Apparent thyrotoxicosis in a patient on high-dose biotin
A 44-year-old woman with relapsing-remitting multiple sclerosis enrolled in a high-dose biotin protocol (300 mg/day) is referred for palpitations and 3 kg of unintentional weight loss. Heart rate is 88/min, blood pressure normal; there is no goitre, tremor or lid lag. Results on the hospital's streptavidin-biotin chemiluminescent platform: TSH 0.02 mIU/L (0.4-4.0), free T4 41 pmol/L (10-22), free T3 9.8 pmol/L (3.1-6.8), TSH-receptor antibody negative, thyroid ultrasound normal. Total T4 measured on a different instrument is 105 nmol/L (66-181). The endocrinologist has prescribed methimazole 10 mg twice daily, with a repeat panel in 6 weeks. The laboratory has no policy on biotin interference.
1. How do you explain this pattern of results, and how would you confirm it before anyone acts on the diagnosis?
Read the worked answer
The pattern is the signature of biotin interference in a streptavidin-biotin immunoassay, not thyrotoxicosis. In competitive assays (free T4, free T3, total T4) excess biotin blocks the streptavidin sites, reduces signal from the labelled tracer, and is read as a spuriously high result. In sandwich assays (TSH) the same blockade reduces signal and is read as spuriously low. The combination of low TSH with high free T4 and free T3 therefore occurs without any thyroid disease.
Supporting clues: a clinically euthyroid patient with a normal gland on ultrasound, negative TSH-receptor antibody, and a normal total T4 on a different platform. Confirmation is by repeating the panel on a biotin-tolerant or non-streptavidin method and after withholding biotin for at least 72 hours (96 hours is safer at 300 mg/day; clearance depends on renal function). Meanwhile, ask the laboratory to perform dilution studies and, if available, assay the sample on an alternative platform: lack of dilution linearity and normalisation after washout are the practical proofs.
Do not accept 'assay interference' as a label without the washout result; the whole point is to prove the analytes are normal off biotin.
2. What should happen to the methimazole prescription, and who is responsible for what?
Read the worked answer
Methimazole should be withheld until the interference question is settled. Starting it on a spurious result exposes a euthyroid woman to agranulocytosis, hepatotoxicity and iatrogenic hypothyroidism, and it will also confound the repeat panel. This requires an active, documented conversation with the prescriber rather than a silent laboratory cancellation of the order.
The laboratory's responsibility: telephone the ordering clinician the same day, state plainly that the results are probably invalid because of biotin, recommend holding the drug, and record the communication. The clinician's responsibility: stop the prescription, explain to the patient that the blood test was misleading, and arrange confirmatory testing off biotin.
A useful discipline is to ask what the pretest probability was. Palpitations and 3 kg of weight loss are nonspecific; there was no goitre, no tremor, no lid lag, and a normal heart rate for a symptomatic thyrotoxic patient. A high free T4 with suppressed TSH would ordinarily show a suppressed but detectable TSH in subclinical disease or a frankly raised free T3 with an abnormal gland; the 'perfectly thyrotoxic' biochemistry in a clinically unconvincing patient is itself the warning.
3. The repeat panel 96 hours off biotin shows TSH 1.9 mIU/L, free T4 16 pmol/L, free T3 5.0 pmol/L. What laboratory policy would you now institute, and what residual risk remains?
Read the worked answer
The washout result confirms interference and closes the case for this patient. Policy should be built so the next patient is not missed. Add a mandatory supplement question to the chemistry request form and to the phlebotomy script, flag 'biotin' in the LIS so that any immunoassay result on that patient carries a comment, negotiate a biotin-tolerant assay or an alternate method for free T4/TSH at contract renewal, and educate clinicians in neurology, dermatology and endocrinology, since high-dose biotin is used for multiple sclerosis and hair/nail disorders.
Residual risks worth stating explicitly: biotin interferes with other streptavidin-biotin assays in both directions, including troponin, hCG, ferritin, parathyroid hormone and some tumour markers, so a single 'normal' troponin in a patient on biotin carries little weight. Washout depends on renal function and body stores. Other interference mechanisms (heterophile antibodies, human anti-animal antibodies, macro-TSH, thyroid hormone autoantibodies) produce different patterns and need separate workup, so a general 'interference policy' should describe the pathway rather than naming biotin alone.