1Department of Anesthesiology, Faculty of Medicine, University of Stellenbosch, PO Box 19063, Tygerberg 7505, South Africa. 2Technology and Business Development Group, Medical Research Council of South Africa, PO Box 19070, Tygerberg 7505, South Africa*Corresponding author
*Supplementary material is available to subscribers with the on-line version of the journal at the journal website.
Accepted for publication: August 30, 2001
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Results. At these flows, desflurane consumption depended on V·F. In contrast, halothane consumption was not influenced by V·F. Isoflurane and enflurane showed differences in consumption between flows of 0.5 and 3 litre min1. Stepwise linear regression suggested that besides V·F, other factors influenced consumption of the more soluble agents (sex, age, weight, height, altitude, and temperature). The partial pressure ratios were independent of V·F for desflurane (end-tidal to fresh gas=0.8), but the ratios of the more soluble agents varied with V·F (end-tidal to fresh gas=0.30.7).
Conclusions. At V·F that involves significant re-breathing, consumption of soluble agents depends only partially on V·F. These results can be explained using Maplesons hydraulic analogue model.
Br J Anaesth 2002; 88: 4655
Keywords: anaesthesia, audit; anaesthetics, volatile; equipment, anaesthesia machines
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The purposes of this study were: (i) to determine consumption of volatile anaesthetic agent using circle-absorber systems at moderate, low and minimal V·F (for the purpose of this study moderate fresh gas flow is defined as a flow of 3 litre min1; low fresh gas flow 1.0 litre min1 and minimal fresh gas flow 0.5 litre min1). (ii) To evaluate the differences between vaporizer settings and end-tidal agent partial pressures that are achieved at these flows. Clinicians administered the anaesthetics as they would have done in clinical practice, according to their judgement of patient requirements.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
No pre-medication was given. Monitoring provided by the Julian anaesthetic machine included ECG, pulse oximetry, nasopharyngeal temperature, non-invasive arterial pressure, capnography, inspired and expired gas partial pressure measurement (oxygen, nitrous oxide, volatile agent), airway pressure, tidal volume, ventilatory frequency, and minute volume. Neuromuscular transmission was monitored by train-of-four nerve stimulation.
After pre-oxygenation with 100% oxygen (V·F 6 litre min1) and establishing an i.v. infusion using local anaesthesia, fentanyl 3.0 µg kg1 was administered intravenously to suppress the response to tracheal intubation. Induction of anaesthesia was with thiopentone 35 mg kg1 and vecuronium 0.08 mg kg1 followed by manually controlled ventilation by mask with V·F settings of nitrous oxide:oxygen (3.0:1.5). Anaesthetic vapour was introduced after tracheal intubation using Table 1 as an initial guide to the dosage that could be adjusted according to patient response. These initial vaporizer settings were based upon the following calculation: FEAN = (1.3FEN2O) x MAC, where FEAN = fractional expired partial pressure of the volatile agent in question, FEN2O = fractional expired concentration of nitrous oxide (1.3 x MAC was assumed to be the volatile agent expired partial pressure sufficient to suppress movement on skin incision in 95% of patients.911 For a partial pressure of 66 kPa for nitrous oxide these initially targeted expired partial pressures (kPa) were: halothane 0.5, isoflurane 0.74, desflurane 3.8, enflurane 1.2.
|
If depth of anaesthesia had to be changed, the following procedures were followed. If a rapid change was not required, the vaporizer setting was adjusted while maintaining the V·F. To increase inspired partial pressure, overpressure was used if deemed necessary and to decrease inspired partial pressure, the vaporizer was temporarily shut off as needed. If a rapid change was required, the V·F was increased to 3 litre min1 in addition to adjusting the vaporizer setting. During low and minimal V·F the vaporizer was shut off towards the end of the procedure (1520 min) to allow the inhaled vapour partial pressure to decrease slowly. Additional neuromuscular blocking agent was administered as necessary.
Residual neuromuscular block was antagonized at the start of skin closure using neostigmine (0.04 mg kg1) mixed with glycopyrrolate (0.006 mg kg1). The vaporizer and nitrous oxide supply was shut off and the V·F increased to 6 litre min1 of oxygen after the last skin stitch had been applied. The ventilatory frequency was reduced to allow the PE'CO2 to increase to a maximum of 6.5 kPa to initiate breathing.
