Computed tomographic (CT) colonography is gaining favor as a frontline screening modality for colorectal cancer ( 1 ). Researchers in a number of stud

Transcription

1 Note: This copy is for your personal, non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at Perry J. Pickhardt, MD Nathan A. Durick, MD B. Dustin Pooler, MD Cesare Hassan, MD Left-sided Polyps Detected at Screening CT Colonography: Do We Need Complete Optical Colonoscopy for Further Evaluation? 1 Purpose: Materials and Methods: To estimate the relative yield of therapeutic flexible sigmoidoscopy compared with complete optical colonoscopy for isolated left-sided polyps ( 6 mm in diameter) identified at screening computed tomographic (CT) colonography. This retrospective institutional review board approved and HIPAA-compliant study included a review of CT colonographic screening results in 6570 consecutive asymptomatic adults (3551 women, 3019 men; mean age, 56.8 years [standard deviation]). Of 887 (13.5%) patients with cases positive for nondiminutive polyps ( 6 mm), a subset of 171 patients met the inclusion criteria (a) of having left-sided only lesions of 6 mm or larger identified at CT colonography (rectum-to-splenic flexure) and (b) of undergoing subsequent evaluation with complete optical colonoscopy. CT colonography optical colonoscopy concordance and proximal-versus-distal diagnostic yield at optical colonoscopy were assessed. The 95% confidence intervals (CIs) were calculated for relevant results. ORIGINAL RESEARCH n GASTROINTESTINAL IMAGING 1 From the Department of Radiology, University of Wisconsin School of Medicine and Public Health, E3/311 Clinical Science Center, 600 Highland Ave, Madison, WI (P.J.P., N.A.D., B.D.P.); and Digestive Endoscopy Unit, Nuovo Regina Margherita Hospital, Rome, Italy (C.H.). Received August 24, 2010; revision requested September 29; fi nal revision received November 18; accepted December 22; fi nal version accepted December 28. Address correspondence to P.J.P. ( ppickhardt2@uwhealth.org ). Results: Conclusion: From the study group of 171 patients, a total of 234 leftsided lesions of 6 mm or larger were prospectively reported at CT colonography, of which 222 (94.9%; 95% CI: 91.3%, 97.0%) were confirmed as true-positive findings at optical colonoscopy. With optical colonoscopy, an additional 17 benign left-sided polyps of 6 mm or larger (13 small, four large) were found in 11 patients, resulting in a total left-sided yield of 239 nondiminutive lesions, including 137 small polyps (6 9 mm) and 102 large lesions ( 10 mm), 160 neoplasms, 82 advanced adenomas, and seven cancers. Evaluation of the colon proximal to the splenic flexure in this cohort yielded eight small and two large benign polyps in nine patients but no proximal cancers or histologically advanced lesions. For patients with left-sided only polyps detected at CT colonography, the additional yield of complete optical colonoscopy beyond the expected reach of flexible sigmoidoscopy is very low and may not justify the added costs, potential risks, and procedural time. q RSNA, 2011 q RSNA, 2011 Radiology: Volume 259: Number 2 May 2011 n radiology.rsna.org 429

2 Computed tomographic (CT) colonography is gaining favor as a frontline screening modality for colorectal cancer ( 1 ). Researchers in a number of studies have validated CT colonography as a safe and effective screening examination ( 2 4 ). Any implementation of primary CT colonographic screening would also lead to an increase in therapeutic endoscopy for removing relevant polyps identified at CT colonography. At our institution, most patients who were referred for endo scopic polypectomy had undergone complete optical colonoscopy to the cecum, regardless of the specific location of the CT colonographic findings. The main disadvantage of flexible sigmoidoscopy relative to optical colonoscopy is that only the rectum and distal colon are evaluated, typically to the level of the splenic flexure or at least 40 cm of scope insertion ( 5 ). Flexible sigmoidoscopy, however, does have several advantages over optical colonoscopy it is generally safer, faster, and less costly compared with optical colonoscopy, requires less extensive bowel preparation, and can often be performed with little or no sedation ( 1 ). Given these relative advantages for flexible sigmoidoscopy, if there is low additional diagnostic yield from performing complete optical colonos copy for cases that are positive only for left-sided (distal, according to Advances in Knowledge n In patients with isolated left- sided polyps of 6 mm or larger