1290 Journal of Atherosclerosis and Thrombosis　Vol.21, No.12

Original Article

Introduction

Nocturnal intermittent hypoxia (NIH) is a sur-rogate marker of Obstructive Sleep Apnea (OSA), underdiagnosed condition characterized by recurrent

episodes of obstruction of the upper airway leading to sleep fragmentation and intermittent hypoxia during sleep1). OSA prevalence in adults has been estimated to be in the range of 3% to 7%2). It has been reported the association of OSA with several cardiovascular dis-eases (CVD), including coronary artery disease3), stroke4), atrial fibrillation5) and heart failure6). Because of the high prevalence of OSA and the growing evi-dence regarding the cardiovascular consequences, screening for OSA in the general population may be an important strategy for the prevention of CVD7).

Relationships between Nocturnal Intermittent Hypoxia, Arterial Stiffness and Cardiovascular Risk Factors in a Community-based Population: The Toon Health Study

Eduardo Campos Alberto1, Takeshi Tanigawa1, Koutatsu Maruyama2, Yuri Kawasaki1, Eri Eguchi1, Hiromi Mori1, Kana Yoshimura1, Sakurako Tanno1, Susumu Sakurai3, Shinichi Hitsumoto4 and Isao Saito5

1Department of Public Health, Ehime University Graduate School of Medicine, Ehime, Japan2Department of Basic Medical Research and Education, Ehime University Graduate School of Medicine, Ehime, Japan3Department of Clinical Laboratory Science, Faculty of Health Care, Tenri Health Care University, Nara, Japan4Total Medical Support Center, Ehime University Hospital, Ehime, Japan5Department of Basic Nursing and Health Science, Ehime University Graduate School of Medicine, Ehime, Japan

Aim: Nocturnal intermittent hypoxia (NIH), a primary marker of obstructive sleep apnea, has increas-ingly been linked with cardiovascular morbidity and mortality. The purpose of this study was to inves-tigate the association between NIH and arterial stiffness as measured according to the cardio-ankle vas-cular index (CAVI) based on cardiovascular risk factors in a Japanese community-dwelling population.Methods: We conducted a cross-sectional study in Toon city among 684 men and 1,241 women 30-79 years of age. The severity of NIH was defined as mild or moderate-to-severe according to five or 15 events/hour on the 3% oxygen desaturation index (ODI), respectively. Increased arterial stiff-ness was diagnosed according to a CAVI of ≥9.Results: The number of subjects with no, mild and moderate-to-severe NIH was 1,348 (70%), 451 (23%) and 126 (7%), respectively. Increased arterial stiffness was detected in 21.9% of the partici-pants. The multivariable-adjusted odds ratio (95% CI) of severe NIH related to an increased CAVI in comparison with a 3% ODI of ＜5 was 1.36 (0.82-2.23). The stratified logistic regression analysis showed that the multivariable-adjusted OR of severe NIH for an increased CAVI was remarkably increased in the individuals with a BMI of ≥25 (OR=2.53, 1.08-5.96; p=0.03). An interaction test showed a trend for an overweight status to be a modifier of the association between OSA and increased arterial stiffness (p=0.05).Conclusions: NIH has a tendency to promote increased arterial stiffness as measured according to the CAVI, especially in overweight subjects.

J Atheroscler Thromb, 2014; 21:1290-1297.

Key words: Cardio-ankle vascular index (CAVI), Arterial stiffness, Nocturnal intermittent hypoxia (NIH), Overweight

Address for correspondence: Takeshi Tanigawa, Department of Public Health, Ehime University Graduate School of Medicine, Shitsukawa, Toon, Ehime 791-0295, JapanE-mail: tt9178tt9178@gmail.comReceived: February 28, 2014Accepted for publication: June 13, 2014

1291Intermittent Hypoxia and Arterial Stiffness

nary disorders were also excluded from the analysis. Written informed consent was obtained from all sub-jects, and the study protocol was approved by the Human Ethics Review Committee of Ehime University Graduate School of Medicine.

Anthropometric and Biochemical MeasurementsThe BMI was calculated according to the follow-

ing formula: weight in kilograms divided by height in meters squared (weight (kg)/[height (m)]2). An over-weight status was defined as a BMI of ≥25 kg/m2. Blood pressure was measured twice in the sitting posi-tion after resting for at least five minutes using an automatic sphygmomanometer (BP-103iII; OMRON Colin Co., Tokyo, Japan). The mean of the two mea-surements was used for the analysis. Hypertension was defined as a systolic blood pressure (SBP) of ≥140 mmHg or diastolic blood pressure (DBP) of ≥90 mmHg and/or the current use of antihypertensive agents. DM was defined as a fasting blood glucose level of ≥126 mg/dL (7.0 mmol/L) or the use of glu-cose-lowering drugs. Dyslipidemia was defined as an LDL-C level of ≥140 mg/dL (3.63 mmol/L), HDL-C level of ＜40 mg/dL (1.04 mmol/L) or TG level of ≥150 mg/dL (1.70 mmol/L) and/or the current use of lipid-lowering drugs, according to the diagnostic crite-ria for dyslipidemia for Japanese individuals27). A self-reported questionnaire was conducted to assess the subjects’ medical history, smoking habits (current smokers, ex-smokers and never smokers) and drinking habits (regular alcohol consumption or none). The level of physical activity was assessed based on a glob-ally used compendium28) comprised of 14 questions regarding occupation, locomotion, housework, sleep time and leisure time spent in physical activity. The questionnaire data were converted to measurements of the intensity of each physical activity expressed in metabolic equivalents (METs) as METs ·h/day.

