August 2009
|
Subclinical Thyroid Dysfunction: To Treat or Not to Treat, That Is the Question
By Saleh A. Aldasouqi, MD, FACE, FACP; and Ved V. Gossain, MD, FACE, FACP
There has been an explosion in the literature about subclinical thyroid dysfunction at both ends of the spectrum, that is, subclinical hypothyroidism (SC-hypo) and subclinical hyperthyroidism (SC-hyper)—and a sharp increase in the diagnosis of these conditions has occurred.1 These developments have been attributed to refinements in thyroid-stimulating hormone (TSH) measurements and ubiquitous lab testing during the past few decades, and all of this has provoked the inevitable debates about screening and treatment of the subclinical thyroid dysfunction.
Subclinical thyroid dysfunction is defined by an abnormal circulating TSH and normal levels of thyroid hormones,1-6 and it is essentially a laboratory diagnosis.6 In other words, although this definition does not take symptomatology into account, most patients are asymptomatic. Thyroid homeostasis is meticulously regulated through the hypothalamic-pituitary-thyroid feedback axis. The more advanced stages of thyroid dysfunction are characterized by abnormal thyroid hormone levels, being below normal in overt hypothyroidism (O-hypo) and above normal in overt hyperthyroidism (O-hyper).
Subclinical thyroid dysfunction is one of the most controversial issues in modern endocrinology. Debate about the need for screening and treatment of this borderline, sometimes referred to as mild, subclinical thyroid disorder has been the subject of ongoing dialogue.1-6 This article discusses these controversies as applied to SC-hypo and SC-hyper.
THYROID FUNCTION TESTS
Among the currently available thyroid function tests (TFTs), serum TSH is the best initial test for thyroid dysfunction.7 TSH has been considered by the American Thyroid Association (ATA) as the single most reliable test to diagnose all forms of hypo- and hyperthyroidism.8 Unless pituitary dysfunction is suspected, typically a normal TSH level indicates normal thyroid function. Aside from pituitary dysfunction, it is prudent to emphasize that physicians should be aware of a few other caveats in the pursuit of correct interpretation of thyroid function tests, as discussed later.
If TSH level is suppressed, hyperthyroidism is suspected, and measurement of tetraiodothyronine (T4) is obtained to confirm the presence of hyperthyroidism. In the presence of a suppressed TSH and normal free T4, the level of free triiodothyronine (T3) should be obtained to exclude T3 toxicosis. If both T4 and T3 levels are normal and TSH is suppressed, a diagnosis of SC-hyper is made. If the TSH level is elevated, free T4 should be obtained. Most often, this will be in the low-normal range. T3 is not necessary in the workup of most cases of SC-hypo; furthermore, T3 may be normal or high-normal, depending on the stage of the disease, and therefore T3 could be misleading.
It should be emphasized that more than 99% of both T4 and T3 is bound to thyroid-binding proteins, and only the free component is metabolically active. Therefore, it is recommended that free hormones should be analyzed—rather than total—unless there is a concern about availability or cost. Free hormone assays are overall more expensive than total assays, but with the availability of different assays, the cost issue may vary from one laboratory to another. Furthermore, some laboratories, especially in remote areas, may not provide free assays. Therefore, physicians can evaluate the availability and cost situation within their local laboratories when ordering thyroid hormone assays.
SUBCLINICAL HYPOTHYROIDISM
Adult hypothyroidism was described as early as the late 19th century.9 Although SC-hypo is a recently introduced entity, a photographic depiction of progression from mild to O-hypo (Figure 1) was published in 1888.10 The first picture of the patient may represent the earliest stage of hypothyroidism (ie, SC-hypo). In strict terms, SC-hypo is defined as elevated TSH with normal thyroid hormones.1-6,9,11,12 SC-hypo is not uncommon, particularly in the elderly population, with prevalence rates in adults of 5% to 17%8 and 16% to 21% in elderly patients.11
Although SC-hypo is defined by elevated TSH, the "normal" level of TSH is also controversial. Most laboratories report the normal range of TSH as 0.3 to 5.5 µU/L, but some investigators13 have recently suggested a lower upper limit of normal for TSH. We prefer to avoid discussion of this issue since it has not yet been validated.
