INTRODUCTION

According to the International Classification of Sleep Disorders by the American Academy of Sleep Medicine, obstructive sleep apnea (OSA) belongs to a group of sleep‐related breathing disorders, along with central sleep apnea, sleep‐related hypoventilation, and sleep‐related hypoxemia disorder (International Classification of Sleep Disorders, 2014). It is a chronic pathological condition characterized by recurrent pauses of breathing during sleep due to obstruction of the upper airways, which leads to sleep disruption (Eckert & Malhotra, 2008; Jordan, McSharry, & Malhotra, 2014) (Figure 19.1). Specifically, in the context of OSA, narrowing or complete obstructions of the upper airways are observed in the pharyngeal region during sleep, when the tone of upper‐airway dilator muscles decreases normally and leads to temporary partial or complete cessation of breathing (respiratory events) despite respiratory effort. Breathing pauses are accompanied by a decrease in blood oxygen saturation (hypoxemia) and carbon dioxide retention (hypercapnia) that lead to sleep disruption (awakening) to restore the tone of the upper‐airway dilator muscles and normal airflow. However, after sleep restoration, upper‐airway obstructions reoccur and repetitive awakenings end up disrupting sleep architecture, duration, and quality (Eckert & Malhotra, 2008; Jordan et al., 2014).

The clinical presentation of OSA includes symptoms that occur during both sleep and daytime. During sleep, OSA is characterized by a distinct pattern of snoring, which involves intervals of loud noises and respiratory efforts that alternate with intervals of silence due to complete breathing cessations. In most patients, snoring symptoms are usually present for many years, but the decision to undergo a sleep diagnostic procedure is usually made due to the aggravation of snoring frequency and/or intensity as reported by patients’ bedpartners. This aggravation is usually reported when alcohol has been consumed before bedtime or as a result of recent weight gain (Lurie, 2011; Stansbury & Strollo, 2015). Patients’ respiratory events can vary in severity, and in the case of severe OSA, they are maintained for several seconds before respiratory movements are observed and cause cyanosis due to hypoxemia. The end of respiratory events is usually accompanied by intense snoring and sounds that resemble “groans” or “squeals”, while awakenings are often accompanied by whole‐body movements, which can be sharp or even violent. The majority of patients are not fully aware of the breathing pauses and repetitive awakenings that occur during sleep but usually report insomnia symptoms and a sense of “restless” sleep (Lurie, 2011; Stansbury & Strollo, 2015). Nocturnal symptoms of the disease also include heavy sweating and nocturia, while early daytime symptoms manifest as fatigue, decreased rejuvenation, disorientation, mental sluggishness, lack of coordination, and headaches that often require analgesics (Lurie, 2011; Stansbury & Strollo, 2015).

DIAGNOSIS

Polysomnography (PSG) is the gold‐standard method for OSA diagnosis (Epstein et al., 2009; Kapur et al., 2017; Qaseem et al., 2014). PSG is non‐invasive and consists of a simultaneous recording of multiple physiologic parameters related to sleep and wakefulness. The procedure is usually performed during night sleep in a laboratory (e.g., a sleep center) under the supervision of specialized medical personnel (sleep technologists and somnologists). In the context of the PSG, brain activity via electroencephalography, eye movements via electrooculography, muscle activity via electromyography, heart rate via electrocardiography, respiratory airflow via pressure transducers, respiratory effort via abdominal movement sensors, and blood oxygen levels via pulse oximetry are assessed during real sleep conditions, and all data are simultaneously recorded on a multi‐channel digital system (Figure 19.2) (Epstein et al., 2009; Kapur et al., 2017; Qaseem et al., 2014). After the procedure is completed, all recorded data are scored and analyzed according to the American Academy of Sleep Medicine guidelines (The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications, 2015), providing a detailed assessment of the patient’s sleep architecture and cardio‐respiratory function.

