Abstract
Study question
Does the risk of neurodevelopmental disorders differ between singletons born after various assisted reproductive techniques (ART) and spontaneous conception (SC) until young adulthood?
Summary answer
ART children had a slightly increased rate of learning and motor functioning disorders, autism spectrum disorders (ASD), and ADHD and conduct disorders.
What is known already
Studies on the impact of ART on offspring have reported both increased risk and comparable incidences of neurodevelopmental disorders between ART and SC offspring. The most studied neurodevelopmental disorders with ART are autism spectrum disorders (ASD.) There is, however, no consensus on the risk of ASD for ART children. The risk for other neurodevelopmental disorders, like attention-deficit hyperactivity disorders (ADHD) or tic disorder among ART children, is also a debated issue, as studies are scarce.
Study design, size, duration
A Nordic register-based cohort study including all singleton live births (N = 5 076 444) after ART (n = 116 909) or SC (n = 4 959 535) between 1995 and 2014 in Denmark and Finland, 1995 and 2015 in Sweden; and 2005 and 2015 in Norway. Children with intellectual disability (ICD-10: F70-F79) are excluded. The children are followed up to young adulthood (the year 2014 in Denmark and Finland, and 2015 in Norway and Sweden).
Participants/materials, setting, methods
Offspring outcomes were defined as following ICD-10 diagnoses: learning and motor functioning disorders (F80-83), ASD (F84), ADHD and conduct disorders (F90-F92), and tic disorders/Tourette (F95). We calculated crude and adjusted hazard ratios (HR) for neurodevelopmental diagnoses using Cox regression. Adjustments were made for the country, maternal age at the delivery, parity, smoking, and maternal psychiatric morbidity.
Main results and the role of chance
The cumulative incidences of neurodevelopmental disorders in the cohort were 1.74% for F90-F92, 1.40% for F80-83, 0.66% for F84, and 0.22% for F95. In crude Cox-regression ART children had an increased likelihood during the follow-up of being diagnosed with F84 (HR 1.12 [95% CI 1.04-1.21]) and F95 (HR 1.21 [95% CI 1.06-1.38]), but not with F80-83 (HR 1.01 [95% CI 0.96-1.07]) or F90-92 (HR 0.82 [95% CI 0.77-0.86]). After adjustments the likelihood was increased for F80-83 (HR 1.20 [95% CI 1.13-1.27]), F84 (HR 1.12 [95% CI 1.03-1.24]), and F90-92 (HR 1.09 [95% CI 1.04-1.19]), but nor for F95 (HR 1.13 [95% CI 0.99-1.30]).
After adjustments, intracytoplasmic sperm injection children compared with in vitro fertilization children had similar likelihood during follow-up for F80-83 (1.06 [95% CI 0.89–1.25]), for F84 (HR 0.92 [95% CI 0.76–1.11]), for F90-92 (HR 0.96 [95% CI 0.83–1.12]), and for F95 (HR 1.16 [95% CI 0.83–1.63]).
After adjustments, frozen embryo transfer children compared with fresh embryo transfer children had similar likelihood during follow-up for F80-83 (HR 1.11 [95% CI 0.90–1.37]), F84 (HR 0.98 [95% CI 0.76–1.27]), F90-92 (HR 0.96 [95% CI 0.78–1.19]), and F95 (HR 0.83 [95% CI 0.51–1.35]).
Limitations, reasons for caution
There may be residual confounding by unknown or unmeasured confounders. We lack information on possible confounders like the reason and length of infertility, maternal substance use other than self-reported smoking status, paternal age, and parental somatic morbidity. Additional limitations are differences in registration practice and data availability between study countries.
Wider implications of the findings
This is the largest singleton cohort and the first multinational study on the risk for neurodevelopmental disorders among ART children. While the rate of some neurodevelopmental disorders was increased among ART children, the absolute risk was moderate. The type of ART did not associate with the incidence of neurodevelopmental disorders.
Trial registration number
ISRCTN11780826
Schizophrenia, autism spectrum disorders and developmental disorders share specific disruptive coding mutations
