This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (23) Citing Articles via Google Scholar Google Scholar Articles by Stabernack, C. R. Articles by Eger, E. I Search for Related Content PubMed PubMed Citation Articles by Stabernack, C. R. Articles by Eger, E. I, II

---

GENERAL ARTICLES

Absorbents Differ Enormously in Their Capacity to Produce Compound A and Carbon Monoxide

Caroline R. Stabernack, MD^*, Ronald Brown, BS^*, Michael J. Laster, DVM^*, Raphael Dudziak, PhD, MD, and Edmond I Eger, II, MD^*

*Department of Anesthesia, University of California, San Francisco, California; and Department of Anesthesiology, Johann Wolfgang Goethe University, Frankfurt, Germany

Address correspondence to Edmond I Eger, MD; Box 0464, University of California at San Francisco, San Francisco, CA 94143-0464.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Concern persists regarding the production of carbon monoxide (CO) and Compound A from the action of carbon dioxide (CO₂) absorbents on desflurane and sevoflurane, respectively. We tested the capacity of eight different absorbents with various base compositions to produce CO and Compound A. We delivered desflurane through desiccated absorbents, and sevoflurane through desiccated and moist absorbents, then measured the resulting concentrations of CO from the former and Compound A from the latter. We also tested the CO₂ absorbing capacity of each absorbent by using a model anesthetic system. We found that the presence of potassium hydroxide (KOH) and sodium hydroxide (NaOH) increased the production of CO from calcium hydroxide (Ca[OH]₂) but did not consistently affect production of Compound A. However, the effect of KOH versus NaOH was not consistent in its impact on CO production. Furthermore, the effect of KOH versus NaOH versus Ca(OH)₂ was inconsistent in its impact on Compound A production. Two absorbents (Amsorb^® [Armstrong Medica, Ltd, Coleraine, Northern Ireland], composed of Ca(OH)₂ plus 0.7% polyvinylpyrrolidine, calcium chloride, and calcium sulfate; and lithium hydroxide) produced dramatically lower concentrations of both CO and Compound A. Both produced minimal to no CO and only small concentrations of Compound A. The presence of polyvinylpyrrolidine, calcium chloride, and calcium sulfate in Amsorb^® appears to have suppressed the production of toxic products. All absorbents had an adequate CO₂ absorbing capacity greatest with lithium hydroxide.

Implications: Production of the toxic substances, carbon monoxide and Compound A, from anesthetic degradation by carbon dioxide absorbents, might be minimized by the use of one of two specific absorbents, Amsorb^® (Armstrong Medica, Ltd., Coleraine, Northern Ireland) (calcium hydroxide which also includes 0.7% polyvinylpyrrolidine, calcium chloride, and calcium sulfate) or lithium hydroxide.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Carbon dioxide (CO₂) absorbents are essential to the economical application of inhaled anesthetics, and thus, to the modern practice of anesthesia. However, their use also enables the production of toxic byproducts, such as carbon monoxide (CO) (from desflurane, enflurane, or isoflurane) and Compound A (from sevoflurane). Several studies have investigated the impact of absorbent composition on the production of CO and Compound A. The results of these studies (1,2) suggest that the production of CO and Compound A results from absorbents with higher percentages of potassium hydroxide (KOH) and, to a lesser extent, sodium hydroxide (NaOH) (1,2).

Proceeding from these results, we now present new information on production of CO and Compound A by several absorbents with various compositions. Our findings show it is possible to minimize production of CO and Compound A by replacing KOH and NaOH with calcium hydroxide (Ca[OH]₂ or by using lithium hydroxide (LiOH) alone. We find that such replacement or use does not compromise absorption of CO₂.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Absorbents were obtained from several sources and contained various bases (Table 1). All contained 12% to 17% water, except LiOH (1% water). For studies of desiccated absorbents, the water content from the normally hydrated absorbents was removed by placing precisely known (weighed) amounts of the absorbents in tared vessels through which we directed a flow of nitrogen (purity 99.998%) in an oven at approximately 80°C. The outflow from the vessel was attached to a vacuum. Drying continued to a constant vessel weight, a process taking approximately 24 h. All eight variants were tested as completely dry absorbents and as absorbents containing 12% to 17% water except for LiOH which was tested with 1% water (i.e., a total of 16 variants).

