Post by Admin on Oct 20, 2024 0:12:17 GMT -8
West J Med. 2000 February; 172(2): 102–106.
PMCID: PMC1070767
How medications affect thyroid function
Betty J Dong1
Abnormal results of thyroid function tests are common in clinical practice. The diagnosis of thyroid dysfunction is easily made, especially if the clinical signs and symptoms of thyroid dysfunction are also present. In elderly patients aged 65 years and older, who often have atypical presentations, the diagnosis of thyroid dysfunction is more difficult. Sometimes laboratory findings are abnormal because of a patient's use of medications or the presence of unrelated and nonthyroidal medical illnesses. Many abnormalities identified by thyroid laboratory tests can be caused by various illnesses that do not directly involve the thyroid gland. This “euthyroid sick syndrome” occurs in as many as 70% of patients who have been admitted to a hospital.1,2 Recognition is critical because therapy is not necessary and may be detrimental. There results of thyroid function tests often revert to normal once the patient recovers from the illness.
In this review, I focus on medications that interfere with the proper interpretation of thyroid function test results, cause thyroid illnesses, influence levothyroxine requirements, and impair absorption of exogenous levothyroxine.
Go to:
THYROID FUNCTION TESTS
Common laboratory tests used in the assessment and diagnosis of thyroid disorders include measuring the circulating thyroid hormone concentrations, evaluating the integrity of the pituitary negative-feedback system, measuring thyroid antibody concentrations, and evaluating radioactive iodine uptake and scans. Tests to measure circulating thyroid hormone concentrations can include total thyroxine (T4), total triiodothyronine (T3), free thyroxine index, free T4 (FT4), and free T3 (FT3).
Total T4 and total T3 measurements are less accurate because several medications can interfere (for example, estrogen and estrogen-containing birth control pills, tamoxifen citrate, heroin, methadone hydrochloride, fluouracil, mitotane, androgens, anabolic steroids, nicotinic acid, and glucocorticoids) scan medical conditions (such as the nephrotic syndrome) that alter thyroid-binding globulin and confuse results. Measurement of the total T3 level is still frequently used because many laboratories cannot perform FT3 measurements.
The measurement of the free T4 index and FT4level should replace measurements of the total T4 level. The free hormone concentrations also accurately represent the thyroid state.
The integrity of the pituitary negative-feedback system is evaluated by measuring levels of thyrotropin (the thyroid-stimulating hormone). This is the most sensitive test for screening, diagnosis, and monitoring of thyroid dysfunction because thyrotropin levels may become abnormal before corresponding changes occur in the circulating free thyroid hormone levels. The diagnosis of primary hypothyroidism is confirmed by raised concentrations of thyrotropin and subnormal FT4 concentrations. Similarly, findings of an undetectable concentration of thyrotropin and raised FT4 concentration are diagnostic of hyperthyroidism. In subclinical thyroid disease, however, the FT4 concentration remains normal whereas in subclinical hypothyroidism, the thyrotropin concentration is raised and, in subclinical hyperthyroidism, it is suppressed below normal.
The measurement of the total or FT3 level is useful to document possible hyperthyroidism, especially if FT4 levels are normal and the thyrotropin level is suppressed. Furthermore, the T3 level should not be monitored in hypothyroidism because many medications and illnesses that do not involve the thyroid can block peripheral conversion of T4 to T3 and produce a low concentration of T3. In addition, T3 concentrations can be normal in hypothyroidism.
Go to:
MEDICATIONS THAT INTERFERE WITH THYROID FUNCTION TESTING
Dopamine agonists and similar agents (table 1) can acutely suppress thyrotropin levels to lower-than-normal but detectable values.3 In patients with true hyperthyroidism, thyrotropin levels are often undetectable. Amphetamines also transiently increase dopamine release for between 1 and 3 weeks. Although atypical antipsychotic agents (such as quetiapine fumarate) possess dopamine-blocking effects, no changes in thyrotropin levels have been reported. Patients who are taking these medications over long periods do not have sustained reductions in thyrotropin levels, and hyperthyroidism does not develop. Similarly, dopamine antagonists, such as metoclopramide hydrochloride, at doses of greater than 1 mg/kg each, can produce slight elevations in thyrotropin levels but not usually greater than 10 mIU/L.
Table 1
Table 1
Medications that alter thyroid function test results in euthyroid people
Go to:
DISPLACEMENT OF THYROID HORMONES FROM THYROID-BINDING GLOBULIN
Clinicians should be aware of medications (for example, nonsteroidal anti-inflammatory drugs4) that can displace thyroid hormones from thyroid-binding globulin and transiently elevate FT4 and FT3 concentrations and depress thyrotropin levels (table 1). During continued medication administration, however, FT4, FT3, and thyrotropin levels return to normal.
The use of heparin increases lipoprotein lipase activity and produces a fivefold increase in FT4 levels because T4 is displaced by free fatty acids. Therefore, to avoid laboratory interference with the test results, FT4 levels should be measured 1 hour or more after intravenous administration or 10 hours or more after administering low-molecular-weight heparin.5
A phenomenon that has confused clinicians for decades is that therapeutic levels of phenytoin and carbamazepine produce sustained reductions in T4 and FT3 levels despite normal thyrotropin levels in a clinically euthyroid patient. These paradoxic findings are caused by interference of these agents with the FT4 assay, producing an underestimation of the FT4 concentrations.6 Therefore, clinicians should rely on the thyrotropin level rather than the FT4 level when assessing thyroid status in patients receiving these agents.
