Obstructive ventilatory dysfunction (OVD) describes a pattern of pulmonary dysfunction characterized by a reduced ratio of the forced expiratory volume in 1 second (FEV₁) to the forced vital capacity (FVC) on spirometry. In addition to the reduced FEV₁/FVC ratio there is a reduction in flow manifested by a reduction in the FEV₁ (Fig. 22–1).¹–⁴ Patients with very mild OVD may have a normal FEV₁ but a reduced FEV₁/FVC ratio. For convenience, the lower limit of normal for the FEV₁ is assumed to be 80% of predicted and the lower limit of normal for the FEV₁/FVC ratio is assumed to be 0.70. However, using a slightly higher value of the FEV₁/FVC ratio for the lower limit of normal is appropriate in younger patients. For example, using 90% of predicted as the lower limit of normal for the FEV₁/FVC ratio may be more sensitive for identification of OVD. Patients with OVD have normal or increased lung volumes (residual volume [RV], function residual capacity [FRC], and total lung capacity [TLC]) (Fig. 22–2). The first lung volume to be affected is the RV. Most patients with OVD have an elevated RV due to airtrapping from small airway narrowing and closure. With more severe disease, the FRC and then the TLC may also be increased (hyperinflation). However, the relative increase in the RV is greater than any increase in the FRC or TLC. In moderate to severe OVD, the increase in RV is large enough (compared with any increase in the TLC) to cause a reduction in the FVC (see Fig. 22–2). The increase in the FRC and TLC is due to loss of lung elastic recoil (emphysema) or airway disease (asthma).

A significant acute response to inhaled bronchodilator is defined by a 12% or greater increase in the FEV₁ or FVC with a minimum of 200 mL absolute increase. Lack of an acute bronchodilator response does not rule out a benefit from chronic bronchodilator therapy.

The common OVD disorders are listed in Table 22–1. These include asthma (bronchospasm = reversible airway obstruction), chronic bronchitis (productive cough/sputum production usually with OVD), and emphysema (destruction of alveoli and increased size of terminal airspaces). Patients often have a mixture of these manifestations and are said to have chronic obstructive pulmonary disease (COPD). Patients with asthma may have normal pulmonary function between exacerbations that can occur after upper respiratory tract infections, exposure to allergens, or exercise. They typically demonstrate a large improvement in the FEV₁ or FVC after inhaled bronchodilator. Asthmatics have bronchial hyperresponsiveness to methacholine challenge (>20% fall in FEV₁ after inhaling low concentrations), allergen, or exercise challenge (>20% fall in FEV₁ after exercise). A diagnosis of asthma can be made in a patient with normal pulmonary function by demonstration of bronchial hyperresponsiveness. Some asthma patients have persistent airflow obstruction and require inhaled or even oral corticosteroids for improvement. Patients with severe asthma can have hyperinflation and even hypercapnic respiratory failure. Patients with COPD and predominantly chronic bronchitis typically have recurrent exacerbations and tend to have lower arterial partial pressure of oxygen (PaO₂) values earlier in the disease course. Some may develop hypercapnia and cor pulmonale (blue bloaters). Other OVD patients have predominantly emphysema with severe hyperinflation (due to loss of lung elastic recoil) and relative preservation of PaO₂ levels until late in the disease course. They present with dyspnea and, because of the relatively spared PaO₂, are called “pink puffers.”

TABLE 22–1

Spectrum of Obstructive Ventilatory Dysfunction

	DIAGNOSIS	DIFFUSING CAPACITY	BRONCHODILATOR RESPONSE	AIRTRAPPING AND HYPERINFLATION
Asthma	Reversible airflow obstruction (physiologic diagnosis)	Normal	Yes >12% often 25%	May be present
Chronic bronchitis	Productive sputum for ≥ 3 mo for 2 consecutive yr (clinical diagnosis)	Normal	Sometimes 10–15%	Air trapping (high RV)
Emphysema	Destruction of gas-exchanging units and enlargement of terminal airspaces (pathologic diagnosis)	Decreased	Rarely	Yes
Mixed chronic bronchitis and emphysema (COPD)	Combination	Decreased	None or small	Yes

In OVD the increase in RV > increase in FRC > increase in TLC.

Air trapping manifested by an increased RV and hyperinflation manifested by an increased FRC and TLC.

COPD = chronic obstructive pulmonary disease; FRC = functional residual capacity; OVD = obstructive ventilatory dysfunction; RV = residual volume; TLC = total lung capacity.

Chronic Obstructive Pulmonary Disease

COPD is the fourth leading cause of death in the United States.⁴ The major cause of COPD is cigarette smoking. Patients with COPD may die from respiratory failure or lung cancer. Although they typically present to physicians with complaints of bronchitis or dyspnea, they also frequently complain of poor-quality sleep. The Global Initiative for Chronic Obstructive Lung Disease (GOLD) has defined severity criteria (Table 22–2).⁴ In general, an FEV₁ less than 30% to 40% of predicted or 1.0 L for a normal-sized individual indicates very severe disease. Daytime hypercapnia is usually associated with an FEV₁ of 30% to 40% of predicted or less. Patients with the overlap syndrome (COPD + obstructive sleep apnea [OSA]) can develop hypercapnia with higher FEV₁ values.

TABLE 22–2

GOLD Criteria for Chronic Obstructive Pulmonary Disease Severity

STAGE	DESCRIPTION	FINDINGS (BASED ON POSTBRONCHODILATOR FEV₁)
0	At risk	Risk factors and chronic symptoms but normal spirometry
I	Mild	FEV₁/FVC ≤ 70% FEV₁ at least 80% of predicted value May have symptoms
II	Moderate	FEV₁/FVC ≤ 70% FEV₁ 50% to < 80% of predicted value May have chronic symptoms
III	Severe	FEV₁/FVC ≤ 70% FEV₁ 30% to < 50% of predicted value May have chronic symptoms
IV	Very severe	FEV₁/FVC ≤ 70% FEV₁ < 30% of predicted value or FEV₁ < 50% of predicted value + chronic respiratory failure Respiratory failure: PaO₂ < 60 mm Hg with or without PaCO₂ > 50 mm Hg while breathing room air at sea level

FEV₁ = postbronchodilator forced expiratory volume in 1 second; FVC = forced vital capacity; PaCO₂ = arterial partial pressure of carbon dioxide; PaO₂ = arterial partial pressure of oxygen.

From Rabe KF, Hurd S, Anzueto A, et al: Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2007;176:532–555.

Sleep in COPD

Sleep in patients with COPD is often impaired in both duration and quality.5–⁷ Many patients also have significant hypercapnia and hypoxemia at night. Ten percent to 15% of patients may also have concomitant OSA that worsens nocturnal gas exchange (overlap syndrome). The typical pattern of nocturnal oxygen desaturation (NOD) in a patient with COPD is shown in Figure 22–3. The baseline sleeping arterial oxygen saturation (SaO₂) falls 2% to 4% from the awake baseline with minor fluctuations until much larger drops are noted during rapid eye movement (REM; stage R) sleep. In contrast, a typical oximetry of a patient with the overlap syndrome is shown in Figure 22–4. There is a low baseline sleeping SaO₂ and a saw-tooth pattern consistent with repeated discrete events. Whereas central sleep apnea could cause a saw-tooth pattern, this is most likely to represent OSA.

FIGURE 22–3 Nocturnal oximetry in a patient with severe chronic obstructive pulmonary disease (COPD). The awake arterial oxygen saturation (SaO₂) is mildly reduced at 92% but falls to the low 80s with sleep. Further severe falls in the SaO₂ occur during rapid eye movement (REM) sleep (*dark bars*). Note that the worst period of desaturation was noted in the early morning hours. *From Berry RB: Sleep Medicine Pearls, 2nd ed. Philadelphia: Hanley & Belfus, 2003, p. 216.*

Etiology of Abnormal Nocturnal Gas Exchange

Patients with COPD often experience exaggerations of normal sleep-related changes in ventilation. The most severe desaturation occurs during REM sleep. The major mechanisms of nocturnal arterial oxygen desaturation are listed in Box 22–1. The relative importance of hypoventilation and ventilation-perfusion () mismatch is still debated.

Box 22–1 Major Mechanisms of Nocturnal Oxygen Desaturation in Chronic Obstructive Pulmonary Disease

• Hypoventilation—believed by most to be the most important factor

• Decreased chemosensitivity (REM < NREM < wake)

• Reliance on accessory respiratory muscles and loss of this assistance during REM atonia

• Increased upper airway resistance

• Lower alveolar ventilation per minute ventilation (high dead space–to–tidal volume ratio—due to high dead space and low tidal volume)

• Ventilation-perfusion mismatching (drop in PaO₂ exceeds the increase in PaCO₂)

• High closing volume and decreases in FRC—especially during REM-associated hypopnea

• Coexisting sleep apnea (12–15%)—especially blue bloaters

• Low awake SaO₂—starting position on the steep portion of the oxygen-hemoglobin saturation curve (larger fall in SaO₂ for a given fall in PaO₂)

FRC = functional residual capacity; NREM = non–rapid eye movement; PaCO₂ = arterial partial pressure of carbon dioxide; PaO₂ = arterial partial pressure of oxygen; REM = rapid eye movement; SaO₂ = arterial oxygen saturation.

Normal Individuals

During non–rapid eye movement (NREM) sleep, CO₂ production falls but alveolar ventilation falls proportionately more and arterial pressure of carbon dioxide (PaCO₂) increases slightly (Fig. 22–5). Recall that PaCO₂ = constant × (CO₂ production)/(alveolar ventilation); see Chapter 10. The fall in ventilation is due to a loss of the wakefulness stimulus to breathe, decreased ventilatory responses (chemosensitivity) to hypoxia and hypercapnia, and increased upper airway resistance.^5,7 The increase in PaCO₂ results in a mild decrease in the PaO₂. However, because the awake PaO₂ value is on the flat portion of the oxygen-hemoglobin saturation curve (Fig. 22–6), minimal drops in the SaO₂ are noted (e.g., from 97% to 95%). The FRC may decrease slightly from wake to NREM sleep (Fig. 22–7).^8,9

FIGURE 22–5 Normal changes in gas exchange during sleep. PaCO₂ = arterial partial pressure of carbon dioxide; PaO₂ = arterial partial pressure of oxygen; SaO₂ = arterial oxygen saturation. *Adapted from Mohsenin V: Sleep in chronic obstructive pulmonary disease. Semin Respir Crit Care Med 2005;26:109–115.*

During REM sleep in normal individuals, ventilation is irregular and periods of reduced tidal volume occur, often during bursts of REMs. Skeletal muscle hypotonia reduces the contribution from the accessory respiratory muscles and respiration depends on the diaphragm.8,9 Two studies did not find a lower FRC during REM compared with NREM sleep in normal individuals (see Fig. 22–7).^8,10 During REM sleep, there is typically a decrease in chest wall movement,⁹ likely due to chest wall hypotonia. These REM-associated physiologic changes result in a slight increase in PaCO₂ and decreases in PaO₂ during REM compared with NREM sleep. The slight fall in PaO₂ results in minimal changes in the SaO₂ due to the position of the oxygen-hemoglobin saturation curve (see Fig. 22–6).

Sleep-Related Changes in Respiration in COPD

Koo and colleagues ¹¹ studied 15 patients with severe COPD (FEV₁ of 0.96 L) with a mean daytime PaO₂ greater than 60 mm Hg and found a mean decrease in PaO₂ of 13.5 mm Hg and a mean increase in PaCO₂ of 8.3 mm Hg during REM sleep (Fig. 22–8). Although patients with daytime hypercapnia or a low awake PaO₂ or SaO₂ are more likely to have worse gas exchange during sleep, up to 25% of patients with COPD exhibit REM-related hypoventilation and NOD, despite having a daytime PaO₂ above 60 mm Hg.¹²

FIGURE 22–8 Changes in arterial partial pressure of carbon dioxide (PaCO₂) and arterial partial pressure of oxygen (PaO₂) during sleep in a group of patients with severe COPD. Here stages 2, 3, and REM represent stages N2, N3, and R. REM = rapid eye movement; SD = standard deviation. *From Koo KW, Sax DS, Snider GL: Arterial blood gases and pH during sleep in chronic obstructive pulmonary disease. Am J Med 1975;58:663–670.*

Non–Rapid Eye Movement

During supine wakefulness, patients with COPD often start with low-normal or slightly decreased SaO₂ values. Therefore, even the normal fall in PaO₂ with sleep will cause a greater decrease in the SaO₂. Koo and colleagues ¹¹ documented a fall in PaO₂ of approximately 8 mm Hg from wake to NREM sleep and an increase in PaCO₂ of about 5 mm Hg. In patients with COPD, the onset of NREM sleep is associated with mild falls in the SaO₂ (4–8%) and PaCO₂ increases. If obstructive apneas or hypopneas are present, the degree of desaturation is worsened. Due to hyperinflation, the diaphragm is at a mechanical disadvantage and there is more dependence on the accessory muscles of respiration. In a study by Becker and coworkers¹³ of a group of hypercapnic patients with COPD, the minute ventilation fell by 16% from wake to NREM.

Rapid Eye Movement

During REM sleep, there are more profound periods of arterial oxygen desaturation compared with NREM sleep in patients with COPD. These episodes of desaturation are characterized by long periods of irregular breathing and reduced tidal volume (Figs. 22–9 and 22–10). In contrast to obstructive hypopnea in OSA patients, the onset and termination of these nonapneic periods of REM desaturation are less well defined. As noted previously, it is believed that REM-associated hypotonia reduces the contribution for the intercostal muscles so that ventilation depends on the diaphragm. Hyperinflation in patients with COPD reduces the effectiveness of the diaphragm. During the periods of REM-associated diaphragmatic inhibition (often during bursts of eye movements),¹³–¹⁵ the tidal volume and minute ventilation often decrease dramatically (see Figs. 22–9 and 22–10). The drop in alveolar ventilation is even larger than the change in minute ventilation due to an increase in the dead space–to–tidal volume ratio (see Chapter 10). In one study of a group of hypercapnic patients with COPD, the minute ventilation fell by 32% from wake to REM sleep (compared with 16% from wake to NREM).¹³ Studies have documented increases in PaCO₂ during these episodes.^11,15,16 Figure 22–10 illustrates a nonapneic period of REM-associated fall in tidal volume.

Mismatch

Whereas hypoventilation definitely occurs during periods of nonapneic desaturation (also called hypopneic breathing), Catterall and associates 16,17 directly measured arterial blood gases and found the PaO₂ dropped more than the PaCO₂ increased (Fig. 22–11). Therefore, increased mismatch is believed to play a role in REM-associated desaturation (in addition to hypoventilation).^15,16 However, as Catterall and associates^16,17 and others have pointed out, the usual blood gas analysis of the relationship between PaO₂ and PaCO₂ depends on an assumption of steady state. This condition does not exist during the transient periods of hypopneic breathing. Because oxygen stores in the body are much smaller than carbon dioxide stores, a change in ventilation may affect PaO₂ more than PaCO₂.

FIGURE 22–11 An episode of rapid eye movement (REM)–associated nonapneic desaturation in a patient with COPD. The decrease in arterial partial pressure of oxygen (PaO₂) was greater than the increase in the arterial partial pressure of carbon dioxide (PaCO₂). Stage 3 and REM denote stages N3 and R. EEG = electroencephalogram; SaO₂ = arterial oxygen saturation. *From Catterall JR, Calverley PMA, MacNee W, et al: Mechanism of transient nocturnal hypoxemia in hypoxic chronic bronchitis and emphysema. J Appl Physiol 1985;59:1698–1703.*

Many patients with COPD have an increase in FRC in the upright position (hyperinflation). In the supine position, the FRC may be slightly lower during wakefulness. Ballard and coworkers ¹⁰ found similar FRC values during wake, NREM, and REM sleep in patients with emphysema. However, in asthmatics, another study found that the FRC does decrease from wake to NREM and from NREM to REM (see Fig. 22–7).¹⁰ COPD patients with a significant component of airways disease could experience a drop in FRC during REM sleep. In a given patient with COPD, the FRC may also be affected by the degree of obesity. Furthermore, assigning a single FRC value to stage R sleep is problematic because breathing during REM sleep is very inhomogeneous. Hudgel and coworkers¹⁴ found no decrease in FRC during transition from stage N2 to stage R sleep in a group of five patients with COPD. However, during episodes of hypopneic breathing in REM sleep, there was a fall in FRC. The same study found that those COPD patients who desaturated during REM sleep had longer periods of hypopneic breathing (Fig. 22–12). Thus, it seems likely that the FRC during hypopneic breathing in COPD is lower than during NREM sleep. Whether the absolute value is below normal may vary between patients.

