© 2008 American Thoracic Society doi: 10.1164/rccm.200801-051UP
Update in Sleep and Control of Ventilation 20071 Department of Medicine, 2 Department of Physiology, and 3 Centre for Sleep Medicine and Circadian Biology at the University of Toronto, Toronto, Canada; and 4 Sleep Research Laboratory of the Toronto Rehabilitation Institute and 5 Department of Medicine, Toronto General Hospital of the University Health Network, Toronto, Canada Correspondence and requests for reprints should be addressed to Richard L. Horner, Ph.D., Room 6368, Medical Sciences Building, 1 Kings College Circle, Toronto, ON, Canada, M5S 1A8. E-mail: richard.horner{at}utoronto.ca CONSEQUENCES OF INTERMITTENT HYPOXIA: ANIMAL MODELS There is epidemiologic evidence for a relationship between obstructive sleep apnea (OSA) and development of insulin resistance, but cause and effect had not been demonstrated largely because of the confounding effects of obesity in human populations. Accordingly, in a mouse model of OSA, Iiyori and colleagues (1) determined whether application of intermittent hypoxia in lean mice caused a reduction in insulin sensitivity. Mice were exposed to recurrent cycles of intermittent hypoxia (nadir FIO2 = 5–6%, 60 cycles/h for 9 h) or intermittent air as controls. The mice were also instrumented for hyperinsulinemic–euglycemic clamps and determination of whole-body insulin sensitivity. To determine if any changes in insulin sensitivity were mediated by autonomic nervous system activation, studies were also repeated after autonomic nervous system blockade using hexamethonium. The results showed that intermittent hypoxia caused a 21% reduction in insulin sensitivity that was not affected by hexamethonium. This was the first demonstration of a direct cause-and-effect relationship between intermittent hypoxia and development of acute insulin resistance in otherwise lean healthy animals. This fundamental determination of the endocrine consequences of stimuli associated with OSA is analogous to the breakthrough observations of the relationship between OSA and hypertension that were of subsequent major impact on the field. In addition to the effects on blood pressure, OSA is linked to other adverse cardiovascular events in humans, such as angina, myocardial infarction, stroke, and congestive heart failure. Although there may be a cause-and-effect relationship between intermittent hypoxia and development of atherosclerosis, however, this had not been demonstrated. Accordingly, Savransky and colleagues (2) determined whether mice fed regular chow or a high-cholesterol diet showed differences in the development of atherosclerotic lesions after 12 weeks of intermittent hypoxia (nadir FIO2 = 5–6%, 60 cycles/h for 12 h). The main finding was that chronic intermittent hypoxia caused atherosclerosis and dyslipidemia in the mice fed the high-cholesterol diet. These data are important because they are the first to indicate a direct cause-and-effect relationship between intermittent hypoxia and development of atherosclerosis in an animal model that is normally resistant to atherosclerosis in the absence of hypoxia. PEDIATRIC SLEEP APNEA Children with OSA have abnormally high levels of C-reactive protein, an inflammatory risk factor associated with cardiovascular disease. Sleep-disordered breathing in children is also associated with cognitive deficits and behavioral abnormalities, but this does not occur in all children and is not fully accounted for by OSA severity. Gozal and colleagues (3) hypothesized that C-reactive protein levels may identify children with OSA who have high susceptibility for cognitive abnormalities. In a community-based study of snoring and nonsnoring school-aged children (5–7 yr, n = 278 subjects), C-reactive protein levels were measured in the morning after full polysomnography, and a battery of neurocognitive tests were also performed. C-reactive protein levels were nearly double in the children with OSA compared with control subjects (0.36 ± 0.11 vs. 0.19 ± 0.07 mg/dl). Importantly, these levels were further elevated in the children with OSA who also exhibited neurocognitive deficits compared with the children with OSA who had normal cognitive assessments, again with the difference being approximately twofold (0.48 ± 0.12 vs. 0.21 ± 0.08 mg/dl). Although cause and effect were not proven in this study, the results highlight that the level of systemic inflammation in pediatric sleep-disordered breathing, as indicated by C-reactive protein, is associated with adverse neurobehavioral outcomes. The effects of OSA on behavioral outcomes may relate to recent findings that exposure to long-term intermittent hypoxia in a mouse model of OSA causes