Prof. Marek Ruchała, M.D., Ph.D., Department of Endocrinology, Metabolism, and Internal Medicine, Poznan University of Medical Sciences, Przybyszewskiego St, 49, 60-355 Poznań, Poland, e-mail: mruchala@ump.edu.pl, phone: + 48 601 748 905; fax: + 48 61 869 16 82

Introduction

Irisin, a cleaved and secreted part of the transmembrane protein FNDC5 (fibronectin type III domain containing 5), is a recently discovered adipo-myokine that is said to have a significant influence on body metabolism [1-3]. This cytokine, first reported by Boström et al. in 2012, is widely discussed in tire literature. Secreted by muscles during physical exercise, it increases thernrogenesis by promoting the process of white adipose tissue “browning” [4–8]. Further studies revealed that irisin in not only a myokine, but also a potential adipokine, since FNDC5 gene expression was also reported in adipose tissue [9-11]. To date, numerous studies have been published evaluating the role of irisin in pathological entities [12, 13] and thyroid dysfunction in metabolic disorders [14-16]. To the best of the authors’ knowledge, the first clinical presentation of the thyroid function impairment influence on serum irisin level was published in 2014. The preliminary study by Ruchala et al. [17] states that changes in thyrometabolic state may alter serum irisin level. However, the results obtained by Samy et al. [18], who described serum irisin concentration in rats with short-term drug-induced hyperand hypothyroidism, were not fully consistent with the previous clinical study. Therefore, the aim of the present research is to evaluate the changes in serum irisin level in long-lasting autoimmune hypothyroidism, short-term hypothyroidism in athyreotic patients after L-thyroxine treatment withdrawal before radioiodine therapy, and a control group with no impairment of thyroid function.

Material and methods

Studied group

The studied group comprised 24 patients (22 women and 2 men) affected by severe overt hypothyroidism, meeting the inclusion criteria, and referred to the Department of Endocrinology, Metabolism, and Internal Medicine. Half of the group constituted newly diagnosed 12 patients with autoimmune thyroid disease, with a presumed long duration of the disorder (AITD). The other 12 subjects were recruited from patients after total thyroidectomy because of differentiated thyroid cancer, who withdrew L-thyroxine four weeks before radioiodine treatment (TC). Inclusion criteria were as follows: age over 18 years, overt hypothyroid state due to autoimmune thyroid disease, or a history of thyroidectomy with no adequate oral L-thyroxine treatment. Exclusion criteria included: any other chronic diseases, particularly metabolic diseases, kidney, heart, muscle, and neurological diseases, as well as those intensively involved in a sport or any strenuous physical activity [19-22]. All patients taking any drugs daily were also excluded from the project. Both groups had similar age and gender distribution.

Control group

A control group (CG) of 12 subjects (11 women and 1 man) with normal thyroid function and no other chronic diseases was engaged, with similar age and gender distribution.

The Bioethical Committee

The Bioethical Committee of Poznan University of Medical Sciences approved the study (Resolution no. 923/13). All participants provided informed written consent to take part in the study.

The study protocol

The study protocol was almost the same as in the research by Ruchala et al. 17 Similarly, all subjects underwent full clinical examination and thyroid ultrasonography. The measured laboratory parameters included irisin level, thyroid and thyroid-related hormone concentrations (thyroid stimulating hormone – TSH, free triiodothyronine – FT3, free thyroxine – FT4), anti-thyroid autoantibody titre assessments (serum anti-thyroid peroxidase antibodies – TPOAb, anti-thyroglobulin antibodies – TgAb), and creatine kinase (CK) level as a biochemical marker of muscle destruction. Additionally, serum fasting glucose and insulin levels were measured. All measurements were performed in venous blood samples collected in a consistent manner after eight hours of fasting. Patients additionally underwent bioelectrical impedance analysis to evaluate their body mass, body mass index (BMI), muscle mass (MM), and fat mass (FM).

