Foreign body aspiration is a life-threatening condition, with children under 3 years of age most at risk.1 Early recognition and prompt treatment for foreign body aspiration are important to minimize the potentially serious and sometimes fatal consequences.2 In China, rigid bronchoscopy under general anaesthesia by the otorhinolaryngologist has remained the procedure of first choice for tracheal or bronchial foreign body removal. For the anaesthesiologist, anaesthetic management can be challenging for sharing the airway with the otorhinolaryngologist, general anaesthesia without tracheal intubation and maintenance of the depth of anaesthesia with spontaneous ventilation.3 Apart from this, duration of emergence from anaesthesia had been identified as one of the risk factors associated with intraoperative or postoperative hypoxemia in rigid bronchoscopy.4 According to these points, both sevoflurane and propofol are appropriate for this procedure. The nonirritant nature of sevoflurane for the respiratory tract makes it the agent of choice for volatile induction/maintenance anaesthesia (VIMA), which is used frequently in paediatric surgery including rigid bronchoscopy.5 Combined with remifentanil, a super short-acting opioid, propofol-based total intravenous anaesthesia (TIVA) is attractive for paediatric procedures based on their properties and synergistic effect.6 Previous studies have demonstrated that sevoflurane VIMA technique offers better haemodynamic control and a better cardiovascular profile than propofol in elective coronary artery bypass surgery,7 and bolus injections of propofol were frequently associated with apneas.8 On the basis of these findings, we hypothesized that sevoflurane VIMA would provide a better haemodynamic profile and more stable respiration for rigid bronchoscopy for foreign body removal in children than propofol–remifentanil TIVA. In this study, we compared sevoflurane VIMA and propofol–remifentanil TIVA, when used for both induction and maintenance of anaesthesia in children undergoing rigid bronchoscopy for tracheal or bronchial foreign body removal under Bispectral Index (BIS) monitor.
Ethical approval for this study (no. 2007-9) was provided by the Biological-Medical Ethical Committee of West China Hospital, Sichuan University, Chengdu, Sichuan, China (President Professor Zhi Zeng) on 21 April 2007, and informed consent was obtained from each patient's legal guardian (most of them were parents) at the preoperative visit.
Sixty-four children, with American Society of Anesthesiologists physical status I or II, age of 1–4 years, undergoing rigid bronchoscopy for removal of tracheal or bronchial foreign body over a period of 28 months (May 2007 to September 2009), were enrolled in this trial, and they were allocated randomly to receive sevoflurane VIMA (Group VIMA; n = 32) or propofol–remifentanil TIVA (Group TIVA, n = 32). None of the patients had signs of hepatic, renal, cardiac or endocrinal impairment. Exclusion criteria were known allergies to any anaesthetic agent, family history of malignant hyperthermia, coagulopathy and the legal guardian's refusal to sign consent. All patients were not premedicated.
Before induction of anaesthesia, ECG, heart rate (HR), noninvasive blood pressure (BP) and pulse oximetry were continuously monitored for each patient. A BIS sensor was attached to each patient in conjunction with the BIS monitor (Aspect Medical Systems, Newton, Illinois, USA). In Group VIMA, anaesthesia was induced with inhalation of sevoflurane (Sevofrane; Abbott Laboratory, North Chicago, Illinois, USA). Before induction, a closed circuit with a 1-l reservoir bag was overflowed by 8 vol% sevoflurane with 0.3 l min−1 fresh oxygen flow for 3 min. During induction, the patient inspired 8 vol% sevoflurane with a fresh gas flow of 5 l min−1 pure oxygen via a sealed face mask till BIS decreased to 40, and laryngoscopy was performed. Anaesthesia was maintained using 2.5–3.5% sevoflurane in fresh gas flow via the side port to keep BIS values between 40 and 60. In Group TIVA, anaesthesia was induced with a target-controlled infusion (TCI) of propofol (Graseby 3500 infusion pump; SIMS Graseby, Watford, UK. Diprifusor Software; Aspect Medical, Inc., Newton, Massachusetts, USA), and the initial plasma concentration target was 3 μg ml−1, combined with continuous infusion of remifentanil (Humanwell Pharmaceutical Co. Ltd., Yichang, China) with a concentration of 0.05–0.10 μg kg−1 min−1, until BIS decreased to 40, and laryngoscopy was performed. Anaesthesia was maintained by continuous infusion of remifentanil (0.05–0.10 μg kg−1 min−1) and the TCI of propofol (plasma target concentration 2–3 μg ml−1) to keep BIS values between 40 and 60.9 Lignocaine (4%) was sprayed on the root of the tongue, epiglottis, larynx, between the vocal cords, into the upper trachea (maximum dose limited to 5 mg kg−1), then the rigid bronchoscope with a side port for fresh gas flow was introduced to the trachea. Time taken to achieve unconsciousness (disappearance of eyelash reflex) and for complete induction when the BIS value decreased to 40 were recorded.
