Hyperthyroidism in Graves Disease Causes Sleep Disorders Related to Sympathetic Hypertonia

Abstract

Context

It is well known that Graves disease (GD) causes sleep disorders (SDs). However, the characteristics and associated factors of SD and its clinical course post hyperthyroidism normalization remain unclear.

Objective

To clarify the characteristics and associated factors of subjective SD and its clinical course after GD treatment.

Methods

From November 2017 to October 2020, we enrolled 72 participants (22 newly diagnosed with GD with untreated hyperthyroidism, 20 previously diagnosed with GD with normal thyroid function, and 30 normal controls) with no other underlying SD-related diseases. We compared the groups at enrollment and conducted prospective observations after 12 months of treatment on participants with newly diagnosed GD. Main outcome measures were differences and changes in the Pittsburgh Sleep Quality Index (PSQI) global and component sleep quality scores.

Results

PSQI global sleep quality scores (P = .036) and sleep disturbance scores (P = .011) were significantly different among the 3 groups, and were highest in the untreated hyperthyroidism group. Multiple regression analysis demonstrated that free thyroxine level, which was positively correlated with sympathetic tone (ST) as evaluated by pulse rate, and urinary total metanephrines was associated with poorer PSQI global sleep quality scores independently of other factors (P = .006). Prospective observation showed that PSQI global sleep quality scores (P = .018) and sleep disturbance scores (P = .011) significantly improved with thyroid function normalization and ST attenuation.

Conclusion

Hyperthyroidism caused by GD augmented ST and exacerbated subjective SD. Normalization of hyperthyroidism caused by GD improved subjective SD.

hyperthyroidism, Graves disease, subjective sleep disorders, Pittsburgh Sleep Quality Index, sympathetic tone

Sleep disorders (SDs) are common clinical problems and have negative effects on quality of life such as physical, mental, and social functions (1). SDs are described by 2 different terms according to the method of evaluation. Subjective SDs are assessed by self-report questionnaires and a sleep diary, while objective SDs are assessed by the manifestations obtained by polysomnography and actigraphy (2, 3). There are many different types of primary SD, including insomnia, circadian rhythm disorder, sleep-disordered breathing, hypersomnia/narcolepsy, parasomnia, and restless legs syndrome/periodic limb movement disorder. Primary SDs affect the nocturnal thyrotropin (TSH) surge and increase the levels of free thyroxine (FT4) and free triiodothyronine transiently (4, 5). Secondary SDs are caused by various medical and psychiatric conditions, such as pulmonary disease, cardiac disease, neurologic disease, depression, anxiety, and thyroid dysfunction. Thus, thyroid dysfunction is a common cause of SD (6).

Hypothyroidism is characterized by sleep-disordered breathing conditions such as sleep apnea syndrome (SAS) (7), symptoms of which include difficulty falling asleep at night and frequent waking during the night (8). Hyperthyroidism causes tremor, fatigue, anxiety, palpitations, and SDs (9). Approximately 43.0% to 71.9% of patients with hyperthyroidism experience SDs, including difficulty falling asleep at night, difficulty staying asleep at night, and reduced sleep efficiency (10-13). However, previous studies have used retrospective designs and have not evaluated the characteristics of SD associated with hyperthyroidism in Graves disease (GD) in detail. For example, 1 prospective study discussed the relationships between hyperthyroidism and mental disorders such as anxiety, depression, and cognitive function (14). However the authors did not examine whether SDs improved following therapeutic intervention. Additionally, the factors involved in underlying SDs associated with hyperthyroidism in GD remain to be clearly understood. Hyperthyroidism increases tissue sensitivity to catecholamines and activates the renin–angiotensin–aldosterone system, resulting in increased heart rate, blood volume, and myocardial contractility (15). Sympathetic nerve activation is likely to correlate with SDs (16). We hypothesized that hyperthyroidism may activate the sympathetic nervous system, resulting in SDs. Clarification of the characteristics and factors involved in underlying SDs in GD may lead to effective therapeutic interventions and improve the quality of life of patients with GD.

We conducted a prospective study with a substudy that involved cross-sectional analysis at baseline to evaluate the relationship between thyroid function and the characteristics of subjective SDs and to clarify that sympathetic hypertonia due to hyperthyroidism affect SDs by assessing surrogate markers of sympathetic tone (ST) in patients with GD.

