Article Text

Download PDFPDF

Asthma and Obstructive Sleep Apnea
  1. Swathy Puthalapattu, MD*,
  2. Octavian C. Ioachimescu, MD, PhD
  1. From the *Topeka Veterans Affairs Medical Center, Topeka, KS; and †Atlanta Veterans Affairs Medical Center and Emory University, Atlanta, GA.
  1. Received November 2, 2013, and in revised form January 7, 2014.
  2. Accepted for publication January 14, 2014.
  3. Reprints: Octavian C. Ioachimescu, MD, PhD, Emory University, Atlanta Veterans Affairs Medical Center, Sleep Medicine (111), 1670 Clairmont Rd, Decatur, GA 30033. E-mail: oioac{at}yahoo.com.
  4. Conflict of interest: none relevant to this study.

Clinical and Pathogenic Interactions

Abstract

Asthma and obstructive sleep apnea (OSA) are among the most prevalent chronic human diseases of the 21st century. They share several risk and aggravating factors such as obesity, smoking, gastroesophageal reflux, sinonasal disease or upper airway involvement, systemic inflammation, etc. Although the association between OSA and chronic obstructive pulmonary disease or “overlap syndrome” is better known and characterized, the association of asthma and OSA or “alternative overlap syndrome” is less clearly defined and understood. Nevertheless, their coexistence has synergistic effects on patient symptoms, response to therapy, and general outcomes. Taxonomically, asthma and OSA are syndromically defined entities that are quite heterogeneous, being characterized by a plethora of clinical phenotypes. The complex interactions between these conditions should take into account more specific etiopathogenic mechanisms or distinct disease endotypes. The potential clinical, pathogenic, and therapeutic significance of the disease endotypes is still emerging and needs further evaluation. We present here a review on the bidirectional relationships between asthma and OSA, including their clinical, pathophysiologic, and therapeutic connections. Furthermore, we propose here to look at these interactions beyond the development of comprehensive inventories of genotypes, clinical and pathophysiologic phenotypes, but in the larger context of obstructive lung and airway disorders, with the goal to reassess meaningful syndromes based on natural history and predictable patient outcomes, which will help us better stratify therapy in an era of personalized medicine.

Key Words
  • asthma
  • airflow limitation
  • airway disease
  • obstruction
  • COPD
  • sleep apnea
  • obstructive sleep apnea
  • overlap syndrome
  • phenotype
  • endotype

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key Words

Bronchial asthma is a major health problem, affecting approximately 300 million individuals of all ages and racial or ethnic groups, and being the most common chronic pediatric disease worldwide. Approximately 250,000 people die each year prematurely because of asthma.1,2Bronchial asthma is a common and complex disease, characterized by chronic inflammation, airway hyperresponsiveness (AHR), and (often reversible) expiratory flow limitation. Various nosologic presentations and clinical manifestations have been lumped into disease “phenotypes,” which are end results of a complex interplay between the host’s genetic making and the environment. Although distinct phenotypes (such as allergic or late-onset asthma) have been traditionally defined by the constellation of clinical features, biomarkers, and/or effective therapeutic modalities available, a disease phenotype generally does not imply a specific underlying pathophysiology or distinct course. Different pathogenic mechanisms carve more specific disease subcategories called endotypes or endopathogenic phenotypes, which are characterized by recognizable and (presumably) specific mechanistic relationships.3

Obstructive sleep apnea (OSA) is a prevalent condition in the general population.4,5It involves the upper airways (nasal, oral, and pharyngeal passages) and manifests with inspiratory flow limitation during sleep (and likely not only). Obstructive sleep apnea syndrome is frequently defined polysomnographically as an apnea-hypopnea index (AHI) of 5 or more, in combination with daytime symptoms such as excessive sleepiness.

Although the association between OSA and chronic obstructive pulmonary disease (COPD) or “overlap syndrome” is better known and characterized, the association of asthma and OSA (or “alternative overlap syndrome”) is less clearly defined and understood.6Several epidemiological studies suggested an “association” between asthma and OSA symptoms or syndrome.7–16The association seems to be seen more often than expected as pure intersection of 2 very prevalent conditions. In this bidirectional interaction, obesity and gastroesophageal reflux may play significant copathogenic roles (Fig. 1). The prevalence of obesity, a major risk factor and comorbidity of OSA, has been increasing significantly over the past decades, especially in the developed countries. Obesity seems to play a role in the symptomatic manifestations (morbidity) and the poor therapeutic response of persons with asthma.17–22

FIGURE 1.

Complex interplay between genetics, environment, local and systemic inflammation, GERD, obesity, and other comorbidities in the relationship between OSA and bronchial asthma.

