Tracheal extubation is associated with a significant risk of complications in both adults and children. Loss of airway control in a patient after the removal of endotracheal tubes or airway devices presents a serious challenge for anesthesiologists. Failure to maintain an unobstructed natural airway and establish spontaneous respiration after extubation can lead to catastrophic respiratory complications, such as cardiovascular collapse or even lasting cerebral damage. According to the Fourth National Audit Project of the Royal College of Anaesthetists and Difficult Airway Society (NAP4)

, one-third of major airway complications occur during emergence from anesthesia in operating rooms and in Postanesthesia care unit (PACU) [1]. However, tracheal extubation has been less focused on and discussed by anesthesiology experts than tracheal intubation, for which detailed guidelines have been published and repeatedly updated [2]. To date, extubation guidelines for adult patients have been published [3]; however, no specific guidelines have been published for extubation in pediatric patients.

In pediatric patients incapable of cooperating with and understanding the medical procedures being applied to them, airway devices often need to be removed while they are, to some degree, under the influence of anesthetics. There is a danger that the lingering effects of the remaining anesthetics can compromise airway patency and respiratory reflex, which increases the risk of perioperative airway complications in this age group [4]. The removal of endotracheal tubes and airway devices is customarily conducted in two ways in pediatric anesthesia: when patients are in a deep anesthesia or after regaining consciousness. According to the Anaesthesia PRactice in Children Observational Trial (APRICOT) [5], majority of anesthesiologists prefer awake extubation in children. Deep extubation is an advanced technique [3]. Some experts are skeptical of this method because removing airway devices from deeply anesthetized patients, whose protective airway reflex is impaired by anesthetics, may present a risk of aspiration. Anesthetized patients may be incapable of maintaining airway patency; therefore, potentially collapsed upper airways may increase the risk of airway obstruction. Conversely, many pediatric anesthesiologists prefer deep extubation and removal of supraglottic airway devices (SAD). [5]. Awake extubation is associated with the risk of coughing and spasmodic expiration with desaturation in children; however, the incidence of serious complications such as laryngospasm and breath-holding does not differ between the two methods [6,7].

In both methods, assessing the readiness of a patient and optimal timing for the removal of airway devices remains crucial for anesthesiologists. However, to date, clear criteria for safe airway management and/or standard protocols for extubation and removal of airway devices in the pediatric population have not been established. Anesthesiologists decide the readiness of a patient for extubation and the method most suitable in each case based on a few clinical cues and their own experiences. This review aims to illustrate issues associated with extubation and the removal of airway devices in children and offers readers some insights on points to consider before taking this important airway management step during emergence from anesthesia to reduce the risk to pediatric patients.

In adults, extubation or removal of SADs is almost always performed after patients are awakened from anesthesia, when they are able to obey commands and their protective airway reflexes and laryngeal muscle tone recover to preanesthetic levels. The main concerns of practitioners at this moment include coughing associated with discomfort and hemodynamic changes induced by airway stimulation. The latter is particularly important in adult patients who are often compromised by pre-existing cardiovascular conditions. However, these principles do not apply to children. Children awakened from anesthesia are often confused and become agitated in unfamiliar operating room environments; therefore, they can hardly understand or obey commands from strangers. Struggling with airway devices and pulling out tubes and catheters is not rare, and these behaviors may potentially cause children to harm themselves, leading to wound hemorrhage or dehiscence, an occurrence dreaded by surgeons.

Ideally, when children emerge from anesthesia, they are calm or asleep. There should be no agitation, straining movements, kicking, or punching. After extubation or removal of airway devices, they are able to breathe without airway complications such as excessive coughing, breath-holding, panting, or airway obstruction and require the application of no more than a moderate degree of continuous positive airway pressure (CPAP) or shoulder rolls to maintain airway patency. However, achieving these clinical goals is often easier said than done, especially in young children. Children are more prone than adults to developing airway complications throughout the perioperative period because their airways are more susceptible and irritable to external stimulation [4]. Additionally, young children have naturally softer and more collapsible pharyngeal airways. Studies have demonstrated that anesthetics significantly influence airway and pharyngeal functions and alter physiological responses in a dose-dependent manner. Even in patients who are at a sub-hypnotic level of anesthesia, the effects of anesthetics are evident [8]. The effects on airway response vary according to the anesthetic used and can be modulated with the use of adjuvant agents such as opioids and lidocaine. The developmental stage of the children and the maturation of the airways should also be considered [9]. To anticipate airway responses and potential complications during emergence from anesthesia (detailed later), understanding anesthetic function and its effects on upper airway physiology in children remains essentials. This will help anesthesiologists achieve better and safer recovery from anesthesia in children.

