Clinical Biochemistry 39 (2006) 511 – 518
Congenital hypothyroidism: From paracelsus to molecular diagnosis Anissa Djemli a , Guy Van Vliet b , Edgard E. Delvin b,c,⁎ a
Department of Laboratory Medicine, Centre Hospitalier de Sorel, Québec, Canada Department of Pædiatrics, CHU Sainte-Justine, Université de Montréal, Montréal, Québec, Canada Department of Clinical Biochemistry, CHU Sainte-Justine, Université de Montréal, Montréal, Québec, Canada b
c
Received 9 February 2006; received in revised form 28 March 2006; accepted 29 March 2006 Available online 20 April 2006
Abstract Endemic cretinism was noted in alpine Europe as early as the 13th century. However, it was only in 1848 that a commission, sponsored by the King of Sardinia, first formally demonstrated its link to goiter. An important landmark was the publication of a report in 1871 describing several cases of nongoitrous hypothyroidism that were clearly distinguished from the endemic form of the disease, for which the author suggested the designation of “sporadic cretinism.” Classification of the hypothyroid status was for a long time solely based on clinical observation. In the second half of the 20th century, the use of radionuclides (iodine radioisotope and technetium pertechnetate) allowed a more precise diagnosis and taxonomy into thyroid dysgenesis and dyshormonogenesis. This brief review summarizes the progress that has been achieved during the last 40 years in diagnosing the multiple variants of congenital hypothyroidism (CH). It becomes evident that while accurate diagnosis for CH is readily available, its exact etiology requires a precise molecular investigation as different genes are implicated in the differentiation, migration and growth of the thyroid gland. © 2006 The Canadian Society of Clinical Chemists. All rights reserved. Keywords: Thyroid function; Thyrotropin; Thyroxin; Congenital hypothyroidism
Historical background Although writers of the 1st century CE have referred to thyromegaly [1], Paracelsus, the 16th century physician and alchemist, was the first to mention hypothyroidism including its association to mental retardation [1]. While endemic cretinism was noted in alpine Europe as early as the 13th century, it was only in 1848 that a commission, sponsored by the King of Sardinia, first formally demonstrated its link to goiter [2]. In 1850, Curling was the first to report sporadic nongoitrous hypothyroidism in 2 female children aged 10 years and 6 months, respectively [3]. In his conclusion, he even suggested an association between a defective brain and the absence of the thyroid gland. Another important landmark was the publication in 1871 of a report describing several cases of nongoitrous hypothyroidism that were clearly distinguished from the ⁎ Corresponding author. Department of Clinical Biochemistry, CHU SainteJustine, 3175 Côte Ste-Catherine, Montréal, Québec, Canada H3T 1C5. Fax: +1 514 345 4803. E-mail address:
[email protected] (E.E. Delvin).
endemic form of the disease, for which the author suggested the designation of “sporadic cretinism” [4]. In 1897, Sir William Osler published the first collection of 60 cases of hypothyroidism in North America, of which only 7 had goiter [5], thereby showing that in America the sporadic form of the disease was the prevailing type. At the same time, it was determined that the insufficiency of the thyroid gland could be cured by injection of sheep thyroid extracts [6,7], and Baumann established that iodine was present in the thyroid gland [8]. However, it took almost 60 years before a synthetic product, ‘thyroxine,’ was produced economically in high yield allowing its use as an efficient treatment [9]. Classification of the hypothyroid status was for a long time solely based on clinical observation. In the second half of the 20th century, the use of radionuclides (iodine radioisotope and technetium pertechnetate) allowed a more precise diagnosis and taxonomy into thyroid dysgenesis and dyshormonogenesis [10–12]. It is essential to note that neonatal screening for CH abides with all the principles of population screening programs set forth in 1968 by the World Health Organization [13]. Without
0009-9120/$ - see front matter © 2006 The Canadian Society of Clinical Chemists. All rights reserved. doi:10.1016/j.clinbiochem.2006.03.015
512
A. Djemli et al. / Clinical Biochemistry 39 (2006) 511–518
doubt, this disease is serious and treatable; the screening procedure is reliable and inexpensive and enables the identification of patients at a presymptomatic stage. More importantly, hormone replacement at an early stage is available and has been shown to be beneficial. Indeed, before the screening era, the diagnosis for CH was often delayed and mental retardation, associated with this condition, was common [14]. In fact, most infants with CH appear normal at birth and the clinical symptoms appear only after a few weeks. Early cross-sectional and longitudinal studies have repeatedly shown that the cognitive deficit in CH is correlated with the lag-time in the diagnosis and initiation of the treatment [15–18]. However, the natural history of CH has improved to such an extent since the implementation of universal newborn mass screening programs [19–21] that it was recognized by the American Office of Technology Assessment as being one of the few programs that had a major impact on public health with a positive cost to benefit ratio [22]. The year 1972 marks the beginning of the screening programs for CH in North America. Seven years later, a retrospective study [23] including the five oldest North American programs (Québec, Ontario, Michigan, Oregon and New England Regional), documented an overall CH incidence of one in 3684 live births with 96% of the neonates having primary and 4% having secondary–tertiary hypothyroidism. Of the infants with primary hypothyroidism, (i) 63% were determined to have aplastic or hypoplastic glands, (ii) 14% had normal or enlarged glands, and (iii) 23% had ectopic thyroid tissue. Also, approximately 15 to 20% of primary CH cases may result from inherited thyroid dyshormonogenesis with defects ranging from thyroglobulin synthesis to iodine transport and organification [24,25]. Biochemical screening for congenital hypothyroidism The spark that gave the impetus to initiate a screening program was the development by Chopra in 1972 [26] of a RIA for the measurement of total thyroxine (TT4). A year later in 1973, Dussault and Laberge [27] showed the feasibility of TT4 measurement on the dried blood samples that were already being collected in most industrialized countries to detect phenylketonuria in newborns In 1974, the Province of Québec officially incorporated mass screening for congenital hypothyroidism together with phenylketonuria and tyrosinemia and thus became the first formal CH screening program based on TT4 measurement. Following the publication of Dussault et al. [28] in 1975, other programs based on this assay sprung throughout the industrialized world. However, the situation became more complex and controversial when other investigators evaluated the measurement of thyrotropin (TSH) on these same filter paper samples for the same purpose [29]. For a number of years, the proponents of the TT4 and TSH approaches argued about the advantages of each method. In order to understand the rationale of the debate, thyroid hormone secretion in the neonatal period is briefly outlined.
As described by Fisher et al. [30], TSH secretion peaks approximately 30 min after birth, rapidly declines thereafter for up to 12 h and stabilizes gradually at about 48 h. The TSH surge is met with more modest but more sustained elevations in T4 and T3 (Triiodothyronine) that peak at approximately 24 h of life followed by a slow decline over several weeks. Therefore, the TSH levels measured within the first 24 h of life are frequently elevated. In 1985, the screening protocol adopted in North America involved the collection of blood on filter paper from children on the 2nd through the 5th day of life and took advantage of the time-related TSH surge and stabilization mentioned above. Under this protocol, specimens with T4 below the 10th percentile were retested in duplicate, and TSH was determined. A confirmatory test on venous blood was requested if one of the following criteria was met: (1) low dried blood spot T4 (
