Thimerosal in Vaccines and Neuropsychological Outcomes?

William T. Basco, Jr, MD, FAAP


January 22, 2008

Early Thimerosal Exposure and Neuropsychological Outcomes at 7 to 10 Years

Thompson WW, Price C, Goodson B, et al, for the Vaccine Safety Datalink Team
N Engl J Med. 2007;357:1281-1292

The public concern that thimerosal in vaccines is linked with autism and other neurologic disorders has been a difficult concern to shake, despite several large epidemiologic studies that do not support the link.

This study was undertaken to more thoroughly evaluate exposure to mercury (thimerosal is ethyl mercury) through vaccines in prenatal, perinatal, and infancy periods to determine whether the extent of mercury exposure in vaccines was associated with neuropsychological outcomes. Data were derived from the Centers for Disease Control and Prevention's Vaccine Safety Datalink, a monitoring program in several large health maintenance organizations (HMOs). Four HMOs provided data for the current study.

Subjects were enrolled at 7-10 years of age and were generally healthy and typically developing children. The authors retrospectively reconstructed their exposures to mercury up until the subjects were 7 months old, reasoning that the bulk of thimerosal exposure (by weight) in vaccines occurs during the first half-year of life. Mothers were interviewed to determine possible fetal exposure to ethyl mercury, in the form of maternal vaccines or immune globulin exposure but also in diet and even dental work.

Data collected included measures of maternal intelligence, measures of the home environment, and other social factors that might influence testing performance. During a half-day assessment, the blinded testers evaluated the children on 42 neuropsychological performance measures, including assessments of language development and performance, fine motor functioning, general intelligence, measures of attention, and frequency of tics, among other domains.

The final analyses included 1047 children, but this was approximately 30% of those who were originally deemed eligible. The authors did not find differences in exposure rates between those who agreed to participate and those who did not. The median mercury exposure level from vaccines in the subjects was 112.5 micrograms. Only 11% of children had been exposed to mercury through maternal vaccination or administration of immune globulin. However, there was positive correlation between maternal IQ and cumulative exposure to thimerosal, suggesting that those children were more likely to receive all vaccines.

Overall, the authors detected no pattern of association with thimerosal exposure and developmental outcome. For example, prenatal exposure had a positive association (meaning that higher levels correlated with higher performance) with one measure and a negative association with another, and this only in the combined (male and female) subjects) and separately in males.

For exposures from birth to 7 months among males, increasing levels of thimerosal exposure were associated with improved letter and word recognition but poorer behavior ratings (parental assessment) and higher likelihood of tics. Females had positive correlations between thimerosal and 2 performance measures.

The authors identify 2 possible associations that have been found in other studies or were consistent enough within their own data to consider for further analyses: (1) the increased risk for tics in boys with postnatal exposure to thimerosal, and (2) the concern of speech and verbal IQ effects associated with prenatal exposure. They note that even these associations were minimal in regard to magnitude of effect, and they were offset by positive correlations in other performance measures.

The authors concluded that this study did not support a causal link between prenatal or postnatal thimerosal exposure and neuropsychological functioning in childhood.

Two commentaries accompany this article. One reviews the process of the Vaccine Injury Compensation Program and the progress of families of autistic children in these legal proceedings[1]; it is worth reading for the insight it provides into the more complex issue of vaccine injury and whether autism should even be considered as such. The second commentary reviews other studies that evaluated any possible link between thimerosal and health and the effects that the concern over thimerosal in vaccines have had on the vaccine industry and vaccine delivery.[2] Both are thought-provoking. Although the authors hint at the number of analyses completed being a statistical problem (42 outcomes, evaluated for the whole cohort, then by gender), the number of comparisons and the significance level of .05 virtually assured that several "statistically significant" associations would be found. The argument would be that these are random, and the fact that the associations are both positive and negative, some found only in one gender, would generally support the contention that these findings are more random than real.



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.
Post as: