Although nitrous oxide (N 2 O) has been used clinically

Transcription

1 Ambulatory Anesthesiology Section Editor: Peter S. A. Glass Can Remifentanil Replace Nitrous Oxide During Anesthesia for Ambulatory Orthopedic Surgery with Desflurane and Fentanyl? Donald M. Mathews, MD Vijay Gaba, MD Bledi Zaku, BS George G. Neuman, MD BACKGROUND: The administration of nitrous oxide (N 2 O) may be associated with side effects and toxicities. Remifentanil shares characteristics with N 2 O, including MAC-reducing and antinociceptive effects and a rapid decrease in clinical effect when discontinued. We compared the outcome after ambulatory orthopedic surgery with desflurane and fentanyl supplemented with clinically equivalent doses of either N 2 O or remifentanil. METHODS: Seventy patients undergoing ambulatory orthopedic surgery were studied. Thirty-five received 66% N 2 O and 35 received remifentanil g kg 1 min 1 in addition to desflurane, titrated to a bispectral index (BIS) value of 50, and a fentanyl infusion. The principle outcome measure was time to awakening to verbal stimulation. Secondary outcome measures included neuropsychological testing, time to orientation, hemodynamic values, pain and nausea visual analog scores, discharge times, and satisfaction scores. The average end-tidal desflurane concentration and fentanyl effect-site concentration were determined. RESULTS: The median time (interquartile range) to awakening to verbal stimulation, 3.0 min ( min) in the remifentanil group and 4.6 min ( min) in the N 2 O group was not significantly different. Median time to orientation was significantly faster in the remifentanil group: 6.0 min ( min) compared with 8.0 min ( min) for the N 2 O group. There was no difference between groups in desflurane or fentanyl administration, neuropsychological testing, or any other outcome measure. CONCLUSIONS: This study demonstrates that a remifentanil infusion of g kg 1 min 1 may be substituted for 66% N 2 O during desflurane/fentanyl anesthesia without any clinically significant change in outcome. (Anesth Analg 2008;106:101 8) Although nitrous oxide (N 2 O) has been used clinically for more than 160 yr, some current practitioners question its role in the era of modern, less soluble volatile anesthetics, especially in the context of its known side effects and toxicities. 1,2 Others maintain that N 2 O, with proper patient selection, offers advantages, primarily economic, over other available anesthetics. 3 When considering possible replacement anesthetics, Smith 4 recently stated perhaps remarkably for an agent so often suggested as an alternative to nitrous oxide, there have to date been no published comparisons between remifentanil and nitrous oxide as supplements to either propofol or inhalational From the Department of Anesthesiology, St. Vincent Catholic Medical Centers, St. Vincent s Manhattan, and New York Medical College, Valhalla, New York. Accepted for publication September 4, Address correspondence and reprint requests to Donald M. Mathews, MD, Department of Anesthesiology, St Vincent s Hospital, 170 West 12th St., NR 408, New York, NY Address to dmathews@svcmcny.org. Copyright 2007 International Anesthesia Research Society DOI: /01.ane anesthesia that have evaluated any indices of relevance to ambulatory anesthesia. N 2 O and remifentanil share similar pharmacokinetic profiles. Both have a rapid onset of action and short half-life after administration: N 2 O because of its low solubility and remifentanil because of rapid hydrolysis by tissue esterases. Neither drug is, by itself, a complete anesthetic; each must be used with other therapeutic drugs. The MAC of N 2 O in humans is 104%, 5 which is physiologically impossible to deliver without a hyperbaric chamber. The maximum MAC-reducing effect of 60% 70% N 2 O is approximately Remifentanil alone, even at very high doses, is unable to suppress movement to stimulation in 100% of patients. The MACreducing effect of remifentanil reaches an asymptote of about 90% with a whole blood concentration between 15 and 32 ng/ml. 7 N 2 O is clinically used as an adjuvant to volatile anesthetics and narcotics. This study was designed to determine whether there was any difference between a clinically equivalent dose of remifentanil and N 2 O when administered in a desflurane/fentanyl anesthetic. The primary outcome measure was time to eye opening to verbal stimulation. Secondary outcomes Vol. 106, No. 1, January

