Selenium supplementation and placebo are equally effective in improving quality of life in patients with hypothyroidism

Camilla Bøgelund Larsen*, Kristian Hillert Winther, Per Karkov Cramon, Åse Krogh Rasmussen, Ulla Feldt Rasmussen, Nils Jakob Knudsen, Jakob Bue Bjorner, Lutz Schomburg, Kamil Demircan, Thilo Samson Chillon, Jeppe Gram, Stinus Hansen, Frans Hjelm Brandt Kristensen, Birte Nygaard, Torquil Watt, Laszlo Hegedüs, Steen Bonnema

*Kontaktforfatter

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

42 Downloads (Pure)

Abstract

Abstract
Purpose: We investigated whether selenium supplementation improves quality-of-life (QoL) in
patients with autoimmune thyroiditis (ID:NCT02013479).
Methods: We included 412 patients ≥18 years with serum thyroid peroxidase antibody (TPOAb)
level ≥100 IU/mL in a multicentre double-blinded randomised clinical trial. The patients were
allocated 1:1 to daily supplementation with either 200 μg selenium as selenium-enriched yeast
or matching placebo tablets for 12 months, as add-on to levothyroxine (LT4) treatment. QoL,
assessed by the Thyroid-related Patient-Reported-Outcome questionnaire (ThyPRO-39), was
measured at baseline, after six weeks, three, six, 12, and 18 months.

Results: In total, 332 patients (81%) completed the intervention period, of whom 82% were
women. Although QoL improved during the trial, no difference in any of the ThyPRO-39 scales
was found between the selenium group and the placebo group after 12 months of intervention.
In addition, employing linear mixed model regression no difference between the two groups
was observed in the ThyPRO-39 composite score (28.8 [95%CI:24.5-33.6] and 28.0 [24.5-33.1],
respectively; P=0.602). Stratifying the patients according to duration of the disease at inclusion,
ThyPRO-39 composite score, TPOAb level, or selenium status at baseline did not significantly
change the results. TPOAb levels after 12 months of intervention were lower in the selenium
group than in the placebo group (1995 [95%CI:1512-2512] vs. 2344 kIU/L [1862-2951]; P=0.016)
but did not influence LT4 dosage or free triiodothyronine/free thyroxin ratio.

Conclusion: In hypothyroid patients on LT4 therapy due to autoimmune thyroiditis, daily
supplementation with 200 μg selenium or placebo for 12 months improved QoL to the same
extent.
Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024 Page 2 of 33
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
Introduction
Autoimmune thyroiditis (AIT) is the leading cause of hypothyroidism in iodine-sufficient
countries [1]. AIT is characterised by lymphocytic infiltration of the thyroid gland and the
presence of autoantibodies against thyroid peroxidase (TPOAb) and/or thyroglobulin [2]. The
prevalence of overt hypothyroidism ranges between 0.2% and 5.3% in Europe, depending on
geography, gender, age, and genetic factors [1]. The standard treatment is life-long
levothyroxine (LT4) substitution with adjusted dosage to normalise circulating thyrotropin (TSH)
levels [3]. However, despite adequate treatment, quality of life (QoL) remains impaired in a
subset of patients [4-6], and excess morbidity and mortality persist [7, 8]. The fact that the
aetiology of AIT is multifactorial, with complex interactions between genetic and environmental
factors, as well as personality traits [9], further emphasises the need to search for ways of
improving treatment [10].
Selenium is a metalloid and an essential micronutrient with fundamental importance to
human health [11]. The selenium concentration in the thyroid gland is higher than in most other
organs. In addition, selenium has been identified in the active site of iodothyronine deiodinases
type 1 and 2, which catalyse the conversion of thyroxin (T4) to triiodothyronine (T3), thus
establishing the importance of selenium status for thyroid metabolism [11, 12]. These findings
have led to investigations into the clinical relevance of selenium status in thyroid diseases, and
several selenium supplementation studies in patients with AIT have been carried out during the
last 20 years. Meta-analyses have supported that supplementation with selenium, organic [13]
as well as inorganic [14], leads to a reduction in TPOAb levels [15, 16]. However, in a systematic
review and meta-analysis, we concluded that there is no evidence of effects on clinically
relevant outcomes such as disease remission, progression, lowered LT4 dose, or improved QoL
[2, 17]. In the current international guidelines, selenium supplementation is not recommended
for the management of hypothyroidism [3, 18].
Information on the use of selenium supplementation in clinical practice has been
investigated in survey studies. In a study among European Thyroid Association (ETA) members
20% of 147 respondents considered that current evidence warrants selenium supplementation
in patients with AIT. If the patient was not receiving LT4, 65% of the respondents would
occasionally recommend selenium supplementation [19]. In a study among members of the
Danish Endocrine Society, 37.5% indicated that supplementation with selenium or iodine could
be used if requested by the patient [20].
Page 3 of 33 Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
The combination of gaps in current evidence regarding clinical efficacy and the apparent
everyday use and recommendation of selenium supplements underlines the need for wellpowered trials evaluating important clinical outcomes to help clinical decision-making. In “The
chronic autoimmune thyroiditis quality of life selenium trial” (CATALYST) [21], we aimed to
investigate the effect of selenium supplementation on disease-specific QoL in patients with
hypothyroidism due to AIT.

