August 12, 2013

expert reaction to induced/augmented childbirth and autism

An analysis of birth and educational records in JAMA Pediatrics suggested that induction and augmentation (increasing the strength, duration, or frequency of uterine contractions) during childbirth appears to be associated with increased odds of autism diagnosis in childhood.

Professor Mike Ludwig, Professor of Neurophysiology, University of Edinburgh, said:

“The effect size is small and authors mention a number of limitations of their own study.  In addition, there is no mention of stress and/or anxiety levels, or nutritional status of the mothers, and many other factors which could have contributed to ASD later in lives of the children.

“Current evidence linking oxytocin and autism is still very weak and therefore these findings do not, in my opinion, outweigh the significant maternal and foetal benefits of labour induction and augmentation.”

Professor Gareth Leng, Professor of Experimental Physiology, University of Edinburgh, said:

“Autism comprises a broad spectrum of developmental disturbances. The causes of these disturbances are still poorly understood, but they include both genetic factors and environmental factors. This study asked whether there was any link between the way that childbirth is managed and autism, by analysing the records of more than 625,000 children born in North Carolina.  The study found a weak association, in that children born to mothers who had been given oxytocin to initiate or augment labor were slightly more likely to subsequently develop autism.

“This study shows an association, it does not demonstrate that there is a causal link, so it is important to consider different explanations for the association. Most obviously, labour is induced or augmented in mothers in whom labour is not proceeding as fast as in mothers for whom no intervention is needed, and they may be more anxious, more tired, and more distressed as a result. In some of these mothers, labour will be progressing slowly for undiagnosed reasons – reasons that may independently be associated with developmental disturbances. Thus it is not necessarily the case that the management of labour is a cause of autism – it may be that both slow labour in mothers and developmental problems in the child have a common cause in some cases, or it may be that experiencing a slow or difficult labour is in some cases a risk factor for autism.

“The authors speculate that the oxytocin infusions that are used to augment or initiate labour may be a direct cause of the autism. This seems extremely unlikely and is not supported by any other evidence that I am aware of. First, oxytocin does not cross either the blood-brain barrier, or the placenta, so oxytocin that is given in these circumstances will not reach the brain of the fetus. Second oxytocin is released into the mother’s circulation in very large amounts during normal labour. The main purpose of this oxytocin release during labor is to strengthen the uterine contractions and so speed up the course of delivery, so women who require induction and augmentation of labour are likely to include some in whom there is a deficiency in normal oxytocin production or release.”

Professor Kevin McConway, Professor of Applied Statistics at The Open University, said:

“Because the number of diagnosed cases of autism has been rising over time, and because its causation remains unclear in many respects, there have been several previous studies to investigate possible associations between events around the time of birth and autism. Two recent major reviews of this work (they are references 13 and 31 in the new paper) both came to similar conclusions: there is some evidence that autism is associated with certain risk factors that might loosely be put under the heading of ‘a difficult birth’. But the reviews point out that the effects of these risk factors are generally quite small, and that it is very far from clear what is causing what.

“This new study is well conducted and generally clearly reported, and it is based on large numbers of births and relatively large numbers of diagnoses of autism. The statistical analysis is careful, and as far as possible tries to clarify the picture by taking account of a range of other factors apart from induced and augment births. But, however well done, no study of this kind can really separate out what is causing what. The researchers found, in male children, a statistically significant association between induction of labour or augmentation and a diagnosis of autism in childhood. However, the association is not very strong. Of the boys in this study, altogether about 13 in 1000 were recorded as having autism. The researchers report that the risk of autism in boy babies who have an induced or augmented labour is 15% to 30% higher than for those who do not, the exact amount depending on whether the labour was induced, augmented or both. That sounds quite impressive, but since the risk of autism is not actually very high, taking away the effect of induction or augmentation would not make a huge difference.

“If, magically somehow, all the mothers of boys who had induction or augmentation could have not had it, while everything else remained the same, then the number of their children with autism diagnoses would reduce by only 2 in 1000. The association in girls is even weaker, too small to be of statistical significance. Maybe a reduction of 2 in 1000 would still be worthwhile, if only we actually knew that we could achieve it by stopping inductions. But we can’t. Firstly, we still don’t know if the induced or augmented labours have anything directly to do with causing autism. The researchers themselves make this very clear. Maybe something else – some genetic factor, or something that occurred earlier in pregnancy or during the birth that the researchers could not take into account – is causing some babies to be more likely to have an indication of labour and separately making it more likely that they have autism. Second, we can’t just stop doing inductions or augmentations. As the researchers very clearly point out, there are very large benefits to the mother and to the baby of carrying out induction of labour or augmentation in appropriate cases. Even if it turns out that we could save a few children from autism by inducing many fewer labours, that would come at a major cost in other complications, including deaths to mothers and babies. All the authors are calling for is more research to try to separate all these factors out. There’s certainly nothing here that pregnant women or new mothers should be particularly concerned about, in my view. The researchers make it explicit that their work gives no reason to change medical practice, and I absolutely agree with them on that.”

‘Association of Autism With Induced or Augmented Childbirth in North Carolina Birth Record (1990-1998) and Education Research (1997-2007) Databases’ by Gregory et al. published in JAMA Pediatrics on Monday 12th August.