2 were time to orientation, performance on neurophysiologic (NP) tests during recovery, pain and nausea visual analog scale (VAS) scores, administration of medication to treat pain and nausea, time to discharge readiness, and satisfaction scores at time of discharge and at 1 day postoperatively. METHODS Determination of Clinical Equivalency Lang et al. 8 demonstrated that the whole blood concentration of remifentanil needed to decrease the MAC of isoflurane by 50% was 1.37 ng/ml, and MAC was decreased by 60% at 2 ng/ml, which is a similar degree of MAC reduction expected by 66% N 2 O. 9 We assumed a similar degree of desflurane MAC reduction and used STANPUMP (Steven Shafer, Palo Alto, CA; simulation mode) to model a remifentanil (Minto kinetics) bolus and infusion that would produce a stable whole blood concentration of 2 ng/ml in 5 min. This time course paralleled the increase in N 2 O seen with the gas flows planned for the study, 1 L/min O 2 and 2 L/min N 2 O. It was determined that a bolus of remifentanil 0.7 g/kg and infusion of g kg 1 min 1 produced the desired blood concentration. Clinical Protocol After approval by the hospital IRB, 70 patients were recruited for the study and signed an informed consent. Inclusion criteria were age between 18 and 55 yr, orthopedic surgery of 1 2 h expected duration, and body mass index between 18 and 30. Exclusion criteria were surgery on the dominant hand or arm; history of hypertension; significant cardiac, pulmonary, or neurological disease; opioid use within 14 days, or consumption of more than five alcoholic drinks per week; allergy to any study medication; or history of significant postoperative nausea and vomiting. Seventy patients were randomized by envelope to either the N 2 O (NITROUS) or the remifentanil (REMI) groups. In the preoperative area, patients were familiarized with the pain and nausea VAS and a blinded research assistant administered the initial NP tests (Psychological Assessment Resources, Lutz FL), which included 10-s finger tap, digit-symbol substitution test, and color trail-making test Part A. In the operating room, routine clinical monitors were applied including a BIS XP sensor (Aspect Medical System, Newton MA, version 3.23, 15 s smoothing) applied according to the manufacturer s specifications. Midazolam 0.05 mg/kg was given for anxiolysis, and beginning 5 min later, three heart rate and arterial blood pressure measurements were obtained during the subsequent 3 min. These were averaged and served as baseline hemodynamic measurements. After administration of oxygen, general anesthesia was induced with propofol 1.5 mg/kg and fentanyl 1.5 g/kg. A laryngeal mask airway (LMA) was inserted without neuromuscular blockade and a fentanyl infusion of 0.75 g kg 1 h 1 was started. An unblinded anesthesiologist then began the study medication. For those in the NITROUS group, the flow meters were set to oxygen 1 L/min, N 2 O 2 L/min and were covered, and the patient received a sham saline bolus and infusion. For patients in the REMI group flow meters were set to oxygen 1 L/min, air 2 L/min and were covered, and the patient received a remifentanil bolus 0.7 g/kg, followed by infusion of g kg 1 min 1. The MAC calculation displayed on the Datex S/5 monitor (Datex- Ohmeda, Helsinki, Finland) was also covered for all patients. Desflurane 4% was begun and ventilation was controlled to maintain an end-tidal carbon dioxide of 32 mm Hg. Regional anesthetics were not used. During the procedure, a second, blinded anesthesiologist made the decisions about clinical care based on the following variables: desflurane was adjusted to maintain a BIS value of 50; a fentanyl bolus 0.2 g/kg was administered as often as every 2 min for heart rate or arterial blood pressure more than 20% above baseline, tearing, sweating, or patient movement. During surgery, the heart rate, BIS values, and end-tidal desflurane concentration were recorded by a computer, every second using RUGLOOP (written by Michel Struys and Tom DeSmet, University of Ghent, Belgium, more information at arterial blood pressure was determined every minute. The area under the end-tidal