We performed this meta-analysis to compare the characteristics of sevoflurane and propofol for the induction of routine anesthesia and for laryngeal mask airway (LMA) insertion. The variables assessed were 1) time to loss of consciousness, 2) incidence of apnea during induction, 3) induction complications, 4) time for successful LMA insertion, 5) success with LMA insertion on first attempt, 6) patient dissatisfaction, and 7) postoperative nausea and vomiting. MEDLINE, Embase, and the Cochrane library databases between January 1992 and October 1999 were reviewed for randomized, controlled trials comparing anesthetic induction between sevoflurane/nitrous oxide and propofol. Data from the 12 randomized, controlled studies were used for the meta-analysis. Sevoflurane induction was associated with a trend toward higher patient dissatisfaction and higher first-time success with LMA. Apnea was less common in the sevoflurane group. The incidence of postoperative nausea and vomiting was significantly more frequent in the sevoflurane group (P < 0.05). This effect was still present when all other variables, except the induction methods, were controlled. The other pooled variables did not show a significant difference between sevoflurane and propofol. Sevoflurane and propofol had similar efficacy for anesthetic induction. However, for routine outpatient surgery, propofol may still be the preferred induction anesthetic because of its favorable induction of anesthesia characteristics, high patient satisfaction, and less frequent incidence of postoperative nausea and vomiting.
Implications: Sevoflurane and propofol had similar efficacy for anesthetic induction. However, for routine outpatient surgery, propofol may still be the preferred induction anesthetic because of its favorable induction of anesthesia characteristics, high patient satisfaction, and less frequent incidence of postoperative nausea and vomiting.
Departments of *Anaesthesia and †Pharmacy, St. Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada
March 17, 2000.
Address correspondence to Hwan Joo, MD, Department of Anaesthesia, St. Michael’s Hospital, 30 Bond St., Toronto, Ontario, Canada, M5B 1W8.
Sevoflurane has been advocated for the routine induction of anesthesia, especially in outpatients. Sevoflurane, when used with nitrous oxide for the induction of anesthesia, was rapid, with minimal complications (1). Its purported advantages were rapid onset, rapid offset, lack of major side effects, ability to induce and maintain anesthesia with one drug, better conditions for laryngeal mask airway (LMA) insertion, and ability to induce anesthesia without IV access. The purported disadvantages of a sevoflurane inhaled induction were possible increased pollution of the operating room with anesthetics, excitatory movements, and laryngospasm during anesthetic induction.
There have been many randomized, controlled clinical studies comparing sevoflurane and propofol for anesthetic induction. Individually, these studies have not shown a definite advantage of one technique over the other. The usual conclusions were that the techniques were comparable.
The objective of this meta-analysis was to compare the induction characteristics of sevoflurane with propofol, the current “gold standard” for anesthetic induction. We wanted to determine the best anesthetic for routine anesthetic induction. We compared characteristics that are desirable for an ideal drug to induce anesthesia, such as 1) speed of induction–time to loss of consciousness, 2) incidence of apnea during induction, 3) induction complications, 4) LMA insertion, 5) success with LMA insertion on first attempt, 6) patient satisfaction with anesthetic induction technique, and 7) postoperative nausea and vomiting (PONV). As a result of the lack of well controlled and accurately reported data, hemodynamic responses to the induction of anesthesia were not reviewed. Our primary hypothesis was that sevoflurane is associated with a higher frequency of PONV.
MEDLINE, Embase, and the Cochrane library databases between January 1992 and October 1999 were searched for randomized, controlled trials comparing anesthetic induction between sevoflurane/nitrous oxide and propofol. Studies performed with midazolam or opioid premedications were included. Studies written in all languages and involving adult subjects of both genders were included. Systematic searches were conducted under the following key words: randomized, controlled trial, sevoflurane, propofol, vital capacity induction, and inhalational induction, as well as their respective drug identification numbers. Hand searching of major anesthetic journals and their references from 1992 to October 1999 resulted in the rest of the included studies. Unpublished studies and abstracts were excluded.
