Association between insomnia and atrial fibrillation in a Chinese population: A cross‐sectional study

Abstract

Background

Insomnia is the most prevalent sleep disorder; however, little research has explored the link between insomnia and atrial fibrillation (AF).

Hypothesis

Insomnia is associated with increased risk of AF in a Chinese population.

Methods

A total of 8371 Chinese participants (4314 males; mean age, 42.4 ± 13.1 years) were enrolled in this cross‐sectional study to investigate the association between insomnia and AF. AF was assessed in a standard supine resting position with a 10‐s 12‐lead electrocardiograph (ECG) or by self‐reported history. Insomnia was assessed using the Athens Insomnia Scale (AIS), and a score of ≥6 was regarded as having insomnia. The association between insomnia and AF was determined by logistic regression analysis.

Results

Among the 8371 subjects, 1074 (12.8%) had different degrees of insomnia, and AF was observed in 50 subjects (0.60%). After adjusting for potential confounders, individuals with insomnia had moderately increased likelihood of suffering from AF compared with those without insomnia (odds ratio [OR]: 1.92, 95% confidence interval [CI]: 1.00‐3.70, P = 0.05). After stratifying data by age, a significant positive association was found in those age <40 years (OR: 6.52, 95% CI: 1.64‐25.83, P = 0.01), and a similar trend existed in males after stratifying by sex, although this relationship was not statistically significant (OR: 2.11, 95% CI: 0.92‐4.83, P = 0.08).

Conclusions

Individuals with insomnia may have a higher risk of AF in the particular Chinese population assessed in this study. Age (<40 years) is a significant factor in the association between insomnia and AF.

1. INTRODUCTION

Insomnia, characterized by difficulty initiating or maintaining sleep followed by a series of symptoms such as drowsiness and fatigue during daytime, is the most prevalent sleep disorder.1 The prevalence of insomnia, according to different definitions, varies from 6% to 33% in >50 epidemiological studies,2 suggesting that individuals with insomnia constitute a considerable proportion on a global scale. Previous studies have indicated that insomnia is a risk factor for the emergence and development of somatic and mental disorders,3, 4, 5 thus significantly contributing to increased healthcare costs worldwide.

Atrial fibrillation (AF) is the most common heart arrhythmia and is associated with substantial increases in morbidity and mortality.6 The prevalence of AF is 0.77% in subjects age 30 to 85 years, and approximately 8 million Chinese were estimated to suffer from AF.6 AF is often accompanied by increased risks of ischemic stroke,7 dementia,8 heart failure,9 and overall mortality.9 Therefore, AF has a significant impact on patient quality of life, and early diagnosis and medical interventions targeting individuals who are at a high risk for AF may significantly reduce the complications and health‐related burden.

Recently, mounting evidence has shown that insomnia is common in patients with cardiovascular diseases and that insomnia is associated with an increased risk of coronary heart disease.10, 11, 12 This indicates that insomnia is emerging as a serious health concern. Indeed, a recent report suggested that sleep disturbances should be included in the top 10 potentially modifiable cardiovascular disease risk factors.13 However, the relationship between insomnia and AF is unknown. Herein, we conducted a population‐based, cross‐sectional study to explore to the relationship between insomnia and AF prevalence in a Chinese community population.

2. METHODS

2.1. Study design and participants

Participant recruitment has been detailed in previous studies.14, 15 Briefly, a total of 9078 residents from the Jidong community who were recruited between 2013 and 2014 agreed to participate in this cross‐sectional study. The Jidong community is located in the city of Tangshan, which is an industrial city in Hebei Province in northern China. Of the 9078 participants, 707 were excluded from our study because of missing electrocardiographic (ECG) data or key variables, cardiovascular diseases, clinical instability, or significant medical comorbidity that was likely to independently impact the outcomes. Thus, 8371 participants were included in this analysis.

Our study was approved by the Ethics Committee of Jidong Oilfield, Inc., and was conducted in accordance with the Declaration of Helsinki. All participants signed informed consent.

