Definitions
Severity of obstructive sleep apnoea is measured as the apnoea-hypopnoea index (AHI) and ranges from mild (5-20), to moderate (>20-<40) to severe (>40). An apnoea, defined as cessation of airflow for at least 10 seconds, is classified as obstructive or central on the basis of presence or absence of respiratory effort. A consensus conference (Chicago criteria) provided a definition of hypopnoea as including one of three features: substantial reduction in airflow (>50%), moderate reduction in airflow (<50%) with desaturation (>3%), or moderate reduction in airflow (<50%) with electroencephalographic evidence of arousal. Previously, the definition of obstructive sleep apnoea syndrome was restricted to patients with increased AHI and symptoms such as excessive daytime sleepiness. However, because obstructive sleep apnoea has been shown to increase cardiovascular risk, many investigators classify non-sleepy patients with high AHI as having this disease.

Epidemiology
Twenty percent of people snore however, 4% of men and 2% of women in middle-aged populations in Hong Kong have symptoms of obstructive sleep apnoea with an AHI of greater than 5 events per hour of sleep. The risk factors for obstructive sleep apnoea are obesity, male sex, and increasing age. Of these factors, obesity is the most important because it is present in roughly 70% of patients with this disorder; it is reaching epidemic proportions, especially in developed countries; and it is the only major reversible risk factor. Pharyngeal airway size is probably diminished with increased weight, thereby increasing the propensity for apnoea. Male sex is also a risk factor for obstructive sleep apnoea, but the magnitude of the effect varies between studies. Sleep laboratory data suggest a five or six-fold increased risk, whereas community based studies suggest a two or three fold increased risk of this disorder in men versus women. Some newer risk factors for apnoea have been identified, including menopausal status and ethnic origin. Although the anatomical and physiological mechanisms underlying this predisposition are unclear, differences in the soft tissue and bony structure of upper airways are the most likely explanations. Finally, some social habits have also been identified as risk factors for obstructive sleep apnoea, including cigarette smoking and alcohol consumption. Data for smoking are largely epidemiological, with the presumed mechanism being upper airway oedema. The effect of alcohol intake is almost certainly due to suppression of pharyngeal dilator muscle activation.

Clinical Presentation
Sleep apnoea should be suspected in individuals with cardinal signs and symptoms. Measures of obesity, witnessed apnoeas, and prominent snoring are the strongest of these associations. Although patients with obstructive sleep apnoea are usually obese older men, more subtle presentations can occur. For example, roughly 30% of patients with obstructive sleep apnoea are not obese, but many physicians do not pursue this diagnosis in individuals who are not overweight. Personality changes and poor performance or concentration may be manifestations in addition to symptoms of daytime sleepiness and headaches. In view of the prevalence of sleep apnoea, physicians should include in a general review of body systems, questions about snoring and sleep quality.

Consequences
Once apnoea or hypopnoea develops, arousal from sleep is generally needed to stop the event. The precise stimulus to arousal is debated, with most investigators suggesting some combination of increasing respiratory effort in association with hypoxia or hypercapnia as the cause. The neurocognitive consequences of recurrent arousals are well established and include sleepiness, reduced performance in neuropsychological tests, lengthened reaction times, reduced creativity, decreased quality of life, and increased accidents. Furthermore, results of randomised controlled trials have recently shown a substantial improvement in cognitive performance and daytime function after treatment with CPAP. The cardiovascular consequences of obstructive sleep apnoea can be acute and chronic. Acutely, many mechanisms probably contribute to the haemodynamic response to obstructive apnoea. Reductions or cessations in breathing can be associated with large reductions in the partial arterial pressure of oxygen (PaO₂), increases in the partial arterial pressure of carbon dioxide (PaCO₂), and increases in sympathetic activity. At the termination of apnoea, increased stroke volume (augmented preload with increased intrathoracic pressure) in the setting of a vasoconstricted circulation (sympathetic activation) can lead to repetitive profound acute increases in systemic blood pressure during the night. Chronically, obstructive sleep apnoea can lead to sustained periods of high blood pressure. Finally, in cross-sectional studies, the combination of acute and chronic haemodynamic effects in obstructive sleep apnoea have been associated with increased risk of myocardial infarction, cerebrovascular accidents, and congestive heart failure.

