Iodide-induced goiter occurs in about 10% of the population of Hokkaido, a Japanese island. The inhabitants of this island, particularly the fishermen and their families, consume large quantities of an iodide-rich seaweed called kombu. The quantity of iodide ingested daily may exceed 200 mg. Despite goiter, hypothyroidism is rare (31,32). In another study, goiter due to iodide-rich drinking water was observed in 10% of the subjects residing in 19 Chinese counties (28). In general, those subjects had normal thyroid function. Other Chinese subjects drinking iodide-rich water had a prevalence of clinical and subclinical hypothyroidism of 2% and 6%, respectively (29). A later study conducted in China reported that subjects with mildly deficient iodine intake (urinary iodine excretion 84 to 88 μg/L), those with more than adequate iodine intake (urinary iodine excretion 214 to 243 μg/L) and those with excessive iodine intake (urinary iodine excretion 634 to 651 μg/L) had a prevalence of subclinical hypothyroidism of 0.9%, 2.9%, and 6.1%, of overt hypothyroidism of 0.3%, 0.9%, and 2.0%, and of autoimmune thyroiditis of 0.5%, 1.2%, and 2.8%, respectively (33). These findings were confirmed in another survey showing that subjects with a median urinary iodine excretion of 200 to 300 μg/L and those with >300 μg/L had a prevalence of subclinical hypothyroidism of 1.99% and 5.03%, respectively (34). These authors reported in the same population that a median urinary iodine excretion of 100 to 200 μg/L reflected a safe dietary iodine intake (35). In American Peace Corps, in workers who drank iodide-rich water due to a faulty iodinator, serum TSH concentrations were above 4.2 mU/L in 29%. This value decreased to 5% after iodide removal (30). An increased prevalence of hypothyroidism (12%), defined by serum TSH concentrations >5 mU/L, was observed in thyroid autoantibody-negative Japanese subjects with an iodide concentration of more than 9.5 mg/L in morning urine samples; in subjects with normal iodide excretion, the prevalence of hypothyroidism was only 2% (36). When the iodide intake was restricted, the increased serum TSH concentrations returned to normal in patients with negative antithyroid antibodies, but not in those with antibodies, suggesting that excessive iodide intake should be considered as a cause of hypothyroidism in addition to chronic thyroiditis in these areas (37). In an elderly Icelandic population, high urinary iodide excretion rates (median 150 μg/L, range 33 to 703 μg/L) were found to be accompanied
by a high prevalence (18%) of serum TSH concentrations >4 mU/L. In subjects residing in Jutland with low urinary iodide excretion (median 38 μg/L, range 6 to 770 μg/L), serum TSH levels were low (<0.4 mU/L) in 10%. The incidence of positive thyroid antibodies was similar in both populations (38). It has been observed in Denmark that after mandatory iodine fortification of salt, the incidence of hypothyroidism slightly increased, but only in young and middle-aged subjects (39).

In Poland, the cumulative prevalence of overt and subclinical hypothyroidism progressively increased in respect to urinary iodide excretion, being 5%, 12%, and 32%, corresponding to a median urinary excretion of 72, 100, and 513 μg/g creatinine, respectively (40). The prevalence of positive antithyroid peroxidase (anti-TPO) and antithyroglobulin (anti-Tg) antibodies was similar in all three groups. In Brazil, it has been observed that after 5 years of excessive iodine intake the prevalence of autoimmune thyroiditis was 16.9%, and hypothyroidism was present in the 8% of subjects with autoimmune thyroiditis (41). In general, populations with high dietary iodine intake have increased serum TSH concentrations, especially with advancing age and possibly related to the presence of thyroid autoimmunity (42).

Skin application of povidone-iodide for 3 to 133 months resulted in subclinical hypothyroidism in 3 of 27 patients with neurologic diseases and negative antithyroid antibodies (43). Gargling with povidone iodine has been reported to induce overt and subclinical hypothyroidism in two subjects without antithyroid antibodies. These subjects became euthyroid after gargling was discontinued (44). These findings suggest that iodide-induced hypothyroidism might occur even in subjects with no underlying thyroid disease, including thyroid autoimmunity.

Histologic examination of the thyroid of patients with iodide-induced hypothyroidism revealed the presence of lymphocytic infiltration in only half the specimens examined. In the other specimens, hyperplastic changes in the follicles with papillary folding, cuboidal or columnar change of cells with clear and vesicular cytoplasm, and markedly reduced colloid in the distended follicles were seen. These changes were reversible after iodide withdrawal (45). In contrast to these findings, the administration of a single dose of 50 to 70 mg of potassium iodide (KI) to children and adults for iodide prophylaxis following the Chernobyl reactor accident was not accompanied by an increment in serum TSH concentrations (46).

