| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The introduction of a new generation of inhaled anesthetics into pediatric clinical practice has been associated with a greater incidence of ED, a short-lived, but troublesome clinical phenomenon of uncertain etiology. A variety of anesthesia-, surgery-, patient-, and adjunct medication-related factors have been suggested to play a potential role in the development of such an event. Restless behavior upon emergence causes not only discomfort to the child, but also makes the caregivers and parents feel unhappy with the quality of recovery from anesthesia. Although the severity of agitation varies, it often requires additional nursing care, as well as treatment with analgesics or sedatives, which may delay discharge from hospital. To reduce the incidence of this adverse event, it is advisable to identify children at risk and take preventive measures, such as reducing preoperative anxiety, removing postoperative pain, and providing a quiet, stress-free environment for postanesthesia recovery. More clinical trials are needed to elucidate the cause as well as provide effective treatment.
Emergence delirium (ED) is not a new phenomenon in clinical practice. In the early 1960s, Eckenhoff et al. (1) were the first to report the signs of hyperexcitation in patients emerging from ether, cyclopropane, or ketamine anesthesia, particularly when administered for tonsillectomy, thyroidectomy, and circumcision. Children experienced postanesthesia agitation more often than adults (12%13% vs 5.3%) (2). Gradually, the above-mentioned anesthetics were either discontinued or significantly reduced as the administration of halothane became more common. In children, halothane was the predominant anesthetic for decades. With the recognition of postoperative pain management in children and the increased use of analgesics, the incidence of emergence agitation (EA) was attenuated. However, with the introduction into clinical practice of the new, short-acting, volatile anesthetics sevoflurane and desflurane, the problem of ED reemerged (3). When children were aroused from anesthesia in a quiet manner, they suddenly entered, often due to an external stimulus, a state of excitation in which they could not be consoled by the usual methods (4). Restless recovery from anesthesia may not only cause injury to the child or to the surgical site, but may also lead to the accidental removal of surgical dressings, IV catheters, and drains. Extra nursing care may often be necessary as well as supplemental sedative and/or analgesic medications, which may delay patient discharge from hospital. This adverse postanesthesia event raises the question about the "quality" of a particular anesthetic. Parents who witness ED in their child may worry about permanent sequelae. The purpose of this article is to review the most relevant advances in this field and to point out certain controversial points that persist despite continuing scientific research.
A child who emerges from anesthesia may experience a variety of behavioral disturbances that are interchangeably described in the literature as postanesthetic excitement, delirium, and agitation (5). To avoid confusion, much effort has recently been made to recognize the difference among these terms. Delirium is a complex psychiatric syndrome that includes perceptual disturbances, hallucinations, and psychomotor agitation (6). At present, there is no unique definition of ED because of its heterogeneous clinical presentation. Some authors have described such an event as a dissociated state of consciousness in which the child is irritable, uncompromising, uncooperative, incoherent, and inconsolably crying, moaning, kicking, or thrashing (7,8). Typically, these children do not recognize or identify familiar objects or people. Paranoid delusions are also possible (9). Combative behavior has been more often described than simple restlessness and incoherence (5). Sikich and Lerman (10) defined ED as "a disturbance in a child's awareness of and attention to his/her environment with disorientation and perceptual alterations including hypersensitivity to stimuli and hyperactive motor behavior in the immediate postanesthesia period." ED usually occurs within the first 30 min of recovery from anesthesia, is self-limited (515 min), and often resolves spontaneously (8,11). However, agitation and regressive behavior that lasted up to 2 days were also described (3). The term "delirium" is often replaced with the descriptive terms "agitation" or "excitation" as it is not feasible to fully evaluate a young child's psychological state during emergence (12). EA is a state of mild restlessness and mental distress that, unlike delirium, does not always suggest a significant change in behavior (13). Agitation can indicate any number of sources, including pain, physiological compromise, or anxiety (14). Delirium may be confused with agitation, but it may also be a cause of agitation. As most of the literature on this subject cannot differentiate between these two terms, we refer to the problem as EA/ED for the purpose of this article. The incidence of EA/ED largely depends on definition, age, anesthetic technique, surgical procedure, and application of adjunct medication. Generally, it ranges from 10% to 50% (1,5,11,1519), but may be as high as 80% (20,21). Despite much scientific work that deals with pediatric EA/ED, its underlying cause remains obscure. Many factors related to anesthesia, surgery, the patient, and adjunct medication have been suggested to play a potential role in its initiation (Table 1).
