ORIGINAL ARTICLE

https://doi.org/10.5005/jp-journals-10049-2025
Journal of Research and Innovation in Anesthesia
Volume 8 | Issue 1 | Year 2023

Predictive Accuracy of Gas Man® Simulation Model in Datex Avance CS2 Anesthesia Work Station using Low Flow Anesthesia with Isoflurane, Sevoflurane, and Desflurane

Kasa Sowmya¹, Madhavi Singh²

^1,2Department of Anesthesiology, Apollo Institute of Medical Sciences and Research, Hyderabad, Telangana, India

Corresponding Author: Kasa Sowmya, Department of Anesthesiology, Apollo Institute of Medical Sciences and Research, Hyderabad, Telangana, India, Phone: +91 7032360899, e-mail: kasa.sowmya28@gmail.com

Received on: 02 November 2022; Accepted on: 03 January 2023; Published on: 22 May 2023

ABSTRACT

Introduction: Gas Man® is a computer simulation program used for understanding the pharmacokinetics of volatile agents. On entering the patient details, fresh gas flow (FGF), volatile anesthetic concentration, and ventilatory details, it can predict the end-tidal concentration of volatile agents. ICOLLECT software is available for collection of real-time data from the workstation as well as hemodynamic parameters as it functions as a multichannel monitor. It is an electronic anesthetic record. We have used both these computer-based programs to study low flow anesthesia (LFA) in clinical practice.

Aims: The primary purpose of this study was to compare the expired volatile anesthetic concentrations predicted by the Gas Man® simulation model with those actually occurring during general anesthesia (GA) using isoflurane, sevoflurane or desflurane in clinical practice using low FGF.

Material and methods: Study area—the study was conducted in the Department of Anesthesiology, CARE Hospital, Hyderabad, Telangana, India. Study population—all patients who are posted for surgery under GA between the age groups 18–65 years, of either sex, and those belonging to the American Society of Anesthesiologists physical status I–II will constitute the study population. Sample size—a total of 30 patients undergoing GA with isoflurane, sevoflurane, and desflurane have undergone the validation trial using LFA. Our sample size calculation is based on a similar validation study. Study design—observational. Study duration—the proposed study was conducted over a period of 1 month (September 2016). Data collection techniques and tools—we collected relevant data directly from the Datex CS2 workstation via the ICOLLECT software for 30 anesthetics (isoflurane, sevoflurane, and desflurane) during the maintenance phase employing LFA. The measured concentration of volatile agent as well as the calculated concentration obtained by the Gas Man® equation were tabulated for each patient at 5-minute intervals. The performance error (PE), divergence, median predictive error, and wobble were determined for all three agents using the actual measured concentration end tidal agent against the predicted concentration. The statistics used for predicting the accuracy of volatile anesthetics uptake have been described by Varvel et al. Their model is based on those described for intravenous drug delivery systems. Multiple studies have validated this as a reliable model for predicting the accuracy of volatile anesthetics too. We calculated the median absolute performance error (MDAPE) median predictive error (MDPE) divergence, and wobble from the PE for all our cases. We calculated the MDAPE, MDPE, divergence, and wobble from the PE for all our cases.

Results: Mann–Whitney U test done for each study group demonstrated that there was no statistically significant difference between the median measured end-tidal concentrations of volatile agents versus that predicted by the Gas Man® anesthesia simulator. The three groups were similar with respect to the measured and predicted end-tidal concentration of the anesthetic agent. Kruskal–Wallis test was undertaken to study the intergroup variability with respect to measured and predicted end-tidal concentration of volatile agents. We obtained a p-value of <0.01, confirms a statistically significant difference between the three groups. This is expected as the minimum alveolar concentration (MAC) requirements of the three agents studied are different.

Discussion: In spite of the oversimplification of volatile kinetics, the Gas Man® simulation is an accurate predictor of the actual volatile agent’s end-tidal concentrations achieved during LFA. It can serve as a useful educational tool for the implementation of LFA.

Conclusion: There is good correlation between measured and predicted end-tidal concentrations of all three volatile anesthetics during LFA.

How to cite this article: Sowmya K, Singh M. Predictive Accuracy of Gas Man® Simulation Model in Datex Avance CS2 Anesthesia Work Station using Low Flow Anesthesia with Isoflurane, Sevoflurane, and Desflurane. J Res and Innov Anesth 2023;8(1):1-5.

Source of support: Nil

Conflict of interest: None

Keywords: Desflurane, Gas man, ICOLLECT, Isoflurane, Low flow anesthesia, Sevoflurane.

INTRODUCTION

Gas Man® is a computer simulation program designed for understanding the pharmacokinetics of volatile agents. On entering patient details, FGF, volatile anesthetic concentration on the vaporizer, and ventilatory details, it can predict the end-tidal concentrations of volatile agents in steady state.¹^,²
ICOLLECT software collect real-time data from the workstation. It is an electronic anesthetic record linked to the workstation which also records all vital parameters via a multichannel monitor. We have used both these computer-based programs to study LFA in clinical practice.

AIMS

The primary aim of this study was to compare the concentrations of expired volatile anesthetics predicted by the Gas Man® anesthesia simulator with concentrations occurring in real-time GA (isoflurane, sevoflurane, or desflurane) in clinical practice using low FGF.

MATERIAL AND METHODS

Study Area

The study was done in the Department of Anesthesiology, CARE Hospital, Hyderabad, Telangana, India.

Study Population

Our study population consisted of all adult patients (18–65 years) scheduled for elective surgical procedures under GA who belonged to the American Society of Anesthesiologists physical status I–II.

Sample Size

After a review of previous validation studies using LFA, we decided to enroll 30 patients undergoing GA with isoflurane, sevoflurane, and desflurane.

Study Design

Observational

Study Duration

The proposed study was conducted over a period of 1 month (September 2016).

Data Collection Techniques and Tools

We collected relevant data directly from the Datex CS2 workstation via the ICOLLECT software for 30 anesthetics (isoflurane, sevoflurane, and desflurane) during the maintenance phase employing LFA (Total gas flow 0.8-1 liter/minute).

The measured concentration of volatile agent as well as the calculated concentration obtained by the Gas Man® equation were tabulated for each patient at 5-minute intervals.

The PE, divergence, median predictive error, and wobble were determined for all three agents using the actual measured concentration (ET AGENT) against the predicted concentration. The acceptable values for these parameters are tabulated below (Table 1).

**Table 1:** Values for MDPE, MDAPE, divergence, and wobble for all 30 cases, separated by the volatile anesthetic used
Variables	Normal values	All cases (n = 30)	Isoflurane (n = 10)	Sevoflurane (n = 10)	Desflurane (n = 10)
MDPE (%)	−12–16%	−0.24 (5.19–4.72)	3.46 (11.74–4.82)	3.36 (2.93–9.11)	3.3 (3–4)
MDAPE (%)	11–24%	13.7 (10.6–16.8)	17.2 (11.7–22.7)	10.9 (8.5–13.4)	11 (5–15)
Divergence (%/h)	−17–2.9%/h	2.29 (3.95–8.53)	3.94 (6.48–14.4)	0.85 (6.7–8.4)	0.9 (5–10)
Wobble (%/h)	7–11.6%	3.96 (2.67–5.25)	3.98 (1.78–6.18)	3.75 (2.225.28)	2.9 (1–3)

Varvel and colleagues have described the statistical methods involved in predicting the accuracy of volatile anesthetics uptake.⁹ Their model is based on those described for intravenous drug delivery systems. Multiple studies have validated this as a reliable model for predicting the accuracy of volatile anesthetics too. We calculated the MDAPE, MDPE, divergence, and wobble from the PE for all our cases.

RESULTS

Demographics
The three groups consisted of 10 patients; each was similar with respect to age and weight (Table 2 and Fig. 1).

**Table 2:** Patient demographics (age, weight)
	Isoflurane	Sevoflurane	Desflurane
Age [mean ± standard deviation (SD)]	138 ± 10.03	41 ± 8.41	45 ± 7.61
Weight (mean ± SD)	64 ± 12.45	65 ± 13.34	63 ± 8.46

Fig. 1: Patient demographics (age, weight)

Relationship between measured and predicted end-tidal concentrations of three volatile agents (isoflurane, sevoflurane, and desflurane) (Table 3).

