Although the large majority of anesthesia critical incidents and catastrophes involve human error, some events involve overt equipment failure or failure of the anesthesia provider to discover an equipment problem. Most equipment problems in anesthesia practice are preventable, and this chapter is intended to help practitioners achieve that goal.
The analogy between administering general anesthesia and piloting a commercial jet may be somewhat overused, but it is singularly relevant in this context. The aviation industry has developed extraordinarily thorough plans involving “acute” and “chronic” interactions with its principal equipment, the commercial passenger jetliner. The acute component is the immediate preflight check to verify that a particular aircraft should fly safely that day on a given trip; the chronic component is the elaborate scheme of scheduled preventive maintenance, repair, exchange of old parts for new, and safety inspections of structural components. These are all oriented toward ensuring that the aircraft will fly safely for the designated interval of weeks or months covered by that particular action. In anesthesia practice, the analogy is obviously appropriate to an anesthesia equipment quality assurance (QA) program. The acute effort is the preanesthetic equipment check, and the chronic component is the vital and ongoing QA mechanism that involves preventive maintenance, testing for safe function, and the detection of expected wear prior to the failure of a piece of equipment.
Preflight checkout procedures in aviation have changed as technology has advanced. Aircraft systems have become more automated, and modern “glass cockpit” instrument displays have reduced the workload involved in operating multiple complex systems. This has allowed two crew members to do work that previously required three or more, with increased reliance on automated systems that function with minimal pilot input. Anesthesia equipment has also evolved rapidly, with increasing reliance on automated systems to function properly with minimal input by the anesthesia professional. Nevertheless, human vigilance will remain important as the ultimate watchdog and guarantor of machine safety.
Because the practice of anesthesia is heavily dependent on the correct functioning of a large number of diverse pieces of equipment, and because anesthesia professionals usually have technical and mechanical proclivities, reports of problems with anesthesia equipment have been prominent in the anesthesia literature virtually since its inception. A great many of the “classic” traditional problems that had been common since recognizable anesthesia machines came into use—such as fresh gas rotameter leaks, ventilator leaks, and disconnections of poorly designed hose or tubing connectors—have been largely eliminated by the adoption and implementation of rigorous so-called voluntary design and fabrication standards by anesthesia machine manufacturers. By no means does this suggest that equipment problems do not arise today. On the contrary, some problems—such as absent, broken, or stuck unidirectional breathing system valves, or failure to remove the wrapper on prepackaged carbon dioxide absorbent inner canisters—are still a concern today. Also, because of the increasing complexity and integration of multiple parts and functions that had been separate (replaceable) components of anesthesia delivery systems, scrupulous attention to acute and chronic QA of anesthesia equipment is more important than ever. However, many of the details and the specific problems have changed, and advances have been made in the study of human interaction with the anesthesia system. Improving the “user friendliness” of anesthesia technology and study of the human factors in both routine and crisis anesthesia situations has been made possible in part by major growth and development of almost frighteningly realistic high-fidelity patient simulators—mannequins in particular, but also screen-based computer programs—that enable trials and modification of equipment and protocols without patient risk. Further, current simulators allow specific training of anesthesia professionals to deal with extremely rare catastrophic situations. Although required to be familiar with and prepared to react to these challenges, everyone hopes to never encounter them in real life, such as crossed oxygen and nitrous oxide pipelines.
Although historic perspectives are valuable with regard to anesthesia equipment, current concerns are more important than reproducing a litany of hundreds of problems that have been identified with anesthesia equipment over the years. Classic treatises on anesthesia equipment provide useful references for reviewing the spectrum of defects and problems that have been reported with anesthesia equipment. Rendell-Baker edited a classic monograph that described 48 specific safety-related problems with anesthesia machines. Among these, in order of frequency, were problems with 1) the vaporizer, 2) the breathing system, 3) the gas flowmeters, 4) the mechanical functions of the machine, and 5) human engineering. In another classic, Spooner and Kirby outlined some of the data collected by the ECRI Institute regarding the role of equipment in anesthesia accidents. In the American Society of Anesthesiologists (ASA) Closed Claims Study, one report noted that only 2% of claims were apparently caused by the gas delivery equipment; however, 76% of those involved catastrophic adverse outcomes. These data again suggest that a combination of the occasional overt device failure along with a large component of various types of human error lead to anesthesia mishaps.