desflurane concentration curve was determined from induction to drug discontinuation. BIS and hemodynamic data were averaged each minute using LABGRAB (also written by Michel Struys and Tom DeSmet) and determined from skin incision until drug discontinuation. For each patient, the percent change in average hemodynamic data from baseline to the surgical period was calculated. At the conclusion of the surgery, the surgeon injected 30 ml 0.25% bupivicaine intraarticularly, the first anesthesiologist discontinued the desflurane, fentanyl, and study drugs simultaneously and administered keterolac 60 mg IV. No attempt was made to titrate or decrease the anesthetic dosing during skin closure. The second, blinded anesthesiologist determined time to eye opening to verbal stimulation, LMA removal, and orientation to date and place. Verbal stimulation was achieved by loudly calling the patient s name every 20 s and the LMA was removed when the patient followed commands and had a respiratory rate of at least 8 breaths per minute. In the postanesthesia care unit (PACU) at 20, 40, and 60 min after awakening, 100-mm VAS pain and nausea scores were recorded and a blinded research assistant repeated the NP tests. Medication administered for nausea and pain was recorded. Pain was treated with fentanyl 25 g every 5 min, until VAS pain was below 30 mm, and nausea was treated with ondansetron 4 mg. Time spent in Stage 1 (acute postoperative care) and Stage 2 (subacute care, preparation for discharge) PACU were determined, and discharge from each unit was based on the usual clinical criteria. At the time of 102 Comparison of Nitrous Oxide and Remifentanil ANESTHESIA & ANALGESIA

3 Table 1. Demographics REMI (n 34) NITROUS (n 35) Significance Age N.S. BMI 26.4 (3.2) 25.8 (2.7) N.S. Height (cm) 173 (9) 175 (10) N.S. Weight (kg) 80 (15) 78 (13) N.S. Gender (M:F) 23:11 25:10 N.S. Surgeries ACL reconstruction N.S. Shoulder arthroscopy ORIF ankle 3 2 ORIF wrist 3 2 ORIF patella 2 1 Ankle arthroscopy 1 2 Mean (SD), BMI body mass index; ACL anterior cruciate ligament; ORIF open reduction, internal fixation. Table 2. Hemodynamics REMI (n 34) NITROUS (n 35) Significance Heart rate Baseline 77.5 (12.7) 69.9 (9.1) Surgery 61.8 (10.0) 60.9 (9.3) N.S. Change from baseline 21% ( 26% to 12%) 17% ( 21% to 2%) 0.01 Systolic blood pressure Baseline (12.5) (10.3) Surgery (16.3) (16.2) N.S. Change from baseline 8% ( 14% to 3%) 3% ( 11% to 7%) N.S. Diastolic blood pressure Baseline 75.5 (8.4) 75.3 (9.5) N.S. Surgery 71.5 (11.4) 72.7 (9.9) N.S. Change from baseline 4% ( 16% to 3%) 4% ( 14% to 6%) N.S. Baseline and surgery values are mean (SD). Change from baseline is median (25th percentile 75th percentile). patient discharge from the hospital, a series of satisfaction questions was administered and the questions were repeated by telephone 24 h later. The fentanyl infusions and all fentanyl boluses, including the induction dose, were simulated post hoc using STANPUMP in simulation mode (Shafer kinetics). For each patient, the area under the concentration time curve was determined from induction to drug discontinuation. Statistical Analysis Parametric data were analyzed by unpaired t-testing; nonparametric data were compared with Mann Whitney rank-sum analysis. Awakening and orientation time were compared by Kaplan Meier survival analysis. NP data were analyzed by Kruskal Wallis repeated measures ANOVA. The proportion of patients between groups was compared by 2 analysis. A P value of 0.05 was considered significant. SigmaStat version 3.0 (Systat, Port Richmond, CA) was used for statistical analysis. Power Analysis The study was powered to detect a difference in time of 2 min to awakening between the groups. Using a group standard deviation, from pilot data, of 4 min with an [alpha] of 0.05 and a power of 0.8, we determined that 32 patients per group were required. We chose to enroll 35 per group to allow for dropouts or technical problems. RESULTS Thirty-five patients were enrolled in each group. One patient in the REMI group was excluded because the remifentanil syringe became disconnected from the IV line sometime during the case. There was no difference in patient demographics between the groups, as displayed in Table 1. Hemodynamic differences between the groups are displayed