| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Opinion |
1 Department of Radiology, Boston University School of Medicine, Boston Medical Center, 88 E. Newton St., Boston, MA 02118-2393.
Received January 24, 2003;
accepted after revision March 17, 2003.
Address correspondence to J. T. Ferrucci
(joseph.ferrucci{at}bmc.org).
Introduction
 Top Introduction PolypsColonoscopyPoliticsReferences |
|---|
All that glisters is not gold.
Geoffrey Chaucer [1]
In reviewing the potential of virtual colonoscopy for colon cancer screening in my Caldwell Lecture to the American Roentgen Ray Society 2 years ago [2], I touched on certain issues distinct from the traditional technical imaging concerns of radiologists, by referring to the topics of "polyps and politics." I suggested that assessing the perceptions of this new method by those outside radiology is a vital strategic effort for those of us on the inside. The increasing attention paid to virtual colonoscopy in the lay media [3-5] and in nonradiology journals [6-8] confirms that virtual colonoscopy is not being introduced into a vacuum. It will be necessary to make the case for virtual colonoscopy within the broader context of colorectal cancer screening initiatives that encompass published national guidelines, scientific controversies, and turf competition.
There is and will be skepticism and resistance, in large part because of the widely held views that colonoscopy is the best (or only) test to screen for colorectal polyps and cancer [9, 10]. The notion of the "total colon examination" has greatly undermined the legitimacy of sigmoidoscopy, which covers only the distal colon and has been likened to "mammography of one breast" [9-11]. Similarly, double-contrast barium enema has been defined for nonradiologists as ineffective for polyp detection by reports in the prestigious New England Journal of Medicine [12, 13]. In addition, the reimbursement for these two older procedures is so poor that neither primary care physicians nor radiologists have much interest in performing them. At the same time, marketing, education, and public relation efforts of the colonoscopy community have largely succeeded, and colonoscopy is accepted as the de facto gold standard for detecting colonic neoplasia.
Proponents of virtual colonoscopy will have to address the arguments used in favor of colonoscopy while at the same time proving that virtual colonoscopy gives accurate results and is acceptable to patients. As virtual colonoscopy becomes more widely available, closer inspection of the attributes and especially the weaknesses of conventional colonoscopy is therefore increasingly justified. To borrow from Chaucer, it is not unfair to notice the tarnish on the gold.
|
TopIntroductionPolypsColonoscopyPoliticsReferences |
|---|
To the contrary, as Glick [14] has observed, "polyp" simply means a bump or an excrescence, and polyps can be found in 50% of adults at 50 years old and 70% by 70 years. Moreover, most polyps are not likely to become a threat; pathologic studies have shown that on histologic examination 50% of "polyps" in the average adult colon are only hyper-plastic normal tissue. Most of the remainder are so small that they are unlikely to evolve into cancerous growths in the lifetime of the individual. Indeed, endoscopists have recently begun to acknowledge that only a very small percentage of polyps—namely, those larger than 1.0 cm or those with villous components—found on histology are likely to pose any threat of malignant change. Hence, the term "advanced adenoma" has been introduced to identify this group [9]. Data on the prevalence of the advanced adenoma in asymptomatic screening populations vary but are rarely much in excess of 10% [9, 15]. Thus, the remaining 40% of patients in whom tubular adenomatous polyps do exist probably have little cause for concern. The choice of an appropriate size target threshold is therefore a key strategy issue when proposing a colon cancer screening test, whether it is sigmoidoscopy, conventional colonoscopy, or virtual colonoscopy [16].
Leading virtual colonoscopy researchers have shown sensitivity of nearly 90% for polyps 10 mm or larger and have suggested this as an appropriate size threshold [17-19]. However, endoscopists have urged a lower threshold, at the 5- to 6-mm level [20], in part because the presence of two or more smaller adenomas has also been shown to be associated with an increased cancer risk [21]. In addition, use of a lower, 6-mm threshold would lengthen the intervals between subsequent screenings and thus increase patient compliance and reduce cost. These arguments are persuasive, and it is now clear that a sensitivity near 80% at the 6-mm-polyp level will have to be achieved for virtual colonoscopy to succeed. It is also clear that the sensitivity of virtual colonoscopy for smaller polyps (e.g., < 5 mm) is somewhat less than that of conventional colonoscopy [17-19]. Although it may be possible to improve the sensitivity for very small polyps with newer multidetector CT techniques using thin slice collimation, the added radiation dose will force the question as to whether it is of any value to do so. Emerging low-dose CT techniques may make this issue less pressing, but they will not resolve it entirely.
