The term endemic goiter is a descriptive diagnosis and reserved for a disorder characterized by enlargement of the thyroid gland in a significantly large fraction of a population group, and is generally considered to be due to insufficient iodine in the daily diet. Since nontoxic goiter also exists when there is abundant iodine in the diet, the distinction between endemic and non endemic goiter is necessarily arbitrary. Endemic goiter may be said to exist in a population when more than 5% of the preadolescent (at 6-12) school-age children have enlarged thyroid glands (2), as assessed by the clinical criterion of the thyroid lobes being each larger than the distal phalanx of the subject's thumb. Detailed criteria are discussed further below.
Most of the significantly mountainous districts in the world have been or still are endemic goiter regions. The disease may be seen throughout the Andes, in the whole sweep of the Himalyas, in the Alps where iodide prophylaxis has not yet reached the entire population, in Greece and the Middle Eastern countries, in many foci in the People's Republic of China, and in the highlands of New Guinea. There are or were also important endemias in non mountainous regions, as for example, the belt extending from the Cameroon grasslands across northern Zaire and the Central African Republic to the borders of Uganda and Rwanda, Holland, Central Europe and the interior of Brazil. An endemic existed in the Great Lakes region in North America two generations ago. Measurements have indicated that these regions have in common a low concentration of environmental iodine. The iodine content of the drinking water is low, as is the quantity of iodide excreted each day by residents of these districts.
Goiter maps of various countries have been repeatedly drawn, requiring modification as successful prophylactic measures have been introduced. Although goiter was an important problem in many regions of the United States in the past (56), more recent US surveys have shown it in no more than 4-11% of schoolchildren, with evidence of continued adequate iodine nutrition in the country since 1988 (57, 57 bis). This finding is a testimony to the effectiveness of iodine prophylaxis in preventing endemic goiter.
The great arc of the Himalayas from West Pakistan across India and Nepal, into Northern Thailand and Vietnam and into Indonesia, is one of the most highly endemic regions of the world. The disease used to be a major problem throughout the Andes. It has been reported from New Zealand and from different places in Europe. The world or regional distribution of goiter was exhaustively reviewed by Kelly and Snedden in 1960 (58) and subsequently by others in 1980 (5), 1993 (59), 1996 (60), 1998 (61), 1999 (62) and 2003 (62 bis).
These surveys reveal striking differences in the rate of goiter in different endemic regions and even in adjacent districts. The geographic unevenness of an endemic undoubtedly has much to do with the habits of the population and their economic resources for the importation of foods. In attempting to account for the variability in the expression of endemic goiter from one locality to the next, the availability of iodine should be investigated before searching for some other subtle dietary or genetic factors. The key to the problem almost always lies in the availability of iodine. One must also consider the possibility that an observed goiter rate may not reflect current conditions, but rather may be a legacy of pre-existing iodine deficiency that has not yet been entirely resolved by an improvement in the supply of iodine. The assessment of goiter in a population is discussed in the section on assessment of the IDD status of the population.
The arguments supporting iodine deficiency as the cause of endemic goiter are four: (1) the close association between a low iodine content in food and water and the appearance of the disease in the population; (2) the sharp reduction in incidence when iodine is added to the diet; and (3) the demonstration that the metabolism of iodine by patients with endemic goiter fits the pattern that would be expected from iodine deficiency and is reversed by iodine repletion. 4) Finally, iodine deficiency causes changes in the thyroid glands of animals that are similar to those seen in humans (60, 63).
Almost invariably, careful assessment of the iodine intake of a goitrous population reveals levels considerably below normal (100-200 μg/day). Severe iodine deficiency was still encountered until recently. From two endemic goiter areas of Zimbabwe, mean urinary iodine excretion from adults was reported to vary between 10 and 21 ug/l (64). In Senegal a mean iodine excretion of 17 ug/g creatinine (roughly equivalent to 24h) was also reported in 1992 (65). In the Eastern part of Germany, a 24h iodine excretion of 16 ug has been reported in 1989 (66).
Although the relation of iodine deficiency to endemic goiter is well established, other factors may be involved. A whole variety of naturally occurring agents have been identified that might be goitrogenic in man (67, 68). Most of these have only been tested in animals and/or have been shown to possess antithyroid effects in vitro. These compounds belong to the following chemical groups : Sulfurated organics (like thiocyanate, isothiocyanate, goitrin and disulphides), flavonoids (polyphenols), polyhydroxyphenols and phenolderivatives, pyridines, phtalate esters and metabolites, polychlorinated (PCB) and polybrominated (PBB) biphenyls, other organochlorines (like DDT), polycyclic aromatic hydrocarbons (PAH), inorganic iodine (in excess), and lithium. Gaitan (67) divides goitrogens into agents acting directly on the thyroid gland and those causing goiter by indirect action. The former group is subdivided into those inhibiting transport of iodide into the thyroid (like thiocyanate and isothiocyanate), those acting on the intrathyroidal oxidation and organic binding process of iodide and/or the coupling reaction (like phenolic compounds) some phalate derivatives (disulfides and goitrin) and those interfering with proteolysis, dehalogenation and hormone release (like iodide and lithium).
Indirect goitrogens increase the rate of thyroid hormone metabolism (like 2,4-dinitrophenol, PCB's and PBB's). Soyabean, an important protein source in many third world countries interrupts the enterohepatic cycle of thyroid hormone (69) and may cause goiter when iodine intake is limited. It should be recognized that goitrogens are usually active only if iodine supply is limited and/or goitrogen intake is of long duration.
Some of these goitrogens are synthetic and are used medicinally. Others occur in certain widely used food plants (70). The initial recognition of dietary goitrogenesis is attributed to Chesney et al.(71) who in 1928 found that rabbits fed largely on cabbage developed goiters. In 1936, Barker (72) found that thiocyanate used in large doses to treat hypertension resulted in goiter. In 1936, Hercus and Purves (73) reported their studies on the production of goiter in rats by feeding the seeds of several species of Brassica (rape, choumoellier, turnip, etc.). Both Mackenzie and MacKenzie (74) and Astwood (75) found in the 1940’s that certain drugs such as thiourea and related compounds caused hyperplasia of the thyroid when administered to rats. Their investigations quickly led to the introduction of the thionamide series of antithyroid drugs, now so familiar in clinical therapeutics.
Thiocyanate and precursors of thiocyanate, such as the cyanogenic glycosides, form another group of widely distributed natural antithyroid substances. They have been found particularly in the widely used tuber cassava (manioc) (76). Cassava causes goiter when fed to rats (77). Certain sulfur-containing onion volatiles are also goitrogenic (78). All of these substances interfere with the accumulation of thyroidal iodide, an effect that usually can be overcome by an increasing iodine intake.
Delange et al. (79) observed a striking difference in incidence of goiter in two regions of an isolated island in the Kivu Lake in Eastern Dem. Rep. of Congo, although the iodine intake of both groups of ethnically identical people was approximately the same. There was a major difference in use of cassava. Delange, Ermans, Thilly and their colleagues have developed a strong case for cassava as playing a definite role in the development of endemic goiter in Zaire (76, 80, 81). In a study of several communities in the Ubangi region of Zaire, they found an interesting relationship between goiter, on the one hand, and thiocyanate and iodide excretion, on the other. The thiocyanate was derived from intestinal breakdown of the cyanogenic glycoside, linamarin, from cassava and its conversion to thiocyanate by the liver. The results indicated a reciprocal relationship between iodide and thiocyanate in that increasing amounts of iodide protected increasingly against the thiocyanate derived from the cassava. It now seems well established that cassava may contribute to the severity of endemic goiter and probably the incidence of endemic cretinism (80), but there are many severe endemics where cassava is not eaten. In these regions, it is possible that other goitrogens in the local food may contribute to the effects of a prevailing iodine deficiency. Thiocyanate may cross the human placenta (80, 82) and affect the thyroid of the fetus.
Excessive intake of iodine may itself cause goiter. A localized endemia has been reported on the coast of Hokkaido in Northern Japan (83). In this district the diet contained a huge amount of seaweed, and excretion of 127I in the urine exceeded 20 mg/day. The uptake of RAI by the thyroid was low, and some of it could be discharged by administration of thiocyanate indicating impairment of organification. Similar findings have been reported from coastal (84) and continental (85) China.
