Article Text

Download PDFPDF

How often should we screen children with Down’s syndrome for hypothyroidism?
  1. G Van Vliet
  1. Correspondence to:
    Dr G Van Vliet
    Endocrinology Service and Research Center, Sainte-Justine Hospital, Department of Pediatrics, University of Montreal, 3175 Sainte-Catherine Road, Montreal H3T 1C5, Quebec, Canada;

Statistics from

Commentary on the paper by Gibson et al (see page 574)

Aside from mental retardation, the prevalence of several health problems is higher in individuals with Down’s syndrome than in the general population. Among these problems, hypothyroidism is often noted.1 At the outset, it is important to distinguish congenital from acquired hypothyroidism. Numerous references continue to state that the prevalence of congenital hypothyroidism is considerably higher in newborns with Down’s syndrome than in the general population.2 However, the evidence for this is unconvincing: in Quebec, there was no case of Down’s syndrome in several hundred children with permanent primary congenital hypothyroidism diagnosed through neonatal screening, suggesting that thyroid dysgenesis (ectopy or agenesis, which account for 85% of cases of congenital hypothyroidism) is not more frequent in Down’s syndrome.3 The coexistence of Down’s syndrome with severe persistent primary congenital hypothyroidism with a normally located gland of normal or increased size (“dyshormonogenesis”) has been reported in a few cases but may represent a chance association.4 On the other hand, the distribution of thyrotrophin (TSH) and thyroxine (T4) measured on neonatal blood spots are slightly shifted to the right and left, respectively, in newborns with Down’s syndrome compared to the general newborn population.5 These shifts are likely present in older individuals with Down’s syndrome but this has not been systematically studied. The bioactivity of thyrotrophin being normal, a state of mild thyroidal resistance to thyrotrophin was proposed;6 however, the genes coding for TSH receptor and Gs-alpha, two proteins known to be involved in mild TSH resistance, are normal in patients with Down’s syndrome.7 Furthermore, it is questionable to label as hypothyroidism this shift of thyroid function values of the whole population with Down’s syndrome, especially in view of the fact that they do not seem to be associated with abnormal myocardial structure and function.8

The higher prevalence of acquired hypothyroidism, mostly from autoimmune thyroid diseases, in individuals with Down’s syndrome is better established. For instance, one of 67 patients younger than 22 years receiving thyroxine in a health region of Scotland had Down’s syndrome, clearly higher than expected;9 however, the type (congenital or acquired) and the severity of hypothyroidism was not known in all these patients. The study by Gibson et al in this issue addresses the question of the frequency of screening that is necessary to detect hypothyroidism at a preclinical stage in young patients with Down’s syndrome.10 The authors measured TSH and T4 levels and the thyroid antimicrosomal antibody titre in 122 children with Down’s syndrome at a median age of 9.8 years (range 6–14). One hundred and three subjects were sampled again at a median age of 14.4 years (range 10–20). At first sampling, the commonest abnormality was isolated hyperthyrotropinaemia, which was observed in 20 patients; however, only one of those was found to be hypothyroid at second testing, thyrotrophin levels had become normal in 14 and isolated hyperthyrotropinaemia persisted in five. Hypothyroidism was also present at second sampling in two patients who were euthyroid at first sampling. Isolated hyperthyrotropinaemia was not associated with reported clinical symptoms. The presence of thyroid autoantibodies on the first test was not significantly associated with hypothyroidism on the second test, although the numbers on which this was evaluated are very small.

The considerations above and the findings of Gibson et al clearly challenge the recommendations of the American Academy of Pediatrics that thyroid function should be evaluated at 6 and 12 months and annually thereafter in all children with Down’s syndrome.11 One should be aware that US based recommendations are sometimes influenced by considerations that are not of a purely medical nature, such as the need to obtain private insurance coverage for laboratory testing or fear of litigation. Rather than accepting these recommendations at face value, it is appropriate that investigators in every jurisdiction evaluate the evidence supporting a given frequency of screening; the study by Gibson et al is a step in the right direction.

There are, however, several limitations to the study of Gibson et al. Its relatively small sample size has already been mentioned. On the other hand, the two patients who developed hypothyroidism by the time of the second screening did not have growth failure. Given the exquisite sensitivity of post-natal linear growth to hypothyroidism, this is surprising and would suggest that hypothyroidism was of recent onset. In fact, as in all children, deceleration of linear growth associated with weight gain is a sensitive indicator of hypothyroidism, and yearly monitoring of height and weight should allow detection of overt hypothyroidism.12,13 Lastly, in the era of ultrasensitive TSH assays, the use of TRH stimulation tests to assess subtle primary hypothyroidism is questionable. Indeed, eight of nine patients with a reportedly abnormal TRH test according to Gibson et al had normal TSH three months later.

The principles of population screening14 imply that, if an abnormality is identified, there is an effective treatment. While this is clearly the case for congenital hypothyroidism or for overt acquired hypothyroidism, the effectiveness of thyroxine administration to children with Down’s syndrome and subtle abnormalities of thyroid function tests has not been rigorously validated. Given the non-specific nature of the symptoms of mild hypothyroidism (constipation, dry skin) and their overlap with manifestations of Down’s syndrome itself, only randomised, double blind controlled studies will settle the issue. The only published trial of this nature did not support thyroxine administration,15 but was woefully small in scope, with only seven patients studied. Thus, further study is clearly needed on the most sensitive clinical hallmarks of hypothyroidism in children with Down’s syndrome, on the minimal biochemical screening frequency, and on the need for thyroxine administration in those with subtle biochemical abnormalities. Meanwhile, clinicians caring for children with Down’s syndrome should remain alert to clinical signs of hypothyroidism but do not necessarily need to prescribe yearly blood tests; they should also refrain from treating on the basis of subtle biochemical abnormalities alone, which would unnecessarily increase medicalisation of these children who already face a number of genuine developmental and health challenges.

Commentary on the paper by Gibson et al (see page 574)



  • Competing interests: none declared

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles