Bunevicius T4+T3 Trial: A Plain-English Overview of What It Established

At a glance

| Detail | Value | |---|---| | N | 33 (crossover, each patient served as own control) | | Intervention | Usual levothyroxine dose minus 50 mcg, plus 12.5 mcg liothyronine | | Comparator | Usual levothyroxine dose (T4 monotherapy) | | Duration | 5 weeks per treatment arm (10 weeks total) | | Primary endpoint | Composite of cognitive, mood, and quality-of-life assessments | | Key result | T4+T3 combination superior to T4 alone on multiple neuropsychological measures |

The Question the Trial Asked

By the late 1990s, levothyroxine monotherapy had been standard care for hypothyroidism for decades. The healthy thyroid gland secretes both thyroxine (T4) and triiodothyronine (T3), but clinicians relied on T4 alone, trusting peripheral tissue conversion to supply adequate T3. Most patients did fine. Some did not.

Robartus Bunevicius and colleagues at the Lithuanian Institute of Endocrinology, working with Arthur Prange Jr. at the University of North Carolina, asked a pointed question: would partially substituting T3 for a portion of a patient's T4 dose improve neuropsychological function and well-being in ways that T4 monotherapy could not?

Who Was Enrolled

The investigators recruited 33 adults (31 women, 2 men) with established hypothyroidism already on stable levothyroxine replacement. Their ages ranged from 24 to 72 years, with a mean around 46. All had been on a consistent T4 dose for at least the preceding several months, and all had TSH levels within the normal reference range at baseline.

Exclusion criteria were straightforward: active psychiatric illness requiring medication changes, pregnancy, and unstable cardiac disease. The population was intentionally narrow. These were euthyroid-on-paper patients, not undertreated ones. That matters because the trial was testing whether biochemical adequacy on T4 alone might still leave room for neuropsychological improvement.

What They Were Given

The trial used a double-blind, randomized crossover design. Each patient completed two five-week treatment periods separated by no washout (immediate crossover):

• T4-only period: Their usual levothyroxine dose, unchanged.
• T4+T3 period: Their usual levothyroxine dose reduced by 50 mcg, with 12.5 mcg of liothyronine added.

The substitution ratio deserves attention. The investigators chose a 50 mcg T4-to-12.5 mcg T3 swap based on the conventional bioequivalence estimate that 1 mcg of T3 is roughly equivalent to 4 mcg of T4 in suppressing TSH. This ratio has since been debated. Some pharmacologists argue the potency ratio is closer to 1:3, which would mean the combination arm was slightly more potent overall. Others note that the short half-life of T3 (roughly 6 to 24 hours versus T4's 7-day half-life) means a single daily dose of liothyronine produces a peak-and-trough pattern quite different from the steady-state achieved with T4 monotherapy.

Patients received identical-appearing capsules for both periods. Neither the subjects nor the clinicians performing assessments knew which period contained T3.

What Was Measured

The investigators deployed an unusually broad battery of neuropsychological and mood instruments:

| Domain | Instruments Used | |---|---| | Mood | Profile of Mood States (POMS), Visual Analogue Scales (VAS) for mood | | Cognition | Digit Span, Digit Symbol, Trail Making B, Controlled Oral Word Association | | Psychiatric symptoms | Beck Depression Inventory (BDI), Hamilton Depression Rating Scale, SCL-90 | | Quality of life / physical | VAS for physical status, clinical symptom questionnaires |

Thyroid function labs (serum TSH, free T4, total T3, total T4) were drawn at the end of each treatment period to confirm the biochemical effect of the substitution.

What Was Found

Neuropsychological Results

On the combination arm, patients scored better on 6 of 17 neuropsychological measures and showed no worse performance on any measure. The specific improvements:

| Measure | Direction of Effect | |---|---| | POMS total mood disturbance | Lower (better) on T4+T3 | | POMS anger | Lower on T4+T3 | | Beck Depression Inventory | Lower on T4+T3 | | Digit Symbol substitution | Higher (better) on T4+T3 | | VAS for composite physical status | Better on T4+T3 | | SCL-90 somatization | Lower on T4+T3 |

The magnitude of mood improvement was clinically meaningful. Mean BDI scores dropped from around 8 on T4 alone to around 5 on combination therapy, a shift that moves many patients from the "mild mood disturbance" range to "minimal." The Digit Symbol improvement suggested faster psychomotor processing speed, not just subjective well-being.

Thyroid Function Labs

Serum TSH remained within normal limits in both treatment periods, though it trended slightly lower on the combination arm. Total T4 and free T4 were predictably lower during the T4+T3 period (because patients were taking 50 mcg less levothyroxine). Total T3 was slightly higher on the combination arm. No patient became biochemically hyperthyroid or hypothyroid during either period.

Patient Preference

When asked at study end, the majority of participants preferred the combination treatment period, though blinding was maintained throughout.