detected at screening CT colonography, the diagnostic yield of right-sided evaluation (proximal to the splenic flexure) with optical colonoscopy may be too low to justify the additional time, risks, and costs compared with flexible sigmoidoscopy alone. n The concordance rate between CT colonography and optical colonoscopy for left-sided lesions was 95%, with matched lesions dominating over CT colonographic false-positive and falsenegative results. the definition below) findings at CT colonography, then flexible sigmoidoscopy may be the more suitable option in this setting because the right-sided (proximal) colon has already been evaluated by using CT colonography. Individuals with left-sided neoplastic lesions, especially when advanced, are at proved higher risk for right-sided (proximal) neoplasia ( 6 ). As such, optical colonoscopy is warranted following a positive result of screening with flexible sigmoidoscopy. With CT colonographic screening, right-sided lesions are reliably detected. Therefore, performing optical colonoscopy after CT colonography in patients with results positive for only left-sided lesions may represent a duplication of right-sided screening efforts. The purpose of this study was to estimate the relative yield of flexible sigmoidoscopy compared with complete optical colonoscopy as the therapeutic procedure for isolated left-sided polyps of 6 mm or larger identified at screening CT colonography. Materials and Methods This Health Insurance Portability and Accountability Act compliant study was approved by our institutional review board (University of Wisconsin School of Medicine and Public Health, Madison, Wis); the need for signed informed consent was waived owing to the retrospective nature of the assessment. Retrospective review was performed in 6570 consecutive asymptomatic adults who underwent primary CT colonographic screening at our institution from April 2004 through April The mean age of this screening population was 56.8 years (standard deviation), with an age range of years, Implication for Patient Care n When only left-sided colorectal polyps are detected at screening CT colonography and relevant lesions proximal to the splenic flexure are excluded, flexible sigmoidoscopy may be a more appropriate therapeutic test than complete optical colonoscopy. and included 3551 women (mean age, 56.6 years; range,40 97 years) and 3019 men (mean age, 57.1 years; range, years). The CT colonographic technique employed at our institution has been previously described ( 3,4,7,8 ). To briefly summarize, patients underwent bowel preparation consisting of a saline laxative (currently, magnesium citrate; previously, sodium phosphate) followed by administration of oral contrast material (2% barium and diatrizoate) for stool and fluid tagging. Colonic distention was achieved with automated low-pressure CO 2 delivery. Supine and prone low-dose acquisitions were obtained with eight and 16 detector row CT scanners (Light- Speed; GE Healthcare, Waukesha, Wis), with images reconstructed with 1.25-mm section thickness at 1-mm intervals. Image interpretation was performed with dedicated software by using a combined three-dimensional and two-dimensional detection strategy ( 7 ). In our CT colonographic screening program, individuals with a positive study result, defined as one or more polyps of 6 mm or larger, are offered same-day endoscopic polypectomy ( 1 ). In addition, individuals with one or two small polyps (6 9 mm) are also eligible for short-term CT colonographic surveillance under a research protocol ( 3,9 ). From the entire screening cohort of 6570 adults, there were 887 (13.5%) cases with positive results at CT colonography, with a positive result defined by Published online before print /radiol Radiology 2011; 259: Abbreviation: CI = confi dence interval Author contributions: Guarantors of integrity of entire study, P.J.P., C.H.; study concepts/study design or data acquisition or data analysis/ interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of fi nal version of submitted manuscript, all authors; literature research, P.J.P., N.A.D.; clinical studies, all authors; statistical analysis, N.A.D., B.D.P.; and manuscript editing, P.J.P., N.A.D., B.D.P. Potential confl icts of interest are listed at the end of this article. 430 radiology.rsna.org n Radiology: Volume 259: Number 2 May 2011