Go to:
INHIBITION OF CONVERSION OF T4 TO T3
Most T3 is produced from the peripheral conversion of T4 to T3 by the enzyme 5-deiodinase. Amiodarone and iodinated contrast media (for example, ipodate, iopanoic acid, and tyropanoate) can inhibit the conversion of T4 to T3 both in the peripheral circulation and in the pituitary gland, and can produce confusing thyroid function abnormalities (table 1 and table 2). Within 1 week of ipodate administration, elevations in FT4 levels, reductions in total T3 levels, and a transient rise in thyrotropin levels occur; these return to normal within 2 weeks. Similar findings are observed during the first week of amiodarone administration. A transient elevation in thyrotropin levels (<10 mIU/L) occurs during the first 3 months of amiodarone treatment but during ongoing therapy, elevations in FT4 and reductions in T3 levels persist in patients who are euthyroid. Therefore, normal FT4 and T3 levels in a patient taking amiodarone are highly suggestive of overt thyroid dysfunction.
Table 2
Table 2
Effects of amiodarone on thyroid function
Beta blockers and corticosteroids that only inhibit the peripheral conversion of T4 to T3 interfere minimally with thyroid function test results. Propranolol hydrochloride (>160 mg/d), atenolol, and metoprolol tartrate produce small reductions in total T3 levels. Large doses of corticosteroids (for example, >4 mg dexamethasone) produce reductions in total T3 levels, which are useful in the management of thyroid storm or severe hyperthyroidism.
Go to:
DRUG-INDUCED THYROID DISEASE
Drug-induced thyroid illness is associated with the use of iodides, iodide-containing preparations, lithium, and interferon alpha treatment.7,8,9,10 Routine monitoring of thyroid function tests at baseline and every 3 to 6 months thereafter is recommended in susceptible people (for example, those with thyroid antibodies or euthyroid goiter) receiving these medications.
Iodides
Iodides are hidden in many preparations, including prescription (for example, amiodarone, radiocontrast dyes, povidone iodine, and iodinated glycerol) and nonprescription items (for example, cough and cold preparations, kelp tablets, herbal preparations, and dietary supplements).5,6 Amiodarone contains 37.5% iodine by weight, exposing patients to an iodine load that is at least 100 times the normal daily intake of 0.5 mg. A dietary supplement, Cellasene (Rexall Sundown, Boca Raton, FL) recently promoted to reduce cellulite, was found to contain 930 μg of iodine per recommended dose of three capsules. Thyroid dysfunction has been reported after the vaginal use of povidone iodine.6 As much as 175 mg of iodide can be released from radiographic contrast media.
Iodide-induced hyperthyroidism (Jod-Basedow disease) usually develops within 3 to 8 weeks after an increase in iodide supplementation (for example, after the administration of radiocontrast dye or amiodarone) in persons with autonomously functioning, nontoxic multinodular goiters. Hyperthyroid symptoms can persist for several months, requiring therapy with thioamides and beta blockers. Radioactive iodine is not effective because the iodine loading prevents effective retention of the radioactive iodine. Iodide-induced hyperthyroidism has been attributed to malfunction of the Wolff-Chaikoff block.5 This autoregulatory block normally protects the gland against excessive hormone production and the development of hyperthyroidism once a critical intrathyroidal iodide concentration is achieved.
Iodide-induced hypothyroidism is most common, but goiter and hyperthyroidism also occur. Risk factors for drug-induced hypothyroidism in people who are not taking thyroid supplements are listed in the box below. Women are at greater risk than men, as are those who show thyroid antibodies before starting therapy. Iodide-induced hypothyroidism develops if the gland is unable to “escape” from the Wolff-Chaikoff block after several days to resume normal hormone production. Thyroxine therapy should be administered for at least 6 months before attempts to stop are considered.
Amiodarone is being increasingly used for the management of a number of cardiac conditions, including atrial fibrillation and congestive heart failure. The overall prevalence of amiodarone-induced thyroid disease is estimated to be 2% to 24%.11 Hypothyroidism is more common than hyperthyroidism. No clear relation exists between the development of thyroid dysfunction and either the cumulative dosage or the duration of amiodarone therapy, although most cases develop within the first 2 years of therapy. Rarely, because of amiodarone's long half-life (40-55 days) and sequestration in adipose tissue, thyroid dysfunction can occur the first year after it is stopped.
Some risk factors for drug-induced hypothyroidism in people not taking thyroid supplements
Autoimmune thyroiditis (such as Hashimoto disease)
Previous thyroid disease
Partial thyroidectomy
History of radioactive iodine administration
History of postpartum thyroid disease
Family history of thyroid disease
Previous thyroid damage
Female sex
Preexisting or de novo development of antithyroid antibodies
Hypothyroidism is reported in 6% to 10% of patients receiving amiodarone and may be difficult to recognize because bradycardia and constipation are also common side effects of amiodarone use. Hypothyroidism should be confirmed by thyroid function tests (table 2). Levothyroxine replacement is often necessary even if amiodarone is stopped or the dosage reduced. After 1 year, thyroxine supplementation can be stopped to evaluate the need for continued therapy. Few guidelines are available for regulating thyroxine replacement. A return of the thyrotropin level to normal might not be possible without aggravating the underlying cardiac status. If amiodarone therapy is continued, thyroxine replacement should be maintained indefinitely.