Materials and Methods

Participants and Study Design

From November 2017 to October 2020, we enrolled 72 participants aged 20-90 years. Of these, 22 were diagnosed as having GD with current hyperthyroidism (HT group), 20 with GD and normal thyroid function following treatment (NF group), and 30 were healthy individuals (normal controls; CR group). Patients with GD were enrolled in the outpatient clinic of Tottori University Hospital. The CR group was recruited by a notice in Tottori University Hospital. The required sample size was calculated using the paired-samples t-test. With alpha = 0.05, beta = 0.2, and effect size = 0.60, the required sample size for each group was estimated to be 24. However, after applying the exclusion criteria (eg, SAS), the final sample size was 22 in the HT group and 20 in the NF group (Fig. 1).

Figure 1.

Enrollment and exclusion of study participants. Patients with GD were divided into HT and NF groups. Abbreviations: CR, normal control group; GD, Graves disease; HT, hyperthyroidism group; NF, normal thyroid function group; SAS, sleep apnea syndrome.

GD was diagnosed in accordance with the diagnostic guideline of the Japan Thyroid Association (17). The NF group was defined by TSH and FT4 levels within the normal range. The CR group comprised volunteers who had no history of thyroid disease, TSH and FT4 levels within the normal range, and absence of TSH receptor autoantibodies (TRAbs). Participants with a history of pregnancy and lactation, diseases related to somnipathy (eg, respiratory disease, heart failure, cerebrovascular disease, psychiatric diseases), conditions that affected sleep (eg, diabetes mellitus, dementia), renal dysfunction, and SAS were excluded. The HT group underwent treatment to normalize thyroid function according to the Japan Thyroid Association guidelines (18) and SDs were reassessed in this group 12 months after treatment. SDs were evaluated using the Pittsburgh Sleep Quality Index (PSQI) global and component sleep quality scores. ST was evaluated using surrogate markers including pulse rate and urinary total metanephrines. The study protocol was in accordance with the ethical principles stated in the Declaration of Helsinki and was approved by the ethics committee of Tottori University Hospital (17B004). Written informed consent was obtained from each participant. Details of the study design are elsewhere (Figure 1 (19)).

Biochemical Assessment

FT4 was measured using electrochemiluminescence immunoassays (ECLIAs) (Roche Diagnostics, Tokyo, Japan). Interassay variability, intra-assay coefficients of variation, measurement range, and reference range were <10%, <15%, 0.023 to 7.77 ng/dL, and 0.8–1.7 ng/dL, respectively. TSH was measured using an ECLIA (Roche Diagnostics, Tokyo, Japan). The interassay variability, intraassay coefficients of variation, measurement range, and reference range were <10%, <18%, 0.005-100 μU/mL, and 0.27–4.2 μU/mL, respectively. TRAb was measured using a third-generation TSH-binding inhibitory immunoglobulin assay that uses the automated Cobas ECLIA (Roche Diagnostics, Tokyo, Japan). Interassay variability, intra-assay coefficients of variation, measurement range, and reference range were <10%, <15%, 0.8 to 40 IU/L, and <2.0 IU/L, respectively. Urinary total metanephrines were measured using high-performance liquid chromatography (SRL, Tokyo, Japan) and liquid chromatography tandem mass spectrometry (SRL, Tokyo, Japan). Interassay variability, intra-assay coefficients of variation, measurement range, and reference range were <15%, <20%, 0.01 to 99999.99 mg/L, and 0.13 to 0.52 mg/L, respectively. When spot urine was collected, the samples were stored at <6°C and analyzed within a few days. Values were expressed in mg/g of creatinine (Cr). The reported reference range for urinary total metanephrines is 0.27 to 0.35 mg/g × Cr (20, 21).

The Pittsburgh Sleep Quality Index

The PSQI is a self-administered questionnaire that assesses subjective sleep quality during the previous month (22, 23). The self-rated 18 items of the PSQI generate 7 component scores (range of subscale scores, 0-3): sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbance, sleep medication, and daytime dysfunction. Sleep quality is a subjective assessment of overall sleep, sleep latency is the ease of falling asleep, sleep duration is the length of total sleep time, habitual sleep efficiency is the ratio of actual sleep time to bedtime, sleep disturbance is the degree of difficulty with sleep maintenance, sleep mediation is the frequency of using medication to sleep, and daytime dysfunction is the assessment of sleepiness associated with sleep problems. The sum of these 7 component scores produces a global sleep quality score (range 0-21); higher scores represent poorer subjective sleep quality. The original version (22) and the Japanese version (23) of the PSQI are comparable and have a cutoff score of 6. Participants completed the Japanese version of the PSQI. A PSQI score of ≥6 was diagnosed as indicating SD.