Extubation and removal of SADs are usually conducted when patients are in one of the following two conditions: when they are at a surgical level of anesthesia and do not respond to external stimuli (deep extubation) or when they become conscious and are able to perform some purposeful movements and obey commands (awake extubation). According to the APRICOT, many anesthesiologists prefer performing awake extubation in children, probably because of concerns about irritable pediatric airways and respiratory complications [5]. However, studies have presented data showing no difference in the incidence of perioperative respiratory complications between awake and deep extubation in children [6,7]. Koo et al. [10] conducted a meta-analysis of 17 randomized control trials, and showed that deep removal of endotracheal tubes and SADs does not necessarily increase the risk of airway complications, such as laryngospasm or breath-holding. Furthermore, they showed that deep removal of endotracheal tubes and SADs reduces overall airway complication risks in children, which was contradictory to previous studies [10].

In daily practice, the tasks of selecting an optimal method and timing of extubation are decided by the anesthesiologists in charge. The tasks include the clinical evaluation of patients, such as the type of surgery performed, patients' clinical conditions, and pre-existing comorbidities, as well as the anesthesiologist’s own experience and confidence in the management of children’s airways (Fig. 1). Furthermore, the decision to remove the airway devices is made based on a careful evaluation of the recovery of a patient from anesthesia. Many studies have concluded that, extubation can be safely conducted as either awake or deeply anesthetized children if the procedure is performed adequately by experienced anesthesiologists. However, experts also agree that when patients are between the two stages of anesthesia, neither deeply anesthetized nor fully awake (i.e., in the light plane of anesthesia), extubation is unsafe and can provoke airway complications, which can lead to catastrophic events in worst-case scenarios. Therefore, the difficulty with extubation in children is not actually associated with the procedure and maneuver but in the assessment of the level of anesthesia and the readiness of the patient for airway manipulation. For either method, assessing the level of anesthesia is undoubtedly the most crucial component for the safe removal of airway devices in children.

The depth of anesthesia was first defined by Guedel in 1937. He classified the levels of anesthesia into four stages: 1 (analgesia), 2 (excitement: disinhibition and hyperreaction to external stimuli), 3 (surgical level of anesthesia), and 4 (overdose: cortex and medullary depression). Diethyl ether was the only anesthetic available during Guedel’s time; therefore, his anesthesia depth chart was based on the observation of the physiological characteristics of patients under ether anesthesia. Eventually, by the 1980s, ether was replaced with other volatile anesthetics and is no longer used. However, since there is no new substitutional classification, Guedel’s classification of anesthesia depth is still referred to today by many anesthesiologists for clinical use and education of residents and trainees. Without revised indicators to assess the level of anesthesia for the next generation of anesthetics, anesthesiologists often rely on their own experience, institutional culture, and “gut feeling” in deciding the depth of anesthesia in patients and the optimal timing for airway manipulation.

Pediatric patients vary widely in age and developmental stages. Some also have various comorbidities. This diversity in patients makes it difficult to evaluate the depth of anesthesia and further complicates determining the optimal conditions and timing for the removal of airway devices in each patient. Even in healthy, non-compromised children, visual assessment may sometimes be confusing and misleading, because lightly anesthetized and awake children sometimes respond similarly to noxious stimuli. Anesthesiologists may consider children’s reflexive responses as purposeful movements and signs that they are awake and ready for removal of airway devices. Objective measures for assessing the depth of anesthesia using various anesthetics must be established.

For awake extubation, Templeton et al. [11] conducted a prospective study on young children anesthetized with sevoflurane and investigated the criteria for safe extubation. They identified five criteria: grimace, conjugate gaze, purposeful movements, eye opening, and tidal volume > 5 ml/kg to be associated with success in awake extubation, and that the success rate increases in a stepwise manner if patients meet multiple or all of these five criteria. However, consensus on the definition of a deep level of anesthesia has never been reached, and clear criteria for the deep removal of airway devices have not been established. Previous studies on deep extubation often have not specified measurements of the depth of anesthesia or documented the protocols for extubation. The lack of criteria for extubation and the definition of “deep” anesthesia has left this method in a somewhat vague status. An end-tidal concentration of > 1 or 1.5 minimal alveolar concentration (MAC) is customarily proposed as a deep and safe level of anesthesia for extubation only when using volatile anesthetics. This is because high concentrations of volatile anesthetics eliminate the respiratory reflex to laryngeal stimulation [12]. Studies have demonstrated that sevoflurane at 0.8 MAC and desflurane 7.7%, which is approximately 1 MAC with nitrous oxide, can achieve smooth extubation in children [13,14]. However, there is no clear definition of an adequate depth of anesthesia or objective criteria for awake or deep extubation when using intravenous anesthetics or when adjuvant drugs are added to volatile anesthetics. Propofol is now one of the most commonly used anesthetics in pediatric anesthesia owing to its antiemetic effects and lower incidence of emergence delirium after anesthesia. However, clear criteria are lacking for the removal of airway devices, and there are few studies on the effect of propofol on the airways [8].

