Little evidence for long‐term harm from antenatal corticosteroids in a population‐based very low birthweight young adult cohort. Issue 5 (16th May 2022)

• Record Type: Journal Article
• Authors: Darlow, Brian A.; Harris, Sarah L.; Horwood, L. John
• Other Names: Anderson Peter J. guestEditor.; Zeitlin Jennifer guestEditor.
• Is Part Of: Paediatric and perinatal epidemiology. Volume 36:Issue 5(2022)
• Page Start: 631
• Page End: 639
• Publication Date: 2022-05-16
• Subjects: antenatal corticosteroids -- follow‐up -- mental health -- physical health -- population‐based cohort study -- preterm young adults -- very low birthweight infants; Pediatrics -- Periodicals; Perinatology -- Periodicals; Pediatric epidemiology -- Periodicals; Infants (Newborn) -- Diseases -- Periodicals; 618.92
• DOI: 10.1111/ppe.12886
• Languages: English
• ISSNs: 0269-5022

Abstract

Abstract: Background: Antenatal corticosteroids (ACS) given to mothers with anticipated very preterm delivery are widely used and improve infant outcomes. Follow‐up studies of the first trials of ACS have shown no adverse effects, but recently there have been concerns about possible longer‐term harms. Objectives: We aimed to assess the relationship of ACS therapy to a range of physical health and welfare measures in a cohort of very low birthweight (VLBW; <1500 g) young adults. Methods: Population‐based cohort follow‐up study. All VLBW infants born in New Zealand in 1986 were included in a prospective audit of retinopathy of prematurity. Perinatal data collection included information on ACS. At 26–30 years, 250 of 323 (77%) survivors participated, 58% having received ACS, with 229 assessed in one centre, including cardiovascular, metabolic, respiratory and neurocognitive measures. Differences in outcome between those receiving/not receiving ACS were summarised by the mean difference for continuous outcomes supplemented by Cohen's d as a standardised measure of effect size (ES), and risk ratios (RRI) for dichotomous outcomes, adjusted for relevant covariates using generalised linear regression methods. Results: There were no or minimal adverse effects of receipt of ACS versus no receipt across a range of health and welfare outcomes, both for the full cohort (adjusted ES range d = 0.01–0.23; adjusted RR range 0.78–2.03) and for individuals with gestation <28 weeks (extremely preterm; EP), except for a small increase in rates of major depression. In EP adults, receipt of ACS was associated with a higher incidence of hypertension, but might have a small benefit for IQ. Conclusions: In this population‐based VLBW cohort, we detected minimal adverse outcomes associated with exposure to ACS by the third decade of life, a similar result to the 30‐year follow‐up of participants in the first ACS trial. However, further follow‐up is warranted.