The Accuracy of the Anesthetic Conserving Device (Anaconda©) as an Alternative to the Classical Vaporizer in Anesthesia : Anesthesia & Analgesia

Secondary Logo

Journal Logo

Technology, Computing, and Simulation: Research Reports

The Accuracy of the Anesthetic Conserving Device (Anaconda©) as an Alternative to the Classical Vaporizer in Anesthesia

Soro, Marina MD, PhD, DEA; Badenes, Rafael MD; Garcia-Perez, Maria Luisa MD; Gallego-Ligorit, Lucia MD; Martí, Francisco J. MD, PhD; Aguilar, Gerardo MD, PhD; Belda, F. Javier MD, PhD

Author Information
Anesthesia & Analgesia 111(5):p 1176-1179, November 2010. | DOI: 10.1213/ANE.0b013e3181f4db38

The Anesthetic Conserving Device, AnaConDa© (ACD; Sedana Medical, Uppsala, Sweden) has been used safely since 2004 for the administration of inhalation anesthetics to sedate critical care patients receiving mechanical ventilation.13 The ACD is a modified bacterial filter or heat and moisture exchanger that is connected to the patient at the Y-piece. The ACD contains a layer of activated carbon fibers. Liquid anesthetic is supplied via a syringe pump to the patient side of the fibers. The liquid is instantly vaporized and delivered to the lungs during inspiration. During expiration, 90% of the exhaled volatile anesthetic molecules condense on the surface of the activated carbon fibers and are released again during the next inspiration.4 As a result, the ACD can be considered to be a disposable vaporizer.

However, there are many differences between a vaporizer and the ACD that make the latter more attractive. The first difference is that in the ACD the anesthetic concentration is adjusted through the infusion of the liquid anesthetic. The second difference is that the time to reach the required concentration depends only on the infusion rate and is independent of the fresh gas flow rate. The third difference is that anesthetic consumption with the ACD is less, constant, and independent of the circuit and fresh gas flow.5

Sturesson et al.5 used the ACD during cardiac surgery, and Nishiyama6 during gastrectomy. We used the ACD during postoperative care.7 The present study was designed to measure the accuracy at different anesthetic concentrations.


The hospital ethics committee approved the study and informed consent was obtained from 30 ASA I–III patients scheduled for elective surgery under general anesthesia and orotracheal intubation with a duration of 60 to 90 minutes. Patients with respiratory diseases, who were overweight (20% over their ideal weight), with muscle diseases and age <18 years were excluded.

Patients were randomly organized into 3 groups of 10 patients each. In each group, the infusion rate was adjusted following the pharmacokinetic scheme, so that a 1 vol%, 1.5 vol%, or 2 vol% alveolar target concentration (Ct) of sevoflurane was reached.


Standard 2-lead electrocardiogram, noninvasive arterial blood pressure, temperature, and pulse oximetry were monitored by means of a Tramscope monitor (Marquette™, Milwaukee, Wisconsin). Patients were premedicated orally with 0.1 mg/kg of midazolam. General anesthesia was induced with fentanyl 2 mcg/kg and propofol 1.5 to 2 mg/kg. Atracurium 0.5 mg/kg IV was given to facilitate orotracheal intubation. The lungs were ventilated with an Evita 4 ventilator (Drager™, Lubeck, Germany) with a FiO2 of 0.35. Respiratory patterns were adjusted according to the patients' needs, with a maximum tidal volume of 10 mL/kg body weight and a respiratory rate of 12 to 16 per minute to obtain PaCO2 between 35 and 45 mm Hg.

The ACD was placed between the Y-piece of the respiratory circuit and the endotracheal tube. A syringe infusion pump (Ivac P7000, Alaris Medical Systems, Basingstoke, UK) was used to infuse liquid sevoflurane into the ACD. The tubing from the syringe to the ACD, which has a volume of 1.2 mL, was purged with 1.1 mL of sevoflurane to accelerate the process of delivery of anesthetic to the patient, while not flooding the ACD.

Anesthesia was maintained with sevoflurane at an end-tidal concentration of 1%, 1.5%, or 2% with supplemental boluses of fentanyl and atracurium when indicated. A bispectral index A-2000 monitor (BIS, Aspect Medical Systems, Newton, Massachusetts) was used to evaluate anesthesia depth. BIS values were recorded only when signal quality was 80%–100%. A gas analyzer (Vamos, Drager™, Lubeck, Germany) measured the end-tidal sevoflurane and CO2 concentrations. The sampling flow (150 mL/min) was returned to the breathing system through a port located between the ACD and the endotracheal tube.

