TSH

If you have been feeling tired, cold, foggy, restless, or noticing your weight or heart rate drift for no clear reason, TSH (thyroid stimulating hormone) is the single most useful blood test to explain what your thyroid is doing. It is the number doctors look at first when they suspect a thyroid problem, because it responds sensitively to even small shifts in your thyroid hormone levels.

TSH is not a thyroid hormone itself. It is a signal from your pituitary gland, the small gland at the base of your brain that manages your thyroid. When the pituitary senses your thyroid hormones drifting low, it pushes TSH up. When it senses them running high, it pulls TSH down. That feedback loop is what makes TSH such an early and revealing marker.

What Your TSH Number Actually Tells You

TSH (thyroid stimulating hormone) is a hormone made by cells in your anterior pituitary called thyrotrophs. Its job is to travel through your bloodstream to your thyroid and tell it how much thyroid hormone to release. The thyroid then makes T4 and T3, the hormones that set the pace for your metabolism, body temperature, heart rate, mood, and mental clarity.

Because of the way TSH and free T4 (the main circulating thyroid hormone) are linked, tiny changes in thyroid output cause much bigger swings in TSH. The relationship is often described as inverse log-linear, though larger population studies show it is actually more complex and nonlinear, shaped by age, sex, and other factors. Either way, the practical point holds: TSH catches subtle thyroid trouble earlier than measuring the thyroid hormones themselves. A high TSH usually means your thyroid is underperforming and your pituitary is shouting louder to compensate. A low TSH usually means your thyroid is overproducing and your pituitary has gone quiet.

One important caveat: TSH is only a clean read on your thyroid if the pituitary itself is healthy. If the pituitary is damaged or diseased, TSH can look normal even when your thyroid hormones are not. That is why a single TSH result is best interpreted with free T4, and sometimes free T3, alongside it.

Hypothyroidism and Its Consequences

The most common condition TSH picks up is hypothyroidism, an underactive thyroid, most often caused by Hashimoto's thyroiditis, an autoimmune attack on the thyroid gland. Untreated hypothyroidism is not a minor inconvenience. It is linked to higher cardiovascular disease, higher overall mortality, and worse quality of life, with the strongest associations seen when TSH is markedly elevated (roughly 10 mIU/L or higher).

The degree of TSH elevation matters. In a large longitudinal study of people with hypothyroidism, being undertreated to the point of markedly elevated TSH was linked to higher risk of ischemic heart disease, heart failure, and fragility fractures compared with people whose TSH sat in the middle of the normal range. Mortality risk climbed too: people with the highest TSH levels had roughly double the mortality risk of those in the reference band.

Subclinical hypothyroidism, where TSH is elevated but free T4 is still normal, is a milder pattern, but not harmless. It can progress to overt hypothyroidism over time, and when TSH climbs markedly higher, it becomes more clearly linked to elevated cholesterol and heart-related risk.

Hyperthyroidism and Its Consequences

A low TSH points the other way, toward an overactive thyroid. The most common cause is Graves' disease, another autoimmune condition, this one where antibodies bind to the TSH receptor and drive the thyroid into overdrive. Symptoms include a racing heart, weight loss, tremor, heat intolerance, and anxiety.

The strongest evidence for harm sits at the very low end. When TSH is markedly suppressed, the risk of atrial fibrillation, a dangerous heart rhythm, is clearly increased, and the odds of progressing to overt hyperthyroidism rise as well. In treated hypothyroid patients whose TSH was pushed to markedly suppressed levels by too much thyroid replacement, mortality risk was higher than in those kept in the middle of the range.

The Sweet Spot in the Middle

A large pooled analysis of about 134,000 adults without known thyroid disease examined how risk of cardiovascular disease and death changes with TSH. The lowest combined risk sat in the middle of the normal range. Low TSH was clearly tied to higher all-cause and cardiovascular mortality. The association at the high end was less consistent in that particular analysis, though other large studies have found a U-shaped pattern with higher risk at both extremes.

This matters because a TSH result that sits at either edge of what a lab flags as "normal" is not necessarily as safe as one that sits in the middle. Statistical reference ranges and outcome-based ranges are not the same thing.

Population-Specific Signals

In people with diabetes, both the low end and the high end of the normal TSH range were linked to higher mortality. Compared with people in the middle third of normal TSH, those in the high-normal group had substantially higher all-cause and cardiovascular mortality, and those in the low-normal group had similarly elevated risk.

In heart failure, both elevated and suppressed TSH predicted worse outcomes, and even mildly elevated TSH was linked to more hospitalizations and higher mortality. After a heart attack, one large study found that higher TSH within the normal range predicted higher long-term mortality.

