Bernie Sunwoo, Nirav P. Patel^†, Aharon Sareli and Richard J. Schwab

Obstructive Sleep Apneas (OSAs)

Obstructive sleep apnea (OSA), a condition of upper airway instability, is present when there are repetitive episodes of cessation of respiration (apnea), decrements in airflow (hypopnea), or both during sleep, associated with sleep fragmentation, arousals, and reductions in oxygen saturation. An apnea (defined as no flow for > 10 seconds) can be obstructive, central, or mixed (i.e., both obstructive and central). A hypopnea is commonly defined as a decrement in airflow of 50% or more for 10 seconds associated with a 4% fall in oxygen saturation or electroencephalographic (EEG) arousal. However, the exact definition of a hypopnea remains debated.

The diagnosis of OSA is most often made in the outpatient setting following an in-laboratory polysomnogram, although this may change with the expansion of coverage for portable, unattended, in-home sleep studies. The apnea/hypopnea index (AHI, referring to the number of apneas plus hypopneas per hour of sleep) is the standard metric used to quantify the severity of obstructive sleep apnea. Although there is no consensus definition of OSA, an AHI > 5 events/hour with associated symptoms (or 15 or greater regardless of associated symptoms) is typically used to define OSA.

Clinical features suggestive of OSA include snoring, snorting or gasping, witnessed apnea, and daytime somnolence. Patients typically have large neck circumference measurements (men > 17 inches, woman > 15 inches), macroglossia, elongated or enlarged soft palate, lateral narrowing of the pharynx, tonsillar hypertrophy, retrognathia, and a crowded upper airway (e.g., class 4 modified Mallampati class; Figure 80.1).

Anatomic narrowing of the upper airway is a major factor in the pathogenesis of SBRDs. In addition, a neural component with reduction in the activity of the upper airway dilator muscles, predisposing to upper airway collapse, during sleep is likely involved. Repeated apneic events often result in oxyhemoglobin desaturation, arousals (change in sleep state from a deeper stage of sleep to a higher stage of sleep or wakefulness, often manifested by the presence of an alpha pattern on the electroencephalogram), sleep fragmentation (disruption of the normal sequence of sleep stages and cycling), and sleep deprivation, especially decreased rapid eye movement (REM) sleep.

OSA is a systemic disorder associated with various clinical consequences including systemic hypertension, cerebrovascular disease, arrhythmias, coronary artery disease, congestive heart failure, mild pulmonary hypertension, and insulin resistance. Although patients may report loud snoring, apneas or choking witnessed by bed partners, excessive daytime sleepiness, unrefreshing sleep, nocturia, intellectual impairment, and irritability, a detailed sleep history is often limited in the ICU and a high index of clinical suspicion is necessary. This is particularly true in the obese patient, as obesity is the major risk factor for OSA. Oropharyngeal examination may be helpful in identifying ICU patients with obstructive sleep apnea. Such patients may have a high Mallampati classification—that is, class 3 or 4 during oral cavity examination (see Figure 80.1)—retrognathia, and enlargement of the tongue, lateral peritonsillar tissue, soft palate, and tonsils.

In inpatients, overnight pulse oximetry and portable sleep testing can assist with the diagnosis of OSA. However, in intubated patients, the assessment for OSA can be challenging. Nonetheless, history (from family) and physical examination combined with an assessment of the airway (typical features stated previously) upon extubation can raise appropriate suspicion. This is important because intubated patients with OSA should be extubated to continuous positive airway pressure (CPAP).

The American Society for Anesthesia has developed a tool for the identification and assessment of OSA in the preoperative period. A significant level of confidence for the diagnosis can be achieved at the bedside by observing sleep-related cyclical oxygen desaturations associated with apnea. First-line treatment of OSA is with CPAP. Pressures can be titrated by observing residual apnea and cyclical desaturations during sleep or by utilizing auto-titrating CPAP units. Of note, third-party payers may require polysomnography-diagnosed OSA to justify coverage of a CPAP apparatus. Therefore, high-risk patients can be discharged from the hospital to the sleep laboratory (for polysomnography) to expedite diagnosis and to initiate treatment.

