Agitation lasted up to 45 min in some cases (range, 3–45 min; mean, 14 ± 11 min) and was associated with a prolonged postanesthesia recovery (117 ± 66 min versus 101 ± 61 min for nonagitated children; P = 0.02). Most of these children were thrashing (86%) and kicking (64%), whereas a smaller percentage (14%) were simply restless and incoherent. Fifty-six (60%) children in the EA group required physical restraint (i.e., held down by a nurse), and many (42%) required two or more nurses until the agitation subsided. EA subsided without pharmacologic intervention in 48% of cases, and the duration of EA in these children was shorter (11 ± 10 min) than those with EA who were treated (16 ± 10 min; P = 0.02). Fifty children (52%) required an opiate (n = 46), benzodiazepine (n = 2), or both (n = 2), whereas only 77 (18%) nonagitated children required intervention with an opiate or benzodiazepine (P < 0.0001). Finally, EA was associated with five adverse events. These included increased bleeding from the surgical site (n = 1), pulling out a surgical drain or an IV (n = 2), increased pain at the operative site (n = 1), and minor injury of the nurse (n = 1).
Table 3 presents the relationship between perioperative factors and the presence of EA. Children who underwent otorhinolaryngology or ophthalmology procedures and those who received sevoflurane or isoflurane experienced EA significantly more frequently than other children. Sodium pentothal, however, was associated with a less-frequent incidence of EA. Children who received a combination of sevoflurane and isoflurane for induction/maintenance were more than twice as likely to have EA compared with those having another anesthetic regimen (P < 0.0001). Almost all children (98%) who had EA had received intraoperative analgesics, compared with 86% of nonagitated children (P = 0.001). There was no difference in the duration of anesthesia between children who experienced EA and those who did not (61 ± 28 min versus 68 ± 48 min, respectively). However, children with EA had a significantly shorter time to awakening (14 ± 14 min versus 26 ± 23 min; P = 0.0001).
Ten variables were found, by univariate analysis, to be associated with an increased incidence of EA. These included young age, no previous surgery, poor adaptability, ophthalmology procedures, otorhinolaryngology procedures, sevoflurane, isoflurane, sevoflurane/isoflurane, analgesics, and short time to awakening. These factors (and their interactions) were subsequently entered into a logistic regression model with backward selection. Multivariate analysis of these factors yielded three that were predictive of EA. These are described in Table 4.
Agitation on emergence from GA is a frequent phenomenon in children that demands increased nursing care in the PACU, delays reunion with parents, and may lead to adverse sequelae in some cases. An understanding of the risk factors for EA is important to minimize contributory factors and to appropriately manage agitation when it occurs. This study identified multiple factors associated with EA in young children and found that the use of isoflurane, short time to awakening, and otorhinolaryngologic procedures were independent risk factors for this outcome. Although EA in this sample was of relatively short duration, pharmacologic intervention was required in 52% of cases, and EA was associated with a prolonged PACU stay and resulted in adverse events in 1% of all cases.
Several studies that have addressed EA as a postoperative complication in children have generally focused on the volatile anesthetics (sevoflurane and desflurane) as primary risk factors (8–11,14,21,31–36). The reported incidence of EA has ranged from 24% to 66% in children who received these anesthetics; however, variations in protocols and in the definition and classification of EA in these studies make it difficult to compare results. Indeed, a recent investigation by Cole et al. (37) showed that when the definition of EA was expanded to include children who were inconsolable and crying, as well as restless and disoriented, the incidence increased from 10% to 30% in one sample. Still, most data suggest that anesthetics with a low solubility are associated with an increased incidence of agitation that is, in some manner, related to abrupt emergence. Our findings are consistent with these studies in that short time to awakening and the use of sevoflurane were associated with a frequent incidence (24%) of EA. We further demonstrated that maintenance with isoflurane was significantly associated with agitation (23%). Cravero et al. (13) suggested that marked agitation on emergence may negate the advantage of rapid emergence, demonstrating that despite a quicker emergence with sevoflurane, the time to discharge was prolonged compared with halothane. Others have similarly found shorter emergence times but no difference in discharge times between sevoflurane and halothane groups (9,31,34). In this study, we found that EA significantly prolonged the PACU stay, which may have resulted from additional pharmacologic treatment and other supportive therapies necessary to manage agitation.
