The Journal of Clinical Endocrinology & Metabolism Vol. 88, No. 10 4540-4542
Copyright © 2003 by The Endocrine Society
Editorial
In Search of the Impossible Dream? Thyroid Hormone Replacement
Therapy That Treats All Symptoms in All Hypothyroid Patients
Michael M. Kaplan, David H. Sarne and Arthur B. Schneider
Associated Endocrinologists and Departments of Medicine and Nuclear Medicine, William Beaumont Hospital (M.M.K), Royal Oak, Michigan 48073; and
University of Illinois at Chicago, Section of Endocrinology and Metabolism (D.H.S, A.B.S.), Chicago, Illinois 60612
Address all correspondence and requests for reprints to: Michael M. Kaplan, M.D., Associated Endocrinologists, 6900 Orchard Lake Road, Suite 203, West Bloomfield, Michigan 48322.
E-mail: mmkallegro@comcast.net.
We tell our patients, "It’s really quite simple, your thyroid is not working (or has been removed or destroyed by our treatment). The tablet contains the natural
hormone that your body cannot make. Don’t worry, you’ll be fine." For many of our patients, T4 therapy resolves their symptoms and they are fine. For some,
however, this therapy remains unsatisfactory, with the persistence of specific symptoms or a failure to regain a normal sense of well-being.
It is unclear how many patients fail to achieve satisfactory results, because dissatisfied patients are more likely to be referred or to seek advice from other physicians.
A survey conducted in Bristol, United Kingdom, attempted to determine the degree of dissatisfaction with T4 therapy (1). Records obtained from general
practitioners’ offices were used to create a roster of patients taking T4 and an equal number of controls. Two survey instruments, a well-established one related to
feelings of general well-being and a second one containing questions related to the symptoms associated with hypothyroidism, were administered. Patients on T4
scored worse on both surveys. For the thyroid-specific survey, 48.6% of the 397 responding patients with a recent normal TSH and 35.0% of the 551 responding
controls had scores indicating dissatisfaction with their health status. In other words, an excess of about 13 in every 100 patients had symptoms that might, in some
way, be related to their T4 therapy.
Why is this? One proposed explanation with support from both clinical and basic research, is that the thyroid produces both T4 and T3, whereas we use T4 alone for
treatment. In a study performed in Kaunas, Lithuania, 33 hypothyroid patients felt better and performed better, according to their scores on standardized tests, when
12.5 µg of T3 was given to replace 50 µg of their usual T4 dose (2). Lending credence to this clinical observation are studies in rats that show that in thyroidectomized
rats, normal tissue levels of T4 and T3 cannot be achieved by an infusion of T4 alone, but require an infusion of T4 and T3 (3, 4). Only the combined treatment with T4
and T3 ensures euthyroidism of the thyroidectomized rat, although cerebral cortical T3 tissue concentrations are normal over a wide range of T4 doses and a wide
range of serum T4 concentrations.
With a problem as important as this and an explanation as straightforward as this, why was the initial T3-substitution study met with so much reserve? As discussed
below, the study had a number of limitations. However, another reason may be that we have the preconceived conviction that there are other explanations and do not
feel the need for additional ones. First, many hypothyroid patients receive or, because of variable compliance, take insufficient or excessive doses of T4 (5). Second,
many hypothyroid patients are relatively young, and T4 may be the only medication that they take. Thus, it is understandable that they might ascribe a variety of
symptoms to their medication. Third, it is generally thought that some patients feel better when they are slightly hyperthyroid (6). It is often difficult to reduce the dose
of T4 in patients who have been overreplaced for a prolonged period of time, because they develop low energy and other symptoms. Finally, because symptoms
such as fatigue, constipation, and difficulty losing weight are common in the general population, it would be expected that they would occur in some hypothyroid
patients even after appropriate thyroid hormone replacement.
In this issue of JCEM, Walsh et al. (7) report on their attempt to replicate the Kaunas study in a group of patients from Western Australia using a similar design, a
blinded, crossover study with randomization of the order of the treatments, and Sawka et al. (8) report a comparison of T4 plus T3 vs. T4 alone, using a different
study design, a randomized double-blinded trial with two patient groups in Hamilton, Ontario. Both studies used a variety of standardized instruments to measure
psychological status, including quality of life, thyroid symptoms, and cognitive function. In contrast to the findings from Kaunas, the results of a series of primary and
subgroup analyses indicate that T3 substitution is no better, and in some comparisons, worse, than T4 alone.
