During sevoflurane anesthesia, compound A can be formed, particularly in low-flow, minimal-flow, and closed-circuit systems. Compound A has a known toxicity in rats (1) which primarily involves renal, hepatic, and cerebral damage. If it is indeed assumed that compound A produces similar toxic effects in humans, patients are likely to be at significant risk when sevoflurane is delivered from a circle absorber breathing circuit with low fresh gas flows (500 mL/min). All evidence suggests that the compound A levels achieved in clinical practice are well below the concentration that is toxic in animals (2,3). Nevertheless, the compound A issue is still being hotly and controversially debated, especially whether compound A has an effect clinically on renal function. As a consequence, some still question the use of sevoflurane in low- or minimal-flow conditions (4). Different CO2 absorbents, however, vary enormously in their capacity to degrade sevoflurane to compound A (5–7). In commercially available soda limes, sodium hydroxide (NaOH) and potassium hydroxide (KOH) are mostly used as initiators in the CO2 chemical binding process. They are highly reactive compounds that cause the breakdown of sevoflurane in the canister of the breathing circuit. Strong alkali-free Amsorb® (Armstrong, Coleraine, Northern Ireland), however, does not produce compound A in experimental conditions (8,9) or in clinical practice (10). Recently a new NaOH- and KOH-free CO2 absorbent, Superia® (Molecular Products, Thaxted, UK), became available. It is a Ca(OH)2-based absorbent containing ±17.5% water as well as 1% MgCl2 and 1% aluminosilicate.
It was our aim to study the use of Superia in minimal-flow anesthesia, in which compound A in large concentrations is eventually expected. Superia was compared with KOH-free Sofnolime® (Molecular Products). Sofnolime contains >75% Ca(OH)2, 3% NaOH, and 19%–21% H2O, but no KOH.
After ethical committee approval and written, informed consent, 18 adult ASA status I–III patients scheduled for elective urological or gynecological surgery and with a predicted anesthesia duration of at least 2.5 h were accepted in the study. They were randomized to receive sevoflurane minimal-flow anesthesia with either Superia or Sofnolime as the CO2 absorbent. As a result of noncompliance to the rigid study protocol, four patients had to be removed from the study (anesthesia time was too short). The remaining patients (eight men and six women), aged 52 ± 11 yr (mean ± sd) and weighing 78 ± 8 kg, were analyzed.
All patients were premedicated with lorazepam 1 mg orally 45 min before the induction of anesthesia. Anesthesia was induced with propofol (Diprivan; Astra-Zeneca, Brussels, Belgium) by using a target-controlled infusion device (Diprifusor; Astra-Zeneca). Blood target concentration was initially set at 4 μg/mL. All patients also received a continuous infusion of remifentanil, initially set at 0.1 μg · kg−1 · min−1. After loss of consciousness, 0.5 mg/kg of vecuronium was given to allow the placement of an endotracheal tube. Anesthesia was maintained with a constant end-tidal (ET) sevoflurane concentration. Only remifentanil administration was increased or decreased stepwise according to clinical necessity. After the induction of anesthesia, controlled ventilation was done with an ADU ventilator (Datex-Ohmeda, Helsinki, Finland), with a tidal volume aiming at a PETco2 of 33–35 mm Hg. In this machine, tidal volume is independent of the fresh gas flow rate. A special vapor admission Aladin cassette (Datex-Ohmeda) was used, whereby the inspiratory concentration can be electronically dialed (20). The amount of liquid sevoflurane was cumulatively calculated by using the A-FGC1 module of the AS/3 ADU ventilator (Datex-Ohmeda). With this module, it is possible to calculate precisely and on-line the flow (mL/min) of the anesthetic liquid by measuring pressure and flow in the different parts on the breathing system. A sevoflurane (Sevorane; Abbott Laboratories, Abbott Park, IL) ET concentration of 2.3–2.5 vol% was aimed at throughout the entire study period. After ventilation equilibration in temporarily high-flow conditions to stabilize fraction of inspired oxygen (Fio2) and PETco2 values during the transition period from manual ventilation before intubation to constant mechanical ventilation (fresh gas flow of 6 L/min during 15 min), minimal flow was started (500 mL/min). An oxygen/air mixture was used (Fio2 = 0.4).
