Emergence delirium and agitation in children

INTRODUCTION — Delirium and agitation can occur as a child awakens, or emerges, from anesthesia. Emergence delirium (ED) may be distressing to the patient, parents, and caregivers, and can result in inadvertent removal of intravenous (IV) catheters, drains, and dressings, and rarely, self-harm. This topic will discuss the definition, risk factors, prevention, and treatment of ED and agitation in children. ED in adults is discussed separately. (See "Delayed emergence and emergence delirium in adults".)

DEFINITION — ED is an altered state of consciousness, which begins with emergence from anesthesia and continues through the early recovery period. ED is a disturbance of awareness of, or attention to, the child's environment, and manifests as disorientation, hyperactive behavior, and hypersensitivity in the immediate period after anesthesia [1]. Emergence agitation (EA) is the umbrella term that includes ED, pain, and several other factors (table 1) [2].

INCIDENCE — The incidence of ED in children is unclear, but is likely low in modern pediatric anesthesia practice. Older studies reported estimates between 10 and 80 percent in young children [3-5]. However, the incidence of ED varies with the anesthetic agent used, age of the child, the procedure or surgery and in particular, with the criteria used to diagnose ED. Current practice routinely includes preemptive administration of analgesics, often sedatives (dexmedetomidine) and other adjunct agents to ensure a smooth emergence from anesthesia, which has reduced the clinical incidence of ED in the PACU. In one single institution observational study, the incidence of ED was <2 percent [6].    

Importantly, in young patients EA due to pain can be confused with ED. In our experience, incomplete pain relief remains the most common cause of inconsolable behavior in the recovery room and is far more common than ED. In order to diagnose ED in the post-anesthetic period the child must be completely pain-free. (See 'Differential diagnosis' below.)

PATHOGENESIS — The mechanism responsible for ED and for its increase in frequency in preschool-age children remain poorly understood. A variety of explanations have been proposed. One theory attributes ED to the unique neurodevelopmental characteristics in this age group and the effects of the newer volatile anesthetics (ie, sevoflurane, desflurane, isoflurane) on them. Another theory associates ED with the unique electroencephalographic (EEG) findings that occur with sevoflurane anesthesia. However, ED occurs with a similar frequency after desflurane and isoflurane even though the EEG tracings during these inhaled anesthetics are quite dissimilar from sevoflurane [7,8].

Several functional brain changes have been observed in children who developed ED, but these findings require verification and further study.

●EEG changes – The EEG patterns that may predict ED have yet to be determined, and EEG monitoring cannot yet be used to predict which children might develop ED in the PACU.

•Two studies reported differences in frontal lobe alpha and delta wave activity in children who developed ED during sevoflurane anesthesia, compared with children who did not develop ED [9,10].

•In one study, interictal spike events (a form of epileptiform discharge) during sevoflurane anesthesia were more common in children who developed ED [11]. This observation is consistent with evidence that such epileptiform discharges are much less common during propofol anesthesia, which does not usually result in ED.

•In another observational study, the occurrence and duration of burst suppression during anesthesia were not associated with ED [12].  

●Brain metabolism – A spectroscopy study in 59 children suggested that sevoflurane but not propofol is associated with altered metabolism in some parts of the brain, which in turn is associated with a greater risk of ED [13].

Animal studies implicate excitation of the locus coeruleus in the pathogenesis of ED [14]. This mechanism appears to be supported by the fact that alpha₂ agonists, which prevent ED, exert their effects primarily at the locus coeruleus.

RISK FACTORS — Risk factors for ED in children include preschool age and the choice of anesthetic agent. Rapid emergence from anesthesia and deeper or more prolonged anesthesia do not predispose to ED. The effect of preoperative anxiety on risk of ED remains unclear. In our experience, a prior episode of ED is not a risk factor for ED with a subsequent anesthetic (table 2).

A systematic review of risk prediction models for developing ED in children extracted only one suitable scale, the Emergence Agitation Risk Scale (EARS), that had low usability and high risk for bias [15].

Age — Studies using different diagnostic tools have reported that preschool-age children (ie, two to six years of age) are more likely to experience ED than older children, with many studies reporting an incidence of 30 to 50 percent [5,16-18]. The incidence of ED decreases with increasing age, with a frequency in adults of 4 to 20 percent depending on the definition and timing of the ED [19-21]. (See "Delayed emergence and emergence delirium in adults", section on 'Clinical features'.)