Portable Sleep Monitoring

All participants underwent portable sleep monitoring at home. SAS was determined using portable sleep monitoring results Type 3 Alice Night One® (Philips Respironics, Tokyo, Japan), Type 3 Stardust® (Philips Respironics, Tokyo, Japan), or Type 3 LS-330G® (Fukuda Denshi, Tokyo, Japan). A nasal cannula was placed at the nares to measure respiratory airflow using a disposable airflow sensor, and a breathing band connected to a sensor monitored respiratory movements of the middle chest wall. Arterial oxygen saturation was continuously monitored using a pulse oxymeter. Trained technicians (M.F. and E.M.) analyzed sleep states according to a clinical guideline (24). Apnea was defined as a continuous cessation of respiratory airflow for ≥10 seconds per hour of sleep; hypopnea was defined as a ≥50% reduction in respiratory airflow with arterial oxygen desaturation ≥3%. The apnea/hypopnea index (AHI) was calculated from the portable sleep monitoring results as the total number of episodes of apnea and hypopnea per hour of sleep. SAS was defined as AHI ≥ 15 (25).

Statistical Analysis

Continuous valuables are presented as the median (interquartile range) and categorical data are summarized as numbers (percentage). Differences among the 3 groups were analyzed using the Kruskal–Wallis test. The chi-squared test was used to evaluate whether the categorical values had a significant group effect. Differences in PSQI global sleep quality and component scores between enrollment and 12 months after therapeutic intervention were analyzed using the Wilcoxon signed rank sum test. We investigated the association between pulse rate, urinary total metanephrines, and FT4 using Pearson’s correlation coefficient. We also performed multiple regression analysis to predict factors related to PSQI global sleep quality score by adjusting for age, sex, body mass index, smoking habits, and drinking habits. All statistical analyses were performed using Bellcurve for Excel version 3.21 (Social Survey Research Information Co., Ltd., Tokyo, Japan), and P < .05 was considered statistically significant.

Results

Cross-sectional Analysis Comparing GD With HT, GD With NF After Treatment, and CR Groups

A total of 72 participants (42 with GD and 30 CR) were divided into 3 groups according to their thyroid function and TRAb titer (Fig. 1). Their clinical backgrounds are summarized in Table 1. There were no significant differences in age, sex, diastolic blood pressure, smoking habits, drinking habits, and AHI. The pulse rate was significantly higher in the HT group than in the other groups (vs NF, P < .001 and vs CR, P < .001). TSH (vs NF, P < .001 and vs CR P < .001) of the HT group was significantly lower, and FT4 (vs NF, P < .001 and vs CR P < .001) and TRAb (vs NF, P = .009 and vs CR P < .001) significantly higher than those of the other groups. Urinary total metanephrines were significantly higher in the HT group than in the other groups (vs NF, P = .047 and vs CR, P = .028). Significant positive correlations between FT4 and surrogate markers of ST such as pulse rate (r = 0.643, P < .001) and urinary total metanephrines (r = 0.387, P < .001), were observed.

Table 1.