Further, classic pioneering work by Cooper and associates suggested that among anesthesia critical incidents, 82% involved human error, and only 14% resulted from overt equipment failure. Of course, many of the human errors involved unrecognized problems with the equipment, such as breathing system disconnection, that were not classic equipment failures—and that would not necessarily have been prevented or mitigated by the preanesthetic checkout (PAC) of the anesthesia equipment. Among the equipment failures, 20% involved the breathing circuit, 18% airway components, 12% laryngoscopes, and 12% the anesthesia machine. Failure to perform a normal checkout was cited 22 times on a list of 481 factors associated with 359 incidents. Follow-up studies of 1089 preventable critical incidents found that only 4% of incidents with substantive negative outcomes involved equipment failure. Of all incidents reported in the various parts of the study, 11% to 19% involved equipment failure. However, 129 (22%) of 583 instances of human error involved anesthesia machine use, indicating that the interaction of the anesthesiologist with normally functioning equipment accounts for many problems. Minimizing this type of problem by eliminating defects in the delivery system prior to the start of any anesthetic is the goal of a thorough preuse anesthesia equipment checkout.
Many other studies have also implicated the failure to perform an adequate PAC of the equipment as a factor in critical incidents and accidents. For example, 1 of the 11 cases reported in the analysis of severe anesthesia injuries at the Harvard teaching hospitals was caused by a misconnection of a vaporizer that had just been returned from servicing, a condition that would have been detected by a thorough preuse checkout of the anesthesia equipment.
It is a well-accepted dictum that a thorough and stepwise anesthesia apparatus checkout should be performed prior to the delivery of anesthesia. Even in surveys that identify poor provider compliance with PAC procedures, most participants feel that such checks improve patient safety. This perception is indeed correct, as it has been clearly demonstrated that a PAC performed with a checklist and protocol is associated with a decreased risk of perioperative morbidity and mortality. It is important to note that a PAC is a checklist . The obvious industry parallel is aviation, in which a strict adherence to pre-event checklists (e.g., before start, at takeoff, on approach) is known to enhance compliance with important steps and procedures and to save lives. The unfortunate difference between anesthesiology and aviation is that pilots seem to use checklists much more regularly, possibly in part because the pilot of an airplane is always the first one to a crash scene. Through leadership and the personal effort of many, great progress has been made over the past decades in enhancing the knowledge of the anesthesia workstation and in introducing anesthesia equipment preuse checkout guidelines. What is striking, however, is that despite these efforts, our overall performance in “preflighting” the anesthesia workstation seems to be less than optimal. This is both troubling and perplexing given the pivotal role that the anesthesia machine plays in anesthesia practice and patient safety, and the fact that anesthesia professionals, by nature, are savvy about and comfortable with technology.
Data predating the first publication of the U.S. Food and Drug Administration (FDA) Anesthesia Apparatus Checkout Recommendations in 1986 demonstrated a low level of proficiency by anesthesiologists in detecting life-threatening machine problems. Using a machine with five intentionally created faults, researchers found that anesthesiologists detected on average only 2.2 serious problems (44%); and 7.3% found no faults at all despite knowing ahead of time that the machine was intentionally altered. At that time, available preuse checkout procedures for anesthesia machines were provided and promoted by individual machine manufacturers. Given the design and engineering perspectives of the manufacturers and their liability concerns, the machine-specific checkout recommendations were not entirely “user friendly,” nor were they well suited for clinical application. This also parallels current times, when manufacturer checkout recommendations still tend to be long and sometimes unwieldy, making them difficult to use routinely.
Prompted by a series of anesthesia machine–related accidents, in 1984 the FDA met with representatives from the ASA, anesthesia equipment experts, and anesthesia machine manufacturers to discuss methods of reducing patient risk during anesthesia. During that meeting the FDA was asked to take the lead in the development of the first generic anesthesia apparatus checkout recommendations. This general guideline was intended to instruct users on how to perform a preuse checkout, to promote the concept of a preuse checkout, and to create a framework that providers could modify to meet their local needs. This first preuse checkout was released in final form in August of 1986. These recommendations contained 25 primary items, with some having up to six subitems. The guideline intended that a comprehensive checkout be accomplished at the beginning of the day (“day check”) followed by an abbreviated check prior to subsequent cases (“case check”). The 1986 PAC guideline was fairly detailed, proved to be very time consuming, and was found to be not extensively used.