It is notable in this regard, that the symptoms usually attributable to hypothyroidism (eg, fatigue, weight gain, constipation, depression, etc.) are nonspecific and variable, and therefore, the accuracy of clinical diagnosis of hypothyroidism in general is limited.8 Regardless of the issue of symptomatology, once a patient is diagnosed with SC-hypo, as defined above, the next question is whether or not to treat this patient (with thyroid replacement). This controversial issue has triggered endless debates and is still unsettled.1-6,11-13 Etiologies of SC-hypo are essentially the same etiologies of O-hypo, with autoimmune thyroid disease being the most common.
To establish the diagnosis of SC-hypo, a few caveats in regard to elevated TSH should be noted. Before assuming a diagnosis of SC-hypo in patients with elevated TSH, one should exclude the following conditions: TSH-producing adenomas; resistance to thyroid hormone; artifactual interference by heterophil antibodies and TSH isoforms; pulsatile TSH secretion; medications (especially amiodarone, which may reset the TSH set point); mutations in TSH receptors; recovering thyroiditis; sick euthyroid disease; and depression.4,11,13,14
To ascertain the need to treat and screen for SC-hypo, debate among investigators focused on (1) the effects of SC-hypo on a host of various pathophysiological consequences of SC-hypo and whether or not these are reversed by thyroid replacement and (2) on the progression of SC-hypo to O-hypo. The first issue of the debate, regarding clinical and laboratory manifestations of SC-hypo, is summarized in Table 1.9 It has been concluded from a large number of relevant studies that there is no consensus regarding the clinical significance of these clinical and laboratory abnormalities, and regarding benefits of thyroid replacement.9
Progression of SC-hypo to overt hypothyroidism also remains controversial. Several longitudinal studies have indicated a rate of progression of approximately 5% per year.5,11,13 Others have reported the rate of conversion from SC-hypo to O-hypo ranges from 7.8% to 17.8%,9 and this is proportionate to thyroid antibody status. The 10-year probability of this conversion at a baseline TSH of 12 mU/L is 27% and 59% in antibody negative and antibody positive patients, respectively, while the chance of normalization of TSH at 1 year is only 5.5%.9
There seems to be agreement about treating SC-hypo when TSH is >10 µU/L, regardless of antibody status, but the controversy is ongoing below this threshold. Cooper11 proposed a useful algorithm, reproduced in Figure 2. There is an emphasis to have a lower threshold for screening for and treating any degree of hypothyroidism in pregnant women.11 Not only that, but it is recommended that women contemplating conception be screened with TSH, and treated to normalize TSH, prior to conception. These measures are to ensure healthy mental and physical development of the baby.
Although the treatment for SC-hypo is essentially the same as that for O-hypo, the urgency to start thyroxine and the starting dose may not be equivalent. It is prudent to emphasize the generally common phrase "start low and go slow" when treating hypothyroid patients, especially for those who are elderly or with cardiac problems.
SC-HYPERTHYROIDISM
SC-hyper is defined as a suppressed TSH with normal T3 and T4.14-16 Unlike SC-hypo, SC-hyper is less controversial, and also the clinical and pathophysiological areas of controversy are less ramified. The main concerns about untreated SC-hyper are the effects of excessive thyroid hormones on bone density and the heart,4,14 most notably the risk of atrial fibrillation (AF), especially when the TSH level is <0.1 µU/L, as compared with a TSH level of 0.1 to 0.45 µU/L.4
SC-hyper may result from endogenous or exogenous thyroid excess, and the etiologies are essentially the same as those of O-hyper, the most common being Graves' disease and toxic goiter/adenomas. It is prudent to keep in mind a few caveats in interpreting thyroid function tests when faced with suppressed TSH. These include:14 central hypothyroidism, sick euthyroid disease, and medications (steroids and dopamine).
SC-hyper is less prevalent than SC-hypo, reported at below 2% overall.4 The prevalence increases with age, at about 6% in the elderly.16 SC-hyper tends to be more common in the elderly,15-17 and in this age group that the disorder has drawn attention and controversy given the higher risks of osteoporosis and AF. The rate of conversion from SC-hyper to O-hyper is 5% per year.14
The association of SC-hyper and AF has been well documented. In a prospective study with a follow-up of 10 years, the cumulative incidence of AF was 28% with a serum TSH level of <0.1 µU/L.17 A recent large prospective study of an elderly population confirmed this risk (adjusted hazard ratio of 1.98), although there was no increased risk of resulting strokes, or mortality or risk of other cardiovascular complications.18 Obviously, because more studies may be needed before this mortality issue is settled, caution should be taken in determining whether to treat SC-hyper based on AF-related mortality.
