JOURNAL HOME CME HOME THIS MONTH PAST ISSUES ETOC COLLECTIONS
AUTHORS REVIEWERS EDITORIAL BOARD FEEDBACK RSS HELP
A&A International Anesthesia Research Society
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a colleague
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (15)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Holak, E. J.
Right arrow Articles by Woehlck, H. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Holak, E. J.
Right arrow Articles by Woehlck, H. J.
Related Collections
Right arrow Equipment
Right arrow Technology
Right arrow Pharmacology

Anesth Analg 2003;96:757-764
© 2003 International Anesthesia Research Society

TECHNOLOGY, COMPUTING, AND SIMULATION

Carbon Monoxide Production from Sevoflurane Breakdown: Modeling of Exposures Under Clinical Conditions

Elena J. Holak, MD PharmD*, David A. Mei, MD PhD*, Marshall B. Dunning, III, PhD{dagger}, Rao Gundamraj, MD*, Randa Noseir, MD*, Lu Zhang, MD PhD*, and Harvey J. Woehlck, MD*

Department of *Anesthesiology and {dagger}Pulmonary and Critical Care Medicine, Medical College of Wisconsin, Milwaukee

Address correspondence and reprint requests to Harvey Woehlck, MD, Department of Anesthesiology, Froedtert Memorial Lutheran Hospital West, 9200 W. Wisconsin Ave., Milwaukee, WI 53226. Address e-mail to hwoehlck{at}mcw.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Isoflurane, enflurane, sevoflurane, and especially desflurane produce carbon monoxide (CO) during reaction with desiccated absorbents. Of these, sevoflurane is the least studied. We investigated the dependence of CO production from sevoflurane on absorbent temperature, minute ventilation (VE), and fresh gas flow rates. We measured absorbent temperature and in vitro CO concentrations when desiccated Baralyme reacted with 1 minimum alveolar anesthetic concentration of (2.1%) sevoflurane at 2.3-, 5.0-, and 10.0-L VE. Mathematical modeling of carboxyhemoglobin concentrations was performed using an existing iterative method. Rapid breakdown of sevoflurane prevented the attainment of 1 minimum alveolar anesthetic concentration with low fresh gas flow rates. CO concentrations increased with VE and with absorbent temperatures exceeding 80°C, but concentrations decreased with higher fresh gas flow rates. Average CO concentrations were 150 and 600 ppm at 2.3- and 5.0-L VE; however, at 10 L, over 11,000 ppm of CO were produced followed by an explosion and fire. Methanol and formaldehyde were present and may have contributed to the flammable mixture but were not quantitated. Mathematical modeling of exposures indicates that in average cases, only patients <=25 kg, or severely anemic patients, are at risk of carboxyhemoglobin concentrations >10% during the first 60 min of anesthesia.

IMPLICATIONS: Sevoflurane breakdown in desiccated absorbents is expected to result in only mild carbon monoxide (CO) exposure. Completely dry absorbent and high minute ventilation rates may degrade sevoflurane to extremely large CO concentrations. Serious CO poisoning or spontaneous ignition of flammable gases within the breathing circuit are possible in extreme circumstances.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Isoflurane, enflurane, and particularly desflurane are known to produce carbon monoxide (CO) in contact with desiccated CO2 absorbents (1). With the absorbent desiccated for 48 h, Frink et al. (2) achieved lethal carboxyhemoglobin (COHb) concentrations in a porcine model. However, before sevoflurane-related CO exposures had been reported in pediatric patients (3), the existing studies on sevoflurane breakdown suggested that it could produce only insignificant concentrations of CO. Sevoflurane is one of the most easily degraded inhaled anesthetics and undergoes measurable degradation, even in normally hydrated soda lime, although its rate of destruction is known to be faster at smaller absorbent water contents and at increased temperatures (4). A series of sevoflurane degradation products has been identified (5,6), but measurements of CO were not reported in these studies.

