Secondary Logo

Journal Logo

Original Papers

Adsorption of desflurane from the scavenging system during high-flow and minimal-flow anaesthesia by zeolites

Jänchen, J.*; Brückner, J. B.; Stach, H.*

Author Information
European Journal of Anaesthesiology: May 1998 - Volume 15 - Issue 3 - p 324-329



All modern inhalational anaesthetics are chlorofluoro-(HCFCs)- or fluoro- (HFCs) compounds. Only a fraction of the applied amount is absorbed by the patient. Consequently, the overwhelming part of the anaesthetic leaves the conventional circuit via the spill into the scavenging system. As a consequence; all these inhalation anaesthetics pollute the atmosphere, and contribute to the greenhouse effect and (HCFCs only) to ozone loss [1,2]. Pollution caused by inhalational anaesthetics can be reduced either by using minimal/low flows systems or efficient recovery devices.

Desflurane is characterized by a very low blood/gas partition coefficient [3]. Therefore, induction of the anaesthesia is rapid and the recovery shorter in comparison with other fluorocarbons. Furthermore, losses of desflurane via skin and intestines are lower compared with other inhalation anaesthetics. This corresponds with the significantly lower coefficients of solubility in the rubber and plastic components of the anaesthetic circuit system. Therefore, desflurane is an ideal anaesthetic for use in the low/minimal flow technique [3].

Adsorption of inhalational anaesthetics, by charcoal filters, from the outlet port of the circuit is sufficient but is not very suitable for recycling the used fluorocarbon as desorption of chemically unchanged compounds in significant quantities is difficult to achieve. Unlike charcoals, zeolites allow the desorption of adsorbed fluorocarbons because of weaker interaction in the zeolite pores [4]. Zeolites are crystalline microporous aluminosilicates which can be modified to give a wide range of chemical compositions, polarities and pore sizes [5,6]. Hence, these materials, often so called molecular sieves, are unique materials which can be tailored for the separation of fluorocarbon vapours from a nitrous oxide/oxygen mixture.

The purpose of the present study was to evaluate the quantity of desflurane adsorbed by high-silica molecular sieves using a standard closed-circuit system under a minimal flow fresh gas regime in comparison with higher inflows.


The present study was performed on 13 patients (ASA I-II) scheduled for standard surgical procedures, mainly gynaecological laparoscopies and breast surgery. The patient data are shown in Table 1. All patients were premedicated with midazolam (0.1 mg kg−1) and atropine (0.5 mg) 1 h before induction of the anaesthetic.

Table 1
Table 1:
Patients data and duration of the anaesthesia

Anaesthesia was induced with thiopentone (3.5-4.0 mg kg−1). All patients received remifentanil (4-6 μg kg−1 h−1) by continuous infusion. Endotracheal intubation was performed after neuromuscular blockade with mivacurium 0.2 mg kg−1 intravenously (i.v.). After intubation, all patients were subject to controlled ventilation with a Dräger Cato® (Dräger Company, Lübeck, Germany). Tests for anaesthetic leakage were performed before the induction of anaesthesia. The maximum tolerated leakage volume was less than 25 mL min−1.

Desflurane was delivered immediately after intubation with a fresh gas flow of 6 L min−1 (N2O/O2: FIO2 0.3) for 5 min, followed by 3 L for another 5 min. The delivered desflurane was adjusted to obtain expiratory concentrations of 6% (v/v). In the 'minimal flow group', the fresh gas flow was adjusted to 0.5 L min−1 10 min after the induction with desflurane. The FIO2 was maintained at 0.3, the expiratory desflurane concentration in the circuit at 6.0% (DATEX AS 3®; Datex-Engstrom Deutschland GmbH, Achim, Germany). The amount of gas used for the DATEX AS 3 was re-injected into the cycle system of the Catomachine. In the 'high-flow-group', a flow of 3 L min−1 was used until the end of the procedure. At the end of the surgical procedure, all patients were ventilated with oxygen for 5-10 min according to the need for extubation.

Two glass adsorbers of different size filled with about 220 or 620 g of hydrophobic zeolite were placed at the outlet of the anaesthesia machine for selective adsorption of the scavenged desflurane [7]. This arrangement did not influence the performance of the anaesthesia machine. Using the larger adsorber (330 mm long and 74 mm in diameter), adsorption was efficient in high-flow anaesthesia up to 110 min. The smaller adsorber (270 mm long and 50 mm diameter) served during minimal flow anaesthesia up to 140 min for adsorption of the anaesthetic. The outlet of the zeolite adsorber was connected to a DATEX Ultima® gas monitor (Datex-Engstrom Deutschland GmbH) in order to monitor the adsorbing capacity of the zeolite and to detect the break-through point of the desflurane. The adsorber was placed on an electronic balance (Mettler-Toledo AG, Greifensee, Switzerland) during the experiments to record the increase in weight with time for determination of the amount of scavenged desflurane. The delivered amount of desflurane was estimated after anaesthesia by weighing the quantity of desflurane needed to refill the vaporizer. Anaesthesia always started with a completely charged vaporizer.

