Abstract
Background
The anesthesiology standard of care requires limiting oxygen (O2) concentration to less than or equal to 30% during monitored anesthetic care for upper chest and head and neck surgery to reduce the fire risk. While Venturi devices are used to regulate O2 concentration, they generate high flow rates, which may increase O2 concentration near the face mask. Even small increases of O2 concentration greater than room air significantly increase combustibility, highlighting the need for careful O2 delivery strategies.
Methods
A face mask was attached to a full-body patient simulator. The mask was connected to 1) the breathing circuit via a 15-mm tracheal tube adapter using the anesthesia machine’s O2 blender set at 30%, or 2) the auxiliary O2 port using a 31% Venturi device. Surgical drapes were placed on intravenous poles (“ether screen”) to separate the anesthesia and surgical side. Inflow O2 flow was 6 L/min for both groups, with the Venturi delivering ≈52 L/min. After room air calibration, ambient O2 concentrations were measured at three sites on the anesthesia side and three on the surgical side of the drape at no flow and then 15, 30, and 60 minutes after O2 flow initiation. Experiments were repeated three times.
Results
The pooled differences in the O2 concentration increases between the two devices were 0.80% (95% CI 0.53 to 1.07, P<.001) on the surgical side and 2.22% (95% CI 2.02 to 2.42, P<.001) on the anesthesia side of the ether screen. At 60 minutes, the mean (SD) concentrations in the Venturi group were 22.99% (0.31%) on the anesthesia side and 21.72% (0.90%) on the surgical side. Corresponding values in the anesthesia circuit group were 21.07% (0.12%) and 21.00% (0.11%).
Conclusions
A 31% Venturi device resulted in clinically relevant increases in ambient O2 concentrations on both sides of the ether screen compared to using the anesthesia machine O2 blender set at 30%. The differences are substantive because each 1% increase in the ambient O2 concentration increases the combustion rate of cotton by 15%. The use of a Venturi device to regulate O2 concentrations in the presence of an ignition source during head and neck surgery is not recommended. We advocate for anesthesia machine manufacturers to provide an integrated O2 blender for the auxiliary gas outlet of all their products as a fire safety measure.
Keywords: equipment safety, fire risk, mask, oxygen inhalational therapy, simulation
Introduction
During monitored anesthesia care for upper chest and head and neck surgery in the presence of an ignition source (e.g., an electrosurgical device), current guidelines recommend that the delivered oxygen (O2) concentration should be limited to no greater than or equal to 30% to mitigate the fire risk [1-4]. We recently described a method for delivering O2 using the anesthesia machine's O2 blender and breathing circuit. In this approach, a nasal cannula or face mask is connected to the Y-piece of the circuit via a 15-mm adapter from a 5.0 mm tracheal tube. The adjustable pressure-limiting (APL) valve is partially closed to generate circuit pressure, which influences the delivered flow rates [5]. Loeb previously suggested using a Venturi device to regulate the O2 concentration [6], but such devices generate high flow rates, raising concern about increasing the O2 concentration near the mask more than when using a face mask at conventional flow rates. For example, based on the device manufacturers' standard entrainment ratios, a 31% Venturi device requiring 6 L/min O2 flow will entrain 45.5 L/min of room air, producing a total flow of 51.5 L/min. It is a misconception that an ambient O2 concentration of 30% is without increased risk compared to room air. Even small increases above 21% produce an increase in combustibility, with the burn rate of cotton rising linearly from 2 cm/sec in a 21% O2 environment to approximately 5 cm/sec when the concentration is increased to 31% [7]. Because cotton is a common component of surgical drapes and dressings in the operative field, a higher ambient O2 concentration may significantly increase the speed at which these materials ignite and burn, elevating the risk of surgical fires. Because Venturi devices generate high flow rates, they may saturate the nearby environment with O2-enriched gas, particularly in the presence of surgical drapes that limit dispersion. In contrast, face masks connected to the anesthesia circuit using the O2 blender operate at lower total flow rates. For this reason, we hypothesized that differences in the ambient O2 concentration near the face mask would be >0.5% higher when using a Venturi device than a face mask connected to the circuit at comparable delivered O2 concentrations. We selected 0.5% as the threshold based on its potential clinical significance in increasing fire risk [3,7].