in Table 2. The baseline resting heart rate and systolic blood pressure were greater in the REMI group than in the NITROUS group. During surgery, there were no differences between the groups but there was a larger decrease in heart rate between baseline and surgery in the REMI group than measured in the NITROUS group. Data regarding the surgery and anesthesia dosing are presented in Table 3. Case time, desflurane usage (area under the concentration time curve), and fentanyl usage (total delivery, infusion rate [baseline infusion plus boluses] in g kg 1 min 1 ), median number of boluses per hour, or area under the concentration time curve were not significantly different. Seven patients in the REMI group and six in the NITROUS group required no Vol. 106, No. 1, January International Anesthesia Research Society 103

4 Table 3. Surgery and Anesthesia REMI (n 34) NITROUS (n 35) Significance Case length (min) (49.0) 99.3 (39.3) N.S. Desflurane AUC (%) 2.8 (0.6) 2.6 (0.5) N.S. Fentanyl Total dose ( g) 315 (143) 306 (107) N.S. Calculated effect-site AUC (ng/ml) 1.60 (0.41) 1.61 (0.33) N.S. Fentanyl bolus/h (median) N.S. Overall delivery (including induction dose) Infusion boluses ( g kg 1 min 1 ) (0.014) (0.010) N.S. Mean (SD ); AUC area under curve. Figure 1. Area under the end-tidal desflurane concentration curve and area under the fentanyl effect-site concentrations curve for patients receiving nitrous oxide (unfilled squares) and remifentanil (filled diamonds) and group means: nitrous oxide (large filled square) and remifentanil (large filled diamond). Errors bars represent standard deviation. fentanyl boluses (N.S.). The scatter plot of average desflurane and average calculated effect-site fentanyl concentration is displayed in Figure 1. The intraoperative BIS values from incision until drug discontinuation (mean sd) were also similar: and (N.S.), for the REMI and NITROUS groups, respectively. The median (interquartile range) time to eye opening on verbal command was 3.0 min ( min) in the REMI group and 4.6 min ( min) in the NITROUS group; the difference was not statistically significant. One patient in the REMI group took 19 min to awaken, 10 min longer than the patient with the next longest time. When this patient was excluded as an outlier, the awakening time in the REMI group was significantly faster (P 0.021). The median (interquartile range) to orientation was 6.0 min ( min) in the REMI group 8.0 min ( min) in the NITROUS group, a statistically significant difference (P 0.034). The survival curve of those not oriented is displayed in Figure 2. Results of the NP tests are shown in Figure 3. At no time point was there a significant difference between the two groups. For each test, performance was better 60 min after eye opening compared with 20 min. In the NITROUS group, the 40-min performance was significantly better than the 20-min performance in the digit Figure 2. Proportion of patients not oriented. The remifentanil group (solid line) was oriented significantly more quickly than the nitrous oxide group (dotted line) (P 0.034). symbol substitution test, and the 60-min scores were significantly better than the 40-min scores on the color trail-making test. There were no differences between the groups for the VAS pain and nausea scores at 20, 40, or 60 min as shown in Table 4. The percentage of patients treated for nausea (26% in the REMI group and 40% in the NITROUS group) was not statistically different, nor was the percentage of patients treated for pain (63% in the REMI group and 54% in the NITROUS group). Time to discharge eligibility from Stage 1 and Stage 2 PACU was also not statistically different. Stage 1 was min and min, and Stage 2 was min and min for the REMI and NITROUS groups, respectively. VAS satisfaction scores at time of discharge are shown in Table 4. There was no significant difference between the groups. There was also no difference in satisfaction scores obtained from the next day follow-up phone call (data not shown). 104 Comparison of Nitrous Oxide and Remifentanil ANESTHESIA & ANALGESIA