2.2. Assessment of insomnia

The entire 8‐item Athens Insomnia Scale (AIS‐8) based on the International Classification of Diseases, Tenth Revision (ICD‐10) diagnostic criteria of insomnia, which was developed by Soldatos and colleagues to evaluate the degree of insomnia,16 was used to assess insomnia in this study. The AIS‐8 is a self‐reported questionnaire containing 8 items. The first 5 items evaluate difficulty with sleep induction, awakening during the night, early‐morning awakening, total sleep time, and overall quality of sleep; the last 3 items focus on sense of well‐being, overall functioning, and sleepiness during daytime. Subjects reported if they experienced the mentioned symptoms ≥3 times every week in the recent 3 months. Each item of the AIS can be scored from 0 to 3, with 0 corresponding to no problem at all and 3 to a very serious problem. The total scores were calculated by adding each item's score, with higher scores suggesting greater severity in insomnia symptoms. Patients with scores of ≥6 were considered as having insomnia.

2.3. Assessment of AF

In the present study, we used either self‐reported history or the standard 12‐lead ECG examination to determine if a participant had AF. Self‐reported history has been shown to be more sensitive for detecting AF than using the standard 12‐lead ECG alone.17 We first asked every participant the following question, “Have you had AF?” If the answer was “yes,” we denoted a self‐reported history of AF. ECGs were recorded in the standard supine resting position and were assessed by experienced physicians. The 10‐s 12‐lead ECG was recorded for all participants using strictly standardized procedures. According to the American Heart Association guidelines,18 AF was diagnosed if one of the following criteria was present: (1) irregular R‐R intervals, (2) absence of repeating P waves, and (3) irregular atrial activity. Two independent experienced cardiologists who were blinded to all clinical data validated the ECGs. When an inconsistency was found, the ECGs were validated by another senior cardiologist for final diagnosis.

2.4. Potential covariates

Information on demographic characteristics (age, sex, marital status, education, income, current smoker, body mass index, heavy alcohol consumption, physical activity) and clinical characteristics (hypertension [HTN], diabetes mellitus [DM], dyslipidemia, depression) was collected via questionnaires and physical examination, as described in previous studies.14, 15

2.5. Statistical analysis

Descriptive analyses were used to summarize the baseline characteristics of participants based on confirmed AF. Continuous variables are presented as mean ± SD and compared using ANOVA analysis. Categorical variables are presented as percentages and compared using the χ² test. The relationship between the existing insomnia symptoms and AF was analyzed by logistic regression analysis based on adjustment for nothing (Model 1); sex and age (Model 2); and sex, age, DM, HTN, dyslipidemia, and depression (Model 3) to calculate the odds ratio (OR) and 95% confidence interval (CI). For subgroup analyses, we estimated the association between insomnia and AF stratified by sex and age, respectively. Participants were divided into 3 subgroups based on age: <40 years, between 40 and 59 years, and ≥60 years. All statistical tests were 2‐sided with the significance level of P < 0.05. The statistical analysis was performed with SAS version 9.4 software (SAS Institute, Inc., Cary, NC).

3. RESULTS

After excluding 707 participants with missing data and vital confounders, the final analysis comprised 8371 participants (4314 men and 4057 women). We identified 50 cases of AF (0.60%) among the 8371 participants. Table 1 provides the comparison of basic characteristics of subjects between AF and non‐AF groups. The AF prevalence in men was significantly higher than in women (0.76% vs 0.42%; P = 0.04). In general, participants with AF were older, had lower education levels and lower income levels, and had higher body mass index. In addition, participants with AF often suffered from HTN, DM, and dyslipidemia, and currently had insomnia. We also found that the prevalence of insomnia was significantly higher in the AF group (26.0%) than in the non‐AF group (12.8%; P = 0.01). The AF group had shorter total sleep time (6.7 ± 1.3 hours) compared with the non‐AF group (7.1 ± 1 .2 hours; P = 0.03). The AIS scores of the AF and non‐AF groups were 3.5 ± 4.8 and 1.9 ± 3.0 (P < 0.05), respectively.

Table 1.

Comparison of baseline clinical characteristics of participants between AF and non‐AF groups