With regard to sleep architecture, PSG analyses provide data on wake time, and time spent in non‐rapid eye movement (NREM) sleep, which consists of stage 1 (N1), stage 2 (N2), and stage 3 (N3) sleep as well as rapid eye movement (REM) sleep; for each sleep stage, the relative contribution to total sleep period is calculated as ([(time spent in sleep stage [min] ÷ total sleep period [min]) × 100]). Table 19.1 presents an overview of adult sleep staging and disturbances observed in OSA patients (Jordan et al., 2014). In general, human sleep consists of distinct sleep circles, each starting from a period of non‐REM sleep (N1 to N3) that is followed by a shorter period of REM sleep, which are repeated several times along with some arousals until the final awakening. Most patients with OSA spend more time awake, in N1, and N2 sleep and less time in N3 and REM sleep due to repetitive airway obstructions that lead to awakenings, disrupt normal sleep architecture, and prevent deeper sleep stages. Sleep latency (min), i.e., time from lights out to sleep onset, is also recorded, and sleep efficiency (%) is calculated as [(sleep time [min] ÷ time in bed [min]) × 100].

Respiratory events are classified as apneas (obstructive, central, or mixed) and hypopneas (The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications, 2015), depending on the severity of airflow reduction (Table 19.2). Based on the assessment of respiratory events, the apnea index (AI), the hypopnea index (HI), and the apnea‐hypopnea index (AHI) (total, non‐REM, and REM) are calculated as the number of corresponding respiratory events per hour of sleep [respiratory events (total number) ÷ sleep period (h)].

The respiratory disturbance index (RDI) can also be calculated; it is based on the number of apneas and hypopneas just like the AHI but also includes respiratory effort‐related arousals and usually receives slightly higher values compared to the AHI. Oximetry indices, such as average and minimum oxygen saturation (SaO₂) are also recorded during PSG, and respiratory events accompanied by a decrease in SaO₂ equal or greater than 4% during sleep, compared to SaO₂ during wake period, are recorded to calculate the oxygen desaturation index (ODI) as [oxygen desaturation events (total number) ÷ sleep period (h)]. Electromyography of the anterior tibialis is also often performed to assess limb movements that might alter sleep stage or respiration and body position is monitored because of the position‐specific nature of OSA in many patients. Other parameters, such as average and highest heart rate, including episodes of bradycardia, asystole, tachycardia, atrial fibrillation, snoring, and any unusual or atypical behavioral events occurring during sleep and/or wakefulness are also recorded during PSG. Figure 19.3 provides an example of the signals recorded during an overnight PSG.

According to the American Academy of Sleep Medicine International Classification of Sleep Disorders (International Classification of Sleep Disorders, 2014), the diagnosis of OSA requires the presence of signs‐symptoms (e.g., arousals accompanied by breathing difficulty or choking, snoring or witness of apneas during sleep, and daytime sleepiness) or co‐morbidities (e.g., hypertension, coronary heart disease, atrial fibrillation, congestive heart failure, stroke, diabetes mellitus, cognitive impairment, or psychiatric disorder), in combination with more than five, mainly obstructive, respiratory episodes per hour of sleep during a sleep study. Alternatively, OSA can be diagnosed in the presence of ≥15 respiratory events per hour of sleep even in the absence of associated signs‐symptoms or co‐morbidities. In clinical practice, the AHI, i.e., the number of apneas and hypopneas per hour of sleep, is the most important PSG measure, based on which OSA diagnosis and severity is evaluated (Sateia, 2014). An OSA diagnosis is established when the AHI receives values ≥5 events/h and the disease is further classified as mild (5–14 events/h), moderate (15–29 events/h) or severe OSA (≥30 events/h). OSA severity evaluation based on ODI and RDI values is similar to AHI classification, but these indexes are not widely used for OSA diagnosis and severity assessment in clinical practice.

Although an overnight attended PSG is the gold‐standard method for OSA diagnosis and provides definitive results, the procedure is expensive, time consuming, burdensome for most patients, and requires specialized personnel and equipment. Simplified home sleep studies using portable devices also provide basic information on sleep architecture and respiratory events and can alternatively be used for OSA diagnosis. Available evidence suggests that a home‐based diagnosis, if properly performed, is not inferior to a laboratory diagnosis and can be of great value for individuals with mobility problems who have difficulty accessing a sleep center or when an attended PSG is not available (Masa et al., 2011; Mulgrew, Fox, Ayas, & Ryan, 2007; Rosen et al., 2012).