The following were automatically recorded to computer disk at 5-s intervals, from the time that pre-oxygenation began until the patient awoke:
Fresh gas flow
Oxygen concentration in the fresh gas
Inspired and end-tidal partial pressures of: volatile agent, nitrous oxide, oxygen, carbon dioxide
Tidal volume, ventilatory frequency
Oxygen saturation (pulse oximetry)
Heart rate (from the pulse oximeter)
Non-invasive arterial pressure (mean) measurements
Nasopharyngeal temperature
Times of adjustments to V·F and vaporizer settings were recorded manually.
After completion of the procedure the mass (g) of volatile agent was measured using mass balances capable of measuring to 0.1 g.*
Calculations
The data were imported to a computer spreadsheet (Microsoft Excel 97) for analysis. For each patient the arithmetic mean of the end-tidal anaesthetic partial pressures that had been recorded every 5 s were calculated for the duration of volatile agent administration (MFA). The following anaesthetic agent partial pressure ratios were calculated and plotted at 5 s intervals: end-tidal to inspired (FA/FI); inspired to fresh gas (FI/FD); end-tidal to fresh gas (FA/FD). These ratios were noted at the following times: at the end of the 10-min wash-in period and at a later time during anaesthetic maintenance when these ratios were not changing (i.e. at steady state). Mean rate of anaesthetic agent consumption was calculated by dividing the total mass used by the time that the anaesthetic vaporizer was switched on.
Data analysis
Statistical analysis was done with Statistical Analysis Support System (SAS), Version 6 Release 12, SAS Institute Inc., BMDP Statistical Software Inc. As the data were not always normally distributed, non-parametric, distribution-free tests were performed. Comparisons between two groups were done using the MannWhitney test or where appropriate, the Wilcoxon signed ranks paired test. Multiple group comparisons were done using Kruskal Wallis one-way analysis of variance followed by post hoc multiple comparisons (Tukeys test). An alpha-value of 0.05 or less was regarded as indicating statistical significance.
To identify factors that influenced rate of consumption of anaesthetic agent, stepwise multiple regression was performed using the following as independent variables: fresh gas flow, duration of agent administration, the inverse of the square root of time (1/t), age, mass, height, sex, ASA status, MFA, body temperature change, FA/FD. A backward selection method was used with a threshold F-ratio of 4 for entry-to/removal-from the linear regression model.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
Stepwise linear regression
Factors associated with agent consumption were sought by stepwise linear regression modelling. Results for volatile agent total consumption and consumption rate are presented in Tables 5 and 6. V·F and duration of administration were included in the regression models for total consumption for all four agents. In addition, enflurane total consumption was influenced by MFA, FA/FD at steady state, patient height, and change in body temperature. Total halothane consumption was also affected by patient height, sex, age, and ASA status; isoflurane by altitude, temperature change, and body mass.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Recognizing that all three V·F allowed significant re-cycling of exhaled gases, these apparent disparities can be explained as follows. To maintain a desired end-tidal (alveolar) partial pressure it is necessary to compensate for tissue uptake. For halothane this is considerable14 because its high tissue solubility increases uptake into peripheral organs (e.g. muscle) and furthermore a considerable proportion of halothane undergoes hepatic metabolism.14 15 It is therefore necessary to deliver a minimum amount of drug per unit time to compensate for uptake into the blood. This can be effected by either supplying the inhaled drug using a greater V·F containing a partial pressure that is equal to (or close to) the desired inhaled value, or by using a greater partial pressure at low V·F. Either way, a certain rate of delivery of agent to the breathing system is needed to maintain inhaled partial pressures.
