Emergence delirium after paediatric anaesthesia: new strategies in avoidance and treatment

Key points.

• •ED is common in preschool children undergoing general anaesthesia with sevoflurane.
• •Symptoms include agitation, hyperactivity with flailing movements, confusion, and failure to recognize/engage with people and the surroundings.
• •ED has known risk factors that include the use of short-acting volatile agents.
• •The use of TIVA with propofol reduces the risks of ED.
• •Treatment is largely drug based and should be undertaken if there is a potential risk of self-injury.

Emergence delirium (ED) was first described by Eckenhoff and colleagues¹ in the1960s, and is defined in paediatric anaesthesia as ‘a disturbance in a child’s awareness or attention to his/her environment with disorientation and perceptual alterations including hypersensitivity to stimuli and hyperactive motor behaviour in the immediate post anaesthesia period’.² ED is a diagnosis of exclusion: clearly, any child who emerges after anaesthesia and surgery in distress could be in pain, and it is essential to ensure the adequacy of analgesia with examination and appropriate treatment. In addition, other pathological causes of postoperative disorientation and agitation, such as hypoglycaemia and hypoxia, need to be considered and eliminated before labelling the symptoms as ED.

The incidence of ED is two to three times more common in children than in adults. Current data would suggest that the incidence of ED varies from 20% to 80% of all paediatric anaesthetics with most of the literature suggesting it to be close to 20%. There are defined risk factors, and usually symptoms occur within 30 min of termination of anaesthesia and last for 15–30 min. However, ED can be persistent and has been reported to continue for up to 2 days.³ While the presence of postoperative pain is highly emotive, ED causes very real distress to the child, parents, and staff in the recovery ward. It also changes the atmosphere within the postoperative unit and can precipitate sympathetic agitation in other postoperative children.

This update will look at what is new in our understanding of the condition, new techniques to reduce the incidence, and available techniques that can treat the problem of ED once it has occurred.

Cause and risk factors

The earliest reports postulated that ED was related to a sense of suffocation after head and neck surgeries. However, while there are identifiable associations with the condition that relate to patient, surgical, and anaesthetic factors, the underlying causation remains unknown. Outside the known risk factors, ED still remains unpredictable in incidence and severity, and can take the anaesthetist entirely by surprise.

Increasing concern and interest in ED have emerged with the transition in mainstream paediatric practice to shorter acting volatile anaesthetic agents and the perceived frequency of ED associated with their use. Sevoflurane and desflurane with their low blood–gas solubility coefficients are associated with rapid washout and emergence from anaesthesia. It remains unclear as to whether rapid emergence per se causes ED, or whether emergence in the presence of surgical pain and before the onset of effective analgesia, is a major factor. Both of these theories have flaws: propofol anaesthesia, which is associated with rapid emergence, has a much lower incidence of ED than volatile agents, and slower emergence from volatile anaesthesia per se does not change the incidence of ED.⁴ Moreover, while delayed onset of analgesia has associations with ED, the condition can occur after non-painful procedures or diagnostic anaesthetics, such as magnetic resonance imaging studies. A reasonable conclusion is that, rather than rapid emergence with the tardy onset of analgesia, it is the agents themselves that have some neuropharmacological stimulus to postoperative agitation in the immature nervous system. Added to this is the effect of preoperative demeanour and anxiety. Kain and colleagues⁵ have shown that children with preoperative anxiety have a 12.5% increase in behavioural changes, and every 10% increase in anxiety scores causes 10% increase in ED. Kain and colleagues⁶ also found parents of anxious children are anxious: underpinning the importance of parental and child education in perioperative setting.

Clinical presentation and diagnostic criteria

ED can be described with neurological symptoms and terms that amount to a discreet postoperative neurological state, characterised by core neurological and behavioural symptoms: motor agitation, confusion, and lack of recognition or appropriate interaction with their surrounding environment. It usually begins at the start of emergence, but the onset can be as long as 45 min after the termination of anaesthesia. Common features are kicking, thrusting their head backward, failing to make eye contact, and an inconsolable demeanour, which correlates to their failure to interact with carers, parents, or the environment. Several descriptive scales, such as the Cravero (Table 1) and Watcha (Table 3) scales, have been used to quantify the severity. One of the most popular is the Paediatric Anaesthesia Emergence Delirium (PAED) scale,² which claims a sensitivity of 64% and a specificity of 86% (Table 2).