PREVALENCE

Even though the first reports of OSA date back to antiquity (Kryger, 1985), epidemiological data on the prevalence of the disease were not available until the early 1990s. The Wisconsin sleep cohort study, a prospective epidemiological study initiated in 1988 with the aim of investigating the prevalence and natural history of sleep‐related breathing disorders, is a landmark for the epidemiology of OSA (T. Young et al., 1993). According to analyses of 602 US men and women aged 30 to 60, the prevalence of OSA—defined as the presence of ≥5 apneas‐hypopneas/h based on PSG combined with increased daytime sleepiness—was 4% in men and 2% in women. Subsequent epidemiological studies in the USA, Europe, Australia, and Asia (Franklin & Lindberg, 2015; Punjabi, 2008; Senaratna et al., 2017) have shown that the prevalence of OSA in the general population, defined as AHI/ODI ≥5 events/h, ranges from 9% to 36%; disease rates are higher in men (from 13% to 33%) compared to women (from 6% to 19%) and increase with age, reaching 90% and 80% in older men and women, respectively. According to a comprehensive review of epidemiological studies examining global estimates of OSA prevalence published in 2019 (Benjafield et al., 2019), 936 million adults aged 30 to 69 years have mild‐to‐severe OSA and 425 million have moderate‐to‐severe disease; the number of affected individuals was highest in China, followed by the USA, Brazil, and India.

It is worth mentioning, however, that despite the widespread recognition of OSA as a major public health problem, available estimates of its prevalence probably underestimate its true extent, given that a large part of the general population (approximately 20%) experiences the respiratory manifestations of the disease in the absence of accompanying symptoms or co‐morbidities that usually lead to diagnostic sleep testing. Moreover, a large proportion of patients, even those with symptoms or those at increased risk for co‐morbidities, remain undiagnosed due to the practical difficulties of PSG.

PATHOPHYSIOLOGY AND RISK FACTORS

OSA has been traditionally considered a disease of abnormal upper‐airway anatomy and neuromuscular pathology, in which abnormal craniofacial structure or excess body fat decreases the size of the pharyngeal airway lumen and, combined with upper‐airway muscle dysfunction, leads to an increased likelihood of pharyngeal collapse during sleep (Eckert & Malhotra, 2008; Horner, 2008; Patil, Schneider, Schwartz, & Smith, 2007). The pharynx is a common organ for the respiratory system and the gastrointestinal tract, and it is involved in many vital body functions, such as swallowing, speaking, and breathing. It is unsupported by bones or cartilaginous tissue, but it contains several groups of muscles and soft tissues, along with a section that can be closed, which extends from the hard palate to the larynx (Ayappa & Rapoport, 2003). This ability of the pharynx to close is essential for speech and swallowing while awake, but it also provides the grounds for an undesirable obstruction under certain conditions during sleep.

During wakefulness, the upper airway is held open by the high activity of the numerous upper‐airway dilator muscles. After the onset of sleep, when muscle activity is normally reduced, the airway is susceptible to obstruction, especially in the presence of a detrimental upper‐airway anatomy. The size of the upper airways in the pharyngeal region depends on many factors, as the pharyngeal lumen is surrounded lengthwise by bones, such as the nasal bones, jaw bones, and the hyoid bone, as well as soft tissues, such as the tongue, the soft palate, the tonsils, the epiglottis, fat particles, and blood vessels. In general, a disproportionately high proportion of soft tissue mass, relative to the space available from the skeletal structures surrounding the pharynx, favors its obstruction (C. M. Ryan & Bradley, 2005). Patients with OSA often present with disorders of the skeletal structures surrounding the pharyngeal cavity, such as hypoplasia and/or dislocation of the jaws and displacement of the hyoid bone as well as enlargement of the soft tissues surrounding the pharynx, such as the thickening of the lateral walls of the pharynx (secondary to the accumulation of fat in the area, swelling of the pharyngeal mucosa due to vascular congestion or inflammation, or swelling of the veins in the area in the presence of fluid accumulation due to heart failure or chronic kidney disease) or the swelling of the soft palate and the tonsils (Eckert & Malhotra, 2008; Horner, 2008; Patil et al., 2007). In addition to this detrimental upper‐airway anatomy, compared to healthy subjects, patients with OSA also experience a greater reduction in genioglossus muscle activity, the most important upper‐airway dilator muscle, and its response to local stimuli during sleep. Upper‐airway muscle fatigue is also thought to be involved in OSA pathogenesis, caused by chronic muscle trauma and inflammation due to continuous low‐frequency vibrations of snoring and recurrent hypoxia (Eckert & Malhotra, 2008; Horner, 2008; Patil et al., 2007). Therefore, although healthy individuals experience only a small degree of respiratory instability at the onset of sleep, a person with a detrimental upper respiratory tract anatomy who relies more on muscle tone to support respiration is particularly prone to developing OSA when the function of the upper‐airway dilator muscles is compromised.

