EVIDENCE-BASED ANSWER
Thyroid-stimulating hormone (TSH) level is the preferred test for initial evaluation of suspected primary hypothyroidism (strength of recommendation [SOR]: C, expert opinion). If TSH is abnormal, a free thyroxine (T4) level will further narrow the diagnosis. Obtain a triiodothyronine (T3) level if TSH is undetectable and free T4 is normal.
When assessing the adequacy of replacement therapy in primary hypothyroidism, the TSH is the most important parameter to monitor (SOR: C, expert opinion). Because TSH levels can’t be used to monitor central hypothyroidism, use free T4 and T3 concentrations (SOR: C, case series).
Clinical commentary
A reasonable approach, yes, but more data are needed
David Schmitz, MD
Rural Director, Family Medicine Residency of Idaho, Inc., Boise
In my practice, some patients ask for more testing than necessary, whereas others can’t afford indicated interval lab tests. Ordering unnecessary screening tests or a batched thyroid panel is, too often, a simple but inappropriate clinician response.
Unfortunately, we must rely solely on expert opinion to guide laboratory testing for hypothyroidism. Nevertheless, the guidelines described in this Clinical Inquiry provide not only an appropriate algorithm for diagnosis, but also a logical basis on which to justify ongoing monitoring intervals.
Inconsistent use of laboratory testing among health care providers can lead to misdiagnosis, inappropriate changes in treatment, patient confusion, and added cost. Given the high incidence of hypothyroidism, evidence-based decision making could help avoid unnecessary testing and wasteful expenditure.
Evidence summary
Hypothyroidism is a common condition, affecting 4.6% of the population in the United States, according to the National Health and Nutrition Examination Survey (NHANES III).1 A statewide study in Colorado found the prevalence of elevated TSH levels to be 9.5%.2 The study population was older and had more women, Caucasians, and high school and college graduates than the general population. Among the general population, the prevalence of unsuspected overt hypothyroidism has been reported to be 0 to 18 cases per 1000.3
No randomized controlled trials or other high-quality studies have addressed the question of what laboratory tests are most useful to diagnose and monitor the treatment of hypothyroidism.
TSH is a cost-effective initial test, but has limitations
Experts recommend TSH level as the most cost-effective initial laboratory test for suspected primary hypothyroidism.4 TSH had a high sensitivity (98%) and specificity (92%) when used to confirm thyroid disease in patients referred to a specialty endocrine clinic, but its positive predictive value is low when used as a screening test in primary care.5
TSH is a poor measure of the clinical severity of hypothyroidism. In one study, no correlation was found between serum TSH and clinical and metabolic markers—such as clinical score, ankle reflex time, total cholesterol, and creatine kinase—when estimating the severity of primary thyroid failure.6