From Gaitain's review (68) it appears that firm evidence for goitrogenic action in humans has only been shown for a few compounds: thiocyanate, goitrin, resorcinol, dinitrophenol, PBB's and its oxides, excess iodine and high doses of lithium. Final proof for a definite role in endemic goiter has only been provided for thiocyanate and sulfurated organics, although substantial and circumstantial evidence favors the view that natural goitrogens, acting in concert with iodine deficiency, may determine the pattern and severity of the condition. An example is the possible role of the consumption of pearl millet in the etiology of endemic goiter in Sudan in spite of iodine sufficiency (86).
Quite recently it has been shown that selenium deficiency may have profound effects on thyroid hormone metabolism and possibly also on the thyroid gland itself (87, 88, 88 bis). In this situation the function of type I deiodinase (a selenoprotein) is impaired. Type I deiodinase plays a major role in T4 deiodination in peripheral tissues. It has been shown that when, in an area of combined iodine and selenium deficiency, only selenium is supplemented, serum T4 decreases (89). This effect is explained by restoration of type I deiodinase activity leading to normalization of T4 deiodination while T4 synthesis remains impaired because of continued iodine deficiency.
Selenium deficiency also leads to a reduction of the selenium containing enzyme glutathione peroxidase. Glutathione peroxidase detoxifies H2O2 which is abundantly present in the thyroid gland as a substrate for the thyroperoxidase that catalyzes iodide oxidation and binding to thyroglobulin, and the oxidative coupling of iodotyrosines into iodothyronines. Reduced detoxification of H2O2 may lead to thyroid cell death (87, 90). Elevated H2O2 levels in thyrocytes may be more toxic under situations of increased TSH stimulation such as is present in areas with severe iodine deficiency. Last, but not least, decreased availability of glutathione peroxidase impairs thyroid hormone synthesis in the thyroid gland, a fact that could also contribute to decreased T4 synthesis.
Selenium deficiency certainly plays a role in the etiology of the type of myxedematous endemic cretinism seen in Central Africa (See endemic cretinism) but does not by itself constitute a cause of endemic goiter. Extensive epidemiological data collected in China indicated that all selenium-deficient areas were IDD-endemic areas. However, the reverse is not true : IDD can be very severe in many selenium-rich areas (91).
There are no gross or microscopic features that distinguish the thyroid of endemic goiter from changes that may appear in simple and sporadic goiter. The changes evolve through stages. In the very young, or in older patients who have lived under constant iodide deprivation, the finding is extreme hyperplasia. In some instances only a cellular organ is found, with little or no colloid. (Fig. 3) The evolution of pathologic findings in humans have been detailed and well illustrated by Correa (92) and Studer and Ramelli (93) and follow the pattern of events first described by Marine (94) and known as the Marine cycle. In this formulation, repeated episodes of hyperplasia induced by iodine deficiency are followed by involution and atrophy, the result being a gland containing a mixed bag of nodules, zones of hyperplasia, and involuting, degenerative, and repair elements.
A diagnosis of endemic goiter implies that the cause is known, or at least strongly suspected. Usually water and food are found to have very low iodine content. The thyroid glands are often diffusely enlarged in childhood, but are almost always nodular in adults. The typical laboratory findings are elevated radioiodine thyroidal uptake (RAIU), normal or low T4 and FT4 levels, normal or elevated T3 levels, normal or elevated TSH levels, and diminished urinary 127I excretion. RAIU is typically suppressible when thyroid hormone is given, but not always. Scanning with radioidide or TcO4- shows a mottled distribution of the isotope. Antithyroglobulin or thyroperoxidase antibodies are usually absent. In an area of endemic goiter, the diagnosis can be presumed if the goiter is a community problem, but one must always be wary of missing individual patients with thyroiditis, thyrotoxicosis or thyroid carcinoma.
When iodine intake is abnormally low, adequate secretion of thyroid hormones may still be achieved by marked modifications of thyroid activity. These adaptative processes include stimulation of the trapping mechanism of iodide by the thyroid as well as of the subsequent steps of the intrathyroidal metabolism of iodine leading to preferential synthesis and secretion of T3. They are triggered and maintained by increased secretion of TSH. The morphological consequence of prolonged thyrotropic stimulation is thyroid hyperplasia (review in ref. 63). Stanbury and his colleagues (95) have synthesized this concept in their classical publication : « Endemic goiter. Adaptation to iodine deficiency ». However, large goiters are characterized by their decreased ability to synthesize thyroid hormones and, therefore, appear rather as signs of failure of adaptation to iodine deficiency.
