History of the Thyroid

According to Chinese legend, in 2697 BCE, the “Yellow Emperor” Hung Ti compiled “The Yellow Emperor’s Classic of International Medicine,” a pharmacopeia consisting of 162 chapters in 18 volumes [1], including the use of seaweed, and in 1596, Li Shizen’s pharmacopeia included iodine as a treatment for goiter [2]. Iodine-deficient, endemic goiter was well-known in the ancient world, is seen in the ancient art of many countries, and a connection between oceanic plants and goiter has been known since antiquity in both Western and Chinese cultures. In 1811, Bernard Courtois, a French chemist involved in the manufacture of saltpeter (potassium nitrate, which is essential for gunpowder), noticed violet fumes while washing seaweed ash (sodium carbonate) with sulfuric acid. When encountering cold surfaces, the fumes crystalized as purple crystals: the discovery of iodine [3]. Courtois’ work was confirmed, imprecisely described, but extended by Joseph Gay-Lussac, who recognized he was dealing with a newly discovered element; he named it iodine, from the Greek Ιώδης “violet-colored” [4].

The term “thyroid” apparently derives from the classic Greek word “thyra” (θύρα), meaning “door” or from the word thyreos (θυρεός), referring to a stone placed against a door to keep it closed, or a large oblong shield shaped like a door [5]. The thyroid cartilage was described by Galen and others in antiquity [5], and the gland itself was described by Leonardo Da Vinci, Vesalius, Realdus Columbus, Michelangelo, and Bartolomeo Eustachio (who also described the adrenals) during the Renaissance [6]. The English name “thyroid,” based on the Greek (above), was apparently coined by Thomas Wharton (also known for Wharton’s jelly) in his “Adenographia” (1656) [7].

Jean-Francois Coindet (1774–1834) reported the efficacy of treating goiter with various iodine preparations (rather than preparations of seaweed or sponges), settling on a tincture (alcoholic solution) whose successful use he detailed in 1821 [8]. It was long known that goiter was more common among residents of the mountains than those living by the sea. The Roman satirical poet Juvenal (2nd century AD) said “Quis tumidum guttur miratur in Alpibus?” [“Who is surprised to see a swollen goitre in the Alps?”] (Juvenal, xiii. 162) [9]. The presence of a goiter, physical deformities, and intellectual disability in individuals in the Alpine regions of Europe provided the first connection of goiter to the function of the thyroid. The term cretinism was first used in the medical literature in 1754 to describe those afflicted, apparently derived from the Latin “Christianus,” as afflicted individuals were thought unable to commit a sin [10] (Fig. 1).

The clinical syndrome we now recognize as hypothyroidism was characterized as “myxoedema” in 1878 by William Ord, a physician at St. Thomas’ Hospital [11] (Fig. 2). The concept of an organ (gland) secreting a chemical substance (hormone) is attributed to the French physician Brown-Séquard, who in 1889 injected himself with testicular extracts from several animals and reported improvement in strength and stamina [12]. With recognition of this principle, in 1891, George Murray, a British physician, reported that injection of thyroid extract from sheep led to improvement in symptoms in a woman with myxedema [13]. Orally administered fresh sheep thyroid gland was first reported by Mackenzie in 1892 [14] Soon thereafter, another case with “before and after” pictures was reported by Shapland (Fig. 3) [15]. Oral thyroid was cheaper, easier to use, and lacked the attendant risk of “tonic spasms” (Mackenzie’s term) apparently from co-injecting PTH, which would have been contained in Murray’s thyroid extracts.

Robert Graves is known for his classic 1835 report of “palpitations, goiter, and exophthalmos” in three women [16], but Caleb Parry observed the same clinical features in 1786 (published posthumously by his son in 1825) [17] – yet the name Graves’ disease, coined by Armand Trousseau in 1861, persists [18]. In 1840, Adolph von Basedow completed the picture of toxic goiter [19]. He noted exophthalmos, and as he practiced in Merseburg, Germany, the triad of goiter, exophthalmos, and palpitation was frequently called Merseburg’s triad. In 1867, Adolph von Graefe described lid lag in thyrotoxicosis and coined it Basedow’s disease [20]. Up until the 1800s, the pathophysiology of Graves’ disease was unclear; cardiac and nervous system disorders were thought to be the cause [21].

