On the Association Between Sleep Quality and Arterial Stiffness: A Population Study in Community-Dwelling Older Adults Living in Rural Ecuador (The Atahualpa Project)

Abstract

Study Objectives:

Evidence of an association between atherosclerosis and sleep quality is limited and has not been studied in remote rural settings, where living conditions are different than in urban centers. We aimed to assess the relationship between the aortic pulse wave velocity (PWV) and sleep quality in older adults living in rural Ecuador.

Methods:

Atahualpa residents aged 60 years or older identified during door-to-door surveys, who consented to participate, underwent face-to-face interviews with the Pittsburgh Sleep Quality Index (PSQI) to assess sleep quality. Aortic PWV determinations were performed for arterial stiffness estimation (as a surrogate of atherosclerosis).

Results:

A total of 303 individuals were included (mean age: 70.3 ± 7.8 years; 59% women). Univariate logistic regression showed a significant association between the aortic PWV and poor sleep quality (odds ratio [OR] 1.22; 95% confidence interval [CI] 1.07–1.39; P = .003). A multivariate logistic regression model, adjusted for demographics, cardiovascular risk factors, oily fish intake and psychological distress showed a significant association between increased PWV and poor sleep quality (OR 1.59; 95% CI 1.12–2.25; P = .009). Similar significance was noted when the model was adjusted for neuroimaging signatures of cerebral small-vessel disease and stroke (OR 1.47; 95% CI 1.07–2.03; P = .019).

Conclusions:

This study shows a significant independent association between the aortic PWV and poor sleep quality in older adults living in rural Ecuador. Results provide more insights into the relevance of the impact of sleep disorders on cardiovascular diseases.

Clinical Trial Registration:

The Atahualpa Project has been registered at ClinicalTrials.gov. The identifier number is NCT01627600, and the date was: 10/02/2012.

Citation:

Del Brutto OH, Mera RM, Peñaherrera E, Costa AF, Peñaherrera R, Castillo PR. On the association between sleep quality and arterial stiffness: a population study in community-dwelling older adults living in rural ecuador (the atahualpa project). J Clin Sleep Med. 2019;15(8):1101–1106.

BRIEF SUMMARY

Current Knowledge/Study Rationale: Nonbreathing sleep-related symptoms have been associated with atherosclerosis, although information is inconclusive and has not been adequately studied in remote rural settings. We aimed to assess this association in community-dwelling older adults living in rural Ecuador.

Study Impact: Multivariate models, fitted with sleep quality as the dependent variable, showed a significant association between the aortic pulse wave velocity (as a surrogate of atherosclerosis) and sleep quality (using the Pittsburgh Sleep Quality Index). This study shows a significant independent association between the aortic pulse wave velocity and poor sleep quality and provides more insights into the relevance of the effect of sleep disorders on cardiovascular diseases.

INTRODUCTION

There is growing evidence favoring a link between sleep-related symptoms and cardiovascular risk factors and diseases.^1–3 Although cardiovascular correlates of sleep disorders have been extensively evaluated in people living in industrialized nations, these associations have been scarcely studied in remote rural settings, where living conditions and cardiovascular risk factors are totally different than in large urban centers.⁴ In addition, most population studies attempting to assess a correlation between sleep-related symptoms and cardiovascular risk factors did so through assessing self-reported sleep duration by a single question, that is, how many hours of sleep do you get on an average weeknight? However, it is plausible that many people—particularly those with low literacy—may not properly answer this subjective question, probably explaining heterogeneity in study results.^5–8

The burden of both cardiovascular diseases and sleep disorders is on the rise in the developing world.^9,10 There is a need for a better understanding if a relationship between these conditions explains their simultaneous increase. Sleep disorders have been associated with atherosclerosis, although information is inconclusive and has not been adequately studied in remote rural settings. Arterial stiffness has proved to be a marker of systemic atherosclerosis.¹¹ Therefore, the study of the association between arterial stiffness and sleep-related symptoms may provide more insights into the relevance of the effect of sleep disorders on cardiovascular diseases. Using the Atahualpa Project cohort, we aimed to assess the association between the aortic pulse wave velocity (PWV)—used as a reliable surrogate of atherosclerosis—and sleep quality, in community-dwelling older adults living in rural Ecuador.