Hyperthyroidism is reported in 1% to 5% of patients. The typical signs and symptoms are often not apparent, and clinical suggestions of hyperthyroidism should be confirmed by laboratory findings (table 2). Cardiac symptoms, weight loss, tremor, sleep disturbances, and myopathy could be attributed to amiodarone use, the existing cardiac disease, or hyperthyroidism. Two types of amiodarone-induced hyperthyroidism have been identified. Type 1 is a Graves disease-like hyperthyroidism, characterized by the presence of antithyroid antibodies. Type 2 is a subacute thyroiditis (no antithyroid antibodies) that is caused by a direct toxic effect on the gland, producing a “dumping” of thyroid hormone into the circulation.7,12 Treatment is complicated, and referral to an endocrinologist is recommended. Treatment is necessary even after amiodarone is stopped because the hyperthyroidism can persist for several months due to amiodarone's long duration of action.
Lithium
Hypothyroidism and subclinical hypothyroidism have been reported in 5% to 20% and as high as 50% of people taking lithium carbonate.3,9,10,13 A smooth, nontender goiter is observed in up to 60% of those receiving lithium for 5 months to 2 years; hypothyroidism may not be present.
As with iodides, lithium is concentrated in the gland and interferes with thyroid hormone synthesis and release, causing a compensatory increase in thyrotropin levels. A twofold increase in the incidence of thyroid antibodies has been found in patients treated with lithium (24%) compared with those not taking lithium (12%). The risk of lithium-induced antibodies increases with the duration of therapy (for example, >2 years) and is more common in women than in men. Antibody titers increased in two thirds of people with thyroid antibodies at baseline.13,14
Lithium-induced hypothyroidism is more frequent in those taking lithium for more than 2 years. Close monitoring of thyrotropin levels, rather than the immediate institution of thyroxine therapy, is reasonable in patients with subclinical hypothyroidism induced by lithium because abnormal thyrotropin levels can return to normal spontaneously. In a review of several prospective studies, a substantial fall in serum hormone levels and a rise in thyrotropin levels within 10 days to a few months of starting therapy were reported.14 The thyroid abnormalities, however, returned to pretreatment levels within the first year, without additional therapy or interruptions or changes in lithium therapy. Thyrotropin levels were less likely to return to normal in people with preexisting antithyroid antibodies.
Levothyroxine supplementation can reverse the hypothyroidism, prevent further growth of an existing goiter, and permit the continued administration of lithium. Once lithium is stopped, the goiter and hypothyroidism do not always resolve. Management with thioamides should be considered, as should the surgical removal of the goiter.
Interferon alpha
Thyroid dysfunction is common after interferon alpha therapy for chemotherapy or long-term treatment of hepatitis C.15,16,17 In contrast, thyroid dysfunction after the administration of interferon beta-1b for the treatment of multiple sclerosis is rare.18
The prevalence of thyroid abnormalities during interferon therapy ranges from 2.5% to 20%.15,16 Symptoms of thyroid dysfunction may be absent or occur as early as 6 to 8 weeks after starting therapy or be delayed until after 6 to 23 months of receiving therapy. Hypothyroidism is more common (40%-50% of patients) than hyperthyroidism (10%-30% of patients). Fortunately, thyroid dysfunction seems to be transient in most patients, and treatment is not always necessary. Levothyroxine treatment is necessary only to alleviate hypothyroid symptoms; hypothyroidism often resolves spontaneously within 2 to 3 months after stopping therapy. Similarly, beta-blockade therapy is only needed for symptomatic hyperthyroidism because the hyperthyroidism is often transient. Thyroid dysfunction may take as long as 17 months after stopping therapy to resolve. Thyroid disease is rarely permanent.
Go to:
MEDICATIONS THAT AFFECT T4 REQUIREMENTS
Thyroid hormones are metabolized primarily by deiodination, but glucuronidation and sulfation are also important routes of elimination. Cytochrome P-450 hepatic enzyme inducers (for example, rifampin, rifabutin, phenytoin, carbamazepine, and phenobarbital) can increase the metabolic elimination of T4 and T3 by 20%, which is not clinically important in people who are euthyroid. Those requiring thyroxine replacement therapy, however, may need higher doses to maintain euthyroidism. Ritonavir, a potent P-450 mixed hepatic enzyme inhibitor and inducer, can increase thyroxine glucuronidation, necessitating a twofold increase in thyroxine dosage to maintain euthyroidism.19
Serotonin reuptake inhibitors may also alter T4 requirements. In nine patients receiving thyroxine therapy, an elevation in thyrotropin levels and a reduction in FT4 levels were noted after the addition of sertraline hydrochloride.20 An increase in thyroxine clearance was thought to have occurred.