The initial studies that focused on CO production from sevoflurane breakdown found only trivial concentrations of CO when reaction was performed at room temperature (7) or in small reactors that were temperature controlled to <50°C, the usual maximum temperature of hydrated absorbents in clinical situations (1). However, reaction of anesthetics with desiccated absorbents is exothermic (8), producing temperatures in excess of this usual range. A peak absorbent temperature >400°C occurred during an experimental attempt to maintain a 1.5 minimum alveolar anesthetic concentration (MAC) of sevoflurane using desiccated absorbent, but CO concentrations were not reported (9). Other studies demonstrated that clinically significant concentrations of CO were produced (10) when sevoflurane breakdown occurred at temperatures >100°C (11,12). Therefore, there may be a temperature threshold in the vicinity of 100°C for production of a clinically significant concentration of CO from sevoflurane.

The present study was designed to quantify CO production under an approximation of clinical conditions and to predict the severity of patient exposures via mathematical modeling. This study also sought to identify clinically useful warnings of otherwise undetectable CO exposures.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
In vitro experiments were performed with an anesthesia machine (Narkomed 2, North American Draeger, Telford, PA), which was equipped with 2 canisters of 1180 g (predesiccation weight) of barium hydroxide lime (Baralyme, Allied Healthcare Products, St Louis, MO). Sevoflurane was administered via a Dräger Vapor 2000 vaporizer (Dräger Medizintechnik GmbH, Lübeck, Germany). A polyethylene breathing circuit was completed with a latex breathing bag at the patient y-connector to substitute for a patient. No human or animal subjects were used. A cannula was inserted into the latex bag for the addition of CO2 to simulate respiration. This method produces a capnogram when measured at the patient y-connector that is similar to actual human ventilation and triggers clinical gas monitors to display inspiratory and expiratory concentrations of anesthetic gases. The fresh gas flow (FGF) rate was selected from nominal rates of 1, 2, 4, or 6 L/min of O2. To approximate the conditions for small adults, the ventilator was set to deliver 10 breaths/min at 500-mL tidal volume for a total of 5.0-L minute ventilation (VE), with the administration of 200 mL/min of CO2. The actual tidal volumes were measured to be within 10% of these nominal settings. To model smaller pediatric patients, experiments were performed with 167-mL tidal volume, 14 breaths/min, and 100 mL/min of CO2 (2.3-L VE). To approximate conditions for large adults, experiments were planned with 1000-mL tidal volume and 10 breaths/min (10-L VE) with 350 mL/min of CO2. The inspiratory to expiratory ratio was 1:2 in all cases.

Control experiments were performed with normally hydrated absorbent, measured to be approximately 17% to 19% water calculated by weight loss on desiccation. The absorbent used in this study was desiccated by FGF from the bottom up (13). Absorbent was desiccated for 24 or 66 h by the flow of 10 L/min of O2 to simulate what would happen if the flow of dry medical oxygen were left on for a day or a weekend in an anesthesia machine. The water contents were within the range of those previously reported at similar desiccation times (14). To determine the maximum CO production possible from sevoflurane, completely desiccated absorbent was prepared by the flow of >15 L/min of O2 for >=1 wk, which resulted in constant weight (±0.2 g per 1180-g canister) for the last 24 h of desiccation.

Infrared measurements of sevoflurane were performed using a clinical gas monitor (Smart Anesthesia Multi-gas Module, General Electric-Marquette Medical Systems, Milwaukee, WI). Gas was sampled at the patient y-connector. Monitor exhaust was returned into the expiratory limb of the breathing circuit to avoid dilutional effects of sample withdrawal. Vaporizer settings were adjusted to maintain the target end-tidal concentration of 2.1%, representing a 1 MAC concentration in adults (15). CO was measured by gas chromatography, with a lower limit of detection of 15 ppm of CO (14). Temperature was monitored at the center of each canister with custom designed probes with a range of 0°C to approximately 110°C. Where indicated, gas composition was also determined with combined gas chromatography or mass spectrometry, as previously described (16).

At time 0, the CO2 and sevoflurane were administered, and ventilation was initiated. At 5-min intervals, measurements were taken of inspiratory and expiratory CO concentrations and absorbent temperature. Experiments were terminated at 60 min. One experiment with completely desiccated absorbent was allowed to continue for 120 min to estimate the duration of reaction with 4 L/min of FGF and 5.0-L VE.

