Association between sensitivity to thyroid hormones and risk of arrhythmia in patients with coronary heart disease: a RCSCD-TCM study in China

Thyroid dysfunction is closely related to arrhythmia. However, the relationship between sensitivity to thyroid hormone and risk of arrhythmia remains unknown. This study aimed to investigate the association between the thyroid system complex index and risk of arrhythmia in patients with coronary heart disease (CHD). This large, multicenter study included 28,413 patients with CHD. Central sensitivity to thyroid hormone was assessed by calculating the thyroid feedback quantile-based index (TFQI). Logistic regression was used to analyze the relationship between sensitivity to thyroid hormone and risk of arrhythmia. This study also assessed the relationship between sensitivity to thyroid hormone and risk of arrhythmia in different sexes, ages, and glucose regulation and blood lipid states. Of the 28,413 participants, 8935 (31.4%) patients with CHD had arrhythmia. There was a remarkable association between TFQI and risk of arrhythmia (odds ratio [OR]: 0.783; 95% confidence interval [CI], 0.735–0.836). Furthermore, the association between the TFQI and risk of arrhythmia in women (OR: 0.731; 95% CI: 0.667–0.802) was stronger than that in men (OR: 0.894; 95% CI: 0.816–0.910), as well as higher in elder (OR: 0.779; 95% CI: 0.721–0.843) than middle-aged (OR: 0.789; 95% CI: 0.703–0.886) patients. Furthermore, the association was strong in the state of diabetes (OR: 0.635; 95% 0.569–0.709) and dyslipidemia (OR: 0.706; 95% CI: 0.644–0.774). There is a remarkable association between sensitivity to thyroid hormone and risk of arrhythmia in patients with CHD, which is more pronounced among women and the elderly. The association is also stronger in CHD patients with dyslipidemia or diabetes.


Introduction
Coronary heart disease (CHD) is a primary public health problem worldwide, accounting for 914 million deaths annually [1]. Sudden cardiac death related to arrhythmia represents a major cause of death in patients with CHD [2]. Patients with CHD are prone to arrhythmia, and CHD is responsible for sudden cardiac death, including arrhythmic and non-arrhythmic deaths [3,4]. Arrhythmia in patients with CHD is often associated with poor prognosis [5]. Therefore, it is of great significance to identify modifiable and possibly reversible risk factors for the prevention and treatment of arrhythmia in patients with CHD.
Animal and clinical studies have shown that thyroid function is closely associated with arrhythmia and that thyroid function may affect susceptibility to arrhythmia by These authors contributed equally: Fanfan Liu, Yijia Liu, Zhu Li influencing cardiac electrophysiology [6][7][8]. The initial evaluation of patients with arrhythmias recommended measurement of thyroid function [9]. However, the influence of thyroid function on the risk of arrhythmia has been inconsistent in previous studies, with some clinical studies indicating that decreased thyrotropin (TSH) is associated with an increased risk of arrhythmia [10,11], while other studies have found that higher free thyroxine (FT4) is associated with increased risk of arrhythmia and that TSH level is not associated with increased risk of arrhythmia [12,13]. This suggests that a single indicator of thyroid function cannot adequately explain the association between the thyroid system and risk of arrhythmia. As a composite indicator of thyroid function, sensitivity to thyroid hormone has been proven to be related to the risk of hypertension, cardiovascular disease [14,15], prediabetes [16], diabetes, and metabolic syndrome [17]. However, currently, there is no study on the association between sensitivity to thyroid hormone and risk of arrhythmia in patients with CHD.
Therefore, this study aimed to investigate the association between sensitivity to thyroid hormone and risk of arrhythmia in patients with CHD and to further explore this association in different ages, sexes, and glucose regulation and blood lipid states.

Study population
This study was approved by the ethics committee of the Tianjin University of Traditional Chinese Medicine (TJUTCM-EC20190008), and the requirement for informed consent was waived. This study was registered with the Chinese Clinical Trial Registry (ChiCTR-1900024535)  Arrhythmia was defined based on the International Classification of Diseases, tenth revision, Clinical Modification, including atrial fibrillation (I48), atrial flutter (I48), tachycardia (I47), conduction block (I44, I45), and ventricular fibrillation (I49). Patients with CHD were grouped based on whether they have arrhythmia (with arrhythmia, N = 8935; without arrhythmia N = 19,478). A flowchart for the enrollment of the study is shown in Fig. 1.

Statistical analyses
Continuous variables are represented median (interquartile range). Categorical variables were expressed as number of events and percentages. The characteristics of the participants in the different groups were analyzed using the independent Mann-Whitney U test. The statistical differences in categorical variables between the two groups were determined by χ 2 test. Adjusting for important risk factors for developing study events, including age, sex, hypertension, dyslipidemia, diabetes, and heart failure. Next, logistic regression analysis was used to calculate the relationship between thyroid hormone sensitivity and arrhythmia, as well as the relationship between the quartile of thyroid hormone sensitivity index and risk of arrhythmia, with the first quartile being the reference group. The results of logistic regression analysis were expressed as odds ratio (OR) and 95% confidence interval (CI). Statistical significance was set at P < 0.05. Statistical analyses were performed using SPSS 25.0 (IBM Corp, New York, NY, USA).

Baseline characteristics
The basic characteristics of the 28,413 eligible patients with CHD are presented in Table 1. The average age of all the patients with CHD was 64 years, and the proportion of men (51.5%) was slightly higher than that of women (48.5%). Among them, 8935 (31.4%) patients with CHD had arrhythmia, and 9383 (33.0%) patients suffered from diabetes. Compared to patients with CHD without arrhythmia, participants with arrhythmia were more likely to be elderly, female, and with a higher prevalence of hypertension and heart failure. In addition, patients with CHD with arrhythmia tended to have higher levels of FT3/FT4, while the levels of TFQI, TSHI, TT4RI, and PTFQI were lower.

Association between sensitivity to thyroid hormones and the risk of arrhythmia
The association between the sensitivity to thyroid hormone and risk of arrhythmia is shown in Table 2. There was a negative association between TFQI (OR: 0.783; 95% CI: 0.735-0.836), PTFQI (OR: 0.745; 95% CI: 0.697-0.797), and TSHI (OR: 0.904; 95% CI: 0.885-0.923), and risk of arrhythmia, whereas there was a positive association between FT3/FT4 (OR: 1.138; 95% CI: 1.113-1.164) and risk of arrhythmia. In addition, there was no statistical relationship between TT4RI and risk of arrhythmia in all models. This study also observed that the association between the sensitivity to thyroid hormones in quartiles and the risk of arrhythmia remains significant after multivariate adjustment in patients with CHD. As shown in Table 2, when Q1 was used as a reference, ORs for Q4 of TFQI, PTFQI, TT4RI, TSHI, and FT3/FT4 quartiles for risk of arrhythmia were 0.804 (95% CI: 0.748-0.865), 0.763 (95% CI: 0.710-0.820), 0.693 (95% CI: 0.645-0.745), 0.747 (95% CI: 0.695-0.803), and 1.560 (95% CI: 1.450-1.321), respectively. All associations remained consistent when the analysis was performed in patients in patients with non-thyroid disease (Supplementary Table 1). In addition, we further assessed the association between atrial arrhythmia, ventricular arrhythmia, tachyarrhythmia and bradyarrhythmia, and thyroid hormone sensitivity, respectively, and the results did not change substantially (Supplementary Table 2).
Association of sensitivity to thyroid hormones with risk of arrhythmia in different age and gender As shown in Table 3, after adjustment for confounding factors, sensitivity to thyroid hormone was associated with risk of arrhythmia in different sexes, and the association was stronger in women than in men. In men, TFQI (OR: 0.894; 95% CI: 0.816-0.910) and PTFQI (OR: 0.869; 95% CI: 0.789-0.957) were negatively associated with the risk of arrhythmia, while there was no association between TT4RI, TSHI, and FT3/FT4 and the risk of arrhythmia. In women, there was a negative association between TFQI (OR: 0.731; 95% CI: 0.667-0.802), PTFQI (OR: 0.641; 95% CI: 0.583-0.704), and TSHI (OR: 0.862; 95% CI: 0.837-0.887) and the risk of arrhythmia, but there was a positive association between FT3/FT4 (OR: 1.167; 95% CI: 1.131-1.203) and the risk of arrhythmia. Regardless of age, this relation between sensitivity to thyroid hormone and arrhythmia remains significant after adjusting for confounding factors.
The association in the elderly was stronger than in middle-aged patients with CHD ( Association of sensitivity to thyroid hormones with risk of arrhythmia in different glucose regulation and blood lipid state As shown in Table 5, the association remains significant between the sensitivity to thyroid hormone and arrhythmia in different glucose metabolic states. After adjusting for confounding factors, the association between the TFQI and risk of arrhythmia in diabetes (OR: 0.635; 95% CI: 0.569-0.709) was higher than in non-diabetes (OR: 0.877; 95% CI: 0.810-0.950).
As shown in Table 6, after adjusting for confounding factors, the TFQI of patients with CHD was remarkably associated with arrhythmia in different lipid states, and the association between the TFQI and risk of arrhythmia in patients with CHD in a state of dyslipidemia (OR: 0.706; 95% CI: 0.644-0.774) was higher than that of patients with CHD in ortholiposis (OR: 0.865; 95% CI: 0.791-0.947).

