Assisted Reproduction Techniques and ASD

Background: Given clinicians working with Autistic Spectrum Disorders (ASD) and some in clinics specialising in Assisted Reproduction Techniques (ART) report an anecdotal association of ASD with the use of an ART, it is important to establish if ART is a concern for ASD.  A number of studies to date have undertaken assessment of the relationship between various assisted reproduction techniques (ART) and the incidence of Autistic Spectrum Disorders (ASD).  While some find an effect, others don’t.  However, none of these studies have controlled for all important confounds, such as paternal age, maternal age, birth weight, and familial causes.

Objectives: We sought to measure if ART increases the likelihood of a later diagnosis of an ASD after controlling for maternal and paternal age, birth weight, and having a relative with a similar diagnosis. We hypothesised that after these variables were controlled, users of ART would be more likely to have a child later diagnosed with an ASD.

Methods: We undertook a retrospective online questionnaire. The instrument asked parents (n=989) to respond if they had a child with an ASD regarding that child (n=520), or if they did not have a child with an ASD (n=469), to respond regarding either their oldest (n=227) or youngest child (n=242), depending upon which version of the questionnaire they were invited to complete. The instrument assessed diagnostic status, maternal age at birth, paternal age at birth, birth weight, biological relationship to others with a similar diagnosis, and ART type. 

Results: We obtained results for 98 women who reported having used an ART.  Of these, 52 had used In Vitro Fertilisation (IVF) or IntraCytoplasmic Sperm Injection (ICSI), 29 had used hormonal or medical ovulatory induction, with the remaining 17 either not detailing the technique or using various other ARTs.  On average, mothers having used an ART were 3.6 years older (p<0.05), fathers were 3.9 years older (p<0.05), and infants born following these techniques were 310 grams lighter (p<0.05).  We found the children who were later diagnosed with ASD had older mothers (1.08 years, p<0.05), but not older fathers (0.61 years, p=ns), and were on average 110 grams lighter (p<0.05).  Interestingly, after controlling for maternal age, paternal age, birth weight, and having a genetic relative with a similar diagnosis, not all ARTs were associated with an increased risk of a later ASD diagnosis.  However, certain ARTs were associated with ASD.  Use of either ICSI or IVF was found to increase the risk of a later diagnosis by 2.12 times (95%CI 1.08 – 4.13; p<0.05).

Conclusions: The use of manipulative ARTs, such as IVF or ICSI increased the risk of later diagnoses of ASD, while use of hormonal or medical techniques did not.  This suggests that the elevated risk from some techniques is not due to teratogenic effects, but may be due to manipulation of gametes prior to implantation.