Immediately before minimal-flow conditions (designated as “pre” and used as the baseline measure), at the stable sevoflurane minimal-flow conditions (designated as “start”), and at 15, 30, 60, 90, 120, 150, and 210 min thereafter, data were recorded. Besides routine cardiovascular variables, Fio2, PETco2, inspired sevoflurane, end-tidal sevoflurane, and the amount of administered sevoflurane were recorded.
Beforehand, two thermistors (Arbo, Yellow Springs, OH) were introduced into the small absorbent lime canister which has a storage capacity for 750 mL of CO2 absorbent granules, to measure the temperature. Because the Datex-Ohmeda canister has two vertically separated compartments, one thermistor was located in the inflow part, measuring inflow temperature (T°in), while the other was in the outflow part (T°out). At all times, fresh absorbent was used. Gas samples (2 mL) were taken, in duplicate, by means of airtight syringes for the determination of compound A concentrations. The syringes were connected to the breathing circuit in a gas-tight manner by means of three-way valves and male Luer-lock connections, two in the inspiratory limb (for compound Ainsp) and two in the expiratory limb (for compound Aexp) of the breathing circuit. The gas samples were immediately transferred into 5-mL glass headspace vials (Gerstel Gmbh, Mülheim-an-der-Ruhr, Germany) and stored briefly at room temperature, awaiting analysis. All these gas samples were taken at the above-mentioned times.
Compound A was assayed by capillary gas chromatography combined with mass-spectrometric detection (HP 6890-5973 MSD; Agilent Technologies, Palo Alto, CA). Injection was fully automated in a technique based on headspace sampling (1 mL) and cryogenic focusing in the injector liner on a Tenax® TA adsorbent (Alltech, Deerfield, IL). Isothermal separation (38°C) was achieved on a thick-film capillary column (CP-select 624, a 6% cyanopropylphenyldimethyl silicone stationary phase; Chrompack, Middelburg, The Netherlands). Helium was used as carrier gas at a flow rate of 1 mL/min. The mass spectrometric detector was operated in the full-scan mode. For more details on the analytical methodology and validation, we refer to Bouche et al. (21).
Statistical analysis was performed with repeated-measures analysis of variance, followed by a post hoc test (Tukey) or a Mann-Whitney U-test, where appropriate. The level of significance was set at 5%. Correction for multiple testing was used. All data are reported as median (range) unless otherwise indicated.
To calculate the power for this study, we hypothesized that the new Superia did not produce compound A. In a previous study (9), a compound A level for Sofnolime was found of approximately 38 ppm maximum with an sd of approximately 10 ppm. If Superia would produce 20 ppm less compound A (mean value) than Sofnolime, with the same sd, then six patients would be required to predict a significance level with 90% power (α error of 0.05) between groups for compound A production.
The anthropometric data of the 14 patients (7 in each group) are shown in Table 1. No difference in age, weight, height, or sex distribution was found between the two groups. During anesthesia, the overall PETco2 remained stable between the targeted values: 34.5 ± 2.4 mm Hg (mean ± sd) in the Superia group and 33.7 ± 2.5 mm Hg in the Sofnolime group (not statistically different between groups). Overall Spo2 values were (mean ± sd) 98% ± 1.6% in the Superia group and 98% ± 1.5% in the Sofnolime group, without a significant difference between groups. The ET sevoflurane concentrations for both groups during anesthesia are shown in Table 2 and were similar in both groups. All patients’ temperatures were between 35°C and 36°C in both groups.
Individual and median (range) inspiratory and expiratory compound A levels during anesthesia are shown in Figure 1 and Table 3, respectively. At all subsequent time points, compound A levels were higher than baseline levels in both groups. However, only very low levels of both inspiratory and expiratory compound A were found in the Superia group. Significantly higher levels of compound A were found in the Sofnolime group. As seen in Figure 1, the largest increase in compound A concentration occurred at the beginning of sevoflurane administration. Thereafter, the levels remained stable.
Individual and median (range) canister temperatures during anesthesia are shown in Figure 1 and Table 3, respectively. Similar temperature increases were seen in both groups. T°in increased from baseline values and remained stable afterward. For T°out, a similar trend was observed; however, because of a large range in baseline results, no significance level was obtained. As shown in Figure 2, a linear correlation was found between the inspiratory compound A levels and T°in. For Sofnolime, the Pearson correlation was 0.442, and for Superia it was 0.668. Both correlations were highly significant (P < 0.0001).
