Children born to mothers with systemic lupus erythematosus (SLE) have more than double the risk of being diagnosed with an autism spectrum disorder, Canadian researchers reported.

Among children whose mothers had lupus, 1.4% were given a diagnosis of autism compared with 0.6% of control children, according to Sasha Bernatsky, MD, PhD, of McGill University in Montreal, and colleagues.

In a multivariate analysis, the odds ratio for autism among children whose mothers had lupus was 2.19 (95% CI 1.09-4.39), the researchers reported online in Arthritis & Rheumatism.

Some research has suggested that children whose mothers have lupus are at greater risk of neurodevelopmental disorders, and that exposure to maternal antibodies and cytokines may contribute to the risk.

“There’s a growing body of literature that suggests that genetics alone is not the total answer to autism, and that in utero environment plays a role,” said Betty Diamond, MD, of the Feinstein Institute for Medical Research in Manhasset, N.Y.

“There are data suggesting that maternal cytokines and brain-reactive antibodies can lead to abnormal brain development. I think there certainly is good reason to think that maternal antibodies can cross the placenta and access the fetal brain before the blood-brain barrier is fully competent,” Diamond told MedPage Today.

To examine the possible association between maternal lupus and autism in the offspring, Bernatsky and colleagues assembled a cohort of 719 children whose mothers had lupus, along with 8,493 matched controls from the general population. Mean maternal age was 30, and duration of lupus was 3.7 years.

The mothers with lupus had more comorbidities and had more complications such as preterm birth.

Overall, the diagnosis of autism spectrum disorder was an infrequent event, with an incidence rate of 75.2 per 100,000 person-years.

The age at the time of autism diagnosis was younger for children whose mothers had lupus, at 3.8 years compared with 5.7 years for controls.

In a sensitivity analysis that adjusted for complications such as gestational diabetes and preterm birth, the OR of autism was increased, at 1.97 (95% CI 0.95-4.08), although the confidence interval crossed 1.

Other factors that were associated with autism included gestational diabetes and being small for gestational age, with ORs of 2.42 and 1.69, respectively, although the confidence intervals again were wide and the number of events was small.

Analysis of a subset of the children whose mothers had records of medication exposure found that none of the children with autism had been exposed to antimalarial drugs or immunosuppressants. One case with an SLE mother had been exposed to corticosteroids and one control child had been exposed to anticonvulsants.

The finding that the autism diagnosis was typically earlier in the children whose mothers had lupus “raises concerns as to whether there might be a different clinical presentation of autism spectrum disorder in children born to SLE mothers (e.g., earlier and/or more severe disease presentation) versus control children,” Bernatsky and colleagues noted.

Adding further support to the possibility of a different phenotype in children whose mothers have lupus was another study of the same cohort in which attention deficit hyperactivity disorder was diagnosed significantly later than in controls, at age 12.5 years vs 7.8 years. That finding suggests that the earlier diagnosis of autism was not simply a result of mothers’ seeking medical advice earlier because of concerns about their babies’ potential health problems related to SLE.

“Our study findings should prompt future research, notably on the role of maternal SLE-related autoantibodies, which could yield important insights into the physiopathology of these complex disorders,” the researchers concluded.

“I think we have a lot of work to do to understand which antibodies might be the perpetrators,” Diamond said.

“And also whether the relationship is multifactorial, that there’s something else going on with the fetus that might need to happen for there to be this long-term pathology,” she said.

“For example, with fetal heart block in children with SLE mothers, there are data that you need not only anti-Ro antibodies from the mother, but you also need some cardiac ischemia in order for the antigen to be exposed in a way for the antibody to do its damage,” she explained.

Limitations of the study included the possibility of unmeasured confounding and the lack of serologic information in the administrative databases.

###
Source Reference: Vinet E, et al “Increased risk of autism spectrum disorders in children born to women with systemic lupus erythematosus: results from the OSLER cohort” Arthritis Rheum 2015; DOI: 10.1002/art.39320.