**Table 3:** Relationship between measured and predicted end tidal concentrations of volatile agents
Variables	Measured expired anesthetic agent concentration (%vol) median (min-max)	Gas Man predicted end tidal concentration (%vol) median (min-max)	p-value
Isoflurane	0.92 (0.88–0.97)	0.97 (0.86–0.98)	0.17	Not significant
Sevoflurane	1.9 (1.8–1.97)	1.89 (1.85–1.96)	0.67	Not significant
Desflurane	5.5 (5.4–5.9)	5.4 (5.4–5.8)	0.27	Not significant

Test of significance (Mann–Whitney U Test) was conducted for the three groups, and we did not find a statistically significant difference in the measured and predicted median values of the three agents (Fig. 2).

Fig. 2: Relationship between measured and predicted end tidal concentrations of volatile agents

The three groups were similar with respect to measured and predicted end-tidal concentration of anesthetic agent (Table 4).

**Table 4:** Intergroup variability with respective to measured and predicted endtidal concentration of anesthetic agents
Variable	Isoflurane	Sevoflurane	Desflurane	p-values
Expired	0.92 (0.88–0.97)	1.9 (1.8–1.97)	5.5 (5.4–5.9)	<0.01
Predicted	0.97 (0.86–0.98)	1.89 (1.85–1.96)	5.4 (5.4–5.8)	<0.01

Kruskal–Wallis test was undertaken to study the inter-group variability with respect to measured and predicted end-tidal concentration of volatile agents.
We obtained a p-value of <0.01, which illustrates that there was a statistically significant difference between the three groups. This is expected as the MAC requirements of the three agents studied are different.

Isoflurane

This is a graphical display of the measured and predicted end-tidal concentration of isoflurane on a linear time scale (Fig. 3).

Fig. 3: Graphically display of the measured and predicted end tidal concentration of isoflurane. Dotted line; measured end tidal anesthetic concentration. Continuous line: predicted end tidal anesthetic concentration

Sevoflurane

This is a graphical display of the measured and predicted end-tidal concentration of sevoflurane on a linear time scale (Fig. 4).

Fig. 4: Graphical display of the measured and predicted end tidal concentration of sevoflurane. Dotted line; measured end tidal anesthetic concentration. Continuous line; predicted end tidal anesthetic concentration

Desflurane

This is a graphical display of the measured and predicted end-tidal concentration of desflurane on a linear time scale (Fig. 5).

Fig. 5: Graphically display of the measured and predicted end tidal concentration of desflurane. Dotted line; measured end tidal anesthetic concentration. Continuous line; predicted end tidal anesthetic concentration

We obtained an overall MDPE of −0.24, MDAPE of 13.7, a divergence of 2.29, and wobble of 3.96. (from Table 1) All these values collectively, as well as individual agents, fall within acceptable limits.