The cumulative consumption of liquid sevoflurane (mL) in both groups is shown in Table 4. An equal amount of sevoflurane was used in both groups.
It has been postulated that CO2 absorbents could be made safer, producing less carbon monoxide and compound A, if NaOH and KOH were excluded as components of the absorbent (5,11). In accordance with these suggestions, our results show that the newly developed NaOH- and KOH-free absorbent Superia is indeed free of significant compound A production and therefore complies with Kharash’s message (11). Rigid minimal-flow conditions (fresh gas flow rate of only 500 mL/min in the breathing circuit and almost complete rebreathing) and ET sevoflurane concentrations of 2.3%–2.5% (inspiratory concentrations of 2.7%–3.1%) were applied in our clinical practice set-up, with the aim that the production of compound A would be maximized. Although the US Food and Drug Administration does not recommend flow rates of <1 L/min, there is no such restriction in most European countries. The particular aspect of generation of carbon monoxide was not examined in this study, nor was the capacity to absorb CO2. In our clinical study using a small absorber of 750 mL capacity and an anesthesia duration of at least 2.5 h, no color indication of CO2 absorbent exhaustion appeared, nor was CO2 rebreathing evidenced on capnography.
Very few clinical studies are published on compound A production during modern clinical minimal-flow conditions. In a recent European study, applying Drägersorb as the CO2 absorbent, with a flow rate of 500 mL/min of oxygen/N2O and an inspiratory sevoflurane concentration of 2.1%, a mean maximal compound A concentration of 40 ± 9 ppm, but also a maximum concentration of 57 ppm, was reported (12). In this study with KOH-free Sofnolime, we noticed maximum median (range) inspiratory compound A concentrations of 25 (16) ppm at 60 minutes, with a maximum of 29.1 ppm. Sofnolime produced less compound A than the classic composition soda lime Drä- gersorb. However, in experimental closed-circuit conditions, we found larger (13) concentrations with Sofnolime than with Sodasorb (13). Recently, it has been reported that in low-flow conditions an ADU machine produced less compound A than other devices tested (Excel® 210 SE; Datex-Ohmeda and Cicero; Dräger, Lubeck, Germany) because of the small volume of the canister/soda lime (14). This might also have been an element why, in this study, with the small-volume canister, less compound A was found than in the Goeters et al. (12) study, which used a Cicero-EM machine (Dräger). It is also interesting that at the very beginning of minimal-flow anesthesia, small concentrations of compound A were found in all but one patient in the Sofnolime group without apparent reason. However, during the initial setup, automatic calibration, and leak-test procedure of the ADU ventilator, a very small amount of sevoflurane is introduced into the system, without traceable sevoflurane concentration afterward (technical bulletin, ref. TB ADV STO 00 015, December 1, 2000, Datex-Ohmeda). It might be that these small aliquots of sevoflurane already induced the production of compound A in the presence of Sofnolime. A similar rapid production of compound A with Sofnolime was also found previously in experimental in vitro conditions (13).
Differences between compound Ainsp and compound Aexp were found, suggesting uptake of compound A in the body. The small amount of compound A found in the Superia group was almost identical to that contained intrinsically in commercial sevoflurane (15).
An interesting aspect is also that in our study no difference was found for the canister temperatures between the two groups. The difference between T°in and T°out was not significant in either group. The temperatures recorded during our minimal-flow technique with a smaller-sized canister of 750 mL were not higher than those reported in the literature (16–18). A significant correlation was found between compound Ainsp and canister T°in for both absorbents. Because the tiny amount of compound A in the Superia group is well below 1 ppm, this correlation is of only theoretical consideration. The Pearson correlation of 0.442 found in the Sofnolime group shows that this absorbent tends to follow the previously reported relationship between canister temperature and generation of compound A (15).
The amount of liquid sevoflurane used to obtain an almost identical ET concentration in the two groups showed no statistical difference, suggesting that the two absorbents behave in the same way concerning eventual absorption. The liquid sevoflurane consumption was in the same range as the amount of 12–14 mL/h reported at a flow rate of 2 L/min and 1 minimum alveolar anesthetic concentration of sevoflurane (19).
In conclusion, the NaOH- and KOH-free CO2 absorbent Superia does not produce compound A, in contrast to the KOH-free Sofnolime.