This phenomenon may be explained using Maplesons hydraulic, pharmacokinetic model for volatile agent uptake.16 In Figure 015F2, uptake and distribution of volatile anaesthetic vapour is shown by a series of interconnected cylinders where the fluid heights represent vapour partial pressures. Volumes of distribution are depicted by the base areas, amounts of drug by fluid volumes, and drug clearances by the diameters of the various PIPES. Fresh gas partial pressure is shown by the height of the cylinder at the extreme left (FD), which is full at all times and V·F is represented by the diameter of its outlet. Inspired partial pressure is portrayed by the height of fluid in cylinder FI from which excess vapour may spill over during high V·F (analagous to the spill-valve of a ventilator). This spill-over is considerable during high V·F, minimal during low V·F and absent during closed-circuit anaesthesia. Two large peripheral compartments (muscle, fat) drain agents away from the lungs (FA), and viscera (including the brain). To maintain a constant level in the brain, it is necessary to supply drug to FA from FI to prevent the level in FA from falling. Removal of drug from FI can in turn be replaced by supplying drug from FD using a high V·F containing a low partial pressure and this is represented in the upper illustration of Figure 015F2 by a short cylinder, FD delivering fluid to FI via a wide diameter pipe of high conductance. On changing to low flow (thin outlet pipe from FD), the same amount of drug per unit time must be delivered to the breathing system and lungs by an increased pressure in FD. This is shown in the lower illustration in Figure 015F2 by the tall cylinder, FD delivering fluid to FI via a small diameter pipe of low conductance. In this manner the fluid level in FI and, therefore, in FA and the viscera is maintained. The dotted lines in the lower illustration in Figure 015F2 show how the partial pressures in FI and FA decrease if the partial pressure in FD is not increased.
|
We suggest that with regard to halothane in this study, continuing uptake of drug was so rapid that at the three V·F studied, there was little venting (spill-over) of vapour via expired gas to the atmosphere, and patient factors became important determinants of drug consumption. This conclusion is strengthened by the following considerations.
FA/FI was 0.75 at all V·F, indicating that during anaesthetic maintenance, transfer from alveoli to blood was a continuing process. The re-circulation of alveolar gas containing reduced amounts of halothane vapour to the breathing system resulted in decreased FI/FD and FA/FD ratios, and these varied with V·F (Tables 3 and 4).
At low and minimal flows, halothane FA/FI and FA/FD ratios were small (Table 4) so that the vented gas contained reduced amounts of vapour. (At reduced V·F, gas vented from a circle-system is composed mainly of expired gas).
No differences in halothane consumption rate were found at the three V·F.
Stepwise linear regression indicated that although V·F did indeed influence halothane usage, patient descriptors were also strongly associated.
At the other extreme, continuing uptake of desflurane, a much less soluble drug is markedly decreased during anaesthetic maintenance. Therefore, V·F strongly influences desflurane consumption at all flow rates because surplus is lost to the atmosphere. This conclusion is strengthened by the following considerations.
FA/FI was 0.94 at all V·F, indicating that at steady state, transfer of desflurane from alveoli to blood was small.
FI/FD and FA/FD ratios were high (0.8 and 0.75, respectively), indicating that re-circulated exhaled gas contained concentrations of desflurane vapour that were close to the inspired values. Furthermore, these ratios were uninfluenced by V·F (Tables 3 and 4). These high FA/FI and FA/FD ratios, resulted in vented gas containing significant amounts of vapour even at low and minimal flows.
Significant differences in desflurane consumption rate were measured at the three V·F.
Stepwise linear regression revealed that V·F was a major determinant of desflurane consumption but patient descriptors were not associated with drug usage.
The drugs of intermediate solubility, isoflurane and enflurane, lie between the two extremes. Agent consumption was significantly less when comparing V·F of 0.5 and 3.0 litre min1, but there were no differences in consumption between V·F of 0.5 and 1.0 litre min1 and between 1.0 and 3.0 litre min1. Consumption of these two drugs was also, like halothane, associated with certain patient descriptors (Tables 5 and 6). Lowe17 has shown that during totally closed-circuit anaesthesia, volatile agent consumption rate is directly related to the square root of time in accordance with the exponentially decreasing uptake of agent over time. This variable played a role in the regression models for enflurane and isoflurane under conditions of partial re-breathing, which suggests that at these low flows, venting of expired vapour from the circle-system was minimal.
Mathematical models have been useful in predicting theoretical uptake and consumption of various agents for comparative purposes,13 but they assume a constant alveolar partial pressure requirement and fixed patient variables that influence agent uptake (e.g. cardiac output, ventilation). In reality, these factors vary during clinical anaesthesia. For example, alveolar partial pressure requirements may increase during surgical stimulations or decreased by supplementary drugs such as nitrous oxide, opioids, and alpha-receptor agonists. Stepwise linear regression revealed statistically significant associations between volatile agent consumption and various patient descriptors. However, it cannot be construed from our data that these are causative. Our results merely indicate that other variables besides V·F influence agent consumption and prospective studies will be necessary to identify the patient characteristics and surgical conditions that determine drug uptake.
