To determine how the Aisys and 2 other currently available workstations detect and respond to breathing circuit obstruction, a fault was created. Layers of plastic material (from the wrap enclosing the facemask) were used to completely occlude the junction between the corrugated expiratory limb of the plastic portion of the breathing circuit, and its mount on the machine (Fig. 3). The automated checkout was performed, noting the ability of the workstation to detect the occlusion, and how it responded. The response of these machines to an inspiratory limb occlusion, created in similar fashion, was also tested. This simulation was performed on the Aisys, ADU, and Apollo (Drager Medical Inc., Telford, PA).
Although it was assumed that all examples of a particular workstation model would respond identically to a breathing circuit obstruction, repeated trials were performed to ensure that the responses observed were not due to the particular workstation used. Five examples of each model were tested with inspiratory and expiratory obstructions. Statistical analyses were not performed.
In the presence of an expiratory limb obstruction, the Aisys displayed an error message in the course of its electronic checkout, indicating that it could not measure circuit compliance and suggested “check flow sensors.” However, it allowed the user to accept this condition and initiate patient care (Fig. 4, Table 1). A simulated case was started using a spare breathing bag attached at the elbow as a test lung, with fresh gas flow 6 L/min, tidal volume 600 mL, and respiratory rate 10 breaths/min in volume control mode (Fig. 5). With a complete expiratory limb occlusion, the mechanical ventilator can inflate the simulated lung, but gas cannot escape from it. Within a very brief interval (<2 minutes), peak, mean, and end-expiratory pressure increased to approximately 40 cm H2O, high-priority alarms sounded, indicating high peak pressure, and the message “PEEP high. Blockage?” appeared (Fig. 5). Note that these alarms did not appear during performance of the automated checkout, but only after simulated patient care had been initiated. Similar results were obtained when initiating a simulated case using the manual limb of the breathing circuit. Inflation of the simulated lung was possible by squeezing the breathing bag, but gas could not escape from the test lung, and the same alarms followed promptly.
With an inspiratory limb occlusion, the Aisys failed its electronic checkout earlier, because the bellows could not empty through the inspiratory limb. On repeating the test, the ventilator check failed again, and the Aisys displayed a visual reminder to “ventilate manually!” However, the Aisys again allowed the user to start a case (Table 1).
The responses of the ADU and Apollo were quite different (Table 1). With the expiratory limb occluded, the ADU error message used the word “failed” twice. With an inspiratory limb occlusion, the error message in the ADU referred to “probable disconnection.” In one of 5 trials of the ADU with an inspiratory limb obstruction, the wording of the error message varied slightly, referring to leakage. In the Apollo, one of 2 error messages occurred whether the inspiratory or expiratory limb was occluded (see Table 1). One of these (“System error. Device cannot be used”) was unrecoverable, requiring the user to turn the workstation off, and then on, before it would clear. The other error message identified the source of the problem as a leak in the ventilator or breathing circuit and allowed the user to repeat the test. The occurrence of one or the other error was not related to which valve was occluded, which valve was occluded first, or which error message had previously appeared. In 3 cases, the same error appeared for both occlusions; in the other 2 cases, inspiratory and expiratory occlusions caused the appearance of one of each of the 2 error messages. Neither the ADU nor the Apollo allowed the user to proceed to the next step in the automated checkout in the presence of occlusion in either limb. Neither the ADU nor the Apollo allowed the user to start a case.
Most current workstations have a checkout that is partly automated, and partly check reminders for actions that must be performed manually by the user. It can be difficult to determine exactly what is checked during the automated portion of the checkout.16,17 If tests for breathing circuit occlusion are not performed before each case, a patient may be induced using a circuit that cannot ventilate a patient or in which a high, sustained pressure builds very quickly. Add this emergent mechanical difficulty to the normal physiologic instability and high task burden associated with the first moments after induction, and a crisis occurs that demands a quick and accurate response.18–20 The differential diagnosis includes complete bronchospasm, esophageal intubation, and a kinked endotracheal tube.
In a recent report from the Closed Claims database (covering closed claims from 1990 to 2011), problems with breathing circuits represented 20% of claims related to gas delivery equipment.21 In half of the breathing circuit claims, ventilation by self-inflating manual ventilation device (Ambu) was not available, or intitiated too late, and the outcome was death or brain damage. In half of these 8 claims, difficult ventilation was misdiagnosed as bronchospasm. Most of the breathing circuit claims (6 of 8) were judged preventable with a proper preanesthesia checkout.21
The current Pre-Anesthesia Checkout (2008 PAC) requires users to check that gas flows properly through the breathing circuit during both inspiration and exhalation. This check must be performed not only at the start of the day but also before each case.1 The 2008 PAC suggests using a test lung attached at the elbow, during both manual and mechanical ventilation. Two other means for detecting occlusions or malfunctioning valves have been suggested. The user can check for obstruction in the circle breathing system by breathing through it.22 However, this presents a danger of disease transmission and increases occupational exposure to gases and vapors. Providers can also check for expected breathing bag movement at the start of denitrogenation, immediately after placing the mask over the patient’s face.23 This approach leaves little time to correct errors, but at least, an obstruction may be caught before induction (and the cessation of spontaneous respiration).