Methods
Study design and participants
CATALYST is a randomised placebo-controlled double-blinded multicentre trial, investigating
the effect of 12 months selenium supplementation on disease-specific QoL in patients with AIT.
The study protocol [21] was approved by the Regional Scientific Ethical Committees for
Southern Denmark (project ID: S-20130123) and registered at ClinicalTrials.gov (ID:
NCT02013479).
Participants were randomised 1:1 at the baseline visit to daily supplementation of either
200 μg selenium as selenium-enriched yeast or matching placebo tablets for 12 months. The
patients attended four planned visits (V1-4): At baseline, and after three, 12, and 18 months.
Anthropometric data, medical history, and current medication including LT4 dosage were
recorded, and project pre-specified blood and urine samples were obtained and stored at -80°C
in the research biobank at Odense University Hospital (OUH) until analysis after study
completion.
The trial intervention was given in addition to the current LT4 treatment. The trial was
conducted according to a pragmatic design, meaning that the participants were followed
routinely at their local outpatient clinic and according to local clinical guidelines. Routine
thyroid function tests were done on a regular basis to adjust the LT4 dose, aiming for a serum
TSH level within the reference range.
Inclusion criteria
Age ≥18 years; diagnosis of AIT with serum TPOAb ≥100 kIU/L measured within the last 12
months; LT4 treatment based on an initial serum TSH ≥4.0 mIU/L; and written informed
consent.
Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024 Page 4 of 33
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
Exclusion criteria
Previous diagnosis of toxic nodular goitre; Graves’ hyperthyroidism or orbitopathy; postpartum
thyroiditis; previous radioiodine therapy or thyroid surgery; antithyroid drug treatment;
comorbidity rendering the participant unable to process patient-reported outcomes or receive
intervention during the trial; immunomodulatory medication; other medication known to
affect thyroid function; pregnancy, breastfeeding, or planned pregnancy within the next 18
months; allergy towards any component in the selenium or the placebo tablets; intake of
selenium supplementation >55 μg/day; inability to read or understand Danish; or lack of
informed consent.
Allocation and randomisation
From March 2014, participants were recruited from six clinical sites in Denmark (Odense
University Hospital, Rigshospitalet, Bispebjerg Hospital, the Hospital of Southwest Jutland,
Herlev/Gentofte Hospital, and the Hospital of Southern Jutland) and randomised 1:1 to daily
supplementation with either 200 μg selenium as selenium-enriched yeast or matching placebo
tablets for 12 months.
The allocation sequences were computer-generated by the central pharmacy at Odense
University Hospital, with varying block size. Randomisation was stratified by clinical site and
duration of LT4 treatment (less or more than three months) (Figure 1). Blinding was maintained
for both investigators and participants throughout all aspects of the trial, until all analyses were
completed.
Primary and secondary outcomes
The primary outcome was health-related QoL after 12 months of intervention, as measured by
the Composite score from the Thyroid-related Patient-Reported-Outcome questionnaire
(ThyPRO-39) assessed at baseline, after six weeks, and after three, six, 12, and 18 months.
Secondary outcomes were the LT4 dosage, serum TPOAb, and serum free T3 index (FT3I)/free
T4 (FT4) ratio [22], assessed at each study visit (V1-4).
Thyroid status
Serum TSH (reference level: 0.3-4.0 mIU/L), FT4 (10.0-22.0 pmol/L), total T3 (1.3-2.2 nmol/L),
and T3 uptake test (0,75-1,25) were determined at Gentofte Hospital, Copenhagen, by two-site
Page 5 of 33 Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
chemiluminescent immunometric assay using Siemens Atellica®IM Analyser. Serum TPOAb
(value <60 kIU/L is considered negative) was determined at Herlev Hospital, Copenhagen, by
two-site chemiluminescent immunometric assay using Siemens Atellica®IM Analyser. The