The desorption of the desflurane from the zeolite was carried out under vacuum (about 10 mbar) at 90-130°C, followed by condensation of the desflurane in a cooling unit at 2-8°C. The desorbed desflurane was analysed for purity using gaschromatography/mass spectrometry (GC/MS SATURN II, VARIAN; Varian GmbH, Darmstadt, Germany).

The results were evaluated with SPSS-ANOVA (n-test, t-test and regression coefficient); P<0.05 was considered to indicate statistically significant differences.


Figures 1 and 2 show selected 'adsorption curves' of scavenged desflurane adsorbed in the zeolite using minimal- and high-flow anaesthesia, respectively. In both cases, the course of the curves is exactly governed by the anaesthesia regime. Whereas three phases in the curves can be identified in Fig. 1 (minimal flow), the traces in Fig. 2 (high flow) show two sections because of the differences in the fresh gas inflow regime. With some delay, the slope of the curves in Fig. 1 becomes less steep after 5 min (inflow reduced from 6 to 3 L min−1 induction period) and much more pronounced after 10 min, the beginning of the maintenance period with 0.5 L min−1. The recovery period (the last 10 min) is characterized by an upswing at the end of the curves caused by adsorption of the desflurane washed out into the zeolite. This amount of the anaesthetic, about 3-5 g, comes from the patient and the circuit-system.

Fig. 1
Fig. 1:
Scavenged desflurane adsorbed by zeolite as a function of time during minimal-flow anaesthesia of some selected cases: (▵) patient A; (◊) patient E; and (○) patient G.
Fig. 2
Fig. 2:
Scavenged desflurane adsorbed by zeolite as function of time of some selected cases: (▵) patient A (minimal flow); (○) patient H (3 L min−1 fresh gas flow); and (□) patient J (3 L min−1 fresh gas flow).

Application of the high-flow regime (Fig. 2), starting with 10 min at 6 L min−1 followed by 3 L min−1 in the maintenance period, and again 6 L min−1 in the recovery period changes the curves significantly. For comparison, a typical 'minimal flow curve' is included in Fig. 2. It is striking that there is up to five times higher adsorption of desflurane during high-flow anaesthesia. Another difference can be found in the recovery period. The amount adsorbed decreases slightly in contrast with the upswing of the curve in the minimal flow regime caused by desorption of some nitrous oxide by the pure oxygen stream during the recovery period. In a separate experiment without desflurane, it was found that less than 9 g of nitrous oxide was adsorbed in the large adsorber and that this was removed almost completely by flushing with pure oxygen. Furthermore it has been taken into account that the incoming desflurane also displaces the weakly bonded N2O in the zeolite. Thus, only residual nitrous oxide which remains is removed in the recovery period and this can be seen in the small down-swing of the curves in Fig. 2.

Figure 3 (minimal flow) and 4 (high flow) summarize the results for all 13 anaesthetic procedures detailing the amount of desflurane delivered during anaesthesia and the quantity of scavenged desflurane which was withheld in the zeolite. Since the concentration of desflurane at the outlet of the zeolite adsorber was nearly zero (0.02%), losses of the anaesthetic were not taken into consideration. Almost all the desflurane leaving the anaesthesia machine was adsorbed. Only for the two operations lasting more than 100 min was some desflurane (about 0.2-0.5%) detected during recovery at the end of the adsorber tube, indicating the exhaustion of adsorbing capacity. Comparing the delivered desflurane with the adsorbed desflurane in the zeolite, 62% and 86% (means ±6.5 and ±3.5 SEM, respectively) of the anaesthetic was recovered during minimal- or high-flow anaesthesia, respectively, at the flow rates used (0.5 or 3 L min−1). The observed differences between the two groups were highly significant.

Fig. 3
Fig. 3:
Amount of desflurane delivered (left-hand column) and adsorbed (right-hand column) by zeolite using minimal-flow anaesthesia.

As shown in Fig. 5, all the desflurane could be separated from the molecular sieve within 2-3 h and recovered in the fluid phase. However, the over-whelming amount of the anaesthetic (about 90%) is desorbed within 1 h (from the smaller adsorber) or after 2 h from the larger adsorber at temperatures of about 90°C. The results from gas chromatography/mass-spectrometry studies in the higher-flow regimes showed very low concentrations of impurities in the desorbed desflurane (Table 2) and some water (≈1 g) swimming on the surface of the liquid desflurane.

Fig. 5
Fig. 5:
Characteristic kinetic curves of the desorption of desflurane from the larger (▿) and the smaller (○) adsorber. The dashed lines denote maximum weight loading of the zeolite during anaesthesia.
Table 2
Table 2:
Amounts of desflurane adsorbed, desorbed and condensed and the purity of the desorbed liquid desflurane


Of course, application of minimal-flow anaesthesia minimizes the consumption of the anaesthetic compared with the high-flow method, as clearly demonstrated in Figs 2-4. Thus, the absolute values in Fig. 3 differ considerably from those in Fig. 4 because of the large difference in the inflow rate of the fresh gas. The same clear idea of this difference is shown in Fig. 2 in which the uptake curves for both methods are compared. Furthermore, Figs 1 and 2 illustrate the well-known fact that most of the anaesthetic is consumed in the induction period in minimal-flow anaesthesia, whereas large amounts of desflurane have to be delivered during the maintenance period in the high-flow system. However, the difference between delivered and adsorbed amounts of desflurane (Figs 3 & 4) was independent of the fresh gas inflow rates used and should represent the uptake of the anaesthetic by the patient. Values between 10.7 g (patient J) and 33.4 g (patient G) of desflurane, respectively were found to be caused by variations in the duration of the anaesthesia and patients body weight. Losses of the anaesthetic by leakage in the used equipment were minimal.

