Kathleen Sarmiento and José S. Loredo
SLEEP-DISORDERED BREATHING
The spectrum of sleep-disordered breathing ranges from intermittent snoring, which is primarily a nuisance, to obesity-hypoventilation syndrome (OHS), which is associated with severe morbidity and high mortality. In between these two extremes are disorders of gradually increasing impact on morbidity and mortality, including chronic snoring, upper airway resistance syndrome (UARS), and sleep apnea. More recently, continuous positive airway pressure (CPAP)-emergent central apnea (complex sleep apnea syndrome) has been described in patients with unequivocal obstructive sleep apnea (OSA) who develop central apnea when exposed to CPAP.
OSA is the most common form of sleep-disordered breathing seen in the sleep laboratory. Both children and adults may be affected; however, the prevalence of obstructive apnea is highest in middle-aged men. In 1993, the prevalence of symptomatic OSA in the middle-age working population was 4% in men and 2% in women and the prevalence of asymptomatic sleep apnea was 24% in men and 9% in women. However, with the rise in obesity in the last 20 years, the prevalence of OSA is higher, especially in patients with cardiovascular or metabolic disorders (≥50%). OSA is characterized by repetitive upper airway obstructions during sleep. The immediate consequences of OSA include microarousals from sleep, full awakenings, hypoxemia, hypercapnia, increased systemic and pulmonary pressures, nocturia due to increased atrial natriuretic peptide levels, and sleep fragmentation. The most common presenting symptoms are excessive daytime somnolence and chronic loud snoring. However, it is not uncommon for the OSA patient to present with complaints of daytime fatigue, decreased cognitive function, sexual dysfunction, depression, and even sleep maintenance insomnia due to frequent nocturnal awakenings. Loud snoring, choking spells, abnormal motor activity during sleep, and, more specifically, observed apneas often are noted by the bed partner. These patients also have higher rates of automobile accidents than the general population. In children OSA often presents with snoring and hyperactivity and can be misdiagnosed as attention-deficit hyperactivity disorder.
The etiology of OSA is not well understood. However, obesity, upper airway narrowing, loss of upper airway motor tone during sleep, abnormalities of central control of ventilation, high CO2 sensitivity, and cardiac dysfunction have been implicated. Major risk factors for OSA include obesity (strongest), upper airway abnormalities, male gender, increasing age, a family history of OSA, and menopause in women. There are strong epidemiologic and experimental evidence-linking OSA with the development of systemic hypertension and other cardiovascular complications. The mechanism for this association is unclear, but chronic intermittent hypoxia and frequent arousals leading to hyperactivity of chemoreceptors and the sympathetic nervous system may be involved. Current evidence suggests that untreated OSA causes systemic hypertension, and its treatment reduces blood pressure.
There are three basic types of sleep-disordered breathing: (1) apnea, 90% or more decrease in airflow greater than or equal to 10 seconds; (2) hypopnea, 30% reduction in airflow greater than or equal to 10 seconds accompanied by 4% oxygen desaturation or, alternately, 50% reduction in airflow greater than or equal to 10 seconds accompanied by 3% oxygen desaturation or an arousal; and (3) respiratory effort related arousals (RERAs), the UARS.
Three types of apnea have been described: (1) obstructive apnea, in which oronasal airflow is blocked and diaphragmatic efforts continue; (2) central apnea, in which diaphragmatic and intercostal muscle activity cease; and (3) mixed apnea, obstructive apnea with an initial central component. In most symptomatic patients with sleep apnea, all three types of apnea are found, but, by far, obstructive events predominate. The syndromes of obstructive and mixed sleep apnea are clinically similar and, therefore, are grouped together. UARS differs in that oxyhemoglobin desaturation and obvious apneas or hypopneas are not evident in the standard polysomnogram. However, airflow limitation associated with crescendo snoring often terminates in an arousal; RERAs can be observed if a pressure transducer is used to measure airflow. The gold standard for diagnosing RERAs is by measuring esophageal pressure during sleep; this is seldom performed clinically.
On physical examination, 70% of patients are overweight or obese; it is not unusual to find them dozing in the waiting room. Hypertrophy of the tonsils and adenoids appears to be the major mechanism of the upper airway obstruction in children but not in adults. Malformations of the jaw and maxilla, such as retrognathia, micrognathia, narrow high-arching palate (long face syndrome), and large tori of the palate and the jaw, are noted occasionally. More commonly, the adult sleep apnea patient presents with erythematous, enlarged and edematous soft palate and uvula, prominent tonsillar pillars, drapelike soft palate, macroglossia with scalloping, and Mallampati class 3 to 4 oropharyngeal opening, resulting in oropharyngeal crowding and reduced caliber. However, in some patients the physical examination may be entirely normal. Despite having a narrow oropharyngeal opening, airway obstruction during the waking state is absent secondary to neuromuscular compensation. During sleep, this neuromuscular compensation is lost, predisposing the patient to upper airway obstruction. In the waking state, pulmonary function tests, arterial blood gases, and ventilatory response to carbon dioxide are usually normal, unless the separate effects of morbid obesity or another disease are present. Systemic hypertension is present in up to 50% of patients. Pulmonary arterial hypertension may occur in up to 40% of cases of uncomplicated OSA; however, this is generally not clinically significant.
The diagnosis of the sleep apnea syndromes can be most accurately made by documenting apneic episodes with polysomnography (type 1 study), which measures EEG, oculogram, oronasal airflow, chin and tibialis anterior electromyogram, ventilatory effort, heart rate, and arterial oxygen saturation. More recently, unattended cardiorespiratory sleep recordings (type 3 studies) have been recommended for patients with a high likelihood of OSA after clinical assessment by a sleep specialist. To be considered significant, apneic episodes must last at least 10 seconds and occur repetitively. The Apnea/Hypopnea Index (AHI) is used to determine the severity of sleep apnea. The AHI represents the number of apneas plus hypopneas per hour of sleep. An AHI less than 5 is considered normal, while an AHI greater than or equal to 30 is considered severe.
