovascular trials: systematic review of randomised controlled trials. BMJ 2007;334:786.
5. Stevens TP, Finer NN, Carlo WA, et al. Respiratory outcomes of the surfactant positive pressure and oximetry randomized trial (SUPPORT). J Pediatr 2014;165:240-9 e4.
6. Zivanovic S, Peacock J, Alcazar-Paris M, et al. Late outcomes of a randomized trial of high-frequency oscillation in neonates. N Engl J Med 2014;370:1121-30.