Assessment of NIHNIH was measured during sleep for one night

using a pulse oximeter (PULSOX-3Si, Minolta Co., Osaka, Japan) in the participant’s home. The sensor probe was fitted to the fourth finger with tape. The data were downloaded to a personal computer via an interface (PULSOX IF-3, Minolta) and analyzed using a proprietary software program (DS-3 ver. 2.0a, Minolta). Because the measurement time for pulse oximetry is often longer than the true total sleep time, an objectively measured sleep log for a single night was utilized to exclude waking time from the analysis in order to minimize the potential for overestimating the total sleep time. The 3% ODI (oxygen desatura-

Arterial stiffness is a major contributor to CVD. Several studies have supported its impact as an early independent marker of cardiovascular morbidity and mortality8). Cardio-Ankle Vascular Index (CAVI) quantitatively reflects arteriosclerosis of the aorta, fem-oral and tibial arteries based on Bramwell-Hill`s equa-tion9) and stiffness parameter10). CAVI has been asso-ciated with risk factors for coronary disease such as diabetes mellitus (DM)11), dyslipidemia12), metabolic syndrome13) and smoking14). Short- and long-term continuous positive airway pressure (CPAP) therapy was found to significantly reduce CAVI15). It has also been shown that short-term weight reduction therapy improves not only metabolic dysfunction, but also the severity of OSA and arterial stiffness, as measured by CAVI16). The measurement of CAVI has demonstrated good reproducibility and was not affected by the blood pressure during measurement17, 18).

To date, the majority of studies linking OSA to arterial stiffness recruited participants from sleep clinic patients19-21), there have been only two community-based studies22, 23), which also recruited patients with sleep disorder breathing and their methodology to measure arterial stiffness was different to CAVI.

The mean body mass index (BMI) of Eastern Asia is much lower than the rest of the world24); OSA was found to affect cardiovascular risk factors more strongly among lower BMI populations25). So far, there have been no studies on the assessment of NIH and arterial stiffness by using CAVI in a community-dwelling population. Therefore, the present study aims to clarify the association between NIH and arte-rial stiffness according to cardiovascular risk factors in a Japanese general community-based population.

Methods

SubjectsWe carried out a cross-sectional study included

in an ongoing longitudinal epidemiological study (Toon Health Study) of the general population living in Toon City, Ehime prefecture, Japan from 2009 to 2012. Toon city is a primarily rural area located on Shikoku Island in southern Japan, with a population of approximately 35,000. Of the 2,033 total partici-pants 30-79 years of age, 1,932 subjects who under-went sleep testing with pulse oximetry and measure-ment of the CAVI were included in this study. Seven subjects were excluded due to having a low ankle-bra-chial index (ABI) of ＜0.9, which indicates severe arte-riosclerosis of the femoral artery related to a falsely low CAVI value26). Subjects with cardio- or cerebro-vascular disease, heart failure, renal failure or pulmo-

1292 Alberto et al.

CAVI=a {(2ρ/DP)×ln (Ps/Pd) PWV2}＋b

where Ps and Pd are the systolic and diastolic blood pressure values, respectively, PWV is the pulse wave velocity between the heart and ankle, a and b are constants, DP is Ps−Pd and ρ is the blood density. This equation was derived from Bramwell-Hill`s equa-tion9) and the stiffness parameter β10). The scale con-version constants were determined so as to match the CAVI with the PWV according to Hasegawa`s method30). After automatically obtaining the measure-ments, the right and left CAVI values were calculated and analyzed using the VSS-10 software program (Fukuda Denshi). The average value of the right and left CAVI measurements was used for the analysis.

In line with the manufacturer’s recommendation, a CAVI of ＜8 was considered normal, 8≤ CAVI ＜9 borderline and CAVI ≥9 abnormal. Patients with a CAVI of ≥9 have been reported to present with signif-icantly higher mean and maximum intima-media thicknesses, carotid beta stiffness indices and levels of carotid plaque31). Furthermore, it has been reported that a cut-off CAVI value of 8.81 yields maximum sensitivity and specificity for detecting coronary artery disease. We therefore defined increased arterial stiff-ness in this study as an abnormal CAVI (≥9), as applied in previous studies31, 32). Further theoretical details of the CAVI method are described elsewhere26).