3 the presence of at least one polyp that was 6 mm or larger. To correspond with the expected coverage of sigmoidoscopy ( 1 ), left-sided or distal lesions were defined as being located anywhere from the rectum to the splenic flexure at CT colonographic examination. This definition is analogous to that adopted in previous optical colonoscopic studies for simulating flexible sigmoidoscopic yield ( 6 ), with the main difference being that CT colonography is superior to optical colonoscopy for precise anatomic localization. Cases with positive results at CT colonography were then classified into two groups: cases in patients with only left-sided polyps detected at CT colonography and cases in patients with proximal (right-sided) polyps, either with or without associated left-sided lesions. The group of patients with left-sided only polyps at CT colonography was further analyzed by using our clinical CT colonographic database and electronic medical record review. Patients from this group were excluded from further consideration if they underwent CT colonographic surveillance instead of polypectomy, if the endoscopic examination did not reach the cecum for any reason (whether intentional or not), if the optical colonoscopic examination was performed more than 1 month after CT colonography, or if they were referred from an outside provider to our CT colonographic screening program and were subsequently treated elsewhere. After the above exclusion criteria were applied, the remaining patients formed the study cohort ( Figure ). For each patient, the pertinent CT colonographic, optical colonoscopic, and histopathologic findings were analyzed. Of greatest interest were any nondiminutive polyps proximal to the splenic flexure detected at optical colonoscopy, because these polyps would presumably be beyond the range of therapeutic flexible sigmoidoscopy. According to the study criteria, these lesions would have been missed at CT colonography, and their localization and size were therefore purely based on endoscopic classification. Diminutive lesions found at optical colonoscopy, which are generally not reported in detail at CT colonography, were also considered, although they do not necessarily represent missed lesions per se at CT colonography. Polyps were characterized by the maximum linear size, segmental location, morphologic characteristics, and histologic findings ( 7,10 ). Optical colonoscopic polyp characteristics were used for all diminutive lesions and CT colonographic false-negative results. Large polyps were defined as 10 mm or larger, small polyps were defined as in the 6 9-mm range, and diminutive polyps were defined as 5 mm or smaller. Advanced adenomas were defined by a large size and/or by the presence of histologic features of a prominent villous component or high-grade dysplasia. A malignant lesion implied invasive carcinoma. For left-sided lesions, a standard CT colonographic optical colonoscopic matching algorithm was applied for lesion size (within 50%) and location (same or adjacent segment) ( 4 ). Matching was generally straightforward, as most patients had a solitary polyp detected at CT colonography. For difficult cases with multiple polyps, a consensus review was performed by two radiologists (P.J.P. and N.A.D., with 15 and 5 years of experience, respectively) to resolve any potential issues with CT colonographic optical colonoscopic correlation. The concordance rate reflects the frequency of an associated optical colonoscopic correlate for a polyp detected with CT colonography. A discordant lesion may represent either a CT colonographic false-positive result or an optical colonoscopic false-negative result. For additional lesions identified at optical colonoscopy that were not detected at CT colonography, lesion size was based on visual estimation at optical colonoscopy. When appropriate, 95% confidence intervals (CIs) were calculated (Excel; Microsoft, Redmond, Wash). Results Flowchart of CT colonography (CTC) screening population and selection of fi nal study cohort of 171 subjects who met all inclusion criteria. OC = optical colonoscopy. From the entire screening cohort of 6570 asymptomatic adults, at least one polyp of 6 mm or larger was detected with CT colonography in 887 (13.5%) individuals. Of these patients with positive results at CT colonography, leftsided only polyps were identifed in 415 (46.8%) ( Figure ). In a total of 472 patients, proximal polyps were detected, either with or without associated leftsided lesions ( Figure ). After applying the exclusion criteria outlined in Materials and Methods to the 415 patients with only left-sided lesions, the final study cohort of 171 patients was formed ( Figure ). The mean age of the study cohort was 57.9 years (range, years), including 93 men (mean age, 58.2 years; range, years) and 78 women (mean age, 57.5 years; range, years). Of the 244 patients with left-sided only lesions who were excluded, 30 underwent endoscopic evaluation that did not reach the cecum (either intentionally related to patient Radiology: Volume 259: Number 2 May 2011 n radiology.rsna.org 431