METHODS

Study Population

Atahualpa is a rural village located in Coastal Ecuador. Residents are homogeneous regarding race/ethnicity, overall living conditions, and dietary habits; almost all men work as artisan carpenters and most women are homemakers.⁴ Shift working is limited and nighttime light pollution is scarce, providing a better scenario for studying sleep-related symptoms. Regarding chronobiology, almost all individuals are “larks,” being awake at 7:00 am or even before (as occurs in most rural villages). These consistencies reduce the risk of unexpected confounders at the time of analyses.

Study Design

Atahualpa residents aged 60 years or older identified during door-to-door surveys were invited to participate. These individuals were offered magnetic resonance imaging (MRI) of the brain and aortic PWV determinations, and those who signed a comprehensive informed consent and had no contraindications for the practice of these examinations were included. In addition, participants were interviewed to assess sleep quality.¹² Using a population-based cross-sectional study design, we assessed whether the aortic PWV was associated with sleep quality (as the dependent variable), after adjusting for relevant clinical and neuroimaging confounders (see next paragraphs). The Institutional Review Board of Hospital-Clínica Kennedy, Guayaquil, Ecuador (Federalwide Assurance 00006867) approved the study.

Arterial Stiffness Evaluation

Examinations were performed with individuals resting in the supine position under comfortable room temperature levels. Participants were instructed to avoid caffeine-containing products, nicotine, and alcohol for 24 hours before the test. Arterial stiffness was evaluated by the use of a Mobil-O-Graph NG (IEM, Stolberg, Germany) device. This device estimates the aortic PWV based on the oscillometric detection of the brachial pressure wave with a single cuff. The Mobil-O-Graph NG does not require operator expertise and has demonstrated good repeatability for PWV assessment and a higher repeatability than devices measuring carotid-femoral PWV in elderly populations.¹³

Sleep Quality Assessment

Sleep quality was assessed by the use of the Pittsburgh Sleep Quality Index (PSQI), as detailed elsewhere.¹² The PSQI basically distinguished between “good” and “poor” sleepers. The instrument was administered by trained field personnel by means of face-to-face interviews. The PSQI consists of 19 items grouped into 7 components, each weighted on a scale of 0 to 3, for a total score of 21 points. A PSQI score ≥ 6 points indicates poor sleep quality. Components of the PSQI include assessment of sleep duration, sleep disturbances, sleep latency, daytime dysfunction due to sleepiness, sleep efficiency, overall sleep quality, and medications needed to sleep.

Neuroimaging Protocol

Examinations were performed by the use of a Philips Intera 1.5T MRI scanner (Philips Medical Systems, Eindhoven, the Netherlands). Following predefined protocols, MRI included two-dimensional multislice turbo spin-echo T1-weighted, fluid-attenuated inversion recovery (FLAIR), T2-weighted, and gradient-echo sequences in the axial plane, as well as a T1-weighted sequence oriented in the sagittal plane; slice thickness was 5 mm with a 1-mm gap between slices.¹⁴ All MRI studies were read by two raters blinded to each other’s assessment and clinical information. Kappa coefficients for interrater agreement were high for all the neuroimaging signatures of interest (0.90 for white matter hyperintensities [WMH], 0.76 for deep cerebral microbleeds [CMB], 0.90 for lacunar infarcts [LI], and 0.83 for the presence of > 10 enlarged basal ganglia perivascular spaces [BG PVS]), and discrepancies were resolved by consensus.

MRI studies were reviewed following research standards for cerebral small vessel disease.¹⁵ In particular, WMH of presumed vascular origin were defined as lesions appearing hyperintense on T2-weighted images that remained bright on FLAIR (without cavitation) and graded according to the modified Fazekas scale.¹⁶ CMB were identified and rated according to the microbleed anatomical rating scale¹⁷; for this study, only CMB located deep in the brain were considered. LI were defined as fluid-filled cavities measuring 3 to 15 mm located in the territory of a perforating arteriole.¹⁵ Enlarged BG-PVS were defined as small (< 3 mm) structures of cerebrospinal fluid signal intensity—assessed on the T2-weighted sequence—that followed the orientation of perforating arteries, and rated as abnormal if more than 10 of these lesions were present in a single slice in one side of the brain (we used the BG slice with the highest number on one side).¹⁸ In addition, other subtypes of strokes (ischemic and hemorrhages) were identified and tabulated.