The infusion pump rate was set to the value obtained from the pharmacokinetic model to achieve 1%, 1.5%, and 2% alveolar sevoflurane concentration. The first rate setting was held for 10 minutes, after which the infusion rate was reset once each hour until the end of surgery. Hemodynamic data, BIS, and end-tidal sevoflurane concentration values were recorded every 2 minutes.

Pharmacokinetic Model

Anesthetic uptake and distribution were modeled with a simplified 9-compartment model.7 Cardiac output was deemed to be constant, steady, and proportional to body weight (cardiac output = 0.2 × kg3/4 L/min); ventilation was assumed to be continuous, in such a way that pulmonary volume was considered constant and equal to the addition of the functional residual capacity (estimated as 4% of body weight) plus half the adjusted tidal volume.810 The lung was considered to have just 1 alveolar compartment with an ideal ventilation/perfusion ratio. The loss of anesthetic through the ACD was calculated with an empirical formula based on bench studies.11 The model was built using a MS Excel spreadsheet in which patient weight, sevoflurane end-tidal Ct, and minute ventilation were entered and the initial and maintenance infusion rates were calculated and displayed.7 The desired time to reach the Ct was set at 10 minutes.

Analysis of Predictive Accuracy of the Infusion Scheme

Performance was measured using the methods described by Varvel et al.12,9,1315 The performance error (PE) is the difference between the measured end-tidal sevoflurane concentration and the Ct divided by the Ct:

The median absolute performance error (MDAPE%) is the median of the absolute values of PE. The median performance error (MDPE%) is the median of the positive and negative values of PE. The wobble is the median of PE − MDPE. The performance at each Ct was calculated using measurements of end-tidal sevoflurance concentration from all patients recorded every 2 minutues, starting 10 minutes after induction and continuing to the end of anesthesia.

Statistical Analysis

SPSS for MS Windows (version 11.0, SPSS Inc.™, Chicago, Illinois) was used to calculate the mean, SD, and 95% confidence interval (95% CI) for each performance measure. The Mann–Whitney U test for independent samples was used to test the difference between the mean value of MDAPE, MDPE, and Wobble at each Ct. The Student t test for independent samples was used to test the difference between the means of MDAPE, MDPE, and Wobble at each Ct. The Mann–Whitney U test for independent samples was applied to detect differences in demographic data among the groups.


Thirty patients consented to participate. Demographic data are shown in Table 1. There were no statistically significant differences among groups. Ten minutes after induction, the median of the absolute difference between the end-tidal sevoflurane concentration and the Ct was 13.5 ± 7.1% of the Ct (median ± SD) when the target was 1.0 vol%. The median absolute difference was 7.2 ± 4.1% of the target at 1.5 vol% and 8.1 ± 5.6% of the target at 2.0 vol%. Table 2 shows the Varvel measures of performance. MDAPE, MDPE, and Wobble were significantly higher at 1 vol% than at 1.5% and 2%.

Table 1:
Table 2:
Performance Accuracies of the Model with the “Two-Stage Approach” for the Data from Minute 10 to the End of Surgery

Figure 1(PE–time course for all cases during the procedure) shows the difference between the end-tidal sevoflurane concentration and the Ct, as a percentage of the Ct, during each procedure for each patient. Except for occasional transients, all differences were between −25% and +25% of the Ct. Table 3 shows hemodynamic and BIS values during anesthesia for each subgroup.

Figure 1:
Performance error (PE)–time course for all cases during the procedure.
Table 3:
Hemodynamic Data and BIS Values Obtained During Anesthesia for Each Subgroup


The ACD achieves and maintains a targeted end-tidal sevoflurane concentration within 13.5% ± 7.1% (median of absolute difference ± SD) of the target in anesthetized patients. Its performance is the same at 1%, 1.5%, and 2% sevoflurane. An earlier study reported an accuracy of 3.9% when used for patient sedation in intensive care.7 The present study adds measurements of accuracy during brief anesthetic procedures in which anesthetic uptake may influence accuracy.

Enlund et al.16 used a nonlinear mixed modeling in NONMEM to deliver sevoflurane to an ACD in a closed breathing circuit. In 38 patients the MDAPE was 27%. The better performance of our model (MDAPE 14%) may be partially due to our use of a nonrebreathing circuit in which losses are constant and equal to 50%–65% of the administered anesthetic. The more accurate calculation of these losses, which include compensation for minute ventilation and end-tidal sevoflurane concentration, increases the accuracy of the model.