Not until 1884, with the first report of thyroidectomy for a patient with Graves’ disease by Ludwig Rehn, was the thyroid suspected to be the etiology [22]. After improvement in symptoms and signs in his patients, Rehn proposed the disease was due to hyperactivity of the thyroid gland. In 1887, neurologist Paul Julius Möbius referred to Rehn’s work and suggested the cause was disordered function of the thyroid gland poisoning the body [23]. With the description of X-rays, external radiation was tried for all manner of disease, posing risks to both patient and physician: after 60 years as a practitioner, Dr. Emil H. Grubbe, who built his own X-ray machine in 1896, underwent 90 surgeries for “cancerous burns” and died of cancer [24].

“Cretinism,” a term that has mostly been replaced in the medical literature by “congenital hypothyroidism,” refers to severe thyroid hormone deficiency early in life. Until the mid-19th century, there was little to link the thyroid to cretinism. Severe hypothyroidism may be congenital or, rarely, acquired during infancy or early childhood. In 1850, Thomas Curling described two children, aged 6 months and 10 years, with the absence of goiter (noted at autopsy) and defective cerebral development [25]. More than 20 years later, C. Hilton Fagge noted similar patients and postulated “…the healthy thyroid body is capable of exerting a counteracting influence [on cretinism]” [26]. In 1883, Felix Semon tried to unify myxedema, cretinism, and cachexia strumipriva (signs and symptoms of hypothyroidism, with or without myxedema resulting from the loss of thyroid tissue) by proposing that all were manifestations of the absence or degeneration of the thyroid [27]. Horsley noted the amelioration of severe, induced hypothyroidism in animals by transplanting parts of the thyroid gland into the peritoneum of the hypothyroid animals [28]. In 1895, Bramswell showed the remarkable transformation of a child with severe hypothyroidism with marked linear growth, greater mental awareness, as well as the loss of subcutaneous edema [29]. Thus, by the end of the 19th century, endemic severe hypothyroidism was established as a condition due to the lack of thyroid hormone function. The commonality of myxedema in adults and severe hypothyroidism in children presaged the more widespread use of thyroid replacement therapy. The link between iodine deficiency and cretinism was confirmed in 1966 following a trial in Papua New Guinea which led to the discovery that administration of iodized oil to the population reduced the incidence of goiter and cretinism [30].

Theodore Emil Kocher, a Swiss surgeon, is credited with discovery of the finding that patients with goiter who underwent complete thyroidectomy developed what he termed cachexia strumipriva, a “bad condition due to removal of a struma (goiter).” Kocher (among others) proposed that surgeons conserve part of the gland to prevent this outcome. He was awarded the Nobel Prize in Physiology and Medicine in 1909 for his work on the physiology, pathology, and surgery of the thyroid gland [31].

Shortly thereafter, Edward Kendall, PhD, working at the Mayo Clinic, a center performing a lot of thyroid surgery, was isolating thyroid hormone. On Christmas day, 1914, Kendall finally succeeded in isolating thyroid hormone crystals, which he named “thyroxin” (“e” later added) [32-35] (Fig. 4). However, it was another decade until Charles Harington and George Barger determined its chemical structure (1926) and then synthesized thyroxine (T4) in 1927 [36, 37]. With this advance, the significance of iodine as a component of “tetraidothyronine” became clear. Synthesis of T4 allowed studies of its physiology; for example, in 1931, DM Lyon demonstrated the impact of T4 on body metabolism and weight homeostasis [38]. Gross and Pitt-Rivers reported biologic activity of tri-iodothyronine (T3) in 1952–1953 [39].

Studies in the 1930s–1940s began to work out regulation of the thyroid gland by a substance produced in the pituitary. In 1931, Loeb and Bassett extracted and purified thyrotrophic substance (TSH) from bovine pituitary glands [40]. Then in 1949, RG Hoskins advanced the concept of the pituitary-thyroid axis function in a paper entitled “The thyroid-pituitary apparatus as a servo (feed-back) mechanism” [41]. Thyrotropin-releasing hormone (TRH) was isolated independently by Andrew Schally and Roger Guillemin in 1970 [42, 43], defining the hypothalamic-pituitary-thyroid axis. Schally and Guillemin shared the Nobel Prize in Medicine for their discovery in 1979.