Risk factors for ED

Preschool children

Male

Sevoflurane/desflurane anaesthesia

Ear, nose, and throat surgery

Preoperative anxiety

Parental anxiety

Child temperament

Table 1.

Cravero scale

Scale	Behaviour
1	Obtunded with no response to stimulation
2	Asleep but responsive to movement or stimulation
3	Awake and responsive
4	Crying (>3 min)
5	Thrashing behaviour that requires restraint

Table 3.

Watcha scale for ED

Behaviour	Score
Asleep	0
Calm	1
Crying, but can be controlled	2
Crying, but cannot be controlled	3
Agitated and thrashing around	4

Table 2.

PAED scale. 1: calm; 2: not calm, but easily consoled; 3: moderately agitated, not easily consoled; and 4: combative, excited, and thrashing around

Behaviour	Not at all	Just a little	Quite a bit	Very much	Extremely
Make eye contact	4	3	2	1	0
Purposeful actions	4	3	2	1	0
Aware of the surroundings	4	3	2	1	0
Restless	0	1	2	3	4
Inconsolable	0	1	2	3	4

Prevention

Given the known relationship of preoperative and operative factors in the incidence of ED, there may be merit in modifying anaesthesia technique in children with the highest risk either by reducing/avoiding exposure to volatile agents or by adding drugs that can modify the postoperative behaviour. Using this strategy, sevoflurane can be avoided completely or limited simply to a short period of gaseous induction until an i.v. cannula is in place. Total i.v. anaesthesia (TIVA) techniques have become popular in recent years with well-described techniques that can be adapted to most surgeries,⁷ and a meta-analysis has shown TIVA to reduce the risks of ED.⁷ Even outside a full TIVA anaesthetic, propofol can be a useful tool to modify ED: a single dose of propofol at a dose of 1 mg kg⁻¹ given at the end of a sevoflurane anaesthetic has been shown to significantly reduce the incidence of ED.⁸ Classical TIVA, using remifentanil and propofol, is associated with a significant reduction in postoperative delirium in children.⁹

A variety of other drugs have been used for the prevention and treatment of ED, and many have been reported in published studies (Table 4). Measures advocated include sedative/anxiolytic pre-emptive treatments that include benzodiazepines and α-2 agonists. While oral midazolam as a premedication does not appear to prevent ED,¹⁰ an i.v. dose of 0.03 mg kg⁻¹ at the end of squint surgery has been reported to be effective.¹¹ In 2013, Zhang and colleagues¹² performed a meta-analysis on a total of 447 children, which saw that the prophylactic oral administration of midazolam before operation 10–45 min before induction significantly decreases the incidence of ED in paediatric patients.¹²

Table 4.

Drugs used to prevent ED

Agent	Route/time of administration
Propofol	I.V. TIVA/end of surgery
Midazolam	P.O./i.v. at end of surgery
Clonidine	P.O. (preoperative)/i.v. (intra-/postoperative)
Dexmedetomidine	I.V. pre-/intraoperative
Fentanyl	I.V./intra-nasal intraoperative
Gabapentin	P.O./preoperative
Magnesium infusion	I.V./intraoperative
Dexamethasone	I.V./preoperative
Ketamine	Intra-nasal preoperative/i.v./intrathecal

Administrations of potent opioids (fentanyl 2 mcg kg⁻¹ or alfentanil 10 mcg kg⁻¹) on induction or during surgery appear to modulate ED frequency and severity.¹³ Intraoperative remifentanil given by infusion can reduce the incidence by 50%. Even a single dose of remifentanil (1 mcg kg⁻¹) given at the end of anaesthesia can reduce ED, although at the expense of an increased frequency of nausea and vomiting.¹⁴

A drug that has received significant attention recently in smoothing emergence from anaesthesia in children is dexmedetomidine. This α-2 agonist has limited availability in the UK at present, but a recent meta-analysis by Pickard and colleagues¹⁵ and a systematic review suggest that both dexmedetomidine and clonidine can have a significant impact on reducing the incidence of ED, but at the expense of prolonged recovery times. An intraoperative infusion of dexmedetomidine at 0.2–1 mcg kg⁻¹ h⁻¹ can significantly reduce the incidence of ED.¹⁶ However, more simply, a dose of 0.3 mcg kg⁻¹ at the end of surgery has been reported to reduce the incidence of ED from 47% to 5%. In comparative studies, dexmedetomidine has been shown to be superior to a propofol bolus of 1 mg kg⁻¹ at the end of surgery.¹⁶ Dexmedetomidine has many features, making it an ideal agent to prevent or treat ED: with analgesic, sedative, and anti-emetic properties.