Discussion
This is the first large-scale study to demonstrate the relation between sensitivity to thyroid hormone and arrhythmia in patients with CHD. The results proved the significant   association between sensitivity to thyroid hormone and arrhythmia in patients with CHD, and this association remained significant in different sexes, ages, and glucose metabolism and blood lipid states.
Patients with CHD are prone to arrhythmia [21]; once arrhythmias are present in patients with CHD, they experience a worse prognosis [2]. Therefore, it is of great clinical significance to understand the relation between sensitivity to thyroid hormone and arrhythmia in patients with CHD. In this population-based study, we found that decreased central thyroid hormone sensitivity (increased TFQI, TSHI, TT4RI, and PTFQI) and increased peripheral thyroid hormone sensitivity (increased FT3/FT4 ratio) were associated with a lower risk of arrhythmia in patients with CHD. This study also stratified the relation between sensitivity to thyroid hormone and arrhythmia in different ages, sexes, and glucose metabolism and blood lipid states. This study found that the association remains significant between the sensitivity to thyroid hormone and arrhythmia in patients with CHD in all ages and sexes and was more pronounced in women and the elderly. In addition, this study observed a significant association between the sensitivity to thyroid hormone and arrhythmia in patients with CHD in all states of glucose metabolism and blood lipid and was stronger in diabetes and dyslipidemia.
The thyroid function has a profound effect on regulating heart rhythm [22,23]. Arrhythmia is defined as an irregular heartbeat, severely exceeding the normal range [24]. Thyroid hormones play a momentous role in cardiac electrophysiology and Ca 2+ processing, which may affect   [25]. Circulating thyroid hormone is regulated by the hypothalamicpituitary-thyroid axis through a negative feedback mechanism. Thyroid hormones also affect the heart in three ways. The active thyroid hormone triiodothyronine (T3) binds to the T3 nuclear receptor, resulting in specific cardiac gene expression. Then, T3 affects the sensitivity of the peripheral sympathetic nervous system. Finally, T3 affects the mechanical state of peripheral blood, leading to increased cardiac filling, thus affecting the cardiac contraction mode [26]. Sawin et al. found a 3.3-fold increase in atrial fibrillation risk when TSH ≤ 0.1 mIU/L [27], and analysis of cardiovascular health studies showed that participants with low TSH concentrations had an increased risk of arrhythmias even when FT4 concentrations were normal [10]. In contrast, a previous cross-sectional study involving 5860 subjects found that FT4 concentration (but not TSH) was associated with arrhythmia [28], while a study of 30,085 participants based on 11 prospective cases found that elevated FT4 was associated with an increased risk of arrhythmia, while TSH levels were not associated with arrhythmia risk [29]. Previous research works on the inconsistent relation between thyroid function and arrhythmia may suggest that a single thyroid indicator is insufficient to better predict the risk of arrhythmia.
