Abstract
We aimed to examine whether hypertension status modified the association between sleep duration and stroke among middle‐aged and elderly population. This cross‐sectional study included 10 516 participants aged ≥45 years from the China Hypertension Survey study. Sleep duration and history of stroke were assessed by questionnaires. Multivariate logistic regression analyses, a generalized additive model (GAM) and smooth curve fitting (penalized spline method) and a two‐piecewise logistic regression models were performed to evaluate the association between sleep duration and stroke in different status of hypertension. 95% confidence interval (CI) for turning point was obtained by bootstrapping. Multiple logistic analyses showed that per 1 hour increase in sleep duration was associated with a 37% increased prevalence of stroke among participants without hypertension and associated with a 8% increased prevalence of stroke among hypertensive participants (without hypertension: odds ratio [OR] = 1.37, 95% CI 1.09‐1.71; with hypertension: OR = 1.08, 95% CI 0.95‐1.21; P Interaction = .029). The fully adjusted smooth curves presented a linear association between sleep duration and stroke among participants without hypertension, but a threshold, nonlinear association among hypertensive participants. The turning point for the curve was found at a sleep duration of 8 (95% CI 5‐9) h among hypertensive patients. The ORs (95% CIs) for stroke were 0.92 (0.79, 1.06) and 1.60 (1.23, 2.08) to the left and right of the turning point, respectively. In conclusion, we found a linear association between sleep duration and stroke among middle‐aged and elderly participants without hypertension, but a threshold, nonlinear association among hypertensive participants.
Keywords: China, hypertension, middle‐aged and elderly population, sleep duration, stroke
1. INTRODUCTION
Sleep is a natural and essential physiological process crucial for homeostasis. However, growing numbers of people suffer from sleep disturbance.1 Growing evidence has indicated short and long sleep durations were associated with a range of adverse health outcomes, including hypertension, diabetes, obesity, cardiovascular disease, and mortality.2, 3, 4, 5, 6 As a major public health concern, stroke is the leading cause of death in China and the second leading cause of death in the world.7 Therefore, the association between sleep duration and stroke has aroused researchers’ attention.
Increasing numbers of prospective studies and meta‐analysis have examined this association between sleep duration and risk of stroke. However, obvious conflicting results could be found among those reports. Some studies showed that both short and long duration of sleep were associated with increased risk of stroke, presenting a U‐shaped relationship.8, 9, 10 Several studies reported that only short sleep duration was positively associated with risk of stroke.11, 12, 13 In contrast, other studies indicated that only long sleep duration was positively associated with risk of stroke, suggesting a J‐shaped relationship.14, 15, 16, 17 These conflicting results might be attributed to the differences in cohort characteristics, race, sample size, and adjustment of confounders. Nevertheless, the optimal range of sleep duration among adults remains unclear. As we know, hypertension is a major risk factor for increased risk of stroke. When combined with known risk factors for stroke, the target sleep duration is still less known. To date, few studies addressed the exact shape of the dose‐response relationship between sleep duration and stroke in different status of hypertension.
Therefore, the aim of this study was to evaluate whether hypertension status modified the association between sleep duration and stroke among middle‐aged and elderly Chinese population form a population‐based cross‐sectional study.
2. METHODS
2.1. Study design and population
The China Hypertension Survey study, encompassed 31 provinces and 262 countries, was a nationally representative cross‐sectional study designed to provide reliable and evidence‐based data on the current status of hypertension and associated factors in Chinese adults. The study design was published previously.18, 19 Briefly, a stratified multistage random sampling method was used to obtain a nationally representative sample of the general Chinese population aged ≥15 years. The first stage of sampling was to select 4 cities in urban areas and four counties in rural areas within each province by the probability proportional to size method. Then by a simple random sampling method, two districts or two townships were selected within each city or county, and three communities or villages were chosen within each district or township, respectively. In the final stage of sampling, a given number of participants from each of the 14 sex/age strata (men and women 15‐24, 25‐34, 35‐44, 45‐54, 55‐64, 65‐74, ≥75 years of age) were selected from communities or villages using lists compiled from local government registers of households. Our present study was conducted between November 2013 and August 2014 in Jiangxi province, China, which was a subset of China Hypertension Survey study. Written informed consent was obtained from each participant and the guardians on behalf of the minors/children aged 15‐18 years enrolled in the study. Ethical approval was obtained from the ethics review boards of the Second Affiliated Hospital of Nanchang University and the Fuwai Cardiovascular Hospital (Beijing, China).
As a result, a total of 15 296 participants out of 15 364 eligible participants completed the investigation. Given the middle‐aged and elderly who might have more sleep problems and higher prevalence of stroke, we excluded participants aged <45 years old (n = 4758) and with missing stroke values (n = 22).