Anesthetics — The order of anesthetics most likely to be associated with ED follows the order: potent ether inhalational anesthetics (ie, sevoflurane, isoflurane, desflurane) > propofol > halothane.

A meta-analysis of 14 randomized trials with 1100 children who had general anesthesia reported that ED was less likely to occur after propofol anesthesia than sevoflurane (pooled odds ratio [OR] 0.25; 95% CI: 0.16-0.39) [22]. Another meta-analysis including nonrandomized and randomized studies (158 studies, greater than 14,000 children) also reported that ED was less likely to occur after propofol than sevoflurane (relative ratio [RR] 0.35) [23].

The mechanism for the increase in ED after sevoflurane compared with other anesthetics is unknown. Rapid emergence from general anesthesia does not increase the incidence of ED. In one study, the incidence of ED in children who were randomly assigned to either rapid emergence after abrupt discontinuation of sevoflurane or controlled, gradual emergence from sevoflurane, was similar [24]. In another study, the incidence of ED after sevoflurane was sixfold greater than with propofol anesthesia (24 versus 4 percent), despite rapid emergence in both groups [25]. The incidence of ED after sevoflurane or desflurane anesthesia is also similar [26,27].

The newer, less-soluble inhalation anesthetics (ie, sevoflurane, desflurane, and isoflurane) are associated with an increase in the incidence of ED compared with halothane, a more soluble agent [23,28,29]. A meta-analysis of 23 randomized trials including 2300 patients comparing halothane with sevoflurane reported an increase in emergence agitation with sevoflurane (pooled OR 2.21) [30]. Halothane is only available outside the United States, in a few countries around the world, and is infrequently used in pediatric anesthesia. In one study, the incidence of ED assessed with the PAED scale was 9 percent in children who had halothane for induction and maintenance of anesthesia, versus 29 percent in children who had propofol for induction and maintenance [31].

Type of surgery — When pain is prevented or adequately treated, the type of surgical procedure probably has no impact on the risk of developing ED. Studies that reported an association between specific surgical procedures and ED (eg, tonsillectomy, urologic surgery without neuraxial analgesia, strabismus surgery) may have conflated postoperative pain-related behavior with ED.

The incidence of ED after a pain-free procedure (eg, magnetic resonance imaging [MRI]) appears to be similar to that which has been reported for pain-free surgical procedures. In a small study, 32 children who underwent MRI were randomized to receive general anesthesia with sevoflurane or halothane [32]. The incidence of ED with sevoflurane when a high diagnostic threshold was applied (similar to the Pediatric Anesthesia Emergence Delirium [PAED] scale) was 33 percent, compared with 0 percent with halothane.

Duration and depth of anesthesia — The duration of deep anesthesia (bispectral index [BIS] <45) does not correlate with the risk for developing ED [33]. ED is as likely to occur after brief procedures (eg, myringotomy and tubes) as it is after surgeries of greater duration.

Neither the depth of sevoflurane anesthesia [34] nor the duration of deep sevoflurane anesthesia [33], as assessed by BIS monitoring, affects the incidence of ED in children.

Preoperative behavior — The association between preoperative mental status and the development of ED is unclear, and studies have yielded conflicting results. One retrospective study including approximately 800 anesthetized children reported a correlation between more preoperative anxiety and the odds of developing ED, as measured by a three level delirium scale [35]. However, two other studies found no relationship between preoperative anxiety and ED, as measured by the PAED scale [36,37].

DIAGNOSIS — ED is a clinical diagnosis that is based on a composite of behaviors that are present during the postoperative period.

Clinical features — During the episode, children with ED exhibit non-purposeful movement and fail to establish eye contact or to interact normally with parents or caregivers [1]. They commonly avert their eyes or stare, and exhibit non-purposeful movement. The child with ED is agitated, inconsolable, squirming, and sometimes flailing his or her arms, and often attempts to remove the monitors, ripping and grabbing at them as if they are irritating or causing pain. However, removing the monitors does not attenuate the child's delirium.

