Abstract
Objective
The effect of midazolam premedication on forestalling postoperative agitation in children is not yet concluded. The purpose of this study was to compare the effects of midazolam premedication and parental presence during anesthetic induction on the incidence of postoperative agitation in pediatric patients.
Methods
One hundred sixty-seven children between 2 years and 7 years of age, undergoing anesthesia for outpatient surgery, were enrolled and randomly divided into four groups: sevoflurane anesthesia with parental presence without premedication, sevoflurane anesthesia with oral midazolam premedication, halothane anesthesia with parental presence without premeditation, and halothane anesthesia with oral midazolam premedication. The children randomized to the premedication groups took oral midazolam 0.5 mg/kg 20–30 minutes before anesthetic induction. For patients in the groups without premedication, one of the parents was present throughout the induction of anesthesia. One recovery room nurse blinded to the group assignment observed the patients and recorded the agitation scores all through their stay in the postanesthesia care unit.
Results
Postoperative agitation was significantly less in patients who received halothane anesthesia with oral midazolam premedication (p < 0.002).
Conclusion
Based on our data, the presence of a parent at induction of sevoflurance anesthesia was as effective as midazolam premedication in decreasing the incidence of postoperative agitation. Midazolam premedication, however, decreased postoperative agitation when halothane was used as the anesthetic agent.
Keywords
inhalation: sevoflurane; parents; postoperative complications: agitation;
1. Introduction
Sevoflurane is often used for anesthetic induction in pediatric patients because it is faster in the induction of anesthesia1, 2 and causes less cardiovascular depression2, 3 and fewer dysrhythmias4 than halothane. Emergence agitation, the excited and disoriented behavior on awakening from general anesthesia, is a common problem in preschool children with a reported incidence of up to 80%.2, 3 This unwanted reaction not only makes communication with the child impossible but also places the child at risk for injury or subjects the child to avoidable treatment such as reinstallment of the loss of venous access.
Many factors may contribute to the occurrence of emergence agitation in children. Some investigators have found that emergence agitation is more likely to occur in preschool (age, 3–6 years) children compared with school-aged (age, 6–10 years) children.2, 5 Up to now, the exact etiology of emergence agitation remains unknown nor is there a clear strategy for its prevention.6 Moreover, there are conflicting data on the effect of midazolam premedication on postoperative agitation. Some studies suggest that it could offer beneficial effect on decreasing the incidence of postoperative agitation,3, 7 whereas others reported other way around that it would not or could even increase the incidence of adverse behavior, which would linger 1 week or more after surgery.8, 9, 10 Parental presence combined with midazolam premedication, on the other hand, was reported recently to be beneficial on emergence behavior of children undergoing general anesthesia.11
Therefore, the purpose of this study was to compare the effects of midazolam premedication and parental presence on the incidence of postoperative agitation after emergence from general anesthesia in the postanesthetic care unit with sevoflurane or halothane as the anesthetic. The incidence of postoperative agitation requiring the intervention of the recovery room personnel during the first hour after awakening would also be used for comparison.
2. Materials and methods
After approval by our institutional review board and obtaining parental written informed consent, 167 ASA Physical Status I or II children between 2 years and 7 years of age, undergoing general anesthesia for short (less than 0.5 hour) outpatient surgeries were enrolled for study. Eligibility criteria included elective subumbilical surgery, compatible with peripheral nerve block. Exclusion criteria included history of chronic illness or developmental delay, mental retardation, attention deficit/hyperactivity disorder, psychiatric illness, or paradoxical excitation with sedatives.
Patients were assigned by a computer-generated randomization program to one of the four anesthetic groups:
1. Sevoflurane anesthesia with parental presence without premedication (S);
2. Sevoflurane anesthesia with oral midazolam premedication (SM);
3. Halothane anesthesia with parental presence without premedication (H);
4. Halothane anesthesia with oral midazolam premedication (HM).
Children randomized to the premedication groups took oral midazolam 0.5 mg/kg mixed with 10 mL of 10% dextrose, 20–30 minutes before anesthetic induction. For patients in the other two groups, one of the parents was present and collaborated with the anesthetic team to console the child during induction of anesthesia with facemask and left the theater when the child had closed the eyes.
