Mask anesthesia with a mixture of oxygen (O2) and nitrous oxide (N2O) carrying a potent volatile anesthetic remains the preferred approach for induction in children (1,2). Pediatric anesthesiologists follow different strategies for inhaled induction, i.e., single-breath vital capacity (3), large inspiratory concentration (4), or incremental increase techniques (1,2,5), most relying on the timed administration of different inspiratory concentrations of the drug. Although understanding the pharmacokinetic properties of the drug allows inhaled anesthesia adjustment, actual gas uptake during the induction of anesthesia in children is not well documented. Graphic data from a study abstract (6) showed that the rate of sevoflurane uptake during an induction in children was apparently faster than that for halothane. Generally, wash-in curves of inhaled anesthetics have been drawn for adults (7–9) and children (10–12) during maintenance of anesthesia after tracheal intubation with artificially controlled ventilation. This situation may differ considerably in awake children undergoing mask anesthesia because they are initially exposed to a very different gas environment that can produce dose-dependent ventilatory depression and airway irritability often managed with airway supporting maneuvers and assisted ventilation, decreasing the inconvenience of spontaneous ventilation. When compared with adults, infants and children present a larger ratio of alveolar ventilation to functional residual capacity (FRC), a larger fraction of cardiac output delivered to vessel rich organs, and different blood/gas partition coefficients for inhaled anesthetics (13,14). These factors account for age-related differences in the pharmacokinetic variables of volatile anesthetics.
The induction is a critical phase in anesthesia, and few studies have illustrated anesthetic gas uptake in the pediatric setting. In this study, the wash-in curves for sevoflurane and N2O were calculated in patients anesthetized with either a large (L-N2O) or a small (S-N2O) inspired N2O concentration to assess the rate of increase in the ratio of estimated alveolar concentration (FA) to inspired concentration (FI) measured simultaneously. This was done to confirm a concentration effect for N2O and to investigate the existence of a second gas effect for sevoflurane during an inhaled induction of anesthesia in healthy children.
Twenty consecutive ASA physical status I and II children aged 4–12 yr old were studied after the approval of the institutional committee on human research and parents’ informed consent. Patients were excluded if respiratory, gastrointestinal, circulatory, or neurological disease was diagnosed or if medications had been administered before surgery. Children scheduled for abdominal, ophthalmological, or cochlear implant procedures were premedicated with sublingual midazolam (0.2 mg/kg). Before the induction of anesthesia, standard clinical monitors (electrocardiogram, automated blood pressure, and pulse oximetry [Sao2]) were applied. An anesthetic machine (Julian®, Dräeger Medizintechnique GmbH, Mainz, Germany) with infrared spectrometer was calibrated according to factory specifications. An adult circle absorber circuit was primed with the desired gas concentration and prewashed for at least 15 min before each procedure. A small piece of the extension tube from a butterfly needle was gently introduced into the patient’s nostril, secured with adhesive tape, and connected to the sampling gas line. This device is usually well tolerated in premedicated children. The sampling flow was set at 200 mL/min. Once acceptable carbon dioxide (CO2) square waves with plateaus were observed, the anesthesia circuit was rapidly attached to the patient through a well-fitted face mask while the patient was encouraged to breathe normally. Beforehand, the fresh gas flow had been adjusted in excess of predicted minute ventilation to maintain the inspiratory gas concentration constant. The anesthesiologist who administered anesthesia was unaware of anesthetic gas data collection. If severe excitement or poor face-mask fit interfered with gas monitoring, the patient was excluded from the study. Ventilation was manually assisted after loss of eyelash reflex to avoid anesthetic dose-related ventilatory depression. An IV line was then started for fluid and drug administration. The research period was terminated before atropine, fentanyl, local anesthetics, or neuromuscular blocking drugs were administered. After the drug administration, tracheal intubation or laryngeal mask insertion proceeded as dictated by anesthetic requirements.
