Summary

Obstructive sleep apnea (OSA) is a common disorder with growing incidence. Major risk factors for OSA are obesity, aging, gender and menopause. As life expectancy lengthens and the obesity epidemic is ongoing, we can assume that OSA will affect an increasing part of the population. Pathological consequences of this disease include an increased risk of arterial hypertension, coronary artery disease, arrhythmia, heart failure as well as cerebrovascular diseases, such as stroke, transient ischemic attack and cognitive dysfunction. The cerebrovascular system differs significantly from other vessels in the body. Brain oxygen demands constitute about 20% of the total oxygen consumed by the body.

OSA significantly affects the cerebral blood flow both during sleep and daily activities. This can have serious health consequences and makes the brain more vulnerable to ischemia. In this review we describe the impact of OSA on cerebral circulation during both sleep and wakefulness and we also outline the pathophysiology of these changes.

In patients with other risk factors for cerebral ischemia, early screening and treatment for OSA should be introduced.

Introduction

The main characteristic of obstructive sleep apnea (OSA) is the obstruction of the upper airways during sleep, which leads to intermittent hypoxia, reoxygenation, hypercapnia, and sleep fragmentation. The prevalence of this disease differs widely in the available data, depending on the size and characteristics of the study group [11, 40, 44, 70, 75, 87]. An analysis by Peppard, Young et al. estimates the incidence of moderate to severe OSA (Apnea+Hypopnea Index – AHI ≥15) in 13% of men and 6% of women between 30-70 years of age. In this study, 14% of men and 5% of women have symptoms of daytime sleepiness and an AHI ≥5 [70]. Some groups of patients show an increased incidence of OSA. These subpopulations include overweight or obese and middle-aged, older, post-menopausal subjects. Epidemiological results do not show substantial differences between developed and developing countries [99]. OSA is associated with numerous vascular complications, increased risk of hypertension, coronary artery disease, arrhythmia, heart failure as well as cerebrovascular diseases, such as stroke and transient ischemic attack [24, 28, 52, 80, 85]. The pathogenesis of the negative impact on vascular system includes endothelial dysfunction, oxidative stress, increased sympathetic activation, and systemic and vascular inflammation, which are caused mainly by intermittent hypoxemia [23]. The use of devices producing positive airway pressure (continuous positive airway pressure, CPAP) is recognized as an effective treatment for OSA. Regular and proper use of CPAP normalizes sleep architecture, significantly reduces snoring, daytime sleepiness and improves the quality of life [20, 33, 62, 68, 90]. OSA is well known as one of the most common causes of secondary arterial hypertension. Most studies show the positive effect of CPAP therapy in lowering blood pressure, although in patients with minimally symptomatic sleep apnea, the effect on blood pressure is not pronounced [8, 12, 58]. Patients with resistant hypertension or those taking antihypertensive pharmacotherapy are shown to have more substantial BP reduction [37]. A decreased risk of stroke is another beneficial effect of CPAP. In Sleep Apnea Cardiovascular Endpoints (SAVE) trial, the risk of a cerebrovascular event was significantly lower in patients using CPAP for more than 4 hours [59, 61]. The cerebrovascular system differs significantly from other vessels in the body. Brain oxygen demands constitute about 20% of the total oxygen consumed by the body. To meet this requirement, constant blood flow and efficient mechanisms for regulation of flow pressure are indispensable. In this work we want to focus on the impact of OSA on cerebral circulation.

