Sleep and Pulmonary Diseases

William McDowell Anderson

Arthur Andrews

David G. Davila

INTRODUCTION

Disorders of the lung and chest wall have traditionally been classified according to their physiologic effects as measured by pulmonary function tests (PFTs) (Table 34-1). Obstructive lung diseases have a disproportionate decrease in the forced expiratory volume that can be exhaled in 1 second (FEV₁) when divided by the total amount exhaled over >6 seconds, or forced vital capacity (FVC). This results in hyperinflation of the lung and elevation of the residual volume (RV), functional residual capacity (FRC), and total lung capacity (TLC) (Fig. 34-1). RV is the volume at maximal exhalation, FRC is the relaxed end-expiratory volume during tidal breathing, and TLC is the amount of air in the lung at the end of maximal inhalation. The RV is the first lung volume to increase (air trapping) with increases in the FRC and TLC in more severe obstructive airway disease. That is, in obstructive lung diseases, the TLC is either normal or increased. The hallmark of obstruction is a reduction in the FEV₁/FVC ratio on spirometry. In contrast, restrictive lung diseases, whether they be intrinsic to the lung parenchyma, such as interstitial fibrosis, or extrinsic, such as chest wall abnormalities, such as kyphoscoliosis (KS), result in a reduced TLC and a normal FEV₁/FVC ratio. In severe parenchymal restrictive lung disease, the FRC, the RV, and the TLC can be decreased. In diseases associated with expiratory muscle weakness, the RV may be increased. Expiratory muscles are required to exhale from FRC to RV. Of note, the vital capacity, = TLC − RV, can be decreased in both obstructive and restrictive lung disease. In obstruction, the RV increases more than the TLC. In restriction, the TLC decreases more than the RV. The differences in impact of disease on pulmonary function help the clinician understand patient complaints during daytime activities as well as sleep.

Considerable research has been done to better understand the effects of sleep on obstructive lung diseases, such as chronic obstructive pulmonary disease (COPD). Improved monitoring has shown that sleep may have deleterious effects on oxygenation in sleep. This has enabled the development of noninvasive ventilation (NIV) to improve morbidity and mortality during acute exacerbations of COPD, as well as chronic respiratory failure.

Other than COPD, the clinician should understand the effects of sleep on other obstructive diseases, such as asthma and cystic fibrosis (CF).

In contrast to COPD, intrinsic restrictive diseases of the lung, such as idiopathic pulmonary fibrosis (IPF), have not reached the same success in therapy, whereas the problems in sleep are well documented. The polio epidemics early in the last century led to expanded knowledge of how extrinsic lung diseases of the chest wall could lead to respiratory insufficiency. This prompted the development of NIV, even before invasive techniques, such as endotracheal intubation, were instituted. These techniques have been applied not only to postpolio syndrome (PPS) and respiratory failure but also to the treatment of patients with most neuromuscular disorders.

OBSTRUCTIVE LUNG DISEASES

Sleep-related Ventilation and Oxygenation

Sleep is associated with a diminished responsiveness of the respiratory center to chemical, mechanical, and cortical inputs. These changes are more pronounced in rapid eye movement (REM) sleep, during which there is more variability in respiratory rate and tidal volume (1).
Therefore, one can expect the mandatory increase in PaCO₂ and reduction in PaO₂ of 2 to 8 mm Hg and 3 to 10 mm Hg, respectively. Regardless of the change in PaO₂, the SaO₂ declines only by <2% in this scenario, given the characteristics of the oxyhemoglobin dissociation relationship (2). In individuals with normal pulmonary function and at sea level, there is relatively little consequence to SaO₂. The impact of sleep in patients with COPD has been long-studied and, indeed, poses several challenging management considerations. Chronic bronchitis patients—the “blue bloaters”—are more likely to be hypoxemic and hypercapnic as compared to patients with advanced emphysema, the so-called “pink puffers” (3). Patients with COPD have been noted to demonstrate significant increase in mean pulmonary artery pressures and PaCO₂ levels when going from wakefulness to sleep (4). The tendency for worsening nocturnal hypoxemia in these patients, coupled with the pulmonary vasculature’s response of hypoxic vasoconstriction, predisposes to the development of pulmonary hypertension and cor pulmonale (5). These consequences may be partly explained by the effects of the potent pulmonary vasoconstrictor endothelin (ET)-1. Patients with COPD who desaturate at night have higher circulating levels of ET-1 in their sera than do those who do not desaturate (6).

FIGURE 34-1 Lung volumes for a normal individual (far left) are compared with inflated lung volumes in a patient with COPD and reduced volumes in patients with various disorders of the thorax. IC, inspiratory capacity; ERV, expiratory reserve volume. (From Bergofsky EH. Respiratory insufficiency in mechanical and neuromuscular disorders of the thorax. In: Fishman AP, ed. Pulmonary diseases and disorders. New York: McGraw-Hill, 1980:1563, with permission.)

Clinical Epidemiology

COPD is now recognized as our nation’s most rapidly growing health problem, ranking as the fourth most common killer (7). Patients with COPD must also endure the expected sleep-related increase in airways resistance, decreased intercostal muscle activity, and fall in FRC (1). Obstructive sleep apnea (OSA) and COPD often exist in the same patient in what was previously commonly termed the “overlap syndrome.” These patients with concurrent OSA and COPD are more prone to nocturnal hypoxia (Sao₂ <85%, >5% of total sleep time) with an odds ratio of 30 compared to 3.15 and 15.8 for COPD and OSA alone (4). Chaouat et al. (8) prospectively studied 256 patients with confirmed OSA. Of these patients, 11 % had evidence of obstruction on spirometry. The overlap patients were male, more hypoxemic, had higher PaCO₂ levels, and were found to have more elevated rest and exercise mean pulmonary artery pressures. Suspicion for significant sleep-disordered breathing (SDB) in patients with COPD should be heightened in the following circumstances (9):

Patients who desaturate during exercise
Nonobese patients with moderate to severe COPD with progressive decline in arterial blood gases (ABGs)
Hypercapnic patients with severe chronic bronchitis
Pulmonary and systemic hypertension
Congestive heart failure (CHF)

Asthma and Other Obstructive Lung Diseases

Asthma affects 14 to 15 million persons in the United States and is the most common chronic disease in childhood (10). In a large study of nocturnal symptoms in 7,729 asthmatic patients in the United Kingdom, 74% reported awakening at least once per week and 64% awakened at least three times per week (11). Patients often underestimated the severity of their asthma and were more prone to
having allergies; no particular asthma drug was associated with a lessening of symptoms (11). Serial measurements of peak expiratory airflow have shown that airflow obstruction in asthmatic patients peaks between 3:00 and 4:00 AM (12). As described in the next section, this is part of the reason why asthmatic patients are more likely to present to the emergency department between midnight and 8:00 AM, 40% of their calls to physicians occur between 11:00 PM and 7:00 AM, and 42% to 53% of fatal asthma exacerbations occur at night (12-15).

TABLE 34-1 A CLINICAL CLASSIFICATION OF ILDS (PARTIAL LIST)

COLLAGEN VASCULAR DISEASES

Rheumatoid arthritis

Polymyositis

Scleroderma

Sjogren’s syndrome

Systemic lupus erythematosus

Ankylosing spondylitis

DRUG-INDUCED

Nitrofurantoin

Sulfasalazine

Amiodarone

Gold

Cyclophosphamide (Cytoxan)

Carmustine (BCNU)

Methotrexate

Bleomycin

Busulfan

Procarbazine

Radiation

PRIMARY DISEASES

Sarcoidosis

Eosinophilic granuloma

Vasculitis

Eosinophilic pneumonias

Diffuse alveolar hemorrhage

Lymphangioleiomyomatosis

Alveolar proteinosis

Acute respiratory distress syndrome

OCCUPATIONAL-ENVIRONMENTAL

Inorganic

Silicosis

Asbestosis

Berylliosis

Coal worker’s pneumoconiosis

Organic

Hypersensitivity pneumonitis (farmer’s lung, bird breeder’s lung, etc.)

Unknown etiology

IPF

Bronchiolitis-associated ILD

Acute interstitial pneumonia (Hamman-Rich syndrome)

Bronchiolitis obliterans organizing pneumonia

Lymphocytic interstitial pneumonitis

A more extreme but much less common form of airway disease is CF. It is the most common life-shortening autosomal recessive disorder in the white population (16). It affects approximately 30,000 persons in the United States. The resulting bronchiectasis, COPD, and chronic recurrent infection account for >90% of fatalities.

Etiology

Chronic Obstructive Pulmonary Disease

The effects of hyperinflation on patients with COPD have been well documented. Factors such as mechanical disadvantage from shortened diaphragmatic muscle fiber length, with resultant decrease in force of contraction, lead to an increased work of breathing. The excessive workload leads to respiratory muscle fatigue and potential failure during periods of increased demands on the system. Patients with COPD have lower than normal PaO₂ and will have a greater fluctuation in SaO₂. These patients have been shown to be hypoxemic during sleep through the mechanisms of hypoventilation, OSA, reduction in FRC, and ventilation and perfusion (V/Q) mismatch (17). An increased incidence of SDB in these patients may further lead to significant nocturnal hypoxemia.

Insomnia is a commonly encountered problem in patients with COPD occurring in as much as 53% of patients with chronic bronchitis. Important factors may be excessive mucous production, cough, increased work of breathing while supine, nocturnal dyspnea, and desaturation-related arousals from sleep (18,19).

Asthma and Other Obstructive Lung Diseases

Whether nocturnal asthma is a distinct entity or just a manifestation of poorly controlled asthma, in general, is not clear. However, there are several mechanisms that have a role in worsening symptoms at night:

Airway inflammation
Circadian rhythm changes in parasympathetic tone and hormone levels
Allergens and airway cooling
Gastroesophageal reflux
Mucociliary clearance
Bronchial hyperactivity

Studies have shown that airway inflammation occurs specifically at night in asthmatic patients with increased airway eosinophils and neutrophils, superoxide, and cytokine concentrations, as well as activated lymphocyte and macrophage levels from bronchoalveolar lavage (BAL) samples and biopsies (20). Genetic factors may also play a role, such as in downregulation of the beta-2 adrenoreceptor number in the Glyl6 polymorphism of this receptor (21).

Normal people have circadian changes in airway caliber, with mild nocturnal bronchoconstriction. This appears to be exaggerated in asthmatic patients. When subjects are given cholinergic blocking agents, such as inhaled ipratropium, tiotropium, or intravenous atropine, this effect can be diminished (22,23). Sympathetic tone does not appear to be involved in these circadian changes (24). Control of allergen exposure has been a mainstay of asthma management. Controlling these factors at night, such as cleaning the bedroom or bedding, may enhance overall improvement in asthma and not just nocturnal symptoms. Avoiding these agents at night does not abolish the circadian changes in bronchoconstriction (24). Cold, dry air may further add nocturnal bronchoconstriction. When a small group of asthmatic patients breathed warm, humidified air (36°C-37°C, 100% saturation) compared to room air, bronchoconstriction could be abolished (25). Gastroesophageal reflux is highly prevalent in patients with asthma, but its role in producing daytime symptoms is not clear. Although gastric acid can enhance previously mentioned bronchoconstriction, acid suppression with histamine type-2 blockers (26) or proton pump inhibitors (27) improves nighttime symptoms with minimal daytime effects (24). In addition, drugs such as theophylline, which are known to decrease distal esophageal sphincter tone (promoting reflux), actually improve nocturnal symptoms (28). Increased mucociliary clearance and bronchial hyperresponsiveness occur during sleep but are not likely to be a major factor in producing symptoms over and above the previously described mechanisms.

CF results in mutations in the CF transmembrane conductance regulator (CFTR) gene, which results in defective chloride transport in the epithelial cells of the respiratory, hepatobiliary, gastrointestinal, and reproductive tracts, as well as the pancreas. This results in the clinical manifestations discussed below (16).

Diagnostic Evaluation

Ancillary Tests

Chest radiography is often of critical importance in the evaluation of pulmonary diseases. A stepwise approach to interpretation of a chest radiograph is often helpful. The large airways such as the glottis, trachea, and mainstem bronchi can be assessed for any suggestion of extrinsic compression by parenchymal or extraparenchymal masses or vascular structures. The remainder of the bronchial tree is evaluated for any evidence of intraluminal obstruction secondary to tumors or tenacious secretions, and diffuse enlargement and dilation as can be seen in bronchiectasis. The skeletal structures are reviewed for rib fractures and spinal scoliosis and kyphosis, both of which may alter the mechanics of breathing. The cardiovascular structures, including the heart, pulmonary vasculature, and aorta, should be examined for signs of enlargement, engorgement, and dilation, respectively. Diaphragmatic contour and relative height may suggest the presence or lack of significant hyperinflation and paralysis. An increase in the retrosternal airspace (>3 cm) on the lateral view may also suggest hyperinflation. Pulmonary hypertension and right ventricular hypertrophy are indicated by prominent hilar vascular shadows and encroachment of the heart shadow on the retrosternal space as the right ventricle anteriorly enlarges (29). The pulmonary parenchyma may demonstrate lucencies representative of emphysematous changes or, indeed, deficient blood supply. Diffuse interstitial infiltrates in a peripheral distribution may point to interstitial lung disease (ILD). Asthma exacerbations with mucus plugging can lead to nodular densities and atelectasis, as well as the hyperinflation described above. The chest radiograph in CF reveals patchy infiltrates, bronchiectasis, and cystic areas resembling small abscesses.

The electrocardiogram (ECG) may supplement the clinician’s suspicion for significant COPD, asthma, or other obstructive lung diseases. As many as 75% of patients with COPD have ECG abnormalities. Hyperinflation of the lungs leads to depression of the diaphragm and, therefore, clockwise rotation of the heart along its longitudinal axis (30). The end result would be deep S waves in the right precordial leads with poor R wave progression and QS waves in a pattern similar to that seen in anterior myocardial infarction. P pulmonale, defined as a P wave ≥2.5 mm in the inferior leads resulting from right atrial enlargement due to hypertrophy or dilation, is often found in COPD. In the presence of cor pulmonale and right ventricular hypertrophy, the ECG may reflect right atrial enlargement with right axis deviation and a prominent R wave in lead V₁

As previously described, PFTs are an essential part of the diagnosis of COPD as well as other obstructive lung diseases. They have a well-founded utility in establishing the severity of the impairment from disease and monitoring the response to treatment. The FEV₁ is the gold standard for the diagnosis of airflow obstruction (29). These patients also typically display a reduction in the FEV₁/FVC ratio, and an increase in TLC and RV, which suggests hyperinflation and air trapping, respectively (Fig. 34-1). ABGs reveal the expected hypoxemia.

Hypercapnia develops as the disease progresses and is more common when the FEV₁ falls below 1 L. Laboratory evaluation is important in the evaluation of patients with obstructive lung diseases and includes not only the ABGs, as noted above, but also the carbon monoxide (CO) as measured by co-oximetry. An evaluation of CO may be helpful in assessing further abuse of tobacco products and promote effective counseling. Erythrocytosis is more commonly seen when the arterial PaO₂ falls to <55 mm Hg. A normal hemoglobin in the face of severe obstruction on PFTs should prompt the physician to search for a source of blood loss. The appropriate patient with family history, early age onset, or the nonsmoker will need an alpha-1 antitrypsin level. If abnormal, then more specific genetic testing may be indicated. For patients with early obstructive airway disease but not limited to an early age at onset, sweat chloride analysis for CF may be warranted.
More specific genetic testing for mutations of the CFTR gene can be obtained if results and clinical information are equivocal (16).

Nighttime Evaluation

The American Sleep Disorders Association recognizes four different levels of recording devices for the assessment of sleep apnea (31). Level I includes standard polysomnography (PSG) performed in the laboratory. Level II studies are performed outside of the laboratory and are capable of recording complete PSG without a technologist present (unattended). Level III devices can record respiratory effort, airflow, oxygen saturation, and pulse or ECG. These devices have been the subject of recent guidelines for use as a diagnostic in patients with a high pretest probability of OSA (32). Level IV devices record only one or two variables such as pulse, oximetry, and ECG.

PSG is the recommended test for the diagnosis of OSA and various disorders of sleep, such as periodic limb movements of sleep (PLMS), upper airway resistance syndrome (UARS), REM behavior disorder, somnambulism, and parasomnias (31).

Interestingly, the sensitivity of PSG in the diagnosis of OSA has not been clearly shown, as PSG is usually assumed to be the gold standard of comparison. However, there can be substantial night-to-night variability in the apnea—hypopnea index (AHI) on PSG studies. It is also well recognized that a negative overnight PSG does not completely exclude the diagnosis of OSA. This is particularly the case for patients who have risk factors for SDB on clinical grounds. Meyer et al. (33) studied 11 such patients who met diagnostic criteria for OSA during a second study, with an increase in the AHI from 3.1 ± 1.0 at baseline to 19.8 ± 4.7 events per hour (mean ± SEM, p < 0.01) on the repeat study. Studies have shown that 11.5% of patients with an AHI of <5 on study night 1 have an increase in the AHI to >5 on study night 2 (34).

Levels II and III devices refer to portable devices that are useful in situations where symptomatic patients who suggest OSA require prompt evaluation and treatment when PSG is not available. These devices have not been adequately studied in patients with COPD and are not yet recommended for routine clinical use in these patients (31,32).

Oximetry, the mainstay of level IV devices, measures the oxygen saturation of hemoglobin by distinguishing deoxyhemoglobin from oxyhemoglobin based on the differential absorption of light (35). The normal value for oxyhemoglobin saturation during sleep in healthy subjects is approximately 96.5% (±1.5%). There are no decreases reported in oxyhemoglobin saturation for different ethnic populations or sex, but it will decrease with altitude.

Historically, pulse oximetry was the key means to identify patients with Pickwickian syndrome with prolonged oxygen desaturation (Figs. 34-2 and 34-3) or severe sleep apnea syndrome with alveolar hypoventilation. The latter may be associated with a saw-toothed pattern (Fig. 34-4) (36). Over the last decade, multiple studies have evaluated the utility of pulse oximetry in screening for SDB. The term SDB encompasses OSA, central sleep apnea (CSA) syndrome, UARS, and sleep hypoventilation syndrome. Sensitivities ranging from 31% to 98% and specificities ranging from 41% to 100% have been reported. Williams et al. (37) used home oximetry and laboratory PSG to evaluate 40 patients suspected of having OSA. Sensitivity of oximetry was only 56%, whereas specificity was 100%. Epstein and Dorlac (38), retrospectively, reviewed overnight sleep studies on 100 consecutive patients. They used either a “deep” pattern of a ≥4% decrease in SaO₂ to <90% or a “fluctuating pattern” without a definite criteria for change or nadir in the SaO₂ to identify events. Oximetry was considered abnormal when there were 10 or more oximetry events per hour. The fluctuating pattern had a greater sensitivity, whereas the deep pattern had a greater specificity in identifying OSA. If patients with only mild symptoms were considered, the fluctuating pattern was still not as sensitive as PSG.

Overnight oximetry can be a useful screening test for SDB when there is a low clinical suspicion of sleep apnea. In a patient able to sleep efficiently for a sufficient time compared with what he or she would consider a routine night of sleep, a normal or negative oximetry can help exclude sleep apnea. Similarly, when there is a high degree of clinical suspicion for OSA and oximetry results are normal, further testing is required. The key characteristics of an abnormal overnight oximetry that suggest SDB are oscillatory variation in saturation and oscillatory desaturation (Fig. 34-4). Epstein and Dorlac (38) suggest that this abnormality can be defined as a >4% change in oxyhemoglobin saturation to ≤90% and a pattern of repetitive short-duration fluctuations in saturation and desaturations. Other investigators (39) have found that resaturations of ≥3% SpO₂ within 10 seconds at the end of a respiratory event were a better detector of respiratory disturbance events (RDEs). Low baseline saturations that may slowly increase or decrease about baseline values without oscillations may be more indicative of gas-exchange abnormalities, such as those encountered in advanced COPD (Figs. 34-2 and 34-3). The desaturations, secondary to COPD, tend to last much longer and have a much lesser degree of slope in the waveform (36).

The accuracy of oximetry can be affected by factors such as vasoconstriction, hypotension, skin pigmentation, nail abnormalities, patient movement, and probe positioning. Significantly, different saturation data are obtained at various acquisition options. Davila et al. (40) studied 75 patients with suspected OSA with simultaneous pulse oxyhemoglobin saturation traces at three recording settings, or averaging times. These investigators found that faster recording settings (shorter averaging times or response times) resulted in lower levels of oxyhemoglobin saturation than did slower settings (longer averaging times or response times) (Figs. 34-4 and 34-5). Several authors
have also found significant differences in the saturation data obtained online real time (high sampling rates) and those values obtained from memory (low sampling rates) in unattended studies. Desaturation indexes obtained from memory have been found to be significantly lower than those obtained online (40). Studies have confirmed lower sensitivities and higher specificities using longer averaging times settings (e.g., 12 seconds) and memory display mode (40). Faster recording settings (shorter averaging times) and online display mode give rise to higher sensitivity and lower specificity. Therefore, the default settings must be known to the user and the interpreter. Faster recording settings and online display are the methodology of choice.

FIGURE 34-2 Periods of alveolar hypoventilation (gradual decline In Sao₂ channel 10) during sleep (30-second epoch) may go undetected as the nasal/oral flow (thermistor) and chest and abdominal effort (piezoelectric belts) signals remain essentially unchanged. Channels 1 and 2, left and right anterior tibialis electromyogram; channels 3 and 4, left and right central electroencephalogram; channels 5 and 6, left and right occipital electroencephalogram; channel 7, chin electromyogram; channels 8 and 9, not used; channel 10, oxygen saturation; channel 11, nasal and oral flow sensors; channels 12 and 13, thoracic and abdominal effort sensors; channel 14, numeric oxygen saturation.

Indeed, oximetry alone is not recommended for the evaluation of sleep apnea in place of PSG, given the lack of standardization in performance and interpretation (32). Oximetry is, however, comparatively inexpensive, less disruptive to sleep, and more readily available than PSG. It has proved to be very useful in following the response to therapies for OSA, such as oral appliances.

Overnight pulse oximetry is also useful for determining the degree to which patients with COPD desaturate during sleep. Patients with COPD and daytime hypoxemia (PO₂ <55 mm Hg) or PO₂ 55 to 59 mm Hg with signs of end-organ damage have improved survival with continuous diurnal and nocturnal oxygen therapy. In the Nocturnal Oxygen Therapy Trial (41), 203 patients with hypoxemic COPD were randomized to continuous or 12-hour nocturnal supplemental oxygen. One-year follow-up demonstrated improved survival and decreased morbidity with continuous supplemental oxygen. However, there was no improvement in morbidity or mortality with nocturnal oxygen therapy alone. COPD patients with FEV₁ of <1 L may spend >90% of the night with saturations well below 90% when breathing room air (3,4). Risk factors for nocturnal desaturation are PaCO₂ ≥45 mm Hg and PaO₂ <65 mm Hg on oxygen (3,4). However, the
severity of nocturnal desaturation in patients with COPD cannot be predicted with certainty. Hence, we can see the potential utility of overnight oximetry. The causes of the nocturnal desaturation include hypoventilation, coexisting sleep apnea, reduction in FRC, and altered ventilation-perfusion matching, as previously mentioned. The benefit of nocturnal oxygen therapy in patients with a daytime PO₂ of >60 mm Hg, but nocturnal desaturation, is unproved. However, many clinicians will prescribe oxygen treatment if there is prolonged desaturation or evidence of end-organ dysfunction (cor pulmonale).

FIGURE 34-3 Periods of alveolar hypoventilation during sleep undetected by flow (thermistor) but demonstrated on the chest and abdominal effort channels (piezoelectric belts). Channels 1 and 2, left and right anterior tibialis electromyogram; channels 3 and 4, left and right central electroencephalogram; channels 5 and 6, left and right occipital electroencephalogram; channel 7, chin electromyogram; channels 8 and 9, not used; channel 10, oxygen saturation; channel 11, nasal and oral flow sensors; channels 12 and 13, thoracic and abdominal effort sensors; channel 14, numeric oxygen saturation.

Approximately 10% to 15% of patients with COPD have sleep apnea (1). In COPD patients with signs and symptoms of OSA and/or polycythemia or pulmonary hypertension with right heart dysfunction or right heart failure, a sleep study is indicated. The American Academy of Sleep Medicine outlines other associated features and laboratory findings in many patients with SDB (42).

Asthma and Other Obstructive Lung Diseases

PSG has demonstrated decreased sleep efficiency, increased arousals and awakenings, decreased sleep time with associated daytime sleepiness, and impaired cognition in patients with asthma (24,28). The resulting sleep deprivation may be associated with impaired ventilatory drive and contribute, along with other factors, to the worsening of hypoxemia and hypercapnia in severe acute attacks. In spite of these findings and in the absence of other symptoms of a specific sleep disorder, PSG is not indicated. Overnight oximetry may be used as described for COPD, to evaluate for nocturnal hypoxemia and plan home oxygen therapy.

Patients with CF may have more severe sleep disruption than in other obstructive lung diseases, with impaired daytime functioning. Dancey et al. (43) reported that 19
patients with severe CF had reduced sleep efficiency (71%) and frequent awakenings, as well as lower mean SaO₂, when compared with 10 healthy controls. Furthermore, patients with CF were sleepier, with reduced sleep latency on multiple sleep latency test (MSLT) (6.7 minutes). These findings correlated with more reported fatigue and lower levels of happiness and activation as well as impaired cognitive function. Sleep-related complaints such as sleep-onset difficulties, sleep-maintenance trouble, and snoring are commonly seen in adults and children with CF. Children and adolescents with CF have decreased sleep efficiency, prolonged REM latency, and decreased REM sleep percentage when compared with controls (44). Children with stable CF have more frequent nocturnal cough than do children without CF. The cough is more severe with advancing disease and occurs most frequently in the first hour of sleep (45).

FIGURE 34-4 Cyclic desaturations in a patient with OSA depicted by oximetry. The oximetry data were recorded either online (OL) or from the oximeter’s memory function (MEM) at three different recording settings (3,6, and 12 seconds), which equate to averaging time and response time. (From Davila DG, Richards KC, Marshal BL, et al. Oximeter performance: the influence of acquisition parameters. Chest 2002;122:1654, with permission.)

Predicting nocturnal desaturation appears problematic, as spirometric parameters, awake SpO₂, were not found to be predictive in 70 patients with CF studied by Frangolias and Wilcox (46). However, Milross et al. (47) reported that evening arterial PaO₂ did contribute to the ability to predict both sleep-related desaturation and elevated transcutaneous CO₂ in 32 stable patients with CF in transitioning from nonrapid eye movement (NREM) sleep to REM sleep. Further studies are needed to determine whether evening ABGs and/or nocturnal oximetry may be useful in the management of these patients with moderate to severe disease (16,39,40 and 41).

Clinical Management

Chronic Obstructive Pulmonary Disease

As previously mentioned, long-term oxygen is the only therapy proved to extend survival in patients with COPD (41). Many practitioners will prescribe nocturnal oxygen to patients with COPD who demonstrate sustained oxygen desaturation of ≤88% at night (48). However, documentation of hypoxemia during sleep is not a proved justification for continuous oxygen therapy if hypoxemia is not present when the patient is awake and at rest. Nocturnal oxygen is indicated in those patients who qualify for long-term oxygen therapy while awake and in those patients with COPD who develop erythrocytosis, cor pulmonale, and right heart failure without awake hypoxemia (48,49). Treatment has been shown to reduce pulmonary artery pressure and reduce mortality.

Anticholinergics (ipratropium bromide and tiotropium), short- and long-acting beta agonists, and steroids are the foundation of therapy for COPD (7,20,22,24). They are helpful in improving nocturnal gas exchange given their impact on obstruction and gas trapping. In a randomized, double-blind, placebo-controlled crossover study of patients with moderate and severe COPD, Ryan et al. (50) showed that salmeterol use twice daily led to improvements in sleeping mean SaO₂ and the percentage of time spent below an SaO₂ of 90%. The long-acting anticholinergic tiotropium may also be a useful nocturnal bronchodilator. Sposato and Franco showed that nightly administration of this drug to patients with stable moderate/severe COPD decreased the severity of nocturnal desaturations. These patients had no to mild OSA with an AHI of ≤ 10 (51).

Theophylline has been of interest due to its known beneficial effects on diaphragmatic contraction, central respiratory stimulation, and reduction of airway obstruction and gas trapping (52). Berry et al. (52) compared the effect of a shorter-acting beta agonist with that of the combination of the beta agonist and sustained-action theophylline on sleep and breathing in patients with COPD. The addition of theophylline was found to improve morning FEV₁, NREM SpO₂, and transcutaneous PCO₂, without impairing sleep quality. Unfortunately, sleep quality can be diminished in individual patients sensitive to the central stimulatory effects of theophylline (52).