The described measurements were performed with the use of the same methods as previously [17]. TPOAb and TgAb levels were assessed by radioimmunological method with commercially available BRAMHS anti-TPO and anti-Tg RIA kits using a scintillation gamma counter (LKB Wallac CliniGamma 1272). Additionally, the glucose level was analysed with the use of a Hitachi Cobas e601 chemiluminescent analyser (Roche Diagnostics), whereas insulin concentration was assessed using an ELISA kit from Phoenix Pharmaceuticals.

Reference ranges

The used reference ranges were as follows: TSH 0.27-4.2 μIU/mL, FT3 3.90-6.70 pmol/L, FT4 11.5-21.0 pmol/L, CK 26.0-140.0 U/L, glucose 70.0-99.0 mg/dl, insulin 2.0-25.0 μIU/mL.

The diagnosis

The diagnosis of overt hypothyroidism was established with TSH level over the normal range and free thyroid hormones concentrations below the normal range. Autoimmune thyroid disease was diagnosed with at least one anti-thyroid autoantibody titre elevated (TPOAb and/or TgAb) and the presence of sonographic signs of chronic inflammation in thyroid ultrasound (decreased, heterogeneous echogenicity, poorly defined hypoechoic areas permeated by fibrous echogenic layers) [23].

Statistical analysis

The gathered data were analysed statistically with the use of STATISTICA software by StatSoft. Normality was analysed by Shapiro-Wilk test, and the equality of variances was assessed with the use of Levene’s test. Due to the small number of analysed cases, the difference in normality and equality of variances between groups, nonparametric tests like Spearman’s RankOrder Correlation and Kruskal-Wallis one-way analysis of variance were applied to evaluate the statistical difference between the groups. In cases of statistically significant differences in Kruskal-Wallis test multiple pairwise comparisons with a Bonferroni correction was be performed. A P-value less than 0.5 indicated statistical significance.

Results

The study included three groups of 12 subjects, that did not differ statistically in gender distribution (11 women and 1 man), body mass, age, BMI, FM, and MM. There was also no statistical difference in serum fasting glucose and insulin level. The statistical difference between the groups was observed in irisin, TSH, fT3, fT4, and CK concentrations. Subsequent post-hoc analysis revealed statistically significant differences in irisin level between AITD and TC, and AITD and CG, and no difference between TC and CG (Fig. 1). TSH level was significantly higher in AITD and TC than in CG, with no difference between AITD and TC. Also, fT3 and fT4 levels were similar in AITD and TC and significantly higher in CG. CK concentration was significantly higher in AITD and CG with no difference between AITD and TC, as well as TC and CG (Table I).

Table I. Median values of performed measurements in patients with long-lasting (AITD) and short-term (TC) overt hypothyroidism and in the control group (CG). The statistical difference between groups is illustrated by the p value. Statistically significant results in Kruskal-Wallis one-way analysis of variance and post-hoc multiple pairwise comparisons with a Bonferroni correction are marked (*). Values of the 25th and 75th percentile are presented in brackets
Tabela I. Mediana wartości wykonywanych pomiarów u pacjentów z długotrwałą (AITD) i krótkotrwałą (TC) jawną niedoczynnością tarczycy oraz w grupie kontrolnej (CG). Różnicę statystyczną między grupami zilustrowano wartością p. Statystycznie istotne wyniki w Teście Kruskala-Wallisa i w teście wielokrotnych porównań post-hoc z korektą Bonferroniego są oznaczone symbolem (*). Wartości 25 i 75 percentyla prezentowane są w nawiasach

Additionally, in the whole group of involved subjects CK concentration was negatively correlated with irisin level (r = -0.58; p < 0.01), as well as TSH level (r = -0.39; p = 0.016). FT3 and fT4 levels were not correlated with irisin concentration.