Usually, serum T₄ and T₃ concentrations are low or low-normal, and serum TSH is increased in patients with iodide-induced hypothyroidism. The thyroid radioactive iodide uptake would be expected to be very low in these patients; however, about 30% have a normal or high thyroid radioactive iodide uptake (47). Similar findings have been observed in European but not US patients who developed iodide-induced hypothyroidism after amiodarone administration (48,49).

Iodide readily crosses the placenta and is concentrated by the fetal thyroid and large quantities of iodide administered to pregnant women may result in goiter in the newborn, probably because the fetal thyroid is inordinately sensitive to the inhibitory effect of iodide on hormone synthesis (50). Whether the inhibitory effect of iodide is also exerted on the release of thyroid hormones is not clear. Studies in rats strongly suggest that the inhibitory effects are exerted in utero as well as in the late neonatal period, which corresponds to the last few weeks of human fetal life (51). The thyroid of the fetus and newborn can be exposed to iodide from various routes. Large quantities of kombu, an iodine-rich seaweed, consumed by pregnant Japanese women caused neonatal transient hypothyroidism. The neonates whose mother consumed kombu had increased urinary iodine excretion ≥300/L, serum TSH concentrations higher than control, and 12 of 15 required temporary l-T₄ treatment (52).

By ultrasonography and cordocentesis, goiter and hypothyroidism have been diagnosed in a fetus whose asthmatic mother consumed two to three spoonfuls per day of a syrup containing 130 mg/15 mL of iodide (53). Severe goitrous hypothyroidism was reported in a newborn infant with a history of iodide exposure in utero derived from an expectorant used by the mother (54).

Vaginal douching with iodide-containing solutions in nonpregnant women results in an increase of serum iodide concentrations and a small increase in serum TSH concentrations (55). In contrast, in nonpregnant women vaginal disinfection with povidone-iodide vaginal pessaries and obstetric cream does not affect serum iodide concentrations and thyroid function (56,57). Transient hypothyroidism of the newborn, as indicated by an elevation of the serum TSH, has been reported to follow the application of vaginal solutions of povidone-iodide and in a few cases after povidone-iodide cream application during the last trimester and during labor (50,58). Drugs containing iodide also may induce hypothyroidism in the fetus. Bartalena et al. (59) reviewed 64 cases of pregnant women treated with amiodarone. Transient hypothyroidism was detected in 17% of the newborns. Other newborns with hypothyroidism due to maternal and direct fetal amiodarone administration have been reported. In these newborns hypothyroidism was transient and, when specified, goiter was not present (60,61,62). Finally, maternal exposure to iodinated contrast media during pregnancy caused neonatal hypothyroidism, especially in preterm neonates, 18.2%, as compared to term neonates, 8.3% (63).

Topical application of povidone-iodide to the skin of the newborns may induce transient neonatal hypothyroidism, more frequently in premature, low-birthweight infants (64). Serum TSH concentrations above 20 mU/L occurred in 25% of the cases, promptly normalizing after the iodide-containing antiseptic was discontinued (65). In a preterm neonate with large scalp skin loss, treatment with topical, prolonged povidone iodine was accompanied by a marked increase of serum TSH, higher than 500 mU/L, and thyroid enlargement by ultrasound (66). Skin disinfection of a single at-term neonate with iodine-containing antiseptic solution caused transient hypothyroidism (67). The injection of small amounts of an iodinated contrast dye through nonradiopaque silastic catheters in premature infants induced hypothyroidism in some and thyrotoxicosis in others (64,68).

The administration of a single dose of 15 mg of KI to newborn infants for iodide prophylaxis after the Chernobyl nuclear reactor accident resulted in a transient increase of serum TSH concentrations in 0.4% of the 3,214 treated infants (46). In the same population, the exposure to iodide in utero due to maternal iodide prophylaxis did not result in an increase in congenital hypothyroidism (46).

Iodides are actively transported by breast tissue and secreted into the milk. Thus, the administration of iodides to nursing
mothers could result in iodide-induced hypothyroidism and goiter in their infants. Perinatal hypothyroidism was observed in Australian Asian newborns whose mothers consumed a large amount of seaweed soup during pregnancy and the postpartum period to promote their breast milk supply (69). Transient iodine-induced hypothyroidism has been diagnosed in breast-fed premature neonates whose mothers were medicated with iodoform gauze because of an abscess or were exposed to topical iodine treatment. In contrast with neonatal thyroid function, maternal thyroid function was normal (70,71).