Diagnosis
The gold standard for diagnosis of obstructive sleep apnoea is overnight polysomnography in a sleep laboratory, which generally incorporates recording of electroencephalogram, electro-oculogram, chin electromyogram, snoring (microphone), thermistor, electrocardiogram, pulse oximetry, and tibialis anterior electromyogram. Use of overnight oximetry in screening for sleep disordered breathing is controversial. Although repetitive desaturation in the appropriate clinical context is highly specific for obstructive sleep apnoea, oximetry frequently does not have enough sensitivity for detection of more subtle disordered breathing, with sensitivity and specificity ranging from 50-90%. Home polysomnography testing is an accurate and increasingly popular alternative to formal testing in a sleep laboratory and provides a natural environment to the patient reproducing more usual sleeping conditions; and the cost is reduced significantly.

Management
CPAP treatment remains the treatment of choice for obstructive sleep apnoea because of its effectiveness in elimination of apnoea and improvements in apnoea sequelae. Results of randomised trials have shown substantial improvements in both sleepiness and neurocognitive performance of patients on nasal CPAP compared with those on placebo or subtherapeutic CPAP. In addition, decrements in blood pressure have been shown with CPAP treatment. However, CPAP adherence is still a difficulty, with the best compliance in patients with severe obstructive sleep apnoea and substantial sleepiness. Strategies including heated humidification or nasal decongestants, steroids, or both to alleviate nasal symptoms plus intensive support with regular follow-up improve CPAP adherence. Other (non-CPAP) treatments should be considered in three situations. First, individuals with clearly reversible causes for obstructive sleep apnoea (eg, anatomical deformities such as adenotonsillary hypertrophy) should be considered for surgery. The most common surgical procedure for obstructive sleep apnoea is uvulopalatopharyngoplasty (UPPP), in which the uvula and redundant soft tissue of the soft palate is resected. Only about 41% of patients who undergo the procedure obtain an AHI of fewer than 20 events per hour. Newer surgical approaches such as laser-assisted palatal procedures, and radiofrequency ablation techniques have also been disappointing. Neither of these techniques has been effective in treatment of obstructive sleep apnoea. However, for patients whose main complaint is snoring, with little or no apnoea found on formal testing, these procedures may be considered. Although obesity is a reversible risk factor, substantial weight loss has been associated with poor success rates. Second, individuals who have failed or refuse CPAP treatment should be considered for other approaches (oral appliances or upper airway surgery including tracheostomy). Mandibular advancing and tongue retaining devices are mechanical devices designed to prevent retroglossal collapse. Although CPAP is clearly more effective than such devices, especially for severe disease oral appliances should be considered for patients who have failed or refused CPAP treatment, for those with snoring or mild obstructive sleep apnoea, and for those who do not respond to surgery. Third, treatment of mild apnoea is a subject of much debate. In such patients, conservative measures should be emphasised, including maintenance of nasal patency, avoidance of depressants including alcohol, and the goal of 7¯8 h sleep per night. In addition, individuals with documented positional apnoea should be encouraged not to sleep on their backs. Despite the potential advantages of conservative measures, treatment with CPAP should also be considered for patients with mild apnoea. Newer positive pressure devices have gained popularity and include bilevel positive airway pressure machines (different inspiratory and expiratory pressure) and autotitrating devices (which provide variable pressure levels based on snoring, flow limitation, vibration, etc).

Conclusion
The rising number of people with obesity will probably make obstructive sleep apnoea an increasingly important public-health problem, especially in view of the neurocognitive and cardiovascular sequelae associated with this disorder. Furthermore, CPAP has been established as the treatment of choice for obstructive sleep apnoea syndrome on the basis of randomised controlled trials, and improvements in our understanding of the underlying mechanisms of obstructive sleep apnoea will hopefully lead to improved therapeutic strategies.