Clinical background of participants with GD and CR

	HT (n = 22)	NF (n = 20)	CR (n = 30)	P value
Age, years (IQR)	41 (35-49)	49 (38-51)	38 (29-48)	.111
Male/Female, n (%)	3/19 (13.6/86.4)	1/19 (5.0/95.0)	6/24 (20.0/80.0)	.323
BMI, kg/m2(IQR)	21.7 (19.9-24.3)	23.0 (21.7-24.0)	20.5 (19.5-21.6)	.008a
Systolic blood pressure, mmHg (IQR)	132 (122-145)	124 (114-138)	116 (106-125)	.004a
Diastolic blood pressure, mmHg (IQR)	75 (65-80)	74 (66-80)	71 (64-78)	.768
pulse rate, bpm (IQR)	70 (67-76)	58 (55-66)	59 (54-62)	<.001a
Drinking habits (never/current/unknown), n (%)	15/7/0 (68.2/31.8/0)	11/8/1 (55.0/40.0/5.0)	23/7/0 (76.7/23.3/0)	.338
Smoking habits (never/current/unknown), n (%)	17/5/0 (77.3/22.7/0)	18/1/1 (90.0/5.0/5.0)	29/1/0 (96.7/3.3/0)	.069
TSH, μU/mL (IQR)	0.0 (0.0-0.0)	2.4 (1.3-3.5)	1.6 (1.1-2.0)	<.001a
FT4, ng/dL (IQR)	4.5 (3.6-6.8)	1.2 (1.1-1.4)	1.3 (1.2-1.3)	<.001a
TRAb, IU/L (IQR)	7.7 (4.6-15.6)	3.6 (1.7-5.8)	0.8 (0.8-0.8)	<.001a
Urinary total metanephrines, mg/g × Cr (IQR)	0.38 (0.27-0.51)	0.30 (0.23-0.34)	0.28 (0.25-0.35)	.018a
Urinary metanephrines, mg/g × Cr (IQR)	0.13 (0.09-0.17)	0.10 (0.08-0.12)	0.10 (0.09-0.13)	.065
Urinary normetanephrines, mg/g × Cr (IQR)	0.24 (0.18-0.34)	0.20 (0.15-0.23)	0.18 (0.15-0.22)	.031a
AHI, times/h (IQR)	2.0 (1.0-4.5)	2.6 (1.3-6.1)	1.2 (0.5-3.1)	.133
Thyroid medication (ATD/ATD and LT4/RAI/OPE), n (%)	0/0/0/0	14/5/1/0 (70.0/25.0/5.0/0)	0/0/0/0	<.001a
ATD
MMI 2.5 mg/day, n (%)		2 (10.0)
MMI 5 mg/day, n (%)		8 (40.0)
MMI 10 mg/day, n (%)		2 (10.0)
MMI 15 mg/day, n (%)		1 (5.0)
MMI 20 mg/day, n (%)		1 (5.0)
ATD and LT4
MMI 5 mg/day and LT4 25 μg/day, n (%)		3 (15.0)
MMI 10 mg/day and LT4 50 μg/day, n (%)		1 (5.0)
MMI 10 mg/day and LT4 75 μg/day, n (%)		1 (5.0)
Beta-adrenergic blocker, n (%)	5 (22.7)	0	0	.002a
Bisoprolol 2.5 mg/day, n (%)	3 (13.6)
Bisoprolol 5 mg/day, n (%)	1 (4.5)
Metoprolol 60 mg/day, n (%)	1 (4.5)

Abbreviations: AHI, apnea hypopnea index; ATD, antithyroid drug; BMI, body mass index; CR, normal control group; Cr, creatinine; GD, Graves disease; HT, hyperthyroidism group; IQR, interquartile range; LT4, levothyroxine; MMI, methimazole; NF, normal thyroid function group; OPE, operation; RAI, radioiodine therapy; FT4, free thyroxine; TRAb, TSH receptor autoantibody.

^aAll P < .05 were considered to indicate significant differences between the HT, NF, and CR groups.

PSQI global sleep quality and PSQI component scores among the 3 groups are summarized in Fig. 2. PSQI global sleep quality score was significantly different among the 3 groups. PSQI global sleep quality scores were higher in the HT group than in the other groups (vs NF, P = .077 and vs CR, P = .053). A comparison of PSQI component scores among the 3 groups showed that habitual sleep efficiency and sleep disturbance were significantly higher in the HT group than in the CR group (habitual sleep efficiency, P = .003 and sleep disturbance, P = .009) and tended to be higher in the HT group than in the NF group. No significant differences were observed in sleep quality, sleep latency, sleep duration, sleep medication, and daytime dysfunction among the 3 groups. Multiple regression analysis demonstrated a significant positive association between PSQI global sleep quality scores and FT4 (R² = 0.143, standardized β = 0.338, P = .006) (Table 2). However, thyroid medication (R² = 0.049, standardized β = –0.142, P = .278) and TRAb titers (R² = 0.031, standardized β = –0.027, P = .830) did not have significant positive association between PSQI global sleep quality scores.

Table 2.