It is important to identify occult or undiagnosed OSA in the acute setting, because many studies describe significant associations with common medical indications for ICU admission, including atrial fibrillation, stroke, myocardial infarction, hypertension, pulmonary hypertension, and pulmonary edema. Several case reports suggest that OSA is the cause of acute presentations of negative pressure pulmonary edema because of large negative swings in intrathoracic pressure. OSA remains associated with an increase in perioperative morbidity (episodic hypoxemia, sudden respiratory arrest, myocardial ischemia, heart block, and delirium), unplanned ICU transfer, and a prolonged length of ICU stay.

Patients with suspected OSA need careful attention to airway management, as both intubation and extubation may be more challenging. Intubating patients with OSA can be difficult because of a crowded upper airway and may require experienced anesthesiologists or fiberoptic intubation. The timing of extubation should be guided by the patient’s alertness, which in turn is influenced by several factors including sedatives and narcotics. Centrally acting medications—for example, benzodiazepines, anesthetic agents, and narcotics—alter normal responses to hypoxia and hypercapnia and facilitate collapse of the upper airway in OSA patients. At the time of extubation, patients with a history or risk for OSA should have positive airway pressure therapy available. Absent contraindications, routine extubation of OSA patients to CPAP is preferred, although no studies address this issue.

Central Sleep Apneas (CSAs)

Central sleep apnea (CSA) is characterized by repeated episodes of apnea during sleep in the absence of any respiratory effort. The pathogenesis of central sleep apnea is related to a transient loss of central nervous system drive to the respiratory muscles (Chapter 1). In contrast, in OSA, respiratory effort is preserved but airflow is prevented by upper airway occlusion. An overnight polysomnography (which monitors the electroencephalogram, extraocular muscles, airflow at mouth and nose, and movements of the chest wall, abdomen, and extremities) can differentiate central from obstructive apnea, but currently their utility in the ICU remains unknown. CSA, but not OSA, produces the absence of nasal-oral airflow and thoracoabdominal excursion observed on the polysomnogram.

Although CSA can be idiopathic, more commonly CSA in the ICU is due to congestive heart failure where Cheyne-Stokes breathing (a waxing-waning pattern of respiration) may be observed, along with brain stem and neuromuscular disease and opioid use.

Cheyne-Stokes respiration (CSR), a subtype of CSA, represents periodic breathing with alternations between apnea and hyperpnea (Figure 80.2). CSR manifests in patients with severe congestive heart failure (CHF) and stroke, with a cited prevalence in CHF varying widely between 30% and 100%. The primary hypothesized pathophysiology is hyperventilation, producing hypocapnia below the threshold required to trigger breathing.

Importantly, CSR-CSA is associated with worse outcomes in CHF patients. Widespread utilization of beta-blocker therapy for CHF was thought to have reduced the prevalence of CSR, until a prospective study failed to validate this conclusion. As with OSA, a presumptive diagnosis of CSR-CSA can be established with bedside observation of an apnea-hyperpnea cycle lasting at least 45 seconds during sleep or sleep onset. The identification of CSR-CSA often signifies a need to optimize the CHF medical regimen as the primary treatment for CSR-CSA.

However, CPAP, bi-level ventilation (bi-level), and adaptive servo-ventilation (ASV) complement medical therapy. ASV provides servo-controlled, independently varying expiratory and inspiratory pressure support, based on the detection of CSR with a backup respiratory rate. The machine servo-controls the patient’s ventilation to achieve a target of 90% of the long-term average ventilation. Studies utilizing ASV in CSR-CSA patients have reduced central events, micro-arousals, daytime sleepiness, and improved compliance compared to patients on CPAP or sham ASV. However, to date, no impact of ASV upon cardiovascular outcomes has been proven.

Obesity Hypoventilation Syndrome (OHS)

The definition of OHS is challenging and hampered by differing nomenclature and criteria. The American Academy of Sleep Medicine developed criteria for sleep hypoventilation syndrome, which subsumes OHS, although current literature commonly utilizes the latter term (Table 80.2). In spite of the discrepant definitions and criteria, the well-established phenotype of OHS includes obesity (especially a body mass index [BMI] > 40 kg/m² that the CDC defines as morbid obesity) and chronic alveolar hypercapnia (daytime Pco₂ > 45 mm Hg) in the absence of other etiologies of alveolar hypoventilation (such as existing pulmonary and neuromuscular disease). Oxyhemoglobin desaturations detected by nocturnal pulse oximetry can clue the diagnosis in an obese patient (Figure 80.3).