Maternal Hypothyroidism Identified as Possible Autism Risk

Pam Harrison

May 18, 2015

SALT LAKE CITY — Subclinical maternal hypothyroidism may contribute to characteristics of autism spectrum disorder (AASD), new research shows.

Investigators found that elevated prenatal levels of thyroid stimulating hormone (TSH) and reduced levels of free thyroxin (fT4) in women are associated with higher scores on the Autism Observation Scale for Infants (AOSI) at 12 months of age.

"If subclinical hypothyroidism causes ASD, then we have potential for prevention," study investigator Igor Burstyn, PhD, Drexel University, Philadelphia, Pennsylvania, told Medscape Medical News.

"Next steps would be to determine if this association is seen with respect to other ASD outcomes, like diagnosis is seen in other study samples, and to better understand why we see the association," he added.

The study was presented at the International Meeting for Autism Research (IMFAR) 2015.

Too Early for Screening Recommendations

Investigators measured thyroid hormone levels in the earliest available prenatal serum samples for 180 mothers in the EARLI cohort.

The EARLI cohort is a high-risk pregnancy cohort made up of mothers of a child with ASD who were followed from the start of a subsequent pregnancy.

The majority of the prenatal serum samples were taken during the second trimester.

"We looked at the AOSI score in several ways, because it was not clear how to best analyze these data," Dr Burstyn explained.

By looking at the data as a continuous variable (parameterized), they were able to evaluate the relationship between a change in thyroid hormone levels and a change in AOSI score.

They also "dichotomized" the data in order to evaluate the relationship between thyroid hormone levels and the chance that the AOSI score was above a particular cut point.

AOSI developers have noted that the higher the AOSI score at 12 months, the greater the likelihood that a toddler will receive an autism diagnosis at an older age.

A score of 7 or greater on the AOSI score has been suggested as a potentially useful initial cut point.

In the overall cohort, investigators identified 56 children who had a total AOSI score of 7 or greater.

"None of the mothers had both hormones outside clinical norms," researchers caution; the TSH levels were from 0.35 to 3.30 µIU/mL, and the fT4 levels were within 0.56 to 1.64 ng/dL.

However, mothers whose children had a total AOSI score of 7 or greater had higher maternal TSH concentrations and lower maternal fT4 concentrations compared with the reference cohort.

After adjusting for multiple potential confounders, "we still observed an increased risk of ASD endophenotype [characteristics] with increased concentration of TSH and decreased concentration of fT4," the authors note.

Indeed, within these normative ranges, a doubling of TSH concentrations increased the risk of a child having an AOSI of 7 or greater by about 35% (adjusted relative risk [RR] = 1.4).

The corresponding effect of halving the fT4 concentration doubled the risk of a child having an AOSI score of 7 or greater (adjusted RR = 1.9).

Findings also suggest that an infant's AOSI score of 7 or greater at 12 months was largely predictive of a "best estimate" clinical diagnosis of ASD at 36 months, with a sensitivity of 70% and a specificity of 90%.

Adjustment for potential confounders had a negligible impact on effect estimates, and results were similar when the AOSI was either parameterized as a continuous variable or dichotomized at an AOSI of 6 or greater.

"We do not think there is sufficient evidence yet to recommend that physicians start looking at TSH and fT4 test results in order to advise mothers about the chances that their baby will go on to have ASD," Dr Burstyn cautioned.

"However, in the EARLI study, we should soon be able to look at how TSH and fT4 relate to different measures of ASD endophenotype and at a gold standard diagnosis of ASD when children reach 3 years of age," he said.

"After that, we will still need to look to other larger studies, ideally with measurements of the fluctuations in thyroid hormones throughout pregnancy, to get to the point where we could make recommendations about treatment."

Dr Burstyn reports no relevant financial relationships.

International Meeting for Autism Research (IMFAR) 2015. Abstract 19238. Presented May 14, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.