coefficient of variation (CV) was 2.5% and 5% at TSH levels of 1.04 and 10.7 mIU/L, respectively;
CVs were 8.0% and 4.9% at total T3 levels of 1.4 and 4.4 nmol/L, respectively; CVs were 10% at
T3-uptake test levels of 0.68 and 1.05, respectively; CVs were 7.7% and 8% at FT4 levels of 12.3
and 28.9 pmol/L, respectively, and CVs were 7.0% and 4.8% at TPOAb levels of 120 and 520
kIU/L, respectively.
Selenium status
Serum selenium was assessed at Charité Universitätsmedizin, Berlin, Germany. Serum samples
were diluted 1:2 with a Gallium (Ga) containing buffer, which served as standard. Eight µL of
the diluted serum samples containing Ga were applied on quartz glass slides (Bruker Nano
GmbH) and left to dry overnight in a 37°C incubator. Subsequently, total selenium was
measured using a benchtop TXRF analyser (T-Star, Bruker Nano GmbH, Berlin, Germany) by
examining the emission spectra [23]. A standard serum was included in all sets of
measurements to serve as control. The inter-assay CV was 7.3% and intra-assay CV was below
7.8%.
Health-related quality of life
QoL was measured by the validated short-form of the disease-specific questionnaire ThyPRO39 consisting of 39 items [24-27]. Scales regarding hyperthyroidism (Hyperthyroid symptoms,
Eye symptoms, and Cosmetic complaints) were not included in this study. The items employ a
recall period of four weeks and are summarised in 10 multi-item scales as well as a single item
scale measuring the overall impact of thyroid disease on QoL. The items are scored from 0 to 4,
following a Likert scale (where “0” is equivalent to “nothing at all” and “4” to “very much”). The
average score of items in each scale is divided by four and multiplied by 100 to generate scores
from 0 to 100, with higher scores indicating poorer health status. A ThyPRO-39 Composite
score, based on 21 items from the Tiredness, Cognition, Anxiety, Depressivity, Emotional
susceptibility, Impaired social life, and Impaired daily life scales, plus the Overall QoL item, was
also computed (23).
Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024 Page 6 of 33
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
Sample size and power estimation
A sample size estimation was made based on the primary outcome (ThyPRO-39 Composite
score). With a correlation between observations on the same participant of 0.50, and a type I
error probability (two-sided α level) of 0.05 it was estimated that 236 participants in each
intervention group were required to identify a difference between the two intervention groups
of four points on the 0 to 100 ThyPRO-39 Composite scale with 80% power [21].
Based on the calculated sample size of 472 in total, a difference in serum TPOAb levels of
138 IU/ml between the two intervention groups could then be identified with 80% power.
Likewise, the probability of finding a true difference in LT4 dosage of 25 μg/day between the
two groups following intervention was calculated to be 92% [21].
Statistical analyses
Statistical analyses were performed with intact blinding, meaning that the experimental
intervention and control intervention were randomly coded as ‘A’ and ‘B’. The blinding was
broken after completion of all statistical analyses.
We assessed normality assumptions by quantile plots. Due to deviation from the normal
distribution TPOAb were log-transformed before calculation. Groups were compared using chisquared test or Student’s t-test for parametric data, and Mann-Whitney U test for
nonparametric data. Data are presented as n (%); means ± standard deviation (SD) or medians
and quartiles (25-75%).
We compared changes in health-related QoL (Composite score, Hypothyroid symptoms,
and Goitre symptoms) between the two intervention groups from baseline to six weeks, and
three, six, 12, and 18 months by linear mixed regression models for longitudinal data, a model
which efficiently deals with missing data. The same method was also used to compare changes
in serum selenium, serum TPOAb, serum FT3I/FT4 ratio, and LT4 dosage. For each of the
dependent variables, we included a participant-specific random intercept as well as a random