Cerebrovascular consequences of OSA

Studies performed in a large group of patients showed a significant association between OSA and cardio- and cerebrovascular diseases, including stroke, coronary artery disease, and heart failure [28, 34, 52, 76, 96]. Minoguchi et al. performed MRI imagining on subjects with OSA and a control group to determine the incidence of silent brain infarctions (SBI) with OSA. The results showed that SBIs are more common in patients with moderate to severe OSA, approaching 25% vs 6.7% in control subjects [60]. Data presenting high association between OSA and stroke should be analyzed carefully, as numerous confounding variables co-exist with OSA. However, even after adjusting for age, sex, race, smoking, alcohol, BMI, diabetes, hyperlipidemia, atrial fibrillation, and hypertension, OSA increases the risk for stroke [96]. The damage caused by brain infarction is more severe in patients with obstructive sleep apnea. There is also an increased risk of subsequent stroke in untreated OSA patients [2, 16, 27, 46, 63, 86]. Longer hospitalization, rehabilitation and worse neurologic recovery are described in literature in comparison to patients without OSA. Higher incidence of delirium, depressed mood, latency in reaction, difficulties in everyday living decrease in attention and problem-solving are also observed [1, 6, 13, 79, 97]. The evidence of beneficial effects of CPAP on reducing negative stroke outcome is not so obvious. Some data shows that early use of CPAP in OSA- stroke- patients has a beneficial effect, as it improves neurologic recovery, delays the appearance of cardiovascular events [6]. Martínez-García et al. suggest that CPAP reduces the excess of incidence of nonfatal cardiovascular events [57]. On the other hand, Hsu et al. report no benefit from CPAP treatment, which was assessed by changes in the Nottingham Extended Activities of Daily Living scale (EADL) after 6 months of CPAP therapy. One of the reasons for this finding may be poor adherence to CPAP therapy, which has been reported in stroke patients [36, 69].

Impact of OSA on functional and structural dysfunction of blood vessels

Vascular damaging agents include both physical (blood pressure) and the cellular and molecular mechanisms (vasoactive substances).

The link between hypertension and OSA is well established [22, 23, 24, 71, 99]. The absence of nocturnal blood pressure dipping, extreme dipping, or reversed dipping have been associated with an increased risk of silent cerebral infarctions and strokes in older hypertensive patients [21]. Fluctuations in blood pressure during apneas and hypopneas can negatively affect cerebral vessels. As the fluctuations can be rapid, autoregulation may not be able to adapt to these changes. That fact exposes especially small arteries and arterioles to damage. Balfors and Franklin measured intra-arterial radial blood pressure during sleep. A gradual increase in blood pressure during apneas and a rapid decline after they terminated was observed. This period immediately after the apnea, when a fall in BP was noticed, is a time of particular vulnerability of the brain to hypoxia [5]. Repetitive surges in blood pressure at the end of apneas and hypopneas result in increased shear stress [31].

Endothelium plays a key role in the regulation of cerebrovascular circulation. A physiologically proper balance between vasoconstrictors and vasodilators must be preserved. The most important substance produced by endothelium is nitric oxide (NO). It is a strong vasodilator of major arteries and arterioles in parenchyma [26]. OSA significantly affects vascular endothelium function, promotes pro-inflammatory and pro-oxidant activation, increased coagulability and disturbed balance between metalloproteases and their inhibitors [26]. Nitric oxide is synthesized by endothelial nitric oxide synthase (eNOS). The main site of action of NO is soluble guanylate cyclase (sGC). Activation of sGC leads to increased production of cyclic guanosine monophosphate (cGMP). Further effects result from cGMP-dependent protein kinase I and include reductions in intracellular calcium, decreases in vascular tone, and changes in gene expression, which results in vessel dilatation [29]. Numerous research reports reveal lower plasma nitrates concentration in subjects with OSA compared to healthy patients. This occurrence is inversely related to the severity of OSA [14, 48, 67, 84]. Successful CPAP treatment, the gold standard in treating OSA, improves plasma nitrates concentration [17, 66, 72]. Other gas mediators, such as hydrogen sulfide (H2S) and carbon monoxide (CO), are also involved in physiological and pathological processes in the vascular system. CO is produced from heme by heme oxygenase (HO). The mechanism of action is similar to NO and also includes the activation of cGMP and calcium-gated potassium channels. Studies in animal models showed the vasodilatory effect of CO [42, 77]. However, in certain situations vasoconstrictor and NO antagonizing effects of CO were also described [38, 41]. Endothelial nitric oxide and prostacyclin interact with CO system in circulatory regulation. In cerebral arterioles in contrast to dilation to acute CO increase, a prolonged exposure to elevated CO produces progressive constriction by inhibiting nitric oxide synthase [55]. Patients with OSA demonstrate higher levels of CO. Azuma et al. evaluated exhaled and blood CO concentrations at night and in the morning. Exhaled CO levels at night in the OSA patients were significantly higher than in the control group. No significant differences in blood CO were observed [4]. Kobayashi et al also measured serum CO concentration in OSA patients before and after sleep. In this study CO levels were significantly increased in the morning, but not in the evening. The change in CO level, which was defined as a gap between the presleep and postsleep CO levels correlated with apnea-hypopnea index and hypoxia duration as a percentage of total sleep time [51].

