Sleep Apnea in Renal Patients

In Western societies, sleep apnea (SA) has emerged as a major medical problem with important social (1) and financial implications. The issue may be even more relevant in nephrology because some of the factors involved in the pathogenesis of renal disease are the same that cause, or are associated with, SA in the general population. The interest in this syndrome is continuously increasing in the renal community, but many nephrologists do not fully appreciate its potential clinical implications. Before reviewing in some detail the particular situation of SA in chronic renal diseases, we therefore briefly review definitions, general pathophysiology, and types of SA first (2). Definitions and Pathomechanisms SA is defined as an intermittent interruption of airflow at the level of nose and mouth during sleep. Episodes of apnea are considered important if they persist for longer than 10 s, but in some cases they may last as long as 2 min. The SA syndrome is the clinical consequence of frequent episodes of apnea during sleep. Usually patients with the full-blown syndrome have at least ten episodes of apnea per h. The syndrome is probably the most important cause of daytime somnolence. Its prevalence is at least 2% in middle-age women and 4% in middle-aged men. The following types of apnea can be distinguished: (1) obstructive, (2) central (i.e. neurally mediated), and (3) mixed. Obstructive SA (OSA) is characterized by a cessation of airflow caused by occlusion of the oropharyngeal tract and central SA (CSA) by a transient abolition of the neural drive to respiratory muscles. Mixed apnea represents a combination of the two forms. Obstructive Sleep Apnea OSA typically occurs in males in an age range between 30 and 60 y and is characterized by a history of snoring, excessive daytime sleepiness, noctural choking or gasping, and moderate obesity. The underlying disturbance in OSA is transient airway occlusion, usually at the level of the oropharynx. The airway occlusion provokes progressive asphyxia, which in turn elicits a brief arousal reaction restoring airway patency. This phenomenon occurs in a cyclic fashion, sometimes up to 500 times per night, and causes sleep fragmentation. The airway collapse in OSA depends on the inability of the airway dilator and abductor muscles to maintain airway patency during inspiration, i.e. when subatmospheric pressure is generated inside the respiratory system. By reducing muscular tone of the upper airways, sleep has a permissive role on airway collapse. Alcohol may be an important contributory factor. Structural abnormalities, such as macroglossia, retrognathia, and adenotonsillar hypertrophy, may generate as well as aggravate OSA. By increasing the fat content of the pharynx, obesity is another frequent condition contributing to OSA. Snoring, i.e. high frequency vibration of the palatal and pharyngeal soft tissues that results from the reduced size of the upper airway lumen, may further exacerbate the process by inducing soft tissue edema. Snoring usually precedes by several years the development of full-blown OSA, but only a minority of snoring patients have OSA. Recurrent episodes of nocturnal hypoxemia and arousal eventually lead to the full-blown OSA syndrome, which is characterized by additional neuropsychiatric and behavioral disturbances. These include excessive and progressive daytime sleepiness which often reaches a disabling degree as well as memory and personality disturbances. In men, impotence is a relatively frequent complaint. OSA induces also important cardiorespiratory complications, which are considered a consequence of the recurrent episodes of nocturnal hypoxemia. Cyclical bradycardia during the apneic episodes, followed by tachycardia during the ensuing ventilatory phases, is frequently found. In a minority of cases, bradycardia may end up in prolonged asystolic periods or in various forms of tachyarrhythmia, including ventricular tachycardia. Arrhythmias in OSA may cause sudden death during sleep. Arterial pressure rises abruptly at the termination of each obstructive event. OSA is typically associated with nocturnal hypertension and with an inverted 24-h arterial pressure profile. OSA is a recognized risk factor for systemic hypertension (3,4) and perhaps also for myocardial infarction, stroke, and death (5–7). It is important to bear in mind that obstructive SA may precipitate left ventricular failure in patients with heart disease. Increased afterload during obstructive events, nocturnal hypoxemia, and the ensuing elevated sympathoadrenal activity all conjure to impair left ventricular function (8). On the other hand, a minority of patients with severe OSA, usually obese patients with sustained daytime hypoxemia, develops frank pulmonary hypertension, right ventricular failure, polycythemia, and chronic hypercapnia and hypoxemia. Upper airway collapse may also induce central sleep apnea by eliciting reflexes that inhibit the respiratory drive. The cardiovascular consequences of OSA will be discussed further in the paragraph dealing with renal diseases. Central Sleep Apnea In contrast to OSA, the apneic events in CSA are associated with a decreased or absent ventilatory effort. CSA depends on a defective central drive to the ventilatory muscles. The resulting apnea has pathophysiologic consequences that are