Statistical AnalysisThe age- and sex-adjusted mean CAVI values and

prevalence of selected cardiovascular risk factors related to the NIH categories were calculated using an analysis of covariance according to three 3% ODI levels: non-NIH, 0 to ＜5; mild NIH, 5 to ＜15 and moderate-to-severe NIH, ≥15 events/hour. A linear regression model was used to test for linear trends across the NIH catego-ries based on a median variable of 3% ODI. Multivari-able-adjusted logistic regression analyses adjusted for age (year), sex, BMI (kg/m2), physical activity (METs ·h/day), smoking (y/n), alcohol consumption (y/n), hypertension (y/n), DM (y/n) and hyperlipid-emia (y/n) were used to estimate the association between the severity of NIH and increased arterial stiff-ness. Furthermore, a stratified analysis of cardiovascular risk factors according to the NIH categories was con-ducted using multivariable-adjusted odds ratios and 95% confidence intervals (CIs). The SAS version 9.2 software program (SAS Institute Inc., NC, USA) was used for all statistical analyses. All probability values for the statistical tests were two-tailed, with a p value of ＜0.05 considered to be statistically significant.

tion index) was employed as an indicator of NIH, and the 3% ODI level recorded during an estimated sleep time of more than four hours was used in the analysis. A 3% ODI of 5 to ＜15 events per hour corresponded to mild NIH and a 3% ODI of ≥15 represented mod-erate-to-severe NIH.

The validity of the pulse oximetry measurements obtained using synchronous overnight recordings acquired with both PULSOX-3Si and standard poly-somnography (PSG) was confirmed in a previous study of 256 consecutive patients. In that study, the sensitivity and specificity were 80% and 95%, respec-tively, for detecting an apnea-hypopnea index (AHI) of ≥5 using PSG with a cut-off threshold of 3% ODI=5. Similarly, the sensitivity and specificity for detecting an AHI of ≥20 using PSG with a cut-off threshold of 3% ODI=15 were 85% and 100%, respectively29). In order to evaluate reproducibility, pulse oximetry was conducted for two nights among 61 men. The median value of 3% ODI was 5.4 on the first night and 4.8 on the second night (p for differ-ence=0.95, Wilcoxon’s signed-rank test). Spearman’s rank correlation coefficient was 0.81 (p＜0.001)25).

Measurement of CAVIThe CAVI reflects the degree of stiffness of the

arteries from the heart to ankles. As arteriosclerosis progresses, the CAVI increases.

The reproducibility of the CAVI measurements was demonstrated by a 3.8% coefficient of variation (CV); this value is within a satisfactory range, as a CV of 5% is generally accepted to be the limit for clinical laboratory testing26). The CAVI was measured accord-ing to a standardized method using the Vasera VS-100 (Fukuda Denshi Co., Ltd, Tokyo, Japan). All CAVI measurements were obtained during morning hours, in which cuffs were applied to the bilateral upper arms and ankles, with the subject lying supine with their head held in the midline position. The ABI was calcu-lated concurrently with the CAVI. The ABI is the ratio of the ankle systolic blood pressure to the bra-chial systolic blood pressure. The examinations were performed after the subject rested for five minutes. The CAVI was calculated based on the stiffness parameter β, which represents the natural vascular stiffness independent of blood pressure, as measured on carotid echography. The pulse wave velocity (PWV) between the heart and ankle was obtained according to the L/T ratio, where L is the distance from the aortic valve to the ankle and T is the time during which the PWV propagates from the aortic valve to the ankle. Scale conversion from the PWV to CAVI was performed using the following formula:

1293Intermittent Hypoxia and Arterial Stiffness

(21.9%), respectively. The age- and sex- and multi-variable-adjusted odds ratios for various cardiovascular risk factors, including age (year), sex, BMI (kg/m2), physical activity (METs ·h/day), smoking (y/n), alco-hol consumption (y/n), hypertension (y/n), DM (y/n) and hyperlipidemia (y/n), were used to evaluate the relationship between increased arterial stiffness and the category of 3% ODI (Table 2)., The multivari-able-adjusted odds ratio (95% CI) of an increased CAVI (≥9) for severe NIH was 1.36 (0.82-2.23)-fold higher than that for the subjects with a 3% ODI of ＜5; however, this association was not statistically sig-nificant (p for trend=0.26).

