Which thyroid function tests (thyroid‑stimulating hormone, free thyroxine, free triiodothyronine, and thyroid antibodies) should be ordered and how should their results be interpreted?

Thyroid Function Testing: Which Tests to Order and How to Interpret Results

Primary Screening Test: TSH First

TSH is the single most important and sensitive screening test for thyroid dysfunction, with sensitivity above 98% and specificity greater than 92%. 1, 2 This should be your initial test in virtually all cases of suspected thyroid disease.

When TSH Alone Is Sufficient

• For routine screening in asymptomatic patients, TSH alone is adequate 3
• Normal TSH (0.45-4.5 mIU/L) with no symptoms requires no further testing 1
• TSH reference ranges shift upward with age—values up to 7.5 mIU/L may be normal in patients over 80 1

Reflex Testing Algorithm: When to Add Free T4

Measure free T4 (FT4) whenever TSH is abnormal to distinguish subclinical from overt thyroid dysfunction. 1, 2

Interpretation Based on TSH and Free T4 Combinations

High TSH + Low Free T4 = Overt Hypothyroidism

• This confirms primary hypothyroidism requiring immediate levothyroxine treatment 1
• Start treatment without delay to prevent cardiovascular dysfunction and quality of life deterioration 1

High TSH + Normal Free T4 = Subclinical Hypothyroidism

• TSH >10 mIU/L: Treat with levothyroxine regardless of symptoms—carries ~5% annual progression risk to overt disease 1
• TSH 4.5-10 mIU/L: Monitor every 6-12 months; consider treatment only if symptomatic, pregnant, or positive anti-TPO antibodies 1
• Confirm with repeat testing after 3-6 weeks, as 30-60% of elevated TSH values normalize spontaneously 1

Low TSH + High Free T4 = Overt Hyperthyroidism

• Confirms hyperthyroidism requiring treatment 1
• Consider measuring free T3 if T3 thyrotoxicosis suspected 4

Low TSH + Normal Free T4 = Subclinical Hyperthyroidism

• TSH <0.1 mIU/L: Consider treatment, especially if age >60, cardiac disease, or osteoporosis risk 1
• TSH 0.1-0.45 mIU/L: Monitor every 3-12 months; treat if symptomatic or high-risk features present 1
• In patients on levothyroxine, this indicates overtreatment—reduce dose by 12.5-25 mcg 1

Free T3 Testing: Very Limited Utility

Free T3 (FT3) measurement has limited clinical utility and should NOT be routinely ordered. 4

The Only Valid Indication for Free T3

• Order FT3 only when TSH is suppressed (<0.01 mIU/L) AND free T4 is normal or low, to diagnose T3 thyrotoxicosis 4
• This scenario occurs in only 0.5% of cases—making routine FT3 testing wasteful 4
• T3 thyrotoxicosis is more common in outpatient settings (34%) than inpatient (14%) when TSH <0.01 mIU/L 4

Why Free T3 Is Not Useful in Most Situations

• 80% of circulating T3 comes from peripheral T4 conversion, not thyroid secretion 5
• T3 levels are heavily influenced by non-thyroidal illness, medications, and nutritional status 5, 6
• In hypothyroidism, FT3 may remain normal even when TSH is elevated and FT4 is low 7
• FT3 does not add diagnostic information for patients on levothyroxine replacement 1

Thyroid Antibody Testing: When and Which Ones

Measure anti-thyroid peroxidase antibodies (anti-TPO) to confirm autoimmune etiology and predict progression risk. 2

When to Order Anti-TPO Antibodies

• Subclinical hypothyroidism (TSH 4.5-10 mIU/L with normal FT4) to guide treatment decisions 1, 2
• Positive anti-TPO predicts 4.3% annual progression to overt hypothyroidism vs 2.6% in antibody-negative patients 1
• Suspected Hashimoto's thyroiditis—anti-TPO is the most sensitive serological marker 2