2.2. Hypertension
Blood pressure (BP) was measured three times with 30 seconds interval on the participants’ right arms positioned at heart level using the standardized electronic monitors (HBP‐1300; Omron) after the participants sitting for at least 5 minutes. The accuracy of Omron HBP‐1300 for BP measurement had been verified in prior study.20 Systolic BP (SBP) and diastolic BP (DBP) were calculated as the average of three readings. Hypertension was defined as SBP ≥140 mm Hg and/or DBP ≥90 mm Hg, and use of antihypertensive drugs within 2 weeks.19, 21
2.3. Sleep duration
Sleep duration was based on the response to the following question: “On a weekday, how many hours of sleep do you get in a 24‐hour period?” and “On the weekend, how many hours of sleep do you get in a 24‐hour period?”. Because our previous study showed that sleep debt measured as the difference in sleep duration between weekdays and weekends was not associated with stroke,10 the weekly mean sleep duration was calculated as the weighted average of the sleep duration on weekdays and weekends using the formula22: (sleep duration on weekday × 5 + sleep duration on weekend × 2)/7.
2.4. Stroke
Self‐reported history of stroke was assessed by the following question: “Have you ever been told by a doctor or other health professional that you had a stroke?” if respondents answered “yes,” they were also asked for symptoms, initial dates, diagnostic units, medical records, and imaging data in order to make a reasonable assessment of the original diagnosis. Stroke included subarachnoid hemorrhage, intracerebral hemorrhage, or cerebral ischemic necrosis, but did not include transient cerebral ischemia, secondary stroke caused by brain tumor, brain metastasis tumor, or trauma.10, 23
2.5. Covariates
At this baseline assessment, standardized questionnaire was used to collect information on demographic information (such as age, gender, marital status, education), lifestyle (such as smoking, alcohol consumption), family history of diseases, and oral antihypertensive drugs by trained health professionals. The anthropometric examinations included weight, height, waist circumference (WC), and rest heart rate (RHR). Body mass index (BMI) was calculated as the body weight in kilograms divided by the square of the height in meters (kg/m2).
2.6. Statistical analysis
Sleep duration was categorized into short (<6 hours), average (6‐8 hours), and long (>8 hours) in accordance with the previous study.10, 16, 24 Data are presented as mean ± SD for continuous variables and as frequency (%) for categorical variables. Baseline characteristics of study population were described by hypertension status and sleep duration categories. Comparisons among different sleep duration groups were performed using one‐way ANOVA test (continuous variables) or chi‐square test (categorical variables), accordingly. Sleep duration was analyzed as a categorical variable with average sleep being the reference group and also as a continuous variable. Multivariate logistic regression analysis was performed to obtain the odds ratio (OR) and 95% confidence interval (CI). We performed testing for linear trends by entering the median value of each category of sleep duration as a continuous variable in the models. To further characterize the shape of the association between sleep duration and stroke in different status of hypertension, a generalized additive model (GAM) and smooth curve fitting (penalized spline method) were conducted. If nonlinearity was detected, we first calculated the turning point using recursive algorithm, and then constructed a two‐piecewise binary logistic regression model on both sides of the turning point. The threshold level was determined by choosing the turning point which provided the maximum model likelihood, using the trial and error method, along with a log‐likelihood ratio test comparing the one‐line logistic regression model with a two‐piecewise logistic model to examine the statistical significance.25 95% CI for turning point was obtained by bootstrapping. Interaction and stratified analyses were also performed to evaluate whether covariates influenced the association between sleep duration and stroke in different status of hypertension.
We considered the following covariates in multivariable models: sex (male/female), age (continuous), area (urban/rural), education status (0‐6, 7‐9, ≥10 years), occupation (employed, retired, unemployed), family history of stroke (yes/no), current smoking (yes/no), current drinking (yes/no), antihypertensive drugs (yes/no), BMI (continuous), WC (continuous), SBP (continuous), DBP (continuous), and RHR (continuous).
To ensure the robustness of data analysis, we did the sensitivity analysis. First, we performed the associations between sleep duration on the weekday and weekend and stroke in different status of hypertension. Second, 6.0% of covariates information in our study was missing. Dummy variables were used to indicate missing covariate values. We imputed the median values for continuous variables (ie, BMI and RHR) and used a missing indicator approach for categorical variables (ie, smoking, drinking, education status, and occupation). The main results were also performed using the imputed datasets.
All the analyses were performed using the statistical package R (http://www.R-project.org, The R Foundation) and Empower (R) (www.empowerstats.com; X&Y Solutions, Inc). A two‐side P value <.05 was considered to be statistically significant.
