Drugs That Distort Your TSH Test

What TSH Measures and Why Accuracy Matters

Thyroid-stimulating hormone (TSH) is a pituitary glycoprotein that drives thyroid hormone production. A single TSH measurement is the frontline screening test for both hypothyroidism and hyperthyroidism, endorsed by the American Thyroid Association and the AACE/ACE 2012 clinical practice guidelines as the most sensitive marker of thyroid status.

The reference range at most laboratories falls between 0.4 and 4.0 mIU/L, though the National Academy of Clinical Biochemistry has argued that the upper limit should be 2.5 mIU/L when autoimmune thyroid disease is excluded. Because clinical decisions hinge on tenths of a mIU/L, even modest drug-induced shifts can trigger inappropriate levothyroxine initiation, dose adjustments, or imaging workups.

TSH secretion is pulsatile and peaks between 11 PM and 4 AM. Blood drawn in the afternoon can read 50% lower than a morning sample from the same patient. Layering pharmacologic interference on top of this diurnal variability creates a narrow window for reliable interpretation.

Drugs That Falsely Lower TSH

Several medication classes suppress pituitary TSH secretion through central mechanisms or assay interference, producing readings that mimic subclinical hyperthyroidism.

Glucocorticoids suppress TSH in a dose-dependent fashion. A single 2 mg dexamethasone dose reduces TSH by approximately 50% within 4 to 6 hours, and chronic prednisone at 10 mg/day or higher keeps TSH below baseline indefinitely. The mechanism involves direct suppression of thyrotropin-releasing hormone (TRH) neurons in the hypothalamus. The Endocrine Society's guidelines on central hypothyroidism explicitly flag glucocorticoid use as a confounder.

Dopamine and dopamine agonists (cabergoline, bromocriptine, ropinirole) inhibit TSH release from thyrotroph cells through D2 receptor activation. ICU patients receiving dopamine infusions at renal or intermediate doses (2 to 10 mcg/kg/min) routinely show TSH values below 0.1 mIU/L. This finding does not indicate thyrotoxicosis and does not require treatment. A 2019 systematic review confirmed that dopaminergic TSH suppression resolves within 24 to 48 hours of discontinuation.

Biotin (vitamin B7) deserves special attention. Patients taking 5 to 10 mg daily for hair or nail growth can generate falsely low TSH and falsely elevated free T4 on competitive streptavidin-biotin immunoassays. The FDA issued a safety communication in 2017 after at least one death was linked to troponin assay interference in a biotin-supplementing patient. Stop biotin for at least 48 hours (72 hours for high-dose biotin therapy) before thyroid function testing.

Metformin appears to lower TSH modestly. A randomized controlled trial (N=66) showed metformin reduced TSH by an average of 0.7 mIU/L in hypothyroid diabetic patients already on levothyroxine. The effect did not reach clinical significance in euthyroid subjects, but it can push a borderline-high TSH into the normal range, delaying appropriate dose escalation.

Somatostatin analogs (octreotide, lanreotide) suppress TSH through direct pituitary inhibition. This is well-characterized and expected in patients treated for acromegaly or neuroendocrine tumors.

Drugs That Falsely Raise TSH

Medications that raise TSH either damage the thyroid gland directly, block hormone synthesis, or interfere at the assay level.

Lithium remains one of the most clinically significant thyroid disruptors. It concentrates in thyroid follicular cells, inhibiting iodine organification and thyroid hormone release. The 2014 ATA/AACE hypothyroidism guidelines recommend baseline and every-6-month TSH monitoring in all lithium-treated patients. Overt hypothyroidism develops in 6 to 52% of lithium users, depending on study duration and iodine status, per a Cochrane-indexed meta-analysis.

Amiodarone contains 37% iodine by weight. Each 200 mg tablet delivers approximately 75 mg of organic iodine. This iodine load can trigger either amiodarone-induced hypothyroidism (AIH) or amiodarone-induced thyrotoxicosis (AIT). AIH occurs in approximately 13% of patients in iodine-sufficient regions, manifesting as elevated TSH within the first 6 to 12 months. A prospective cohort (N=354) published in the Journal of Clinical Endocrinology & Metabolism documented thyroid dysfunction in 15.3% of amiodarone-treated patients over 3 years.

Tyrosine kinase inhibitors (sunitinib, sorafenib, imatinib, lenvatinib) cause hypothyroidism through thyroid capillary regression and reduced glandular blood flow. Up to 53% of sunitinib-treated patients develop elevated TSH requiring levothyroxine, according to a meta-analysis of 26 studies. TSH should be checked every 4 to 6 weeks for the first 6 months on these agents.