Materials and methods
Institutional review board approval was not indicated because this was a laboratory investigation not involving patients, animals, or subjects. This manuscript adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.
Experimental design
The simulation experiments were conducted on September 17, 2024, at the Center for Patient Safety (CPS), a joint venture of the University of Miami Miller School of Medicine and the Jackson Health System. In the CPS operating room, a SimMan® Essential Bleeding patient simulator (Laerdal, Wappingers Falls, NY, USA) was used on an operating room table with the face mask secured by a strap. A standard, disposable surgical drape was fastened between two poles at a height of approximately 5 feet on either side of the patient simulator (i.e., an “ether screen”), separating the testing area into two zones. We refer to the area between the anesthesia machine and the drape as the anesthesia side of the ether screen, corresponding to the area surrounding the face mask. We refer to the area on the other side of the drape as the surgical side of the ether screen. The face mask was connected to the anesthesia circuit using the 15-mm adapter from a 5.0 mm tracheal tube, as previously described [5], or to a Venturi device (Vyaire Medical, Inc., Mettawa, IL, USA). A Datex-Ohmeda Aisys anesthesia machine (GE HealthCare Technologies, Chicago, IL, USA) was used with its digital O2 blender set to 30% and a 6 L/min flow rate for experiments using the anesthesia circuit as the delivery portal. A 31% Venturi device was used with an O2 flow rate of 6 L/min for the Venturi experiments. An annotated photograph of the experimental setup is shown in Figure 1.
Figure 1. Photograph of the experimental setup.
In the upper panel (A), the sampling sites are noted by the red circles, labeled with an S. An ether screen, lowered in the photo from the 5-foot height used during the experiments, separated the anesthesia and the surgical side. Those on the anesthesia side were approximately 6 inches equidistant from the center of the mask, whereas those on the surgical side were just beyond the ether screen at the center and in line with the shoulders. The bottom left panel (B) shows the connection of the supply tubing to the face mask to the anesthesia circuit using a 15-mm adapter from a 5.0 mm tracheal tube inserted into the tubing. The oxygen (O2) concentration was regulated using the anesthesia machine O2 blender. The bottom right panel (C) shows the connection of the Venturi device. The ether screen was removed in panels B and C for visualization.
O2 concentrations were measured with an Expedition-X Oxygen Analyzer (OxyCheq, Marianna, FL, USA), calibrated in room air to 20.9% before each set of replicates. The analyzer concentrations were linear between 21% and 50% (slope =1.00, intercept = 0.00, Pearson R=0.999), and the 98% response time in 50% O2 (i.e., to 48%) was 9.05 sec (standard deviation 0.40 sec). For each ambient O2 measurement, the sensor was allowed to stabilize for 60 seconds before recording the value. Baseline measurements were made near the face mask without O2 flow at the start of each sequence of recordings. Additionally, prior to each sequence of measurements, the baseline O2 concentration inside the mask was measured for each O2 level (30%, 40%, and 50%) to confirm the accurate delivery of the intended O2 concentration. Simulations with the anesthesia circuit or Venturi device were conducted three separate times, with each experiment lasting one hour. During each sequence, measurements were made at 15, 30, and 60 minutes after the start of O2 flow at the center, right, and left of the midline adjacent to the surgical side of the drape and on the anesthesia side of the drape at a radial distance of approximately 6 inches from the center of the face mask at angles of 0°, 90°, and 180°.
Differences between 20.9% and the measured values were computed to determine the increases in the ambient O2 concentration from baseline.