Combinations of absorbent desiccation and FGF rates were identified, which resulted in the ability to maintain 1 MAC end-tidal sevoflurane concentrations in the absence of a patient. The mean CO data from these conditions were applied to a mathematical model to calculate COHb versus time. COHb was used as a clinically relevant indicator of patient exposure severity and was calculated based on the production data and patient size. In brief, the iterative model uses a spreadsheet to calculate COHb based on a previous model developed by Coburn et al. (17) that uses equilibrium and diffusion constraints and then recalculates these values after accounting for blood and alveolar gas CO concentrations (18,19). This initial model was based on environmental exposures, where the inspiratory concentrations do not change because the environmental reservoir of CO is essentially infinite compared with the patient absorption capacity. The rebreathing model compensates for the finite quantity of CO produced in an anesthesia machine by performing a mass balance to recalculate the CO remaining in the gas phase after patient absorption over very small time increments. It simultaneously incorporates incremental CO production, CO removal by fresh gas, and rebreathing of previously expired CO based on FGF rates, dead space, and VE. This model was minimally modified from the published version, which contained approximations consistent with average sized adult variables (20), and the current model better accommodates the small tidal volumes required for pediatric patients and the large FGF rates required to compensate for the rapid destruction of sevoflurane.

An unpaired t-test with the Bonferroni correction for multiple comparisons was used to compare CO concentrations of appropriate groups. The values are reported as mean ± SD.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Control experiments were performed at 5.0-L VE using fresh, normally hydrated Baralyme. After a brief loading phase, the sevoflurane concentration increased to 2.1% in the breathing circuit at high vaporizer settings. Thereafter, the vaporizer dial could be adjusted to values between 2.0% and 2.5% to maintain this end-tidal concentration at all tested FGF rates. No CO was detected using normally hydrated absorbents.

Partially desiccated or completely desiccated absorbents produced measurable sevoflurane breakdown, which required increased FGF rates to produce 1 MAC end-tidal concentrations of sevoflurane, even in the absence of uptake by a patient. With 5.0-L VE and Baralyme desiccated for >=24 h, 1 MAC end-tidal sevoflurane concentrations could not be maintained using <=2 L/min of FGF, despite maximal vaporizer output (8%). At 2.3-L VE, 2 L/min of FGF sufficed to maintain 1 MAC sevoflurane concentrations provided that the vaporizer was set at or near 8%; however, 1 L/min of FGF could not produce 1 MAC. No data are reported for conditions that were unable to produce 1 MAC because they did not represent clinically useful situations for the use of sevoflurane as the sole anesthetic. The average CO concentrations for the 60-min experiments and the associated conditions are shown in Table 1.


View this table:
[in this window]
[in a new window]
  		Table 1. The Mean Carbon Monoxide (CO) Concentrations Between 15 and 60 Min are Shown in parts per million ± SD for Each Condition.
 
After determining that 6 L/min of FGF were required to produce 1 MAC at 10-L VE, an experiment was performed under these conditions using completely desiccated absorbent. The upper canister of absorbent was heated to >110°C and turned blue in <10 min. The outside of the upper absorbent canister became too hot to touch within 15 min, although the metal pipe leading away from the bottom of the absorber did not feel warm until the bottom canister was heated to 80°C at 25 min into the experiment. An exponential increase in CO concentration began at 30 min, as shown in Figure 1, and exceeded 11,000 ppm at 50 min. Temperature increased beyond the range of the temperature monitors, and a glass-mercury thermometer held against the outside of the upper absorbent canister registered 137°C at 20 min and >200°C at 45 min. At 53 min into the experiment, an audible explosion occurred, accompanied by a flash of light in the breathing circuit. The rubber ventilator bellows and the rubber seal between the two absorbent canisters were ignited by the explosion and burned. Gas chromatography and mass spectrometry analysis of the gas composition measured at the patient connector just before the ignition revealed the presence of formaldehyde and methanol. Because the gas sampling methodology was intended for noncondensing gases like CO, it was impossible to quantitate the concentrations of volatile liquids like methanol or to confirm the presence of compounds that are poorly volatile at room temperature. Fire safety considerations precluded further experiments at 10-L VE. At no time did the infrared gas monitor indicate multiple anesthetics or the wrong anesthetic when CO was present. At the end of this experiment, the blue color had disappeared from the absorbent.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 1. Carbon monoxide (CO) concentrations are plotted against time for experiments conducted with completely desiccated absorbent and the lowest fresh gas glow (FGF) that attained 1 minimum alveolar anesthetic concentration (MAC) (Table 1). The middle curve represents 5-L minute ventilation (VE), and the lower curve represents 2.3-L VE. The upper curve represents the single experiment before the explosion and fire at 10-L VE. Error bars are omitted for clarity.

