Sleep and Breathing

, Volume 16, Issue 3, pp 677–684

Coexistence of obstructive sleep apnoea and metabolic syndrome is independently associated with left ventricular hypertrophy and diastolic dysfunction

Authors

    • Department of CardiologyTokyo Medical University
  • Yoshifumi Takata
    • Department of CardiologyTokyo Medical University
  • Yuichi Inoue
    • Department of SomnologyTokyo Medical University
  • Katsunori Shimada
    • Second Department of Hygiene and Public HealthTokyo Women’s Medical University
  • Hirofumi Tomiyama
    • Department of CardiologyTokyo Medical University
  • Yosuke Nishihata
    • Department of CardiologyTokyo Medical University
  • Kota Kato
    • Department of CardiologyTokyo Medical University
  • Kazuki Shiina
    • Department of CardiologyTokyo Medical University
  • Akira Yamashina
    • Department of CardiologyTokyo Medical University
Original Article

DOI: 10.1007/s11325-011-0557-2

Cite this article as:
Usui, Y., Takata, Y., Inoue, Y. et al. Sleep Breath (2012) 16: 677. doi:10.1007/s11325-011-0557-2

Abstract

Purpose

This study was conducted to investigate the impact of the severity of obstructive sleep apnoea (OSA) and metabolic syndrome (MS) on left ventricular (LV) hypertrophy and LV diastolic function.

Methods

Echocardiography for evaluation of LV hypertrophy (defined by relative wall thickness (RWT) and LV mass index (LVMI)) and for diastolic function (defined by the early rapid/atrial filling velocity (E/A ratio)) was performed on 660 OSA patients.

Results

In patients with both MS and severe OSA, LVMI and RWT were significantly higher and the E/A ratios were significantly lower compared to patients with neither MS nor severe OSA. Multivariate analysis after adjustment for other descriptive variables demonstrated that (1) coexistent MS and severe OSA was independently associated with increased LVMI and RWT and (2) severe OSA, MS and coexistence of both disorders were independently associated with a decreased E/A ratio. Significant interaction between MS and severe OSA was not observed with respect to LVMI and RWT, but was observed for the E/A ratio.

Conclusions

Coexistent severe OSA and MS can exacerbate LV concentric hypertrophy. However, not only the coexistence of these two disorders, but also either severe OSA or MS can impair LV diastolic function.

Keywords

Left ventricular diastolic functionLeft ventricular hypertrophyMetabolic syndromeObstructive sleep apnoea

Introduction

Metabolic syndrome (MS) is defined as the existence of insulin resistance and/or visceral fat deposition with clinically complex abnormalities that include hypertension, dyslipidaemia and abnormal glucose metabolism. MS is frequently associated with increased cardiovascular morbidities and associated mortality [1, 2]. Severe obstructive sleep apnoea (OSA) is also recognised as an independent risk factor for cardiovascular events such as myocardial infarction, heart failure and ischaemic stroke [3]. Both syndromes are well-known to be associated with obesity, and about half of OSA patients are affected with metabolic syndrome [4, 5]. Furthermore, previous studies have shown an independent association between OSA and the component of MS [6].

In particular, left ventricular (LV) hypertrophy (an important indicator of future cardiac events [7]) and severely impaired LV diastolic function can develop into heart failure with preserved LV systolic function [8]. Although several investigators have reported that patients with OSA or MS frequently have both LV hypertrophy and impaired LV diastolic function [911], differences and similarities of the impact of these two disorders on LV hypertrophy and LV diastolic function have not been clarified. The purpose of this study was to investigate with echocardiography the relationship between severe OSA and MS on LV hypertrophy and LV diastolic function.

Materials and methods

Subjects

This retrospective cross-sectional study was performed on 1,135 consecutive patients who were suspected to have OSA because of their heavy snoring, witnessed apneic episodes by their family members and/or excessive daytime sleepiness, and they were polysomnographically diagnosed with the disorder at our institution from November 2005 to August 2009. Patients who met the following conditions were excluded: <20 years old (n = 9), previous history of cardiovascular and cerebrovascular diseases (n = 305), atrial fibrillation (n = 45), left ventricular ejection fraction (LVEF) of less than 50% (n = 69), receiving chronic haemodialysis (n = 5) and those with an apnoea–hypopnoea index (AHI) of less than 5 events/h (n = 42). Finally, 660 patients (558 men, 102 women; 51.9 ± 14.2 years old) were enrolled. No patients had previously received continuous positive airway pressure (CPAP) treatment. This study was approved by the ethics committee of our institution, and written informed consent was obtained from each patient.