Go to:
DRUGS THAT IMPAIR THE ABSORPTION OF EXOGENOUS THYROXINE
Patients should take levothyroxine on an empty stomach for optimal absorption. Several medications, including iron, aluminum-containing products (such as sucralfate, antacids, and didanosine), sodium polystyrene sulfonate, resin binders, and calcium carbonate have been reported to impair the absorption of exogenous thyroxine and decrease its efficacy.21,22,23,24
Not all calcium carbonate preparations have been implicated. Patients should take levothyroxine at least 4 hours before or after taking any medication that might interfere with absorption to minimize this interaction. To maintain euthyroidism, the levothyroxine dosage may need to be adjusted or the offending agent stopped.
Go to:
CONCLUSIONS
The accurate interpretation of abnormal thyroid function test results may be complicated by the concomitant presence of medications and nonthyroidal illnesses. It is important that clinicians recognize the effects of drugs on laboratory interpretation, drug-induced thyroid illnesses, and exogenous thyroid requirements to prevent medical treatments that may be dangerous or that inappropriately increase the cost of caring for patients.
Go to:
Acknowledgments
I would like to thank Andrew Leeds for his review of this manuscript.
Go to:
References
1. DeGroot LJ. Dangerous dogmas in medicine: the nonthyroidal illness syndrome. J Clin Endocrinol Metab 1999;84:151-164. [PubMed]
2. Camacho PM, Dwarkanathan AA. Sick euthyroid syndrome: what to do when thyroid function tests are abnormal in critically ill patients. Postgrad Med 1999;105:215-219. [PubMed]
3. Davies PH, Franklyn JA. The effects of drugs on tests of thyroid function. Eur J Clin Pharmacol 1991;40:439-451. [PubMed]
4. Bishnoi A, Carlson HE, Gruber BL, et al. Effects of commonly prescribed nonsteroidal anti-inflammatory drugs on thyroid hormone measurements. Am J Med 1994;96:235-238. [PubMed]
5. Stevenson HP, Pooler G, Archbold R, et al. Misleading serum free thyroxine results during low molecular weight heparin treatment. Clin Chem 1998;44:1002-1007. [PubMed]
6. Surks MI, DeFesi CR. Normal serum free thyroid hormone concentrations in patients treated with phenytoin or carbamazepine: a paradox resolved. JAMA 1996;275:1495-1498. [PubMed]
7. Stanbury JB, Ermans AE, Bourdoux P, et al. Iodine-induced hyperthyroidism: occurrence and epidemiology. Thyroid 1998;8:83-98. [PubMed]
8. Silva JE. Effects of iodine and iodine-containing compounds on thyroid function. Med Clin North Am 1985;69:881-898. [PubMed]
9. Surks MI, Sievert R. Drugs and thyroid function. N Engl J Med 1995;333:1688-1694. [PubMed]
10. Gittoes NJ, Franklyn JA. Drug-induced thyroid disorders. Drug Saf 1995; 13:46-55. [PubMed]
11. Harjai KJ, Licata AA. Effects of amiodarone on thyroid function. Ann Intern Med 1997;126:63-73. [PubMed]
12. Bartalena L, Brogioni S, Grasso L, et al. Treatment of amiodarone-induced thyrotoxicosis, a difficult challenge: results of a prospective study. J Clin Endocrinol Metab 1996;81:2930-2933. [PubMed]
13. Lazarus JH. The effects of lithium therapy on thyroid and thyrotropin-releasing hormone. Thyroid 1998;8:909-913. [PubMed]
14. Kleiner J, Altshuler L, Hendrick V, et al. Lithium-induced subclinical hypothyroidism: review of the literature and guidelines for treatment. J Clin Psychiatry 1999;60:249-255. [PubMed]
15. Koh LKH, Greenspan FS, Yeo PPB. Interferon-α induced thyroid dysfunction: three clinical presentations and a review of the literature. Thyroid 1997;7:891-896. [PubMed]
16. Amenomori M, Mori T, Fukuda Y, et al. Incidence and characteristics of thyroid dysfunction following interferon therapy in patients with chronic hepatitis C. Intern Med 1998;37:246-252. [PubMed]
17. Schuppert F, Rambusch E, Kirchner H, et al. Patients treated with interferon-α, interferon-β, and interleukin-2 have a different thyroid autoantibody pattern than patients suffering from endogenous autoimmune thyroid disease. Thyroid 1997;7:837-842. [PubMed]
18. Schwid SR, Goodman AD, Mattson DH. Autoimmune hyperthyroidism in patients with multiple sclerosis treated with interferon beta-1b. Arch Neurol 1997;54:1169-1170. [PubMed]
19. Tseng A, Fletcher D. Interaction between ritonavir and levothyroxine [letter]. AIDS 1998;12:2235-2236. [PubMed]
20. McCowen KC, Garber JR, Spark R. Elevated serum thyrotropin in thyroxine-treated patients with hypothyroidism given sertraline [letter]. N Engl J Med 1997;337:1010-1011. [PubMed]
21. Schneyer CR. Calcium carbonate and reduction of levothyroxine efficacy [letter]. JAMA 1998;279:750. [PubMed]
22. Sherman SI, Malecha SE. Absorption and malabsorption of levothyroxine sodium. Am J Ther 1995;2:814-818. [PubMed]
23. Campbell NR, Hasinoff BB, Stalts H, et al. Ferrous sulfate reduces thyroxine efficacy in patients with hypothyroidism. Ann Intern Med 1992;117:1010-1013. [PubMed]
24. Liel Y, Sperber AD, Shany S. Nonspecific intestinal adsorption of levothyroxine by aluminum hydroxide. Am J Med 1994;97:363-365. [PubMed]
PMCID: PMC1070767
How medications affect thyroid function
Betty J Dong1
Abnormal results of thyroid function tests are common in clinical practice. The diagnosis of thyroid dysfunction is easily made, especially if the clinical signs and symptoms of thyroid dysfunction are also present. In elderly patients aged 65 years and older, who often have atypical presentations, the diagnosis of thyroid dysfunction is more difficult. Sometimes laboratory findings are abnormal because of a patient's use of medications or the presence of unrelated and nonthyroidal medical illnesses. Many abnormalities identified by thyroid laboratory tests can be caused by various illnesses that do not directly involve the thyroid gland. This “euthyroid sick syndrome” occurs in as many as 70% of patients who have been admitted to a hospital.1,2 Recognition is critical because therapy is not necessary and may be detrimental. There results of thyroid function tests often revert to normal once the patient recovers from the illness.