Model System
Desflurane Degradation to CO. As described previously (3), we sealed the open ends of barrels of 30-mL syringes with rubber stoppers, each pierced with a needle. Gases were directed through the needle. With the exception of LiOH, we placed an amount of dry absorbent in each barrel, approximately equaling the active components of 25 g of standard absorbent (e.g., for dry absorbent, we placed 21 g in each barrel). We placed only 15 g of dry LiOH in each barrel (the density of the LiOH was such that 15 g filled the barrel). The Luer lock at the exit from the syringe barrel was connected to Teflon tubing terminating in a three-way stopcock allowing sampling of exiting gas. Barrels were immersed in a constant temperature water bath at 45°C for all absorbents. In addition, we studied barrels containing LiOH in a water bath at 80°C, doing so because we found that this temperature was reached in a model circle absorption system to which CO₂ was added when LiOH was used, however, not when the other absorbents were used. The rubber stopper also was pierced with a temperature probe with the probe tip placed midabsorbent.
Desflurane-containing gas flowed through each syringe at 12.5 mL/min. Flow rates were calibrated by using a bubble flowmeter. Entering and exiting concentrations of desflurane and exiting concentrations of CO were measured by using gas chromatography (see below). At 5, 10, 15, 30, 60, 120, 180, and 240 min, we sampled and analyzed gases from the inflow and outflow of all syringes. All studies were performed in quadruplicate (four absorbent systems).

Sevoflurane Degradation to Compound A. Identical conditions were used to test the degradation of sevoflurane to Compound A except for substitution of 1.5% sevoflurane for desflurane, and except that we studied both desiccated and hydrated absorbents. We did not measure the capacity of absorbents to degrade sevoflurane to CO because previous results indicated such degradation does not occur (3).

CO and Anesthetic Analyses
For CO analysis, we used a thermal conductivity detector gas chromatograph (Gow-Mac 580; Bridgewater, NJ) equipped with a 5.8-m long, 0.22-mm internal diameter column containing a 5A washed molecular sieve maintained at a column temperature of 126°C and a detector temperature of 153°C with a helium carrier flow. Sensitivity permitted detection of <10 ppm CO. Calibration standards (99.9% purity) contained 1000 ppm CO. We demonstrated linearity of the gas chromatographic response over the entire range of concentrations studied (up to 60,000 ppm).

For anesthetic analyses, we used a Gow-Mac 580 flame-ionization detector gas chromatograph equipped with a 4.6-m long, 0.22-mm internal diameter column containing 10% SF96 on Chromosorb WHP (Alltech, Deerfield, IL) maintained at a column temperature of 144°C and a detector temperature of 172°C, by using a carrier flow of nitrogen. Linearity was obtained over the concentration ranges studied.

Capacity of Absorbents to Remove CO₂
The capacity of absorbents to remove CO₂ was determined by using a standard anesthetic machine and circuit. With the exception of LiOH, the lower of two absorbent canisters was filled with 1000 g fresh absorbent. The upper canister was left empty. We conducted two studies with LiOH. In one, the lower canister was filled with 500 g, and in the second, the lower and upper canisters were filled with a total of 1000 g. The circuit was assembled as in clinical practice, with the ventilator attached to the rebreathing bag port. A 3-L rebreathing bag connected to the patient Y-piece of the circuit acted as a compliant "lung." Oxygen at a flow rate of 0.5 L/min was delivered to the usual inflow port as the fresh gas inflow (i.e., this inflow rate approached that of a closed system). A flow rate of 300 mL/min CO₂ (flow calibrated with a bubble flowmeter) was directed into the "lung." The lung was ventilated 12 times per minute producing a minute ventilation of approximately 8 L/min. The initial end-tidal concentration of CO₂ was 30–35 mm Hg measured with an analyzer. Each study continued until the inspired CO₂ reached 30 mm Hg.

We calculated the percentage of absorbent exhausted at inspired CO₂ partial pressures of 5 and 30 mm Hg. This required a knowledge of the base content of the absorbent. In all cases except for Amsorb^® (Armstrong Medica, Ltd, Coleraine, Northern Ireland), this information was available from the manufacturer. For Amsorb^®, we determined the base capacity by grinding Amsorb^® granules with a mortar and pestle, then, dissolving a measured (weighed) amount of the ground Amsorb^® in distilled water. By using a pH meter, we titrated the solution to a pH of 7 with a known concentration of hydrochloric acid, performing the determination in triplicate.

