Armour Thyroid vs Levothyroxine: What the Trials, Dr Broda Barnes, + Ray Peat PhD Found

What is in each The conversion question Three questions to ask Head-to-head trials Barnes and Peat Dose, timing, cost Cautions on record FAQ Sources

Armour Thyroid and levothyroxine both replace thyroid hormone, but they do it differently. Levothyroxine is synthetic T4 only, and the body has to convert it to the active hormone T3. Armour Thyroid is dried pig thyroid that supplies T4 and T3 together, at a fixed ratio of about four to one.¹ In the two randomized crossover trials that compared them directly, the group averages came out the same, while a subset of patients who still felt unwell on levothyroxine did better and preferred a T3-containing option.^2,3 This article lays out what is in each tablet, why conversion is the whole argument, the three questions that decide the comparison for most readers, what the trials found, and what Broda Barnes and Ray Peat wrote about it.

Key takeaways

• One grain (60 mg) of Armour Thyroid contains 38 mcg of T4 and 9 mcg of T3. Levothyroxine tablets contain T4 alone.¹
• People on levothyroxine with a normal TSH run about 15 to 20 percent lower T3-to-T4 ratios than untreated people with the same TSH, and about 15 percent of thyroidectomy patients on levothyroxine have free T3 below the control range.^4,5
• In a 70-patient blinded crossover, symptom and cognitive scores did not differ, but 48.6 percent preferred desiccated thyroid against 18.6 percent for levothyroxine, with a 3 lb (1.4 kg) average weight difference.²
• In a 75-patient three-arm crossover, the third of patients most symptomatic on levothyroxine strongly preferred T3-containing therapy and scored better on symptom, mood, and memory measures.³
• A 2021 joint statement from the American, British, and European thyroid associations concluded there is enough uncertainty to justify a new trial of T4 plus T3, and set a 13:1 to 20:1 ratio for that combination.⁶
• Barnes dosed desiccated thyroid against the morning basal temperature; Peat argued the T3:T4 ratio in the pig gland is closer to what the human gland secretes than T4 alone.^9,11

This article is for education only. It does not diagnose, treat, or replace care from your physician. I am a clinical herbalist, not a medical doctor, and I do not prescribe thyroid medication. Do not start, stop, switch, or change the dose of any thyroid hormone based on what you read here. Thyroid hormone has a narrow dosing range, interacts with many medications and supplements, and requires lab monitoring, with added caution in pregnancy, heart disease, osteoporosis, and adrenal conditions. Individual responses vary. If you are in crisis, contact your local emergency number or a crisis line.

What is in Armour Thyroid vs levothyroxine

Evidence: strong. These figures come from the manufacturers' prescribing information.

The human thyroid makes two hormones. T4 (thyroxine) is the storage form, and T3 (triiodothyronine) is the active form that cells use. The gland releases mostly T4, and about 80 percent of the T3 in the blood is made later by enzymes that strip one iodine atom from T4, mainly in the liver.¹ Any thyroid medication is a bet on which of those two steps the body can be trusted with.

Levothyroxine (Synthroid, Levoxyl, Unithroid, Euthyrox, Tirosint, and generics) is synthetic T4. It has a half-life of about a week, so blood levels stay flat with one daily dose, and the pituitary's TSH signal responds to it predictably, which is why it is the standard of care in every professional guideline.⁶ It supplies no T3. Whatever T3 the patient ends up with, the patient's own enzymes made.

Armour Thyroid is desiccated porcine thyroid, standardized so that each grain (60 mg) supplies 38 mcg of T4 and 9 mcg of T3.¹ Tablets run from a quarter grain (15 mg) to five grains (300 mg). NP Thyroid, Adthyza, and the compounded products are the same category with the same per-grain standard. Because it is a ground gland, it also contains thyroglobulin and other tissue components that a synthetic tablet does not, which is the basis of the regulatory dispute covered in my FDA status article.

Feature	Levothyroxine	Armour Thyroid
Hormones	T4 only	T4 + T3, about 4.2:11
Source	Synthetic	Pig thyroid gland
T3 supply	Made by the patient's own conversion	9 mcg per grain, direct
Half-life	About 7 days (T4)	T4 about 7 days; T3 about 1 day
Dose units	Micrograms (25 to 300 mcg)	Grains or milligrams (15 to 300 mg)
FDA status	Approved	Marketed without approval under enforcement discretion
Guideline position	First-line	Not recommended by major societies

The ratio matters. A pig gland runs about four parts T4 to one part T3. The human gland secretes closer to 14 or 15 parts T4 to one part T3, which is why the 2021 consensus document set 13:1 to 20:1 as the starting ratio for any synthetic combination.⁶ Desiccated thyroid gives more T3 per dose than the human gland would, which is the source of both its reputation for making people feel better and the T3 peak that its critics point to.