Several investigators have argued that pain during impaired consciousness contributes to severe EA in some children (35,36); however, a clear relationship has not been established (3–5,12). The administration of ketorolac, acetaminophen, tramadol, or fentanyl has been shown to reduce the agitation associated with sevoflurane anesthesia in children undergoing otorhinolaryngology surgery (10,18–21), suggesting a potential relationship between pain and EA. However, Murray et al. (34) demonstrated that preemptive oxycodone reduced postanesthesia agitation for children who had received halothane but not for those who received sevoflurane. Furthermore, several studies have demonstrated a clinically significant incidence of EA in presumably pain-free patients (3,5,9,12,14,22,34,37,38), suggesting that analgesics cannot completely attenuate postanesthetic agitation. Although it remains difficult to differentiate pain-related agitation from other sources, several studies have attempted to do so by incorporating pain scales, in addition to agitation scales, into the methodology (9,20,35). In this study, children were classified as having EA if they demonstrated agitation behaviors but did not localize or complain of pain. Furthermore, 98% of children who experienced EA had received a preemptive analgesic, and in 48%, the agitated behaviors resolved without pharmacologic intervention. Although pain cannot be entirely excluded as a contributory factor for EA, these data suggest the influence of another mechanism. Consideration of pain as a potentiating factor for agitation is important, particularly in children undergoing short surgical procedures for which the peak effects of analgesics may be delayed until well after they awake.
There are no recent data to help clarify the relationship between surgical procedure and EA; however, early anecdotal and descriptive reports suggested that EA was encountered more frequently in young people who underwent tonsillectomy or head and neck surgery (7,39). Furthermore, most recent investigations that address EA have selectively studied children undergoing ear, nose, and throat procedures (10,11,13,19–21,33,35,36,40,41). The choice of this study sample may have been at least partially influenced by the premise that this population is at risk for postanesthetic agitation. Our data indeed demonstrate that otorhinolaryngology procedures pose an independent risk for EA, but the explanation for this remains unclear. Eckenhoff et al. (7) and Bastron and Moyers (39) speculated that a “sense of suffocation” may contribute to EA in patients undergoing head and neck procedures, yet there are no scientific data to support this.
Several patient-related factors have also been associated with an increased incidence of EA, including young age (9,13,22,31) and anxiety or distress (11). Recent studies demonstrated that preoperative administration of midazolam reduced the incidence of EA compared with placebo (11) and with clonidine premedication (40); however, this may have been related to slowed awakening rather than anxiety. Indeed, other data have contradicted this notion (33,37,42,43). One study found that children who received midazolam experienced EA more frequently than those who did not and that the observed agitation lasted longer (37). Although Kain et al. (42) demonstrated a decreased incidence of maladaptive behaviors at two postoperative weeks after midazolam, there was no difference in EA between children who received midazolam and those who did not. Furthermore, benzodiazepines have themselves been associated with paradoxical reactions and agitation (15,44–46) that have been reversed with flumazenil (47). Our data demonstrate a similar incidence of EA in those who received preoperative midazolam compared with those who did not (15% versus 19% respectively). Furthermore, there was no relationship between behaviors at separation and induction and those at emergence. Taken together, these data raise doubts about a potential relationship between preoperative anxiety and EA.
Although the child’s temperament has not been previously studied in relation to EA, several studies have shown a relationship between certain traits and the child’s response to medical procedures or hospitalization. Kain et al. (48) demonstrated that children who were not enrolled in daycare, those with no siblings, and those who were very impulsive were at greater risk for developing negative behavior changes, such as separation anxiety, nightmares, and bedwetting, at two or more weeks after surgery. Other studies have reported that children with lower thresholds (i.e., sensitivity) (25) and those with low adaptability (i.e., inability to readily adapt to new situations) (26) displayed more distress behaviors during venipuncture or immunization. Low adaptability has also been associated with sedation failures (27). Our data demonstrated that low adaptability was associated with but was not an independent risk factor for EA. Further study in this area may provide more insight into the interaction between temperament and emergence behaviors.
This study was intended to describe emergence phenomenon in a routine clinical setting, and data are therefore subject to the limitations posed by observational, noncontrolled study designs. Because observers were not blinded to many of the factors under investigation, these data are subject to observer bias. Additionally, the MAC of anesthetics was not controlled, nor were the doses and timing of analgesics. Although consecutive sampling over two separate time periods was used to reduce the possibility of secular trends in practice, one cannot overlook the possibility of a selection bias in our sample. Our sample represented only 20% of our annual patient population with these inclusion criteria. However, our sample size was more than sufficient to detect an expected 20% incidence of EA with a CI of 99% (30). Although these limitations reduce the generalizability of our findings, the fact that our data are consistent with previous studies, many of which used randomized, controlled designs, lends external validity to our results and reduces the likelihood that these associations were due to bias.