In this review, I focus on medications that interfere with the proper interpretation of thyroid function test results, cause thyroid illnesses, influence levothyroxine requirements, and impair absorption of exogenous levothyroxine.
Go to:
THYROID FUNCTION TESTS
Common laboratory tests used in the assessment and diagnosis of thyroid disorders include measuring the circulating thyroid hormone concentrations, evaluating the integrity of the pituitary negative-feedback system, measuring thyroid antibody concentrations, and evaluating radioactive iodine uptake and scans. Tests to measure circulating thyroid hormone concentrations can include total thyroxine (T4), total triiodothyronine (T3), free thyroxine index, free T4 (FT4), and free T3 (FT3).
Total T4 and total T3 measurements are less accurate because several medications can interfere (for example, estrogen and estrogen-containing birth control pills, tamoxifen citrate, heroin, methadone hydrochloride, fluouracil, mitotane, androgens, anabolic steroids, nicotinic acid, and glucocorticoids) scan medical conditions (such as the nephrotic syndrome) that alter thyroid-binding globulin and confuse results. Measurement of the total T3 level is still frequently used because many laboratories cannot perform FT3 measurements.
The measurement of the free T4 index and FT4level should replace measurements of the total T4 level. The free hormone concentrations also accurately represent the thyroid state.
The integrity of the pituitary negative-feedback system is evaluated by measuring levels of thyrotropin (the thyroid-stimulating hormone). This is the most sensitive test for screening, diagnosis, and monitoring of thyroid dysfunction because thyrotropin levels may become abnormal before corresponding changes occur in the circulating free thyroid hormone levels. The diagnosis of primary hypothyroidism is confirmed by raised concentrations of thyrotropin and subnormal FT4 concentrations. Similarly, findings of an undetectable concentration of thyrotropin and raised FT4 concentration are diagnostic of hyperthyroidism. In subclinical thyroid disease, however, the FT4 concentration remains normal whereas in subclinical hypothyroidism, the thyrotropin concentration is raised and, in subclinical hyperthyroidism, it is suppressed below normal.
The measurement of the total or FT3 level is useful to document possible hyperthyroidism, especially if FT4 levels are normal and the thyrotropin level is suppressed. Furthermore, the T3 level should not be monitored in hypothyroidism because many medications and illnesses that do not involve the thyroid can block peripheral conversion of T4 to T3 and produce a low concentration of T3. In addition, T3 concentrations can be normal in hypothyroidism.
Go to:
MEDICATIONS THAT INTERFERE WITH THYROID FUNCTION TESTING
Dopamine agonists and similar agents (table 1) can acutely suppress thyrotropin levels to lower-than-normal but detectable values.3 In patients with true hyperthyroidism, thyrotropin levels are often undetectable. Amphetamines also transiently increase dopamine release for between 1 and 3 weeks. Although atypical antipsychotic agents (such as quetiapine fumarate) possess dopamine-blocking effects, no changes in thyrotropin levels have been reported. Patients who are taking these medications over long periods do not have sustained reductions in thyrotropin levels, and hyperthyroidism does not develop. Similarly, dopamine antagonists, such as metoclopramide hydrochloride, at doses of greater than 1 mg/kg each, can produce slight elevations in thyrotropin levels but not usually greater than 10 mIU/L.
Table 1
Table 1
Medications that alter thyroid function test results in euthyroid people
Go to:
DISPLACEMENT OF THYROID HORMONES FROM THYROID-BINDING GLOBULIN
Clinicians should be aware of medications (for example, nonsteroidal anti-inflammatory drugs4) that can displace thyroid hormones from thyroid-binding globulin and transiently elevate FT4 and FT3 concentrations and depress thyrotropin levels (table 1). During continued medication administration, however, FT4, FT3, and thyrotropin levels return to normal.
The use of heparin increases lipoprotein lipase activity and produces a fivefold increase in FT4 levels because T4 is displaced by free fatty acids. Therefore, to avoid laboratory interference with the test results, FT4 levels should be measured 1 hour or more after intravenous administration or 10 hours or more after administering low-molecular-weight heparin.5
A phenomenon that has confused clinicians for decades is that therapeutic levels of phenytoin and carbamazepine produce sustained reductions in T4 and FT3 levels despite normal thyrotropin levels in a clinically euthyroid patient. These paradoxic findings are caused by interference of these agents with the FT4 assay, producing an underestimation of the FT4 concentrations.6 Therefore, clinicians should rely on the thyrotropin level rather than the FT4 level when assessing thyroid status in patients receiving these agents.
