eMedicine Specialties > Otolaryngology and Facial Plastic Surgery > Palatal & Maxillofacial Surgery

Snoring and Obstructive Sleep Apnea, Physiologic Approach

Author: Veena Sankar, MD,, George Washington University School of Medicine
Coauthor(s): Philip E Zapanta, MD, Assistant Professor of Surgery, Associate Director of Otolaryngology Residency Program, Division of Otolaryngology-Head and Neck Surgery, George Washington University Medical Center; Consulting Staff, Division of Otolaryngology-Head and Neck Surgery, Medical Faculty Associates; John M Truelson, MD, FACS, Chairman, Division of Head and Neck Surgery, Associate Professor, Department of Otorhinolaryngology, University of Texas Southwestern Medical Center at Dallas
Contributor Information and Disclosures

Updated: Jan 14, 2010

Introduction

Sleep-disordered breathing (SDB) encompasses a spectrum of disorders with implications in many fields of medicine. In its most recognizable and ubiquitous form, SDB is represented by snoring. On occasion, snoring is totally benign, a consequence of a removable stimulus (ie nasal congestion, excess fatigue, central nervous system depressants, abnormal sleep positions), or both. More strikingly, snoring indicates an underlying pathologic change that is more significant than the auditory annoyance of one's bed partner. The spectrum of SDB (in increasing significance) is as follows: primary snoring, upper airway resistance syndrome (UARS), and obstructive sleep apnea hypopnea syndrome (OSAH). Further advancements in the study of this disease spectrum have blurred the distinctions between the different syndromes.

The purpose of this article is to provide a concise but thorough review of SDB (most notably obstructive sleep apnea [OSA]), its pathogenesis and sequelae, and current recommendations for diagnostic evaluation and treatment. OSA is a major health issue in the Western Hemisphere, and its symptoms and pathology have a wide range of consequences. For example, daytime somnolence, the cardinal symptom of OSA, can result in everything from minor reductions in daytime functioning to fatal motor vehicle accidents.

Because of the increased morbidity and mortality in patients with OSA and concurrent cardiovascular disease, researchers have focused on the association of cardiovascular disease and OSA. Consequently, the Sleep Heart Health Study, a large prospective cohort study, is examining OSA and other SDB as an independent risk factor for the development of cardiovascular diseases.¹ Because of the ubiquity of OSA and the variability of its consequences, physicians must have a cursory understanding of this disease process.

The image below details an algorithm for the treatment of snoring and obstructive sleep apnea syndrome.

For further reading, please see the eMedicine articles Snoring and Obstructive Sleep Apnea, Upper Airway Evaluation; Snoring and Obstructive Sleep Apnea, CPAP; Snoring and Obstructive Sleep Apnea, Surgery; and Snoring and Obstructive Sleep Apnea, Prosthetic Management.

Normal Sleep

Sleep studies

These studies are the cornerstone of diagnosis and evaluation of obstructive sleep apnea (OSA). Polysomnography (PSG), for example, is an overnight study that records a multitude of physiological factors, including brain waves, muscle tone, eye movements, respiratory effort, and oxygen saturation (see below). Individual-patient studies are often performed in a sleep laboratory, where the state, quality of sleep in each stage, and the apnea-hypopnea index (AHI) can be determined.

The following studies are used to evaluate and diagnose OSA:

• EEG – To assess brain waves
• EMG – To assess muscle tone
• Electromyelogram (EOG) – To assess eye movements
• Airflow, chest, and abdominal efforts – To assess respiratory effort
• ECG – To assess heart rate
• Pulse oximetry – To assess oxygen saturation

Sleep stages

Sleep is broadly divided into rapid eye movement (REM) sleep and non-REM (NREM) sleep. Based on differences noted by PSG, NREM sleep is further subdivided into 4 stages. The 5 stages of REM and NREM constitute one sleep cycle, which typically lasts about 90 minutes. A person has 3-4 sleep cycles during an average night of sleep.

NREM sleep is loosely associated with a quiet brain in a moveable body and constitutes about 75-80% of sleep. It begins with stage I sleep, and the EEG distinguishes sleep from wakefulness. Table 1 outlines the sleep stages, percentage of total sleep, characteristic EEG waves, and defining traits. REM sleep is often called paradoxical sleep. Despite significant brain activity, the patient is difficult to arouse and there is a loss of postural motor tone. Dreaming also occurs during REM sleep. Transitions between NREM and REM sleep are often accompanied by brief movements, arousals, or both. These movements and arousals are physiologic and are not considered abnormal or sleep disruptions. NREM sleep is dominant in the first third of the total sleep episode. REM sleep increases in percentage of the sleep cycle as the sleep episode progresses and dominates the last third of the total sleep episode.

Table 1. Sleep Stages

Open table in new window

REM sleep accounts for approximately 50% of total sleep in the neonatal period, 20-30% of total sleep in childhood, and 20% of total sleep throughout the rest of life. Delta sleep decreases steadily from childhood to old age, but REM sleep remains fairly constant. The continuity of sleep declines with age. The sleep of the elderly is often disrupted by many arousals (sometimes hundreds) and may lead to symptoms of poor sleep quality. The need for sleep does not decrease with age.

Understanding the normal sleep cycle is helpful in evaluating the effects of sleep-disordered breathing (SDB). A common sign of severely disrupted sleep is an increase in the percentage of stage I sleep, resulting in poor sleep quality and excessive daytime sleepiness (EDS).

Sleep apnea often worsens during REM sleep. As discussed previously, muscle tone of the hypopharyngeal and oropharyngeal musculature decreases, resulting in upper airway collapse. Airway obstruction, oxygen desaturation, and (in severe cases) retention of carbon dioxide cause the characteristic arousals from sleep that are associated with sleep apnea.

Sleep physiology

The following physiologic changes occur during normal sleep:

• Cardiovascular changes (These changes reflect that the cardiovascular system is controlled by the autonomic nervous system. Generally, vagal tone increases and sympathetic input decreases throughout the night.)
  • Heart rate (HR) and blood pressure (BP) decrease during NREM sleep.
  • Further decreased HR and BP occurs during tonic REM.
  • HR and BP increase during phasic REM.
  • Some cardiac dysrhythmias may diminish or disappear during sleep, but premature ventricular contractions (PVCs) increase during REM sleep.
• Central nervous system changes
  • The blood flow to the brain increases during sleep and is higher during REM sleep.
  • Intracranial pressure and temperature increase during REM sleep and decrease during NREM sleep. These differences may indicate variations in the metabolic activity of the brain.
• Pulmonary changes
  • Respiration becomes completely involuntary. Specifically, ventilatory control becomes mainly metabolic in that breathing is triggered by CO₂ levels. Also, the sensitivity of the respiratory chemoreceptors decreases during sleep. Thus, a higher concentration of CO₂ is needed to stimulate ventilation. PCO₂ typically rises by 4 mm Hg during sleep. In a healthy person, this increase is physiologic; however, in a patient with lung disease, this increase may result in significant oxygen desaturation. Chemical responses to hypoxia and hypercapnia decrease in NREM and decrease further in REM. This is another reason why OSA is most severe during REM.
  • A decrease in tidal volume leads to an overall decrease in ventilation.
  • In NREM sleep, respiratory rate, tidal volume, and minute ventilation decrease, leading to an increase in end-tidal CO₂ and a decrease in O₂ saturation.
  • During REM sleep, respiration may be rapid and irregular.
  • Upper airway resistance increases up to 7 times that of waking levels. Muscle tone is lost, especially in the intercostal and pharyngeal muscles; however, the diaphragm maintains its tone.
  • Mucociliary clearance, alveolar oxygen tension (PAO₂), and arterial oxygen tension (PaO₂) decrease. The drop in PaO₂ is normally less than 2%.
  • Patients with SDB experience arousals due to labored breathing. The exact stimulus for arousals is unknown, but mechanoreceptors in the upper or lower airway or the diaphragm may be responsible.
• Gastrointestinal changes
  • Gastric acid secretion generally increases during sleep and peaks in REM sleep.
  • Swallowing and esophageal motility decrease during sleep.
  • Other effects have been minimally studied and have yielded conflicting results.
• Changes in thermoregulation
  • During REM sleep, thermoregulation and perspiration are absent; the body becomes poikilothermic (cold blooded).
  • Overall body temperature decreases during sleep.
• Endocrine changes
  • Growth hormone levels peak during the early hours of sleep and gradually decline.
  • Prolactin levels are affected by sleep and increase during both nocturnal and daytime sleep.
  • Thyroid-stimulating hormone (TSH) levels tend to decrease during sleep, which may coincide with the decreased metabolic needs during sleep.
  • Melatonin and cortisol levels are affected by circadian rhythms and not by sleep itself.

Relevant Anatomy

Oropharynx and soft palate

The airway is passively opened without direct muscular action. In contrast, closure is active and is accomplished by the muscles of the oropharynx. Intrinsic muscular tone keeps the airway in its normal, patent configuration.

Muscular anatomy

The levator palatini muscles originate at the apex of the petrous bone and the cartilaginous portion of the eustachian tubes. These muscles obliquely extend into the posterior superior palate, where they blend at the midline into the palatal aponeurosis.

If the tensor palatini muscles are relaxed during the contraction of levator palatini muscles, 2 dimples appear on the oral surface of the palate. If the tensor stiffens the palate, the palate raises without altering its shape. In either case, the nasopharynx is effectively shut off by the valve action of the levator palatini.

The soft palate superiorly elevates and comes into contact with the Passavant ridge at the anterior part of the atlas. The palatoglossus originates at the anterior palatal aponeurosis and inferiorly extends to the tongue. The palatoglossus comprises the anterior pillar, and its function is to raise the tongue and narrow the oropharynx.

The palatopharyngeus originates as 2 heads at the palatal aponeurosis and the hard palate with the insertion of the levator palatini. The 2 heads join to form the posterior tonsillar pillar and extend to insert into the larynx at the thyroid cartilage near the base of the superior cornu. This muscle elevates the larynx, depresses the soft palate, and constricts the pharynx. Some anatomic descriptions indicate that the fibers from the palatopharyngeus originate at the hard palate and posteriorly extend inside the fibers of the superior constrictor to form the Passavant ridge. According to other descriptions, the Passavant ridge is formed only by the superior constrictor.

The musculus uvulae, which raises the uvula, originates at the palatine spine and inserts into the uvula. The tensor veli palatini arises from the pterygoid plate, the sphenoid bone, and the cartilaginous eustachian tube. The tensor veli palatini inferiorly extends, passes under the hamulus, medially progresses, and then inserts into the palatal aponeurosis. It principally tenses the soft palate.

The 10th cranial nerve innervates all the oropharyngeal muscles except the tensor veli palatine. This muscle is innervated by the fifth cranial nerve.

Craniofacial relevance

Jamieson used cephalometric roentgenograms to evaluate 155 nonselected patients with obstructive sleep apnea.² These patients were compared with 41 subjects who had consulted orthodontists for malocclusion and had no history compatible with sleep apnea. 

Only 2 patients with sleep apnea had normal cephalometric findings. Of the 155 patients studied, 150 had at least 2 landmarks that significantly differed from the normative data. The most common findings were retropositioned mandible, a nasion-sella-basion angle that was more acute than expected, a lower-than-expected displacement of the hyoid bone, and a normal position of the maxilla.

Mouth breathing may result in worsening apnea. An interincisor opening of 1.5 cm causes an associated 1 cm dorsal displacement of the mandibular angle. Assuming that the genioglossus retains its shape, this action results in a 1-cm decrease in the oropharyngeal diameter. Additionally, muscular relaxation during REM sleep results in worsening tongue retrodisplacement. The combination of a decreased diameter of the airway and the decreased muscle tone contributes to airway obstruction.

Anthropomorphic measurements

Almost all anthropomorphic measurements have been associated with the apnea-hypopnea index (AHI). Flemons studied multiple body parameters for their predictive values and documented that 26 anthropomorphic measurements were significantly associated with the AHI.³ Only height and thigh circumference were not associated with the AHI.

Flemons’ study concluded that neck circumference, hypertension, habitual snoring, and gasping or choking were the most important parameters.³ The use of these 4 parameters approximated the more complex models.

The authors noted that the most important factor was neck circumference. It was equivalent in its predictive effect to a combination of other factors, such as body mass index (BMI), age, or gender.

None of these parameters are perfect, but they are used to identify which patients are appropriate candidates for polysomnography (PSG). Most importantly, none of these parameters are indicated to be predictive of treatment success.

Bony and soft-tissue abnormalities

Both bony and soft-tissue abnormalities within the proximity of the upper aerodigestive tract can contribute to OSA in patients.

Pathophysiology of Airway Obstruction

Airflow

Anatomic and physiologic factors combine to cause the pathognomonic pharyngeal collapse associated with sleep apnea. An imbalance between forces that promote airway dilation and forces that promote collapse is associated with sleep apnea. 

The following factors promote collapse:

• Small airway size
• Upper airway resistance
• Negative inspiratory pressure
• Extraluminal tissue pressure
• Small mandible
• Supine position
• Upper airway inflammation

The following factors promote patency:

• Pharyngeal dilator muscles
• Larger airway and mandible
• Higher lung volume

If the physical and spatial relationship of the facial skeleton creates a large airway, the patient can tolerate excessive soft tissue (such as adipose tissue seen in obese patients). Otherwise, the flow of air may be impeded, resulting in snoring and obstructive sleep apnea syndrome (OSAS). If a patient has a congenitally narrower airway, excessive soft tissue, increased extraluminal pressure due to obesity, and increased neck circumference, he or she is at a very high risk for developing obstructive sleep apnea (OSA).

Two basic principles of fluid flow, the Bernoulli principle and the Venturi effect, can be applied to give additional insight to the effects of airway narrowing.

The Bernoulli principle describes fluid flow in a column. A partial vacuum exists at the outer edges of a column of moving fluid. As airflow speed increases, the partial vacuum pressure increases. The smaller the columnis, the faster the flow. This principle is illustrated by a drinking straw: if too much negative pressure is generated within the straw, it collapses; as the negative pressure decreases, the straw becomes more rigid and does not collapse.  

The Venturi effect describes the acceleration of airflow that occurs as a current of air enters a narrow passageway. The wind blowing between buildings or the water spraying out of a hose that is partially occluded by the thumb are examples of this effect. In the human anatomy, the pharynx is the passageway, and the distensibility and moveable walls of the pharynx govern the width of the named passageway; thus, inhaled air accelerates through the pharynx.

Because of the different anatomy and pliability of the pharyngeal walls, patients with OSA and individuals without sleep apnea differ in their closing pressures and airway resistance. In a nonsnoring adult, the negative pressure required to close the upper airway is less than (more negative than) -25 cm water. Snoring adults have a much more pliable airway, with closure during sleep occurring at pressures that range from -2 to -10 cm water.

Physiologically, the pharyngeal muscles work to keep the airway open. During sleep, the cortical input to these muscles decreases and therefore increases the chance of collapse. In a normal person, negative intrapharyngeal pressures during sleep reflexively stimulate pharyngeal dilator tone to keep the airway patent. In a patient with OSA, whose airway is likely narrower than that of the normal patient, the reflexive increase in tone does not counteract the increased pressure of the narrowed airway of sleep and is not enough to prop the airway open. This results in pharyngeal collapse and causes either a hypopneic or an apneic event, which self-terminates in an arousal as the body strives to open the airway and restore ventilation.

There is significant person-to-person variability in chemoreceptor and mechanoreceptor sensitivity, and this impacts the duration of apneic events and the likelihood of significant oxygen desaturation, hypoxemia, and hypercapnia. For example, if a patient has a low sensitivity to high levels of CO₂, the apnea associated with a pharyngeal collapse may last for several seconds before arousal is stimulated.

Cardiac and pulmonary consequences of sleep-disordered breathing

The long-term cardiovascular consequences of OSA are of major concern in sleep medicine today. To date, no definitive association has been made, and numerous studies have yielded conflicting results; however, a strong association undoubtedly exists between OSA and cardiovascular or cerebrovascular disease.

Acutely, a single apneic event has many cardiovascular consequences. During an apneic episode, there is often associated bradycardia, an initial drop then increase in blood pressure, increased pulmonary artery pressure, decreased cardiac output, decreased left ventricular end-diastolic pressure (LVEDP) blood pressure, and increased afterload. Bradycardia is due to increased vagal tone from hypoxemia.

Following an apneic episode, sympathetic nerve activity increases, resulting in tachycardia, increased cardiac output, and a rise in blood pressure.

The cardiovascular sequelae and other sequelae of OSA include the following: pulmonary hypertension, diabetes mellitus, hypertension, heart disease, and stroke. OSA has been shown in studies to cause mild pulmonary hypertension (mean pulmonary arterial pressure [MPAP] of 20-30 mm Hg).⁴ Concurrent chronic obstructive pulmonary disease (COPD) also plays an important role in the development of pulmonary hypertension. OSA has also been associated with glucose intolerance, which is thought to be secondary to sleep fragmentation, repetitive hypoxemia, and higher levels of proinflammatory cytokines associated with insulin resistance.

Whether OSA is a cause of or a common comorbidity of hypertension is undetermined. This distinction is difficult to make because the 2 syndromes occur in similar demographics. Studies show that 30-50% of patients with hypertension also have OSA. The incidence of hypertension in patients with OSA linearly increases with respect to increases in AHI.⁴

Other studies have shown an extremely high prevalence of OSA in refractory hypertension (defined as hypertension, excluding secondary hypertension that is not controlled by more than 3 antihypertensive medications despite compliance). The associated increase in sympathetic tone following the repetitive apneas and hypopneas (as documented by the AHI) is currently believed to possibly be a causal factor for hypertension in some patient populations. In fact, some studies have shown that patients with OSA have an increase in serum and urine catecholamines.⁴

Definitive data that suggest a causal relationship between OSA and hypertension do not exist, and numerous studies conflict with each other and themselves. Bassetti demonstrated an increase in long-term mortality with an increased AHI, but he found no associated increased recurrence of vascular events or stroke.⁵ Currently, ruling out OSA in a hypertensive patient is accepted practice, especially if the patient has additional risk factors.

Some suggest that OSA should be a separate risk factor for heart disease (along with hyperlipidemia, hypertension, diabetes, and smoking). The Sleep Heart Health Study showed that the incidence of stroke was higher in OSA than in any other cardiovascular disease.⁶ Additionally, some studies suggest that sleep-disordered breathing (SDB) is associated with an increased long-term mortality after cerebrovascular accidents. OSA was 5 times more frequent in stroke patients than in healthy patients of the same age and gender.⁵

A further suggested link between heart disease and OSA is the inflammatory effects of apneic events. Hypoxemia is a trigger for the activation of neutrophils, which adhere to endothelial cells and are responsible for the release of oxygen free radicals. Repeated oxidative stress can lead to endothelial dysfunction, increased vasoconstriction, decreased vasodilation, and, ultimately, increased risk of atherosclerosis. Inflammatory markers, such as C-reactive protein, are often elevated in patients with OSA.⁷

Recent studies by Suzuki have shown an increase in interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) in patients with OSA.⁸ After initiation of continuous positive airway pressure (CPAP) therapy, IL-6 levels were noted to decrease. This may suggest that OSA contributes to the oxidative stress and systemic inflammation present in atherosclerosis. Additional studies have shown a reduction in objective sleepiness when immune modulators such as TNF-α were neutralized. Thus, these cytokines may be possible targets of symptomatic treatment of OSA.

Still, many of the studies regarding cardiovascular disease and OSA and/or SDB are contradictory or inconclusive because of the overlap of disease in many patients. OSA may be regarded as an immunologic disease with multi-system consequences because immune modulators have been associated with OSA. However, the cardiovascular system seems to be the most affected.

Syndromes and Symptoms

Snoring, upper airway resistance syndrome (UARS), and sleep apnea are generally regarded as stages in a single process of airway obstruction. Sleep-disordered breathing (SDB) is a generic term that classifies these stages of disease on a spectrum, and these stages differ only in severity and presentation.

Snoring is a respiratory sound, typically occurring during inspiration or expiration, that is generated in the upper airway during sleep. Simple snoring, usually noted by a patient’s bed partner, has no clinical sequelae. In this scenario, patients do not experience daytime somnolence and usually present to the doctor only at the urging of their bed partner or family member. Suggestions to reduce snoring include weight reduction,⁹ positional modification during sleep, or an oral appliance.

If a patient presents with snoring and comorbid hypertension or daytime somnolence, he or she should be further evaluated. Respiratory effort–related arousal (RERA) occurs when upper airway narrowing has led to an increased respiratory effort. As noted on the EEG of a polysomnogram (PSG), this extra effort could stimulate arousal. The clinical relevance of RERA is not as well documented as the apnea-hypopnea index (AHI), but RERA is assumed to impact daytime somnolence. When patients are noted to have several RERAs but have a clinically insignificant AHI, they are often thought to have UARS.