In an attempt to evaluate compliance with the 1986 checklist, the FDA surveyed anesthesiology providers at 125 hospitals in four states and found that only 70% of facilities acknowledged they had a documented anesthesia machine preuse checkout at their site. Only 73% of the sites reported that a preuse checkout was routinely carried out at the beginning of each day or shift, and only 59% reported that a preuse checkout was accomplished between cases. Of those hospitals that acknowledged having a documented checkout, only 26% used the FDA’s version. The issuance of the 1986 recommendations also did not appear to improve the ability of anesthesiologists to detect anesthesia machine faults. Testing this hypothesis, March and Crowley showed that 188 anesthesiologists, using their own methods, could detect only one of four preset faults. When the same subjects used the FDA checklist to asses a different set of failures, detection performance improved only modestly. The authors concluded that the introduction of the FDA checklist did not improve the ability of anesthesiologists to detect anesthesia machine faults.
Recognizing that there was poor compliance with the 1986 guideline, the FDA revised the PAC in the early 1990s, working once again with anesthesiology professional organizations and industry. Other factors, including the development of monitoring standards by the ASA, the retirement of many older machines, and the introduction of newer-generation anesthesia machines following the new American Society for Testing and Materials (ASTM) specifications also provided impetus for the update of the guidelines. The revised checklist was issued in 1993 and included only 14 major steps, many with several bulleted instructions ( Fig. 32-1 ). Although the checklist was fairly comprehensive and universal, a stated intent of the authors, similar to the 1986 version, was to encourage users to “modify [the guideline] to accommodate differences in equipment design and variations in local clinical practice” and to subject modifications to local peer review. It also encouraged users to “refer to the operator’s manual for the manufacturer’s specific procedures and precautions,” particularly when addressing the machine’s low-pressure system leak test. Even without modification, the 1993 PAC was applicable to most machines of the day and was nicely formatted to fit onto a single page. It is interesting to note that like the 1986 guideline, the FDA did not mandate the use of the 1993 Anesthesia Apparatus Checkout Recommendations. It was made clear in the Federal Register in 1994 that the FDA recommendations only offered guidance and encouraged modification to accommodate differences in equipment design and variations in local clinical practice. Even though the FDA authorized the matter, it did not undertake direct regulatory action.
Although the data are limited, evidence that the 1993 PAC recommendations led to improved user compliance and better detection of machine faults was not forthcoming. When anesthesiology providers of different backgrounds and experience levels were asked to use their own anesthesia preuse checkout procedures to check a fault-laden machine, and then went on to check another sabotaged machine using the 1993 FDA checkout procedure, researchers detected no difference in the rate of fault detection using either method. In fact, despite having the FDA checklist in hand, 41% of the participants could not identify more than 50% of the faults. Using a prospective crossover design, Blike and Biddle found that anesthesiology providers missed “easy” anesthesia machine faults 30% of the time and “difficult” anesthesia faults 62% of the time when provided with the FDA checklist. Larson and colleagues observed 87 participants at a “nationally attended anesthesia meeting” when they were asked to perform a checkout on an anesthesia machine with preset faults. The average number of faults detected by all participants was 3.1 of 5 total faults. Interestingly, the authors showed a negative correlation between level of experience and the ability to detect faults. In other controlled settings, where anesthesiologists would be anticipated to be thorough and accurate regarding the checkout procedure, the data also indicate poor performance. Olympio and colleagues observed anesthesiology residents checking out the machine and noted a low performance rate (69%), which improved by only 12% after focused instructional review. It is important to note that the residents knew in advance that their performance would be evaluated. In another experimental setting, Armstrong and colleagues observed anesthesiologists in a simulator who knew only that they were involved in a study to evaluate the simulator as a testing tool, and they were aware that simulated patient or technical problems would be presented during the case. The researchers quietly graded the quality of the anesthesia checkout and found that the subjects, on average, checked 50% or fewer of 20 key items. Performance was noted to be poor regardless of the age or experience of the anesthesiologist.
As tempting as it may be to implicate the checklists in these failures, human factors and training issues are more likely to blame. In particular, a lack of cultural discipline in the routine, proper use of a PAC is seemingly the principal problem. As noted, pivotal work by Cooper and colleagues demonstrated that in 22% of equipment-related mishaps, a failure to check or inspect was identified as an associated factor. Similarly, in a 1981 survey of anesthetic misadventures, human error was found to be more often responsible than equipment failure, and a failure to perform a machine checkout was the factor most likely associated with an equipment-related issue. In 1992 Mayor and Eaton found that almost 41% percent of anesthesia providers admitted to performing inadequate machine checks, and few followed published guidelines. In an Internet-based survey of anesthesiology providers and anesthesia technicians published in abstract form in 2005, 29% of respondents rated their competence in performing the 1993 FDA preuse checkout as poor. Reasons cited in the same survey for skipping the checkout included unfamiliarity with the procedure, a belief that the machine self-check alone was sufficient, and that checkout took too long to perform. Finally, in a 2007 survey conducted in the United Kingdom, researchers found that most anesthetists admitted to only partially checking the anesthesia machine; only 12% performed a check between cases, and only 27% identified an alternate means for ventilation prior to anesthesia.