irreversible impairments in the neuronal systems regulating wakefulness, especially selective loss of catecholaminergic arousal neurons via NADPH oxidase–mediated neuronal injury (4). The prevalence of OSA is increasing in the general population, and this trend is occurring in parallel with an increasing incidence of obesity, especially among the young. After a review of the literature, Ievers-Landis and Redline (5) concluded that obesity is modestly associated with OSA in young children but more strongly associated in older children and adolescents. They also concluded that the rising incidence of pediatric obesity will likely impact on the efficacy of adenotonsillectomy, the most common treatment for childhood OSA, as well as the prevalence of adverse outcomes such as systemic inflammation, vascular disease, neurobehavioral abnormalities, and diabetes via the mechanisms discussed above (1–4). Central congenital hypoventilation syndrome (CCHS) is characterized by autonomic nervous system disturbances and sleep-disordered breathing, as well as greatly diminished ventilatory responses to hypercapnia and respiratory sensation. However, whether the sensory and/or motor components of the cough reflex and sensations associated with the urge to cough are also impaired in patients with CCHS has been little studied. Lavorini and colleagues (6) recruited seven children with CCHS (age range, 9–16 yr) and seven age- and sex-matched control subjects, and used ultrasonically nebulized distilled water to induce cough. All the control subjects and six of the seven patients with CCHS coughed normally in response to the nebulized distilled water. The latency and threshold to cough were almost identical between groups. In addition, the peak expiratory flows and integrated abdominal muscle activity recorded during cough were also similar, and corresponded to 80% of those recorded during voluntary cough (i.e., indicating no impairments in the motor components of the cough reflex). A major difference between groups, however, was that although all of the control subjects reported the sensation of an urge to cough at the threshold level for the cough reflex, only one patient with CCHS reported the perception of an urge to cough in response to the ultrasonically nebulized distilled water, and this sensation was "low intensity" and was notably absent in all the other subjects with CCHS. Moreover, the immediate precough period was associated with increased tidal volume in the control subjects that did not occur, or were trivial, in the subjects with CCHS. These interesting findings show that cough-promoting stimuli elicit minimal sensation and alterations in breathing pattern in patients CCHS, suggesting a deficit in afferent processing of the stimuli that may increase the risk of aspiration.
Although OSA has been implicated as a risk factor for metabolic syndrome in adults, its relationship to metabolic syndrome in adolescents has not been examined. To examine for such a relationship, Redline and colleagues (7) performed sleep studies, anthropometric measurements, and bioassays in 270 adolescents with a mean age of 13.6 years. Metabolic syndrome was identified if threshold levels were exceeded in three of five areas: waist circumference, blood pressure, triglyceride level, high-density lipoprotein cholesterol level, and glucose levels. Seventy percent of children with OSA (apnea–hypopnea index [AHI]
Other aspects of health may also be adversely affected by OSA in children. For example, during the year before OSA is diagnosed in preschool children, health care use is higher. To explore morbidity and health care use among children with OSA, Tarasiuk and associates (8) monitored children from birth until the diagnosis of OSA, and compared health care utilization with healthy control subjects matched for age, sex, primary care physician, and geographic location. They found that the 156 patients diagnosed with OSA between the ages of 3 and 5 years had 40% more (P = 0.048) hospital visits, 20% more repeated ( VENTILATORY AND ADAPTIVE RESPONSES TO HYPERCAPNIA AND HYPOXIA Brainstem CO2 chemosensitive neurons are activated by increased intracellular H+ levels, which are then extruded by Na+/H+ exchangers (NHEs). Of the nine subtypes of NHEs currently identified, NHE3 is expressed in regions of brainstem containing respiratory neurons. Kiwull-Schöne and colleagues (9) determined the effect of chronic acid-base disturbances on NH3 mRNA levels in the obex region of the brainstem in rabbits, and also measured the levels of ventilation in the same animals. The rabbits were exposed to prolonged hypercapnia (6% FICO2 for 3 d) to produce respiratory acidosis, or provided with a