Anesthesia was induced with sevoflurane (or halothane) and 60% nitrous oxide in oxygen at a flow rate of 10 L/min. After anesthetic induction, an intravenous cannula was placed. Sevoflurane was started at 1% and gradually increased up to 70% at intervals of every three breaths. Halothane was started at 0.5% and increased to 4% with increments of 0.5% after every three breaths. Thereafter, patients received 2-mg/kg rectal diclofenac sodium and an appropriate peripheral field block by the surgeon with 0.25% bupivacaine depending on the surgical procedure. A standard level of general anesthesia was provided to maintain the patients’ heart rate and blood pressure within 80% of basal values.
Heart rate; noninvasive blood pressure; oxygen saturation (Cardiocap; Datex, Helsinki, Finland); and inspired and expired concentrations of the anesthetic agents (Vamus; Drager, Lubeck, Germany) were monitored continuously throughout the procedure. No opioid analgesics were administered during the operation. The pediatric circle system (system F) was used to convey the volatile anesthetic at a flow rate of 10 L/min in all cases. Spontaneous or assisted ventilation by means of facemask was maintained throughout the operation. At the end of skin suture, FiO2 was increased to 100%, and the anesthetic agent was discontinued. The total anesthetic duration was considered as the time from the start of induction (start of inhalational anesthetic) to the time of termination of administration of inhalational agents. For the purpose of assessing recovery, the time at which the inhalational anesthetics were turned off was considered as Time 0. While still asleep, the children were transferred to the recovery unit. The designated time points for recording were as follows: at the start of the anesthetic induction, at the discontinuation of inhalational agents, at first opening of eyes (asking the children every 30 seconds), when the Steward score12 of 6 was reached, and at departure from the recovery unit.
The emergence agitation scale was measured every 5 minutes after admission to the recovery room (1 = awake and calm, cooperative; 2 = crying, requires consoling; 3 = irritable/restless, screaming, inconsolable; 4 = combative, disoriented, thrashing). An agitation score of 3 or 4 was classified was being agitated.2 Pulse rate and oxygen saturation were recorded until the child was fully alert. Rescue medication for agitation was intravenous fentanyl 1 μg/kg if the agitation score was 3 or 4.
2.1. Statistical analysis
Assuming an incidence of postoperative agitation of 0.5, a two-sided Type I error of 0.05, a power of 0.85, and an effect size of 0.6, at least 41 patients in each group were required to find a significant difference in the incidence of postoperative agitation. The Chi-square test and two-way analysis of variance were used to test for demographic differences among groups. The one-way analysis of variance and Tukey post hoc test were used for comparison of agitation incidence between the four groups. The Student t test was used to perform a subgroup analysis comparing the time data between sevoflurane and halothane groups. Continuous data (age, anesthesia time, surgery time, and awakening time) are presented as mean and standard deviation. Proportions (sex, postoperative agitation) are presented as frequency. A p value <0.05 was considered statistically significant. All statistical comparisons were accomplished with SPSS software version 11.5 (SPSS Inc., Chicago, IL, USA).
3. Results
A total of 167 children were included in the present study, of whom 44 belonged to the group S. Forty, 41, and 42 children belonged to the groups SM, H, and HM, respectively. The four groups were comparable with respect to age, gender distribution, weight, duration of anesthesia, and surgery (Table 1). One patient each from the H and SM groups was excluded from the study because of ventricular arrhythmia and regurgitation of gastric content, respectively.
Table 2 lists the average anesthetic and recovery times with standard deviation by groups. Differences in the rates of emergence from anesthesia and the rate at which a Steward score of 6 is reached between the four groups were recorded (p < 0.0001). To examine the effects of premedication and anesthetics on emergence and recovery times, a two-subgroup analysis was performed for sevoflurance and halothane. Time to eye opening and time to a Steward score of 6 were not significantly different between premedicated and non-premedicated children receiving sevoflurane. However, the times to eye opening and a steward score of 6 were significantly shorter in the sevoflurane groups compared with those in the halothane groups (p < 0.14 and <0.0001, respectively). With halothane, the time to eye opening was significantly shorter in non-premedicated children (p < 0.003), but no significant difference was detected in the time to get a Steward score of 6 (p < 0.21). Hence, the cause for the significant difference observed in the discharge time between the four groups was significant prolongation of recovery in the HM groups compared with that in the sevoflurane groups.
Finally, to compare the degrees of postoperative agitation in the recovery room among the four groups, a four-by-two contingency table (Table 3) was created. The resulting p value was 0.017. After one-by-one cross-tabulation and subgroup analysis, there was a statistically significant difference between the four groups; postoperative agitation was significantly less when midazolam premedication was used before halothane anesthesia (p < 0.002).
