Iodine-Induced Hypothyroidism in Full-term Infants With Congenital Heart Disease: More Common Than Currently Appreciated?

Abstract

Context:

Iodine is a micronutrient essential for thyroid hormone synthesis. Thyroid hormone is critical for normal neurocognitive development in young infants, and even transient hypothyroidism can cause adverse neurodevelopmental outcomes. Both iodine deficiency and excess can cause hypothyroidism. Although iodine-induced hypothyroidism is well recognized in premature infants, full-term neonates have received less attention. Infants with congenital heart disease (CHD) are commonly exposed to excess iodine from administration of iodinated contrast agents during cardiac catheterization as well as topical application of iodine-containing antiseptics and dressings; hence, this is a vulnerable population.

Objective:

We report three cases of iodine-induced hypothyroidism in full-term neonates with CHD after cardiac angiography and topical application of iodine-containing antiseptics and dressings in the operative setting.

Results:

Three neonates with CHD and normal thyroid function at birth developed hypothyroidism after exposure to excess iodine. Two of these infants had transient hypothyroidism, and one had severe hypothyroidism requiring ongoing thyroid replacement therapy. All infants were asymptomatic, with hypothyroidism detected incidentally in the inpatient setting due to repeat newborn screening mandated by the long duration of hospitalization in these infants.

Conclusions:

Iodine-induced hypothyroidism may be under-recognized in infants with CHD exposed to excess iodine. Systematic monitoring of thyroid function should be considered to avoid potential long-term adverse neurodevelopmental effects of even transient thyroid dysfunction in this susceptible population.

Iodine is an essential micronutrient for thyroid hormone synthesis, and iodine deficiency as well as excess can result in thyroid dysfunction with potential adverse neurocognitive sequelae in children, particularly in those <3 years of age in whom brain development is thyroid hormone-dependent (1–4). The thyroid gland is efficient in handling iodine deficiency and excess within limits (2). In the setting of excess iodine exposure, the thyroid gland actively inhibits organification of iodine by an autoregulatory mechanism known as the acute Wolff-Chaikoff effect (5). Excess intrathyroidal iodine results in decreased thyroid peroxidase activity, thereby inhibiting thyroid hormone synthesis. With continued iodine exposure, transport of excess iodine into the thyroid gland is reduced, resulting in escape from the acute Wolff-Chaikoff effect and resumption of normal thyroid hormone synthesis (6). This escape may be delayed in the fetus and neonate due to immaturity of the thyroid gland, resulting in the development of hypothyroidism (7–9).

Hypothyroidism has been described most commonly in premature infants exposed to excess iodine by transcutaneous absorption of iodine-containing skin antiseptics (10), and many newborn intensive care units have abandoned the routine use of these agents. Less frequent causes of iodine-induced hypothyroidism include direct ingestion of breast milk containing high iodine levels (8, 9) and iv administration of iodine-containing radiological contrast agents (10–13). Hypothyroidism in full-term infants is unusual but can occur if the iodine load is sufficiently large.

Infants with congenital heart disease (CHD) are a potentially vulnerable population. They commonly receive large iv contrast loads during cardiac catheterization and undergo surgical procedures where topical iodine-containing antiseptics and dressings are frequently applied. For example, a typical dose of 3–5 mL/kg of the contrast agent Ioversol 350 administered to these infants during cardiac catheterization contains 350 mg/mL of organically bound iodine, providing an iodine load that far exceeds the tolerable upper limit of 200 μg/d for infants (14). Two small prospective studies have documented hypothyroidism in infants with CHD exposed to iodinated contrast (11, 12). Currently there are no generally accepted guidelines for screening of thyroid function in these infants.

We report a series of three cases of unsuspected iodine-induced hypothyroidism in full-term neonates with CHD after cardiac angiography and surgery that was detected during mandated screening.

Subjects and Methods

Three full-term infants with CHD were evaluated for thyroid dysfunction detected during hospitalization. All infants underwent routine newborn screening by the New England Newborn Screening Program (NENSP), which uses a primary T₄ screening strategy. TSH measurements are performed in newborns whose blood T₄ values are in the lowest 10th percentile and in all infants who are hospitalized in special care nurseries or neonatal intensive care units at the time of sampling (15). Repeat filter-paper specimens are also routinely obtained on all infants hospitalized in special care nurseries or neonatal intensive care units at 2 weeks, at 1 month, and monthly thereafter, until discharge or transfer to another facility. Both TSH and T₄ are assayed on all repeat screens. Total T₄ and TSH are assayed on filter-paper blood specimens by a time-resolved fluoroimmunoassay procedure (AutoDELFIA system; Perkin-Elmer; normal range for T₄ in infants, >5 μg/dL; normal range for TSH, <20 mU/L at 24–96 h of age and <15 mU/L at >96 h of age).