All drugs were discontinued after completion of the procedure, and dexamethasone 0.1 mg kg−1 was administered. Children were allowed to breathe 100% oxygen spontaneously with the aid of face mask. The time from discontinuation of anaesthetics until the patient opened his eyes spontaneously or on verbal command (emergence time) were recorded. Patients were regarded as ready to be discharged from the operating theatre when the BIS value was above 85 and the patient could open his eyes or respond purposefully to nonpainful stimuli. All patients were kept for observation in the paediatric ward for at least 24 h, and the incidence of nausea and vomiting were recorded.
Before induction, HR, mean BP (MBP), respiratory rate, SpO2 and BIS value were recorded as the baseline value (T0). HR, MBP and respiratory rate were recorded at the time points as follows: laryngoscopy (Tlary) when BIS value decreased to 40; intubation of the rigid bronchscopy (Tbron); 5, 10 and 20 min during procedure (T5min, T10min and T20min); extubation of the rigid bronchscope at the end of procedure (Tend) and discharge (Tdis).
The quality of rigid bronchoscopic intubation was assessed according to the score of the intubating conditions (Table 1)10 and recorded. During the procedure, if anaesthetic depth was not deep enough (e.g. patient's movement or BIS value >60 with increase of HR or BP), procedure was paused and the patient was treated with sevoflurane 1% increase (in Group VIMA) or 1 μg ml−1 increase of plasma target concentration of propofol (in Group TIVA) until the BIS value returned to <60 with the returning of HR and BP, and the procedure was then restarted. If the anaesthetic depth was too great (e.g. BIS value <40 with decrease of HR or BP), the patient was treated with sevoflurane 1% decrease (in Group VIMA) or 1 μg ml−1 decrease of plasma target concentration of propofol (in Group TIVA). Adverse events during procedure, including breath holding, cough, laryngospasm and desaturation (SpO2 <93%), and excitement after procedure were recorded. In case of breath holding lasting more than 1 min or if SpO2 decreased to below 90%, ventilation was manually assisted. After completion of the procedure, if SpO2 could not be maintained more than 93% by breathing 100% oxygen spontaneously with the aid of a facemask, 1 mg kg−1 suxamethonium was administered, tracheal intubation was performed, and the patient was transferred to the ICU with controlled ventilation.
Data were expressed as mean ± SD. Statistical analysis was performed by using SPSS 13.0 software (SPSS Inc., Chicago, Illinois, USA). Age, weight, HR, MBP, respiratory rate, SpO2 and BIS value at baseline were compared by using unpaired, two-tailed Student's t-test, and sex distribution was analysed by using the χ2 test. Respiratory rate, HR and MBP during observing period were compared by using a repeated-measures general linear model with a Huynh–Feldt correction for sphericity. Duration of anaesthesia, time for loss of consciousness, time for the BIS value to decrease to 40, emergence time and bronchoscopic intubation condition scores were compared by using independent-samples t-test, and the incidence rates of adverse events were analysed by using the χ2 test. A P value of less than 0.05 was considered significant.
Complete recordings were obtained in all 64 children. The two study Groups did not differ significantly in demographic variables (Table 2), and all the foreign bodies were organic (including peanuts, melon seeds, sunflower seeds, pine nuts, walnuts and corn kernels).
Time for loss of consciousness, time of BIS value decreased to 40 and emergence time in Group VIMA were significantly shorter than those in Group TIVA (P < 0.05). Duration of anaesthesia, duration of surgery and mean intubating condition scores between the two groups were comparable (P > 0.05). The incidence rates of breath holding and desaturation in Group VIMA were significantly lower than those in Group TIVA (P < 0.05), and excitement after procedure in Group VIMA were significantly higher than that in Group TIVA (P < 0.05) (Table 3). In none of the patients was breath holding lasting more than 1 min observed or SpO2 decrease to below 90% recorded, both of which require manually assisted ventilation.
Changes in respiratory rate, HR and MAP at different time points during the study period are shown in Fig. 1. Respiratory rate was decreased slightly but had no significant difference compared with T0 (P > 0.05) in Group VIMA from Tlary to Tdis during the study period, and was decreased significantly after anaesthetic induction and returned to the baseline level at Tdis in Group TIVA. Respiratory rate was significantly higher in Group VIMA than that in Group TIVA from Tlary to Tend. HR was significantly decreased in the two groups after anaesthetic induction, and returned to baseline levels when discharged (Tdis). HR was significantly higher in Group VIMA than that in Group TIVA from Tlary to Tend. MAP was decreased after intubation of the rigid bronchscopy (Tbron) in Group VIMA, and decreased after anaesthetic induction in Group TIVA. MAP was significantly higher in Group VIMA than that in Group TIVA after anaesthetic induction, and returned to the baseline levels when discharged in both groups.
