Postnatal steroids for chronic lung disease – a lesson learned

“Don't just stand there, do something”. And we did. Neonatology is full of heroic efforts using potent therapies in novel ways to save smaller and sicker babies. Postnatal corticosteroids for the prevention and treatment of chronic lung disease (CLD) are among these therapies. CLD (or bronchopulmonary dysplasia) is an inflammatory condition of the premature lung with an etiology that is multifactorial. Defined as a supplemental oxygen requirement in a preterm infant at either 28 days of age or 36 weeks' postmenstrual age (PMA), it is often associated with lower birth weight and gestation. The Canadian Neonatal Network (CNN) of 17 neonatal intensive care units reported that 26% of very low birthweight infants (less than 1500 g) have CLD at 36 weeks' PMA (1). Its prevalence in infants who have received mechanical ventilation supports the hypothesis that CLD is promoted by the inflammatory response to oxygen-related radicals or ventilator-related volutrauma – hence, the proposed role of anti-inflammatory corticosteroids. In the present issue of Paediatrics & Child Health, a joint statement by the Fetus and Newborn Committee of the Canadian Paediatric Society (CPS) and the American Academy of Pediatrics (AAP) reviews the evidence for postnatal systemic corticosteroids (pages 20–28). Although ineffective in altering the course of respiratory distress syndrome, systemic corticosteroids reduce the oxygen and ventilatory needs of small, sick, premature infants in intensive care units. Dozens of randomized controlled trials attest to the effectiveness of systemic corticosteroids, principally dexamethasone, in reducing ventilator days and days using supplemental oxygen. By using both oxygen requirements and survival to 36 weeks' PMA, several meta-analyses have reinforced the short term benefits of preventative therapy in the first few days of life and ‘rescue’ treatment at a later stage. The effect was so convincing to neonatologists that in 1996/97, the CNN reported that 25% of very low birth weight infants received dexamethasone during their hospital stay. It appeared that CLD could be treated, admittedly with an assortment of short term complications (eg, hypertension, hyperglycemia, transient growth impairment and gastrointestinal side effects). So, why should the CPS Fetus and Newborn Committee and the AAP recommend that routine use of systemic corticosteroids, particularly dexamethasone, be curtailed? The answer was derived from systematic reviews that included longer term outcomes. The CPS and the AAP concluded that the survival to discharge of infants with CLD was not significantly altered by steroid therapy. On the other hand, follow-up data demonstrated an important and significant increase in neurodevelopmental disability in infants who received dexamethasone postnatally. It appears that our miracle cure has turned into a curse. Cerebral palsy is the most common major neurodevelopmental disability in very low birth weight infants. The data suggest that the odds ratio for cerebral palsy in dexametha-sone-treated infants compared with controls may be as high as 2.89 (95% CI 1.96 to 4.27), indicating a two- to fourfold increase in the risk of this adverse outcome! In addition, the CPS and the AAP strongly recommend that further research into alternative steroids and modes of delivery be performed before their routine use. The study by Jangaard et al (pages 13–19) in this issue exemplifies this principle, demonstrating the ineffectiveness of an inhaled steroid (beclomethasone) in preventing CLD, as well as a lack of long term benefits on mortality and morbidity. The authors' conclusion that more research is required before recommending this intervention is wholly appropriate. So, what went wrong with postnatal corticosteroids? How did neonatal medicine develop an established North American practice, only to find, through hindsight, that it was of questionable benefit? It is clear that we considered the ‘short term gain’ rather than the ‘long term pain’, perhaps with a view that more is always better. It seems that modern medicine has an unfaltering readiness to introduce new therapies before they are adequately appraised. A lesson in evidence-based practice should be learned from the experience with dexamethasone and CLD – we should be aware of our failures and successes. Perhaps the most thoroughly evaluated intervention, surfactant replacement therapy, should be a model for future therapeutic evaluations. A scan of any neonatal unit will reveal the difficulties in meeting this rigorous standard with a number of common interventions, such as blood transfusion for anemia of prematurity, caffeine or theophylline for apnea of prematurity and antireflux medications. The reality is that most therapies in newborn medicine may be considered ‘investigational’ and, therefore, are worthy of further, well-designed studies, with appropriate long term outcomes. It is not too late to learn this lesson and, perhaps, until the appropriate studies are completed, the maxim should be: “Don't just do something, stand there”.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.