I say contentious because obesity is, pardon the pun, big news in health circles as we are all bombarded with news about how rates are rising and how obesity increases risk for lots of other conditions including an early exit from this mortal coil. Whilst not wishing to belittle the 'obesity crisis', I hark back to a previous post last year on how the projections about obesity rates are still a little fuzzy; to quote Physicist Niels Bohr: "prediction is very difficult, especially about the future".
The Krakowiak study has generated quite a few headlines. Probably the most direct, and at the same time inaccurate, is this one 'Obesity in pregnancy causing rise in autism cases' which makes so many assumptions that I really don't know where to begin. The latest CDC figures on autism, 1 in 88, have the potential to be contorted to fit lots of arguments about autism causation it seems.
What did the study actually say:
- It was another side to the CHARGE initiative which included over 500 children diagnosed with an autism spectrum condition, an additional 170 or so children with a developmental delay/disorder and 315 asymptomatic controls. The aim was to ascertain whether various metabolic conditions (obesity, diabetes, hypertension) forming part of the so-called metabolic syndrome might show association with either autism or elements related to offspring development. Children were all diagnosed with autism and importantly the diagnosis was independently confirmed as part of the initiative. Additional data on things like maternal health and weight were derived from either medical records or structured interview with mums.
- Based on the collected data, all metabolic conditions were more prevalent in mums of children with autism compared with controls; for example, cases of type-2 diabetes or gestational diabetes were present in 9.3% and 11.6% of autism and developmental delay groups respectively compared with only 6.4% of controls. The elevated odds ratio (OR) reflected this greater frequency both for autism and developmental disorders compared with controls. Interestingly, the OR was actually greater for cases where developmental disorders were present (2.35) over autism cases (1.61). Mums with maternal obesity (body mass index, BMI over 30) had a 60% increased risk of having a child with autism compared with non-obese mothers and risk increased further if diabetes or hypertension were present before or during pregnancy.
- Based on measures of cognitive and adaptive development, the Mullen Scales of Early Learning (MSEL) and an old favourite, the Vineland Adaptive Behaviour Scales (VABS), the presence of maternal diabetes in the autism group correlated with more significant 'deficits' in expressive language. Having said that, excluding children with autism from the analysis also showed that metabolic conditions collectively were associated with poorer functioning in areas of motor skills and receptive and expressive language.
Where do we go with this I thought I heard you ask? Well, I'm pretty sure that if you are what might be described as a slim-figured mother with a child with autism, you probably aren't all that impressed with this latest contribution. Indeed this is a consistent feature when reporting studies of association in that association and heightened risk is not an absolute measure, so not every mum of a child with autism is obese and neither is every obese mum going to have a child with autism.
The authors do speculate on one or two ideas as to why they found an association between mum's metabolic health and risk of autism, mainly centred on insulin and the regulation of maternal glucose levels in relation to effects on offspring. The theory goes that foetal exposure to high levels of glucose results in an sustained overproduction of insulin which increase the requirement for oxygen inducing chronic intrauterine tissue hypoxia. Glucose has previously been mentioned on this blog (here).
Interestingly a similar process has also been suggested for schizophrenia** (which again has also been linked to maternal diabetes). Indeed the paper by Van Lieshout & Voruganti** (full-text) offers a couple of other possible explanations which might also be relevant to cases of autism including a role for those dastardly proinflammatory cytokines and/or oxidative stress. Similar findings also go for ADHD.
* Krakowiak P. et al. Maternal metabolic conditions and risk for autism and other neurodevelopmental disorders. Pediatrics. April 2012
** Van Lieshout RJ. & Voruganti LP. Diabetes mellitus during pregnancy and increased risk of schizophrenia in offspring: a review of the evidence and putative mechanisms. Journal of Psychiatry & Neuroscience. 2008; 33: 395-404.