Furthermore, we conducted a stratified analysis of cardiovascular risk factors associated with the CAVI, such as BMI32) and age33), using a multivari-able-adjusted logistic regression analysis (Table 3). The combination of severe NIH and an overweight status was found to be significantly associated with an increased CAVI (p for trend=0.03). However, no sig-nificant associations were found in the stratified analy-sis of age, sex, hypertension, DM, hyperlipidemia, smoking and alcohol consumption in the subjects

Results

The age- and sex-adjusted mean CAVI values and prevalence of cardiovascular risk factors related to NIH are shown in Table 1. The number of subjects with no, mild and moderate-to-severe NIH was 1,348 (70%), 451 (23%) and 126 (7%), respectively. The mean age and proportion of men were significantly higher in the severe NIH group. The mean BMI was higher among the subjects with an increased 3% ODI, albeit most of the subjects were not overweight. Waist circumference, glucose, HbA1c, triglycerides, hyper-tension and hyperlipidemia were also significantly greater in the participants with an increased 3% ODI. In contrast, the HDL-C and physical activity levels were significantly lower in the severe NIH group than in the mild and non-disease groups. No differences were observed in the rates of alcohol consumption and smoking or the mean total cholesterol or LDL-C lev-els. However, the CAVI values and prevalence of DM were higher in the severe NIH group.

The number of subjects without and with increased arterial stiffness was 1,503 (78.1%) and 422

Table 1. Age and sex-adjusted mean and prevalence of cardiovascular risk factors and CAVI according to 3% ODI levels

Characteristics3% ODI, events/hour

p for trend0 to ＜5 5 to ＜15 ≥15

Number of subjectsAge, y (SD)Males, No.(%)BMI, kg/m2

Waist circumference, cmSystolic blood pressure, mm HgDiastolic blood pressure, mm HgFasting blood glucose, mg/dL (SI)HbA1c, %Total Cholesterol, mg/dL (SI)LDL-Cholesterol, mg/dL (SI)HDL-Cholesterol, mg/dL (SI)Triglycerides, mg/dL (SI)Physical activity (mets · hr/day)Hypertension, %DM, %Hyperlipidemia, %Current smokers, %Regular alcohol drinkers, %CAVI

134855.4 (12.8)384 (28.5)

22.782.3

125.175.9

94.2 (5.23)5.11

202.9 (5.26)118.4 (3.07)

60.2 (1.56)104.6 (1.18)

36.733.8

5.044.011.057.58.16

45163.2 (9.9)205 (45.5)

24.386.3

130.179

96.2 (5.34)5.23

205.8 (5.33)121.8 (3.15)

58 (1.50)118 (1.33)

35.248.5

6.453.011.654.38.16

12664.3 (9.9)

95 (75.4)25.288.7

128.477.84

97.4 (5.41)5.27

198.4 (5.14)115.7 (3.00)

56.4 (1.46)128.2 (1.45)

34.356.9

8.052.710.061.88.26

＜0.01＜0.01＜0.01＜0.01＜0.01＜0.01＜0.01＜0.01

0.500.93

＜0.01＜0.01＜0.01＜0.01

0.10＜0.01

0.890.690.28

Values given as mean or No. (%), unless otherwise indicated. BMI=body mass index; HbA1c=hemoglobin A1c; LDL= low-density lipoprotein; HDL=high-density lipoprotein; Mets=metabolicequivalents; DM=diabetes mellitus; CAVI=cardio-ankle vascular index; SD=Standard Devia-tion; SI= International System of Units.

1294 Alberto et al.

an association between obstructive sleep apnea and increased arterial stiffness associated with aging, but not BMI. This discrepancy may be explained by the higher prevalence of obesity in the above study of Caucasians than that observed in our study of Japa-nese individuals (mean BMI: 32 kg/m2 and 24 kg/m2, respectively) as well as the use of the pulse wave veloc-ity (PWV) to assess the degree of arterial stiffness, which tends to be overestimated in patients with hypertension and autonomic dysfunction26, 36). In comparison to the PWV, the CAVI has distinctive fea-tures. For example, the CAVI changes over a short period of time in response to alterations in circulatory conditions, reflecting the state of smooth muscle con-tractions, rather than changes in blood pressure37). Moreover, the CAVI is not related to systolic or dia-stolic blood pressure17, 36); therefore, it is considered to be superior to PWV as an index of arterial stiffness.

Yim-Yeh et al.23) reported that both OSA and aging impair the endothelial function and increase arterial stiffness in patients with obesity. Although that study included patients with OSA, the mean patient age was lower (mean age: 40 versus 60 years of age) and the mean BMI was higher (37 kg/m2 versus 24 kg/m2) than that observed in our population. Never-theless, both studies indicate that subjects with obesity and OSA have a higher risk of arterial stiffness. Iguchi et al.16) recently reported a significant correlation

with severe NIH and increased arterial stiffness. The interaction test showed a trend for an overweight sta-tus to be a modifier of the association between OSA and increased arterial stiffness (p=0.05).