Starting in the 1950s, an understanding of the various steps involved in the synthesis of thyroid hormone began to emerge. In 1950, John Stanbury reported two brothers described as “goitrous cretins” (Fig. 5) [44]. Radioactive iodine uptake by the thyroid was increased, though protein-bound iodine content was decreased, pointing toward a defect beyond trapping, such as a defect in oxidation/organification of iodide. Thyroid peroxidase was cloned and sequenced in 1987 [45]. The role of DUOX2 as a co-factor providing hydrogen peroxide for thyroid peroxidase activity was reported in 1999 [46]. Cloning and characterization of the Na⁺/I^- symporter was reported in 1996 [47]. The etiology of Pendred syndrome, named after British physician Virginia Pendred, who reported deafness and goiter in an Irish family in 1896 [48], was discovered a century later (1997) to be caused by a mutation in SLC26A4, a gene coding for a chloride-iodine protein that transports iodide from the follicular cell into the colloid [49]. Characterization of iodotyrosine deiodinase (IYD or DEHAL1) was reported by Rosenberg et al. [50]. Taken together, these studies confirmed the concept that there is a gene encoding a protein controlling each step in the biosynthesis of thyroid hormone, information used to identify specific defects in subjects with dyshormonogenesis.

The next step was to determine the mechanism of thyroid hormone action. The first clue came from Jack Oppenheimer’s group in 1972, who reported T3 binding to a nuclear receptor [51]. Two groups, Sap et al. [52] and Weinberger et al. [53], reported that the c-erb-A oncogene encodes a thyroid hormone receptor (published back-to-back in Nature in 1986). Thyroid hormone action was further clarified with the report in 1987 that there are two classes of T3 nuclear receptors: TRα and TRβ [54].

The remaining piece of the puzzle was to determine how thyroid hormone enters the cell. Friesema et al. [55] reported discovery of monocarboxylase transporter 8 (MCT8), the first of several thyroid hormone transporters. This discovery came full circle when MCT8 mutations were shown to be the cause of Allen-Herndon-Dudley syndrome, a disorder first reported in 1944 and characterized by severe intellectual disability [56]. This X-linked disorder is seen in boys only; the tipoff on thyroid function testing was an elevated serum T3 level, with low–normal T4 and TSH concentrations.

In the early 19th century, iodine was added to food and water after it was suggested as prophylaxis for goiter. However, this practice was abandoned until Dr. David Marine’s landmark experiment in 1916–1920 where iodide administration to 5,000 schoolgirls in Akron, Ohio, prevented development of endemic goiter [57]. Several subsequent studies produced similar results, and in 1924, iodized salt became commercially available in Michigan, followed by the remainder of the USA. Iodine deficiency remained a significant public health issue throughout the 20th century. UNICEF has called iodine deficiency the greatest preventable cause of cognitive deficits. Through public health efforts, the number of iodine-deficient regions around the world has been declining. In 2021, nearly 90% of the global population were consuming iodized salt, a significant rise from 25% in 1990.

In 1948, Drs. Jan Wolff and Israel Lyon Chaikoff discovered that excess intrathyroidal iodine levels inhibit thyroid peroxidase, leading to inhibition of organification and subsequently causing a transient decrease in thyroid hormone production. This phenomenon, now known as the acute Wolff-Chaikoff effect, protects against overproduction of thyroid hormone due to iodine excess [58-60]. Wolff and Chaikoff went on to describe that in adults, an escape from downregulation usually occurs after a few days of exposure to excess iodine, leading to resumption of normal thyroid hormone production [61, 62]. However, the immature neonatal thyroid gland is unable to escape from the acute Wolff-Chaikoff effect, making infants and children more susceptible to iodine-induced hypothyroidism [63, 64]. Further work has led to the important discovery that excess iodine exposure in utero and in the postnatal period, in the form of iodine-containing antiseptics, contrast agents, medications, and supplements, can lead to transient or permanent neonatal hypothyroidism [63, 65, 66].