The conversion question: why T4 alone works for most people and not all

Evidence: strong for the lab findings; the link between low T3 and symptoms is still being tested.

Levothyroxine works on the assumption that a person given enough T4 will convert it to enough T3. For most patients that holds. Two large datasets show where it does not.

In 2011, endocrinologists in Catania compared 1,811 patients who had their thyroid removed and were on levothyroxine with a normal TSH against 3,875 untreated people with normal thyroid function. The treated group had higher free T4 and lower free T3 across the board. About 15 percent of them had free T3 below the control range, and the spread of T3-to-T4 ratios was wide, which the authors read as a large difference between individuals in how well they convert. Their conclusion was that levothyroxine alone cannot guarantee normal thyroid status in every patient.⁴

In 2016, Antonio Bianco's group at Rush pulled 9,981 people with a normal TSH from the US national health survey, 469 of them on levothyroxine, and matched the treated group to controls by age, sex, race, and TSH. The levothyroxine group had 15 to 20 percent lower T3-to-T4 ratios. They also had a higher body mass index despite reporting fewer calories per kilogram, were more likely to be taking statins, beta-blockers, and antidepressants, and reported less physical activity and worse self-rated health, with 12 of 52 measures differing from controls.⁵ The study is cross-sectional, so it cannot show which way the arrow points, but it puts numbers on what patients had been reporting.

Part of the variation is inherited. About 16 percent of people carry a common variant in the type 2 deiodinase gene (Thr92Ala-DIO2), the enzyme that makes T3 inside tissues. In a British trial of 552 patients, carriers had worse baseline wellbeing on levothyroxine and improved more on combination therapy than non-carriers.⁸ The Walter Reed three-arm trial genotyped its patients and did not find the variant predicted outcome, so the genetic story is unsettled.³

If you already read my article on cholesterol and thyroid, the statin finding above is the same relationship from the other side: T3 drives LDL receptor activity, and a person who is under-converted can carry a "normal" TSH and a stubborn cholesterol number at the same time.

Three questions that decide the comparison

Evidence: these questions are drawn from the trial subgroups and the 2021 consensus criteria. They are for a conversation with your prescriber, not a self-treatment guide.

1. Do I still feel hypothyroid on levothyroxine with a normal TSH?

This is the group both crossover trials singled out. In the 2013 trial, the patients who preferred desiccated thyroid were the ones whose symptom scores improved on it, and they were identifiable in advance by a set of five baseline variables.² In the 2021 trial, the third of patients with the worst symptom scores on levothyroxine showed a strong preference for either desiccated thyroid or T4 plus T3, with measurable gains in symptom, mood, and visual memory scores.³ The 2021 consensus document makes the same distinction: a patient who has no improvement on an adequate levothyroxine dose of at least 1.2 mcg per kilogram, after other causes of the symptoms have been ruled out, is a candidate for combination therapy.⁶ A patient who feels well on levothyroxine has no reason in the research to switch.

2. Is my free T3 low while my TSH is normal?

The consensus document lists a low serum T3 on levothyroxine as the second criterion.⁶ The Catania and NHANES data explain why: a normal TSH on T4 alone can sit on top of a below-range T3, and TSH tracks T4 more closely than it tracks T3 in treated patients.^4,5 A prescriber who is willing to run free T3 alongside TSH and free T4 has the number this question needs. Many labs and many clinicians order TSH alone, so it is worth asking for the full panel before assuming the answer.

3. Will my doctor prescribe desiccated thyroid, and if not, what is the alternative?

Some will not, because the major guidelines do not recommend it, because it is unapproved, or because of the T3 peak discussed below. That does not close the door on T3. The consensus document endorses a trial of synthetic levothyroxine plus liothyronine (Cytomel) at a 13:1 to 20:1 ratio, roughly 5 or 10 mcg of T3 added to 100 to 200 mcg of T4, reassessed after three to six months.⁶ In the Walter Reed trial the synthetic combination and desiccated thyroid performed the same, so a patient who cannot get Armour has a guideline-backed route to the same hormone.³ Availability and coverage of desiccated thyroid in 2026 are covered in the FDA status article.