 
In all experiments using desiccated absorbent, a blue color developed throughout much of the upper canister except in those desiccated for only 24 h where only a faint color change occurred in the lower canister. At 4 L/min of FGF and 5.0-L VE, a sudden decrease in sevoflurane breakdown occurred at 50 to 60 min for Baralyme desiccated for 24 and 66 h and at 100 to 120 min with completely desiccated Baralyme. The CO concentrations are plotted against peak absorbent temperatures in Figure 2. Despite the high absorbent temperatures, the metal parts of the breathing circuit did not feel warm to the touch until >30 min into each experiment. In some cases at 5-L VE and in all cases at 2.3-L VE, the metal parts never felt warm to the touch. The sides of the canisters became warm to the touch, but this was less evident when prepackaged inserts containing Baralyme were used.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 2. Carbon monoxide (CO) concentrations in parts per million (ppm) are plotted against absorbent temperature measured in the center of the hottest canister. Most clinically relevant CO concentrations do not occur until absorbent temperature exceeds 80°C. The single experiment at 10-L minute ventilation (VE) is excluded from this plot because the data would be off scale.

 
Mathematical relationships involving CO production were derived. CO production increased with VE, as shown in Table 1. Inspiratory CO concentrations decreased with larger FGF provided that 1 MAC sevoflurane concentrations were maintained. CO production data and mathematical modeling data are summarized in Figure 3 for 10-kg patients and Figure 4 for larger patients. Higher FGF rates are required to maintain 1 MAC at higher VE when the absorbent is desiccated. Regardless of the degree of desiccation of the absorbent, the production of CO is related to the delivery of sevoflurane to the breathing circuit in the following exponential equation: equation




View larger version (24K):
[in this window]
[in a new window]
  		Figure 3. Carboxyhemoglobin (COHb) concentrations are plotted against time for the first 60 min of an anesthetic with completely desiccated Baralyme and a 10-kg patient, 40% inspired oxygen, and a 35% hematocrit. Curves labeled as anemic represent a hematocrit of 20%. Note the difference in predicted COHb concentration between carbon monoxide (CO) production data obtained at 2.3-L minute ventilation (VE) and those data obtained at 5-L VE, which would represent significant hyperventilation in this size patient.

 


View larger version (27K):
[in this window]
[in a new window]
  		Figure 4. Carboxyhemoglobin (COHb) concentrations are plotted against time for the first 60 min of an anesthetic with completely desiccated Baralyme, 5-L minute ventilation (VE). Twenty-five-, 50-, and 100-kg patients are represented at 42% hematocrit unless indicated as anemic (20% hematocrit).

 
Where L CO = the liters CO produced = ((FGF rate x duration) + breathing circuit volume) x ppm CO x 10-6 and L sevo = the liters sevoflurane gas delivered in fresh gas = FGF rate x duration x %sevoflurane vaporizer setting/100. All gases are measured at room temperature and pressure.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
After an initial loading phase under control conditions using normally hydrated absorbent, no measurable anesthetic concentration differences were evident between the breathing circuit and fresh gas to the limits of measurement (0.1%). We found no clinically significant degree of anesthetic breakdown or CO production with normally hydrated absorbents. No attempt was made to measure Compounds A-E or to calculate adsorption of sevoflurane.

When CO production occurred at 5.0-L VE, higher FGF rates resulted in smaller CO concentrations, suggesting dilution of reacted, rebreathed gas by fresh unreacted gas. This relationship was not sought for other conditions, and only the smallest FGF rate that allowed maintenance of 1 MAC was reported.