The laryngeal stimulation test is an objective test that assesses the depth of anesthesia according to the response of a patient to laryngeal stimulation and is applicable to any patient prior to extubation. Templeton et al. [11] described this simple method to determine the level of anesthesia, which involves gently moving the endotracheal tube in the cephalad and caudad directions while examining the laryngeal response of the patient. Maneuvers, such as airway suctioning or deflation of endotracheal tube cuffs, also stimulate the larynx and provoke a similar response. Gently moving supraglottic devices or deflating a cuff, if any, may provide stimuli. If a patient is in stage one, namely the awakened stage of anesthesia, laryngeal stimulation induces coughs followed by a brief pause (< 5 s) of respiration and quick recovery of spontaneous respiration. In stage two, the same stimulus provokes successive coughs followed by prolonged apnea (> 5 s), which indicates that the laryngeal systems of the patient are still under the influence of anesthetics, and extubation at this stage can provoke a strong airway response with a risk of laryngospasm and breath-holding. When patients are in the deep stage of anesthesia, namely stage 3, they usually do not respond to laryngeal or oral stimulation. Furthermore, no cessation in spontaneous respiration, or only a brief pause, is observed.

Since the end of the 20th century, patient consciousness level monitors using electroencephalograms (EEGs) have been developed and become available in anesthesia and intensive care unit (ICU) practice. The bispectral index (BIS), introduced by Aspect Medical Systems, Inc. in 1994, was one of the first developed and commercially available monitors, and is probably one of the most commonly used monitors in current practice. Its value correlates well with the consciousness level of patients under both anesthesia and sedation in ICU settings. However, certain anesthetics, including ketamine, nitrous oxide, and remimazolam [15,16], render the BIS assessment unreliable. Additionally, because young children are neurologically immature, evaluating the depth of anesthesia with the BIS in this age group is different from that in the older population. Users should acknowledge that BIS values are age-dependent. Sinha and Sood investigated BIS values in children and concluded that a BIS of approximately 60 is suitable for deep removal of a laryngeal mask airway when sevoflurane is used as an anesthetic [17]. Although the BIS can distinguish between the two stages of anesthesia, i.e. maintenance and recovery, in children, Bannister et al. [18] reported that the BIS employed in infants and young children made no difference in their recovery from anesthesia. The usefulness of EEG-based monitoring for airway management during recovery from anesthesia remains inconclusive and requires further investigation.

As shown by previous studies, there is no difference in the incidence of serious perioperative complications between awake and deep extubation [6,7,19]. Minor complications may occur but they rarely alter patient outcomes. Therefore, the choice of extubation method is left to the anesthesiologists, their preferences, and on-site clinical judgements [20]. How should anesthesiologists decide on the method of extubation or SAD removal? Several reviews have been published on extubation in pediatric patients and children with difficult airways [21,22]. The consensus reached by the authors was that either method is reasonably safe in patients with low-risk airways if the procedure is performed by experienced anesthesiologists. However, there are cases in which one method is generally preferred over another (Fig. 2).

Deep extubation is generally the choice for extubation in patients with reactive airways due to recent upper airway infection or asthma or after surgery in which coughing or cardiovascular changes induced by extubation are undesirable at the time of emergence from anesthesia. Patients with recent upper airway infections exhibit bronchial hyperreactivity similar to that in patients with asthma and an enhanced coughing reflex that lasts as long as four weeks after infection. Coughing increases intracranial and ocular pressure and should be avoided, especially after eye or intracranial surgery. Hypertension and tachycardia are undesirable outcomes after cranial surgery and in patients with cardiovascular conditions. The elevated cranial venous pressure induced by coughing and breath-holding also increases the risk of bleeding from the surgical site after cleft surgery and tonsillectomy. When children have a loose primary tooth, deep extubation may be suitable because biting tubes or bite blocks during emergence are common in awake extubation and may lead to dental injury and, at worst, aspiration of a dislodged tooth. Biting a reinforced endotracheal tube can lead to tube occlusion and airway obstruction, resulting in negative-pressure pulmonary edema (NPPE).

