Statistics from Altmetric.com
Editor,—We would like to comment on the recommendations on the management of congenital hypothyroidism by Razaet al. 1 Although thyroxine dosage recommendations have been available for the treatment of congenital hypothyroidism for many years2 it has not been possible to answer the crucial question as to what concentrations of circulating thyroid hormones and degree of thyroid stimulating hormone (TSH) suppression are required to provide the optimal environment for maximising neurobehavioural and intellectual development. The working group on congenital hypothyroidism of the European Society for Paediatric Endocrinology has emphasised that it has not been possible to measure a dose–response relation, and it is coordinating a prospective study to assess the adequacy of treatment and outcome using two different dosage schedules for the first 24 months after birth.
In the absence of randomised, prospective outcome studies, previous authors have recommended that any treatment strategy should aim to “achieve euthyroidism as soon as possible”3 4 and there is a persuasive argument that TSH suppression is the only and most relevant neurobiological marker of effective or optimal thyroid hormone concentrations and hypothalamic feedback.5 6
Surprisingly, Raza et al chose an unusual neonatal dosage regimen based on body surface area (100 μg/m2/day) with all its inaccuracies, and sought to evaluate whether the lack of TSH suppression was influenced by the underlying thyroid disease or basal TSH concentrations, studies of which have been reported previously.3 It would have been more helpful to have chosen a dosage schedule similar to those recommended previously (8–15 μg/kg/day) and to have accepted TSH suppression to < 10 mU/l to compare with other studies.
When screening for congenital hypothyroidism started in the Trent Region in 1980 several consultant paediatricians discussed a “best guess” dose of thyroxine; in Leicester we decided that as suspensions of thyroxine have questionable stability7 and the smallest tablet available is 25 μg we would initially use 25 or 50 μg on alternate days (that is 37.5 μg/day), which would be near to 10 μg/kg/day for most infants.
We have looked at the results of our last 29 cases of congenital hypothyroidism to compare TSH suppression with Razaet al’s and other reports (table1).
The starting dose in our infants weighing 2.5–4.7 kg ranged from 8–15 μg/kg/day (mean 10.5 μg/kg/day) compared with a calculated dose of 5–8 μg/kg/day based on Raza et al’s recommendation using body surface area. Using our regimen the concentrations of circulating total thyroxine at the time of TSH suppression ranged from 103 to 279 nmol/l (mean 174 nmol/l) and we saw no clinical evidence of hyperthyroidism. These data and a recent French study,6 where frequent dose titration was used, demonstrate that TSH suppression is related to thyroxine dosage and that there is considerable variation in thyroxine concentrations, presumably owing to variability in thyroxine absorption and metabolism. We therefore agree with Touati et al that early and regular individualisation of dosage is required to achieve TSH suppression.6 Thyroid hormone concentrations often seem to be high but fall within reported normal ranges for infants.9
We are concerned that following Raza et al’s recommendations will lead to an acceptance that significant non-suppression of TSH is unimportant, whereas it almost certainly represents undertreatment and so may not encourage the best achievable long term intellectual development.
Drs Brook and Hindmarsh comment:
We agree with Dr Swift and colleagues that a prospective randomised trial of different regimens of thyroxine in congenital hypothyroidism is needed. As the European Society for Paediatric Endocrinology trial he mentions will not look at long term sequelae it will not answer the question; we are trying to set up a UK multicentre trial that will address the problem.
One factor that Swift et al has not considered is whether feedback mechanisms of thyroxine on TSH are the same in infancy as in older patients. We never use suspensions of any hormone replacement and it seems intrinsically inappropriate to overtreat one day and undertreat the next.
We remain as concerned as ever about the effects of pursuing TSH suppression by increasing free thyroxine concentrations: we reiterate that the long term effects of clinically undetectable hyperthyroidism are much more damaging than the same degree of hypothyroidism. There are simply no data to support the contention of Swiftet al’s last paragraph.