Fig. 1. Factors regulating cerebral blood flow

Endogenous hydrogen sulfide is synthesized from cysteine by beta-cystathionine synthase (CBS) or gamma-cystathionine lyase (CSE). CBS plays a major role in the brain. Recent research proves that the another source of H2S can also be 3-mercaptopyruvate sulfurtransferase (3MST) and cysteine aminotransferase (CAT) which are both localized to vascular endothelium [82]. Its vasodilatory effect is mainly dependent on acting on the KATP-channels, but a direct impact on angiotensin converting enzyme (ACE) cannot be excluded [10, 54, 100]. Jain et al. in their research observed a significant decrease in circulating H2S in diabetic patients with sleep apnea compared with diabetic patients without sleep disorders [43]. Reports on H2S in patients with OSA are still few and further studies are needed.

Intermittent hypoxia is one of the main features of OSA. The carotid bodies situated at the bifurcation of the common carotid arteries, as well as aortic bodies located in the aortic arch and at the thoraco-abdominal level, are the sensory organs which detect changes in oxygen levels in arterial blood and induce cardiovascular adaptation to provide sufficient oxygen supply. When these peripheral chemoreceptors are activated by hypoxia, a rise in the sympathetic nervous system occurs (leading to BP rise). Recurrent episodes of hypoxia activate pro-inflammatory molecules increase oxidative stress markers production, reactive oxygen species and reactive nitrogen species levels. These changes promote long-lasting stimulation of arterial chemoreflexes and elevated chronic sympathetic nerve activation [81]. The other factor contributing to sympathetic nerve activation is sleep fragmentation, a common feature of OSA. Additionally, recurrent episodes of hypoxia cause the activation of renin- angiotensin system, elevation of endothelin I, pro-inflammatory transcription factors such as hypoxia-inducible factor (HIF)-1α, and nuclear factor that determine an activation of some genetic factors and an induction of adhesion molecules, tumor necrosis factor-α, interleukin-6, chemokines, C-reactive protein, cyclooxygenase, and thromboxane synthase [29, 31].

Cerebrovascular circulation

As constant blood flow in the brain is necessary, multilevel regulation is needed. The most important mechanisms controlling the blood flow in the brain are vessel resistance, myogenic response in cerebral arteries and arterioles and neurogenic factors. The large and small arteries include perivascular nerves in their adventitial layer that originate from the peripheral nervous system but innervation of parenchymal arterioles and cortical microvessels comes directly from the brain tissue. In peripheral circulation, vessel resistance is mainly dependent on small vessels and arterioles. In the cerebrovascular system, the role of large arteries is more pronounced. When an increase of the blood flow is needed in some areas, vasodilation from distal to proximal arterial segments occurs. Upstream vessels must dilate in order to avoid reductions in downstream microvascular pressure. Myogenic response is the ability of smooth muscle of large and small vessels to constrict or dilate in response to changes in pressure. Increased blood pressure leads to constriction and decreased one to dilatation. Although response is myogenic, some factors released from the endothelium or perivascular nerves can affect this autoregulation [18]. Endothelium secretes a number of biologically active substances, which are involved in the regulation of vascular tone, pro-, anti- inflammatory and coagulation homeostasis. Substances regulating vascular tone can be divided in two groups with contradicting properties. Endothelial derived relaxing factors (ERDF), represented especially by nitric oxide (NO), are responsible for dilatation. On the other hand, endothelium-derived contracting factors (EDCF), represented among others by endothelin, cause the contraction of smooth muscles of blood vessels [98]. Some data suggests that cerebral arteries, as well as other cells of the cardiovascular system, have a functional circadian clock and show a diurnal rhythm in response to vasoactive substances. In animal models OSA affects these rhythms in cerebral arteries, which may contribute to adverse cerebrovascular outcomes [25].