Iodide contamination is the major cause of transient neonatal hypothyroidism (72), responsible for 3% of recalls at screening for congenital hypothyroidism (73). Recently, it has been reported in Bosnia and Herzegovina that 5.5% of 8,105 newborns had screening serum TSH concentrations higher than 5 mIU/L, especially in those who were born by cesarian section and exposed to iodine-containing antiseptics (74). Because these reports emanate primarily from continental Europe, where mild iodide deficiency is present, it is possible that iodide deficiency might predispose the fetal and neonatal thyroid to the inhibitory effect of iodide on hormone synthesis. Thus, the supplementation of 150 μg iodide/day to pregnant women in Denmark, a mild-to-moderate iodide deficiency area, resulted in a significant increase in the percentage of cord serum TSH values above 10 mU/L, 41% in neonates whose mothers were supplemented with iodide and 31% in the control group (75). Brown et al. reported that transient hypothyroidism is not a common sequela of routine skin cleansing with iodide in premature newborn infants in the United States, an iodide-sufficient area (76). Furthermore, Momotani et al. (77) reported that only 2 of 35 newborns whose mothers had been treated with 6 to 40 mg iodide daily from 11 to 37 weeks of gestation for Graves’ disease had elevated cord serum TSH concentrations. It is possible that the lack of fetal iodide-induced hypothyroidism in these newborns was due to the concomitant presence of autoimmune thyroid hyperfunction.

Endemic iodide-induced goiter has also been observed in 64% of children residing in a village in central China (26). These children drank water containing 462 mg iodide per liter. No increased prevalence of lymphocytic thyroiditis was found in these children. Thyroid autoantibodies as well as immunoglobulins that inhibited TSH binding were negative. Thyroid growth–stimulating immunoglobulins were found in 60% of goitrous children but were absent in children without goiter who resided in an area with increased iodide concentrations in the drinking water (27). This finding has yet to be confirmed.

In Chinese children, consuming iodine-rich drinking water, urinary iodine concentration reached values higher than 1,500 μg/L and goiter prevalence was 3.7 times higher than that in those children with urinary iodine levels of 100 to 199 μg/L (78). The administration of 40 to 65 mg iodide daily to euthyroid children residing in Greece resulted in serum TSH concentrations above 4.2 mU/L in 75%. In contrast, adult subjects did not have an increase in serum TSH concentrations (79). These findings suggest that the autoregulatory mechanisms are immature in children, and therefore the thyroid is susceptible to the inhibitory effects of excess iodide on hormone synthesis and goiter may develop.

Patients with chronic nonthyroidal illness usually are not susceptible to the inhibitory effect of iodide despite the multiplicity of thyroid dysfunction. However, certain diseases may predispose the patient to iodide-induced thyroid dysfunction.

Iodide-induced hypothyroidism has been reported in patients with a variety of chronic lung diseases, including asthma, treated for prolonged periods with iodide-containing expectorants (80,81). However, underlying Hashimoto’s thyroiditis predisposing these patients to the inhibitory effect of iodide was not ruled out (see later).

Children with cystic fibrosis, especially those treated with sulfisoxazole, are particularly susceptible to iodide-induced hypothyroidism (82). No evidence of underlying thyroid dysfunction was found in these children, although the accumulation of lipofuscin has been observed in the thyroids of patients with cystic fibrosis. The significance of the latter finding is unclear.

In children and adults with thalassemia major and requiring chronic blood transfusions, iodide administration (60 mg/day) resulted in subclinical hypothyroidism (TSH >5 mU/L) in 60%. TSH returned to basal levels 2 to 3 weeks after iodide withdrawal. It appears that hemosiderosis renders the thyroid of these patients susceptible to the inhibitory effects of iodide (83).

Patients with chronic renal failure frequently have thyroid dysfunction, including thyroid enlargement and abnormal thyroid function tests. Some of these abnormalities may be due to chronic disease, although iodides have been suspected as a potential pathogen because they are used as antiseptics in these patients. In one study, however, no relationship was found between thyroid abnormalities and the application of iodide-containing skin antiseptics (84). In another study, iodide-induced hypothyroidism was diagnosed in 3% of patients on chronic dialysis treatment (85). In these patients the thyroid was enlarged, thyroid radioactive uptake was normal or elevated, the iodide-perchlorate discharge test was positive, and no lymphocytic infiltration was present on cytologic examination. After restriction of iodide intake in the 83% of patients with renal dysfunction and increased serum TSH concentrations, thyroid function tests returned to normal (86).

The use of mucolytic expectorants containing iodinated glycerol is particularly frequent in elderly subjects, although this may no longer be a problem in the United States because iodide has been removed from these products. The occurrence of mild hypothyroidism after iodinated glycerol administration was observed in an elderly patient with a previous episode of severe hypothyroidism induced by KI administration. This also occurred in subjects without known thyroid disorders (87). In these elderly subjects, the abnormalities of thyroid function resolved spontaneously after the therapy was withdrawn.

