The 1993 Food and Drug Administration (FDA) anesthesia apparatus checkout recommendations provide a systematic approach to checking anesthesia circle systems (1). We believe that the FDA recommendation to visually inspect unidirectional valves (UDVs) during the ventilator check for “proper action of UDVs” is inadequate. A more thorough evaluation may be warranted based on the reported incidence of UDV incompetence (Table 1) ranging from 3% to 16% (2,3).
Currently, UDV function can be tested using either the breathing method (3,4), the anesthesia machine valve tester (2) (AMVT), capnography, or the pressure decline method (5). UDVs can malfunction by either leaking (an incompetent valve) or sticking. A sticking or incompetent inspiratory valve (IV) or an incompetent expiratory valve (EV) may cause hypoventilation, whereas a sticking EV can lead to high airway pressures (4,6).
The breathing test was described by Kim et al. (3) and referenced by Dorsch and Dorsch (4). First, the tester exhales into a piece of corrugated tubing attached to the inspiratory port while occluding the expiratory port. Inflation of the reservoir bag at the bag mount indicates IV incompetence. Next, the reservoir bag is moved to the expiratory port and the corrugated tubing is moved to the bag mount site. The tester exhales into the corrugated tubing at the bag mount site while occluding the inspiratory port. Inflation of the bag indicates EV incompetence. This test evaluates only for valve incompetence, potentially contaminates the circle system, relies on respiratory effort, and has unclear end-points in terms of time or volume needed to recognize reservoir bag filling.
The AMVT (2) (Wolf Tory Medical, Inc., Salt Lake City, UT) is a product consisting of a self-inflating bulb with a 22-mm fitting. The bulb is squeezed and released appropriately to test UDV function. A strong squeeze of the bulb can produce high positive pressure (135–165 cm H2O) when testing the IV. High negative pressure (−94 to −108 cm H2O) may be generated when testing the EV. These high pressures may artificially seat a warped UDV (2). The AMVT evaluates both UDVs for incompetence and sticking. However, extra equipment must be purchased and the long-term effect of these high pressures on UDV function is unknown.
Capnography can be used to identify UDV incompetence intraoperatively. An incompetent IV creates a plateau that extends beyond end expiration (prolonged phase III) (5). This subtle finding can be difficult to detect. The capnograph tracing does not return to zero when the EV is incompetent (5). The differential diagnosis also includes exhaustion or channeling through CO2 absorbent. Capnography cannot be used during the pre-use checkout and is insensitive and nonspecific for identifying incompetent UDVs.
Many anesthesia machines have reverse flow alarms that can identify UDV incompetence of one or both valves intraoperatively. The alarm sounds when a flowmeter or spirometer detects reverse flow through the inspiratory or expiratory limbs of the breathing circuit. Although these alarms might be helpful in detecting reverse flow, their sensitivity and specificity are unknown and they are not present on both limbs of most anesthesia machines.
Kitagawa et al. (5) described a two-step procedure to test for UDV incompetence which was referenced by Dorsch and Dorsch (4) as the “pressure decline method.” First, the inspiratory port is occluded and the system pressurized to 30 cm H2O. The airway pressure gauge is monitored for declining pressure which indicates a leaking EV. Next, a reservoir bag filled with air is connected to the inspiratory port and, if possible, another pressure gauge placed between the IV and reservoir bag. Declining pressure indicates a leaking IV. Dorsch and Dorsch (4) modified the second step, advising: “a reservoir bag is connected to the inhalation port and filled using the oxygen flush. The pressure in the bag should be maintained.”
We independently developed a technique to test UDV function that is essentially a modified version of the pressure decline method. The modified pressure decline method (MPDM) uses two reservoir bags, does not require port occlusion, uses the adjustable pressure limiting (APL) valve, and does not require a second pressure sensor. The pressure decline method and the MPDM test both valves for incompetence but they test only the IV (not the EV) for sticking.
The MPDM test requires an extra 3-L reservoir bag (Fig. 1) and takes <1 min to perform. The test must be performed after routine system leak checks.
- Step 1: Place an extra 3-L reservoir bag at the inspiratory port.
- Step 2: Minimize fresh gas flow and close the APL valve. Pressurize the circle system to 30 cm H2O using the oxygen flush. Both reservoir bags should inflate. A stuck IV prevents bag inflation at the inspiratory port. For 20 s, observe the reservoir bag at the bag mount site for wrinkles indicating deflation secondary to EV incompetence.
- Step 3: Open the APL valve to decompress the circle system. For 20 s, observe the extra reservoir bag at the inspiratory port for wrinkles indicating deflation secondary to IV incompetence.