To be considered for the review, the study must have had a minimum of 2 of a maximal score of 5 on the Quality Assessment Score according to Jadad (2). This meant that adequate randomization was an absolute requirement for selection. However, double-blinding was not a requirement, because adequate blinding was not felt to be possible in most studies. Each study was evaluated independently by both primary authors (HJ and WP). Disagreements were resolved by consensus. The Methods and Results sections were reviewed for data that would contribute to one of the seven characteristics being studied. As all indices for comparison may not have been available in each of the studies, data were to be selectively obtained for each category according to availability. For time to loss of consciousness (LOC), only data from studies that used primed 7–8% sevoflurane vital capacity induction technique with 50%–75% nitrous oxide were used. Time to LOC was defined as the interval from induction time to loss of lid lash reflex. Induction complications were defined as presence of oxygen desaturation, coughing, laryngospasm, patient movement, and any other event that required abortion of the induction technique or pharmacologic intervention. Success with LMA insertion was defined as the ability to insert the LMA for oxygenation and ventilation without the need for other rescue medications such as additional propofol or muscle relaxants. Patient satisfaction was defined as unfavorable if patients stated either that the induction was unpleasant or that they would not choose the method of induction for their next general anesthetic.
Data analysis was performed by using Revman 3.1 (The Cochrane Library, Oxford, UK) and Sigmastat 2.0 (SPSS Inc., Chicago, IL).
To verify homogeneity in the effects of the combined analysis, a formal test of homogeneity using the χ2 test was performed. A P value of >0.10 was defined as homogeneous If this procedure demonstrated homogeneity, a simple fixed-effects model was used. If homogeneity was not demonstrated, a random-effects model was used. Continuous outcomes were reported as weighted mean difference, and categorical outcomes were reported as odds ratio with numbers needed to harm. Effects were considered significant if P value was < 0.05.
Thirteen studies complied with the previously mentioned criteria for possible use in the meta-analysis. Data from the 12 included studies were used for the meta-analysis (Table 1). One study by Dashfield et al. (15) was excluded because of the unconventional method of propofol administration (slow infusion via syringe pump) and the method of reporting PONV (3-point PONV scale). Because of nonhomogeneity of five of the seven characteristics (all characteristics except LOC and induction complications), the studies were deemed heterogeneous. The random-effect model was used for all variables.
Five studies met the criteria for time to LOC (Figure 1A). Time to LOC was similar in both the sevoflurane and propofol groups with a weighted mean (95% confidence interval) difference of 2.85 s (−12.36, 18.05) favoring propofol.
Three studies met the criteria for apnea during anesthetic induction (Figure 1B). There was a more frequent rate of apnea in the propofol group with an odds ratio of 0.10 (0.04, 0.27).
Six studies met the criteria for inclusion in assessing the complication rate during induction (Figure 1C). Induction complications were similar in two groups with an odds ratio of 0.72 (0.44, 1.18) in favor of sevoflurane.
Only three studies met the criteria for time to successful LMA insertion (Figure 1D). Time to successful LMA insertion was slightly slower by 19.09 s (−21.09, 59.26) in the sevoflurane group This was not statistically significant.
Four studies met the criteria for success with LMA on the first attempt (Figure 1E). There was a trend toward higher success rates with sevoflurane with an odds ratio of 2.37 (0.52, 10.88).
Seven studies measured patient satisfaction/dissatisfaction with the method of induction technique (Figure 1F). The incidence of patient dissatisfaction was not significantly different. However, with an odds ratio of 2.26 (0.73, 6.94), there was a trend toward more dissatisfaction with the sevoflurane inhaled induction technique than with propofol.