Difluoromethyl ether anesthetics (isoflurane, enflurane, and desflurane) produced large amounts of CO before temperatures exceeded 50°C (14). The data from sevoflurane suggest that 80°C may be the temperature threshold for a clinically relevant rate of CO production. This finding resembles the breakdown of hexafluoroisopropanol, a fragment of sevoflurane, which reacts with dry absorbent to produce CO at 70°C but not lower temperatures (21). Partially desiccated absorbents produced decreasing CO concentrations by the end of the first hour with isoflurane, desflurane, (14) and sevoflurane. When completely desiccated absorbents were used, desflurane also produced peak CO concentrations in the first 30 minutes and decreased thereafter. In contrast, Figure 1 shows that CO concentrations from sevoflurane increased steadily throughout the first 60 minutes with completely desiccated absorbent. As a result, the total CO production is not identified in this study. Extreme heating, as occurred during the experiment at 10-L VE, was associated with CO concentrations exceeding 11,000 ppm, the largest yet reported from sevoflurane breakdown, and in the range previously associated only with desflurane breakdown.

The dependence of CO production on water content in the clinical setting is different for sevoflurane than for other volatile anesthetics. The COHb concentration and the quantity of CO produced by the reaction of isoflurane, enflurane, and desflurane with desiccated absorbents are highly dependent on the absorbent water content (1,2,4,14). As an example, desflurane produced roughly 10 times more CO with completely desiccated absorbent than absorbent desiccated for only 24 hours (14). In contrast, sevoflurane, at 5.0-L VE, and either 24 hours of absorbent desiccation or complete desiccation resulted in similar average CO concentrations during the first 60 minutes, although the duration of CO production was approximately twice as long with completely desiccated absorbent.

The small CO production at 66 hours of desiccation can easily be explained in terms of the dependence of CO production on absorbent temperature. With 24 hours of absorbent desiccation, temperatures >80°C occurred exclusively in the bottom canister. With completely desiccated absorbent, temperatures >80°C occurred sequentially in the top and then the bottom canister, presumably as a reaction zone moved down through the absorbent. With 66 hours of absorbent desiccation, some reaction occurred in both canisters, and a lesser but simultaneous temperature increase occurred in both canisters. This temperature may not have been sufficient for optimal CO production. One should be careful in generalizing that relatively small CO production results after 66 hours of desiccation because the thermal increase in different anesthesia machines or with only one canister of absorbent may be highly dependent on factors that cannot be adequately predicted by this study.

There was an exponential relationship between CO production and the delivered dose of sevoflurane, even when the dryness of the absorbent was ignored. Higher VE presumably increased the contact of sevoflurane with the absorbent and resulted in more chemical reaction. The dryer the absorbent, the more the breakdown occurred. Both of these factors resulted in a larger required mass of sevoflurane to be delivered to the circuit to maintain target concentrations, making the delivered dose of sevoflurane a dependent variable but one that can be identified by the anesthesiologist. More chemical reaction resulted in higher temperatures, which facilitated CO production because of the temperature dependency of this reaction. Ideally, comparisons should be made between a far greater number of experiments to identify confounding factors such as loss of unreacted sevoflurane at high FGF rates through the exhaust despite the lack of breakdown in wet absorbent. In clinical settings, patient absorption would also impact this mathematical relationship.

The inability to produce at least 1 MAC sevoflurane concentrations with low FGF could be a clinically useful indicator of sevoflurane breakdown, although patient uptake must also be considered when making this determination during an actual anesthetic. The requirement for high vaporizer settings and FGF rates in this study are very similar to case reports of human CO exposures from sevoflurane breakdown (3). Because abnormally high delivery rates of sevoflurane can be identified by the anesthesiologist via high FGF rates and vaporizer settings, this by itself may be a reasonable warning of sevoflurane breakdown and the potential for CO production. In clinical practice, it may be nearly impossible to predict exact settings at which a warning of breakdown should be inferred without detailed calculations because of the absorption of sevoflurane by a patient.