Impact of OSA on cerebral blood flow (CBF) during apnea

During an episode of apnea, oxygen partial pressure (PO2) decreases and carbon dioxide partial pressure (PCO2) increases. Such changes in gas pressures lead to vasodilatation. The elevation of PCO2, among other factors (such as the arousal and activation of the sympathetic nervous system) additionally raises systemic blood pressure. Although hypercapnia leads to a relatively uniform vasodilation effect in the whole cerebrovascular area, the role of sympathetic tone in the brain is not so pronounced as in the other vessels of the body. Sympathetic stimulation constricts large cerebral arteries, but at the same time this is compensated by the dilation of resistance arterioles [7, 18, 53, 78, 91, 93]. Kuznetsova and Kulikov measured the responses of middle cerebral arteries velocity (MCAv) and mean arterial blood pressure (MAP) to CO2 in humans. Hypercapnia in healthy volunteers was obtained by a rebreathing test. A breathing circuit with 1000 ml of dead space extra volume was connected to the face mask. This led to an increase in end-tidal partial pressure of CO2 (PETCO2) by 10-15 mmHg. During this maneuver, PETCO2 as well as middle cerebral arteries velocity (assessed by transcranial Doppler) started increasing during the first 10 s and continued to increase for 60 s, then remained at the level reached with minor alterations. No sharp growth in MAP at the beginning of rebreathing was observed. MAP started increasing after about 30 seconds of rebreathing and continued to 2 min, then remained at the level reached. SpO2 decreased gradually during 2-3 min and remained reduced until the end of rebreathing [53]. Similar results were reported by Battisti-Charbonney et al. In this research, the cerebrovascular response to increasing PCO2 at hyperoxic, hypoxic and isoxic pressures were measured using the Duffin rebreathing method. Dynamic rebreathing was implemented by programming the gas mixing device to supply a flow of gas with a pCO2 equal to the PETCO2 of the previous breath and O2 sufficient to maintain an isoxic PETCO2 at either 150 mmHg (hyperoxic test) or 50 mmHg (hypoxic test). An increase in MCAv and MAP was observed. Additional comparison of isoxic hyperoxic with isoxic hypoxic tests enabled the determination of the pO2 effect. Hypoxia significantly increased the maximum percentage rise in MCAv in the majority of subjects [7]. An important limitation of these studies is that responses to hypoxia have been assessed during wakefulness. OSA patients do not typically experience hypoxia during wakefulness, but during sleep. Research done by Staudacher et al. assessed CBF during sleep. TCD was performed continuously during the whole night and was co-registered with polysomnography. Blood flow was minimal before episodes of apneas or hypopneas and maximal closely after their termination [32]. Another mechanism that affects CBF (for example perivascular innervation as mentioned above) could also contribute to alterations in cerebral blood flow during episodes of apnea but none data evaluating that influence is now available [26].

Fig. 2. Pathophysiology of vascular complications of OSA

Impact of OSA on resting cerebral blood flow (CBF)