Prof. H. Förster of Johann Wolfgang Goethe University, Frankfurt, Germany, suggested (oral communication, 1999) that in clinical practice, calcium chloride and calcium sulfate might absorb exhaled moisture and thereby decrease the capacity to absorb CO₂. We tested this possibility in an additional study in which we inserted a heated humidifier into the inspiratory limb of the anesthetic circuit described in the first paragraph of this section. As with the other studies of absorbent efficacy, this trial was performed in triplicate.

Values were presented as the average of the triplicate or quadruplicate determinations made at each time point. By using the trapezoid rule, we determined the area under the curve for the CO and Compound A concentrations produced over time. For statistical analysis, we performed multiple unpaired Student’s t-tests, correcting for multiple comparisons (Bonferroni method) by assuming statistical significance at P < 0.001.

	Results

Top Abstract Introduction Methods Results Discussion References

 
CO
Except for LiOH, all desiccated absorbents eliminated desflurane from the outflow for 10–30 min (Fig. 1). An appreciable concentration of desflurane appeared in the first outflow sample from LiOH. Desflurane output as a fraction of input at 240 min was between 0.95 and 1.0, except for Baralyme^® (Chemetron, St. Louis, MO) where it was 0.85. Temperatures in all absorbents remained within 1°C of the target temperatures of 45°C and 80°C. Output of desflurane for the study of LiOH at 80°C was not remarkably different from that at 45°C.

View larger version (20K): [in this window] [in a new window]   Figure 1. Desflurane, at a concentration of approximately 4.3% and a flow rate of 25 mL/min, was directed through approximately 21 g of desiccated absorbents (15 g of lithium hydroxide [LiOH]) at a temperature of 45°C (except for LiOH where an additional study at 80°C was done) and the outflow concentration of desflurane was measured. Except for LiOH, all desiccated absorbents eliminated desflurane (i.e., completely degraded the anesthetic) from the outflow for 10–30 min. Desflurane output as a fraction of input at 240 min was between 0.95 and 1.0, except for Baralyme® (Chemetron) where it was 0.85.

View larger version (23K): [in this window] [in a new window]   Figure 2. Under the circumstances outlined in the legend to Fig. 1, degradation of desflurane produced carbon monoxide (CO) with peak CO concentrations appearing in the outflow in 10–30 min and decreasing thereafter, to minimal concentrations with all but Baralyme® (Chemetron) by 240 min. The highest peak values were found with Baralyme® and Carbolime® (Chemetron), and soda lime, with lower values for Grace 1M, 2M, and 3M. Minimal or no CO was produced by Amsorb® (Armstrong Medica) and lithium hydroxide (LiOH).

View larger version (16K): [in this window] [in a new window]   Figure 3. Sevoflurane, at a concentration of approximately 1.5% and a flow rate of 25 mL/min, was directed through approximately 21 g of desiccated absorbents (15 g for lithium hydroxide [LiOH]) at a temperature of 45°C (except for LiOH where an additional study at 80°C was done) and the outflow concentration of sevoflurane was measured. Except for LiOH, all desiccated absorbents eliminated sevoflurane from the outflow for 15–240 min (Fig. 3). An appreciable concentration of sevoflurane appeared in the first outflow sample from LiOH. The order of appearance of sevoflurane from other absorbents was Amsorb® (Armstrong Medica), Grace 3M and 2M, and finally soda lime and Carbolime® (Chemetron). Both Baralyme® (Chemetron) and Grace 1M contained the greatest amounts of KOH and prevented the appearance of sevoflurane at any time. Despite this degradation of sevoflurane, only minimal concentrations of Compound A appeared in the outflow (data not shown).

View larger version (20K): [in this window] [in a new window]   Figure 4. Sevoflurane, at a concentration of approximately 1.5% and a flow rate of 25 mL/min, was directed through approximately 25 g of moist absorbents (15 g for lithium hydroxide [LiOH]) at a temperature of 45°C (except for LiOH where an additional study at 80°C was done) and the outflow concentration of sevoflurane was measured. For moist absorbents, sevoflurane degradation was greatest with Baralyme® (Chemetron) (i.e., the effluent concentration was smallest) and was least with Amsorb® (Armstrong Medica) and LiOH. The other absorbents produced results intermediate to those produced by Baralyme® (Chemetron) versus Amsorb® (Armstrong Medica) and LiOH.

