| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Endocrinology and Metabolism (B.C.A., E.F., W.M.W.), Psychiatry (E.M.W., A.H.S., J.H.), Cardiology (J.G.P.T.), Clinical Chemistry (E.E.), and General Practice (H.C.P.M.v.W.), Academic Medical Centre, University of Amsterdam, 1100 DE Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: Bente C. Appelhof, M.D., Academic Medical Center of the University of Amsterdam, Department of Endocrinology and Metabolism, F5-161, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands. E-mail: b.c.appelhof{at}amc.uva.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
2 test for trend, P = 0.024). This linear trend was not substantiated by results on any of the secondary outcome measures: scores on questionnaires and neurocognitive tests consistently ameliorated, but the amelioration was not different among the treatment groups. Median end point serum TSH was 0.64 µU/ml (mU/liter), 0.35 µU/ml (mU/liter), and 0.07 µU/ml (mU/liter), respectively [ANOVA on ln(TSH) for linear trend, P < 0.01]. Mean body weight change was +0.1, –0.5, and –1.7 kg, respectively (ANOVA for trend, P = 0.01). Decrease in weight, but not decrease in serum TSH was correlated with increased satisfaction with study medication. Of the patients who preferred combined LT4/LT3 therapy, 44% had serum TSH less than 0.11 µU/ml (mU/liter). Patients preferred combined LT4/LT3 therapy to usual LT4 therapy, but changes in mood, fatigue, well-being, and neurocognitive functions could not satisfactorily explain why the primary outcome was in favor of LT4/LT3 combination therapy. Decrease in body weight was associated with satisfaction with study medication. |
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Studies in thyroidectomized rats have shown that replacement therapy with levothyroxine (LT4) alone does not ensure euthyroidism in all tissues. Euthyroidism in all tissues could be achieved only by combined treatment with LT4 and liothyronine (LT3). These findings implicate that, in humans, standard LT4 therapy might not be sufficient to restore euthyroidism in all tissues either. The cerebral cortex, however, is able to maintain T3 homeostasis over a wide range of plasma T4 and T3 levels (2, 3).
In 1999 the results of a crossover trial investigating combined treatment of hypothyroidism with LT4 and LT3 were published. The authors concluded that substitution of 50 µg LT4 by 12.5 µg LT3 daily resulted in improved scores on mood scales and neurocognitive tests (4). These remarkable findings elicited quite a bit of discussion: is the current standard replacement therapy with LT4 alone the optimal treatment for hypothyroidism? In 2003 the results of three more trials investigating the value of combined treatment with LT4 and LT3, compared with LT3 alone, were published, but none of these replicated the finding of any advantageous effects of LT4/LT3 combined therapy on measures of well-being and neurocognitive functioning (5, 6, 7). An extensive review of the qualities and pitfalls of these studies was published in this journal (8).
A consistent limitation of these trials was that a fixed amount of LT4 was substituted with a fixed amount of LT3, leading to very variable ratios of LT4 and LT3 that are unlikely to have comparable effects. A more recent and equally negative trial did supply a fixed LT4 to LT3 molar ratio of 14:1 to a small group of patients with mainly postsurgery and postradioiodine hypothyroidism (35). In the present larger trial (n = 141), we studied whether combined treatment with LT4 and LT3 in any of two different weight ratios (5:1 and 10:1) was preferred over LT4 monotherapy in a homogeneous group of patients with primary autoimmune hypothyroidism.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subjects
Patients could participate if they were between 18 and 70 yr of age and had been on an adequate dose of LT4 replacement therapy for primary autoimmune hypothyroidism for at least 6 months. An adequate dose of LT4 was defined as resulting in a serum TSH between 0.11 and 4.0 µU/ml (mU/liter), as measured in the morning before LT4 intake.
Participants were excluded if they: 1) had a history of congenital hypothyroidism, hyperthyroidism, thyroidectomy, 131I-therapy, or thyroid cancer; 2) had angina pectoris (New York Heart Association functional class II or greater), paroxysmal supraventricular tachycardia, or any serious unstable medical condition; 3) were pregnant or within 6 months postpartum; and 4) had insufficient understanding of the Dutch language.
