Introduction: Autoimmune thyroiditis seems to be associated with increased cardiometabolic risk. Statins, the mainstay of cardiovascular risk reduction and prevention, were found to reduce thyroid antibody titers. The aim of this study was to investigate plasma markers of cardiometabolic risk in statin-treated women with thyroid autoimmunity. Methods: We compared two matched groups of euthyroid women with hypercholesterolemia receiving atorvastatin treatment: subjects with autoimmune (Hashimoto’s) thyroiditis (group A, n = 29) and subjects without thyroid pathology (group B, n = 29). Plasma lipids, glucose homeostasis markers, as well as circulating levels of uric acid, high-sensitivity C-reactive protein (hsCRP), fibrinogen, homocysteine, and 25-hydroxyvitamin D were measured before atorvastatin treatment and 6 months later. Results: At entry, both groups differed in antibody titers, insulin sensitivity, and plasma levels of uric acid, hsCRP, fibrinogen, homocysteine, and 25-hydroxyvitamin D. Atorvastatin-induced reduction in hsCRP and homocysteine, but not in total cholesterol and LDL-cholesterol, was more pronounced in group B than in group A. Only in group B, the drug decreased uric acid and fibrinogen and increased 25-hydroxyvitamin D. In group A, atorvastatin reduced insulin responsiveness. Conclusion: The obtained results indicate that euthyroid women with Hashimoto’s thyroiditis may benefit to a lesser degree from atorvastatin treatment than other populations of women with hypercholesterolemia.

© 2023 The Author(s). Published by S. Karger AG, Basel

Introduction

Autoimmune (Hashimoto’s thyroiditis) is one of the most frequent human disorders, the most common cause of thyroid hypofunction in developed countries, as well as the most prevalent organ-specific autoimmune disease [1, 2]. There is increasing evidence suggesting that autoimmune thyroiditis may predispose to the development of atherosclerosis and that this susceptibility is observed even if thyrotropin and thyroid hormones are within the reference range [3-7]. Carotid-femoral pulse wave velocity, a marker of arterial stiffness, was significantly higher in women with Hashimoto’s thyroiditis and normal thyroid function than in matched healthy controls, and the between-group difference was greatest in premenopausal women [3]. Flow-mediated arterial dilation was significantly lower in euthyroid women with autoimmune thyroiditis than in controls [4]. In comparison to healthy subjects, euthyroid patients with Hashimoto’s thyroiditis had a lower value of the reactive hyperemia index, an operator-independent marker of endothelial function, and the severity of endothelial dysfunction was independently negatively associated with titers of thyroid peroxidase antibodies (TPOAb) [5]. The intima-media thickness of the common carotid artery was greater in obese and overweight women with thyrotropin and free thyroid hormones within the reference range if they were affected by Hashimoto’s thyroiditis, and this association was independent of age, the body mass index, and hypertension [6]. Euthyroid women with autoimmune thyroiditis were characterized by low-grade systemic inflammation and increased production of pro-inflammatory cytokines by activated human monocytes and lymphocytes [7]. Lastly, despite restoration of euthyroidism, patients with Hashimoto’s thyroiditis aged over 50 years had a threefold increase in cardiovascular admissions compared to controls [8].

Patients with autoimmune thyroiditis seem to benefit from treatment with 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins) [9-11]. High-intensity rosuvastatin therapy reduced elevated TPOAb titers in women with Hashimoto’s thyroiditis [9]. Moderate-dose simvastatin therapy potentiated the impact of vitamin D on TPOAb titers in levothyroxine-treated women with Hashimoto’s thyroiditis and vitamin D insufficiency [10]. High-intensity atorvastatin treatment enhanced the impact of selenomethionine on titers of TPOAb and thyroglobulin antibodies (TgAb) [11]. Benefits resulting from statin therapy may be attributed to anti-inflammatory and immune-suppressive properties of HMG-CoA reductase inhibitors. Statins were found to stimulate apoptosis of peripheral blood T lymphocytes and to change the distribution of the lymphocyte subpopulations in patients with Hashimoto’s thyroiditis, which were accompanied by a reduction in circulating levels of thyrotropin and an increase in free thyroid hormone concentrations [12]. Moreover, statins were reported to repress human leukocyte antigen D-related expression on thyrocytes obtained from patients with autoimmune thyroiditis [13].

Despite a reduction in thyroid autoimmunity, no previous study investigated the impact of statin therapy on plasma lipids, glucose homeostasis markers, and cardiovascular risk factors in individuals with thyroid autoimmunity. Although Krysiak et al. [14] observed that lipid-lowering and extra-lipid effects of atorvastatin were impaired in individuals with untreated subclinical hypothyroidism, their study population was unselected, including patients with thyroid hypofunction of both autoimmune and non-autoimmune origin. To better understand the effectiveness of HMG-CoA reductase inhibitors in subjects with Hashimoto’s thyroiditis, the present study was aimed at investigating whether thyroid autoimmunity by itself modifies the impact of statin therapy on cardiometabolic risk factors in women with hypercholesterolemia.

Methods

The study was approved by the Institutional Review Board prior to initiation, as well as was conducted in accordance with the 1964 Declaration of Helsinki and its later amendments. All patients provided written informed consent before participating in the research after the nature of the study had been explained and each participant had the opportunity to ask questions. The paper was prepared in accordance with the Enhancing the Quality and Transparency of Health Research (EQUATOR) Network guidelines for observational studies (STROBE).

Patients

The participants were selected among women, aged between 18 and 55 years, with hypercholesterolemia (total cholesterol levels above 200 mg/dL and low-density lipoprotein (LDL)-cholesterol levels above 115 mg/dL), qualified for statin therapy. They were required to have a 5% or greater risk of developing coronary death, nonfatal myocardial infarction, fatal stroke, or nonfatal stroke within the next 10 years and the presence of risk-enhancing factors (family history of premature atherosclerotic cardiovascular disease, premature menopause, history of hypertensive disorders of pregnancy, persistently elevated LDL-cholesterol ≥160 mg/dL; metabolic syndrome; ethnicity factors; persistently elevated triglycerides ≥175 mg/mL; and hsCRP >2.0 mg/L). The 10-year risk for atherosclerotic cardiovascular disease was calculated using the freely available ASCVD Risk Estimator Plus (http://tools.acc.org/ASCVD-Risk-Estimator-Plus/#!/calculate/estimate/). If participants were younger than 40 years, an age of 40 was assigned, as the pooled cohort equations for estimation of the 10-year cardiovascular risk are only applicable to individuals aged between 40 and 79 years. The study included only patients with plasma levels of thyrotropin and free thyroid hormones within the reference range (thyrotropin between 0.4 and 4.5 mU/L, free thyroxine between 10.0 and 21.2 pmol/L and free triiodothyronine between 2.1 and 6.5 pmol/L).

The study population consisted of two groups of patients. Group A included 29 women with autoimmune thyroiditis while group B included 29 matched women without thyroid pathology. Autoimmune thyroiditis was diagnosed in patients with TPOAb titers above 100 U/mL and the diffusely enlarged thyroid gland with hypoechoic thyroid parenchyma on ultrasonography. The sample size calculation showed that a sample of 26 subjects per study group was required to detect a 20% difference between baseline and follow-up values of the measured cardiometabolic risk factors with 80% power and a significance level of 5%. This means that the study population exceeded the necessary number of patients. Because both groups were required to be matched for age, the body mass index, blood pressure, plasma lipids and plasma levels of thyrotropin and free thyroid hormones, subjects without thyroid pathology were chosen among 82 women with hypercholesterolemia requiring statin therapy on the basis of a computer algorithm. To minimize the impact of seasonal fluctuations in the measured variables, similar numbers of patients were recruited between April and May (28) and between November and October (30).

The exclusion criteria were as follows: cardiovascular disease, other endocrine disorders, diabetes, other autoimmune or inflammatory diseases, malabsorption syndromes, impaired renal or hepatic function, other serious disorders, levothyroxine or liothyronine treatment, pregnancy or lactation, and poor patient compliance. The subjects were also excluded if they were treated with other hypolipidemic drugs, angiotensin-converting enzyme inhibitors, sartans, glucocorticoids, nonsteroidal anti-inflammatory drugs, drugs reducing thyroid antibody titers (vitamin D, selenium, or myoinositol), or with drugs known to interact with statins.

Study Design

Over the entire study period (6 months), all patients were treated with atorvastatin (20 mg), which was taken once daily at bedtime. In addition to medical treatment, the participants were requested to comply with the therapeutic lifestyle changes diet (total fat intake <30% of total energy intake, saturated fat intake <7% of energy consumed, cholesterol intake <200 mg per day, an increase in fiber intake to 15 g per 1,000 kcal), as well as to do at least 150 min of moderate-intensity aerobic physical activity per week. Short-term use of new prescription or over-the-counter drugs (not longer than 7 days) was allowed only if such treatment was stopped at least 6 weeks before the end of the study. Participants were seen every 6 weeks to ensure adherence to rosuvastatin treatment and to boost compliance with the study protocol. Medication adherence was measured by counting the number of residual tablets. Compliance with dietary recommendations was assessed by analysis of individual dietary questionnaires and of diaries in which the patients recorded all their activities.

Laboratory Assays

Venous blood samples were drawn in the fasting state between 8 and 9:00 a.m. on the first and last study day, and all measurements were performed in duplicate. Plasma levels of total cholesterol, LDL-cholesterol, high-density lipoprotein (HDL)-cholesterol, triglycerides, uric acid, glucose, and creatinine were assessed by an automated system (COBAS Integra 400 Plus, Roche Diagnostics, Basel, Switzerland). Direct chemiluminescence using acridinium ester technology was used to measure plasma levels of insulin, thyrotropin, free thyroxine, triiodothyronine, homocysteine, and 25-hydroxyvitamin D, as well as titers of TPOAb and TgAb (ADVIA Centaur XP Immunoassay System, Siemens Healthcare Diagnostics, Munich, Germany). Concentrations of high-sensitivity C-reactive protein (hsCRP) were measured using immunoassay with chemiluminescent detection (Immulite 2000XPi, Siemens Healthcare, Warsaw, Poland), while fibrinogen levels were measured by the Clauss technique using an automated BCS XP analyzer (Siemens Healthcare, Warsaw, Poland). The homeostasis model assessment 1 of insulin resistance index (HOMA1-IR) was calculated as fasting glucose (mg/dL) × fasting insulin (mU/L)/405, while the glomerular filtration rate was estimated by the Cockcroft-Gault equation.

Statistical Analysis

Because the raw data were skewed, all variables were log-transformed to stabilize the variance and attain normality. Baseline values, follow-up values, and percent changes from baseline after adjustment for baseline values in both study groups were compared using Student’s t test for independent samples. Baseline and follow-up values within the same group were compared using Student’s paired t tests. The χ2 test with Yates’ correction was employed to compare nominal data. Correlations were assessed using Pearson’s r tests for two continuous variables; Phi coefficient for one continuous and one categorical variable; and point-biserial for two categorical variables. p values corrected for multiple testing below 0.05 were considered statistically significant.

Results

At study entry, both groups were comparable with respect to age, smoking, the body mass index, blood pressure, circulating levels of thyrotropin, free thyroid hormones, lipids and glucose, and the glomerular filtration rate. Expectedly, TPOAb and TgAb titers were higher in group A than group B. Both groups also differed in plasma levels of uric acid, hsCRP, fibrinogen, homocysteine, 25-hydroxyvitamin D, and HOMA1-IR (Tables 1, 2).

Table 1.

Baseline characteristics of patients

Table 2.

The effect of atorvastatin on thyroid antibodies, hormones, lipids, glucose homeostasis markers, and the remaining risk factors in the investigated population

During the follow-up period, no patient developed serious or unexpected adverse events. One patient (from group A) withdrew consent because of personal reasons, while another patient (from group B) prematurely terminated the study because of noncompliance with the study protocol. Fifty-six patients (28 in each group) completed the study. All these patients complied with treatment and dietary recommendations and were included in the final analysis. A post hoc power calculation based on the primary outcome data and the given sample size showed that the study had sufficient statistical power (0.85). Atorvastatin did not affect the body mass index and blood pressure (data not shown). There were no differences between both groups in physical activity.

In both study groups, atorvastatin decreased total cholesterol, LDL-cholesterol, hsCRP, and homocysteine. The percentage changes in hsCRP and homocysteine were more pronounced in group B than group A. Only in group A, atorvastatin increased HOMA1-IR and tended to reduce TPOAb titers. Only in group B, the drug reduced uric acid and fibrinogen levels and increased 25-hydroxyvitamin D levels. Atorvastatin treatment did not affect TgAb, HDL-cholesterol, triglycerides, glucose, thyrotropin, free thyroid hormones and the estimated glomerular filtration rate. There were differences between both groups in follow-up titers of thyroid antibodies, follow-up values of HOMA1-IR and follow-up concentrations of uric acid, hsCRP, fibrinogen, homocysteine and 25-hydroxyvitamin D (Tables 2, 3).

Table 3.

Percentage changes in biochemical variables from baseline in the investigated population

In group A, there were positive correlations between antibody titers and hsCRP levels (TPOAb: r = 0.46, p = 0.0002; TgAb: r = 0.42, p = 0.0007), positive correlations between antibody titers and HOMA1-IR (TPOAb: r = 0.35, p = 0.0120; TgAb: r = 0.31, p = 0.0388), as well as inverse correlations between antibody titers and 25-hydroxyvitamin D levels (TPOAb: r = −0.40 [p = 0.0014]; TgAb: r = −0.38 [p = 0.0023]). In both groups, (a) baseline HOMA1-IR, uric acid, hsCRP, fibrinogen and homocysteine positively correlated with baseline hsCRP levels (group A: r values between 0.34 [p = 0.0147] for HOMA1-IR and 0.41 [p = 0.0007] for fibrinogen; group B: r values between 0.35 [p = 0.0120] for HOMA1-IR and 0.42 [p = 0.0006] for homocysteine); (b) baseline HOMA1-IR, uric acid, hsCRP, fibrinogen and homocysteine inversely correlated with baseline 25-hydroxyvitamin D (group A: r = −0.29 [p = 0.0401] for HOMA1-IR and r = −0.38 [p = 0.0023] for uric acid; group B: r = −0.31 [p = 0.0388] for HOMA1-IR and r = −0.40 [p = 0.0012] for fibrinogen); and (c) baseline hsCRP levels inversely correlated with baseline 25-hydroxyvitamin D (group A: r = −0.37 [p = 0.0020]; group B: r = −0.38 [p = 0.0017]). There were correlations between the effect of treatment on uric acid, fibrinogen and homocysteine and: (a) baseline hsCRP levels (group A: r values between −0.25 [p = 0.0498] and −0.40 [p = 0.0011]; group B: r values between −0.29 [p = 0.0392] and −0.47 [p = 0.0002]); (b) baseline 25-hydroxyvitamin D (group A: r values between 0.32 [p = 0.0206] and 0.41 [p = 0.0007]; group B: r values between 0.34 [p = 0.0194] and 0.46 [p = 0.0002]); (c) treatment-induced changes in hsCRP (group A: r values between 0.26 [p = 0.0498] and 0.40 [p = 0.0015]; group B: r values between 0.30 [p = 0.0402] and 0.44 [p = 0.0004]); and (d) treatment-induced changes in 25-hydroxyvitamin D (group A: r values between 0.30 [p = 0.0355] and 0.39 [p = 0.0014]; group B: r values between 0.35 [p = 0.0154] and 0.47 [p = 0.0002]). There were also positive correlations between treatment-induced changes in hsCRP and in 25-hydroxyvitamin D (group A: r = 0.38 [p = 0.0017]; group B: r = 0.40 [p = 0.0008]). In group A, the impact on hsCRP and 25-hydroxyvitamin D inversely correlated with baseline titers of TPOAb (hsCRP: r = −0.41 [p = 0.0006]; 25-hydroxyvitamin D: r = −0.37 [p = 0.0028]) and TgAb (hsCRP: r = −0.39 [p = 0.0017]; 25-hydroxyvitamin D: r = −0.32 [p = 0.0295]). The remaining correlations were insignificant.

Conclusion

In the current study, the effect of atorvastatin on thyroid antibody titers was limited to an insignificant reduction in TPOAb titers, observed in women with autoimmune thyroiditis. This finding is in line with previous observations indicating that the impact of HMG-CoA reductase inhibitors on thyroid autoimmunity depended on baseline antibody titers, that only intensive but not moderate-dose statin monotherapy decreased thyroid antibody titers [9, 10], as well as that TPOAb titers are a more sensitive and specific marker of autoimmune thyroid disease than TgAb titers [15]. It may be assumed that the effect on thyroid autoimmunity could have been greater if a higher dose of atorvastatin was used. However, the current study included only young and middle-aged women without cardiovascular disease or diabetes, which justified using in this population atorvastatin at the daily dose of 20 mg.

Another important finding was between-group differences in levels of all cardiometabolic risk factors assessed in the study. They cannot be explained by the impact of confounding factors, including hypercholesterolemia itself, because the study groups were well matched. Moreover, strict inclusion and exclusion criteria limited the impact of comorbidities and other drugs. Because elevated levels of uric acid, hsCRP, fibrinogen, and homocysteine, decreased levels of 25-hydroxyvitamin D and high values of HOMA1-IR are associated with an increased risk of cardiovascular disease, metabolic syndrome, and type 2 diabetes [16-21], women with Hashimoto’s thyroiditis seem to be susceptible to the development of atherosclerosis and metabolic disorders. This observation is in line with previous findings indicating that subjects with autoimmune thyroiditis more frequently develop atherosclerosis and its complications [3-8]. Although it is generally accepted that hypothyroidism is a well-known cardiometabolic risk factor [22], thyroid hypofunction cannot explain our findings because the participants were recruited among subjects with thyrotropin and free thyroid hormones within the reference range. Moreover, the selection procedure eliminated the impact of discrete abnormalities in hypothalamic-pituitary-thyroid axis activity resulting from thyroid autoimmunity. Consequently, contrary to other studies including euthyroid women with Hashimoto’s thyroiditis [3, 6], no between-group differences in thyrotropin and free thyroid levels have been found.

However, the most important finding of the present study was that women with autoimmune thyroiditis were partially resistant to cardiometabolic effects of atorvastatin. In this group of patients, the impact on hsCRP and homocysteine was less pronounced than in women without thyroid pathology, and the treatment did not affect circulating levels of uric acid, hsCRP, and 25-hydroxyvitamin D. Based on these findings, some practical conclusions can be drawn. First, because clinical benefits result from both hypolipidemic and extra-lipid properties of statins [23], untreated women with Hashimoto’s thyroiditis may be worse candidates for treatment with atorvastatin and possibly also with the remaining HMG-CoA reductase inhibitors than other women. Second, because lipid-lowering properties of atorvastatin did not differ between both groups, it seems that measurement of only plasma lipids has a limited value as a prognostic marker of statin effectiveness in subjects with autoimmune thyroid disease. Third, the unfavorable impact of thyroid autoimmunity was observed in patients without thyroid hypofunction. This means that even euthyroid Hashimoto’s thyroiditis, particularly coexisting with other disorders predisposing to cardiovascular and metabolic disorders, may require specific treatment with levothyroxine or with other drugs reducing thyroid antibody titers (vitamin D, selenium, or myo-inositol) [24, 25]. Moreover, the presence of inverse correlations between changes in hsCRP and 25-hydroxyvitamin D levels and baseline antibody titers suggests that pleiotropic effects of atorvastatin are determined by the severity of autoimmune thyroid disease. Lastly, because, similarly to thyroid autoimmunity, even mild hypothyroidism impairs cardiometabolic effects of HMG-CoA reductase inhibitors [14], it is likely that cardiometabolic risk may be greater in individuals with autoimmune hypothyroidism because of a possible overlapping effect of thyroid autoimmunity and thyroid hypofunction.

Although HMG-CoA reductase inhibitors are the first-line therapy for reducing the risk of cardiovascular mortality and morbidity, their use seems to be associated with a slightly increased risk of development of diabetes [26]. In the current study, women with thyroiditis were more insulin-resistant than the remaining participants, which cannot be explained by differences in body weight. Although the role of thyroid autoimmunity in insulin resistance is not yet well defined, some authors reported raised TPOAb titers in clinically euthyroid patients with type 2 diabetes mellitus compared with the nondiabetic controls [27], as well as an association of thyroid autoimmunity in euthyroid women with higher glycated hemoglobin levels and the increased prevalence of metabolic syndrome [28]. The current study, assessing for the first time the impact of HMG-CoA reductase inhibitors on glucose homeostasis in subjects with positive thyroid antibodies, showed that moderate-intensity atorvastatin treatment deteriorated insulin sensitivity in women with Hashimoto’s thyroiditis but not in individuals without thyroid pathology.

Similar changes in plasma lipids clearly indicate that between-group differences in atorvastatin action cannot be attributed to hypolipidemic effects of atorvastatin. It is generally accepted that extra-lipid properties of HMG-CoA reductase inhibitors result from their inhibitory effect on synthesis of nonsterol isoprenoids (farnesyl-pyrophosphate and geranylgeranyl-pyrophosphate), the nuclear factor-κB pathway, as well as from an inhibitory effect on the leukocyte function-associated antigen-1 intercellular adhesion molecule-1 interaction [29]. The opposite effects on protein prenylation, the nuclear factor-κB pathway, and the leukocyte function-associated antigen-1 intercellular adhesion molecule-1 interaction are induced by cytokines and other inflammatory mediators [30-32]. Our findings may suggest that differences in atorvastatin action are, at least in part, associated with the pro-inflammatory state, characterizing Hashimoto’s thyroiditis [2, 7]. Throughout the study, circulating levels of hsCRP, which is an established marker of systemic inflammation [17], were higher in women with thyroid autoimmunity than in subjects without thyroid disorder. Relatively small changes in levels of this protein in the affected women suggest that systemic inflammation in subjects with this disorder is not fully reversed by moderate-dose atorvastatin treatment. Moreover, hsCRP was one of only few factors correlating with the severity of thyroiditis, and the decrease in its levels correlated with the remaining cardiometabolic risk factors, assessed in the current study.

Dimorphism in atorvastatin action may be also associated with differences in vitamin D status. Lower 25-hydroxyvitamin D levels in women with thyroid autoimmunity than in control subjects, observed in the present study, support a widely accepted view that autoimmune thyroiditis is accompanied by low vitamin D status and that thyroid antibody titers correlate with the severity of vitamin D deficiency [24, 33]. In line with this explanation, we have found that baseline 25-hydroxyvitamin D and atorvastatin-induced changes in 25-hydroxyvitamin D positively correlated with the impact of treatment on uric acid, hsCRP, fibrinogen, and homocysteine. Moreover, pleiotropic effects of atorvastatin were less pronounced in subjects with vitamin D insufficiency than in individuals with normal vitamin D status, even if 25-hydroxyvitamin D levels within the reference range were a consequence of vitamin D supplementation [34]. Interestingly, despite the same daily dose and treatment duration, atorvastatin improved vitamin D status only in women without thyroid pathology. This dimorphism in atorvastatin action was also observed in previous studies carried out by our research team. Although the drug was found to have a neutral effect on 25-hydroxyvitamin D levels in subjects with autoimmune thyroiditis [11, 35], the treatment improved vitamin D status in women without endocrine abnormalities [36] or with non-classic congenital adrenal hyperplasia [37]. Statin-induced inhibition of HMG-CoA reductase may lead to an increase in 7-dehydrocholesterol, providing a substrate for the synthesis of 25-hydroxyvitamin D [38]. Alternatively, because atorvastatin and 25-hydroxyvitamin D are both metabolized by CYP3A4, occupation of this enzyme by a statin may slow down vitamin D metabolism, resulting in an increase in 25-hydroxyvitamin D levels [39]. This improvement in vitamin D status seems to be absent in individuals with autoimmune thyroid disease probably because cytokines and other pro-inflammatory mediators interfere with steroidogenesis [40] and dysregulate cytochrome P450 expression and activity [41].

Lastly, weak pleiotropic effects of atorvastatin in women with thyroiditis may be attributed to the unfavorable effect of atorvastatin on insulin sensitivity. Both groups differed in HOMA1-IR, which is a validated surrogate marker of insulin resistance [42], and HOMA1-IR correlated with antibody titers. One can assume that deterioration in insulin sensitivity in one study group may partially counterbalance the direct effect on the remaining risk factors assessed in the present study. This explanation seems, however, less likely than the previous ones. Contrary to previous studies of our research group [43, 44], at entry there were no correlations between HOMA1-IR and uric acid, fibrinogen and homocysteine, while correlations with hsCRP and 25-hydroxyvitamin D were weaker than correlations between both these factors and other cardiometabolic risk factors. Moreover, there were no relationships between changes in uric acid, hsCRP, fibrinogen, homocysteine and 25-hydroxyvitamin and insulin sensitivity. This discrepancy between results of the present and of previous ones may be associated with greater insulin sensitivity in euthyroid patients with Hashimoto’s thyroiditis than in subjects with hyperprolactinemia [43] or early onset androgenic alopecia [44].