What the head-to-head trials found

Evidence: moderate. Two randomized, double-blind crossover trials from the same military center, one older T4 plus T3 trial, one systematic review, and one large observational cohort.

Walter Reed, 2013: desiccated thyroid vs levothyroxine

Seventy adults with primary hypothyroidism, stable on levothyroxine for at least six months, were randomized to desiccated thyroid or levothyroxine for 16 weeks, then switched for another 16 weeks. Neither the patients nor the investigators knew which they were taking, and doses were adjusted to keep TSH in range. The conversion used was one grain of desiccated thyroid for 88 mcg of levothyroxine. On the primary outcomes, symptom questionnaires and a battery of memory and cognitive tests, the two treatments did not differ. Patients weighed an average of 3 lb (1.4 kg) less at the end of the desiccated thyroid period. Asked which they preferred, 34 chose desiccated thyroid, 13 chose levothyroxine, and 23 had no preference. The patients who preferred desiccated thyroid had lost 4 lb (1.8 kg) on it and had better symptom scores during that period.² Serum T3 was higher on desiccated thyroid, which a letter to the journal criticized as a non-physiological peak.

Walter Reed, 2021: three arms

The same group then ran 75 patients through all three options, levothyroxine, levothyroxine plus liothyronine, and desiccated thyroid, for 22 weeks each. Across the whole group there was no difference on symptom score, quality of life, memory testing, or depression score, and no difference in preference. Desiccated thyroid produced a small rise in heart rate. The result changed when the investigators looked at the third of patients who had scored worst on levothyroxine at baseline: that subgroup strongly preferred the two T3-containing arms and improved on the symptom questionnaire, the general health questionnaire, the depression inventory, and the visual memory index.³ Antonio Bianco, a former president of the American Thyroid Association, was a co-author.

The earlier combination trials

The trial that started the modern debate was published in 1999 in the New England Journal of Medicine. In 33 patients, replacing 50 mcg of levothyroxine with 12.5 mcg of liothyronine improved mood and cognitive scores on six of seventeen measures, and 20 of the 33 preferred the combination.⁷ Most of the trials that followed did not reproduce that benefit at the group level, which is why the 2021 joint consensus of the American, British, and European thyroid associations counted fourteen trials without a consistent group benefit and still concluded that a new, better-designed trial is justified, in part because patients keep reporting improvement the group averages do not capture.⁶

The review and the cohort

A 2024 systematic review in Thyroid found the two randomized trials of desiccated thyroid showed no group difference in symptoms or quality of life, while the nonrandomized studies favored it, and flagged the heart rate increase and a possible reduction in HDL as the findings to watch.⁹ A 2026 analysis of 1.26 million hypothyroid patients found that treatment containing T3, whether synthetic combination or desiccated thyroid, was associated with lower dementia and mortality risk than levothyroxine alone. It is observational, which means it can show association only, and it is discussed in more detail in the FDA status article.¹⁰

Read together, the trials say the same thing three ways. For the average patient the two approaches are equivalent on every objective measure. For the patient who does not feel well on T4 alone, the blinded data support trying T3, in either form.

What Broda Barnes and Ray Peat wrote about it

Evidence: clinical writing from a physician and a physiologist. Barnes's basal temperature method has never been validated in a modern trial against TSH; Peat's ratio argument is mechanism and observation, not a controlled study.

Barnes: the basal temperature and the grain

Broda Barnes was a physician and physiologist who practiced for decades in Fort Collins, Colorado, and published his basal temperature method in the Journal of the American Medical Association in 1942, before any blood test for thyroid function existed. The test is an underarm temperature taken for ten minutes on waking, before getting out of bed. He put the normal range at 97.8 to 98.2°F (36.6 to 36.8°C) and treated a repeated reading below 97.8°F (36.6°C) as a sign of low thyroid function when the symptoms fit.^11,12 His 1976 book Hypothyroidism: The Unsuspected Illness laid out the case that blood tests were missing a large share of hypothyroid patients, and that the temperature, the pulse, and the patient's own account should guide dosing.¹²

Barnes prescribed desiccated thyroid, dosed in grains and raised slowly until the morning temperature and the symptoms normalized. He did not use levothyroxine, partly because it did not exist for most of his career and partly because he considered the whole gland the better match for the patient. His method has one advantage that is still relevant: it measures an outcome in the tissues rather than a signal from the pituitary. Its weakness is that morning temperature is moved by many things other than thyroid, which is where Peat picked up the thread.