1. Kandel L, Laster MJ, Eger EI II, et al. Nephrotoxicity in rats undergoing a one-hour exposure to compound A. Anesth Analg 1995; 81: 559–63.
2. Ebert TJ, Messana LD, Uhrich TD, Staacke TS. Absence of renal and hepatic toxicity after four hours of 1.25 minimum alveolar anesthetic concentration sevoflurane anesthesia in volunteers. Anesth Analg 1998; 86: 662–7.
3. Karash ED, Frink EJ Jr, Zager R, et al. Assessment of low-flow sevoflurane and isoflurane effects on renal function using sensitive markers of tubular toxicity. Anesthesiology 1997; 86: 1238–53.
4. Mazze RI, Jamison RL. Low-flow (1 l/min) sevoflurane: is it safe? Anesthesiology 1997; 86: 1225–7.
5. Neumann MA, Laster MJ, Weiskopf RB, et al. The elimination of sodium and potassium hydroxides from desiccated soda lime diminishes degradation of desflurane to carbon monoxide and sevoflurane to compound A but does not compromise carbon dioxide absorption. Anesth Analg 1999; 89: 768–73.
6. Stabernack CR, Brown R, Laster MJ, et al. Absorbents differ enormously in their capacity to produce compound A and carbon monoxide. Anesth Analg 2000; 90: 1428–35.
7. Ikeuchi Y, Bito H, Katoh T, Sato S. Quantification of the degradation products of sevoflurane using four brands of CO2
absorbent in a standard anesthetic circuit. J Anesth 2000; 14: 143–6.
8. Murray JM, Renfrew CW, Bedi A, et al. Amsorb: a new carbon dioxide absorbent for use in anesthetic breathing systems. Anesthesiology 1999; 91: 1342–8.
9. Versichelen LFM, Bouche MPLA, Rolly G, et al. Only carbon dioxide absorbents free of both NaOH and KOH prevent formation of compound A during in vitro closed-system sevoflurane: evaluation of five absorbents. Anesthesiology 2001; 95: 750–5.
10. Bito H, Sato S. Quantification of degradation product of sevoflurane (compound A) using a new CO2
absorbent (Amsorb). In: Book of abstracts, 12th World Congress of Anaesthesiology, Montreal, Canada, 2000. 4.4.21.
11. Kharash ED. Putting the brakes on anesthetic breakdown. Anesthesiology 1999; 91: 1192–4.
12. Goeters C, Reinhardt C, Gronau E, et al. Minimal flow sevoflurane and isoflurane anaesthesia and impact on renal function. Eur J Anaesthesiol 2001; 18: 43–50.
13. Versichelen L, Bouche MP, Struys M, et al. Compound A production from sevoflurane is not less when KOH-free absorbent is used in a closed-circuit lung model system. Br J Anaesth 2001; 86: 345–8.
14. Yamakage M, Kimura A, Chen X, et al. Production of compound A under low-flow anesthesia is affected by type of anesthetic machine. Can J Anaesth 2001; 48: 435–8.
15. Munday IT, Ward PM, Foden ND, et al. Sevoflurane degradation by soda lime in a circle breathing system. Anaesthesia 1996; 51: 622–6.
16. Liu J, Laster MJ, Eger EI II, Tahari S. Absorption and degradation of sevoflurane and isoflurane in a conventional anesthetic circuit. Anesth Analg 1991; 72: 785–9.
17. Frink EJ, Malan TP, Morgan SE, et al. Quantification of the degradation products of sevoflurane in two CO2
absorbants during low-flow anesthesia in surgical patients. Anesthesiology 1992; 77: 1064–9.
18. Bito H, Ikeda K. Long-duration, low-flow sevoflurane anesthesia using two carbon dioxide absorbents. Anesthesiology 1994; 81: 340–5.
19. Eger IE. Compound A: does it matter? Can J Anaesth 2001; 48: 427–30.
20. Hendrickx JFA, De Cooman S, Deloof T, et al. The ADU vaporizing unit: a new vaporizer. Anesth Analg 2001; 93: 391–5.
© 2002 International Anesthesia Research Society
21. Bouche MPLA, Van Bocxlaer JFP, Rolly G, et al. Quantitative determination of vapor phase compound A in sevoflurane anesthesia using gas chromatography-mass spectrometry. Clin Chem 2001; 47: 281–91.