5 Figure 3. Neurophysiologic testing. Percent change from baseline measurement. Remifentanil (solid diamonds) and nitrous oxide (open squares). Error bars represent 95% confidence intervals. (A) 10-s finger tap test, (B) color trails test Part A, (C) digit symbol substitution test. *Significantly more than 20-min result, significantly more than 40-min result, significantly more than 20-min result. DISCUSSION This study demonstrates that, during ambulatory orthopedic surgery with a desflurane fentanyl general anesthetic, an infusion of remifentanil g kg 1 min 1 compared with 66% N 2 O resulted in faster time to orientation with no differences in other recovery measurements, including NP testing. N 2 O use is associated with unwanted side effects, which include distention of gas-filled cavities and postoperative nausea and vomiting. 10 The greater potential consequence of N 2 O exposure is inhibition of the enzyme methionine synthetase. 9 The clinical implications of this inhibition are not entirely clear but have been associated with hematological and neurological toxicity, as well and possible effects on the fertility of operating room personnel. 15 N 2 O use has also been associated with increased myocardial ischemia after carotid endarterectomy, probably related to the N 2 O-induced increase of homocysteine. 16 In contrast, N 2 O has been shown to have N-methyl-daspartate receptor antagonist properties in rats and could potentially benefit the prevention of tolerance to opioids. 17 Ultimately, each practitioner has to decide how he or she interprets the relative risks and benefits of N 2 O use. Practitioners, who choose to eliminate N 2 O from their practice, need to reconfigure their anesthetic technique to replace its contribution to the anesthetic state. One strategy is to increase the delivery of the primary anesthetic. Arellano et al. 18 reported that elimination of N 2 O in a propofol fentanyl anesthetic resulted in an increase in propofol administration of 15% in patients undergoing termination of pregnancy and 25% in gynecological laparoscopy patients with no change in discharge times. When using a volatile anesthetic, one may replace N 2 O by increasing volatile anesthetic delivery. With more soluble anesthetics, such as isoflurane, such an increase would lead to prolonged emergence. With less soluble anesthetics, such as desflurane or sevoflurane, the time for patients brain concentration to decrease below MAC-awake would not be significantly prolonged. However, there are important recovery milestones, such Table 4. Pain, Nausea, and Satisfaction Visual Analog Scores REMI (n 34) N 2 O(n 35) Significance Nausea 20 min 0 (0 22) 4 (0 24) N.S. 40 min 7 (0 28) 4 (0 26) N.S. 60 min 3 (0 31) 5 (0 27) N.S. Pain 20 min 40 (15 36) 40 (17 58) N.S. 40 min 40 (24 57) 42 (23 56) N.S. 60 min 28 (7 55) 32 (23 50) N.S. Pain satisfaction 80 (51 100) 89 (72 100) N.S. Nausea satisfaction 95 (70 100) 93 (49 100) N.S. Drug satisfaction 100 (97 100) 100 (84 100) N.S. QOR satisfaction 99 (86 100) 92 (71 100) N.S. Overall satisfaction 100 (98 100) 98 (94 100) N.S. Median (25th 75th percentile) score on 100 mm scale. QOR quality of recovery score. Vol. 106, No. 1, January International Anesthesia Research Society 105

6 as ability to swallow, that require a lower drug concentration. McKay et al., 19 for example, demonstrated that patients who received desflurane were significantly more successful at swallowing liquids upon awakening compared with those who received sevoflurane, an effect most likely due to the lower solubility of desflurane. Eger and Shafer 20 calculated that the time required for a 92% 95% decrement of volatile anesthestic level was prolonged with increasing length of exposure even with modern, less soluble drugs. It stands to reason that some recovery milestones may be delayed with the increased volatile anesthetic concentration that a non-n 2 O containing anesthetic would require, even when low solubility anesthetics are used. Another strategy is to replace the N 2 O with remifentanil. Lee et al. 21 demonstrated that a remifentanil infusion of 0.17 g kg 1 min 1 could replace 70% N 2 Oin patients receiving isoflurane undergoing colorectal surgery. In the current study, we used a lower remifentanil infusion, g kg 1 min 1 and a fentanyl infusion plus bolus design. In the study by Lee et al., all patients received 0.15 mg/kg morphine sulfate before skin incision, and remifentanil infusion was adjusted for criteria similar to those used in the current