The first functional consequence of iodine deficiency is an increase in the uptake of iodide by the thyroid mediated via a transmembrane protein, the sodium iodide symporter (NIS) (96). There is a clear inverse relation between iodine supply and thyroidal uptake of radioiodide. The increased uptake may be accompanied by and may result from an increase in the serum levels of TSH. However, elevated TSH in endemic goiter is almost systematically found only in conditions of extreme iodine deficiency but in only a small fraction of subjects, usually the youngest (7), in conditions of mild iodine deficiency. It could be that it is the sensitivity of the thyroid to TSH rather than the TSH levels themselves which mainly varies with iodide supply. However, whatever the relative roles of TSH levels and sensitivity to TSH, the thyroid is stimulated as demonstrated by increased secretion and elevated serum levels of thyroglobulin.
For any adequate adjustment of iodine supply to the thyroid, iodide trapping must fullfil two conditions. First, it must reduce the amount of iodide excreted in the urines to a level corresponding to the level of iodine intake in order to preserve the preexisting iodine stores. Second, it must ensure the accumulation in the thyroid of definite amounts of iodide per day estimated at at least 100 μg/day in adolescents and adults. The increase in the iodide clearance by the thyroid despite the decrease in the serum concentration of iodide maintains a normal absolute uptake of iodide by the thyroid and an organic iodine content of the thyroid which remains within the limits of normal (i.e., 10-20 mg) as long as the iodine intake remains above a threshold of about 50 μg/day. Below this critical level of iodine intake, despite a further increase of thyroid iodide clearance, the absolute uptake of iodide diminishes and the iodine content of the thyroid decreases with functional consequences resulting in the development of a goiter (97, 98).
Thyroid hyperplasia induced by iodine deficiency is associated with an altered pattern of thyroid hormonogenesis : the abnormal configuration of the poorly iodinated thyroglobulin in the thyroid colloid is accompanied by an increase in poorly iodinated compounds, monoiodotyrosine (MIT) and T3, and a decrease in diiodotyrosine (DIT) and T4. The increase of the MIT/DIT and T3/T4 ratios is closely related to the degree of iodine depletion of the gland (99).
Similarly, the T3/T4 ratio in the serum is elevated in conditions of iodine deficiency probably because of thyroidal secretion of T4 and T3 in the proportion in which they exist within the gland and/or because preferential secretion of T3 or increased peripheral conversion of T4 to T3. The shift to increase T3 secretion plays an important role in the adaptation to iodine deficiency because T3 possesses about 4 times the metabolic potency of T4 but requires only 75 % as much iodine for synthesis.
However, efficient adaptation to iodine deficiency is possible in the absence of goiter as demonstrated in non goitrous patients in endemic goiter areas such as New Guinea (100) and the Congo (101). Moreover, adequate adaptation to iodine deficiency has been demonstrated in areas of severe iodine deficiency in the absence of endemic goiter (79, 102). This clearly indicates that goiter is not required for achieving efficient adaptation to iodine deficiency. Rather, in these conditions, efficient adaptation to iodine deficiency is possible thanks to a high iodide trapping capacity but with only a slight enlargement of the thyroid. At this stage, the characteristic hyperplastic picture includes abundant parenchyma, high follicular epithelium and rare colloid. On the contrary, in large goiters, the major part of the gland is occupied by extremely distended vesicles filled with colloid with a flattened epithelium. The mechanism responsible for the development of colloid goiter is not fully understood (103). It is not TSH hyperstimulation. It must be the consequence of an imbalance between thyroglobulin synthesis and hydrolysis, i.e. secretion. In these conditions, iodide is diluted while thyroglobulin is in excess, resulting in a lesser degree of iodisation of thyroglobulin and, consequently, in a decrease in iodothyronines synthesis and secretion (104). Hydrolysis of large amounts of poorly iodinated thyroglobulin will result in an important leak of iodide by the thyroid and enhanced urinary loss of iodide, further aggravating the state of iodine deficiency (105).