3. RESULTS
3.1. Baseline characteristics of study participants
Based on the inclusion and exclusion criteria, a total of 10 516 participants aged 45‐97 years (mean age: 62.8 ± 11.1 years; 40.8% men) were selected for final data analysis (Figure 1). Overall, the prevalence of stroke was 2.1% (224/10516). The average sleep duration per night was 7.3 ± 1.3 hours. The baseline characteristics of study participants by hypertension status and sleep duration categories were presented in Table 1. Regardless of hypertension status, compared with short and average sleep duration, participants with long sleep duration were more likely to be from rural, to have a lower mean age, to have lower educational level, and to be employed (all P < .05). No significant differences were found between the three groups in terms of BMI, WC, SBP, RHR, smoking status, use of antihypertensive drugs, family history of stroke, regardless of hypertension status.
Figure 1.
Flowchart of study participants
Table 1.
Baseline characteristics of study participants by hypertension status and sleep duration categories
| Variablesa | Without hypertension | With hypertension | ||||||
|---|---|---|---|---|---|---|---|---|
| Sleep duration (h) categories | P valueb | Sleep duration (h) categories | P valuec | |||||
| <6 | 6‐8 | >8 | <6 | 6‐8 | >8 | |||
| N | 691 | 4446 | 1251 | 490 | 2914 | 724 | ||
| Male, n (%) | 302 (43.7) | 1779 (40.0) | 496 (39.7) | .158 | 179 (36.5) | 1236 (42.4) | 295 (40.8) | .046 |
| Age, y | 63.1 ± 11.2 | 59.9 ± 10.6 | 60.1 ± 11.1 | <.001 | 68.3 ± 9.9 | 66.4 ± 10.6 | 66.2 ± 10.8 | .001 |
| Urban, n (%) | 344 (49.8) | 2257 (50.8) | 465 (37.2) | <.001 | 243 (49.6) | 1867 (64.1) | 360 (49.7) | <.001 |
| BMI, kg/m2 | 22.4 ± 3.4 | 22.7 ± 3.4 | 22.7 ± 3.5 | .107 | 23.9 ± 4.3 | 23.8 ± 3.9 | 23.6 ± 3.7 | .334 |
| WC, cm | 78.6 ± 8.7 | 79.0 ± 8.7 | 79.1 ± 9.1 | .518 | 82.6 ± 9.4 | 82.2 ± 10.0 | 81.7 ± 9.9 | .214 |
| SBP, mm Hg | 119.4 ± 11.7 | 119.2 ± 11.3 | 119.6 ± 10.8 | .407 | 146.2 ± 19.1 | 146.7 ± 18.9 | 147.4 ± 19.0 | .552 |
| DBP, mm Hg | 70.8 ± 8.5 | 71.5 ± 8.2 | 71.6 ± 8.2 | .074 | 78.7 ± 12.2 | 81.5 ± 11.7 | 81.3 ± 12.1 | <.001 |
| RHR, bpm | 76.6 ± 11.4 | 76.9 ± 10.7 | 77.2 ± 10.6 | .543 | 78.9 ± 12.3 | 78.3 ± 11.7 | 78.6 ± 12.0 | .538 |
| Sleep duration, h | 4.9 ± 0.7 | 7.3 ± 0.7 | 9.0 ± 0.7 | <.001 | 4.9 ± 0.7 | 7.2 ± 0.7 | 9.0 ± 0.8 | <.001 |
| Current smoking, n (%) | 142 (20.6) | 907 (20.4) | 241 (19.3) | .615 | 77 (15.7) | 555 (19.1) | 124 (17.1) | .119 |
| Current drinking, n (%) | 193 (27.9) | 1080 (24.3) | 361 (28.9) | .003 | 100 (20.4) | 665 (22.8) | 179 (24.7) | .220 |
| Education status, y, n (%)d | ||||||||
| 0‐6 | 418 (60.5) | 2583 (58.1) | 806 (64.4) | .002 | 322 (65.7) | 1906 (65.4) | 505 (69.8) | .015 |
| 7‐9 | 258 (37.3) | 1689 (38.0) | 406 (32.5) | 158 (32.2) | 877 (30.1) | 193 (26.7) | ||
| ≥10 | 11 (1.6) | 108 (2.4) | 27 (2.2) | 4 (0.8) | 78 (2.7) | 13 (1.8) | ||
| Occupation, n (%)d | ||||||||
| Employed | 211 (30.5) | 1426 (32.1) | 456 (36.5) | <.001 | 107 (21.8) | 579 (19.9) | 170 (23.5) | <.001 |
| Retired | 155 (22.4) | 702 (15.8) | 150 (12.0) | 146 (29.8) | 650 (22.3) | 138 (19.1) | ||
| Unemployed | 323 (46.7) | 2265 (50.9) | 641 (51.2) | 235 (48.0) | 1642 (56.4) | 408 (56.4) | ||
| Antihypertensive drugs, n (%)e | —e | —e | —e | 139 (28.4) | 679 (23.3) | 163 (22.5) | .035 | |
| Family history of stroke | 33 (4.8) | 184 (4.1) | 48 (3.8) | .766 | 26 (5.3) | 162 (5.6) | 32 (4.4) | .104 |
| Stroke, n (%) | 2 (0.3) | 28 (0.6) | 17 (1.4) | .010 | 27 (5.5) | 112 (3.8) | 38 (5.3) | .090 |
Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; RHR, rest heart rate; SBP, systolic blood pressure; WC, waist circumference.