It seems that no matter how well conceived and heavily promoted PAC recommendations have been, their adoption and routine use has not been consistent. Underlying this issue seems to be an inconsistent understanding of the anesthesia machine. What currently compounds the issue is the growing assortment of anesthesia machine designs and features that depart significantly from the more generic, older-generation gas machines and workstations. When these factors are combined with a misunderstanding of and overreliance on “automated” machine checkout functions, the potential for a suboptimal PAC is significant.
To improve PAC compliance and performance, it is recommended that individual anesthesia departments align the ASA’s most recent 2008 Guideline for Designing Pre-Anesthesia Checkout Procedures with their respective manufacturer’s suggested checkout procedures in order to develop their own effective, workstation-specific PAC checklists ( Table 32-1 ). In addition to developing effective PACs, it is also important that anesthesia providers remain knowledgeable about their equipment and embrace a “checklist culture.” The 2008 guideline can be found at www.asahq.org for members within the “Standards, Guidelines, and Statements” tab (under Recommendations and Clinical Management Tools, ASA Committees: Anesthesia Machine Preoperative Checkout Procedures section) along with workstation-specific PACs from individual departments that can be used for PAC development. 28
|To Be Completed Daily|
|1||Verify auxiliary oxygen cylinder and self-inflating manual ventilation device are available and functioning||Provider and technician|
|2||Verify patient suction is adequate to clear the airway||Provider and technician|
|3||Turn on anesthesia delivery system and confirm that AC power is available.||Provider or technician|
|4||Verify availability of required monitors, including alarms||Provider or technician|
|5||Verify that pressure is adequate on the spare oxygen cylinder mounted on the anesthesia machine||Provider and technician|
|6||Verify that the piped gas pressures are ≥50 psig||Provider and technician|
|7||Verify that vaporizers are adequately filled and, if applicable, that the filler ports are tightly closed||Provider only|
|8||Verify that there are no leaks in the gas supply lines between the flowmeters and the common gas outlet||Provider or technician|
|9||Test scavenging system function||Provider or technician|
|10||Calibrate, or verify calibration of, the oxygen monitor and check the low-oxygen alarm||Provider or technician|
|11||Verify carbon dioxide absorbent is not exhausted||Provider or technician|
|12||Perform breathing system pressure and leak testing||Provider and technician|
|13||Verify that gas flows properly through the breathing circuit during both inspiration and exhalation||Provider and technician|
|14||Document completion of checkout procedures||Provider and technician|
|15||Confirm ventilator settings and evaluate readiness to deliver anesthesia care (anesthesia time out)||Provider only|
|To Be Completed Before Each Procedure|
|1||Verify patient suction is adequate to clear the airway||Provider and technician|
|2||Verify availability of required monitors, including alarms||Provider or technician|
|3||Verify that vaporizers are adequately filled and, if applicable, that the filler ports are tightly closed||Provider|
|4||Verify carbon dioxide absorbent is not exhausted||Provider or technician|
|5||Perform breathing system pressure and leak testing||Provider and technician|
|6||Verify that gas flows properly through the breathing circuit during both inspiration and exhalation||Provider and technician|
|7||Document completion of checkout procedures||Provider and technician|
|8||Confirm ventilator settings and evaluate readiness to deliver anesthesia care (anesthesia time out)||Provider|
2008 Recommendations for Preanesthesia Checkout Procedures
The 2008 recommendations were developed with the knowledge that the existing PAC was not well understood, nor was it reliably used by anesthesia providers, and that anesthesia delivery systems have evolved to the point where one checkout procedure is no longer universally applicable. Although anesthesia providers were encouraged to modify the 1993 PAC to meet their own equipment needs, it was essentially applicable to almost any machine when it was published. As stated, it was a nearly universal checklist. However, diligent users of the 1993 PAC came to realize that it became increasingly difficult to strictly apply as a newer generation of machines began to emerge. The newer generation machines differed in their functions and features and in their checkout procedures, and they have become increasingly diverse even among themselves. For precisely these reasons, the authors of the ASA’s 2008 Recommendations for Pre-Anesthesia Checkout Procedures created a template to develop “checkout procedures that are appropriate for each individual anesthesia machine design and practice setting,” instead of a detailed PAC. Their goal was to provide guidelines applicable to all machines, so that individual departments could develop their own PAC, which could be performed consistently and expeditiously. In fact, the footer of the document reads “Guideline for Designing Pre-Anesthesia Checkout Procedures.”
