Recovery Characteristics of Sevoflurane and Halothane in Preschool-Aged Children Undergoing Bilateral Myringotomy and Pressure Equalization Tube Insertion

Abstract

This double-blinded study was undertaken to prospectively evaluate the role of halothane and sevoflurane and the use of IV ketorolac on the anesthetic emergence in a group of children undergoing bilateral myringotomy with pressure equalization tube procedures. Two-hundred ASA physical status I and II patients were premedicated with nasal midazolam (0.2 mg/kg) and randomized to one of four groups (Group 1 - halothane and ketorolac; Group 2 - halothane and placebo; Group 3 - sevoflurane and ketorolac; Group 4 - sevoflurane and placebo). A blinded nurse observer characterized the quality of the anesthetic emergence and recorded the incidence of emesis and the use of pain medications in the recovery room. There were no differences in age, weight, previous anesthetic experience, or duration of anesthesia among the four groups. There was no difference in the incidence of emergence agitation for patients anesthetized with sevoflurane compared with halothane, regardless of whether they received ketorolac or placebo. Regardless of the anesthetic, the incidence of emergence agitation was significantly less in patients who received ketorolac compared with patients who received placebo. The incidence of emesis in the recovery room, the total 24-h incidence of emesis, and the use of at-home pain medications were similar in all four groups. Implications: We conclude that the incidence of emergence agitation in children undergoing ultrashort anesthetic procedures is similar for sevoflurane and halothane and that ketorolac markedly diminishes emergence agitation and/or pain behavior.

(Anesth Analg 1999;88:34-8)

Section Editor: William J. Greeley.

As the practice of anesthesia introduces newer drugs that can provide for rapid, smooth anesthetic inductions; reliable intraoperative analgesia and amnesia; and rapid emergence, the role of the anesthetic emergence has taken on clinical significance. This is especially true for pediatric anesthesia, in which the quality of the anesthetic emergence may be a dilemma for both the patient and the anesthesia care provider. Sevoflurane, with its low blood-gas solubility rapid-induction and emergence characteristics, coupled with its nonirritating airway properties and stable patient hemodynamic characteristics, make the drug a desirable anesthetic for use in children. However, recent reports suggest that, compared with halothane, sevoflurane may be associated with a higher incidence of postoperative delirium ^[1-6]. For the child, a rapid return to consciousness in an unfamiliar area, in the presence of strangers, along with concerns of bodily injury and pain, may exacerbate a child's underlying sense of fear and apprehension and increase the chance of a delirious emergence. Because pain and discomfort in preverbal and preschool-aged children may be indistinguishable from the behavior associated with delirium on emergence from anesthesia, this double-blinded study was undertaken to evaluate the role of the anesthetic and the use of an IV nonsteroidal analgesic on the quality of the anesthetic emergence in a group of pediatric patients at risk of emergence agitation.

Methods

After institutional review board approval and informed, written consent from a patient's parent, 200 ASA physical status I and II patients, 1-5 yr of age undergoing ambulatory surgery for bilateral myringotomy and tube insertion (BMT) were randomly entered into a placebo-controlled, double-blinded study involving four treatment groups. The patients were randomized to their assigned group using a computer-generated random number code. For all children, parents provided information about any previous surgical experience. All children were premedicated with intranasal midazolam (0.2 mg/kg). All patients underwent an inhaled induction of anesthesia with either sevoflurane or halothane using a coaxial pediatric circuit. Once the patients were adequately anesthetized, and while the surgeon was performing the procedure, the patients received an IV injection of an equal volume of either saline or ketorolac (1 mg/kg). These injections were prepared by the hospital pharmacist and placed in specially labelled syringes. Patients, physicians, and the research nurse were blinded to the syringe's contents. Anesthesia in patients in Groups 1 and 2 was induced with nitrous oxide, oxygen, and incremental increases in the inspired halothane concentration; anesthesia in patients in Groups 3 and 4 was induced with 8% sevoflurane, nitrous oxide, and oxygen. Patients in Groups 1 and 3 received an IV injection of ketorolac (1 mg/kg), whereas patients in Groups 2 and 4 received IV isotonic sodium chloride solution. Spontaneous ventilation was maintained and anesthetic depth was adjusted to meet each individual's clinical needs. At the time of the tympanotomy tube insertion, the surgeons reported the condition of the middle ear on a scale of 1-4: 1 = no fluid; 2 = serous fluid; 3 = pus; and 4 = thick tenacious mucus (glue ear). At the completion of the surgical procedure, the anesthetics were discontinued, and the patient was allowed to breathe 100% oxygen. The duration of anesthesia was recorded and noted as the difference between the time from the start of the anesthetic induction and the time the anesthetics were discontinued.

