Down Syndrome Abstract
of the Month:
Aug 2005
Longitudinal study of thyroid function in Down's syndrome in the first two decades.
Gibson PA, Newton RW, Selby K, Price DA, Leyland K, Addison GM
Arch Dis Child. 2005 Jun;90(6):574-8
Department of Paediatrics, Royal Lancaster Infirmary, Lancaster, UK.
Abstract:
AIMS AND METHODS: Thyroid function tests were initially carried out on 122 children with Down's syndrome aged 6-14 years and then repeated four to six years later in 103 adolescents (85% of the group of 122) when they were aged 10-20 years (median 14.4 years). At the second test two were hypothyroid and two with isolated raised thyroid stimulating hormone (IR-TSH) were receiving thyroxine. RESULTS: At the first test there were 98 (80%) euthyroid children: 83 were retested and four (5%) had IR-TSH. At the first test 24 had IR-TSH: 20 were retested and 14 (70%) had become normal. Seventeen with IR-TSH on initial testing had a thyrotrophin releasing hormone test within three months; TSH had become normal in eight (47%) of these children. There was no association between reported clinical symptoms and IR-TSH, but there were clear symptoms in one of the two with definite hypothyroidism. CONCLUSIONS: The likelihood ratio for a positive result on second testing when raised TSH and positive antibody status on first testing are combined is 20. This suggests initial testing results could be used as a basis to select a subgroup for further testing at say five yearly intervals unless new symptoms emerge in the interim. It also suggests that yearly screening (as recommended by the American Academy of Pediatrics, 2001) is probably not justified in the first 20 years of life.
My comments:
Two months ago, I reviewed this study that recommended starting all infants with Down syndrome on thyroid hormone replacement to ensure maximal growth and development. Now we have a study that suggests that the need for thyroid replacement is very uncommon. So what's going on here?In this study, 103 children with DS had two tests for thyroid disease, approximately four to six years later. In the initial testing, 83 children had normal tests and 20 had elevated TSH levels but normal thyroxine (T4) levels (this condition is referred to as "IR-TSH" by the authors). None had hypothyroidism in the initial test. In the second test, of the 83 children with normal initial testing, 2 developed hypothyroidism, 4 developed IR-TSH and the rest continued to have normal tests. When the 20 children who had initial IR-TSH were tested a second time, 1 had developed hypothyroidism, 14 became normal and the other 5 continued to have IR-TSH.
On the basis of only 3 children of the original 103 developing hypothyroidism over the space of four to six years, the authors state that yearly testing of all children with DS for thyroid disease is unnecessary. They do note that of the three children with DS that developed hypothyroidism, none had any signs of slowing growth.
The authors did spend some time discussing the children that had elevated TSH and normal T4 tests (in my essay on thyroid disease I call this "idiopathic hyperthyrotropinemia"). Since only one in this group developed hypothyroidism, it shows that this condition does not always mean impending thyroid failure. While they did do thyroid antibody testing, the authors do not mention if the one child with IR-TSH had positive antithyroid antibodies or not. However, 20 children is a very small group from which to draw assumptions. In fact, the study as a whole is too small to use as a reason to make any changes in the current recommendations.
In an accompanying commentary, one of the editors of the journal comments that the study does show yearly testing is unnecessary. However, the author makes two assumptions that are probably false: one is that low thyroid tests in children with DS may not be an actual abnormal condition since heart function is not adversely affected the way it is usually in the general population. I hasten to point out that low thyroid does have other measurable effects on children with DS. Second, the author points out that the most sensitive indicator of low thyroid production is a decrease in growth; yet this study showed that the three children who did become hypothyroid during the study had no such decrease in growth.
The bottom line is that thyroid disease is common in people with DS, not diagnosing it will cause a significant health risk, and the blood test is reliable. While I agree with the authors that isolated raised TSH levels do not need to be treated with thyroid replacement, I do not agree that decreased screening of thyroid disease is warranted.
Home Page | List of Past Abstracts | Contact Me