How many patients were in the Bunevicius T4+T3 trial?

Thirty-three patients completed the crossover study. All were women with hypothyroidism due to autoimmune thyroiditis or thyroidectomy, recruited from a single Lithuanian institution.

Did subsequent trials confirm the Bunevicius results?

No. At least 11 subsequent RCTs and one major meta-analysis failed to show consistent benefits of T4/T3 combination over T4 monotherapy in unselected hypothyroid populations. The reasons for non-replication remain debated.

Do any guidelines recommend T4/T3 combination therapy?

The 2012 European Thyroid Association guidelines allow a 3-month experimental trial in persistently symptomatic patients. The ATA considers it experimental and does not recommend routine use, while acknowledging it may be considered on an individual basis.

What is the DIO2 polymorphism and why does it matter here?

The DIO2 Thr92Ala polymorphism may reduce intracellular conversion of T4 to T3 in brain tissue. Homozygotes (about 16% of the population) could theoretically benefit more from exogenous T3 supplementation, explaining why only a subgroup responds.

What T4:T3 ratio should be used if prescribing combination therapy?

Current ETA guidance recommends a 13:1 to 20:1 ratio by weight. The original Bunevicius substitution was approximately 4:1, which is considered more aggressive than what most guidelines now suggest.

Is liothyronine FDA-approved for hypothyroidism?

Yes. Cytomel (liothyronine sodium) is FDA-approved for hypothyroidism. However, combination T4/T3 therapy as studied by Bunevicius is an off-label application of standard approved drugs rather than a separately approved regimen.

What are the cardiac risks of adding T3?

Liothyronine produces serum T3 peaks that may increase heart rate and cardiac workload. Patients with atrial fibrillation, coronary artery disease, or heart failure require careful risk-benefit discussion before any T3 supplementation.

How long should a T4/T3 combination trial last before deciding it failed?

Most experts recommend 8 to 12 weeks minimum. The original Bunevicius trial used only 5-week arms. If no subjective or measurable improvement occurs by 12 weeks, reverting to T4 monotherapy is reasonable.

Why do many patients report feeling better on T3 despite negative trial data?

Possible explanations include placebo response (mood endpoints are highly susceptible), genetic subgroup effects (DIO2 variants), the psychoactive quality of T3 peaks, and selection bias (patients who seek T3 are often those most dissatisfied with current treatment).

Should I ask my doctor about T3 if my TSH is normal but I still feel tired?

Fatigue with normal TSH has many causes unrelated to thyroid hormone conversion. Iron deficiency, sleep disorders, depression, and adrenal insufficiency should be evaluated first. If all other causes are excluded and cognitive or mood symptoms persist, discussing a monitored T3 trial with your endocrinologist is reasonable.

What Bunevicius T4+T3 Actually Changes in Clinical Practice

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |-----------|--------| | N | 33 (crossover, each patient served as own control) | | Intervention | Levothyroxine dose reduced by 50 mcg, replaced with 12.5 mcg liothyronine | | Comparator | Levothyroxine monotherapy (usual dose) | | Duration | 5 weeks per arm, 5-week crossover | | Primary endpoint | Composite cognitive, mood, and quality-of-life battery | | Key result | T4/T3 combination superior on 6 of 17 neuropsychological measures; patients preferred combination 2:1 | | Publication | NEJM 1999; 340:424-429 |

Why This Trial Still Gets Cited in Clinic

Before February 1999, combination T4/T3 prescribing was essentially a relic of desiccated thyroid extract. Synthetic levothyroxine monotherapy had become standard practice based on the assumption that peripheral deiodinase conversion supplies adequate T3 to all tissues. Bunevicius et al. challenged that assumption directly, and the results appeared in the most influential medical journal in the world.

The clinical question was simple: do hypothyroid patients on adequate T4 replacement still have residual symptoms because their brains are not getting enough T3? The answer, at least in this cohort, was yes. That answer set off a chain reaction in endocrinology that has not fully resolved.

Methodology Worth Understanding

Design Strengths

The crossover structure was the right choice for a small sample. Each patient received both treatments in random order, eliminating inter-individual variability. Washout was handled by the 5-week treatment periods themselves, given T4's long half-life (~7 days) and the relatively short half-life of T3 (~1 day). The investigators maintained TSH within reference range during both arms, which meant any cognitive or mood differences could not be attributed to overt thyroid status changes.

