American College of Radiology White Paper on Radiation Dose in Medicine

Page created by Amy Lambert
 
CONTINUE READING
American College of Radiology White
                 Paper on Radiation Dose in Medicine
                    E. Stephen Amis, Jr, MDa, Priscilla F. Butler, MSb, Kimberly E. Applegate, MDc,
                     Steven B. Birnbaum, MDd, Libby F. Brateman, PhDe, James M. Hevezi, PhDf,
                       Fred A. Mettler, MDg, Richard L. Morin, PhDh, Michael J. Pentecost, MDi,
                         Geoffrey G. Smith, MDj, Keith J. Strauss, MSk, Robert K. Zeman, MDl

    The benefits of diagnostic imaging are immense and have revolutionized the practice of medicine. The
    increased sophistication and clinical efficacy of imaging have resulted in its dramatic growth over the past
    quarter century. Although data derived from the atomic bomb survivors in Japan and other events suggest that
    the expanding use of imaging modalities using ionizing radiation may eventually result in an increased
    incidence of cancer in the exposed population, this problem can likely be minimized by preventing the
    inappropriate use of such imaging and by optimizing studies that are performed to obtain the best image quality
    with the lowest radiation dose. The ACR, which has been an advocate for radiation safety since its inception in
    1924, convened the ACR Blue Ribbon Panel on Radiation Dose in Medicine to address these issues. This white
    paper details a proposed action plan for the college derived from the deliberations of that panel.
    Key Words: Radiation dose, radiation safety, radiation risk, radiation exposure, radiations, exposure to
    patients and personnel
    J Am Coll Radiol 2007;4:272-284. Copyright © 2007 American College of Radiology

INTRODUCTION                                                                    lution of imaging has also resulted in a significant
                                                                                increase in the population’s cumulative exposure to
Ionizing radiation has been used for diagnostic pur-
                                                                                ionizing radiation. Will this cause an increased inci-
poses in medicine for more than a century. The bene-
                                                                                dence of cancer years down the line? Although the
fits are immense and certainly exceed the risks. The
                                                                                answer to that question is currently under debate, the
more recent development of remarkable equipment
                                                                                presumption is that it will.
such as multidetector row computed tomography and
                                                                                   Consequently, there is increasing international and
the increased utilization of x-ray and nuclear medicine
                                                                                federal interest in, and scrutiny of, radiation dose from
imaging studies have improved the lives of our patients
                                                                                imaging procedures. Although there has been recent
and, along with other new modalities, revolutionized
                                                                                widespread interest in patient safety issues [1], the
the practice of medicine. However, this dramatic evo-
                                                                                possible hazards associated with radiation exposure
                                                                                generally have not been brought into clear focus by the
a
  Albert Einstein College of Medicine/Montefiore Medical Center, Bronx, NY.     public or members of the medical community other
b
  American College of Radiology, Reston, Va.                                    than radiologists. The ACR, pursuing its commitment
c
  Riley Hospital for Children, Indianapolis, Ind.                               to radiation safety, currently supports the following
d
  Associated Radiologists, P.A., Nashua, NH.                                    activities: accreditation programs, practice guidelines
e
  University of Florida College of Medicine, Gainesville, Fla.                  and technical standards [2], Appropriateness Crite-
f

g
  South Texas Radio Surgery Associates, San Antonio, Tex.                       ria® [3], a dose index registry (in progress), educa-
  New Mexico Federal Regional Medical Center, Albuquerque, NM.
h
                                                                                tional programs, the RadiologyInfo public informa-
  Mayo Clinic Jacksonville, Jacksonville, Fla.
i
                                                                                tion Web site (jointly developed with the Radiological
  Kaiser Permanente, Washington, DC.
j
                                                                                Society of North America [RSNA]) [4], collaborations
  Casper Medical Imaging, Casper, Wyo.
k
                                                                                with government and legislators on safety issues, and
  Children’s Hospital Boston, Boston, Mass.
l                                                                               research activities such as the ACR Imaging Network
  George Washington University, Washington, DC.
    Corresponding author and reprints: Priscilla F. Butler, MS, American Col-
                                                                                (ACRIN)®. To further enhance radiology’s leadership
lege of Radiology, 1891 Preston White Drive, Reston, VA 20191; e-mail:          role in the arena of patient safety, the chairman of the
pbutler@acr.org.                                                                ACR Board of Chancellors convened the ACR Blue

272                                                                                                           © 2007 American College of Radiology
                                                                                                0091-2182/07/$32.00 ● DOI 10.1016/j.jacr.2007.03.002
Amis et al/Radiation Dose in Medicine 273

