Centennial Review

The Pathogenesis of Obstructive Sleep Apnea

Advances in the Past 100 Years
David P. White
Division of Sleep Medicine, Harvard Medical School, and Brigham and Womens Hospital, Boston, Massachusetts

Obstructive sleep apnea (OSA), in the medical scheme of things, is a relatively new disorder and has barely been in the medical literature for the last 100 years. Thus, virtually everything we know about its pathogenesis has been discovered over this time frame. Although Sidney Burwell is often credited with the rst description of a patient with OSA in an article in the American Journal of Medicine in 1956 (1), more careful review reveals this passage from a Lancet (2) paper of 1877 by a London physician, W. H. Broadbent:
When a person, especially advanced in years, is lying on his back in heavy sleep and snoring loudly, it very commonly happens that every now and then the inspiration fails to overcome the resistance in the pharynx of which stretor or snoring is the audible sign, and there will be perfect silence through two, three, or four respiratory periods, in which there are ineffectual chest movements; nally, air enters with a loud snort, after which there are several compensatory deep inspirations before the breathing settles down to its usual rhythm. In the case to which I allude there was something more than this. The snoring ceased at regular intervals, and the pause was so long as to excite attention, and indeed alarm.

Thus, at least a few physicians have been aware of this disorder for over 100 years and, as the passage above would suggest, recognized that it resulted from a sleep-induced obstruction of the upper airway.

SLEEP PHYSIOLOGY
It must be recognized that our increasing knowledge of sleep apnea and its pathogenesis did not occur in a vacuum, but evolved in association with a rapid growth in our knowledge about sleep itself. Thus, before addressing the specics of sleep apnea pathophysiology, a brief discussion of some of the major milestones in our understanding of sleep seems in order. However, it must be stated from the start that we still do not have a meaningful explanation for the actual function of sleep. Despite this limitation, considerable progress has been made. For instance it was just over 50 years ago that rapid eye movement (REM) sleep was rst recorded (3), with the relationship of these eye movements to body movements and dreaming being outlined a few years later (4). It was 30 years ago that Hobson and McCarley rst described their reciprocal interaction model

of non-REM (NREM) and REM sleep cycling (5), much of which is still accepted today. This model recognized that there was an oscillation in the ring of monoaninergic neurons (neurons in the locus coeruleus [noradrenergic neurons] and the dorsal raphe [serotonergic neurons]) that are active during wakefulness, less active during NREM sleep, and are silent during REM sleep, with pontine reticular neurons (cholinergic neurons in the pedunculopontine tegmentum and laterodorsal tegmentum) that are active during REM sleep. The concept was that the monoaminergic neurons (noradrenergic and serotonergic) inhibit the cholinergic neurons, thus maintaining NREM and suppressing REM sleep. As the ring frequency of the monoaminergic neurons decreases during NREM sleep, the cholinergic neurons begin ring, leading to the onset of REM sleep. Thus, a cycle is established that recurs throughout the night. This model was improved upon in 1996 when Sherin and Saper found and reported the ventrolateral preoptic (VLPO) neurons (6), which, increasing evidence suggests, may be the master switch that turns on the sleep process. The VLPO neurons, which are silent during wakefulness, begin ring at sleep onset, which initiates the inhibition of many of the known arousal mechanisms (noradrenergic, serotonergic, and histaminergic neurons) in the brainstem, allowing NREM sleep to emerge (see Figure 1). As these neurons progressively decrease their ring frequency, cholinergic neurons are no longer inhibited (see above), and REM sleep emerges again, establishing a cycle. Finally, although there is still considerable argument as to the molecule in the brain that accumulates during wakefulness, thus producing sleepiness or a drive to sleep, recent evidence suggests that this may be extracellular adenosine (7). As adenosine levels rise in certain areas of the brain (basal forebrain), wake-active neurons are inhibited, and the VLPO neurons described previously become activated. Thus, the sleep process begins, which cycles through NREM and REM episodes approximately every 90 min, based on the reciprocal interaction model previously described. Thus, state (sleep/wakefulness) can be regulated in a precise manner through the interactions between these neurons/neuromodulators. Although this is a gross simplication of the neurobiology of sleep, it does demonstrate some of the progress made in the last 100 years.