The patients were subsequently transferred to the recovery room. In the recovery room, the anesthetic emergence was evaluated by the research study nurse who was blinded to the patient's anesthetic and analgesic grouping. The study nurse continuously assessed the quality of each patient's recovery from anesthesia. Emergence from anesthesia was characterized on a 3-point scale: 1 = asleep, calm, or mildly agitated but easily consolable; 2 = moderately agitated or restless but inconsolable; and 3 = hysterical, crying inconsolably, or thrashing. The research nurse assessed anesthetic emergence until the time the patient was discharged from the hospital. For purposes of the study, patients were considered agitated if they had scores of 2 or 3.

For patients in whom simple comfort measures (presence of parents, oral fluids, and physically holding the child) did not ameliorate the symptoms, oral acetaminophen was administered (15 mg/kg) and recorded. At our institution, patients undergoing BMT are discharged home from the recovery room. The criterion for discharge home is based on our institution's 10-point scale for recovery with patients requiring a minimum of 9 points ^[7,8]. Although oral fluids are offered to patients, the taking and retaining of oral fluids is not required for hospital discharge. The incidence of emesis in the postanesthesia care unit was also recorded by the study nurse. An episode of emesis was defined as the expulsion of stomach contents through the mouth. An episode of retching (dry heaves) was an attempt to vomit that was not productive of any stomach contents. An emetic episode was defined as a single vomit or retch. Because nausea is difficult to quantify in this age group, it was not assessed. In addition to assessing the quality of the emergence, the research nurse also telephoned the patients' parents 24 h later to ascertain the incidence of at-home emesis, the need for pain medications, and the overall quality of the patient's and family's surgical experience.

Data were analyzed using descriptive statistics, chi squared, and the Cochran-Mantel-Haenszel test. Statistical significance was assumed for P < 0.05. A sample size of 50 patients per treatment group was calculated to detect a 30% difference in the incidence in agitation between anesthetics (55% vs 25%) with an alpha of 0.05 and a beta of 0.2 (power 80%).

Results

There were no significant differences in the ages, weights, previous anesthetic experience, or duration of anesthesia among the four groups (Table 1). There was no association of emergence agitation with age, weight, recovery room emesis, or ear score. There was no difference in emergence agitation for patients anesthetized with sevoflurane compared with halothane, regardless of whether they received the ketorolac or placebo. Regardless of the anesthetic, the incidence of emergence agitation was significantly less in patients who received ketorolac than in patients administered placebo (Table 2). The incidence of rescue medication use (oral acetaminophen) in the recovery room was greater for patients who received placebo compared with those who received ketorolac. The incidence of the 24-h at-home use of analgesics was similar for all four groups. There was no difference in the use of rescue medication in the recovery room for halothane-versus sevoflurane-anesthetized patients. There was no difference among the four groups in the recovery-room incidence of emesis and 24-h incidence of emesis (Table 2).

Table 1: Patient Demographics

Table 2: Recovery Characteristics

Discussion

The results of this study show that, for children undergoing ultrashort anesthetic procedures (i.e., <10 min), the incidence of emergence agitation seems to be similar for both halothane and sevoflurane. In addition, the use of the nonopioid, nonsteroidal analgesic ketorolac markedly decreased the incidence of emergence agitation or pain associated with both halothane and sevoflurane.

Sevoflurane, with its low blood-gas solubility, rapid induction, and rapid emergence characteristics, coupled with its nonirritating airway properties and stable patient hemodynamic characteristics, is desirable for use in children. Although low soluble anesthetics allow for faster emergences, they have also been associated with higher incidences of emergence agitation ^[1-6,7,9,10]. Consequently, the advantage of a rapid emergence may be more than offset by the quality of the anesthetic emergence. In pediatric studies of desflurane and halothane emergence, excitement was noted to be more common in patients anesthetized with desflurane than in patients anesthetized with halothane ^[7,9,10]. In a study of pediatric ambulatory patients undergoing a variety of surgical procedures, Lerman et al. ^[1] noted that emergence excitement was 3 times more common in sevoflurane-anesthetized patients than in those patients receiving halothane. Aona et al. ^[3] also noted that the incidence of emergence excitement was greater with sevoflurane than with halothane. However, when their data were stratified by age, this difference in incidence of emergence agitation occurred only in preschool-aged children undergoing circumcisions, not in the older school-aged children undergoing inguinal hernia repairs. Whether this difference in emergence agitation was a function of age, type of surgery, psychological patient preparation, or postoperative pain is unclear. Not all studies have demonstrated an increase in emergence agitation with sevoflurane. Wellborn et al. ^[9] compared the emergence and recovery characteristics of desflurane, halothane, and sevoflurane in children undergoing outpatient adenoidectomy with BMT who had also received intraoperative opioids. Although the addition of opioids attenuates emergence agitation, the incidence of emergence excitement was similar for both the halothane- (25%) and sevoflurane- (10%) anesthetized patients but was markedly greater in the desflurane-anesthetized children (55%).

