This map reflects what the world was thinking when Christopher Columbus set out on his voyage to “discover” the New World. That’s kind of like where we are with Autism Spectrum Disorder (ASD). But there are many folks collaboratively engaged – parents, clinicians, educators, advocates, and scientists – who are working to add details to the map.
On the first of September researchers at Seattle Children’s and UW published a paper in Autism Research that identified a connection between exposure to diagnostic ultrasound in the first trimester of pregnancy and severity of autism symptoms observed in children diagnosed with autism. This connection was strongest when the children had certain genetic variations associated with ASD (called copy number variants). The paper has received some recent attention in the media so we wanted to take a moment to provide our take on the paper without any spin (one important caveat is that this blog author is one of the authors of the paper).
First and foremost, this study does not look at whether ultrasound causes autism or not. That is an important statement that should be clear. Given the way this study was designed, this question cannot be answered. Second, a key finding in this paper concerns the interaction between the genetic variations identified in the children, ultrasound, and the timing of exposure. The key point of the paper suggests there may be an interactive relationship between genetics and ultrasound exposure and the time of the exposure.
One more quick comment: this is the first paper to describe this interaction between ultrasound, genetics, timing and autism severity and therefore requires replication (repeat studies). However, science requires independent replication. Additional studies are needed and, importantly, should be conducted by different research groups. Independent replication is critical to moving science forward.
So, what should we do with this information? We should put it on our map in pencil. We should consider it one step in our ongoing move toward a better understanding of ASD, which we’ll color in a little more with each independent replication – or we can erase it if there isn’t any replication. We can only keep filling in the landscape if we continue working together. As we’ve said in other studies on autism, we recommend a healthy dose of caution along with curiosity when reading about this study.
How large was the sample size? Frequency and duration of ultrasound in first trimester that was correlated with increase in copy errors?
Thanks for your question. These are important questions in understanding the limitations and impact of this paper. The final sample size consisted of 1749 children diagnosed with ASD who had both medical history information and array sequencing completed. Information regarding pregnancy and general medical history for all probands was collected by a clinician in a semi-structured interview. Specific to this study, ultrasound exposure was assessed via this information and labeled as whether or not the biological mother received at least one ultrasound during the first trimester. The duration of ultrasound was unknown. I hope that provides some clarification. Thanks!