The 2008 recommendations warn against an overreliance on automated machine checkouts, alerting that anesthesiology providers may be unaware of what is actually assessed by these features and that they may omit important preuse checkout items if they place all their faith in an automated checkout. When developing a local PAC, a detailed understanding of what is actually checked by the machine is required. However, this is not always easy to ascertain by simply reviewing user manuals.
The 1993 version of the PAC placed all of the responsibility of the preuse checkout on the anesthesia provider. The authors of the 2008 guidelines recognized that using anesthesia technicians and/or biomedical technicians to perform some aspects of the checkout procedures may improve compliance with a department’s PAC and could add redundancy to critical steps. Although the 2008 guidelines suggest which steps may be checked by “a qualified anesthesia technician, biomedical technician, or manufacturer-certified technician,” this should indeed be an institutional decision, because skill levels, work flow patterns, and training requirements vary greatly. The 2008 guidelines did not intend to make the use of technician checks mandatory. Regardless of who participates in the PAC, the anesthesia care provider is ultimately responsible for the proper and safe functioning of the equipment.
The items and rationale statements listed below, excerpted directly from the 2008 Recommendations for Pre-Anesthesia Checkout Procedures, are intended to describe a basic approach to developing sound institution-specific PAC procedures designed “for the equipment and resources available.” They identify items that need to be checked as part of a complete PAC. The method used to check each item will be dependent upon the specific equipment. Also identified in the recommendations are the suggested frequency of the checks, and the suggested responsible parties either individually, alternatively (“or”), or redundantly (“and”). It is important to recognize that the guidelines are not all inclusive; they simply suggest the minimum machine-related items that should be assessed prior to use. A local PAC checklist should represent a workable merger between these guidelines and the manufacturer’s checkout recommendations. As in the prior PAC guidelines, items that require checkout prior to each procedure are distinguished from those that need only to be checked daily.
Minimum PAC Checklist
Item 1: Verify auxiliary oxygen cylinder and self-inflating manual ventilation device are available and functioning
Responsible Parties: Provider and technician
Failure to be able to ventilate is a major cause of morbidity and mortality related to anesthesia care. Because equipment failure with resulting inability to ventilate the patient can occur at any time, a self-inflating manual ventilation device (e.g., Ambu bag) should be present at every anesthetizing location for every case and should be checked for proper function. In addition, a source of oxygen separate from the anesthesia machine and pipeline supply, specifically an oxygen cylinder with regulator and a means to open the cylinder valve, should be immediately available and checked. After checking the cylinder pressure, it is recommended that the main cylinder valve be closed to avoid inadvertent emptying of the cylinder through a leaky or open regulator.
This step was item 1 on the 1993 PAC and remains so in the 2008 recommendations. It is the most important item on the checklist. No matter what happens to the machine, you should always be prepared to keep the patient alive without it. The auxiliary ventilation device should be self-inflating, which would exclude the disposable Mapleson circuits often found in and out of the operating room (OR); these devices should be located at “every anesthetizing location,” and the guideline further recommends that they be checked for proper function ( Fig. 32-2 ). The recommendation also states that the auxiliary oxygen source should be separate from the machine and its pipeline supply, “specifically an oxygen cylinder.” Ensuring that properly filled portable cylinders with attached flowmeters are available at specific locations requires an institutional logistic commitment and careful attention to detail by support staff. Incorporating technicians into this step would likely be very useful.
Item 2: Verify patient suction is adequate to clear the airway
Frequency: Prior to each use
Responsible Parties: Provider and technician
“Safe anesthetic care requires the immediate availability of suction to clear the airway if needed.”
This step moved up from the last position on the 1993 PAC recommendations. Suction is critically important to anesthesia care, because it is the only major piece of equipment used routinely by anesthesiologists whose function cannot be replaced in a life-threatening crisis by the anesthesiologist’s own body. An anesthesiologist can monitor, ventilate, and even intubate if necessary without any equipment at all. An anesthesiologist cannot, however, adequately clear a pharynx full of secretions or vomitus without an adequately functioning suction; thus suction is a genuinely vital piece of anesthesia equipment. One simple way to check the suction is to determine whether there is enough negative pressure for the tubing to attach to the operator’s finger and support its own weight while suspended in the air ( Fig. 32-3 ).