Seven studies included data for PONV (Figures 1, G and H). Patients were significantly more likely to have nausea with an odds ratio of 4.24 (1.90, 9.47) and/or vomiting with an odds ratio of 3.18 (1.38, 7.32) if they were in the sevoflurane group The numbers needed to harm for nausea was 4.00 inductions and for vomiting 6.76 inductions The more frequent incidence of PONV in the sevoflurane group remained, even when we restricted the analysis to studies in which the only variable was the difference in induction drugs (all patients in this subgroup received the same maintenance anesthetic consisting of sevoflurane/N2O) (Figures 1, I and J). The relative risk for nausea/vomiting for these patients was 3.75 and 8.77 inductions, respectively.
Induction characteristics, including time to LOC, induction complications, and time to LMA insertion, were similar in both groups. All complications reported were minor, and the consequences were immaterial. The lack of any severe anesthetic induction complications in the studies reviewed may attest to the safety of both induction techniques.
In our review, the incidence of transient apnea during induction was also more frequent in the propofol group as compared with the sevoflurane group. The importance of transient apnea is unclear, as the incidence of pulmonary aspiration of gastric contents was not found to be different between patients breathing spontaneously and patients whose lungs were ventilated with positive pressure (16).
Weighted mean first time success rate with LMA was 89% in the sevoflurane group and 84% in the propofol group. The success rates in the studies in this review were lower than previously reported success rates by experienced users after propofol induction. Success rates have been reported as high as 95.5% on first attempt (17). These low success rates may be results of the strict methodology used for anesthetic induction in both groups.
Patient satisfaction/dissatisfaction should play an important role in the choice of anesthetic technique in healthy patients who undergo minor operations. In the studies examined, patient dissatisfaction was not significantly different with either induction method. There was a trend toward more patient dissatisfaction in the sevoflurane induction group. However, there may be bias built into the studies. Patients who disliked inhaled induction because of perceived or previous adverse experiences would not have enrolled in inhaled studies. Patients may be dissatisfied with sevoflurane induction because of the phobia of suffocation brought on by mask placement, the smell of sevoflurane, or PONV. Patients may be dissatisfied with the propofol induction because of the pain associated with propofol injection The exact reasons for overall patient preferences are unclear. However, with the trend toward more patient dissatisfaction with sevoflurane induction, further studies on satisfaction with anesthetic induction technique may show patient preference with IV propofol induction.
PONV has been a major morbidity for outpatient surgery. In fact, vomiting is the most important factor from a patient’s point of view (18). PONV was almost twice as common in the sevoflurane induction group. This was true even when both groups received the same maintenance anesthetic and postoperative care. Patients who receive sevoflurane for induction were at more frequent risk of PONV. Even when compared with thiopental, sevoflurane for induction was associated with a more frequent rate of postoperative vomiting (10). So far, no adequate explanation has been given for the higher incidence of PONV in patients induced with sevoflurane. One explanation may be that these patients did not receive propofol, which has antiemetic properties (19). Another explanation may be that sevoflurane causes a greater incidence of PONV. The PONV may be a function of the initial high concentration of sevoflurane or it may be caused by air and gases, which may be swallowed into the stomach during induction. It may be cogent to give prophylactic antiemetics to patients receiving inhaled induction with sevoflurane.
Sevoflurane may be an excellent induction drug in needle-phobic patients, pediatric patients, and patients with a potentially difficult airway. Its role in these special situations has been extremely helpful. In fact, sevoflurane has replaced halothane as the gold standard for inhaled anesthetic induction. However, it seems that, for routine outpatient anesthetic induction, sevoflurane has few definitive advantages, such as maintenance of spontaneous ventilation and has some important disadvantages when compared with propofol. Anesthetic induction with sevoflurane results in more frequent incidence of PONV and may be associated with higher patient dissatisfaction when compared with propofol IV induction.