Peat: the ratio and the pulse

Ray Peat took Barnes's method and added the resting pulse. He wrote that during a hot Oregon summer he saw hypothyroid people whose morning temperature looked normal by Barnes's standard while their pulse was very low, and that in cool weather an electric blanket could produce the same false reading. His conclusion was that temperature and pulse together are far more useful than either alone, and that a healthy resting pulse runs higher than most people assume.¹³

On the choice of medication, Peat's argument was about the ratio. He noted that people taking desiccated thyroid have much higher T3-to-T4 ratios in their blood than people taking thyroxine, that untreated healthy people also have higher ratios than thyroxine patients, and that the human gland secretes proportionally more T3 than serum levels suggest because T3 is used up faster. He called the serum ratio a poor argument for thyroxine over a product that resembles what the gland secretes, and pointed to brain tissue, which concentrates T3 and tends to expel T4, as the reason. He also described a pattern he saw in women given thyroxine alone: as the dose went up, some felt worse, which he attributed to excess T4 acting against T3 in tissues that convert it slowly.¹⁴

Peat did not treat desiccated thyroid as sacred. He wrote that synthetic combinations at four to one work, that many people think more clearly at about three to one, and that because T3 has a short half-life it is better taken in small amounts through the day than in one morning dose. His practical tool was the same as Barnes's with the pulse added: a 14-day chart of morning temperature and resting pulse, watched for a slow rise as the dose settles.¹⁴ Both writers were dosing to the person rather than to a lab range, which is the exact question the Walter Reed subgroups reopened fifty years later. Peat's broader thyroid framework, including his food list and his reading of TSH, is collected on the Ray Peat guide, and his view of how low thyroid function shows up in body weight is covered in hypothyroidism and weight loss.

Start here The Ray Peat Diet Guide Every food he mentioned, his hormone framework in order, and links to his interview audio transcripts, and articles. Open the guide →

Handmade in Montana

← →

Frequently asked questions

Short answers, with the longer discussion above.

Is Armour Thyroid better than levothyroxine?

Not for the average patient. In two blinded crossover trials, symptom and cognitive scores were the same on both. The patients who did better on Armour were the ones who still felt hypothyroid on levothyroxine, and that subgroup also did as well on synthetic T4 plus T3.^2,3

What is the difference between Armour Thyroid and Synthroid?

Synthroid is synthetic T4 only. Armour is dried pig thyroid that supplies T4 and T3 together, 38 mcg and 9 mcg per grain.¹ Synthroid depends on the body converting T4 to T3; Armour supplies some T3 directly.

How much Armour Thyroid equals 100 mcg of levothyroxine?

Conversions range from about 1 grain (60 mg) per 74 to 100 mcg of levothyroxine depending on the method. The randomized trial used 88 mcg per grain.^1,2 The dose is set by follow-up labs, not by the conversion chart.

Can I have a normal TSH and still be low in T3?

Yes. In a series of 1,811 thyroidectomy patients on levothyroxine with a normal TSH, about 15 percent had free T3 below the range of untreated controls.⁴ Asking for free T3 alongside TSH is how that shows up.

Why do some doctors refuse to prescribe Armour Thyroid?

The major guidelines recommend levothyroxine first, Armour is marketed without FDA approval, and its T3 produces a peak after each dose. A prescriber who will not use Armour may still consider adding synthetic liothyronine to levothyroxine, which the 2021 joint consensus supports as a trial in patients who do not improve on T4 alone.⁶

Is Armour Thyroid still available in 2026?

Yes. The FDA withdrew its August 2026 removal deadline in March 2026 and is applying case-by-case enforcement while manufacturers pursue approval. Coverage is the current obstacle, since CVS Caremark dropped it from its standard formulary in April 2026. Details are in the FDA status article.

What did Broda Barnes use to dose thyroid?

The morning underarm temperature, taken for ten minutes before rising, with 97.8 to 98.2°F (36.6 to 36.8°C) as his normal range, along with the pulse and the patient's symptoms.^11,12 Peat added the resting pulse and a 14-day chart.¹⁴

Sources

Every citation was checked against its PubMed or DOI record or the manufacturer's published label. Funding and affiliations are noted where they apply.