After the completion of the procedure, all the patient could maintain SpO2 more than 95% by breathing 40% oxygen spontaneously with the aid of a facemask, and none of the patients needed endotracheal intubation for controlled ventilation. None of the patients experienced nausea or vomiting during the 24-h observing period in the paediatric ward.
Our results demonstrate the comparison of sevoflurane VIMA and propofol–remifentanil TIVA for rigid bronchoscopy for tracheobronchial foreign bodies in paediatric patients with spontaneous ventilation. We collected and analysed data during the procedure from induction until discharged from the operating theatre, and we verified our hypothesis that sevoflurane VIMA provides better haemodynamic characteristics and more stable respiration than propofol–remifentanil TIVA for this procedure in children.
In our study, time for loss of consciousness, time of BIS value decrease to 40 and emergence time in Group VIMA were significantly shorter than those in Group TIVA (P < 0.05), suggesting that sevoflurane VIMA provide a more rapid induction and recovery than propofol–remifentanil TIVA. In previous studies, comparisons of induction and recovery times between sevoflurane and propofol in paediatric patients are limited and inconclusive. Lopéz Gil et al.11 reported that emergence was more rapid with sevoflurane than propofol in children undergoing minor surgery below the umbilicus, which is consistent with our results. Joo and Perks12 demonstrated no statistical difference in the time of induction or the complications occurring in the induction between sevoflurane and propofol, with similar efficacy for anaesthetic induction by performing a meta-analysis in adult patients. The discrepancy in this study and our data may be due to the combined use of other agents such as nitrous oxide in the studies they included, and the results were anaesthetic regimen of balanced anaesthesia. From our data, 8% sevoflurane VIMA inhalation provides a more rapid induction than 3 μg ml−1 propofol TCI combined with continuous infusion of remifentanil in paediatric patients.
Our data demonstrate that respiratory rate was decreased slightly and had no significant difference compared with baseline levels in Group VIMA. In this study, it was impossible to monitor tidal volume or end-tidal concentration of sevoflurane or carbon dioxide via the side port of rigid bronchoscope, so we could not collect these data, and this was a limitation of our study. It was reported that sevoflurane administered below 1.3 minimum alveolar concentration (MAC) maintained spontaneous respiration at acceptable levels during surgical stimulation.13 Another study14 showed that respiratory frequency increased to compensate the decreased minute volume when the sevoflurane level was more than 1.4 MAC. In our study, the end-tidal sevoflurane was not monitored, and the respiratory rate was not increased, but the inhalational concentration of 2.5–3.5% sevoflurane after induction was observed to maintain spontaneous respiration with preserved respiratory rate. In Group TIVA, respiratory rate was decreased significantly after anaesthetic induction and returned to the baseline level at discharge, and respiratory rate was significantly higher in patients anaesthetized with sevoflurane VIMA as compared with patients receiving propofol–remifentanil TIVA. Successful use of propofol–remifentanil TIVA in paediatric patients undergoing gastrointestinal endoscopy or flexible bronchoscopy has been reported in previous studies.9,15 Although decreased respiratory rate was observed, hypoxemic events were not found in these trials. In our study, unlike these studies, breath holding was observed in 10 patients and desaturation was seen in 12 patients in Group TIVA, and this discrepancy could be due to the stimuli of tracheobronchial procedure. In Group VIMA, breath holding occurred in only two patients and desaturation was seen in five patients. These adverse events occurred less often in Group VIMA than in Group TIVA. Preserved respiratory frequency and a lower incidence of breath holding and desaturation in patients receiving sevolurane VIMA suggested that this technique provides a more stable respiratory level and relieves the intratracheal procedure stimuli better than propofol–remifentanil TIVA. Similar intubating conditions between the two groups suggested that both anaesthetic techniques provide satisfactory bronchoscopic intubating conditions for foreign body removal in paediatric patients.
Differences between VIMA and TIVA were observed in HR and BP. Sevoflurane VIMA was associated with higher HR than propofol–remifentanil TIVA. Sevoflurane VIMA has been proven to provide a satisfactory depth of anaesthesia for paediatric procedures,16 and our purpose was to compare the two anaesthetic techniques, VIMA and TIVA, so extra analgesic was not given to patients in Group VIMA, whereas remifentanil was given to patients in Group TIVA. Lack of opioids in Group VIMA may be attributed to the higher HR, and this was a limitation of our study. After anaesthetic induction, MAP was significantly higher in Group VIMA than that in Group TIVA. During procedure, depth of anaesthesia was monitored by BIS and maintained at the same levels between 40 and 50. Nishiyama et al.17 reported rapid induction of anaesthesia with sevoflurane 7% and tidal volume breathing induced minor haemodynamic changes and has no inhibitory effect on sympathetic activity, and propofol–remifentanil TIVA had stronger inhibition of the neuroendocrine reaction to stress than sevoflurane.18 This could be the explanation of the higher MAPs in patients receiving sevoflurane VIMA.