Regarding osteoporosis, studies have shown reduced bone mineral density in women with SC-hyper.19,20 Furthermore, treatment of SC-hyper has resulted in improvement of bone density in few recent studies.16,20,21 Fracture risk is currently not known, and so similar to the case with AF, one can understand the debate about whether or not to treat SC-hyper in regard to the risk of bone loss. Greenland et al recently showed improved weight and muscle mass upon treating elderly women with SC-hyper,16 but the real patient benefits were unclear.
Treatment SC-hyper is essentially the same as for O-hyper, including antithyroid medications, surgery, and radioiodine, as appropriate.
THE DEBATE CONTINUES
Given the ongoing debate, a few years ago and almost simultaneously, expert panels (comprising the American Association of Clinical Endocrinologists [AACE], the ATA, The Endocrine Society [TES],4,5 and the US Preventive Services Task Force [USPSTF]) published guidelines on the issue of screening for and treating subclinical thyroid dysfunction.4,5,22
The panels recommended against general screening, as well as treatment of subclinical thyroid disease, and limited that to special situations, leaving the final decision to the discretion of treating physicians—citing lack of evidence for benefits, and risks of treatment.
This triggered a debating statement by clinicians representing AACE, ATA, and TES2 advocating screening and treatment, citing that a lack of evidence for benefit (of treatment) does not mean evidence for a lack of benefit. This yet triggered a response by the panel,3 and the debate continues!
Ringer and Mazzaferri1 commented on the aforementioned debate in an article, "Can there be a consensus about the consensus?" by asserting that this debate is an example of how differently clinicians2 and epidemiologists3 look at this issue. Ringer and Mazzaferri concluded their editorial writing:1 "This dilemma is a clarion for further clinical research that fulfills the epidemiologists' high standards: prospective long-term randomized studies aimed at evaluating the efficacy of treatment of subclinical thyroid dysfunction. Until this occurs, we will continue to vigorously debate this issue."
CLINICAL IMPLICATIONS
Given the controversies about SC-hypo and SC-hyper, we recommend the following guidelines:
- TSH is the single most reliable screening test for thyroid dysfunction.
- Caveats about interpreting TSH results should be kept in mind, and should always be anlyzed within the clinical context of each case. Most important are pituitary dysfunction and medications known to interfere with thyroid function tests.
- Treatment of SC-hypo when TSH is >10 µIU/mL.
- Consider treatment of SC-hypo when TSH is 5 to 10 µIU/mL, in the presence of goiter, or positive thyroid-specific antibodies.
- For patients with TSH 5 to 10 µIU/mL, the option to treat is left to an informed discussion between the patient and his/her physician.
- Active screening and treatment of SC-hypo in pregnant and preconceptual women.
- Remember the rule of thyroid replacement in the elderly and in the presence of cardiac disease: "Start low and go slow."
- Although it will remain controversial, the trend regarding SC-hyper is probably leaning toward treatment, especially in the elderly, because the data regarding morbidity and mortality from ensuing (potential) AF are still unsettled.
CONCLUSION
Subclinical thyroid dysfunction and treatment of these disorders if detected, remains controversial. Aggressive case finding is appropriate in special situations, such as pregnant women, the elderly, and others at high risk for thyroid dysfunction. We believe that physicians' clinical judgment and active participation of patients can achieve an informed and effective approach regarding screening and treatment of suclinical thyroid dysfunction.
Saleh A. Aldasouqi, MD, FACE, FACP, is Clinical Associate Professor, Michigan State University College of Human Medicine, and in the Department of Medicine, Saint Francis Medical Center, Cape Girardeau, Mo. He may be reached at saldasouqi@pol.net; phone: 573-339-8718; or fax: 573-339-9543.
Ved V. Gossain, MD, FACE, FACP, is the Swartz Professor of Medicine and Chief of Endocrinology in the Department of Medicine, Michigan State University College of Human Medicine. He may be reached at ved.gossain@ht.msu.edu; phone: 517-353-3730; or fax: 517-353-1326.
The authors thank Faith Crutscheon (Saint Francis Medical Center Medical Library, Cape Girardeau, Mo.) for assistance in references retrieval. We title for this paper was borrowed from an article published in 1995 by Dr. Jayne Franklyn,23 "Subclinical hypothyroidism: To treat or not to treat, that is the question."
|