Circle-systems are often used at inappropriately high V·F and repeated recommendations have been made for the reduction of V·F in order to reduce costs. 2 47 1821 This study confirms the well known phenomenon that during low V·F there are discrepancies between vaporizer settings and inspired (or end-tidal) vapour partial pressures22 23 which is more marked with soluble agents and at low flows. Low flow anaesthesia should, therefore, be practised using agent monitors. In developing countries, halothane is the mainstay of anaesthesia and agent monitors are seldom available. This study, suggests that a V·F of less than 3 litre min1 should not be used without agent monitors. Furthermore, a potential danger of halothane overdose is present when using low V·F without an agent monitor, if the anaesthetist increases V·F and forgets to reduce the vaporizer setting.
In contrast to the more soluble drugs, FA/FD and FI/FD for desflurane were similar, predictable, and independent of V·F (0.70.8). Furthermore this study supports previous work that desflurane usage can be reduced when given using minimal V·F.1 6 23 24 Considering the predictablity of desflurane partial pressures, perhaps agent monitoring is not mandatory because the setting of the precision vaporizer can be used to indicate the alveolar partial pressure.1 This is not recommended because under- or overdosage could result if the vaporizer was faulty.
A wash-in period of 10 min at high V·F (4.5 litre min1) was used. This is less than the usually recommended 1520 min for minimal flow.25 If this were inadequate, a marked difference between FA/FI at the end of the wash-in period and during anaesthetic maintenance at low V·F would have occurred. No difference was found with desflurane. Differences were found with the other three drugs (Table 4) but these differences were small and were easily compensated for by adjustments to the vaporizers. We conclude that even for the most soluble drug, halothane, a 10-min wash-in period was sufficient. For desflurane, an even shorter wash-in period will suffice 3 14 23 24 26 with even greater cost savings.
Three of the participating hospitals were in coastal cities whereas the remainder were situated inland, more than 1300 m above sea level at reduced atmospheric pressures (Table 7). It was therefore possible to study the influence of altitude on MFA and on agent consumption. On pooling and comparing the patients mean end-tidal partial pressures of volatile agents according to whether they were anaesthetized at sea level or altitude, it was found that those who had undergone surgery at altitude required greater end-tidal partial pressures of all four volatile agents. In the present study, anaesthesia was with a combination of approximately 66% nitrous oxide and a volatile agent. Furthermore, anaesthesiologists in the seven different institutions were requested to administer anaesthesia as they normally would in their everyday practices, titrating volatile agent partial pressures according to patient requirements. Our findings therefore possibly reflect the reduced effectiveness of nitrous oxide at altitude. It has been demonstrated that at decreased ambient pressures, nitrous oxide causes less analgesia27 and is a less efficient anaesthetic,28 and nitrous oxide is also probably less effective as an adjuvant. As far as we are aware, this may be the first documented evidence for this hypothesis, considering that the only other anaesthetic adjuvant administered was a small dose of fentanyl.