Patients were randomly assigned by a computer program to receive one of the following gas mixtures: 16% N2O and 6% sevoflurane (S-N2O) or 53% N2O and 6% sevoflurane (L-N2O), both with balance O2. Respiratory rate (RR), as well as end-tidal plus inspiratory gas concentrations for O2, CO2, N2O, and sevoflurane were continuously measured during 6–8 min from initiating anesthesia (time 0), digitized in 10-s epochs, and recorded in a portable computer for off-line data analysis. End-tidal and inspired concentrations of O2, N2O, and sevoflurane were summarized as the ratio of the estimated FA to the FI, measured simultaneously. Individual FA/FI ratios for O2, N2O, and sevoflurane together with RR and end-tidal CO2 were obtained off-line from the recorded data by an independent observer at 1, 2, 3, 4, 5, and 6 min after the start of anesthesia and were used for statistical analysis. Gas uptake during the first minute of the induction was assumed to represent the filling of FRC by anesthetics (8) and was not included in data analysis.
The unpaired t-test was used to determine the significance of differences between groups for patient characteristics and cardiovascular variables. Differences between preinduction (control) and postinduction (anesthesia) heart rate (HR), mean arterial blood pressure (MAP), and Sao2 values were assessed with a paired t-test. Repeated-measures analysis of variance was used to examine the behavior of individual FA/FI ratios, end-tidal CO2, and RR measured at 1-min intervals throughout the study period. Probability values for multiple comparisons were corrected with the Bonferroni multiple comparison test. SPSS (version 8.0) statistical software (SPSS, Chicago, IL) was used for data analysis. Data were expressed as mean ± sd, and P < 0.05 was considered significant.
Two patients receiving the gas mixture with the small N2O concentration developed severe excitement and breath-holding during the induction, making the mask difficult to secure and allowing a brief entrainment of air; this was retrospectively recognized by the sudden decrease in anesthetic gas concentration. These findings were considered to produce unacceptable data for gas analysis, and both patients were eliminated from the study. No other complications including airway obstruction, secretions, or laryngospasm were observed in both groups of children. Patient characteristics with respect to age (S-N2O, 8.2 ± 3.5 yr; L-N2O, 6.8 ± 2.8 yr), weight (S-N2O, 33.5 ± 20.1 kg; L-N2O, 29.5 ± 16.2 kg), height (S-N2O, 125.6 ± 25.1 cm; L-N2O, 132.1 ± 25.3 cm), and sex (S-N2O, 4:4; L-N2O, 6:4 M:F) did not differ between groups. After the mask induction of anesthesia with 6% sevoflurane and N2O, HR increased more than control values, but this was not significant (S-N2O, HRcontrol = 95.5 ± 11.36 bpm; HRanesthesia = 100.0 ± 26.08 bpm;P = 0.65, not significant; L-N2O, HRcontrol = 95.8 ± 9.47 bpm; HRanesthesia = 114.9 ± 35.76 bpm;P = 0.08; not significant). However, MAP decreased from control values (S-N2O, MAPcontrol = 83.25 ± 8.38 mm Hg; MAPanesthesia = 71.40 ± 14.95 mm Hg;P = 0.08; not significant; L-N2O, MAPcontrol = 84.90 ± 11.73 mm Hg; MAPanesthesia = 69.00 ± 16.16 mm Hg;P = 0.007). Sao2 presented a small, but significant, increase in the S-N2O group (1.88% ± 1.13%;P = 0.002) as a result of breathing the O2-enriched gas mixture (Table 1).
End-tidal CO2 remained virtually unchanged throughout the study period (37.02 ± 0.87 mm Hg) despite a nonstatistically significant trend to hyperventilate observed in the L-N2O group at the first minute (Fig. 1, top two curves). RR was reduced at the beginning of the induction (Fig. 1, bottom two curves), typically representing breath-holding events during inhaled anesthesia, but then gradually increased (18.72 ± 0.94 bpm). Differences between groups were not significant.
In the study period, the FA/FI for O2 increased significantly (F = 5.92, P < 0.001). The rate was faster at the beginning, and then values remained steady. The increase in FA/FI for O2 was similar in both groups (L-N2O, 0.92 ± 0.06; S-N2O, 0.92 ± 0.08) and did not differ significantly (Fig. 2).