Statistical analysis
Analysis of variance was performed using Stata v18.5 (College Station, TX, USA), followed by Tukey’s honestly significant difference test for comparative measurements if the ANOVA F value was statistically significant at P<0.05.
Based on a concentration of 20.9% O2 in room air and a standard deviation of 0.1%, as measured during 15 room air determinations, a sample size of 6 would be required to demonstrate at least a 0.5% increase in the concentration at P <0.05 using a two-sided, two-sample t-test comparing independent means.
Results
Averages of the nine replicates at each combination of measurement location, delivery device, and time are presented in Table 1.
Table 1. Oxygen concentrations by delivery device, location, and time.
a There were N=9 measurements at each combination of side, device, and time.
| % Ambient oxygen, mean (standard deviation)a | |||||
| Ether Screen Side | Delivery Device | Baseline | 15 min | 30 min | 60 min |
| Anesthesia | Anesthesia Circuit | 20.92 (0.07) | 20.94 (0.07) | 21.11 (0.12) | 21.07 (0.12) |
| Venturi | 20.84 (0.07) | 23.20 (0.53) | 23.59 (0.47) | 22.99 (0.31) | |
| Surgical | Anesthesia Circuit | 20.88 (0.04) | 20.91 (0.03) | 20.94 (0.05) | 21.00 (0.11) |
| Venturi | 20.88 (0.11) | 21.84 (0.58) | 21.68 (0.58) | 21.72 (0.90) | |
At the anesthesia side of the ether screen, in the anesthesia circuit group, the ambient O2 concentrations adjacent to the face mask increased from 20.92% at time 0 (baseline) to 21.07% at 60 minutes. In the Venturi device group, concentrations increased from 20.84% to 22.99%.
At the surgical side of the ether screen, in the anesthesia circuit group, the ambient O2 concentrations increased from 20.88% at baseline to 21.00% at 60 minutes. In the Venturi device group, concentrations increased from 20.88% to 21.72%.
By two-way analysis of variance, there was no interaction between the time of measurement after the start of gas flow within each delivery device group and side of the ether screen and the increase in the O2 concentration from baseline at 15, 30, and 60 minutes (Figures 2, 3).
Figure 2. Change from baseline oxygen concentration using a face mask connected to the anesthesia circuit.
Mean differences between the baseline (room air) O2 concentration and measurements at the surgical and anesthesia side of the ether screen near the drape and the face mask during administration of 30% O2 by face mask at a 6 L/min flow rate via the anesthesia circuit. The error bars represent the standard deviations. Comparison p-values derived from data in Figures 2, 3 are described in the text below.
Figure 3. Change from baseline oxygen concentration using a Venturi device.
Differences between the baseline (room air) O2 concentration and measurements at the surgical and anesthesia side of the ether screen near the drape and the face mask, respectively, during the administration of O2 by face mask via a 31% Venturi device connected to the auxiliary gas outlet on the anesthesia machine set at 6 L/min of 100% O2. The error bars represent the standard deviations. Comparison p-values derived from data in Figures 2, 3 are described in the text below.
Differences between the two measuring sites (the surgical versus the anesthesia side of the ether screen) were significant (P<.001). Therefore, the three replicated time values at the three measurement times (n=9 values) were pooled, and the differences in the increases from baseline for each delivery device were compared for each side of the ether screen. At the surgical side of the screen, the mean difference in the O2 concentration increase was 0.80% (95% CI 0.53% to 1.07%, P<.001, Figure 2). At the anesthesia side of the ether screen (i.e., adjacent to the face mask), the mean increase was 2.22% (95% CI 2.07% to 2.42%, P<.001).
Discussion
Using the Venturi device resulted in a substantively greater (mean 2.2%) increase in the O2 concentration on the anesthesia side of the ether screen adjacent to the face mask over the baseline value compared to using the anesthesia circuit when delivering comparable concentrations of O2. This enhancement of the ambient O2 concentration has important ramifications when the surgical site is close to the face mask and an ignition source is used. The increase on the surgical side of the ether screen was less dramatic (mean 0.8%) but substantive because even small increases in ambient O2 can significantly increase fire risk. For example, cotton, a common material in surgical drapes and dressings, becomes more flammable in oxygen-enriched environments. Each 1% increase in the ambient O2 concentration increases the rate of combustion of cotton by 15% as compared to room air (up to ambient O2 concentrations of 35%) [7]. During the experiments, we were careful to maintain the integrity of the ether screen, but in clinical practice, that might not always be the case, with the potential for more O2 leakage into the surgical field. Thus, the 0.8% increase observed on the surgical side of the ether screen should be considered conservatively low. The nominal 1% difference between the delivered O2 concentrations cannot account for the large difference on the anesthesia side of the screen because even a 40% delivered concentration via the anesthesia circuit at 6 L/min only increased the local ambient concentration at the same distance from the center of the face mask used during the experiments by 0.60% (95% CI 0.53% to 0.67%, unpublished data). We emphasize that an ambient concentration of 30% O2 is not inherently safe; any increase above 21% substantially increases the combustion rate should a fire occur [7].
In our previous report, using the anesthesia circuit to deliver a regulated O2 concentration from the anesthesia machine O2 blender to a face mask is straightforward because high circuit pressure does not need to be achieved by closing the APL valve [5]. Because the consequences of an operating room fire are dire [1], preference for an alternative delivery method that increases the risk, even marginally, is questionable. We continue to advocate for anesthesia machine manufacturers to provide an O2 blender for the auxiliary gas outlet of all their products as a fire safety feature. Among the major producers, this feature became commercially available from Mindray (Mindray, North America, Mahwah, NJ) in 2011, followed by GE HealthCare (GE HealthCare, North America, Chicago, IL) in 2016 for some of their products. An integrated O2 blender is not currently available for Dräger anesthesia machines (Drägerwerk, Lübeck, Germany). A blender for the auxiliary gas port incorporated into the anesthesia machine is preferable to adapting O2 mask tubing to connect to the anesthesia circuit or connecting a Venturi device to a source delivering 100% O2. Venturi devices are not suitable when supplemental O2 needs to be administered via a nasal cannula because the high pressure from the tubing will limit the amount of entrained air and thus raise the O2 concentration generated by the device [8].
Limitations
Our study has several limitations. First, the tests were conducted in a simulated environment with a lower room air exchange rate than is typically provided in operating rooms [9]. This difference may have resulted in overestimating ambient concentrations measured in a clinical environment. Second, although our O2 analyzer had a fast response time (i.e., <10 seconds to 98% of the final value), it might have missed transient, localized increases in local O2 concentration. Consequently, our measurements should be considered as a minimum estimate of the actual concentrations that can occur when 30% O2 is delivered. Third, we did not test multiple Venturi devices from multiple manufacturers. Oxygen concentrations delivered will vary from the nominal concentration of the device, depending on the manufacturing tolerances of the orifice, and are also affected if an incorrect O2 flow rate is provided or if there is back pressure in the circuit distal to the device [6]. If the outflow from the Venturi device becomes inadvertently obstructed, for example, if it is located under the drapes, then room air will not be entrained and the 100% O2 feeding the Venturi device from the auxiliary O2 supply will spill out through the vents in the Venturi device, markedly increasing the local O2 concentration. These considerations create additional concern about using Venturi devices when there are fire safety risks compared to the expected, more uniform performance of digital O2 blenders on anesthesia machines.
Conclusions
When supplemental O2 via a face mask is required during upper chest and head and neck surgical procedures with an ignition source, using a Venturi device to control the delivered O2 concentration generates additional, substantive fire risk compared to using the O2 blender from the anesthesia machine. While, ideally, that would be provided via an integrated blender controlling the O2 concentration supplied by the auxiliary gas outlet of the machine, the use of the anesthesia circuit and the standard machine blender provides a viable alternative to the regulation of the administered O2 concentration. Notwithstanding these considerations, when providing sedation for procedures at an increased risk of fire, avoiding supplemental O2 if not necessary, limiting the concentration to provide a minimally acceptable pulse oximetry saturation, and considering general anesthesia with a sealed airway should be considered part of the overall fire safety strategy.