In 1963, Robert Guthrie developed a test to detect phenylketonuria on blood spotted on a filter paper card [67]. The “Guthrie test” led to the novel concept of screening newborn populations, with the premise that early detection and treatment of selected disorders would prevent death or intellectual disability. In 1973, Jean Dussault developed a radioimmunoassay to measure T4 in dried blood filter paper specimens, then applied to screening newborns for congenital hypothyroidism [68]. In 1975, Dussault reported the results of screening 47,000 newborns in Quebec, with detection of 7 cases of congenital hypothyroidism (∼1:7,000) [69].

In 1978, Delbert A. Fisher and collaborators reported detection of 277 patients with congenital hypothyroidism in 1 million infants screened in North America, an incidence of 1:3,684 [70]. The success of newborn screening (NBS) was confirmed 1985 when the New England Congenital Hypothyroidism Collaborative reported similar cognitive outcome at age 6 years in children detected by NBS versus sibling controls [71]. A year later, in 1986, Cheryl Hanna and colleagues were the first to report children with congenital central hypothyroidism detected by the Northwest Regional NBS Program, made possible by collection of two routine specimens and follow-up of cases with persistently low T4 – nonelevated TSH levels. Most infants had congenital hypopituitarism, with an estimated incidence of 1:29,000 [72]. The most common etiology for congenital hypothyroidism is thyroid dysgenesis, resulting in permanent hypothyroidism. In 1995, Pacaud et al. [73] reported 3 siblings detected by NBS with transient congenital hypothyroidism caused by transplacental passage of maternal TSH receptor blocking antibody.

Four decades’ experience with NBS programs led to discoveries that challenged previous knowledge about congenital hypothyroidism. In 2000, collaborators from the New England NBS Program were the first to report “delayed TSH elevation” in preterm infants [74]. This discovery led many NBS programs to change their protocol and routinely collect 2nd or even 3rd specimens in preterm infants. In 2000, another study by Fisher and colleagues reported that up to 30 percent of infants with congenital hypothyroidism treated with thyroid hormone had a pattern similar to that of patients with thyroid hormone resistance; this “resistance” appeared to decrease over time [75]. While thyroid dysgenesis resulted in permanent hypothyroidism, other milder forms of thyroid dysfunction were found to be transient [76]. As NBS programs expanded, a reported increase in incidence was noted compared to the prescreening era [77].

Hashimoto’s thyroiditis, the most common autoimmune disease in humans and the most frequent cause of hypothyroidism [78, 79], was first reported by Hakaru Hashimoto, who in 1912 described a thyroid disorder discovered in 4 women who had undergone thyroidectomy [80]. The common pathologic findings among the thyroid glands of these patients were diffuse lymphocytic infiltration of follicles with prominent germinal centers, destruction of follicular epithelial cells, and fibrosis of the gland. This constellation of pathologic findings was coined by Hashimoto as struma lymphomatosa. At the time, many etiologies of this thyroid disorder were postulated, including infection, premalignancy, or goiter from stress [80].

In the early 1930s, Graham and McCullagh [81] used the term “Hashimoto” when distinguishing between struma lymphomatosa and Riedel’s thyroiditis; the term Hashimoto disease was later coined by Cecil Joll [82]. Riedel’s thyroiditis, initially described in 1896 [83], is characterized by inflammatory fibrosis which may extend into adjacent tissues and is now considered exceptionally rare, with an estimated incidence of 0.06% and occurring primarily in adults [84]. This was separate from the third variety of thyroiditis, subacute thyroiditis, described in 1936 by De Quervain and Giordandengo [85].

The association of Hashimoto’s thyroiditis with autoimmunity was first made in 1956, when Noel Rose and Ernest Witebsky reported lymphocytic infiltration in the thyroids of rabbits injected with rabbit thyroid extracts [86, 87]. This concept was developed further by Robert Blizzard et al. [88] who in 1960 demonstrated the association of autoimmunity with thyroid gland destruction [89]. However, it was not until the 1980s that thyroperoxidase was identified as the dominant antigen targeted by the immune system, leading to the finding that circulating antithyroid peroxidase antibody is the most sensitive and specific test for autoimmune thyroid disease [90].