Therefore, there is no doubt that goiter, and especially large colloid goiters in endemic iodine deficiency represents maladaptation instead of adaptation to iodine deficiency because it leads to a vicious cycle of iodine loss and defective thyroid hormones synthesis (104).
When Sir Robert McCarrison described cretinism in north-western India during the first decade of this century (106), he delineated a neurologic form, with predominantly neuromotor defects, including strabismus, deaf-mutism, spastic diplegia, and other disorders of gait and coordination. The patients usually had a goiter. The other form, which he called the myxedematous form, showed evidence of severe hypothyroidism, short stature, and markedly delayed bone and sexual maturation. The patients usually had a thyroid normal in size and position, and were seldom deaf.
The three characteristic features of neurological endemic cretinism in its fully developed form are extremely severe mental deficiency together with squint, deaf mutism and motor spasticity with disorders of the arms and legs of a characteristic nature. (Fig. 4). As would be expected with a deficiency disease there is a wide range in the severity of the clinical features in the population affected. Studies by De Long et al (107), Boyages et al. (108) and by Halpern et al (109) have provided new observations and insights.
Figure 4. Male from Ecuador about 40 years old, deaf/mute, unable to stand or walk. Use of the hands was strikingly spared, despite proximal upper-extremity spasticity. From DeLong et al (107).
Mental deficiency is characterized by a marked impairment of the capacity for abstract thought but vision is unaffected. Autonomic, vegetative, personal, social functions and memory appear to be relatively well preserved except in the most severe cases.
Deafness is the striking feature. This may be complete in as many as 50% cretins. It has been confirmed by auditory brain stem evoked potential studies which showed no cochlear or brain stem responses even at the highest sound frequencies. These findings suggest a cochlear lesion. In subjects with reduced hearing a high tone defect is apparent. Deafness is sometimes absent in subjects with other signs of cretinism. All totally deaf cretins were mute and many with some hearing had no intelligible speech.
The motor disorder shows a characteristic proximal rigidity of both lower and upper extremities and the trunk. There is a corresponding proximal spasticity with markedly exaggerated deep tendon reflexes at the knees, adductors and biceps. Spastic involvement of the feet and hands is unusual or, if present, is much milder than that of the proximal limbs. Function of the hands and feet is characteristically preserved so that most cretins can walk. This observation is very useful in differentiating cretinism from other forms of cerebral palsy commonly encountered in endemic areas, such as cerebral palsy from birth injury or meningitis.
In addition to frank cretinism, a larger proportion of the population (estimated to be 3-5 times as great) suffers from some degree of mental retardation and coordination defect. Comparative population based neuropsychological assessments of children in areas of iodine deficiency compared with areas with adequate iodine intake confirm a shift of the intelligence curve to the left in the iodine deficient areas. Careful examination of affected individuals in such areas reveals a pattern of neurological involvement similar to that seen in frank cretins, although of milder degree. In assessing these less severe defects, nonverbal tests are most helpful and school progress is a good indicator. After the age of 3 years drawings are very useful, indicating a defect in visual motor integration. Finally, elevated hearing thresholds have been reported in children with no other signs of endemic cretinism in conditions of mild iodine deficiency (110).
On the basis of his clinical observations, De Long (111) suggests that the neuropathological basis of the clinical picture includes underdevelopment of the cochlea for deafness; maldevelopment of the cerebral neocortex for mental retardation; and maldevelopment of the corpus striatum (especially putamen and globus pallidus) for the motor disorder. The cerebellum, hypothalamus, visual system, and hippocampus are relatively spared. Studies of human cretin brains by modern techniques are badly needed.
The frequency of goiter and thyroid dysfunction in these defectives is similar to the ones observed in the general population.
Developmental neuropathology and available epidemiologic data suggest that the period from about 12-14 weeks until 20-30 weeks of gestation may be the critical period during which damage occurs (10). Cortical and striatal neuron proliferation, migration, and early formation of neuropil occur between 12 and 18 weeks. Cochlear development occurs at the same time. These data correlate well with the data from the Papua New Guinea trial which indicated that iodine repletion must occur by three months of pregnancy to prevent cretinism (35).