The FA/FI for N2O increased significantly (F = 16.81, P < 0.001) after a similar pattern in both groups. The FA/FI for S-N2O was 0.87 ± 0.09 and was significantly larger in L-N2O (0.92 ± 0.08; F = 10.73, P < 0.01). Pair-wise comparisons between groups demonstrated significant differences at 3 (t = 5.2, P < 0.001), 4 (t = 3.6, P < 0.004), and 5 min (t = 3.2, P < 0.005) (Fig. 3).
The FA/FI for sevoflurane increased significantly during the induction (F = 3.45, P < 0.01) but at a slower rate than for N2O. The FA/FI of sevoflurane was 0.78 ± 0.07 (S-N2O) and increased to 0.80 ± 0.06 (L-N2O) (F = 1.2; not significant). Pair-wise comparisons in both groups demonstrated a significant difference only at 3 min (FA/FI, 85.03 ± 0.06 [L-N2O]; FA/FI, 77.67 ± 0.04 [S-N2O];t = 2.99, P < 0.01). This transitory increment in the FA/FI for sevoflurane when delivered with 53% of N2O vanished at 5 min, increasing in both groups of patients at a comparable rate during the rest of the induction (Fig. 4).
When the N2O concentration delivered was increased, the wash-in curves for N2O (Fig. 3) and sevoflurane (Fig. 4) shifted to the left, reaching maximum shifts of 125.8 s and 94.3 s, respectively.
This study documents the steep rate of increase in the ratio of estimated FA to FI for N2O and sevoflurane during an inhaled induction of anesthesia. Increasing the inspired N2O content produced a concentration effect for N2O and a brief second gas effect for sevoflurane. Essentially, as in earlier studies (7,8,10), both FA/FI ratios for N2O and sevoflurane increased rapidly at the beginning of the induction and then increased more slowly. Consistent with their respective blood/gas partition coefficients, the FA/FI for N2O increased more rapidly than that for sevoflurane but more slowly than for O2. Salanitre and Rackow (10) measured the uptake of N2O in mechanically ventilated children and confirmed the existence of a short-lived concentration effect when the inspired N2O concentration increases from 3% to 75%. The difference in FA/FI between the 3% and 75% curves was 0.07 at two minutes and 0.06 at five minutes and had faded by approximately 20 minutes (10).
The concentration effect theory suggests that the level of N2O (the first gas during the induction) would influence the rate at which the FA approaches the FI (8,17). The classic idea was that the net uptake of large volumes of N2O would increase the inspired ventilation and concentrate both N2O and any other simultaneously administered gas (second gas effect). However, the validity of this principle has been questioned in an adult study (9); second, Salanitre and Rackow (10) did not statistically analyze the differences between the wash-in curves for N2O at different N2O concentrations in their halothane-anesthetized children. Finally, breath-by-breath data supporting the existence of concentration and second gas effect during the sevoflurane/N2O induction of anesthesia in children are scarce. More recently, Taheri and Eger (8), using discrete sample analysis, proved in mechanically ventilated adults that the administration of 65% N2O increased the FA/FI for N2O compared with the administration of 5% N2O by a moving average of 0.08. In the present study, the difference observed was 0.04 at two minutes, peaked to 0.06 at three minutes, and decreased to 0.03 at six minutes after initiating anesthesia. Although differences in methodology preclude comparison of results, smaller increments in the FA/FI ratios would be expected in this spontaneous breathing study because the administration of 53% N2O presumably produces less shrinkage of lung volume than the administration of 75% N2O and thereby would decrease the likelihood of creating a recognizable concentration effect (8,10). However, according to Korman and Mapleson (18), the compensation of net gas uptake occurring when large volumes of N2O are absorbed depends on the ventilation mode. During controlled ventilation with a predetermined inspired volume, the constant inflow principle together with a decreased expired ventilation to compensate the volume lost by FRC contraction would better describe gas uptake. On the contrary, a constant outflow model conforms to the scenario of spontaneous ventilation if end-tidal CO2 remains steady. Here, gas inflow is actively generated by chest expansion and passively drawn into the alveoli to compensate for net gas uptake. Salanitre and Rackow (10) postulated that because their data were obtained during mechanical ventilation with a volume-limited apparatus, they represented a smaller effect than would be seen with spontaneous breathing. Despite the observed early decrease in RR and the switch from spontaneous to assisted ventilation, end-tidal CO2 was well maintained throughout the measurement period (Fig. 1). Thus, spontaneous breathing and assisted ventilation may have compensated for the anticipated smaller effect on the FA/FI for N2O consistent with its smaller inspiratory concentration. The selected inspiratory concentrations of N2O were justified by the need to avoid extreme values in the N2O concentration because the cardiovascular-stimulating properties of N2O could modify the distribution of blood flow, consequently affecting gas uptake and FA/FI ratio.
