| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Commentary |
1 Department of Radiology, Brooke Army Medical Center, 3851 Roger Brooke Dr.,
FSH, San Antonio, TX 78234.
2 Department of Radiology, USUHS, Bethesda, MD 20814-4799.
3 Department of Radiology, Wake Forest University School of Medicine,
Winston-Salem, NC 27157-1088.
Received February 10, 2005; accepted after revision February 16, 2005.
Address correspondence to L. T. Bui-Mansfield
(liem_mansfield{at}hotmail.com).
Materials and Methods
The list of the 50 most frequently cited articles published in the AJR was obtained from Chew [1]. Using the Web of Science search engine (Institute for Scientific Information) [2], the number of citations for each article was obtained. A citation occurs when one scientific article (the citing article) lists another (the cited article) as a reference. The Science Citation Index is a reference text that lists cited articles along with their citing articles that were published in scientific journals. The database covers 1945 to the present.
The articles were ranked based on the number of citations that they have received since 1945. In addition, the articles were classified according to subspecialty (e.g., abdominal imaging, basic science, breast imaging, chest imaging, gastrointestinal imaging, interventional radiology, neuroradiology, nuclear medicine, pediatric radiology, and radiation oncology), type of article (e.g., clinical application, complication, concept, technique, and treatment), clinical problem addressed, and types of imaging techniques or treatment (e.g., CT, MRI, radiography, sonography, and radiation therapy).
The Institute for Scientific Information was commissioned to provide to the ARRS a current list of the 100 most cited articles published in the AJR as of September 30, 2004.
Results
The new ranking of Chew's list of 50 most frequently cited articles published in the AJR [1] is shown in Table 1. The number of citations ranges from 136 to 592, with a mean of 232. Three articles have not changed in ranking, 24 articles have dropped, and 23 articles have risen in ranking since 1987. The increase in ranking ranges from +1 to +34. The decrease in ranking ranges from -2 to -25. "Observations on growth rates of human tumors," "Clinical NMR imaging of the brain: 140 cases," and "Percutaneous catheter drainage of abdominal abscesses guided by ultrasound and CT" [3-5] did not change in their ranking since 1988.
|
Figure 1 shows the ranking of the subspecialties represented on the list. Interventional radiology, radiation oncology, and abdominal imaging account for more than half (56%) of the articles on the list. Figure 2 shows the ranking of the types of articles on the list. Papers on clinical application of imaging, techniques, and new concepts account for the majority (82%) of the articles on the list. Figure 3 shows the 13 most common clinical problems addressed, which accounts for 64% of the articles, and the number of articles written about them. Figure 4 shows the frequency of the imaging techniques used in the articles. Radiography was more than twice as common as CT, the next most commonly used imaging technique. Figure 5 plots the number of citations of four selected articles at a 10-year interval since 1945. Articles by Collins et al. [3] and Caffey [6] were conceptual papers. The article by Gianturco et al. [7] was about a technique for arterial occlusion. The article by Sheedy et al. [8] was on the clinical application of CT in the imaging of the body.
|
|
|
|
|
The articles that remain frequently cited have one common feature. They were about topics that appealed to a broad audience of physicians, not just to radiologists, and were of clinical importance. "Observations on growth rates of human tumors" [3] is of interest to radiologists, pathologists, oncologists, and surgeons. Research conducted on tumor growth in any medical discipline is likely to cite this article. For the last 2 decades, breast cancer screening and treatment has become an important public issue. Thus, research conducted by primary care physicians, surgeons, and radiologists frequently cited Wolfe's article [9]. Before Caffey's report [6], nonaccidental trauma was not well recognized in the medical literature. Subsequent research on this topic invariably cited his article. The trend of medicine is toward minimally invasive procedures. Consequently, radiologically guided fine-needle aspiration and biopsy of masses became popular because they are less invasive, safer for the patients, and less costly. Again, this topic would be appealing to a wide audience, some of whom would perform their own research that cited Ferrucci's article [10].
Conversely, the greatest decrease in ranking of some articles is due to the rapid advance of technology in radiology, which makes old technology obsolete, or to the development of a new technique. For example, "Prolactin-secreting pituitary microadenomas: roentgenologic diagnosis" [11] had the sharpest drop in ranking, from 9 to 34, because MRI has since replaced roentgenography as the imaging technique of choice for the diagnosis of pituitary masses. With each new generation of CT, it is less likely that "Computer tomography of the body: initial clinical trial with the EMI prototype" [8] would be cited.