similar to those of OSA. Normally, the rate and depth of breathing are regulated by a negative feedback system that maintains the partial pressure of arterial CO2 (pCO2) within a narrow range. Changes in pCO2 lead to changes in ventilation: the greater the sensitivity to CO2, the greater the ventilatory response. Among normal persons, there is considerable variation in the sensitivity to CO2, which may in part be related to genetic influences. Diminished sensitivity to CO2 increases the risk of chronic hypercapnia in patients with established pulmonary disease. Conversely, an increase in CO2 sensitivity minimizes perturbations in the partial pressure of arterial CO2. In theory, this should offer protection against the long-term sequelae of hypercapnia in patients with cardiopulmonary disease. Although this protective mechanism is advantageous during waking hours, the increased sensitivity may destabilize breathing during sleep when ventilation decreases and the partial pressure of CO2 rises (approximately 5 mmHg). In persons with increased sensitivity to CO2, the negative feedback system that controls breathing elicits a large ventilatory response when the pCO2 rises; the consequent hyperventilation, by driving the pCO2 below a certain level (the apneic threshold) then results in central apnea. As a consequence of apnea, the pCO2 rises again, which leads to an increase in ventilation. In this fashion, cycles of central apnea and hyperventilation recur during sleep (Cheyne-Stokes respiration). Patients with hyperventilation (driven by cardiopulmonary disease, metabolic acidosis, fever, and other causes) often develop CSA because during the awake state hyperventilation reduces pCO2 below the critical value, which elicits CSA. Heart failure (see below) is frequently associated with CSA (8). Primary hypoventilatory syndromes that are caused by altered metabolic control of respiration or disease of respiratory muscles are a rare cause of CSA. Diagnosis The diagnostic criterion standard for SA and for differentiating OSA from CSA is polysomnography, i.e. an overnight sleep study that includes (1) neurophysiologic variables (EEG, electrooculogram, and submental myogram) that allow a clear identification of sleep stages, (2) measurements of respiratory effort (respiratory inductive plethysmography or esophageal pressure measurements) that are aimed at detecting and classifying apnea as of central (CSA) or obstructive (OSA) origin, (3) arterial O2 saturation by pulse oxymetry, (4) heart rate, and (5) transcutaneous pCO2. The main informations derived from polysomnography are schematically summarized in Table 1. The diagnosis of OSA is established when episodes of airflow cessation at the nose and mouth (at least ten episodes of apnea per h) are documented despite simultaneous evidence of continuing respiratory effort. Because of the cost of polysomnography, there is considerable interest in simplified, unattended, ambulatory sleep monitoring for home studies. The most useful test is the recording of arterial O2 saturation by oxymetry. The reliability of overnight oxymetry in the diagnosis of OSA depends on the pretest probability of the disorder. In patients with a high pretest probability (based on clinical symptoms and witnessed episodes of apnea during sleep), overnight oxymetry is of proven usefulness to confirm the diagnosis by documenting episodes of arterial O2 desaturation (at least ten episodes per h). However, negative results in those with a high clinical probability do not exclude the diagnosis, and polysomnography in these cases is mandatory to exclude OSA. When the pretest probability of OSA is low (in the occasional snorer with rare episodes of daytime disturbance), the absence of arterial O2 saturation virtually excludes the diagnosis, rendering polysomnography unnecessary. Table 1: Main information derived from polysomnography (also see text)aThe key element for the diagnosis of CSA is the documentation of recurrent episodes of apnea (at least five per h) that are not accompanied by respiratory effort. Relation of Sleep Apnea to Hypertension, Heart Failure, and Renal Disease SA alters several mechanisms that regulate extracellular fluid volume and vascular tone. It is apparently also related to systemic hypertension. Furthermore, SA can be a complication of advanced renal insufficiency. As alluded to above, SA may be of particular relevance in heart failure because it aggravates ventricular afterload by direct and indirect mechanisms. Recognizing the link between SA and cardiorenal physiology is important because it has implications for the prognosis and the treatment of patients with primary and secondary forms of the syndrome. Body Fluid Homeostasis By profoundly altering cardiopulmonary dynamics, SA induces reflex circulatory responses that critically affect renal function and body fluid volume homeostasis. Nocturnal secretion of atrial natriuretic peptide (ANP) is increased and renin secretion decreased, suggesting increased cardiac preload as in a state of hypervolemia (19). When patients are awake, however, hypervolemia is not demonstrable, indicating that apneic episodes promote a volume shift from the peripheral to the central circulation, i.e. central hypervolemia. Hypertension and Heart Failure There are important links between OSA or CSA and hypertension or heart failure, respectively. OSA