In addition to the choice of anesthetics, other factors can influence the incidence of emergence agitation ^[3,6,11]. Pain, patient temperament, age, and developmental maturity are all factors that can affect emergence agitation. Pain, especially in preverbal children, can be difficult to quantify and may mimic the signs of emergence agitation from anesthesia.

Although pain scales have been devised for preverbal children, we are not convinced that, in preschoolage children emerging from anesthesia, the signs of pain are distinguishable from emergence agitation. We did not separate pain scores from agitation; thus, our emergence score measurements may have assessed pain. Lack of a preemptive analgesic resulted in a higher incidence of emergence-type agitation behavior compared with controls. When similar groups are compared (i.e., halothane and ketorolac with sevoflurane and ketorolac or halothane placebo with sevoflurane placebo), the incidence of agitation is similar. This further supports that pain in children may be a major factor that distorts the emergence due to anesthetics. Frequently, it is the retrospective evaluation of the patient's response to analgesics that helps to separate agitation from pain. Lerman et al. ^[12], has noted that the intraoperative administration of opioids or the placement of regional blocks profoundly reduced the incidence of emergence agitation in patients anesthetized with both sevoflurane and halothane.

We used a placebo for the control group. Watcha et al. ^[13] showed that 70% of BMT patients not receiving intraoperative analgesics required postoperative analgesics in the immediate postoperative period. Our study noted a 46%-48% incidence of the placebo group patients requiring rescue analgesics in the recovery room. This incidence was significantly different from the ketorolac-treated patients (14%-22%). Our institution's postoperative pain management for children undergoing BMT has been variable; some children receive 10-30 mg/kg rectal acetaminophen intraoperatively, whereas others are treated in the recovery room with either oral or rectal acetaminophen if symptoms of pain or agitation occur. The reported effectiveness of various pain management strategies for children undergoing BMT has been quite variable. Bennie et al. ^[14] and others have shown that preoperative administration of oral acetaminophen (15 mg/kg) and oral ibuprofen (10 mg/kg) were no more efficacious for pain relief than placebo. Tobias et al. ^[16] noted that the combination of preoperative oral acetaminophen (10 mg/kg) and codeine (1 mg/kg) was better than acetaminophen (15 mg/kg) alone. In a study comparing oral ketorolac (1 mg/kg) with both acetaminophen (10 mg/kg) and placebo, Watcha et al. ^[13] noted the analgesic superiority of ketorolac and the lack of difference between placebo and oral acetaminophen.

Although oral administration provides a readily acceptable route of drug administration for children, optimizing the timing of a preoperative oral analgesic's therapeutic effects to coincide with postoperative pain relief can be difficult. In adults, peak plasma concentrations after oral ketorolac administration occurs in 30-50 min ^[16,17]. The pharmacokinetic data after oral ketorolac administration are not available in children. For ambulatory pediatric patients undergoing BMT procedures in which both the perioperative and intraoperative time may be extremely brief, both oral and rectally administered analgesics may not be beneficial. For this reason, we chose to administer the ketorolac IV. In addition, because peak plasma levels occur instantaneously, we thought that the analgesic effect should be present by the time the children were recovering from anesthesia, thereby mitigating the effects of pain on the quality of emergence.

Our choice of patients 1-5 yr of age undergoing BMT to evaluate emergence agitation addresses some of the design flaws of previous studies that reported the emergence characteristics of the newer inhaled anesthetics in children. Previous studies have included pediatric patients with large age spans and significantly different levels of developmental maturation, children who were undergoing a wide variety of surgical procedures, and children who had received intraoperative opioids and/or adjunct regional anesthesia. Because preverbal and preschool-aged children seem more likely to have emergence agitation than older children and adolescents, we chose to limit the study to children who were in an age-group most at risk of emergence agitation, whose procedure was well defined, and in whom a placebo control group might mimic our standard of care.

In summary, our study demonstrated that, in pediatric patients undergoing ultrashort surgical procedures, both halothane and sevoflurane produced a similar incidence of emergence agitation. In addition, for both anesthetics, the incidence of emergence agitation was markedly and similarly attenuated by the intraoperative administration of ketololac. Although BMT procedures are generally considered to be associated with minimal pain, it is clear that nonsteroidal drugs administered in a pharmacologically rational way can attenuate emergence agitation and/or pain and make children more comfortable in the immediate postoperative period.

The authors thank Lisa Goetz, PhD, for her editorial assistance; Susan Danfelt for her secretarial support; David Thomson, PhD, for his statistical assistance; and Paul Hoffman, BS, RPh, for his assistance in preparing the study syringes.