Discussion

The results of this study demonstrated that severe NIH, as evaluated based on the 3% ODI, has a ten-dency to be associated with increased arterial stiffness, as measured according to the CAVI, especially in over-weight subjects. Previous studies have linked BMI with OSA, although most of the participants in these reports were obese1, 34, 35). It has also been shown that subjects with OSA have a high prevalence of CVD3-6). Therefore, we sought to investigate the relationships between NIH, a surrogate indicator of OSA, arterial stiffness and cardiovascular risk factors in a Japanese general community-dwelling population. Conse-quently, NIH was detected in 30% of our study popu-lation, albeit most of the subjects had mild disease. In addition, the stratified analysis of cardiovascular risk factors using a multivariable-adjusted logistic regres-sion model showed only the combination of a 3% ODI of ≥15 and BMI of ≥25 to be significantly asso-ciated with a CAVI of ≥9.

The present findings differ from those of a previ-ous Caucasian community-based study22) that reported

Table 2. Age-sex and multivariable-adjusted odds ratios (95% confidence intervals) of increased arterial stiffness (CAVI ≥9) accord-ing to 3% ODI levels

3% ODI, events/hourp for trend

0 to ＜5 5 to ＜15 ≥15

TotalNumber of subjectsNumber of subjects with CAVI ≥9 (%)Age and sex-adjusted modelMultivariable-adjusted model

MaleNumberNumber of subjects with CAVI ≥9 (%)Age-adjusted modelMultivariable-adjusted model

FemaleNumber of subjectsNumber of subjects with CAVI ≥9 (%)Age-adjusted modelMultivariable-adjusted model

1348234 (17.36)

1.001.00

38497 (25.26)

1.001.00

964137 (14.21)

1.001.00

451137 (30.38)

0.99 (0.74-1.32)1.20 (0.75-1.39)

20570 (34.15)

0.91 (0.59-1.41)0.95 (0.60-1.49)

24667 (27.24)

1.05 (0.71-1.54)1.07 (0.71-1.64)

12651 (40.48)

1.20 (0.76-1.92)1.36 (0.82-2.23)

9542 (44.21)

1.31 (0.76-2.26)1.43 (0.79-2.56)

319 (29.03)

0.87 (0.34-2.21)0.99 (0.35-2.80)

0.500.26

0.370.24

0.920.88

Multivariable-adjusted for age, sex, BMI (kg/m2), physical activity (mets · hr/day), smoking (y/n), alcoholconsumption (y/n), hypertension (y/n), DM (y/n), and hyperlipidemia (y/n)

1295Intermittent Hypoxia and Arterial Stiffness

and release free oxygen radicals. The phenomenon of hypoxia/reoxygenation, which frequently occurs dur-ing episodes of NIH, may elicit increased vascular oxi-dative stress39), thus inducing arterial stiffness.

An increased severity of OSA is associated with an increased serum leptin concentration40), although the hyperleptinemia observed in subjects with OSA can be normalized with treatment with nasal CPAP41). Moreover, hyperleptinemia is common in patients with obesity, reflecting increased adiposity and leptin resistance, which may contribute to the development of hypertension, impaired glucose metabolism and a pro-atherogenic state42). Therefore, it can be specu-lated that pathways leading to leptin resistance play an important role in the interactions between OSA, ath-erosclerotic diseases and cardiovascular risk factors.

The strengths of this study include the large sam-ple size in a general community-based population of middle-aged and elderly subjects and the fact that most of our participants were non-obese, whereas pre-vious studies primarily evaluated obese individuals.

between the severity of OSA and that of metabolic syndrome and arterial stiffness in a small Japanese population of overweight and obese outpatients. Although no cardiovascular endpoints were examined in that study, the data also support the relationship between obesity and both OSA and arterial stiffness.

Obesity appears to affect the control of the upper airway during sleep via several mechanisms, including alterations in the upper airway structure and function, reductions in the resting load volume and negative effects on the respiratory drive and load compensa-tion34). Additionally, neuromuscular control of the upper airways is negatively impacted by alterations in several key cytokines related to obesity, including leptin, tumor necrosis factor-alpha and interleukin-6 38). The hypoxia, hypercapnia and pressure surges accom-panying intermittent hypoxia events may also serve as potent stimuli for the release of vasoactive substances, leading to impairment of the endothelial function. Moreover, a low oxygen level is a trigger for the activa-tion of neutrophils, which adhere to the endothelium

Table 3. Age-sex and multivariable-adjusted odds ratios (95% confidence intervals) of increased arterial stiffness (CAVI ≥9) accord-ing to 3% ODI levels and stratified by some cardiovascular risk factors (age and BMI)

3% ODI, events/hour p fortrend

p forinteraction0 to ＜5 5 to ＜15 ≥15

BMIBMI ＜25

Number of subjectsNumber of subjects with CAVI ≥9 (%)Age and sex-adjusted ORMultivariable-adjusted OR

BMI ≥25Number of subjectsNumber of subjects with CAVI ≥9 (%)Age and sex-adjusted ORMultivariable-adjusted OR