Design Limitations That Matter Clinically

The 33-patient sample came from a Lithuanian population with autoimmune thyroiditis or post-thyroidectomy hypothyroidism. Several methodological issues reduce the trial's generalizability:

No blinding verification. Liothyronine has a noticeable onset (patients sometimes report feeling "sharper" within hours). The investigators did not assess whether patients could guess their treatment arm.
Fixed substitution ratio. Replacing 50 mcg T4 with 12.5 mcg T3 assumes a 4:1 potency ratio. Actual bioequivalence is debated and likely varies by individual.
Short duration. Five weeks may capture acute mood effects but cannot address durability, long-term cardiac safety, or bone metabolism consequences.
Composite outcome inflation. With 17 separate neuropsychological tests, 6 reaching significance represents a concerning multiple-comparisons problem. No correction (Bonferroni or otherwise) was applied.
Population homogeneity. All subjects were women ages 28 to 64, all from one institution. Men, elderly patients, and those with comorbid psychiatric illness were absent.

Results in Detail

Neuropsychological Battery

| Measure | T4 Alone (mean) | T4+T3 (mean) | p-value | |---------|-----------------|---------------|---------| | Profile of Mood States (total) | Higher (worse) | Lower (better) | <0.05 | | Visual scanning speed | Slower | Faster | <0.01 | | Digit Symbol (WAIS-R) | Lower | Higher | <0.05 | | Beck Depression Inventory | Higher | Lower | <0.05 | | Spielberger Anxiety (state) | Higher | Lower | <0.05 | | Analog scales (well-being) | Lower | Higher | <0.001 |

The remaining 11 measures (memory tasks, reaction time, some attention subtests) showed no significant difference. This pattern suggests T3 supplementation may preferentially affect mood and processing speed rather than declarative memory.

Patient Preference

When unblinded at study end, 20 of 33 patients preferred the combination period. Eleven preferred T4 alone. Two had no preference. This 2:1 ratio is clinically meaningful but could reflect placebo response or unblinding artifact.

Thyroid Biochemistry

TSH remained within normal range in both arms. Free T4 was predictably lower during the combination arm. Free T3 was not significantly different between arms in most samples, raising questions about whether the mechanism is truly a T3 supply issue or something else entirely (timing of T3 peaks, intracellular T3 availability, or ratio effects).

What Changed in Guidelines

ATA/AACE Response

The 2012 ATA/AACE guidelines for hypothyroidism acknowledged the Bunevicius data but recommended against routine T4/T3 combination therapy. The reasoning: subsequent larger trials (Sawka 2003, Walsh 2003, Clyde 2003, Saravanan 2005) failed to replicate the benefits. The guidelines assigned a "weak recommendation, moderate-quality evidence" against combination therapy.

However, the 2014 ATA guidelines softened slightly, stating that combination therapy could be considered as an "experimental" approach in patients who remain symptomatic despite optimal TSH on T4, provided they are informed of the limited evidence and monitored carefully.

ETA Position

The 2012 European Thyroid Association guidelines took a more permissive stance: a 3-month therapeutic trial of T4/T3 combination could be offered to patients with persistent complaints, using a T4:T3 ratio of 13:1 to 20:1 by weight, with close monitoring. This was the first major society guideline to create an explicit pathway for combination prescribing.

The DIO2 Polymorphism Question

The most scientifically productive legacy of the Bunevicius trial was stimulating research into why some patients respond to combination therapy while most do not. The type 2 deiodinase gene (DIO2) Thr92Ala polymorphism, present in roughly 16% of the population as homozygotes, may impair intracellular T4-to-T3 conversion in brain tissue.

A 2009 study by Panicker et al. found that patients homozygous for Thr92Ala showed greater psychological improvement on combination therapy. This finding has not been consistently replicated, but it offers a biological rationale for why the Bunevicius results could be real in a subpopulation while failing in larger, unselected cohorts.

What This Means for Prescribing Today

Patients Who Resemble the Trial Population

The strongest (though still limited) case for a combination trial exists in:

Women with autoimmune hypothyroidism or post-thyroidectomy status
TSH well-controlled on levothyroxine monotherapy
Persistent cognitive complaints or mood symptoms despite euthyroid labs
No cardiac contraindications (atrial fibrillation, coronary artery disease, osteoporosis)
Willingness to accept an evidence base of small, conflicting trials

Practical Prescribing Considerations

If attempting combination therapy based on this evidence:

Ratio: The ETA recommends starting at 13:1 to 20:1 (T4:T3 by weight). The Bunevicius substitution was approximately 4:1, which is more aggressive than current recommendations.
Formulation: Liothyronine's short half-life (~6 hours to peak) creates T3 spikes. Splitting into twice-daily dosing or using sustained-release compounded T3 (not FDA-approved) partially addresses this. Cytomel (liothyronine sodium) is the only FDA-approved T3 product.
Monitoring: Check free T3 4 to 6 hours post-dose, TSH at trough. Suppress TSH only with patient consent after discussing bone and cardiac risks.
Duration: Give at least 8 to 12 weeks before judging efficacy. The Bunevicius trial used only 5 weeks per arm.
Exit plan: If no subjective or objective benefit by 12 weeks, revert to T4 monotherapy.

Who Should Not Be Started on Combination Therapy Based on This Trial

Patients with uncontrolled atrial fibrillation or unstable angina (T3 peaks increase cardiac workload)
Osteoporotic patients or those with low bone density (suppressed TSH accelerates bone loss)
Patients whose primary complaint is fatigue without cognitive or mood specificity (fatigue has dozens of causes unrelated to T3 supply)
Anyone with an unstable TSH who has not yet been optimized on T4 alone

The Replication Problem

Between 2003 and 2009, at least 11 randomized controlled trials attempted to replicate the Bunevicius findings. The majority showed no benefit of combination therapy over T4 alone. A 2006 meta-analysis by Grozinsky-Glasberg found no statistically significant advantage for combination therapy on any outcome.

Why the discrepancy? Possible explanations include:

Dose and ratio differences. Many replication attempts used lower T3 doses or different substitution ratios.
Population heterogeneity. Bunevicius enrolled symptomatic patients; several replication trials included patients with no residual complaints.
Genetic subgroups. If only DIO2 Thr92Ala homozygotes benefit, a trial enrolling the general hypothyroid population would dilute the signal below detection in a typical sample size.
Placebo response magnitude. Mood and cognition endpoints are highly susceptible to placebo effects, which may have inflated the Bunevicius results in the absence of verified blinding.

Limitations the Authors Acknowledged

Bunevicius et al. were transparent about the small sample size and preliminary nature of the findings. They explicitly called for larger, longer studies. They noted the all-female cohort and acknowledged that the fixed-dose substitution was a simplification. The paper's discussion section warned against premature changes in clinical practice.

This measured tone was largely ignored by media coverage and patient advocacy groups, who interpreted the trial as definitive proof that millions of hypothyroid patients were being undertreated. The resulting patient demand for T3 prescriptions outpaced the evidence by a wide margin.

The Bottom Line for 2026

Twenty-seven years after publication, Bunevicius T4+T3 remains a hypothesis-generating study, not a practice-changing one. It asked the right question. It produced a signal that could not be reliably reproduced in larger populations. It spawned productive research into deiodinase genetics and tissue-specific thyroid hormone action.

For the clinician sitting across from a patient who has read about T3 online: this trial supports a time-limited, carefully monitored therapeutic trial in selected patients. It does not support routine combination prescribing for all hypothyroid patients. The gap between those two statements is where most clinical disagreement lives.

Frequently asked questions

›

References

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. PubMed
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. PubMed
Grozinsky-Glasberg S, Fraser A, Nahshoni E, Weizman A, Leibovici L. Thyroxine-triiodothyronine combination therapy versus thyroxine monotherapy for clinical hypothyroidism: meta-analysis of randomized controlled trials. J Clin Endocrinol Metab. 2006;91(7):2592-2599. PubMed
Wiersinga WM, Duntas L, Fadeyev V, Nygaard B, Vanderpump MP. 2012 ETA guidelines: the use of L-T4 + L-T3 in the treatment of hypothyroidism. Eur Thyroid J. 2012;1(2):55-71. PubMed
Jonklaas J, Bianco AC, Bauer AJ, et al. Guidelines for the treatment of hypothyroidism. Thyroid. 2014;24(12):1670-1751. PubMed
Cytomel (liothyronine sodium) prescribing information. U.S. Food and Drug Administration. FDA Label