Screening is "the application of a test to detect a potential disease or condition in a person who has no known signs or symptoms of that condition at the time the test is done" [1]. Screening with thyroid function tests can identify clinically inapparent subclinical thyroid dysfunction. Subclinical hypothyroidism is diagnosed when a patient has an elevated thyroid-stimulating hormone (TSH) level as determined by a sensitive TSH test and a normal thyroxine level. Subclinical hyperthyroidism is diagnosed when a patient has an undetectable TSH level and a normal thyroxine level. Screening also detects overt thyroid dysfunction, which is diagnosed when a patient has an elevated TSH level and a low thyroxine level (overt hypothyroidism) or an undetectable TSH level and an elevated thyroxine level (overt hyperthyroidism).

In 1990, the American College of Physicians published guidelines on screening for thyroid disease [2, 3]. These guidelines recommended the use of thyroid function tests to identify symptomatic but clinically unrecognized overt hypothyroidism and thyrotoxicosis in women older than 50 years of age. Since 1990, studies of screening have confirmed that even in clinic patients who receive regular care, screening can identify symptomatic overt hypothyroidism or hyperthyroidism that was not suspected in the usual course of care.

The 1990 guidelines recommended against screening for subclinical thyroid failure in the general population because evidence that such patients benefit from earlier detection and treatment was insufficient. Subclinical hypothyroidism (also called mild thyroid failure) may be associated with nonspecific symptoms, hypercholesterolemia, and progression to overt hypothyroidism. Studies performed since 1990 have added to the ability to estimate the risk for these complications in relation to age, sex, and TSH level. However, screening remains controversial because the results of randomized trials of treatment in symptomatic patients have been inconclusive and because asymptomatic persons have not been shown to benefit from treatment [5.9-5.20].

To help providers assess the potential benefits and harms of screening, we focused on screening with thyroid function tests among patients who come to physicians for unrelated reasons, an approach called case finding. We reviewed studies of screening for thyroid dysfunction and studies of the effectiveness of early treatment for subclinical thyroid dysfunction. We make recommendations for each of the four conditions detected by screening, summarize the rationale for each recommendation, and describe the strength of the evidence underlying the rationale. In the accompanying background paper, we present detailed results of the literature review and estimates of the potential benefits of screening in different groups of patients. We also provide more detailed recommendations for screening in the primary care setting, including specific suggestions for interpreting and acting on abnormal test results.

Recommendations

Top Recommendations Author & Article Info References

1. It is reasonable to screen women older than 50 years of age for unsuspected but symptomatic thyroid disease. The preferred screening method is a sensitive TSH test. A free thyroxine test should be done when the TSH level is undetectable or is 10 mU/L or more. Patients who have an undetectable TSH level and an elevated free thyroxine level have overt hyperthyroidism. Patients who have a TSH level higher than 10 mU/L and a low free thyroxine level have overt hypothyroidism. Patients with either of these conditions are likely to benefit from appropriate treatment.

Well-designed prospective clinical and epidemiologic studies have found that 1 in 71 women older than 50 years of age has unsuspected but symptomatic overt hypothyroidism or overt hyperthyroidism that will respond to treatment [3.3]. Screening in women younger than 50 years of age and in men is not warranted because of the low prevalence of unsuspected but symptomatic thyroid dysfunction among these persons [3.4].

2. The treatment of patients found by screening to have persistent subclinical hyperthyroidism has not been studied. Persistent subclinical hyperthyroidism is associated with an increased risk for atrial fibrillation and osteoporosis. Patients who are found to have relatively specific symptoms and signs (such as goiter, nodule, eye findings of Graves disease, or tremor) should be referred to an endocrinologist for consideration of treatment. The management of patients without clinical findings is not clear because most of these persons remain healthy. Information about factors that could help predict which patients are likely to develop complications is needed to develop a practical strategy for managing this condition.

3. The available evidence is not sufficient to recommend for or against treatment of subclinical hypothyroidism. Potential complications of subclinical hypothyroidism are reversible symptoms, hypercholesterolemia, and progression to overt hypothyroidism. Results from randomized trials of treatment to relieve symptoms have been equivocal, and these trials have been too small to show statistically significant effects on elevated lipid levels [5.9-5.17]. Larger, well-designed randomized trials are needed to determine whether treatment would be effective in clinic-based screening of otherwise healthy primary care patients.

Among persons with subclinical hypothyroidism, women older than 50 years of age with markedly elevated TSH levels ( 10 mU/L) have the highest risk for complications. Either treatment or observation is a reasonable approach in this group. One option is to treat patients who have symptoms that may be caused by hypothyroidism with the understanding that only about one in four will benefit and that follow-up evaluation will be needed to decide whether to discontinue therapy in patients whose condition does not improve. Small, sometimes poorly controlled observational studies suggest that, on average, treatment with L-thyroxine may reduce serum cholesterol by 8% in women who have a TSH level of 10 mU/L or more and an elevated serum cholesterol concentration ( 6.2 mmol/L) [5.31]. It is reasonable to offer a trial of L-thyroxine therapy to patients who have these findings. The efficacy of treating patients with subclinical hypothyroidism to prevent progression to overt hypothyroidism is not known. However, physicians who believe that significant morbidity accompanies progression to overt hypothyroidism may wish to prescribe L-thyroxine to all asymptomatic patients who have a TSH level of 10 mU/L or more.

Other groups, including younger women, men, and patients with a mildly elevated TSH level (6 to 9 mU/L), have a lower risk for complications [5.6, 5.21-5.24]. In these groups, no strong evidence indicates that treatment is effective or ineffective in relieving symptoms. The question can only be resolved by a large controlled trial of treatment in persons identified by screening to have mild thyroid failure. Quality of life, response to treatment, and rates of adverse effects among these patients must be known before wider use of L-thyroxine is recommended.

Patients who have a mildly elevated TSH level present a challenge to the physician. Concerns about the medicolegal or ethical consequences of failing to act on abnormal test results on the one hand and the uncertain benefit of targeted follow-up or treatment on the other may lead some physicians not to screen at all. There is no clear answer to this dilemma, but clinicians might consider a trial of therapy in selected, more symptomatic patients and a specified plan for follow-up in the remaining patients. Follow-up could be done by taking a history, performing a physical examination, and repeating thyroid function tests at regular intervals. How frequently this evaluation should be done has not been established. Follow-up every year or every 2 years is probably not necessary, because otherwise healthy persons who have a mildly elevated TSH level seldom progress within 2 years to the point of being identified by screening. On the basis of this observation, follow-up every 2 to 5 years may be adequate.

4. For screening to be cost-effective, the laboratory, clinic, or office should be equipped to identify patients who meet age and sex criteria for screening and to perform appropriate follow-up tests on the same serum sample without requiring a second phlebotomy. In particular, the clinic must follow patients who have overt but clinically unrecognized disease because these patients derive the most definite benefit from treatment. These patients, who have an elevated TSH level and a low free thyroxine level or an undetectable TSH level and an elevated free thyroxine level, can be identified by the laboratory, and special effort can be made to trace them for follow-up.

A formal system to identify these patients and arrange for follow-up evaluation should be part of the screening program. Published reports describe carefully organized programs that have used efficient, reliable methods to identify candidates for screening and patients who require follow-up and treatment [3]. Inefficient or haphazard identification of candidates for screening and inappropriate or incomplete responses to a screening test can reduce the effectiveness of screening in actual practice. The results of published studies apply best to health systems that use an organized, systematic approach to identifying and following patients.