1	Screening virtual colonoscopy -ready for prime time? Yiping Rao, MD, GI fellow Donald M. Switz, MD, FACP, Professor of Medicine
2	Why screen colorectal cancer ? Colorectal cancer is the second leading cause of cancer-related mortality in US (Fatality 56,600 in 2002) Colorectal cancer screening is effective and cost-effective Only a third of Americans older than 50 years old have been screen with colonoscopy or Flex+ACBE 90% women have had at least one Papanicolaou test for cervical cancer, even though considerably fewer women dies from cervical cancer (20,000 in 2002)
3	Limitations of Colonoscopy as a screen test Iv sedations, side effects Recovery time and designated driver Limited experienced specialist Risks of bleeding, perforations and others Relatively low-yield for invasive test
4	Virtual colonoscopy (CT colonoscopy) * was first described in 1994 * required bowel-cleansing, insertion of a rectal tube and the insufflation of air * required approximately 10 to 15 minutes for CT, with an additional 15 to 30 minutes for the interpretation * radiation exposure
5	Recent virtual colonoscopy studies. Symptomatic patients or patients with a moderate-to-high risk of colonic neoplasia. Pineau BC et al. Gastroenterology 2003, 125:304 Johnson CD et al. Gastroenterology 2003, 123;311 Summers RM et al. Radiology 2002, 225:391 Fenlon HM et al. N Engl J Med 1999, 341:1496 Mixed patient populations. Cotton PB et al. A Multicenter Comparison With Standard Colonoscopy for Detection of Colorectal Neoplasia. JAMA.. Published yesterday
6	Important conclusion In high risk patient for CRC, virtual and conventional colonoscopy have similar efficacy for the detection of polyps 5- 6 mm or more in diameter . In mixed population, virtual colonoscopy was less sensitive and specific than conventional colonoscopy
7	Current clinical uses of virtual colonoscopy * Patients who have undergone incomplete conventional colonoscopy Patients with obstructing colorectal cancer Patients whose medical problems make them unsuitable for conventional colonoscopy
8	Concerns 1. flat polyps (>10 mm) may be missed by virtual colonoscopy, especially if they located in the right colon, which appear to be difficult to diagnose with the use of CT, are rarely missed on conventional colonoscopy.
9	Concerns It is also possible that the incident cancers arose from flat lesions. National Polyp Data showed that about 25% of cancers were not prevented despite of screening, which could have been a consequence of missed lesions, such as flat lesions.
10	Concerns 2. Small polyps (<5mm) may not be identified or patient may not be referred for polypectomy. Some of the small lesions may harbor high- grade dysplasia and cancer despite their smaller size.
11	Consensus form the studies Virtual colonoscopy may not be used for screening of high risk patient: Familial Adenomatous Polyposis IBD First degree relatives
12	Question What is the role of virtual colonoscopy in the asymptomatic, average-risk screening population ?
13	Computed Tomographic Virtual Colonoscopy to Screen for Colorectal Neoplasia in Asymptomatic Adults
14	Hypothesis CT colonoscopy could provide an attractive alternative for use in widespread screening.
15	Characteristics of this study A prospective study. Typical asymptomatic screening population. Three-dimensionalapproach to the study and interpretation of the radiographicimages.
16	Methods Study Group: 1233 asymptomatic adults at three medicalcenters. Adults between 50 and 79 years of age with an average risk ofcolorectal cancer made up the primary study group.
17
18	Study Design Image processing and interpretation: Viatronix software: 1. extractsthe images of the air-filled colon. 2.generates an automated centerlinefor luminal navigation. 3. electronically cleansing: removesfrom images the residual fluid in postprocessingstep. 4. interface creates a three-dimensional image.
19
20	Study Design Polyps were measured with electronic calipers on thethree-dimensional view, recorded according to the segment.
21	Study Design Same day colonoscopy was performed by endoscopiest. Scope was inserted to the cecum and sequentially withdrawn segment by segmentfor the detection of polyps. Polyps were photographedand measured with a calibrated linear probe
22	Segmentalunblinding Astudy coordinator revealed the results of the virtual colonoscopyfor the previously examined segment. If a polyp measuring 5mm or more in diameter was seen on virtual colonoscopy but noton the initial optical colonoscopy, the colonoscopist closelyreexamined that segment and was allowed to review the imagesobtained on virtual colonoscopy for guidance.
23	Time recorded/Questionnaire The time was recorded: in the CT suite in the endoscopysuite in recovery area for the interpretation of virtual colonoscopic studies All study patients were given a questionnaire to complete at home and return by mail.