Higher incidence of excitement in Group VIMA than Group TIVA could be attributed to the more rapid emergence from anaesthesia with sevoflurane,19 and this finding is similar to previous studies.20 In this study, none of the patients were premedicated, and the isolated effects of the two anaesthetic techniques, VIMA and TIVA, were evaluated with little confounding factors that may influence anaesthetic profiles. For operating team, including anaesthesiologists, surgeons and nurses, gas contamination is unavoidable due to the open ventilation system during VIMA. This issue should be considered when choosing the anaesthetic technique for paediatric airway procedures.
In summary, compared with propofol–remifentanil TIVA, sevoflurane VIMA provides more stable haemodynamics and respiration, faster induction and recovery and higher incidence of excitement in paediatric patients undergoing tracheal/bronchial foreign body removal.
There are no conflicts of interest.
1 Baharloo F, Veyckemans F, Francis C, et al
. Tracheobronchial foreign bodies: presentation and management in children and adults. Chest 1999; 115:1357–1362.
2 Kosloske AM. Bronchoscopic extraction of aspirated foreign bodies in children. Am J Dis Child 1982; 136:924–927.
3 Motoyama EK. Anaesthesia for ear, nose, and throat surgery. In: Motoyama EK, Davis PJ, editors. Smith's anaesthesia for infants and children. St. Louis, Missouri, USA: Mosby; 1996. pp. 653–676.
4 Chen LH, Zhang X, Li SQ, et al
. The risk factors for hypoxemia in children younger than 5 years old undergoing rigid bronchoscopy
for foreign body removal. Anesth Analg 2009; 109:1079–1084.
5 Jöhr M, Berger TM. Paediatric anaesthesia and inhalation agents. Best Pract Res Clin Anaesthesiol
6 Drover DR, Litalien C, Wellis V, et al
. Determination of the pharmacodynamic interaction of propofol
and remifentanil during esophagogastroduodenoscopy in children. Anesthesiology 2004; 100:1382–1386.
7 Bharti N, Chari P, Thingnam SKS, et al
. Comparison of haemodynamic and cardiovascular effects of VIMA with sevoflurane
versus TIVA with propofol
in patients undergoing coronary artery bypass surgery. Indian J Anaesth 2008; 52:805–812.
8 Stokes DN, Hutton P. Rate-dependent induction phenomena with propofol
: implications for the relative potency of intravenous anesthetics. Anesth Analg 1991; 72:578–583.
9 Abu-Shahwan I, Mack D. Propofol
and remifentanil for deep sedation in children undergoing gastrointestinal endoscopy. Paediatr Anaesth 2007; 17:460–463.
10 Steyn MP, Quinn AM, Gillespie JA, et al
. Tracheal intubation without neuromuscular block in children. Br J Anaesth 1994; 72:403–406.
11 Lopéz Gil ML, Brimacombe J, Clar B. Sevoflurane
for induction and maintenance of anaesthesia with the laryngeal mask airway in children. Paediatr Anaesth 1999; 9:485–490.
12 Joo HS, Perks WJ. Sevoflurane
for anesthetic induction: a meta-analysis. Anesth Analg
13 Doi M, Takahashi T, Ikeda K. Respiratory effects of sevoflurane
used in combination with nitrous oxide and surgical stimulation. J Clin Anesth 1994; 6:1–4.
14 Doi M, Ikeda K. Respiratory effects of sevoflurane
. Anesth Analg 1987; 66:241–244.
15 Berkenbosch JW, Graff GR, Stark JM, et al
. Use of a remifentanil–propofol
mixture for pediatric flexible fiberoptic bronchoscopy sedation. Paediatr Anaesth 2004; 14:941–946.
16 Keleş GT, Toprak V, Ok G, et al
. Volatile induction and maintenance of anaesthesia using laryngeal mask airway in pediatric patients. J Anesth 2004; 18:51–54.
17 Nishiyama T, Aibiki M, Hanaoka K. Haemodynamic and catecholamine changes during rapid sevoflurane
induction with tidal volume breathing. Can J Anaesth 1997; 44:1066–1070.
18 Ledowski T, Bein B, Hanss R, et al
. Neuroendocrine stress response and heart rate variability: a comparison of total intravenous versus balanced anaesthesia. Anesth Analg 2005; 101:1700–1705.
19 Walker SM, Haugen RD, Richards A. A comparison of sevoflurane
with halothane for pediatric day care surgery. Anaesth Intensive Care 1997; 25:643–649.
20 Uezono S, Goto T, Terui K, et al
. Emergence agitation after sevoflurane
in pediatric patients. Anesth Analg