Congenital Hypothyroidism

In newborns, TSH is used to screen for congenital hypothyroidism, one of the few endocrine emergencies. If it is missed, the child suffers irreversible loss of cognitive function. This is why nearly every developed country tests every baby's TSH shortly after birth, and it is one of the strongest arguments for TSH as a screening tool.

Why One Reading Is Not Enough

TSH is one of the most biologically variable tests you can order. Levels swing throughout the day, following a circadian rhythm with a nocturnal surge around 2 to 4 a.m. and a daytime nadir, and are secreted in pulses that vary minute to minute. The within-person coefficient of variation for TSH over 24 hours is roughly 31%, far higher than for free T4 or free T3. Season plays a role too, with levels tending to run higher in winter.

This means a single TSH result, especially one near the edge of the reference range, can easily mislead you. Reviews of within-person variation are explicit that clinicians who react to one number risk over- or underdiagnosing thyroid disease. The most accurate picture of your thyroid comes from tracking TSH over time under similar conditions, ideally in the morning before eating, at the same lab, and away from acute illness.

A practical trending cadence: get a baseline now, retest in 3 to 6 months if you are making lifestyle changes, starting or adjusting medication, or if the first result was borderline. Once stable, retest at least once a year. If you are already treated for a thyroid condition, follow the retesting cadence your clinician recommends, usually every 6 to 12 weeks after any dose change and then every 6 to 12 months when stable.

When Results Can Be Misleading

TSH is sensitive, which is a strength, but it also means several everyday factors can shift the number without a real thyroid problem. The most common issues:

• Time of day: TSH is highest at night and lowest in the afternoon, and can more than double between afternoon and shortly after midnight. Blood drawn at 2 p.m. can look meaningfully lower than blood drawn at 8 a.m., enough to change how the result is classified.
• Recent food: A randomized study found that TSH fell by about 30% two hours after eating breakfast or drinking glucose, versus a 5% drop with continued fasting. A borderline TSH taken after a meal can look artificially normal or low.
• Acute illness: During and just after a bad infection, injury, or hospitalization, TSH can drop sharply and then rebound above normal for days to weeks. It is best not to test during acute illness unless a clinician needs an urgent read.
• Assay interference: High-dose biotin supplements can falsely lower TSH and raise free T4 and T3. Antibodies in your blood can also produce spurious values. A rare condition called macro-TSH can make TSH look dramatically elevated when your actual thyroid function is fine, seen in roughly 0.25% to 1.6% of people with elevated TSH.

Several commonly used medications also shift TSH without necessarily reflecting real thyroid disease. Glucocorticoids like prednisone or dexamethasone suppress TSH by acting on the pituitary. Dopamine agonists and somatostatin analogs do the same. High-dose biotin distorts the assay itself. If your result looks off and you are taking any of these, that context needs to be weighed before drawing conclusions.

Medications That Genuinely Alter Thyroid Function

Some drugs do cause true thyroid disease, not just a shifted lab value. Lithium, prescribed for bipolar disorder, blocks thyroid hormone release and can cause real hypothyroidism. Amiodarone, a heart rhythm medication, can cause both hypo- and hyperthyroidism because of its iodine content. Certain antiseizure drugs, particularly valproate, are associated with rising TSH over months of use. These effects belong in the conversation about who genuinely needs closer TSH monitoring.

Decision Pathway for an Out-of-Range Result

If your TSH is elevated, the standard next step is a confirmatory retest in a few weeks, drawn in the morning, along with free T4 to determine whether this is subclinical hypothyroidism (high TSH, normal free T4) or overt hypothyroidism (high TSH, low free T4). Adding TPO (thyroid peroxidase) antibodies helps identify whether the cause is autoimmune Hashimoto's thyroiditis, which raises the likelihood of progression.

If your TSH is suppressed, retest with free T4 and free T3 to confirm and classify hyperthyroidism, and consider TSH receptor antibodies (TRAb) or thyroid stimulating immunoglobulin (TSI) to check for Graves' disease. Third-generation TRAb assays and TSI have high accuracy for Graves', with sensitivity and specificity both generally reported in the high 90s in recent studies. An endocrinologist referral makes sense for a suppressed TSH with symptoms, a large goiter, or eye involvement.

If your TSH sits in the upper part of the normal range but you have symptoms, cardiovascular risk factors, high cholesterol, family history of thyroid disease, or you are considering pregnancy, that is worth discussing with a clinician rather than ignoring. A mildly high TSH is not automatically benign, and reference ranges do not fully capture personal risk.