 Table 1. Growth of computed tomography (CT)                mSv for a single study [15-18], and some patients have
 and nuclear medicine examinations in the United            multiple studies; thus, it would not be uncommon for
 States (approximate) [5-8]                                 a patient’s estimated exposure to exceed 50 mSv. In
   Examination             1980          2005               further validation of this concern, the International
 CT                     3,000,000     60,000,000
                                                            Commission on Radiological Protection has reported
 Nuclear medicine       7,000,000     20,000,000            that CT doses can indeed approach or exceed levels
                                                            that have been shown to result in an increase in cancer
                                                            [19].
                                                               Although there are currently no data showing that
Ribbon Panel on Radiation Dose in Medicine to assess        high-dose medical diagnostic studies such as com-
the current situation and to develop an action plan for
                                                            puted tomography and nuclear medicine have actually
the ACR that would further protect patients and in-
                                                            increased the incidence of cancer, a 2004 study (using
form the public. Panel members included private prac-
                                                            survey data from 1991 to 1996) suggested that medi-
tice and academic diagnostic radiologists, medical
                                                            cal exposure might be responsible for approximately
physicists, representatives of industry and regulatory
                                                            1% of the cancer in the United States [20]. This rate
groups, and a patient advocate. This white paper is the
result of the panel’s deliberations.                        can be expected to increase on the basis of the higher
                                                            number of examinations performed today. On the
SCOPE OF THE PROBLEM                                        other hand, as the use of medical radiation has in-
                                                            creased, the incidence of some cancers has actually
Over the past quarter century, there has been a rapid       decreased. For example, lung cancer is decreasing in
growth in both the number of diagnostic x-ray exam-         men because of smoking cessation, and breast cancer is
inations and the introduction of newer, very valuable,      leveling off [21], possibly because of less estrogen use.
but also relatively high-dose technologies (see Table 1     Does this mean that current radiation exposure can be
[5-8]). In 1987, medical x-rays and nuclear medicine        neglected? The answer is no. Radiation-induced can-
studies contributed less than 15% of the average yearly     cers typically do not occur until 1 or 2 decades or
radiation exposure received by the US population; the       longer after exposure. Thus, any increase in cancer
large majority was attributable to radon and other          occurrence due to burgeoning medical exposures in
natural sources [9]. Two decades later, because of the      the past 2 decades, as is the case for CT and nuclear
dramatic increase in the number of diagnostic exami-        medicine studies, may not be expected to be evident
nations performed each year as well as the higher doses     for many years. In addition, because radiation is a
associated with these examinations, this fraction has
                                                            relatively weak carcinogen, it is difficult to isolate ra-
most likely increased.
                                                            diation-induced cancers (1/1,000 per 10-mSv effec-
   Before a new technology or procedure that uses
                                                            tive dose) that are superimposed on the normal back-
ionizing radiation is introduced, there should be gen-
                                                            ground risk for other cancers (approximately 40% of
eral agreement that the benefits exceed the risks and
                                                            the population will be diagnosed as having cancer at
that an attempt has been made to reduce the potential
risks as low as practicable [10]. Ionizing radiation,       some point in their lives) [22].
especially at high doses, has long been known to in-           To put the issue in perspective, the current annual
crease the risk for developing cancer. In fact, x-rays      collective dose estimate from medical exposure in the
have recently been officially classified as a “carcino-     United States has been calculated as roughly equiva-
gen” by the World Health Organization’s Interna-            lent to the total worldwide collective dose generated
tional Agency for Research on Cancer [11], the              by the nuclear catastrophe at Chernobyl [23,24].
Agency for Toxic Substances and Disease Registry of         Therefore, one can assume, using International Com-
the Centers for Disease Control and Prevention [12],        mission on Radiological Protection [25] risk factors
and the National Institute of Environmental Health          and other data cited above, that this annual collective
Sciences [13]. The most comprehensive epidemiologic         dose may likely result in an increase in the incidence of
study supporting the carcinogenic effect of radiation is    imaging-related cancer in the US population in the
that of the atomic bomb survivors in Japan. The data        not-too-distant future.
from this study show a statistically significant increase      The preceding information poses many challenges
in cancer at dose estimates in excess of 50 mSv [14].       for referring physicians, radiologists, radiologic tech-
Whether there is a cancer risk at lower doses remains       nologists, medical physicists, equipment vendors, reg-
controversial. It is worth noting that many computed        ulators, and patients. Subsequent sections in this
tomographic (CT) scans and nuclear medicine studies         white paper recommend actions the ACR should take
have effective dose estimates in the range of 10 to 25      to manage this potential problem.
274 Journal of the American College of Radiology/ Vol. 4 No. 5 May 2007

MEASUREMENTS                                                         calculations currently provide only an estimate for a
                                                                     single anatomic model. Furthermore, when estimating
 Recommendations                                                     population doses, these calculated values are multi-
 ●   The ACR should adopt the policy of expressing quanti-           plied again by weighting factors [25] that change as
     tative radiation dose values as dose estimates and replace      scientific and professional organizations do their best
     the term dose with dose estimate as ACR publications are        to examine past exposures and subsequent biologic
     revised.                                                        effects. Hence, it is important to understand that the
 ●   The ACR should support the development of a national            reported numerical values for individual radiation
     database for radiation dose indices to address the actual       doses may vary by factors of 5 to 10 depending on
     range of exposures for x-ray examinations.                      individual patients and the manner of image acquisi-
                                                                     tion. Although there is little doubt that the absorbed
   There are references to radiation dose and radiation              radiation dose for an abdominal CT examination is
exposure throughout this paper. It is critical for the               larger than that for a radiograph of the ankle, the
reader to understand the terminology, especially as it               precise numeric quantity (particularly for an individ-
pertains to the difference between the easily measured               ual) is quite problematic. Therefore, the ACR should
radiation exposure from the various imaging modali-                  adopt a policy of expressing quantitative values regard-
ties and the actual amount of radiation absorbed by a                ing radiation dose as “dose estimates.” Furthermore,
patient.                                                             in future ACR publications, the term dose should be
   The determination of ionizing radiation dose to a                 replaced with dose estimate.
living human from an x-ray examination or nuclear                       For population exposure measurements, a range of
medicine study is very complex. Even during radiation                values would more accurately describe the situation.
oncology treatment, the radiation dose is actually                   Currently, there are few data on which to base such
measured only in extreme cases. This never occurs in                 estimations. However, with the growing applications
diagnostic imaging. To determine the absorbed radia-                 of digital imaging, such data can now be easily ac-
tion dose, the initial x-ray beam exposure and the                   quired. Digital x-ray imaging systems, such as com-
absorption in each organ must be known. It is the
                                                                     puted radiography, digital radiography, and com-
latter quantity that complicates this determination.
                                                                     puted tomography provide an index related to the
This absorption is dependent on the amount and
                                                                     amount of radiation that was generated to form an
properties of each tissue encountered by the x-ray
                                                                     image. Currently, these quantities are either displayed
beam, and these parameters vary widely among pa-
tients. The situation is further complicated because it              at the operator’s console or embedded with the image
is not practical to insert radiation detectors into each             itself. If these data were to be extracted and a central
organ of every patient.                                              database established for dose indices as a function of
   For x-ray examinations, the radiation dose is esti-               patient parameters (eg, gender, age, size) and exami-
mated by a calculation that involves an actual physical              nation type (eg, lateral lumbar spine, pelvic computed
measurement multiplied by various factors (Table 2):                 tomography), the relative range of radiation doses
                                                                     could be analyzed. The feasibility of such a database is
 Dose Estimate ⫽ Radiation Measurement ⫻ Factors.                    currently being examined by the ACR, and continued
   The various factors [25-29] are generated by Monte                support of this initiative is essential. The value of such
Carlo simulations [29] using mathematical represen-                  a database lies in its ability to demonstrate changes in
tations of the human body. Although the latest re-                   dose indices due to technologic advances and practice
search work in this area seeks to provide closer repre-              modifications. These trends would be useful to advi-
sentations for the range of body sizes [30-32], these                sory radiation safety agencies as well as to individual

