Cough is a desired physiological response to protect the airway from aspiration. However, severe cough causes discomfort and can induce bronchospasm in patients with hyperreactive airways such as those undergoing fiberoptic bronchoscopy (FOB) [1,2]. Bronchoscopy may precipitate cough, which may last for several hours after the procedure. [3,4]. These effects may lead to agitation, which is a common postanaesthetic problem in children.
Fiberoptic bronchoscopy, used as a diagnostic tool in patients with pulmonary disease, is usually performed under a deep and relatively short duration of anaesthesia in order to minimize irritation of the airway by the scope. Postoperatively, rapid emergence is desirable to return full airway control to the child as soon as possible. Remifentanil, a short-acting opioid, confers analgesia and suppresses airway reflexes when used for bronchoscopy . Sevoflurane, lacking airway irritation but having a pleasant smell and airway dilating effect, is the preferred inhalation anaesthestic agent for induction and maintaince of anaesthesia in children undergoing FOB. However, sevoflurane is associated with agitation during the recovery phase in up to 24% of children undergoing fast emergence [6,7].
In this study, we determined the effects of remifentanil on cough and agitation during emergence and recovery when added to sevoflurane anaesthesia in children undergoing FOB.
Our protocol received approval from our institutional ethics committee. Parents of 53 consecutive children aged 2–6 years of age scheduled to undergo elective diagnostic FOB were approached to participate in the study. Patients scheduled for FOB who had hemoptysis (n = 2) or who had previously undergone FOB (n = 1) were excluded from participation. Parents of the remaining 50 consecutive children gave consent for inclusion of his/her child in the study. None of these had acute exacerbations of cough at the time of study enrollment or just prior to the FOB procedure.
Thirty minutes before the induction of anaesthesia, children were premedicated with oral midazolam (0.5 mg/kg) and nebulized lidocaine 4% (4 mg/kg) via facemask (standard procedures at our institution). At the time of separation from parents when going to the operating room, a separation score [1 = excellent (separates easily); 2 = good (not clinging, whimpers, calms with reassurance); 3 = fair (not clinging, cries, will not calm or quiet); and 4 = poor (crying, clinging to parent)] was recorded. A separation score of one or two was considered satisfactory, whereas a score of three or four was considered unsatisfactory. Tolerance of the anaesthetic mask during induction was graded on a four point scale: 1 = excellent (unafraid, cooperative, accepts mask readily); 2 = good (slight fear of mask, easily calmed); 3 = fair (moderate fear, not calmed with reassurance); 4 = poor (terrified, crying, agitated). Scores of one or two were considered satisfactory, whereas scores of three or four were considered unsatisfactory.
Before induction of anaesthesia, children were randomized by sealed envelope into the sevoflurane group (group S) or sevoflurane and remifentanil group (group SR). Routine patient monitoring included oxygen saturation (SpO2), heart rate (HR), noninvasive blood pressure (NIBP), continuously measured end-tidal carbon dioxide (EtCO2), and continuously measured end-tidal sevoflurane (EtSevo). Additionally, an A2000 BIS (bispectral index) monitor (Aspect Medical Systems, Inc., Norwood, USA) was used. EtCO2 and EtSevo levels were measured through a capnograph sensor placed between the L piece and Bain circuit. EtSevo concentration was recorded at 2-minute intervals. Adverse events were defined as arrhythmias, oxygen desaturation (SpO2 <90% for ≥30 s), and CO2 retention (EtCO2 >50 mmHg).
Anaesthesia was induced via mask using 6–8% sevoflurane in 100% oxygen in all children. An intravenous catheter was placed and IV atropine 10 μg/kg was administered. Ventilation was controlled manually to keep the EtCO2 between 35 and 45 mmHg in a Bain breathing circuit. Maintenance of anaesthesia was continued with sevoflurane in 100% oxygen, the sevoflurane concentration being adjusted to maintain a BIS value of 40 (in a pilot study, children coughed and moved when we titrated sevoflurane delivery to a BIS of 50, therefore we used a BIS of 40). In addition to the above medications, patients in group SR received remifentanil infusion, started immediately after the intravenous catheter was placed, in a nonblinded fashion at a bolus dose of 1 μg/kg over 2 min followed by an constant infusion of 0.15 μg/kg/min for maintenance. Total remifentanil consumption was recorded.
