Editor's Note: J.S. Gravenstein, M.D., a much-respected leader of
the APSF for the decade of its existence, is honored in an editorial
on the following page. Dr. Gravenstein shares here some of his
remarkably incisive current thoughts about and insight on anesthesia
safety in our evolving practice environment.
by J. S. Gravenstein, M.D.
Not a week passes without mention in one or the other medical
publication of "cost containment," not a day without reference
in a hospital meeting about the economic vicissitudes of managed and
capitated care. "Fewer resources, fewer personnel, longer hours,
reduced income, do more with less" are the key words that ring in
our ears. In the clamor for reducing costs we listen in vain for calls
to invest in measures to increase or at least maintain safety in
How Safe is "Safe"?
Safety, of course, is relative. We all participate in activities that
are not 100% safe. For example, we know that we face risks when we
travel by air. Society's concerns about safety in aviation can be
judged by a recent cover story in Newsweek (April 24, 1995). The
magazine introduced the story with a stirring: "How safe is this
flight? Hundreds of Americans died in plane crashes in 1994, sounding
a wake-up call for an industry lulled into complacency."
How terribly unsafe is it to fly? Newsweek says that in the past
decade with U.S. carriers the "death risk" (the probability that
someone who randomly flew on one of the flights would be killed en
route) ranged from zero deaths in 10 years of airline operation to 1
in 1 million flights. Imagine that: with some carriers in 10 years no
death attributable to crashes!
But because other carriers had statistics with many deaths in the
decade, the magazine published "Ten ways to make flying safer."
- Modernize now, (get billion dollar modernization program for
the air-traffic-control back on track)
- Speed up the flight-data program (upgrading of data recording
equipment to analyze patterns and problems on all flights, not
- Set the highest standards for pilots (insist on rigorous and
continuous training for pilots to insure that they know how to use the
- Hire more air-traffic controllers (modernization may mean that the
system can do with fewer people - but until then, hire more)
- Install the terminal doppler weather radar (the FAA has bought 44
but has fully activated only 3)
- Take fliers at their word (71% of respondents said they would pay
higher airfares for safety 94% claimed they would put up with
delays or flight cancellations)
In comparison to commercial aviation, how safe is anesthesia?
Relative Risk of Anesthesia Compared to Air Travel
Clearly, we cannot point to a record of no anesthesia related death in
the last decade. Even a death rate of 1 in 1 million anesthetics would
be far better than we can boast. The death risk is sometimes
calculated as the deaths attributable to accidents in 100 million
hours of exposure. Assuming a death risk of 1 in 10 million for
commercial aviation and assuming an average of 2 hours per domestic
flight, the death risk would be about 5 per 100 million hours of
exposure. If we assume a preventable anesthetic mortality of 1 in
100,000 and assuming the average anesthetic to last about 2 hours, the
anesthesia death risk would be 500 per 100 million hours of
exposure. Feel free to play with the data. If you think the average
anesthetic lasts longer or shorter, or if you believe anesthetic
mortality to be higher or lower than these data used here just plug
them into the formula. You won't be able to get away from the fact
that anesthesia is far less safe than flying as a passenger with one
of the big commercial airline companies.
Of course, flying and undergoing anesthesia have nothing in common
except that both are not entirely safe, that in both examples the
victim does not contribute to a disaster, and that in both examples
the passenger or patient has every right to expect that he or she will
not be harmed by the trip - be it a flight or anesthetic. One
might, therefore, reasonably ask, "What anesthetic death risk is
acceptable?" And if the death risk in anesthesia is deemed to be
unacceptable, what is society willing to invest in improving safety in
anesthesia? Or, in other words, how much (in money and resources)
should we commit to saving a life?