The pathophysiology and hemodynamic consequences of the sleep apnea syndromes have been studied extensively. During an obstructive episode, the posterior wall of the hypopharynx collapses and the strap muscles of the neck become hypotonic as documented by electromyographic recordings. As the apneic episode continues, hypercapnia and hypoxemia develop. Progressive increases in negative intrathoracic pressure develop with increasing efforts to breathe against the obstruction. Systemic and pulmonary hypertension, sinus bradycardia, and a variety of arrhythmias and conduction disturbances may occur. A loud snort may signal the end of the obstruction and correlates with an EEG pattern of arousal from sleep. Subsequently, abnormalities of gas exchange and hemodynamics resolve rapidly, unless repetitive apneic episodes occur, a frequent situation in severely affected individuals.
Treatment of OSA patients should always include behavioral interventions: weight reduction, smoking cessation, and avoidance of alcohol, sedatives, sleep deprivation, and sleeping in the supine position. Weight reduction to optimum levels can be curative in some cases. However, even modest weight reduction may result in significant reductions in sleep apnea severity. More recently, in small randomized clinical trials, physical exercise, even without weight reduction, and playing the didgeridoo have been shown to decrease AHI by about 25% to 50%. Patients with positional sleep apnea (supine AHI >10/hour and nonsupine AHI <10/hour) may benefit from sleeping in the lateral position, which can be achieved with antisnore shirts, bumper belts, or specialized pillows. Pharmacological therapy for sleep apnea has been disappointingly ineffective in reducing AHI. Nocturnal nasal oxygen therapy can improve arterial oxygen saturation during sleep in OSA, but it does not significantly change the AHI. CPAP is the most effective treatment for sleep apnea. CPAP maintains upper airway patency during sleep by creating a pneumatic splint. In most cases, it can effectively control sleep apnea, reduce arousals, and reverse oxyhemoglobin desaturation. CPAP is usually well tolerated. Complications are mild, usually associated with mask-fit problems. In rare cases, aerophagia can be problematic. CPAP remains a cumbersome therapy despite significant technological advances, including integrated heated humidifiers, auto-CPAP, bi-level positive airway pressure (PAP) modalities, expiratory pressure relief, and ultraquiet blowers. However, with proper introduction and follow-up, CPAP compliance can be significantly higher than the usually quoted 50% nightly use after 4 months of therapy. Overnight CPAP titration in the laboratory is still the preferred method of identifying the optimal pressure level to treat OSA. However, in selected patients, a trial of auto-CPAP at home with minimal pressure estimated from a CPAP prediction formula may be effective and practical in treating OSA and controlling the night-to-night variability of AHI. Oral appliances provide an alternative to CPAP in patients with mild to moderate sleep apnea, although these are not as effective as CPAP in controlling apneic events. Many types of appliances are available; the most effective are those that are adjustable and advance the jaw, increasing upper airway patency. Temporomandibular joint pain is a common complaint from the use of an oral appliance. Surgical procedures that increase upper airway size (e.g., uvulopalatopharyngoplasty) have been found to effectively eliminate snoring, but frequently fail to control sleep apnea, especially when sleep apnea is severe. Except in highly selected cases, uvulopalatopharyngoplasty is no longer recommended to treat OSA. Unfortunately, there is no way to predict which patients will benefit from surgical treatment. Extensive surgery to advance the entire mouth forward (maxillary mandibular advancement) has been as successful as CPAP in controlling sleep apnea in highly selected populations; however, convalescence can be prolonged, frequently resulting in prolonged facial numbness and cosmetic changes. More recently, several minimally invasive operations have been proposed to treat OSA in patients intolerant of CPAP. These include radiofrequency volume reduction of the base of the tongue and genioglossus advancement surgery, in combination with the pillar procedure. However, this type of surgical intervention has not been evaluated systematically. Hypoglossal nerve stimulation to treat OSA remains experimental. Improving upper airway muscle tone during sleep corrects part of the pathophysiology of OSA. In some studies, the AHI has decreased by 50% with hypoglossal nerve stimulation during sleep. Tracheotomy is a last resort for patients with life-threatening conditions exacerbated by untreated OSA and who are not able to use CPAP. Tracheotomy is consistently effective in relieving signs and symptoms of OSA; however, its morbidity makes it a rarely used procedure. Finally, an effort should be made to identify associated conditions that may worsen sleep apnea (e.g., hypothyroidism, use of testosterone, alcohol or sedatives before sleep). Correcting these can sometimes cure or improve sleep apnea. In children, removal of obstructing tonsils and adenoids is usually curative, although the rise in obesity in children has somewhat reduced the effectiveness of this procedure.
Pure central sleep apnea is often associated with advanced age, CNS disorders, congestive heart failure, and sleeping at high altitude. It may be associated with high sensitivity to carbon dioxide, conditions that promote hypocapnia, and chronic opioid therapy. Oral acetazolamide and nasal CPAP have both been partially successful in treating patients with central sleep apnea. Supplemental oxygen may also be beneficial, especially in central sleep apnea associated with high altitude. More recently, adaptive servo-ventilation therapy has been effective in controlling central sleep apnea, especially when associated with Cheyne–Stokes ventilation in patients with heart failure. Successful treatment can result in improved cardiac function.
ALVEOLAR HYPOVENTILATION