Data in the literature indicates that CBF in patients with OSA is altered not only during sleep. One of effective and non-invasive methods assessing cerebral blood flow is a transcranial Doppler ultrasonography of cerebral blood flow velocity (CBFV) in the middle cerebral artery. Navarro et al. investigated 152 subjects; all participants underwent a daytime transcranial Doppler study of the right middle cerebral artery to record cerebral blood flow velocity. Results showed a significant reduction in mean cerebral blood flow velocity in OSA patients in comparison to non-OSA subjects [19]. The reasons why opposite changes in CBF occur during wakefulness and sleep are not fully understood. Some authors suggest a possible impact of an increase in peripheral resistance associated with small vessel disease or significant increase in middle cerebral artery diameter induced by nocturnal hypercapnia and hypoxia. [19, 39, 47, 95]. Arterial spin labeling (ASL) procedures are non-invasive MRI-based methods that quantify regional CBF values with greater sensitivity and specificity than Doppler techniques. Carrie and colleagues investigated cerebral perfusion in people with untreated obstructive sleep apnea. Results showed no significant difference in regional perfusion between controls and people with mild OSA, but patients with moderate-severe OSA had decreased perfusion, while awake [17, 39]. These findings are in line with results from studies carried out by Yadav et al. In this research, which included 11 OSA patients (with mean AHI = 32.9/h), ASL MRI showed significantly reduced regional CBF values [95]. In addition in changes in CBF, Chen H.L. et al. examined leukocyte apoptosis as a marker of systemic inflammation. The patients with moderate-severe OSA showed decreased global and regional CBF and also significantly increased percentages of total leucocyte apoptosis and granulocyte late apoptosis, suggesting increased systemic inflammation [95]. Joo Y. et al. used the 99mTc-ethylcysteinate dimer (ECD) single photon emission computed tomography (SPECT) images to assess cerebral blood flow during wakefulness. This study showed reduced blood flow in the parahippocampal and lingual gyri and a negative correlation between AHI and the CBF during the daytime awake state [47]. Using the same technique, Kim at al. showed improvement in resting CBF after CPAP treatment in the brain regions with significant hypoperfusion in the pretreatment examination [50]. Regardless of the method chosen to assess cerebral blood flow numerous studies have shown that resting daytime CBF is significantly decreased in a subset of OSA patients with severe disease. This can contribute to the negative impact of OSA on attention, memory, learning and overall cognitive function [49].

Impact of OSA on autoregulation of CBF

Autoregulation is indispensable to maintain stable blood flow in the brain, when blood pressure changes occur. Data suggests that this mechanism is also affected in subjects with OSA. Macey and colleagues assessed acute BOLD (blood oxygen level dependent signal, a magnetic resonance imaging procedure) responses in OSA subjects to pressor challenges that elicit cerebral blood flow changes – Valsava maneuver, hand grip and cold pressor. The results showed that the amplitude of the global BOLD signal indicative of brain blood volume and oxygenation is altered in OSA patients compared to healthy control subjects during hand grip and cold pressor autonomic challenges [56]. Nasr et al. measured mean cerebral blood flow velocity (CBFV) in the middle cerebral artery and mean arterial blood pressure (ABP) with transcranial Doppler. Autoregulation was assessed with the Mx autoregulatory index. Mx is a moving correlation coefficient between mean CBFV and mean ABP (the higher the Mx the worse autoregulation). Patients and controls were examined in the next day morning after an overnight complete polysomnography. Results showed that cerebral autoregulation was impaired in OSA patients compared to controls and the severity of autoregulation impairment was correlated to the severity of OSA [64]. Urbano and colleagues assessed cerebral artery blood flow velocity (CBFV) by using a transcranial Doppler ultrasound and arterial blood pressure during orthostatic hypotension and recovery as well as during 5% CO2 inhalation. Autoregulatory response was expressed by measuring the time to recovery of CBFV after a decrease in mean arterial pressure. Although there was no significant difference in response to CO2 inhalation, OSA patients had a lower rate of recovery of cerebrovascular conductance after orthostatic challenge [88]. In the study of Gregori-Pla et al., CBF measured at different head-of-bed positions showed a similar response after changing position from supine to 30° in OSA and healthy subjects. After being tilted back to the supine position, patients with severe OSA, contrary to control and the mild OSA group, did not recover to the initial baseline. After 2 years of CPAP treatment this altered cerebral vasoreactivity response was normalized [35]. The mechanisms for alterations in autoregulatory response can be partially explained by myogenic response of brain arteries or endothelium activity. Data in the literature indicate that the intermittent hypoxia leads to disruption of endothelial function, oxidative stress, inflammation, atherosclerosis and hypertension. All these factors contribute to the dysregulation of autoregulatory mechanisms related to cerebral blood flow. The impaired cerebral autoregulation significantly contributes to increased risk of stroke in OSA patients [39].

Conclusion

OSA significantly affects brain circulation. Cerebral blood flow both during sleep and daily activities is decreased and autoregulation is impaired. Endothelial dysfunction, oxidative stress, increased the carotid body response and sympathetic activity, systemic inflammation have a major role in cerebrovascular complications. These changes can have serious health consequences and make the brain more vulnerable to ischemia. This suggests that in the group of patients with other risk factors for cerebral ischemia, early screening for OSA should be considered.

No funding was received for this research.

All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this manuscript.

Full text