While the availability of dexmedetomidine is limited, clonidine, an α-2 agonist that has been used in children for many years, is both inexpensive and readily available. Premedication with clonidine also decreases the incidence of ED¹⁷ and does not appear to cause a delay in leaving post-anaesthesia care unit (PACU). A recent meta-analysis involving 669 patients looked at the efficacy of α-2-adrenergic agonists given as premedication in the management of ED. The results indicated that these drugs significantly reduce the incidence of ED, but do not eliminate the phenomenon completely.¹⁷ Head-to-head comparisons between clonidine and dexmedetomidine have not yet been undertaken, but would be useful. Clonidine has advantages over dexmedetomidine; it has a high bioavailability (more than 90%), allowing reliable use by oral, rectal, and transdermal routes. Moreover, because clonidine has some α-1 agonist activity, it may have advantages in maintaining blood pressure.

Ketamine, when administered intra-nasally in a dose of 2 mg kg⁻¹ as a premedication, was found to reduce the incidence of emergence agitation. A single bolus dose of 0.25 mg kg⁻¹ given intravenously 10 min before the end of surgery also appears to be effective, but may prolong recovery time and also cause an increase in postoperative nausea and vomiting.¹³

Reduction in the incidence of ED was seen after a bolus dose of magnesium sulphate (30 mg kg⁻¹) followed by 10 mg kg⁻¹ h⁻¹ infusions for adenotonsillectomy,¹⁸ but this has been attributed to improved analgesia. Dexamethasone 0.2 mg kg⁻¹ before operation has also been found effective in reducing ED.¹⁸

Non-pharmacological strategies

A technique known as ADVANCE (Anxiety-reduction, Distraction, Video modelling and education, Adding parents, No excessive reassurance, Coaching, and Exposure/shaping) performed in the waiting room and at induction has shown to reduce the incidence of ED.

Pre-emptive education and preparation may have some value with parent’s information leaflets, short films, and interactive games in reducing ED.¹⁸ Reducing parental and child preoperative anxiety impacts on ED, and therefore techniques to modulate this are sensible in attempts to reduce the incidence of postoperative ED.¹⁸

Treatment

The routine use of the PAED scale can be helpful in identifying and monitoring the severity of ED once the alternative causes have been eliminated. Treatment at this stage is largely pharmacological and includes propofol 0.5−1 mg kg ¹, fentanyl 1−2 mcg kg⁻¹, or midazolam 0.1 mg kg⁻¹ i.v.¹³ Although these measures have been studied as preventive strategies at the end of surgery, it is surprising that they have not been evaluated as treatments. All of these measures can delay discharge from PACU.

Conclusion

While there has been considerable progress in the neuroscience of anaesthesia and the application of new pharmacological agents, the mystery behind the exact mechanism of ED is elusive. ED is a diagnosis of exclusion once other causes have been dismissed. The frequency of ED phenomenon can be greatly reduced given that there are identifiable pre-surgical risks, which should prompt the adoption of a low ED anaesthetic technique. Nevertheless, there are some unexpected ED events that require early identification and drug treatment. Of these, the evidence base for the best treatment is α-2 agonists¹⁵ and propofol as a single dose⁸ or as TIVA.⁷

Declaration of interest

None declared.

MCQs

The associated MCQs (to support CME/CPD activity) can be accessed at www.bjaed.org/cme/home by subscribers to BJA Education.

Biographies

Professor Andrew Wolf MA MBBChir FRCA MD is a former Consultant in Paediatric Cardiac Anaesthesia and Intensive Care at Bristol Royal Children's Hospital and Immediate Past President of the Association of Paediatric Anaesthetists.

Shrijit Nair MBBS FCAI is a specialty registrar at Our Lady's Hospital for Sick Children, Dublin.

Matrix codes: 1H02, 2D02, 3J02