Serum TSH was measured in patients with abnormal results on NENSP screening by a third-generation chemiluminescent immunoassay (Access, hypersensitive human TSH; Beckman Coulter; normal range for infants 1–20 wk of age, 1.7–9.1 mIU/L). Free T₄ (FT4) in serum was measured by an electrochemiluminescence immunoassay (Roche Diagnostics; normal range for infants 1–20 wk of age, 0.9–2.3 ng/dL). Whole blood iodine concentrations were assayed on filter-paper blood specimens by spectrophotometry using 0.5-inch punch samples after a modification of the method of Benotti et al (16) (Technicon Autoanalyzer; Technicon Instrument Inc; normal range, <0.01 μg) (8). Spot urinary iodine concentrations were measured using inductively coupled plasma mass spectrometry (ARUP Laboratories; normal range, 150–220 μg/L).

Case Reports

Case 1

A full-term male infant (birth weight, 3200 g) was diagnosed with critical aortic stenosis, hypoplastic aortic arch, restrictive atrial septal defect, restrictive ventricular septal defect (VSD), and aberrant right subclavian artery. Dysmorphic features present on physical examination included retrognathia, wide nasal bridge, and long fingers and toes. Genetic testing revealed a chromosome Xq28.1 megabase duplication of unknown significance that was subsequently identified in his mother. The newborn thyroid screen was normal on day 3 of life (Table 1).

Table 1.

Clinical Characteristics of Neonates with Iodine-Induced Hypothyroidism

Case No.	Age at Cardiac Angiography, d	Age at Surgical Procedures, d	Initial Newborn Screen Filter-Paper Blood-Specimen Concentrations			Newborn Screen Filter-Paper Blood-Specimen Concentrations at Diagnosis of Hypothyroidism			Serum Concentrations at Diagnosis of Hypothyroidism			Spot Urine Iodine, μg/Le	Blood Iodine, μgf	Follow-Up
			TSH, mIU/La	TT4, ng/dLb	Age, d	TSH, mIU/La	TT4, ng/dLb	Age, d	TSH, mIU/Lc	FT4, ng/dLd	Age, d
1	3	4, 5, 6, 8	<2.5	9.1	3	18.5	4.1	12	175	0.3	25	835	0.02	Continues on thyroid replacement therapy (4 μg/kg/d) at 15 mo of age without any dose adjustment with weight gain
2	2, 3	7, 10, 15	30.9g	11.4	<1g	—	—	—	42.7	1.44	13	2664	0.09	Euthyroid after discontinuation of thyroid hormone supplementation on day 23 of life
3	18	1, 8, 24	5.1	7.7	6	30.1	12.3	31	13.6	0.88	36	13 827	—	Euthyroid by day 45 of life without thyroid hormone supplementation

Abbreviation: TT4, total T₄.

^aNormal range in infants, <2.5–20 mIU/L at 24–96 h of age; <2.5–15 mIU/L at >96 h of age.

^bNormal range in infants, >5 ng/dL.

^cNormal range in infants 1–20 weeks of age, 1.7–9.1 mIU/L.

^dNormal range in infants 1–20 weeks of age, 0.9–2.3 ng/dL

^eNormal range, 150–220 μg/L.

^fNormal range, <0.01 μg.

^gHeel-stick specimen obtained at 10 h of age; TSH level consistent with appropriate postnatal surge.

The infant underwent cardiac angiography with 6.2 mL/kg of Ioversol 350 (containing 6900 mg of organically bound iodine) on day 3 of life, followed by stage 1 palliation surgery on day 5 of life, with placement of a right ventricle to pulmonary artery shunt. An iodine-containing surgical dressing was applied to the open sternum until sternal closure on day 8 of life. Peripherally inserted central catheters (PICCs) were placed on days 4 and 6 of life.