Rapid Emergence Postanesthesia agitation has been noted more often with the newer, less soluble, inhaled anesthetics, such as desflurane and sevoflurane, than with other volatile ones. It has been postulated that rapid awakening after the use of the insoluble anesthetics may initiate EA/ED by worsening a child's underlying sense of apprehension when finding himself in an unfamiliar environment (9,2224). Some parents claim the patient's behavior upon emergence was the same as when he was suddenly awakened from deep sleep (4). Older children and adults usually become oriented rapidly, whereas preschool-aged children, who are less able to cope with environmental stresses, tend to become agitated and delirious. However, recovery from propofol anesthesia which is also rapid, is smooth and pleasant. Several studies have shown that sevoflurane anesthesia is associated with a higher incidence of EA/ED compared with propofol (2528). Similar results were obtained when desflurane/nitrous oxide anesthesia was compared with propofol/remifentanil anesthesia (29). Delaying emergence by a slow, stepwise decrease in the concentration of inspired sevoflurane at the end of surgery did not reduce the incidence of EA, and thus questions the role of abrupt awakening in the development of EA (30). Furthermore, Cole et al. (16) found a similar incidence of EA among children who entered the postanesthesia care unit (PACU) still asleep and those who entered awake. This study should be interpreted with caution, in that it was not a randomized, controlled trial.
Intrinsic Characteristics of an Anesthetic Some authors have speculated that two unique, intrinsic characteristics of sevoflurane might account for the development of EA/ED (3). First, this anesthetic exerts an irritating side effect on the central nervous system (CNS). Second, although sevoflurane degradation products appear to cause no organ damage themselves, data are lacking on their possible interactions with other types of medications. Epileptiform activity has been reported during the use of sevoflurane anesthesia in both patients (4043) and volunteers (44) with no medical history of seizures. However, such cases have been sporadic and have had an uneventful recovery. Furthermore, desflurane, which has no proconvulsant properties (45,46), has been associated with a similar (47), if not a greater incidence of EA when compared with sevoflurane (22). These facts suggest that the causality between the CNS effects of sevoflurane and EA/ED is unlikely. As for the eventual neurotoxic influence of sevoflurane degradation products, there is no supporting scientific evidence. Postanesthesia agitation has been described not only with sevoflurane and desflurane, but also with isoflurane (5,16,19,48) and, to a much lesser extent, with halothane (16,49). Przybylo et al. (19) found significantly higher postanesthesia behavior assessment scores in children who were anesthetized for strabismus surgery with isoflurane than in those who were anesthetized with remifentanil. Other studies recorded the same incidence of ED with sevoflurane and isoflurane in children who were tracheally extubated while deeply anesthetized (25% and 40%, respectively) (48). In a prospective trial that included 521 children aged 37 yr who were scheduled for elective outpatient surgery, isoflurane was identified as an independent risk factor for EA (5). Children who received sevoflurane/isoflurane for the induction/maintenance of anesthesia were twice as likely to develop EA when compared with children who had any other anesthetic regimen (5). Considering that sevoflurane-induced electroencephalogram changes are similar to those observed during the administration of either desflurane or isoflurane (50,51), but different from changes recorded with halothane (52), EA/ED may be related to the similar CNS effects of these anesthetics, which may affect brain activity by interfering with the balance between neuronal synaptic inhibition and excitation in the CNS (42).
Pain Postoperative pain has been the most confounding variable when assessing a child's behavior upon emergence because of the overlapping clinical picture with EA/ED. Inadequate pain relief may be the cause of agitation, particularly after short surgical procedures for which peak effects of analgesics may be delayed until the child is completely awake (17).
