The Challenges and Complexities of Thyroid Hormone Replacement

Shayri M. Kansagra, BS; Christopher R. McCudden, PhD; Monte S. Willis, MD, PhD


Lab Med. 2010;41(6):229-348. 

In This Article

Challenges to Combination Therapy

These initial reports suggested that a change in the treatment for primary hypothyroidism should be considered. Consequently, a number of other investigators sought to confirm these findings (Table 1). Throughout these papers are key differences in study design. In particular it is important to note the type of hypothyroidism in the study population and the time period for different treatments. We therefore provide a high level of detail necessary to allow the reader to adequately compare the various reports.

One of the first follow-up studies was a randomized, double-blind, placebo-controlled trial performed by Clyde and colleagues. The study included 46 active-duty military personnel and members of their families (age range: 24–65 years) with primary hypothyroidism who had been treated for at least 6 months with levothyroxine (T4 mono-therapy).[40] In this study, patients received either their usual dose of levothyroxine or T3/T4 combination therapy, which was achieved by reducing the levothyroxine dose by 50 μg daily and replacing it with liothyronine (the l-isomer of T3) at a dosage of 7.5 μg twice daily for 4 months.[40] These investigators intentionally avoided the crossover design used by the Bunevicius team in their 1999 study to preclude the possibility of a "testing effect" influencing patient performance. The testing effect refers to the relative improvement patients can display simply as a result of repeating the same cognitive test. Thus, in the study by Clyde and colleagues, half of the patients were assigned to levothyroxine alone and the other half to T4/T3 combination therapy. Hypothyroid-specific health-related QOL, as well as body weight, serum lipids, and neuropsychological factors, were evaluated before and after treatment.[40] No significant difference between levothyroxine monotherapy and combination levothyroxine/liothyronine therapy were identified for any of these factors. Notably, the lack of improvement in psychological test results[40] was in sharp contrast to the findings of the Bunevicius team.[19,44] It was discussed that these differences may be explained by significant testing effects in the Bunevicius study. Specifically, 76% of the thyroid cancer patients were randomized to the control group for the first phase of the Bunevicius crossover study. This group was therefore subject to repeat testing after crossing to the combined therapy increasing the likelihood of improvement on T4/T3 simply by experimental design.

In another study, Sawka and colleagues studied 40 individuals with symptoms of depression who were taking levothyroxine for primary hypothyroidism (Table 1).[41] The investigators randomized these patients into either T4 plus a placebo or T4/T3 combination therapy for 15 weeks. Combination therapy was achieved by reducing their pre-study T4 dose by 50% and adding 12.5 μg of T3 twice daily. The T4 and T3 doses were adjusted to maintain target TSH levels. Extensive assessments of self-rated mood and well-being again did not identify any differences in symptoms between the 2 treatment groups.[41] The study design of the Sawka group was considered an improvement over that of the original Bunevicius paper in several respects: (1) stable doses of T4 prior to the study; (2) patients with symptoms of depression who would be the most likely to exhibit improvement; and (3) avoidance of overtreatment (many of the Bunevicius study patients had suppressed TSH indicating excess thyroid hormone). With this study, evidence was beginning to mount against T3/T4 combination therapy. Other studies that also failed to detect improvement in mood and neurocognitive measures after combination therapy are summarized in Table 1 and discussed below.

At the same time as the report by Sawka and colleagues described above, there was a double-blind, randomized, controlled trial with a crossover design published by Walsh and colleagues (Table 1).[43] In this study, patients with poor QOL scores and relatively severe symptoms were selected. Of the 110 patients enrolled in this study, 85% (n=94) had autoimmune or idiopathic hypothyroidism, and the remaining 15% (n=12) had postsurgical hypothyroidism, a history of Graves' disease, or Hashimoto's disease. The daily T4 dose taken by each patient was reduced by 50 μg and replaced with either 50 μg of T4 (monotherapy) or 10 μg liothyronine (combination therapy).[43] Each treatment was given for a 10-week period, and these 2 periods were separated by a 4-week washout period, during which patients resumed their usual T4 dosage. No significant difference in cognitive function, QOL, Thyroid Symptom Questionnaire, subjective satisfaction with treatment, or 8 of 10 VAS scores was identified between treatment groups.[43] In the discussion, Walsh and colleagues highlighted the importance of the treatment and washout period lengths. They stated that the study by Bunevicius did not allow a long enough time period (only 5 weeks) to achieve equilibrium or to allow the effects of the previous therapy to clear (no washout).

In another trial, Siegmund and colleagues conducted a double-blind, randomized, controlled crossover trial[42] in 23 patients with hypothyroidism due to surgery/radioablation (n=21) or autoimmune thyroiditis (n=2) (Table 1). The patients received either 100% of their previous T4 dose or 95% of the previous dose with the remaining 5% replaced by T3 (combination therapy). They found that TSH levels were suppressed to a greater degree with combination therapy than with T4 therapy alone, but measurements of mood and cognitive function did not differ between the 2 treatment groups.[42] Interestingly, mood was significantly impaired (n=8) and subclinical hyperthyroidism (characterized by an increase in steady-state free T3 levels) was identified in the patients taking combination therapy.[42] These authors again cited the short treatment period and thyroid cancer population used in the Bunevicius study to explain the differences between their results and the landmark report. Finally, Rodriguez and colleagues focused on a reduction in fatigue as a clinical end point of T4/T3 combination therapy,[45] with secondary end points consisting of improvement in depressive symptoms and working memory, as well as the serum thyroid hormone profile and other physical parameters (Table 1).[45] They selected 30 patients with a diagnosis of primary hypothyroidism stabilized with T4 monotherapy from diabetes and endocrinology clinics. The patients were screened for evidence of significant fatigue and symptoms of depression or anxiety. The investigators assigned patients into the following 2 categories: (1) 14 patients to a normal dose of T4 plus a placebo and (2) 16 patients to a normal T4 dose minus 50 μg combined with 10 μg of T3.[45] Importantly, they used a 6-week treatment period (versus 5 weeks in the Bunevicius trial) allowing T4 hormone levels to reach a steady state, and added a washout period for patients receiving combination therapy before crossing over to the opposite treatment. While only 27 patients completed the trial, there were no significant differences in fatigue or symptoms of depression identified between treatment groups suggesting that combination therapy was not significantly better than T4 monotherapy.[45] Collectively, this group of studies largely refuted the initial landmark report that combination T4/T3 therapy improves neurocognitive measures.