study for fentanyl bolus administration. The difference in the MACreducing ability of the background opioid probably accounts for the difference in remifentanil administration between the studies. Several comments should be made about our study design. First, the study was designed to deliver to both groups an equivalent depth of anesthesia, which can be conceptualized as separate, but related, depth of hypnosis and depth of antinociception. In our study, desflurane was the primary hypnotic drug, and although N 2 O and opioids can produce hypnosis, their primary role in the anesthetic was their antinociceptive properties. We attempted to obtain an equivalent depth of hypnosis by titrating the desflurane concentration to the BIS monitor (see below), and an equivalent depth of antinociception was attempted by treating autonomic or somatic signs of nociception with small fentanyl boluses. Second, the clinical equivalency between remifentanil and N 2 O was determined using their ability to reduce MAC. By extrapolating data from the interaction of isoflurane and remifentanil in reducing MAC, an infusion rate was determined with pharmacokinetic modeling. There are confidence intervals associated with each step, and there is a possibility of cumulative error, which could have lead to drug delivery that was less clinically equivalent than intended. The specific interaction of desflurane and remifentanil has not been described. Albertin et al. 22 recently described the reduction in desflurane MAC-BAR by remifentanil in the presence of 60% N 2 O. They found a 57% reduction with a remifentanil target-controlled infusion of 1 ng/ml and a 68% reduction at 3 ng/ml, data that were neither available during study design nor directly applicable because of the presence of N 2 O. It should be noted that the average anesthetic in each group, represented in Figure 1, was indistinguishable, supporting that the method used to determine clinical equivalency during study design was valid. The study was designed to result in a variation in fentanyl delivery. The baseline fentanyl administration (bolus 1.5 g/kg and infusion 0.75 g kg 1 h 1 )was designed to result in a plasma fentanyl concentration of approximately 1 ng/ml, which would be adequate for only a percentage of the population when desflurane was titrated to a BIS of 50. Indeed, the number of patients who required no additional fentanyl bolus in each group was similar and were approximately 18% of the study population; the majority of the patients required additional fentanyl boluses to maintain an acceptable anesthetic state. The calculated average fentanyl plasma concentration (1.6 ng/ml) delivered in this study would be expected to decrease volatile anesthetic MAC by approximately 50%. 23 The average desflurane AUC of 2.8% and 2.6% for the REMI and NITROUS groups, respectively, is in concert with this expected degree of MAC reduction. With the addition of 0.6 MAC-reducing equivalents from the remifentanil or N 2 O, the average patient received an overall anesthetic of approximately 1.6 MAC, which is approximately MAC-BAR. 24 There was variation in both fentanyl and desflurane dosing required to maintain an adequate anesthetic state among patients (Fig. 1). This degree of population variation is not different from that seen in other studies of opioid/volatile anesthetic interaction. 25 Although they have clinical similarities, N 2 O and remifentanil are not identical; they have different sites of action in the central nervous system and different interactions with volatile anesthetics. Remifentanil exerts its effects on the -opioid receptor, primarily those located in the dorsal horn of the spinal cord, periaqueductal gray, and locus coeruleus. N 2 O has been reported to have several sites of action. In the rat model, N 2 O releases norepinephrine in the spinal cord, which causes analgesia through activation of -1 adrenergic receptors. 26 Spinal -aminobutyric acid receptors 27 and brainstem opioid and -aminobutyric acid receptors 28 )may also be involved in analgesic effects. The interaction of N 2 O and volatile anesthetics is best described as additive. 29,30 The interaction of opioids and volatile anesthetics, however, is synergistic. 