When an adequate depth of anaesthesia was achieved, the fiberoptic bronchoscope (passed through an elbow adapter into the mask) was inserted orally while ventilating via a facemask. FOB was performed by same paediatric pulmonologist with a 3.6 mm Olympus BF3C30 bronchoscope (Olympus Corporation, Orangeburg, USA) with a video-adapter. Upon direct visualization of the vocal cords and after passing through the cords, 1% lidocaine (maximum 7 mg/kg) was injected as needed to ensure patient comfort via the bronchoscope into trachea and major bronchi. Bronchoalveolar lavage was performed using aliquots of nonbacteriostatic 0.9% NaCl (maximum 5 mg/kg).
When the bronchoscope was removed at the end of the procedure, the remifentanil infusion and sevoflurane inhalation were terminated. The patient was allowed to breathe 100% oxygen spontaneously via facemask. The emergence period was calculated as the time from discontinuation of sevoflurane and remifentanil until the patient moved, opened the eyes spontaneously, or coughed. After emergence was complete, the patient was transferred to the postanaesthesia care unit (PACU), given 6–10 l/min humidified oxygen via facemask, and monitored (SpO2 and HR). Children were observed for coughing, laryngospsm, bronchospasm, or desaturation (SpO2 <90% for ≥30 s) during emergence and recovery.
Patient recovery was assessed by Aldrete scoring (0–10 points). When the patient had fully emerged from anaesthesia (time 0), the first coughing and agitation scores were recorded. Scores were measured every 5 min thereafter until recovery was complete (when an Aldrete score of nine was reached). Cough was graded as 0 (no coughing), 1 (minimal coughing: once or twice), 2 (moderate coughing: 3–4 times), or 3 (severe coughing: 5–6 times). Children with a cough score of two or three and an SpO2 less than 95% were classified as severe coughing and were given cold mist with oxygen via facemask. Children were observed in the PACU until the cough score was below two.
Agitation was measured as one (sleeping), two (awake, calm, and cooperative) three (crying, requires consoling), four (irritable/restless, screaming, inconsolable), or five (combative, disoriented, thrashing) . Children with an agitation score of four or five were classified as agitated and were given 0.07 mg/kg of midazolam IV. Persons scoring the children for agitation and cough were unaware of group allocation regarding medications during FOB. Anaesthesia time, procedure time (the time from introduction of the bronchoscope until its removal), lidocaine dose, emergence time, and recovery time were recorded.
SPSS (Statistical Package for Social Sciences) for Windows (version 15.0) was utilized for statistical analysis and Statistica for Windows (Version 6.0, StatSoft, Inc., Oklahoma City, USA) was used for power analysis. All descriptive data are reported as ‘mean ± SD' or ‘median ± interquartile range’. Normally distributed parametric variables were compared with Student's t-test. Differences in categorical variables between the groups were analysed with two-tailed Fisher's exact test. As an a-priori multiple-comparison procedure, Friedman ANOVA by ranks method was utilized for within group comparison of repeated measures of coughing and agitation scores; followed by Wilcoxon matched pairs test for pairwise assessments in a post hoc manner. Kruskal–Wallis ANOVA by ranks method was employed to analyse any scoring differences between the groups. Results were evaluated within 95% CI and a P value of less than 0.05 was considered statistically significant. When interpreting the results of multiple-comparison procedures however, the level of significance was readjusted downward to 0.01 by dividing the conventional α level (0.05) by four (number of repeated measurements for each group) for Wilcoxon matched pairs test and to 0.006 by dividing 0.05 by eight (4 repeated measurements × 2 groups) for Kruskal–Wallis ANOVA test. We anticipated that 25 patients in each group would give a power of 0.75 using two-way ANOVA testing to demonstrate a 20% difference between groups with an α significance level of 0.05.
Demographic characteristics and indications for FOB were similar in groups S (n = 25) and SR (n = 25; Table 1). Preoperative separation scores were similar in group S and group SR (1.6 and 1.6, respectively, P = 0.8). Tolerance of the anaesthetic mask was similar in both groups (1.6 and 1.7, respectively, P = 0.5). Mean EtSevo concentrations were 4.4 ± 0.8% in group S and 3.8 ± 0.4% in group SR (P = 0.001). The mean dose of remifentanil administered to patients in group SR was 22 ± 8 μg. The procedure, anaesthesia, emergence, and recovery times for each group are listed in Table 2; only the recovery time was significantly different, being shorter (7 ± 6 min) in group SR than in group S (13.0 ± 3.5 min; P < 0.0001).
Severe coughing was not observed in any child in group S or group SR during emergence, but during recovery, 24% of group S (n = 6) and 4% of group SR (n = 1) had severe cough (two-tailed Fisher's exact test, P = 0.1). Coughing scores were not significantly different between the two groups during emergence and recovery (Kruskal–Wallis ANOVA test, P > 0.05). A-priori within group comparisons showed statistically significant differences in coughing scores during both emergence and recovery in group S (Friedman ANOVA multivariate test, P < 0.0001). In contrast, the difference between coughing scores during the recovery period within group SR was not significantly different (Friedman ANOVA, multivariate test; P = 0.3; Table 3).
Severe agitation was not observed during emergence in either group. Frequency of agitation during recovery was 8% (n = 2) in group S and 16% (n = 4) in group SR (two-tailed Fisher's exact test, P = 0.4). Agitation scores at emergence (T = 0 min.), 10, and 15 min were not significantly different between the groups (Kruskal–Wallis ANOVA test, P = 0.09, P = 0.56 and P = 0.16, respectively). However, the mean agitation score at 5 min into recovery in group SR was significantly higher than that of group S (P = 0.003). Similar to cough, a-priori within group comparisons also found significant differences between agitation scores at emergence and during the recovery period in group S (Friedman ANOVA multivariate test, P < 0.0001). In group S, agitation scores at 10 and 15 min into recovery were significantly higher than those at emergence and 5 min into recovery (posthoc comparison with Wilcoxon matched pairs test, P < 0.002, P < 0.001 and P < 0.002 and P < 0.001, respectively; Table 4). In group SR, no difference was found between agitation scores during the recovery period (Friedman ANOVA multivariate test, P = 0.2; Table 4). No child exhibiting severe agitation had severe coughing.
Thoracic wall rigidity was observed in one patient in group SR during bolus infusion of remifentanil before FOB, resulting in difficult manual ventilation of the lungs. SpO2 fell to 54%. Within 3 min, rigidity resolved with 100% oxygen administered by a mask under pressure. Nondepolarizing neuromuscular blockers were not required and the procedure was completed as planned. No haemodynamic instability occurred in any child during the emergence or recovery periods.
We expected remifentanil to have favourable effects on cough because its analgesic and antitussive effects continue well after completion of the procedure. However, in our study, remifentanil did not decrease coughing during emergence and recovery from sevoflurane anaesthesia in children undergoing FOB for diagnostic bronchoalveolar lavage. Similarly, Shajar et al. found that 1 μg/kg of remifentanil given at the end of anaesthesia did not decrease coughing during emergence. In contrast, others have found remifentanil to decrease cough during emergence in adults when used in total intravenous anaesthesia .
We thought that our results might have been caused by the short duration of action of remifentanil. Egan et al. and Westmoreland et al. calculated the computer-derived context sensitive half time of infused remifentanil as 3.6 min and 3.0 min, respectively. This is the time required for a 50% reduction in the effect site concentration of remifentanil . In addition, Ross et al. found that remifentanil's terminal elimination was similar in children of all ages, a mean of 3.4–5.7 min .
Although coughing and agitation scores increased significantly with time during the recovery period in group S, no such increase was seen in group SR. Although differences between the groups were not statistically different at particular points in time during recovery, within group changes in scores were significant in group S but not in group SR. This finding is not easily explained, and should be studied again in future observational projects.
The mean agitation score in group S was higher than that of group SR 5 min into recovery. We attribute this to a quicker recovery in group SR patients due to the lower concentration of sevoflurane used in these patients [6,7]. This was also the time that many of these children began to recover, just as they were being transported from the OR to the ICU, or during their first minutes in the ICU. The quickly changing environment during this time may have exacerbated any tendency for agitation to occur.
Although the mean sevoflurane concentration in group SR patients was lower than that of group S patients, emergence times were similar in both groups. We had expected the emergence time in group SR to be shorter than that of group S, but we surmise that our results were due to the short FOB procedure time and to the sedative effects of midazolam continuing into the emergence period. The short context-sensitive half time of remifentanil and the lower concentration of sevoflurane in group SR patients resulted in short recovery times in group SR.
In our study, although the bolus dose of remifentanil was given over 2 min, thoracic wall rigidity occurred in one child during bolus injection. The incidence of rigidity has been reported to be 0.3–17% in various studies and is dependent on total dose and rate of administration [13,14]. Doses less than 2 μg/kg given over 1 min have not been reported to cause rigidity . Thoracic wall rigidity can potentially cause serious hypoxia.
Readiness for tracheal intubation during sevoflurane anaesthesia can be gauged by titrating sevoflurane delivery to a target Bispectral Index (BIS) of 35 ± 5 in children. Intubating conditions during sevoflurane anaesthesia in children were found to be improved by a single bolus dose of remifentanil 1.0 μg/kg . BIS has been found to correlate with , and not correlate with , the depth of anaesthesia with sevoflurane. In another study, the optimal bolus dose of remifentanil required for successful tracheal intubation was 0.56 ± 0.15 μg/kg in 50% of children during inhalation induction using 5% sevoflurane, in the absence of neuromuscular blocking drugs . We used similar doses in our study.