In summary, EA remains a significant postanesthetic problem that interferes with the child’s recovery and challenges the PACU care provider in terms of assessment and treatment. An understanding of potential risk factors is important to appropriately differentiate and treat agitation in the pediatric PACU. This study identifies multiple factors associated with EA, of which short time to awakening, use of isoflurane, and otorhinolaryngologic procedures were independent risk factors. Further investigation of these factors as potential predictors is necessary before data can be generalized to other settings or populations.
Funded in part by a grant from Sigma Theta Tau International, Honor Society of Michigan, Rho Chapter.
1. Jerome EH. Recovery of the pediatric patient from anesthesia. In: Gregory GA, ed. Pediatric anesthesia. 2nd ed. New York: Churchill Livingstone, 1989: 629.
2. Olympio MA. Postanesthetic delirium: historical perspectives. J Clin Anesth 1991; 3: 60–3.
3. Wells LT, Rasch DK. Emergence “delirium” after sevoflurane anesthesia: a paranoid delusion? Anesth Analg 1999; 88: 1308–10.
4. Veyckemans F. Excitation phenomena during sevoflurane anaesthesia in children. Curr Opin Anaesthesiol 2001; 14: 339–43.
5. Holzki J, Kretz FJ. Changing aspects of sevoflurane in paediatric anaesthesia: 1975–99. Paediatr Anaesth 1999; 9: 283–6.
6. Smessaert A, Schehr CA, Artusio JFJ. Observations in the immediate postanaesthesia period. II. Mode of recovery. Br J Anaesth 1960; 32: 181–5.
7. Eckenhoff JE, Kneale DH, Dripps RD. The incidence and etiology of postanesthetic excitement. Anesthesiology 1961; 22: 667–73.
8. Davis PJ, Cohen IT, McGowan FX Jr, Latta K. Recovery characteristics of desflurane versus halothane for maintenance of anesthesia in pediatric ambulatory patients. Anesthesiology 1994; 80: 298–302.
9. Beskow A, Westrin P. Sevoflurane causes more postoperative agitation in children than does halothane. Acta Anaesthesiol Scand 1999; 43: 536–41.
10. Davis PJ, Greenberg JA, Gendelman M, Fertal K. Recovery characteristics of sevoflurane and halothane in preschool-aged children undergoing bilateral myringotomy and pressure equalization tube insertion. Anesth Analg 1999; 88: 34–8.
11. Lapin SL, Auden SM, Goldsmith LJ, Reynolds AM. Effects of sevoflurane anaesthesia on recovery in children: a comparison with halothane. Paediatr Anaesth 1999; 9: 299–304.
12. Cravero J, Surgenor S, Whalen K. Emergence agitation in paediatric patients after sevoflurane anaesthesia and no surgery: a comparison with halothane. Paediatr Anaesth 2000; 10: 419–24.
13. Cravero JP, Beach M, Dodge CP, Whalen K. Emergence characteristics of sevoflurane compared to halothane in pediatric patients undergoing bilateral pressure equalization tube insertion. J Clin Anesth 2000; 12: 397–401.
14. Uezono S, Goto T, Terui K, et al. Emergence agitation after sevoflurane versus propofol in pediatric patients. Anesth Analg 2000; 91: 563–6.
15. Doyle WL, Perrin L. Emergence delirium in a child given oral midazolam for conscious sedation. Ann Emerg Med 1994; 24: 1173–5.
16. Hollister GR, Burn JM. Side effects of ketamine in pediatric anesthesia. Anesth Analg 1974; 53: 264–7.
17. Weinger MB, Swerdlow NR, Millar WL. Acute postoperative delirium and extrapyramidal signs in a previously healthy parturient. Anesth Analg 1988; 67: 291–5.
18. Fan K-T, Lee T-H, Yu K-L, et al. Influences of tramadol on emergence characteristics from sevoflurane anesthesia in pediatric ambulatory surgery. Kaohsiung J Med Sci 2000; 16: 255–60.
19. Finkel JC, Cohen I, Hannallah RS, et al. The effect of intranasal fentanyl on the emergence characteristics after sevoflurane anesthesia in children undergoing surgery for bilateral myringotomy tube placement. Anesth Analg 2001; 92: 1164–8.
20. Galinkin JL, Fazi LM, Cuy RM, et al. Use of intranasal fentanyl in children undergoing myringotomy and tube placement during halothane and sevoflurane anesthesia. Anesthesiology 2000; 93: 1378–83.
21. Johannesson GP, Floren M, Lindahl SG. Sevoflurane for ENT-surgery in children: a comparison with halothane. Acta Anaesthesiol Scand 1995; 39: 546–50.
22. Aono J, Ueda W, Mamiya K, et al. Greater incidence of delirium during recovery from sevoflurane anesthesia in preschool boys. Anesthesiology 1997; 87: 1298–300.