We can only speculate about mechanisms by which autoimmune thyroiditis impairs insulin responsiveness to statin therapy. High concentrations of HMG-CoA reductase inhibitors have been found to promote insulin resistance through at least two mechanisms: a decrease in translocation of GLUT-4 to plasma membranes and activation of the Nod-like receptor family pyrin domain containing 3 (NLRP3)/caspase-1 inflammasome [45, 46]. The GLUT-4 transporter plays a pivotal role in the uptake of glucose in peripheral cells, while NLRP3 is an upstream inhibitor of Akt, an important amplifier of multiple pathways in insulin action [46, 47]. Contrary to high doses, the impact of moderate doses may be observed only in some settings, and one of them is probably thyroid autoimmunity. It has been found that translocation and membrane expression of GLUT-4 are inhibited by tumor necrosis factor-α, interleukin-1β and interferon-γ [48-50], cytokines secreted in increased amounts by monocytes and lymphocytes of euthyroid women with autoimmune thyroiditis [7]. Moreover, euthyroid patients with autoimmune thyroiditis were characterized by increased production of NLRP3, correlating with thyroid expression of pro-inflammatory cytokines [51]. This explanation, based only on indirect pieces of evidence, requires further verification in experimental models.

To obtain a homogenous population, the study included only women younger than 55 years old. However, prevalence of cardiovascular and metabolic disorders raises with age [52]. Moreover, thyroid autoimmunity is most frequently observed in women over 60 years of age [53]. This means that a significant proportion of statin-treated women with coexisting Hashimoto’s thyroiditis are older than the participants of our study. However, two observations indirectly suggest that the presence of autoimmune thyroiditis may impair cardiometabolic effects of statins independently of age. First, the impact of atorvastatin on the measured variables did not correlate with age. Second, between-groups differences in percent changes from baseline in the measured cardiometabolic risk factors were also observed if we analyzed only data of 10 women (5 in each group) who were after menopause (data not shown). This suggests that postmenopausal changes in hypothalamic-pituitary-ovarian axis activity do not modify statin action on cardiovascular and metabolic risk factors. Because our study does not allow us to conclude about the relationship between thyroid autoimmunity and statin action in patients older than 55 years, future research should address this issue.

Our results need to be interpreted also in light of other study limitations. Although the study had sufficient statistical power, a small number of participants and short treatment duration make drawing any strong conclusions difficult. The study assessed only surrogate parameters as a substitute for clinically relevant endpoints. Because the study comprised patients with low selenium intake [54] and sufficient iodine intake (due to iodine prophylaxis) [55], it cannot be completely ruled out that the effect of atorvastatin may be different in individuals with adequate selenium and/or low iodine intake. Lastly, the participants received moderate-intensity atorvastatin therapy, and therefore, it is uncertain whether similar effects are observed in patients treated with high doses of this drug and in patients receiving other HMG-CoA reductase inhibitors.

In conclusion, young and middle-aged euthyroid women with Hashimoto’s thyroiditis exhibited a more proatherogenic profile and were more insulin-resistant than age-, body mass index-, blood pressure-, plasma lipid- and plasma hormone-matched hypercholestrolemic women without thyroid disorder. The effect of atorvastatin treatment on hsCRP and homocysteine was greater in patients without thyroid pathology than in women with autoimmune thyroiditis and only in the former group of patients the drug affected uric acid, fibrinogen, and 25-hydroxyvitamin D levels. Despite a small reduction in TPOAb titers, atorvastatin administered to women with thyroiditis reduced insulin responsiveness. The obtained results indicate that euthyroid women with Hashimoto’s thyroiditis may benefit to a lesser degree from atorvastatin treatment than other women with hypercholesterolemia. Owing to numerous study limitations, the obtained results should be, however, interpreted with caution and need to be confirmed in large prospective studies.

Statement of Ethics

The study was conducted in accordance with the 1964 Declaration of Helsinki and its later amendments and was approved before its beginning by the Institutional Review Board (the Bioethical Committee of the Medical University of Silesia (KNW/0022/KB/209/15; October 7, 2015). Written informed consent to participate in the study was obtained from all participants.

Conflict of Interest Statement

The authors have no conflicts of interest to disclose.

Funding Sources

This work did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author Contributions

Robert Krysiak: conceived of the study, participated in its design, performed the statistical analysis, as well as drafted and edited the manuscript. Karolina Kowalcze: conducted the literature search, carried out the assays, and performed the statistical analysis. Bogusław Okopiń: participated in its design and coordination, and provided critical input during manuscript preparations. All authors read and approved the final manuscript.

Data Availability Statement

All data generated or analyzed during this study are included in this manuscript. Further inquiries can be directed to the corresponding author.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC). Usage and distribution for commercial purposes requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.