slope. Robust standard errors were used to compute confidence intervals (CI) and p-values.
Figures are showing plots with predicted mean values and 95% CI.
The analyses were performed using Stata software (V17.0; StataCorp, College Station, TX,
USA). Statistical significance was defined as p<0.05. All statistical tests were two-sided.
Page 7 of 33 Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
Results
Study participants
The trial was terminated before the intended number of participants was achieved, because
the expiry date of the trial medication was reached by January 2020. From March 2014 until
that point, 412 patients were included. The mean age at baseline was 49±14 years and women
constituted 85% of the participants. Three-hundred and thirty-two patients (81%) completed
12 months of intervention of whom 82% were women, and 80 patients (19%) dropped out
(Figure 1). The reasons for leaving the study before completion were pregnancy (n=6), change
from LT4 to liothyronine+LT4 combination treatment (n=14), great desire for selenium
supplementation (n=3), adverse effects attributed to the trial medication (n=13), or withdrawn
consent for unknown reason or absence from trial visits (n=44). Differences between patients
who completed the trial and those who dropped out are shown in Table 1. The non-completers
were younger than the completers (45±13 vs. 40±13 years; P=0.002). No other differences were
observed.
Patient characteristics at baseline and after 12 months of intervention are shown in Table
2. No significant differences between the selenium group and the placebo group were
observed. Further, no significant differences between the two groups were observed at three
and at 18 months of follow-up (data not shown).
Biochemical results
At baseline, no significant treatment group differences were found for serum levels of TSH, FT4,
FT3I, TPOAb, and selenium (Table 3). At 12 months, serum selenium was significantly higher in
the selenium group than in the placebo group (140.4±32.8 vs. 84.1±15.9 µg/L; P<0.001), while
the serum TPOAb level was lower (median and quartiles: 1862 [724-4365] vs. 2455 kIU/L [977-
6760]; P=0.032).
Similar results were found by linear mixed model regressions, shown in Figure 2 (and in
supplementary Table 1). Serum selenium was significantly higher in the selenium group
compared to the placebo group after 12 months of intervention (139.9 [95%CI: 133.2-146.4] vs.
84.3 µg/L [79.9-88.6]; P < 0.001; Figure 2a), and the TPOAb level was lower (1995 [95%CI: 1512-
2512] vs. 2344 kIU/L [1862-2951]; P=0.016; Figure 2b). No significant differences were found
between the two groups regarding changes in the FT3I/FT4 ratio or the LT4 dosage (Figure 2c
and 2d). At 18 months of follow-up (six months off intervention) the only difference between
Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024 Page 8 of 33
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
the two groups was that serum selenium was still significantly higher in patients initially
randomised to receive selenium (P<0.001; Figure 2a).
Health-related quality of life
No group differences in QoL measured by ThyPRO-39 were found at baseline or after 12 months
(Figure 3). Similarly, in linear mixed model regressions no significant difference between the
two groups was found at any time point in the ThyPRO-39 Composite score and in the Goitre
symptoms score (Figure 4a and 4b, supplementary Table 1). At 12 months, participants in the
placebo group had a lower score (better QoL) in the Hypothyroid symptoms scale compared to
the selenium group (P=0.016; Figure 4c, supplementary Table 1).
Subgroup analyses
Post-hoc subgroup analyses stratifying the patients according to duration of the disease (LT4
treatment longer or shorter than three months, Figure 5), median baseline QoL (ThyPRO-39
Composite score <30 or >30), median TPOAb level at baseline (<2950 or >2950 kIU/L), or median
selenium status at baseline (<83.3 or >83.3 µg/L) did not significantly change the results, nor
did we find any effect of selenium supplementation, according to these stratifications, on QoL,