Iodide-induced hypothyroidism has been reported in patients with anorexia nervosa (88,89). In one patient, excessive iodide intake resulting from kombu ingestion induced severe hypothyroidism despite negative antithyroid antibodies (89). After withdrawal of kombu, thyroid function returned to normal. Kombu is used in Japan as a low-calorie food, and one package contains approximately 13 mg of iodide.

In Japanese hypothyroid patients with negative thyroid antibodies, restriction of iodide intake resulted in a decrease of urinary iodide excretion from a median value of 834 μg I/g creatinine to 123 μg I/g creatinine a week later and a parallel decrease in the serum TSH concentrations from 123 to 4 mU/L. Furthermore, during long-term follow-up, these patients remained euthyroid, except for one subject, who resumed excessive iodide intake (90).

Patients with chronic lymphocytic (Hashimoto’s) thyroiditis often develop hypothyroidism due to thyroid destruction by the autoimmune process and/or the presence of TSH-blocking antibodies. Administration of pharmacologic quantities of iodide (180 mg/day) resulted in hypothyroidism in more than 60% of the patients in one study. The iodide-perchlorate discharge test was positive, indicating a defect in the intrathyroidal organification of iodide, in patients who developed iodide-hypothyroidism and negative in those who did not (91). The failure of the thyroid to escape from the inhibitory effect of iodide is probably due to a persistent Wolff–Chaikoff effect and not due to inhibition of the release of T₄ and T₃ from the thyroid. In Bio-Breeding/Worcester (BB/Wor) rats, which are genetically susceptible to chronic lymphocytic thyroiditis, pharmacologic quantities of iodide enhanced the development of lymphocytic thyroiditis without inducing hypothyroidism (92). In BB/Wor sublines with the most extensive lymphocytic thyroiditis, however, iodide administration did induce hypothyroidism (93). Contradictory results have been obtained when patients with Hashimoto’s thyroiditis were exposed to a small increase in iodide intake. The administration of 1.5 mg iodide daily for 3 months to patients with Hashimoto’s thyroiditis did not induce hypothyroidism (94). A few Japanese patients with primary hypothyroidism due to lymphocytic thyroiditis and a high dietary iodide intake became euthyroid when the iodide intake was restricted (95,96). Interestingly, some of these patients had an unexpectedly high thyroid radioactive iodine uptake and a positive iodine perchlorate discharge test. Resumption of normal thyroid function after iodine restriction has been confirmed in other studies (90,97). However, a large number of these patients had a relapse of their hypothyroidism even in the presence of normal serum nonhormonal iodide concentrations (90).

Small quantities of iodide given chronically to four patients with Hashimoto’s disease who resided in an area of sufficient iodide intake did not induce any changes in thyroid function (13). Eight of forty patients with high TPO antibody levels residing in an area of mild iodide deficiency developed subclinical or overt hypothyroidism following the ingestion of only 250 μg KI/day for 4 months (98). These patients had serum TSH concentrations >3 mU/L before iodide supplementation. During iodide supplementation, the TPO antibody titer did not change. In another study conducted in patients with autoimmune thyroiditis residing in an area of mild iodide deficiency, small quantities of iodide caused a transient increase in serum T₄ and T₃ concentrations (99). Small quantities (150 μg/day) of iodide given to moderately iodide-deficient TPO-positive pregnant women did not induce or worsen postpartum thyroid disease (100).

Before the discovery of antithyroid drugs, the sole medical treatment of Graves’ thyrotoxicosis was the chronic administration of large quantities of iodide. Most patients were reasonably well controlled on this regimen, but thyrotoxicosis recurred in many, and a few patients developed reversible hypothyroidism. When patients with Graves’ disease treated with iodide-¹³¹I were given iodide (250 mg/day) 1 to 2 weeks after ¹³¹I therapy, 60% developed transient hypothyroidism (101). Euthyroid patients treated years earlier either with ¹³¹I or thyroidectomy developed severe hypothyroidism during the administration of pharmacologic quantities of iodide. The hypothyroidism was transient, and thyroid function returned to normal after iodide withdrawal (102). All patients who developed hypothyroidism had a positive iodide-perchlorate discharge test. In euthyroid subjects previously treated with antithyroid drugs for Graves’ disease, the chronic administration of 10 drops of a saturated solution of potassium iodide (SSKI) induced an increase in basal or TRH-stimulated serum TSH concentrations irrespective of the iodide-perchlorate discharge test (103). Basal and TRH-stimulated serum TSH concentrations returned to normal 60 days after SSKI withdrawal.