Immune checkpoint inhibitors (nivolumab, pembrolizumab, ipilimumab) trigger autoimmune thyroiditis in 5 to 10% of patients. The typical pattern is a transient thyrotoxic phase (low TSH) followed by permanent hypothyroidism (elevated TSH). A 2018 meta-analysis in Lancet Oncology (N=7,551) found hypothyroidism in 6.6% of anti-PD-1 recipients and 13.2% of combination immunotherapy patients.

Iodinated contrast media deliver a massive iodine bolus (typically 13,500 to 45,000 mcg per CT scan). In patients with autonomous thyroid nodules or latent Graves' disease, this can precipitate thyrotoxicosis. In those with Hashimoto's thyroiditis, the Wolff-Chaikoff effect can cause transient hypothyroidism. The ACR recommends waiting 6 to 8 weeks after contrast administration before interpreting thyroid function tests.

Drugs That Affect Thyroid Hormone Binding (Indirect TSH Effects)

Several medications alter thyroxine-binding globulin (TBG) or displace T4 from carrier proteins without directly affecting TSH secretion. The indirect TSH shift occurs because the pituitary responds to transient changes in free hormone levels.

Estrogens (oral contraceptives, hormone replacement therapy) increase hepatic TBG synthesis. Total T4 rises, but free T4 and TSH typically remain stable. In hypothyroid patients on fixed-dose levothyroxine, however, the increased TBG sequesters more T4, potentially raising TSH. The Endocrine Society recommends rechecking TSH 6 weeks after initiating or changing estrogen therapy.

Androgens and anabolic steroids reduce TBG. This lowers total T4 while free T4 remains normal. TSH usually stays unchanged, but automated free T4 estimates using total T4 correction may produce confusing results.

Heparin (unfractionated and LMWH) activates lipoprotein lipase, generating free fatty acids that displace T4 from binding proteins in vitro. This creates artifactually elevated free T4 on analog immunoassays, potentially suppressing TSH if the lab draw occurs during active heparin infusion. The effect is an assay artifact rather than a true physiologic change.

Phenytoin, carbamazepine, and rifampin induce hepatic CYP enzymes that accelerate T4 clearance. Patients on stable levothyroxine doses may see TSH creep upward after starting these drugs, requiring a 20 to 50% levothyroxine dose increase. This represents a true pharmacokinetic interaction, not merely an assay artifact.

Timing and Practical Interpretation

The timing of blood draw relative to medication dosing matters as much as knowing which drugs interfere. Draw TSH before the morning levothyroxine dose. A blood sample taken 1 to 3 hours after ingestion may show a transient free T4 spike, producing a reflexively lower TSH reading on the same day's afternoon labs.

For biotin: discontinue 48 to 72 hours before testing. For glucocorticoids: note the dose and timing on the lab requisition. For dopamine agonists: document whether the patient took the medication that morning.

The 2017 ATA guidelines for hypothyroidism in pregnancy explicitly state that trimester-specific reference ranges and concurrent medications must both be considered before adjusting levothyroxine.

A practical framework for clinicians: when TSH does not match the clinical picture, ask three questions. Is the patient taking biotin or a multivitamin containing biotin? Was the blood drawn during a glucocorticoid burst, dopamine infusion, or within 8 weeks of iodinated contrast? Is the patient on a medication known to alter TBG or accelerate T4 metabolism?

How to Lower TSH Safely

Elevated TSH indicates the thyroid gland is underperforming relative to pituitary demand. The standard intervention is levothyroxine sodium, titrated in 12.5 to 25 mcg increments with TSH reassessment every 6 weeks. The AACE 2012 clinical practice guidelines recommend a target TSH of 0.4 to 2.5 mIU/L for most non-pregnant adults.

Before attributing an elevated TSH to primary hypothyroidism, exclude drug-induced elevation. Lithium, amiodarone, tyrosine kinase inhibitors, and checkpoint inhibitors are the most common culprits. A patient started on sunitinib three months ago with a TSH of 12 mIU/L needs levothyroxine replacement, not drug discontinuation, because the thyroid damage is often permanent.

For mildly elevated TSH (4.5 to 10 mIU/L) in older adults, the TRUST trial (N=737) published in the New England Journal of Medicine showed no benefit from levothyroxine treatment in adults over 65 with subclinical hypothyroidism. This finding underscores why accurate TSH measurement, free of drug artifact, is essential before initiating lifelong therapy.