Multiple regression analysis of factors affecting PSQI global sleep quality among the 3 groups

	Standardized β (SE)	P value
Age	−0.007 (0.025)	.951
Male	0.020 (0.794)	.864
BMI	0.145 (0.112)	.207
Smoking habits	0.145 (0.958)	.220
Drinking habits	−0.131 (0.685)	.255
FT4	0.338 (0.143)	.006a

Abbreviations: β, regression coefficient; BMI, body mass index; PSQI, Pittsburgh Sleep Quality Index; SE, standard error; FT4, free thyroxine

^aAll P < .05 were considered to indicate significant association.

Figure 2.

PSQI global sleep quality (A) and PSQI component (B-H) scores. The PSQI comprises 7 components: sleep quality (B), sleep latency (C), sleep duration (D), habitual sleep efficiency (E), sleep disturbance (F), sleep medication (G), and daytime dysfunction (H). There were significant differences among the 3 groups on A, E, and F. *All P < .05 were considered to indicate significant differences between the HT, NF, and CR groups. Abbreviations: CR, normal control group; HT, hyperthyroidism group; NF, normal thyroid function group; PSQI, Pittsburgh Sleep Quality Index.

Prospective Observational Analysis of GD With HT Group to Compare Pre- and Post-treatment Characteristics

Of the 22 patients with hyperthyroidism, 14 were treated and re-evaluated following 12 months of treatment. Clinical characteristics at enrollment and 12 months after therapeutic intervention were compared (Table 3). Systolic blood pressure (P = .024), diastolic blood pressure (P = .020), pulse rate (P = .013), and urinary total metanephrines (P = .002) significantly reduced after therapeutic intervention, whereas body mass index (P = .002) and AHI (P = .009) significantly increased.

Table 3.

Comparison between enrollment and 12 months after hyperthyroidism treatment

	Enrollment	12 months	P value
BMI, kg/m2(IQR)	21.2 (20.0-22.6)	22.8 (21.5-25.1)	.002a
Systolic blood pressure, mmHg (IQR)	132 (123-145)	120 (109-128)	.024a
Diastolic blood pressure, mmHg (IQR)	77 (72-80)	70 (65-78)	.020a
Pulse rate, bpm (IQR)	69 (65-75)	63 (55-67)	.013a
TSH, μU/mL (IQR)	0.0 (0.0-0.0)	2.3 (1.3-4.1)	.002a
FT4, ng/dL (IQR)	4.3 (3.6-5.2)	1.0 (1.0-1.4)	.002a
TRAb, IU/L (IQR)	7.6 (4.6-12.8)	2.8 (1.9-4.5)	.075
Urinary total metanephrines, mg/g × Cr (IQR)	0.42 (0.27-0.52)	0.27 (0.25-0.37)	.002a
Urinary metanephrines, mg/g × Cr (IQR)	0.15 (0.10-0.17)	0.09 (0.07-0.12)	<.001a
Urinary normetanephrines, mg/g × Cr (IQR)	0.26 (0.18-0.34)	0.20 (0.16-0.23)	.004a
AHI, times/h (IQR)	1.6 (1.0-6.0)	5.6 (2.0-13.0)	.009a
Thyroid medication (ATD/ATD and LT4/RAI/OPE), n (%)	0/0/0/0	10/3/1/0 (71.4/21.4/7.1/0)	<.001a
ATD
MMI 2.5 mg/day, n, (%)		1 (7.1)
MMI 5 mg/day, n, (%)		5 (35.8)
MMI 10 mg/day, n, (%)		3 (21.4)
PTU 150 mg/day, n, (%)		1 (7.1)
ATD and LT4
MMI 5 mg/day and LT4 25 μg/day, n (%)		1 (7.1)
MMI 5 mg/day and LT4 100 μg/day, n (%)		1 (7.1)
MMI 10 mg/day and LT4 25 μg/day, n (%)		1 (7.1)
Beta-adrenergic blocker, n, (%)	4 (28.6)	2 (14.3)	.645
Bisoprolol 0.625 mg/day, n (%)	0 (0)	1 (7.1)
Bisoprolol 2.5 mg/day, n (%)	2 (14.3)	0 (0)
Bisoprolol 5 mg/day, n (%)	1 (7.1)	1 (7.1)
Metoprolol 60 mg/day, n (%)	1 (7.1)	0 (0)

Abbreviations: AHI, apnea hypopnea index; ATD, antithyroid drug monotherapy; BMI, body mass index; Cr, creatinine; LT4, levothyroxine; MMI, methimazole; OPE, operation; PTU, propylthiouracil; RAI, radioiodine therapy; FT4, free thyroxine; TRAb, TSH receptor autoantibody.