 Table 2. Radiation quantities and units
    Quantity                             Unit                                                     Determination
 Exposure             Coulomb per kilogram (C/kg), roentgen (R)                         Measurement
 Dose                 Gray (Gy), rad                                                    Multiply exposure by f-factor or a
                                                                                         conversion factor
 Equivalent dose             Sievert (Sv), rem                                          Multiply dose by a quality factor
 Effective dose              Sv, rem                                                    Multiply equivalent dose by a
                                                                                         tissue weighting factor
Amis et al/Radiation Dose in Medicine 275

practices wishing to compare their own doses against             Liaison Committee on Medical Education, the accred-
established benchmarks.                                          iting body for medical schools, with a proposal to
                                                                 incorporate such a requirement into the standards that
REFERRING PHYSICIANS                                             schools must meet for accreditation. The Association
                                                                 of American Medical Colleges may prove an effective
 Recommendations                                                 ally in accomplishing this goal. Furthermore, the ACR
                                                                 should offer to develop learning materials on radiation
 ●   The ACR should work to convince the Liaison Commit-         dose in medicine and should approach the Alliance of
     tee on Medical Education and the Association of Amer-       Medical Student Educators in Radiology for input in
     ican Medical Colleges of the need for a standard meth-      support of this initiative.
     odology of introducing medical students to radiation           The issue of radiation exposure associated with di-
     exposure in medical imaging and offer to prepare learn-     agnostic imaging should be reinforced for nonradiol-
     ing materials in support of this initiative.                ogy residents during their postgraduate years. Al-
 ●   The ACR should work with the American Medical Asso-
                                                                 though core competencies for residents typically
     ciation to ensure the wide dissemination and enactment
                                                                 address patient safety issues, they generally lack the
     of its Council Report on Diagnostic Radiation Exposure.
 ●   The ACR should request that the Council of Medical
                                                                 curriculum content specific to radiation safety. Aca-
     Specialty Societies (CMSS) address the critical issue of    demic practices should be encouraged to provide ap-
     radiation exposure during medical imaging with its          propriate guidance for nonradiology house staff re-
     member societies.                                           garding both the risks associated with radiation and
 ●   The ACR should add relative radiation dose levels to its    the appropriate clinical indications for imaging in an
     Appropriateness Criteria® and work to ensure that the       effort to develop knowledgeable referral patterns on
     Appropriateness Criteria can be integrated into physician   the part of these physicians as they enter practice.
     order entry systems for real-time guidance in ordering         Education regarding the risks of radiation exposure
     imaging examinations.                                       provides a good foundation for physicians as they con-
 ●   The ACR should sponsor a summit meeting with leaders        sider imaging preferences, but it is always beneficial to
     from emergency medicine to discuss developing consen-       reinforce theory with tangible information. The ACR
     sus guidelines for imaging common conditions for which      Appropriateness Criteria® [33,34] offers an important
     computed tomography may be overutilized.                    decision-making framework for radiologists, referring
                                                                 physicians, and trainees at all levels. Incorporating
   Although some referring physicians are very knowl-            relative radiation dose levels into these guidelines will
edgeable regarding radiation safety issues and incorpo-          produce a powerful tool for influencing the proper
rate such information into their imaging decisions,              utilization of imaging examinations, based not only on
others have had little or no training in radiation expo-         efficacy but also on potential radiation exposure of the
sure and do not routinely consider this factor when              patient. The ACR should enact this revision and work
ordering imaging examinations. Furthermore, non-                 to ensure that the Appropriateness Criteria are in a
physician health care providers (eg, physician assis-            compatible electronic format that would allow them
tants, nurse practitioners) may be granted the author-           to be used within, or linked to, hospital information
ity to order imaging studies, and their ordering                 systems and physician order entry systems. By promi-
patterns are likely to reflect the behavior of their su-         nently displaying the relative radiation exposure level
pervising physicians. Subsequent references to refer-            for a particular examination during order entry, a cli-
ring physicians in this paper assume that their imaging          nician may be steered toward an imaging regimen that
preferences would likely be reflected in those they              minimizes radiation. Furthermore, the display of the
supervise.                                                       appropriateness rating for an examination (on a scale
   Educating future referring physicians on radiation            of 1 to 9, with 9 being the most appropriate) at the
exposure to patients during diagnostic imaging must              time of order entry is designed to influence an order-
begin during medical school. Radiology clerkships, in            ing physician to choose the most efficacious examina-
which students can be easily introduced to radiation             tion for a given clinical condition. It is likely, though
safety issues, are not required by all medical schools. A        not proved, that more appropriate utilization of imag-
more practical solution might include integrating                ing will reduce overall radiation exposure of patients.
mandatory material on radiation safety into the gen-             When imaging guidelines such as those just described
eral curriculum. The method of instruction, clerkship,           are built into order entry systems, regularly posting
or general curriculum is not as important as the goal of         individual physician ordering patterns (both appropri-
inculcating the awareness of radiation exposure in stu-          ate and inappropriate) may positively influence physi-
dents during training. The ACR should approach the               cian ordering behavior through peer pressure.
276 Journal of the American College of Radiology/ Vol. 4 No. 5 May 2007