Although the pungency of sevoflurane may limit the rate at which the inspired concentration increases, this rate would be presumably faster than with a more soluble anesthetic, such as halothane. Effectively, the FA/FI for sevoflurane approached 0.83 ± 0.05 in both groups at five minutes, whereas that for halothane in small infants was dramatically less at the same time (11). In vivo measurements of halothane wash-in curves and computer simulations for halothane uptake that incorporate age-related anatomic and physiologic variables predict these differences (11), and one may assume that sevoflurane uptake would be faster in children than in adults. Interestingly, it has been shown that an FA/FI ratio of 0.80 was achieved at approximately 10 minutes of sevoflurane administration in adults (7) in contrast with the curves shown in Figure 4, which reached the same equilibrium rate in remarkably less time at between 2.2 to 4 minutes.
Qualitatively, the increase in the FA/FI ratio for the different gases could be predicted from their respective blood/gas solubility. Accordingly, the uptake curve for sevoflurane exceeds those of halothane and isoflurane but was less than the one for N2O. It is also difficult to interpret the short-lived second gas effect for sevoflurane when compared with the well-defined concentrating effect for N2O. Experimental FA/FI curves obtained in dogs with ethylene, cyclopropane, and halothane (17) support the notion that the intensity of a second gas effect is also predictable from its respective partition coefficients. If this interpretation holds true in the clinical setting, it would not only explain the brief second gas effect for sevoflurane, but also its absence when a more soluble drug, such as enflurane, is used (8,9).
Butler et al. (15) using end-tidal O2 sampling determined that the duration of preoxygenation required to reach a FA/FI end-point of 0.9 was achieved within 80 seconds in all their children. They also confirmed that this point was achieved more rapidly than in adult patients. Although their data are not strictly comparable because they used 100% O2 as the inspired gas, the same 0.9 end-point was observed in the present study at 75 seconds (Fig. 2), which is close to their results. Arterial oxygenation increases in adults when nitrogen is replaced by N2O in the inspired gas mixture and the fraction of inspired O2 remained constant (16,19). It is feasible that a concentration effect for O2 was not detected here because it balanced the N2O content in the inhaled gas mixture. The agreement of O2 wash-in curve with published studies supports the use of this methodology for gas uptake assessment during the induction of anesthesia.
Although alterations in respiratory and hemodynamic variables may affect anesthetic gas uptake, only small changes that affect both groups similarly have been observed. Initially, RR was slower, consistent with the breathing pattern often observed during mask induction, and end-tidal CO2 showed a nonsignificant trend to hyperventilation in the L-N2O group; however, both events vanished after the first minute (Fig. 1). Assisted ventilation effectively cancelled the otherwise expected decreased ventilatory drive in patients spontaneously breathing large-dose sevoflurane, explaining the lack of change in end-tidal CO2. Ventilation influences gas uptake, especially during wash-in of FRC, but thereafter, increasing alveolar ventilation will move the FA/FI curve upward resulting in a decreased fraction of uptake, whereas the patient uptake remains unchanged. Because ventilation in both groups was increasingly preserved throughout the study period excluding the first minute of the induction, the observed differences in FA/FI are unlikely to be explained by ventilatory changes.