LT4 dosage, or the FT3I/FT4 ratio, as compared to placebo supplementation (data not shown).
Differentiated effects were seen on the serum TPOAb level (Supplementary Table 2).
Stratifying the patients according to duration of the disease (as defined above), the effect of
selenium on TPOAb after 12 months was only significant in those receiving LT4 for more than
three months, and not among those with a shorter duration of the disease; P=0.006 and
P=0.599, respectively. By stratification according to baseline QoL (as defined above), the effect
of selenium on TPOAb after 12 months was only significant among those with baseline
Composite score >30 (i.e. lowest QoL) and not among those with baseline Composite score <30;
at 12 months: P=0.001 vs. P=0.795; and 18 months: P=0.017 vs. P=0.581 respectively. Similarly,
the effect of selenium on TPOAb after 12 months of intervention was only significant in the “low
level” serum selenium group (as defined above); P=0.048 vs. P=0.176, respectively. Finally, the
effect of selenium supplementation on the TPOAb level was only significant in the “high level”
TPOAb group (as defined above) and not in the “low level” TPOAb group; P=0.019 vs. P=0.439,
respectively.
Page 9 of 33 Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
Adverse events
No serious adverse events were observed during the trial. No significant differences were found
in adverse effects between the selenium group and the placebo group (P=0.261). Thirteen
patients (nine in the selenium group and four in the placebo group) dropped out during the
intervention period because of adverse effects, including skin rash and itching, stomach pain,
coughing, or general discomfort and tiredness. In the group of participants who completed the
study, 31 participants (14 in the placebo group and 17 in the selenium group, P=0.712) had
adverse effects like skin rash and itching, stomach pain, and tiredness.
Discussion
The present study is, to our knowledge, the largest randomised controlled trial of selenium
supplementation in patients with hypothyroidism due to AIT, including more than 400 patients,
and the first study using QoL as the primary outcome. The main finding of our study was that
selenium supplementation had no effect on QoL in patients with hypothyroidism due to AIT, as
compared to placebo supplementation. Both intervention groups showed a marked
improvement in QoL, compared to baseline, which emphasises the placebo effect, as
demonstrated in numerous trials in the past. This improvement persisted even after the trial
medication was discontinued. Approximately 20% of patients dropped out, of which only a
minority were due to adverse events, being almost equally distributed between the selenium
group and the placebo group. No serious adverse events were observed.
The majority of previous selenium supplementation trials in patients with AIT have
focused on circulating thyroid autoantibody levels, with only a few studies reporting the effect
on clinically relevant outcomes such as QoL [14, 28-32]. Three of these studies reported
improved well-being with selenium supplementation [14, 28, 29] while the remaining studies
showed no effects [30-32]. Different generic questionnaires for measuring QoL have been used,
hindering these data to be synthesised into a meta-analysis [33, 34]. The composite score of
ThyPRO-39 was chosen as the primary outcome in the present trial [21]. This questionnaire is
a well-validated tool specifically designed to explore and monitor QoL among patients with
benign thyroid diseases [24-27].
To optimize the concentration of selenoprotein P, considered one of the most valid
biomarkers of the selenium status, serum selenium should be in the range 120-130 μg/L. To
reach this level, the required selenium intake is at least 100 μg per day [2, 35]. Further, the
Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024 Page 10 of 33
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
impact of selenium deficiency may be modified by autoantibodies to selenoprotein P [36].
Marginal selenium deficiency in Europe as well as lower selenium status in patients with AIT