4 Proximal (Right-sided) versus Distal (Left-sided) Diagnostic Yield at Colonoscopy in 171 Individuals with Left-Sided Only Findings at Screening CT Colonography Data Proximal Yield * Distal Yield No. of patients Total polyps Large ( 10 mm) Small (6 9 mm) Diminutive ( 5 mm) Total neoplasms Nondiminutive ( 6 mm) Tubular adenoma Tubulovillous adenoma 0 41 Villous adenoma 0 3 Invasive cancer 0 7 Advanced neoplasia 2 82 Serrated adenoma 1 7 Other benign neoplasm 0 4 Total nonneoplastic or not retrieved Hyperplastic Other 0 26 Not retrieved 2 35 * Implies a location proximal to the splenic fl exure. Subcategories are not all mutually exclusive. All lesions not retrieved (eg, cauterized, lost, or left in place) were diminutive in size except for four small (6 9-mm) left-sided polyps. tical colonoscopy, and these polyps included 13 small and four large polyps, for a total left-sided yield of 239 nondiminutive lesions. This would correspond to a CT colonographic sensitivity for left-sided nondiminutive polyps of 222 of 239 or 92.9% (95% CI: 88.9%, 95.5%), including 124 of 137 or 90.5% (95% CI: 84.4%, 94.4%) for small and 98 of 102 or 96.1% (95% CI: 90.4%, 98.5%) for large lesions. The 239 left-sided nondiminutive lesions consisted of 137 small polyps (6 9 mm) and 102 large lesions ( 10 mm). The size range was 6 mm to 8 cm. Polyp location included the rectum ( n = 74; 31.0%), sigmoid colon ( n = 130; 54.4%), and descending colon to include the splenic flexure ( n = 35; 14.6%). There were 14 masses ( 3 cm). Histologic findings consisted of 155 total neoplasms (benign and malignant), 82 advanced adenomas, and seven cancers ( Table ). Ninety-six benign left-sided diminutive lesions were seen at optical colonoscopy ( Table ), of which 53 (55.2%) were clustered in six patients. A total of 31 leftcomorbidities or unintentionally from incomplete optical colonoscopy). A total of 209 patients did not undergo endoscopy but instead underwent CT colonographic surveillance for small (6 9-mm) unresected polyps. Five patients were excluded because the interval between CT colonography and optical colonoscopy was longer than 1 month. The yield of optical colonoscopy for proximal versus distal lesions for the study cohort of 171 patients is summarized in the Table. There was a high level of agreement between the left-sided polyps detected at CT colonography and at subsequent optical colonoscopy. Of the 234 total left-sided nondiminutive lesions reported at CT colonography, 222 (94.9%; 95% CI: 91.3%, 97.0%) were confirmed as true-positive results at optical colonoscopy and 12 (5.1%) were false-positive results (or possibly optical colonoscopic falsenegative results in some cases). An additional 17 benign left-sided polyps of 6 mm or larger (CT colonographic falsenegative results) were found with op- sided diminutive lesions were cauterized or otherwise not retrieved, and histologic findings were therefore not available for these lesions. Proximal (right-sided) evaluation by using optical colonoscopy yielded 10 nondiminutive polyps in nine patients from the entire cohort of 171 patients with left-sided only CT colonographic findings ( Table ), corresponding to a perpatient miss rate of nondiminutive polyps of 5.3% (95% CI: 2.4%, 9.7%). None of these 10 proximal polyps missed at CT colonography was histologically advanced or malignant; eight were small (6 9 mm) and two were large (12 and 14 mm). Histologic findings included five tubular adenomas, four hyperplastic polyps, and one serrated polyp. In addition, 26 diminutive lesions scattered proximal to the splenic flexure were found in 14 patients, and these lesions included 17 tubular adenomas and five hyperplastic polyps (some lesions were not resected or retrieved). Discussion The use of flexible sigmoidoscopy as a primary screening tool has diminished considerably in the United States during the past 2 decades, because of relatively low reimbursement rates, lack of trained personnel, and the rise of reimbursed complete optical colonoscopy for screening ( 11,12 ). Renewed interest in flexible sigmoidoscopy, however, may arise owing to the recent randomized controlled flexible sigmoidoscopy trial by Atkin et al ( 13 ) in the United Kingdom, which for the first time, to our knowledge, demonstrated substantially decreased mortality related to endoscopic screening. In particular, the protective benefit against left-sided colorectal cancer, coupled with the apparent lack of benefit related to right-sided cancers at colonoscopic screening ( 14 ), further improve the future prospects for primary screening with flexible sigmoidoscopy. Our study considers the use of flexible sigmoidoscopy not as a primary screening test but rather as a therapeutic procedure following screening CT colonography with positive results for left-sided polyps. Currently, complete 432 radiology.rsna.org n Radiology: Volume 259: Number 2 May 2011