The present study failed to show a statistically significant increase in HR during the sevoflurane induction of anesthesia in children. Similar trends were seen by Lerman et al. (2) in infants and children. Because in most induction studies HR variability is large, the nonsignificant increase in the L-N2O group, hypothetically related to the stimulating properties of N2O, should be interpreted with caution. Sevoflurane generally decreased MAP in a dose-related manner in infants and children as a result of decreases in myocardial contractility and systemic vascular resistance (20–22). The modest decrease from preanesthetic MAP values compares favorably with the 26% reduction in systolic blood pressure reported by Wodey et al. (20) during the sevoflurane induction. Thus, the hemodynamic changes observed in both groups probably affect uptake in a similar manner, but the lack of data in humans make it difficult to quantify the influence of cardiac output. Nevertheless, hypotension may occur in any patient during the induction period, making appropriate fluid therapy and titration anesthesia a requirement.
In accordance with earlier reports validating this measurement technique (23,24), preliminary work demonstrated that end-tidal gas sampling through a nasal cannula preserved the CO2 curve without evidence of mixing with inspiratory gas inflow except during the first minute where irregular breathing occasionally occurred. Thereafter, the return of the CO2 curve to negligible values during inspiration confirmed that mixing of inspiratory and expiratory gas flows, although possible, was irrelevant in both patient groups. Because data recorded during the first minute of the induction were not considered for gas analysis, it is conceivable that results were probably unaffected. Although this breath-by-breath methodology was not specifically validated as an estimate of the FA of anesthetic under dynamic conditions seen during the mask induction, it was accepted for the minimum alveolar anesthetic concentration (23) and for end-tidal CO2 measurements in neonates (25), provided that the airway was patent. This is critical in children where nasopharyngeal obstruction caused by hypertrophic adenoid tissue is very likely. Breath-by-breath analysis of volatile anesthetics was early assessed with mass spectrometry, and more recently, concentration effect was also documented (16). In addition, CO2 output was validated in neonates sampling respiratory gases through a nasal catheter (25). Thus, breath-by-breath analysis is appropriate for patients with small tidal volume not requiring modifications of the anesthesia circuit to allow sampling of expired gas, as recommended with discrete sampling techniques in adults (8).
This study is the first to characterize the wash-in curves of N2O and sevoflurane during the mask induction of anesthesia in pediatrics. The concomitant use of N2O allowed the confirmation of a concentrating effect with a comparatively brief impact on the second gas, probably explained by the higher solubility of sevoflurane. Consistent with the concentration and second gas effects, the wash-in curves for N2O and sevoflurane exhibited a leftward shift when delivered with a large N2O content. The small concentration and second gas effects observed here in children agree with the recognized trend toward a more rapid equilibration between FA and FI than in adults. Pediatric anesthesiologists, convinced of the benefits of a quick and smooth induction of anesthesia, usually administer increased concentrations of potent and low-soluble anesthetics, taking advantage principally of the additive properties of N2O to the minimum alveolar anesthetic concentration of sevoflurane. Because of its weak effect, concentration and second gas effect are less important as a means to produce a clinically significant increase in anesthetic alveolar concentration. However, the fast equilibrium rate achieved between FA and FI and, to a minor extent, concentration and second gas effects, may subject some children to unnecessarily increased concentrations of the anesthetic and increase the risk for hypotension. Computer simulation of anesthetic gas exchange (26) predicts that the concentrating effect could be exacerbated in the presence of worsening ventilation-perfusion inhomogeneity. Whereas the clinical impact of this phenomenon is not yet fully established, children with characteristic inequalities in ventilation-perfusion from cardiopulmonary disease may require careful titration of their volatile anesthetics dose according to pharmacokinetic findings to lessen hemodynamic repercussions.
1. Sarner JB, Levine M, Davis PJ, et al. Clinical characteristics of sevoflurane in children: a clinical comparison with halothane. Anesthesiology 1995; 82: 38–46.
2. Lerman J, Sikich N, Kleinman S, Yentis S. The pharmacology of sevoflurane in infants and children. Anesthesiology 1994; 80: 814–24.
3. Agnor RC, Sikich N, Lerman J. Single-breath vital capacity rapid inhalation induction in children: 8% sevoflurane versus 5% halothane. Anesthesiology 1998; 89: 379–84.