We propose here to characterize the endotype defined by the comorbid association of OSA and an obstructive lung disease such as asthma (alternative overlap syndrome). The asthma-OSA endotype’s existence is supported by several arguments of dual interaction, common biologic pathways, and unique therapeutic responses. The use of an asthma-OSA endotype has the potential to lead us in the near future to personalized and, hopefully, more effective therapies for persons with asthma.

EPIDEMIOLOGY

OSA in Asthma

Several studies examined the association between OSA and asthma. These studies could be subclassified as populational or clinic based, whereas OSA is defined by self-reported symptoms or by objective testing (eg, polysomnography). Understandably, the study of an association between OSA and asthma based on self-reported symptoms has its inherent flaws because nocturnal symptoms of breathlessness, wheezing, and cough are not specific and sometimes difficult to discern in the setting of standardized questionnaires administered to subjects from general population or specialized clinics, with 1, 2, or none of these conditions.

Population-Based Studies With OSA Defined by Self-Reported Symptoms

Persons with asthma of all ages and body weights snored more often than the controls in a British population23(29% vs 9%). In the European Community Health Respiratory Survey, self-reported habitual snoring and apnea were seen significantly more often in persons with asthma versus in controls without asthma (14% vs 9% and 3% vs 1%, respectively), independent of age, body weight, sex, or smoking status.24Among the 677 atopic women, 34% reported snoring at least once a week and 13% reported snoring 5 to 7 nights per week.25Snoring was associated with symptoms or current diagnosis of asthma (odds ratio [OR], 1.8) and African American race (OR, 1.6) after controlling for income level and smoking status.25Furthermore, in a population-based study that included 2713 subjects with asthma symptoms, health-related quality of life was adversely affected by snoring and witnessed apneas.26

Population-Based Studies With OSA Defined by Polysomnography

One of the more conclusive studies that looked at this association is the recent longitudinal analysis of Wisconsin Sleep Cohort study.15Among the 773 adults enrolled in the study who had baseline polysomnography with an AHI less than 5 (in 1988) and who underwent repeated polysomnography 8 years later, adult-onset and childhood-onset asthma predicted the risk of developing OSA with an OR of 1.48 and 2.34, respectively. A 5-year unit of asthma duration correlated with a 10% increase in odds (OR, 1.1) of developing subsequent OSA.15

Clinic-Based Studies With OSA Defined by Self-Reported Symptoms

Several clinic-based studies on patients with asthma11–13,27showed high prevalence of self-reported occasional, habitual, or frequent snoring and witnessed apneas. For example, the high prevalences of self-reported snoring (86%), habitual snoring (38%), and witnessed apneas (31%) were found by Teodorescu et al.11in a study on 115 adult persons with asthma. In another study,13high OSA risk was associated with 2.8 times higher odds of not-well-controlled asthma after adjusting for obesity and other factors relevant for asthma control. In a cohort of persons with asthma,12they were more likely to report frequent snoring (18.5% vs 8.0%) and chronic daytime sleepiness (46.1% vs 34.3%) versus patients without a diagnosis of asthma. In a large group (n = 752) of persons with asthma seen in specialized clinics at 2 academic institutions,27high OSA risk was associated with persistent daytime and nighttime asthma symptoms. After adjusting for covariates, diagnosed OSA remained as a predictor of persistent daytime (OR, 2.08) but not of nighttime asthma symptoms. Similarly, in univariate analysis, continuous positive airway pressure (CPAP) use was associated with lower odds of persistent daytime asthmatic symptoms (OR, 0.5; P = 0.049) but not of nighttime symptoms.27

Clinic-Based Studies With OSA Defined by Polysomnography

A number of studies found a high prevalence of OSA (diagnosed polysomnographically) in patients with severe asthma. In a study of 22 persons with difficult-to-control asthma receiving long-term or frequent bursts of oral corticosteroids, 21 (95%) were diagnosed with OSA.9More recently, OSA was found to be significantly more prevalent among patients with severe asthma (88%) versus patients with moderate disease (58%) or controls (12%).16In both studies, the very high prevalence of OSA was unexpected for the degree of obesity.9Obstructive sleep apnea was also linked to worse asthma flares (frequent exacerbations and difficulty in achieving disease control). In 63 patients with difficult-to-treat asthma, OSA was shown to be an important risk factor (OR, 3.4) for frequent exacerbations in the prior year.10