Thirty-six UDVs in 18 anesthesia circle systems were tested with the MPDM test (Table 2). Visual deflation was defined as the appearance of wrinkles in the reservoir bag. Trident Medical-Meridian Medical Systems (Indianapolis, IN) manufactured the reservoir bags and pressure tested a random sample of them at 50 cm H2O in accordance with industry standards (ISO 5326).
Disk Flow Leak Calibration
We constructed Draeger (Draeger Medical Inc., Telford, PA) and Ohmeda (Datex-Ohmeda Inc., Madison, WI) machine-specific incompetent valve discs by drilling 1- to 2-mm holes into standard valve discs (Fig. 2). These valve discs were used only for testing and were stored outside the clinical area. The degree of valve incompetence (“flow leak”) was defined by the fresh gas flow allowed through the incompetent valve disk. The Draeger incompetent valve disk was placed in the expiratory limb of a Draeger Narkomed 2B and the gas flow maximized as a starting point. Gas flow was then incrementally decreased until the circle system pressure was just above zero. This fresh gas flow equaled the gas leak through the valve at a pressure differential across the valve of just above zero. This gas flow exactly replaced the loss through the leaking valve and was termed “flow leak.” To expose the Ohmeda pressure sensor to the circle system, the IV was removed, the inspiratory port occluded, and testing of the Ohmeda incompetent valve disk performed as above.
The MPDM test was performed on one Draeger and one Ohmeda machine using the machine-specific incompetent valves.
Thirty-six UDVs in 18 anesthesia machines were tested using the MPDM. One incompetent EV (3%) was identified in a Draeger Julian machine (Table 2). No incompetent or sticking IV was detected.
The Draeger and Ohmeda-specific incompetent valve disk flow leaks were respectively 0.6 and 0.9 L/min.
When the machine-specific incompetent valve discs were placed in one Draeger and one Ohmeda machine, the MPDM test identified the leak in the inspiratory and expiratory positions.
We believe the MPDM has several advantages compared with other methods for detecting incompetent UDVs. The MPDM is quick and easy to perform during the pre-use machine checkout, requires only an extra reservoir bag, and avoids contamination. The extra reservoir bag may be returned to the set and used later, similar to the extra bag the FDA recommends for the ventilator check. The MPDM end-point is clearly defined as wrinkles in the bag and this test is specific to valve function in a leak-free system. Although the true sensitivity of the test is unknown, we have shown it is sensitive to an incompetent valve with a flow leak of 0.6 L/min. It tests the bidirectional function of the IV but only competence of the EV. The true sensitivity of the test is unknown, therefore it is possible that a minimal but acceptable leak during the leak test (e.g., <100 mL/min leak is typically accepted clinically) could produce a false-positive EV result. A false-negative leak test or MPDM test result is possible for machines that have a minimal fresh gas flow (e.g., 200 mL/min for the Narkomed 2B and Narkomed GS), which may be more than the circle system leak or the incompetent valve flow leak. The accuracy and sensitivity of the MPDM are dependent on the magnitude of the leak deemed acceptable for the leak test and the minimal gas flow for a given machine.
Pressurizing the circle system to 30 cm H2O produces a relatively fast leak through the incompetent valve and tests the valves at a pressure sufficient to ventilate most patients. This is the same pressure the FDA recommends during the leak test (1). Excessive pressure could seat and seal a warped disk, thereby masking UDV incompetence.
The MPDM test provides a simple, safe, and effective way to regularly test UDVs. It is important to remember that this test is intended as an adjunct to the current FDA anesthesia apparatus checkout recommendations. Because the 1993 FDA checkout recommendations already specifically require a second reservoir bag to perform the ventilator check, adding one more use for this bag can easily produce a more thorough circle system checkout.
The authors thank Dr. Cecil Borel and the reviewers at A&A for their significant contributions to the article in both scientific content and presentation.
1. Anesthesia apparatus checkout recommendations 1993. Anesthesia Patient Safety Foundation newsletter. Park Ridge, IL: APSF, Fall, 1994:35.
2. Eappen S, Corn SB. The anesthesia machine valve tester: a new device and method for evaluating the competence of unidirectional anesthetic valves. J Clin Monit 1996;12:305–9.
3. Kim JM, Kovac AL, Hugh MS. A method for detection of incompetent unidirectional dome valves: a prevalent malfunction. Anesth Analg 1985;64:745–7.
4. Dorsch JA, Dorsch SE. Understanding anesthesia equipment. 4th ed. Baltimore: Williams & Wilkins, 1999:953–6.
5. Kitagawa H, Sai Y, Nosaka S, Amakata Y. New leak test for specifying malfunctions in the exhalation and inhalation check valve [letter]. Anesth Analg 1994;78:611.
6. Ehrenwerth J, Eisenkraft JB. Anesthesia equipment principles and applications. 1st ed. St. Louis: Mosby, 1993:239–40.