A 55-year-old female with a BMI of 42 kg·m⁻², presented for total thyroidectomy for a long-standing multinodular goiter. She appeared clinically euthyroid and her thyroid function tests were normal. Although mild retrosternal extension, high tracheal deviation and compression were demonstrated on the CT scan, the patient did not exhibit any compressive symptoms. On preoperative screening using the STOP-Bang Questionnaire, the patient was deemed to be at high risk for OSA in view of the presence of loud snoring, daytime sleepiness, history of hypertension, BMI more than 35 kg·m⁻², and age above 50 years old. She offered the information that she was told “it was difficult to insert a breathing tube” during her previous surgery 10 years ago, but could not recall further details. Airway examination demonstrated good mouth opening, a short neck but good cervical range of movement, Mallampati Class IV and thyromental distance of 6 cm. Referral to sleep physician for sleep study evaluation was offered but the patient declined due to financial reasons. She also adamantly refused awake intubation despite a thorough explanation of the indications. After discussion with the surgeon and patient, the plan was to proceed with the surgery with risk mitigating strategies in view of patient’s refusal for further investigations.

Obstructive sleep apnea (OSA) is the most common sleep-disordered breathing and is a growing problem with substantial economic cost. The presence of OSA in patients negatively influences postoperative outcomes.¹ Severe perioperative complications (i.e., death and anoxic brain injury) directly related to OSA are being increasingly reported as the central contention of medical malpractice suits, with a substantial medico-legal burden.^2,3 The prevalence of moderate to severe OSA among the general population is estimated at 13% in men and 6% in women between the ages of 30 and 70 years, with increased prevalence in the older age group of 50 to 70 years (17% in men, 9% in women). This reflects a substantial relative increase of OSA in the general population by 14% to 55% over the last two decades.⁴ The prevalence of OSA in surgical patients is even higher, with a quarter of the surgical population at high risk of OSA.^5–7 Up to 90% of these patients may have undiagnosed OSA, and 40% of them may have moderate to severe OSA if subjected to testing by polysomnography (PSG).^6,7

What Are the Perioperative Risks of a Patient with OSA?

OSA is an independent predictor of uncontrolled hypertension in patients less than 50 years old. Chronic nocturnal hypoxemia and hypercarbia trigger increased sympathetic activity, and the subsequent hemodynamic stress may lead to cardiovascular diseases through multiple mechanisms.⁸ Moderate to severe OSA is also associated with multiple medical comorbidities, including metabolic syndrome, obesity, insulin resistance, uncontrolled hypertension, heart failure, arrhythmias, coronary heart disease, and stroke.^8–10 As such, patients with OSA are at increased risk of the following postoperative complications:

1. 
   

   Cardiovascular complications—atrial fibrillation, myocardial infarction/ischemia, arrhythmia, cardiac arrest

   
   
2. 
   

   Respiratory complications—postoperative oxygen desaturation and hypoxemia (secondary to airway obstruction, opioid-induced central apnea, etc.), acute respiratory distress syndrome (ARDS), aspiration pneumonia, respiratory failure

These result in an increased incidence of noninvasive ventilation (NIV), emergency reintubation, unplanned ICU admission, and mechanical ventilation in both general and bariatric surgical populations.^11–18 There is also a possible association between OSA and postoperative delirium.¹⁹ Hence, it is imperative to identify patients at risk of OSA preoperatively to allow for adequate optimization and implementation of risk mitigation strategies.

What Is the Pathophysiology of OSA?

In normal subjects, pharyngeal muscle tone decreases during sleep, particularly during rapid eye movement (REM) sleep. The consequent airway narrowing and increased airflow resistance contributes to hypoventilation and an increase of PaCO₂ of 3 to 5 mm Hg. This response is exaggerated in patients with sleep-related disorders, including OSA.²⁰

OSA is characterized by the closure and collapse of the pharyngeal airspace during sleep. This results in airflow cessation and bouts of hypoxemia and hypercapnia with consequent repeated arousals from sleep. The narrowest part of the airway lies posterior to the soft palate, where the tongue and the soft palate are in close proximity and may be in apposition. The most common site of upper airway obstruction during non-REM sleep is typically located at this point. The posterior movement of the tongue may in addition, further occlude the retroglossal space in approximately half of the patients with sleep apnea. This segment of the oropharyngeal obstruction may further increase in length during REM sleep when the pharyngeal muscle activities are more suppressed.²⁰ Identification of the sites of obstruction during sleep is important in order to guide the implementation of appropriate therapy, for example, uvulopalatopharyngoplasty, base of tongue surgery, noninvasive positive pressure ventilation, etc.

Risk factors for developing OSA include obesity (with increased fatty deposits around the pharynx), anatomical factors such as hypertrophied adenoids and tonsils, macroglossia, retrognathia, and micrognathia.

How Is OSA Diagnosed?

Screening tests can be used to identify patients at risk of OSA, who may then be referred for diagnostic testing. The Adult Obstructive Sleep Apnea Task Force of the American Academy of Sleep Medicine (AASM) recommends the diagnosis of OSA based on clinical findings from history, physical examination, and results from objective sleep testing.²¹

Objective testing is most commonly achieved by an in-lab PSG, although home testing with portable monitors might be suitable and a reliable alternative for some patients. The frequency of obstructive events is reported as an apnea–hypopnea index (AHI) or respiratory disturbance index (RDI), in accordance with the AASM Manual for the Scoring of Sleep and Associated Events.²² Apnea is defined as cessation of airflow for at least 10 seconds. Hypopnea occurs where there is reduced airflow with oxygen desaturation of ≥3% or is associated with an arousal. According to the AASM guidelines, a diagnosis of OSA is confirmed if there are ≥15 obstructive events/hour (apnea, hypopnea, respiratory event related arousals) on PSG, or ≥5 events/hour in a patient who reports any symptoms such as unintentional sleep episodes during wakefulness, daytime sleepiness, insomnia, loud snoring described by bed partner, or observed obstruction during sleep.²¹ Severity of OSA is classified as mild (AHI ≥5 and <15), moderate (AHI 15–30), and severe (AHI >30 events/hour).

How Do We Perform a Preoperative Evaluation of a Patient with Known OSA?

A thorough history and physical examination should be performed. Targeted questions on OSA symptoms should be asked and sleep study results should be reviewed to ascertain the severity of OSA.

For patients with a known history of OSA, the following should be identified and optimized:

1. 
   

   Presence of significant comorbidities such as morbid obesity, metabolic syndrome, poorly controlled hypertension, coronary artery disease, arrhythmia, heart failure, and cerebrovascular diseases.

   
   
2. 
   

   Systemic complications of long-standing OSA include hypercarbia, hypoxemia, cor pulmonale, polycythemia, and pulmonary hypertension. The prevalence of pulmonary arterial hypertension in sleep apnea is estimated to be 20% to 34%.^23–26

A simple bedside observation of a resting oxygen saturation less than 94% on pulse oximetry, in the absence of other possible causes of hypoxemia, may be suspicious for OSA.^27,28 As the degree of pulmonary arterial hypertension associated with OSA is usually mild, the American College of Chest Physicians Guidelines Committee does not recommend routine evaluation for pulmonary hypertension in patients with OSA.²⁹

A history of therapies that have been initiated, for example, surgical treatment and positive airway pressure (PAP) therapy should be taken. PAP therapy includes continuous positive airway pressure (CPAP), bilevel positive airway pressure (BPAP), and autotitrated positive airway pressure (APAP) therapy. The recommended PAP settings and compliance of patients to the PAP treatment should be determined. Patients on home PAP therapy should be advised to bring their PAP devices to the medical facility on the day of the surgery and to continue with PAP therapy during the perioperative period. Patients who have undergone surgical treatment, as well as those who have been prescribed alternative therapy such as oral appliances or nasal resistive valves, but have not had a repeat sleep study to document the improvement or resolution of OSA, should be regarded as having a high probability of untreated OSA.² Continued use of these appliances in the perioperative setting should be encouraged.²

How Do We Perform a Preoperative Evaluation of a Patient with Suspected OSA?

Preoperative identification of patients at high risk of OSA allows for heightened awareness and the implementation of targeted interventions, which may reduce perioperative complications. Patients suspected clinically to have OSA should be referred for an early preanesthetic evaluation, to allow ample time for appropriate referral to a sleep physician and preparation of a perioperative plan. Focused evaluation includes a comprehensive medical record review, patient/family interview and screening and physical examination. Medical record review should include obtaining a history of a difficult airway, hypertension, other cardiovascular comorbidities, and other congenital/acquired medical conditions. History of observed apnea and snoring offered by the bed partner is useful. Physical examination should include the respiratory, cardiovascular, and neurologic systems. Particular attention should be paid to BMI, evaluation of the airway, and neck circumference. Serum bicarbonate level may be elevated in patients with moderate to severe OSA ( HCO 3 − ≥ 28 mmol ⋅ L − 1 ) due to renal compensatory mechanisms to chronic hypercapnia and respiratory acidosis.³⁰

Multiple screening tools and questionnaires have been developed to simplify and improve the identification of patients with possible OSA, such as the Berlin Questionnaire,³¹ the Sleep Apnea Clinical Score,³² the ASA checklist,³¹ and the STOP-Bang Questionnaire.³³ An effective screening tool should be designed with high sensitivity to capture as many suspected cases as possible with a low false-negative rate, although specificity is often compromised. The STOP-Bang Questionnaire was found to have the highest sensitivity in predicting moderate to severe OSA with high methodological validity.³⁴

