"Steroid injections for premature babies could raise ADHD risk," reports The Daily Telegraph after a Finnish study found a link between steroid use (corticosteroids) in premature babies and developmental conditions such as attention deficit hyperactivity disorder (ADHD).
Steroids are sometimes given to pregnant women if they go into premature labour (particularly before 35 weeks) as they can help stimulate the development of the baby's lungs. This significantly reduces the risk of premature babies developing a serious and potentially fatal breathing condition known as neonatal respiratory distress syndrome (NRDS).
Because of the use of steroids and breathing equipment, deaths from NRDS are now extremely rare in the UK. However, concerns have been raised that using steroids at such an early stage of a child's development could cause problems in later life, such as ADHD.
The study involved children born in Finland in 1986 who were followed up at the ages of eight and 16, when they were assessed using various behavioural scales.
The research included 37 children who were exposed to corticosteroids before birth. They were matched by sex and gestational (pregnancy) age at birth with around 6,000 unexposed children.
The researchers found associations between pre-birth steroid exposure and ADHD-like behavioural scores at the age of eight, but the associations were not significant at 16.
A significant limitation of the study was its small sample size – it involved just 37 children from one region of Finland. For this reason, the research should be treated as exploratory. Further research into the potential risk is needed, but it is likely that any risk associated with steroid use is likely to be outweighed by the benefits of preventing NRDS.
Where did the story come from?
The study was carried out by researchers from Imperial College London, the University of Oulu, Finland, and Mid Sweden University, Östersund, Sweden. Funding was provided by the Academy of Finland, the Sigrid Jusélius Foundation, Finland, the Thule Institute at the University of Oulu, Finland, the National Institute of Mental Health, US, and euro-blcs at Imperial College London.
Both The Daily Telegraph and BBC News accurately and responsibly reported the findings of this study. BBC News included an important quote from Professor Vivette Glover, Imperial College London, who said that, "These are really important and lifesaving drugs. These findings shouldn't affect clinical practice and parents shouldn't worry".
The Telegraph also makes it clear that the benefits of using steroids are thought by most clinicians to outweigh any risk.
What kind of research was this?
This study looked at whether there is a link between pre-birth exposure to steroids and long-term mental health outcomes.
Babies who are born prematurely (prior to 37 completed weeks of pregnancy) have an increased risk of various problems, with generally a greater risk the earlier the baby is born.
One risk is that the baby has breathing problems because their lungs aren't yet fully developed. This problem is more likely if the baby is born at less than 35 weeks of pregnancy. Corticosteroids may be given to try to reduce the risk of the baby developing lung problems such as neonatal respiratory distress syndrome (NRDS).
However, the researchers discuss how levels of corticosteroid hormone – both naturally occurring in the body and synthetic steroid hormones – have been demonstrated to have an effect on the developing brain in animal models.
They further suggest that although previous research has found an association between increased maternal stress during pregnancy and ADHD in the child, it could be that higher levels of stress hormones, such as cortisol, could actually be responsible for this association.
The researchers say that very little is known about corticosteroid hormones administered to the mother prior to preterm birth and what effect they could have on child behaviour, including ADHD symptoms.
This study aimed to examine this by comparing a small sample of children exposed to synthetic corticosteroid hormones (glucocorticoids) while in the uterus with children of the same average gestational age who had not been exposed to these drugs.
What did the research involve?
This study used participants in the Northern Finland Birth Cohort, which recruited pregnant women in 1986. The study included 8,954 live single-born babies with consent to use their data.
Women completed self-report questionnaires during pregnancy, and antenatal and birth information was made available through hospital records. The researchers screened for the use of pre-birth synthetic glucocorticoids (sCGs) by performing a systematic review of the records.
The researchers say how in 1986, the use of sCG in pregnancy was controversial, so they only identified 37 children who had been exposed during pregnancy.
They excluded 11 children who had been exposed to sCG less than four days before birth, as this would be unlikely to have an effect on foetal brain development. They also excluded the slightly different steroid hormones that had been used to treat maternal inflammatory or allergic conditions.