Finally, CPAP therapy for coexistent OSA was found to improve asthma control. Chan et al.28assessed CPAP therapy in 9 persons with asthma with severe nocturnal symptoms and comorbid OSA diagnosed by polysomnography.29A 2-week course of CPAP therapy was associated with decreased frequency of asthma symptoms, bronchodilator use, and improved peak expiratory flow rates. The cessation of CPAP returned flow rates to levels before therapy. Guilleminault et al.30studied 10 men with OSA and predominantly nocturnal asthma; more than 6 months of CPAP therapy abolished nocturnal asthma attacks. Five subjects with both daytime and nighttime asthma symptoms had regular snoring and more negative inspiratory esophageal pressures during sleep (indicative of increased upper airway resistance). In these patients, 6 months of CPAP therapy eliminated nocturnal asthma and reduced daytime attacks. In a more recent study on 16 patients with nocturnal asthma symptoms, Ciftci et al.31found that 2 months of CPAP therapy for coexistent OSA resulted in significant reduction in nocturnal asthma, without significant concomitant spirometric changes. Furthermore, Lafond et al.32reported improved asthma-specific quality of life after 6 weeks of CPAP for OSA but no reduction in bronchial reactivity in 20 patients with stable asthma.

Asthma in OSA

In a study involving 606 subjects, the prevalence of asthma in patients with sleep apnea was 35%.33In the Australian Busselton Health Study, asthma emerged as an independent risk factor for the development of habitual snoring.14The study included 967 adults who reported “no snoring” in the 1981 Busselton Health Survey and who participated again in the 1994 to 1995 follow-up survey. Approximately 13% became habitual snorers at follow-up. Male sex (OR, 3.5) and baseline body mass index (BMI; OR, 1.4 per additional 3.4 kg/m2) were significant predictors of new habitual snoring. A change in BMI over the 14-year period (OR, 1.5 per additional 2.3 kg/m2), incident asthma (OR, 2.8), and starting smoking (OR, 2.2) were other significant independent risk factors for development of habitual snoring.

Positive airway pressure (mainly CPAP) therapy for OSA syndrome has been shown to improve asthmatic symptoms in patients with both conditions,28,30–32pulmonary function tests,28,34,35and disease-related quality of life.32Nevertheless, these benefits still await definitive confirmation in larger studies.

PATHOPHYSIOLOGY

The bidirectional interactions between asthma and OSA could be classified as direct or indirect effects, mechanical, neural, or biologic in nature (Figs. 2 and 3).

FIGURE 2.

Several theories in support of a connection between asthma and OSA in this direction, that is, asthma leading to or exacerbating/aggravating OSA (further details in text).

FIGURE 3.

Several theories in support of a connection between OSA and asthma in this direction, that is, OSA leading to or exacerbating/aggravating asthma (further details in text).

Asthma Effects on OSA

The asthma effects on OSA are shown in Figure 2.

Direct Effects

Asthma could detrimentally affect the upper airway patency in several ways, as explained by several theories—sleep restriction, mechanical effects, respiratory phase interdependence, inflammatory theory, etc.

Sleep Architecture Theory

If uncontrolled, asthma can lead to frequent arousals, sleep fragmentation, and sleep loss, which have been shown to worsen upper airway collapsibility.36,37Additionally, rapid eye movement (REM) sleep, a sleep stage during which OSA is generally more severe, may aggravate further airflow obstruction. Ballard et al.38found that functional residual capacity falls during sleep in both normal subjects and patients with asthma and that the hyperinflation of awake subjects with asthma is diminished during non-REM sleep and virtually eliminated during REM sleep. As such, sleep-related reductions in functional residual capacity may contribute to but do not account for all the nocturnal increase in airflow resistance observed in patients with asthma with nocturnal worsening. Overnight decreases in forced expiratory volume in 1 second and peak expiratory flow rate were significantly related to REM sleep in a study suggesting that patients with asthma may experience more severe bronchoconstriction during REM sleep.39Other studies have failed to show a significant REM sleep stage effect.40,41

In a recent study using forced oscillation technique (FOT) to measure airway resistance and by altering lung volumes in subjects with and without asthma, the airway resistance was increased and the reactance was decreased in persons with asthma, indicating airflow limitation or obstructive physiology. Interestingly, the increased total resistance seen in persons with asthma was not only due to elevated lower airway resistance but also increased upper airway resistance (ie, supraglottic airway). Changes in reactance during REM sleep confirm that the lungs are becoming stiffer with sleep, particularly in those with asthma. When bilevel positive airway pressure (BPAP) was applied, respiratory system resistance was decreased as compared with baseline in both healthy subjects and subjects with asthma. However, this effect was driven by a decrease in upper airway resistance, particularly in those with asthma.

Furthermore, some authors42found that the peak oxygen desaturation during REM, but not during NREM, was significantly higher in children with asthma, independent of other variables. This suggests that children with asthma may have a REM-related vulnerability trait that affects oxygenation independently of OSA.42

Tracheal Tug Theory

Generally, persons with asthma have greater reductions in lung volumes (functional residual capacity and end-expiratory lung volume) during sleep, especially during REM sleep stage,38which could attenuate the stiffening effect of tracheal tug on the pharyngeal upper airway segment, similarly to the effect of recumbent position or abdominal obesity.43This leads to more collapsible upper airway (“tracheal tug theory”).

Respiratory Phase Interdependence Theory

The size of the upper airway is the smallest during sleep. Intrarespiratory changes in pharyngeal cross-sectional area are highly dependent on the upper airway compliance, which is known to be higher in patients with OSA and obese individuals.44Tamisier et al.45have shown that the upper airway resistance increases during the latter part of the expiratory phase of the 3 breaths preceding an upper airway inspiratory collapse (apnea), suggesting that the inspiratory-expiratory flow interdependence plays a greater role than thought before. By extrapolation, this means that asthma, which is characterized by expiratory flow limitation and increased airway resistance during exhalation, can predispose the upper airway to greater collapsibility during inspiration (while asleep).

Inflammatory Theory

Asthma is associated with acute and chronic systemic inflammation,46which could affect the strength or force generation of the respiratory muscles,47including the upper airway dilator muscles. Similar to the connection between asthma and upper airway diseases such as rhinosinusitis and nasal polyposis, the mechanisms linking lower airway inflammation with sleep-related upper airway collapse are likely multiple and may explain a unified airway hypothesis. Several factors could be invoked for this link; these are as follows: spillover of inflammatory cytokines into systemic circulation, selective chemotaxis and preferential recruitment of specific pathways of defense (eg, neutrophilic inflammation), drainage of the respiratory secretions containing prophlogistic mediators, mechanical vibration (snoring)–induced pervasive airway changes, etc.

Indirect Effects

Corticosteroid Effects

Patients with hypercortisolism, such as Cushing syndrome or disease (eg, corticotropin-secreting pituitary adenomas), have been found to have a high OSA prevalence (50%).48Yigla et al.9studied prospectively a cohort of 22 consecutive patients with difficult-to-control asthma for a mean of 8.9 years, 14 on continuous and 8 on “bursts” of oral corticosteroids for asthma exacerbations. All subjects had polysomnography, and OSA was defined as respiratory disturbance index of less than 5 and typical complaints. In this cohort, 21 of 22 patients had OSA (95% prevalence), which was more severe in continuous versus intermittent use of oral corticosteroids.9Possible explanations for the observed findings are as follows: fat deposition in and around the upper airway as a result of systemic absorption,49myopathy of the airway muscle dilators,50worsening obesity, etc. Another study suggested that the use of inhaled corticosteroids in patients with asthma is correlated with the risk of OSA in a dose-dependent manner.51

Nasal Disease

The nose is the preferred breathing route during sleep, and nasal obstruction contributes to sleep disordered breathing in predisposed individuals.52,53Chronic or allergic upper airway disorders (eg, rhinosinusitis, nasal polyposis) are very common among persons with asthma (up to 95% of them have associated nasal disease),54whereas nasal congestion is a known risk factor for snoring in general population55and in persons with asthma.56Patients with allergic rhinitis have more often comorbid allergic asthma, whereas patients with nonallergic or perennial rhinitis tend to have associated nonallergic asthma.57By using acoustic rhinometry and peak nasal inspiratory flow measurements, a recent publication found significantly smaller nasal airway cross-sectional area and volume in persons with asthma versus controls, irrespective of allergy status.58These findings are concordant with other studies suggesting bidirectional relationships that feed into the “unified airway” model of respiratory inflammation, based on anatomical and histologic upper and lower airway connections. In support of this theory is the fact that nasal and oral pharyngeal lymphoid tissues are frequently hypertrophied in both asthma and OSA (especially in children), in the form of adenoidal or tonsillar hypertrophy; in these patients, adenotonsillectomy is frequently curative for OSA and ameliorative for asthma.59Several mechanisms may explain this relationship, which are as follows: aspiration of “inflamed” sinonasal secretions into the lower airways, vagal stimulation in the upper airways leading to reflex bronchospasm, nasal obstruction leading to mouth breathing and excessively dry air in the lower airways, bacterial colonization and toxin production (microbiota), and spillover of proinflammatory and bronchoconstrictive cytokines from the infected sinuses into systemic circulation. The most recent evidence seems to support the latter mechanism, that is, implication of the systemic inflammation.60

Smoking

Upper airway inflammation caused by smoking and manifested by neutrophilic cell infiltration, mucus production, and edema represents an additive factor to the chronic upper and lower airway inflammation of persons with asthma. The added inflammatory “burden” could thus increase the upper airway resistance and collapsibility during sleep, with worsening sleep apnea. A smokers’ asthma endotype may emerge, although little is known about its pathobiology at this time.61–63

OSA Effects on Asthma

The OSA effects on asthma are shown in Figure 3.

Direct Effects

Several connecting pathways from OSA to asthma exist; as such, a few pathophysiologic theories have been postulated.

Mechanical Theory

The inspiratory flow limitation seen during sleep in OSA combined with the expiratory flow impairment of asthma has multiplicative effects on respiratory function and patients’ symptoms. As such, OSA increases the resistive load on the lower airways,64fact superimposed on an already challenged airway, especially during sleep.65Furthermore, distally transmitted mechanical stress from snoring may detrimentally affect lower airways. As discussed earlier, the respiratory system resistance already elevated in asthma is in part due to the increased upper airway resistance (supraglottic area).

Neural Reflex Theory

Upper airway closure and snoring could trigger vagally mediated bronchoconstriction30,66and worsen AHR, for example, by altering chemical arousal threshold or to resistive loading.30An additional postulated trigger for reflex bronchoconstriction is stimulation of the carotid body by the accompanying hypoxia.67

Sleep Architecture Theory

Sleep deprivation and sleep fragmentation caused by OSA may also exacerbate nocturnal asthma,31potentially by increasing airway resistance and blunting the arousal response to bronchoconstriction.38

Intermittent Hypoxia Theory

In patients with OSA, repetitive airway occlusions during sleep lead to intermittent hypoxia and reoxygenation, with subsequent production of reactive oxygen species and complex oxidative stress cascade downstream, inflammation, sympathetic tone surcharges, and endothelial dysfunction.68,69Patients with OSA have increased exhaled 8-isoprostane and interleukin (IL) 670,71and a neutrophilic type of inflammation, both well correlated with OSA severity and the hypoxic burden.72These observations suggest that OSA may contribute to the noneosinophilic inflammatory phenotype seen in patients with difficult-to-control or severe asthma.73

Local Inflammatory Theory

Obstructive sleep apnea is associated with an upper airway inflammatory process that has the potential to influence lower airways. One proposed mechanism for airway inflammation in OSA is the mechanical stress exerted on the mucosa by snoring and the high negative pressures transmitted against a closed airway during obstructive apneas. This may enhance one’s susceptibility to bronchospasm, a major element in the pathogenesis of asthma.74,75

Systemic Inflammatory Theory

Obstructive sleep apnea leads to both systemic and local (upper airway) inflammation.69,76Additionally, sleep deprivation77–79and obesity80lead to an inflammatory “diathesis.” In general, asthma is characterized by bronchial mucosa inflammatory cell infiltration with eosinophils, T lymphocytes, and mast cells. Several cytokines have a central role in this inflammatory process, such as IL-4, IL-5, IL-6, IL-13, and tumor necrosis factor α (TNF-α). Tumor necrosis factor α increases the production of TH2 cytokines and is involved in the chemotaxis and recruitment of neutrophils, eosinophils, and additional T lymphocytes. Although the TNF-α–centered inflammatory cascade seems to be activated in both obesity and severe or uncontrolled asthma, it is possible that this pathway is up-regulated in some asthma-obesity phenotypes.

Conceivably, independent of obesity, systemic inflammation induced by OSA might promote lower airway inflammation and asthma. Indeed, a recent study found increased central airway bronchial wall thickness in patients with nonasthmatic OSA.81Mehra et al.82suggested that OSA could cause oxidative stress and inflammation in the lower airways and that the proinflammatory effects of 1 disorder could influence the expression of the other. Indeed, IL-6, a proinflammatory cytokine secreted by T cells and macrophages, is elevated in obese patients and may play a role in both OSA and in asthma.83Finally, TNF-α, a cytokine involved in systemic inflammation, is also elevated in OSA, independent of body weight, and may play a role in the pathogenesis of asthma in some cases.84

Vascular Theory

Vascular endothelial growth factor (VEGF) is a hypoxia-sensitive glycoprotein that stimulates vessel growth.85An increasing body of evidence indicates that VEGF may also play an important role in the asthma pathogenesis and may contribute to bronchial inflammation and hyperresponsivenes85and/or vascular remodeling.86Furthermore, increased VEGF levels in patients with asthma seem to be correlated with the degree of airway obstruction.87Recent studies also indicate that patients diagnosed with OSA syndrome have elevated concentrations of VEGF that correlate with disease severity, as assessed by AHI and the degree of nocturnal oxihemoglobin desaturation.88,89Free oxygen radicals and endothelin, which were found to be elevated in patients with OSA, may also enhance the gene expression of VEGF.89,90Although the connection is plausible, no conclusive data exist yet on the relationship between elevated VEGF levels in patients with OSA with coexistent asthmatic airway inflammation.