Subsequent testing revealed an elevated newborn screen TSH level and low FT4 level on day 12 of life, with concomitantly elevated blood and spot urine iodine levels (Table 1). On day 25 of life, TSH rose further on serum testing, and FT4 was critically low. Thyroid hormone replacement therapy (12 μg/kg/d) was initiated on day 26 of life. The infant continues to receive thyroid hormone replacement therapy at 15 months of age without a need for subsequent dose adjustment with weight gain (current dose, 4 μg/kg/d).

Case 2

A full-term female infant (birth weight, 3240 g) was diagnosed with critical aortic stenosis and a hypoplastic left ventricle. The initial newborn thyroid screen at 10 hours of life was normal. Cardiac catheterizations were performed on days 2 and 3 of life, with a cumulative dose of 9.7 mL/kg of Ioversol 350 (10 900 mg of organically bound iodine). On day 7 of life, the infant underwent placement of a PICC line. Stage I palliation surgery was performed on day 10 of life, with right ventricle to pulmonary artery conduit placement. Iodine-containing dressings were applied to the open sternum that was closed after mediastinal exploration and washout on day 15 of life. An elevated TSH level was incidentally noted on serum testing on day 13 of life, with concomitant elevation in blood and spot urine iodine levels (Table 1). Thyroid hormone replacement therapy (12 μg/kg/d) was initiated on day 14 of life. Thyroid function normalized by day 23 of life, and thyroid replacement therapy was gradually discontinued.

Case 3

A full-term female infant (birth weight, 2870 g) was diagnosed with coarctation of the aorta, hypoplastic transverse arch, hypoplastic aortic and mitral valves, and multiple VSDs. The initial newborn screen was normal. Repair of the coarctation with closure of the VSD was performed in two stages, on days 1 and 24 of life. Cardiac catheterization with 5.6 mL/kg of Ioversol 350 (5500 mg of organically bound iodine) was performed on day 18 of life. PICC lines were placed on days 8 and 24 of life.

Follow-up newborn screen revealed elevated TSH and low FT4 levels on day 31 of life; spot urine iodine level was also elevated (Table 1). Serum blood urea nitrogen increased transiently to 60 mg/dL (normal, <5) from days 15 to 17 of life; serum creatinine remained normal. Serum TSH concentration on day 36 of life declined but remained slightly elevated; FT4 was low. Thyroid function normalized by day 45 of life without thyroid hormone supplementation.

All three infants in this series received dopamine for inotropic support during the perioperative period: case 1 on days 5–16, case 2 on days 1–12, and case 3 on days 24–25 of life. Cases 1 and 2 also received glucocorticoids during most of this time period. None of the infants received amiodarone.

Discussion

We describe three full-term neonates with CHD who developed hypothyroidism after exposure to iodinated contrast during cardiac angiography and topical iodine-containing dressings and antiseptics in the operative setting. One infant developed severe primary hypothyroidism and continues to receive thyroid hormone replacement therapy at 15 months of age, another required treatment transiently, and the third experienced spontaneous resolution. Hypothyroidism was detected incidentally in all three cases at mandated testing during the extended hospital stays. The infants described in this report did not display typical signs or symptoms of hypothyroidism, such as prolonged jaundice, poor feeding, hypothermia, enlarged fontanelles, hypotension, hypoactivity, constipation, myxedema, or hoarse cry. Thus, in the absence of a policy of repeat thyroid function screening, the diagnoses would have been missed entirely in those with transient hypothyroidism or delayed in the case with severe hypothyroidism until the emergence of overt signs. Because screening and follow-up of thyroid function in infants undergoing angiography or cardiac surgery may not be routinely performed in all centers caring for infants with CHD, these three neonates may represent only a minority of cases of iodine-induced thyroid dysfunction in infants with CHD.

Thyroid hormone is critical for brain development in the first three years of life, and even transient hypothyroxinemia may increase the risk of adverse neurodevelopmental outcomes in neonates (17). In a population of infants with CHD who already bear a high risk of long-term developmental delay (18), detection of iodine-induced hypothyroidism, even of a transient nature, may be even more consequential, and routine periodic monitoring of thyroid function may be necessary.