	Discussion

Top Abstract Introduction Methods Results Discussion References

View larger version (21K): [in this window] [in a new window]   Figure 5. Under the circumstances outlined in the legend to Fig. 4, degradation of sevoflurane produced various concentrations of Compound A. Unlike the result for degradation of desflurane to CO (Fig. 2), the outflow concentrations of Compound A tended to be sustained. The highest peak Compound A values were reached in the outflow from Grace 1M and soda lime. In contrast Amsorb® (Armstrong Medica) and lithium hydroxide (LiOH) produced only small or negligible peak concentrations of Compound A. The other absorbents, including Baralyme® (Chemetron) (which produced the greatest degradation of sevoflurane—see Fig. 4) produced results intermediate to those produced by Grace 1M and soda lime versus Amsorb® (Armstrong Medica) and LiOH.

Similarly, there was a tendency for CO production to correlate with desflurane degradation. Baralyme^® produced the greatest degradation and produced the most CO, whereas Amsorb^® and LiOH produced the least degradation and produced the least amount of CO. However, as with sevoflurane’s degradation to Compound A, there were exceptions to this tendency. Despite producing greater degradation, Grace 2M produced less CO than did Carbolime^®, Grace 3M, and soda lime.

Likewise, the capacity to degrade anesthetics to toxic products was not as clearly related to the composition of the absorbents as we predicted from previous work (1,2). We had anticipated that absorbents without KOH, and, to a lesser extent NaOH, would produce the least CO and Compound A. Consistent with this notion, Grace 1M produced more CO and Compound A than Grace 3M; Grace 1M and 3M produced more CO than Grace 2M; and Baralyme^® produced more CO than Carbolime^® and soda lime (Tables 1–3). However, contradicting this notion, although Baralyme^® had the largest concentration of KOH, production of Compound A was less than that produced by Carbolime^® and soda lime, absorbents which had no KOH and had concentrations of NaOH less than the KOH concentration found in Baralyme^® (Tables 1–3). Similarly, Grace 2M produced more Compound A than Grace 3M. Furthermore, although Carbolime^® and soda lime have no KOH, they produced as much CO as did Grace 1M which contains 3% KOH as well as 1.5% NaOH. All of these absorbents have essentially the same Ca(OH)₂ content.

Thus, we would conclude that although there is a tendency suggesting a greater importance of KOH than NaOH to the degradation of anesthetics and production of CO and Compound A, the difference in the effect is not consistent or large. Indeed, it is not clear that either KOH or NaOH are important to the production of Compound A, because the results with Grace 2M (containing only Ca[OH]₂) do not differ appreciably from those with several absorbents containing KOH and NaOH in addition to Ca(OH)₂ (Table 3). The variability of this finding may relate to the fact that absorbents not only degrade sevoflurane to Compound A, they also degrade Compound A (4). Thus, the amount of Compound A appearing in the effluent gas is the result of the difference between the production of Compound A and its destruction, and this relationship is influenced in unpredictable ways by the presence of NaOH and KOH.

Similarly, the importance of KOH versus NaOH to the production of CO is not as consistent as we previously thought. Carbolime^®, soda lime, and Grace 1M produce the same average amount of CO (Table 2) however, only Grace 1M has KOH. Further, Grace 3M has the same composition (including NaOH) as Carbolime^® and soda lime; however, Grace 3M produces a lower average CO.

The most glaring difference among the absorbents not explained by base composition is found in a comparison of Amsorb^® and Grace 2M. The base composition of and base concentration in these absorbents are essentially identical, 75% to 85% Ca(OH)₂. It appears that the presence of 0.7% polyvinylpyrrolidine, calcium chloride, and calcium sulfate markedly decreases or eliminates anesthetic degradation and the conversion of anesthetics to CO or Compound A. Differences in the grain or tablet shape of the absorbent may provide an alternative explanation for these differences, indeed for the discrepancies for all absorbents that are not explained by the composition of the absorbents.

Since submission of this report, Murray et al. (5) have described the development and properties of Amsorb^®. Our results confirm their findings for desflurane and sevoflurane suggesting that Amsorb^® causes minimal or absent production of CO and Compound A by the degradation of these anesthetics. Similarly, they and we found that Amsorb^® was modestly less efficient in removing of CO₂.