4. Discussion
Our results showed that midazolam premedication as compared with parental presence could not reduce postoperative agitation with sevoflurane as anesthetic agent for induction and maintenance of anesthesia for short outpatient surgeries. However, if halothane was the anesthetic agent, postoperative agitation was significantly less prevalent with midazolam administered as premedication.
The influence of midazolam on emergence behavior seems to be somewhat controversial. In comparison with placebo, Lapin et al3 and Ko et al7 reported a reduction of agitation in premedicated children, whereas McGraw and Kendrick observed that no difference was found at all, irrespective of the use of midazolam premedication.8 We found a similarly high incidence of emergence agitation in the sevoflurane groups. It meant that both parental presence and midazolam premedication were incapable of reducing the incidence of postanesthetic agitation.
There are several possible reasons for this finding. Regarding other factors, disorientation resulting from anesthetic drugs may be a more important contributor to emergence behavior than separation from parents. Another possibility is that the midazolam premedicated and non-premedicated groups were too similar. Perhaps, a difference might be detected if there was a control group where neither parental presence nor premedication at induction was applied. However, such a study design is neither moral nor free from blemish in current practice.
One possible explanation for midazolam failure is that midazolam, as reported, causes nightmares and sleep disturbances in other settings.8 Midazolam also suppresses rapid eye movement sleep and disrupts sleep cycles.13 Furthermore, the amnesia of surroundings by midazolam premedication might increase postoperative anxiety.8 Therefore, on awakening, premedicated children could be disoriented and agitated, believing that surgery was not over. Naturally, this would more readily occur in children who often do not yet understand their surroundings, especially when not fully alert.14
We found a significantly lower incidence of emergence agitation when midazolam was used before anesthesia with halothane (47.6% vs. 63.4%). This deferent response of preventive effect of midazolam on emergence agitation after sevoflurane and halothane anesthesia might be explained by different mechanisms of agitation with these agents.15
We absolutely noted a high incidence of emergence agitation in midazolam-premedicated children who received halothane anesthesia (47.6%). This is consistent with the study of Cole et al.16 They found a ninefold higher risk of developing postoperative delirium in children who received midazolam as premedication in a group of 260 children undergoing anesthesia for outpatient lower abdominal surgery.
We also found no significant difference in the incidence of postoperative agitation after sevoflurane or halothane anesthesia in the absence of premedication (70.5% vs. 63.4%, respectively). This was not in accordance with the previous studies except for those by Wellborn et al and Davis et al.17, 18 As far as we know, these studies were conducted in children with midazolam premedication. This discrepancy, although difficult to explain, might be really attributable to the use of volatile anesthetics in non-premedicated children or when a low-threshold scoring scale was used for evaluation.
When comparing midazolam-premedicated groups, we found a significantly lower incidence of postoperative agitation in the halothane group (80 vs. 47.6%). This was in accordance with some previous studies19 and reflects the inability of midazolam to decrease postoperative agitation after sevoflurane but not halothane anesthesia.
Recognition of and defining of postoperative agitation among previous studies varied greatly. Many investigators used crying for 3–5 minutes on emergence from anesthesia as their threshold for agitation,20 whereas others used a higher threshold involving thrashing behavior requiring restraint.3 The higher incidence of agitation in our study might reflect a relatively low threshold we used for defining agitation (agitation scores of 3 and 4).
Although operation time and surgery time were not different among the groups, recovery time was significantly shorter with sevoflurane anesthesia. Midazolam premedication did not delay discharge from recovery room in sevoflurane groups but caused delayed eye opening in the recovery room and delay of the time of discharge from recovery room in the halothane groups.
Controversy still exists as to whether or not late recovery or delayed discharge is actually less with sevoflurane compared with that with halothane. Several studies showed no significant difference between them,1, 19, 21 whereas others determined a shorter time to meet the criteria with sevoflurance.3, 22, 23
Based on our findings, one may recommend that the presence of a parent as the child’s consort at the time of induction of anesthesia is a better clinical practice when sevoflurane is the chosen anesthetic. Further studies are needed to find more effective ways for prevention of emergence agitation in outpatient pediatric anesthesia.
Acknowledgments
The authors would like to appreciate The Office for Clinical Research Development of Nemazee Hospital for advice on data analysis and editorial assistance. They also thankfully acknowledge the kind help of N. Khosrownejad and S. Shokrizadeh for data collection. Supported by the local distributor of Abbott Ltd., Company, Tehran, Iran.