^aAll P < .05 were considered to indicate significant differences between enrollment and 12 months after treatment.

Figure 3 shows changes in PSQI global and component sleep quality scores. PSQI global sleep quality scores were significantly reduced at 12 months after therapeutic intervention (P = .018). However, PSQI global sleep quality scores of 4 patients with hyperthyroidism were not improved after therapeutic intervention. Sleep quality and sleep disturbance scores were significantly improved at 12 months after therapeutic intervention (sleep quality, P = .028 and sleep disturbance, P = .011). Of the PSQI components, sleep disturbance showed the highest rate of improvement after therapeutic intervention (P < .001). No significant changes were observed in sleep latency, sleep duration, habitual sleep efficiency, sleep medication, and daytime dysfunction after therapeutic intervention. The rate of improvement in PSQI global sleep quality (r = 0.375, P = .186) and the improvement in sleep disturbance (r = 0.468, P = .091) correlated with the reduction in FT4 but was not significant. A significant positive correlation was observed between FT4 and pulse rate (r = 0.754, P = .003).

Figure 3.

Comparisons of PSQI global sleep quality (A) and PSQI component scores (B) between enrollment and 12 months (12 M) after GD treatment. (A) The solid lines indicate patients with improved PSQI global sleep quality scores and the dotted lines indicate patients with worse PSQI global sleep quality scores. Bars indicate the median. (B) Each portion of the graph indicates the distribution of component scores: gray portion, score of 0; dot portion, score of 1; halftone dot mesh portion, score of 2; oblique line portion, score of 3. *All P < .05 were considered to indicate significant differences between enrollment and 12 months after treatment. Abbreviations: GD, Graves’ disease; M, months; PSQI, Pittsburgh Sleep Quality Index.

Discussion

Our study demonstrated 2 novel findings. First, we prospectively demonstrated that SDs caused by GD improved after therapeutic intervention for hyperthyroidism. Second, SD problems such as poor sleep quality and sleep disturbance were correlated with the augmented ST associated with hyperthyroidism.

It is well known that thyroid hormones enhance catecholamine sensitivity and myocardial contractility by increasing β1-adrenergic receptor levels, G protein coupling, and catecholamine synthesis (15, 26-28). Hyperthyroidism is a sympathovagal unbalanced state, which is associated with both increased sympathetic activity and reduced vagal modulation of heart rate. These autonomic dysfunctions can be restored to normal if hyperthyroidism is treated (27-29). We found a reduction in pulse rate and urinary total metanephrines after hyperthyroidism treatment (Table 3), in accord with previous findings. Previous reports suggest that hyperthyroidism is associated with SDs, including difficulty in falling asleep at night, difficulty in maintaining sleep at night, and reduced sleep efficiency (9-14). However, the relationship between SDs and ST augmented by hyperthyroidism in GD has not been discussed prospectively.

Both subjective and objective SDs are not necessarily correlated with each other. Subjective SDs, which are directly related to quality of life, can be evaluated by PSQI (2). On the other hand, evaluations of objective SDs are defined by the findings of polysomnography. Sleep state misperception was reported to cause discrepancy between subjective sleep and objective sleep assessment in patients with SDs (30, 31).

We used portable sleep monitoring in this study not to evaluate sympathetic nerve activity as a mechanism of SDs in GD, but to exclude sympathetic nerve hyperactivity caused by SAS. Our findings expand on previous work by prospectively demonstrating the effect of hyperthyroidism-induced augmentation of ST by pulse rate and urinary total metanephrines on subjective SDs (32, 33).