Data are presented as number (%) or mean ± standard deviation.
Comparisons among different sleep duration categories groups in participants without hypertension.
Comparisons among different sleep duration categories groups in participants with hypertension.
Numbers that do not add up to 100% were attributable to missing data.
The model failed because of the none sample size.
3.2. Association between sleep duration and stroke in different status of hypertension
Table 2 showed the effect modification of hypertension on the association between sleep duration and stroke. In fully adjusted model (model II), each 1 hour increase in sleep duration was associated with a 37% higher prevalence of stroke among participants without hypertension (OR = 1.37, 95% CI 1.09‐1.71; P = .007). In contrast, each 1 hour increase in sleep duration was associated with a 8% higher prevalence of stroke among hypertensive subjects (OR = 1.08, 95% CI 0.95‐1.21; P = .232). There was a significant interaction between hypertension status and sleep duration on prevalence of stroke (P Interaction = .029). We also converted sleep duration from a continuous variable to a categorical variable. Among participants without hypertension, compared to participants in 6‐8 hours of sleep duration, sleep duration <6 hours was associated with lower prevalence of stroke (OR = 0.36, 95% CI 0.09‐1.54), but there did not reach statistical significance. Moreover, sleep duration >8 hours was significantly associated with a 1.21 times higher prevalence of stroke (OR = 2.21, 95% CI 1.19‐4.14). P for trend in the all models was significant and consistent with the P value when sleep duration was used as a continuous variable, suggesting the linear association between sleep duration and stroke. Among participants with hypertension, compared to average sleep duration, sleep duration <6 hours was not significantly associated with increased prevalence of stroke, but sleep duration >8 hours was significantly associated with increased prevalence of stroke (OR<6h = 1.21, 95% CI 0.73‐2.01; OR>8h = 1.59, 95% CI 1.07‐2.38; P trend = 0.309). These results suggested that the association between sleep duration and stroke among hypertensive subjects was likely to be nonlinear.
Table 2.
Effect modification of hypertension on the association between sleep duration and stroke
| Sleep duration, h | Events, n (%) | Crude model | Model I | Model II | |||
|---|---|---|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | OR (95% CI) | P value | ||
| Without hypertension | |||||||
| Per 1 h increase | 47 (0.7) | 1.34 (1.06, 1.69) | .014 | 1.36 (1.08, 1.70) | .009 | 1.37 (1.09, 1.71) | .007 |
| Categories | |||||||
| <6 | 2 (0.3) | 0.46 (0.11, 1.93) | .287 | 0.39 (0.09, 1.64) | .198 | 0.36 (0.09, 1.54) | .169 |
| 6‐8 | 28 (0.6) | Ref. | Ref. | Ref. | |||
| >8 | 17 (1.4) | 2.17 (1.19, 3.98) | .012 | 2.14 (1.17, 3.93) | .014 | 2.21 (1.19, 4.14) | .013 |
| P for trend | .005 | .003 | .002 | ||||
| With hypertension | |||||||
| Per 1 h increase | 177 (4.3) | 1.00 (0.89, 1.12) | .972 | 1.00 (0.90, 1.13) | .933 | 1.08 (0.95, 1.21) | .232 |
| Categories | |||||||
| <6 | 27 (5.5) | 1.46 (0.95, 2.25) | .086 | 1.45 (0.94, 2.24) | .090 | 1.21 (0.73, 2.01) | .460 |
| 6‐8 | 112 (3.8) | Ref. | Ref. | Ref. | |||
| >8 | 38 (5.3) | 1.39 (0.95, 2.02) | .090 | 1.40 (0.96, 2.05) | .080 | 1.59 (1.07, 2.38) | .022 |
| P for trend | .898 | .941 | .309 | ||||
| P value for interaction* | .027 | .024 | .029 | ||||
Model I: regression was done with adjustment for sex, age. Model II: regression was done with adjustment for sex, age, area, smoking, drinking, education status, occupation, antihypertensive drugs, family history of stroke, BMI, WC, SBP, DBP, and RHR.
P value for interaction test: 2‐way interaction of hypertension status and sleep duration (continuous) on stroke.
In addition, we reported the association between sleep duration on weekday or weekend and stroke in different status of hypertension. Table S1 showed that per 1 hour increase in sleep duration on weekday was associated with a 34% increased prevalence of stroke among participants without hypertension and associated with a 7% increased prevalence of stroke among hypertensive participants [without hypertension: odds ratio (OR) = 1.34, 95% CI 1.07‐1.68; with hypertension: OR = 1.07, 95% CI 0.95‐1.20; P Interaction = .033]. Table S2 also showed that hypertension status could modified the association between sleep duration on weekend and stroke (P Interaction = .038). Sensitivity analysis showed that the similar trend about effect modification of hypertension on the association between sleep duration and stroke using post‐imputation data (Table S3).
