Let No Patient be Harmed by Anesthesia

J.S. Gravenstein, M.D.

On October 16, 1846, Dr. Morton, a dentist, gave the first public and successful anesthetic. He used diethyl ether. The operation took place in the Massachusetts General Hospital. In a matter of months, the idea of rendering patients unresponsive through the use of inhaled drugs had spread to Europe and beyond.

On January 18, 1848, Hannah Greener required the removal of a toenail, a simple operation probably performed in the "office" of the surgeon. Hannah was a 15-year-old English girl. We do not know what her surgeon told her about the hazards of anesthesia. Maybe he said nothing about anesthetic mortality because there were no statistics he might have quoted. Had he said that she had a 99% chance of surviving the anesthetic, she might have considered anesthesia extraordinarily safe. She would have had a vivid image of the excruciating pain she would have experienced during an operation without anesthesia. Alas, she was unlucky. She inhaled chloroform from a cloth. In 1/2 minute her arm became rigid. The operation started. She gave a kick and sputtered at the mouth. Breathing continued for another 1/2 minute. Two minutes from the commencement of the operation she was quite dead (Snow, J: Edinburgh Medical Journal 1849,72; 75-87). The physician administering the anesthetic had no monitors other than perhaps a finger on the pulse and whatever information he could gather by inspection of the patient. He also had little knowledge about the pharmacology of the anesthetic nor of effective methods and drugs employed for resuscitation. He had no access to experts who might have stepped in to help before it was too late.

Today we would classify the death of Hannah Greener as a death during office-based anesthesia (OBA).

Safety Features in Anesthesia

Figure 1: Particularly over the last fifty years, the introduction of new safety measures has reduced the chance of harming patients during anesthesia.

In the meantime, we have gathered statistics on anesthetic mortality. A century after the death of Hannah Greener, Beecher and Todd reported an overall anesthetic mortality of about 1 in 2,000 anesthetics given in a hospital. Such mortality did not cause a public outcry of concern, even though annually anesthesia killed more patients than did poliomyelitis in the United States. Beecher and Todd commented that society was prepared to spend great sums in an attempt to conquer polio, but very little toward reducing anesthetic mortality. Perhaps society was not aware of anesthetic mortality but had been swayed by the highly visible March of Dimes campaign to raise funds for polio research. The enormous efforts to conquer polio eventually proved to be dramatically successful. However, even without a major investment into safety in anesthesia, we have made progress, particularly in the last half-century (Figure 1). Interestingly enough, many of the important efforts to make anesthesia safer derive from two disparate, and in the past often quite independent sources, namely industry designing and building equipment and then clinicians using it. The designers and manufacturers of anesthesia equipment have added many features to the equipment that help reduce the chance of human error causing disasters. Table 1 lists many of these features. Observe that a divine anesthetist who would never commit an error would do well without most of these features. Simply always attaching the correct cylinder to the appropriate yoke of the anesthesia machine would obviate the need for a pin index system. Always adjusting the flow rates so that hypoxic mixtures are not administered to the patient would obviate the need for an automatic oxygen proportioning system. Always watching the patient’s chest rise and fall and always listening to breath sounds and observing the breathing circuit would obviate the need for disconnect alarms in patients depending on mechanical ventilation. In short, the manufacturers of anesthesia equipment have recognized that most disasters in anesthesia (which is also true for accidents in aviation, rail and road traffic, shipping, atomic power plants and others) must be attributed to human error. Good design can help reduce the frequency with which such errors occur or lead to disasters. Manufacturers of anesthesia machines, ventilators, and monitors have also equipped their devices with alarms to alert the user to developing trouble.

Industry and Safety

  • Minimize Change of Hypoxic Gas Mixture
  • Minimize Chance of Ventilator Disconnect
  • Minimize Chance of Excessive Pressure
  • Minimize Chance of Machine Failure
  • Introduce Many Better Drugs
  • Introduce Better Monitors and Cover Many Variables With Alarms
  • Provide Trend Data ¥ Etc., etc., etc.

While many of the refinements of anesthesia equipment were initiated by industry without the urging of the medical profession, clinicians have not been idle either. When we analyze the information that is processed by the clinician, we must be impressed by the wealth of data presented. Most of these data are not captured by instruments, but are culled from many diverse sources. Their capture depends on an intimate knowledge of the system in which we operate, on an appreciation of the strength and weaknesses of ourselves and our coworkers, on not-recorded information about the patient, the procedure, the clinical setting and the team. Instruments, however, offer data on many variables we cannot assess with our senses such as gas concentrations and the electrocardiogram.

All monitoring efforts can be grouped under two headings: Monitors having to do with titration (e.g. drug effects and ventilation) and those serving primarily safety concerns. Of course, all monitors serving titration also have indirect safety functions, but many safety monitors are never used for titration. Table 2 shows a rather astonishing fact: the majority of variables we monitor during anesthesia have less to do with titrating anesthesia and much more to do with safety. Apparently, during the preceding 1 1/2 century, the profession worried more about safety than about titrating the anesthetic to an ideal level.

What Do Monitors Do?

For Titration: For Safety:
  • BIS monitor
  • arterial pressure
  • heart rate
  • carbon dioxide
  • muscle function
  • (temperature)
  • arterial pressure
  • ECG
  • central venous pressure
  • blood gases
  • respired gases
  • muscle function
  • PA pressure
  • cardiac output
  • temperature
  • EEG
  • TEE
  • inspiratory pressure
  • tidal volume
  • minute volume
  • multiple alarms on system
  • multiple alarms on physiologic variables
  • automated records
  • safety features on anesthesia machine
  • safety features on ventilator

Table 2: The efforts to monitor variables having to do with safety far outstrip efforts to monitor the titration of anesthetic drugs.

Whether or not in response to these developments, today’s anesthesia mortality is at least an order of magnitude lower than it was in the 1950s. But even at that level, all too many patients perish because of anesthetic mishaps. Two recent Australian studies have projected this to their respective populations. In New South Wales, anesthesia annually kills 4.4 patients per million population (Warden JC, Horan BF: Anaesth Intensive Care 1996; 66-73) while in Western Australia, anesthesia annually kills 1 to 2 patients (in a population of 1.75 million) (Eagle CC, Davis NJ: Anaesth Intensive Care 1997; 51-599). For the United States with a population of 260 million, that would translate to approximately between 200 and 1100 people losing their lives annually attributable to anesthetic mishaps.

Polio has been conquered and can no longer serve for comparison. Aviation is one of the many fields in which human errors more than any other factor cause preventable fatalities. The National Transportation Safety Board reports that annually on average approximately 150 people die in accidents involving major US carriers. In some years there are no deaths, in others several hundred passengers may be killed. These accidents are highly visible. The government responds. On April 4, 1998, Vice President Gore announced that the FAA "…based upon a comprehensive review of the causes of aviation accidents, adopted a focused priority safety agenda designed to bring about a five-fold reduction in fatal accidents." Very large sums are earmarked to accomplish that. Aviation has the wonderful advantage of having a National Transportation Safety Board, an agency that can develop the underpinning for a "safety agenda." The NTSB routinely investigates aviation accidents and issues recommendations as to what needs to be done to help prevent them.

Costly safety measures protect patients requiring anesthesia in hospitals and freestanding ambulatory surgery centers. Recent developments in medicine have caused many to worry that safety in anesthesia at the threshold to a new century is losing ground (Figure 2). The economic pressures that are buffeting medicine in the United States do not spare anesthesia. In order to reduce these costs, surgical procedures under anesthesia are now often performed in physicians’ offices. Deaths under such anesthesia circumstances have been reported. A recent article in Newsweek (August 9, 1999) illustrates such a case. The patient wanted to lose a few wrinkles. She was sedated for the procedure and something went wrong. The surgeon failed to notice that her heart rate and blood pressure dipped dangerously low. Her breathing had almost completely stopped before the clinic staff called 911. The patient died on the way to the hospital.

Safety Features in Anesthesia (with OBA)

Figure 2: Recent reports about fatalities during office-based anesthesia have intensified the worry that the current efforts to reduce the cost of medical care in general have worsened mortality statistics in anesthesia.

When in 1848 Hannah Greener died during OBA, the physician administering the anesthetic had no monitors other than perhaps a finger on the pulse and whatever information he could gather by inspection of the patient. He also had little knowledge about the pharmacology of the anesthetic nor of effective methods and drugs employed for resuscitation. He had no access to experts who might have stepped in to help before it was too late. Today this is different. We have the knowledge, the instruments, and the resources that we could bring to bear.

We face the imperative to reduce costs in healthcare without sacrificing safety. This will require a concerted effort of all who contribute to safety in anesthesia, that is, all who are represented in the Anesthesia Patient Safety Foundation. It includes those who administer anesthesia as well as those who make our tools and drugs, regulate what we do and where we work, represent the patients and administer the insurance coverage.

The time has come for the Anesthesia Patient Safety Foundation to do no less than the aviation community. We should not be satisfied with halting a worsening of the mortality statistics but should aim to improve even the best statistics. APSF might paraphrase the statement published by the FAA as follows: "The Board of Directors of the Anesthesia Patient Safety Foundation has announced that based upon a comprehensive review of the causes of anesthesia accidents, the anesthesia profession is adopting a focused priority safety agenda designed to bring about a five-fold reduction in fatal anesthesia accidents." Following the example set by the National Transportation Safety Board, APSF might also publish a list of ‘Most Wanted Improvements." For example, it might call for a penalty-free, anonymous system that would enable the collection and analysis of all anesthetic mishaps, for research and development of a system suitable for OBA and for a training system to be offered all working in the OBA setting. The profession will have to decide whether these or other improvements deserve the highest priority.

Much remains to be done before we can be sure that "No Patient is Harmed by Anesthesia."

Dr. Gravenstein, a founding member of the APSF Executive Committee and internationally recognized senior sage regarding anesthesia patient safety, is Graduate Research Professor at the University of Florida, Gainesville.