nonalkogenic feed combined with ammonium chloride in the drinking water to produce metabolic acidosis. Brainstem NHE3 mRNA levels and ventilation were both unchanged after chronic respiratory acidosis, but chronic metabolic acidosis led to a 2.5-fold increase in brainstem NHE3 mRNA and reduced alveolar ventilation by approximately 25%. That these changes in NHE3 occurred in conditions of chronic metabolic acidosis but not prolonged hypercapnia led the authors to suggest that the elevated brainstem NHE3 expression effectively limits the maladaptive hyperventilation during metabolic acidosis. There is also evidence that the neural and ventilatory responses to hypercapnia are modulated by reactive oxygen species. Zakynthinos and colleagues (10) therefore performed a randomized, double-blind, placebo-controlled study to determine if antioxidant treatment (vitamins E, A, and C for 2 mo, allopurinol for 15 d, and N-acetylcysteine for 3 d) affected the ventilatory response to hyperoxic hypercapnia in healthy humans, with and without resistive loading. Antioxidant treatment had no effect on resting breathing pattern, but there was an almost doubling of the ventilatory responses to hypercapnia (group mean, 1.7–3.2 L/min/mm Hg) that persisted for up to 2 hours after resistive loading. The authors implicate a role of oxidative stress in carbon dioxide chemoreception during hyperoxia in normal subjects, and suggest that antioxidant treatment may be beneficial to minimize hypoventilation in conditions associated with increased respiratory load, such as chronic obstructive pulmonary disease combined with oxygen therapy. Individuals who normally reside at sea level and ascend to altitude without adequate acclimatization are susceptible to develop acute mountain sickness, with symptoms such as insomnia, headache, nausea, and fatigue. The carbonic anhydrase inhibitor acetazolamide is the first-line pharmacologic treatment for this condition, in which the symptomatic improvements are believed to be due to stimulation of ventilation and increased cerebral oxygenation. However, the effects of acetazolamide on the cerebral blood flow response to hypoxia have not been clearly defined and so this was the focus of a study by Teppema and colleagues (11). In addition, because high-altitude pulmonary edema is also a serious consequence of rapid ascent to altitude, these authors also addressed the effects of acetazolamide on hypoxic pulmonary vasoconstriction. A randomized, double-blind, placebo-controlled study was performed in human subjects to determine if acetazolamine (250 mg taken every 8 h for 3 d) altered the ventilatory, pulmonary vascular resistance, and cerebral blood flow responses to hypoxia. Acetazolamide increased baseline ventilation but did not affect the ventilatory responses to isocapnic hypoxia. Importantly, however, acetazolamide caused a large (34–57%) reduction in the rise of pulmonary vascular resistance in both short- (20 m) and long-duration (4 h) hypoxic exposures. Normal resting cerebral blood flow was not different between acetazolamide and placebo, and the cerebral blood flow responses to both the short- and long-duration hypoxic exposures were unaltered. These findings are important because they show that clinically relevant doses of acetazolamide cause a large reduction in the pulmonary vascular resistance response to short- and long-term hypoxia, a finding that may be relevant to prevention or treatment of high-altitude pulmonary edema. Epidemiologic studies show that OSA is an independent risk factor for development of stroke, but the potential mechanisms underlying this association are not yet defined. In this context, Foster and colleagues (12) hypothesized that the cerebral blood flow responses to hypoxia are reduced in patients with OSA, and that this deficit would be normalized by treatment with nasal continuous positive airway pressure (CPAP). Measurements of the cerebral blood flow response to 20 minutes of isocapnic hypoxia were performed in eight awake male patients with OSA before and after 4 to 6 weeks of CPAP treatment. The cerebral blood flow response to hypoxia was markedly reduced (by 43%) in the patients with untreated OSA compared with age- and weight-matched control subjects, but this deficit was normalized after CPAP. Correlational analyses also suggested that patients with the most severe OSA had the lowest initial cerebral blood flow response to hypoxia, and that these individuals with the most severe OSA had the largest increase in cerebral blood flow response to hypoxia after CPAP. These observations led the authors to speculate that the blunted cerebrovascular vasodilator response to hypoxia in patients with