|
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Baxter AD. Low and minimal flow inhalational anaesthesia. Can J Anaesth 1997; 44: 64352[Abstract]
3 Mapleson WW. The theoretical ideal fresh-gas flow sequence at the start of low-flow anaesthesia. Anaesthesia 1998; 53: 26472[ISI][Medline]
4 Body SC, Fanikos J, DePeiro D, Philip JH, Segal BS. Individualized feedback of volatile agent use reduces fresh gas flow rate, but fails to favorably affect agent choice. Anesthesiology 1999; 90: 11715[ISI][Medline]
5 Cotter SM, Petros AJ, Dore CJ, Barber ND, White DC. Low-flow anaesthesia. Practice, cost implications and acceptability. Anaesthesia 1991; 46: 100912[ISI][Medline]
6 Eger EI. Economic analysis and pharmaceutical policy: a consideration of the economics of the use of desflurane. Anaesthesia 1995; 50 (Suppl): 458[ISI][Medline]
7 Pedersen FM, Nielsen J, Ibsen M, Guldager H. Low-flow isoflurane-nitrous oxide anaesthesia offers substantial economic advantages over high- and medium-flow isoflurane-nitrous oxide anaesthesia. Acta Anaesthesiol Scand 1993; 37: 50912[ISI][Medline]
8 Logan M. Breathing systems: effect of fresh gas flow rate on enflurane consumption. Br J Anaesth 1994; 73: 7758[Abstract]
9 de-Jong RH, Eger EI. MAC expanded: AD50 and AD95 values of common inhalation anesthetics in man. Anesthesiology 1975; 42: 3849[ISI][Medline]
10 Schwilden H, Stoeckel H. Quantitative EEG analysis during anaesthesia with isoflurane in nitrous oxide at 1.3 and 1.5 MAC. Br J Anaesth 1987; 59: 738[Abstract]
11 Stevens WD, Dolan WM, Gibbons RT, et al. Minimum alveolar concentrations (MAC) of isoflurane with and without nitrous oxide in patients of various ages. Anesthesiology 1975; 42: 197200[ISI][Medline]
12 Harper M, Eger EI. A comparison of the efficiency of three anesthesia circle systems. Anesth Analg 1976; 55: 7249[Abstract]
13 Zbinden AM, Feigenwinter P, Hutmacher M. Fresh gas utilization of eight circle systems. Br J Anaesth 1991; 67: 4929[Abstract]
14 Yasuda N, Lockhart SH, Eger EI,II, et al. Kinetics of desflurane, isoflurane and halothane in humans. Anesthesiology 1991; 74: 489[ISI][Medline]
15 Carpenter RL, Eger EI,II, Johnson BH, Unadkat JD, Sheiner LB. Pharmacokinetics of inhaled anesthetics in humans: measure ments during and after simultaneous administration of enflurane, halothane, isoflurane, methoxyflurane and nitrous oxide. Anesth Analg 1986; 65: 57582[Abstract]
16 Mapleson WW. Pharmacokinetics of inhaled anaesthetics. In: Prys-Roberts C, Hug, CC jr., eds. Pharmacokinetics of Anaesthesia. Oxford: Blackwell Scientific Publishers, 1984; 89111
17 Lowe HJ, Ernst EA. The Quantitative Practice of Anesthesia: The Use of Closed Circuit. Baltimore: Williams & Wilkins, 1981
18 Bailey CR, Ruggier R, Cashman JN. Anaesthesia: cheap at twice the price? Staff awareness, cost comparisons and recommend ations for economic savings. Anaesthesia 1993; 48: 9069[ISI][Medline]
19 Becker KE jr, Carrithers J. Practical methods of cost containment in anesthesia and surgery. J Clin Anesth 1994; 6: 38899[ISI][Medline]
20 McKenzie AJ. Reinforcing a low flow anaesthesia policy with feedback can produce a sustained reduction in isoflurane consumption. Anaesth Intens Care 1998; 26: 3716[ISI][Medline]
21 Watcha MF, White PF. Economics of anesthetic practice. Anesthesiology 1997; 86: 117096[ISI][Medline]
22 Baum J. Low Flow Anaesthesia: The Theory and Practice of Low Flow, Minimal Flow and Closed System Anaesthesia, 1st Edn. Oxford: Butterworth-Heinemann, 1996
23 Hargasser S, Hipp R, Breinbauer B, Mielke L, Entholzner E, Rust M. A lower solubility recommends the use of desflurane more than isoflurane, halothane and enflurane under low-flow conditions. J Clin Anesth 1995; 7: 4953[ISI][Medline]
24 Nel MR, Ooi R, Lee DJ, Soni N. New agents, the circle system and short procedures. Anaesthesia 1997; 52: 3647[ISI][Medline]
25 Virtue RW. Minimal-flow nitrous oxide anesthesia. Anesthesiology 1974; 40: 1968[ISI][Medline]
26 Gowrie-Mohan S, Muralitharan V, Lockwood GG. The estimation of inspired desflurane concentration in a low-flow system. Anaesthesia 1996; 51: 9047[ISI][Medline]
27 James MFM, Manson EDM, Dennett DE. Nitrous oxide analgesia and altitude. Anaesthesia 1982; 37: 285[ISI][Medline]
28 Powell JN, Gingrich TF. Some aspects of nitrous oxide analgesia at an altitude of one mile. Anesth Analg 1969; 48: 680[Medline]