Introduction

Justification and optimization are two of the three fundamental principles of radiation protection (the 3rd is dose limitation) that apply to medical imaging [1]. It has been 10 years since the original pediatric CT ALARA conference organized by Tom Slovis in 2001 (published in 2002) [2], so it is fitting that we look back at our progress. The following material, then, will look at metrics—that is, how we might quantify performance, for both the justification and optimization arms of radiation protection—and it will also summarize some of the efforts and initiatives that have likely contributed to these changes. Because Dr. Slovis will detail this history in the introductory material for this ALARA conference, the discussion will focus on measures of success.

Background

Although pediatric radiologists have long been invested in patient safety, especially radiation protection, most of the rest of medicine was relatively unconcerned. Much of the current emphasis on radiation protection in diagnostic imaging, especially with CT, started with three articles in the American Journal of Roentgenology in 2001 [3–5]. These were accompanied by an editorial by the editor, Lee Rogers, who reaffirmed in his opening statement [6], “Sorry to say, but kids can get overlooked.” Based on this series, Steve Sternberg in USA Today authored a front page article stating, based on an estimate of 1.6 million pediatric CTs per year, that “about 1,500 of those (children) will die later in life from radiation induced cancer” [7]. One of the results of this uninformed reporting was the 2001 ALARA Conference [2], where leaders in imaging, medical physics, radiation biology, epidemiology, engineering and regulatory agencies participated in what might have been the first conference specifically addressing CT radiation risks and dose reduction strategies for pediatric CT. Now, 10 years later, there has been a marked increase in the number of publications addressing this issue. Educational forums have included those through the Society for Pediatric Radiology, the Radiological Society of North America, as well as many other professional organizations. Other commitments to CT radiation dose reduction came from industry to technology improvement (notably tube current modulation and more recently iterative reconstruction) and protocol refinements (especially weight- and size-based protocols for children), and regulatory engagement such as the Food and Drug Administration [8] and the National Electrical Manufacturers Association (NEMA) [9] addressing the need for more accountability for CT radiation doses, such as dose alerts on CT scanners. This is in addition to a great number of efforts by professional societies and other organizations such as the Alliance for Radiation Safety in Pediatric Imaging, the group behind the Image Gently campaign [10] and more recently the Image Wisely campaign [11]. Other recent professional endorsements for radiation protection include statements and articles by the National Academy of Sciences [12] and recent reports from the National Council on Radiation Protection & Measurements [13] to the American Board of Radiology Foundation Summit from 2009 [14] in addition to international efforts by the International Atomic Energy Agency (SmartCard Project) [15] and the World Health Organization [16]. This list is by no means exhaustive but illustrates the breadth and depth of response through scientific and medical organizations with a variety of missions. There is no single issue during the last decade in radiology that has generated such intense scrutiny as medical radiation. These efforts largely began from the work of pediatric imagers and their colleagues. But what has been the impact? How have we, in fact, been “performing”?

Performance evaluation is challenging and one can look at many measures and each in great detail. An easily justifiable categorization is to consider the principles of radiation protection consisting of justification and optimization for CT radiation. For justification, are we in fact better in the last decade at doing only appropriate examinations? For the optimization aspect, assuming the exam is indicated, are we optimizing image quality and radiation dose?

Imaging utilization: justification

Justification for CT use in children is extremely difficult to define. Although original statements from the 2001 ALARA conference [17] have been widely cited, this statement by Slovis sums up the discussion and opinions of the time: “One of the things that we all talk about is appropriate imaging. I think you agree that somewhere between 10% and 30% of the examinations don’t have to be done, and if we would get rid of those, we would significantly reduce dose” (T. Slovis, page 306). The data certainly support that medical imaging is over-utilized [18]. Accepting this, we should see whether we are in fact being more appropriate. This assumes that we can define the range of appropriate, and I will offer that this is extremely difficult. The reasons examinations are performed are varied and include science- and other evidence-based approaches (what we would prefer), inexperience, anecdotal experience, variable availability of resources, self-referral, reimbursement, legal pressures (i.e. defensive medicine), media influences, and the opinions and wants of parents and guardians. We know that the practice of medicine as tracked through Medicare reimbursement varies tremendously region to region [19] and radiology is certainly a subset of this medical practice. How do we measure justification to see whether there have been changes?

Should we consider number of CT studies performed in children as an indicator? We know based on very recent data from Dorfman et al. [20] that CT can be a relatively frequent examination in children [20]. CT accounted for 11.9% of all procedures in five healthcare markets representing more than 350,000 children younger than 18 years. CT was performed two or more times in 3.5% of all children. While this paints a picture of a relatively large number of CT examinations, matched with general indicators that CT examinations have been increasing during the last 2–3 decades [13], we must still ask ourselves whether this increase is justifiable. We know that CT use has increased with respect to cardiovascular applications, given ability to obtain faster scans with 3D and multiplanar capabilities (e.g., gated cardiac CT, other CT angiography) and evaluation especially in the older pediatric population for whom appendicitis has increased. For pediatric CT angiography, scanning has decreased the need for catheter-based procedures with attendant decrease in potential morbidity as well as improved use of resources [21]. CT has been demonstrated to result in improved patient care, including improved diagnosis and management strategies [12]. Perhaps CT use has peaked, and we are not doing as many unnecessary examinations. Recent data from Townsend et al. [22] in AJR from a survey of SCORCH 2 years ago indicate a fall in percentage of CT examinations (function of the total number of CT, MR and sonography examinations) from about 41% in 2003 to 35% in 2007. A new survey is underway, through the American College of Radiology, to assess pediatric CT use at practices and institutions that are not dedicated children’s hospitals but that do care for children to see how this community is with respect to CT use (personal communication, Michael Callahan, MD, Boston, MA). Perhaps this decrease in CT use in the last few years in children is a result of increasing use of MR and US [22]. Perhaps the reported decreased use is a result of economic factors, including the recent international economic downturn. Perhaps business has shifted from children’s hospitals to community hospitals and in fact the volume at these practices is going up. All of these considerations might be further impacted by governmental healthcare changes. What is clear is that we don’t know whether the numbers, whether they are increasing, staying the same, or decreasing, reflect a difference in the quality of care delivered to children. The bottom line is that changes in utilization are difficult to link with changes in justification.