OSA may contribute to the increased risk of stroke in untreated OSA. VASCULAR AND CEREBROVASCULAR DISEASE IN SLEEP APNEA There is increasing evidence that OSA increases the risk for various cardiovascular diseases. One pathway through which OSA contributes to this increased risk is metabolic syndrome. McArdle and colleagues (13) examined metabolic factors in 42 patients with OSA and 42 control subjects matched for age, body mass index, and current smoking status. They were free of diabetes, clinically demonstrable cardiovascular disease, marked hypertension, and dyslipidemia. Compared with control subjects, patients with OSA had a higher median (interquartile range) homeostasis model assessment (HOMA) score for insulin resistance (OSA, 1.7 [0.8–4.1]; control, 1.0 [0.7–1.8] mU · mmol/L; P = 0.02), total cholesterol (OSA, 5.6 [4.8–6.2]; control, 4.8 [4.3–5.4] mmol/L; P = 0.049), and low-density lipoprotein cholesterol (OSA, 3.8 [2.9–4.2]; control, 3.1 [2.6–3.6] mmol/L; P = 0.04). Multiple linear regression, adjusting for central obesity, age, and alcohol consumption, confirmed an independent association between OSA and metabolic risks (all, P < 0.05). These findings indicate that patients with OSA have increased insulin resistance and other metabolic changes independent of confounding factors that could contribute to the previously reported increased risk of cardiovascular disease. This study, together with the two articles discussed above (1, 7), also provides strong evidence of an association between OSA and dysglycemia that is present in both children and adults. Endothelial injury or dysfunction is another potential mechanism through which OSA might contribute to the development of cardiovascular diseases, especially atherosclerosis. Accordingly, El Solh and coworkers (14) examined the relationship between circulating apoptotic vascular endothelial cells (CD146+ annexin V+) and vasomotor function in 14 subjects with OSA and 10 healthy control subjects. Brachial artery flow–mediated dilation and the concentration of circulating apoptotic endothelial cells, measured via flow cytometry, were determined at baseline and 8 weeks after CPAP therapy. Patients with OSA had higher numbers of circulating CD146+ annexin V+ cells than the healthy subjects (39.2 ± 13.6 vs. 17.8 ± 9.4 cells/ml, respectively; P < 0.001). The numbers of CD146 annexin V+ cells correlated directly with the AHI (r = 0.56, P = 0.004), and inversely with vascular function (r = –0.61, P = 0.001). After 8 weeks of treatment with CPAP, the numbers of circulating apoptotic endothelial cells were reduced significantly from 39.2 ± 13.6 to 22.3 ± 12.9 per milliliter (P < 0.001). This reduction correlated with improvement in endothelium-dependent vasodilation (r = 0.49, P = 0.07). These findings suggest that endothelial dysfunction, manifest by increased circulating apoptotic endothelial cells and impaired flow-mediated vasodilation, is an additional mechanism through which OSA may predispose to premature vascular disease and appears to be at least partially reversible by CPAP therapy. More direct evidence that OSA can contribute to the development of atherosclerosis comes from the work of Drager and associates (15). In a previous study, these investigators had shown that patients with OSA have increased subclinical signs of early atherosclerosis (i.e., increased carotid intima-medial thickness and arterial pulse-wave velocity) compared with healthy subjects. In this study, they tested the hypothesis that treating OSA could alleviate signs of early atherosclerosis. To this end, they randomly assigned 24 patients with severe OSA who were free of cardiovascular comorbidities to receive no treatment (control, n = 12) or CPAP (n = 12) for 4 months. Carotid intima-medial thickness and arterial stiffness (evaluated by pulse-wave velocity) were evaluated at baseline and after 4 months. In contrast to the control group, in whom no changes in these variables occurred, the CPAP-treated group experienced a significant decrease in carotid intima-medial thickness (707 ± 105 vs. 645 ± 95 µm, P = 0.04) and pulse-wave velocity (10.4 ± 1.0 vs. 9.3 ± 0.9 m/s, P < 0.001) after 4 months of CPAP. The reduction in carotid intima-medial thickness was similar in degree to that induced by lipid-lowering agents shown to reduce the risk of myocardial infarction and strokes. Accordingly, these findings are very important because they are the first to demonstrate that early signs of atherosclerosis in patients with OSA can be reversed by CPAP. They further support the concept that OSA is an independent risk factor for atherosclerosis, and raise the possibility that treating OSA might reduce cardiovascular risk.