Studies already cited above on the effect of iodine deficiency on brain cell development in the newborn rat, sheep and marmoset suggest that iodine deficiency has an early effect on neuroblast multiplication. Brain weight is reduced with a reduced number of cells as indicated by lowered DNA, a greater density of cells in the cerebral cortex and reduced cell acquisition in the cerebellum. In the light of the recent evidence summarized above (Iodine deficiency in the foetus) that maternal thyroxine crosses the placenta, it is now envisaged that neurological cretinism is predominantly caused by maternal hypothyroidism due to iodine deficiency. It has been suggested that an autosomal recessive predisposition, besides maternal iodine deficiency, may play an etiological role in neurological cretinism (112).
Administration of PTU to pregnant mice also causes abnormalities in their pups in the tectorial membrane of the organ of Corti and results in deafness (113). These experiments strongly suggest that it is the hypothyroidism per se that somehow damages the developing auditory system, causing deafness and other neurologic defects that are found characteristically in the endemic cretin. Patients with Pendred's syndrome also have goiter and deaf-mutism. Characteristically, when given SCN- or C104-, trapped but not yet oxidized iodide is discharged. There is no evidence that the Pendred syndrome can account for the deafness associated with endemic goiter.
The typical myxedematous cretin (Fig 5) has a less severe degree of mental retardation than the neurological cretin. It has all the features of extremely severe hypothyroidism present since early life, as in non recognized sporadic congenital hypothyroidism (114, 115) : severe growth retardation, incomplete maturation of the features including the naso-orbital configuration, atrophy of the mandibles, puffy features, myxedematous, thickened and dry skin, dry and rare hair, eyelashes and eyebrows and much delayed sexual maturation.
Figure 5. Myxedematous endemic cretinism in the Democratic Republic of Congo. Four inhabitants aged 15-20 years : a normal male and three females with severe longstanding hypothyroidism with dwarfism, retarded sexual development, puffy features, dry skin and hair and severe mental retardation. From Delange (216).
Contrasting with the general population and with neurological cretinism, goiter is usually absent and the thyroid is often even not palpable, suggesting thyroid atrophy. This diagnosis is confirmed by thyroid scans with showed thyroids in normal location but of small volume with a very heterogeneous and patchy distribution of the tracer (101). Thyroidal uptake of radioiodine is much lower than in the general population. The serum levels of T4 and T3 are extremely low, often undetectable and TSH is dramatically high. Markedly enlarged sella turcicae have been demonstrated, suggesting pituitary adenomas (116).
Myxedematous cretinism used to be particularly common in Zaire. The early reports by the Belgian teams indicated limited neurological abnormalities in the cretins in this country (115). The movements are torpid and the reflex relaxation is usually much prolonged. However hyper reflexia and Babinsky signs were occasionally reported while knocked knees, flat feet were obvious from the photographs of these patients reported in the literature. However, subsequent neurological examination of some of these patients by De Long (107) evidenced in some of them the neurological signs reported in the neurological type of cretinism, partly obscured by the status of severe hypothyroidism.
Iodine deficiency is, as for the neurologic type, a perequisite in the etiology of the disorder. Its role is demonstrated by 1) the correlation between the degree of iodine deficiency and the frequency of the condition (79, 80), 2) the preventive action of iodine supplementation on its incidence(34, 117), 3) the reemergence of cases in populations previously affected but submitted to programs of salt iodization after abrupt interruption of these programs as recently reported from Central Asia (61).
Three additional factors, acting alone or in combination, have been proposed for explaining the particularity of thyroid atrophy characteristic of the myxedematous type of cretinism :
1) Thiocyanate overload resulting from the chronic consumption of poorly detoxified cassava (80). Its role has been suggested in Zaire from the observation that populations in areas with severe but uniform iodine deficiency exhibit cretinism only when a critical threshold in the dietary supply of SCN is reached. SCN crosses the placenta and inhibits the trapping of iodide by the placenta and fetal thyroid (41, 82).