Imaging performance: optimization

What about our performance with optimization? First, optimization assumes a somewhat definable balance between image quality and radiation dose. This is not so easy to do; as with justification, quality can be defined in many ways [23]. If we assume, however, there has been relatively little attendant change in both our objective measures and needs for quality as well as subjective needs, looking at changes in the technical parameters that determine CT radiation dose could be an indication of how we are performing. With that said, are we currently providing lower radiation exposure, through changes such as peak kilovoltage (kVp) and tube current (milliamperes—mA), than we were 10 years ago? It does depend on what user population we are looking at. If you look at the Pediatric Radiology subspecialty, through Society for Pediatric Radiology data, survey data from 2001 to 2006 by Arch and Frush [24] showed significant changes in the tube current (Fig. 1) and kVp (Fig. 2) for various age categories in children. We can also look at other investigations when technological advances such as automatic tube current modulation were applied to children [25–27] and more recently dose savings with the use of iterative reconstruction [28], with published benefits to the pediatric population imminent. While dose reductions have been reported to maintain diagnostic quality, we still don’t know whether changing the dose changes risk (see discussion later in this supplement about risk). Nevertheless, it is reasonable to assume this.

Fig. 1
figure 1

Peak kilovoltage used by members of the Society for Pediatric Radiology for pediatric chest (a) and abdomen (b) MDCT from a 2006 survey compared with a 2001 survey. kVp used decreased from 2001 to 2006, with 100% of those in the 2006 survey reporting known techniques of 120 kVp or less (P < 0.001) (Reprinted with permission [24])

Full size image
Fig. 2
figure 2

Mean tube current used by members of the Society for Pediatric Radiology for pediatric chest (a) and abdomen (b) MDCT over several age ranges from a 2006 survey compared with a 2001 survey. For chest MDCT, mean tube current used decreased 32–61 mA for each age range. For abdomen MDCT, mean tube current decreased 31–55 mA for each age range (Reprinted with permission [24])

Full size image

There are innumerable investigations that say radiation dose reductions can provide similar diagnostic accuracy. These all support that we are performing better at reducing dose in CT in children. If we look at the Image Gently campaign, has this campaign changed the exposure and risk from CT overall to children? We have heard that assessing dose is problematic (recall the material in this ALARA conference by Keith Strauss detailing the difficulties with CT dose estimation through the CT dose index [CTDI] and dose length product [DLP] in children), so meaningfully measuring change is also difficult. One current effort at assessing reports of using lower radiation dose for pediatric CT is a survey sponsored by the American College of Radiology, of institutions and practices that are not children’s hospitals but do care for children. These data will address practice patterns for CT that can be compared with data existing from children’s hospitals [22].

Performance improvement through awareness

If we measure performance by awareness, then I believe that there is no question that we are doing better. Returning to the original ALARA conference 10 years ago and resulting efforts, we can show that radiologists, vendors, regulatory agencies, government, other healthcare providers and the public are more aware of CT radiation dose and risks. Just this year, attention was heightened through several recent events with radiation injury reported in The New York Times [29, 30]. The CT equipment is better at helping the imaging community to perform lower-dose CT, the protocols for pediatric radiologists are refined and more available, educational material is more prevalent through numerous vendors, and the public is more vested in this topic. As a measure of performance, awareness is greater—a success. As a measure of performance, has utilization improved? This is difficult to say but with assumptions previously outlined, using what we currently know about the number of pediatric CT examinations as an indicator, we may be heading in that direction. As a measure of performance, are we optimizing parameters? This is supported by the literature but we must continue to develop newer technologies and continue educational and awareness efforts, especially providing guidance for protocol design and availability. We must remember that whether or not we can define appropriate, we must be able to say that we are doing everything we can to reduce dose. It is irresponsible to perform a CT examination that is unwarranted, and it is unsupportable to use more radiation than necessary to provide diagnostic information.

Are we, as radiologists, performing better? Effects of the widely recognized educational efforts, technological improvements and scientific publications have been difficult to measure. I believe most experts would argue that we are doing better, recognizing that this is somewhat circumstantial. Use might be different in children but could be related to influences other than more appropriate strategies for CT imaging. Are children exposed to less radiation than a decade ago? Are we performing at reasonable diagnostic levels if radiation is lower? What are appropriate levels of radiation from CT in children? Efforts are underway, and answers to these queries from consensus initiatives will be important in determining where we have come from, where we are now and where we need to go. Are we performing better? It is a cautious yes, understanding that there is no clarification needed for the fact that pediatric radiologists and their imaging partners (e.g., technologists, medical physicists) have made many changes in the last decade in the manner in which CT examinations are perceived and performed worldwide.