Item 3: Turn on anesthesia delivery system and confirm that AC power is available
Responsible Parties: Provider or technician
Anesthesia delivery systems typically function with back-up battery power if AC power fails. Unless the presence of AC power is confirmed, the first obvious sign of power failure can be a complete system shutdown when the batteries can no longer power the system. Many anesthesia delivery systems have visual indicators of the power source showing the presence of both AC and battery power. These indicators should be checked, and connection of the power cord to a functional AC power source should be confirmed. Desflurane vaporizers require electrical power and recommendations for checking power to these vaporizers should also be followed.
Most anesthesia machines provide some indication that the machine is plugged into AC power, or that it is not and is on battery power ( Fig. 32-4 ). Ensuring that the machine is plugged into AC power should be a checklist item. Some newer generation machines perform a battery check automatically and report problems to the operator during start-up checks, whereas some machines require a manual assessment of battery power, such as unplugging the machine from the outlet and pressing a battery test button. If not an automated function, checking the battery power is another example of how an anesthesia technician could unburden anesthesia providers.
Item 4: Verify availability of required monitors and check alarms
Frequency: Prior to each use
Responsible Parties: Provider or technician
Standards for patient monitoring during anesthesia are clearly defined. The ability to conform to these standards should be confirmed for every anesthetic. The first step is to visually verify that the appropriate monitoring supplies (BP cuffs, oximetry probes, etc.) are available. All monitors should be turned on and proper completion of power-up self-tests confirmed. Given the importance of pulse oximetry and capnography to patient safety, verifying proper function of these devices before anesthetizing the patient is essential. Capnometer function can be verified by exhaling through the breathing circuit or gas sensor to generate a capnogram, or verifying that the patient’s breathing efforts generate a capnogram before the patient is anesthetized. Visual and audible alarm signals should be generated when this is discontinued. Pulse oximeter function, including an audible alarm, can be verified by placing the sensor on a finger and observing for a proper recording. The pulse oximeter alarm can be tested by introducing motion artifact or removing the sensor. Audible alarms have also been reconfirmed as essential to patient safety by ASA, American Association of Nurse Anesthetists, Anesthesia Patient Safety Foundation, and the Joint Commission. Proper monitor functioning includes visual and audible alarm signals that function as designed.
Verifying the availability and proper functioning of standard and other required monitors is a relatively straightforward task. However, the process of checking alarm thresholds, and possibly resetting them, can be tedious. It is possible for alarm settings on monitors to vary within individual facilities, because of provider manipulation of alarms for case requirements, a lack of standard default settings, and a failure to routinely reset alarm limits. Departmental alarm default settings can be established and programmed into anesthesia workstation monitors. Alarm limit settings also include anesthesia machine alarms such as volume, pressure, and inspired oxygen concentration limits ( Fig. 32-5 ). It is advisable to ensure that critical alarm limits are set to values that allow them to do what they were intended to do. Here, anesthesia technicians can improve the quality of the preuse checkout by checking the function of standard monitors and confirming that critical alarm thresholds are set to established default values.
Item 5: Verify that pressure is adequate on the spare oxygen cylinder mounted on the anesthesia machine
Responsible Parties: Provider and technician
Anesthesia delivery systems rely on a supply of oxygen for various machine functions. At a minimum, the oxygen supply is used to provide oxygen to the patient. Pneumatically powered ventilators also rely on a gas supply. Oxygen cylinder(s) should be mounted on the anesthesia delivery system and determined to have an acceptable minimum pressure. The acceptable pressure depends on the intended use, the design of the anesthesia delivery system, and the availability of piped oxygen.
Verification of oxygen cylinder pressure is accomplished by opening the oxygen cylinder(s) on the back of the machine and evaluating the tank gauge pressure located on the front of the machine, although some newer machines may also have a tank gauge located on the back of the machine. The 1986 PAC guideline recommended to “replace any cylinder with less than 600 psig.” The 1993 PAC guideline recommends that the oxygen cylinder be “at least half full (about 1000 psig)” during checkout. The current recommendations do not provide a specific value, but some manufacturer’s manuals still suggest the 1000 psig minimum.
It is important to understand that it is theoretically possible for the tank gauge pressure to read higher than the actual pressure in the tank; this is because on many machines, the tank gauge pressure reflects the pressure in the pipeline segment between the yoke check valve and low-pressure side of the pressure-reducing regulator ( Fig. 32-6 ). When the tank pressure falls below the pressure it contained when it was previously opened, the gauge will continue to reflect the higher pressure in the segment unless the wall supply pressure within the machine dips low enough for the pressure-reducing regulator to open to the tank supply route. A situation like this could occur, theoretically, if the provider were to leave the tank valve open during the PAC, and the tank were to drain down low or even empty through a leak at the yoke assembly. If a provider or technician then opened the tank valve, the gauge pressure would read the pressure remaining in the piping segment, not within the tank, because the segment pressure is higher than the pressure in the tank. This is due to the one-way nature of the yoke check valve. Unless tank pressure can overcome the upstream segment pressure, actual tank pressure will not be reflected on the gauge. In fact, even if the tank is completely removed , segment pressure will still be reflected on the gauge ( Fig. 32-7 ).