2) Selenium deficiency. Severe selenium deficiency has been reported in Zaire in populations where myxedematous cretinism is endemic (87-89). As indicated earlier, selenium is present in Glutathione Peroxidase (Gpx) that detoxifies H2O2 produced in excess in thyroid cells hyperstimulated by TSH because of iodine deficiency. Accumulation of H2O2 within the thyroid cells could induce thyroid cell destruction, and finally thyroid fibrosis resulting in myxedematous cretinism. It has been recently proposed that the combination of deficiencies in iodine and Selenium and SCN overload are required for explaining the occurrence of severe thyroid failure during the perinatal period with as longterm consequence the development of the myxedematous type of endemic cretinism (87).
3) Immunological mechanisms. Some authors (118, 119) but not others (120) provided evidence of immunological factors causing destruction of the thyroid, both in endemic and sporadic congenital hypothyroidism. The role of autoimmunity in the etiology endemic cretinism remains controversial.
The “ traditional ” distinction between the neurologic and myxdematous types of endemic cretinism derived from the observations of McCarrisson at the term of the century (106) have been maintained in the epidemiologic definition of the condition given by the World Health Organization (WHO) in 1986 (121) and has resulted in vivid, occasionally passionate debates on the definition of endemic cretinism. One of the reasons is that the frequency distribution of the symptoms markedly varied from one geographical area to another with “ mixed ” forms (122) The mean reason of course for this confusion is that the etiology of the disorder was unknown.
During the past five to ten years, major progress has been made in our understanding of endemic cretinism, mostly based on new epidemiological studies and on experimental aspects of thyroid homeostasis during the perinatal period in conditions of iodine deficiency, largely due to the outstanding group of Morreale de Escobar et al. (18).
During their extensive study of endemic cretinism in Western China, Boyages et al. (108) and Halpern et al. (109) reported an almost similar degree of mental retardation and frequency of neurological abnormalities in both types of cretinism. Similarly, a few years later, rather similar observations were reported from Thailand : Rajatanavin et al. (123) reported similar frequencies of low intelligence, defects in visual-perceptive neuromanual ability, sensorineural hearing loss and neurological defects in 57 neurologic, 19 myxedematous and 36 mixed cretins investigated by a multidisciplinary team in Northern Thailand.
For Boyages et al. (108) and Halpern et al. (109), the similar frequency of intelligence and neurological deficiency in all types of cretinism point to a primary pathophysiologic event which probably occurred in utero during early gestation. These authors suggested that all neurological abnormalities occurred in utero due to both maternal and fetal hypothyroxinemia secondary to severe iodine deficiency. Postnatally, the persistence of hypothyroidism, both from continuing iodine deficiency and other mechanisms causing thyroid failure, entails the development of myxedematous cretinism.
However, there is no doubt that the word myxedematous endemic cretinism used in the studies performed in China and in Zaire apply to patients with different conditions : retardation in bone maturation and in height are much less pronounced in China than in Zaire, suggesting that hypothyroidism was acquired rather late in life in the first country while it was already present in very early life in the second.This difference can be explained on the basis of the experimental data reported by the group of Morreale de Escobar (15, 18) considering that Zaire is not only iodine but also selenium deficient (88) and exposed to thiocyanate (SCN) overload (80). The key point evidenced by the Spanish group is the role of thyroid hormones produced not only by the foetus but also by the mother in the development of the brain since early fetal life, even before the onset of fetal thyroid function.
The role of selenium deficiency can be understood as follows : on the one hand, as indicated earlier, selenium deficiency can be involved in the process of thyroid atrophy causing neonatal and postnatal hypothyroidism as in sporadic congenital hypothyroidism. SCN overload triggers follicular cell necrosis and thyroid atrophy (87). On the other hand, selenium is also present in the type I 5’deiodinase responsible for the peripheral conversion of T4 into T3. Selenium deficiency could decrease the catabolism of T4 to T3 and thus increase the availability of T4 of maternal origin for the brain of the fetus during early gestation, and thus preventing part of the neurological damage (87). Thus, selenium deficiency and thiocyanate overload superimposed to iodine deficiency explain why the neurologic picture is mitigated and the signs of thyroid failure predominate in the African cretins.
Although a certain number of issues still remain to be clarified, this unifying hypothesis on the etiopathogenesis of endemic cretinism constitutes one of the important steps forward in the field of IDD.