5 optical colonoscopy represents the standard therapeutic complement following a positive CT colonographic examination result. As an evolving tool that has been used in the screening realm for less than a decade, it was a reasonable first step to rescreen the entire colon with optical colonoscopy during this initial phase of CT colonographic implementation. This factor was also related to the evidence for a higher risk of right-sided advanced neoplasia in those with left-sided lesions ( 6 ). However, as CT colonographic technique has now matured and its diagnostic accuracy for screening has been repeatedly demonstrated ( 2,4,15 ), the need for complete optical colonoscopy (versus flexible sigmoidoscopy) following a CT colonographic study with positive results only for left-sided polyps deserves consideration. Our findings show that the redundant evaluation of the proximal colon when nondiminutive polyps have already been excluded by using CT colonography is a very low-yield activity. In this large screening cohort, redundant right-sided evaluation with optical colonoscopy produced only five additional nondiminutive polyps and no advanced adenomas or cancers, which in our opinion does not appear to justify the additional costs, procedure time, and potential complications of a complete lower endoscopy. The yield of right-sided diminutive lesions found at complete optical colonoscopy was also quite low and substantially less than the number of left-sided diminutive lesions (18 versus 72).To our knowledge, researchers in only one prior study have looked at the diagnostic yield of full optical colonoscopy following a positive result at CT colonography for distal polyps ( 16 ). In this work, Young et al ( 16 ) studied 203 patients with only rectosigmoid polyps detected at CT colonography and also showed a relatively small additional gain from right-sided evaluation at optical colonoscopy, with nondiminutive proximal advanced adenomas uncovered in three patients (1.5%). Young et al concluded that the CT colonographic error rate for detecting important proximal lesions was comparable to the optical colonoscopic error rate, perhaps justifying the use of flexible sigmoidoscopy in this setting. One major difference between the study of Young et al and ours was that 63 (31%) of the 203 patients with isolated rectosigmoid polyps identified at CT colonography in the study of Young et al did not have matching lesions at subsequent optical colonoscopy. This discordant rate is much higher than we observed in our study, where 95% of left-sided lesions were matched. In general, the positive predictive value of CT colonography in our screening program was more than 90% for all polyps 6 mm or larger ( 17 ). This disparity may also help to explain the lower rate of proximal polyps found at optical colonoscopy in our CT colonogaphic cohort with positive results for left-sided lesions. Although flexible sigmoidoscopy is intended for evaluation of the entire left colon to the level of the splenic flexure, insertion beyond 40 cm has been suggested as a reasonable surrogate that is more reproducible and quantifiable, because identification of the splenic flexure at endoscopy is imprecise ( 1,5 ). Not surprisingly, the perforation rate at flexible sigmoidoscopy is considerably lower compared with that at complete optical colonoscopy ( ). Because flexible sigmoidoscopy is typically performed without sedation, the associated risks, costs, and recovery time are further reduced. Not surprisingly, the actual procedure length is typically shorter as well. The theoretical risk of combustion with flexible sigmoidoscopy is essentially eliminated in the therapeutic scenario after CT colonography, because all patients have undergone cathartic bowel preparation for CT colonography ( 22 ). Standard Medicare reimbursement rates for optical colonoscopy and flexible sigmoidoscopy are $ and $125.63, respectively ( ). At our institution, the charge for complete optical