Leptin Theory

A relatively novel hypothesis is about a potential role of leptin in the pathogenesis of asthma. Airway epithelial cells express receptors for the adipokines leptin and adiponectin, suggesting that airway epithelium may be biologically modulated by these mediators. Indeed, in an elegant study on subjects undergoing bariatric surgery for obesity, visceral fat leptin expression was significantly correlated with AHR.91In an animal model, the administration of leptin to mice increased both AHR and serum IgE levels, suggesting a role played by leptin in mast cell activation.92Even after controlling for body weight, serum leptin level was noted to be increased in male children with asthma compared with controls without asthma.93Furthermore, plasma and airway leptin levels were found to be the highest in obese adults with asthma versus controls with asthma who are not obese, obese, overweight, or lean.94In ex vivo studies, human alveolar macrophages from overweight and obese patients with asthma were found to be uniquely sensitive to leptin, inducing significantly the production of proinflammatory cytokines.95Consequently, in the context of higher levels of soluble leptin, it was posited that this macrophage phenotype may contribute to the pathogenesis of airway disease associated with obesity.95

Obese male patients with OSA were also found to have serum leptin levels higher than those of obese men without OSA.96One possible mechanism may be that elevated leptin levels are the result of intermittent hypoxia seen in OSA, although the exact mechanism is not clear.97Coupled with the increased levels of serum leptin observed in OSA, the proinflammatory effects of leptin suggest that this hormone might be relevant to asthma exacerbations in OSA. The association of increased leptin levels, AHR, and (local or systemic) inflammation in patients with OSA could be revealing of important pathogenic links between this condition and asthma.

Cardiac Dysfunction Theory

It is well documented that OSA induces or worsens cardiac dysfunction.98Congestive heart failure could cause airway obstruction and, therefore, may worsen preexisting asthma conditions (“cardiac asthma”). One crucial component of bronchial narrowing in heart failure is hyperresponsiveness to cholinergic stimuli, with subsequent constriction of airway smooth muscles.99Possible explanations for this phenomenon include the following: down-regulation of pulmonary β-receptors with concomitant decreases in adenylcyclase activity, which results in significant attenuation of cyclic adenosine monophosphate–mediated airway relaxation100; nonspecific bronchial C-fiber activation; thickening of bronchial walls; changes in epithelial sodium and water transport; and increased endothelin levels.101

Indirect Effects

Nasal Disease

It has been shown that mechanical trauma on the upper airway produced by snoring and repetitive closure reopening of the airway triggers local inflammation of the nasal and pharyngeal mucosa, with cellular infiltration and vascular congestion.76This in turn will lead to an increase in upper airway and total respiratory resistance, with potential deleterious effects on lower airways during expiration and worsening airflow limitation.

Smoking

Although smoking is an independent risk factor for OSA and gastroesophageal reflux disease (GERD), tobacco or marijuana smoke may trigger symptoms of wheezing, cough, and sputum production, all suggestive of asthma or, more commonly, chronic bronchitis.102Interestingly, prenatal maternal smoking has been consistently associated with early-life wheezing,103and this effect seems to be augmented by continuous postnatal exposure. A dose-response relationship was found between prenatal maternal smoking intensity and decrease in airways’ caliber during infancy and early life.104Similarly, air pollution may adversely affect adult asthma but more frequently worsens preexistent disease rather than causing new-onset asthma.105

THE “INNER CIRCLE” OR TRIANGLE OF ASTHMA, OSA, AND OBESITY

By far, obesity is the most important risk factor for OSA,106as multiple epidemiological studies show a strong and consistent association between these conditions. Two major etiopathogenic theories have been postulated, which are as follows: (1) abdominal obesity causes upper airway collapsibility (OSA) through a loss of tracheal tug, and (2) disturbed sleep architecture and intermittent hypoxia of OSA lead to neurohormonal abnormalities such as leptin-ghrelin hormonal changes, subsequently more voracious appetite, increased oral intake, and worsening obesity (and subsequently more symptomatic/difficult to control asthma).

Mounting evidence also implicates obesity as a major risk factor for asthma (Fig. 1).56,107,108Almost all cross-sectional studies published to date show that asthma is more prevalent in obese or overweight individuals.109Some longitudinal studies suggest that obesity precedes the onset of asthma110–112and that the relative risk of incident asthma increases with higher body weight.113,114Obesity also leads to more frequent or more severe exacerbations and more difficult-to-control disease.17,115Conversely, weight loss seems to improve asthma control and lung function.116–118In human subjects and in mice, obesity seems to predispose individuals to AHR, an important component of asthma.109In addition, in obese persons with asthma, weight loss was associated with a reduction in the severity of both asthma and OSA.116Although the exact mechanistic basis of this association is still unknown, several factors may contribute, such as psychologic effects, stress, exercise intolerance, diet, metabolism, etc. Obesity may also affect the small airways in the dependent portions of the lungs, leading to some airways to close and others to overdistend and ovestretch the bronchial smooth muscle cells, potentially leading to AHR. The connections between obesity and inflammation or AHR have been explored, although the results are mixed.109

A recent American Thoracic Society workshop stated that “asthma in the obese may represent a unique phenotype of asthma, with more severe disease that does not respond well to conventional therapy.”112In our quest for more specific phenotypes and endotypes, there seem to exist at least 2 distinct subtypes of asthma in obese; these are as follows: (1) early-onset (younger than 12 years and seen in both sexes), atopic, eosinophilic asthma with elevated titers of IgE that is in fact complicated by obesity, and (2) late-onset, nonatopic, noneosinophilic asthma that may be caused by obesity and found predominantly in middle-aged women119; the latter is increasingly shaping up as a distinct inflammatory, noneosinophilic endotype.118,120–123Although asthma-obesity phenotypes and endotypes are relatively new and still emerging, it would be of interest to explore in a systematic fashion the contribution of sleep-disordered breathing to a special asthma-obesity endotype.

THE OTHER INNER CIRCLE OR TRIANGLE OF ASTHMA, OSA, AND GERD

Obstructive sleep apnea is often associated with gastroesophageal reflux.124Conversely, asthma and (silent or overt) GERD are strongly correlated because one may induce or aggravate the other.125

The prevalence of GERD is very high in patients with OSA. Green et al.126and Valipour et al.127reported that GERD was a comorbidity in 62% and 58% of patients with OSA syndrome, respectively. Although GERD may be exacerbated by OSA, it also seems to be improved by CPAP therapy.128,129The disproportionately high prevalence of reflux in patients with OSA could be explained by several mechanisms, which are as the following: increased transdiaphragmatic pressure gradients or deep negative intrathoracic pressures swings,130,131vagally mediated bronchoconstriction,132,133transient lower esophageal sphincter relaxations caused by autonomic nervous abnormalities or arousals induced by apneas, or by adverse effects of antiasthma medications (β-agonists, corticosteroids, theophylline, etc) on the lower esophageal sphincter. Contrary to an earlier theory, recent studies found that the lower esophageal sphincter actually contracts during apneic episodes, thereby inhibiting gastroduodenal content reflux into the esophagus. Transient lower esophageal sphincter relaxations (possibly associated with arousal-induced deglutition and esophageal peristalsis) seem more likely to be the mechanism of reflux in patients with OSA.134–137

Among the 244 patients seen in specialty clinics for asthma,5637% reported habitual snoring and 40% were deemed high risk for OSA. Interestingly, independent predictors of habitual snoring included GERD (OR, 2.1) and use of inhaled corticosteroids (OR, 2.6), whereas high OSA risk was predicted by asthma severity step (OR, 1.5), GERD (OR, 2.7), and use of inhaled corticosteroids (OR, 4.0).56A recent prospective study from northern Europe138found that individuals with persistent nocturnal reflux were approximately 2 times more likely to report incident asthma (adjusted OR, 2.3) and new onset of (daytime or nighttime) respiratory symptoms such as wheezing, dyspnea, chest tightness, or cough, when compared with participants who never had nocturnal reflux symptoms. Of note, in this study, OSA was assessed by self-reported symptoms and not by polysomnography. Subjects with new-onset or persistent nocturnal GERD symptoms reported symptoms of OSA more often than those without reflux during the 9-year study period, independent of changes in BMI or age. The association between new nighttime reflux and incident OSA symptoms was stronger in men, whereas the association between persistent nocturnal GERD and new respiratory symptoms was stronger in women.138These differences may be important in explaining some of the gender-related distinctions seen in other publications, which investigated the relationship between asthma and obesity or other comorbidities, leading again toward more specific and nuanced phenotypes and endotypes. In an analysis of the Busselton population study,129which compared the prevalence of nocturnal reflux symptoms in patients with untreated OSA, patients with OSA using CPAP therapy, and general population, GERD symptoms were found to be related to the presence and severity of OSA, whereas CPAP therapy decreased significantly the prevalence of reflux symptoms.129

Gastroesophageal reflux may induce asthma directly by microaspiration, with respiratory mucosal injury by gastric (acid and pepsin) or duodenal (bile acids and trypsin) contents, and indirectly, via vagally mediated mechanisms or reflex bronchospasm.

In summary, as discussed previously, the interactions between host factors, environment, local or systemic inflammation, GERD, and obesity are complex; each one of these factors may play a role (directly or indirectly) in the phenotypic characterization and/or outcomes of patients with what was called6 alternative overlap syndrome, that is, comorbid OSA and asthma (Fig. 1).