Respiratory stimulants have been shown to have a short-lived impact on SDB. Progesterone has been studied in patients with COPD and hypoxemia and shown to be beneficial in reducing awake PaCO₂ after 4 weeks of therapy.
Dolly and Block (53) performed a randomized, placebocontrolled study on the effects of medroxyprogesterone acetate at 20 mg t.i.d. in patients with COPD and SDB. Four weeks of therapy was associated with increased awake mean PaO₂ and reduced PaCO₂ but no change in number of apneas, hypopneas, desaturation events, or lowest SpO₂.

FIGURE 34-5 A RDE with pulse oximetry monitoring at three different recording settings (RS) at averaging times of 3, 6, and 12 seconds. Channels: EMG, electromyogram; EEG, electroencephalogram; EOG, electrooculogram; ECG, electrocardiogram; ABD (abdominal) and chest effort (piezoelectric belts); SONA, sonogram; flow, thermistor sensors; SpO₂, oxygen saturation.

As stated earlier, patients with COPD typically experience worsening hypoventilation and more profound desaturation in REM sleep. Protriptyline is a tricyclic antidepressant that reduces total REM sleep and may be helpful in patients with significant REM-associated SDB (17). However, the medication causes severe urinary hesitancy or retention in many older male patients.

NIV is often prescribed for patients with advanced COPD and respiratory failure, those with concomitant hypoventilation, and in patients with OSA intolerant of continuous positive airway pressure (CPAP) (54,55). Stable patients with concurrent OSA and COPD have been shown to benefit from NIV, as determined by de Miguel et al. (56). These investigators studied patients with eucapnic and hypercapnic concurrent OSA and COPD during 6 months of CPAP treatment. Significant increases in PaO₂, FEV₁, and FVC, with decreases in PaCO₂, resulted. Nasal bilevel positive airway pressure has been shown to decrease daytime PaCO₂ in patients with COPD (57). Other investigators have found no benefits from extended bilevel therapy in hypercapnic COPD patients. Automatic self-titrating CPAP (auto-CPAP) systems are designed to respond to changes in upper airway resistance by variably increasing or decreasing the positive pressure in response to changes in pressure, flow, and/or the presence of snoring sounds (58). Therefore, the effectiveness would theoretically be equivalent to that of constant pressure treatment with a lower mean treatment pressure. Ficker et al. (59) evaluated auto-CPAP in comparison with conventional CPAP in the treatment of symptomatic OSA. There was no significant difference in mean AHI, sleep architecture, arousal index, Epworth Sleepiness Scale (ESS)score, or vigilance tests. The mean pressure during auto-CPAP treatment was significantly higher than that used in conventional therapy. This mode of NIV has not been adequately tested in patients with COPD. Inappropriate ventilation could occur in these patients with a decreased capacity to trigger the device.

Acute exacerbations of COPD are often managed with NIV with a significant reduction in the requirements for intubation and mechanical ventilation. Brochard et al. (60) undertook a prospective, randomized study of 85
patients with COPD admitted to the intensive care unit for acute exacerbation. The standard treatment group was treated with up to 5 L/min supplemental oxygen to maintain the SpO₂ at >90%, in addition to bronchodilators, antibiotics, and steroids. The group treated with NIV received similar therapy. The individuals treated with NIV experienced reduced morbidity, mortality, and length of hospital stay.

Asthma and Other Obstructive Pulmonary Diseases

The NIH defined asthma as a chronic inflammatory disease in the 1997 guidelines for diagnosis and therapy (10). With this designation, it has laid out a stepped approach to management centered around the decrease in acute and chronic airway inflammation and control of symptoms. In 2007, the Expert Panel Report (61) asthma treatment guidelines stressed how treatment strategies impact patient outcomes. Improvement in nocturnal asthma may be a marker of control in these two areas. The long-term goal of therapy is to prevent airway remodeling and fixed airway obstruction as measured by PFTs (20,61).

Only gold members can continue reading. Log In or Register to continue