24	Study interests Primary interest: adenomatous polyps measuring6 mm or more in diameter. advanced neoplasia was defined asany adenoma measuring 10 mm or more in diameter or/and high-grade dysplasia, or cancer. Secondary interest : hyperplastic polypsand polyps measuring 5 mm or lesser in diameter
25	True positive Match between virtual and opticalcolonoscopy: * appearing within the samesegment or in adjacent segments. ** two recorded diametershad to be the same, within a 50 percent margin of error.
26	Results
27
28
29	Interobserver agreement Some of colonoscopicstudies were read twice, with a segmental agreement rate: 99.6 percent (797 of 800 polyps) for polyps 10 mm or larger, 99.1 percent (793 of 800) for polyps 8 mm or larger, 97.6percent (781 of 800) for polyps 6 mm or larger ( = 0.75 to 0.80). Agreement according to the patient was 95 percent for polyps8 mm or larger ( = 0.79).
30	Time spent In the CT suite = 14.1 minutes. In the endoscopy suite =31.5 minutes (P<0.001)
31	Satisfaction with virtualcolonoscopy Excellent= 40.6 percent Very good= 32.8percent Good=17.9 percent Fair= 6.3 percent Poor= 2.4 percent
32	Preferences for future screening 500 patients (49.8percent) preferred virtual colonoscopy. 413 patients (41.1percent) preferred optical colonoscopy (P=0.004) 92 patients(9.2 percent) had no preference or were undecided
33	Discussions
34	Size matters Studies of virtual colonoscopy have focused on three categoriesof polyp size: 5 mm or smaller 6 to 9mm 10 mm or larger
35	Size matters Polyps < 5mm shouldbe regarded as clinically “insignificant”. Only a minority of these smalllesions are adenomatous, less than 1 % arehistologically advanced and virtually none are malignant.
36	Size matters The 10-mm polyp has been the benchmark for virtualand optical colonoscopy, relatively few advancedlesions measure less than 10 mm. A 10-mm threshold would allow the use of virtual colonoscopyfor screening cost effective, only about 1 of every 13average-risk patients would be referred for optical colonoscopy.
37	Size matters 6 to 9 mm polyps may be considered into one category. 8 mm might be a reasonable threshold fortriggering immediate optical colonoscopy. Intermediate-size lesions (perhaps 5 to 7 mm) might be bestserved by the use of short-term virtual surveillance.
38	The percentage of asymptomatic patients who undergo virtualcolonoscopy for screening but would not require optical colonoscopy
39	Blind spot ^{Most of the polyps that were found on virtual colonoscopy but not on the initial optical colonoscopy were situated behind a colonic fold, which is a relative blind spot for optical endoscopy but not for virtual endoscopy.}
40	2 CRCs found in 1233 patients Two carcinomaswere detected in our asymptomatic population. one was detected by both tests. one located on a fold near the hepaticflexure was detected on optical colonoscopy only after the revelationof the results from the virtual colonoscopy.
41	Technique limit factors suboptimal preparation of the colon. the limited experience with this new procedure among radiologists. operator dependence as reflected by high interobserver variability
42	Three-dimensional CT Three-dimensional approach for the detection resolves some uncertainties between the radiologist.
43	Electronic cleansing The vigorous bowel preparation + electronic cleansing result excellent results and reduce the interobserver variability.
44	Hyperplastic polyps The authors classified nonadenomatous polyps as false positive results, because “they are not associated with the risk of cancer and are not the target of screening”. As results, high sensitivity and low specificity were seen in this study.
45	Unresolved issues Threshold for optical colonoscopy? offer virtual colonoscopy as an initial screening procedure. refer all patients with polyps larger than a predetermined size limit (6mm? 7mm?) for same-day conventional.
46	Unremoved polyps Some physicians and patients might want smaller polyps to be removed for fear of delaying the removal of a small cancer. The current study does not provide any data that would support an evidence-based surveillance strategy for the follow-up of smaller polyps that are not removed.
47	Patient may the real thing 70 % of the patients found their virtual colonoscopic examination more acceptable than conventional colonoscopy. However, for a future screening test, only 50 % chose virtual colonoscopy (41 % chose conventional colonoscopy). Patient interested in screening may question why they should get a virtual colonoscopy when they could get the real thing.
48	Virtual colonoscopy: ready for prime time? 1. If the results of this study are reproducible on large scale 2. If the appropriate size threshold and the intervals for surveillance of smaller polyps can be resolve 3. If the insurers willingly to pay the screen virtual colonoscopy 4. Radiation issues