Go to:
INHIBITION OF CONVERSION OF T4 TO T3
Most T3 is produced from the peripheral conversion of T4 to T3 by the enzyme 5-deiodinase. Amiodarone and iodinated contrast media (for example, ipodate, iopanoic acid, and tyropanoate) can inhibit the conversion of T4 to T3 both in the peripheral circulation and in the pituitary gland, and can produce confusing thyroid function abnormalities (table 1 and table 2). Within 1 week of ipodate administration, elevations in FT4 levels, reductions in total T3 levels, and a transient rise in thyrotropin levels occur; these return to normal within 2 weeks. Similar findings are observed during the first week of amiodarone administration. A transient elevation in thyrotropin levels (<10 mIU/L) occurs during the first 3 months of amiodarone treatment but during ongoing therapy, elevations in FT4 and reductions in T3 levels persist in patients who are euthyroid. Therefore, normal FT4 and T3 levels in a patient taking amiodarone are highly suggestive of overt thyroid dysfunction.
Table 2
Table 2
Effects of amiodarone on thyroid function
Beta blockers and corticosteroids that only inhibit the peripheral conversion of T4 to T3 interfere minimally with thyroid function test results. Propranolol hydrochloride (>160 mg/d), atenolol, and metoprolol tartrate produce small reductions in total T3 levels. Large doses of corticosteroids (for example, >4 mg dexamethasone) produce reductions in total T3 levels, which are useful in the management of thyroid storm or severe hyperthyroidism.
Go to:
DRUG-INDUCED THYROID DISEASE
Drug-induced thyroid illness is associated with the use of iodides, iodide-containing preparations, lithium, and interferon alpha treatment.7,8,9,10 Routine monitoring of thyroid function tests at baseline and every 3 to 6 months thereafter is recommended in susceptible people (for example, those with thyroid antibodies or euthyroid goiter) receiving these medications.
Iodides
Iodides are hidden in many preparations, including prescription (for example, amiodarone, radiocontrast dyes, povidone iodine, and iodinated glycerol) and nonprescription items (for example, cough and cold preparations, kelp tablets, herbal preparations, and dietary supplements).5,6 Amiodarone contains 37.5% iodine by weight, exposing patients to an iodine load that is at least 100 times the normal daily intake of 0.5 mg. A dietary supplement, Cellasene (Rexall Sundown, Boca Raton, FL) recently promoted to reduce cellulite, was found to contain 930 μg of iodine per recommended dose of three capsules. Thyroid dysfunction has been reported after the vaginal use of povidone iodine.6 As much as 175 mg of iodide can be released from radiographic contrast media.
Iodide-induced hyperthyroidism (Jod-Basedow disease) usually develops within 3 to 8 weeks after an increase in iodide supplementation (for example, after the administration of radiocontrast dye or amiodarone) in persons with autonomously functioning, nontoxic multinodular goiters. Hyperthyroid symptoms can persist for several months, requiring therapy with thioamides and beta blockers. Radioactive iodine is not effective because the iodine loading prevents effective retention of the radioactive iodine. Iodide-induced hyperthyroidism has been attributed to malfunction of the Wolff-Chaikoff block.5 This autoregulatory block normally protects the gland against excessive hormone production and the development of hyperthyroidism once a critical intrathyroidal iodide concentration is achieved.
Iodide-induced hypothyroidism is most common, but goiter and hyperthyroidism also occur. Risk factors for drug-induced hypothyroidism in people who are not taking thyroid supplements are listed in the box below. Women are at greater risk than men, as are those who show thyroid antibodies before starting therapy. Iodide-induced hypothyroidism develops if the gland is unable to “escape” from the Wolff-Chaikoff block after several days to resume normal hormone production. Thyroxine therapy should be administered for at least 6 months before attempts to stop are considered.
Amiodarone is being increasingly used for the management of a number of cardiac conditions, including atrial fibrillation and congestive heart failure. The overall prevalence of amiodarone-induced thyroid disease is estimated to be 2% to 24%.11 Hypothyroidism is more common than hyperthyroidism. No clear relation exists between the development of thyroid dysfunction and either the cumulative dosage or the duration of amiodarone therapy, although most cases develop within the first 2 years of therapy. Rarely, because of amiodarone's long half-life (40-55 days) and sequestration in adipose tissue, thyroid dysfunction can occur the first year after it is stopped.
Some risk factors for drug-induced hypothyroidism in people not taking thyroid supplements
Autoimmune thyroiditis (such as Hashimoto disease)
Previous thyroid disease
Partial thyroidectomy
History of radioactive iodine administration
History of postpartum thyroid disease
Family history of thyroid disease
Previous thyroid damage
Female sex
Preexisting or de novo development of antithyroid antibodies
Hypothyroidism is reported in 6% to 10% of patients receiving amiodarone and may be difficult to recognize because bradycardia and constipation are also common side effects of amiodarone use. Hypothyroidism should be confirmed by thyroid function tests (table 2). Levothyroxine replacement is often necessary even if amiodarone is stopped or the dosage reduced. After 1 year, thyroxine supplementation can be stopped to evaluate the need for continued therapy. Few guidelines are available for regulating thyroxine replacement. A return of the thyrotropin level to normal might not be possible without aggravating the underlying cardiac status. If amiodarone therapy is continued, thyroxine replacement should be maintained indefinitely.