Previous studies demonstrated that CO production resulting from the breakdown of isoflurane and desflurane was accompanied by trifluoromethane production, which interfered with anesthetic gas monitoring. The resulting display of mixed anesthetics or the wrong anesthetic provided a potential warning of CO production by suitable gas monitors (16,22,23). During the production of CO by sevoflurane breakdown, no interference with anesthetic gas monitoring occurred, and no suitable gas species were identified by gas chromatography or mass spectrometry. There is data that suggest that methanol or other similar compounds may only be produced in extreme situations, such as that which resulted in fire. Therefore, indirect warning of CO production via clinical gas monitors is not possible during sevoflurane breakdown.

If absorbent heating is noticed to be extreme, sevoflurane breakdown is likely, but thermal insulation may cause the external surface of the absorbent canister to remain only warm to the touch despite CO production and high internal temperatures. Heating of the absorbent itself seems to be a requirement for CO production. Temperature monitoring of the absorbent may be a reasonable way to identify sevoflurane breakdown with inexpensive existing technology.

The rapid appearance of blue coloration in the absorbent may be another indicator of CO production. However, exhaustion may also produce a blue coloration and renders this finding nonspecific. The blue coloration may be transitory because the extreme temperatures in the experiment at 10-L VE seemed to destroy the ethyl violet. Both false positives and false negatives will result from the use of blue coloration as an indicator of CO production.

Of potential concern is that we present the first report of spontaneous combustion in a breathing circuit using modern nonflammable inhaled anesthetics. Sevoflurane, although not flammable at clinical concentrations, is flammable at a concentration of 11% in oxygen (24). We used infrared monitoring to control sevoflurane concentrations at 2.1% in the breathing circuit; therefore, sevoflurane could not have been the sole fuel. Formaldehyde, methanol, and formate have been identified as intermediates in the chemical reaction of sevoflurane (6) (J. Baum, Associate Professor of Anaesthesiology, Medical Faculty of the University of Muenster, Germany, personal communication, 2002), and we detected two of these just before ignition. The concentration of methanol can far exceed the concentration of other products (25). We suggest that methanol, formaldehyde, CO, and sevoflurane all contributed to the development of a flammable gas mixture. Formaldehyde gas spontaneously ignites at 300°C (26), a temperature that has been exceeded in similar experiments (9). It is possible that explosion and fire have never been reported during clinical anesthetics because patient absorption of these components may reduce their concentrations below the lower limit of flammability. However, one patient in a published report of sevoflurane breakdown developed significant impairment of pulmonary gas exchange for no apparent reason requiring prolonged postoperative ventilation (3) (J. Baum, personal communication, 2002). Breakdown products such as formaldehyde may have caused pulmonary injury in these patients and may be more harmful than the CO produced by sevoflurane breakdown.

No cases of severe CO poisoning have been reported in large adults. Therefore, the CO production data from 5-L VE were used to model 100-kg patients. We recognize that this VE would probably result in hypoventilation in a clinical setting, but it is more representative of reported clinical episodes of CO exposures. Mathematical modeling predicts that patients who possess small hemoglobin quantities, including anemia and small patient size, result in more severe exposures. However, the small CO production at proportionately small VE is expected to reduce the severity of CO exposure. Figure 3 shows that in 10-kg patients, 5.0-L VE was required to exceed 10% COHb, but this VE would hyperventilate a 10-kg patient. Figure 4 shows that larger patients also are unlikely to attain COHb concentrations over 10% despite the larger CO production at higher VE. Excluding exceptional circumstances, our data suggest that CO poisoning by sevoflurane should only be a concern where high VE rates are used in proportion to the patient size. Comparison of these data to COHb concentrations reported in the European literature prompts one to speculate that European anesthesia machines, which allow anesthetic flow through the absorbent in the absence of rebreathing, may have a higher risk of CO exposures than anesthesia machines in which FGF enters the inspiratory limb directly. Some machines with this gas flow design are now widely available and may become popular in the North American market.

