A maternal rheumatoid arthritis (RA) diagnosis before delivery was linked to an increased risk of offspring with autism spectrum disorder, particularly for those with seronegative RA.
In a cohort of Swedish mothers, a rheumatoid arthritis (RA) diagnosis before delivery was linked to an increased risk of offspring Autism Spectrum Disorder (ASD), according to a study published in Cambridge University Press.1 The association between maternal arthralgia and ASD indicated other pathways of risk than autoimmunity and inflammation, acting jointly or independently of RA.
The etiology of ASD, a chronic neurodevelopmental disorder that affects 1-2% of children worldwide, has been associated with both genetic factors and non-genetic factors. Although inflammatory and immune-related factors, such as immune activity during pregnancy, are a part of the non-genetic etiology portion, maternal RA diagnosis during pregnancy and ASD in children are limited and have shown mixed results.2
“RA is a heterogeneous condition, and the presence of rheumatoid factor or antibodies to citrullinated peptides (‘seropositive’) defines an entity that is distinct from the absence of these markers (‘seronegative’),” wrote Weiyao Yin MD, PhD, Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden, and colleagues. “The current study is the first to examine RA serostatus in relation to offspring ASD. To examine the specificity of RA in the association with ASD, never done before, we included a large comparison group with similar symptomology as RA, ie, arthralgia, which includes other non-inflammatory states of joint problems such as unspecific joint pain.”
The prospective, population-based cohort study included all children born alive in Sweden between 1995 and 2015, with a follow-up period through 2017. Diagnoses of either RA and ASD were clinically ascertained from the National Patient Register. After adjustments for confounders, investigators quantified the association by hazard ratios (HR) and 95% confidence intervals (CIs). The etiological subgroups, the timing of exposure, and RA serostatus were analyzed. A negative control group for RA, arthralgia, which had similar symptoms but were free from autoimmunity and inflammation, was used to determine the underlying mechanism for this association.
Of the 3629 children born to mothers with an RA diagnosis, 1.94% (n = 70) were ultimately diagnoses with ASD, compared with 1.92% (n = 28,892) of 1,503,908 children born to mothers without RA. A maternal RA diagnosis before delivery was linked to an increased risk of offspring with ASD (HR = 1.43, 95% CI 1.11–1.84), particularly for those with seronegative RA (HR = 1.61, 95% CI 1.12–2.30). A similar association was not demonstrated for paternal RA, RA diagnosis after delivery, or for maternal sisters with RA. Maternal arthralgia was observed as high risks for offspring ASD as well as maternal RA (HR = 1.41, 95% CI 1.24–1.60).
Investigators noted that strengths of the study included the large population-based cohort design with prospective ascertainment of both RA and ASD, which were diagnosed by experienced specialists. Additionally, selection bias was minimized by using data from a publicly financed health care system with a nearly complete follow-up and equal access. The database allowed for addressing confounding and modifying factors, as well as determining the specificity of RA by including a negative control group.
However, investigators stated that statistical precision may have been limited in subgroups. They were also unable to access information on RA disease activity or medications received during pregnancy, which may also play a role in the causal path between offspring ASD and maternal RA.
“Collectively, this indicates an increased ASD risk specific to RA exposure before delivery,” investigators concluded. “Maternal arthralgia a non-inflammatory condition, displayed comparably high risks for ASD as did maternal RA, irrespective of the timing of the arthralgia diagnosis. This suggests that alternative, or additional, pathways of risk than autoimmunity/inflammation are present in RA.”