Featuring Contributions from the Society for Technology in Anesthesia
The Anesthesia Patient Safety Foundation’s Committee on Technology would like to thank those members of the Society for Technology in Anesthesia (STA) who contributed to the discussion below. This string was originally posted to their listserv, and the STA Board of Directors graciously allowed the APSF to edit and publish the following commentary.
Dear Q&A,
In looking at the schematics of some of the newest anesthesia machines (e.g., Aisys and Avance, GE Healthcare, Inc., and the former Julian, Dräger Medical, Inc.), I am struck by how they have finally eliminated the hydraulic flow controls (i.e., needle valves) for computer controlled valves. This means, in the absence of power, no gas will flow, except for the redundant oxygen system. The question I’d like opinions on from the group is, “What has driven this change in anesthesia machine design?” I have several hypotheses (locally generated), but I’d like input from a wider group. I’ll share my ideas afterwards so as not to bias anyone’s thoughts.
James F. Szocik, MD
University of Michigan
Dear Dr. Szocik,
One can postulate many reasons for the change. The primary rationale is likely the ability to control and measure fresh gas flow by software. In recent years the most notable change in the anesthesia delivery system design has revolved around the anesthesia ventilator. Looking forward, designs that allow more efficient delivery of anesthetic vapor would be the next evolutionary step. Platforms which utilize mechanical fresh gas flow delivery will be more difficult or impossible to evolve into designs that can manipulate the fresh gas/anesthetic vapor concentration relationship to achieve more efficient vapor delivery. Electronic fresh gas control opens up the possibility of engineering that relationship.
From a safety point of view, all of the machines on the market provide some means of direct oxygen delivery in the event of electrical failure even if it is an oxygen flow meter mounted to the machine. One question is whether or not you want to deliver vapor if electricity fails. Interesting question to raise!
Jeffrey M. Feldman, MD
University of Pennsylvania School of Medicine
Dear Dr. Szocik,
I am reminded of the story, a few years ago, in which the anesthesia machine engineers from a certain company went out to various hospitals to ask anesthesiologists what they did not like about their anesthesia machines and what new features they wanted.
The engineers WANTED people to tell them they needed electronic flow meters, electronic vaporizers, electronic ventilators, new ventilator modes, etc. Instead, the anesthesiologists complained the wheels did not roll very well—they should be redesigned and equipped with “cow-catchers” to push cables and hoses out of the way. The drawer space was inadequate. A pull-out writing desk was needed. A small auxiliary light was needed for endoscopy cases so the anesthesia record could be seen in the dark. And so on.
Of course, the engineers departed in a mood of frustration, because what the anesthesiologists asked for was NOT what they wanted to build.
It should be noted that one CAN design controls that can be both manually operated and electronically operated. For instance I can increase or decrease the volume on my home stereo system by manually turning the big volume knob on the receiver, or by pushing the buttons on the remote control and watching the knob “turn itself” from across the room. It is easy to imagine a combination manual-automated system to adjust gas flows and vaporizer settings. The precise gas flow can be measured by electronic flow meters (or even by a simple system that measures the pressure drop across a known resistance). This measurement can be used to provide feedback for the flow meter controller.
One challenge for the new machines is that the old, mechanical machines had become nearly 100% safe—nearly 100% failure-free. With new complexity, there are new opportunities for failure, new opportunities that we may not have even thought of yet!
Frank Block, MD
University of Arkansas for Medical Sciences
Dear Dr. Szocik,
I also agree that there is much to be gained by adding electronic controls to the anesthesia machines if they enhance patient safety and improve the delivery of anesthesia. We need to have a machine that defaults to a safe basic machine, at least until the new controllers, software, and hardware have a very low probability of failure; and clinicians must feel comfortable using them under all circumstances. I favor parallel controls, like the electronic flow control in line with the needle valve. Permitting clinicians to use as much or as little of the new technology, in the beginning, is a wise marketing strategy in addition to providing a safe environment for the patient.
We don’t need additional catastrophic failure modes, which represent the worst outcome of replacing needle valves with electronic valves. Clinicians need to have enough familiarity with more advanced machines, that they can use it out of the box, like making Xerox copies or driving away in a new rental car, and then move into the more advanced features as they gain confidence.
I would like to see an affordable machine that makes the successful transition to a new and safer technology, even for the Luddites* among us. Don’t we have a responsibility to the lowest common denominator of user?
Bill Paulsen, MMSc, PhD, CCE, AAC
South University School of Health Professions
Mercer University School of Medicine
Dear Dr. Szocik,
The important aspect of your last paragraph is “affordable.” Every feature costs money and at the end of the day the companies always wonder if people will pay for what is included in the machine. In an ideal world an evolutionary approach that incorporates the old and the new would be appealing. Economics speak against it. Dr. Block made a good point that what excites an engineer is not always what excites a user.
Jeffrey M. Feldman, MD
University of Pennsylvania School of Medicine
Dear Dr. Szocik,
I agree with Dr. Feldman that one of the main reasons for considering computer controlled valves is control, but I think the other is something else. For billing purposes, vendors want users to be able to measure how much vapor they use, measure how much gas they use, and then bill for those as they would for any other service. To make it even more efficient, you transfer control to the machine itself so that it can adjust gas flows and agent concentration while monitoring “MAC” with other parameters.
Obviously, inherent machine control is not so obvious. I want such mechanisms to be proven reliable. And, I would rather buy a machine where the underlying control platform was already well established, before it controlled newly developed computerized valves.
Ryan Forde
Massachusetts General Hospital
Dear Dr. Szocik,
I think the reasons for going to electronically controlled flow include 1) a pathway to automated control of flow, 2) input for the electronic record and cost calculations, 3) input for fresh gas flow compensation for the ventilator and vaporizer, and 4) decreased maintenance requirement for electronic versus glass flow meters. It is interesting that anesthesiologists are so wary of electronic flow meters. Microprocessor-based ICU ventilators have been the standard for almost 20 years. All of the new anesthesia machines have battery backup, and most hospitals have emergency generators, so power loss at the electrical outlet is not really a concern.
I am more worried about the complex ventilation modes that are found in the new machines. This is where the “Luddites” will get into trouble, in my opinion.
Robert “Butch” Loeb, MD
University of Arizona
In Reply to All,
Our 2 local hypotheses for the progression to complete electronic control are 1) the continuing search for good information to put into the electronic medical record, and 2) “Technological Inertia,” analogous to Newton’s law, whereby the system and engineering were already on a roll in this direction, and would likely take a large amount of energy to stop or redirect it.
The biggest issue I have with the electronic control is the “failure” mode with prolonged power outage. With the great Eastern power outage a few years ago, we came within hours of running out of fuel for the generators. Cylinder oxygen is finite as well, but can be rationed better than generator power, and TIVA pump batteries will last about 3 days. I’d be interested to know what plans others have for regional wide disasters, wherein the infrastructure is also disrupted.
James F. Szocik, MD
University of Michigan
*Lud.dite [luhd-ahyt] – noun: A member of any of various bands of workers in England (1811-16) organized to destroy manufacturing machinery, under the belief that its use diminished employment. [Dictionary.com Unabridged (v1.1)]
Numerous questions to the Committee on Technology are individually and quickly answered each quarter by knowledgeable committee members. Many of those responses would be of value to the general readership, but are not suitable for the Dear SIRS column. Therefore, we have created this simple column to address the needs of our readership.
The information provided in this column is for safety-related educational purposes only, and does not constitute medical nor legal advice. Individual or group responses are only commentary, provided for purposes of education or discussion, and are neither statements of advice nor the opinions of APSF. It is not the intention of APSF to provide specific medical or legal advice or to endorse any specific views or recommendations in response to the inquiries posted. In no event shall APSF be responsible or liable, directly or indirectly, for any damage or loss caused or alleged to be caused by or in connection with the reliance on any such information.