Minoguchi and colleagues (16) also provided evidence that OSA can contribute to cerebrovascular disease. Because silent brain infarction (SBI) on magnetic resonance imaging (MRI) as well as increased levels of soluble CD40 ligand (sCD40L) and soluble P-selectin (sP-selectin) are associated with an increased risk of cerebrovascular disease, they assessed these variables in 50 male patients with OSA and 15 obese male control subjects. They found that the percentage of patients with moderate to severe OSA (AHI In a prospective observational study, Buchner and associates (17) investigated the incidence of nonfatal (myocardial infarction, stroke, and acute coronary syndrome requiring revascularization procedures) and fatal (death from myocardial infarction or stroke) cardiovascular events in patients with treated and untreated OSA. Of 449 patients enrolled, 364 patients received OSA treatment (positive airway pressure or an oral appliance) and 85 patients remained untreated during a median follow-up period of 72.0 months. For the whole group, events were more frequent in untreated than in treated patients (24 [28.3%] in untreated vs. 52 [14.3%] in treated patients, P = 0.009; estimated event-free survival at 10 yr, 51.8 vs. 79.7%, log-rank test: P < 0.001; absolute risk reduction, 27.9%). Similar results were observed in the subset of patients with mild–moderate OSA (AHI, 5–15; n = 288). After adjustment for age, sex, cardiovascular risk factors, and comorbidities at baseline, OSA treatment was an independent predictor of reduced risk for events (hazard ratio, 0.36; 95% confidence interval, 0.21–0.62; P < 0.001). The authors concluded that OSA treatment should be considered for primary and secondary cardiovascular event prevention, even in milder OSA. However, because this was a nonrandomized observational trial, these results must be interpreted with caution. Because treatment of OSA by a mandibular advancement splint (MAS) is effective in some, but not all, patients, there is a need for a clinical tool to assist in predicting treatment responses. Zeng and colleagues (18) hypothesized that abnormalities of flow–volume curves would help to predict MAS treatment outcome. In 54 patients with OSA, they performed expiratory and inspiratory flow–volume curves in the erect and supine positions to derive midinspiratory flow (MIF50) and the ratio of expiratory to inspiratory flow at 50% of vital capacity (MEF50:MIF50). They found that in patients in whom the MAS alleviated OSA (responders), the MIF50 was lower (6.04 ± 1.80 vs. 6.88 ± 1.08 L/s, P = 0.035) and the MEF50:MIF50 ratio was higher (0.82 ± 0.23 vs. 0.61 ± 0.15, P = 0.001) than in those in whom OSA was not alleviated (nonresponders). Logistic regression analysis revealed that the MEF50:MIF50 ratio was the most important predictive factor for MAS treatment outcome, but that body mass index, age, and baseline AHI were also contributory. These findings suggest that a combination of flow–volume curve data, body mass index, age, and baseline AHI may be useful in predicting response to MAS. Because CPAP is not accepted by all patients with OSA, new approaches to treatment are needed. McGinley and colleagues (19) tested the effect of nasal insufflation in 11 patients with OSA. They found that administration of warm and humidified air at 20 L/minute through an open nasal cannula reduced the AHI from 28 ± 5 to 10 ± 3 events per hour (P < 0.01). The mechanism of action appeared to be through an increase in end-expiratory pharyngeal pressure, which alleviated upper airway obstruction and improved ventilation. These findings provide proof of concept that nasal insufflation at a high flow rate can alleviate OSA in some patients. Randomized trials will now be required to determine in which patients nasal insufflation may be most effective, whether it can be tolerated over long periods of time, and how it compares with other treatments. The clearest indication to treat OSA is a symptom of excessive daytime sleepiness. However, many patients with OSA do not