7,8 The study design assumes that the desflurane fentanyl N 2 O interaction 31 is similar to desflurane fentanyl remifentanil interaction. The latter has not been specifically described, nor, in fact, has the interaction of fentanyl and remifentanil. The lack of data about these interactions does not prevent them from being used clinically in combination. Our purpose in study design was not to determine the nature of these interactions, but rather to simply determine whether replacing N 2 O with remifentanil in an otherwise standard anesthetic would result in a similar or improved outcome. Desflurane was administered by titrating to the BIS monitor. N 2 O and opioids have different effects on the 106 Comparison of Nitrous Oxide and Remifentanil ANESTHESIA & ANALGESIA

7 electroencephalogram, which may result in unintended differences in desflurane administration. Although the effect of the drug combinations used in this study on the BIS have not been specifically characterized, it has been demonstrated that neither opioid 32 nor N 2 O, 33 when given individually, significantly changes the BIS value. Both N 2 O 34 and opioid, 35 however, shift the relationship of BIS value to responsiveness to stimulation to the right. That is, in the presence of N 2 O or opioid, the BIS value increases in 50% of a population that responds to a particular stimulus. We did not find a significant difference in the primary end-point, time to eye opening. The study was powered to detect a difference of 2 min between the groups; the actual median difference in time was 1.6 min. Excluding the outlier in the REMI group would have led to a significant difference, however. The statistically significantly faster time to orientation is interesting, but may lack clinically relevance, as we could demonstrate no difference in NP testing. Nor did the difference in time to orientation lead to significant differences in discharge time or patient satisfaction. The study demonstrates that the clinical outcome after an infusion of remifentanil g kg 1 min 1 compares favorably with 66% N 2 O in a desflurane/ fentanyl anesthetic during ambulatory orthopedic surgery. Patients receiving remifentanil were oriented more rapidly with no difference in any recovery or satisfaction measurement; however, the study was not designed nor powered to consider these as primary outcome measures. Substituting remifentanil for N 2 O may be an effective strategy for those who want to achieve similar rapid recovery profiles as that of N 2 O, while avoiding the potential toxicities of the latter. REFERENCES 1. Shaw AD, Morgan M. Nitrous oxide: time to stop laughing? Anaesthesia 1998;53: Stenqvist O, Husum B, Dale O. Nitrous oxide: an ageing gentleman. Acta Anaesthesiol Scand 2001;45: Hopkins PM. Nitrous oxide: a unique drug of continuing importance for anaesthesia. Best Pract Res Clin Anaesthesiol 2005;19: Smith I. Nitrous oxide in ambulatory anaesthesia: does it have a place in day surgical anaesthesia or is it just a threat for personnel and the global environment? Curr Opin Anaesthesiol 2006;19: Hornbein TF, Eger EI II, Winter PM, Smith G, Wetstone D, Smith KH. The minimum alveolar concentration of nitrous oxide in man. Anesth Analg 1982;61: Eger EI II. Age, minimum alveolar anesthetic concentration, and minimum alveolar anesthetic concentration-awake. Anesth Analg 2001;93: Manyam SC, Gupta DK, Johnson KB, White JL, Pace NL, Westenskow DR, Egan TD. Opioid-volatile anesthetic synergy: a response surface model with remifentanil and sevoflurane as prototypes. Anesthesiology 2006;105: Lang E, Kapila A, Shlugman D, Hoke JF, Sebel PS, Glass PS. Reduction of isoflurane minimal alveolar concentration by remifentanil. Anesthesiology 1996;85: Koblin DD, Waskell L, Watson JE, Stokstad EL, Eger EI II. Nitrous oxide inactivates methionine synthetase in human liver. Anesth Analg 1982;61: Tramer M, Moore A, McQuay H. Omitting nitrous oxide in general anaesthesia: meta-analysis of intraoperative awareness and postoperative emesis in randomized controlled trials. Br J Anaesth 1996;76: Felmet K, Robins B, Tilford D, Hayflick SJ. Acute neurologic decompensation in an infant with cobalamin deficiency exposed to nitrous oxide. J Pediatr 2000;137: Lee P, Smith I, Piesowicz A, Brenton D. Spastic paraparesis after anaesthesia. Lancet 1999;353: Hadzic A, Glab K, Sanborn KV, Thys DM. Severe