SLEEP APNEA PATHOGENESIS

As stated previously here, it has been known for some time that sleep apnea is characterized by recurrent collapse of the upper airway during sleep (8). However, why this occurs, why it only occurs in certain individuals, and why it only occurs during sleep is still only partially understood. That being said, considerable progress has been made over the last 30 years in our understanding of this disorder. Most evidence would suggest that a variety of traits can lead to OSA, and that the combination of traits leading to this disorder in any particular patient may vary

(Received in final form August 22, 2005) Correspondence and requests for reprints should be addressed to David P. White, M.D., Brigham and Womens Hospital, Division of Sleep Medicine, Gerald E. McGinnis Professor of Sleep Medicine, Sleep Research Program at BI, 75 Francis Street, Boston, MA 02115. E-mail: [email protected] Am J Respir Cell Mol Biol Vol 34. pp 16, 2006 DOI: 10.1165/rcmb.2005-0317OE Internet address: www.atsjournals.org

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 34

2006

Figure 1. Non-rapid eye movement (NREM) and rapid eye movement (REM) sleep pathways. Neurons of the ventrolateral preoptic area (VLPO) produce -aminobutyric acid (GABA) and galanin, and inhibit all the arousal systems during NREM sleep. Many of these cells are active during REM sleep as well. REM sleep is driven by a distinct population of cholinergic pedunculopontine tegmental (PPT) and laterodorsal tegmental (LDT) areas neurons. During wakefulness and NREM sleep, these cells are inhibited by norepinephrine, serotonin, and histamine, but during REM sleep, the aminergic neurons fall silent, thus disinhibiting the LDT/PPT REM-generating neurons. These cholinergic neurons also produce the atonia of REM sleep by activating the medial medulla, which inhibits motor neurons. The medial medulla also reduces excitatory signals from the locus coeruleus (LC) that normally increase motor tone. SN/VTA, substantia nigra/ventral tegmental area; TMN, tuberomammillary nucleus.

concept was initially put forth by Haponik and colleagues (9), who used computer tomographic imaging during wakefulness to compare airway luminal size in 20 patients with OSA with 10 normal control subjects. They reported reduced cross-sectional airway size in the patients with apnea in the nasopharynx, oropharynx, and hypopharynx, with the site of the smallest airway lumen varying between patients. These ndings were amplied considerably by Richard Schwab and colleagues using magnetic resonance imaging techniques (10). These investigators focused not as much on the airway lumen as on the pharyngeal structures themselves, and reported patients with apnea to have larger lateral pharyngeal walls, tongues, and soft palate tissue, with increased tongue and lateral wall size independently increasing the risk for sleep apnea. Despite this evidence, the argument persisted as to whether the airway lumen was actually anatomically smaller in patients with apnea than in normal control subjects. The counter argument was always that decreased pharyngeal dilator muscle activity or function in the patient with OSA could lead to the same reduced luminal dimensions/conguration seen on computer tomography/magnetic resonance imaging. This discussion was put to rest when Isono and Remmers quantied airway size and collapsing pressure in patients with OSA and control subjects under general anesthesia with complete muscle paralysis (11). Thus, the impact of muscle activation could be eliminated and pure anatomy examined. They found patients with apnea to have higher (less negative or more positive) collapsing pressures and a smaller airway size than the normal control subjects. This concept was further supported by Smith, Schwartz, and Permut, who developed the technique by which the passive closing pressure of the upper airway could be measured in sleeping subjects. Using this approach, they reported progressively less negative or more positive closing pressures as one moved from normal control subjects to snorers to patients with obstructive hypopneas to patients with obstructive apneas (Figure 2) (12). Thus, it seems clear that most patients with sleep apnea have an anatomically small pharyngeal airway.
Upper Airway Motor Control

considerably (Table 1). Thus, the history of the evidence supporting the importance of each trait will be outlined subsequently.
Upper Airway Anatomy

Most evidence would not suggest that the majority of patients with OSA have an anatomically small pharyngeal airway. This

TABLE 1. PHENOTYPIC TRAITS CONTRIBUTING TO APNEA PATHOGENESIS

Anatomic/Physiologic Trait Upper airway anatomy Upper airway motor control Ventilatory control stability Lung volume Arousal threshold Apnea Predisposition Small pharyngeal airway Poor muscle responsiveness during sleep High loop gain Low sleeping lung volume Low respiratory arousal threshold