Although upper‐airway anatomy and muscle function are crucial for the development of OSA, accumulated research over the last decades has also identified several non‐anatomical and non‐neuromuscular factors that contribute to OSA pathophysiology in patients with apparently normal upper‐airway anatomy and very responsive upper‐airway dilator muscles. These factors include the instability of the central respiratory control system, increased tendency to wake up (low arousal threshold), and low lung volume (Eckert & Malhotra, 2008; Horner, 2008; Patil et al., 2007). In addition, the close relationship between OSA and cardio‐metabolic diseases has led the scientific community to reconsider its traditional view as merely an anatomical disease of the upper respiratory system and investigate cardio‐metabolic factors as potential pathogenetic mechanisms. In this context, the landmark theory of Vgontzas et al. (Vgontzas, Bixler, & Chrousos, 2003, 2005; Vgontzas, Gaines, Ryan, & McNicholas, 2016) suggests that cardio‐metabolic disorders in the context of the metabolic syndrome, i.e., central obesity, insulin resistance, dyslipidemia, and hypertension are implicated in OSA pathogenesis. In particular, while the majority of researchers argue that inflammation and oxidative stress resulting from intermittent hypoxia in the context of OSA lead to cardio‐metabolic morbidity, Vgontas et al. (Vgontzas et al., 2003, 2005; Vgontzas et al., 2016) suggest that the pathophysiology of the metabolic syndrome actually precedes the onset of OSA, with central fat deposition triggering upper‐airway dysfunction through insulin resistance, inflammation and oxidative stress, and that OSA should be essentially be considered as the manifestation of the metabolic syndrome in the respiratory system of obese individuals. This hypothesis is based on evidence that: i) obesity, and diseases in which insulin resistance and inflammation play a central role, such as diabetes mellitus, are associated with symptoms of OSA (e.g., daytime sleepiness) in the absence of a formal OSA diagnosis; ii) treatment of OSA‐related respiratory disorders is not accompanied by benefits in glucose metabolism, inflammation, and oxidative stress, suggesting that these disorders precede OSA rather than manifest as a result of the disease; iii) exercise can reduce OSA severity through improvements in inflammation and insulin resistance in the absence of weight loss; iv) the presence of the metabolic syndrome is associated with sarcopenia and skeletal muscle dysfunction and may consequently contribute to upper‐airway dilator muscle dysfunction; v) elevated leptin levels, a key feature of obesity, can disrupt central respiratory control; vi) the prevalence of OSA and the metabolic syndrome has a similar age distribution in the general population, reaching a maximum in middle age, a fact that supports their causal relationship; and vii) inflammation is present in patients with OSA regardless of the presence of obesity (Vgontzas et al., 2003, 2005; Vgontzas et al., 2016).

Risk factors for developing OSA resemble those of cardiovascular diseases and include increased age, obesity, male gender, smoking, and heavy alcohol consumption (Gharibeh & Mehra, 2010; Punjabi, 2008). OSA also exhibits a hereditary pattern, with up to 35% of the variation in disease severity being attributed to genetic factors regardless of age and body‐weight status (Gharibeh & Mehra, 2010; Punjabi, 2008).