Fig. 4
Fig. 4:
Amount of desflurane delivered (left-hand column) and adsorbed (right-hand column) by zeolite using 3 L min−1 fresh gas flows during anaesthesia.

In Fig. 6, the amount of desflurane in g kg−1 body weight which remained in the patients (or was partially lost) is shown in relation to the used fresh-gas flow. As it might be expected, there were no differences to be found between the high-flow and minimal-flow systems in relation to the uptake of desflurane, which was between 0.2 and 0.4 g kg−1 at anaesthesia times greater than 70 min. The uptake curves of desflurane [8] are in accordance with these results. The observed differences in desflurane consumption are caused by individual patient variations.

Fig. 6
Fig. 6:
Desflurane uptake of all patients in g kg−1 and duration of the anaesthesia for minimal- (○) and higher-flow (▵) anaesthesia.

Desorption at elevated but still mild temperatures of about 90°C under vacuum allows a careful removal of most of the adsorbed desflurane from the molecular sieve. Complete desorption in an acceptable period of time needs a slight increase in the desorption temperature up to 130°C. The results of the purity of the desorbed desflurane by gas chromatography/mass spectrometry indicate the good quality of the recovered desflurane (Table 2). At the moment, further work is in progress in an attempt to identify the remaining chemical impurities.

It can be concluded that adsorption of scavenged desflurane by zeolites and almost complete desorption can be achieved. Thus, recovering between 62% and more than 86% of the delivered desflurane from the scavenged gas stream was achieved depending on the fresh gas rate. Because of some losses during condensation of the desfurane, about 85% of the adsorbed anaesthetic has been obtained as liquid only (see Table 2), corresponding to 53-73% of the delivered desflurane. These values are higher than those given by Marx et al.[9] for isoflurane, enflurane and halothane.

In the past, several attempts were made to use charcoal filters for adsorption of inhalational anaesthetics from the outlet of the anaesthesia machine. These adsorbers did not allow efficient desorption because of the higher affinity of the charcoal and different pore sizes. Therefore, those charcoal filters were used in a one-way procedure to protect the personnel; however, this method was without environmental benefits. On the contrary, high silica zeolites allow efficient adsorption and desorption of the desflurane, and thus, recycling of the scavenged inhalation anaesthetic.

At present, anaesthetics play only a minor part in the total air pollution by fluorocarbons and it is not yet proved to what extent the 'stable' fluorocarbon desflurane has environmental side-effects, but the public discussion of those effects will continue to increase in the future and our speciality will have to face this. The use of low/minimal flow systems will decrease air pollution by inhalational anaesthetics. Additionally, the use of recovery systems based on zeolites opens opportunities to recycle inhalational anaesthetics and/or avoid environmental pollution almost completely. We do not yet believe that it will be possible to reuse the recovered desflurane in patients without additional purification procedures. If this could be done in cooperation with the manufacturers of the anaesthetics, almost complete recycling would be possible.


1 Brown AC, Camosa-Mas CE, Parr AD, Pierce JHT, Wayne RP. Tropospheric lifetimes of halogenated anaesthetics. Nature 1989; 341: 635-637.
2 Solomon S, Albritton D. Time-dependent ozone depletion potentials for short and long-term forecasts. Nature 1992; 357: 33-37.
3 Yasuga N, Lockhart SH, Eger E I, II et al. Kinetics of desflurane, isoflurane, and halothane in humans. Anesthesiology 1991; 74: 489-498.
4 Stach H, Sigrist K, Radeke K-H, Riedel V. Investigations of the adsorption of halogenated hydrocarbons on microporous adsorbents. Part 3. Gaschromatographic measurement of the initial heat of adsorption Q0 on charcol. Chem Tech 1995; 47: 55-61.
5 Jansen JC, Stöcker M, Karge HG, Weitkamp J. Advanced zeolite science and applications. Stud Surface Sci Catal 1994; 85.
6 Puppe L. Zeolites-properties and technical applications. Chemie in unserer Zeit 1986; 20: 117-127.
7 Stach H, Bachert H-J, Welke W. DE-OS 195 49 271.
8 Eger EI, II. Uptake and distribution. In: Miller RD, ed. Anesthesia, 4th Edn. Edinburgh: Churchill Livingstone, 1994.
9 Marx T, Gross-Alltag F, Ermisch J, Haehnel J, Weber L, Friesdorf W. Recovery of anaesthetic waste gases-an experimental approach. Anaesthesist 1992; 41: 99-102.

DESFLURANE, adsorption and desorption, zeolites, scavenging system

© 1998 European Society of Anaesthesiology