Characteristic	Total, N = 8371	AF Group, n = 50	Non‐AF Group, n = 8321	P Value
Age, y	42.2 ± 13.0	55.4 ± 12.0	42.1 ± 12.9	<0.001
Sex				0.0401
M	4314 (51.5)	33 (66.0)	4281 (51.4)
F	4057 (48.5)	17 (34.0)	4040 (48.6)
Marital status				0.0484
Married	7774 (92.9)	50 (100)	7724 (92.8)
Others	597 (7.1)	0 (0.0)	597 (7.2)
Education				<0.001
Illiteracy/primary school	308 (3.7)	5 (10.0)	303 (3.6)
Middle school	2916 (34.8)	32 (64.0)	2884 (34.7)
College/university	5147 (61.5)	13 (26.0)	5134 (61.7)
Income, ¥/mo				0.0039
≤3000	3140 (38.1)	29 (60.4)	3111 (37.9)
3001–5000	4482 (54.3)	15 (31.3)	4467 (54.5)
>5000	626 (7.6)	4 (8.3)	622 (7.6)
Current smoker	2133 (25.5)	18 (36.0)	2115 (25.4)	0.0869
BMI, kg/m2	24.5 ± 3.7	25.3 ± 2.8	24.5 ± 3.7	0.0384
Heavy alcohol consumption	597 (7.1)	4 (8.0)	593 (7.1)	0.8109
Physical activity				0.0500
Inactive	3222 (38.5)	13 (26.0)	3209 (38.6)
Moderately active	797 (9.5)	9 (18.0)	788 (9.5)
Very active	4352 (52.0)	28 (56.0)	4324 (52.0)
HTN	2608 (31.2)	25 (50.0)	2583 (31.0)	0.0039
DM	549 (6.6)	9 (18.0)	540 (6.5)	0.0010
Dyslipidemia	2969 (35.5)	25 (50.0)	2944 (35.4)	0.0312
Insomnia	1074 (12.8)	13 (26.0)	1061 (12.8)	0.0052
AIS score	1.9 ± 3.0	3.5 ± 4.8	1.9 ± 3.0	0.0385
Total sleep time, hours	7.1 ± 1.2	6.7 ± 1.3	7.1 ± 1.2	0.0271
Depression	540 (6.5)	2 (4.0)	538 (6.5)	0.4792

Abbreviations: AF, atrial fibrillation; AIS, Athens Insomnia Scale; BMI, body mass index; DM, diabetes mellitus; F, female; HTN, hypertension; M, male; SD, standard deviation.

Data are presented as n (%) or mean ± SD.

Table 2 presents age, sex, and several multivariable factors, and adjusted OR for AF. Participants who had insomnia had an increased risk of AF compared with those who did not have insomnia. After adjusting for potential confounders including sex, age, DM, HTN, dyslipidemia, and depression, the effect sizes of associations were slightly attenuated (OR: 1.92, 95% CI: 1.00‐3.70, P = 0.05). In the age subgroup analysis, a positive relationship between insomnia and AF was more remarkable in those age <40 years (OR: 6.52, 95% CI: 1.64‐25.83, P = 0.01), and a similar trend was also observed in those age ≥60 years, although the relationship was not significant (OR: 2.29, 95% CI: 0.95‐5.55, P = 0.07). In addition, a similar trend was found in males after stratifying for sex, although the relationship was not significant (OR: 2.11, 95% CI: 0.92‐4.83, P = 0.08).

Table 2.

ORs With 95% CIs for AF associated with insomnia in subgroups of sex and age

	Total	Sex		Age, y
		M	F	<40	40–59	≥60
Model 1	2.40 (1.27‐4.54)	3.01 (1.35‐6.71)	2.18 (0.77‐6.20)	5.40 (1.00‐29.60)	0.86 (0.25‐2.94)	2.28 (0.95‐5.46)
P value	0.01	0.01	0.15	0.05	0.81	0.06
Model 2	1.91 (1.00‐3.66)	2.24 (1.00‐5.04)	1.54 (0.53‐4.45)	4.60 (0.83‐25.49)	0.82 (0.24‐2.81)	2.33 (0.97‐5.59)
P value	0.05	0.05	0.43	0.08	0.75	0.06
Model 3	1.92 (1.00‐3.70)	2.11 (0.92‐4.83)	1.63 (0.56‐4.74)	6.52 (1.64‐25.83)	0.79 (0.23‐2.78)	2.29 (0.95‐5.55)
P value	0.05	0.08	0.37	0.01	0.72	0.07

Abbreviations: AF, atrial fibrillation; CI, confidence interval; DM, diabetes mellitus; F, female; HTN, hypertension; M, male; OR, odds ratio.

Model 1: Unadjusted.

Model 2: Sex, age.

Model 3: The following potential confounders were adjusted for each OR: sex, age, DM, dyslipidemia, HTN, and depression.

4. DISCUSSION

To the best of our knowledge, our study is the first to assess the association between insomnia and AF. We demonstrated a significant association between insomnia and AF, particularly in those age <40 years. This increased risk was independent of comorbid conditions frequently associated with AF and insomnia, including sex, age, DM, HTN, dyslipidemia, and depression. Among different age groups, participants age <40 years showed the most robust association between insomnia and AF.