AgeAge ＜65

Number of subjectsNumber of subjects with CAVI ≥9 (%)Age and sex-adjusted ORMultivariable-adjusted OR

Age ≥65Number of subjectsNumber of subjects with CAVI ≥9 (%)Age and sex-adjusted ORMultivariable-adjusted OR

1106207 (18.72)

1.001.00

24227 (11.15)

1.001.00

96665 (6.73)

1.001.00

382169 (44.24)

1.001.00

28196 (34.16)

0.91 (0.65-1.28)0.93 (0.65-1.33)

17041 (24.12)

1.75 (0.96-3.21)1.48 (0.79-2.78)

23129 (12.55)

1.00 (0.60-1.64)1.04 (0.60-1.81)

220108 (49.09)

0.96 (0.69-1.39)1.01 (0.70-1.46)

6530 (46.15)

0.99 (0.54-1.81)1.01 (0.53-1.90)

6121 (34.43)

2.56 (1.15-5.68)2.53 (1.08-5.96)

5410 (18.52)

1.52 (0.68-3.39)1.80 (0.73-4.42)

7241 (56.94)

1.12 (0.65-1.93)1.25 (0.70-2.24)

0.820.90

0.020.03

0.390.25

0.750.49

0.05

0.44

Multivariable-adjusted for age, sex, BMI (kg/m2), physical activity (mets · hr/day), smoking (y/n), alcohol consumption (y/n), hypertension (y/n), DM (y/n), and hyperlipidemia (y/n)

1296 Alberto et al.

Sleep Apnea. Pro Am Thorac Soc, 2008; 5: 136-1433) Kent BD, Garvey JF, Ryan S, Nolan D, Dodd JD,

McNicholas WT: Severity of obstructive sleep apnoea pre-dicts coronary artery plaque burden: a coronary CT angi-ography study. Eur Respir J, 2013; 42: 1263-1270

4) Elwood P, Hack M, Pickering J, Hughes J, Gallacher J: Sleep disturbance, stroke and heart disease events: evi-dence from the Caerphilly cohort. J Epidemiol Commu-nity Health, 2006; 60: 69-73

5) Latina JM, Estes III NA, Garlitski AC: The relationship between obstructive sleep apnea and atrial fibrillation: a complex interplay. Pulm Med, 2013; 2013: 1-11

6) Kato M, Adachi T, Koshino Y, Somers VK: Obstructive sleep apnea and cardiovascular disease. Circ J, 2009; 73: 1363-1370

7) Tanigawa T: Obstructive sleep apnea: its prevention and screening may contribute to the prevention of hyperten-sion, diabetes and cardiovascular diseases. EPMA J, 2011; 2: 83-89

8) Dolan E, Thijs L, Li Y, Atkins N, McCormack P, McClory S, O’Brien E, Staessen JA, Stanton AV: Ambula-tory arterial stiffness index as a predictor of cardiovascular mortality in the Dublin Outcome Study. Hypertension, 2006; 47: 365-370

9) Bramwell JC, Hill AV: The velocity of the pulse wave in man. Proc Roy Soc of London Series B, 1922; 93: 298-306

10) Hayashi G, Sato M, Niimi H, Handa H, Moritake K: Analysis of vascular wall constitutive law with finite defor-mation theory. Med Electron Biol Eng, 1975; 13: 293-298

11) Ibata J, Sasaki H, Hanabusa T, Wakasaki H, Furuta H, Nishi M, Akamizu T, Nanjo K: Increased arterial stiffness is closely associated with hyperglycemia and improved by glycemic control in diabetic patients. J Diabetes Invest, 2013; 4: 82-87

12) Soska V, Frantisova M, Dobsak P, Dusek L, Jarkovsky J, Novakova M, Shirai K, Fajkusova L, Freiberger T: Cardio-Ankle Vascular Index in Subjects with Dyslipidaemia and Other Cardiovascular Risk Factors. J Atheroscle Thromb, 2013; 20: 443-451

13) Satoh N, Shimatsu A, Kato Y, Araki R, Koyama K, Oka-jima T, Tanabe M, Ooishi M, Kotani K, Ogawa Y: Evalu-ation of the Cardio-Ankle Vascular Index, a New Indica-tor of Arterial Stiffness Independent of Blood Pressure, in Obesity and Metabolic Syndrome. Hypertens Res, 2008; 31: 1921-1930

14) Kubozono T, Miyata M, Ueyama K, Hamasaki S, Kusano K, Kubozono O, Tei C: Acute and chronic effects of smoking on arterial stiffness. Circ J, 2011; 75: 698-702

15) Kato M, Kumagai T, Naito R, Maeno K, Kasagi S, Kawana F, Ishiwata S, Narui K, Kasai T: Change in car-dio-ankle vascular index by long-term continuous positive airway pressure therapy for obstructive sleep apnea. J Car-diol, 2011, 58: 74-82

16) Iguchi A, Yamakage H, Tochiya M, Muranaka K, Sasaki Y, Kono S, Shimatsu A, Satoh-Asahara N: Effects of weight reduction therapy on obstructive sleep apnea syndrome and arterial stiffness in patients with obesity and meta-bolic syndrome. J Atheroscler Thromb, 2013; 20: 807-820

Furthermore, we used the CAVI, a reliable, reproduc-ible and novel marker of arterial stiffness26, 31).