Acknowledgments
We thank the University of Miami/Jackson Memorial Hospital Center for Patient Safety (CPS), and specifically Ilya Shekhter, Roxanna Araya, Tiffany Carter, and Sebastian Estuardo Leon Zambrano for their help with the logistics of this experiment and for allowing the use of the CPS facilities.
Disclosures
Human subjects: All authors have confirmed that this study did not involve human participants or tissue.
Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.
Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following:
Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work.
Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work.
Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.
Author Contributions
Acquisition, analysis, or interpretation of data: Danielle B. Horn, Ken D. Nguyen, Richard H. Epstein, Joni M. Maga
Drafting of the manuscript: Danielle B. Horn, Ken D. Nguyen, Richard H. Epstein, Joni M. Maga
Critical review of the manuscript for important intellectual content: Danielle B. Horn, Ken D. Nguyen, Richard H. Epstein, Joni M. Maga
Supervision: Danielle B. Horn, Richard H. Epstein
Concept and design: Richard H. Epstein
References
- 1.Operating room fires: a closed claims analysis. Mehta SP, Bhananker SM, Posner KL, Domino KB. Anesthesiology. 2013;118:1133–1139. doi: 10.1097/ALN.0b013e31828afa7b. [DOI] [PubMed] [Google Scholar]
- 2.Ehrenwerth J. UpToDate. Waltham, MA: UpToDate; 2024. Fire safety in the operating room. [Google Scholar]
- 3.Reducing the incidence of surgical fires: supplying nasal cannulae with sub-100% O2 gas mixtures from anesthesia machines. Lampotang S, Gravenstein N, Paulus DA, Gravenstein D. Anesth Analg. 2005;101:1407–1412. doi: 10.1213/01.ANE.0000180215.50589.02. [DOI] [PubMed] [Google Scholar]
- 4.Practice advisory for the prevention and management of operating room fires: an updated report by the American Society of Anesthesiologists Task Force on Operating Room Fires. Apfelbaum JL, Caplan RA, Barker SJ, et al. Anesthesiology. 2013;118:271–290. doi: 10.1097/ALN.0b013e31827773d2. [DOI] [PubMed] [Google Scholar]
- 5.Nasal cannula and face mask gas flow rates when connecting to the Y-piece of the anesthesia circuit. Salas AS, Epstein RH. J Clin Anesth. 2024;95:111463. doi: 10.1016/j.jclinane.2024.111463. [DOI] [PubMed] [Google Scholar]
- 6.Supplying sub-100% oxygen gas mixtures during monitored anesthesia care: respiratory monitoring and use of a venturi device. Loeb RG. Anesth Analg. 2006;103:1048. doi: 10.1213/01.ane.0000239037.72107.a0. [DOI] [PubMed] [Google Scholar]
- 7.Fire hazards of oxygen and oxygen-enriched atmospheres. [ Nov; 2024 ]. 2018. https://www.eiga.eu/uploads/documents/DOC004.pdf. https://www.eiga.eu/uploads/documents/DOC004.pdf [PubMed]
- 8.Performance of venturi oxygen delivery devices. Fracchia G, Torda TA. Anaesth Intensive Care. 1980;8:426–430. doi: 10.1177/0310057X8000800406. [DOI] [PubMed] [Google Scholar]
- 9.Operating room air pollution: influence of anaesthetic circuit, vapour concentration, gas flow and ventilation. Mehta S, Cole WJ, Chari J, Lewin K. Can Anaesth Soc J. 1975;22:265–274. doi: 10.1007/BF03004835. [DOI] [PubMed] [Google Scholar]