This possible error can be overcome by bleeding the oxygen tank gauge down to zero prior to checking oxygen tank pressure. This can only be accomplished by disconnecting the wall oxygen source, closing the O 2 tank valve, and draining down all machine oxygen pressure using the oxygen flush button or the oxygen flow control valve. Disconnection of the oxygen pipeline supply was a recommended step in the 1986 FDA PAC but not in the 1993 version. Likewise, there is no mention in the 2008 guideline, although many machine users’ manuals recommend this step in their respective daily checkout procedures. During development of the 1993 PAC, it was noted that provider failure to reconnect the main oxygen supply line during the PAC was not a rare occurrence. There was also a concern that daily removal and reconnection of the oxygen supply line connector could contribute to wear or breakdown. Given these concerns, disconnection of the oxygen pipeline supply is not a specified recommendation within the 1993 and 2008 guidelines. Some newer generation machines measure tank pressure prior to the outlet check valve, which eliminates this concern completely. Additional bulleted comments in this item of the 2008 guidelines include:
Typically, an oxygen cylinder will be used if the central oxygen supply fails.
Auxiliary oxygen cylinders will be used if the pipeline supply of oxygen fails or becomes contaminated . During a simulated gas pipeline crossover accident, researchers found that several participants used the machine’s auxiliary oxygen flowmeter as a presumed external source of oxygen, yet none properly disconnected the wall oxygen line while the inspired oxygen concentration declined in the face of sustained pipeline pressure.
If the cylinder is intended to be the primary source of oxygen (e.g., remote-site anesthesia), then a cylinder supply sufficient to last for the entire anesthetic is required.
The amount of oxygen required for a case begins with estimating the patient’s anticipated needs and then determining the requirements of the mechanical ventilator if driven by gas (see below). It is always wise to estimate finite-source oxygen needs (e.g., a tank) by applying a wide margin on the side of safety.
If a pneumatically powered ventilator that uses oxygen as its driving gas will be used, a full ‘E’ oxygen cylinder may provide only 30 minutes of oxygen. In that case, the maximum duration of oxygen supply can be obtained from an oxygen cylinder if it is used only to provide fresh gas to the patient in conjunction with manual or spontaneous ventilation. Mechanical ventilators will consume the oxygen supply if pneumatically powered ventilators that require oxygen to power the ventilator are used. Electrically powered ventilators do not consume oxygen so that the duration of a cylinder supply will depend only on total fresh gas flow.
Generally speaking, mechanical ventilators using a bellows are typically gas driven, with either oxygen or air, and piston driven ventilators are electrically driven. This underscores the importance of machine familiarity.
The oxygen cylinder valve should be closed after it has been verified that adequate pressure is present, unless the cylinder is to be the primary source of oxygen (i.e., piped oxygen is not available). If the valve remains open and the pipeline supply should fail, the oxygen cylinder can become depleted while the anesthesia provider is unaware of the oxygen supply problem.
The interface between the oxygen tank and the yoke assembly is very vulnerable to leaking. As alluded to above, if the oxygen tank pressure were to steadily decrease during the day, the provider would possibly be unaware, because real-time tank pressure measurement could be blocked by the yoke check valve. If tank pressure on the machine was measured prior to the yoke check valve, which is currently the exception, actual tank pressure would be continuously displayed, and this problem would be immediately recognized.
Other gas supply cylinders (e.g. Heliox, CO 2 , Air, N 2 O) need to be checked only if that gas is required to provide anesthetic care.
Item 6: Verify that piped gas pressures are 50 psig or greater
Responsible Parties: Provider and technician
A minimum gas supply pressure is required for proper function of the anesthesia delivery system. Gas supplied from a central source can fail for a variety of reasons. Therefore the pressure in the piped gas supply should be checked at least once daily.