Children were followed up at eight and 16 years. Childhood mental health was screened at eight years using the validated Rutter's Behaviour Scale (B2), which included subsections for neurotic, antisocial and inattention hyperactivity.
At 16 years, parents reported adolescent behaviour using the Strengths and Weaknesses of ADHD symptoms and Normal behavior (SWAN) scale. The SWAN scale consists of 18 items based on the symptoms of ADHD listed in the Diagnostic and Statistical Manual of Mental Disorders IV (DSM-IV). Adolescents also completed the Youth Self-Report (YSR), a widely used questionnaire for 11 to 18-year-olds derived from the Child Behavior Checklist (CBCL).
The researchers took various potential confounders into account that could be related to both sCG use and child mental health problems:
- child sex
- mother's age, education and marital status (measured in 1986 at recruitment only)
- gestational age at birth
- total pre-birth sGC dose (mg)
- interval between pre-birth sGC exposure and birth (days)
- mother's number of previous pregnancies
- mother's pre-pregnancy body mass index (BMI)
- mother's smoking status during pregnancy
- pregnancy complications of high blood pressure, pre-eclampsia or placenta praevia (placenta positioned over the cervix)
The researchers matched children exposed to sCG with those who were not exposed on the basis of sex and gestational age. They did this because preterm birth is itself associated with increased risk of poor mental health outcomes, and males are more vulnerable to mental health problems in childhood.
The researchers carried out linear models looking for the association between sCG use and child mental health problems, adjusting their analyses for the confounders.
At eight years, they had a total of 6,116 children available for analysis: 37 exposed to sGC cases and 6,079 controls. At 16 years, they had 5,108 adolescents available: 29 sGC cases and 5,079 controls.
What were the basic results?
There were reportedly no differences between sex and gestational age-matched cases and controls in terms of sociodemographic or other maternal medical factors.
There was a significant association between pre-birth sGC exposure and the total Rutter score and inattention hyperactivity scores at eight years. They also found consistent associations between pre-birth sGC exposure and each of the outcomes measured at 16 years, though none reached statistical significance.
How did the researchers interpret the results?
The researchers say that, "This study is the first to explore the long-term associations between prenatal exposure to sGC and mental health in childhood and adolescence.
"We found that both children and adolescents prenatally exposed to sGC scored consistently higher on internationally validated screening instruments of mental health, by teacher, parental and self-reports, than controls."
This research is exploratory and on its own does not prove that pregnancy exposure to corticosteroids causes ADHD.
The research does have strengths in that it matched exposed and non-exposed children on the basis of sex and gestational age. Matching for gestational age and prematurity in particular is important – prematurity is associated with pre-birth corticosteroid use, as well as an increased risk of adverse effects upon brain development and mental health. This could therefore confound the relationship.
The researchers further adjusted for various possible socioeconomic, medical and pregnancy-related factors. They also benefit from the use of validated scales to assess child mental health outcomes, as well as a large birth cohort.
However, despite the large birth cohort, which included almost 9,000 babies, only 37 were exposed to corticosteroids. It is not known for certain why the mothers were given corticosteroids.
It is most likely this was in expectation of premature birth, but the researchers excluded babies who were given steroids within four days of their birth, as they thought that this wouldn't have an effect on the baby's brain.
These children were also all born in one region of Finland during 1986. Therefore, how applicable the results are to pregnant women given corticosteroids prior to premature birth today is unclear. With a small sample of 37 children, it is possible that the results could be because of chance associations that would not be observed in a different sample of children exposed to steroids in the modern era.
The research also reported the scores on a symptom scale and did not say whether or not children had confirmed diagnoses of ADHD.
It is also worth noting that significant associations could be because of other unmeasured confounders and not directly from pregnancy corticosteroids.
Overall, corticosteroids have an important role in maternity care. It is not hyperbole to say that steroids have saved thousands of premature babies' lives.
For now, most experts would agree that the benefits – in terms of reducing babies' risk of breathing problems – outweigh the possible risk of ADHD symptoms developing in the future.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Daily Telegraph, 22 November 2013
BBC News, 23 November 2013
Links to the science
PLoS One. Published online November 22 2013