Sleep onset is associated with a decline in catecholamine plasma levels. Difficulty in sleep maintenance is associated with increases in nocturnal catecholamines in patients with insomnia (16, 34). Zhang et al. reported that short sleep duration and poor sleep quality were associated with increased 24-hour urinary catecholamines in healthy adults (35). Thus, high catecholamine levels are strongly related to poor sleep quality and sleep disturbance. Our findings suggest that hyperthyroidism exacerbated sleep quality and sleep disturbance (as reflected in PSQI component scores) because of ST augmentation and that therapeutic intervention for hyperthyroidism improved SD. We found that 4 of 14 patients did not show an improvement in PSQI global sleep quality scores after therapeutic intervention (Fig. 3), and 2 of these 4 patients experienced worse AHI. The thyroid function of 1 patient did not improve to within the normal range and their sleep disturbance worsened. One patient was still taking beta-adrenergic blockers after follow-up. The patient showed no change in PSQI global sleep quality scores after therapeutic intervention. SAS causes insomnia and deterioration in PSQI global sleep scores (36, 37). Pulse rate as a marker of ST in this study can be lowered by the administration of beta-adrenergic blockers. Therefore, we hypothesized that beta-adrenergic blockers may have reduced PSQI global sleep quality scores. SAS, persistent hyperthyroidism, and beta-adrenergic blockers might prevent improvements in SD after therapeutic intervention for hyperthyroidism. Although PSQI component scores showed a disturbance in habitual sleep efficiency in the HT group, there was no significant improvement in this component after therapeutic intervention. This may be because most scores on this component were 0 or 1 even before the treatment. Hyperthyroidism causes sleep deprivation by stimulating the mitochondria to increase adenosine triphosphate production. This phenomenon leads to increased purinergic neurotransmission and synaptic velocity, which cause sleep deprivation (38, 39). Hyperthyroidism may enhance central nervous system excitability. Therefore, our results suggest that the PSQI component of sleep disturbance in the HT group was exacerbated by the excitability of the central nervous system, resulting in poor sleep quality. Thus, medications that reduce sympathetic activity and suppress arousal, such as orexin receptor antagonists, may be an alternative for managing SDs related to hyperthyroidism caused by GD. We showed that hyperthyroidism in GD caused sleep disturbance and reduced habitual sleep efficiency; these results are in accord with those of previous studies (11, 13, 39). We found that therapeutic intervention for hyperthyroidism improved sleep disturbance but not habitual sleep efficiency. Although hyperthyroidism and ST are involved in sleep disturbance, other factors may be related to habitual sleep efficiency. Additional studies are needed to investigate these factors using objective tests such as polysomnography.

The present study had some limitations. First, we prospectively evaluated only the HT group. The NF and CR groups were not prospectively evaluated. PSQI was evaluated only at enrollment and 12 months after treatment. We did not assess the clinical outcome after 12 months. Second, the sample was relatively small and drawn from a single center. Prospective studies with long follow-ups and large samples are needed to confirm our findings. Third, the evaluation of ST was limited to pulse rate and urinary total metanephrines. We did not assess the sympathetic nerve activity by microneurography during sleep. However this requires invasive and specialized methods that we deemed unsuitable for participants with no symptoms or signs of SDs. In addition, we believe that a combination of central sleep measures and assessment of neurohormonal drivers, such as orexin, dopamine, thyrotropin releasing hormone activity, and parameters obtained by polysomnography, would help to elucidate SDs. Additional studies are needed to elucidate SDs using these parameters.

In conclusion, we demonstrated that hyperthyroidism can cause SDs, especially sleep quality and sleep disturbance, by augmentation of ST. Treatment for hyperthyroidism can improve poor sleep quality and sleep disturbance.

Abbreviations

Abbreviations
 
• AHI

  apnea/hypopnea index

 
• Cr

  creatinine

 
• CR

  normal controls group

 
• ECLIA

  electrochemiluminescence immunoassay

 
• FT4

  free thyroxine

 
• GD

  Graves disease

 
• HT

  hyperthyroidism group

 
• NF

  normal thyroid function group

 
• PSQI

  Pittsburgh Sleep Quality Index

 
• SAS

  sleep apnea syndrome

 
• SD

  sleep disorder

 
• ST

  sympathetic tone

 
• TRAb

  thyrotropin receptor autoantibody

 
• TSH

  thyrotropin

Acknowledgments

We acknowledge support for this work from the Tottori University Hospital exploratory research funds and a scholarship donation from Kyowa Kirin Co. We thank Diane Williams, PhD, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.

Financial Support

This work was supported by the Tottori University Hospital funds for exploratory research and scholarship from Kyowa Kirin Company, Limite (KHKS20200429004).

Disclosure summary

The authors have nothing to disclose.

Data Availability

Some or all of the datasets generated and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.