Physiologically, thyroid hormone and TSH are negatively correlated, showing a negative feedback loop. TSH values alone do not truly measure the severity of hypopituitarism because TSH secretion is still inhibited by a negative feedback of FT4 concentration [30]. The thyroid hormone resistance index is a measure of central sensitivity/resistance, that is, the degree of pituitary inhibition, by the levels of TSH and FT4. Recently, the composite indices TFQI, PTFQI, TT4RI, and TSHI based on the interaction estimation of FT4 and TSH have been used to estimate the central sensitivity of thyroid hormones. Increased thyroid hormone resistance index seems to be an indicator of decreased thyroid hormone sensitivity [14]. FT3 is converted in peripheral tissues by deiodinase from FT4. Therefore, FT3/FT4 ratio (FT3/FT4) can be used not only as an indicator of deiodinase activity, but also as an indicator of peripheral sensitivity to thyroid hormone. The comprehensive index can systematically reflect the regulation of thyroid hormone homeostasis compared to the single thyroid index. As a composite indicator of thyroid function, sensitivity to thyroid hormone has been proven to be associated with cardiovascular metabolic parameters such as diabetes, metabolic syndrome, and hypertension in different populations [14][15][16][17]. This study showed that among patients with CHD, there was a significant relation between sensitivity to thyroid hormone and arrhythmia. It is interesting to note that the association was stronger in patients with diabetes than those without diabetes, which may be due to the elevated blood glucose levels associated with risk of arrhythmia and change in thyroid function [31,32]. Furthermore, this study also found a stronger association between sensitivity to thyroid hormone and arrhythmia in women patients with CHD than in men, possibly because women are more prone to changes in thyroid function [33]. In addition, CHD usually coexists with dyslipidemia [34]. The present study found that in different blood lipid states, sensitivity to thyroid hormone was strongly related to arrhythmia in patients with CHD with dyslipidemia than in those with ortholiposis, which may be mainly due to the fact that dyslipidemia itself is a risk factor for arrhythmia [35]. Further studies may be needed in future to elucidate the mechanisms of these results.
This study has several limitations. First, in multicenter studies, there may be deviations in different measurement methods. However, external quality assessments were conducted between clinical laboratories at each center. Second, this study did not have a track record of whether participants received thyroid disease-specific treatment. In addition, a cross-sectional design based on a single measurement was insufficient to establish a causal association between sensitivity to thyroid hormone and risk of arrhythmia in patients with CHD, despite adjusting for multiple confounders. The exact mechanism between thyroid hormone sensitivity and arrhythmia requires further large-scale prospective studies.

Conclusion
This study demonstrated that sensitivity to thyroid hormone, as a comprehensive indicator of thyroid function expression, was significantly associated with arrhythmia in patients with CHD, and the relation was more pronounced among females than males, as well as higher in elder than middle-aged patients with CHD. Furthermore, under the condition of dyslipidemia and diabetes, the given association was stronger. These findings may highlight the demand for risk management strategies tailored to different populations of patients with CHD to prevent arrhythmia.

Compliance with ethical standards
Conflict of interest The authors declare no competing interests.
Ethics approval This study was approved by the ethics committee of the Tianjin University of Traditional Chinese Medicine (approval number TJUTCM-EC20190008) and registered with the Chinese Clinical Trial Registry on July 14, 2019 (registration number ChiCTR-1900024535) and in ClinicalTrials.gov on July 18, 2019 (registration number NCT04026724). Written informed consent for participation was not required for this study in accordance with national legislation and institutional requirements.