Normal pipeline pressures in the United States for common gases (O 2 , air, N 2 O) are 50 to 55 psig (345 to 380 kPa). Pipeline pressure gauges on anesthesia machines include standard numeric analog gauges, analog gauges that highlight acceptable ranges, and digital pressure gauges ( Fig. 32-8 ). Although the guideline suggests verifying gauge pressures, an inspection of the supply hoses and connections is also recommended by some manufacturers. Checking that “hoses are connected” was a checklist item on the 1993 PAC. Despite gas-specific connectors, misconnections of gas hoses have been reported. Likewise, medical gas supply lines behind the walls of the OR are not immune from misconnection or contamination. A preuse check that includes a quick daily inspection of connections, supply hoses, gas pressures, and the presence of more than 90% oxygen in the inspiratory limb will greatly minimize risk.
An important safety item on all machines is an audible and visual alarm that warns the operator of diminishing oxygen supply pressure. The only way to evaluate this item is to disconnect the wall oxygen supply and shut off any oxygen supply tanks. Likewise, an evaluation of the machine’s oxygen failure protection device or fail-safe feature would also require disconnection of the O 2 supply hose. These two checks were not included in the 1993 guideline or in the current version, presumably because they are too time consuming relative to their risk-preventative value. Also, and as alluded to above, daily removal of the oxygen supply hose may introduce a risk greater than that posed by the potential for failure of these features. An evaluation of these features is usually part of routine preventative maintenance.
Item 7: Verify that vaporizers are adequately filled and, if applicable, that the filler ports are tightly closed
Frequency: Daily Responsible Parties: Provider and also the technician if redundancy is desired
If anesthetic vapor delivery is planned, an adequate supply is essential to reduce the risk of light anesthesia or recall. This is especially true if an anesthetic agent monitor with a low agent alarm is not being used. Partially open filler ports are a common cause of leaks that may not be detected if the vaporizer control dial is not open when a leak test is performed. This leak source can be minimized by tightly closing filler ports. Newer vaporizer designs have filling systems that automatically close the filler port when filling is completed. High and low anesthetic agent alarms are useful to help prevent over- or under-dosage of anesthetic vapor. Use of these alarms is encouraged, and they should be set to the appropriate limits and enabled.
Although not part of the 2008 guideline, some manufacturers recommend a check of their machine’s vaporizer interlock system, which if present prevents more than one vaporizer from being activated simultaneously. If this step is added to a local checklist, make sure that when one vaporizer handwheel is turned to a setting greater than zero that any other vaporizers remain locked in the zero position. Test the system for all mounted vaporizers, and then ensure all vaporizers are placed back to the zero position. This is also a good time to make certain that the vaporizers are firmly mounted.
Item 8: Verify that there are no leaks in the gas supply lines between the flowmeters and the common gas outlet
Frequency: Daily and whenever a vaporizer is changed
Responsible Parties: Provider or technician
The gas supply in this part of the anesthesia delivery system passes through the anesthetic vaporizer(s) on most anesthesia delivery systems. In order to perform a thorough leak test, each vaporizer must be turned on individually to check for leaks at the vaporizer mount(s) or inside the vaporizer. Furthermore, some machines have a check valve between the flowmeters and the common gas outlet, requiring a negative pressure test to adequately check for leaks. Automated checkout procedures typically include a leak test but may not evaluate leaks at the vaporizer, especially if the vaporizer is not turned on during the leak test. When relying upon automated testing to evaluate the system for leaks, the automated leak test would need to be repeated for each vaporizer in place. This test should also be completed whenever a vaporizer is changed. The risk of a leak at the vaporizer depends upon the vaporizer design. Vaporizer designs where the filler port closes automatically after filling can reduce the risk of leaks. Technicians can provide useful assistance with this aspect of the machine checkout, since it can be time consuming.
This step checks the integrity of the so-called low-pressure system (LPS) of the anesthesia machine, which is traditionally defined as the section downstream from the flow control valves to the common gas outlet. Leaks in this section of the machine are associated with hypoxemia or patient awareness under anesthesia. Leaks here are commonly related to the anesthetic vaporizer, the vaporizer mounting, or the flowmeter tubes.
Because of significant machine design differences, several tests have been described to check for leaks within the LPS. These tests use either positive pressure, assessing either leak flow or system pressure stability, or negative pressure to facilitate leak detection in this vulnerable part of the anesthesia machine ( Figs. 32-9 and 32-10 ). Historically, selecting the proper test was confusing, because some machines have an outlet check between the common gas outlet and the vaporizers (many Ohmeda machines), but others do not. The check valve is meant to minimize the effects of intermittent backpressure on vaporizer output. For machines without an outlet check valve, positive-pressure tests of the LPS are generally sufficient. These include simple pressurization of the patient breathing circuit or more complex positive-pressure testing of the LPS using specialized bulbs, manometers, and/or flowmeters.