23. Chung F. Recovery pattern and home-readiness after ambulatory surgery. Anesth Analg 1995; 80: 896–902.
24. McDevitt SC, Carey WB. The measurement of temperament in 3–7 year old children. J Child Psychol Psychiatry 1978; 19: 245–53.
25. Lee LW, White-Traut RC. The role of temperament in pediatric pain response. Issues Compr Pediatr Nurs 1996; 19: 49–63.
26. Schechter NL, Bernstein BA, Beck A, et al. Individual differences in children’s response to pain: role of temperament and parental characteristics. Pediatrics 1991; 87: 171–7.
27. Voepel-Lewis T, Malviya S, Prochaska G, Tait AR. Sedation failures in children undergoing MRI and CT: is temperament a factor? Paediatr Anaesth 2000; 10: 319–23.
28. Altman DG. Practical statistics for medical research. New York: Chapman and Hall, 1991.
29. Munro BH. Statistical methods for health care research. 4th ed. Philadelphia: Lippincott, 2001.
30. Hulley SB, Gove S, Browner WS, Cummings SR. Choosing the study subjects: specification and sampling. In: Hulley SB, Cummings SR, eds. Designing clinical research. Baltimore: Williams & Wilkins, 1988: 18–30,appendix 13E.
31. Lerman J, Davis PJ, Welborn LG, et al. Induction, recovery, and safety characteristics of sevoflurane in children undergoing ambulatory surgery: a comparison with halothane. Anesthesiology 1996; 84: 1332–40.
32. Michalek-Sauberer A, Wildling E, Pusch F, Semsroth M. Sevoflurane anaesthesia in paediatric patients: better than halothane? Eur J Anaesthesiol 1998; 15: 280–6.
33. Viitanen H, Annila P, Viitanen M, Tarkkila P. Premedication with midazolam delays recovery after ambulatory sevoflurane anesthesia in children. Anesth Analg 1999; 89: 75–9.
34. Murray DJ, Cole JW, Shrock CD, et al. Sevoflurane versus halothane: effect of oxycodone premedication on emergence behaviour in children. Paediatr Anaesth 2002; 12: 308–12.
35. Cohen IT, Finkel JC, Hannallah RS, et al. The effect of fentanyl on the emergence characteristics after desflurane or sevoflurane anesthesia in children. Anesth Analg 2002; 94: 1178–81.
36. Cohen IT, Hannallah RS, Hummer KA. The incidence of emergence agitation associated with desflurane anesthesia in children is reduced by fentanyl. Anesth Analg 2001; 93: 88–91.
37. Cole JW, Murray DJ, McAllister JD, Hirshberg GE. Emergence behaviour in children: defining the incidence of excitement and agitation following anaesthesia. Paediatr Anaesth 2002; 12: 442–7.
38. Kulka PJ, Bressem M, Tryba M. Clonidine prevents sevoflurane-induced agitation in children. Anesth Analg 2001; 93: 335–8.
39. Bastron RD, Moyers J. Emergence delirium. JAMA 1967; 200: 883.
40. Fazi L, Jantzen EC, Rose JB, et al. A comparison of oral clonidine and oral midazolam as preanesthetic medications in the pediatric tonsillectomy patient. Anesth Analg 2001; 92: 56–61.
41. Viitanen H, Tarkkila P, Mennander S, et al. Sevoflurane-maintained anesthesia induced with propofol or sevoflurane in small children: induction and recovery characteristics. Can J Anaesth 1999; 46: 21–8.
42. Kain ZN, Mayes LC, Wang SM, et al. Parental presence during induction of anesthesia versus sedative premedication: which intervention is more effective? Anesthesiology 1998; 89: 1147–56.
43. Kain ZN, Mayes LC, Wang SM, Hofstadter MB. Postoperative behavioral outcomes in children: effects of sedative premedication. Anesthesiology 1999; 90: 758–65.
44. Honan VJ. Paradoxical reaction to midazolam and control with flumazenil. Gastrointest Endosc 1994; 40: 86–8.
45. Khan LC, Lustik SJ. Treatment of a paradoxical reaction to midazolam with haloperidol. Anesth Analg 1997; 85: 213–5.
46. McGraw T. Oral midazolam and post-operative behaviour in children [letter]. Can J Anaesth 1993; 40: 682–3.
47. Thurston TA, Williams CG, Foshee SL. Reversal of a paradoxical reaction to midazolam with flumazenil. Anesth Analg 1996; 83: 192.
© 2003 International Anesthesia Research Society
48. Kain ZN, Mayes LC, O’Connor TZ, Cicchetti DV. Preoperative anxiety in children: predictors and outcomes. Arch Pediatr Adolesc Med 1996; 150: 1238–45.