The fully adjusted smooth curve fitting presented a linear association between sleep duration and stroke among participants without hypertension, but a threshold, nonlinear association among hypertensive participants (Figure 2). Whether it was a weekday or a weekend, we also observed that there was a linear association between sleep duration and stroke among participants without hypertension, but a threshold, nonlinear association among hypertensive participants (Figure S1). Therefore, we further performed the threshold effect analysis of sleep duration on stroke among hypertensive participants to calculated the turning point using the standard binary logistic regression model and the two‐piecewise binary logistic regression model, respectively (Table 3, Tables S4 and S5). As shown in Table 3, the P for log‐likelihood ratio test was less than 0.05, indicating that the two‐piecewise binary logistic regression was more suitable for fitting the association between sleep duration and stroke. The turning point that we identified for sleep duration among hypertensive patients was 8 (95% CI 5‐9) hour. The ORs (95% CIs) for stroke were 0.92 (0.79, 1.06) and 1.60 (1.23, 2.08) to the left and right of the turning point, respectively. Threshold effect analysis of sleep duration on stroke in a sensitivity analysis showed the same turning point of sleep duration (Tables S4 and S5).
Figure 2.
Association between sleep duration and stroke stratified by hypertension status. The smooth curve fitting presented a linear association between sleep duration and stroke among participants without hypertension, but a threshold, nonlinear association among hypertensive participants. Adjustment factors included sex, age, area, smoking, drinking, education status, occupation, antihypertensive drugs, family history of stroke, BMI, WC, SBP, DBP, and RHR
Table 3.
Threshold effect analysis of sleep duration on stroke among participants with hypertension using piecewise binary logistic regression models
| Sleep duration, h | Crude model | Adjusted modela | ||
|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | |
| Standard logistic regression model | 1.00 (0.89, 1.12) | .972 | 1.07 (0.95, 1.20) | .293 |
| Two‐piecewise logistic regression models | ||||
| Turning point | 8 (5‐9)b | |||
| <8 | 0.88 (0.76, 1.02) | .088 | 0.92 (0.79, 1.06) | .250 |
| >8 | 1.54 (1.17, 2.02) | .002 | 1.60 (1.23, 2.08) | <.001 |
| P for log likelihood ratio test | .003 | .002 | ||
Adjusted for sex, age, area, smoking, drinking, education status, occupation, antihypertensive drugs, family history of stroke, BMI, WC, SBP, DBP, and RHR.
95% CI for turning point was obtained by bootstrapping.
3.3. Subgroup analysis
Given the above results showing the significant association between sleep duration more than 8 hours and stroke among hypertensive subjects, we only analyzed hypertensive subjects with sleep duration ≥8 hours. We performed stratified analyses by subgroups to explore the role of other covariables on the association between sleep duration and stroke. As shown in Figure 3, the association between elevated sleep duration and stroke was consistent in the following subgroups: sex, age, area, BMI, smoking, and drinking (all P for interaction >.05).
Figure 3.
Effect size of sleep duration on stroke in each subgroup stratified by hypertension status. Adjusted for sex, age, area, smoking, drinking, education status, occupation, antihypertensive drugs, family history of stroke, BMI, WC, SBP, DBP, and RHR, if not be stratified
4. DISCUSSION
In the present study, we found that hypertension status modified the association between sleep duration and stroke among middle‐aged and elderly population. There was a linear association between sleep duration and stroke among participants without hypertension, but a threshold, nonlinear association among hypertensive participants. Sleep duration >8 hours was positively associated with the prevalence of stroke in hypertensive patients.
