Correspondence
Virtual Colonoscopy
N Engl J Med 2000; 342:737-739March 9, 2000
- Article
To the Editor:
Like Fenlon et al. (Nov. 11 issue),1 we undertook a prospective study to establish the sensitivity, specificity, and accuracy of virtual colonoscopy. Of 25 patients (median age, 71 years) with large-bowel symptoms (mainly rectal bleeding) who underwent virtual colonoscopy and then conventional colonoscopy immediately afterward (complete to the cecum in 24 patients), 5 had a polyp or cancer larger than 10 mm (range, 10 to 50), and 1 had an amyloid deposit. Only one of these lesions was identified with virtual colonoscopy. Four patients had false positive findings of polyps (polyp size, 6 to 40 mm); in two of these cases a suspicion of carcinoma was reported (one in the rectum and one in the descending colon).
Virtual colonoscopy was performed in a manner similar to that described by Fenlon et al., in which helical computed tomography was used. All patients received intravenous contrast material and a smooth-muscle relaxant and underwent scanning in both supine and prone positions. Images were acquired with the use of 5-mm collimation and reconstructed at 3-mm intervals. The radiologist who viewed the scans had experience with approximately 50 virtual colonoscopic procedures in patients known to have polyps or cancer.
It is difficult to understand why our results were so poor in relation to those of Fenlon et al. However, patient selection may be important. To avoid biasing our results, we excluded patients from the study who had previous or recent sigmoidoscopic evidence of one or more polyps. Furthermore, 52 percent of our patients had substantial diverticular disease both in the sigmoid and throughout the colon. This condition may adversely affect the results of virtual colonoscopy because of narrowing of the colonic lumen, thickening of the bowel wall, and the presence of solid fecal matter adhering to the bowel wall at the site of diverticula.
1 ReferencesScott Mackenzie, M.B., Ch.B.
Ramsay Vallance, M.B., Ch.B., D.M.R.D.
Patrick J. O'Dwyer, M.Ch.
University of Glasgow, Glasgow G11 6NT, United Kingdom1
Fenlon HM ,Nunes DP ,Schroy PC III ,Barish MA ,Clarke PD ,Ferrucci JT . A comparison of virtual and conventional colonoscopy for the detection of colorectal polyps. N Engl J Med 1999;341:1496-1503
Full Text | Web of Science | Medline
To the Editor:
In the article by Fenlon et al. and the accompanying editorial by Bond,1 which discussed the relative merits of virtual and conventional colonoscopy for screening, the use of a double-contrast barium enema was casually dispatched. Fenlon et al. contend that their data “also suggest that virtual colonoscopy is superior to double-contrast barium enema,” and they cite a sensitivity for the latter of 65 to 75 percent for the detection of polyps larger than 7 mm. However, neither of the two papers that are cited mention the sensitivity of barium enema. Two articles cited elsewhere report sensitivities of 81 percent2 and 98 percent3 for polyps larger than 10 mm. Fenlon et al. state that “the results of double-contrast barium enema are highly dependent on the skill of the operator.” Once again, the two references cited are unrelated to the issue. Furthermore, the accuracy of any procedure depends on the skill of the operator. Because most important neoplasms that are missed with a double-contrast barium enema are visible when the results are reexamined, concentration and perception are the keys to enhancing proficiency. The American College of Radiology will soon implement an accreditation program for the performance of double-contrast barium enema, which is likely to improve the accuracy of the procedure.
On the other hand, the data on virtual colonoscopy have been produced only in highly dedicated centers, where 20 to 30 minutes are devoted to the interpretation of each examination. Speculation about the likelihood of widespread reproducibility has no basis. Even in this controlled setting, only 87 percent of the studies were complete, whereas the double-contrast barium enema rarely fails to permit visualization of the entire colon.
At present, virtual colonoscopy is not competitive in terms of costs with double-contrast barium enema. As compared with conventional colonoscopy, the double-contrast barium enema is one third to one fourth as expensive and much safer. The rate of perforation with double-contrast barium enema is 1 in 25,000.4 Bond also states that “in most community hospitals barium-enema imaging is probably insufficiently sensitive.” However, in the reference that Bond cites,5 this point is not addressed; rather, after analyzing the literature, the author of the study concludes that there is no significant difference between the two procedures in detecting polyps larger than 10 mm.