4. Morimoto Y, Mayhew JF, Knox SL, Zornow MH. Rapid induction of anesthesia with high concentrations of halothane or sevoflurane in children. J Clin Anesth 2000; 12: 184–8.
5. Epstein RH, Stein AL, Marr AT, Lessin JB. High concentration versus incremental induction of anesthesia with sevoflurane in children: a comparison of induction times, vital signs and complications. J Clin Anesth 1998; 10: 41–5.
6. Lerman J, Sikich N, Levine M. Wash-in of sevoflurane in healthy children. Anesthesiology 1995; 83: 3A–1133.
7. Yasuda N, Lockhart SH, Eger EI II, et al. Comparison of kinetics of sevoflurane and isoflurane in humans. Anesth Analg 1991; 72: 316–24.
8. Taheri S, Eger EI II. A demonstration of the concentration and second gas effect in humans anesthetized with nitrous oxide and desflurane. Anesth Analg 1999; 89: 774–80.
9. Sun X-G, Feng S, Shi Y-Q, Lee C. The “second gas effect” is not a valid concept. Anesth Analg 1999; 88: 188–92.
10. Salanitre E, Rackow H. N2O volumes absorbed and excreted during N2O anesthesia in children. Anesth Analg 1976; 55: 95–9.
11. Brandom BW, Brandom RB, Cook DR. Uptake and distribution of halothane in infants: in vivo
measurements and computer simulations. Anesth Analg 1983; 62: 404–10.
12. Steward DJ, Creighton RE. The uptake and excretion of nitrous oxide in the newborn. Can Anaesth Soc J 1978; 25: 215–7.
13. Lerman J, Schmitt-Bantel BI, Gregory GA, et al. Age and solubility of volatile anesthetics in blood. Anesthesiology 1984; 61: 139–43.
14. Lerman J, Schmitt-Bantel BI, Gregory GA, et al. Effect of age on the solubility of volatile anesthetics in human tissues. Anesthesiology 1986; 65: 307–11.
15. Butler PJ, Munro HM, Kenny MB. Preoxigenation in children using expired oxygraphy. Br J Anaesth 1996; 77: 333–4.
16. Nishikawa K, Kunimoto F, Isa Y, et al. Second gas effect of N2O on oxygen uptake. Can J Anaesth 2000; 47: 506–10.
17. Stoelting RK, Eger EI II. An additional explanation for the second gas effect: a concentrating effect. Anesthesiology 1969; 30: 273–7.
18. Korman B, Mapleson WW. Concentration and second gas effects: can the accepted explanation be improved? Br J Anaesth 1997; 78: 618–25.
19. Bojrab L, Stoelting RK. Extent and duration of the nitrous oxide second-gas effect on oxygen. Anesthesiology 1974; 40: 201–3.
20. Wodey E, Pladys P, Copin C, et al. Comparative hemodynamic depression of sevoflurane versus halothane in infants: an echocardiography study. Anesthesiology 1997; 87: 795–800.
21. Holzman RS, van der Velde ME, Kaus SJ, et al. Sevoflurane depresses myocardial contractility less than halothane during induction of anesthesia in children. Anesthesiology 1996; 85: 1260–7.
22. Kern C, Erb T, Frei FJ. Haemodynamic responses to sevoflurane compared with halothane during inhalational induction in children. Paediatr Anaesth 1997; 7: 439–44.
23. Nishina K, Mikawa K, Shiga M, et al. Oral clonidine premedication reduces minimum alveolar concentration of sevoflurane for tracheal intubation in children. Anesthesiology 1997; 87: 1324–7.
24. Kihara S, Inomata S, Yaguchi Y, et al. The awakening concentration of sevoflurane in children. Anesth Analg 2000; 91: 305–8.
25. Evans JM, Hogg MI, Rosen M. Measurement of carbon dioxide output, alveolar carbon dioxide concentration and alveolar ventilation in the neonate. Br J Anaesth 1977; 49: 453–6.
26. Peyton PJ, Robinson GJB, Thompson B. Effect of ventilation-perfusion inhomogeneity and N2O on oxygenation: physiological modeling of gas exchange. J Appl Physiol 2001; 91: 17–25.