Patients were recruited from 13 general practices in the cities of Amsterdam and Almere. At these practices, prescribing records were checked to identify all patients receiving LT4 treatment (n = 590). Subsequently clinical records were checked for obvious/apparent inclusion and exclusion criteria, after which 303 seemingly eligible patients were invited by a letter from the general practice to participate in this clinical trial. Of these, 246 patients (81%) responded to the letter, 237 of whom agreed to receive detailed information about the trial. Three patients could not be reached after the initial contact, and 56 would or could not participate for personal reasons (e.g. not able to combine trial visits with work, no means of transportation, no interest in trial any longer). The remaining 178 were invited for a screening visit. If not adequately supplied with LT4, patients received a dose adjustment and were invited for a new screening visit 6 weeks or more later. Six patients failed to reach adequate LT4 supplementation before closure of the inclusion period. Another 31 had to be excluded from participation for the following reasons: no primary autoimmune hypothyroidism (n = 9), no current LT4 treatment (n = 5), LT4 treatment less than 6 months (n = 3), pregnancy (n = 5), insufficient understanding of Dutch language (n = 6), age (n = 1), angina pectoris (n = 1), or paroxysmal supraventricular tachycardia (n = 1). One hundred forty-one participated in the trial, 46 (33%) of these after dose adjustment(s) resulting in a serum TSH level within inclusion criteria.
Study design
During a screening visit at the Academic Medical Centre inclusion and exclusion criteria were checked, an electrocardiogram was made, and the presence of current major depressive disorder was assessed by means of the Structured Clinical Interview for Axis I Diagnostic and statistical manual of mental disorders-IV Disorders (9). If eligible, patients were randomly assigned to one of three treatment arms by means of a computer-generated list such that for every six patients, two were assigned to each treatment arm. Randomization was stratified for patients with current major depressive disorder (n = 10). Patients in arm 1 were assigned to receive LT4 only, patients in arm 2 received LT4 and LT3 in a ratio of 10:1, and patients in arm 3 received LT4 and LT3 in a ratio of 5:1. For patients in the combination treatment arms, the study medication dosage was calculated by subtracting 25 µg from their LT4 dose at time of inclusion, then adding the amount of LT3 conform treatment arm ratio, e.g. a patient adequately substituted with 100 µg LT4 at inclusion, assigned to the 5:1 ratio treatment arm, would receive 75 µg LT4 (100–25 µg) and 15 µg of LT3 (75:15 µg, equivalent with 5:1 ratio), whereas one assigned to the 10:1 ratio treatment arm would receive 75 µg LT4 (100–25 µg) and 7.5 µg LT3 (75:7.5 µg, equivalent with 10:1 ratio).
On the first day of the trial (baseline visit), patients arrived in the morning in fasting state. After laboratory testing and physical measurements, patients took their usual dose of LT4 and had breakfast at the Academic Medical Centre. About an hour later, they proceeded with the neurocognitive testing session, which lasted for approximately 1.5 h, and handed over the baseline set of questionnaires that they had filled out on the day before the baseline visit.
From the next day on, patients were instructed to take the first portion of study medication in the morning before breakfast and the second portion 12 h later. The daily dosage (LT4 or both LT4 and LT3) was divided into two portions and filled out in weekly medication blister packs.
After 5 wk of study medication, serum TSH was measured again at our laboratory in the morning. If needed, the study medication dose was adjusted: the LT4 dose was decreased with 12.5 µg if serum TSH was between 0.01 and 0.11 µU/ml (mU/liter) or with 25 µg if TSH was 0.01 µU/ml (mU/liter) or less. If serum TSH was more than 4.0 µU/ml (mU/liter), the LT4 dose was increased with 25 µg LT4. The LT3 dose was subsequently adjusted according to the ratio to which the participant was randomized.
In the LT4 group, a dose adjustment was needed in 15 of 45 participants (33%); in the 10:1 LT4/LT3 group, this was 20 of 44 (46%) and in the 5:1 LT4/LT3 group 27 of 46 (59%).
The adjusted study medication was given from wk 7 to 15. At both 5 and 10 wk after the baseline visit, patients filled out the set of questionnaires, and the main outcome, side complaints, and resting heart rate were assessed.
For the end point visit after 15 wk, the same time schedule was maintained as for the baseline visit, repeating all measurements in the same order and at the same time of day.
Neither the investigators nor the patients were aware of the treatment assignments throughout the trial.