Some of the limitations of our study are as follows: Measurements were performed for only one hour, although CO concentrations increased throughout this time. Clinical exposures may become more severe at longer durations because the single experiment of two hours in duration showed increases in CO concentration for most of the second hour. CO2 input to the circuit was chosen to be proportional to the VE to simulate ventilation of appropriately sized patients. Because CO production is dependent on absorbent temperature, the absorbent heating from CO2 absorption may have contributed to a greater temperature increase and indirectly affected the CO production. Although CO production increased with VE, the minimum FGF rate that allowed maintenance of 1 MAC sevoflurane concentrations also increased with VE. Therefore, these variables are linked, and individual effects of each variable on CO production cannot be distinguished in this study. Limitations of the mathematical model have been described (20).

In summary, we conclude that specific requirements of FGF rate and VE are required for CO production from the reaction of sevoflurane and desiccated CO2 absorbents. In most cases, the quantity of CO produced is sufficient to poison only children and small adults. However, in extreme cases with completely desiccated absorbents, large quantities of CO may be produced, and flammable gases may be produced and ignited. The production of respiratory irritants deserves further study. In the absence of specific monitors for CO, the extreme heating of the CO2 absorbent remains a useful clinical warning. An inability to maintain 1 MAC or larger sevoflurane concentrations unless unusually high FGF rates and vaporizer concentrations are used should also alert the anesthesiologist to the potential of sevoflurane breakdown.