In several studies, the preemptive analgesic approach successfully reduced EA/ED, suggesting that pain may be its major source (53). Intraoperative administration of IV ketorolac 1 mg/kg for minor otorhinolaryngological procedures decreased the incidence of EA three to four times after both halothane (42% vs 12%) and sevoflurane anesthesia (38% vs 14%). (17) This prospective, randomized study included 200 children in the age group 15 yr who were premedicated with intranasal midazolam 0.2 mg/kg. Fentanyl, given either IV 2.5 µg/kg (12,47) or intranasally 2 µg/kg (38,54) during moderately painful surgery, also decreased EA. On the other hand, postanesthesia agitation has been observed when pain was efficiently treated (15,16,35) or even when absent (20,26). Weldon et al. (35) studied 80 premedicated children aged 12 mo to 6 yr undergoing inguinal hernia repair, whose postoperative pain was managed with a preemptive caudal block. At 5 min after arrival in the PACU, agitation was significantly more frequent in sevoflurane-anesthetized children compared with halothane-anesthetized children (26% vs 6%). A higher incidence of EA was also recorded in patients who received sevoflurane for nonpainful interventions, such as magnetic resonance imaging scanning (20) and eye examinations (26). In contrast, children anesthetized with halothane and propofol for the same procedures, respectively, were free of agitation. These findings clearly suggest that EA/ED may be a clinical phenomenon that is separate from pain.
Surgery Type
Age Aono et al. (15) found that ED appeared more often with sevoflurane than with halothane in preschool boys aged 35 yr (40% vs 10%). The difference was not observed in the school-aged population. All children received oral diazepam for premedication and a caudal block for perioperative pain control. The authors speculated that the psychological immaturity of preschool children, coupled with the rapid awakening in a strange environment, may have been the main cause of ED.
Generally, younger children are more likely to show altered behavior upon recovery from anesthesia (5,19). The subpopulation of those aged 25 yr seems to be the most vulnerable as they are easily confused and frightened by unexpected and unpredictable experiences (9). In a recent commentary on the diagnosis of delirium in pediatric patients, Martini (59) addressed the role of brain maturation in the genesis of this phenomenon. He pointed out that the pediatric brain is almost a mirror image of a normal age-related regressive process with a consequent decline in norepinephrine, acetylcholine, dopamine, and
Preoperative Anxiety
Temperament Patient-related factors are an important source of variability among studies in the incidence of EA/ED as they are most difficult to control when investigating this phenomenon.
Numerous drugs, including anticholinergics, droperidol, barbiturates, opioids, benzodiazepines, and metoclopramide, may contribute to behavioral disturbances after anesthesia (13). In summary, none of the above-discussed factors has been proven to be the sole underlying cause of EA/ED. However, each factor, especially when combined with the others, may influence the behavior of a child emerging from anesthesia. The greatest barrier to a better understanding of the EA/ED etiology arises from the difficulty in comparing studies that used different definitions, age spans, surgical procedures, and measuring tools, as well different sedative, analgesic, and other adjunct medication.