compared with healthy controls has been reported [12, 35, 37]. This is in line with the present
study where we found a baseline median selenium level of 83.3 μg/L. Patients randomised to
selenium supplementation showed an increase of approximately 70% in serum selenium (from
84 to 140 μg/L) during the intervention period of 12 months. Taking the relatively low standard
deviation (33 μg/L) of serum selenium into account this indicates a good adherence to the
intervention in the majority of our participants.
As the conversion of T4 to T3 in the organism is mediated by iodothyronine deiodinases
type 1 and 2, selenium supplementation might theoretically result in a higher conversion rate,
in case of suboptimal selenium levels. We found no effect of selenium supplementation on the
serum FT3I/FT4 ratio, or the LT4 dose needed to maintain serum TSH within the reference
range. These results are not surprising since the deiodinases are prioritized over other
selenoproteins in case of selenium deficiency. Thus, the thyroid hormone conversion was most
likely unaffected in our patients at baseline, even though the selenium intake may have been
in the lower range [2]. The only significant finding in our trial was that the level of TPOAb
decreased significantly more in patients randomised to selenium than in those receiving
placebo. Although we found no effects on QoL, LT4 dosage, or the FT3I/FT4 ratio, compared to
placebo, a significant decrease in the TPOAb level was indeed observed in the “low level”
selenium group. These results are consistent with previous studies which demonstrated that
LT4-treated patients with high TPOAb or low selenium levels are those showing the most
pronounced reduction of TPOAb level by selenium supplementation [2, 15, 17]. Nevertheless,
whether supplementing selenium deficiency can prevent the development of AIT remains to
be shown in clinical trials.
In addition to the importance for thyroid hormone conversion, the biological effects of
selenium are mediated by selenoproteins with antioxidant and immunoregulatory capacity.
However, a deeper understanding of how thyroid autoimmunity is reduced by selenium
supplementation is lacking. Likewise, the clinical implication of this effect is uncertain [17]. As
thyroid diseases seem to carry a risk of increased burden of oxidative stress [38-40],
selenoproteins with antioxidant properties, like glutathione peroxidase and thioredoxin
reductases, may be of particular importance. We have recently reported that treatment of
hyperthyroidism caused by Graves’ disease or toxic nodular goiter reduces systemic oxidative
Page 11 of 33 Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
stress load by 10-25% [41], and mostly among patients with the former condition. Oxidative
stress is reported to be increased also in patients with AIT [42, 43], suggesting that thyroid
autoimmunity per se may be the key driver. The effect of selenium may be mediated, at least
in part, through inflammatory pathways. Thus, selenium supplementation seems to upregulate
activated T cells [44] and to inhibit the release of several pro-inflammatory cytokines, including
interleukin 2, interferon gamma (IFNγ), and tumor necrosis factor [2, 45]. A recent study found
that 100 μg daily selenium supplementation in 29 women with AIT reduced IFNγ and increased
interleukin 1β concentrations [46]. Although the present trial found no positive effects on
clinical outcomes, such as QoL and the LT4 dosage, further research is needed to explore
whether the potential effects of selenium on the immune system, the oxidative stress burden,
and low-grade inflammation are of clinical significance in patients with thyroid diseases. In
addition, future studies may reveal if specific gene variants predict increased responsiveness
to selenium supplementation [47-49].
Besides being a risk factor for developing AIT [2, 50] selenium deficiency as well as low
dietary selenium intake [51] have been linked to increased risk of prostate cancer, mortality,
decreased immune function, and impaired fertility [2, 50]. Importantly, overexposure to
selenium can have unintended health effects in humans. A randomised controlled trial found