It is important to point out that in our study the infusion rate was readjusted after 10 minutes and once every hour thereafter. These manipulations can be compared with those involved in clinical anesthesia with a vaporizer to which it is common to perform a larger number of adjustments in the first hour of anesthesia.

Finally, we must point out that our results were obtained in normal-weight patients. Perhaps in obese patients the error of the model could be different because fat uptake in the first 2 hours is approximately half of the volume taken by the tissues and one third of the volume lost through the ACD.

In conclusion, we have shown that liquid anesthetic infusion through the ACD may be a valid alternative to the conventional vaporizer under circumstances in which an anesthesia machine with a vaporizer is not available. The ACD is applicable to clinical practice because its accuracy is in the range of that of the target-controlled infusion systems used in IV infusions. The system is very simple and requires few adjustments.


1. Sackey PV, Martling CR, Granath F, Radell PJ. Prolonged isoflurane sedation of intensive care unit patients with the Anesthetic Conserving Device. Crit Care Med 2004;32:2241–6
2. Migliari M, Bellani G, Rona R, Isgrò S, Vergnano B, Mauri T, Patroniti N, Pesenti A, Foti G. Short-term evaluation of sedation with sevoflurane administered by the anesthetic conserving device in critically ill patients. Intensive Care Med 2009;35:1240–6
3. Röhm KD, Wolf MW, Schöllhorn T, Schellhaass A, Boldt J, Piper SN. Short-term sevoflurane sedation using the Anaesthetic Conserving Device after cardiothoracic surgery. Intensive Care Med 2008;34:1683–9
4. Enlund M, Wiklund L, Lambert H. A new device to reduce the consumption of a halogenated anesthetic agent. Anaesthesia 2001;56:429–32
5. Sturesson LW, Johansson A, Bodelsson M, Malmkvist G. Wash-in kinetics for sevoflurane using a disposable delivery system (AnaConDa) in cardiac surgery patients. Br J Anaesth 2009;102:470–6
6. Nishiyama T. Saving sevoflurane and hastening emergence from anaesthesia using an anaesthetic-conserving device. Eur J Anaesthesiol 2009;26:35–8
7. Belda JF, Soro M, Badenes R, Meiser A, García ML, Aguilar G, Martí FJ. The predictive performance of a pharmacokinetic model for manually adjusted infusion of liquid sevofluorane for use with the Anesthetic Conserving Device (AnaConDa): a clinical study. Anesth Analg 2008;106:1207–14
8. Kennedy RR, Baker AB. The effect of cardiac output changes on end-expired volatile anaesthetic concentrations. A theoretical study. Anaesthesia 2001;56:1034–40
9. Kennedy RR, French RA, Spencer C. Predictive accuracy of a model of volatile anesthetic uptake. Anesth Analg 2002;95: 1616–21
10. Lerou JG, Booij LH. Model-based administration of inhalation anaesthesia: 1. Developing a system model. Br J Anaesth 2001;86:12–28
11. Meiser A, Bellgardt M, Belda J, Röhm K, Laubenthal H, Sirtl C. Technical performance and reflection capacity of the anaesthetic conserving device—a bench study with isoflurane and sevoflurane. J Clin Monit Comput 2009;23:11–9
12. Varvel JR, Donoho DL, Shafer SL. Measuring the predictive performance of computer-controlled infusion pumps. J Pharmacokinet Biopharm 1992;20:63–94
13. Boller M, Moens Y, Kästner SB, Bettschart-Wolfensberger R. Closed system anaesthesia in dogs using liquid sevoflurane injection: evaluation of the square-root-of-time model and the influence of CO2 absorbent. Vet Anaesth Analg 2005;32:168–77
14. Lockwood GG, Chakrabarti MK, Whitwam JG. A computer-controlled closed anaesthetic breathing system. Anaesthesia 1993;48:690–3
15. Hans P, Coussaert E, Cantraine F, Pieron F, Dewandre PY, d'Hollander A, Lamy M. Predictive accuracy of continuous propofol infusions in neurosurgical patients: comparison of pharmacokinetic models. J Neurosurg Anesthesiol 1997;9: 112–7
16. Enlund M, Kietzmann D, Bouillon T, Züchner K, Meineke I. Population pharmacokinetics of sevoflurane in conjunction with the AnaConDa: toward target-controlled infusion of volatiles into the breathing system. Acta Anaesthesiol Scand 2008;52:553–60
© 2010 International Anesthesia Research Society