Many previous studies have reported the association between sleep duration and risk of stroke with inconsistent results. Cappuccio et al9 performed a systematic search including 474 684 participants and found that both short and long sleep durations were positively associated with risk of stroke. Leng et al16 conducted a prospective cohort study of 9692 subjects (4842 participants with SBP ≥137 mm Hg) and found that long sleep was significantly associated with an increased risk of stroke. While another meta‐analysis indicated that only short sleep duration was associated with increased risk of stroke.26 These conflicting results might be attributed to population with different comorbidities which were risk factor of stroke. Several studies also reported the association between sleep duration and stroke in hypertensive patients.11, 27 Eguchi et al11 conducted a cohort study of 932 subjects (mean age: 70.4 years; 93.5% hypertensives) in Japan and followed them for 50 months. They found that a sleep duration <7.5 hours was independently and positively associated with the risk of stroke. Chen et al27 conducted a prospective study on 93 175 older women (aged 50‐79 years; 38.6% hypertensives) in the Women's Health Initiative Observational study cohort and indicated that among women with hypertension, only sleep duration ≥8 hours per night was positively associated with risk of ischemic stroke. Pan et al8 used data from 63 257 Chinese adults aged 45‐74 years and reported that the significant associations with short and long sleep durations were only observed among hypertensive subjects, but not among those without hypertension. In our study, we also found that hypertension status could modify the association between sleep duration and stroke and we further observed a nonlinear association between sleep duration and stroke in hypertensive patients. Sleep duration >8 hours was positively associated with the prevalence of stroke in hypertensive patients. However, the presence of obstructive sleep apnea syndrome in the different groups of the study was not assessed although there is a huge amount of evidence supporting the strong relationship between sleep breathing disorders, BP profile and cardiovascular risk.28, 29, 30 Furthermore, the cross‐sectional evaluations did not allow to assess additional hemodynamic variables, as for example the variability of BP, to improve the characterization of the hypertensive status of patients. Indeed, BP variability has been demonstrated to be a reproducible index31 and meaningful predictor of cerebrovascular disease.32, 33, 34 Therefore, further research is needed to examine the relationship between obstructive sleep apnea syndrome, BP variability and cerebrovascular disease.
Our study indicated that hypertension status could modified the association between sleep duration and stroke, suggesting that pathways on the association between sleep duration and stroke varied by different hypertension status. The mechanism driving this association is still unclear. However, several possible reasons could account for the modification of hypertension status on the association between sleep duration and stroke. Previous studies have showed long sleep duration was related to arterial stiffening,35 which are known to be induced by high BP as well. Moreover, long sleep duration was associated with increased risks of hypertension.36 Therefore, on one hand, hypertension and sleep duration may have a synergistic effect on stroke risk through the common pathways of inducing arterial stiffening and atherosclerosis. On the other hand, long sleep duration could lead to increased BP and prevalence of hypertension, which, in turn, might lead to prevalence of stroke. In addition, it has been hypothesized that long sleep duration might represent an epiphenomenon of comorbidity, sleep‐disordered breathing, depression, and unmeasured sociobehavioral attributes, environmental factors, or biophysical constructs that are proximal causes of stroke.9, 27 This finding suggests that optimal sleep duration is particularly important for hypertensive patients. Further longitudinal studies are needed to assess the association between sleep duration and stroke in hypertensive patients and elucidate their mechanism. In our study, we also found a linear association between sleep duration and stroke which was independent of hypertension, suggesting that other physiological mechanisms may contribute to the increased risk of stroke death. Individuals with insufficient sleep may also be more likely to have sleep disorders or mental distress,37 which may mediate the association with stroke risk. Short sleep duration could lead to increased risk of stroke through several biological pathways by activating pro‐inflammatory pathways. Of course, the specific biological mechanism underlying this association is needed to be fully elucidated. Further researches should mainly focus on the association between sleep duration and risk of stroke in healthy population.
According to STROBE statement, subgroup analysis can make better use of data to reveal underlying truths. Subgroup analyses showed that regardless of hypertension status, the association between sleep duration and stroke was consistent in different subgroups, which yield stable conclusion.
This study has several strengths of note. First, this study was the first report to explore whether hypertension status modified the association between sleep duration and stroke among a middle‐aged and elderly Chinese population. Second, we addressed the nonlinearity between sleep duration and stroke in different status of hypertension and further explained this nonlinearity. Third, we performed sensitivity analyses to enhance the robustness of results.
Several limitations are also noteworthy. First, as a cross‐sectional design, it was less power to infer the causal the association of sleep duration and stroke in different status of hypertension. Second, hypertension was diagnosed during office BP measurement, but “white coat hypertension” and “masked hypertension and masked uncontrolled hypertension” were not assessed. Third, sleep duration was measured by questionnaires, which may not fully capture the actual amounts of sleep per night. Nevertheless, assessments of self‐reported sleep duration have been shown to correlate well with values obtained by autographic monitoring.38 Moreover, stroke was measured by self‐reported questionnaires. If population with history of stroke might have been included in the reference group, this would have biased results toward the null. Fourth, in our questionnaire, we did not collect specific stroke subtypes so that we could not perform sensitivity analyses to evaluate whether results were consistent across stroke subtypes. Fifth, we did not collect information about social factors or demands (such as childcare, work schedules, bed partners, sleeping environment) that may impact the sleep duration.
5. CONCLUSION
We found a linear association between sleep duration and stroke among participants without hypertension, but a threshold, nonlinear association among hypertensive participants. The findings suggested that hypertension status could modify the association between sleep duration and stroke among middle‐aged and elderly population.
CONFLICT OF INTEREST
None.