1. AbbVie Inc. Armour Thyroid (thyroid tablets, USP) prescribing information. rxabbvie.com/pdf/armour_thyroid_pi.pdf. Manufacturer document; labeling not FDA-approved.
2. Hoang TD, Olsen CH, Mai VQ, Clyde PW, Shakir MK. Desiccated thyroid extract compared with levothyroxine in the treatment of hypothyroidism: a randomized, double-blind, crossover study. J Clin Endocrinol Metab. 2013;98(5):1982-1990. doi:10.1210/jc.2012-4107. PMID 23539727. Walter Reed National Military Medical Center; no industry funding declared.
3. Shakir MKM, Brooks DI, McAninch EA, Fonseca TL, Mai VQ, Bianco AC, Hoang TD. Comparative effectiveness of levothyroxine, desiccated thyroid extract, and levothyroxine+liothyronine in hypothyroidism. J Clin Endocrinol Metab. 2021;106(11):e4400-e4413. PMID 34185829. Walter Reed, Rush, and University of Chicago; Bianco has disclosed consulting for thyroid product manufacturers in other publications.
4. Gullo D, Latina A, Frasca F, Le Moli R, Pellegriti G, Vigneri R. Levothyroxine monotherapy cannot guarantee euthyroidism in all athyreotic patients. PLoS One. 2011;6(8):e22552. doi:10.1371/journal.pone.0022552. PMID 21829633. University of Catania and an Italian cancer research grant; funders had no role.
5. Peterson SJ, McAninch EA, Bianco AC. Is a normal TSH synonymous with "euthyroidism" in levothyroxine monotherapy? J Clin Endocrinol Metab. 2016;101(12):4964-4973. doi:10.1210/jc.2016-2660. PMID 27700539. Cross-sectional NHANES analysis, Rush University.
6. Jonklaas J, Bianco AC, Cappola AR, et al. Evidence-based use of levothyroxine/liothyronine combinations in treating hypothyroidism: a consensus document. Thyroid. 2021;31(2):156-182. PMID 33276704. Joint statement of the American, British, and European Thyroid Associations.
7. Bunevicius R, Kazanavicius G, Zalinkevicius R, Prange AJ Jr. Effects of thyroxine as compared with thyroxine plus triiodothyronine in patients with hypothyroidism. N Engl J Med. 1999;340(6):424-429. doi:10.1056/NEJM199902113400603. PMID 10022110. Thirty-three patients; crossover design.
8. 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 in hypothyroid patients. J Clin Endocrinol Metab. 2009;94(5):1623-1629. PMID 19190113. Secondary analysis of a UK trial (n = 552).
9. Riis KR, Larsen CB, Bonnema SJ. Potential risks and benefits of desiccated thyroid extract for the treatment of hypothyroidism: a systematic review. Thyroid. 2024;34(6):687-701. doi:10.1089/thy.2023.0649. Odense University Hospital, Denmark.
10. Beltrão FEL, Carvalhal G, Meneghini V, et al. Treatment of hypothyroidism that contains liothyronine is associated with reduced risk of dementia and mortality. J Clin Endocrinol Metab. 2026;111(2):561-571. doi:10.1210/clinem/dgaf367. Observational cohort; senior author Bianco.
11. Barnes BO. Basal temperature versus basal metabolism. JAMA. 1942;119(14):1072-1074. Original description of the basal temperature method; no DOI for this era.
12. Barnes BO, Galton L. Hypothyroidism: The Unsuspected Illness. New York: Harper & Row; 1976. Clinical book by a physician; not peer-reviewed research.
13. Peat R. TSH, temperature, pulse rate, and other indicators in hypothyroidism. raypeat.com. raypeat.com/articles/articles/hypothyroidism.shtml. Essay by a physiologist; includes his own clinical observations.
14. Peat R. Thyroid: therapies, confusion, and fraud. raypeat.com. raypeat.com/articles/articles/thyroid.shtml. Essay by a physiologist; the ratio and dosing passages are his interpretation, not a controlled study.
15. AbbVie Inc. Synthroid (levothyroxine sodium tablets, USP) prescribing information, including boxed warning, administration instructions, and drug interactions. Available on DailyMed. Manufacturer document; FDA-approved labeling. Levothyroxine subpotency recalls (Accord Healthcare, June 2025; a further multi-strength recall, July 2026) are documented in the FDA status article sources.