Circulation 36,475 • Volume 16, No. 2 • Summer 2001

Anesthesia Quality Not Yet Captured by Raw Data

J.S. Gravenstein, MD

In 1894, Codman and Cushing, two medical students, wanted to compare their skills as “etherizers.” The better anesthetist would win a dinner. They “determined to let the test of satisfactory anaesthesia rest with the patient’s behavior in the ward… it was particularly due… to the detailed attention which we had to put upon the patient by the careful recording of the pulse rate throughout the operation.”1 This “careful recording of the pulse rate throughout the operation” became the prototype of anesthetic records for the next century, which is essentially plotting physiologic signals over time. Over the years this type of anesthesia record served many clinical and administrative functions.

Through a very narrow window of information, a few vital signs describe a surgical patient and how that patient responded to whatever was done to him/her, which include the surgeon’s and the anesthetist’s actions. While we have enriched the Codman-Cushing type of records with more physiologic parameters, comments on events, and information on drugs, I am not aware of any modern, commercially available automated anesthesia record that deviates from their basic design. The archetype original anesthesia records did not describe what the anesthetists did (other than to note the amount of ether used) and, therefore, actually speak to the quality of the anesthetic only in a rather indirect way.

Patient vs. System Data

For much of the last 100 years, we have continued to concentrate our efforts on signals that we gather from the patient. Only relatively recently have we begun to monitor data that directly relate to the anesthetist’s activities and thus to the quality of anesthesia. I remember that development from the mid-1970s. It came in the form of a short letter from our insurance carrier insisting that we monitor the oxygen in the breathing circuit. With that letter in hand, it was not difficult to persuade the hospital administration to invest in oxygen analyzers for all anesthesia machines. The insurance industry had taken this step because it was confronted by all too many malpractice claims alleging that hypoxemia had harmed or killed a patient. In the intervening years we have expanded the monitoring of inspired gases to include anesthetic agents, pressures, and volumes. Yet, even today, we lavish far more on monitoring and recording physiologic signals than on the things we actually control and impose on the patient.

Of course, there is a relationship between our actions and the patient’s responses, and there should be a relationship between the patient’s physiologic signals and our actions. Indeed, experts examine that interplay between the anesthetist’s recorded observations and actions and also between the anesthetist’s action’s and the patient’s responses to assess the quality of care when an injured patient sues. While the typical anesthesia record does not thwart such an analysis, it does not cater to it.

Even before clinicians take an action, they must consider the wisdom of such actions in the light of the current situation, the patient’s disease, the surgeon’s plans, existing guidelines, and standards. Here, even the best anesthesia record leaves a large void. It does not show the quality of care reflected by the clinician’s adherence to established practices or standards, not to mention the skills with which he or she initiates these steps.

During the last decade, we have learned that in anesthesia, as in other human endeavors, most accidents have more than one root cause. When analyzing the contributing factors to an anesthetic disaster, euphemistically called a mishap or adverse outcome, we are now aware that we need to look beyond the final link occurring at a specific single time and place that sealed the eventual result. Factors such as training, familiarity with equipment and procedures, fatigue, production pressure, and the availability of consultation can influence the outcome of an anesthetic procedure. The state of the apparatus and its currency, of back-up equipment, access to drugs and other supplies, all have played roles in preventing or bringing about a critical event. Again, typically we keep no record of such things. Nor can we reconstruct from an anesthesia record the atmosphere in the operating room, which may have been adverse to optimal care by lackadaisical procedures or tension between members of the team.

Quality Documentation

Modern technology gives us great flexibility in collecting, collating, and analyzing data. I suggest that Codman-Cushing’s basic design has served us well for over 100 years. For the next century, technology should present us with data reflecting both the clinicians’ actions as well as the patient’s responses, and it should do so in a variety of formats. Some will utilize the traditional plots; others will find new ways to satisfy the many demands we have, one of which is to document the quality of anesthetic care.

Dr. Gravenstein is Graduate Research Professor, Emeritus, Department of Anesthesiology, University of Florida, Gainesville.


1. Beecher HK, The first anesthesia records (Codman and Cushing). Surg Gynecol Obstet. 1940; 71:689-693.