Genetic links to autoimmunity were first reported in the 1970s; however, initial reports of hypothyroidism linked to gene mutations dated back to reports by J.B. Stanbury in 1950 [44]. The pathogenesis of Hashimoto’s thyroiditis is now understood to be a complex interplay of genetic susceptibility and various endogenous and environmental factors [91, 92]. Several immune-regulatory genes, including HLA, CTLA-4, and PTPN22, have been identified as playing a role in the development of autoimmune thyroiditis [45, 92]. While it is most common for autoimmune thyroiditis to occur in isolation, associations between multiple autoimmune conditions have been made, now known as autoimmune polyglandular syndromes. The constellation of autoimmune adrenal insufficiency and chronic lymphocytic thyroiditis, now recognized as Schmidt syndrome, was first reported in 1926 [93].

While hyperthyroidism was not clearly delineated until 1907 by American surgeon Charles H. Mayo, who introduced the term “hyperthyroidism,” exophthalmic goiters were commonly studied [94]. In 1905, American surgeon Robert Abbe treated exophthalmic goiter with implantation of radium into the enlarged thyroid gland [95]. In 1907, David Marine proved iodine is necessary for thyroid function and recommended iodine in the treatment of Graves’ disease in 1911, but this was not widely accepted [96]. In 1916, Bennet Allen and Philip Smith proved a TSH existed and was present in the anterior pituitary [97, 98]. X-rays were widely used to treat exophthalmic goiter, hyperplastic toxic goiter, and toxic adenoma until the 1930s [99]. In 1922, Henry Plummer, Walter Boothby, and Lewis Wilson at Mayo Clinic advanced the view that the exophthalmic goiter was caused by intensive stimulation of unknown source acting on the entire gland [100]. Plummer extensively studied the effect of iodine in exophthalmic goiter. This Mayo group used iodine therapy in preoperative management of exophthalmic goiter [101]; this decreased mortality and became widely accepted as treatment of hyperthyroidism in the 1930s. In 1933, Frederic Joliot and Irene Curie created novel “artificial” radionuclides, leading to development of useful radioactive elements by others, including radioactive iodine [102].

From 1937, several physicians and physicists used radioactive iodine to study thyroid physiology and treat hyperthyroidism. In 1941, Saul Hertz and Arthur Roberts [103] were the first to use radioiodine with the intention to treat hyperthyroidism. In 1943, Edwin B. Astwood [104] published his landmark paper on treatment of hyperthyroidism with the antithyroid drugs thiourea and thiouracil. He continued to study goitrogenic substances and introduced the more potent 6-n-propylthiouracil in 1945 [105] which was approved by the FDA in 1947. Methimazole was discovered to be an even more potent and less toxic thiourea analog in 1949 [106]. In 1951, carbimazole was introduced with the hope that the prodrug would have less toxicity than methimazole [107]. Also in the 1950s, with improvements in ion exchange chromatography, purer preparations of TSH are isolated. While studying TSH, another substance was discovered that stimulated the thyroid gland but differently from TSH. Duncan Adams and H.D. Purves first reported in 1956 the detection of this long-acting thyroid stimulator in patients with thyrotoxicosis [108]. In 1964, Kriss et al. [109] identified long-acting thyroid stimulator as an immunoglobulin that binds to and activates the thyrotropin receptor on thyroid cells as the etiology of Graves’ hyperthyroidism.

Pediatric thyroid cancer has features that distinguish it from the adult form; however, the history of thyroid neoplasia is indistinguishable. The first documentation of thyroidectomy was in 952 AD by Moorish physician Albucasis for large goiters [110, 111]. Thereafter, several surgeons attempted surgical treatments of goiters, but prior to the mid-19th century, it was difficult, bloody, rarely successful, and largely avoided even by skillful surgeons [112, 113]. In 1791, French surgeon Pierre-Joseph Desault performed the first successful partial thyroidectomy; however, surgery continued to have a high mortality, and the French Academy of Medicine even banned all thyroid operations in 1850 [114, 115]. Wilhelm Röntgen discovered X-rays in 1896 [116] followed by Marie and Pierre Curie discovering radium in 1898 [117]. The Curies suggested using X-rays to treat tumors in 1899 [118]. Roentgen rays were used in cases of inoperable thyroid carcinomas and postoperatively in the early 1910s [119-121]. Radioactive iodine use in endocrine research started in the late 1930s, with Samuel Siedlin being the first to use ¹³¹I to treat thyroid cancer in 1946 [122]. In 1948, Polish physicians Tempka et al. [123] published a preliminary report using thyroid smears of fine needle punctures. This is regarded as the first publication of using fine-needle biopsy or aspiration (FNA) as a diagnostic method [124].

