Normal TSH Levels
For non-pregnant adults, the normal TSH reference range is 0.45 to 4.5 mIU/L, though this increases with age, particularly in older adults where the upper limit extends to approximately 5.8 mIU/L for ages 60-79 and 6.7 mIU/L for those ≥80 years 1, 2.
Standard Adult Reference Range
The most authoritative guideline establishes the normal TSH range as 0.45 to 4.5 mIU/L for the general adult population 1. This comes from the NHANES III study analyzing over 13,000 disease-free individuals after excluding those with thyroid antibodies, medications affecting thyroid function, or existing thyroid disease. The geometric mean TSH in healthy adults is 1.4 mIU/L 1.
Important Caveat on Upper Limit Controversy
Some experts advocate for a lower upper limit of 2.5 mIU/L, arguing that values between 2.5-4.5 mIU/L may represent early hypothyroidism 1. However, the consensus guideline rejects this lower threshold because:
- No evidence exists for adverse consequences in the 2.5-4.5 mIU/L range
- This would misclassify many normal individuals as hypothyroid
- Technical assay variations can produce values in this range 1
Age-Specific Variations in Older Adults
TSH physiologically increases with advancing age, and applying standard adult ranges to elderly patients causes overdiagnosis of subclinical hypothyroidism 2, 3.
Recommended Age-Adjusted Ranges:
- Ages 20-59 years: 0.4-4.3 mIU/L
- Ages 60-79 years: 0.4-5.8 mIU/L
- Ages ≥80 years: 0.4-6.7 mIU/L 2
Recent longitudinal data over 22 years confirms that mean TSH increases modestly in older men (by 0.13 mIU/L over 22 years, and 0.5 mIU/L in men ≥70 years), while women show minimal age-related change 3. The TSH distribution widens significantly at older ages in both sexes 3.
Clinical Impact: Using standard adult ranges without age adjustment would misdiagnose 6.5% of individuals aged 60-79 and 12.5% of those ≥80 years as having elevated TSH when they are actually normal for their age 2.
Older Adults and Metabolic Risk
A critical finding is that only moderate-to-severe subclinical hypothyroidism (TSH ≥7.0 mIU/L) shows associations with adverse outcomes like cerebral microbleeds and sarcopenia in older adults, while mild elevations (TSH 4.23-7.0 mIU/L) show no significant associations with geriatric conditions 4. This supports using higher upper limits in elderly populations.
Pregnancy-Specific Reference Ranges
During the first trimester of pregnancy, the normal TSH range is substantially lower: 0.1-2.5 mIU/L 5. This reduction occurs due to the thyrotropic activity of elevated human chorionic gonadotropin (hCG) 5.
Trimester-Specific Considerations:
First Trimester (11-14 weeks):
- Singleton pregnancies: 0.16-3.43 mIU/L 6
- Twin pregnancies: 0.02-2.95 mIU/L (significantly lower than singletons) 6
The standard non-pregnant range of 0.4-4.0 mIU/L is inappropriate during pregnancy and will miss women with slight TSH elevations or falsely diagnose hyperthyroidism in normal pregnant women with physiologically suppressed TSH 5.
Clinical Significance in Pregnancy:
Evidence demonstrates that TSH >2.5 mIU/L in the first trimester increases spontaneous abortion risk 3.7-fold compared to TSH 0.1-2.5 mIU/L 7. Each 1 standard deviation increase in TSH raises miscarriage risk by 35% 7. This strongly supports using 2.5 mIU/L as the upper limit during first trimester, not the general population cutoff of 4.0-4.5 mIU/L 5, 7.
Key Clinical Pitfalls to Avoid
Don't apply non-pregnant ranges to pregnant women - this causes both underdiagnosis of hypothyroidism and false diagnosis of hyperthyroidism 5
Don't use standard adult ranges for elderly patients - age-adjusted ranges prevent overdiagnosis and unnecessary treatment 2
Ensure laboratory functional sensitivity - TSH assays should have functional sensitivity of at least 0.02 mIU/L 1
Exclude confounding factors before diagnosing thyroid dysfunction:
- Recent levothyroxine dose adjustments
- Recovery from severe illness
- Heterophilic antibodies causing falsely elevated TSH
- Primary adrenal insufficiency 1
Consider twin pregnancies separately - they have significantly lower TSH than singleton pregnancies 6