The Cost for a Life:
Recently, one segment of our society heavily invested in the saving of
a single life. The world and certainly all Americans heard about the
brave rescue of a pilot shot down over Bosnia. Newsweek reported (June
19, 1995) that two CH-53 E Sea Stallions (cost: $26 million apiece),
two AH-1W Sea-Cobra gunships ($12.5 million apiece), four AV-8B Sea
Harriers ($24 million apiece), F/A-18 fighter bombers ($ 30 million
apiece), F-16s ($20 million apiece), F-15Es (35 million apiece),
EF-111s ($60 million), and AWACs ($ 250 million apiece) participated
in the rescue, not to mention dozens of Marines and scouts. The
investment of resources and funds to save one life was enormous and it
was spectacularly successful. All participants realized the risks
faced by rescuers and Captain O'Grady. After the widely hailed
success, I have heard no administrator or bureaucrat suggest that to
expend millions of dollars and risk many millions more to rescue
Captain O'Grady was irresponsible.
Depending on Experience and Advice
Many factors and many uncertainties confronted the planners of the
rescue. No scientific data guided them; no controlled experiment
predicted the probability of failure. The military had to rely on the
opinions of their experts.
When issuing safety related regulations, the Federal Aviation Agency
also must deal with uncertainties and opinions based on experience
rather than science, and rely on the advice of experts and make
decisions without the benefit of controlled, prospective studies. It
is similar in anesthesia; we, too, have not been able to and cannot
hope to obtain scientific evidence that this or that safety strategy,
behavior, or device will pay for itself. We also (like the military
and the FAA) find ourselves forced to rely on the best advice we can
garner. The comparisons among safety in aviation, safety in
anesthetic, and the rescue of Captain O'Grady bring to light an
important fact: The expectations of society - private or military
- and the opinion of experts will be all we have to guide us.
Safety on Trial
Imagine what would happen if we were to demand scientific proof of
measurable cost effectiveness for all safety measures we employ. The
following (invented) lawsuit paints the picture:
It was not a good day for the anesthesiologist.
- Plaintiff's Attorney (PA): Now, doctor, isn't it a fact that
you did not use any of the modern monitors, even though they are
available in your hospital?"
Defendant (D) nods.
Judge: "Doctor, please speak up. The stenographer cannot catch a nod."
PA: "And you gave anesthesia with a Schimmelshrub mask."
D: "With a Schimmelbusch mask, yes."
PA: "And you used ether, correct?"
D: "Yes, diethyl ether. That is what I used."
PA: "Doctor, is that anesthetic used routinely for anesthesia in
D: "By me, yes."
PA: "But not by the other 18 anesthesiologists in your hospital?"
D: "I have not checked that recently, but I believe that I am the
only one using ether."
(A lengthy interrogation followed during which the PA established that
the patient was a 70 year old man who had been in good health and who
had been anesthetized by the defendant so that the surgeon could
remove an inflamed gall bladder. During anesthesia the patient had
suffered a cardiac arrest. Resuscitation had reestablished circulation
but the patient had never regained consciousness and had died 4 weeks
later. An autopsy had been performed. The pathologist had described
extensive coronary as well as cerebral arterial atherosclerosis and
the ravages of hypoxemic brain damage and recent as well as old
PA: "Doctor, you did keep an anesthesia record. Please explain to
the jury how often you checked and recorded the vital signs of the
D: " At all times did I keep my finger on the pulse and at all
times did I watch the patient's spontaneous ventilation and the
color of his mucous membranes. It was necessary to keep anesthesia
fairly deep because the patient was obese and it was not easy for the
surgeon to dissect the gall bladder. I did not use muscle relaxants
nor mechanical ventilation. I was able to check the blood pressure
about every 10 minutes at which times I recorded the systolic
pressure, the heart rate and the respiratory rate."
PA: "Doctor, have you heard of pulse oximetry?"
D: "Yes, but..."
PA: "Just answer the question, please. Have you heard of capnography?"
D: "Of course, I..."
PA: "Please, doctor, just answer the question with yes or no."
PA: "Would you not agree that the use of these devices is by now
well established, indeed pulse oximetry and capnography could now be
called time honored standards of care, endorsed by your own
profession, and adopted throughout the developed world?"
Old "Proven" Ways: Just as Safe?
When the time came for D's attorney to summarize the position of
the defense, he said:
And so ended the closing statement of defense counsel.
- "Instead of standing here accused of negligence, my client should
be praised for having saved many lives. In his practice he uses only
what can be defended by prospective, controlled, and scientifically
valid studies. There are no studies that can document with scientific
rigor that pulse oximetry or capnography will improve the chances of a
patient to make it through anesthesia without suffering harm. There
are no studies to show that ether anesthesia is less safe than any of
the newer drugs in use today. There are no scientific, controlled
studies establishing that the surgeon will do a better job when muscle
relaxants are used during a cholecystectomy. Indeed, we can show you
reports by distinguished experts that we lack scientific evidence that
anesthesia is safer when the newer drugs and methods are used than was
true in years gone by, before these new and expensive drugs and
devices had been available. Well, you might say, but if the profession
accepts these new drugs and devices and they are established as the
standard of care, should my client not have used what every one else
"I tell you, no, no, and a thousand times no. The history of
medicine is replete with 'time honored' procedures that were
accepted as dogma not just for years, not just for decades, but for
centuries until scientific, and I stress scientific, insight proved
them to be utterly without merit. I imagine that someday a man like my
client will be able to demonstrate that the use of fancy muscle
relaxants - which have their own morbidity and mortality rates -
and the use of capnography and pulse oximetry, etc. is not
justified. Indeed, these elaborate monitors may be harmful; they may
distract and confuse the clinician, may present deceptive artifacts,
and may give rise to misinterpretations leading to unnecessary and
even injurious interventions.
"No, my client has not been blinded by the deceptive concepts of
'standard of care' and 'time honored' usage. Instead, he
has searched for the evidence to defend the use of these new devices
and drugs. When he could find neither evidence that such drugs and
devices improved or worsened the outcome of anesthesia, he asked the
second, essential question: `Can I defend the use of something that
has no proven benefit but causes no measurable harm?' `Not', he
concluded, `if it costs money. The sums squandered on useless devices
and drugs in anesthesia could save lives elsewhere, for example, they
could be used to lower the unacceptable high infant mortality in the
"We deeply regret the loss of the life of this elderly patient. But
instead of focusing on his death, focus instead on the lives of the
children that could be saved if throughout anesthesia, my client's
practice patterns were adopted and many millions of dollars were saved
and invested in prenatal care!"
Cost Pressures on Safety
Even ardent defenders of the need to show scientific proof and
favorable cost/benefit ratios for everything we do will not urge us to
adopt the practice here caricatured. But I sense an attitude among
some administrators that would push us in the direction of reducing
rather than enhancing the steps designed to increase safety in
anesthesia. Imagine we were to go before the public and say: "We
are now satisfied with our accomplishments in anesthesia. We have made
enormous progress. In the study by Beecher and Todd, anesthetic deaths
were about 1 in 2000 or, assuming again a 2 hour average duration,
about 25,000 deaths per 100 million hours of exposure.1 To have
wrestled 25,000 down to the vicinity of 500 or less is a great
accomplishment. True, the mortality in anesthesia is still 10 or 100
times as great than that of flying in a commercial jet, but we think
it is pretty good. Certainly, there is no need to spend more money on
safety for anesthesia. Our administrators remind us that we cannot
increase our expenditures for safety unless we can show that it raises
income or measurably improves outcome. With an anesthetic mortality as
low as 500 or maybe even only 250 per 100 million hours of exposure,
we would need millions of comparable cases in a balanced, prospective
study to show an effect on mortality. It cannot be done. So, we are
going to keep the status quo, more or less. Perhaps we can even
squeeze out some savings. We are sure you won't mind."
Safety as Top Priority
Of course, we won't say that. Instead we should raise our voices in
support of safety. If we don't, safety will take a backseat to
economy, and our mortality statistics will eventually show our
patients doing worse rather than better.
In the magazine of Delta Airlines (June 1995), the ruddy face of
Delta's CEO appeared with an article: "Safety is Delta's top
priority every hour of every day."We should say no less. Not only
in publications such as this Newsletter devoted to safety, but also in
speaking with our patients. They deserve to hear how we work to make
anesthesia safe for them. Even safer than it is today. We have a long
way to go to make anesthesia as safe as possible.
Dr. Gravenstein is Graduate Research Professor, Department of
Anesthesiology, University of Florida at Gainesville.
1. Beecher HK, Todd DP: A study of the deaths associated with
anesthesia and surgery. Annals of Surgery 1954;140:2-34.