In 1950, Benedict Duffy and Patrick Fitzgerald reported the connection of irradiation of an enlarged thymus and thyroid cancer when they found 10 out of 28 patients who had been subjected to chest irradiation between 4 and 16 months of life had thyroid cancer [125]. In 1955, Dwight Clark reported the increasing incidence of thyroid carcinoma since 1900 in children and adolescents, correlating with greater use of radiation of benign conditions of head, neck, and upper thorax [126]. Many large studies since have confirmed the greatest risk factor is radiation. This curbed the use of irradiation for nonmalignant conditions in pediatrics after its peak in mid 1950s.

In 1959, with progressive improvements in methods of histology, Hazard et al. [127] identified medullary carcinoma as a distinct entity. This opened the door for identification of multiple endocrine neoplasia syndromes. In 1961, a medical resident, John Sipple, published on the association of pheochromocytoma and medullary thyroid carcinoma and later earned the eponym of Sipple syndrome for MEN2A in 1968 [128].

Ultrasonography became an integral part of the clinical evaluation of the thyroid. Application of ultrasound for diagnostic thyroid imaging began in the late 1960s [129]. Although thyroid biopsies were routinely obtained in thyroiditis, it was not until 1971 that thyroid biopsies were used routinely in thyroid nodules [130]. In 1977, Paul Walfish recommended combining FNA with ultrasound to improve the accuracy of biopsy specimens, which has now become standard of care [131]. In 1985, the rearranged during transfection (RET) proto-oncogene was identified [132]. The analysis of RET mutation became clinically relevant with its role in diagnosis, prognosis, and therapeutic decision-making for childhood MTC and PTCs. In 1990s, the RET gene was found to be associated with MEN2 [133].

The expansion of knowledge regarding genetic alterations furthered the use of molecular markers on biopsies and targeted cellular therapy. Initially, while surgical resection was the only treatment option, in the early 2000s, tyrosine kinase inhibitors were initially studied as treatment for thyroid carcinomas [134]. In 2009, Steven Waguespack et al. [135] reported the first successful use of multi-kinase inhibitor to treat pediatric PTC in a 14-year-old girl.

While the use of cytopathology from FNA was a major advance in diagnosis, there was inconsistency in interpretation of results. The Bethesda System for Reporting Thyroid Cytopathology was initially proposed to address terminology and clarity of communication of FNA results in 2007. Published in 2010 and revised in 2018, it has since been adopted worldwide [136]. In 2014, The Cancer Genome Atlas project, originally launched by NIH in 2005 to comprehend the genomic alterations underlying all major cancers, published a paper on the genomic landscape of PTCs [137]. The use of molecular profile testing has increased in pediatrics, and oncogene panels are shown to have clinical utility to predict increased risk of malignancy in patients less than 19 years old [138]. In 2015, the American Thyroid Association published its inaugural guidelines on pediatric thyroid cancer. This acknowledged the difference in pathophysiology, clinical presentations, and long-term outcomes in children and need for unique guidelines [139]. In 2018, the American College of Radiology published TI-RADS, a new risk stratification system for thyroid nodules on basis of ultrasonography [140]. TI-RADS has not been validated in children, but studies on pediatric adaptations have been published.

Not applicable (no research in article).

The authors declare no conflicts of interest. Kara J. Connelly and Stephen H. LaFranchi: UpToDate author with financial relationship.

No funding was provided for this review.

All three authors Kara J. Connelly, Julie J. Park, and Stephen H. LaFranchi, equally made substantial contributions to the design, acquisition of pertinent references, development of the initial draft, and critical revision of the article.

There were no data generated for this report.