Interventional radiology, radiation oncology, abdominal imaging, and neuroradiology account for two thirds of the 50 most frequently cited articles published in AJR, reflecting the greatest growth in radiology in these subspecialties. CT is primarily responsible for the advancement in abdominal imaging and neuroradiology. A subspecialty that is conspicuously absent is musculoskeletal imaging. Not a single article about the musculoskeletal system was on the list. The most logical explanation is that the most recent advance in musculoskeletal imaging was the introduction of MRI, which did not come into widespread use until the 1990s, well after the publication of Chew's article [1].
The types of articles reflect both the types of research conducted in radiology and the emphasis of past AJR editors. AJR tended to publish articles with practical applications over basic science or early research in imaging. The 2003 AJR Subscriber Research Study has supported this observation. Ninety percent of respondents to the survey said that the scientific content in AJR is valuable to their day-to-day practice. More than half of the respondents rated the publication as informative (88%), easy to read (71%), practical (64%), and timely (54%) [12].
The clinical problems that were addressed in the most frequently cited articles published in AJR continue to be important disease processes of daily interest to general radiologists. They include effects of radiation exposure, Hodgkin's disease, gallbladder disease, liver mass, pituitary mass, abdominal abscess, adrenal mass, pulmonary embolism, nonaccidental trauma, and lung cancer. Again, this reflects the publishing philosophy of the AJR editorial office and the interest of AJR readers [12].
Despite the recent introduction of imaging techniques such as sonography, CT, and MRI, papers centered on radiography still accounted more than half of the most frequently cited articles published by AJR, more than all the articles combined that highlight new imaging techniques. This illustrates the established tradition of radiography in the diagnosis of diseases. Also, it takes time for the newer imaging techniques to be assimilated into the clinical practice of radiology and accepted by the rest of the medical community.
When the number of citations that each article received was plotted against time at 5-year intervals, four typical patterns of curve were identified. Articles on clinical application of imaging techniques or technical procedures had similar curves. Both had a rapid rise and precipitous drop in citations within the first 10 years. This reflected the introduction of new imaging techniques and interventional techniques that made the imaging techniques less useful and the procedures less popular. After the first decade of publication, articles on technical procedures had a slower decrease in the number of citations than articles on imaging techniques. Articles about new concepts had two patterns of citations. The article by Collins et al. [3] has had a near constant number of citations over time since its publication. This probably reflects the general acceptance of this concept by the scientific and medical communities. On the other hand, Caffey's article [6] has had a gradual but steady increase in the number of citations since its publication. It had the highest number of citations in the 1980s and since then, the number of citations has stabilized in the 30s over time. This pattern suggests gradual acceptance of a new concept, which over time was supported by subsequent research in other fields of medicine.
Because the database of the Institute for Scientific Information only went back to 1945, seminal articles published before 1945 may not be accurately represented. Although this is a valid argument, as long as the articles are cited beyond their publication date, the Institute for Scientific Information database will keep track of their number of citations. Therefore, it is highly unlikely that frequently cited articles published before 1945 were not duly recognized.
The articles on Chew's list [1] continued to be frequently cited. Forty-four articles (88%) on Chew's list appeared on the current list of 100 most cited articles published in AJR, a list that will be published later [13]. Of the top 10 most cited AJR articles, eight came from Chew's list [1].
Since 1988, 94% of the 50 most frequently cited articles published in AJR have changed their ranking. The articles that remained frequently cited were about topics that appealed to a broad audience of physicians and were of clinical importance. The clinical problems that were addressed in the most frequently cited articles published in AJR continue to be important disease processes of daily interest to practicing radiologists.
Acknowledgments
I want to thank Cindy deSa of the AJR editorial office and the entire staff at the ARRS headquarters for help during my Figley Fellowship and their assistance in the preparation of this manuscript. I also appreciate Dr. Chew for being an exemplary mentor and for his counsel during and since my fellowship.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) = technique, Sheedy et al.
[
) = application,
Collins et al. [
) =
concept, Caffey [