is associated with hypertension and CSA with heart failure. These two conditions are challenging problems in patients with end-stage renal disease. It is well documented that BP rises in a very consistent fashion during apneic episodes. The mechanisms responsible for this phenomenon are complex because the direct effects of apnea (hypoxemia and low intrathoracic pressure) are modified by cardiopulmonary reflexes. Undoubtedly, the rapid increase in arterial pressure that occurs at the end of an apneic episode is mainly mediated by surges in sympathetic function during the arousal reaction. Whether periodic nocturnal hypoexmia induces sustained hypertension during daytime as well is somewhat controversial. The problem has considerable epidemiologic relevance given the high frequency of OSA in the adult population. To clarify the relationship between OSA and hypertension, it is necessary to address the following issues. What is the nature of this relationship; is the risk of hypertension in patients with OSA truly independent of other risk factors? Is it possible to reverse hypertension by specifically treating SA with continuous positive airway pressure (CPAP)? As to the first question, age, gender, body mass index, and alcohol as well as tobacco consumption are major confounders of the relationship between SA and BP. In some studies, the link was markedly attenuated when these factors were appropriately accounted for (9). Furthermore, in some early studies, the techniques applied to diagnose SA, to measure arterial pressure, and to characterize patients were not adequate. Recent large-scale surveys and cohort studies have produced a convincing answer that is based on solid scientific evidence. The Wisconsin Sleep Cohort Study (3) included 709 individuals of both sexes and was based on polysomnography and well-standardized arterial pressure measurements. This study showed that there was a dose response relationship between sleep-disordered breathing at baseline and BP at follow-up. Importantly, this response was independent of known confounding factors such as obesity and alcohol and tobacco consumption. The contribution of SA to hypertension in this cohort was weak (odds ratios for hypertension at the 4-y follow-up: 1.4 [no apneic events] to 2.9 [>15 episodes of hypopnea or apnea per h]). Similar observations were reported in a large cross-sectional analysis of 6132 subjects enrolled in the Sleep Heart Study (4). In cross-sectional studies that are designed to investigate whether SA is related to increased mortality, bias will lead to underestimation of the true relative risk of hypertension because survival of subjects with disordered breathing during sleep is less (prevalence/incidence bias). As to the second question there is little doubt that CPAP virtually eradicates cyclic BP surges. However, the chronic effects of CPAP on daytime hypertension remain uncertain because most studies addressing this problem suffer from the methodological flaws discussed above. Further shortcomings are low compliance and/or short follow-up. A recent study by Dimsdale et al. (10) examined the BP effect of CPAP treatment for 1 wk in 39 patients with obstructive SA. The strength of this study is that it is the only one that includes a placebo arm, i.e. CPAP administered at an ineffective pressure. Nighttime mean arterial pressure levels decreased to a much greater extent over time in the patients who received active CPAP treatment. However, the daytime decrease was not significantly greater in the active treatment group than in the placebo group, suggesting that the response was in part nonspecific. The very short treatment period and the tendency for BP to decrease toward the end of the active treatment period are obvious limitations of this study. There is obviously a need for properly designed investigations to estimate the true effects of CPEP on arterial pressure. Despite the methodological difficulties encountered in this area, there is agreement that hypertensive patients with apneic episodes are a highly heterogenous population. Some of them retain normal nocturnal BP dipping, and others develop frank nocturnal hypertension. These nondippers may be at higher cardiovascular risk. At variance with systemic hypertension, which is typically associated with OSA, patients with heart failure display a high prevalence of CSA (up to 40%). Low cardiac output causes respiratory instability because it prolongs the time lag between changes in blood chemistry induced by ventilation (pO2, pCO2, and pH) on the one hand and detection of these changes by the chemoreceptors on the other hand. Consequently, the ventilatory drive remains inappropriately high because pCO2 in the central nervous system lags behind pCO2 in the circulation. When pCO2 is reduced below the apneic threshold, respiration ceases. It starts again as CO2 accumulates in blood, thus generating periodic breathing. Individual sensitivity to CO2 plays an important role in precipitating CSA in patients with heart failure because CSA episodes are far more common in patients with relatively low CO2 sensitivity (11). As mentioned above, patients with OSA may develop heart failure because of the detrimental effects of episodes of obstructive apnea on myocardial performance. In these cases, CPAP, by alleviating OSA, reduces left ventricular afterload and