5 ReferencesSeth N. Glick, M.D.
MCP Hahnemann University, Philadelphia, PA 191021
Bond JH . Virtual colonoscopy -- promising, but not ready for widespread use. N Engl J Med 1999;341:1540-1542
Full Text | Web of Science | Medline2
Steine S ,Stordahl A ,Lunde OC ,Loken K ,Laerum E . Double-contrast barium enema versus colonoscopy in the diagnosis of neoplastic disorders: aspects of decision-making in general practice. Fam Pract 1993;10:288-291
CrossRef | Web of Science | Medline3
Williams CB ,Hunt RH ,Loose H ,Riddell RH ,Sakai Y ,Swarbrick ET . Colonoscopy in the management of colon polyps. Br J Surg 1974;61:673-682
CrossRef | Web of Science | Medline4
Blakeborough A ,Sheridan MB ,Chapman AH . Complications of barium enema examinations: a survey of UK Consultant Radiologists 1992 to 1994. Clin Radiol 1997;52:142-148
CrossRef | Web of Science | Medline5
Schrock TR . Colonoscopy versus barium enema in the diagnosis of colorectal cancer and polyps. Gastrointest Endosc Clin N Am 1993;3:585-610
To the Editor:
It is important to clarify how the nonvisualized portion of the colon was factored into the analysis by Fenlon et al. Was the nonvisualized portion, whether in virtual or conventional colonoscopy, removed from the analysis, or was it assumed to have a false negative or false positive result? In addition, no reference is made to any attempt to examine the nonvisualized portion of the colon with, for example, a pediatric colonoscope1 or air-contrast barium enema. In addition, the use of conventional colonoscopy, which had a crude completion rate of 89 percent, as the gold standard may be suboptimal.
1 ReferencesJoseph C. Anderson, M.D.
Bonnie J. Pollack, M.D.
Robert D. Shaw, M.D.
State University of New York at Stony Brook, Stony Brook, NY 117941
Kozarek RA ,Botoman VA ,Patterson DJ . Prospective evaluation of a small caliber upper endoscope for colonoscopy after unsuccessful standard examination. Gastrointest Endosc 1989;35:333-335
CrossRef | Web of Science | Medline
To the Editor:
It is worth noting that the position of conventional colonoscopy as the gold standard in visualizing the colon is questionable. Fenlon et al. report rates of false negative results with virtual colonoscopy (9 percent for large polyps and 18 percent for medium-sized polyps) that are similar to the rates reported in studies of tandem colonoscopy performed by experienced endoscopists (up to 6 percent for polyps larger than 10 mm and as high as 13 percent for 6-to-9-mm polyps).1,2
Arguments will continue about the relative diagnostic accuracy of conventional and virtual colonoscopy, but if the latter could increase the rates of patient compliance by only 10 to 20 percent, it would be an effective screening examination.3 It might be possible to increase this rate, because virtual colonoscopy is less costly, less invasive, and less time-consuming from the patient's perspective, and it requires no sedation.
Therapeutic colonoscopy in patients with positive results on virtual colonoscopy should be performed the same day as the virtual colonoscopy, thus sparing the patient a second bowel preparation. This will surely be possible in large screening centers, where the length of time required for interpreting images from virtual colonoscopy will be reduced as technology and expertise improve.
3 ReferencesMartina M. Morrin, M.D.
Richard J. Farrell, M.D.
Jonathan B. Kruskal, M.D., Ph.D.
Beth Israel Deaconess Medical Center, Boston, MA 022151
Rex DK ,Cutler CS ,Lemmel GT , et al. Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies. Gastroenterology 1997;112:24-28
CrossRef | Web of Science | Medline2
Hixson LJ ,Fennerty MB ,Sampliner RE ,McGee D ,Garewal H . Prospective study of the frequency and size distribution of polyps missed by colonoscopy. J Natl Cancer Inst 1990;82:1769-1772
CrossRef | Web of Science | Medline3
Sonnenberg A ,Delco F ,Bauerfeind P . Is virtual colonoscopy a costeffective option to screen for colorectal cancer? Am J Gastroenterol 1999;94:2268-2274
CrossRef | Web of Science | Medline
To the Editor:
I can tell you that from the patient's point of view, virtual colonoscopy will not be perceived as a major advance unless or until it can be preceded by a virtual bowel preparation, in which virtual GoLytely is used. Then and only then would I embrace the world of virtual reality with genuine enthusiasm.