Methods

Sleep study

Fully attended overnight polysomnography (PSG) monitoring was performed by using the Alice 4 Sleep SystemTM (Respironics Inc., Murrysville, PA, USA). The precise protocol of PSG monitoring has been described previously [12]. Oronasal airflow was measured with a thermistor, thoracoabdominal movements were monitored with a strain gauge and percutaneous oxyhaemoglobin saturation was monitored by pulse oximetry. PSG data were analysed by technologists who were blinded to the status of the patients. Apnoea was defined as a cessation of airflow for 10 s or more, and hypopnoea was defined as a 50% or greater decrease in airflow lasting for more than 10 s associated with arousal or a 3% or greater decrease in arterial oxygen saturation from baseline. AHI was calculated as the total number of apnoea and hypopnoea episodes per hour of sleep. A 3% oxygen desaturation index hour (ODI 3%) was set. A patient was classified as having OSA when the obstructive component was dominant and AHI was ≥5 events/h. According to the American Academy of Sleep Medicine criteria [13], the severity of OSA was classified into mild to moderate OSA (5 ≤ AHI < 30 events/h) and severe OSA (AHI ≥ 30 events/h), respectively.

Blood sampling, measurement of blood pressure and definition of metabolic syndrome

Early in the morning after PSGs, fasting blood samples were taken from all enrolled patients, and blood pressure measurements were taken in a supine position after at least 15 min rest. For the diagnosis of MS, fasting plasma glucose (FPG), triglycerides (TG), high density and low density lipoprotein cholesterol (HDL-C and LDL-C, respectively) levels and waist circumference were measured. All patients with diagnosis and treatment of hypertension, dyslipidaemia or diabetes were instructed to continue their medications at the time of blood sampling.

MS was defined according to the diagnostic criteria of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III; ATP III), i.e., abdominal circumference >102 cm in men, >88 cm in women, TG ≥150 mg/dL, HDL-C <40 mg/dL in men, <50 mg/dL in women, blood pressure ≥130/≥85 mm Hg, and FPG ≥110 mg/dL [14].

Echocardiography

In the afternoon just before PSG monitoring, all the enrolled patients underwent a transthoracic echocardiogram (Sequoia, Siemens Medical Solutions USA Inc., Mountain View, CA, USA) by echocardiographers who were blinded to the status of the patients. The left ventricular internal dimensions and wall thicknesses were measured at end diastole and end systole in accordance with the recommendations of the American Society of Echocardiography [15]. LV mass was calculated according to the anatomically validated Devereux formula [16]. LV mass index (LVMI) was calculated as the value of LV mass divided by body surface area. Relative wall thickness (RWT) was calculated by the following formula: RWT = (2 × thickness of left ventricular posterior wall / left ventricular diameter at diastole). Based on LVMI and RWT, LV geometry was divided into four groups (normal, LVMI <125 g/m2 in men, <110 g/m2 in women and RWT <0.44; concentric remodelling, LVMI <125 g/m2 in men, <110 g/m2 in women and RWT ≥0.44; eccentric hypertrophy, LVMI ≥125 g/m2 in men, ≥110 g/m2 in women and RWT <0.44; and concentric hypertrophy, LVMI ≥125 g/m2 in men, ≥110 g/m2 in women and RWT ≥0.44) [17]. The inter-observer and intra-observer coefficients of variation for the measurement of left ventricular mass were 7.2% and 13.5%, respectively. Pulsed Doppler measurements on LV diastolic inflow were carried out under two-dimensional echo guidance as described previously [12]. From this measurement, the peak velocity of early rapid filling (E velocity) and the peak velocity of atrial filling (A velocity) were recorded, and the E/A ratio and the deceleration time were calculated.

Analysis

Results are expressed as mean ± standard deviation (SD) for continuous variables. Student’s unpaired t test was used to compare the means of continuous variables, and categorical variables were analysed using the chi-square test. Patients were divided into four groups according to the presence or absence of MS and severe OSA, i.e. patients with neither MS nor severe OSA (mild to moderate OSA without MS), patients with severe OSA (severe OSA without MS), patients with MS (mild to moderate OSA with MS) and patients with both severe OSA and MS (coexistent severe OSA and MS). Analysis of variance followed by the Bonferroni correction was used to evaluate the differences of descriptive variables among the four groups.