Patients with UARS often present with daytime somnolence and snoring. These patients may display anthropomorphic abnormalities and decreased posterior airspace with retrodisplacement of the tongue. Airflow is limited, but any subsequent apnea or hypopnea is minor enough to cause no concomitant oxygen desaturation. This condition is associated with esophageal pressure changes as noted during a sleep study conducted with an esophageal pressure monitor. Daytime somnolence is a consequence of disturbed, but not apneic, sleep.

An apnea is defined as a cessation of breathing for 10 seconds. In obstructive sleep apnea, the apnea is accompanied by observed ventilatory effort (ie, a chest rise/fall.) In central apnea, no ventilatory effort is seen. Pure central apnea is rare. A mixed apnea is a disordered breathing event that begins as a central apnea and ends as an obstructive one. A hypopnea is a partial reduction in ventilation with continued effort for at least 10 seconds.

Although the criteria for hypopnea may vary from laboratory to laboratory, a generally accepted definition is a 30% reduction in airflow from baseline level plus a 4% or greater decrease in oxygen saturation. The AHI is the cornerstone of placing a specific patient on the sleep apnea spectrum. AHI is the sum of apneas plus hypopneas in 1 hour of sleep. Although the severity of apnea is not measured by AHI alone (clinical presentation [daytime somnolence], hypoxemia, sleep fragmentation, and presence of arrhythmias are also factors), the following is a simple classification system of OSA severity:¹⁰

• An AHI of less than 5 is classified as normal.
• An AHI of 5-14 is classified as mild apnea.
• An AHI of more than 30 is classified as severe apnea.

The Respiratory Disturbance Index (RDI) is similar to AHI except for a few nuances. AHI measures total sleep time and averages the number of apneas and hypopneas per hour. RDI uses total recording time (this includes awake time in the sleep lab). RDI may also include RERA, which AHI strictly excludes. Although no superior index for measuring SDB currently exists, RDI is not without its shortcomings. For example, AHI does not factor in the degree of oxygen desaturation or the number of arousals throughout the night. Both of these could be indices for daytime symptoms and pathogenic sequelae of the disorder.

Physicians should familiarize themselves with the terminology and definitions used in the specific sleep lab of their patients. As delineated above, definitions for specific indices may vary from lab to lab. Efforts are being made to standardize such measurements, but ubiquity has not been established.

Obstructive sleep apnea syndrome

Obstructive sleep apnea syndrome (OSAS) is defined as a chronic respiratory sleep disorder typified by recurrent episodes of partial or complete upper airway obstruction during sleep that cause cessation of airflow in the presence of respiratory effort. These episodes cause repeated arousals and fragmented sleep and are due to the various anatomic and physiologic dysfunctions described above.

• Restless sleep and snoring (observed in 90% of patients)
• Sleep disruptions
• Choking
• Esophageal reflux
• Nocturia
• Heavy sweating

Daytime symptoms include the following:

• Excessive daytime somnolence: This symptom is observed in 70% of patients. Motor vehicle accidents are a significant concern in patients with OSA.
• Morning headaches
• Mood alterations
• Sexual dysfunction
• Hearing loss
• Automatic behavior
• Short-term memory loss
• Hypnogenic hallucinations

Diagnosis and Evaluation

History and Physical
 
The following are risk factors for sleep apnea:

• Obesity (body mass index [BMI] >30): This is a primary risk factor for obstructive sleep apnea syndrome (OSAS). According to the Wisconsin Cohort Sleep study, a 10% increase in weight is associated with a 6-fold risk of developing sleep-disordered breathing (SDB). About two thirds of the patients with OSAS are more than 20% above their ideal body weight.
• Gender: Obstructive sleep apnea (OSA) affects up to 4% of men and 2% of women. The incidence in women increases after menopause.
• Obstruction: The upper airway may be obstructed at any level. Some physicians may use the Mallampatti classification scale to help communicate this idea of airway obstruction.
• Associated anthropomorphic measurements (especially the neck circumference): A neck circumference of more than 43 cm (17 in) is a predictor for an increased apnea-hypopnea index (AHI).³
• CNS depressants: Alcohol and sleeping pills cause a more relaxed airway, which results in its collapse.
• Tobacco use
• Pulmonary disease: Chronic obstructive pulmonary disease (COPD) and restrictive or neuromuscular diseases of the lungs can cause airway problems.
• Hypothyroidism
• Acromegaly (when associated with macroglossia)
• Supine position during sleep
• Craniofacial anomalies and previous trauma/surgery that affects upper airway function or patency.

• Cardiopulmonary (myocardial infarction [MI], arrhythmia, cor pulmonale)
• Memory (short-term memory loss)
• Dyslipidemia (elevated cholesterol, triglycerides)
• Red blood cells (polycythemia)
• Blood pressure (independently associated with snoring)
• Bronchitis
• Gastroesophageal reflux
• Ulcer
• Headache (Morning headaches may result from carbon dioxide buildup that occurs during apneic episodes.)

• 0 = Would never doze
• 1 = Slight chance of dozing
• 2 = Moderate chance of dozing
• 3 = High chance of dozing

Open table in new window

• A score of 0-5 should be interpreted as supernormal.
• A score of 5-10 should be interpreted as normal.
• A score of 10-15 should be interpreted as sleepy.
• A score of 15-20 should be interpreted as very sleepy.
• A score of more than 20 should be interpreted as dangerously sleepy. (Arrange transportation for patient.)