What are the strengths and weaknesses of the three studies and what factors could explain the different results? Because the Kaunas investigation can be looked
upon as a pilot study, it may be understandable that power calculations were not performed at the outset. We must assume, however, that the study size was
predetermined and not extended during the study until positive observations were made. Because many measurements were made, another concern is that no
mention is made of correcting the statistical tests for multiple endpoints. However, many of the endpoints were statistically significant in the positive direction, making
it unlikely that the findings arose by chance.
Several cogent concerns have been raised about the Kaunas study. The patients were heterogeneous. One of the participants in the study was overtly hypothyroid,
and three had marked depression (9). About half were taking T4 for hypothyroidism arising from thyroid cancer treatment, and about half of these had fully
suppressed TSH levels. In a post hoc analysis (9), 11 patients with autoimmune thyroiditis had less mental improvement than 15 thyroid cancer patients. The average
baseline dose of T4, 175 µg/d, was high. Because the amount of T3 was fixed at 12.5 µg/d, the T3/T4 ratio varied according to the baseline dose of T4. T3 was given
once a day, resulting in a large diurnal variation. T3 levels were measured near the peak of the administered dose. There was no washout period between the two
doses, and the duration of treatment was short (5 wk).
The Western Australia study, an attempt to replicate the Kaunas findings but using 10 µg instead of 12.5 µg of T3 for 50 µg of T4, avoids many of these problems.
First, it is larger than the one from Kaunas, 101 vs. 33 patients. It is notable that the number of subjects was determined by power calculations for the principal
outcome analyses. Without this preplanning, negative findings, such as those observed in this case, would not be as convincing. However, the reported power
calculations were made at the outset of the study, based on certain assumptions about the variance of the measurements. Ideally, when a study is completed, these
assumptions should be checked with the actual observations, and power should be calculated again. The patients in this study were more homogeneous, but not
completely so, and the treatments were given for 10 wk, compared with 5 wk in the Kaunas study. The study avoided the problem of including patients being treated
for thyroid cancer with supraphysiological T4 doses, but it included about equal numbers of patients assessed to be satisfied and not satisfied with their T4 therapy.
This distinction was made by one observer, not by the patients, by criteria that are not presented. Also, the inclusion of such a high fraction of patients who were
classified as not satisfied suggests that the recruitment methods encouraged such patients to participate. The authors also included measurements of thyroid hormone
action in peripheral tissues. Of most concern to the integrity of the findings is that the average TSH level was higher after T3 substitution. The authors, realizing that
these factors could affect the findings, either before (satisfaction classification) or after (TSH levels) the study, performed subgroup analyses that confirm and
strengthen their conclusion. However, when subgroup analyses are performed, it is almost certain that the power will be reduced. Decreased statistical power may
not occur if the variance of the observations decreases, but it is best to calculate and report this in quantitative terms.
The Ontario study had a relatively homogeneous population of T4-treated subjects with noniatrogenic primary hypothyroidism, but specifically selected patients with
depressive symptoms not severe enough to meet criteria for major depression. In the T4 + T3 group, a T3 dose of 12.5 µg twice a day was substituted for half of the
prestudy T4 dose, and the T3 dose was adjusted in an unspecified manner to maintain normal TSH levels by an investigator who had no contact with the patients or
those performing the psychological tests. The study duration was 15 wk. No significant differences between the groups were found in any of the subscales of test
instruments used to access mood, depression, and general health status.
Although the Ontario study had only a few more subjects than the Kaunas study, the Ontario selection criteria presumably increased the likelihood of detecting
psychological benefit from combining T3 with T4. However, no power analysis was performed, several subjects in both groups did not complete the study, and there
was a substantial placebo effect, with significant improvement in symptoms in both groups during the study. Also, between-patient variability would make a
therapeutic difference harder to detect than a study in which patients can be their own controls.
In view of the results of the Western Australia study, the Ontario study, and the post hoc analysis of the Kaunas study, evidence is fading that adding T3 to T4 is
beneficial in the long-term treatment of hypothyroid patients with autoimmune thyroiditis. In addition, the possible long-term risks of elevated or fluctuating T3 levels
have not been evaluated. We do not believe that the current evidence supports the use of T3 for these patients, who are probably the largest group of hypothyroid
patients. The Kaunas study raises the question of whether patients who have had near-total thyroid ablation might respond differently, but the findings of this
short-term pilot study of a small group of patients, analyzed post hoc, are also insufficient to justify adopting a new long-term treatment regimen.