FIGURE 22–12 Hypopnea and hyperpnea in REM sleep in a COPD patient. Raw intercostal and diaphragmatic electromyogram (EMG), esophageal pressure, and arterial oxygen saturation (SaO₂) are shown. This is the end of an episode of REM hypopneic breathing. Note the increase in end-expiratory volume (*) at the end of the episode. The *solid bar* shows that the end-expiratory volume after the hypopnea ends is higher than during the hypopneic breathing. Also note the decreased intercostal EMG and greater reduction in chest wall movement. *From Hudgel DW, Martin RJ, Capehart M, et al: Contribution of hypoventilation to sleep oxygen desaturation in chronic obstructive pulmonary disease. J Appl Physiol 1983;55:669–677.*

A fall in the FRC during hypopneic breathing in COPD patients may have greater consequences than in normal individuals. In COPD, the closing volume (CV), the volume at which small airway closure is significant, is closer to the FRC than in normal individuals. During the hypopneic breathing of REM sleep, the FRC falls below the CV (Fig. 22–13). This means that during tidal breathing, more alveoli are either not ventilated or underventilated. This results in worsening of mismatch (already abnormal awake) and worsening hypoxemia. Hudgel and coworkers¹⁴ concluded that hypoventilation, and to a lesser extent mismatch, causes REM-associated desaturation. Fletcher and associates¹⁵ found an increase in venous admixture during REM hypopnea episodes and concluded that mismatch was a significant cause of hypoxemia. The relative roles of hypoventilation and mismatch in inducing nocturnal desaturation during hypopneic breathing during REM sleep are still debated. The analysis is complicated by the non–steady-state condition of transient hypopneic breathing.

FIGURE 22–13 In patients with chronic obstructive pulmonary disease (COPD), the closing volume (CV) is nearer the functional residual capacity (FRC) than in normal individuals. The fall in FRC associated with hypopneic breathing during rapid eye movement (REM) sleep results in tidal breathing below the CV. This worsens ventilation-perfusion mismatch. Note that, in this figure, the FRC for COPD during hypopneic breathing in REM sleep is shown as approximately equal to that of a normal individual during REM sleep. The actual FRC in COPD patients may depend on other factors such as obesity and vary between individuals.

Time of Night and Circadian Variation in Lung Function

REM episodes in the early morning have greatest REM density and the greatest variation in ventilation even in normal individuals.¹⁹ These REM periods also are typically longer. In the early morning hours, there is greater lower airway resistance due to circadian changes in bronchomotor tone that are exaggerated in many patients with COPD.^5,6 Ballard and colleagues¹⁸ found an increase in upper airway resistance but not lower airway resistance during the night in a group of patients with emphysema. Conversely, lower airway resistance does increase in asthmatics²⁰ and likely in some COPD patients with a significant component of airway disease. These factors help explain why the most severe and longest REM-associated desaturation typically occurs in the early morning hours.

COPD Types and Respiration during Sleep

Several studies have tried to find factors predicting more severe desaturation during sleep in COPD. As might be expected, a lower awake PaO₂ and higher PaCO₂ predict more dramatic changes in gas exchange during sleep. One study compared blue bloaters and pink puffers and found the former were more likely to desaturate during sleep (Table 22–3).²¹ Analysis of the Sleep Heart Health data suggested that the odds of having desaturation more than 5% of total sleep time was minimally increased until the FEV₁/FVC ratio was less than 60% (Table 22–4).²² A plot of awake SaO₂ versus minimum sleeping value is shown in Figure 22–14. Patients with lower awake SaO₂ or PaO₂ values tend to have lower values at night.²³ However, there is considerable variability and oximetry or polysomnography (PSG) is needed to reliably exclude NOD.

TABLE 22–3

Arterial Oxygen Saturation during Wake and Sleep—Variation with Chronic Obstructive Pulmonary Disease Type

	PINK PUFFER (%)	BLUE BLOATER (%)
SaO₂ awake and sitting	93	88.4
SaO₂ awake and supine	92.3	80.5
SaO₂ sleep	92.3	77.3
Maximal fall in SaO₂	4.7	29.5
Episodes of arterial oxygen desaturation	3.5	22.7
SaO₂ < 80% (min)	0.08	120.0