The concept that upper airway muscles might demonstrate respiratory modulation and be important in keeping the airway open during wakefulness and sleep was initially presented by Sauerland and Harper in 1976 (13). They demonstrated clear inspiratory activation of the genioglossus muscle, with considerably reduced activity of this muscle during expiration. Muscle activity was also substantially diminished during REM sleep compared with wakefulness. The potential role of sleep-induced decrements in pharyngeal dilator muscle activity in the development of sleep apnea was rst proposed by John Remmers in 1978 (14). In this article, he beautifully demonstrated large decrements in genioglossal muscle activity at the onset of apnea. Thus, it became important to understand the control of these muscles, and how sleep affects these mechanisms. Original studies in animals by Mathew (15) suggested that negative pressure in the upper airway was a major stimulus to pharyngeal dilator muscle activation, and that the afferent signal for this reex was carried in the superior laryngeal nerve. Similar observations were made in humans by Richard Horner and colleagues (16), who applied rapid onset pulses of negative pressure to the upper airway while monitoring genioglossal activity. These studies demonstrated robust activation of the muscle in response to negative pressure. Anesthesia of the superior laryngeal nerve and the nose substantially reduced this reex. The precision of this control mechanism across a breath in humans was ultimately demonstrated by Akahoshi and colleagues using a negative-pressure ventilator (Figure 3) (17). They observed millisecond-by-millisecond

Centennial Review

Figure 2. This figure compares the values of pharyngeal passive closing pressure of the upper airway (Pcrit) for normal subjects, snorers, and patients with obstructive hypopnea and obstructive apnea. The data are presented as both individual values and as means SD for the four groups. Note the progressive increase in the mean pharyngeal Pcrit with increasing levels of pharyngeal collapsibility and airway obstructions (*P 0.01).

leading in turn to augmented muscle activity during wakefulness. He convincingly demonstrated that this was the case (19). He further proposed that loss of this reex during sleep would lead to substantial decrements in pharyngeal dilator muscle activity in these patients. Mezzanotte and colleagues (20) and Fogel and colleagues (21) have now demonstrated this to be the case. Thus, through reex mechanisms, the patient with apnea can compensate for decient airway anatomy during wakefulness, but not during asleep. Several nal issues regarding pharyngeal dilator muscles and apnea pathogenesis are worthy of comment. First, Younes has recently proposed that pharyngeal dilator muscles must maintain some ability to respond to negative pressure and compensate for poor anatomy during sleep, as virtually all patients with apnea have periods of stable breathing while asleep (22). Thus, variable ability of these muscles between individuals to compensate during sleep for decient anatomy may dictate, to some extent, who develops apnea and who does not. Second, there is increasing evidence that sleep apnea may, over time, lead to neural/muscle damage, thus further exacerbating the disorder. Kimoff and colleagues (23) have demonstrated reduced sensation in the pharyngeal airway of patients with apnea and Petrof and colleagues reported evidence of muscle damage (24). However, the actual role that these mechanisms play in apnea pathogenesis or propagation is unclear at this time.
Ventilatory Control

modulation of genioglossal activation by airway negative pressure, thus providing a mechanism by which airway patency could be carefully regulated. That this reex control mechanism was affected by sleep was rst demonstrated by Wheatley and colleagues, who showed a substantial reduction in muscle responsiveness to negative pressure during NREM sleep (18). How these mechanisms related to apnea pathogenesis was originally addressed by Mezzanotte and colleagues (19). He proposed that the decient pharyngeal anatomy in the patient with apnea would lead to activation of this negative-pressure reex,

That sleep can affect the mechanisms controlling ventilation has become quite clear over the last 3050 years (2528). This was rst reported by Bulow (25), who observed apparent decreases in carbon dioxide (CO2) sensitivity at sleep onset that led to episodes of hypopnea or apnea with breathing instability. Colin Sullivan (26) subsequently, using a variety of approaches, demonstrated that ventilation during NREM sleep is precisely regulated by metabolic mechanisms, and that removal of these inputs led to a near cessation of ventilation. Eliot Phillipson, in a landmark review (27), brought all of the literature together at that time, proposing that ventilation while awake was inuenced by both behavioral and tight metabolic mechanisms, while breathing during NREM sleep, as stated above, had only metabolic regulation. During REM sleep, he proposed, behavioral inuences

Figure 3. The group mean relationship between negative Pepi and GGEMG are shown. In each condition, there was a highly significant relationship between negative Pepi and GGEMG throughout inspiration. This can be seen both in the plot of both signals against time (upper panels) and in the x-y plots (lower panels). The mean slopes of the relationship between negative Pepi and GGEMG were very similar among conditions within each experiment (x and y scales have the same ratio among x-y plots). Nonetheless, there was invariability some degree of hysteresis, whereby GGEMG changed relatively more for a given change in negative Pepi early in inspiration. The hysteresis loops were generated in a clockwise direction throughout inspiration in each plot. A similar hysteresis was observed in all individual subjects data (not shown).

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 34

2006

reemerge against a background of decreased metabolic control, thereby explaining the irregular breathing pattern that characterizes this sleep state. That REM sleep is associated with reduced chemical (CO2) control was subsequently demonstrated by Douglas and colleagues (28). Building on the information described previously, Skatrud and Dempsey (29), in 1983, elegantly demonstrated evidence for a partial pressure of CO2 (Pco2) apnea threshold during NREM sleep that is often only a few millimeters of mercury below the waking partial Pco2 level. Thus, small decrements in Pco2 can lead to a cessation of respiratory output during sleep. Badr and colleagues (30) subsequently demonstrated that such loss of respiratory output during sleep (central apnea) is commonly associated with a substantial decrement in the size of the pharyngeal airway, even to the point of occlusion in some subjects. This was almost certainly a product of reduced pharyngeal dilator muscle activation resulting from the loss of respiratory output to these muscles. One would have to conclude, therefore, that any process that leads to a loss of respiratory output can lead to transiently decreased pharyngeal dilator muscle activation and, subsequently, a smaller or absent pharyngeal lumen. Thus, ventilatory control instability with waxing and waning of respiratory output could lead to OSA in individuals with a susceptible airway. This has been shown to be the case, as hypoxia-induced periodic breathing can lead to cycling obstructive apneas and hypopneas in individuals with a collapsible airway (31). Despite all of the information described previously here, it was still unclear if individual differences in ventilatory control or ventilatory control stability played an important role in the pathogenesis of OSA. However, several studies have now put this into perspective. First, Younes and colleagues developed the methodology by which loop gain, a measure of ventilatory control stability, could be quantied in humans during sleep. Using this technique, he demonstrated that patients with severe sleep apnea have a higher loop gain (greater ventilatory control instability) than individuals with lower loop gain (32). Wellman and colleagues (33) subsequently reported that loop gain measured during NREM sleep was an important predictor of apnea severity (respiratory disturbance index), but only in patients with intermediate collapsibility of the pharyngeal airway. Individuals with extreme airway collapsibility will have obstructive events regardless of loop gain, whereas those with low collapsibility are not particularly susceptible to the effects of cycling respiratory output. Thus, it would appear that ventilatory control instability may play a pathophysiologically important role in some patients with OSA.
Lung Volume

lung volume and apneic events in these patients. However, the described studies suggest that these changes may be important.
Arousal

It has been known for some time that lung volume decreases during sleep (34). That this might be important in sleep apnea pathogenesis was not considered until 1988, when Van de Graaff demonstrated in dogs that lung ination applies a caudal traction (through the trachea and larynx) on the upper airway, thereby stiffening or rendering the airway less collapsible (35). Thus, upper airway resistance goes down during inspiration as the lung inates. If the trachea is severed, there is no such effect. As a result, lung volume can importantly inuence pharyngeal mechanics. That this might be important in apnea pathogenesis was recently demonstrated by Heinzer and colleagues, who reported that relatively small changes in lung volume during sleep in patients with OSA led to relatively large changes in the continuous positive airway pressure required to prevent ow limitation (high upper airway resistance) in the upper airway (36). No one to date has measured the effect of sleep on lung volume in patients with OSA or the association between decrements in

As an apnea progresses, oxygen saturation falls, CO2 rises, and ventilatory effort steadily increases. This leads to one of three possible outcomes. The patient can arouse, activate the upper airway dilator muscles, open the airway, and hyperventilate, correcting the hypoxia and hypercapnia. Apnea termination can also occur without arousal, either leading to a rhythmic respiratory pattern during stable sleep or continued cycling below the arousal threshold. Finally, the patient could succumb to the event, which is fortunately exceedingly rare. Eliott Phillipson, in 1978 (37), brought to our attention the importance of the respiratory arousal response, and the critical role it plays in apnea termination, with the vast majority of apneas and hypopneas requiring arousal to reestablish ventilation. The primary stimulus to arousal across an apnea was debated for years, with evidence supporting both hypoxia and hypercapnia. However, Gleeson and colleagues, in 1990 (38), reported convincing evidence that neither of the direct chemical stimuli (CO2 or O2) was likely leading to arousal, but that the associated incremented respiratory output or effort was likely responsible. A number of subsequent studies support this conclusion. How output of the respiratory system produces arousal has not been mechanistically delineated, although two theories have been proposed. One suggests that afferents from the upper airway, lungs, or respiratory muscles are activated in response to the increasing ventilatory effort, and modulate arousal in the brain stem. The other posits that the increased activity of respiratory neurons in the medulla, resulting from the obstructed ventilation, directly synapse with neurons controlling state (sleep) and lead to arousal without the requirement for afferent stimulation. Which proposal is correct remains unresolved. Despite these observations, Younes recently pointed out that arousal is not always required for apnea termination, and that when arousal does occur, it often leads to hyperventilation, hypocapnia, and further destabilization of the respiratory control system (39). Therefore, arousal may, in many instances, be needed for apnea termination, but it may also contribute to the development of the next apnea. As a result, a low arousal threshold may be destabilizing to respiration and contribute to apnea pathogenesis. Thus, an individual who falls asleep, has a substantial increase in airow resistance, and always arouses after two to three modest respiratory efforts (low arousal threshold) will be unlikely to establish stable sleep and rhythmic respiration. This may well be the case in patients with what was called the upper airway resistance syndrome (40). These individuals develop high pharyngeal resistance at sleep onset and quickly arouse, despite quite modest hypoventilation. As a result, efforts to increase the respiratory arousal threshold may be therapeutic in some cases.
Other Apnea Pathogenic Mechanisms

Although the principle mechanisms believed to be important in apnea pathogenesis are outlined above, at least one additional pathophysiologic process has also been proposed to have a role. Hudgel and colleagues, in 1990 (41), demonstrated a loss of coordination between upper airway and pump muscle in patients with obstructive apnea. Generally the pharyngeal dilator muscles activate 50100 ms before the diaphragm, presumably stiffening or dilating the airway in preparation for the negative pressure that will develop when the pump muscles begin ring. A loss of this preactivation in the upper airway muscles would render the pharynx vulnerable to partial or complete collapse on inspiration. However, the role that this mechanism plays in apnea

Centennial Review

pathogenesis is unclear at this time, as this has not been a consistent observation.
Basic Neural Mechanisms in Apnea Pathogenesis

Of the pathophysiologic mechanisms described here, most believe that decrements in pharyngeal dilator muscle activity during sleep play the largest role in airway collapse in patients with OSA. As a result, there has been considerable interest over the last 15 years in understanding the neural mechanisms by which changes in state (sleep) might affect the control of the upper airway musculature, with a particular focus on the genioglossus muscle. It is well known, as demonstrated in Figure 1, that the transition from wakefulness to NREM sleep is associated with decrements in the ring frequency of a number of neural systems known to play a role in maintaining wakefulness. These include histaminergic, serotonergic, noradrenergic, and cholinergic neurons. For the most part, the neurons further decrement their ring or cease all activity during REM sleep. As a result, a number of investigators have sought to determine how sleepinduced changes in these neural systems might affect genioglossal activation. Serotonin was pursued rst, and it was demonstrated that raphe serotonergic neurons project monosynaptically to hypoglossal motoneurons (that control genioglossus motor units) and are quite excitatory to these motoneurons (42). The serotonin 2A receptor seems to the receptor subtype primarily responsible for this excitatory response (43). Thus, decrements in the ring frequency of raphe serotonergic neurons during sleep could lead to disfacilitation of hypoglossal motoneurons, thereby contributing to pharyngeal collapse. The same principles likely apply to the noradrenergic neurons in locus coeruleus as well, although they have been less studied. Thus, disfacilitation of upper airway motoneurons, due to withdrawal of excitatory inputs from wakeactive neural systems, likely plays a role in the sleep-induced decrement in pharyngeal dilator muscle activity. The question arises as to whether there is also active inhibition of these upper airway motoneurons during sleep via glycine and/or -aminobutyric acid, as is the case in most spinal motoneurons, particularly during REM sleep. The original study by Kubin and colleagues (44) addressing this question suggests little role for such inhibition. However, a more recent study suggested an important role for a glycinergic inhibitory system on hypoglossal motoneurons (45). Both of these studies were conducted in somewhat reduced animal preparations during cholinergically induced REM sleep. The only study to date conducted in freely moving animals demonstrated the presence of active inhibitory mechanisms (both glycine and -aminobutyric acid) at the hypoglossal motor nucleus during REM sleep (46). However, this inhibitory effect was relatively small when compared with the total decrement in muscle activity observed during REM sleep. Thus, disfacilitation appears to be the primary mechanism by which hypoglossal motoneurons lose activity during NREM and REM sleep. To date there has been little work specically addressing the neural mechanisms mediating the decrement in reex-mediated genioglossal activation during sleep, which seems to play such an important role in apnea pathogenesis. The only such study suggested that serotonin had little effect on the upper airway negative-pressure reex (47). However, this study was conducted in decerebrate animals, and it remains unclear whether serotonin would have inuenced this reex in a free-moving animal. Thus, this is an area in need of further investigation.
Genetics and Sleep Apnea Pathogenesis

in which there were many members with OSA and one with sudden infant death syndrome (48). A subsequent systematic assessment of 51 rst-degree relatives of nonobese patients with OSA compared with 51 control subjects reported more apneas/ hypopneas, oxygen desaturation, and sleep fragmentation in the rst-degree relatives (49). They also found these relatives to have narrower upper airways, with posteriorly positioned maxillae and mandibles. Finally, Redline and colleagues (50) studied the distribution of sleep apnea in families identied through a family member with OSA as opposed to community control families. She reported an adjusted odds ratio of 1.58, 2.51, and 3.97 for having OSA syndrome if an individual has one, two, or three affected relatives (i.e., with OSA). Thus, a strong familial aggregation for sleep apnea is suggested by these studies. This has led a number of laboratories to actively search for the actual genes that contribute to the development of this disorder.
Conclusions

Since Broadbents original description of an individual struggling to breathe during sleep in 1877, we have made considerable progress in our understanding of the pathogenesis of OSA. Much of the physiology has been delineated and early studies addressing the neurobiologic correlates of this physiology have begun. However, treatments for this disorder remain crude, most are not specically focused on individual physiology/phenotype, and many patients remain untreated. Thus, physiologists, neurobiologists, and engineers must work together if additional, meaningful progress is to be made.
Conflict of Interest Statement : D.P.W. is a consultant for Respironics Inc. ($20,000/yr), Alfred E. Mann Foundation ($10,000/year), Aspire Medical ($10,000/yr), Itamar Medical ($10,000/yr), and WideMed ($4,000/yr). He occasionally consults for Cephalon ($5,000) and Organon ($4,000). He has research grants from Respironics Inc. (about $200,000/yr), Alfred E. Mann Foundation (about $30,000/yr), Cephalon (about $125,000), and WideMed (about $10,000/yr).

References
1. Bickelmann AG, Burwell CS, Robin ED, Whaley RD. Extreme obesity associated with alveolar hypoventilation: a Pickwickian syndrome. Am J Med 1956;21:811818. 2. Broadbent WH. Cheyne-Stokes respiration in cerebral hemorrhage. Lancet 1877;3:307309. 3. Aserinsky E, Kleitman N. Regularly occurring periods of eye motility, and concomitant phenomena, during sleep. Science 1953;118:273274. 4. Dement W, Kleitman N. Cyclic variations in EEG during sleep and their relation to eye movements, body motility, and dreaming. Electroencephalogr Clin Neurophysiol 1957;673690. 5. Hobson JA, McCarley RW, Wyzinski PW. Sleep cycle oscillation: reciprocal discharge by two brainstem neuronal groups. Science 1975;189: 5558. 6. Sherin JE, Shiromani PJ, McCarley RW, Saper CB. Activation of ventrolateral preoptic neurons during sleep. Science 1996;271:216219. 7. Porkka-Heiskanen T, Strecker RE, Thakkar M, Bjorkum AA, Greene RW, McCarley RW. Adenosine: a mediator of the sleep-inducing effects of prolonged wakefulness. Science 1997;276:12651268. 8. Guilleminault C, Tilkian A, Dement WC. The sleep apnea syndromes. Annu Rev Med 1976;27:465484. 9. Haponik EF, Smith PL, Boldman ME, Allen RP, Goldman SM, Bleecker ER. Computerized tomograph in obstructive sleep apnea: correlation of airway size with physiology during sleep and wakefulness. Am Rev Respir Dis 1983;127:221226. 10. Schwab RJ, Pasirstein M, Pierson R, Mackley A, Hachadoorian R, Arens R, Maislin G, Pack AI. Identication of upper airway anatomic risk factors for obstructive sleep apnea with volumetric magnetic resonance imaging. Am J Respir Crit Care Med 2003;168:522530. 11. Isono S, Remmers JE, Tanaka A, Sho Y, Sato J, Nishino T. Anatomy of pharynx in patients with obstructive sleep apnea and in normal subjects. J Appl Physiol 1997;82:13191326. 12. Gleadhill IC, Schwartz AR, Schubert N, Wise RA, Permutt S, Smith PL. Upper airway collapsibility in snorers and in patients with obstructive hypopnea and apnea. Am Rev Respir Dis 1991;143:13001303.

The concept that sleep apnea may have a familial basis was rst raised in 1978 by Strohl and colleagues, who reported a family

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 34

2006

13. Sauerland EK, Harper RM. The human tongue during sleep: electromyographic activity of the genioglossus muscle. Exp Neurol 1976;51:160 170. 14. Remmers JE, deGroot WJ, Sauerland EK, Arch AM. Pathogenesis of upper airway occlusion during sleep. J Appl Physiol 1978;44:931938. 15. Mathew OP. Upper airway negative-pressure effects on respiratory activity of upper airway muscles. J Appl Physiol 1984;56:500505. 16. Horner RL, Innes JA, Murphy K, Guy A. Evidence for reex upper airway dilator muscle activation by sudden airway negative pressure in man. J Physiol 1991;436:1529. 17. Akahoshi T, White DP, Edwards JK, Beauregard J, Shea S. Phasic mechanoreceptor stimuli can induce phasic activation of upper airway muscles in humans. J Physiol 2001;531:677691. 18. Wheatley JR, Mezzanotte WS, Tangel DJ, White DP. Inuence of sleep on genioglossal muscle activation by negative pressure in normal man. Am Rev Respir Dis 1993;148:597605. 19. Mezzanotte WS, Tangel DJ, White DP. Waking genioglossal electromyogram in sleep apnea patients versus normal controls (a neuromuscular compensatory mechanism). J Clin Invest 1992;89:15711579. 20. Mezzanotte WS, Tangel DJ, White DP. Inuence of sleep onset on upperairway muscle activity in apnea patients versus normal controls. Am J Respir Crit Care Med 1996;153:18801887. 21. Fogel RB, Trinder J, White DP, Malhotra A, Raneri J, Schory K, Kleverland D, Pierce RJ. The effect of sleep onset on upper airway muscle activity in patients with sleep apnea versus controls. J Physiol 2005; 564:549562. 22. Younes M. Contributions of upper airway mechanics and control mechanisms to severity of obstructive apnea. Am J Respir Crit Care Med 2003;168:645658. 23. Kimoff RJ, Sforza E, Champagne V, Oara L, Gendron D. Upper airway sensation in snoring and obstructive sleep apnea. Am J Respir Crit Care Med 2001;164:250255. 24. Petrof BJ, Henricks JC, Pack AI. Does upper airway muscle injury trigger a vicious cycle in obstructive sleep apnea? A hypothesis. Sleep 1996;19: 465471. 25. Bulow K. Respiration and wakefulness in man. Acta Physiol Scand 1963;59(Suppl 209):1110. 26. Sullivan CE, Kozar LF, Murphy E, Phillipson EA. Primary role of respiratory afferents in sustaining breathing rhythm. J Appl Physiol 1978;45: 1117. 27. Phillipson EA. Control of breathing during sleep. Am Rev Respir Dis 1978;118:909939. 28. Douglas NJ, White DP, Weil JV, Pichett CK, Zwillich CW. Hypercapnic ventilatory response in sleeping adults. Am Rev Respir Dis 1982;126: 758762. 29. Shatrud JB, Dempsey JA. Interaction of sleep state and chemical stimuli in sustaining rhythmic ventilation. J Appl Physiol 1983;55:813822. 30. Badr MS, Toiber F, Shatrud JB, Dempsey J. Pharyngeal narrowing/ occlusion during central sleep apnea. J Appl Physiol 1995;78:1806 1815. 31. Warner G, Shatrud JB, Dempsey JA. Effect of hypoxia-induced periodic breathing on upper airway obstruction during sleep. J Appl Physiol 1987;62:22012211.

32. Younes M, Ostrowski M, Thompson W, Leslie C, Shewchuk W. Chemical control stability in patients with obstructive sleep apnea. Am J Respir Crit Care Med 2001;163:11811190. 33. Wellman A, Jordan AS, Malhotra A, Fogel RB, Katz ES, Schory K, Edwards JK, White DP. Ventilatory control and airway anatomy in obstructive sleep apnea. Am J Respir Crit Care Med 2004;170:1225 1232. 34. Hudgel DW, Devadatta P. Decrease in functional residual capacity during sleep in normal humans. J Appl Physiol 1984;57:13191322. 35. Van de Graaff WB. Thoracic inuence on upper airway patency. J Appl Physiol 1988;65:21242131. 36. Heinzer RC, Stanchina ML, Malhotra A, Fogel RB, Patel SR, Jordan AS, Schory K, White DP. Lung volume and continuous positive airway pressure requirement in obstructive sleep apnea. Am J Respir Crit Care Med 2005;172:114117. 37. Phillipson EA, Sullivan CE. Arousal: the forgotten response to respiratory stimuli. Am Rev Respir Dis 1978;118:807809. 38. Gleeson K, Zwillich CW, White DP. The inuence of increasing ventilatory effort on arousal from sleep. Am Rev Respir Dis 1990;142:295300. 39. Younes M. Role of arousals in the pathogenesis of obstructive sleep apnea. Am J Respir Crit Care Med 2004;169:623633. 40. Guilleminault C, Stooks R, Clerk A, Cetel M, Maistros P. A cause of excessive daytime sleepiness: the upper airway resistance syndrome. Chest 1993;104:781787. 41. Hudgel DW, Harasick T. Fluctuation in timing of upper airway and chest wall inspiratory muscle activity in obstructive sleep apnea. J Appl Physiol 1990;69:443450. 42. Kubin L, Tojima H, Davies RU, Pack AI. Serotonergic excitatory drive to hypoglossal motoneurons in the decerebrate cat. Neurosci Lett 1992;139:243248. 43. Iham G, Shaheen F, Mackiewicz M, Fesnik P, Veasey SC. Single cell laser dissection with molecular beacon polymerase chain reaction identies 2A as the predominant serotonin receptor subtype in hypoglossal motoneurons. Neuroscience 2002;113:145154. 44. Kubin L, Kimura H, Tojima H, Daires RO, Pack AI. Suppression of hypoglossal motoneurons during the carbachol-induced atonia of REM sleep is not caused by fast synaptic inhibition. Brain Res 1993; 611:300312. 45. Yamuy J, Fung SJ, Xi M, Morales FR, Chase M. Hypoglossal motoneurons are post-synaptically inhibited during carbachol-induced rapid eye movement sleep. Neuroscience 1999;94:1115. 46. Morrison JL, Sood S, Liu H, Park E, Liu X, Nolan P, Horner RL. Role of inhibitory amino acids in control of hypoglossal motor outow to genioglossus muscle in naturally sleeping rats. J Physiol 2003;552:975 991. 47. Douse MA, White DP. Serotonergic effects on hypoglossal neural activity and reex responses. Brain Res 1996;726:213222. 48. Strohl KP, Saunders NA, Feldman NT, Hallett M. Obstructive sleep apnea in family members. N Engl J Med 1978;299:969973. 49. Mathur R, Douglas NJ. Family studies in patients with the sleep apnea hypopnea syndrome. Ann Intern Med 1995;122:174178. 50. Redline S, Tishler PV, Tosteson TD, Williamson J, Kump K, Brownes J, Ferrette V, Krejci P. The familial aggregation of obstructive sleep apnea. Am J Respir Crit Care Med 1995;151:682687.