Hypoactive delirium — Delirium may also occur without signs of agitation restlessness or non-purposeful movement. This type of delirium is called hypoactive or quiet delirium. In a single institution prospective observational study of over 4400 children (age 0 to 19 years) who received general anesthesia and were assessed with both the Pediatric Anesthesia Emergence Delirium (PAED) and The Cornell Assessment of Pediatric Delirium (CAPD) scales, approximately one quarter of cases of emergence delirium were hypoactive. Conclusions from this study are limited by the very low overall incidence of ED detected, which was <10 percent of the incidence reported in many other studies [6]. (See 'Emergence delirium rating scales' below.)

The clinical significance, cause and management of hypoactive delirium in the post operative setting are unknown.  

Differential diagnosis — Other conditions that may require treatment can produce the signs and symptoms of ED. Most agitation and inconsolable behavior in young children in the recovery room have pain as their provenance, rather than ED. After many surgeries, ED and agitation from pain can occur concomitantly [38]. The differential diagnosis for agitation in the recovery room includes ED and several physiologic derangements (table 1), some of which warrant immediate intervention.

ED usually resolves within 20 minutes of onset; one study reported a mean duration of delirium of approximately six minutes after sevoflurane anesthesia for magnetic resonance imaging (MRI) [32].

Emergence delirium rating scales — A number of rating scales have been developed for the diagnosis of ED to differentiate ED from postoperative pain and to allow accurate assessment of prevalence, predisposing factors, and prevention and treatment strategies. Currently, the Pediatric Anesthesia Emergence Delirium (PAED) scale is recognized as the standard for diagnosing ED in children [38]. The PAED scale consists of five behaviors, each of which is rated on a five-level scale with a score of zero to four (table 3) [2]. The scores are summed to achieve a total scale score (maximum value is 20). The PAED scale has been validated, with a score ≥10 or 12 required to diagnose ED. Research has shown that a PAED score ≥10 yields 64 percent sensitivity and 86 percent specificity [2], and a score of >12 yields 100 percent sensitivity and 94.5 percent specificity for the diagnosis of ED [39]. A psychometric assessment of the internal consistency and reliability of the PAED scale concluded that it is a valid tool for assessing ED in children recovering from general anesthesia [40].

The PAED scale is not a continuous, linear scale. A PAED score of ≥10 is the minimum score that is required to diagnose ED; a score <10 rules out a diagnosis of ED. A score of 5 does not infer that the diagnosis of ED is half as likely as a score of 10. Similarly, a child with a score of 18 is no more likely to have ED than a child with a score of 12 [2]. When the PAED scale is used to assess the incidence of ED in a study population of children, the only valid metric is the incidence of scores greater than a threshold value (ie, 10 or 12). Reporting a mean PAED score for a study population or over time is relevant for neither clinical care and management nor as a research metric. The diagnostic accuracy of the PAED score is the subject of ongoing research.

Other rating scales, including the Watcha and Cravero scales, have not been validated. A study that compared the Watcha, Cravero, and PAED scales with each other and with ED diagnosed by an experienced anesthesiologist reported that all these methods correlated well for diagnosing ED [39]. Observer bias may have affected the results of this study, as the PAED score was consistently scored first, and a single unblinded observer performed all ratings.

Some of the behavioral factors that are included in ED rating scales may have greater predictive value than others. The first three factors in the PAED scale (ie, child makes eye contact, purposeful actions, and awareness of surroundings) track very closely with ED and poorly with pain, whereas the last two factors (ie, restlessness and inconsolability) track poorly with both ED and pain and are therefore minor factors in the diagnosis of ED [1,41]. Both the Watcha and Cravero scales require one or more of these minor factors for the diagnosis of ED, thus reducing the accuracy of these scales.

The Cornell Assessment of Pediatric Delirium (CAPD) scale is a tool that detects both active and hypoactive delirium [42]. The CAPD scale includes the five elements of the PAED scale, and in addition three others: does the child communicate needs, is the child underactive when awake, and does the child take a long time to respond. The CAPD scale has been widely used in the Pediatric Intensive Care Unit, but rarely in the postoperative setting [6,43].

PREVENTION — The most effective measure for prevention of ED is to supplant potent inhalation anesthetics (ie, sevoflurane, desflurane, and isoflurane) during general anesthesia with total intravenous anesthesia (TIVA). This would constitute a significant departure from the usual practice of pediatric anesthesia in some centers where inhalation agents are the mainstay of anesthesia for surgery in children, for a potentially brief side effect that is unlikely to occur.

We do not routinely administer prophylactic medications specifically for ED because the incidence of ED in our practice is small. We make every effort to achieve a smooth emergence from anesthesia after surgery that is likely to be painful, by ensuring adequate analgesia during emergence with an appropriate multimodal strategy. (See "Approach to the management of acute perioperative pain in infants and children".)

In addition, we do not provide prophylaxis when a parent or caregiver reports previous ED in the child; ED is age-dependent and in our experience, may not recur. Others administer a variety of medications as prophylaxis for ED. For clinicians in centers where the prevalence of ED in the recovery room is substantive, reasonable preventative strategies include a dose of propofol, midazolam, ketamine, melatonin, or an alpha₂ agonist intravenous (IV) before the end of surgery (table 4).

A number of measures to prevent ED after sevoflurane anesthesia have been investigated, which are used during induction, maintenance, or emergence from anesthesia. The choice of which preventive strategies should be used may be based on patient factors and clinician preference [44,45]. Doses and timing of administration of medications that have been studied, and the effectiveness for prevention of ED are shown in a table (table 4).

A combination of preventive medications may be more effective than individual medications alone. A 2022 network meta-analysis of 70 randomized trials (6900 patients) of children undergoing sevoflurane anesthesia compared the incidence of ED after administration of one or more of 30 preventive medications, versus placebo or active control [46]. Most combination therapies reduced ED more effectively than monotherapy; the most effective combination consisted of dexmedetomidine, midazolam, and an antiemetic. The addition of midazolam or an antiemetic enhanced the efficacy of other agents. Among single drugs, high dose melatonin (0.2 to 0.4 mg/kg) was the most effective preventive measure for ED. This study, however, like all those seeking to synthesize the ED literature, needs to be interpreted with caution due to the heterogeneity in the characteristics of the participants and surgical procedures as well as the use of different scoring systems for emergence delirium.

●Propofol – Administration of propofol near the end of sevoflurane anesthesia is a reasonable intervention to prevent ED in either pain-free procedures or procedures in which pain has been controlled intraoperatively. In one study, 230 children undergoing general anesthesia for magnetic resonance imaging (MRI) were randomly assigned to transition from sevoflurane to propofol at the end of anesthesia, or to emerge from anesthesia without propofol [29]. The incidence of ED in the children who were transitioned to propofol was reduced from 29 percent to 7 percent as were the duration and severity of ED.

Most, but not all, studies have reported that a single dose of IV propofol (3 mg/kg over three minutes, or 1 mg/kg) at the end of surgery attenuated the frequency of ED [47-52].

●Midazolam – Studies that have investigated the effects of oral and IV midazolam on the frequency of ED have yielded inconsistent results, with some studies showing a reduction in ED with midazolam [53-55] and others showing no benefit [23,56].

The timing and route of administration may be important determinants of the effect of midazolam on ED. Preoperative rectal administration of midazolam does not decrease the incidence of ED [56], whereas IV midazolam at the end of surgery may be effective. In one study, administration of midazolam 0.03 mg/kg IV just before the end of strabismus surgery reduced the incidence of ED using the Pediatric Anesthesia Emergence Delirium (PAED) scale compared with saline (17 versus 43 percent, respectively) [53] but was no more effective than when it was given at induction of anesthesia [57].

Inconsistency in the literature may in part be explained by the use of non-validated ED scales and the presence of postoperative pain.

●Opioids – Fentanyl in a dose of 0.5 to 1 mcg/kg IV or remifentanil 0.05 to 0.15 mcg/kg/minute may prevent ED. A meta-analysis of 19 randomized trials with 1500 patients concluded that prophylactic intranasal or IV fentanyl or intraoperative IV remifentanil, sufentanil, or alfentanil decreased the incidence of ED in children after sevoflurane anesthesia (pooled relative risk [RR] 0.49) [58]. Another meta-analysis of studies of ED after sevoflurane anesthesia also reported that intraoperative opioids decreased the incidence of ED [23]. Intraoperative IV fentanyl administered near the end of surgery reduced the incidence of ED, but to avoid extending the post-anesthesia care unit stay and increasing the incidence of postoperative nausea and vomiting, it should be administered at least 10 to 20 minutes before the end of surgery [59].

●Alpha₂ agonists – Oral administration of dexmedetomidine (1 mcg/kg) does not reduce the incidence of ED [60]. IV bolus administration of clonidine 2 mcg/kg or dexmedetomidine 0.3 to 1 mcg/kg reduces the incidence of ED but may also prolong the time to discharge from the hospital [23,61-64]. In a network meta-analysis of studies of ancillary drugs used to reduce ED after sevoflurane anesthesia for ophthalmic surgery, dexmedetomidine was superior to other drugs [65].

