Postnatal corticosteroids and developmental outcomes in extremely preterm or extremely low birth weight infants: The Victorian Infant Collaborative Study 2016–17 cohort

Abstract Aim Systemic postnatal corticosteroids are used to treat or prevent bronchopulmonary dysplasia (BPD) in extremely preterm (EP) or extremely low birth weight (ELBW) infants but are associated with long‐term harm. We aimed to assess the relationship between cumulative postnatal corticosteroid dose and neurodevelopmental outcomes. Methods Longitudinal cohort study of all EP/ELBW livebirths in Victoria, Australia 2016–2017. Perinatal data were collected prospectively. Neurodevelopmental assessment was performed at 2 years' corrected age. Linear and logistic regression were used to determine relationships between cumulative corticosteroid dose and neurodevelopment, adjusted for gestational age, birth weight, sex and major intraventricular haemorrhage. Results Seventy‐six EP/ELBW infants received postnatal corticosteroids to treat or prevent BPD, 62/65 survivors were seen at 2 years. Median (IQR) cumulative postnatal corticosteroid dose was 1.36 (0.92–3.45) mg/kg dexamethasone equivalent. Higher cumulative corticosteroid dose was associated with increased odds of cerebral palsy, adjusted OR (95% CI) 1.47 (1.04, 2.07). Higher cumulative corticosteroid dose was also associated with lower cognitive and motor developmental scores, however, this weakened after adjustment for confounding variables: cognitive composite score adjusted coefficient (95% CI) −1.3 (−2.7, 0.1) and motor composite score adjusted coefficient (95% CI) −1.3 (−2.8, 0.2). Conclusion Higher cumulative postnatal corticosteroid dose in EP/ELBW infants is associated with increased odds of cerebral palsy at 2 years' corrected age. Adequately powered studies are needed to assess the independent effects of cumulative steroid dose on neurodevelopmental outcomes.


| INTRODUC TI ON
Extremely preterm (EP, born <28 weeks' gestation) and extremely low birth weight (ELBW, birth weight < 1000 g) infants are at risk of bronchopulmonary dysplasia (BPD). 1,2 BPD, usually defined as the need for supplemental oxygen or respiratory support at 36 weeks' postmenstrual age, 3 is a chronic lung disease that affects more than 50% of surviving EP infants. 1 BPD is independently associated with higher mortality, increased incidence of cerebral palsy (CP) and higher rates of cognitive impairment. [4][5][6][7] Postnatal systemic corticosteroids are used to treat or prevent BPD in EP infants. 8 Corticosteroids facilitate weaning from mechanical ventilation. 9 However, some studies suggest that the neonatal administration of systemic corticosteroids may increase the risk of CP and neurodevelopmental disability in childhood, particularly when given in the first week after birth. 9,10 There are large variations between studies in the type, timing and dose of systemic corticosteroids used, [11][12][13] and clinical practice also varies.
A common treatment regimen follows the 'DART' protocol that administered a total of 0.89 mg/kg of dexamethasone, tapered over 10 days. 12 However, it is common for EP/ELBW infants to receive doses that exceed those in the 'DART' protocol, or to receive more than one course.
The balance between the shorter-term benefits of postnatal systemic corticosteroids versus the longer-term risks is complex, and the safest and most effective postnatal corticosteroid regimen remains unclear. Furthermore, the relationship between cumulative postnatal corticosteroid dose and long-term neurodevelopmental outcomes is not well understood.
This observational study in EP or ELBW infants aimed to determine the association between the cumulative postnatal corticosteroid dose given to treat or prevent BPD and neurodevelopment at 2 years of age (corrected for prematurity).

| Perinatal data collection
All maternal, perinatal and neonatal data were collected prospectively. BPD was defined as supplemental oxygen dependency at 36 weeks' postmenstrual age. Postnatal corticosteroid use was defined as any systemic corticosteroids (either dexamethasone or hydrocortisone). The total cumulative dose of corticosteroid was calculated in milligrams per kilogram (mg/kg) dexamethasone equivalent, using the most recent infant weight available (usually recorded within the previous 3-7 days). Where corticosteroid courses were first administered with no corresponding weight recorded, either birth weight (if corticosteroid administration was within first 2 weeks after birth) or the closest recorded weight (administration after the first 2 weeks of life) was used. To standardise corticosteroid dosing, the cumulative dose of hydrocortisone was converted to a dexamethasone equivalent by dividing the cumulative hydrocortisone dose by a factor of 26.7. 15 For the purpose of this analysis, infants who received ≤5 doses of corticosteroid were excluded, as these brief courses were likely prescribed for treatment of hypotension or upper airway obstruction, and not for BPD.