Among modifiable risk factors, obesity has emerged as the most important factor leading to sleep‐disordered breathing and the progression of OSA. Epidemiological studies suggest that being overweight occurs in approximately 60% of patients undergoing diagnostic sleep testing. The Wisconsin sleep cohort study was the first large‐scale prospective study to test longitudinal associations between weight change and OSA indices (Peppard, Young, Palta, Dempsey, & Skatrud, 2000). In a sample of 690 participants from Wisconsin, USA who were followed for 4 years, compared to weight‐stable conditions, a 5%, 10%, and 20% weight gain predicted a 15%, 32%, and 70% increase in the AHI, respectively, while weight loss was associated with analogous decreases in the AHI (Peppard et al., 2000). Similar observations were subsequently reported by the sleep heart health study among 2968 middle‐aged men and women from several US communities during a 5‐year follow‐up period, compared to weight‐stable individuals; those who gained weight experienced increases in the RDI in a dose‐response manner and vice versa (Newman et al., 2005). Several case‐control studies have also revealed a significant positive association between the presence of overweight/obesity and odds of OSA both in children and adults (Dong et al., 2020). Results of available interventional studies reinforce the epidemiological observations, as weight loss has been shown to improve the severity of OSA in most patients (these data are discussed below in detail). The involvement of obesity in the pathophysiology of OSA can be explained through several mechanisms, including: i) central fat accumulation, mainly in abdominal viscera and around the neck, which leads to increased mechanical load on upper airways; ii) increased leptin resistance, which can impair central respiratory control and lead to abnormal hypercapnic ventilatory response; iii) expression of pro‐inflammatory cytokines, which negatively impact upper‐airway neuromuscular control; and iv) a state of increased oxidative stress, which reduces the force‐generating capacity of upper‐airway muscles (Figure 19.4) (Dobrosielski, Papandreou, Patil, & Salas‐Salvado, 2017). It is also worth noting that, regardless of total body mass, an increasing number of studies highlight the potential aggravating role of upper body fat deposition in the development of OSA, in particular visceral adipose tissue (Kritikou et al., 2013; Vgontzas et al., 2000), a metabolic active tissue associated with cardio‐metabolic disorders, and the adipose tissue that accumulates in the area of ​​the upper airways (Mortimore, Marshall, Wraith, Sellar, & Douglas, 1998; Pahkala, Seppa, Ikonen, Smirnov, & Tuomilehto, 2014), which, as mentioned above, contributes to the thickening of the lateral walls of the pharynx.

COMPLICATIONS

Disruption of the normal sleep architecture is the main short‐term complication of OSA. As shown in Table 19.1, sleep is divided into NREM and REM sleep, while NREM sleep is further subdivided into three stages (N1 to N3), which represent the successive transition from lighter to deeper sleep (Kales, Rechtschaffen, University of California, Brain Information, & Network, 1968). Under normal conditions, in adults, N1 sleep accounts for 5%, N2 sleep for 50% to 65%, N3 sleep for 15% to 20%, and REM sleep for 15% to 20% of total sleep (Karacan, 1970). In OSA, these patterns are significantly disturbed, with the duration of N1 and N2 sleep being increased, and the duration of N3 (slow‐wave sleep) and REM sleep being significantly reduced (Jordan et al., 2014) due to repeated interruptions of breathing leading to awakenings. Given the significant disruption of normal sleep architecture, OSA is typically characterized by a wide variety of daytime symptoms, including sleepiness, fatigue, low vitality, inability to concentrate, impaired reflexes, and memory deficits (Bucks, Olaithe, & Eastwood, 2013; Wallace & Bucks, 2013). The most prevalent daytime symptom is sleepiness, which is evident in most patients when at a relaxed state, e.g., after lunch, when reading, or watching TV, and in severe OSA manifests as involuntary sleep episodes during social circumstances (e.g., theater or cinema), work environment (e.g., meetings), and driving (Black, 2003; Pagel, 2008). OSA daytime symptoms collectively lead to reduced functionality and productivity that can negatively affect personal relationships or work efficiency and contribute to psychological disorders (Brown, 2005), while most patients are prone to self‐injuries and accidents (Ellen et al., 2006; Rodenstein, 2009).