•In a 2020 meta-analysis of 58 randomized trials and five case-controlled trials (7714 subjects) that compared dexmedetomidine versus other medications for prevention of emergence agitation or ED, dexmedetomidine was superior to placebo, midazolam, and opioids but was not superior to propofol, ketamine, clonidine, chloral hydrate, melatonin, or ketofol [66].

•In a randomized trial of 156 children undergoing strabismus surgery, intranasal dexmedetomidine (0.2 mcg/kg) resulted in a four-fold reduction in ED compared with oral midazolam (0.5 mg/kg) or placebo [67].

•In another trial, 86 children aged 8 to 24 months who had sevoflurane based general anesthesia for cleft palate repair were randomly assigned to intraoperative IV infusion of dexmedetomidine 0.5 mcg/kg/hour, propofol 2 mg/kg/hour, or saline [68]. The incidence of ED using the Pediatric Anesthesia Emergence Delirium (PAED) scale was lower in children who received dexmedetomidine than in those who received propofol or saline (20.1 percent, 58.6 percent, and 85.7 percent, respectively).

●Ketamine – Ketamine reduces the incidence of ED in children who were anesthetized with sevoflurane [69] either after a bolus of 1 mg/kg IV followed by an infusion at 1 mg/kg/hour, or as a single bolus of 0.25 mg/kg IV near the end of surgery [23,70].

●Magnesium sulfate – Whether intravenous magnesium reduces the incidence of ED is unclear. Two meta-analyses were published in 2022 with conflicting results [71,72]. The included studies were small, there was high statistical and clinical heterogeneity, and various scales were used to diagnose ED. Thus, further study is needed before magnesium can be recommended for this purpose.

•In a 2022 meta-analysis of six randomized trials, administration of IV magnesium (by intraoperative bolus with or without infusion) did not reduce the incidence of ED or PAED scores in the PACU [71]. There was high statistical and clinical heterogeneity among the studies, and various scales were used to diagnose ED.

•In contrast, in another 2022 meta-analysis of eight randomized trials, including the six in the meta-analysis described above, IV magnesium reduced the incidence of ED (29.7 versus 50.5 percent, odds ratio 0.31 [95% CI 0.15 – 0.64]) [72].

At the doses used for adjunct to anesthesia, magnesium toxicity is very rare in patients with normal renal function [73]. However, magnesium may increase time to emergence [71], and can potentiate nondepolarizing neuromuscular blocking agents. (See "Clinical use of neuromuscular blocking agents in anesthesia", section on 'Factors that affect the response to neuromuscular blocking agents'.)

●Melatonin – Administration of oral melatonin may reduce the incidence of ED.

•A 2015 meta-analysis of three randomized trials including children who had general anesthesia with sevoflurane, preoperative oral melatonin (0.05 to 0.5 mg/kg) reduced the incidence of ED compared with placebo (risk ratio 0.31, 95% CI 0.16 - 0.60) [74]. Larger doses of melatonin (≥0.2 mg/kg) reduced the incidence of ED compared with midazolam premedication, although the smaller dose melatonin did not. It should be noted that different and non-validated scales were used to diagnose ED in these studies and combined with the sample sizes in most of the studies and the small number of events, the quality of evidence was graded as low.  

•In a 2021 trial of 132 children who underwent sevoflurane anesthesia and were randomly assigned to oral melatonin (0.3 mg/kg), midazolam (0.3 mg/kg), or placebo (honey) premedication, melatonin reduced the incidence of ED compared with placebo (27 versus 50 percent) and compared with midazolam (27 versus 56 percent) [75]. The incidence of ED was similar in patients who received midazolam versus placebo. ED was assessed using the Watcha scale. (See 'Emergence delirium rating scales' above.)

•In the network meta-analysis described above, melatonin was the most effective monotherapy for reduction of ED in children who underwent sevoflurane anesthesia [46].

●Others – Other strategies that may reduce the incidence of ED include listening to the sound of the mother’s voice during emergence from anesthesia [76,77], exposure to monochromatic blue light upon entry to the postanesthesia care unit (PACU) [78], and preoperative visual conditioning (wearing an eye patch) before ophthalmic surgery [79]. In randomized trials, virtual preoperative visits [80] and acupuncture HT7 (heart 7) stimulation throughout abdominal surgery [81] did not reduce the incidence of ED.