colonoscopy is even higher relative to flexible sigmoidoscopy, even when additional sedation-related charges for therapeutic flexible sigmoidoscopy are included. In truth, this marked disparity represents a substantial financial disincentive for endoscopy suites to switch over from performing flexible sigmoidoscopy instead of optical colonoscopy. Prior to the recent landmark randomized controlled trial by Atkin et al ( 13 ), flexible sigmoidoscopy was shown to substantially reduce left-sided colorectal cancer in two well-known casecontrol studies ( 26,27 ). Additional prospective randomized controlled trials are ongoing in the United States and Europe ( ), with some results likely in the near future. Coupled with the apparent absence of benefit with complete optical colonoscopy in reducing mortality from right-sided cancer from the recent study by Baxter et al ( 14 ), a key policy question relates to whether the incremental costs and risks of optical colonoscopy over flexible sigmoidoscopy are justified ( 31 ). In patients requiring therapeutic endoscopy following positive results for left-sided polyps at CT colonography, our results indicate that flexible sigmoidoscopy may prove to be more cost effective than optical colonoscopy. Furthermore, the inherent advantages associated with virtual right-sided evaluation with CT colonography over physical endoscopy could potentially render a combined CT colonography flexible sigmoidoscopy screening strategy more effective than complete optical colonoscopy alone at perhaps a similar or reduced program cost ( 32 ). For surveillance after polypectomy, follow-up evaluation is typically determined by the gastroenterologist and depends on the size and histologic findings of resected lesions. Regardless, future screening should consist of total colonic examination, whether with CT colonography or optical colonoscopy. We acknowledge a number of potential limitations to our study. It is possible that a satisfaction of search phenomenon could be present with some endoscopists after detection of the left-sided polyps reported at prior CT colonography. However, this might be balanced out by the competing challenge or desire to find potential CT colonographic misses at optical colonoscopy. In our study, flexible sigmoidoscopy was necessarily simulated by employing the distal results from CT colonography or complete optical colonoscopy, as others have done ( 6 ). Given that localization of the splenic flexure at optical colonoscopy is somewhat imprecise, this likely resulted Radiology: Volume 259: Number 2 May 2011 n radiology.rsna.org 433

6 in some cases of misregistration. In addition, it is conceivable that some leftsided lesions in patients with very tortuous colons may not have been reached with flexible sigmoidoscopy. Finally, our algorithm may not have accounted for all covariates that could affect the results. In summary, the yield for proximally located polyps at optical colonoscopy in patients with only left-sided polyps detected at CT colonography was very low and may not justify the increased costs and risks of rescreening the proximal colon in this cohort. A more targeted examination such as flexible sigmoidoscopy would decrease patient risk, cost less, and save time. Disclosures of Potential Conflicts of Interest: P.J.P. Financial activities related to the present article: none to disclose. Financial activities not related to the present article: is a consultant for Medicsight, Viatronix, and Philips and is cofounder of VirtuoCTC. Other relationships: none to disclose. N.A.D. No potential conflicts of interest to disclose. B.D.P. No potential conflicts of interest to disclose. C.H. No potential conflicts of interest to disclose. References 1. Levin B, Lieberman DA, McFarland B, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi- Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin 2008 ; 58 ( 3 ): Johnson CD, Chen MH, Toledano AY, et al. Accuracy of CT colonography for detection of large adenomas and cancers. N Engl J Med 2008 ; 359 ( 12 ): Kim DH, Pickhardt PJ, Taylor AJ, et al. CT colonography versus colonoscopy for the detection of advanced neoplasia. N Engl J Med 2007 ; 357 ( 14 ): Pickhardt PJ, Choi JR, Hwang I, et al. Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. N Engl J Med 2003 ; 349 ( 23 ): Levin TR, Farraye FA, Schoen RE, et al. Quality in the technical performance of screening flexible sigmoidoscopy: recommendations of an international multi-society task group. Gut 2005 ; 54 ( 6 ): Imperiale TF, Wagner DR, Lin CY, Larkin GN, Rogge JD, Ransohoff DF. Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colorectal findings. N Engl J Med 2000 ; 343 ( 3 ): Pickhardt PJ. Screening CT colonography: how I do it. AJR Am J Roentgenol 2007 ; 189 ( 2 ): Pickhardt PJ, Taylor AJ, Kim DH, Reichelderfer M, Gopal DV, Pfau PR. Screening for colorectal neoplasia with CT colonography: initial experience from the 1st year of coverage by third-party payers. Radiology 2006 ; 241 ( 2 ): Zalis ME, Barish MA, Choi JR, et al. CT colonography reporting and data system: a consensus proposal. Radiology 2005 ; 236 ( 1 ): Pickhardt PJ, Lee AD, McFarland EG, Taylor AJ. Linear polyp measurement at CT colonography: in vitro and in vivo comparison of two-dimensional and three-dimensional displays. Radiology 2005 ; 236 ( 3 ): Tangka FK, Molinari NAM, Chattopadhyay SK, Seeff LC. Market for colorectal cancer screening by endoscopy in the United States. Am J Prev Med 2005 ; 29 ( 1 ): Lewis JD, Asch DA. Barriers to office-based screening sigmoidoscopy: does reimbursement cover costs? Ann Intern Med 1999 ; 130 ( 6 ): Atkin WS, Edwards R, Kralj-Hans I, et al. Once-only flexible sigmoidoscopy screening in prevention of colorectal cancer: a multicentre randomised controlled trial. Lancet 2010 ; 375 ( 9726 ): Baxter NN, Goldwasser MA, Paszat LF, Saskin R, Urbach DR, Rabeneck L. Association of colonoscopy and death from colorectal cancer. Ann Intern Med 2009 ; 150 ( 1 ): Graser A, Stieber P, Nagel D, et al. Comparison of CT colonography, colonoscopy, sigmoidoscopy and faecal occult blood tests for the detection of advanced adenoma in an average risk population. Gut 2009 ; 58 ( 2 ): Young PE, Gentry AB, Cash BD. The utility of flexible sigmoidoscopy after a computerized tomographic colonography revealing only rectosigmoid lesions. Aliment Pharmacol Ther 2008 ; 27 ( 6 ): Pickhardt PJ, Wise SM, Kim DH. Positive predictive value for polyps detected at screening CT colonography. Eur Radiol 2010 ; 20 ( 7 ): Gatto NM, Frucht H, Sundararajan V, Jacobson JS, Grann VR, Neugut AI. Risk of perforation after colonoscopy and sigmoidoscopy: a population-based study. J Natl Cancer Inst 2003 ; 95 ( 3 ): Iqbal CW, Cullinane DC, Schiller HJ, Sawyer MD, Zietlow SP, Farley DR. Surgical management and outcomes of 165 colonoscopic perforations from a single institution. Arch Surg 2008 ; 143 ( 7 ): ; discussion Levin TR, Conell C, Shapiro JA, Chazan SG, Nadel MR, Selby JV. Complications of screening flexible sigmoidoscopy. Gastroenterology 2002 ; 123 ( 6 ): Levin TR, Zhao W, Conell C, et al. Complications of colonoscopy in an integrated health care delivery system. Ann Intern Med 2006 ; 145 ( 12 ): Monahan DW, Peluso FE, Goldner F. Combustible colonic gas levels during flexible sigmoidoscopy and colonoscopy. Gastrointest Endosc 1992 ; 38 ( 1 ): Centers for Medicare and Medicaid Services. The guide to Medicare preventive services for physicians, providers, suppliers, and other health care professionals, January /downloads/psguid.pdf. Accessed January 28, Hospital outpatient prospective payment system Addendum B, October outpatientpps/downloads/2010october _AddB.zip. Accessed January 28, Centers for Medicare and Medicaid Services. Physician fee schedule relative value files, October FeeSched/PFSRVF. Accessed October 18, Selby JV, Friedman GD, Quesenberry CP Jr, Weiss NS. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992 ; 326 ( 10 ): Thiis-Evensen E, Hoff GS, Sauar J, Langmark F, Majak BM, Vatn MH. Population-based surveillance by colonoscopy: effect on the incidence of colorectal cancer. Telemark Polyp Study I. Scand J Gastroenterol 1999 ; 34 ( 4 ): Segnan N, Senore C, Andreoni B, et al. Baseline findings of the Italian multicenter randomized controlled trial of once-only sigmoidoscopy SCORE. J Natl Cancer Inst 2002 ; 94 ( 23 ): Gondal G, Grotmol T, Hofstad B, Bretthauer M, Eide TJ, Hoff G. The Norwegian Colorectal Cancer Prevention (NORCCAP) screening study: baseline findings and implementations for clinical work-up in age groups years. Scand J Gastroenterol 2003 ; 38 ( 6 ): Weissfeld JL, Schoen RE, Pinsky PF, et al. Flexible sigmoidoscopy in the PLCO cancer screening trial: results from the baseline screening examination of a randomized trial. J Natl Cancer Inst 2005 ; 97 ( 13 ): Levin TR, Zhao W, Conell C, et al. Complications of colonoscopy in an integrated health care delivery system. Ann Intern Med 2006 ; 145 ( 12 ): Pickhardt PJ, Kim DH, Hassan C. The effectiveness of colonoscopy in reducing mortality from colorectal cancer. Ann Intern Med 2009 ; 150 ( 11 ): ; author reply radiology.rsna.org n Radiology: Volume 259: Number 2 May 2011