Multivariate linear regression analysis was employed to ascertain whether categorical variables including severe OSA without MS, mild to moderate OSA with MS and coexistent severe OSA and MS, as referenced with mild to moderate OSA without MS, appeared to be independently associated with LVMI or RWT. These analyses were performed at first without adjustment (crude model), and then with adjustment for age, gender, body mass index (BMI), medications for hypertension, those for dyslipidaemia, those for diabetes mellitus and systolic and diastolic blood pressures (adjustment model). When targeting the E/A ratio, univariate analyses followed by multivariate analysis with heart rate, in addition to these independent variables, were also performed [18]. Analysis of covariance (ANCOVA) was used to evaluate the interaction between severe OSA and MS for RWT, LVMI and the E/A ratio. For these analyses, RWT, LVMI and the E/A ratio were set as dependent variables, and severe OSA, MS and severe OSA accompanied with MS were covariated, respectively. In these analyses, it was determined that when severe OSA accompanied with MS was significantly independent, there was a significant interaction, indicating that the effect of the combination of the two disorders (severe OSA and MS) was not simply additive but synergistic [19] for LV measures. The sample of 660 patients had more than 95% power to reject the null hypothesis, which meant that the regression coefficient was 0, with 0 of a two-sided P value.

Computations were performed using SPSS (Version 11.0; SPSS, Chicago, IL, USA). A P value of <0.05 was considered to indicate a statistically significant difference.

Results

Patients’ characteristics

The mean AHI was 38.1 ± 24.0 events/h. Of those, the number of patients who met the diagnostic criteria of MS was 290 (44%), and severe OSA was observed in 359 (54%) patients. A total of 191 patients (29%) had both severe OSA and MS. In 359 patients with severe OSA, MS was observed in 190 (53%) patients.

Differences in descriptive variables according to the presence or absence of MS and severe OSA

Table 1 shows differences in the descriptive variables among the four groups according to the presence or absence of MS and severe OSA. Compared with the mild to moderate OSA without MS group, the severe OSA without MS, mild to moderate OSA with MS and coexistent severe OSA and MS groups were significantly older, and BMI and systolic and diastolic blood pressures were significantly higher. In the two groups with MS and the severe OSA without MS group, values of TG and FPG were significantly higher than those of mild to moderate OSA without MS group.
Table 1

Differences in descriptive variables among the four groups, according to the presence or absence of MS and severe OSA

Variables

Mild to moderate OSA without MS

Severe OSA without MS

Mild to moderate OSA with MS

Coexistent severe OSA and MS

Number

202

168

99

191

Age (years)

45.8 ± 14.3

50.5 ± 14.4*

55.4 ± 12.0*

55.0 ± 13.4*

Gender (M/F)

169/33

154/14

77/22

158/33

BMI (kg/m2)

24.3 ± 3.5

26.0 ± 4.0*

26.4 ± 4.3*

29.7 ± 4.9*,**

Waist circumference (cm)

84.5 ± 13.0

90.1 ± 11.6*

91.7 ± 13.7*

99.0 ± 13.1*,**

BMI ≥30 kg/m2 (%)

5

13

14

39

Hypertension (%)

19

27

62

74

Dyslipidaemia (%)

21

23

84

81

Hyperglycaemia (%)

4

5

25

25

FPG (mg/dl)

87.2 ± 14.6

90.7 ± 16.7

100.4 ± 28.2*

104.1 ± 29.6*

HbA1c (%)

5.2 ± 0.5

5.2 ± 0.9

5.9 ± 1.6*,**

6.0 ± 1.4*,**

TG (mg/dl)

130.0 ± 62.0

159.1 ± 112.0*

185.9 ± 99.2*

202.5 ± 112.7*,**

HDL-C (mg/dl)

53.2 ± 14.9

52.6 ± 14.5

48.9 ± 13.9

47.4 ± 14.2*

LVEF (%)

66.8 ± 4.1

66.9 ± 3.9

67.6 ± 3.4

68.1 ± 4.8*,**

DCT (ms)

195.2 ± 34.9

200.3 ± 35.5

214.1 ± 33.7*

206.4 ± 39.6*

SBP (mm Hg)

119.2 ± 13.7

125.2 ± 14.0*,**

132.1 ± 16.4*,**

134.5 ± 14.3*,**

DBP (mm Hg)

70.5 ± 10.2

76.2 ± 10.4*

79.1 ± 10.7*,**

79.8 ± 10.1*,**

AHI (events/h)