It is difficult to determine the relative contributions of excess iodine from iodinated contrast agents and topical exposures in the infants described here. Although iv contrast administration during cardiac angiography was indeed the largest quantifiable source of iodine load, prolonged exposure of deep tissue wounds to iodine-impregnated dressings in the cases of delayed sternal closure after cardiac surgery may represent another significant source of exposure. Furthermore, all three infants underwent multiple cardiac and noncardiac surgical procedures preceded by topical skin exposure to iodine-containing antiseptics, and significant quantities of iodine may have been absorbed due to the relatively small amount of sc tissue and higher surface area-to-weight ratio in neonates (19). Use of an alternative topical antiseptic, such as chlorhexidine, was not a viable option for these infants due to the risk of irritation and chemical burns (20). Hence, in some instances, iodine-containing antiseptics must be utilized in the very population that is at highest risk of iodine-induced hypothyroidism and adverse neurodevelopmental outcomes.

Mild transient hypothyroidism has been described in adults after elective cardiac catheterization and iodinated computed tomography scans (21, 22); however, in infants, the hypothyroidism tends to be more severe and prolonged. In a prospective study performed in Israel, Linder et al (11) reported primary acquired hypothyroidism (TSH > 25 mIU/L) in seven of 21 (33%) infants with congenital cardiac defects exposed to iodinated contrast and topical iodine-containing antiseptics; two of these infants continued to require thyroid hormone supplementation at 6 and 10 months, respectively. Hypothyroidism (TSH > 10 mIU/L) was also reported in 19 of 99 (19%) infants with congenital cardiac malformations 2 weeks after coronary arteriography in a Spanish prospective study; persistent hypothyroidism beyond 3 weeks was described in six of these infants (12).

Iodine-induced hypothyroidism has also been described in association with delayed sternal closure (DSC) in infants after cardiac surgery; chest wounds are bathed in povidone-iodine dressings for several days until chest closure can be completed. In a study of 93 infants, those with DSC exhibited more profound hypothyroidism and ioduria during the immediate postoperative period compared to infants with primary sternal closure (23). Because the infants with DSC had a longer duration of cardiopulmonary bypass and required more inotropic support, it is unclear whether the hypothyroidism was solely related to topical iodine exposure.

Congenital hypothyroidism due to thyroid dysgenesis is not likely to be a plausible explanation for the thyroid dysfunction described in the infants in this case series because all three infants had normal initial thyroid function at birth, and the rise in TSH was temporally associated with excess iodine exposure and resolved completely in two of the three infants. An increased incidence of severe congenital hypothyroidism as well as “atypical hypothyroidism” with delayed thyrotropin elevation has been recognized in patients with CHD (24, 25). The relative contribution of excess iodine to this phenomenon of “atypical hypothyroidism” has not been previously explored, and our study suggests that this may be more common than previously appreciated. The one infant in this series with persistent hypothyroidism (case 1) was investigated for commonly known genetic causes of both congenital hypothyroidism and CHD, including 22q11.2 deletion (DiGeorge syndrome), but only an unrelated X-chromosome anomaly was found. It is possible that this infant with multiple congenital anomalies may also have a morphologically abnormal thyroid gland (26) that decompensated under the stress of an iodine load.

The significantly elevated TSH levels in our cases are not likely to have been due to recovery from the “euthyroid sick syndrome.” This syndrome, which is well documented in infants and children after cardiac surgery and cardiopulmonary bypass (27, 28), is characterized by a decrease in the serum TSH, FT4, and T₃ concentrations during critical illness, often followed by a subsequent transient rise in serum TSH above normal values, but rarely above 10 mIU/L. In all of our cases, the serum TSH level rose well above 10 mIU/L, coinciding in time with high serum and urine iodine levels after known iodine exposure. Similarly, the timing and degree of TSH elevation in our cases cannot be explained by the initial suppression and mild rebound elevation described in association with dopamine and glucocorticoid exposure (29, 30).

Other factors related to these patients' complex medical conditions and hospital courses, including changes in renal function and in perfusion status after cardiac surgery, may also have contributed to the development of hypothyroidism. In addition, unknown genetic factors may increase susceptibility to thyroid dysfunction in some infants and may interact with various environmental factors to influence the risk and severity of dysfunction.

In conclusion, these cases emphasize the need to monitor thyroid function in infants with CHD exposed to exogenous iodine. Further research is needed to elucidate the incidence as well as the temporal relationship between excess iodine exposure and thyroid dysfunction, the relative contributions of iv vs topical sources of exposure, and other risk factors that may impact the duration and severity of hypothyroidism in this population.