improves arterial oxygenation during sleep. On this basis, CPAP has been proposed as a non-pharmacologic adjunct for reducing afterload during nighttime in patients who are on treatment for heart failure. When heart failure is associated with reduces the of apneic episodes and may an or to the of Sleep Apnea in Renal Disease sleep is common in to of chronic patients of sleep and reduced daytime Sleep is a problem of that is by In time in to per for induces in metabolic and including increased sympathetic tone and a state of are well known of chronic There are only a studies, i.e. based on in the sleep which documented the of sleep apnea in patients Furthermore, these studies were in patients who of problems and are therefore not of the true prevalence of this disturbance in the population. The reported prevalence rate in cases from to A very recent estimate that the prevalence may be Because sleep apnea is common in the to have a high degree of in patients of the symptoms and/or the in Table Table and that to sleep high frequency of SA in renal failure is in part by the that the most common conditions of and are also associated with this syndrome. Although apnea in the general is of the obstructive the obstructive (OSA) and the central (CSA) types are frequent in patients with patients with heart failure are to a central SA Despite the confounding effect of cardiovascular disease, there is little doubt that per is associated with SA and that this disturbance plays a major role in sleep in observations that SA is at least renal is convincing that SA is a direct consequence of renal failure The factors responsible for SA in are metabolic on an important for respiration by inducing a in pCO2. However, in the study by et al. relationship was between metabolic and apneic episodes in The pCO2 level below which the breathing i.e. the apneic threshold, increases during sleep. It has been that this is increased in chronic which increase the risk of CSA. To pCO2 measurements during sleep have not been reported in chronic renal failure. is another but an by et al. that of has effect on sleep apnea. The has been advanced that of destabilize but have not been Central may in airway tone during sleep or destabilize respiratory but again the issue remains a of The level of several which may is elevated in patients As discussed sleep apnea on the of by the sleep. important it should not be by the is its with various cardiovascular complications, from cardiac left ventricular hypertrophy, heart failure, and to cardiorespiratory It is to that SA to cardiovascular and in these but solid evidence has been in of this In a recent study in a group of patients with sleep periodic than sleep apnea death However, the of this study is because of the of the that they were and the failure to control for confounding The issue is a one because SA may be a for sequelae of sleep apnea may be caused by direct sympathetic secondary to by hypoxemia and hypercapnia and from chronic of sleep. is and increased sympathetic activity is a response that is aimed at by increasing cardiac blood and of O2 to the causes sympathetic that the et al. that and is high in patients with SA, and this was true not only during nighttime but also during In the general the role of sympathetic as a cardiovascular risk factor involved in the of hypertension and cardiac is well established Increased sympathetic activity the of SA with which is also frequent in chronic are another of SA because the of a which may induce cardiac and vascular and impair the of a and vascular is defective in hypertensive patients with obstructive sleep and this is independent of hypertension The link between obstructive SA and may in part the in The evidence that SA induces cardiovascular in is It has been that nocturnal hypoxemia is associated with nocturnal hypertension in patients It is even more associated with cardiac in SA is associated with control of the cardiovascular system and this true also in patients as well and studies are to test the that SA and the nocturnal hypoxemia are related to the high cardiovascular risk in Although the diagnosis of SA studies, nocturnal pulse at home is a in patients with because of the high pretest probability of SA. A positive test most the of true SA, but a negative test has relatively low negative Nocturnal hypoxemia by ambulatory pulse oxymetry is with several and in patients home techniques for the diagnosis (4) and of SA are these and will allow diagnosis in many patients in it remain and The treatment of SA depends on whether one with OSA or CSA. The treatment of OSA should address the underlying In obese the patency of may be by if or should be of alcohol and may is useful in to moderate SA, but it has effects that range from excessive to There is in patients with OSA. In moderate to severe OSA, of soft and CPAP during sleep are increases the pharyngeal is reported in of the cases, but evidence on the of this in patients is not CPAP the pharyngeal airway by positive pressure a It is the most SA is in of the CPAP has been in patients and to be The effect of on SA has been in only patients and was in all patients with CSA usually to nocturnal CPAP is not only in OSA, but also in CSA. the increase in pCO2 by the in part the effect of this treatment in OSA. As discussed by et al. CPAP and are if CSA is secondary to heart failure.