17.5 ± 6.9

50.5 ± 17.2*

19.8 ± 6.6**

58.7 ± 21.8*

ODI 3% (events/h)

11.8 ± 6.1

42.6 ± 18.2*

14.0 ± 7.1

50.4 ± 21.5*

ArI (events/h)

25.9 ± 9.1

50.5 ± 17.3*

27.6 ± 9.1

56.4 ± 21.2*

Lowest SpO2 (%)

85.5 ± 5.4

76.7 ± 8.8*

83.4 ± 6.0

74.0 ± 9.5*

Values are given as mean ± SD, except for categorical variables

BMI body mass index, FPG fasting plasma glucose, HbA1c glycosylated haemoglobin, TG triglycerides, HDL-C high density lipoprotein cholesterol, LVEF left ventricular ejection fraction, E velocity early filling velocity, A velocity atrial filling velocity, DCT deceleration time, SBP systolic blood pressure, DBP diastolic blood pressure, AHI apnoea hypopnoea index, ODI oxygen desaturation index, ArI arousal index

*P < 0.05 compared to the mild to moderate OSA without MS group; **P < 0.05 compared to the severe OSA without MS group

Differences in LV geometry and the E/A ratio according to the presence or absence of MS and severe OSA

Figure 1 shows the percentage of each LV geometry category among the four groups. Normal geometry was most frequently observed in the mild to moderate OSA without MS group (61%). In contrast, concentric hypertrophy was the most frequent in coexistent severe OSA and MS group (46%). Figure 2 shows comparisons of LVMI, RWT and the E/A ratio among the four groups. There were significant differences among the four groups regarding LVMI, RWT and the E/A ratio (F value = 31.0, 31.7, 50.2; P < 0.0001, P < 0.0001, P < 0.0001, respectively). Compared with the mild to moderate OSA without MS group, the LVMI of the severe OSA without MS, mild to moderate OSA with MS and coexistent severe OSA and MS groups were significantly higher (113.2 ± 23.6 vs. 121.5 ± 22.6, 124.8 ± 24.1, 136.8 ± 27.3 g/m2; P = 0.007, P = 0.001, P < 0.0001, respectively). Moreover, the LVMI of coexistent severe OSA and MS group was significantly higher than that of the severe OSA without MS group (P < 0.0001) (Fig. 2a). The RWT of the mild to moderate OSA with MS group and that of coexistent severe OSA and MS group were significantly higher than that of the mild to moderate OSA without MS group (0.40 ± 0.06 vs. 0.44 ± 0.07, vs. 0.46 ± 0.06, P < 0.0001, P < 0.0001, respectively), whereas that of severe OSA without MS group did not statistically differ from that of the mild to moderate OSA without MS group. The RWT of the mild to moderate OSA with MS group and that of the coexistent severe OSA and MS group were also significantly higher than that of the severe OSA without MS group (P = 0.002, P < 0.0001, respectively) (Fig. 2b). Regarding the E/A ratio, the values of the severe OSA without MS group, the mild to moderate OSA with MS group and the coexistent severe OSA and MS group were significantly lower than that of the mild to moderate OSA without MS group (1.4 ± 0.5 vs. 1.1 ± 0.4, 1.0 ± 0.3, 0.9 ± 0.3, P < 0.0001, P < 0.0001, P < 0.0001, respectively). The value of the coexistent severe OSA and MS group was significantly lower than that of the severe OSA without MS group (P < 0.0001) (Fig. 2c).
https://static-content.springer.com/image/art%3A10.1007%2Fs11325-011-0557-2/MediaObjects/11325_2011_557_Fig1_HTML.gif
Fig. 1

Percentage of each LV geometry category in the four groups. Normal geometry was most frequently observed in the mild to moderate OSA without MS group (61%). In contrast, concentric hypertrophy was the most frequent in the coexistent severe OSA and MS group (46%)

https://static-content.springer.com/image/art%3A10.1007%2Fs11325-011-0557-2/MediaObjects/11325_2011_557_Fig2_HTML.gif
Fig. 2

Differences in LVMI, RWT and E/A ratios in the four groups. a The LVMI of the mild to moderate OSA with MS group, of the MS without severe OSA group and of the coexistent severe OSA and MS group were significantly higher than the LVMI of the mild to moderate OSA without MS group. b The RWT of the mild to moderate OSA with MS group and of the coexistent severe OSA and MS group were significantly higher than the RWT of the mild to moderate OSA without MS group. The RWT of the severe OSA without MS group did not statistically differ from the RWT of the mild to moderate OSA without MS group. The RWT of the severe OSA without MS group was significantly lower than the RWT of the mild to moderate OSA with MS group and of the coexistent severe OSA and MS group. c The E/A ratios of the severe OSA without MS, of the mild to moderate OSA with MS and of the coexistent severe OSA and MS groups were significantly lower than the E/A ratio of the mild to moderate OSA without MS group. *P < 0.01 compared with mild to moderate OSA without MS; †P < 0.01 compared with severe OSA without MS

LVMI, RWT and the E/A ratio and the interaction between MS and severe OSA

When LVMI and RWT were set as dependent variables on ANCOVA, both severe OSA without MS and mild to moderate OSA with MS appeared as significantly independent variables, whereas severe OSA accompanied with MS did not become a statistically significant variable. On the other hand, not only severe OSA without MS and mild to moderate OSA with MS, but also severe OSA accompanied with MS appeared to be independent variables for the E/A ratio (Table 2).
Table 2

Analysis of covariance (ANCOVA) for LVMI, RWT and the E/A ratio

Covariates

LVMI

RWT

E/A ratio

Severe OSA

0.0001

0.006

0.0001

MS

0.0001

0.0001

0.0001

Severe OSA * MS

0.36

0.63

0.001

Values are expressed as P values

LVMI left ventricular mass index, RWT relative wall thickness, E/A ratio ratio between the early filling velocity and the atrial filling velocity, OSA obstructive sleep apnoea, MS metabolic syndrome, Severe OSA * MS severe OSA accompanied with MS

Multivariate linear regression analyses for LVMI, RWT and the E/A ratio

Table 3 shows the results of multivariate linear regression analyses for LVMI, RWT and the E/A ratio. In the crude model, severe OSA without MS, mild to moderate OSA with MS and coexistent severe OSA and MS were independently associated with LVMI. Mild to moderate OSA with MS and coexistent severe OSA and MS were also independently associated with RWT, while severe OSA without MS did not appear as an independent factor. In the adjustment model, only coexistent severe OSA and MS was independently associated with both LVMI and RWT. In contrast, severe OSA without MS, mild to moderate OSA with MS and coexistent severe OSA and MS appeared as independent factors not only in the crude model but also in the adjustment model for the E/A ratio.
Table 3

Multivariate regression analysis on the associated factors for LVMI, RWT and E/A ratio

Variables

Crude model

Adjustment model

β

P value

β

P value

LVMI

Severe OSA without MS

0.14

0.001

0.03

NS

Mild to moderate OSA with MS

0.16

0.0001

0.01

NS

Coexistent severe OSA and MS

0.41

0.0001

0.15

0.0001

RWT

Severe OSA without MS

0.03

NS

0.03

NS

Mild to moderate OSA with MS

0.22

0.0001

0.05

NS

Coexistent severe OSA and MS

0.39

0.0001

0.09

0.001

E/A ratio

Severe OSA without MS

−0.28

0.0001

−0.12

0.0001

Mild to moderate OSA with MS

−0.32

0.0001

−0.09

0.004

Coexistent severe OSA and MS

−0.48

0.0001

−0.13

0.0001

Β standard coefficient, NS not significant, LVMI left ventricular mass index, SBP systolic blood pressure, DBP diastolic blood pressure, MS metabolic syndrome, BMI body mass index, OSA obstructive sleep apnea, RWT relative wall thickness, E/A ratio ratio between the early filling velocity and the atrial filling velocity, crude model without adjustment, adjustment model (LVMI and RWT) with adjustment for age, gender, BMI, medications for hypertension, medication for dyslipidaemia, medication for diabetes mellitus, SBP and DBP, adjustment model (E/A ratio) adjustment model for LVMI and RWT plus heart rate

Discussion

In the present study, the severity of OSA and the prevalence of MS in the OSA patients were similar to previous reports [4, 5]. Moreover, the enrolled patients did not have cardiovascular disease or reduced LVEF, both of which are strongly related to LV hypertrophy and/or diastolic function. Therefore, the patients in this study were thought to be suitable for evaluating OSA and/or MS-related changes in LV morphology and diastolic function.

LV hypertrophy is an independent risk factor for cardiovascular events [20] and is strongly associated with poor prognosis [21]. Several independent reports have demonstrated that both MS and OSA contribute directly to LV hypertrophy [22, 23]. In this regard, the present results are consistent with previous studies. However, the most important differences between previous reports and the present report were that the significance of the coexistence of severe OSA and MS was studied. In the present study, about 30% of subjected patients had both severe OSA and MS. In particular, over 50% of severe OSA patients had MS. The present study demonstrated that only coexistent severe OSA and MS appeared as a significantly associated factor for increased LVMI and RWT, respectively, after adjustment for age, gender, BMI, medications for hypertension, dyslipidaemia, and diabetes mellitus and systolic and diastolic blood pressures. This finding could indicate that severe OSA and MS play additive rather than individual roles on the formation of LV concentric hypertrophy in this patient population. Hypertension and hyperglycaemia are individually important risk factors for LV concentric hypertrophy, and treatment for these disorders has been also known to be effective for the prevention of LV hypertrophy [24]. However, the present study demonstrated that only when severe OSA and MS coexist, concentric LV hypertrophy can appear irrespective of the existence of the above haemodynamic and metabolic disorders.

Heart failure with preserved LVEF, mainly caused by impaired LV diastolic function, is common and its long-term outcome is known to be similar to reduced LVEF [25, 26]. Previous studies have shown that OSA and MS can affect LV diastolic function [911]. Several candidates, such as activation of sympathetic nerve activity, increased transmural pressure during apnoeic episodes and elevated systemic arterial blood pressure brought about by OSA, have been suggested to play roles in impairment of LV diastolic function [27]. On the other hand, CPAP treatment can improve the early stages of LV diastolic dysfunction in patients with OSA [10, 27]. However, the relationship between the impact of these two disorders on LV diastolic function has been unclear. In the present study, severe OSA and MS were independently associated with decreased E/A ratio. Moreover, coexistent severe OSA and MS was also thought to independently impair LV diastolic function. This finding could indicate that severe OSA and MS, individually as well as synergistically, impair LV diastolic function. Thus, intensive treatments for each of these two disorders could be desirable for the prevention of the exacerbation of LV diastolic dysfunction, since these two disorders could impair LV diastolic function not only individually but also synergistically.

Although the precise mechanisms by which the coexistence of severe OSA and MS negatively affects LV morphology and function have not been clarified, early induction of CPAP treatment in patients with coexistent severe OSA and MS should be recommended for preventing the development of heart failure with preserved LVEF. Previous researches has demonstrated that OSA and MS additively increased pulse wave velocity and plasma levels of C-reactive protein [28, 29], both of which have been known to be independently associated with LV hypertrophy and diastolic dysfunction [29, 30]. This might suggest that systemic inflammation and arterial stiffness play causative roles on LV hypertrophy and LV diastolic dysfunction in patients with severe OSA and MS. However, further investigation is needed to clarify this issue.

The present study has several limitations. First, because our institution accepts referred OSA patients on a preferential basis, it is very difficult to collect patients with neither OSA (AHI <5 events/h) nor MS. Therefore, these patients were not enrolled in this study as controls. However, since long-term cardiovascular outcomes are not different between men without OSA and those with mild to moderate OSA [31], this limitation is thought to have a minimum impact on the results of the present study. Second, tissue Doppler imaging, which enables precise detection of LV diastolic function, was not used in this study. Third, because the present study was performed only on Japanese patients, the influence and differences among races cannot be investigated. Fourth, although the regression model was adjusted by BMI, visceral obesity evaluated with abdominal computed tomography was not used. Fifth, in the multivariate model, since patients without OSA could not be included due to an insufficient sample size, the severity of OSA manifested on AHI was not considered a continuous variable but as a categorical variable. Smoking habits and alcohol consumption were also not included in the multivariate model due to a lack of information. However, the results of the present study revealed that there were differences in the pathological roles of severe OSA and MS between for the development of LV hypertrophy and for that of diastolic dysfunction. In conclusion, clinicians should pay attention to the significance of the coexistence of these disorders so as to prevent the development of heart failure with preserved LVEF. Further prospective study on whether CPAP treatment can improve LV morphologies and diastolic function in this kind of patient population is desirable.

Acknowledgements

The authors are indebted to Mr. Roderick J. Turner and Prof. J. Patrick Barron of the Department of International Medical Communications of Tokyo Medical University for their review of this manuscript.

Funding

This work was supported by a fellowship for young clinical sleep researchers from the Japanese Society of Sleep Research.

Conflict of interests

None.

Copyright information

© Springer-Verlag 2011