Why This Trial Changed the Conversation

Before 1999, combination T4+T3 therapy was an almost forgotten relic of pre-synthetic thyroid hormone practice. Desiccated thyroid (Armour Thyroid, Nature-Throid) contained both hormones but had fallen out of favor with endocrinologists who considered levothyroxine monotherapy more precise and predictable.

The Bunevicius paper, published in the New England Journal of Medicine, carried the weight of the journal's prestige. It gave scientific legitimacy to a claim that patient advocacy groups had been making for years: that some hypothyroid patients feel unwell on T4 alone despite "normal" lab values. The publication generated significant media coverage and prompted a wave of replication attempts.

Limitations the Authors Acknowledged

Bunevicius and colleagues were transparent about several weaknesses:

Small sample size. Thirty-three patients is enough to detect a large effect in a crossover design but leaves the study underpowered for smaller, clinically relevant differences. Any single outlier patient could meaningfully shift the averages.

Short duration. Each treatment period lasted only five weeks. Whether the improvements would persist, amplify, or fade over months was unknown. Five weeks is also barely enough time for full steady-state equilibration of thyroid hormone levels in peripheral tissues.

No washout period. The immediate crossover introduces the possibility of carryover effects. If mood or cognition changes from period one lingered into period two, the results for the second period could be contaminated.

Single daily T3 dosing. Liothyronine's short half-life means a once-daily dose produces a serum T3 spike within hours, followed by declining levels. This does not replicate the relatively steady T3 supply from peripheral conversion of T4 in healthy individuals. Whether the cognitive improvements were driven by the peak T3 level, the average T3 level, or simply having any T3 present remains unclear.

Homogeneous population. The cohort was predominantly female, Lithuanian, and middle-aged. Generalizability to other demographics was untested.

What Happened After: The Replication Problem

The Bunevicius trial prompted at least 13 subsequent randomized controlled trials testing T4+T3 combination therapy. The results were mixed, and the majority failed to replicate the original findings.

A 2006 meta-analysis by Grozinsky-Glasberg et al. pooled data from 11 RCTs totaling over 1,000 patients and found no consistent benefit of combination therapy on mood, cognition, or quality of life. A 2012 European Thyroid Association (ETA) guideline acknowledged that while a subset of patients might benefit from combination therapy, the evidence did not support routine use.

The 2014 American Thyroid Association (ATA) guidelines took a similar position: T4 monotherapy remains the standard of care. The ATA noted that combination therapy could be considered as an experimental approach in patients who remain symptomatic despite adequate TSH normalization, but stopped short of a firm recommendation.

Several explanations have been proposed for why the replications diverged from the original:

• Dose ratios varied widely. Some follow-up trials used different T4:T3 substitution ratios, making direct comparison difficult.
• Outcome measures differed. Not all trials used the same neuropsychological batteries, reducing the ability to pool results.
• Population genetics. The DIO2 gene, which encodes the type 2 deiodinase enzyme responsible for converting T4 to T3 in the brain, has a common polymorphism (Thr92Ala) carried by roughly 16% of the population. Some researchers have hypothesized that carriers of this variant may convert T4 to T3 less efficiently and therefore derive greater benefit from exogenous T3. A 2009 study by Panicker et al. found that DIO2 genotype influenced psychological well-being on thyroid replacement, though the finding has not been consistently replicated.

What This Means for Clinical Practice Today

The Bunevicius trial does not prove that all hypothyroid patients should receive T3. It does establish that the question is worth asking. For the clinician managing a patient who remains symptomatic on levothyroxine despite textbook-normal thyroid labs, the trial provides a rationale for a carefully monitored T4+T3 trial, particularly when other causes of persistent symptoms have been excluded.

Current guidelines from both the ATA and ETA do not endorse routine combination therapy. They do leave room for individualized trials. In practice, many endocrinologists will attempt a 3-month combination trial for persistently symptomatic patients, using a low-dose T3 substitution (typically 5 to 10 mcg) while reducing T4 proportionally, with close TSH monitoring to avoid iatrogenic thyrotoxicosis.

Sustained-release T3 formulations, which would smooth the peak-and-trough problem of current liothyronine tablets, remain in development but are not yet commercially available in the United States. If such a formulation reaches market, it could change the risk-benefit calculus for combination therapy substantially.

The trial's most durable contribution may be philosophical rather than pharmacological: it forced the endocrinology community to reckon with the possibility that a normal TSH does not always equal a well patient.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. 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
2. 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
3. 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
4. Jonklaas J, Bianco AC, Bauer AJ, et al. Guidelines for the treatment of hypothyroidism. Thyroid. 2014;24(12):1670-1751. PubMed
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 in hypothyroid patients. J Clin Endocrinol Metab. 2009;94(5):1623-1629. PubMed
6. Levothyroxine sodium prescribing information. U.S. Food and Drug Administration. FDA Label