Carl Silverman, M.D.
20 S. Park St., Madison, WI 53711Author/Editor Response
The authors reply:
To the Editor: We share the concern of Mackenzie et al. regarding the results of their study. However, other centers have published peer-reviewed data1 with results that more closely approached ours. It is difficult to resolve this discrepancy without having further information on the specific techniques used, including the method of colon cleansing and gas insufflation, image-rendering algorithms, methods of image analysis (two-dimensional fly through, three-dimensional fly through, both, or other), and interpretive criteria. We would hope that there might be an explanation in some of these methods used.
Glick rises to the defense of the double-contrast barium enema, but the data he cites from previous reports would be difficult to reproduce today (even if there were an adequate number of procedures) by either academic or private radiology practitioners in the United States. We do, however, support the plan of the American College of Radiology to reemphasize the techniques for double-contrast barium enema, and we certainly await the results of that effort. Anderson et al. ask about the analysis of the nonvisualized portion of the colon. We assure them that it was included in the data in terms of sensitivity both per polyp and per patient. Whether our 89 percent completion rate with conventional colonoscopy is substandard is beside the point, because that was the rate obtained in this study. Morrin et al. properly point to the well-known rates of missed polyps in tandem colonoscopy studies, which can be up to 6 percent for polyps larger than 10 mm. We also note with interest the proposal in Bond's editorial to link the availability of virtual and conventional therapeutic colonoscopy in the context of large screening centers.
Finally, Silverman is justified in wishing for a virtual colon preparation, and indeed, the development of possible dietary fecal tagging agents is under way. These advances would make colon cleansing unnecessary and permit enhanced image contrast between stool and polyp tissue. For the present, however, a clean colon will remain the sine qua non for anatomical inspection of the colon mucosa, whether by barium enema, endoscopy, or virtual imaging.
1 ReferencesHelen H. Fenlon, M.D.
Matthew A. Barish, M.D.
Joseph T. Ferrucci, M.D.
Boston Medical Center, Boston, MA 021181
Hara AK ,Johnson CD ,Reed JE , et al. Detection of colorectal polyps with CT colography: initial assessment of sensitivity and specificity. Radiology 1997;205:59-65
Web of Science | Medline
Author/Editor Response
I disagree with the contention of Glick that double-contrast barium enema was “casually dispatched” in my editorial on the current status of virtual colonoscopy. Comparison studies and the accumulated experience of clinicians who specialize in digestive disease clearly indicate that double-contrast barium enema — performed in either community hospitals or academic centers — is substantially less accurate than conventional colonoscopy for detecting both colorectal cancer and polyps of all sizes.1,2 The accreditation program planned by the American College of Radiology for the performance of the procedure appears to acknowledge this fact. Whether the program will help eliminate perceptual errors in image interpretation remains to be seen. Virtual colonoscopy is a promising new radiographic advance that has the potential to overcome most of the inherent limitations of double-contrast barium enema and to detect most colorectal cancers and advanced adenomatous polyps.
Although colonoscopy is a useful convention in clinical studies of new technology, it should not be considered a gold standard in clinical practice. As pointed out by Morrin et al., even when performed competently by well-trained, experienced physicians, conventional colonoscopy fails to detect an appreciable number of small polyps and about 1 to 3 percent of large polyps and malignant tumors.2,3
3 ReferencesJohn H. Bond, M.D.
Minneapolis Veterans Affairs Medical Center, Minneapolis, MN 554171
Hogan WJ ,Stewart ET ,Geenen JE ,Dodds WJ ,Bjork JT ,Leinicke JA . A prospective comparison of the accuracy of colonoscopy vs air-barium contrast exam for detection of colonic polypoid lesions. Gastrointest Endosc 1977;23:230-230 abstract.
Web of Science2
Rex DK ,Rahmani EY ,Haseman JH ,Lemmel GT ,Kaster S ,Buckley JS . Relative sensitivity of colonoscopy and barium enema for detection of colorectal cancer in clinical practice. Gastroenterology 1997;112:17-23
CrossRef | Web of Science | Medline3
Rex DK ,Cutler CS ,Lemmel GT , et al. Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies. Gastroenterology 1997;112:24-28
CrossRef | Web of Science | Medline
