Author + information
- Naoki Serizawa, MD,
- Dai Yumino, MD⁎ (, )
- Atsushi Takagi, MD,
- Keiko Gomita, MD,
- Katsuya Kajimoto, MD,
- Yukio Tsurumi, MD and
- Nobuhisa Hagiwara, MD
- ↵⁎Department of Cardiology, Tokyo Women's Medical University, 8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666, Japan
To the Editor:
Obstructive sleep apnea (OSA), characterized by partial or complete occlusion of the pharynx during sleep, results in persistent inspiratory effort and interruption of airflow. During each episode of apnea, OSA patients develop increased transmural pressure in the aortic wall. Accordingly, to test the hypothesis that the presence of OSA would be associated with greater thoracic aortic size, we prospectively assessed 150 consecutive patients, newly referred to the sleep clinic in our institution, in a cross-sectional study to confirm OSA. The patients underwent sleep study and chest computed tomography (CT)-derived thoracic aortic diameter. In this particular period, a chest CT was performed within 3 months of the sleep study upon informed consent to our protocol. Exclusion criteria included: 1) prior history of aortic dissection, aortic valvular disease, and clinical characteristics of Marfan's syndrome; 2) central sleep apnea; 3) treatment for sleep apnea; and 4) dialysis. The outer diameter of the ascending aorta was measured by caliper within the CT image. Overnight sleep study was performed using cardiopulmonary monitoring (Morpheus, Teijin Inc., Tokyo, Japan). The apnea-hypopnea index (AHI) was quantified as the frequency of apneas and hypopneas per hour of bed time. OSA was defined as AHI ≥10/h. The data are presented as mean ± SD or frequencies. To determine the independent factors, the multiple linear regression model with backward elimination technique was used, including older age, male gender, blood pressure, hypertension, dyslipidemia, diabetes mellitus, ischemic heart disease, smoking, and AHI. Comparisons between the 2 groups were performed by Student t test for the unadjusted aortic diameter and by analysis of covariance for the adjusted aortic diameter.
The patients' mean age was 60 ± 11 years; 125 (83%) were men, with mean body mass index of 24.7 ± 3.1 kg/m2. Fifty-nine percent of patients had hypertension. At assessment, 91% of patients had already taken prescribed antihypertensive medications. One hundred ten patients (73%) had OSA. On univariate analysis, older age, male gender, body mass index, systolic blood pressure, pulse pressure, hypertension, dyslipidemia, ischemic heart disease, and AHI were positively correlated with thoracic aortic size. On multivariate analysis, older age (per 10-year increase, coefficient 1.82, 95% confidence interval [CI]: 1.13 to 2.34, p < 0.001), male gender (coefficient 3.25, 95% CI: 1.63 to 4.87, p < 0.001), and AHI (per 10-event/h increase, coefficient 0.62, 95% CI: 0.25 to 0.98, p < 0.001) remained as factors associated with greater thoracic aortic diameter. Contrarily, there was no significant independent relationship between blood pressure/hypertension and thoracic aortic size. Additionally, patients with OSA had a significantly greater thoracic aortic size than those without OSA (p < 0.001) (Fig. 1), even if adjusted for older age and male gender (p < 0.001).
Increased thoracic aortic size is known to be related to aging, male, genetic mutation including Marfan's syndrome, hypertension, and atherosclerosis. There are few published papers comparable to this study. Cistulli et al. (1) reported 2 cases of Marfan's syndrome in which treatment of OSA by continuous positive airway pressure associated with a marked attenuation in the aortic dilatation. Sampol et al. (2) reported 19 patients with thoracic aortic dissection who had a high prevalence of severe OSA. The effect of hypertension on greater thoracic aortic size is accepted. For example, Dapunt et al. (3) reported that a history of hypertension was correlated with greater aortic size. However, Masuda et al. (4) and we did not find a significant independent correlation of blood pressure/hypertension on multivariate analysis; it may be difficult to exclude such a correlation as an important contributor to greater thoracic aortic size in already medicated patients. In addition, OSA can contribute to the development of hypertension. Our data suggested that OSA may contribute to increased thoracic aortic size not only by causing hypertension but also through mechanical stress on the aortic wall from repeated episodes of apneas and hypopnea. Inspiratory effort against occluded upper airway during an OSA episode results in negative intrathoracic pressures as low as −80 cm H2O, which can affect intrathoracic hemodynamics. Peters et al. (5) reported that, in rats, decreased intrathoracic pressure during diastole can distend the intrathoracic aorta. Thus, repeated episodes of sudden increments and changes in the transmural pressure of the aortic wall could contribute to increased thoracic aortic size. Our observations had a main limitation: because the present study was based on cross-sectional data, it is difficult to prove that OSA accelerates aortic expansion.
The present study indicates that OSA could contribute to greater thoracic aortic size.
The authors thank Katsunori Shimada (STATZ Institute Inc., Tokyo, Japan) for expert statistical assistance and are indebted to Koichi Takeuchi and Reiko Ohnishi for their excellent technical assistance.
- American College of Cardiology Foundation