DIAGNOSTIC STRATEGIES

Clinically, patients with asthma should be proactively assessed for coexistent sleep-disordered breathing. As such, asking the patient and the bed partner about snoring, witnessed apneas, choking, and gasping during sleep and complaints of excessive fatigue or sleepiness during wakefulness should be part of a comprehensive history in patients with asthma. As of today, specific questionnaires for this evaluation have not been validated yet. Vice versa, patients with sleep apnea may complain of nocturnal symptoms such as cough, paroxysmal dyspnea, or arousals with choking and gasping sensation, which may be difficult to differentiate from asthma symptoms. A functional evaluation with polysomnography or home sleep testing (oligosomnography139) may be warranted in these cases. Similarly, performance and utility validation of oligosomnography in this setting has not been done yet.

The presence of airflow obstruction or expiratory flow limitation is a defining feature of asthma. Its assessment can be done with spirometry (with or without methacholine challenge or bronchodilator response), techniques such as esophageal balloon catheterization (invasive and dependent on lung volume history), or by plethysmographic techniques. The development of technologies based on applied negative expiratory pressure have made possible the identification of airflow obstruction by comparing normal breathing with the breaths under negative pressure of −3 to −5 cm H2O during exhalation.140Any increase in flow suggests an expiratory airflow reserve, such as seen in asthma or COPD. If negative pressure is applied during inspiratory phase in a sleeping subject, the same technique can identify upper airway collapsibility or an equivalent assessment of airway propensity to collapse (inspiratory flow limitation). Another potentially useful technological advance is the use of FOT,141,142which is essentially the analysis of flow-volume curves under small forced oscillations imposed during tidal breathing and using an array of frequencies. Thus, the impedance of the respiratory system (Zrs) is basically a combination of airway resistance (Rrs) and reactance (Xrs); these measurements can be assessed during both respiratory phases and provide additional insight into the inspiratory or expiratory flow limitation of a particular patient.143

THERAPEUTIC IMPLICATIONS

In patients with significant OSA, CPAP remains the main therapeutic modality, improving daytime sleepiness and quality of nocturnal sleep. Barach and Swenson144observed that patients with asthma treated with CPAP had an increase in airway caliber of small and larger bronchi by approximately 1 and 2 mm, respectively; this lead to a heightened interest in its use in patients with acute asthma. Other studies reported a reduction in respiratory rate and dyspnea when using CPAP in patients with acute asthma.145,146Several small studies on adult patients with asthma and OSA revealed that CPAP therapy ameliorates nighttime asthmatic symptoms and improves their quality of life, without necessarily improving their respiratory function or AHR.28,31,32,107These effects could be attributed to an overall improved airway resistance, by increasing the mean airway pressure and minute ventilation, recruiting underventilated alveoli, stabilizing the upper airway, and/or by improving inspiratory muscle loading.147,148In overweight and obese persons with asthma, the application of BPAP during sleep leads to improvements in overall and upper airway resistance, as measured by FOT.35Interestingly, lower airway resistance was relatively unchanged after the application of BPAP, possibly because of small changes in lung volumes in this study.35

It is also well documented that CPAP therapy for OSA improves gastroesophageal reflux. This effect is thought to be the result of passive elevation of the intrathoracic pressure, constriction of the lower esophageal sphincter reflex, autonomic nervous system deactivation,128or simply averting respiratory events associated arousals. In patients with OSA syndrome, CPAP has been shown to reduce plasma levels of C-reactive protein and other proinflammatory cytokines after only 1 month of therapy.149Treatment of OSA with CPAP for 1 year has also been shown to be effective in reducing plasma VEGF concentrations, which may play a pathogenic role in bronchial asthma and in flow limitation of patients with asthma. Continuous positive airway pressure prevents the OSA-induced hypoxia, which may explain why CPAP may be effective in decreasing serum leptin levels in patients with OSA.150,151Further studies are necessary to analyze bronchial inflammation over longer periods and should include serial assessments of lower airway structure and function, local cytokines, and oxidative stress biomarkers.

CONCLUSIONS

The association of COPD and OSA or overlap syndrome and the comorbid existence of asthma and OSA or alternative overlap syndrome could likely be lumped together as obstructive lung disease and OSA (OLDOSA syndrome).6Beyond taxonomical reasons (ie, simplifying and better splitting disease entities in the future by using more advanced computational methods such as cluster analysis, systems and networks biology), this may also have diagnostic and therapeutic justifications. It is important to know that when asthma and sleep apnea coexist, they have a synergistic effect (asthma worsens OSA and vice versa). From a therapeutic point of view, several studies showed improvement in asthma symptoms and respiratory function after the initiation of positive airway pressure therapy. Current asthma guidelines recommend testing for OSA in overweight or obese patients with poorly controlled asthma. We have outlined here several potential pathogenic mechanisms explaining the interactions between asthma, OSA, and various risk factors or comorbidities. Practitioners should be aware of the relationship between asthma and OSA as a potential distinct asthma-OSA endotype and how important it is that sleep apnea be considered, diagnosed, and treated in patients with asthma.

References

View Abstract