   There are other ways in which the ACR can raise and               tive process should be extended to include other
redirect awareness of referring physicians regarding the             specialties.
relative radiation doses associated with various imaging
examinations and the risks associated with diagnostic
radiation exposure. The ACR should approach the
                                                                     RADIOLOGISTS
American Medical Association, as a follow-up to its
Council Report on Diagnostic Radiation Exposure [35],                 Recommendations
with an offer to develop collaborative programs or Web
content directed at nonradiologist physicians. The                    ●   The ACR should support the current multiorganiza-
Council of Medical Specialty Societies is another organi-                 tional effort to improve radiology resident training in
zation that could provide an appropriate venue for rais-                  medical physics.
ing the awareness of referring physicians about imaging-              ●   The ACR should include additional questions on radia-
related radiation exposure of patients. The CMSS vision                   tion safety and patient dose in its Annual In-Training
calls for convening specialty societies to identify critical              Examination.
issues, defining common policies and positions, and in-               ●   The ACR should request that the American Board of
fluencing decisions in the best interest of the public [36].              Radiology consider requiring at least 1 self-assessment
The ACR, which is a member in good standing of                            module on patient safety, to include radiation dose, every
CMSS, should request that CMSS make its other mem-                        10 years as an integral part of the maintenance of certifi-
                                                                          cation.
ber societies aware of this increasingly urgent problem.
                                                                      ●   The ACR should develop and implement maximum ra-
   Influencing referral patterns from emergency depart-
                                                                          diation dose estimate pass/fail criteria for the ACR CT
ments, for example, poses a major challenge. Patients                     Accreditation Program.
arriving at emergency departments are acutely ill or in-              ●   The ACR should review and update the CT Accredita-
jured, may have incomplete medical records, and may                       tion Program’s recommended scanning protocols on a
not be able to communicate their medical histories accu-                  routine basis and make them available on its Web site.
rately. The Emergency Medical Treatment and Labor                     ●   The ACR should request that the editor of the Journal of
Act (part of the Consolidated Omnibus Budget Recon-                       the American College of Radiology (JACR) add a monthly
ciliation Act of 1986) [37] requires that any patient who                 column on patient safety (to include radiation exposure
comes to an emergency department be provided with “an                     issues).
appropriate medical screening examination” to deter-                  ●   The ACR should create a prominent safety link on its
mine if the patient is suffering from an emergency med-                   Web site’s home page to facilitate access to this informa-
ical condition. If such a condition is present, the institu-              tion and to demonstrate the priority given to patient
tion has the duty to treat the patient. Emergency                         safety.
department physicians are aware that computed tomog-                  ●   The ACR should include in its Practice Guidelines and
raphy can provide quick, reliable answers to many clini-                  Technical Standards additional considerations for special
cal questions. The Emergency Medical Treatment and                        radiosensitive populations, such as children and pregnant
Labor Act and other concerns regarding medical liability,                 and potentially pregnant women.
along with the known clinical efficacy of CT scans, have              ●   The ACR should encourage radiology practices to record
been significant drivers of CT referrals.                                 all fluoroscopy times, compare them with benchmarks,
                                                                          and evaluate outliers as part of ongoing quality assurance
   Merely providing educational programs for emer-
                                                                          programs.
gency physicians regarding the radiation risks associated
                                                                      ●   The ACR should encourage radiology practices to define
with computed tomography is likely to do little to alter
                                                                          a surveillance mechanism to identify patients with high
their referral patterns. Rather, the ACR should proac-                    cumulative radiation doses due to repeated imaging.
tively sponsor a summit meeting between leaders of ra-
diology and emergency medicine to address the growing
issue of increasing radiation exposure to patients due to               It has been suggested that radiologists embody 3 prin-
the overutilization of medical imaging. This group                   cipal attributes: clinical acumen, mastery of technology,
should discuss developing consensus statements and                   and dedication to safety and quality [38]. A compelling
guidelines for evaluating common clinical presentations,             argument exists that mastery of imaging technology is the
such as suspected pulmonary embolism or ureteral calcu-              linchpin to these attributes, and that one cannot master
lus, and for reviewing a patient’s history to identify ex-           the technology without learning the principles and appli-
cessive imaging that may have unnecessarily resulted in              cations of the physics underlying the technology. It is
high levels of radiation exposure. Such consensus guide-             then essential that all radiologists continually refresh
lines could be powerful tools for improving the appro-               their knowledge regarding the basics of radiation safety
priate utilization of imaging. If successful, this collabora-        to be able to effectively apply them when imaging and
Amis et al/Radiation Dose in Medicine 277

when consulting with their patients and referring physi-        nificantly reduce radiation exposure. The Committee on
cians.                                                          CT Accreditation currently provides recommended
   To ensure that every radiologist is committed to the         scanning protocols as part of the accreditation program.
safe practice of radiology, especially in the daily applica-    The committee should review and update these routinely
tion of radiation safety principles, a more standardized        and post them on the ACR’s Web site.
approach to physics education at the resident level is             More information on general patient safety, as well as
necessary. The panel was pleased to learn that the Amer-        radiation safety, needs to be made available to the prac-
ican Association of Physicists in Medicine (AAPM) held          ticing diagnostic radiologist. The aim of the JACR is to
a forum on physics education in January 2006, to address        “fill the need for information on clinical parameters,
the issue [39]. The RSNA sponsored a multiorganiza-             practice management, education, health policy, and re-
tional follow-up meeting in February 2007. The draft            search on radiology health services” [43]. In keeping
curriculum being considered by the participants in this         with this mission, the ACR should request that the
educational initiative covers radiation safety in detail.       editor of JACR add a monthly column on patient
Definitive planning is under way for improving training         safety (to include radiation exposure issues) in the
in medical physics by strengthening residency program           journal. Because JACR is widely read by ACR mem-
accreditation requirements, linking physics topics to pro-      bers, this column could provide an effective vehicle for
fessional certification examinations, and reinforcing clin-     keeping the issue of radiation dose squarely in front of
ically relevant physics issues through continuing educa-        practicing radiologists.
tion and the maintenance of certification. The panel fully         Although the ACR’s Web site allows members and
endorses this process as a means of better incorporating        other interested parties access to many resources related
the tenets of radiation safety into the training of all radi-   to patient safety (eg, Practice Guidelines and Technical
ology residents. To supplement these efforts, the ACR           Standards [2], a magnetic resonance imaging safety white
should include additional questions on radiation safety         paper [44], accreditation guidance [45]), these resources
and patient dose in the ACR Annual In-Training Exam-            are scattered throughout the site and may not be readily
ination.                                                        apparent. The ACR should create a prominent safety link
   The maintenance of certification in radiology, admin-        on its Web site’s home page to facilitate finding all ACR
istered by the American Board of Radiology, requires            safety information. This link would also serve to demon-
completion of a minimum of 20 self-assessment modules           strate the priority the ACR places on patient safety.
during each 10-year recertification period [40]. In light          There are special patient populations, notably chil-
of the above comments regarding physics education and           dren and pregnant and potentially pregnant women, that
the need to continually reinforce the lessons learned, it       radiology departments and organizations should target
seems reasonable for the ACR to approach the American           for additional radiation protection. The ACR is currently
Board of Radiology with the request that diplomates be          developing a practice guideline for appropriate screening
required to complete at least 1 module addressing patient       for pregnancy before radiology testing that uses ionizing
safety, including material specific to radiation dose,          radiation. Because children have greater risks from ion-
among the 20 needed every 10 years.                             izing radiation than adults, the ACR should also develop
   Since its development, the ACR CT Accreditation              additional practice guidelines and technical standards to
Program [41] has been attracting a steadily increasing          protect them.
number of practices to seek accreditation. This is occur-          Although the ACR plans to be a resource for imple-
ring for several reasons, including: the desire for a public    menting many new programs designed to help protect
display of practice excellence, as a marketing strategy,        patients, it is incumbent on radiology practices and de-
and to meet the requirements that some payers impose            partments to accept responsibility for minimizing radia-
for reimbursement. The program has gathered credible            tion dose. In this regard, increased oversight of fluoros-
data on the range of radiation exposure factors associated      copy time will promote acute awareness of radiation
with various CT examinations and therefore is able to           exposure levels. Although the ACR currently recom-
define good practice regarding radiation exposure [42].         mends that radiation dose-related information or fluoro-
The ACR should now take this a step further and imple-          scopic exposure times be recorded in patients’ medical
ment maximum radiation dose estimate pass/fail criteria         records for procedures involving more than 10 minutes
for its CT Accreditation Program.                               of fluoroscopic exposure [46], it should now encourage
   Default settings and vendor-supplied protocols for           practices to record actual fluoroscopy time for all fluoro-
computed tomography may be designed to provide op-              scopic procedures. The fluoroscopy time for various pro-
timal imaging quality. However, it is often the case that       cedures (eg, upper gastrointestinal, pediatric voiding cys-
sufficient image quality for the examination may be             tourethrography, diagnostic angiography) should then
maintained by using alternative protocols that also sig-        be compared with benchmark figures, such as those pub-
278 Journal of the American College of Radiology/ Vol. 4 No. 5 May 2007