In conclusion, when added to sevoflurane anaesthesia for fiberoptic bronchoscopy in children, remifentanil significantly shortened the time to recovery. Remifentanil used with sevoflurane anaesthesia did not result in a lower incidence of post-FOB cough, but was associated with higher agitation scores during the recovery period.
We would like to thank Dr Guven Olgac, FEBTCS for his statistical analysis.
1 Irwin RS. Complications
: ACCP evidence-based clinical practice guidelines. Chest 2006; 129:54S–58S.
2 Hohlrieder M, Tiefenthaler W, Klaus H, et al
. Effect of total intravenous
anaesthesia and balanced anaesthesia on the frequency of coughing during emergence from the anaesthesia. Br J Anaesth 2007; 99:587–591.
3 Guler G, Akın A, Tosun Z, et al
. Single-dose dexmedetomidine reduces agitation
and provides smooth extubation after pediatric adenotonsillectomy. Pediatric Anesthesia 2005; 15:762–766.
4 Schwarz Y, Greif J, Lurie O, et al
. Dextromethorphan premedication
requirement: objective and subjective parameters in peribronchoscopy. Respiration 2007; 74:314–319.
5 Berkenbosch JW, Graff GR, Stark JM, et al
. Use of remifentanil
-propofol mixture for pediatric flexible fiberoptic bronchoscopy sedation. Paediatr Anaesth 2004; 14:941–946.
6 Weldon VC, Bell M, Craddack T. The effect of caudal analgesia on emergence agitation
after sevoflurane versus halothane anesthesia. Anesth Analg 2004; 98:321–326.
7 Voepel Lewis T, Malvia S, Tait AR. A prospective cohort study of emergence agitation
in the pediatric postanaesthesia care unit. Anesth Analg 2003; 96:1625–1630.
8 Cole JW, Murray DJ, McAlister JD, et al
. Emergence behaviour in children
: defining the incidence of excitement and agitation
following anaesthesia. Paediatr Anaesth 2002; 12:442–447.
9 Shajar MA, Thompson JP, Hall AP, et al
. Effect of a remifentanil
bolus dose on the cardiovascular response to emergence from anaesthesia and tracheal extubation. Br J Anaesth 1999; 83:654–656.
10 Egan TD, Minto CF, Hermann DJ, et al
versus of remifentanil
during bolus administration reported alfentanil: comparative pharmacokinetics and pharmacodynamics in healthy adult male volunteers. Anesthesiol 1996; 84:821–833.
11 Westmoreland CL, Hoke JF, Sebel PS, et al
. Pharmacokinetics of remifentanil
(GI87084B) and its major metabolite (GI90291) in patients undergoing elective inpatient surgery. Anesthesiol 1993; 79:893–903.
12 Ross AK, Davis PJ, deL Dear G, et al
. Pharmacokinetics of remifentanil
in anesthetized pediatric patients undergoing elective surgery or diagnostic procedures. Anesth Analg 2001; 93:1393–1401.
13 Joshi GP, Warner DS, Twersky RS, Fleisher LA. A comparison of the remifentanil
and fentanyl adverse effect profile in a multicenter phase IV study. J Clin Anesth 2002; 14:494–499.
14 Schüttler J, Albrecht S, Breivik H, et al
. A comparison of remifentanil
and alfentanil in patients undergoing major abdominal surgery. Anaesthesia 1997; 52:307–317.
15 Glass PS, Gan TJ, Howell S. A review of the pharmacokinetics and pharmacodynamics of remifentanil
. Anesth Analg 1999; 89:7–14.
16 Weber F, Füssel U, Gruber M, Hobbhahn J. The use of remifentanil
for intubation in paediatric patients during sevoflurane anaesthesia guided by Bispectral Index (BIS) monitoring. Anaesthesia 2003; 58:749–755.
17 Denman WT, Swanson EL, Rosow D, et al
. Pediatric evaluation of the bispectral index (BIS) monitor and correlation of the BIS with endtidal sevoflurane concentration in infants and children
. Anesth Analg 2000; 90:872–877.
18 Constant I, Leport Y, Richard P, et al
and changes of Bispectral Index and electroencephalographic-derived variables during sevoflurane induction in children
: clonidine premedication
compared with midazolam
. Br J Anaesth 2004; 92:504–511.
19 Min SK, Kwak YL, Park SY, et al
. The optimal dose of remifentanil
for intubation during sevoflurane induction without neuromuscular blockade in children
. Anaesthesia 2007; 62:446–450.