The patient should relax the tongue and remain quiet so that the tongue and palate relationship can be determined. The Mallampatti score is as follows:

Class II: The soft palate, fauces, and uvula are visualized.

Class III: The soft palate and base of the uvula are visualized.

Class IV: The soft plate is not visualized.

Pulse oximetry

Nocturnal pulse oximetry is often used as a screening tool to identify the patients who have a high probability of a positive polysomnography (PSG) result. Oximetry alone is sensitive but not specific enough to make a definitive diagnosis of OSA.

When combined with the Epworth sleepiness scale, pulse oximetry is a very good screening tool for deciding which patients should undergo PSG. A number of algorithms have been used with pulse oximetry in order to decide who is likely to have a positive PSG result.

Gyulay reported that the percentage of time a patient is saturating 92% or 90%, CT92 and CT 90, respectively, are critical landmarks with regard to predicting presence of obstructive sleep apnea and likelihood of positive polysomnography.¹¹ In normal sleep, the CT92 should be less than 2% and the CT90 less than 1% of the total sleep time. For example, if the patient sleeps for 100 minutes and has an oxygen saturation of less than 92% for 2 minutes, then sleep apnea is likely. In this case, the 2 minutes need not be (and usually are not) contiguous. For those patients with a respiratory disturbance index (RDI) of greater than 15 and symptoms severe enough to warrant use of continuous positive airway pressure (CPAP), the CT90 has 100% sensitivity in predicting OSAS, according to Gyulay.¹¹

Counting the number of desaturation events (rather than the total time of desaturation) is no more predictive than the clinical assessment. Respiratory events that cause sleep fragmentation but not arterial desaturation probably lessen the predictive value of this count. No connection exists between desaturations and sleep fragmentation; therefore, oximetry readings are markers for SDB and not measurements of SDB.

Polysomnography

Full-night PSG is the criterion standard for diagnosing OSA. Split-night PSG is performed when the first half of the total sleep episode is dedicated to diagnosis and the latter half is dedicated to a CPAP trial. The patient must be asleep for at least 2 hours before the initiation of CPAP.

Details of the PSG study are outlined above (see Normal Sleep).

Other objective diagnostic tests

The Multiple Sleep Latency Test (MSLT) measures the length of time required for a person to fall asleep. The patient is given the opportunity to fall asleep 5 times throughout the day, separated by 2 hours between each test. The patient is allowed up to 20 minutes to fall asleep. The interpretation of this test is as follows:

6-10 minutes - indeterminate sleepiness

More than10 minutes - normal sleepiness

Treatment

As always, the clinician must fit the appropriate treatment to the needs of the patient. This requires a detailed patient history and physical examination, as well as an exhaustive explanation of the problem, treatment options, and possible complications. Understanding the pathophysiology and the statistical probabilities associated with each treatment is essential for adequate treatment of patients with SDB. The resolution of symptoms with a surgical procedure does not necessarily equate to the resolution of apnea. Resolution of daytime somnolence does not necessarily mean obliteration of SDB. Because of this potential problem and the progressive nature of the disease, patients should be followed in clinic with periodic PSG to ensure efficacy of treatment.

Nonsurgical behavioral therapy

The following are nonsurgical behavioral therapies:

• Weight loss (very effective in patients who are very obese)
• Improved sleep hygiene (Avoid alcohol or sedative-hypnotic medication prior to sleep. Alcohol and sedative-hypnotic medications aid in more rapid sleep onset, but the sleep in not as deep or "refreshing.")Positional therapy (avoidance of the supine position, to decrease snoring and potentially obstruction)
• Continuous positive airway pressure (CPAP)
  • CPAP behaves as a pneumatic splint that props open the pharyngeal walls of the upper airway to prevent collapse.
  • CPAP is the criterion standard of therapy for obstructive sleep apnea (OSA) and is the most commonly used initial treatment.
  • Patients may feel improvement after just one night; however, the major cause of failure of treatment is noncompliance.
  • The specific pressure needed to maintain patency of the patient’s airway is determined during a sleep study. CPAP is titrated upward until all sleep-disordered breathing (SDB) events cease and oxygen saturation is maintained above 90%.
  • Autotitrating CPAP: CPAP self-adjusts against various aspects of the patient’s airflow patterns based on an algorithm within the unit. Overall, this reduces the pressure at which a patient is maintained, compared with the standard CPAP. On the other hand, if the machine does not titrate properly, treatment may be subtherapeutic. Patients may tolerate autotitration better than standard CPAP, although this has not been directly measured. 
  • Bilevel positive airway pressure (BiPAP) is the delivery of different pressures during inspiration and expiration. It may be used in patients with concomitant hypoventilatory syndromes that require more ventilation at night or in patients who require very high CPAP pressures (>16 cm water) but cannot tolerate the high pressure. There is no evidence that BPAP has greater compliance than CPAP.
  • CPAP may not simply act mechanically to reduce the deleterious effects of OSA. Increased sympathetic tone associated with OSA has been shown to resolve after CPAP treatment. Studies of whether CPAP has been shown to resolve hypertension have yielded conflicting results.
  • Dry mouth, nasal congestion, and rhinorrhea (CPAP rhinitis) are common side effects of CPAP therapy that may lead to noncompliance.
• Oral appliances
  • Dental devices are simple, nonsurgical alternatives that are generally effective for simple snoring and mild sleep apnea. These devices reposition the upper airway anatomy to prevent obstruction by increasing the air space, either by stabilizing the mandible anteriorly, advancing the soft palate and tongue, or both (see the eMedicine article Snoring and Obstructive Sleep Apnea, Prosthetic Management for more details). Ferguson assessed the efficacy of mandibular advancing splints, which improved apnea slightly.¹² Subjective reports, such as reduction of daytime somnolence, were greater than objective measures.
  • Interest and usage of dental devices has increased since the 1990s. With advancements in appliances and design, patients tolerate these devices relatively well. Oral appliances fall under the following 2 categories: mandibular advancement devices (MAD) and tongue-retaining devices.
  • Compared with CPAP, the primary advantages of oral appliances are portability, increased tolerance, ease of care, and quietness. The disadvantages are that they are not as effective in patients with severe apnea, they may be uncomfortable, and they may cause temporary temporomandibular joint (TMJ) pain.
  • Coruzzi focused on the effectiveness of oral appliances and suggested that personalized oral jaw-positioning devices improve SDB.¹³ In turn, this may reduce cardiac variability, a marker for cardiovascular risk.
  • Minimal risks to the teeth are present when making the impression at fitting. Occasionally, the jaw thrust device and CPAP must be used in concert to control sleep apnea.
  • Contraindications are TMJ arthritis, poor dentition, nasal obstruction, poor motivation, and pharyngeal stenosis.
• As a solitary treatment, oxygen is minimally effective.
• Drugs are largely ineffective for obstructive apnea.