have this symptom, and therefore lack the usual indication to treat. This presents a dilemma to physicians because it is unclear whether subjects found to have OSA on a sleep study, even if the AHI is high, should be treated. Montserrat and his associates (20) therefore reviewed the diagnostic and therapeutic approach to the nonsleepy individuals with sleep apnea. Some studies have shown that treating asymptomatic patients with OSA does not improve symptoms, quality of life, neurocognitive function, or blood pressure. This would suggest that the nonsleepy patient with OSA need not be treated. On the other hand, some studies have shown that in patients with coexisting cardiovascular diseases, treating nonsleepy patients with OSA can improve cardiovascular function, but there is no evidence that this leads to improvements in morbidity or mortality. The main issue is whether treating nonsleepy patients with OSA would lower the risk of developing cardiovascular diseases, because this would appear to be the main risk associated with untreated OSA in such patients. The authors conclude that there is therefore a need for well-designed randomized trials to determine whether treating nonsleepy patients with OSA prevents cardiovascular diseases or, in patients with coexisting cardiovascular diseases, whether treating OSA reduces morbidity and mortality. In the main analysis of the Canadian Continuous Positive Airway Pressure for Patients with Central Sleep Apnea and Heart Failure Trial (CANPAP), CPAP had no effect on heart transplant–free survival; however, CPAP only reduced the mean AHI to 19 events per hour of sleep, which remained above the trial inclusion threshold of 15. Arzt and colleagues (21) therefore undertook a stratified analysis of CANPAP to test the hypothesis that suppression of central sleep apnea (CSA) below this threshold by CPAP would improve heart transplant–free survival. Of the 258 heart failure patients with CSA in CANPAP, 110 in the control group and 100 in the CPAP-treated group had sleep studies 3 months later. CPAP patients were divided post hoc into those whose AHI was or was not reduced below 15 at this time (CSA suppressed, n = 57, and CSA unsuppressed, n = 43, respectively). Their heart transplant–free survivals were compared with those in the control group. Despite similar CPAP pressure and hours of use in the two groups, CSA-suppressed subjects experienced significantly better transplant-free survival (hazard ratio, 0.371; 95% confidence interval, 0.142–0.967; P = 0.043) than control subjects, whereas the CSA-unsuppressed group did not (hazard ratio, 1.463; 95% confidence interval, 0.751–2.850; P = 0.260). These results suggest that, in patients with heart failure, CPAP might improve transplant-free survival if CSA is suppressed soon after its initiation. These findings provide a strong rationale to test the effects on morbidity and mortality of newer positive airway pressure devices specifically designed to eliminate CSA in patients with heart failure. FOOTNOTES Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form January 9, 2008; accepted in final form January 9, 2008 REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
5) were overweight and 59% had metabolic syndrome. After adjusting for confounders, children with OSA had a 6.49 (95% confidence interval, 2.52–16.70) increased odds of metabolic syndrome compared with children without OSA. Indices of OSA associated with metabolic syndrome included AHI, degree of oxygen desaturation, and sleep efficiency. After adjustment for body mass index, OSA was associated with systolic and diastolic blood pressure, low-density lipoprotein cholesterol, and fasting insulin levels. These findings confirmed that OSA is strongly associated with metabolic syndrome in adolescents, just as it is in adults. Accordingly, there may be a need to develop screening, prevention, and treatment strategies for both disorders in high-risk, overweight adolescents. These data are also in accordance with the findings of Iiyori and coworkers (