Clinical Covariables Investigated

Demographics, cardiovascular risk factors, and psychological distress were chosen as clinical confounding variables, and were assessed through interviews and procedures previously described in the Atahualpa Project.^19,20 These variables were selected because they have been shown to modify either sleep-related symptoms of arterial stiffness in Atahualpa residents. To assess cardiovascular risk factors, we used the American Heart Association criteria of smoking status, physical activity, diet, body mass index, blood pressure, fasting glucose, and total cholesterol blood levels.²¹ Psychological distress was evaluated by the use of a Spanish version of the Depression-Anxiety-Stress Scale-21, a reliable field instrument (previously validated in Atahualpa) comprising three sets of questions evaluating symptoms of depression, anxiety and stress.²⁰ In addition, the amount of dietary oily fish intake was calculated in servings per week according to interviews and procedures previously detailed for the Atahualpa Project.²²

Statistical Analysis

Data analyses were carried out by using STATA version 15 (College Station, Texas, USA). In univariate analyses, continuous variables were compared by linear models and categorical variables by the χ² or Fisher exact test as appropriate. To assess the independent association between the aortic PWV and sleep quality (as the dependent variable), we fitted logistic regression models, adjusted for demographics, cardiovascular risk factors, oily fish intake, manifestations of psychological distress, and neuroimaging signatures of interest.

RESULTS

Of 437 community-dwelling individuals aged 60 years or older identified during door-to-door surveys and enrolled in the Atahualpa Project up to 2017, 303 (69%) underwent sleep interviews, adequate aortic PWV measurements, and brain MRI. Reasons for not obtaining these studies included refusal to consent (n = 68), severe disability (n = 15), contraindications for the practice of MRI (n = 9), and impossibility to perform aortic PWV measurements due to poor collaboration (n = 29); 13 additional persons had died or emigrated between the time of the survey and the invitation.

Mean age of the 303 participants was 70.3±7.8 years (median age: 69 years) and 178 (59%) were women. Six participants (2%) were current smokers, 14 (5%) had a poor diet, and 22 (7%) had poor physical activity. A body mass index ≥ 30 kg/m² was noted in 69 persons (23%), blood pressure ≥ 140/90 mmHg in 132 (44%), fasting glucose ≥ 126 mg/dL in 91 (30%), and total cholesterol levels ≥ 240 mg/dL in 40 (13%). The mean dietary oily fish intake was 9 ± 5.2 servings per week. Symptoms of depression were present in 32 individuals (11%), anxiety in 42 (14%), and stress in 16 (5%). Moderate to severe WMH were noted in 60 individuals (20%), deep CMB in 22 (7%), LI (overt or silent) in 42 (14%), > 10 enlarged PVS in 74 (24%), and other stroke subtypes (all overt strokes) in 13 (4%). The mean value of the aortic PWV was 10.4 ± 1.8 m/s (95% confidence interval [CI] 10.2–10.6). The median aortic PWV was 9.9 m/s (range 7.7 to 16.2 m/s). A poor sleep quality was noticed in 91 (30%) individuals, and the mean sleep duration was 7.2 ± 1.2 hours (median sleep duration: 7 hours).

Characteristics of participants across categories of sleep quality are summarized in Table 1. As noticed, individuals with a poor sleep quality had more often poor physical activity (P = .002), anxiety (P < .001), and stress (P = .019) than those with a good sleep quality. In addition, mean values of the PWV were significantly higher among patients with a poor sleep quality (P = .002). There were marginal (nonsignificant) association between poor sleep quality and increasing age (P = .051), symptoms of depression (P = .074), and presence of moderate to severe WMH (P = .059).

Table 1.

Characteristics of Atahualpa residents aged 60 years or older across categories of sleep quality (univariate analyses).

