eaders of the ASA NEWSLETTER are now using a generation of anesthesia machines with so many functions and features that they are no longer called “machines” but rather “workstations” or “systems.” The old-timers may wish for the return of simpler days; yet the demand for safety and functionality in modern anesthesia requires greater sophistication of systems to allow integration of variable ventilator modes, lower gas flows with safety, controllable anesthetic delivery, features for preventing various forms of user error, alarms and self-checking. This did not all emerge overnight, but evolved over several decades.

Many of the original concepts were initially introduced in the Boston Anesthesia System (BAS), a prototype of the first fully electronic, integrated, microprocessor-controlled anesthesia workstation. First publicly displayed in a scientific exhibit at the ASA 1976 Annual Meeting and described in Anesthesiology in 1978,¹ the BAS was one of the first medical devices of any kind to utilize a microprocessor.

As we celebrate the recent donation of the BAS to the Wood Library-Museum of Anesthesiology (WLM), we will describe a short history of how it was conceived, designed and financed and how the ideas were disseminated to industry.

In the Beginning
The development of the BAS emerged from a collaboration of engineers headed by Jeffrey B. Cooper, Ph.D., with insightful clinicians in the Anesthesia Bioengineering Unit (ABU) of the Department of Anesthesia and Critical Care at the Massachusetts General Hospital in the 1970s. This was a time when anesthesia machines were simple plumbing devices with a few flow meters, typically one or two metered vaporizers such as a Copper Kettle, a calibrated vaporizer for halothane and perhaps one for ethrane as well. The only common monitor in use was the electrocardiogram; clinicians who wished to monitor arterial blood pressure were required to make special advance arrangements. Blood pressure was measured manually since this was before the time of the automated noninvasive technique.

ASA Award-Winning BAS and Team, 1976 The Boston Anesthesia System and its team of engineers at the 1976 Annual Meeting when it was awarded a prize in the Scientific Exhibit category. From left to right: Josh Tolkoff, Jeffrey B. Cooper, Ph.D., Ronald S. Newbower, Ph.D., and Jeff W. Moore. Not shown are Edwin D. Trautman and W. Reynolds “Renny” Maier, M.D.

Despite the existence of the earliest form of safety features in those machines, including color-coded gas tanks, the pin-indexed system and an oxygen-pressure fail-safe mechanism that offered some safeguard against accidental delivery of 100-percent nitrous oxide, there was little else that afforded protection against the many device and human errors that were still possible.

Safety Changes
E. M. Papper, M.D., Ph.D. (1915-2002), Robert D. Dripps, M.D. (1911-1973) and other leaders in anesthesiology recognized this condition. They characterized anesthesia as a public health hazard and, in 1964-65, so testified before Congressional Committees chaired by Lister Hill in the Senate and John Fogarty in the House. The influential medical research philanthropist, Mary Lasker, supported their efforts. They were successful in securing enhanced federal National Institutes of Health (NIH) funding for anesthesiology research and training.

The National Institutes of General Medical Sciences (NIGMS) convened a task force to consider those areas of anesthesia practice that could be improved with targeted research funds. I was tasked to discuss the inadequacies of anesthetic equipment as I was then convinced that equipment failure was largely responsible for anesthetic morbidity and mortality.

At that time, the typical anesthesia machine was a plumbing appliance, the basic architecture of which had not changed in decades. It provided a scaffold for hanging stand-alone devices that had no common readout, did not communicate with one another and had different alarms and failure modes. The machines were bulky and top-heavy, and the small wheels made them difficult to move and easy to tip over. Dr. Cooper later characterized these machines as “accidents waiting to happen.”²

In part for these reasons, I recruited engineers to the Massachusetts General Hospital in 1970 to lead the ABU, which was supported by one of the new NIH-funded Anesthesia Research Centers. My intent was to have engineers available to support the efforts of clinician-scientists in their quest to unravel the mysteries of anesthesia, its mechanisms and related physiology. The ABU team also launched projects completely of its own conception, including the BAS.

Clinician/Engineer Collaboration
In 1972, Dr. Cooper, a biomedical engineer, joined the ABU, which consisted of several engineers and, most importantly, Ronald S. Newbower, Ph.D., a Massachusetts Institute of Technology (MIT)/Harvard-trained solid-state physicist. Edwin D. Trautman, then an undergraduate at MIT, and W. Reynolds (Renny) Maier, M.D., a clinical-research fellow at Harvard and MGH, also were to be key players. Dr. Maier gave the engineers a perceptive view into the world of anesthesia, recruiting them as intraoperative observer members of the anesthesia care team. On-site potential and real-time problems were identified, possible efficiencies discussed, ideas germinated and solutions considered.

This kind of clinician and engineer collaboration is critical to multiple novel innovations in medical technology. The engineering-clinician team, led by Dr. Cooper, conceived the many technology solutions that were to be embodied in the BAS. In so doing, they were the first in our specialty, perhaps in medicine, to apply the techniques of human-factors engineering into the design, fabrication and integration of all the elements of a medical device.

Hard Work Pays Off
Among the team’s initial concerns were ways to secure project funding since the Anesthesia Research Center provided only seed monies. As so often happens unexpectedly, Mr. Trautman, as an MIT student attending a campus engineering dinner, was seated next to a philanthropist, Julius Rippel of the Fannie E. Rippel Foundation. While it was not in his foundation’s charter, Mr. Rippel was persuaded by Mr. Trautman’s presentation to provide $4,000 to construct a demonstration of microprocessor functionality in a medical device.

Under Dr. Cooper’s leadership, the ABU team submitted an application to NIH seeking funds for “A New Anesthesia Delivery System.” This was not the sort of research funded by NIGMS; but, in part due to the foresight and eloquent comments by Robert Epstein, M.D., the study section was convinced of the worthiness of the effort and provided funds for three years. The Cooper team finished the project on time and on budget, introducing the BAS to the anesthesiology community at the ASA 1976 Annual Meeting and, in the process, won a prize in the Scientific Exhibit category.

Innovative Design
The BAS demonstrates many innovations in anesthesia design. At its heart lies one of the earliest microprocessors, an 8-bit Intel 8080, which at that time cost $360. The idea was Ed Trautman’s, who had access, through his MIT window, to the cutting-edge of computer technology. The microprocessor enabled computer control of new digital effectors. The gases were metered by a digital device consisting of a series of individual fixed-flow valves (Dr. Cooper’s chemical engineering background proved essential here). For the liquid anesthetics, a fuel injector, acquired from the local Volkswagen dealer, was another innovation. While it was necessary to substitute O-rings that would withstand the corrosive effect of halocarbons and also modify the injector for lower flows, it worked well. The liquid anesthetics were held in agent-specific, magnetically keyed and pre-filled containers that were to be disposable. The system was designed to accept only one container at a time, nullifying the possibility of using an unselected agent.

This prototype of the Boston Anesthesia System was recently graciously donated to the Wood Library-Museum of Anesthesiology by Jeffrey B. Cooper, Ph.D., and his colleagues.

Programming was critical to the integration of these devices and the safety features that they would enable. Mr. Trautman, joined by Jeff W. Moore, another MIT engineer, created the primary code using very rudimentary tools and working under severe constraints of computing power and memory that were inherent in the early 8080 chips. Perhaps even more important was the conception of the safety features and their integration. Dr. Maier and Dr. Newbower were the key contributors here. Their efforts included not only control of the oxygen/nitrous oxide ratio and alarms for pressure integrated into the breathing system but also the oxygen concentration and circuit pressure measurement with automatic calibration.

Dr. Newbower’s artistic and human-design insights were key to the layout of an electronic message board that displayed all sensor information in a clear and simple format. Audible alarms with programmable limits, verbal and printed warnings that appeared on the message board, and system take-over capabilities for uncorrected faults were all features. Electronic and plasma bar graphs were used to display the gas flows and anesthetic concentrations. At the time, designing and integrating such capabilities by the use of human-factors engineering principles were unknown in medical devices and eventually may well have proved critical to the reduction of human error in all of medicine.

While the BAS was well praised, it was not easy to transfer the technology to common use. Harvard University did not then allow patent applications for medical inventions, which provided a challenging barrier for any manufacturer to risk the investment that would be needed. Dr. Cooper and the team published the concepts,¹ but they also took the unusual step of holding a workshop for all manufacturers who might be interested in learning the details of the design, which they shared openly. Attempts to work with two different manufacturers never reached fruition. The anesthesia market was perhaps not yet ready for such a radical departure. Many features and concepts of integrated functions did, however, work their way into the designs of newer generations of machines. It was not until the late 1980s that fully electronic machines began to appear. Now quite commonplace, they are rapidly replacing the traditional designs.

The BAS was used in an animal laboratory but never on humans. It was designed to explore novel engineering concepts integrated into a unified system capable of better aiding the clinician by reducing the likelihood of human and machine error in caring for anesthetized patients. That is still its most laudable achievement.

The Department of Anesthesiology and Critical Care of the Massachusetts General Hospital and Harvard Medical School is most pleased that the WLM has invited the Boston Anesthesia System to repose among their superb collection of novel contributions to our specialty. We are happy to include the associated laboratory notebooks and original correspondence in the hope that others may find them of interest.

Thanks go to Jeffrey B. Cooper, Ph.D., who provided historical information for this article.


References:
1. Cooper JB, Newbower RS, Moore JW, Trautman ED. A new anesthesia delivery system. Anesthesiology. 1978; 49:310-318 (with editorial).
2. Cooper JB, Newbower RS. The anesthesia machine – An accident waiting to happen. In: Pickett M, Triggs TJ, editors. Human Factors in Health Care. Lexington, MA: DC Heath & Co; 1975:345-358.

Richard J. Kitz, M.D., is Anesthetist-in-Chief, Emeritus, Massachusetts General Hospital; Faculty Dean for Clinical Affairs, Emeritus, Harvard Medical School; Henry Isaiah Door Distinguished Professor, Harvard University, Boston, Massachusetts.