Outcome measures
The main end point of the trial was the subjective appreciation of the study medication by the patient, which was rated on a 5-point scale as much better, somewhat better, the same, somewhat worse, or much worse, compared with their usual LT4 medication from before the trial at every visit. For the main outcome analysis, a dichotomy was made between those who preferred study medication over their usual medication (i.e. somewhat or much better) and those who did not. Patients were encouraged to volunteer positive and negative effects of the study medication at each visit.
Questionnaires
Well-being of patients was measured by means of a set of self-report questionnaires at baseline and after 5, 10, and 15 wk of treatment, filled out the day before the study visit. The set included the 32-item Profile of Mood States Dutch shortened version (POMS), designed to monitor changes in mood states (10); the original Dutch version of the Multidimensional Fatigue Inventory (MFI-20), consisting of 20 questions and designed to measure (changes in) fatigue (11); the mental health and vitality subscales of the Rand 36-item health survey (Rand-36) (12); and the (Dutch version of) the Symptom Checklist (SCL-90), a 90-item self-report scale containing eight subscales measuring multidimensional psychopathology (13).
Neurocognitive tests
Neurocognitive functioning was measured at baseline and after 15 wk of study medication and included tests of attention and working memory [Digit Span subtest of the Wechsler Adult Intelligence Scale (14)], learning and memory [Dutch version of the Story Recall from the Rivermead Behavioral Memory Test (RBMT) (15, 16) and the Dutch adaptation of the California Verbal Learning Test (CVLT) (17, 18)], psychomotor speed (Dutch adaptation of the Digit Symbol subtest of the Wechsler Adult Intelligence Scale-III) (14, 19), speed of memory processing [Memory Comparison Task (MCT), computer version (20) and paper-and-pencil version (21)], and attention [Dutch adaptation of the Paced Auditory Serial Attention Task (22, 23)]. All tests were administered by a trained psychometrician under supervision of a clinical neuropsychologist.
Biochemical measurements
All blood samples were collected in the morning before medication was ingested, approximately 24 h after the last usual medication dose (baseline measurements) and 12 h after the last study medication dose (at 5 and 15 wk).
Baseline and end point fasting laboratory tests included levels of serum TSH, free T4 (fT4), T3, SHBG, and anti-thyroid peroxidase antibodies (TPO-Ab), cholesterol and triglycerides, osteocalcin, and bone fraction of alkaline phophatase (skeletal AP).
Serum TSH and fT4 were measured by time-resolved fluoroimmunoassay (Wallac Oy, Turku, Finland), serum T3 by in-house RIA methods (24), SHBG by immunoradiometric assay (Farmos Diagnostica, Turku, Finland), TPO-Ab by chemiluminescence immunoassay (Brahms, Berlin, Germany), osteocalcin by immunoradiometric assay (INCSTAR, Stillwater, MN), skeletal AP by enzyme immunoassay (Alkphase-B; Metra Biosystems Inc., Mountain View, CA), and cholesterol and triglycerides by enzymatic colorimetric methods (Modular p800; Roche Diagnostics, Indianapolis, IN).
Intra- and interassay coefficients of variation were, respectively, 1–2 and 3–4% (TSH), 4–6 and 5–8% (fT4), 3–4 and 7–8% (T3), 2–5 and 3–6% (SHBG), 3–7 and 8–12% (TPO-Ab), 4–6 and 4–9% (osteocalcin), 4–7 and 6–9% (skeletal AP), 1–3 and 1–2% (cholesterol), and 1–2 and 1% (triglycerides). Detection limits were 0.01 µU/ml (mU/liter) for TSH, 0.2 ng/dl (2 pmol/liter) for fT4, 20 ng/dl (0.3 nmol/liter) for T3, 5 nmol/liter for SHBG, 30 kU/liter for TPO-Ab, 0.5 µg/liter for osteocalcin, 0.7 U/liter for skeletal AP, 1.0 mmol/liter for cholesterol, 0.1 mmol/liter for triglycerides.
Statistical analysis
Analyses for the primary outcome were performed according to the intention-to-treat principle. For all other analyses, last observations were carried forward; if no follow-up measurement was available for a certain parameter (because of drop-out or not completing the test for other reasons), the patient was excluded from the analyses of that particular parameter. Within-group baseline and end point scores were analyzed by means of paired t tests.
