The effect of sleep in patients with neuromuscular and chest wall disorders

INTRODUCTION — Physiologic alterations during normal sleep are clinically insignificant in most adults, but can induce hypoventilation and hypoxemia in patients with neuromuscular and chest wall disorders, who characteristically have challenged ventilation when awake due to one or more of the following: perturbed chest wall mechanics, respiratory muscle weakness, chemoreceptor sensitivity abnormality, and blunted central neural output ("drive to breathe"). In this topic review, the following issues are discussed:

●Normal sleep-related physiologic alterations that affect ventilation, with special attention given to their impact on patients with neuromuscular or chest wall disorders. (See 'Ventilation' below.)

●Sleep-related alterations in blood gases, with special attention on the differences between healthy adults and patients with neuromuscular or chest wall disorders. (See 'Blood gases' below.)

●The major points of this topic review. (See 'Summary and recommendations' below.)

Evaluation and treatment of sleep-disordered breathing in patients with neuromuscular or chest wall disorders are reviewed separately. (See "Evaluation of sleep-disordered breathing in adult patients with neuromuscular and chest wall disorders" and "Noninvasive ventilation in adults with chronic respiratory failure from neuromuscular and chest wall diseases: Patient selection and alternative modes of ventilatory support".)

VENTILATION — Ventilation can be impacted during sleep through a variety of normal sleep-related alterations:

●Decreased central neural output (ie, "drive to breathe")

●A shift to greater ventilatory burden on the diaphragm during rapid eye movement (REM) sleep, due to reduced contributions from the intercostal and accessory muscles

●Increased upper airway resistance

●Decreased responsiveness to hypoxemia and hypercapnia (ie, chemosensitivity)

The consequences of these alterations on healthy adults and patients with neuromuscular or chest wall disorders is discussed in this section.

Central neural output — Central neural output, or drive, decreases during sleep. This is clinically inconsequential in healthy adults, but can be detrimental in patients with neuromuscular or chest wall disorders because they rely on an increased central neural output to compensate for weak ventilatory muscles, maintain their increased work of breathing, or both [1,2]. Thus, these patients may fail to maintain a normal level of ventilation when central neural output decreases during sleep [3].

The increased work of breathing may be a consequence of abnormally low lung and chest wall compliance (figure 1) [4-6]. The reduced lung compliance may be due to microatelectasis and focal pulmonary parenchymal fibrosis, while the reduced chest wall compliance may be due to increased stiffness of the rib cage. Ankylosis of rib cage joints, fibrosis or spasticity of rib cage muscles, or loss of parasternal intercostal activity may all contribute to stiffening of the rib cage in these patients [5,6].

Diaphragmatic burden — Inspiratory intercostal and accessory respiratory muscle activity increases during non-rapid eye movement (NREM) sleep compared with wakefulness, but the activity of the diaphragm remains unchanged [7,8]. In contrast, inspiratory intercostal and accessory muscle activity decreases during REM sleep, while diaphragmatic activity increases [7-10]. This shifts the burden of maintaining ventilation to the diaphragm during REM sleep.

The increased diaphragmatic burden does not have significant clinical implications in healthy adults. However, patients with neuromuscular or chest wall disorders may have a weak or mechanically disadvantaged diaphragm that is incapable of overcoming this load. This increases the risk of hypoventilation during REM sleep in these patients. These findings emphasize that hypoventilation is typically manifested first in stage REM sleep. Recording of sleep thus can facilitate early identification of hypoventilation.

Upper airway resistance — Upper airway resistance may increase during normal NREM sleep, compared to wakefulness [7,11-17]. In addition, immediate load compensation (the physiologic changes elicited to maintain normal ventilation in presence of increased upper airway resistance) decreases during sleep.

The combination of increased upper airway resistance and reduced ventilatory muscle function may lead to hypoventilation during sleep in patients with neuromuscular or chest wall disorders. The risk for hypoventilation may be enhanced by decreased load compensation that occurs in conjunction with sleep. Deficits in regulation of respiratory timing to both elastic and resistive loading have been found in small samples of awake patients with spinal muscular atrophy and Duchenne muscular dystrophy [18].

There are, in fact, conflicting data regarding the importance of sleep-related increases in upper airway resistance and whether load compensation exists. Some studies have found minimal or no immediate response to a resistive load [15,19,20]. Others have found augmented ventilatory muscle activity [11-13]. When present, this augmentation may represent a response to a resistive load itself or the elevation of arterial carbon dioxide tension (PaCO₂) that results when load compensation is incomplete [11-13]. The differing observations across studies may be related to differences in the magnitude of the resistive loads used in the various studies.

In addition, thoracoabdominal paradox is typically identified as upper airway collapse during polysomnography. In the setting of neuromuscular disease, this paradox may simply reflect the weakness of the diaphragm. This may drive some of the variability in interpreting data during sleep recordings of those with neuromuscular disease.

Chemosensitivity — Normal sleep is associated with changes in the ventilatory response to chemostimuli (hypoxemia, hypercapnia). Combined with reduced ventilatory muscle strength, abnormal lung or chest wall stiffness, or both, these changes enhance the risk for pathologic hypoventilation during sleep in patients with neuromuscular or chest wall disorders.

●The ventilatory response to hypoxic chemostimulation is decreased during NREM sleep in men, compared with wakefulness [21,22]. Such a change has not been consistently observed in women [23].

●The ventilatory response to hypercapnic chemostimulation is decreased during NREM sleep in men, compared with wakefulness [24]. A similar change has not been consistently observed in women [25].

●Ventilatory responsiveness to both hypoxic and hypercapnic chemostimulation is less during rapid eye movement (REM) sleep, compared with NREM sleep [21-27].

●An apnea "threshold" to hypocapnia is unmasked during sleep, such that apnea occurs at a higher PaCO₂ than expected from awake physiology [28-30].

These changes may predispose patients with neuromuscular or chest wall disorders to substantial ventilatory and/or gas exchange disturbances during sleep.

BLOOD GASES — Based upon the physiologic responses described above, many adults experience an increase in the arterial carbon dioxide tension (PaCO₂) of 4 to 8 mmHg and a reduction in arterial oxygen tension (PaO₂) of 3 to 10 mmHg during the transition from wakefulness to non-rapid eye movement (NREM) sleep (figure 2) [31,32]. Such a change in PaO₂ typically reduces oxyhemoglobin saturation by less than 3 percentage points.

Oxyhemoglobin desaturation — Abnormalities in oxyhemoglobin saturation during sleep may be accentuated in patients with neuromuscular or chest wall disorders (figure 3) [33-35]. In one study of 92 patients with amyotrophic lateral sclerosis (ALS) and a mean vital capacity of 49 to 56 percent of predicted, the mean sleep time with an oxyhemoglobin saturation <90 percent was 23 to 27 percent [35]. The decreased oxyhemoglobin saturation was most pronounced during rapid eye movement (REM) sleep. In another study of patients with ALS and a mean forced vital capacity of <50 percent, the mean nadir oxyhemoglobin saturation was 85 percent during the use of nocturnal noninvasive ventilation [36].

Hypoventilation is a likely cause of abnormal decreases in oxyhemoglobin saturation during sleep. This explanation is supported by an observational study of 26 patients with a mixture of respiratory disorders, which found that a reduced minute volume due to decreased tidal volume was associated with oxyhemoglobin desaturation during sleep [34]. Other etiologies, including ventilation-perfusion mismatching and rightward shift of the oxyhemoglobin dissociation curve, are also possible contributors that have not been investigated.

Hypercapnia — Awake hypercapnia may occur when lung stiffness, chest wall stiffness, or both are increased relative to the strength or endurance of the inspiratory muscles. This was demonstrated by an observational study that compared 20 patients with neuromuscular diseases to 17 healthy controls [37]. Decreased sniff nasal inspiratory pressure (indicating respiratory muscle weakness) and increased dynamic lung elastance (indicating poor compliance) were associated with rapid, low tidal volume breathing and daytime hypercapnia. This has not been specifically examined during sleep; however, the increase in PaCO₂ is likely to be initially manifested during sleep in view of the other sleep-related physiologic changes described above (altered ventilatory muscle activity, increased upper airway resistance, altered chemosensitivity).

Sleep-related hypercapnia may itself contribute to awake hypercapnia. This occurs because a persistently elevated PaCO₂ during sleep may lead to increased serum bicarbonate and buffering capacity, which persists into wakefulness. This is supported by two key observations:

●Intermittent positive pressure ventilation during sleep was associated with an augmented ventilatory response to carbon dioxide and a decreased PaCO₂ during wakefulness in an uncontrolled trial that included 16 patients with neuromuscular or chest wall disorders [38]. The findings during wakefulness were likely a consequence of improved ventilation and decreased PaCO₂ during sleep, although arterial blood gases were not reported (decreased serum bicarbonate provided evidence that PaCO₂ decreased). Changes in pulmonary function or respiratory muscle strength did not explain the findings and the potential contribution of relief of nocturnal hypoxia was not addressed.

●Overnight hypercapnia may precede daytime hypercapnia. In an observational study of six patients with a neuromuscular or chest wall disorder (in whom elevated awake PaCO₂ decreased following initiation of nocturnal noninvasive positive airway pressure), positive airway pressure ventilation was withheld for an average of one week [39]. Nocturnal oxyhemoglobin saturation and transcutaneous carbon dioxide tension worsened, without a change in daytime blood gases, pulmonary function, or ventilatory muscle strength. Presumably, the daytime blood gases would have worsened to pre-treatment levels if the nocturnal noninvasive ventilation was withheld for a longer duration.

Nocturnal hypoventilation in persons with neuromuscular disorders not only occurs in the absence of awake hypercapnia but also can be present in the absence of concurrent nocturnal hypoxemia or awake symptoms of nocturnal hypoventilation. In an observational study of 46 consecutive young patients (mean age 15 years) with pediatric onset neuromuscular disorders and forced vital capacity <60 percent predicted, nocturnal hypoventilation (as defined as transcutaneous partial pressure of carbon dioxide [PCO₂] >50 mmHg for >25 percent nocturnal recording time) was present in 29 patients, and nearly half of these patients were without evidence of concurrent nocturnal hypoxemia, symptoms of nocturnal hypoventilation, or awake elevated PCO₂ levels [40].

An additional or alternative hypothesis regarding sleep hypercapnia contributing to awake hypercapnia is that blunting of the awake ventilatory response to carbon dioxide is contributed to by sleep deprivation and/or fragmentation. According to this proposal, the improved ventilatory response to hypercapnia during wakefulness that follows intermittent positive pressure ventilation during sleep is due to improved sleep continuity [41]. However, subsequent studies in normal subjects provide compelling evidence that neither experimental sleep deprivation [42] nor fragmentation [43] is associated with a reduced ventilatory response to hypercapnia. It is unclear, however, how these data apply to humans with neuromuscular or chest wall disorders.

SUMMARY AND RECOMMENDATIONS

●Ventilatory changes during sleep – Normal sleep-related changes that impact ventilation include:

•Decreased central neural output (see 'Central neural output' above)

•A shift of the respiratory burden onto the diaphragm (see 'Diaphragmatic burden' above)

•Increased upper airway resistance (see 'Upper airway resistance' above)

•Decreased responsiveness to hypoxemia and hypercapnia (chemosensitivity) (see 'Chemosensitivity' above)

These perturbations are clinically insignificant in healthy adults but can induce or exacerbate hypoventilation and hypoxemia during sleep in patients with neuromuscular or chest wall disorders. (See 'Ventilation' above.)

●Blood gas changes during sleep – Abnormal blood gases are common during sleep. Many healthy adults experience:

•An increase in the arterial carbon dioxide tension (see 'Oxyhemoglobin desaturation' above)

•A reduction in arterial oxygen tension during sleep (see 'Hypercapnia' above)

Such sleep-related blood gas and oxyhemoglobin saturation abnormalities may be accentuated in patients with neuromuscular or chest wall disorders. (See 'Blood gases' above.)

●Clinical implications – The clinical imperatives regarding such sleep-related perturbations in this diverse patient population are discussed separately:

•(See "Evaluation of sleep-disordered breathing in adult patients with neuromuscular and chest wall disorders".)

•(See "Noninvasive ventilation in adults with chronic respiratory failure from neuromuscular and chest wall diseases: Patient selection and alternative modes of ventilatory support".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Mark Sanders, MD, who contributed to an earlier version of this topic review.