|
TopIntroductionPolypsColonoscopyPoliticsReferences |
|---|
In the context of virtual colonoscopy, it is relevant to note that conventional colonoscopy encompasses two separate functions, a diagnostic procedure and a therapeutic technique for removing polyps. In practice, once a colonoscope is inserted, any additional discomfort or any real risk in removing visualized polyps is minimal. The separate functions of diagnostic procedure and therapeutic technique are merged. However, the actual benefit of removing the many tiny subcentimeter polyps encountered in asymptomatic adults is in fact totally unknown. In a recent multicenter trial of screening colonoscopy in 3196 patients conducted by the Department of Veterans Affairs, 54% of subjects had a polypectomy [9]. Although invasive cancer was found in 1%, because of the 8-10% likelihood of advanced adenomas it would appear that at least 43% of the patients underwent removal of a polyp that was "below the legal limit." Adding in the 46% of patients in whom no polyp was found, it could be argued that 90% of colonoscopies in that population were unnecessary. When pushed, even leading colonoscopists have conceded that "eradication of all adenomas is not necessary, desirable or possible" [23].
In stressing this negative view, I am not attempting to undermine conventional wisdom. But colonoscopy is an invasive and costly intervention, and there is no direct evidence as to the cost benefit of mortality reduction from colon cancer by colonoscopic screening in average-risk populations [16]. As Glick [24] has so aptly stated, "The ultimate goal is an acceptable balance between the reduction of disease-specific mortality, the efficient allocation of limited resources, and the potential dangers to the overwhelming majority of patients in whom the disease will never develop."
Colonoscopy payment rates have declined sharply in recent years, but even so, large-scale population screening by colonoscopy is not fiscally realistic. The reason is that 70 million Americans are estimated to be eligible for colorectal cancer screening simply by virtue of being 50 years old or older, and 12 million more reach 50 each year. Moreover, the reimbursement figures of $679 [7] that colonoscopists quote for colonoscopy, and of $697 [11] in theoretic computerized cost-benefit analyses of outcomes, tend to gloss over the inevitable cost incurred by roughly 50% of patients having a polypectomy. These additional procedures usually result in additional charges for the excision and for pathologic examination. Thus, the net cost for screening colonoscopy can vastly exceed the plain vanilla reimbursement schedules used in computer outcome models [11, 25]. For example, a recent colonoscopy performed in Boston on a 62-year-old member of my own family yielded two 2-mm polyps. (Not much benefit there!) The total fees paid by the insurance carrier (professional, facility, pathology) were $1906. That is nearly 2.5 times more expensive than the usually quoted cost for colonoscopy and a price tag that hardly fits the usual definition of an affordable test for a large-scale screening program.
From the foregoing it is clear that the frequently offered proviso that if virtual colonoscopy shows a polyp, the patient will have to come back for polypectomy is inappropriately discouraging. Given the low 1-9% false-positive rate recorded for 1.0-cm polyps in large series [17-19] and the 8-10% likelihood that an advanced adenoma will be found, the recall rate for colonoscopy and polypectomy should be less than 15%. Whether this level of risk for requiring a second procedure—that is, follow-up colonoscopy—seems a high or a low figure becomes more a matter of individual patient choice. Avoiding the unpleasantness of a second bowel preparation remains the major disincentive. Yet virtual colonoscopy is often performed in a medical setting in which a positive finding could result in an immediate same-day referral to a colonoscopist for polypectomy, thus obviating the re-preparation. By the same token, in most facilities in the United States where virtual colonoscopy is being performed, endoscopists now refer failed or incomplete colonoscopies for immediate same-day, single-preparation add-on virtual colonoscopy. Accommodation for reciprocal same-day conventional colonoscopy should therefore be reasonably straightforward to arrange. However, for this integrated same-day workup to succeed, radiologists will have to provide online interpretation of virtual colonoscopy studies. Ultimately, marketing pressures and patient preference may require it.
|
TopIntroductionPolypsColonoscopyPoliticsReferences |
|---|
As part of evolving competition between conventional and virtual colonoscopy, comparisons of patient preference have been studied and predictably have yielded mixed results [26-28]. Intestinal cramping is common during virtual colonoscopy, but it must be balanced against patient reluctance to undergo sedation and its possible aftereffects.
Virtual colonoscopy has the additional obvious advantage of reassurance. Fenlon et al. [17] and Yee et al. [18] report the negative predictive value for polyps at the 1.0-cm level to be approximately 95%. Even if a small polyp or two were overlooked, for most thoughtful adults exclusion of the possibility of a lurking carcinoma of the bowel is reassurance enough. Indeed, reassurance—the Clean Bill of Health!—is a well-advertised benefit of any medical surveillance effort, especially entrepreneurial CT. If a polyp of intermediate size of 5-8 mm is detected, the patient then has a choice. Increasingly, in the era of consumer control of health matters, patients seek data, facts, and information. Whether over the Internet or from consultative visits with their physician, patients are increasingly accustomed to making medical choices, especially when it comes to therapeutic options. Faced with a report of a possible cancer of the prostate or breast, most patients are likely to seek several opinions and do careful soul-searching before making a choice about therapy. Virtual colonoscopy now gives patients a new option to choose how and when to purge their colon of polyps, and radiologists should not be timid about explaining it.
Reimbursement for virtual colonoscopy as a screening procedure is not yet available through Medicare, although private carriers are beginning to reimburse selectively for specific diagnostic indications. Most are awaiting more evidence from larger clinical trials and reassurance that the technique can be used in community practices with similar performance results similar to those recorded in academic reports, and several multicenter studies are under way.
Although this caution is appropriate, it is instructive to examine the precedent of screening mammography in terms of how long and for what level of evidence the carriers actually waited before permitting reimbursement. Clearly, screening mammography is far from a perfect test, and yet it has been reimbursed as a separate procedure for 20 years. No doubt, emotions added to the results of science in prompting reimbursement for mammography.
Inevitably second-generation technologies for virtual colonoscopy will improve results further. These include multidetector CT, techniques that do not require preparation, and computer-aided detection. The issue will be how long the carriers will wait before approving reimbursement. Radiologists must be willing to make the case and, more important, must be willing and able to mobilize their patients to take up the cause at a grassroots level. Even now, virtual colonoscopy appears twice as accurate as the double-contrast barium enema, which is reimbursable by Medicare for colon screening [12, 17-19]. In the case of mammography, public advocacy was instrumental in gaining reimbursement approval. In the case of colon cancer, many lay patient advocacy groups actively support research efforts for colon cancer prevention [29]. Their support could be pivotal in making virtual colonoscopy more widely available for colon screening.
Finally, the impact of entrepreneurial whole-body CT screening on the dissemination of virtual colonoscopy into clinical practice poses a threat and a concern. Many radiologists who are performing virtual colonoscopy are also offering whole-body CT screening. As Baker [30] and others, including the American College of Radiology [31], have noted, the scientific rationale for screening with whole-body CT is yet unproven. On the other hand, the value of screening for colorectal cancer is quite clear. The issues with colorectal cancer screening are which test or combination of tests should be performed and when will virtual colonoscopy be accepted as a legitimate option. It would be near-disaster if the promise of virtual colonoscopy is lost because of the taint of whole-body CT screening. Radiologists must be sensitive to the distinctions between the scientific uncertainty about the efficacy of whole-body screening versus the more focused uncertainties about the merits of specific tests for colorectal cancer screening. The difference is subtle but crucial if virtual colonoscopy is to emerge and "glister."
|
TopIntroductionPolypsColonoscopyPoliticsReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. J. Pickhardt, V. T. Lehman, T. C. Winter, and A. J. Taylor Polyp volume versus linear size measurements at CT colonography: implications for noninvasive surveillance of unresected colorectal lesions. Am. J. Roentgenol., June 1, 2006; 186(6): 1605 - 1610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Kuehle, P. Veit, G. Antoch, F. Grabellus, P. Robert, T. Beyer, and C. U. Herborn Contrast-Enhanced Dark Lumen PET/CT and MR Colonography in a Rodent Polyp Model: Initial Results with Histopathologic Correlation Am. J. Roentgenol., October 1, 2005; 185(4): 1045 - 1047. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ozgun, E. Rollven, L. Blomqvist, S. Bremmer, R. Odh, and A. Fransson Polyp Detection with MDCT: A Phantom-Based Evaluation of the Impact of Dose and Spatial Resolution Am. J. Roentgenol., April 1, 2005; 184(4): 1181 - 1188. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
