Redefining Congenital Hypothyroidism?

In high-income countries, the historical memory of the often devastating permanent neuropsychological disabilities due to a late clinical diagnosis of congenital hypothyroidism (CH) (which affected 1 child in 6500) is receding into the distant past. However, their occurrence in children of families immigrating from countries where there is no newborn screening (NBS) (1) reminds us that only 30% of the world’s newborns benefit from this public health triumph. The problem with NBS for CH is that case finding is based on a quantitative biomarker (typically thyrotropin (TSH), generally expressed in milliunits per liter [mU/L] of whole blood on a sample collected on filter paper at 2-5 days of life). The distribution of TSH at NBS is continuous, and the cutoffs used have decreased substantially over time and vary greatly between jurisdictions. Before referring a newborn with borderline results for a full diagnostic evaluation, most programs collect a second NBS sample, which is a reasonable way of decreasing the number of false positives and thus of infants who are unduly medicalized. However, collecting a second NBS sample has now become routine in some programs, regardless of the results on the first sample, to detect delayed elevations in TSH. These have been mostly described in newborns born prematurely and/or with a very low birth weight, or after a severe illness on discharge from intensive care. The limited evidence available suggests that the problem is transient and that most affected infants may not benefit from treatment (2). For confirming CH, for establishing its etiology through imaging, and for documenting its permanence and its impact on neurodevelopmental outcomes, practice also varies widely. In the United States, analysis of insurance claims data has shown that one-third of the 1 in 1500 children labeled as having CH at NBS no longer had a prescription for levothyroxine filled after age 3 years, regardless of insurance status. In privately vs Medicare-insured children, only 71% and 53%, respectively, had a serum TSH after treatment discontinuation. Imaging at diagnosis or later, mostly by ultrasound, was performed in an even lower proportion (24% and 13%) (3). Multiple studies have shown the predictable result that lower TSH cutoffs and collection of multiple samples result in increased case finding. However, it is not known how many of the children diagnosed with CH as a result of lowered cutoffs or repeat sampling clearly benefit from treatment to the same extent as those infants meeting conventional cutoffs. A recent debate among experts concluded that it is important to define optimal TSH cutoffs but disagreed on the current state of evidence (4). The new study by Caiulo et al (5) reports that CH was diagnosed in 1 in 911 children born in the Italian region of Lombardy in 2014 to 2017, a birth prevalence 7-fold higher than in the pre-NBS era. The study does not address whether the additional infants diagnosed with CH would have been harmed if they had remained undiagnosed and untreated. The marked difference in estimated prevalence of CH before and after NBS contrasts with the similar prevalence of 21-hydroxylase deficiency during the 2 eras. The explanation lies in part in the fact that 21-hydroxylase deficiency is a relatively homogeneous mendelian condition, in which the clinical and biochemical diagnosis can ultimately be confirmed by detecting biallelic mutations inactivating CYP21A2, whereas CH is a highly heterogeneous condition defined by an arbitrary number on the continuous distribution of a biomarker. The most common etiology of overt permanent CH is a defect in thyroid migration during embryogenesis resulting in a sublingual ectopy, which can be reliably evidenced only by scintigraphy with sodium pertechnetate. CH due to sublingual thyroid ectopy is essentially a nonmendelian disease, has a marked female (3 to 1) and ethnic (White) predominance, and occurs sporadically in 99% of cases; its prevalence is stable and does not appear to be influenced by environmental factors (6). Less common are athyreosis (which is itself heterogeneous) and dyshormonogenesis with goiter (a group of autosomal recessive conditions). The additional patients with CH identified by lower TSH cutoffs and multiple samples generally have milder, often transient functional disorders, the consequences of which for neuropsychological development are unknown. Most of these children should be diagnosed as having isolated hyperthyrotropinemia and not CH. There is no consensus that mild isolated neonatal hyperthyrotropinemia is deleterious for neurocognition, with some large population-based studies suggesting an association and others not (4, 7). Confounders include the possible inclusion of children with conditions leading to developmental disabilities unrelated to neonatal TSH levels (8) and the low iodine intake of the population considered (4). A randomized controlled trial (RCT) comparing levothyroxine with placebo given during the first 2 years of life (when the brain is crucially dependent on a state of euthyroidism) to otherwise normal infants with mild TSH elevations at NBS would be the most rigorous way of establishing causation. For example, such an RCT of thyroxine (T4) treatment in infants with trisomy 21, who on average have slightly elevated TSH and decreased T4 at NBS, found no significant advantage of T4 over placebo on intellectual outcome at age 10 years (8). That finding allowed clinicians to decide T4 was no longer needed for all these infants. Although yielding less rigorous evidence than that obtained from RCTs, observational studies comparing outcomes between jurisdictions with different protocols for NBS and for diagnostic confirmation could help clarify the issue. Countries with regions that have comparable health care and educational services but different NBS protocols, such as Italy, the United Kingdom, or Australia, could establish such cohort studies, either retrospectively or prospectively, and help determine which cutoff makes a clinically important difference. This would be particularly important to address the most important challenge in this area, which is the establishment of rational NBS protocols focusing on overt CH for the 70% of the world’s newborns who currently undergo no NBS at all. In low- and middle-income countries, the development of a point-of-care test (POCT) for the TSH assay should be followed by pilot screening projects once an accurate, affordable assay is in place to demonstrate feasibility of implementation of POCT testing. Reporting of results both to health providers and the public health system should ensure that newborns are treated as early as possible, that their outcome is documented, and that the program is monitored. congenital hypothyroidism newborn screening point-of-care test randomized controlled trial thyrotropin. The authors thank Dr Scott D. Grosse (senior health economist, Centers for Disease Control and Prevention, Atlanta, Georgia, USA) for critical reading of this commentary. Disclosures: The authors have nothing to disclose. Some or all data generated or analyzed during this study are included in this published article or in the data repositories listed in “References.” Data sharing is not applicable to this article because no data sets were generated or analyzed during the present study. Restrictions apply to the availability of some or all data generated or analyzed during this study to preserve patient confidentiality or because they were used under license. The corresponding author will on request detail the restrictions and any conditions under which access to some data may be provided.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.