In conclusion, sevoflurane and propofol had similar efficacy for anesthetic induction. However, because of the more frequent incidence of PONV and a trend toward patient dissatisfaction in the sevoflurane-inhaled induction group, propofol still seems to have advantages as an ideal drug to induce anesthesia.
1. Muzi M, Robinson BJ, Ebert TJ, O’Brien TJ. Induction of anesthesia and tracheal intubation with sevoflurane in adults. Anesthesiology 1996; 85:536–43.
2. Jadad AR. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17:1–12.
3. Smith I, Ding Y, White PF. Comparison of induction, maintenance, and recovery characteristics of sevoflurane-N2
O and propofol-sevoflurane-N2
O with propofol-isoflurane-N2
O anesthesia. Anesth Analg 1992; 74:253–9.
4. Fredman B, Nathanson MH, Smith I, et al. Sevoflurane for outpatient anesthesia: a comparison with propofol. Anesth Analg 1995; 81:823–8.
5. Jellish WS, Lien CA, Fontenot HJ, Hall R. The comparative effects of sevoflurane versus propofol in the induction and maintenance of anesthesia in adult patients. Anesth Analg 1996; 82:479–85.
6. Lien CA, Hemmings HC, Belmont MR, et al. A comparison: the efficacy of sevoflurane-nitrous oxide or propofol-nitrous oxide for the induction and maintenance of general anesthesia. J Clin Anesth 1996; 8:639–43.
7. Hall JE, Stewart JI, Harmer M. Single-breath inhalational induction with sevoflurane anaesthesia with and without nitrous oxide: a feasibility study in adults and comparison with an intravenous bolus of propofol. Anaesthesia 1997; 52:410–5.
8. Thwaites A, Edmends S, Smith I. Inhalational induction with sevoflurane: a double-blind comparison with propofol. Br J Anaesth 1997; 78:356–61.
9. Nelskyla K, Korttila K, Yli-Hankala A. Propofol-alfentanil-N2O is preferable to sevoflurane-N 2O for minor gynaecological surgery. Br J Anaesth 1999; 83:576–9.
10. Fleischmann E, Akca O, Wallner T, et al. Onset time, recovery duration, and drug cost with four different methods of inducing general anesthesia. Anesth Analg 1999; 88:930–5.
11. Ti LK, Chow MY, Lee TL. Comparison of sevoflurane with propofol for laryngeal mask airway insertion in adults. Anesth Analg 1999; 88:908–12.
12. Molloy ME, Buggy DJ, Scanlon P. Propofol or sevoflurane for laryngeal mask airway insertion. Can J Anaesth 1999; 46:322–6.
13. Tang J, Chen L, White Pf, et al. Recovery profile, costs, and patient satisfaction with propofol and sevoflurane for fast-track office-based anesthesia. Anesthesiology 1999; 91:253–61.
14. Philip BK, Lombard LL, Roaf ER, et al. Comparison of vital capacity induction with sevoflurane to intravenous induction with propofol for adult ambulatory anesthesia. Anesth Analg 1999; 89:623–7.
15. Dashfield DJ, Birt J, Thurlow I, et al. Recovery characteristics using single-breath 8% sevoflurane or propofol for induction of anaesthesia in day-case arthroscopy patients. Anaesthesia 1998; 53:1062–6.
16. Keller C, Sparr HJ, Luger TJ, Brimacombe J. Patient outcomes with positive pressure versus spontaneous ventilation in non-paralysed adults with the laryngeal mask. Can J Anaesth 1998; 45:564–7.
17. Brimacombe J. Analysis of 1500 laryngeal mask uses by one anaesthetist in adults undergoing routine anaesthesia. Anaesthesia 1996; 51:76–80.
18. Marcario A, Weinger M, Carney S, Kim A. Which clinical anesthesia outcomes are important to avoid? The perspective of patients. Anesth Analg 1999; 89:652–8.
19. Borgeat A, Stirnemann HR. Antiemetic effect of propofol. Anaesthesist 1998; 47:918–24.