Surgical therapy

Surgical intervention for treating OSA is typically reserved for patients in whom CPAP treatment fails, whether due to noncompliance, intolerance, or ineffectiveness. Broadly speaking, the effectiveness of surgical treatment is quite controversial and studies have yielded conflicting results. Uvulopalatopharyngoplasty (UPPP), the most common surgical procedure for the correction of OSA, has only a 40% success rate, at best, and has even been shown to worsen OSA in several cases.¹⁴

Interventions can be divided into those that bypass the upper airway and those that reduce upper airway obstruction. The former is much more effective but is typically too extreme and not the treatment of choice in most situations.

Tracheostomy

Tracheostomy is the creation of a direct opening of the trachea (via tracheotomy) into which a tube is placed to bypass potential upper airway obstruction. During waking hours, the patient can cap the tracheotomy tube, and upon sleeping, the patient may uncap the tracheotomy tube to allow unobstructed ventilation. Although this is a very effective treatment for OSA, it is reserved for special cases, such as life-threatening disease secondary to cor pulmonale, arrhythmias, or hypoxemia that can otherwise not be treated by CPAP.

LAUP, RFUP, and UPPP

More commonly, surgical correction seeks to relieve upper airway obstruction. Soft palate surgery is the most common type of attempted surgical correction of OSA. Laser-assisted uvuloplasty (LAUP), radio frequency uvuloplasty (RFUP), and uvulopalatopharyngoplasty (UPPP) have been used as the first step in surgical intervention. UPPP consists of removal of the palatine tonsil, uvula, a portion of the soft palate, and the lateral pharyngeal wall. Outcomes of soft palate surgery are less than ideal. With respect to long-term results, in 2000, Boot et al suggested that resolution of OSA declined years after surgery.¹⁵

Using LAUP, RFUP, or both may be effective in mild sleep apnea but is not proven in moderate or severe sleep apnea. The RFUP has a 41% recurrence rate of simple snoring after 14 months. Repeat treatment in 8 of 9 patients resulted in a significant decrease in snoring. LAUP has also been shown to improve exercise tolerance in OSA patients during cardiovascular stress tests.¹⁶

Friedman classification

The Friedman classification was developed to predict the successful outcome of surgery, specifically UPPP, for OSA patients. The system includes an evaluation of palate position, tonsil size, and body mass index (BMI) to predict the success of UPPP. A modified version of the aforementioned Mallampatti classification is used to evaluate palate position relative to oropharyngeal size. In Friedman’s classification, the patient keeps his tongue in a neutral position and 4 stages are used to describe the airway.¹⁷

The palate position is graded from I-IV, as follows:

• I: The uvula, soft palate, and tonsils/pillars are clearly visible.
• II: The uvula and soft palate are visible, but the tonsils are not.
• III: Only part of the soft palate is visible.
• IV: Only the hard palate is visible.

The tonsil size is graded from 0-4, as follows:

• 0+: A previous tonsillectomy has been performed.
• 1+: The tonsils are hidden within the tonsillar pillars.
• 2+: The tonsils extend to the tonsillar pillars.
• 3+: The tonsils extend beyond the pillars but not to the midline.
• 4+: The tonsils extend to the midline.

BMI is loosely included in the Friedman classification. The stages of disease are defined as follows:

• Stage I disease includes patients with palate position I or II, tonsil size 3 or 4, and a BMI of less than 40.
• Stage II disease includes patients with palate position I or II and tonsil size 0, 1, or 2--or palate position III and IV and tonsil size 3 or 4--and BMI less than 40
• Stage III disease includes patients with palate position III or IV and tonsil size 0, 1, or 2.

