Why Levothyroxine Stops Working (The Conversion Problem Nobody Tests)

Most people on levothyroxine assume their thyroid problem is solved once their TSH comes back normal. Their doctor adjusts the dose until that number lands in range, and the treatment is considered complete. But that assumption skips over something the test doesn't actually measure, and understanding what it misses starts with understanding how thyroid hormone works in the first place.

Your thyroid gland produces two hormones: T4 and T3. T4 is the storage form, mostly inactive on its own, and the thyroid releases it in large quantities. T3 is the active form, the one that actually binds to receptors inside your cells and drives metabolism, energy production, temperature regulation, and cognitive function. The thyroid gland itself only produces about 20% of the T3 circulating in your body at any given time. The other 80% comes from your body converting T4 into T3 after it leaves the thyroid, primarily in the liver, kidneys, and gut.

That conversion is done by enzymes called deiodinases, which are proteins that physically remove an iodine atom from T4 to produce T3. The most relevant ones are called type 1 and type 2 deiodinase, and they are something called selenoproteins, which means they require selenium as a core structural component to function at all. Without adequate selenium, these enzymes cannot do their job at the molecular level.

So when a doctor prescribes levothyroxine, which is synthetic T4, they are giving you the raw material and assuming the conversion machinery will handle the rest. That assumption works for many people. But it is not universally true, and the standard lab panel does not check whether it is true for you specifically.

TSH, or thyroid stimulating hormone, is released by the pituitary gland in response to how much thyroid hormone the pituitary is sensing. When T4 levels are sufficient, TSH drops. When T4 is low, TSH rises. This makes TSH a useful signal for whether you have enough T4 in circulation, but it tells you nothing about how well that T4 is being converted downstream. A patient can have a perfectly normal TSH and still have low T3 throughout the rest of their body if conversion is impaired.

This is exactly what the research shows. Studies on levothyroxine-treated patients with normal TSH values have found that their T3 to T4 ratios run 15 to 20% lower than those seen in people with intact thyroid function, and that up to 40% of these patients remain symptomatic despite their labs appearing normal. Those symptoms, fatigue, brain fog, cold intolerance, weight difficulty, are often dismissed because the TSH looks fine. But the TSH was never measuring what the patient was actually asking about.

The problem becomes more layered when you consider what impairs conversion in the first place. Selenium deficiency is one pathway. Research in pediatric patients with selenium deficiency showed measurably elevated free T4 to free T3 ratios, which is the biochemical signature of conversion failure. The deiodinase enzymes are not functioning, T4 accumulates, and T3 drops. Chronic inflammation is another pathway, and this matters especially in Hashimoto's thyroiditis, which is the autoimmune condition responsible for most cases of hypothyroidism in developed countries. In Hashimoto's, the immune system attacks thyroid tissue, and the ongoing inflammatory signaling downregulates deiodinase activity in peripheral tissues. So the disease does double damage: it reduces the 20% of T3 the thyroid makes directly, and it simultaneously impairs the 80% that comes from conversion. The medication addresses the T4 deficiency. The conversion problem remains unaddressed.

Then there is the genetic layer. Somewhere between 12 and 36% of people carry a variant in a gene called DIO2, which codes for the type 2 deiodinase enzyme. This variant reduces the activity of that enzyme, meaning the conversion from T4 to T3 is structurally less efficient in that person regardless of their selenium status, regardless of their inflammation levels, regardless of their dose. Research has linked this DIO2 variant to worse psychological wellbeing and reduced response to T4-only therapy. For these individuals, levothyroxine is not a complete solution because their body cannot efficiently convert what the drug delivers.

Liver dysfunction and gut dysbiosis are additional variables that can compound the problem. A significant portion of T4 to T3 conversion happens in the liver, and impaired liver function reduces that capacity. The gut is also an active site of conversion and plays a role in thyroid hormone reabsorption, so disrupted gut microbiome composition can affect how much T3 ultimately reaches circulation.

When you put all of this together, you get a situation where a patient can be doing everything right, taking their medication consistently, following up regularly, and still feeling unwell, not because the medication is failing in a simple sense, but because the medication was designed to replace T4 and the measurement used to guide dosing only confirms that T4 is present. Neither the drug nor the test was designed to evaluate conversion.

The most direct fix is to ask your doctor to test free T3 alongside TSH. Free T3 measures the active, unbound form of the hormone available to your cells and gives you a direct readout of whether conversion is happening adequately. It is a simple blood test that is not routinely ordered but is not exotic either.

If free T3 is consistently low despite normal TSH, there are two main levers. Selenium supplementation at around 200 micrograms per day supports the enzymatic machinery responsible for conversion and is a reasonable first step to discuss with your provider, particularly if your diet is low in selenium-rich foods like Brazil nuts, seafood, and organ meats. The second lever, relevant especially for people with confirmed DIO2 variants or persistent symptoms despite optimization, is combination therapy using both T4 and T3, typically as levothyroxine plus a small dose of liothyronine. This bypasses the conversion step entirely by delivering the active hormone directly.

The standard model of thyroid treatment is built around a single assumption: that T4 in equals T3 out. For a significant portion of patients, that assumption is wrong, and no one ever tested whether it was right.

---

References

1. Bianco AC, Kim BW. Deiodinases: implications of the local control of thyroid hormone action. J Clin Invest. 2006;11610:2571-2579. PMID: 17016550. Source
2. Endotext NCBI Bookshelf. Thyroid Hormone Synthesis and Secretion: "The thyroid gland produces approximately 20% of total daily T3 production, with the remaining 80% arising from peripheral deiodination of T4." Source
3. Salvatore D, Porcelli T, Ettleson MD, Bianco AC. The relevance of T3 in the management of hypothyroidism. Lancet Diabetes Endocrinol. 2022;10(5):366-372. DOI: 10.1016/S2213-8587(22)00004-3.
4. Peterson SJ, Cappola AR, Castro MR, et al. An online survey of hypothyroid patients demonstrates prominent dissatisfaction. Thyroid. 2019;295:707-721. PMID: 29620972. Source
5. Panicker V, Saravanan P, Vaidya B, et al. Common variation in the DIO2 gene predicts baseline psychological well-being and response to combination thyroxine plus triiodothyronine therapy. JCEM. 2009;945:1623-1629. PMID: 19190113. Source
6. Kobayashi R, Hasegawa Y, Kawaguchi T, et al. Thyroid function in patients with selenium deficiency exhibits high free T4 to T3 ratio. Clin Pediatr Endocrinol. 2021;30(1):19-26. DOI: 10.1297/cpe.30.19.
7. Ventura M, Melo M, Carrilho F. Selenium and thyroid disease: from pathophysiology to treatment. Int J Endocrinol. 2017;2017:1297658. PMID: 28255299. Source

---

Join the free community:
Men: Iron Forge Brotherhood
Women: Powerhouse Fitness