TREATMENT — Many children with ED require no treatment other than support and prevention of harm. Approximately 95 percent of cases of ED in children resolve spontaneously within 20 minutes of onset without lasting sequelae.

Children with purported agitation in the recovery room should first be assessed for potentially dangerous causes of agitation (ie, hypoxia, hypotension, hypercarbia, hypoglycemia) (table 1) and for pain and, if pain is a problem, treated appropriately with analgesics.

In theory, delirious children may attempt to pull out their intravenous (IV) cannulae, remove dressings, or hurt themselves, but these events are often easy to manage in small children. Occasionally, treatment may be indicated to prevent disruption of the surgical site (eg, after reconstructive plastic surgery).

Parents or caregivers should be included in treatment decisions whenever possible. When ED is explained, including the expected 15- to 20 minute duration, many parents or caregivers agree to wait for resolution without administration of a sedative medication.

Sedatives and opioids may be administered when treatment is required (table 5). In a single center randomized trial of 53 children with ED diagnosed by a pediatric anesthesia emergence delirium (PAED) score ≥15, a single dose of dexmedetomidine (0.5 mcg/kg IV) was more effective than propofol (1 mg/kg) for treating ED (PAED score <12 in 100 percent versus 70 percent of patients, respectively) [82]. Postanesthesia care unit (PACU) recovery times were similar.

A survey of Canadian anesthesiologists reported that interventions used most frequently to terminate ED in children were propofol (42 percent), midazolam (31 percent), fentanyl (10 percent), morphine (7 percent), and dexmedetomidine (5 percent) [44].

In our experience, a single IV dose of any of these treatments (eg, propofol 1 to 2 mg/kg, midazolam 0.05 mg/kg, fentanyl 1 to 2 mcg/kg, dexmedetomidine 0.3 mcg/kg) sedates the child, who then emerges from the sedation without evidence of ED. Opioids may be effective because of their sedative effects or by relieving pain that was misdiagnosed as ED.

Once the episode of ED has resolved, whether by spontaneous resolution or an intervention, recrudescence of ED has not been reported.

SUMMARY AND RECOMMENDATIONS

●Definition and incidence – Emergence delirium (ED) is an altered state of consciousness that may occur as a child awakens, or emerges, from anesthesia, manifesting as disorientation, hyperactive behavior, and hypersensitivity. (See 'Definition' above.)

ED may occur in as many of 50 percent of pediatric patients after general anesthesia. (See 'Incidence' above.)

It is most common in preschool-age children and after anesthesia with the potent inhaled anesthetics (eg, sevoflurane) (table 2). (See 'Risk factors' above.)

●Diagnosis – ED is a clinical diagnosis based on a composite of behaviors and exclusion of other conditions that may occur in the immediate postoperative period. Most agitation and inconsolable behavior in young children in the recovery room is related to pain, rather than ED (table 1). (See 'Clinical features' above.)

The Pediatric Anesthesia Emergence Delirium (PAED) scale is a validated scoring system that may be used to diagnose ED and differentiate it from pain-related agitation (table 3). (See 'Emergence delirium rating scales' above.)

●Prevention – The most effective preventative measure for ED is to administer total intravenous anesthesia (TIVA) instead of the potent inhalation anesthetics (ie, sevoflurane, desflurane, and isoflurane). We do not routinely use TIVA or administer prophylactic medications for ED since the incidence of ED in our practice is small.

In centers where the frequency of ED is substantial, a variety of medications may be administered in the perioperative period to prevent ED associated with sevoflurane anesthesia, including propofol, midazolam, opioids, ketamine, and alpha₂ agonists (table 4). (See 'Prevention' above.)

●Treatment – Children with purported agitation in the recovery room should first be assessed for potentially dangerous causes of agitation (ie, hypoxia, hypotension, hypercarbia, hypoglycemia) (table 1) and for pain, with immediate and appropriate treatment for any of these causes. (See 'Treatment' above.)

Most episodes of ED last up to 20 minutes, and resolve spontaneously, without treatment other than support and strategies to prevent harm. When necessary, sedatives and opioids may be administered to terminate an episode (table 5). (See 'Treatment' above.)