Hyperthyroidism is reported in 1% to 5% of patients. The typical signs and symptoms are often not apparent, and clinical suggestions of hyperthyroidism should be confirmed by laboratory findings (table 2). Cardiac symptoms, weight loss, tremor, sleep disturbances, and myopathy could be attributed to amiodarone use, the existing cardiac disease, or hyperthyroidism. Two types of amiodarone-induced hyperthyroidism have been identified. Type 1 is a Graves disease-like hyperthyroidism, characterized by the presence of antithyroid antibodies. Type 2 is a subacute thyroiditis (no antithyroid antibodies) that is caused by a direct toxic effect on the gland, producing a “dumping” of thyroid hormone into the circulation.7,12 Treatment is complicated, and referral to an endocrinologist is recommended. Treatment is necessary even after amiodarone is stopped because the hyperthyroidism can persist for several months due to amiodarone's long duration of action.
Lithium
Hypothyroidism and subclinical hypothyroidism have been reported in 5% to 20% and as high as 50% of people taking lithium carbonate.3,9,10,13 A smooth, nontender goiter is observed in up to 60% of those receiving lithium for 5 months to 2 years; hypothyroidism may not be present.
As with iodides, lithium is concentrated in the gland and interferes with thyroid hormone synthesis and release, causing a compensatory increase in thyrotropin levels. A twofold increase in the incidence of thyroid antibodies has been found in patients treated with lithium (24%) compared with those not taking lithium (12%). The risk of lithium-induced antibodies increases with the duration of therapy (for example, >2 years) and is more common in women than in men. Antibody titers increased in two thirds of people with thyroid antibodies at baseline.13,14
Lithium-induced hypothyroidism is more frequent in those taking lithium for more than 2 years. Close monitoring of thyrotropin levels, rather than the immediate institution of thyroxine therapy, is reasonable in patients with subclinical hypothyroidism induced by lithium because abnormal thyrotropin levels can return to normal spontaneously. In a review of several prospective studies, a substantial fall in serum hormone levels and a rise in thyrotropin levels within 10 days to a few months of starting therapy were reported.14 The thyroid abnormalities, however, returned to pretreatment levels within the first year, without additional therapy or interruptions or changes in lithium therapy. Thyrotropin levels were less likely to return to normal in people with preexisting antithyroid antibodies.
Levothyroxine supplementation can reverse the hypothyroidism, prevent further growth of an existing goiter, and permit the continued administration of lithium. Once lithium is stopped, the goiter and hypothyroidism do not always resolve. Management with thioamides should be considered, as should the surgical removal of the goiter.
Interferon alpha
Thyroid dysfunction is common after interferon alpha therapy for chemotherapy or long-term treatment of hepatitis C.15,16,17 In contrast, thyroid dysfunction after the administration of interferon beta-1b for the treatment of multiple sclerosis is rare.18
The prevalence of thyroid abnormalities during interferon therapy ranges from 2.5% to 20%.15,16 Symptoms of thyroid dysfunction may be absent or occur as early as 6 to 8 weeks after starting therapy or be delayed until after 6 to 23 months of receiving therapy. Hypothyroidism is more common (40%-50% of patients) than hyperthyroidism (10%-30% of patients). Fortunately, thyroid dysfunction seems to be transient in most patients, and treatment is not always necessary. Levothyroxine treatment is necessary only to alleviate hypothyroid symptoms; hypothyroidism often resolves spontaneously within 2 to 3 months after stopping therapy. Similarly, beta-blockade therapy is only needed for symptomatic hyperthyroidism because the hyperthyroidism is often transient. Thyroid dysfunction may take as long as 17 months after stopping therapy to resolve. Thyroid disease is rarely permanent.
Go to:
MEDICATIONS THAT AFFECT T4 REQUIREMENTS
Thyroid hormones are metabolized primarily by deiodination, but glucuronidation and sulfation are also important routes of elimination. Cytochrome P-450 hepatic enzyme inducers (for example, rifampin, rifabutin, phenytoin, carbamazepine, and phenobarbital) can increase the metabolic elimination of T4 and T3 by 20%, which is not clinically important in people who are euthyroid. Those requiring thyroxine replacement therapy, however, may need higher doses to maintain euthyroidism. Ritonavir, a potent P-450 mixed hepatic enzyme inhibitor and inducer, can increase thyroxine glucuronidation, necessitating a twofold increase in thyroxine dosage to maintain euthyroidism.19
Serotonin reuptake inhibitors may also alter T4 requirements. In nine patients receiving thyroxine therapy, an elevation in thyrotropin levels and a reduction in FT4 levels were noted after the addition of sertraline hydrochloride.20 An increase in thyroxine clearance was thought to have occurred.
Go to:
DRUGS THAT IMPAIR THE ABSORPTION OF EXOGENOUS THYROXINE
Patients should take levothyroxine on an empty stomach for optimal absorption. Several medications, including iron, aluminum-containing products (such as sucralfate, antacids, and didanosine), sodium polystyrene sulfonate, resin binders, and calcium carbonate have been reported to impair the absorption of exogenous thyroxine and decrease its efficacy.21,22,23,24
Not all calcium carbonate preparations have been implicated. Patients should take levothyroxine at least 4 hours before or after taking any medication that might interfere with absorption to minimize this interaction. To maintain euthyroidism, the levothyroxine dosage may need to be adjusted or the offending agent stopped.
