As radiologists, we all want to provide the best care for our patients. This means, among other things, that we provide an accurate and timely interpretation of appropriate imaging studies. It also means that we interact with referring physicians to help select the most useful, accurate, safe, and cost-efﬁcient imaging to verify or refute a diagnosis.
is the Editor-in-Chief of this journal and a Professor of
Radiology, Diagnostic Imaging Department, University of Maryland
Medical Center, Baltimore, MD.
As radiologists, we all want to provide the best care for our
patients. This means, among other things, that we provide an
accurate and timely interpretation of appropriate imaging studies.
It also means that we interact with referring physicians to help
select the most useful, accurate, safe, and cost-efficient imaging
to verify or refute a diagnosis.
In today's medical environment, radiologists perform an abundant
number of studies, both appropriate and not appropriate (not the
best clinical choice or simply not indicated at all). In the U.S.,
the equipment is generally readily available to perform most
high-tech imaging studies, interpret them, and provide final
reports rapidly. If there is any doubt about the diagnosis after
the clinical assessment and lab work, imaging is often obtained to
confirm, exclude, or offer alternative diagnoses. Unfortunately,
there are almost always some doubts about the diagnosis. Many
presenting symptoms can be caused by myriad etiologies-some benign,
some distinctly not. The very low threshold for requesting imaging
stems from a combination of a litigious medicolegal environment,
the accuracy of imaging information, and the ready availability of
With subspeciality radiology expertise, appropriateness
guidelines for imaging have been developed for many clinical
I doubt that many radiologists are familiar with, use, or promote
them. They may, in the future, be incorporated into computer-based
order entry systems for radiologic studies.
Many of us have also experienced occasional encounters with
referring physicians in which we offer an opinion that a certain
test is inappropriate for the clinical indication (assuming one is
provided). I hope that your suggestion is at least considered. On
the other hand, you may hear, depending on your relationship with
the referrer, that "you are not taking care of the patient," or
"have you examined my patient?" Of course, you
taking care of both the patient and referring physician. Our
judgments about the use of certain imaging studies are based on the
distillation of the history, physical examination, and lab work
provided to us (or not) by the referring physician. We trust
his/her medical judgment to form our opinion about whether a
certain test is appropriate. We would expect them to trust us in
our area of expertise.
As a specialty of medicine, we profess-at least at meetings and
in the literature-that many imaging studies are not appropriate to
the clinical issue. A large number of studies (particularly in the
ER) are requested based on minimal symptoms and produce a very high
rate of negative results. Indeed, a very wide net is cast to catch
very few fish. Often, we know (as does the requesting physician)
that the study is performed to protect the collective gluteus
Let's postulate that a group of physicians in a hospital,
including radiologists, could sit down and, based on current
recommendations in the literature and published guidelines from
specialty organizations, determine which imaging studies were
appropriate, in which order, to work-up common clinical
presentations. This group could meet twice a year to observe the
effects/results of their local guidelines and to digest any new
literature that might be relevant to their approaches. This is no
panacea, since many clinical presentations are complex and
difficult to resolve into neat categories. While cookbook
approaches to medical decision making have their limitations, they
have certainly become a mainstay of practice today.
Let's further assume that a patient presents with a complaint
that fits into a clinical scenario covered by the institution's
imaging guidelines and they are followed to the letter. Still the
patient's medical outcome is suboptimal, and a lawsuit is filed. Is
it likely that a physician following documented guidelines who did
not order a "nonindicated" study (based on the majority of the
literature) would be guilty of malpractice? It would seem to be
Such an approach should decrease the number of unnecessary
imaging studies, now often obtained as a reflex, and allow for more
appropriate study selection. Physicians would feel less compelled
to "unload the shotgun" of tests in the search for diagnostic
certainty. This concept runs quite contrary to the current use of
and literal addiction to diagnostic imaging.
Perhaps, the potential for a real decrease in imaging studies
and its impact on income is a powerful disincentive to move in the
proposed direction. Perhaps, diagnostic doubt is not something
today's direct patient care physicians are willing to tolerate. At
least we should make use of contemporary, well-founded imaging
guidelines. It is in the best interest of patients and ourselves.
There are undoubtedly far more draconian and clinically insensitive
ways to limit the volume of diagnostic imaging, which contributes
to a rapidly rising national medical bill.
- Blackmore CC, Medina LS. Evidence-based radiology and the ACR
Appropriateness Criteria(R).J Am Coll Radiol.2006;3:505-509.
- Rothenberg BM, Korn A. The opportunities and challenges posed
by the rapid growth of diagnostic imaging. J Am Coll