In conclusion, although manufacturers’ automated electronic checkout procedures in 3 current anesthesia workstations all detected an occluded inspiratory or expiratory limb of the breathing circuit, they varied in how clearly they alerted the user and whether they allowed a case to be initiated. The Aisys, unlike ADU and Apollo, allowed the user to accept the condition and initiate care. Users should be aware of the differences between checkout routines and develop local guidelines to ensure that anesthesia providers can properly interpret the results of the self-tests and can perform a test for breathing circuit occlusion independent of the automated checkout.
Name: Michael P. Dosch, CRNA PhD.
Contribution: This author helped design and conduct the study, analyze the data, and write the manuscript.
Attestation: Michael P. Dosch approved the final manuscript.
Conflicts of Interest: I have no ongoing financial relationship with the manufacturers of any devices mentioned here. I have presented at state and national meetings with sponsorship by GE in the past. I served as an invited consultant to Dräger Medical on 1 occasion.
This manuscript was handled by: Maxime Cannesson, MD, PhD.
3. Monteiro JN, Ravindran MN, D’Mello JB. Three cases of breathing system malfunction. Eur J Anaesthesiol. 2004;21:743–5
4. Chacon AC, Kuczkowski KM, Sanchez RA. Unusual case of breathing circuit obstruction: plastic packaging revisited. Anesthesiology. 2004;100:753
5. Joyal J, Vannucci A, Kangrga I. High end-expiratory airway pressures caused by internal obstruction of the Draeger Apollo(R) scavenger system that is not detected by the workstation self-test and visual inspection. Anesthesiology. 2012;116:1162–4
6. Ramarapu S, Ramakrishnan U. An unusual cause of anesthesia circuit blockade by Sodasorb Pre-Pak Anesth Analg. 2012;115:477
7. Wise HJ. ABC: back to basics with anaesthetic breathing components. Anesthesia. 2002;57:86
8. Eckhout GV, Bhatia S. Another cause of difficulty in ventilating a patient. J Clin Anesth. 2003;15:137–9
9. . ECRI. Hazard report: Injuries—one fatal—highlight the need for pre-use testing of disposable breathing circuits Health Devices. 2000;29:188–9
11. Thorpe CM. Plastic in anaesthetic circuit. Anesthesia. 2002;57:85–6
12. Wagner K, Loy J. A strange place to find a cable tie. Anesth Analg. 2006;102:655–6
14. Datex-Ohmeda. Datex-Ohmeda Anesthesia Delivery Unit user’s reference manual, Catalog No. 8502304-2 Madison, WI: Datex-Ohmeda. 2005
15. GE Healthcare Aisys 6.x Carestation Training Document DOC0675696 REV A. Madison, WI: GE Healthcare.
16. GE Healthcare. Aisys User’s Reference Manual Software Revision 8.X: Document No. M1203934. 2011;11 Madison, WI GE Healthcare
17. Venticinque S, Eichhorn JEhrenwerth J, Eisenkraft J, Berry J. Machine checkout and quality assurance. Anesthesia Equipment: Principles and Applications. 20132nd ed Philadelphia, PA Elsevier/Saunders,:653–81
18. Chaney MA. Delivery of excessive airway pressure to a patient by the anesthesia machine. Anesth Analg. 1993;76:1166–7
19. Bourke DL, Tolentino D. Inadvertent positive end-expiratory pressure caused by a malfunctioning ventilator relief valve Anesth Analg. 2003;97:492–3
20. Rieker M. Anesthesia machine perils and pitfalls: how well do you know your best friend? Curr Rev Nurse Anesth. 2005;28:81–92
21. Mehta SP, Eisenkraft JB, Posner KL, Domino KB. Patient injuries from anesthesia gas delivery equipment: a closed claims update. Anesthesiology. 2013;119:788–95
22. Dorsch JA, Dorsch SE Understanding Anesthesia Equipment. 20085th ed Philadelphia, PA Wolters Kluwer Health-Lippincott Williams and Wilkins,
© 2014 International Anesthesia Research Society
23. Dosch MNagelhout J, Plaus K. Anesthesia equipment. Nurse Anesthesia. 2014;2425th ed St. Louis, MO Elsevier/Saunders,:91