Developed as a screening tool for OSA in surgical patients, the STOP-Bang Questionnaire is easy to administer to patients in a busy clinical setting and has been validated in various populations, including surgical, obese, and morbidly obese patients.³⁵ It consists of eight questions in a Yes/No format.³³ The acronym STOP-Bang assesses for snoring, tiredness, observed apnea, high blood pressure, BMI >35 kg·m⁻², age >50 years, neck circumference >40 cm, and male gender (Table 43–1). Each positive answer is allocated a score of 1. Patients are deemed to be at low risk of OSA with a STOP-Bang score of 0 to 2, at risk of OSA if their STOP-Bang score is ≥3, or at high risk of moderate to severe OSA if their score is ≥5.³⁶ For ease of administration, all items in the STOP-Bang Questionnaire are treated equally. However, not all items have an equal predictive weightage for OSA.^33,37 A weighted model for the questionnaire may predict better than a linear model.³⁸ BMI and male gender were found to have heavier weightage than neck circumference and age. Further risk stratification via a two-step strategy has been recently proposed for the group of patients with a STOP-Bang score of 3 to 4 (Figure 43–1), in which the subgroup at higher risk of OSA may be identified by examining specific combinations of the STOP-Bang Questionnaire. Compared to patients with an indiscriminate score of 3 and 4 (any 3 or 4 items positive), the probability of OSA in patients with a STOP score ≥2 + male gender and/or BMI >35 kg·m⁻² was increased by 64%.^39,40 The addition of serum bicarbonate level ( HCO 3 − ≥ 28 mmol ⋅ L − 1 ) further increases the specificity of the STOP-Bang screening in predicting moderate to severe OSA.³⁰ The 2016 guidelines on preoperative screening and assessment of patients with OSA, published by the Society of Anesthesia and Sleep Medicine, suggested that the updated STOP-Bang tool adds clinical value in being an easy method of dichotomous risk stratification of high or low risk of OSA.² It can guide the need for further assessment and optimization.

Should Surgery be Canceled to Allow for the Evaluation of Suspected OSA?

For patients with known OSA, reassessment by their sleep physician preoperatively should be considered if they have been noncompliant to treatment, defaulted follow-up, report worsening symptoms, or have undergone recent therapeutic airway surgery.

There is insufficient evidence to support cancellation or postponement of surgery in patients identified to be at high risk of OSA for advanced screening techniques or sleep testing, in the absence of evidence of significant cardiopulmonary disease.² To date, evidence is still inconclusive regarding the benefit of PAP therapy in the preoperative setting, and the duration of therapy needed to reduce perioperative risks in patients with suspected OSA.⁴¹ Hence, the subsequent management of this subset of patients is dependent on various factors: (1) urgency of surgery; (2) invasiveness of the procedure; (3) presence of significant systemic diseases (congestive heart failure, atrial fibrillation, refractory hypertension, stroke, pulmonary hypertension, resting hypoxemia not attributable to other cardiopulmonary disease); (4) presence of hypoventilation syndromes; (5) postoperative opioid requirements; and (6) planned bariatric surgery.^21,42 Taking these factors into consideration, the surgeon, the anesthesia practitioner, and the patient should jointly decide whether to: (1) proceed with surgery and manage the patient based on clinical criteria alone, taking the necessary perioperative OSA precautions with risk mitigation or (2) defer the surgery to allow time for a referral to a sleep physician, confirmation of diagnosis with a sleep study, and institution of PAP therapy if necessary.^41,42 Figure 43–2 illustrates a comprehensive algorithm for the preoperative evaluation of a patient with known or suspected OSA.

What Are the Goals of Anesthetic Management of Patients with OSA?

Severe OSA-related perioperative complications (i.e., death and anoxic brain injury) are most commonly due to difficulty in airway management (usually in the form of difficult reintubation after premature extubation) and respiratory arrest in an unmonitored setting.³ Perioperative management of patients with OSA focuses on the rapid restoration of consciousness and baseline cardiorespiratory functions after general anesthesia to minimize adverse outcomes. Risk mitigating strategies in the immediate perioperative and postoperative periods are summarized in Table 43–2 and include maintaining oxygenation, limiting the use of long-acting sedative-hypnotics, using short-acting anesthetics for rapid recovery, multimodal analgesia for opioid-sparing effect, performing a safe extubation, and appropriate postoperative disposition.

Is Regional and Local Anesthesia with Sedation Safer Than General Anesthesia?

The use of regional anesthesia may be preferred over general anesthesia⁴³ as it avoids the need for airway manipulation and the use of sedative-hypnotics and analgesics that impair awakening and postoperative neurorespiratory functions. Conscious sedation is generally safe but patients with OSA may still exhibit increased tendency for oxygen desaturation due to apnea–hypopnea.¹ If moderate sedation is required, capnography is recommended to allow for monitoring of airway patency and ventilation. However, if deep sedation is required, general anesthesia should be considered as a safer option in view of the high risk of airway obstruction during deep sedation without a secure airway.⁴²

Is OSA Associated with a Difficult Airway?

It has been suggested that a history of difficult intubation is associated with a high risk of OSA, and such patients should be screened for signs and symptoms of sleep apnea and PSG considered.^44,45 OSA has been demonstrated to be associated with difficult mask-ventilation^46–49 and difficult laryngoscopic intubation.^50–53 Difficult laryngoscopic intubation may occur as high as eight times more frequently in patients with OSA than those without OSA, and there may be a correlation between the severity of OSA and difficult laryngoscopic intubation.⁵¹

Extraglottic devices (EGDs) such as the laryngeal mask airway are effective rescue airway devices in unanticipated difficult intubation. There is little evidence available for OSA being a risk factor for difficult EGD insertion, but large neck circumference may be associated with increased difficulty in laryngeal mask insertion⁵⁴ and surgical airway access.

How Should We Prepare for a Difficult Airway?

As OSA is associated with difficult bag-mask-ventilation and laryngoscopic intubation, careful planning of airway management is required. Difficult airway adjuncts, difficult airway trolley, and skilled assistants must be available. Backup plans should be thought out (based on specific difficult airway management guidelines as recommended by the American Society of Anesthesiologists,⁵⁵ the Canadian Airway Focus Group,^56,57 or the Difficult Airway Society⁵⁸). These plans should be communicated to the team prior to induction of general anesthesia.

Maintenance of oxygenation throughout airway manipulation is vital. Adequate denitrogenation increases pulmonary oxygen reserves and delays onset of hypoxemia, hence allowing more time for safe airway management without desaturation. Denitrogenation is achieved with high flow 100% oxygen for 3 to 5 minutes until the end-tidal oxygen concentration is 90% or more. This can be further facilitated by the use of CPAP at 10 cm H₂O and with the patient propped in a 25-degree head-up position.^58–63 Obese patients may require further “ramping” to achieve the optimal alignment for laryngoscopy (Head Elevated Laryngoscopy Position [HELP]). This can be achieved by the stacking of blankets or towels, or by using specially designed elevation devices such as the Troop Elevation Pillow (Mercury Medical, Clearwater, FL) (see Figure 51–2).

Gastroesophageal reflux disease (GERD) is commonly associated with OSA.⁶⁴ Aspiration prophylaxis can be achieved with preoperative proton pump inhibitors, non-particulate antacids, rapid sequence induction, and cricoid pressure. However, if cricoid pressure interferes with bag-mask-ventilation and glottis visualization, the cricoid pressure should be reduced or released.⁵⁸

Awake intubation should be considered if the patient has had a history of difficult intubation. The early use of a video-laryngoscope is preferred as the team is able to participate more effectively through shared visualization of the patient’s airway, in addition to the benefits of improved glottic view and first pass success. If a difficult airway is suspected, the use of succinylcholine to achieve muscle paralysis may be considered due to its short duration of action. Rocuronium may now be preferred due to the lack of muscle fasciculation and its paralytic effects can be rapidly antagonized with sugammadex.

Continuous oxygenation using oxygen insufflation via a nasal cannula further increases apnea time during laryngoscopy by utilizing the principle of apneic oxygenation.^65,66 During apnea, oxygen consumption is approximately 250 mL·min⁻¹. Conversely, only 8 to 20 mL·min⁻¹ of carbon dioxide diffuses from blood into the alveoli. The differential movement of oxygen and carbon dioxide across the alveolar membrane creates a net negative pressure in the alveoli, generating a mass flow of air from the pharynx to the alveoli. This bulk flow of air allows passive oxygenation in the absence of ventilation. By increasing the oxygen flow rate to 15 L·min⁻¹ (hence increasing oxygen delivery), Nasal Oxygen During Efforts Securing a Tube (NO DESAT) has been shown to extend the apnea time in obese patients and patients with a difficult airway.⁶⁷ The administration of oxygen should persist throughout the entire duration of airway manipulation. Although high-flow oxygen through a nasal cannula can be uncomfortable due to the desiccating effects on the nasopharyngeal mucosa, it should not cause deleterious effects when used for a short period of time during airway manipulation. There is emerging evidence on the benefits of transnasal humidified high-flow oxygen of up to 70 L·min⁻¹ (Transnasal Humidified Rapid-Insufflation Ventilatory Exchange [THRIVE]) in prolonging apnea time during airway manipulation.^68,69 It must, however, be emphasized that a patent airway must be ensured for apneic oxygenation to be effective.