How to Raise TSH (When Overtreatment Is Suspected)

If TSH is suppressed below 0.1 mIU/L in a patient on levothyroxine, the first step is dose reduction. Subclinical hyperthyroidism from overreplacement increases atrial fibrillation risk by 1.6-fold and vertebral fracture risk by 3.2-fold, per a 2015 meta-analysis (N=70,298) in JAMA Internal Medicine.

However, confirm that suppression is not artifactual. Biotin supplementation, concurrent glucocorticoids, and dopamine agonists can all produce spuriously low TSH. Reduce the levothyroxine dose only after ruling out these confounders.

Raise TSH by reducing levothyroxine in 12.5 mcg decrements. Recheck in 6 weeks. In patients with differentiated thyroid cancer, the target TSH depends on risk stratification and should follow ATA 2015 thyroid cancer management guidelines.

Special Populations: ICU, Pregnancy, and Oncology

Critical illness produces the "euthyroid sick syndrome" or nonthyroidal illness syndrome (NTIS). TSH may be low, normal, or mildly elevated despite severe physiologic stress. Dopamine, dobutamine, high-dose glucocorticoids, and octreotide, all common in ICU settings, compound this picture. The recommendation: do not measure TSH in acutely ill patients unless there is strong pretest suspicion of thyroid disease. The European Thyroid Association advises postponing thyroid function testing until at least 6 weeks after ICU discharge.

Pregnancy shifts TSH reference ranges downward, particularly in the first trimester when hCG stimulates the TSH receptor. The lower limit drops to approximately 0.1 mIU/L and the upper limit to 2.5 to 4.0 mIU/L depending on the trimester and population. Medications such as metoclopramide (which raises TSH transiently) and progesterone (minimal effect) should be documented when interpreting prenatal thyroid panels.

Oncology patients on checkpoint inhibitors or TKIs require TSH monitoring every 4 to 6 weeks for the first 6 months, then every 3 months. A rising TSH trend over 2 or more measurements is more informative than any single value, given the multiple confounders present in cancer treatment regimens.

Assay-Specific Considerations

Third-generation TSH immunoassays have a functional sensitivity of 0.01 to 0.02 mIU/L, making them accurate enough to distinguish true suppression from subnormal TSH. Yet no assay is immune to interference.

Heterophilic antibodies (human anti-mouse antibodies, or HAMA) can cause either falsely high or falsely low TSH depending on assay architecture. Patients who have received monoclonal antibody therapies, worked with animals, or received certain immunizations are at higher risk. When clinical presentation and TSH grossly disagree, request an alternative-platform reanalysis or a serial dilution test.

The National Academy of Clinical Biochemistry recommends that laboratories report known interferences specific to their assay platform in the comment section of thyroid function results. Not all laboratories comply.

Summary Table of Key Drug-TSH Interactions

| Drug/Class | Direction of TSH Shift | Mechanism | Clinical Action | |---|---|---|---| | Biotin (>5 mg/day) | Falsely low | Assay interference (streptavidin-biotin) | Hold 48-72 hours pre-draw | | Glucocorticoids | Low (true suppression) | Central TRH/TSH inhibition | Document dose; recheck off steroids | | Dopamine/dobutamine | Low (true suppression) | D2 receptor pituitary inhibition | Do not treat; self-resolves | | Lithium | High (true effect) | Inhibits thyroid hormone synthesis/release | Monitor q6 months; start LT4 if needed | | Amiodarone | High or low | Iodine excess; direct thyroid toxicity | Monitor q3-6 months; evaluate AIT vs AIH | | Checkpoint inhibitors | High (after initial low) | Autoimmune thyroiditis | Monitor q4-6 weeks; permanent LT4 often needed | | TKIs (sunitinib) | High (true effect) | Thyroid capillary regression | Monitor q4-6 weeks; start LT4 early | | Estrogen (oral) | Mildly high in treated patients | Increased TBG sequesters T4 | Recheck TSH 6 weeks after estrogen change | | Phenytoin/carbamazepine | Mildly high in treated patients | CYP induction accelerates T4 clearance | Increase LT4 dose 20-50% | | Heparin | Falsely low (via elevated fT4) | In vitro FFA displacement | Draw TSH before heparin or use equilibrium dialysis | | Metformin | Mildly low | Unclear; possibly increased T4 bioavailability | Consider when borderline TSH normalizes |