Our results have possible clinical implications. Presuming it is desirable to minimize the production of CO and Compound A, Amsorb^® or LiOH might be preferred to the other absorbents. The decreased destruction, and thus, cost of anesthetic also might recommend the use of either Amsorb or LiOH relative to other absorbents. LiOH might be preferred over Amsorb^® because the capacity of LiOH to remove CO₂ is 3 times greater per kg (Table 4). Indeed, Amsorb^® is the least efficient of all tested absorbents in its capacity to remove CO₂. This might affect both acquisition costs and labor costs associated with the need to change absorbent. On the other hand, LiOH is much more corrosive than Ca(OH)₂ (Amsorb^®), and thus, one might have to take greater care in the handling of LiOH. This difference might mandate the use of prepackaged containers for LiOH. Cost considerations also might mandate the recycling of LiOH (e.g., Li₂CO₃ can be converted to LiOH by heating the Li₂CO₃ to 800°C in the presence of water). Recycling also might be desirable in that it would eliminate concerns regarding disposal of spent absorbent. The final clinical choice of absorbent may be dictated by the premium placed by the manufacturers of Amsorb^® and LiOH.

	Footnotes

 
Baxter Pharmaceutical Products (New Providence, NJ) donated desflurane and EIE, II is a paid consultant to Baxter PPI. Otherwise; this study was not funded by Baxter PPI.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Neumann MA, Laster MJ, Weiskopf RB, et al. The elimination of sodium and potassium hydroxides from desiccated soda lime diminishes degradation of desflurane to CO and sevoflurane to compound a but does not compromise carbon dioxide absorption. Anesth Analg 1999;89:768–73.[Abstract/Free Full Text]
2. Förster H, Dudziak R. Über die ursachen der reaktion zwischen trockenem atemkalk und halogenierten inhalationsanästhetika. Anaesthesist 1997;46:1054–63.[Web of Science][Medline]
3. Fang ZX, Eger EI II, Laster MJ, et al.. Carbon monoxide production from degradation of desflurane, enflurane, isoflurane, halothane, and sevoflurane by soda lime and Baralyme^®. Anesth Analg 1995;80:1187–93.[Abstract]
4. Fang ZX, Kandel L, Laster MJ, et al. Factors affecting production of compound a from the interaction of sevoflurane with Baralyme^® and soda lime. Anesth Analg 1996;82:775–81.[Abstract]
5. 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.[Web of Science][Medline]

This article has been cited by other articles:

				M. J. Laster and E. I. Eger II Temperatures in Soda Lime During Degradation of Desflurane, Isoflurane, and Sevoflurane by Desiccated Soda Lime Anesth. Analg., September 1, 2005; 101(3): 753 - 757. [Abstract] [Full Text] [PDF]

				M. Laster, P. Roth, and E. I Eger II Fires from the Interaction of Anesthetics with Desiccated Absorbent Anesth. Analg., September 1, 2004; 99(3): 769 - 774. [Abstract] [Full Text] [PDF]

				E. Knolle, W. Linert, and H. Gilly The Color Change in CO2 Absorbents On Drying: An In Vitro Study Using Moisture Analysis Anesth. Analg., July 1, 2003; 97(1): 151 - 155. [Abstract] [Full Text] [PDF]

				M.-P. L. A. Bouche, L. F. M. Versichelen, M. M. R. F. Struys, J. F. P. Van Bocxlaer, A. P. De Leenheer, E. P. Mortier, and G. Rolly No Compound A Formation with Superia(R) During Minimal-Flow Sevoflurane Anesthesia: A Comparison with Sofnolime(R) Anesth. Analg., December 1, 2002; 95(6): 1680 - 1685. [Abstract] [Full Text] [PDF]

				E. Knolle, G. Heinze, and H. Gilly Small Carbon Monoxide Formation in Absorbents Does Not Correlate with Small Carbon Dioxide Absorption Anesth. Analg., September 1, 2002; 95(3): 650 - 655. [Abstract] [Full Text] [PDF]

				H. Higuchi, Y. Adachi, S. Arimura, M. Kanno, and T. Satoh The Carbon Dioxide Absorption Capacity of Amsorb(R) is Half That of Soda Lime Anesth. Analg., July 1, 2001; 93(1): 221 - 225. [Abstract] [Full Text] [PDF]

				R. I. Mazze and H. Higuchi Composition of CO2 Absorbents Anesth. Analg., May 1, 2001; 92(5): 1356 - 1357. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (23) Citing Articles via Google Scholar Google Scholar Articles by Stabernack, C. R. Articles by Eger, E. I Search for Related Content PubMed PubMed Citation Articles by Stabernack, C. R. Articles by Eger, E. I, II