   Footnotes
 
The senior author has applied for a patent for inhibitors of CO formation from desiccated absorbents.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Fang ZX, Eger II EI, 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]
  2. Frink EJ, Nogami WM, Morgan SE, Salmon RC. High carboxyhemoglobin concentrations occur in swine during desflurane anesthesia in the presence of partially dried carbon dioxide absorbents. Anesthesiology 1997; 87: 308–16.[ISI][Medline]
  3. Baum J, Sitte T, Straub JM, et al. Die reaktion von sevofluran mit trockenem atemkalk: überlegungen anläßlich eines aktuellen zwischenfalls. Anasthesiol Intensivmedizin 1998; 39: 11–6.
  4. Strum D, Eger II EI. The degradation, absorption, and solubility of volatile anesthetics in soda lime depend on water content. Anesth Analg 1994; 78: 340–8.[Medline]
  5. Morio M, Fujii K, Satoh N, et al. Reaction of sevoflurane and its degradation products with soda lime: toxicity of the byproducts [see comments]. Anesthesiology 1992; 77: 1155–64.[ISI][Medline]
  6. Hanaki C, Fujii K, Morio M, Tashima T. Decomposition of sevoflurane by sodalime. Hiroshima J Med Sci 1987; 36: 61–7.[Medline]
  7. Baxter PJ, Garton K, Kharasch ED. Mechanistic aspects of carbon monoxide formation from volatile anesthetics. Anesthesiology 1998; 89: 929–41.[ISI][Medline]
  8. Strauß JM, Baum J, Sumpelmann R, et al. Zersetzung fon halothan, enfluran und isofluran an trockenem atemkalk zu kohlenmonoxid. Anaesthesist 1996; 45: 798–801.[Medline]
  9. Steffey EP, Laster MJ, Ionescu P, et al. Dehydration of Baralyme increases compound A resulting from sevoflurane degradation in a standard anesthetic circuit used to anesthetize swine. Anesth Analg 1997; 85: 1382–6.[Abstract]
  10. Wissing H, Kuhn I, Warnken U, Dudziak R. Carbon monoxide production from desflurane, enflurane, halothane, isoflurane and sevoflurane with dry soda lime. Anesthesiology 2001; 95: 1205–12.[ISI][Medline]
  11. Förster H, Dudziak R. Über die ursachen der reaktion zwischen trockenem atemkalk und halogenierten inhalationsanaesthetika. Anaesthesist 1997; 46: 1054–63.[ISI][Medline]
  12. Wissing H, Kuhn I, Dudziak R. Zur temperaturentwicklung von inhalationsanästhetika auf trockenem atemkalk. Anaesthesist 1997; 46: 1064–70.[ISI][Medline]
  13. Woehlck HJ, Dunning MBIII, Connolly LA. Reduction in the incidence of carbon monoxide exposures in humans undergoing general anesthesia. Anesthesiology 1997; 87: 228–34.[ISI][Medline]
  14. Woehlck HJ, Dunning MBIII, Raza T, et al. Physical factors affecting the production of carbon monoxide from anesthetic breakdown. Anesthesiology 2001; 94: 453–6.[Medline]
  15. Abbott Laboratories NC, IL 60064. USA: ultane sevoflurane, volatile liquid for inhalation. Package insert 1999 June, 1999:1–18.
  16. Woehlck HJ, Dunning MBIII, Nithipatikom K, et al. Mass spectrometry provides warning of carbon monoxide exposure via trifluoromethane. Anesthesiology 1996; 84: 1489–93.[ISI][Medline]
  17. Coburn RF, Forster RE, Kane PB. Considerations of the physiological variables that determine the blood carboxyhemoglobin concentration in man. J Clin Invest 1965; 44: 1899–910.
  18. Peterson JE, Stewart RD. Absorption and elimination of carbon monoxide by inactive young men. Arch Environ Health 1970; 21: 165–71.[ISI][Medline]
  19. Peterson JE, Stewart RD. Predicting the carboxyhemoglobin levels resulting from carbon monoxide exposures. J Appl Phys 1975; 39: 633–8.[Abstract/Free Full Text]
  20. Woehlck HJ, Mei D, Dunning MBIII, Ruiz F. Mathematical modeling of carbon monoxide exposures from anesthetic breakdown. Anesthesiology 2001; 94: 457–60.[Medline]
  21. Warnken UH, Wissing H, Förster H. Sevoflurane and dry lime: is the carbon monoxide formed a reaction product from hexafluoroisopropanol? Anesthesiology 2000; 93: A1165.
  22. Woehlck HJ, Dunning MBIII, Gandhi S, et al. Indirect detection of intraoperative carbon monoxide exposure by mass spectrometry during isoflurane anesthesia. Anesthesiology 1995; 83: 213–7.[ISI][Medline]
  23. Woehlck HJ, Dunning MBIII, Kulier AH, et al. Performance of an electrochemical carbon monoxide monitor in the presence of anesthetic gases. J Clin Monit 1997; 13: 357–62.[ISI][Medline]
  24. Wallin RF, Regan BM, Napoli MD, Stern IJ. Sevoflurane, a new inhalational anesthetic agent. Anesth Analg 1975; 54: 758–66.[Abstract/Free Full Text]
  25. Warnken UH, Asskali F, Förster H. Compound A, methanol, and dimethoxymethane production by decomposition of sevoflurane with dry CO2 absorbents Anasthesiol Inten-sivmed Notfallmed Schmerzther 1998; 33:FP-E 304.8 S415.
  26. Budavari S. The merck index. 11th ed. Rahway, NJ: Merck, 1989.
Accepted for publication November 7, 2002.




This article has been cited by other articles:


Home page
 Contin Educ Anaesth Crit Care PainHome page
G. Nunn
Low-flow anaesthesia
CEACCP, February 1, 2008; 8(1): 1 - 4.
[Full Text] [PDF]

Home page
 Anesth. Analg.Home page
C. D. Barth, M. B. Dunning III, L. Bretscher, and H. J. Woehlck
Barium Hydroxide Lime Turns Yellow After Desiccation
Anesth. Analg., September 1, 2005; 101(3): 748 - 752.
[Abstract] [Full Text] [PDF]

Home page
 Anesth. Analg.Home page
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]

Home page
 Anesth. Analg.Home page
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]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a colleague
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (15)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Holak, E. J.
Right arrow Articles by Woehlck, H. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Holak, E. J.
Right arrow Articles by Woehlck, H. J.
Related Collections
Right arrow Equipment
Right arrow Technology
Right arrow Pharmacology

Lippincott, Williams & Wilkins Anesthesia & Analgesia® is published for the International Anesthesia Research Society® by Lippincott Williams & Wilkins with the assistance of Stanford University Libraries' HighWire Press®. Copyright 2006 by the International Anesthesia Research Society. Online ISSN: 1526-7598   Print ISSN: 0003-2999 HighWire Press