T3 combination therapy

Some people treated for an underactive thyroid still feel unwell even when their blood tests are normal, and adding the active hormone T3 to standard levothyroxine is the main option that has been tested. On average it helps a little, the best-known gene test for picking who benefits did not hold up, and at correct doses the safety data are broadly reassuring.

Treat T3 as a supervised, time-limited experiment with a thyroid specialist, only after other causes of tiredness have been ruled out, and stop it if nothing improves within about three months. Don't get a DIO2 gene test to decide, and don't take it if you feel fine on levothyroxine alone.

What it is

The thyroid makes two hormones: thyroxine (T4), the main circulating form, and triiodothyronine (T3), the active one. A healthy gland secretes them in a ratio of roughly 14 to 1. The standard treatment for an underactive thyroid, synthetic T4 (levothyroxine), has been used worldwide for over half a century. It relies on enzymes in the body's tissues, called deiodinases, to turn T4 into T3 where it is needed. The assumption is that a normal thyroid-stimulating hormone (TSH) result on levothyroxine means the tissues are getting enough T3.

"Combination therapy" means adding synthetic T3 (liothyronine) to levothyroxine. Desiccated thyroid extract, dried animal thyroid sold as Armour Thyroid or NP Thyroid, is an alternative that contains both hormones.

Who still feels unwell, and why

Roughly 5–15% of people on levothyroxine with a normal TSH report lasting symptoms. The complaints cluster: energy and fatigue (reported by 80–90% of them), weight (70–75%), memory (60–80%) and mood (40–50%).^[1]

The first step is ruling out other causes: iron or vitamin B12 shortage, sleep apnoea, depression or another autoimmune disease. After that, the leading but unproven explanation is that some people don't convert enough T4 to T3 inside their tissues, so their blood tests look fine while the tissues run short.^[2] People on levothyroxine alone do tend to run a lower T3-to-T4 ratio in their blood than untreated people with healthy thyroids, and a low ratio goes with classic underactive-thyroid complaints.^[3]

The gene test that didn't hold up

A common variant in the gene for one of the converting enzymes (DIO2, the Thr92Ala variant) was once the favoured explanation. An early secondary analysis of one trial suggested that people with two copies felt worse on levothyroxine and improved more when T3 was added; its authors said it needed replication.^[4]

It did not replicate. In a double-blind crossover trial of 75 patients the genotype made no difference to response,^[5] and among nearly 19,000 people treated with levothyroxine in UK Biobank, a large British health study, carriers had no worse well-being or thinking than non-carriers.^[6] It remains a plausible mechanism, not a predictor. The American Thyroid Association recommends against using genetic tests to choose treatment.^[7]

What the trials show

A small average benefit. A 2024 pooled analysis of 16 randomised trials of adding T3 or of desiccated thyroid, against levothyroxine alone, found the expected blood changes (free T4 down, total T3 up), and adding T3 gave a statistically significant improvement of about 3 points on a 28-item psychological distress questionnaire, a real but modest shift. Fatigue scales, cholesterol, resting heart rate and general quality of life often did not differ.^[8]

Some individuals may gain more. In the crossover trial above, people did equally well on either treatment as a group, but in a subgroup analysis the third with the most symptoms on levothyroxine alone preferred, and did better on, T3-containing therapy.^[9]

Timing is part of the problem. Oral T3 is short-acting: it produces an unnaturally high spike in blood T3 a few hours after a dose and a dip by afternoon. Splitting it into more than one daily dose helps; slow-release forms would help more but are not widely available.

Desiccated thyroid carries T4 and T3 at roughly 4 to 1, far richer in T3 than a human gland makes. Patients often like it, probably because of the morning T3 surge, and in the same pooled analysis it gave higher total T3 and slightly higher TSH than levothyroxine alone. But the trials are low quality, and the hormone ratio does not match human physiology.^[10]

Safety

Too much thyroid hormone is the main danger: over-replacement drives atrial fibrillation and bone loss, so TSH has to be monitored to make sure it isn't pushed below the normal range. The wider evidence on over-treatment is in Thyroid management.

At correct doses the data are broadly reassuring. In pooled trials, side effects were not clearly more common on combination therapy than on levothyroxine alone, and in observational data on over 600,000 people, T3 users showed no significant rise in atrial fibrillation, though the heart-failure estimate was borderline. The serious harms in case reports came from unregulated or compounded T3.^[11]

What to do

Rule out the usual suspects first: iron or B12 shortage, sleep apnoea, depression, insomnia and overtraining.
If they are clean, a trial is reasonable but not first-line. The European Thyroid Association advises that a specialist starts it at a T4-to-T3 ratio of about 13:1 to 20:1 by weight, splits the T3 into two daily doses, and stops it if nothing improves after 3 months.^[12]
Keep TSH in range. A suppressed TSH means too much hormone.
Use regulated products, not unregulated or compounded T3.
Skip the DIO2 gene test, and don't switch if you feel well on levothyroxine.