lished by the AAPM [47]. More complete patient radia-                that a duplicate or questionably indicated examination
tion dose data should be recorded for all high-dose inter-           has been ordered for a patient or that a patient has un-
ventional procedures, such as embolizations, transjugular            dergone multiple similar examinations. An alert technol-
intrahepatic portosystemic shunts, and arterial angio-               ogist should notify the radiologist in such a case and
plasty or stent placement anywhere in the abdomen and                thereby possibly avoid exposing the patient to unneces-
pelvis [48]. Excessive exposure times and doses should               sary radiation.
warrant further evaluation as part of practices’ quality                Technologists are responsible for determining the
assurance programs and Joint Commission sentinel event               need for additional radiation safety actions before a radi-
policy [49].                                                         ation exposure. Considerations include the identification
   One of the more important processes that can be im-               of high-risk patients (ie, children and pregnant or poten-
plemented in any radiology practice, whether outpatient              tially pregnant women) and the imaging of body parts in
or hospital based, is review of the imaging histories of all         which sensitive tissues (ie, fetus, thyroid, breasts, bone
patients presenting for studies. For example, it is not              marrow, gastrointestinal tract, gonads, and eyes) might
uncommon since the advent of noncontrast computed                    be irradiated. For such patients, technologists may need
tomography for evaluating patients with suspected ure-               to use individualized shielding or collimation; for others,
teral calculus to find patients who have had multiple                technologists may need to consult with radiologists about
scans over a period of months or years. In some of these             substituting lower dose examinations (eg, fewer images
cases, and at least on some occasions, the combination of            or radiography instead of computed tomography). In
a kidney, ureter, and bladder radiograph (KUB), and                  addition, technologists are responsible for limiting radi-
ultrasound might also be diagnostically effective and                ation exposure to patients by ensuring that proper pro-
therefore provide a safer imaging regimen. Routine re-               cedures and techniques are followed to prevent the need
views of patients’ detailed imaging histories will alert             for repeated imaging because of suboptimal image qual-
radiologists that such alternatives should be considered.            ity.
However, the imaging records must be scrupulously re-                   It is generally accepted that technologists are appropri-
viewed to identify these patients. The ACR should en-                ately trained and expected by radiology practices to per-
courage practices to make this review a standardized ele-            form the functions described above. To enhance perfor-
ment of practice policy.                                             mance in this arena, the ACR should encourage practices
                                                                     to provide regular in-service training in radiation safety
TECHNOLOGISTS                                                        issues for their technologists and to make sure they know
                                                                     to communicate any related problems to radiologists for
 Recommendations                                                     resolution.
 ●   The ACR should encourage radiology practices to pro-               Patient exposure considerations are more critical in
     vide in-service training on radiation safety issues for their   computed tomography and nuclear medicine because the
     technologists on a regular basis.                               doses are typically much higher than for radiographic or
 ●   The ACR should phase in the requirement that at least           fluoroscopic procedures. To produce high-quality im-
     1 technologist per accredited CT site hold the American         ages with the lowest possible patient doses, CT technol-
     Registry of Radiologic Technologists advanced registry in       ogists need to understand and use well-established pro-
     computed tomography and that at least 1 technologist            tocols. In nuclear medicine, there has been increasing
     per accredited nuclear medicine site hold the advanced          complexity of cardiac procedures and positron emission
     registry in nuclear medicine or certification by the Nu-        tomographic examinations. Such examinations require
     clear Medicine Technology Certification Board.                  more expertise on the part of technologists. Therefore,
 ●   The ACR should continue to support the Consistency,             both CT and nuclear medicine technologists should be
     Accuracy, Responsibility and Excellence in Medical Im-          specially trained, as is currently required by the ACR’s
     aging and Radiation Therapy ACT (known as the CARE              CT Accreditation Program and Nuclear Medicine Ac-
     bill).                                                          creditation Program. Furthermore, the ACR should
                                                                     phase in a requirement that at least 1 technologist per
   A radiologic technologist is typically the first, and may         accredited CT site hold the American Registry of Radio-
be the only, health care professional to interact with a             logic Technologists advanced registry in computed to-
patient presenting for a radiologic procedure. To re-                mography and at least 1 technologist per accredited nu-
spond to patients’ imaging-related questions, technolo-              clear medicine site hold the advanced registry in nuclear
gists need to be familiar with all components of the                 medicine or certification by the Nuclear Medicine Tech-
particular examinations, including not only the technical            nology Certification Board.
aspects but also the associated radiation dose and risk.                The American Society of Radiologic Technologists
Furthermore, it may be a technologist who recognizes                 has supported the introduction of legislation in the US
Amis et al/Radiation Dose in Medicine 279

House of Representatives (HR 583, the Consistency,                   Although patients frequently want to know the radia-
Accuracy, Responsibility and Excellence in Medical Im-           tion “dose” they will receive during examinations, they
aging and Radiation Therapy Act, known as the CARE               are generally unfamiliar with radiation terminology and
bill) [50] as a means of providing safer medical imaging         may not understand the level of risk involved. Conse-
examinations by setting federal standards for personnel          quently, any dose estimates should include information
who perform them. This bill specifically requires certifi-       on comparative risks. An example of such comparative
cation, licensure, testing, training, or experience for in-      information is available on RadiologyInfo [53]. Patients
dividuals who will be involved in performing medical             should also be informed of the risks to their health should
imaging services. The ACR should continue its current            alternative examinations, or perhaps even no examina-
support for this legislation.                                    tion, be performed.
                                                                     A review of several patient support Web sites found
                                                                 little information on the benefits and risks associated
PATIENTS                                                         with imaging examinations. That said, there are many
                                                                 patient advocacy groups (eg, the American Cancer Soci-
 Recommendations                                                 ety, the Susan G. Komen Foundation) with sincere in-
                                                                 terest in providing accurate information on various dis-
 ●   The ACR should, in collaboration with the RSNA, install
     a prominent patient safety link on the RadiologyInfo
                                                                 eases. These groups are familiar with their constituencies
     home page and regularly review and update information       and have well-established communication systems to
     on the Web site regarding the risks and benefits of imag-   reach them. The ACR should work with such patient
     ing procedures.                                             advocacy groups to improve availability and accessibility
 ●   The ACR should install a prominent patient safety link      of information on radiation exposure associated with
     on the ACRIN® home page that will lead patients to          medical imaging. For example, direct links to Radiology-
     information on risks and benefits associated with partic-   Info on appropriate advocacy Web sites would be of
     ipation in current ACRIN research protocols.                significant benefit to patients.
 ●   The ACR should work with patient advocacy organiza-
     tions to more effectively communicate the potential ra-
     diation risks and health benefits of imaging procedures.    MEDICAL PHYSICISTS

                                                                  Recommendations
   Radiologists understand the potential dangers from
ionizing radiation far better than patients do, yet not           ●   The ACR should work with the AAPM to develop a
every radiologist provides a balanced assessment of the               credentialing program for nonradiologist physicians who
risks and benefits of imaging when patients undergo test-             use fluoroscopy.
ing [51]. It is incumbent on radiologists to assume the           ●   The ACR should task its Commission on Education and
responsibility for their patients’ safety with regard to              Commission on Medical Physics to develop more effec-
radiation exposure. They should also educate their pa-                tive teaching methodologies for medical physics in sup-
tients on these issues so they may make informed deci-                port of the AAPM-RSNA initiative on physics education
sions about their health care. To facilitate this educa-              for radiology residents.
tional process, the ACR should continue collaborating             ●   The ACR should implement a periodic review and up-
with the RSNA in reviewing and updating their cospon-                 date of its primer on radiation risk [54].
sored patient education Web site, RadiologyInfo [4], to
make sure it includes current and understandable infor-             Although much of the ACR’s effort regarding radi-
mation on the risks and benefits of imaging procedures.          ation safety is focused on computed tomography and
This material should be easy to customize, download,             nuclear medicine, other modalities such as fluoros-
and print for use by practices and patients. The ACR             copy also contribute to the public’s radiation burden.
should advertise the availability of this resource to its        Fluoroscopy is frequently performed outside radiology
members. As with the ACR’s Web site, there should be a           departments in cardiac catheterization laboratories,
prominent patient safety link on the RadiologyInfo               endoscopy suites, operating rooms, pain management
home page allowing easy patient access to safety informa-        clinics, and orthopedic procedure rooms. Medical
tion for all types of imaging examinations. The ACR              physicists need to routinely survey these systems to
should work with the RSNA to accomplish this. In ad-             ensure that they are producing acceptable image qual-
dition, a safety link on the ACRIN Web site’s home page          ity at the lowest possible dose. In addition, medical
[52] should lead patients to easily understood informa-          physicists, working with radiology practices and their
tion on the risks and benefits associated with participa-        facilities’ radiation safety officers, should implement
tion in any of the current ACRIN protocols.                      programs for credentialing and monitoring the activ-
280 Journal of the American College of Radiology/ Vol. 4 No. 5 May 2007

ities of nonradiologist physicians using radiation. The              tion, the Contrast Media Manual [57], is now in its
purpose of these programs would be to maximize pa-                   fifth revision and has become an invaluable resource
tient safety, not to limit the use of fluoroscopy, and               for radiologists.
they should therefore enjoy strong support by hospital
administration. Such programs typically involve pre-                 VENDORS
paring brief primers on radiation safety pertaining to
fluoroscopy, administering tests on the material in the               Recommendations
primers that must be passed by physicians requesting
                                                                      ●   The ACR should work with the National Electrical Man-
such privileges, and awarding certificates document-                      ufacturers Association (NEMA) to encourage vendors to
ing compliance with these educational requirements.                       ensure that their application specialists are familiar with
This process should also be incorporated into the over-                   imaging protocols that emphasize the standard of as low
all credentialing program of the hospital. Several in-                    as reasonably achievable for their new equipment.
stitutions have already successfully implemented such                 ●   The ACR should work with NEMA to encourage ven-
programs [55,56]. Involving medical physicists in this                    dors to adopt a standardized approach describing expo-
process is essential, and the AAPM is currently devel-                    sure indices for computed radiography and digital radi-
oping such a program. Therefore, the ACR should                           ography.
offer to work with the AAPM on a model credentialing                  ●   The ACR should continue working with NEMA to en-
program that includes educational and testing materi-                     courage vendors to standardize digital equipment using
als. Ideally, the program would be available through                      ionizing radiation so that it automatically captures com-
the ACR’s Web site when complete so that ACR mem-                         plete dose information for each examination.
bers could share it with their hospitals’ credentialing
boards and encourage them to adopt the program. If                      The operational parameters of imaging devices should
such a program is successful, it should be expanded to               always be optimized to achieve patient doses that are as low
credential physicians other than radiologists, radiation             as reasonably achievable. Understanding the imaging de-
oncologists, or nuclear medicine physicians who are                  mands of the multitude of clinical examinations for multi-
responsible for the operation of CT or nuclear medi-                 ple sizes of patients, from neonates to large adults, is a chal-
cine equipment in the hospital setting.                              lenge for vendors’ design engineers and physicists. For
   As mentioned previously in the section on radiolo-                example, in the fluoroscopy mode of an interventional unit,
gists, there is a current multiorganizational effort to              voltage, tube current, pulse width, pulse rate, filter material
improve the quality of physics education for radiology               and thickness, focal spot size, exposure level at the image
residents that the panel finds worthy of ACR support.                receptor, field of view, and at least a half dozen image-
A comprehensive curriculum on radiologic physics has                 processing parameters can affect not only image quality but
been developed. In academic institutions with resi-                  also radiation dose. Although vendors automate many of
dency programs, the responsibility for teaching this                 these parameter choices into a single default selection, this
curriculum will belong in large part to medical physi-               automation needs to be based on the need to minimize
cists. For this initiative to succeed, the faculty must be           radiation exposure as well as the desire to produce an opti-
qualified and the material well structured. To this end,             mized image. To address the safety concerns of a procedure,
the ACR should task its Commission on Education                      appropriately trained radiologists and medical physicists
and Commission on Medical Physics to develop and                     need to be involved in the selection of imaging parameters.
disseminate effective teaching methodologies to med-                 Vendors typically accomplish this collaboration by intro-
ical physicists to ensure that they present their courses            ducing new equipment at designated clinical sites, often
clearly, concisely, and in a clinically relevant manner.             academic departments with which the vendors have formal
Such an initiative will hopefully result in radiology                research agreements. A review of several major computed
residents’ discovering that instruction in medical                   tomography vendors’ Web sites found references to proto-
physics is interesting and memorable, so that they will              cols for reducing radiation dose in pediatric patients or dur-
subsequently be able to apply the principles learned                 ing cardiac studies in all cases.
throughout their careers.                                               Too often, practicing radiologists unrealistically expect
   Because radiology technology and practice are still               vendors’ application specialists to have “all the answers.”
evolving, and more information on the risk of radia-                 Vendors need to continually train and update their applica-
tion is constantly being reported, the ACR should task               tion specialists to make sure that improved, validated pa-
its Commission on Medical Physics to update the                      rameter choices are introduced to its entire customer base.
1996 primer on radiation risk [54] and keep this ma-                 To this end the ACR should work with NEMA to encour-
terial current by creating a mechanism to regularly                  age vendors to ensure such preparation of their application
review and update this document. A similar publica-                  specialists.
Amis et al/Radiation Dose in Medicine 281

   Imaging equipment needs to display and record more             Although there is a literal “alphabet soup” of regulatory
information about patient dose. Although the US Food and          agencies with some degree of responsibility over ionizing
Drug Administration (FDA)-mandated display of radiation           radiation, the FDA, the NRC, and individual state radio-
air kerma rate and cumulative air kerma from new fluoro-          logic health agencies have the primary responsibility for
scopic equipment [58] was a good first step, a single, stan-      medical radiation.
dardized exposure indicator has not been adopted by ven-             The FDA regulates the manufacturers of medical de-
dors, particularly across digital radiographic and computed       vices and other electronic products that emit radiation
radiographic systems. This confuses imaging staff members,        and has promulgated regulatory standards for medical
prevents comparison of exposure indices among rooms or            x-ray equipment safety and performance. In general, the
institutions, and impedes image quality analysis. The             FDA has no legislative authority to regulate the users of
AAPM, with input from vendors, has drafted a standardized         these devices, with the exception of facilities performing
exposure indicator for digital radiographic equipment [59].       mammography (the Mammography Quality Standards
An International Electrotechnical Commission standard is          Act of 1992 [63] gave the FDA authority in this case).
also being developed through user-vendor cooperation [60].        The FDA takes an active role in providing radiation
The ACR should work with the NEMA to encourage adop-              protection guidance through the Center for Devices and
tion of this standard by all vendors.                             Radiological Health. For example, in 2001, in response
   The displayed radiation dose information should auto-          to concerns in the pediatric radiology community, the
matically be recorded without operator intervention to            center issued a public health notification warning of the
eliminate errors and incomplete records that may result           potential risks of radiation-induced cancer from helical
from the manual recording of this information. Interna-           computed tomography in young patients [64].
tional standards that enable such automatic dose data pres-          The NRC regulates the use of radioactive materials in
ervation have already been issued for digital radiography and     medicine, as well as for other applications, except in
fluoroscopy [61] and will be issued for computed tomogra-         so-called agreement states, where that regulatory author-
phy in 2007 [62]. Vendors, radiologists, and medical phys-        ity has been assumed by the state. The NRC and agree-
icists have been collaborating on these standards. The ACR        ment states need to continue focusing efforts on physician
should work with the NEMA to ensure this collaboration            and patient education concerning radiation exposure from
continues.                                                        nuclear medicine procedures. Given the vital involvement
                                                                  of both the FDA and the NRC in minimizing unnecessary
REGULATORY AGENCIES, ACCREDITING                                  radiation exposure, the ACR should approach both organi-
BODIES, AND THIRD-PARTY PAYERS                                    zations seeking input on how it can better support their
                                                                  efforts.
 Recommendations                                                     State regulations on ionizing radiation are variable in
                                                                  scope and enforcement, with a traditional focus on radi-
 ●   The ACR should approach the FDA and the US Nuclear
                                                                  ation-producing machines and radioactive materials. Al-
     Regulatory Commission (NRC) and seek input on how it
     can better support their efforts to minimize unnecessary
                                                                  though not a regulatory agency, the CRCPD plays an
     radiation exposure.                                          important role in the content of state regulations. This
 ●   The ACR should continue working with the Conference          nonprofit organization of state regulators serves as a
     of Radiation Control Program Directors (CRCPD) task          “common forum for the many governmental radiation
     force developing the document “Suggested State Regula-       protection agencies to communicate with each other and
     tions for Control of Radiation” and encourage its mem-       to promote uniform radiation protection regulations and
     ber states to uniformly adopt appropriate regulations.       activities” [65]. The ACR should continue working with
 ●   The ACR should encourage the Joint Commission to             the CRCPD to encourage its member states to adopt
     apply its existing credentialing and privileging standards   appropriate regulations covered in its document “Sug-
     to nonradiologist physicians who wish to use fluoros-        gested State Regulations for Control of Radiation.” In
     copy.                                                        addition, the CRCPD has a partnership with the FDA to
 ●   The ACR should encourage third-party payers to develop       periodically characterize radiation exposures to patients
     a process for identifying patients who have frequent im-     from current diagnostic imaging procedures. This pro-
     aging examinations using ionizing radiation and to pro-      gram, the Nationwide Evaluation of X-Ray Trends [15],
     vide feedback regarding these patients to their referring    is partially funded by the ACR and has tracked x-ray
     physicians.                                                  exposure trends for the past 30 years. The ACR should
                                                                  continue its support of this program.
   In the United States, regulatory agencies, accrediting            The Joint Commission (formerly the Joint Commis-
bodies, and third-party payers all have important roles to        sion on Accreditation of Healthcare Organizations) has
play in reducing unnecessary radiation dose to patients.          been active in the nationwide promulgation of extensive
282 Journal of the American College of Radiology/ Vol. 4 No. 5 May 2007

patient safety initiatives through its hospital accredita-           (T. Dehn and F. Apgar, NIA, personal communication,
tion process. However, it has not routinely addressed                January 2007). Anthem also is exploring a pay-for-per-
radiation safety or patient exposure issues occurring out-           formance component of quality-based reimbursement
side of radiology. A recently published Joint Commission             for hospitals and providers centered around the setup and
reviewable sentinel event, an estimated exposure of more             maintenance of meaningful radiation safety programs
than 15 Sv (1,500 rads) during fluoroscopy, is indicative            focused on dose reduction in multidetector row com-
of interest in this arena [49]. More oversight of radiation          puted tomography (E. Malko and R. LaFleur, medical
use at diagnostic levels might serve to prevent such an              directors, Anthem New Hampshire Blue Cross Blue
event. Just as its emphasis on hand washing has positively           Shield, personal communication, 2006). To expand
affected the rate of nosocomial infections, the Joint                these programs beyond the current, somewhat isolated
Commission, by further application of its current cre-               regional efforts, the ACR should encourage third-party
dentialing standard requiring relevant training to per-              payers to develop a process, such as that planned by
form a requested privilege [66], could help prevent the              Anthem and the NIA, for identifying patients who have
misuse of radiation in hospitals. Therefore, the ACR                 frequent imaging examinations using ionizing radiation
should encourage the Joint Commission to apply this                  and to provide feedback regarding these patients to their
standard during the accreditation process by requiring               referring physicians.
nonradiologist physicians who use fluoroscopy to be
privileged to do so on the basis of documented training in
the safety aspects of this procedure. A credentialing pro-           CONCLUSIONS
gram that could provide relevant training in the safe use            Many questions remain unanswered regarding the fun-
of fluoroscopy for nonradiologist physicians was de-                 damental mechanisms of radiation injury. Deoxyribonu-
scribed above in the section on medical physicists.                  cleic acid breakage, chromosomal aberrations, and gene
    Third-party payers have played only a small active role          mutations caused by radiation exposure, as well as the
in reducing unnecessary radiation exposure until re-                 potential for deoxyribonucleic acid to repair itself be-
cently. The efforts of third-party payers to control the             tween radiation exposures, are important avenues for
utilization of imaging have been primarily economic,                 further investigation. The intensive study of actual oc-
first with the concept of requiring preauthorization for             currences, such as workplace exposure of radiologists, the
certain examinations and, more recently, with pay-for-               long-term fate of patients treated with radiation therapy,
performance incentives. Despite these measures, the                  and the events at Hiroshima, Nagasaki, and Chernobyl
number of CT studies performed has risen significantly               have provided some answers and form the basis for the
over the past few years. At the national level, the Blue             assumptions in this paper regarding the carcinogenic ef-
Cross Blue Shield Association recently commissioned a                fects of ionizing radiation.
special report from its Technology Evaluation Commis-                   There is no question that the benefits of diagnostic
sion titled Potential Health Risks Associated with Radia-            imaging are immense. However, information gleaned
tion Exposure from Diagnostic Imaging. On November 2,                from the above events suggests that the rapid growth of
2006, the Medical Advisory Panel of the Blue Cross Blue              CT and certain nuclear medicine studies over the past
Shield Association approved publication of this report               quarter century may result in an increased incidence of
[67]. This report, now in draft form and due to be pub-              radiation-related cancer in the not-too-distant future.
lished in the near future, reviews the current literature            Pending future studies that contradict this assumption,
and issues nonbinding recommendations to the various                 there should be special attention paid to the practical
state Blue Cross/Blue Shield plans (B. Rothenberg, se-               suggestions set forth in this paper, such as education for
nior scientist, Blue Cross Blue Shield Association, per-             all stakeholders in the principles of radiation safety, the
sonal communication, 2007).                                          appropriate utilization of imaging to minimize any asso-
    There are, however, individual state health insurers             ciated radiation risk, the standardization of radiation
that have proactively implemented radiation safety pro-              dose data to be archived during imaging for its ultimate
grams. In the past year, Anthem Blue Cross/Blue Shield               use in benchmarking good practice, and, finally, the
of New Hampshire, Maine and Connecticut, in conjunc-                 identification and perhaps alternative imaging of patients
tion with National Imaging Associates (NIA), has em-                 who may have already reached threshold levels of esti-
barked on a comprehensive dose reduction initiative that             mated exposure from diagnostic imaging.
is scheduled for rollout in 2007. This program consists of              The recommendations derived from the deliberations
an educational component, available on the Anthem                    of the blue ribbon panel are felt to be feasible, apolitical,
Web site [68], and a preauthorization tracking software              and efficient methods to further minimize unnecessary
system developed by the NIA, which identifies patients               radiation exposure to patients. That said, there is little
who have reached certain radiation exposure levels                   question that some of these recommendations, were they
You can also read