Thyroglobulin Antibodies (TgAb)

• Add TgAb measurement for comprehensive evaluation of Hashimoto's thyroiditis 2
• Some patients with Hashimoto's have positive TgAb but negative anti-TPO 2

When Antibodies Are NOT Helpful

• Do not routinely measure antibodies in overt hypothyroidism—the diagnosis is already established by TSH and FT4 1
• Antibody levels do not guide treatment decisions or dosing 1

Critical Pitfalls to Avoid

Don't Treat Based on Single Abnormal Values

• 30-60% of mildly abnormal TSH levels normalize spontaneously on repeat testing 1
• Always confirm with repeat TSH and free T4 after 3-6 weeks before initiating treatment 1

Don't Order Total T4 or Total T3

• Total hormone measurements are influenced by binding protein variations (TBG, albumin, transthyretin) 7
• Free hormone measurements (FT4, FT3) are superior because they reflect biologically active hormone 7, 8
• Total hormones remain useful only for research or severe hyperthyroidism 5

Recognize When TSH Is Unreliable

• Acute illness, hospitalization, and recovery from severe illness transiently suppress TSH 1
• Recent iodine exposure (CT contrast) can temporarily affect thyroid function 1
• In central hypothyroidism (pituitary/hypothalamic disease), TSH may be low or inappropriately normal despite low FT4 1
• During the first months of thyroid treatment or dose changes, TSH lags behind—use FT4 to assess adequacy 5

Age-Adjusted Reference Ranges Matter

• 12% of persons aged 80+ with no thyroid disease have TSH >4.5 mIU/L 1
• Consider age-adjusted upper limits: up to 7.5 mIU/L may be normal in elderly patients 1
• Avoid overdiagnosis and overtreatment in asymptomatic elderly individuals 3

Special Populations Requiring Modified Approach

Pregnancy and Preconception

• Target TSH <2.5 mIU/L in first trimester; treat any TSH elevation immediately 1
• Untreated hypothyroidism increases risk of preeclampsia, low birth weight, and neurodevelopmental deficits 1
• Check TSH every 4 weeks during pregnancy after dose stabilization 1

Patients on Levothyroxine

• Monitor TSH every 6-8 weeks during dose titration 1
• Once stable, check TSH annually or when symptoms change 1
• If TSH is suppressed (<0.1 mIU/L), reduce dose immediately—25% of patients are unintentionally overtreated 1
• Overtreatment increases risk of atrial fibrillation (3-5 fold), osteoporosis, fractures, and cardiovascular mortality 1

Patients on Immune Checkpoint Inhibitors

• Thyroid dysfunction occurs in 6-9% with anti-PD-1/PD-L1 therapy, 16% with combination immunotherapy 1
• Monitor TSH every 4-6 weeks for first 3 months, then every second cycle 1
• Consider treatment even for subclinical hypothyroidism if fatigue or symptoms present 1

Suspected Central Hypothyroidism

• Always measure free T4 alongside TSH—TSH may be low or inappropriately normal 1
• Rule out adrenal insufficiency before starting levothyroxine to prevent adrenal crisis 1
• Check morning cortisol and ACTH if central hypothyroidism suspected 1

Evidence Quality Considerations

The evidence supporting routine screening for thyroid dysfunction in asymptomatic adults is insufficient. 3

• USPSTF found inadequate evidence that screening improves quality of life, cardiovascular outcomes, or mortality 3
• Screening leads to frequent false-positives, psychological effects of labeling, and substantial overdiagnosis 3
• Aggressive case finding is appropriate in pregnant women, women >60 years, and high-risk individuals 3

For treatment decisions in subclinical hypothyroidism:

• Evidence for treating TSH >10 mIU/L is rated "fair" by expert panels 3, 1
• Evidence for treating TSH 4.5-10 mIU/L is rated "insufficient"—randomized trials show no symptom improvement 1