More needs to be done to understand why some patients do not feel completely well on what, according to current standards, is adequate thyroid hormone
replacement. First, careful cross-sectional and/or case-control epidemiological studies are needed to develop a standardized definition of cases, determine their
prevalence, and generate testable hypotheses. Second, efforts should continue to identify molecular measurements that indicate, directly or as surrogate markers,
whether tissue levels of thyroid hormone are normal.
Third, additional clinical studies with the following characteristics are needed: 1) The study population should be homogeneous. Although this may prevent
generalizing the findings, the current uncertainties necessitate starting in this way. 2) The sample size should be large enough to assure that either positive or negative
findings can be accepted with confidence. The principal psychological, physiological, and molecular endpoints should be selected with care, and the analysis should
take into account the effects of multiple testing. 3) If practical, a random order, double-blind crossover design should be used. 4) In studies to test T3, sustained
release preparations, if available, or divided doses should be used. Consideration needs to be given to the implications of a fixed vs. variable T4/T3 ratio in
combination therapy, both in regard to study design and therapeutic effect. 5) TSH should be monitored dynamically and study medications adjusted according to the
results, to maintain normal serum TSH concentrations.
Footnotes
This work was supported in part by National Cancer Institute Grant CA-21518 (to A.B.S.).
Received August 14, 2003.
Accepted August 14, 2003.
References
1.Saravanan P, Chau F, Roberts N, Vedhara K, Greenwood R, Dayan CM 2002 Psychological well-being in patients on ‘adequate’ doses of
L-thyroxine: results of a large, controlled community-based questionnaire study. Clin Endocrinol (Oxf) 57:577–585[CrossRef][Medline]
2.Buneviius R, Kaanaviius G, alinkeviius R, Prange Jr AJ 1999 Effects of thyroxine as compared with thyroxine plus triiodothyronine in patients with
hypothyroidism. N Engl J Med 340:424–429[Abstract/Free Full Text]
3.Escobar-Morreale HF, Obregón MJ, Escobar del Rey FE, Morreale de Escobar G 1995 Replacement therapy for hypothyroidism with thyroxine
alone does not ensure euthyroidism in all tissues, as studied in thyroidectomized rats. J Clin Invest 96:2828–2838[Medline]
4.Escobar-Morreale HF, Escobar del Rey FE, Obregón MJ, Morreale de Escobar G 1996 Only the combined treatment with thyroxine and
triiodothyronine ensures euthyroidism in all tissues of the thyroidectomized rat. Endocrinology 137:2490–2502[Abstract]
5.Canaris GJ, Manowitz NR, Mayor G, Ridgway EC 2000 The Colorado thyroid disease prevalence study. Arch Intern Med
160:526–534[Abstract/Free Full Text]
6.Carr D, McLeod DT, Parry G, Thornes HM 1988 Fine adjustment of thyroxine replacement dosage: comparison of the thyrotrophin releasing hormone
test using a sensitive thyrotrophin assay with measurement of free thyroid hormones and clinical assessment. Clin Endocrinol (Oxf) 28:325–333[Medline]
7.Walsh JP, Shiels L, Lim EM, Bhagat CI, Ward LC, Stuckey BGA, Dhaliwal SS, Chew GT, Bhagat MC, Cussons AJ 2003 Combined
thyroxine/liothyronine treatment does not improve well-being, quality of life, or cognitive function compared to thyroxine alone: a randomized controlled trial in
patients with primary hypothyroidism. J Clin Endocrinol Metab 88:4543–4550[Abstract/Free Full Text]
8.Sawka AM, Gerstein HC, Marriott MJ, MacQueen GM, Joffe RT 2003 Does a combination regimen of thyroxine (T4) and 3,5,3'-triiodothyronine
improve depressive symptoms better than T4 alone in patients with hypothyroidism? Results of a double-blind, randomized, controlled trial. J Clin Endocrinol
Metab 88:4551–4555[Abstract/Free Full Text]
9.Buneviius R, Prange AJ 2000 Mental improvement after replacement therapy with thyroxine plus triiodothyronine: relationship to cause of hypothyroidism.
Int J Neuropsychopharmacol 3:167–174[CrossRef][Medline]