In contrast, awake extubation is recommended when difficult ventilation and/or intubation is identified before or at anesthesia induction, or when difficult reintubation is suspected after surgery and in full stomach cases. Patients with obesity and those with a history of apnea (sleep apnea, premature infants) are susceptible to the effects of the remaining anesthetics, especially opioids, which increase the risk of airway complications after the removal of airway devices. Most importantly, awake extubation is generally preferred in cases where anesthesiologists are uneasy about performing deep extubation.

The Difficult Airway Society guidelines recommend awake extubation for patients with difficult airways and urge anesthesia providers to plan possible reintubation scenarios prior to extubation [2]. If anesthesiologists decide that patients have airway risks for extubation, extubation can be delayed until patients regain full consciousness, when protective airway reflexes and mechanisms to maintain airway patency return to pre-anesthesia levels. If the concern for patient airway failure is high or if anesthesiologists are concerned about possible difficult reintubation, extubation with more resources, including the availability of experienced personnel and pediatric intensive care unit (PICU) admission, should be considered.

Although the endotracheal tube remains the gold standard for airway management, the use of laryngeal mask airways and other types of SADs is increasing in children. SADs have advantages over endotracheal tubes such as no contact with the vocal cords and no need for laryngoscopy or muscle relaxants for placement. By avoiding direct stimuli to the perilaryngeal structures, SADs have been reported to decrease the incidence of sore throat, cough, and laryngospasm in adult patients. However, because SADs are positioned above the larynx, concerns exist regarding aspiration and mispositioning of the device, especially in small children. Although recent studies favor the use of SADs in children and infants in terms of reducing perioperative respiratory complications [23], in a systematic review of children with upper airway infection, de Carvalho et al. [24] reported that laryngeal mask airway did not decrease the incidence of laryngospasm, breath-holding, and desaturation, although it significantly decreased coughing events. Koo et al. [10] conducted a meta-analysis and compared deep and awake removal of endotracheal tubes and laryngeal mask airways in children. They showed that deep removal of both the laryngeal mask airways and endotracheal tubes similarly reduced the overall airway complications, despite an increased risk of upper airway obstruction. No significant difference was observed between the laryngeal mask airway and endotracheal tube with regard to the risk of desaturation and laryngospasm, regardless of the level of anesthesia when removed. The only difference observed in their study was that while deep extubation significantly reduced the risk of cough, the level of anesthesia did not affect coughing associated with laryngeal mask airway removal [10]. Ramgolam et al. [19] conducted a survey of children with more than one respiratory risk factor and assessed the influence of the timing of laryngeal mask airway removal on the occurrence of adverse respiratory events. The incidence of laryngospasm and desaturation did not differ between early and late removal; however, late removal of the laryngeal mask was associated with more adverse events in the PACU [19]. Furthermore, awake removal may increase the risk of biting and damaging both the devices and teeth. Brain, the inventor of the original laryngeal mask airway, wrote that the laryngeal mask airway should not be removed until the patient regains consciousness and airway-protective reflexes. However, in children, deep removal of the laryngeal mask airway has some benefits and is associated with fewer adverse events than awake removal.

SADs can also be used as substitutes for endotracheal tubes or Guedel (oropharyngeal) airways in intubated patients. First introduced by Bailey in 1995 [25], this method aimed to reduce cough and cardiovascular instability in intubated patients upon emergence from anesthesia. The exchange of endotracheal tubes for SADs is conducted in patients who are deeply anesthetized; therefore, cough and cardiovascular changes associated with extubation can be avoided. The SADs are removed once the patients regain consciousness. However, this is an advanced technique and its efficacy in pediatric patients has not been verified.

Coughing and airway manipulation generate airborne particles and droplets, spread pathogens, and increase the chances of transmission of infectious diseases to healthcare providers. Extubation generates more aerosol particles than intubation, particularly when extubation is associated with coughing. The risk of occupational health hazards has been highlighted during the recent coronavirus pandemic, making anesthesiologists aware of the health risks to themselves and colleagues posed by their routine airway practices. Anesthesiologists must prioritize the prevention of the spread of airborne diseases among healthcare providers and keep the medical environment safe for workers. Reducing the spread of pathogens should always be considered when managing airways in perioperative and ICU settings.

The removal of airway devices is a crucial step in recovery from anesthesia. It is often associated with adverse respiratory events. In children, whose oxygen reserve is low and airway response is different from that in adults, it sometimes leads to severe complications. Careful assessment of the readiness of a patient and their ability to maintain an unobstructed natural airway is mandatory prior to the removal of airway devices. Maneuvers and resources to manage airway complications during emergence from anesthesia or extubation failure need to be considered as part of perioperative airway management.