Discussion

The available data concerning the relationship between serum irisin concentrations and thyroid function is limited. In 2014 our research team published a preliminary study evaluating the irisin level in hypothyroid and hyperthyroid patients [17]. It indicated that the irisin concentrations may vary according to the thyrometabolic state, and it may be related to the degree of muscle damage, presented by elevated CK level. Recently, Samy et al. [18] reported the up-regulation of serum irisin in male rats with hyper- and hypothyroidism, which may be associated with oxidative damage and/or myopathy observed in both conditions. In the described study, the drug-induced thyroid impairment lasted for only three weeks. The irisin and CK levels were elevated both in hyper- and hypothyroid rats, with no statistical difference between these groups. In the study by Ruchala et al. [17], most hypothyroid patients were diagnosed with an autoimmune thyroid disease that usually lasts for a couple of months [24]. Due to partial inconsistencies between both studies, the authors of the present study wanted to evaluate whether the described differences in irisin concentrations in hypothyroidism are associated with the duration of the thyroid dysfunction. In the present study, the studied groups with hypothyroidism represent the clinical model of long- and short-term thyroid dysfunction, which was compared with the control group. The obtained data suggests that only long-lasting hypothyroidism significantly decreases the irisin level, while short-term dysfunction does not change irisin concentrations. The wide resemblance between irisin and thyroid hormone functions, with their profound impact on metabolism and energy production, underline the possible influence of thyroid hormones and TSH on irisin levels. Although much lower than in muscles or adipose tissue, some expression of FNDC5 gene was reported in thyroid [5]. However, the difference in irisin levels in the hypothyroid groups is statistically significant despite no statistical difference in their thyrometabolic state (fT3, fT4, TSH). This lack of statistical difference in their thyrometabolic state can be reflected by no correlation between fT3, fT4, and irisin level. Additionally, irisin probably has no in vitro effect on cell proliferation and malignant potential of thyroid cancer cell lines [25]. According to Samy et al. [18], the irisin level may be even transiently elevated after the induction of hypothyroidism due to reported muscle damage and leaks from damaged muscle cells. Disturbance or disintegration of striated muscle tissue with concomitant leakage of intracellular muscle constituents into the circulation causes a rise in CK level. The mechanism(s) by which CK is metabolised has not been fully elucidated, and it is likely that the observed serum CK levels reflect complex interactions associated with the energy status and the scale of muscle disturbance [26]. The negative correlation between CK and irisin levels additionally underline a possible relation between irisin level and muscle-damage in patients with overt hypothyroidism. Moreover, in reported muscle biopsies from patients with severe prolonged disorder and significantly elevated CK activity we may observe myopathic features. In patients with short duration of metabolic disturbance and slightly elevated CK concentrations, histological changes were absent in muscle biopsy [27]. The authors hypothesise that myopathy presented by patients with severe, long-term hypothyroidism might impair the function of muscle tissue as a hormone-secreting organ, resulting in a gradual decrease of irisin level. However, these assumptions are a matter of debate and may be treated only as an introduction to further studies and need confirmation. In the newest study by Ateş et al. [28] the authors analysed the irisin level changes in patients with Hashimoto’s thyroiditis. They report that hypothyroid patients had higher irisin levels than the control subjects. However, the described results cannot be compared with our current findings because the average TSH level was much lower in the manuscript by Ateş et al. (13.1 μIU/mL). Moreover, the obesity was not concerned as an exclusion criterion [28]. Since the hypothyroidism group had higher median levels of irisin in obese patients than those in the control group, the results are not reliable.

Nevertheless, the present research has some limitations. The studied groups were small and the measurement has not yet been performed after the restoration of euthyroidism. We were also unable to construct a clinical model of long- and short-term hyperthyroidism. It is worth mentioning a recent discussion about the measurement accuracy of available ELISA kits [29]. However, the studied groups were strictly homogenic in various crucial parameters, and thus constitute a good clinical model of the problem described.

Conclusions

To sum up, irisin levels seem to change with the timespan of hypothyroidism. It may result from initial muscle damage and leaks from damaged muscle cells and following a decrease in irisin production due to prolonged myopathy. Hence, patients with long-lasting hypothyroidism have lower irisin levels than those with short-term disorder. Nevertheless, the presented data are only an introduction to a further exploration of this issue.

Acknowledgements

This research was funded from the Statutory Funds of Department of Endocrinology.