Univariate logistic regression showed a significant association between the aortic PWV and poor sleep quality (odds ratio [OR] 1.22; 95% CI 1.07–1.39; P = .003). This association remained significant in multivariate logistic regression models, after adjusting for clinical (OR 1.59; 95% CI 1.12–2.52; P = .01) and neuroimaging covariables (OR 1.47; 95% CI 1.07–2.03; P = .019); in these models, covariables remaining independently significant were poor physical activity, anxiety and stress (Table 2).

Table 2.

Logistic regression models showing a significant association between the aortic pulse wave velocity and poor sleep quality (as the dependent variable), after adjusting for clinical (upper panel) and demographics plus neuroimaging covariables (lower panel).

DISCUSSION

This study shows a significant association between aortic arterial stiffness and poor sleep quality—as assessed by the PSQI—in community-dwelling older adults living in a remote rural setting. This association persisted after adjusting for clinical and neuroimaging covariables, strongly suggesting a link between atherosclerosis and nonbreathing sleep-related symptoms.

As noted, studies assessing a potential relationship between atherosclerosis and sleep-related symptoms relying on a single question about sleep duration, gave contradictory results. Some studies have shown that both short and long sleep duration are associated with atherosclerosis,^7,23,24 others have only found association between short (but not long) sleep duration and atherosclerosis,^8,25,26 and yet others found the contrary (association confined to long sleep duration),^2,27 In addition, other studies have revealed no association between short or long sleep duration and atherosclerosis.^5,28 It is recommendable that population or hospital-based studies stop using this single question to assess the association between sleep-related symptoms and cardiovascular diseases because of the probability of dissimilar findings, which may create confusion at the time of analysis or may induce biases in the results of systematic reviews.¹

A poor sleep quality—assessed by a structured questionnaire, such as the PSQI—has been more coherently associated with atherosclerosis. Although the cause and effect of this relationship is not well understood, prospective studies suggest that the direction goes from poor sleep quality to atherosclerosis.²⁹ Pathogenesis of the association between atherosclerosis and poor sleep quality is complex, and potentially related—but not limited—to modifications in inflammatory markers, hormonal factors, lipid metabolism, dysfunction of the sympathetic nervous system, and endothelial dysfunction.^1,30,31 In addition, psychological distress may also play a role in this association.³² Results from our study are in line with this hypothesis because covariables of psychological distress, particularly anxiety and stress, remained independently significant in multivariate models assessing the association between the aortic PWV and poor sleep quality. Symptoms of depression were not in the path of this association in contrast to what has been found in other studies. However, it must be noticed that our study was conducted in a rural setting where life is more peaceful than in urban settings and the prevalence of depressive symptoms is lower. In general, responsibilities and expectations of persons living in these areas are not as high as in developed urban centers, where pressure for attaining a higher socio-economic status often comes at the price of increased levels of depression.³³

The cross-sectional design (precluding assessment of causation) is a limitation of the present study. It is also possible that Atahualpa residents may not be representative of people living in other villages of the region. However, the major strength of our study is the unbiased inclusion of a homogeneous population of older adults living in a quiet rural village near the Equator, providing an optimal scenario for studying the association between atherosclerosis and sleep quality. Knowledge of the effect of risk factors involved in this association will help to implement more informed public health strategies to reduce catastrophic consequences of atherosclerosis in people living in these underserved populations.

CONCLUSIONS

The current study shows a strong independent association between the aortic PWV and poor sleep quality in older adults living in a remote rural setting. Further longitudinal studies are needed to assess the direction of this relationship.

DISCLOSURE STATEMENT

All authors have read and approved the final draft of the manuscript. This study was supported by Universidad Espíritu Santo – Ecuador. Work for this study was performed at The Atahualpa Project, Universidad Espíritu Santo – Ecuador. The authors report no conflicts of interest.

ABBREVIATIONS

BG

basal ganglia

CI

confidence interval

CMB

cerebral microbleeds

FLAIR

fluid-attenuated inversion recovery

LI

lacunar infarction

MRI

magnetic resonance imaging

OR

odds ratio

PSQI

Pittsburgh Sleep Quality Index

PVS

perivascular spaces

PWV

pulse wave velocity

WMH

white matter hyperintensities