| |
September 2002
Volume 66 |
Number 9
|
| |
VENTILATIONS
|
| Using the ASA
Physical Status Classification May Be Risky Business
|
|
|
Mark J. Lema, M.D., Ph.D. Editor
|
|
It is part of our daily practice to estimate the severity of
surgical patients' medical conditions prior to anesthetizing them.
For that purpose, the ASA Physical Status (PS) Classification
has been used since its inception in 1941.1
However, the purpose of this simple taxonomic guide for
assessing co-existing disease has been obfuscated, exalted, distorted
and misrepresented, largely by those outside of our specialty,
to fill the need for an operative risk barometer. While anesthesiologists
blithely use this scale to indicate the patient's overall physical
health preoperatively, it is regarded by hospitals, law firms,
accrediting boards and other health care groups as a scale to
predict risk and thus decide if a patient should have – or
should have had – an operation.
You may be mildly surprised that ASA headquarters is perennially
inundated with inquiries from around the world asking for clarification
regarding the link between "physical status" and "operative
risk." In fact the preoperative medical condition/anesthetic
technique/surgical procedure triad has never been studied extensively
by ASA or anesthesiology researchers for all classifications and
for all types of surgeries. It has been applied to retrospective
analysis of thousands of cases, but the data are often skewed
in favor of the lower classifications in large studies or higher
classifications with specific types of surgery. Moreover, the
assignment of classification by anesthesiologists is somewhat
subjective and further biased by the foreknowledge of the proposed
surgery.2 Finally, a retrospective
review really implies "potential" risk predictors, not
"real" risk predictors as are elucidated from a prospective
study.
In 1940-41, ASA asked a committee of three physicians (Meyer
Saklad, M.D., Emery Rovenstine, M.D., and Ivan Taylor, M.D.);
"… to study, examine, experiment and devise a system
for the collection and tabulation of statistical data in anesthesia
… that would be applicable under any circumstances"
(emphasis mine).1 While their mission
was to determine predictors for operative risk, they quickly dismissed
this task as being impossible to devise. They state:
"In attempting to standardize and define what has heretofore
been considered 'Operative Risk,' it was found that the term
… could not be used. It was felt that for the purposes
of the anesthesia record and for any future evaluation of anesthetic
agents or surgical procedures, it would be best to classify
and grade the patient in relation to his physical status only."
(emphasis mine)
|
|
Trends in two separate retrospective
studies4,5 suggest that information on surgical mortality
rates with respect to ASA physical status is similar
despite coming from disparate practices.
|
|
So, the ASA PS classification was always meant to globally assess
the degree of "sickness" or "physical state"
prior to selecting the anesthetic or prior to performing surgery.
One has no business applying it as a measure of operative risk.
However, one can estimate higher or lower medical risk when factoring
anesthetic technique and the extent of surgical trauma.
In 1978, William D. Owens, M.D., and colleagues tested the consistency
of PS assessment by sending a questionnaire to 255 anesthesiologists
that presented 10 hypothetical patients.3
Of the 10 hypothetical scenarios, six atients were rated identical
to the authors' assessments. The other four elicited a wide range
of responses. They concluded that the PS scale is a "workable
classification" but "suffer(s) from a lack of scientific
definition." A three-page editorial written by Arthur Keats,
M.D., in the same issue of Anesthesiology simultaneously defended
both the classification and the criticisms of the classification.
Dr. Keats prophetically states:
"At issue then is the expectation – what the classification
system is supposed to do and not do. Progress requires periodic
repetition to renew what is forgotten by the sliding scale of
memory."2
As recently as last year, Dr. Owens was compelled to address
this issue again in Anesthesiology. Commenting on a previous article
in the journal about variability in surgical procedure times,
Dr. Owens succinctly clarified why the ASA classification system
does not predict risk, saying, "The kind of operative procedure
is not a part of the classification system because a physical
status 3 patient is still in that
status if scheduled for an excision of a skin lesion with monitored
anesthesia care or if scheduled for a pancreatectomy with general
anesthesia. The operative risk is different because of the surgery,
but the physical condition of the patient is the same preoperatively."4
The following questions beg for answers: Of what value is the
physical status classification today? Should it be abandoned,
modified, maintained via committee assignment or remain unchanged
to allow for the different interpretations by ASA members and
other organizations?
It seems to me that we have an opportunity to add to our reputation
of being the safest (high-risk) medical specialty. ASA might revisit
the PS classification and attempt to expand it into the realm
of operative risk predictability. With super computers in the
palm of our hands and on every desktop, this generation of anesthesiologists
could computerize the data to prospectively assign relative risks.
Consider the variables for determining outcome:
- preoperative medical condition, be it acute or chronic
- selection of anesthetic techniques
- the nature and severity of the operation
- surgical skill (experience)
- anesthetic skill (experience)
We could surely set up a matrix that assigns a relative risk
to the first three variables while assuming that quality assurance
mechanisms and human interactions take care of the last two issues.
|
|
The combined risk of the patient's
medical condition with the selected anesthetic and
the severity of the surgery can be depicted as a three-point
(low/moderate/high) medical risk scale.
|
|
Ironically, the old data may provide the foundation for a new
classification. In the Keats' editorial, one table cited two large
studies by Vacanti and Marx that retrospectively showed surgical
mortality rates with respect to physical status. I noticed a trend
in both studies – one using 68,000 patients and the other
using 34,000 patients.4,5
I generated two graphs of the studies: Figure
1 and a second graph using semilog paper (not shown here).
What I found was there is a similarity to the shapes of the curves
and their slopes, suggesting that like information is obtainable
from disparate practices. In addition, low (PS 1) and high (PS
5) classifications offer little change when compared with PS 2
and PS 4, respectively. Thus, the current five-point scale might
simply be reduced to a three-point scale of low/moderate/high
medical risk. Assuming that less (i.e., more safe) anesthetics
are given as patients' medical conditions worsen, the PS classification
in this age of low anesthetic morbidity and mortality might serve
as the anesthetic risk factor. Using the same three-point scale
for assessing the risk of a surgical procedure, low/moderate/high,
a matrix can be formed to assign the combined risk of the patient's
medical condition with the selected anesthetic and the severity
of the surgery as depicted in Figure 2. I
used the combined mortality rates for PS 1 + 2, PS 3 and PS 4
+ 5 in the two 1970s studies to roughly predict the surgical risks
(as they existed in the 1960s).
My purpose in the brief exercise is not to come up with the "New
ASA Operative Risk Classification" but to show that patterns
exist among the data garnered over 60 years of assessment to embark
on developing a more meaningful assessment scale. Our current
PS assessment scale applied to a 1:250,000 mortality risk is meaningless
if one was to estimate a 10-fold risk of dying (10:250,000 anesthetics).
Perhaps it's time for ASA, the Society of Academic Anesthesiology
Chairs, the Association of Anesthesiology Program Directors, the
Association of University Anesthesiologists and even the American
College of Surgeons to collectively attempt to construct a more
useful risk assessment scale for anesthesia/surgery.
In the meantime, my advice to those who must answer questions
at headquarters about the predictability of the ASA Physical Status
Classifications with operative mortality is simple. The current
classification has evolved into a ceremonial exercise engaged
by all anesthesiologists in memory of those pioneer physicians
who set out to define anesthetic risk in a bygone era. It has
little meaningful clinical application in today's practice of
anesthesia.
– M.J.L.
References:
1. Saklad M. Grading of patients
for surgical procedures. Anesthesiology. 1941; 2(3):281-284.
2. Keats AS. The ASA classification of physical
status – A recapitulation. Anesthesiology. 1978; 49(4):233-236.
3. Owens WD, Felts JA, Spitznagel Jr EL. ASA physical
status classifications: A study of consistency of ratings. Anesthesiology.
1978; 49(4):239-243.
4. Owens WD. American Society of Anesthesiologists
physical status classification system is not a risk classification
system. Anesthesiology. 2001; 94(2):378.
5. Vacanti CJ, Van Houten RJ, Hill RC. A statistical
analysis of the relationship of physical status to postoperative
mortality in 68,388 cases. Anesth Analg. 1970; 49:564-566.
6. Marx GF, Mateo CV, Orkin LR. Computer analysis
of past anesthetic deaths. Anesthesiology. 1973; 39:54-58.
return to top
|