AUTHOR CONTRIBUTIONS
HHB XSC contributed to conceptualization and methodology, reviewed the manuscript, and edited the manuscript; LHH developed software, wrote the original draft manuscript, and performed formal analysis; LHH XH WZ CJY JXL PL YQW QHW ZWW RLG HHB XSC involved in validation; LHH XH WZ CJY participated in investigation; HHB XSC involved in data curation; HHB XSC supervised the manuscript; JXL PL YQW QHW ZWW RLG HHB XSC performed project administration; ZWW RLG XSC involved in funding acquisition.
Supporting information
ACKNOWLEDGMENTS
We acknowledge the contribution the all staff who participated in this study as well as the study participants who shared their time with us.
Hu L, Huang X, Zhou W, et al. Effect of hypertension status on the association between sleep duration and stroke among middle‐aged and elderly population. J Clin Hypertens. 2020;22:65–73. 10.1111/jch.13756
Funding information
This research was supported by the National Key R&D Program in the Twelfth Five‐year Plan (Nos. 2011BAI11B01, and 2014ZX09303305) from the Chinese Ministry of Science and Technology and National Natural Science Foundation of China (No. 81560051).
Contributor Information
Huihui Bao, Email: huihui_bao77@126.com.
Xiaoshu Cheng, Email: xiaoshumenfan126@163.com.
REFERENCES
- 1. Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, Hla KM. Increased prevalence of sleep‐disordered breathing in adults. Am J Epidemiol. 2013;177(9):1006‐1014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Buxton OM, Marcelli E. Short and long sleep are positively associated with obesity, diabetes, hypertension, and cardiovascular disease among adults in the United States. Soc Sci Med. 2010;71(5):1027‐1036. [DOI] [PubMed] [Google Scholar]
- 3. Kita T, Yoshioka E, Satoh H, et al. Short sleep duration and poor sleep quality increase the risk of diabetes in Japanese workers with no family history of diabetes. Diabetes Care. 2012;35(2):313‐318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Cizza G, Skarulis M, Mignot E. A link between short sleep and obesity: building the evidence for causation. Sleep. 2005;28(10):1217‐1220. [DOI] [PubMed] [Google Scholar]
- 5. Qureshi AI, Giles WH, Croft JB, Bliwise DL. Habitual sleep patterns and risk for stroke and coronary heart disease: a 10‐year follow‐up from NHANES I. Neurology. 1997;48(4):904‐911. [DOI] [PubMed] [Google Scholar]
- 6. Cappuccio FP, D'Elia L, Strazzullo P, Miller MA. Sleep duration and all‐cause mortality: a systematic review and meta‐analysis of prospective studies. Sleep. 2010;33(5):585‐592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380(9859):2095‐2128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Pan A, De Silva DA, Yuan JM, Koh WP. Sleep duration and risk of stroke mortality among Chinese adults: Singapore Chinese health study. Stroke. 2014;45(6):1620‐1625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Cappuccio FP, Cooper D, D'Elia L, Strazzullo P, Miller MA. Sleep duration predicts cardiovascular outcomes: a systematic review and meta‐analysis of prospective studies. Eur Heart J. 2011;32(12):1484‐1492. [DOI] [PubMed] [Google Scholar]
- 10. Hu L, Zhang B, Zhou W, et al. Sleep duration on workdays or nonworkdays and cardiac‐cerebral vascular diseases in Southern China. Sleep Med. 2018;47:36‐43. [DOI] [PubMed] [Google Scholar]
- 11. Eguchi K, Hoshide S, Ishikawa S, Shimada K, Kario K. Short sleep duration is an independent predictor of stroke events in elderly hypertensive patients. J Am Soc Hypertens. 2010;4(5):255‐262. [DOI] [PubMed] [Google Scholar]
- 12. Petrov ME, Howard G, Grandner MA, Kleindorfer D, Molano JR, Howard VJ. Sleep duration and risk of incident stroke by age, sex, and race: the REGARDS study. Neurology. 2018;91(18):e1702‐e1709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Eguchi K, Pickering TG, Schwartz JE, et al. Short sleep duration as an independent predictor of cardiovascular events in Japanese patients with hypertension. Arch Intern Med. 2008;168(20):2225‐2231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Li W, Wang D, Cao S, et al. Sleep duration and risk of stroke events and stroke mortality: a systematic review and meta‐analysis of prospective cohort studies. Int J Cardiol. 2016;223:870‐876. [DOI] [PubMed] [Google Scholar]
- 15. Jike M, Itani O, Watanabe N, Buysse DJ, Kaneita Y. Long sleep duration and health outcomes: a systematic review, meta‐analysis and meta‐regression. Sleep Med Rev. 2018;39:25‐36. [DOI] [PubMed] [Google Scholar]