In the present study, we used 8 items to evaluate the degree of insomnia and demonstrated that insomnia had a positive association with AF, consistent with previous studies in which a similar conclusion concerning the link between ≥1 items and AF was reached.19, 20 Previously, a cohort study containing 18 775 US male physicians, with a mean follow‐up of 6.9 ± 2.1 years, revealed that a longer sleep duration moderately increased the likelihood of suffering from AF, compared with normal sleep duration.19 A shorter sleep duration also showed a similar trend in people with prevalent sleep apnea.19 These studies demonstrate that there is an association between abnormal total sleep time and AF. Also, a large multiethnic study reported that higher sleep efficiency was significantly associated with a lower likelihood of AF, which was independent of the Apnea‐Hypopnea Index (AHI).20 We also identified an association between insomnia and other diseases including obesity,21 HTN,22 DM,23 dyslipidemia,24 coronary heart disease,25 and myocardial infarction,10 all of which are generally considered important risk factors for AF.

In our study, after we stratified the data by age, insomnia remained a significant association for AF in those age <40 years, and a similar trend was found in those age ≥60 years, although that association was not statistically significant. Surprisingly, the association between insomnia and AF was not found in participants between the ages of 40 and 59 years at baseline. Currently available data do not elucidate potential mechanisms that could link insomnia within different age groups to AF, although a previous study demonstrated an association between nonapnea sleep disorders and type 2 DM among subjects age <40 years26; type 2 DM is considered as an important risk factor for AF.27 Because the incidence of AF increases with age, we should pay special attention to the individuals age <40 years who have symptoms of insomnia.

In addition, it appears that slow‐wave sleep contributed to the association between insomnia and AF. Slow‐wave sleep is a sleep state with the highest parasympathetic activity during the sleep process. Long slow‐wave sleep may play an important role in cardioprotection,28 whereas shorter slow‐wave sleep may potentially promote AF in individuals who suffer from insomnia. However, we found no significant association between insomnia and AF in participants age 40 to 59 years at baseline, and the underlying mechanisms are not clear and need to be elucidated in future studies.

Mechanisms mediating adverse effects of insomnia on AF risk have not well been established. The biological evidence supporting a possible pathway between insomnia and AF was partly deduced from experimental studies concerning controlled sleep deprivation. Recent studies showed that acute SD is associated with increased P‐wave dispersion and QT dispersion and higher values of inter‐intra‐atrial electromechanical delay that are known to be associated with AF development and/or recurrence.29, 30, 31 Acute sleep deprivation in healthy adults was associated with a reduction in left atrial early diastolic strain rate without geometric alterations or functional impairment of the left atrium, suggesting that chronic sleep deprivation may play a more profound role in left atrium function and thereby accelerate the occurrence of AF.32 Mechanistically, insomnia acts as an important risk factor of increased cardiovascular diseases likely through its induction of autonomic dysregulation and inflammation. This could be the important underlying pathophysiological mechanism of AF,33, 34, 35, 36 as revealed in a clinical experiment in which patients with primary insomnia exhibited constant sympathetic hyperactivation, including altered heart rate variability and nighttime levels of catecholamines.37, 38

4.1. Study limitations

The present work has some limitations. First, similar to other relevant studies, we did not assess sleep objectively, for example, by performing a polysomnography.39 Second, even though ASI is a widely used epidemiological tool to assess insomnia, it is not used for clinical diagnosis of insomnia. Third, our study was a cross‐sectional study, which was not able to explore a cause‐effect association between insomnia and AF.

5. CONCLUSION

We demonstrated that insomnia is associated with an increased risk of AF in a Chinese population, particularly in those age <40 years.

Author contributions

Xu Han and Yiheng Yang contributed equally to this work. Xu Han, Xiaomeng Yin, and Yunlong Xia designed this study. Xu Han and Yiheng Yang wrote this article. Jing Qiu and Yue Chen conducted the data analysis. Yong Zhou and Youxin Wang provided the database and reviewed the article. Huihua Li, Lianjun Gao, and Yunlong Xia conducted the quality assurance and reviewed and edited the article.

Conflicts of interest

The authors declare no potential conflicts of interest.

Han X, Yang Y, Chen Y, Gao L, Yin X, Li H, Qiu J, Wang Y, Zhou Y and Xia Y. Association between insomnia and atrial fibrillation in a Chinese population: A cross‐sectional study. Clin Cardiol. 2017;40:764–768. 10.1002/clc.22731