All patients with a BMI of greater than 40 are considered stage III. Patients with stage I classification are predicted to have successful results with UPPP. Stage II and III patients should have adjunctive treatment if they undergo UPPP because the procedure does not have a high likelihood of cure. The strength of this classification system is still under review, but it has consistently shown to be a reliable indicator of prognosis after surgery.

Advancement procedures

Advancement procedures are often attempted in patients who do not respond to soft palate procedures. These procedures have a 40-70% cure rate. Genioglossus, hyoid, and suture tongue advancement are common advancement procedures. Bimaxillary advancement is used when all other attempts have failed but can be attempted at any time, particularly in patients with craniofacial abnormalities.

Other surgical procedures

Other attempted surgical procedures include nasal and tongue surgery. Correction of nasal obstruction alone does not usually correct sleep apnea, particularly because patients with apnea often breathe through their mouths. Nasal procedures are often used as an adjunct to make CPAP available in the postoperative period. Tongue resection and submucosal radiofrequency surgery have been shown to be efficacious in patients who do not respond to soft palate procedures.¹⁴

Radiofrequency surgery is a minimally invasive technique used to reduce the bulk of the tongue at its base. The beneficial effects of radiofrequency surgery are likely a result of changes in upper airway collapsibility and not alteration of the upper airway anatomy. Tongue-base suspension procedures, in combination with UPPP, have been shown to be more efficacious than UPPP alone.¹⁸

Surgical treatment for OSA is often managed with a multistep approach. According to Jacobowitz, combined palatal and hypopharyngeal surgery effectively reduced symptoms (eg, daytime somnolence) and the apnea-hypopnea index (AHI) in patients with apnea who are CPAP intolerant and have diffuse airway narrowing and severely elevated AHI.¹⁴ Multilevel surgery in patients with Friedman stages 2 and 3 yielded a decrease in symptoms and AHI.

Using a staged surgical approach, Riley and Powell have demonstrated a long-term surgical success rate of 90%. Phase I consists of UPPP, genioglossus advancement, and hyoid advancement. Phase II is reserved for phase I failures and consists of bimaxillary advancement. The amount of skeletal advancement in the procedures is the most significant predictor of outcome. The study used RDI and LSAT (oxyhemoglobin desaturation nadir) to compare preoperative CPAP control of OSA with postoperative outcomes. The differences were not statistically significant in 90% of cases. The average follow-up duration was 50.7 months. In other words, OSA was managed equivocally by both advancement procedures and CPAP.

In a recent publication by Verse, whose study focused on creating a protocol for multilevel surgery for OSA, all patients underwent uvula flap, tonsillectomy, and radiofrequency treatment of the tongue base.¹⁹ One group additionally underwent hyoid suspension while the other did not. In both groups, nasal surgery was performed if necessary. The group that received the additional hyoid suspension yielded better short-term results. Generally, surgical treatment of OSA has increasing efficacy with increasing number of procedures and a multistep approach. Note that most surgical candidates have a relatively low BMI compared with the classic OSA patient.

Palatal implants have recently attracted attention as a potential treatment for mild-to-moderate OSA. This is a minimally invasive approach in which 3 implants are placed inside the soft palate centered around the uvula. The morbidity is minimal because this approach can be performed in a clinic and no mucosa is cut. If successful, the patient avoids both the operating table and the chronic use of the CPAP machine. The palatal implant method has the potential to significantly improve AHI in patients with mild-to-moderate OSA, minimal tonsillar hypertrophy, uvula elongation, and a BMI of less than 30. Short-term results are comparable with those reported for UPPP without the associated morbidity. In a recent study by Friedman, palatal implants were found to be a valuable mode of treatment for patients whose symptoms failed to improve after surgery, namely UPPP.²⁰ The subjective improvement of symptoms was significant.

CAPSO

Cautery-assisted palatal stiffening operation (CAPSO) is another outpatient procedure being studied for the treatment of OSA. Unlike other oropharyngeal procedures (UPPP and LAUP), CAPSO is a short, minimally invasive procedure with low morbidity. It has, however, been linked to increased pain compared with UPPP and LAUP.²¹

Multimedia

(Enlarge Image)

Media file 1: Airway collapse.

(Enlarge Image)

Media file 2: Snoring and obstructive sleep apnea syndrome (OSAS) treatment algorithm. ESS - Epworth sleepiness scale; LAUP – laser-assisted uvuloplasty; RFUP - radio frequency uvuloplasty; CPAP - continuous positive airway pressure; UPPP - uvulopalatopharyngoplasty.

Keywords

snoring, sleep apnea, obstructive sleep apnea, OSA, obstructive sleep apnea syndrome, OSAS, obstructive sleep apnea-hypopnea syndrome, OSAHS, snoring, sleep physiology, sleep-disordered breathing, SDB, REM sleep

• Search for CME/CE on This Topic »

• Snoring and Obstructive Sleep Apnea, Surgery (Otolaryngology and Facial Plastic Surgery)
• Snoring and Obstructive Sleep Apnea, CPAP (Otolaryngology and Facial Plastic Surgery)
• Snoring and Obstructive Sleep Apnea, Prosthetic Management (Otolaryngology and Facial Plastic Surgery)
• Snoring and Obstructive Sleep Apnea, Upper Airway Evaluation (Otolaryngology and Facial Plastic Surgery)

• GI Tract Findings Common With Obstructive Sleep Apnea, Snoring

• New Daily Persistent Headache Related to Sleep Apnea?