Key Notes
• Although postnatal corticosteroids effectively treat bronchopulmonary dysplasia in extremely preterm or extremely low birthweight infants, they are associated with long-term harm.
• Little is known about the association of higher cumulative postnatal steroid dose and long-term outcomes.
• In this longitudinal cohort study of extremely preterm/ extremely low birth weight infants, higher cumulative postnatal corticosteroid dose was associated with increased odds of cerebral palsy at 2 years' corrected age. and severity of CP, blindness (visual acuity <6/60 in the better eye), deafness (hearing loss requiring amplification or a cochlear implant, or worse) and developmental delay. CP was determined by abnormal tone and reflexes, and a loss of motor function, with severity determined by the Gross Motor Function Classification System (GMFCS). 16 Moderate-severe CP was defined as GMFCS level 2-5.
Cognitive, language and motor development were assessed with the Bayley Scales of Infant and Toddler Development, 3rd edition (Bayley-III). 17 Major developmental delay was defined as greater than 2 standard deviations (SD) below the mean of term-born controls for either the cognitive of language composite scores on the Bayley-III. A child who was unable to complete the psychological testing because of severe developmental delay was assigned a score of −4 SD. Major neurodevelopmental disability was defined as the presence of any one or more of: moderate-severe CP, blindness, deafness or major developmental delay. 14

| Statistical analysis
Data were analysed using STATA v17 (StataCorp). Participant characteristics were summarised using means (SD) for normally distributed continuous data, medians (interquartile ranges, IQR) for skewed continuous data and number and proportion (percent) for categorical data. To explore the associations between postnatal corticosteroid cumulative dose and outcomes, linear or logistic regressions were performed, fitted using generalised estimating equations and reported with robust (sandwich) estimates of standard errors to account for non-independence of outcomes related to multiple births within the same family.
The analyses were performed firstly unadjusted, and then adjusted for potential confounding perinatal characteristics (gestational age at birth, sex, birth weight and major (Grade 3 or 4) intraventricular haemorrhage). Adjustments were based on a directed acyclic graph in order to identify the possible confounding variables ( Figure 1). Associations were reported as coefficients or odds ratios (ORs) with 95% confidence intervals (CIs). We acknowledge multiple comparisons and focus on the overall strength of the evidence rather than presenting p-values.

| RE SULTS
There were 309 EP/ELBW infants who were offered intensive care. Infants in the 'steroid' group were more immature (mean 24.9 vs. 26.7 weeks' gestation) and had a lower birth weight (mean 720 vs. 879 g) than infants in the 'no-steroid' group ( Table 1)

| DISCUSS ION
In a population cohort of infants born EP/ELBW, higher cumula- While the mean reduction in Bayley scores were in the range of 1.6-2.3 points for each of the cognitive, language and motor domains, the confidence intervals extended to more than 4 points for the language domain which is a clinically important reduction in developmental scores. 17 It is important to note that infants who received higher cumulative doses of postnatal corticosteroids were generally 'sicker' and thus were already at greater risk of adverse neurodevelopmental outcomes including cerebral palsy, developmental delay and major disability at 2 years' corrected age. The strengths of this study include the granularity of data available which enabled us to ascertain the true cumulative dose received by individual EP/ELBW infants rather than an intended dose regi-  Program. The funding sources had no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report and in the decision to submit the paper for publication.

CO N FLI C T O F I NTER E S T S TATEM ENT
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
The deidentified data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions Abbreviations: CI, confidence interval; OR, odds ratio. a Adjusted for gestational age at birth, sex, birth weight and major intraventricular haemorrhage.