that 300 μg/day selenium supplementation significantly increased all-cause mortality by 11.3%
after 10 years [52]. Further, a US study linked high selenium levels to an increased mortality
rate [53]. This suggests a high intake of selenium supplements should be avoided unless a solid
clinical indication exits, and the treatment is based on evidence from clinical trials.
The major strength of this study is the randomised, controlled, and double-blinded design,
and use of the extensively validated ThyPRO-39 for measuring patient-reported outcomes [2,
24]. Both men and women across a wide age range were included, and only those with AIT and
positive TPOAb above 100 kIU/L were eligible. The trial was conducted according to a pragmatic
approach to mimic normal clinical practice, meaning that the patients were followed by their
usual clinician in parallel with their study visits to the investigators. The study was well-powered
with more than 400 patients, which makes it the largest trial of selenium supplementation in
this patient population. The trial was terminated before the intended number of 472 included
patients was reached. However, we believe that the risk of a type II error is very low, as no
trends of differences between the two intervention groups were observed in any of the major
outcomes. Moreover, the ThyPRO instrument has been further validated since the trial was
Accepted Manuscript published as ETJ-23-0175.R1. Accepted for publication: 12-Jan-2024 Page 12 of 33
Copyright © 2023 the authors
Downloaded from Bioscientifica.com at 01/15/2024 09:29:54AM
via Open Access. This work is licensed under a Creative Commons
Attribution-NonCommercial-NoDerivatives 4.0 International License
http://creativecommons.org/licenses/by-nc-nd/4.0/deed.en_GB
launched in 2014, including studies to determine minimal important change (MIC) values [54].
MIC is defined as the smallest change in score that patients perceive as important. For the
ThyPRO composite score, MIC between groups is 9.1 [54]. Using this figure in the sample size
calculation for the present trial, only 75 patients in each group would be needed.
A few limitations to this study exist. We included patients from six different sites over a
period of seven years. The total number of patients with hypothyroidism followed at the
hospital units during this period was clearly higher than the number of patients enrolled in the
trial. For practical reasons, we only obtained complete clinical data on patients assessed for
eligibility at Odense University Hospital. We consider this subgroup as representative of the
entire study population, as Odense University Hospital was the main site, including more than
half of the participants. Restricting our analyses to patients recruited only at this site did not
significantly change the results (data not shown).
Although QoL in our patients clearly was lower than in the background population it may
be a point of criticism that QoL at baseline was not sufficiently impaired, thereby attenuating
any positive effect of the intervention, if indeed present. We do not find this to be of any major
concern as no differences between the intervention groups could be found even if we restricted
the analyses to patients with the lowest QoL at baseline.
In conclusion, selenium supplementation for 12 months, compared to placebo, did not
improve QoL, LT4 dosage, or FT3I/FT4 ratio in patients with hypothyroidism due to AIT. The
results from this large randomised controlled trial do not justify the routine use of selenium
supplementation in patients with AIT.
OriginalsprogEngelsk
Artikelnummere230175
TidsskriftEuropean Thyroid Journal
Vol/bind13
Udgave nummer1
Antal sider13
ISSN2235-0640
DOI
StatusUdgivet - feb. 2024

Emneord

  • Hypothyroidism
  • Autoimmune thyroiditis
  • Health-related quality of life; Selenium supplementation

Fingeraftryk

Dyk ned i forskningsemnerne om 'Selenium supplementation and placebo are equally effective in improving quality of life in patients with hypothyroidism'. Sammen danner de et unikt fingeraftryk.

Citationsformater