DISCUSSION

Our observational study consisted of 30 consecutive anesthetics with three commonly employed volatile anesthetics. No randomization was undertaken for sample selection. We analyzed these three groups and found them similar with respect to age and weight, as displayed in Table 2 and Figure 1.
The relationship between MAC and body weight is more complex. It is an important factor in the computation of the multicompartment model. Brody’s formula for oxygen consumption during LFA depends on the subject’s weight.³
Analysis of actual versus calculated end-tidal concentrations of all three volatile agents by comparing the median percent volumes (Table 3 and Fig. 2) resulted in no statistically significant difference. The p-value obtained for the three agents are 0.17, 0.67, and 0.27, respectively (p > 0.05). This indicates that the clinical concentration of the volatile agents matches that calculated by Gas Man multicompartmental model.
Our results are similar to that obtained by Jeffrey M, Feldman, Philip JH⁴ used computer simulation to study volatile agents washed out with varying fresh gas flows. They concluded that Gas Man® could be used as a clinical tool for the maintenance of anesthesia in daily practice. These Gas Man® simulations done by them confirmed that the use of hyperventilation to hasten recovery is marginally beneficial with the newer, less soluble agents.⁵
The Gas Man® simulation model has been used by investigators to prove that changing volatile agents at the end of surgery may not hasten recovery in patients (Neuman et al.).⁶
Screen-based simulation has been employed to study the differences and similarities between intravenous induction as compared to volatile agent-based induction of GA (Philip et al.).⁷ The importance of expired end-tidal concentration of volatile agents has been stressed by various investigators. The expired concentration of volatile anesthetic agent is a direct reflection of the level achieved within the target organ, that is, the human brain (Van Zundert et al.).⁸
To attain 1 MAC, isoflurane, sevoflurane, and desflurane end-tidal anesthetics concentration are 1.2, 2, and 6 vol%, respectively. This concept is demonstrated in Table 4 and Figures 3 to 5.
Kruskal–Wallis test was undertaken to study the intergroup variability with respect to measured and predicted end-tidal concentration of volatile agents. We obtained a p-value of <0.01 for both measured and predicted tidal volatile agent concentrations. This implies that there is a statistically significant difference between the three groups. This is expected as the MAC requirements of the three agents studied are significantly different.
Figures 3 to 5 illustrate the effect of high flow during induction for the first 15 minutes. There is a significant variation between the measured and predicted end-tidal concentration of volatile agents during this period. As time progress, the gap between the measured and predicted end-tidal concentration of volatile agents in the graph decreases. The equilibration of volatile agents is dependent on multiple factors. The 15–20 minutes of GA is needed to ensure that the expired concentration of the volatile agent is closer to the inspired concentration of the volatile agent; FGF can then be reduced to LFA recommendation for maintenance.
Varvel et al. have described the statistics used for predicting the accuracy of volatile anesthetics uptake. Their model is based on those described for intravenous drug delivery systems.⁹ Multiple studies have validated this as a reliable model for predicting the accuracy of volatile anesthetics too. We calculated the MDAPE, MDPE, divergence, and wobble from the PE for all our cases.

CONCLUSION

Using the Gas Man® simulation model for calculating the predicted end-tidal concentration of the volatile agents is a reliable method for delivery of GA as it has a good correlation with the measured end-tidal concentration of isoflurane, sevoflurane, and desflurane during LFA.

REFERENCES

1. Philip JH. Using screen-based simulation of inhaled anaesthetic delivery to improve patient care. Br J Anaesth 2015;115(S2):ii89–ii94. DOI: 10.1093/bja/aev370

2. Nickalls RWD, Mapleson WW. Age-related iso-MAC charts for isoflurane, sevoflurane and desflurane in man. Br J Anaesth 2003;91(2):170–174. DOI: 10.1093/bja/aeg132

3. Brody S. Bioenergetics and Growth. New York: Reinhold, 1945.

4. Philip JH. Gas Man: an example of goal oriented computer assisted teaching which results in learning. Int J Clin Monit Comput 1986;3(3):165–173. DOI: 10.1007/BF01716358

5. De Wolf Andre M, Tom C, et al. Theoretical effect of hyperventilation on speed of recovery and risk of rehypnotization following recovery - a GasMan® simulation. BMC Anesthesiol 2012;12:22. DOI: 10.1186/1471-2253-12-22

6. Neumann MA, Weiskopf RB, Gong DH, et al. Changing from isoflurane to desflurane toward the end of anesthesia does not accelerate recovery in humans. Anesthesiology 1998;88(4):914–921. DOI: 10.1097/00000542-199804000-00010

7. Philip BK, Lombard LL, Roaf ER, et al. Comparison of vital capacity induction with sevoflurane to intravenous induction with propofol for adult ambulatory anesthesia. Anesth Analg 1999;89(3):623–627. DOI: 10.1097/00000539-199909000-00014

8. van Zundert T, Hendrickx J, Brebels A, et al. Effect of the mode of administration of inhaled anaesthetics on the interpretation of the FA/FI curve – a GasMan® simulation. Anaesth Intensive Care 2010;38:76–81. DOI: 10.1177/0310057X1003800114

9. Varvel J, Donoho D, Shafer S. Measuring the predictive performance of computer-controlled infusion pumps. JPharmacokinet Biopharm 1992;20(1):63–94. DOI: 10.1007/BF01143186

10. Kennedy RR, French RA, Spencer C. Predictive accuracy of a model of volatile anesthetic uptake. Anesth Analg 2002;95(6):1616–1621. DOI: 10.1097/00000539-200212000-00027

________________________
© The Author(s). 2023 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and non-commercial reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.