Go to:
CONCLUSIONS
The accurate interpretation of abnormal thyroid function test results may be complicated by the concomitant presence of medications and nonthyroidal illnesses. It is important that clinicians recognize the effects of drugs on laboratory interpretation, drug-induced thyroid illnesses, and exogenous thyroid requirements to prevent medical treatments that may be dangerous or that inappropriately increase the cost of caring for patients.
Go to:
Acknowledgments
I would like to thank Andrew Leeds for his review of this manuscript.
Go to:
References
1. DeGroot LJ. Dangerous dogmas in medicine: the nonthyroidal illness syndrome. J Clin Endocrinol Metab 1999;84:151-164. [PubMed]
2. Camacho PM, Dwarkanathan AA. Sick euthyroid syndrome: what to do when thyroid function tests are abnormal in critically ill patients. Postgrad Med 1999;105:215-219. [PubMed]
3. Davies PH, Franklyn JA. The effects of drugs on tests of thyroid function. Eur J Clin Pharmacol 1991;40:439-451. [PubMed]
4. Bishnoi A, Carlson HE, Gruber BL, et al. Effects of commonly prescribed nonsteroidal anti-inflammatory drugs on thyroid hormone measurements. Am J Med 1994;96:235-238. [PubMed]
5. Stevenson HP, Pooler G, Archbold R, et al. Misleading serum free thyroxine results during low molecular weight heparin treatment. Clin Chem 1998;44:1002-1007. [PubMed]
6. Surks MI, DeFesi CR. Normal serum free thyroid hormone concentrations in patients treated with phenytoin or carbamazepine: a paradox resolved. JAMA 1996;275:1495-1498. [PubMed]
7. Stanbury JB, Ermans AE, Bourdoux P, et al. Iodine-induced hyperthyroidism: occurrence and epidemiology. Thyroid 1998;8:83-98. [PubMed]
8. Silva JE. Effects of iodine and iodine-containing compounds on thyroid function. Med Clin North Am 1985;69:881-898. [PubMed]
9. Surks MI, Sievert R. Drugs and thyroid function. N Engl J Med 1995;333:1688-1694. [PubMed]
10. Gittoes NJ, Franklyn JA. Drug-induced thyroid disorders. Drug Saf 1995; 13:46-55. [PubMed]
11. Harjai KJ, Licata AA. Effects of amiodarone on thyroid function. Ann Intern Med 1997;126:63-73. [PubMed]
12. Bartalena L, Brogioni S, Grasso L, et al. Treatment of amiodarone-induced thyrotoxicosis, a difficult challenge: results of a prospective study. J Clin Endocrinol Metab 1996;81:2930-2933. [PubMed]
13. Lazarus JH. The effects of lithium therapy on thyroid and thyrotropin-releasing hormone. Thyroid 1998;8:909-913. [PubMed]
14. Kleiner J, Altshuler L, Hendrick V, et al. Lithium-induced subclinical hypothyroidism: review of the literature and guidelines for treatment. J Clin Psychiatry 1999;60:249-255. [PubMed]
15. Koh LKH, Greenspan FS, Yeo PPB. Interferon-α induced thyroid dysfunction: three clinical presentations and a review of the literature. Thyroid 1997;7:891-896. [PubMed]
16. Amenomori M, Mori T, Fukuda Y, et al. Incidence and characteristics of thyroid dysfunction following interferon therapy in patients with chronic hepatitis C. Intern Med 1998;37:246-252. [PubMed]
17. Schuppert F, Rambusch E, Kirchner H, et al. Patients treated with interferon-α, interferon-β, and interleukin-2 have a different thyroid autoantibody pattern than patients suffering from endogenous autoimmune thyroid disease. Thyroid 1997;7:837-842. [PubMed]
18. Schwid SR, Goodman AD, Mattson DH. Autoimmune hyperthyroidism in patients with multiple sclerosis treated with interferon beta-1b. Arch Neurol 1997;54:1169-1170. [PubMed]
19. Tseng A, Fletcher D. Interaction between ritonavir and levothyroxine [letter]. AIDS 1998;12:2235-2236. [PubMed]
20. McCowen KC, Garber JR, Spark R. Elevated serum thyrotropin in thyroxine-treated patients with hypothyroidism given sertraline [letter]. N Engl J Med 1997;337:1010-1011. [PubMed]
21. Schneyer CR. Calcium carbonate and reduction of levothyroxine efficacy [letter]. JAMA 1998;279:750. [PubMed]
22. Sherman SI, Malecha SE. Absorption and malabsorption of levothyroxine sodium. Am J Ther 1995;2:814-818. [PubMed]
23. Campbell NR, Hasinoff BB, Stalts H, et al. Ferrous sulfate reduces thyroxine efficacy in patients with hypothyroidism. Ann Intern Med 1992;117:1010-1013. [PubMed]
24. Liel Y, Sperber AD, Shany S. Nonspecific intestinal adsorption of levothyroxine by aluminum hydroxide. Am J Med 1994;97:363-365. [PubMed]