Overall, published research has reported a distortion of sleep architecture, including decreased sleep efficiency, increased sleep fragmentation, and sleep-related hypoventilation in patients with IPF, while OSA was also frequently observed [7,8]. The reported prevalence of OSA in the literature is 6–91%, and several reports have suggested its close association with quality of life and prognosis in patients with IPF [9]. After the IPF diagnostic criteria were more clearly defined in 2002 by the American Thoracic Society/European Respiratory Society, studies on sleep disorders in IPF patients began to be conducted in earnest. Since the Cleveland Clinic Sleep Disorders Center reported an OSA prevalence of about 60% in a retrospective study of 18 IPF patients who underwent polysomnography (PSG), several additional studies have been conducted on sleep-disordered breathing (SDB) in IPF [8]. In most of these studies, the prevalence of OSA was significantly higher in IPF patients than in a healthy population of the same age [10,11]. The exact mechanism underlying the high prevalence of OSA in patients with IPF has not yet been established. Reduced lung volume has been considered one of the main causes [12]. In general, upper-airway resistance increases during sleep, while functional residual capacity (FRC) decreases. When seeking to elevate the lung volume beyond the decreased FRC, increased respiratory effort exerted via negative extrathoracic pressure frequently leads to collapse of the upper airway [13]. Therefore, OSA is more common in IPF patients than in the general population. Moreover, OSA is associated with various comorbidities such as pulmonary hypertension, gastroesophageal reflux disease, and diabetes; these comorbidities are important factors driving disease progression [14-16]. Recently, a single-center observational study was conducted in Italy that enrolled 35 IPF patients after PSG. They found that the more severe the OSA and hypoxemia, the higher the rates of mortality and clinical deterioration [17]. Although the small number of study patients was a limitation, there is growing evidence to indicate that OSA induces IPF aggravation and is closely associated with poor outcomes.

The epidemiologic evidence for an interaction between COPD and OSA is still under debate. Zhao et al. [18] showed that the prevalence of COPD was lower among patients with SDB than among those without SDB, while Soler et al. [19] reported a higher prevalence of OSA in patients with moderate-to-severe COPD. These contrasting findings are due to the nature of COPD, which is usually diagnosed when an irreversible airway obstruction is noted in a pulmonary function test. However, there is a wide spectrum of disease phenotypes associated with such abnormal airway obstruction. The heterogeneity of these diseases, which include an emphysema phenotype, chronic bronchitis phenotype, and elderly COPD, appears to be the reason for the difficulty in clarifying the relationship between COPD and OSA. It has recently been reported that the prevalence of OSA varies according to body mass index (BMI) and COPD type. Patients with COPD, those with dominant emphysema, and those with a low BMI may have a lower likelihood of OSA. On the other hand, patients with a higher BMI, those with a chronic bronchitis phenotype, and those with cor pulmonale are predisposed to OSA [20,21]. There are several pathological mechanisms linking OSA and COPD. Upper-airway inflammation associated with cigarette smoking [22] and myopathy of the upper-airway muscles caused by inhaled corticosteroids [23] may induce OSA in patients with COPD. Moreover, OSA exacerbates COPD by promoting inflammation of the lower respiratory tract [24]. Recently, combined COPD and OSA has been referred to as overlap syndrome, which constitutes a distinct disease entity; such patients show poor clinical characteristics and outcomes compared to COPD-only patients in general. The Spanish Sleep Cohort Study showed that COPD-OSA overlap syndrome patients not undergoing continuous positive airway pressure (CPAP) treatment had increased rates of hospitalization and death due to COPD exacerbation compared to a COPD-only group [25]. There is therefore a growing consensus that certain phenotypes of COPD are responsible for a high proportion of OSA, and that OSA could impact significantly on COPD deterioration and prognosis.

The most important clinical feature of IPF and COPD, and the accepted index of disease progression, is hypoxia. V/Q mismatch and diffusion impairment are the main mechanisms of hypoxia in the two diseases [26,27]. Fibroblast and myofibroblast proliferation and excessive deposition of disorganized collagen and extracellular matrix are the main pathological features of IPF, resulting in distortion of the normal lung alveoli, interstitial spaces, and vasculature structures [28]. In COPD, on the other hand, progressive airflow limitation due to airway inflammation and emphysematous destruction of the alveolar and pulmonary capillaries are observed during disease progression [27]. These changes gradually induce chronic hypoxia and, in advanced cases, eventually lead to pulmonary vasoconstriction and pulmonary hypertension.

Various mechanisms exist that may explain IPF and COPD deterioration. The most important mechanism is systemic inflammation arising from chronic hypoxia. In IPF, persistent hypoxia activates cytokines such as protein kinase C and interleukin (IL)- 1, which stimulate vasoconstriction, fibroblast expression, and leukocyte recruitment [29]. Recruited inflammatory cells release matrix metalloproteinases, transforming growth factor beta, and tumor necrosis factor alpha (TNF-α), and cause collagen deposition leading to IPF progression [30]. This vicious cycle between chronic hypoxia and fibrosis accelerates IPF progression. Several studies have also reported that endothelial cell dysfunction and remodeling due to chronic hypoxia cause microvascular injury, which is associated with IPF progression [31]. Similarly, chronic hypoxia in COPD patients induces systemic inflammation via increased levels of circulatory cytokines, such as TNF-α and IL-6 [32], and causes excessive oxidative stress [33]. Such systemic inflammation and excessive oxidative stress due to chronic hypoxia cause cardiovascular morbidity, as well as progression of COPD itself.

IH, a condition in which periods of normoxia alternate with hypoxia, is a representative pathognomic mechanism induced by OSA. Patients with OSA experience repetitive episodes of apnea and hypopnea during the sleep cycle, leading to excessive systemic inflammation and overproduction of reactive oxygen species. These critical processes during IH in OSA cause endothelial damage and further tissue damage [34]. Various comorbidities of OSA are linked to IH. Metabolic dysfunction and vascular complications in patients with OSA are usually caused by this effect [35] which is also closely related to systemic inflammatory markers such as TNF-α and IL-8, which are the major cytokines associated with disease progression in IPF and COPD [36]. The presence of such overlapping molecules may represent evidence that OSA is closely related to IPF and COPD progression, and that patients with combined OSA have a larger disease burden and increased systemic inflammation compared to those without OSA in IPF or COPD. In addition, as in OSA, IPF and COPD patients show sympathetic overactivity and endothelial dysfunction. Cardiovascular disease and pulmonary hypertension are frequently observed in IPF and COPD patients, as in OSA [9,37]. However, it remains unclear whether IH or chronic hypoxia promotes more severe systemic inflammation and vascular remodeling. In IPF and COPD patients, who are already predisposed to chronic hypoxemia and sympathetic over-activity, exposure to additional IH increases the duration and frequency of hypoxia and promotes further systemic inflammation, sympathetic over-activity, and vascular remodeling (Figure 1). Obviously, IPF and COPD patients with OSA have poorer clinical outcomes, such as higher levels of cardiovascular disease and decreased survival, compared to those without OSA [38,39].