| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
1 Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Avenue, Boston, MA 02215 e-mail: fhall@caregroup.harvard.edu
Editor:
I very much enjoyed the animated exchange of views in the recent Viewpoint article in Radiology (1–4) regarding computed tomographic (CT) screening for lung cancer. However, I was disappointed that the participants made no mention of screening mammography for breast cancer. I believe the political, economic, and statistical lessons that radiologists, and the entire medical profession, have learned from breast screening in the past 2 decades are very applicable to screening for lung cancer.
On the basis of my experience as a mammographer, I strongly suspect that, eventually, the earlier diagnosis of lung cancer made possible with CT lung screening will prove to reduce the mortality from this disease in high-risk individuals. However, as with mammographic screening, there will be intense disagreement along the way as to the actual amount of benefit; who, when, and how persons should be screened; and, perhaps most important, the cost-benefit repercussions of both the screening and the less readily quantifiable downstream expenses from false-positive findings. We will probably have several contentious nonconsensus conferences at the National Institutes of Health about CT screening for lung cancer. Twenty years from now, there will still remain, as is the case of screening mammography today, a few outlier studies and meta-analyses in which the benefits of CT screening will be questioned. To paraphrase Yogi Berra, CT screening for lung cancer promises a bit of déjà vu all over again.
The cogent arguments by Patz et al (1,3) and Miettinen and Henschke (2,4) are equally applicable to the larger, and probably vastly more important, discussion about radiologic screening in general (5). Is lung cancer screening much different than screening for liver and renal cancer, coronary and carotid artery calcifications, bone mineral measurements, or colonic polyps with virtual colonoscopy? Adding these examinations to CT lung screening may soon be only a matter of using several more breath holds and plugging the data into a computer-aided diagnosis software program. Will society once again need to expend the resources necessary for multiple randomized control trials in an attempt to prove the benefit of each of these screening examinations?
Our experience as mammographers with screening mammography suggests that it will be difficult to prove benefits of low-yield examinations, such as mammographic screening at ages 40–50 years, or when the disease has a prolonged course prior to symptoms or mortality, such as ductal carcinoma in situ and other low-grade breast cancers. In the meantime, it will be hard to convince many of "us," particularly those who can afford it, that diagnosing and treating disease in its preclinical stage versus waiting for symptoms, as is the historic model in medicine, will not potentially provide some survival benefit. False-positive findings are a problem with any screening or diagnostic test, particularly in our litigious society but, as happened with mammography, they will decrease with experience and common sense. Information in medicine and science is inherently useful. It is only what we do with that knowledge that is potentially harmful. I would be very interested in comments by the authors.
REFERENCES
and
Philip C. Goodman, MD*
Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 27710* e-mail: patz0002@mc.duke.edu
Department of Radiology, Dartmouth-Hitchcock Medical Center, Lebanon, NH
Our Viewpoint article (1) was intended to specifically address current issues in lung cancer screening. Nevertheless, Dr Hall raises several issues about screening in general, and we welcome the opportunity to respond.
There are certainly many similarities in principle and theory between screening for breast cancer with mammography and screening for lung cancer and other diseases with CT. However, we suspect that there are many different opinions about what we should have learned from screening mammography during the past 2 decades. We believe the most important lesson is that screening is much more complicated than it appeared at first, and that one cannot determine if it is more effective than clinical observations alone. The misconception that longer survival as a result of early diagnosis invariably implies mortality reduction is indeed "déjà vu all over again" and can be traced back long before the advent of screening mammography. In fact, according to an article in the New York Times in 1924, Dr Bloodgood from Johns Hopkins University made the following claim based on his observations of cancer patients: "Deaths from cancer would be practically eliminated and cures accomplished if persons afflicted sought medical aid immediately upon discovery of a foreign growth in any part of their body" (2).
Dr Hall implies that randomized clinical trials designed to prove the effectiveness of screening are expensive. We agree that it is very expensive to closely observe the thousands of participants that are usually required to produce a meaningfully significant result in a screening trial. However, this expense must be compared with the cost of screening millions of people in perpetuity under the assumption that screening must be worthwhile. Furthermore, it needs to be recognized that the reason so many participants are required in randomized trials of screening is that the expected effect of screening per participant is very small. This is generally true even when screening is highly effective, because the target disease usually affects only a small proportion of the population that is eligible for screening.
We believe that the demand for most forms of screening is often artificially inflated by the overestimation of benefits and underestimation of harms (3). As CT and other screening technologies continue to improve, the diagnosis of clinically unimportant disease could become a major source of harm and confusion (4). Nevertheless, if we could learn to put the potential benefits and harms of screening into proper perspective, then we could rationally allocate resources to appropriately investigate the most promising possibilities and fund those that provide a measurable net benefit.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  T. Schroeder, S. G. Ruehm, J. F. Debatin, M. E. Ladd, J. Barkhausen, and S. C. Goehde Detection of Pulmonary Nodules Using a 2D HASTE MR Sequence: Comparison with MDCT Am. J. Roentgenol., October 1, 2005; 185(4): 979 - 984. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