Besides daytime symptomatology, OSA is strongly linked to cardiovascular morbidity. This association is partly because OSA and cardiovascular disease share common risk factors, as discussed earlier in this chapter. However, OSA has also been suggested as an independent risk factor for cardiovascular disorders, a theory based on its adverse effects on the cardiovascular system, which is exposed to a vicious cycle of hemodynamic, oxidative, inflammatory, and metabolic disorders during sleep‐disordered breathing (Figure 19.5) (Bradley & Floras, 2009). Overall, the literature agrees that OSA is a major cause of cardiovascular morbidity and mortality and not just a respiratory disorder (Bradley & Floras, 2009; Kasai, Floras, & Bradley, 2012; Sanchez‐de‐la‐Torre, Campos‐Rodriguez, & Barbe, 2013). Several health organizations, such as the European and the American Heart Association, recognize OSA as a modifiable risk factor for cardiovascular disease (Piepoli et al., 2016; Somers et al., 2008).

Intermittent hypoxia plays a central role in the pathogenesis of cardio‐metabolic complications of OSA, due to its effects on the central nervous system and its various oxidative, inflammatory, and metabolic consequences (Figure 19.6) (Bonsignore, McNicholas, Montserrat, & Eckel, 2012; Bradley & Floras, 2009). During normal sleep, the parasympathetic activity of the central nervous system outweighs the sympathetic one; however, intermittent hypoxemia and the subsequent awakenings in the context of OSA cause an abnormal increase in sympathetic activity during sleep, which is followed by an increase in catecholamine levels. Hyperactivity of the sympathetic nervous system is accompanied by an increase in blood pressure, a disorder that often persists during the day and, in the long term can lead to the establishment of hypertension. This increase in blood pressure levels, combined with the fluctuating intrathoracic pressure due to respiratory effort and the reduction of oxygen supply to body tissues due to hypoxemia, significantly impairs myocardial and vascular function and increases the risk of cardiac arrhythmias, heart failure, and ischemic stroke (Bonsignore et al., 2012; Bradley & Floras, 2009). In addition, recurrent episodes of oxygen desaturation and re‐oxygenation within OSA lead to the production of oxygen free radicals and the establishment of an environment of increased oxidative stress, which in turn results in macromolecular peroxidation, changes in cellular homeostasis, and increased expression of inflammatory genes controlled by the nuclear factor κB. This change in the expression of inflammatory genes results in an increased production of inflammatory markers, mainly tumor necrosis factor‐α, C‐reactive protein, interleukin‐6, and interleukin‐8 as well as a decreased production of anti‐inflammatory markers, such as adiponectin. The combination of sympathetic nervous system hyperactivation, oxidative stress, and chronic subclinical inflammation leads to long‐term endothelial dysfunction and arterial stiffness, factors that play a central role in the establishment and development of atherosclerosis, the background of cardiovascular disease (Bonsignore et al., 2012; Bradley & Floras, 2009). Finally, chronic sleep deprivation or poor sleep quality, combined with oxidative stress and chronic inflammation, also adversely affect the body’s metabolic homeostasis, triggering insulin resistance. Insulin resistance, combined with obesity, which is present in the majority of patients with OSA, lead to increased adipose tissue lipolysis and increased availability of free fatty acids to the circulation, which are stored ectopically in the liver and lead to hepatic steatosis, increased de novo lipogenesis, and eventually to atherogenic dyslipidemia, which manifests as elevated triglycerides (TG), total cholesterol (TC), low‐density lipoprotein cholesterol (LDL‐cholesterol), and very low‐density lipoprotein cholesterol (VLDL‐cholesterol), and low levels of high‐density lipoprotein cholesterol (HDL‐cholesterol) (Bonsignore et al., 2012; Bradley & Floras, 2009).

The hemodynamic, oxidative, inflammatory, and metabolic complications of OSA lead to an overall increased cardio‐metabolic risk. This hypothesis is supported by numerous cross‐sectional epidemiological studies showing an increased prevalence of cardiovascular diseases in patients with OSA and vice versa (Floras, 2018; Sarkar, Mukherjee, Chai‐Coetzer, & McEvoy, 2018) as well as prospective epidemiological studies that reveal a strong positive relationship between the presence and severity of OSA and the risk of hypertension, diabetes mellitus, arrhythmias, coronary heart disease, heart failure, stroke, cardiovascular, and total mortality (Hou et al., 2018; Wang, Bi, Zhang, & Pan, 2013; Xie, Zhu, Tian, & Wang, 2017).