Because the primary outcome showed a linear trend over the three treatment groups in accordance with an increasing proportion of LT3, all statistical comparisons among the three treatment groups were analyzed by 2 for trend or ANOVA for linear trend. Serum TSH was log transformed to normalize the distribution before statistical analysis. Statistical significance was defined as a two-tailed P < 0.05. All statistical analyses were performed using SPSS (version 11.5; SPSS, Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). One participant (5:1 group) withdrew because of unexpected travel abroad for family matters and was excluded from all analyses. Seven patients withdrew because of side effects, four in the LT4 group and three in the 10:1 group. Various side effects were mentioned (e.g. fatigue, dizziness, muscle aches, irritability), but no specific complaints could be identified for those on combination therapy. Patients who withdrew because of side effects were considered not to prefer study medication to their usual treatment and were all included in the primary outcome analysis (n = 140). Three patients did not show up for the end point visit, one because of illness unrelated to study medication (5:1 group) and two because of personal time constraints (both 10:1 group). When available, follow-up data were carried forward. Baseline characteristics of the three groups were similar (Table 1). Ten patients with current major depressive disorder were included (three, four, and three in the LT4, 10:1, and 5:1 group, respectively).
|
|
End point serum TSH and fT4 were decreased in the combination therapy groups, compared with baseline, whereas serum T3 increased. The changes showed a clear and significant relationship with LT3 proportions in the study medication, with the most prominent changes in the 5:1 combination therapy group (Table 2). Regarding markers for peripheral thyrometabolic state, serum SHBG increased, cholesterol decreased, and triglycerides did not change in the combined treatment groups. Changes in osteocalcin and skeletal AP did not show a significant relationship with LT3 proportion in the study medication. However, skeletal AP and osteocalcin significantly increased for patients in the 5:1 ratio group (paired t tests).
|
). Blood pressure did not significantly change. Treatment preference, questionnaires, and neurocognitive function tests
Primary outcome analysis (Fig. 2) shows that study medication was preferred to usual treatment by 14 of 48 patients (29.2%) in the LT4 group, 19 of 46 patients (41.3%) in the 10:1 ratio group, and 24 of 46 patients (52.2%) in the 5:1 ratio group (2 test for trend, P = 0.024). This suggests a linear increase in satisfaction with study medication with an increasing proportion of LT3.
|
shows the baseline and change scores on the questionnaires. Compared with baseline, all three treatment groups showed improvements in all subscale scores of all questionnaires at end point, with the only exceptions of the MFI-20 physical fatigue subscale score for the 10:1 group and the SCL-90 agoraphobia subscale score for the LT4 group. Many within-group improvements were significant when baseline and end point scores were compared by means of paired t tests. There was no significant difference among the three groups in the mean improvement on any of the subscale scores.
|
shows the raw baseline and change scores of the neurocognitive tests. One of the test results (CVLT immediate recall) showed a significant linear relation with treatment but with the largest improvement in the LT4 group and the least in the 5:1 LT4/LT3 group. Many of the within-group test results were improved at end point, compared with baseline, several of them significantly.
|
Patients with current major depressive disorder (n = 10) had higher baseline scores on the questionnaires in comparison with those without depression (or lower in the case of vigor/vitality subscales), indicative of more complaints. The primary outcome was in favor of LT4/LT3 combination therapy: none of the three depressed patients having received LT4 only preferred study medication over usual therapy, whereas all three depressed patients who received LT4/LT3 in a 5:1 ratio preferred study medication over usual LT4 therapy. Of four patients in the 10:1 group, one preferred study medication, whereas three did not. Like in the total group, there were improvements on almost all questionnaire subscale scores but without clear differences among the treatment groups. As for the neurocognitive tests, most, but not all, baseline test results were worse for depressed patients, compared with nondepressed patients, and most scores were somewhat improved at end point. Again, there were no clear differences in change scores among the treatment groups.
To check whether the group with relatively many complaints would especially benefit from LT4/LT3 therapy, another subgroup analysis was performed on the tertile of patients with the highest SCL-90 total scores. However, results for this subgroup (n = 47: 18, 16, and 13 in the LT4, 10:1, and 5:1 groups, respectively) remained similar with those of the whole group.