- 16. Leng Y, Cappuccio FP, Wainwright NWJ, et al. Sleep duration and risk of fatal and nonfatal stroke: a prospective study and meta‐analysis. Neurology. 2015;84(11):1072‐1079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wen YE, Pi F‐H, Guo PI, et al. Sleep duration, daytime napping, markers of obstructive sleep apnea and stroke in a population of southern China. Sci Rep. 2016;6:34689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Wang Z, Zhang L, Chen Z, et al. Survey on prevalence of hypertension in China: background, aim, method and design. Int J Cardiol. 2014;174(3):721‐723. [DOI] [PubMed] [Google Scholar]
- 19. Wang Z, Chen Z, Zhang L, et al. Status of hypertension in China: results from the China Hypertension Survey, 2012–2015. Circulation. 2018;137(22):2344‐2356. [DOI] [PubMed] [Google Scholar]
- 20. Chen Z, Wang X, Wang Z, et al. Assessing the validity of oscillometric device for blood pressure measurement in a large population‐based epidemiologic study. J Am Soc Hypertens. 2017;11(11):730‐736. [DOI] [PubMed] [Google Scholar]
- 21. Hu L, Huang X, You C, et al. Prevalence and risk factors of prehypertension and hypertension in Southern China. PLoS ONE. 2017;12(1):e170238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Kong AP, Wing Y‐K, Choi KC, et al. Associations of sleep duration with obesity and serum lipid profile in children and adolescents. Sleep Med. 2011;12(7):659‐665. [DOI] [PubMed] [Google Scholar]
- 23. Hu L, Huang X, Zhou W, et al. Associations between resting heart rate, hypertension, and stroke: a population‐based cross‐sectional study. J Clin Hypertens (Greenwich). 2019;21(5):589‐597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Lou P, Chen P, Zhang L, et al. Interaction of sleep quality and sleep duration on impaired fasting glucose: a population‐based cross‐sectional survey in China. BMJ Open. 2014;4(3):e4436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Yu X, Cao L, Yu X. Elevated cord serum manganese level is associated with a neonatal high ponderal index. Environ Res. 2013;121:79‐83. [DOI] [PubMed] [Google Scholar]
- 26. Itani O, Jike M, Watanabe N, Kaneita Y. Short sleep duration and health outcomes: a systematic review, meta‐analysis, and meta‐regression. Sleep Med. 2017;32:246‐256. [DOI] [PubMed] [Google Scholar]
- 27. Chen J‐C, Brunner RL, Ren H, et al. Sleep duration and risk of ischemic stroke in postmenopausal women. Stroke. 2008;39(12):3185‐3192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Lattanzi S, Brigo F, Silvestrini M. Obstructive sleep apnea syndrome and the nocturnal blood pressure profile. J Clin Hypertens (Greenwich). 2018;20(6):1036‐1038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Lattanzi S, Brigo F, Silvestrini M. Sleep and blood pressure. J Clin Hypertens (Greenwich). 2018;20(12):1721‐1723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Lattanzi S, Brigo F, Silvestrini M. Blood pressure profile and nocturnal oxygen desaturation. J Clin Hypertens (Greenwich). 2018;20(4):656‐658. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Lattanzi S, Brigo F, Vernieri F, Silvestrini M. Visit‐to‐visit variability in blood pressure and Alzheimer's disease. J Clin Hypertens (Greenwich). 2018;20(5):918‐924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Mehlum MH, Liestøl K, Kjeldsen SE, et al. Blood pressure variability and risk of cardiovascular events and death in patients with hypertension and different baseline risks. Eur Heart J. 2018;39(24):2243‐2251. [DOI] [PubMed] [Google Scholar]
- 33. Kang J, Hong J‐H, Jang MU, et al. Change in blood pressure variability in patients with acute ischemic stroke and its effect on early neurologic outcome. PLoS ONE. 2017;12(12):e189216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Divani AA, Liu XI, Di Napoli M, et al. Blood pressure variability predicts poor in‐hospital outcome in spontaneous intracerebral hemorrhage. Stroke. 2019;50(8):2023‐2029. [DOI] [PubMed] [Google Scholar]
- 35. Yoshioka E, Saijo Y, Kita T, et al. Relation between self‐reported sleep duration and arterial stiffness: a cross‐sectional study of middle‐aged Japanese civil servants. Sleep. 2011;34(12):1681‐1686. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Hu L, Huang X, You C, et al. Relationship of sleep duration on workdays and non‐workdays with blood pressure components in Chinese hypertensive patients. Clin Exp Hypertens. 2019;41(7):627‐636. [DOI] [PubMed] [Google Scholar]
- 37. Liu Y, Croft JB, Wheaton AG, et al. Association between perceived insufficient sleep, frequent mental distress, obesity and chronic diseases among US adults, 2009 behavioral risk factor surveillance system. BMC Public Health. 2013;13:84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Lockley SW, Skene DJ, Arendt J. Comparison between subjective and actigraphic measurement of sleep and sleep rhythms. J Sleep Res. 1999;8(3):175‐183. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials