© 2005 American Thoracic Society doi: 10.1164/rccm.2506001
Obstructive Sleep Apnea and Atherosclerosis"Guilt by Association"University Hospital, Zurich, Switzerland Over the last 10 to 20 years, the association between obstructive sleep apnea (OSA) and such cardiovascular diseases as hypertension (1), stroke (2, 3), ischemic heart disease/heart failure (4), and sudden death in sleep (5) has been recognized. The mechanisms linking OSA with cardiovascular diseases are complex and may include hemodynamic, neural, metabolic, endothelial, coagulatory, and inflammatory consequences of nocturnal respiratory events and hypoxia (6). The frequent association of OSA with obesity, increasing age, hypertension, and hypercholesterolemia challenges the hypothesis that OSA represents an independent risk factor for cardiovascular diseases. Several lines of evidence support, however, the existence of such a direct (causative) link. First, epidemiologic studies that controlled for risk factors confirmed the existence of an independent association between OSA and cardiovascular diseases (7). Second, markers/early signs of atherosclerosis, including carotid intima-media thickness (IMT) (8), carotid-femoral pulse wave velocity (PWV) (9), carotid diameter (10), and circulating inflammatory factors (e.g., C-reactive protein [CRP], interleukin [IL]-6 [11]) are increased in patients with OSA. Third, treatment with continuous positive airway pressure reverses some of the detrimental cardiovascular effects associated with nocturnal respiratory events (including elevated arterial blood pressure values [12]), reduces surrogate markers of atherosclerosis (e.g., CRP [11]), and improves long-term cardiovascular morbidity and mortality of patients with OSA (13). Two articles in the current issue of the AJRCCM give further support to the hypothesis of a direct link between OSA and cardiovascular diseases. Both reports demonstrate, in fact, the presence in patients with untreated OSA, free of overt cardiovascular diseases, of early signs of atherosclerosis of the carotid artery. In the first study by Minoguchi and coworkers (pp. 625–630), 36 patients with OSA and 16 obese control subjects matched for age, sex, and body mass index, and free of hypertension, diabetes, and hyperlipidemia, were studied (14). A significant increase in carotid IMT, percentage of patients with carotid plaques, and serum levels of CRP, IL-6, and IL-18 was found in patients with OSA. In addition, some of these atherosclerotic signs were more severe in patients with an apnea–hypopnea index (AHI) of 30 or greater and directly linked with the duration of nocturnal hypoxia. Unfortunately, most of these correlations, and particularly those between carotid IMT and IL levels, were rather weak (r < 0.5), although statistically significant. In addition, IL-6 and IL-18 were assessed only at a single early-morning time point (5:00 A.M.), excluding the possibility of knowing if the elevated levels persist throughout the day. In the second study, Drager and colleagues (pp. 613–618) assessed 30 patients with OSA and 12 control subjects, matched for age, sex, and body mass index (15). As in the first study, associated cardiovascular diseases and risk factors (including hypertension, diabetes, smoking, hypercholesterolemia) were excluded in both populations. A significant increase in carotid IMT, carotid diameter, and PWV was found in patients with severe, but not with mild–moderate, OSA (AHI > 30). In addition, some of these atherosclerotic signs were directly linked with the duration of nocturnal hypoxia. In the multivariate analysis, however, only the AHI was significantly and independently associated with carotid IMT and PWV (but not with the carotid diameter). The main results of these two studies are similar and confirm the association between OSA and carotid atherosclerosis (8), a marker of more generalized atherosclerosis and a predictor of cardiovascular diseases (including stroke). In line with these data, we observed an increased frequency of large-vessel disease (including carotid disease) in stroke patients with sleep-disordered breathing (Bossetti C, Milanova M, Gugger M, submitted for publication). Both studies have the merit, compared with previous reports, of controlling for most cardiovascular risk factors and diseases in both patients with OSA and control subjects. Limitations of both studies are the small number of patients studied, the almost complete absence of data on women, and the lack of repeated measurements throughout the day. The latter is necessary to rule out the influence of OSA-independent, circadian factors. More importantly, several basic questions concerning the link between OSA and atherosclerosis were not addressed or remain unclear: (1) Which factors promote atherosclerosis in OSA: nocturnal hypoxia, hemodynamic consequences of repeated arousals, or both? (2) Are the observed early signs of atherosclerosis a direct consequence of OSA or do they arise from a common, underlying, and possibly even genetically determined pathophysiology? (3) Can the observed early atherosclerotic signs be reversed by continuous positive airway pressure treatment? In the absence of convincing data on these three questions, it is not surprising that different mechanisms, including serum inflammatory markers and changes in mechanical and reparative properties of the vessels, were evoked by the authors of the two articles to explain the same observation of early atherosclerosis in OSA. In conclusion, the evidence for a direct, independent link between OSA and cardiovascular diseases, including stroke, is increasing, and additional useful pieces of this complex mosaic have been offered by Drager and colleagues (15) and Minoguchi and coworkers (14). Although the exact extent and the nature of this association remain unclear at this point, it is reasonable to suggest that the presence of cardiovascular risk factors and diseases in patients with OSA should favor the decision to initiate treatment. Large-scale treatment studies that demonstrate improvement of markers and signs of atherosclerosis and, more importantly, reduction in cardiovascular morbidity and mortality are obviously needed to justify systematic treatment of patients with OSA even in the absence of prominent sleepiness-related daytime symptoms. FOOTNOTES Conflict of Interest Statement: C.L.B has received $2,000 for speaking at conferences sponsored by ResMed and Orphan over the past 3 years. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