Applying more than 15 different rating scales to measure EA/ED in clinical investigations suggests that none are sufficiently specific and sensitive to assess children's behavior upon emergence (10,63). In addition, it is difficult to interpret behavior in small children who are not able to verbalize pain, anxiety, hunger, or thirst. Finally, opinions diverge on the point at which emergence extends beyond "normal" (57). Agitation, which is simple to assess, is the most frequently used descriptor for emergence behavior in children (15,20,35). Most authors developed 35-point rating scales that used either crying (17) or thrashing requiring restraint (16,32) as their threshold for agitation, which had a significant influence on the calculated incidence of the event. Thus, Cravero et al. (20) recorded EA in 80% of sevoflurane patients when crying was used as a threshold, but in 33% of patients only when thrashing was applied as the threshold for agitation. Several studies have tried to distinguish pain-related agitation from other sources by incorporating both pain and agitation scales into the methodology (12,38). Postoperative agitation alone, however, does not always indicate significant behavioral changes associated with delirium (13). An accurate differentiation of delirium from other sources of agitation requires the identification of more complex symptoms of an acute mental disturbance. This differentiation may be difficult in young children who are often oppositional and unable or unwilling to answer. Przybylo et al. (19) described an assessment tool that is based on the items listed in the Diagnostic and Statistical Manual of Mental Disorders-IV for the diagnosis of delirium (6), but eliminated signs and symptoms that required verbalization or skill demonstration. Their scoring system studied perceptual disturbances, hallucinations, and psychomotor agitation in 25 children aged 29 yr who were premedicated with midazolam and randomized to receive either isoflurane or remifentanil for strabismus surgery. Rectal acetaminophen was given for postoperative pain relief. The authors concluded that while 44% of children showed altered behavior upon emergence, only 20% had complex symptoms that were consistent with delirium. Furthermore, none of these children either verbalized pain or received pain medication during the assessment period, reflecting the measurement of the phenomenon that was independent of pain-induced agitation. Sikich and Lerman (10) developed the pediatric anesthesia emergence delirium (PAED) rating scale that consists of five psychometric items for the measurement of ED in children. According to the Diagnostic and Statistical Manual of Mental Disorders -IV, three of these items are an important part of delirium and may be crucial to its differentiation from pain (6). A decreased ability of the child to make eye contact with the caregiver and a declined awareness of his surroundings reflect disturbances in consciousness with a reduced ability to focus, sustain, or shift attention. Less purposeful actions suggest cognitive changes that include perception and memory impairment as well as disorganized thinking patterns. Two other items, restlessness and inconsolable crying, reflect a disturbance in psychomotor behavior and emotion, although they may also suggest pain or apprehension. The PAED scale score correlated negatively with the child's age and time to awakening and was significantly greater in children who received sevoflurane than in those who received halothane. These results and those of another study (64) support the reliability and validity of the PAED scale. Unfortunately, the authors did not define the ED threshold, which makes the calculation of its incidence impossible and the development of therapeutic approaches rather difficult.
Given that the EA/ED etiology is still unknown, a clear-cut strategy for its prevention has not been developed. Table 2 summarizes some measures that are considered to be helpful in its prevention.
Data on the possible role of premedication in reducing EA/ED have been conflicting. Preoperative administration of midazolam decreased postoperative agitation after both sevoflurane (32) and desflurane anesthesia (66), with no delay in discharge from hospital. When analyzing those rare trials that failed to reveal the difference between sevoflurane and halothane in the incidence of EA/ED, it was striking to note that midazolam was used consistently, although at different dosages and via different routes of administration (17,22,38,39). Sevoflurane at high concentrations has been shown to enhance, and at low concentrations to block, GABAA receptor-mediated inhibition of neurotransmission in the CNS (67). Olsen et al. (68) suggested that midazolam may improve recovery after sevoflurane administration by enchancing the inhibitory effects of GABAA receptors. This proposition is supported by findings of a calmer recovery from sevoflurane-maintained anesthesia when induced with propofol but not with sevoflurane (69). Another possible explanation is that stressful induction and/or a rapid return to consciousness in nonpremedicated children may result in more behavioral disturbances upon emergence (32). On the other hand, there are studies in which midazolam premedication did not show any benefit on the quality of recovery from anesthesia (7072). This finding may possibly be the result of applying a nonspecific measuring tool or a provision of inadequate pain control (70,72). The combination of midazolam and a small dose of diazepam may extend the beneficial effects of premedication until the recovery phase, which