In several cases end point TSH was suppressed, less than 0.11 µU/ml (mU/liter). This occurred in seven of 45 (16%) patients in the LT4 group, 13 of 44 (30%) in the 10:1 group, and 25 of 46 (54%) patients in the 5:1 ratio group, so there was a clear linear relation with treatment (2 test for trend, P < 0,01). Of the patients who preferred combined LT4/LT3 therapy, 44% had serum TSH less than 0.11 µU/ml (mU/liter). At 15 wk, serum TSH was above 4.0 µU/ml (mU/liter) in two (4%), three (7%), and three (7%) patients in the LT4, 10:1, and 5:1 groups, respectively.
A post hoc subgroup analysis (n = 90) was carried out in subjects in whom end point serum TSH was not suppressed. For this subgroup the median TSH at 15 wk was 0.84 µU/ml (mU/liter) for those who had received LT4 and 0.82 µU/ml (mU/liter) and 1.2 µU/ml (mU/liter) for the patients in the 10:1 and 5:1 LT4/LT3 combination groups, respectively (ANOVA for linear trend, P = 0.61). Mean fT4 was 1.15 ng/dl (14.8 pmol/liter), 0.92 ng/dl (11.9 pmol/liter), and 0.78 ng/dl (10.1 pmol/liter) and mean T3 levels 110 ng/dl (1.69 nmol/liter), 114 ng/dl (1.75 nmol/liter), and 127 ng/dl (1.95 nmol/liter) for patients in the LT4, 10:1 LT4/LT3, and 5:1 LT4/LT3 groups, respectively (ANOVA for linear trend, P < 0.01 and P = 0.01, respectively). SHBG did not differ among the groups. For the primary outcome, the results remained similar, with 29, 45, and 48% in the LT4, 10:1, and 5:1 treatment groups, respectively, preferring study medication over their usual mediation, but this was not significant anymore (2 test for trend, P = 0.13). Again, this linear trend was not found on any of the secondary outcome measures.
For the whole group, satisfaction with study medication was not correlated with either end point TSH or change in TSH. Increased preference for study medication as measured on the original five-point scale was correlated with weight loss (Pearson correlation between primary outcome and baseline-end point difference in weight: –0.188; P = 0.03).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This clinical trial investigating combined LT4/LT3 therapy is the largest thus far and the first to confirm a certain beneficial effect of combination therapy with both LT4 and LT3 since the publication by Bunevicius et al. (4) on this matter. Several of the methodological issues that have been raised in relation to that trial were avoided in the present trial. Bunevicius and Prange (25) included both patients with autoimmune hypothyroidism and those with thyroid carcinoma. In a later subgroup analysis, it appeared that only patients who had been treated for thyroid cancer benefited from LT4/LT3 therapy, whereas beneficial effects were absent in the group of patients with autoimmune hypothyroidism. In the present trial, we included a relatively large and homogeneous group of patients, all with autoimmune pathogenesis of hypothyroidism.
Because in the Dutch health care system, all inhabitants are enlisted with a family physician who will diagnose and treat most patients with hypothyroidism, we avoided selection bias by not recruiting from a second-line health care institution but from primary care. We invited all eligible patients to participate, regardless of their satisfaction with treatment.
Patients were treated for a period of 15 wk, which is long enough to reach a steady-state balance after an eventual dose adjustment (26). To attenuate the effect of the rapid absorption and short half-life of LT3, all study medication was divided into two equal daily portions. It has been argued that in studies on the effects of LT4 plus LT3, only sustained-release LT3 preparations should be used to avoid nonphysiological T3 peaks. We agree that this would probably better mimic the continuous physiological thyroid gland secretion. However, the fact is that sustained-release preparations are not commercially available as yet. A first report on the endocrine results of treatment with an in-house slow-release LT3 preparation has very recently been published, concluding that no T3 serum peaks are present with this preparation (27). Nevertheless, no chemical characteristics of the preparation were given, and even with this preparation, LT4/LT3 ratios were not truly physiological (28). Furthermore, endocrine data were available on the first 9 h post ingestion only, leaving uncertainties about the slow-release properties over 24 h.
We considered it crucial to supply fixed proportions of LT4 to LT3 because is seems unlikely that the effect of combination therapy would be independent of the proportion of LT3, which ranged from 3:1 to 15:1 in the trial by Bunevicius et al. (4). A molar ratio of 14:1 has been reported to approach the physiological production of the human thyroid gland (29), but we cannot be certain about the comparability of bioavailability of secreted vs. absorbed thyroid hormones due to the first-pass effect of the liver for one reason. Therefore, we chose to include two different ratios to further elucidate what proportion of LT3 is most adequate.