neurologic deficit after nitrous oxide anesthesia. Anesthesiology 1995;83: Selzer RR, Rosenblatt DS, Laxova R, Hogan K. Adverse effect of nitrous oxide in a child with 5,10-methylenetetrahydrofolate reductase deficiency. N Engl J Med 2003;349: Rowland AS, Baird DD, Shore DL, Weinberg CR, Savitz DA, Wilcox AJ. Nitrous oxide and spontaneous abortion in female dental assistants. Am J Epidemiol 1995;141: Badner NH, Beattie WS, Freeman D, Spence JD. Nitrous oxideinduced increased homocysteine concentrations are associated with increased postoperative myocardial ischemia in patients undergoing carotid endarterectomy. Anesth Analg 2000;91: Richebe P, Rivat C, Creton C, Laulin JP, Maurette P, Lemaire M, Simonnet G. Nitrous oxide revisited: evidence for potent antihyperalgesic properties. Anesthesiology 2005;103: Arellano RJ, Pole ML, Rafuse SE, Fletcher M, Saad YG, Friedlander M, Norris A, Chung FF. Omission of nitrous oxide from a propofol-based anesthetic does not affect the recovery of women undergoing outpatient gynecologic surgery. Anesthesiology 2000;93: McKay RE, Large MJ, Balea MC, McKay WR. Airway reflexes return more rapidly after desflurane anesthesia than after sevoflurane anesthesia. Anesth Analg 2005;100: Eger EI II, Shafer SL. Tutorial: context-sensitive decrement times for inhaled anesthetics. Anesth Analg 2005;101: Lee LH, Irwin MG, Lui SK. Intraoperative remifentanil infusion does not increase postoperative opioid consumption compared with 70% nitrous oxide. Anesthesiology 2005;102: Albertin A, Dedola E, Bergonzi PC, Lombardo F, Fusco T, Torri G. The effect of adding two target-controlled concentrations (1 3 ng ml 1 ) of remifentanil on MAC BAR of desflurane. Eur J Anaesthesiol 2006;23: Sebel PS, Glass PS, Fletcher JE, Murphy MR, Gallagher C, Quill T. Reduction of the MAC of desflurane with fentanyl. Anesthesiology 1992;76: Roizen MF, Horrigan RW, Frazer BM. Anesthetic doses blocking adrenergic (stress) and cardiovascular responses to incision MAC BAR. Anesthesiology 1981;54: Katoh T, Kobayashi S, Suzuki A, Iwamoto T, Bito H, Ikeda K. The effect of fentanyl on sevoflurane requirements for somatic and sympathetic responses to surgical incision. Anesthesiology 1999;90: Orii R, Ohashi Y, Guo T, Nelson LE, Hashimoto T, Maze M, Fujinaga M. Evidence for the involvement of spinal cord alpha1 adrenoceptors in nitrous oxide-induced antinociceptive effects in Fischer rats. Anesthesiology 2002;97: Hashimoto T, Maze M, Ohashi Y, Fujinaga M. Nitrous oxide activates GABAergic neurons in the spinal cord in Fischer rats. Anesthesiology 2001;95: Ohashi Y, Guo T, Orii R, Maze M, Fujinaga M. Brain stem opioidergic and GABAergic neurons mediate the antinociceptive effect of nitrous oxide in Fischer rats. Anesthesiology 2003;99: Swan HD, Crawford MW, Pua HL, Stephens D, Lerman J. Additive contribution of nitrous oxide to sevoflurane minimum alveolar concentration for tracheal intubation in children. Anesthesiology 1999;91: Kihara S, Yaguchi Y, Inomata S, Watanabe S, Brimacombe JR, Taguchi N, Komatsuzaki T. Influence of nitrous oxide on minimum alveolar concentration of sevoflurane for laryngeal mask insertion in children. Anesthesiology 2003;99: Daniel M, Weiskopf RB, Noorani M, Eger EI II. Fentanyl augments the blockade of the sympathetic response to incision (MAC-BAR) produced by desflurane and isoflurane: desflurane and isoflurane MAC-BAR without and with fentanyl. Anesthesiology 1998;88:43 9 Vol. 106, No. 1, January International Anesthesia Research Society 107

8 32. Glass PS, Bloom M, Kearse L, Rosow C, Sebel P, Manberg P. Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 1997;86: Barr G, Jakobsson JG, Owall A, Anderson RE. Nitrous oxide does not alter bispectral index: study with nitrous oxide as sole agent and as an adjunct to i.v. anaesthesia. Br J Anaesth 1999;82: Karalapillai D, Leslie K, Umranikar A, Bjorksten AR. Nitrous oxide and anesthetic requirement for loss of response to command during propofol anesthesia. Anesth Analg 2006;102: Lysakowski C, Dumont l, Pellégrini M, Clergue F, Tassonyi E. Effects of fentanyl, alfentanil, remifentanil and sufentanil on loss of consciousness and bispectral index during propofol induction of anaesthesia. Br J Anaesth 2001;86: Comparison of Nitrous Oxide and Remifentanil ANESTHESIA & ANALGESIA