decreases the incidence of EA/ED (73). Paradoxically, Cole et al. (16) reported an almost ninefold higher risk of the development of EA in children who were premedicated with midazolam over those who were not premedicated before outpatient surgery under either isoflurane or halothane anesthesia. Benzodiazepines themselves are associated with paradoxical reactions and agitation that are reversed with flumazenil (13,74). Furthermore, the antianalgesic effects of midazolam might worsen pain and increase the incidence of nonspecific agitation that resembles ED (75). However, the results of this study may be disputed because of a lack of randomization in the selection process. Premedication with melatonin has proven to be a good alternative to midazolam in reducing postoperative excitement (76). Oxycodone has been shown to decrease the frequency of agitation in children undergoing halothane, but not sevoflurane, anesthesia for myringotomy procedures (18). Both oral ketamine (77) and oral transmucosal fentanyl citrate (78) were also helpful, although the latter increased the incidence of side effects. Some authors have advocated switching anesthetics after induction, despite a lack of scientific evidence supporting this practice (79). Various preemptive analgesic approaches, including caudal block (35,80), fentanyl (12,47,54), ketorolac (17), clonidine (55,56), and dexmedetomidine (57,58), have been recommended to eliminate pain as a potential source of discomfort and agitation. One trial suggests that IV fentanyl 1 µg/kg before the conclusion of sevoflurane anesthesia decreases EA even after nonpainful procedures, while leaving the time of discharge unchanged (65). On the other hand, parental presence in the operating theater appeared to have no influence on the incidence and/or severity of distress behavior upon emergence (81). As for managing EA/ED, certain steps should be taken to protect the child from self-injury. Holding, as a means of providing physical restraint, and engaging more than one caregiver are often necessary. As the child may be upset by environmental stimuli, it is important to provide a quiet, darkened recovery room (82,83). The decision of whether to treat EA/ED with additional medication depends upon the severity and duration of symptoms. Many studies have shown that EA/ED is self-limited, resolving without pharmacological intervention over time (5,16,38). "Rescue" medication includes analgesics, benzodiazepines, and hypnotics. Fentanyl IV 12 µg/kg (22), propofol IV 0.51.0 mg/kg (39), and midazolam IV 0.020.10 mg/kg (12,84) have all been used for the treatment of ED. A single bolus dose of dexmedetomidine 0.5 µg/kg was also shown to be efficient in the PACU for ED (85). Voepel-Lewis et al. (86) found that PACU nurses caring for agitated children most often administered analgesics/sedatives as the first intervention, irrespective of the primary cause of agitation. However, the most effective intervention was reuniting with a parent. The authors also developed the postoperative agitation algorithm, which may serve as a guide for the assessment and treatment of ED in the PACU.
Since Eckenhoff (87) reported his findings, the contribution of anesthesia to the development of postoperative behavioral changes in children has been recognized (8890). However, there is no evidence that EA/ED has any impact on long-term outcome. Kain et al. suggested (61) that ED and new-onset postoperative maladaptive behavior changes are closely associated. The authors found that the odds ratio of having one or more new-onset postoperative maladaptive behavior changes is 1.43 for children with marked emergence status when compared with children with no symptoms of ED, but did not suggest a causeeffect relationship between these two phenomena.
It appears that almost half a century after the first cases of emergence agitation/emergence delirium (EA/ED) were reported, we do not know much more about its etiology, nor do we have a reliable assessment tool or clear-cut preventive strategy for this short-lived but troublesome clinical event. Both sevoflurane and desflurane (and possibly isoflurane) are indisputably associated with a higher incidence of altered behavior upon emergence than are either halothane or propofol. It is doubtless that younger age, preoperative anxiety, and pain are all important contributory factors. However, there are still many questions that deserve answers. What links the above-mentioned inhaled anesthetics to EA/ED? Are we able to identify a child who is at risk for the development of ED without engaging psychologists to interpret extensive questionnaires about their temperaments? Will the use of complicated psychometric tests for the evaluation of a child's mental status upon emergence change our therapeutic approach when either agitation or delirium occurs? Finally, shall we wait for this behavior to resolve spontaneously and risk a child's self-injury and parental distress, or treat it routinely and face potential adverse effects of the given medication? Obviously, further trials are necessary to discover the underlying causes of EA/ED and to determine which factors might help predict and potentially prevent it. For consistency, it seems reasonable for studies to target the most vulnerable population of preschool-aged children scheduled for similarly painful surgical procedures and to stratify patients with regard to their temperament. In addition, pain must be tightly controlled and standardized measuring tools used to evaluate postanesthesia behavior.
Accepted for publication October 9, 2006.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|