Early Determinants of Thyroid Function Outcomes in Children with Congenital Hypothyroidism and a Normally Located Thyroid Gland: A Regional Cohort Study.

BACKGROUND: An increase in the incidence of congenital hypothyroidism (CH) with a normally located gland has been reported worldwide. Affected individuals display transient or permanent CH during follow-up in childhood. This study aimed to determine the prevalence of transient CH and to investigate the possibility of distinguishing between transient and permanent CH in early infancy.

METHODS: This observational cohort study included all patients identified by systematic neonatal screening for CH in the northern Parisian region between 2002 and 2012 and treated for CH with a normally sited gland. A standardized data collection form was completed prospectively at diagnosis. Patients were classified during follow-up as having transient or permanent CH.

RESULTS: Of the 92 patients initially treated for CH with a normally located gland during the neonatal period, 49 (54%) had a transient form of CH after the cessation of levothyroxine (LT4) treatment at 1.5 (0.6-3.2) years of age. Multivariate analysis revealed that transient CH was associated with a lower likelihood of having a first-degree family history of CH (p = 0.03) and a lower LT4 dose at six months of age (p = 0.03) than permanent CH. Sex, ethnicity, neonatal problems (e.g., prematurity, being small for gestational age, and/or neonatal distress), iodine status, coexisting malformations, initial CH severity, and thyroid morphology at diagnosis had no effect. Receiver operating characteristics curve analysis showed that a cutoff of 3.2 μg/kg/day for LT4 dose requirement at six months of age had a sensitivity of 71% and a specificity of 79% for predicting transient CH, with values below this threshold considered predictive of transient CH.

CONCLUSION: In patients with CH and a normally located gland, these findings highlight the need to evaluate LT4 dose requirements early, at six months of age, particularly in patients with no family history of CH, for early identification of the approximately 50% of patients for whom treatment should be stopped.