A limitation of this study is that the combined therapy medication regimens led to overtreatment in many patients, e.g. the median end point TSH in the 5:1 group was 0.07, which implies that serum TSH was suppressed less than 0.07 µU/ml (mU/liter) in half of the patients in this group. The elevated heart rate in the 5:1 LT4/LT3 group and the weight loss in both combination groups are consistent with some degree of overreplacement. Apparently the problem of overtreatment was not sufficiently avoided by the dose adjustment after 5 wk according to the study protocol. It has been reported that some patients achieve the desired sense of well-being only when taking LT4 in a dose of 50 µg in excess of that necessary to restore serum TSH to normal (30), so one might conclude that a certain degree of overtreatment results in satisfaction with study medication and might fully explain the primary outcome results. It is of interest that also in the study by Bunevicius et al. (4), a significant number of patients had suppressed TSH levels. However, in this study we found no correlation between change in serum TSH and the primary outcome.
The choice of outcome measures was complicated by the fact that, although the clinical notice that a substantial minority of hypothyroid patients remain with complaints is widespread, only sparse literature was available on the precise nature and prevalence of these complaints (31). By consequence, one could not be sure in what domains improvements were to be expected. We therefore chose to regard the subjective satisfaction with study medication as the primary outcome instead of one of the questionnaire or neurocognitive test outcomes because that choice would have been an arbitrary one. And in fact, it is the subjective dissatisfaction of biochemically adequately treated patients that forms the main incentive for this line of research. Only recently a community-based study confirmed impairment in psychological well-being of patients on adequate LT4 replacement, compared with controls of similar age and sex (1). Putting on weight was one of the four complaints that were clearly more prevalent in hypothyroid patients than controls. It is therefore interesting that we found weight loss to be correlated with satisfaction with study medication.
Another notable finding is the fact that there was an improvement in scores on virtually all subscales of all outcome measures in all three groups. Because improvements were not consistently related to the treatment arm, this is likely to result from a practice effect or Hawthorne effect: benefit from improved routine care within a trial and from being investigated (32). It has been noted before that many hypothyroid patients find their physicians to be unsympathetic and dismissive of their symptoms (1), which may explain the extent of these aspecific trial effects in this study investigating the possibility to relief these uncomprehended complaints.
As for the safety of LT4/LT3 combination therapy, the majority of patients who dropped out because of side complaints appeared to belong to the LT4 monotherapy group (four of seven), whereas no one in the 5:1 LT4/LT3 group dropped out because of side complaints. However, in the 5:1 group, one patient known to have recurrent episodes of atrial premature beats absent at baseline was found to have atrial premature beats after the first 5 wk of study medication, which disappeared again after a dose adjustment at 5 wk. Although the apparent overtreatment with our LT4/LT3 medication schedules did not result in an excess of withdrawals in these groups, one should be aware that exogenous subclinical thyrotoxicosis is a risk factor for atrial fibrillation and may possibly lead to osteoporosis (33).
Although the primary outcome was in favor of LT4/LT3 combination therapy, we believe the results of this study do not currently support LT4/LT3 therapy as a standard treatment of patients with hypothyroidism, given the fact that the subjective preference for combination therapy was not embodied by more objective secondary outcomes, whereas loss of body weight, the plausible explanation for the subjective preference, may be due (at least partly) to overtreatment. Nevertheless, the outcome of this study does not preclude the possibility that a certain subgroup of patients may benefit from combined LT4/LT3 therapy. Recently identified polymorphisms, i.e. in type 2 deiodinase, important in the regulation of T3 availability, may help to identify subgroups more likely to benefit from LT4/LT3 therapy (34).