Limitations of this study include the fact that the cross-sectional design cannot be used to establish cau-sality between NIH, arterial stiffness and an over-weight status. The Toon Health Study is an ongoing longitudinal epidemiological investigation; hence, we will continue to assess our population for cause-effect relationships. In addition, we measured oxygen desat-uration during sleep using pulse oximetry to estimate the presence of NIH. Because apneic episodes do not always induce oxygen desaturation, we were unable to accurately detect the presence of very mild NIH. However, the sensitivity and specificity of pulse oxim-etry are high29), as well as the reproducibility25). In addition, the convenience of pulse oximetry for assess-ing intermittent hypoxia at home and its utility in providing an early diagnosis and treatment for patients with NIH have been documented7, 43).

In conclusion, we herein demonstrated that NIH has a tendency to promote increased arterial stiffness, as measured according to the CAVI, especially in over-weight or obese subjects. Therefore, in order to reduce cardiovascular risks, it is important to educate patients diagnosed with OSA about CPAP treatment, as well as weight reduction therapy when necessary.

Acknowledgements

The authors wish to thank the staff and partici-pants of the Toon Health Study.

This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Edu-cation, Culture, Sports, Science and Technology of Japan [No. 22390134] and Health and Labour Sci-ences Research Grants from the Japanese Ministry of Health, Welfare and Labour [No. 201021038A].

Conflicts of Interest

None.

Disclosures

None.

References

1) Drager LF, Togeiro SM, Polotsky VY, Lorenzi-Filho G: Obstructive sleep apnea: a cardiometabolic risk in obesity and metabolic syndrome. J Am Coll Cardiol, 2013; 62: 569-576

2) Punjabi NM: The Epidemiology of Adult Obstructive

1297Intermittent Hypoxia and Arterial Stiffness

401-406 (in Japanese)30) Hasegawa M: Fundamental research on human aortic

pulse wave velocity. Jikei Medical Journal, 1970; 85: 742-760

31) Kim KJ, Lee BW, Kim HM, Shin JY, Kang ES, Cha BS, Lee EJ, Lim SK, Lee HC: Associations Between Cardio-Ankle Vascular Index and Microvascular Complications in Type 2 Diabetes Mellitus Patients. J Atheroscler Thromb, 2011; 18: 328-336

32) Shimizu Y, Nakazato M, Sekita T, Kadota K, Yamasaki H, Takamura N, Aoyagi K, Maeda T: Association between hemoglobin levels and arterial stiffness for general Japa-nese population in relation to body mass index status: The Nagasaki Islands study. Geriatr Gerontol Int, 2013; doi: 10.1111/ggi.12171

33) Choi SY, Oh BH, Bae Park J, Choi DJ, Rhee MY, Park S: Age-associated increase in arterial stiffness measured according to cardio-vascular index without blood pressure changes in healthy adults. J Atheroscler Thromb, 2013; 20: 911-923

34) Ong CW, O’Driscoll DM, Truby H, Naughton MT, Hamilton YS: The reciprocal interaction between obesity and obstructive sleep apnoea. Sleep Med Rev, 2013; 17: 123-131

35) Leong WB, Arora T, Jenkinson D, Thomas A, Punamiya V, Banerjee D, Taheri S: The prevalence and severity of obstructive sleep apnea in severe obesity: the impact of ethnicity. J Clin Sleep Med, 2013; 9: 853-858

36) Takaki A, Ogawa H, Wakeyama T, Iwami T, Kimura M, Hadano Y, Matsuda S, Miyazaki Y, Hiratsuka A, Matsu-zaki M: Cardio-ankle vascular index is superior to bra-chial-ankle pulse wave velocity as an index of arterial stiff-ness. Hypertens Res, 2008; 31: 1347-1355

37) Kim B, Takada K, Oka S, Misaki T: Influence of blood pressure on cardio-ankle vascular index (CAVI) examined based on percentage change during general anesthesia. Hypertens Res, 2011; 34: 779-783

38) Calvin AD, Albuquerque FN, Lopez-Jimenez F, Somers VK: Obstructive sleep apnea, inflammation, and the met-abolic syndrome. Metab Syndr Relat Disord, 2009; 7: 271-278

39) Shamsuzzaman AS, Gersch BJ, Somers VK: Obstructive Sleep Apnea. Implications for Cardiac and Vascular Dis-ease. JAMA, 2003; 290: 1906-1914

40) Tokuda F, Sando Y, Matsui H, Koike H, Yokoyama T: Serum level of adipocytokines, adiponectin and leptin, in patients with obstructive sleep apnea syndrome. Inter Med, 2008; 47: 1843-1849

41) Ip MS, Lam KS, Ho C, Tsang KW, Lam W: Serum Leptin and Vascular Risk Factors in Obstructive Sleep Apnea. Chest, 2000; 118: 580-586

42) Correia ML, Rahmouni K: Role of leptin in the cardio-vascular and endocrine complications of metabolic syn-drome. Diabetes Obes Metab, 2006; 8: 603-610

43) Muraki I, Tanigawa T, Yamagishi K, Sakurai S, Ohira T, Imano H, Kitamura A, Kiyama M, Sato S, Shimamoto T, Konishi M, Iso H: Nocturnal intermittent hypoxia and the development of type 2 diabetes: the Circulatory Risk in Communities Study (CIRCS). Diabetologia, 2010; 53: 481-488

17) Kumagai T, Kasai T, Kato M, Naito R, Maeno K, Kasagi S, Kawana F, Ishiwata S, Narui K: Establishment of the cardio-ankle vascular index in patients with obstructive sleep apnea. Chest, 2009, 136: 779-786

18) Kubozono T, Miyata M, Ueyama K, Nagaki A, Otsuji J, Kusano K, Kubozono O, Tei C: Clinical significance and reproducibility of new arterial distensibility index. Circ J, 2007; 71: 89-94

19) Tonini M, Khayi H, Pepin JL, Renard E, Baguet JP, Lévy P, Romanet JP, Geiser MH, Chiquet C: Choroidal blood-flow responses to hyperoxia and hypercapnia in men with obstructive sleep apnea. Sleep, 2010; 33: 811-818

20) Drager LF, Bortolotto LA, Figueiredo AC, Silva BC, Krieger EM, Lorenzi-Filho G: Obstructive sleep apnea, hypertension, and their interaction on arterial stiffness and heart remodeling. Chest, 2007; 131: 1379-1386

21) Kohler M, Craig S, Nicoll D, Leeson P, Davies RJ, Stra-dling JR: Endothelial function and arterial stiffness in minimally symptomatic obstructive sleep apnea. Am J Respir Crit Care Med, 2008; 178: 984-988

22) Korcarz CE, Gepner AD, Peppard PE, Young TB, Stein JH: The effects of sleep-disordered breathing on arterial stiffness are modulated by age. Sleep, 2010; 33: 1081-1085

23) Yim-Yeh S, Rahangdale S, Nguyen AT, Jordan AS, Novack V, Veves A, Malhotra A: Obstructive sleep apnea and aging effects on macrovascular and microcirculatory func-tion. Sleep, 2010; 33: 1177-1183

24) Balkau B, Deanfield JE, Després JP, Bassand JP, Fox KA, Smith SC, Barter P, Tan CE, Van Gaal L, Wittchen HU, Massien C, Haffner SM: International Day for the Evalu-ation of Abdominal Obesity (IDEA): a study of waist cir-cumference, cardiovascular disease, and diabetes mellitus in 168,000 primary care patients in 63 countries. Circula-tion, 2007; 116: 1942-1951

25) Muraki I, Tanigawa T, Yamagishi K, Sakurai S, Ohira T, Imano H, Kitamura A, Kiyama M, Sato S, Shimamoto T, Konishi M, Iso H: Nocturnal intermittent hypoxia and metabolic syndrome; the effect of being overweight: the CIRCS Study. J Atheroscler Thromb, 2010; 17: 369-377

26) Shirai K, Utino J, Otsuka K, Takata M: A Novel Blood Pressure-Independent Arterial Wall Stiffness Parameter; Cardio-Ankle Vascular Index (CAVI). J Atheroscler Thromb, 2006; 13: 101-107

27) Teramoto T, Sasaki J, Ishibashi S, Biro S, Daida H, Dohi S, Egusa G, Hiro T, Hirobe K, Iida M, Kihara S, Kinoshita M, Maruyama C, Ohta T, Okamura T, Yamashita S, Yokode M, Yokote K: Executive Summary of the Japan Atherosclerosis Society (JAS) Guideline for the Diagnosis and Prevention of Atherosclerotic Cardiovascular Diseases in Japan – 2012 Version. J Atheroscler Thromb, 2013; 20: 517-523

28) Ainsworth BE, Haskell WL, Herrmann SD, Meckes N, Basset DR, Tudor-Locke C, Greer JL, Vezina J, Whitt-Glover MC, Leon AS: 2011 Compendium of Physical Activities: a second update of codes and MET values. Med Sci Sports Exerc, 2011; 43: 1575-1581

29) Nakamata M, Kubota Y, Sakai K: The limitation of screening test for patients with sleep apnea syndrome using pulse oximetry. J Jpn Soc Respir Care, 2003; 12: