© 2003 American Thoracic Society
Cycling Sleep ApneaThe Balance of Compensated and Decompensated BreathingDepartment of Respiratory Medicine Alfred Hospital and Monash University Melbourne, Victoria, Australia Speech requires flexibility of the upper airway, including laryngeal and hyoid mobility and separation of the hard palate from the epiglottis. As such, all humans are at risk for obstructive apneas. A balance of neural and anatomic factors maintains pharyngeal patency during sleep. With sleep onset, neural drive to the upper airway muscles falls, muscle tone is reduced, and pharyngeal resistance may increase even in individuals who do not snore (1). Anatomic factors like bony (such as retrognathia) or soft tissue (such as enlarged tonsils) may further elevate pharyngeal resistance (2, 3) and lead to obstructive sleep apnea. In patients susceptible to apnea, increased activity of upper airway muscles may overcome and compensate, partially or completely, for the added anatomic load and thus maintain normal ventilation for varying periods of sleep. Greater basal tone of upper airway muscles seen in patients with apneas as compared with patients without apneas (4, 5) supports this argument. Traditionally, the metric used to assess severity is the apnea–hypopnea index (6), although this index does not take into account periods of stable ventilation. The index has been used by scientific and healthcare providers to establish whether a patient has sleep apnea and to compare studies conducted in one center with those conducted in another. The apnea–hypopnea index, however, is limited by inconsistencies in its definition (7), the lack of a dose-dependent relationship with sleepiness (8), and failure to take into account the periods when breathing is stable. The apnea–hypopnea index is a product of the frequency of events when breathing is cycling (so-called cycling frequency) and the fraction of time spent cycling. An apnea–hypopnea index of 30 events/hour may represent 8 hours of apneas occurring at 30 events/hour or 4 hours of apneas occurring at 60 events/hour combined with 4 hours of stable breathing. As such, the apnea–hypopnea index fails to capture periods of stable breathing when upper airway patency is maintained. In this issue of the AJRCCM (pp. 645–658), Younes (9) reports on the relationships between the percentage of stable breathing (as a marker of compensated breathing) and variables known to exacerbate upper airway instability (increasing age, male gender, increasing body mass index, supine body position, and REM sleep). He applied nasal continuous positive airway pressure (CPAP) to a group of 82 patients who were undergoing polysomnography and had varying degrees of apnea (9). When ventilation was stable on CPAP for at least 3 minutes, he acutely reduced the level of CPAP, over 1–4 seconds, to 1 cm H2O, and measured flow within the first few breaths before an arousal. Flow, body position, sleep state, mask pressure, and leak were continuously measured. This procedure was repeated satisfactorily about 600 times. The therapeutic level of CPAP, a measure of the force generated by upper airway dilators to enable stable breathing in the absence of CPAP, was taken as an estimate of upper airway mechanical load. The reduction in airflow with the rapid dial down of CPAP (from about 6.5 to 1.0 cm H2O) as a proportion of airflow at therapeutic CPAP was taken as a measure of collapsibility of the passive pharynx. After dial down of CPAP, while patients were still asleep, Younes (9) observed that about 80% of patients with apneas had periods of stable breathing without arousals for at least 3 minutes, interspersed with hypopneas and apneas at other times. The mean critical closing pressure was -1.5 cm H2O for hypopneas and 3.9 cm H2O for apneas. Periods of stable breathing were twice as likely to occur when patients were lying on their side versus lying in the supine position and 1.3 times more likely in non-REM versus REM sleep. Younes observed that both collapsibility and mechanical load were correlated with the percentage of stable breathing (9). Moreover, the percentage of stable breathing diminished with increasing age, body mass index, REM sleep, and supine position. Male gender had a borderline effect, and the transition from stage 2 to slow wave sleep had no effect. One third of the variance in the percentage of stable breathing could be explained by these variables, implicating a major role for other factors. What might these factors be? First, the shape and size of the facial bony structure (unfortunately not measured in the study of Younes [9]) are likely to explain much of the additional variance. Patients with apneas are known to have smaller upper airways than healthy subjects (10, 11). Men have larger soft palates and longer airways (from the hard palate to the base of the epiglottis) than women, but men also have larger airway volume, which may be compensatory (12). A short maxillary and mandibular length, caudally placed hyoid bone, retrognathia, and a long face have been shown to explain up to half the variance in the apnea–hypopnea index (2, 3). Patients with Marfan's syndrome highlight the association between apnea and high-arched palate and increased nasal resistance (13). Thus, considerable evidence suggests that the bony structure surrounding the collapsible upper airway is responsible for much of the variance in the percentage of stable breathing. Second, the sleeping position may confer compensation and prevent apneas with the head up (14), lateral position (15), or an extended cervical spine, akin to the recovery position after anesthesia. An appropriate ventilatory response to upper airway collapse, without overshooting, might also prevent cycling of apneas. Surfactant has been used experimentally to maintain upper airway patency during anesthesia (16), and its loss (via upper airway inflammation or infection) may contribute to upper airway instability. Not withstanding the limitations of Younes' study (9) (absence of measures of upper airway anatomy, muscle activity, autonomic disturbance and symptoms), identification of periods of stable breathing during sleep in patients with sleep apnea is an important observation. The identification that well known factors (weight, gender, body position, and sleep state) make only a small contribution to the overall variance of stable breathing should encourage the search for alternative mechanisms that compensate against obstructive breathing during sleep. FOOTNOTES Conflict of Interest Statement: M.T.N. has served on the Australian Medical Advisory Board for ResMed and receives an honorarium for attendance. M.T.N. has received an unconditional grant for 2 years to employ staff and conduct a clinical trial. This grant was applied for after partial funding was awarded from the Australian National Health and Medical Research Council. He has no other conflicts of interest. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