In summary, we found that patients treated for autoimmune hypothyroidism preferred combination therapy with LT4 and LT3 over usual substitution therapy with LT4 alone. This was not substantiated by larger improvements on questionnaires measuring mood, fatigue, and well-being or on neurocognitive tests concerning attention and memory functions. However, we did find decrease in weight to be associated with the proportion of LT3 in substitution treatment as well as satisfaction with study medication, which might explain the satisfaction with LT4/LT3 combination therapy. A limitation of the study is that many patients in the combined therapy groups had end point TSH levels below the normally accepted range. We recommend that future studies assure that TSH levels are maintained within the reference range. These studies should take into account whether LT4/LT3 combination therapy improves control of body weight in patients with hypothyroidism. In addition, it may be that LT4/LT3 combination therapy proves beneficial for only a certain subgroup.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online February 10, 2005
Abbreviations: CVLT, California Verbal Learning Test; fT4, free T4; LT3, liothyronine; LT4, levothyroxine; MCT, Memory Comparison Task; MFI-20, Multidimensional Fatigue Inventory consisting of 20 questions; POMS, Profile of Mood States; Rand-36, Rand 36-item health survey; RBMT, Rivermead Behavioral Memory Test; skeletal AP, bone fraction of alkaline phophatase; TPO-Ab, thyroid peroxidase antibody.
Received October 26, 2004.
Accepted February 1, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Jonklaas, B. Davidson, S. Bhagat, and S. J. Soldin Triiodothyronine Levels in Athyreotic Individuals During Levothyroxine Therapy JAMA, February 20, 2008; 299(7): 769 - 777. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Slawik, B. Klawitter, E. Meiser, M. Schories, O. Zwermann, K. Borm, M. Peper, B. Lubrich, M. J. Hug, M. Nauck, et al. Thyroid Hormone Replacement for Central Hypothyroidism: A Randomized Controlled Trial Comparing Two Doses of Thyroxine (T4) with a Combination of T4 and Triiodothyronine J. Clin. Endocrinol. Metab., November 1, 2007; 92(11): 4115 - 4122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. T. Joffe, M. Brimacombe, A. J. Levitt, and A. Stagnaro-Green Treatment of Clinical Hypothyroidism With Thyroxine and Triiodothyronine: A Literature Review and Metaanalysis Psychosomatics, October 1, 2007; 48(5): 379 - 384. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Cooper-Kazaz, J. T. Apter, R. Cohen, L. Karagichev, S. Muhammed-Moussa, D. Grupper, T. Drori, M. E. Newman, H. A. Sackeim, B. Glaser, et al. Combined Treatment With Sertraline and Liothyronine in Major Depression: A Randomized, Double-blind, Placebo-Controlled Trial Arch Gen Psychiatry, June 1, 2007; 64(6): 679 - 688. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Saravanan, T. J. Visser, and C. M. Dayan Psychological Well-Being Correlates with Free Thyroxine But Not Free 3,5,3'-Triiodothyronine Levels in Patients on Thyroid Hormone Replacement J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3389 - 3393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Walsh, L. C. Ward, V. Burke, C. I. Bhagat, L. Shiels, D. Henley, M. J. Gillett, R. Gilbert, M. Tanner, and B. G. A. Stuckey Small Changes in Thyroxine Dosage Do Not Produce Measurable Changes in Hypothyroid Symptoms, Well-Being, or Quality of Life: Results of a Double-Blind, Randomized Clinical Trial J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2624 - 2630. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Grozinsky-Glasberg, A. Fraser, E. Nahshoni, A. Weizman, and L. Leibovici Thyroxine-Triiodothyronine Combination Therapy Versus Thyroxine Monotherapy for Clinical Hypothyroidism: Meta-Analysis of Randomized Controlled Trials J. Clin. Endocrinol. Metab., July 1, 2006; 91(7): 2592 - 2599. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M Wekking, B. C Appelhof, E. Fliers, A. H Schene, J. Huyser, J. G P Tijssen, and W. M Wiersinga Cognitive functioning and well-being in euthyroid patients on thyroxine replacement therapy for primary hypothyroidism Eur. J. Endocrinol., December 1, 2005; 153(6): 747 - 753. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Appelhof, R. P. Peeters, W. M. Wiersinga, T. J. Visser, E. M. Wekking, J. Huyser, A. H. Schene, J. G. P. Tijssen, W. J. G. Hoogendijk, and E. Fliers Polymorphisms in Type 2 Deiodinase Are Not Associated with Well-Being, Neurocognitive Functioning, and Preference for Combined Thyroxine/3,5,3'-Triiodothyronine Therapy J. Clin. Endocrinol. Metab., November 1, 2005; 90(11): 6296 - 6299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. F. Escobar-Morreale, J. I. Botella-Carretero, F. E. del Rey, and G. M. de Escobar Treatment of Hypothyroidism with Combinations of Levothyroxine plus Liothyronine J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4946 - 4954. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
