| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
1 From the Department of Radiology, McMaster University, 1200 Main St West, Hamilton, Ontario, Canada L8N 3Z5. Received April 28, 2003; accepted May 5. Address correspondence to the author (e-mail: stolberg@mcmaster.ca).
Index terms: Aneurysm, abdominal • Editorials • Technology assessment
The comparison between the results of elective endovascular repair (stent-graft placement) and open surgical repair for abdominal aortic aneurysms has received much attention recently (1,2). It is the subject of an evidence-based review in this issue of Radiology (3). The review follows the process of conducting a systematic review in defining the questions, conducting the literature search, applying inclusion and exclusion criteria, creating data for abstraction, and conducting an analysis (4). The review provides our readers with the current best evidence from nonrandomized controlled trials performed at the efficacy level. Beyond that, it offers a detailed discussion concerning the introduction of endovascular repair at the "effectiveness" level (as opposed to the "efficacy" level) to facilitate the development of solutions for individual patients and departmental policy.
The concepts of efficacy and effectiveness are essential to a comparison of new and established procedures. Brook and Lohr define efficacy as the "probability of benefit to individuals in a defined population from a medical technology applied for a given medical problem under ideal conditions" (5). Effectiveness, on the other hand, reflects the performance of medical technology under ordinary rather than ideal conditions (5). In the present review, the authors apply principles of evidence-based radiology and technology assessment to the assessment of interventional procedures, as they suggested in an earlier publication: "Much of the clinical research published in the diagnostic imaging literature is technology assessment in the broadest sense" (6). The foundations for systematic technology assessment were provided by Fryback and Thornbury, who developed the "conceptual continuum," which refers to the hierarchical model of technology assessment in diagnostic radiology (7,8). The global perspective that Thornbury proposes encourages radiologists to go beyond the performance quality of a test or procedure to determine how such tests or procedures affect outcomes in terms of patient survival and quality of life (9). This perspective is supported by the review of Maher et al in this issue of Radiology (3).
To understand the differences between the present publication and earlier reviews of the same topic, we must clearly distinguish between evidence-based practice as proposed by the Agency for Health Care Policy and Research, or AHCPR, and evidence-based practice according to the McMaster and Oxford University Centres for Evidence-Based Practice in Medicine (10–12). The term "evidence-based practice," used by the AHCPR, designates a series of centers in North America that are charged with the production of evidence-based reports and technology assessment to support development of guidelines. A good commentary is provided by Eisenberg (11).
Evidence-based practice, according to the McMaster and Oxford University Centres for Evidence-Based Practice in Medicine, on the other hand, integrates the best research evidence with clinical expertise and patient values (12,13). It is not the purpose of this editorial to discuss the development of evidence-based health care, its application to radiology, and the development of evidence-based radiology. This was done in earlier publications (5,14). Suffice it to say that the core of evidence-based methodology is explicitness and explicit criteria that are used to make an objective assessment. Standard questions are used to assess methods and establish validity of evidence. Standard calculations are then used to assess the strength and relevance of results. This explicitness gives the results of evidence-based health care evaluations transparency and reproducibility.
Furthermore, evidence-based health care is meant to provide practitioners with the rules and tools to perform their own evaluations and consider them part of their own practice. It is essential to realize that practitioners are thus empowered to confidently develop solutions for individual patients and departmental problems, taking into account the current best evidence of research. That is different from the process used to develop most guidelines in radiology, which are highly dependent on the opinion of experts, become out of date rapidly, are expensive, and may not be applicable to individual patients and individual practice. The principles of evidence-based health care are a potential solution to problems that have long been encountered, such as requirement for valid up-to-date information for clinical and policy-making decisions. Evidence-based decision making can be defined as (a) the systematic application of the best evidence to evaluate the available options and (b) the decision making in clinical management and policy-making settings (14,15). The authors of the present review use evidence-based methods to discuss policy-making and management decisions and to evaluate whether endovascular repair currently being investigated at the efficacy level is ready for use at the effectiveness level. Evidence-based methodology is equally important for assessment of the equivalence of different imaging modalities. The recent article by Visser and associates is a good example (16).
The present review strictly applies the evidence-based medicine steps to evaluate elective endovascular repair. It includes a clearly defined "evidence-based question" by using the PICO ("Patient, Intervention, Comparison intervention, and Outcomes of interest") format. A comprehensive literature search was performed, followed by a detailed critical appraisal of the literature. The authors also retrieved raw data from the results sections of the articles selected to perform evidence-based medicine calculations of benefit and harm (17).
An essential feature of the present review is that a clear distinction is made between critical appraisal of diagnostic and interventional studies (benefit and harm of "therapy") according to basic evidence-based principles (15). Many publications have addressed the advance of statistical methods in diagnostic radiology (18–21). Members of the Standards for Reporting of Diagnostic Accuracy, or STARD, initiative recently published aims toward "complete and accurate reporting of studies in diagnostic accuracy" (22). By comparison, there is no standardized format for reporting the results of interventional radiology studies (5).
Interventional radiologists’ most compelling questions involve choosing the optimal treatment strategy for their patients. When they consider clinical trials, they are interested in the association between the procedure and the outcome. This is not always addressed adequately in the studies being considered; for example, a study may or may not demonstrate a decrease in the risk of adverse events in patients undergoing a new procedure. In the presentation of the effects of interventions, authors generally include the proportion of patients who experienced adverse events. To determine the "constant relative risk" without consideration of varying risk differences does not adequately capture the effects of the intervention (23).
To fully appreciate the effects of the intervention, available data must express the strength of the association as a relative risk, a relative risk reduction, an absolute risk reduction, a risk difference, or an odds ratio, as well as the number needed to treat or the number needed to harm. These data convey a variety of information (23). The benefits of therapy include relative risk reduction, absolute risk reduction, and number needed to treat to achieve one additional favorable outcome, compared with a group of patients who underwent the control treatment (ie, surgery). The procedural harm is determined by calculating the relative odds for case control studies and relative risks for cohort studies. The number needed to harm is calculated to assess the number of patients who would have to undergo the new treatment (ie, an interventional procedure) for one additional person to be harmed when compared with a group of patients who underwent control treatment (ie, surgery) (6,17). The clinical importance, then, can only be determined when numbers needed to treat and harm are calculated. We must keep in mind at this time that most currently available data address only short-term outcomes. There are much fewer data regarding mid- and long-term outcomes. Long-term prospective studies are therefore essential (3,24).
All recent publications addressed in the reviews published in Radiology during the past year on the comparison of endovascular and surgical repair are efficacy studies (1–3). To make evidence-based policy and management decisions, it is necessary to determine the effects of the introduction of endovascular repair in radiology departments other than those that offer ideal conditions for the development and appraisal of this interventional procedure. Beyond this, there are some other limitations. The morphologic characterization of abdominal aortic aneurysms or the measurements used could not be assessed in the reviews simply because such information was not available in sufficient detail. The importance of the marked morphologic variability present in abdominal aortic aneurysms, the importance of accurate and reproducible methods for sizing the aortoiliac arteries prior to device deployment, and the assessment of the success of endovascular stent-graft deployment and follow-up examinations are discussed in the publications by Rubin (25,26).
The preprocedural assessment and follow-up of stent-graft placement under ordinary rather than ideal conditions present a major challenge for radiologists and organizations involved. By today’s standards, these can be done only with helical computed tomography (CT), but use of multi–detector row CT scanners will result in further improvement. With 16-section multi–detector row CT scanners, "there is no preferred plane for image reconstructions" (27). The fact that the images may have been acquired in the actual plane is irrelevant. All planes have equal resolution, and the viewer can "choose" a plane that best shows the anatomy (27). It is reasonably expected that this will allow radiologists to adjust CT measurements to the morphologic variability present in abdominal aortic aneurysms and facilitate the "accurate and reproducible method for sizing the aorto- and iliac arteries prior to device employment" (23,27).
To develop a departmental policy for endovascular repair, there must be unequivocal appreciation about the problems associated with this procedure under ordinary conditions of practice. The workload will substantially increase, and so will manpower requirements. Interventional radiologists involved with the procedure must have highly developed skills and an ever-increasing amount of time as eligibility rates increase. The number of patients requiring lifetime follow-up will increase, as will the need for management of endovascular treatment complications. Workload and manpower requirements will also affect technologists because "the planning of endovascular repair puts greater requirements on preoperative imaging than any other prior application" (23) and also on imaging following stent-graft deployment. Furthermore, resource requirements are substantial in terms of both scan time and the number of graft combinations with different proximal and distal diameters that must be readily available. All this raises the important issue of developing the capacity for evidence-based decision making in departmental policy and management (28). One of the authors of the present review (D.E.M.) and his colleagues have made such an evidence-based policy decision. Their approach should help radiologists make such decisions based on a careful appraisal of the best evidence available. This can be achieved by developing not only individual skills but also cultures, systems, and structures within organizations (28).
The present review shows us how to use evidence-based review techniques in practice (3). It is only appropriate that the article also offers "caveats for potential users" (3). The lack of secondary literature in radiology that is of good methodologic quality is certainly an issue that must be addressed. It is a good example of what is needed, and one of the authors participated in another publication that should be of help to all of us (29).
In conclusion, the article by Maher and colleagues (3) offers our readers an evidence-based review of the recent literature concerning the comparison of stent-graft placement and conventional surgery for abdominal aortic aneurysms. The information offered will allow interventional radiologists to not only integrate available research evidence with the care of individual patients (30) but also make evidence-based policy decisions for individual departments (28). Any review such as this can only be a snapshot in time of current best evidence. At present, there are not only exciting developments in multi–detector row CT scanners, but there are also randomized controlled trials in progress in England and the Netherlands (3) that will help determine the future best practice. All new information is welcome and should be appraised with the use of evidence-based methods.
FOOTNOTES
See also the article by Maher et al in this issue.
REFERENCES
Related Article
This article has been cited by other articles:
|
|
 |
|
  M. Staunton, J. D. Dodd, P. A. McCormick, and D. E. Malone Finding Evidence-based Answers to Practical Questions in Radiology: Which Patients with Inoperable Hepatocellular Carcinoma Will Survive Longer after Transarterial Chemoembolization? Radiology, November 1, 2005; 237(2): 404 - 413. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
