Colorectal Cancer Screening Scientific Review

SCIENTIFIC REVIEW AND CLINICIAN’S CORNER CLINICAL APPLICATIONS

Judith M. E. Walsh, MD, MPH Context Screening for colorectal cancer clearly reduces colorectal cancer mortality, Jonathan P. Terdiman, MD yet many eligible adults remain unscreened. Several screening tests are available, and OLORECTAL CANCER IS THE various professional organizations have differing recommendations on which screen- second leading cause of can- ing test to use. Clinicians are challenged to ensure that eligible patients undergo co- cer death in the United States. lorectal cancer screening and to guide patients in choosing what tests to receive. In women, it ranks third af- Objective To critically assess the evidence for use of the available colorectal cancer terC lung and breast cancer, and in men, screening tests, including fecal occult blood tests, sigmoidoscopy, colonoscopy, double- it ranks third after lung and prostate contrast barium enema, and newer tests, such as virtual colonoscopy and stool-based cancer. Incidence and mortality from molecular screening. colorectal cancer are similar in both Data Sources All relevant English-language articles were identified using PubMed men and women. In 2001, an esti- (January 1966-August 2002), published meta-analyses, reference lists of key articles, mated 135400 cases were diagnosed and expert consultation. and an estimated 56700 deaths oc- Data Extraction Studies that evaluated colorectal cancer screening in healthy in- curred in the United States.1 dividuals and assessed clinical outcomes were included. Evidence from randomized con- Death from colorectal cancer is pre- trolled trials was considered to be of highest quality, followed by observational evidence. Diagnostic accuracy studies were evaluated when randomized controlled trials ventable. Effective, safe, and relatively and observational studies were not available or did not provide adequate evidence. inexpensive methods for screening for Studies were excluded if they did not evaluate colorectal screening tests and if they the disease have been available for de- did not evaluate average-risk individuals. cades, and screening is championed by Data Synthesis Randomized controlled trials have shown that fecal occult blood a large number of public, private, and testing can reduce colorectal cancer incidence and mortality. Case-control studies have professional organizations (Multidis- shown that sigmoidoscopy is associated with a reduction in mortality, and observa- 2 ciplinary Expert Panel, US Preven- tional studies suggest colonoscopy is effective as well. Combining fecal occult blood tive Services Task Force,3 the Ameri- testing and sigmoidoscopy may decrease mortality and can increase diagnostic yield. 4 can College of Gastroenterology, and Conclusion The recommendation that all men and women aged 50 years or older 5 the American Cancer Society ). This ar- undergo screening for colorectal cancer is supported by a large body of direct and in- ticle provides practicing physicians with direct evidence. At present, the available evidence does not currently support choos- an evidence-based review of the litera- ing one test over another. ture of the current status of colorectal JAMA. 2003;289:1288-1296 www.jama.com cancer screening, including the methods of screening that are likely to be barium enema or colonoscopy with the Author Affiliations: Division of General Internal Medi- widely available within the next sev- term screening. To supplement the cine, Departments of Medicine and Epidemiology and eral years. Biostatistics (Dr Walsh) and Division of Gastroenter- search, we reviewed published meta- ology, Department of Medicine (Dr Terdiman), Uni- analyses and personal files and bibliog- versity of California, San Francisco. METHODS Corresponding Author and Reprints: Judith M. E. raphies from published articles, con- In addition to reviewing published sys- Walsh, MD, MPH, Women’s Health Clinical Re- ducted hand searches, including the search Center, University of California San Francisco, tematic reviews, we conducted a litera- Campus Box 1793, 1635 Divisadero Suite 600, San proceedings of recent national profes- ture search using PubMed for English- Francisco, CA 94115 (e-mail: [email protected] sional organization meetings, and con- .edu). language articles from January 1966 Financial Disclosure: Dr Terdiman is on the scientific sulted with experts on colorectal can- through August 2002. For the search, we advisory board at Exact Laboratories, Maynard, Mass. cer screening. Eligibility criteria for Scientific Review and Clinical Applications Section combined search terms colorectal neo- Editor: Wendy Levinson, MD, Contributing Editor. articles that evaluated colorectal screen- plasm or occult blood or sigmoidoscopy or We encourage authors to submit papers to “Scien- ing tests, such as fecal occult blood tests tific Review and Clinical Applications.” Please con- (FOBTs), sigmoidoscopy, colonos- tact Wendy Levinson, MD, Contributing Editor, JAMA; See also p 1297 and Patient Page. phone: 312-464-5204; fax: 312-464-5824; e-mail: copy, and double-contrast barium en- [email protected].

1288 JAMA, March 12, 2003—Vol 289, No. 10 (Reprinted) ©2003 American Medical Association. All rights reserved. COLORECTAL CANCER SCREENING ema, were that they include asymptom- of cancers arise.6-9 The majority of these tion for colorectal cancer death rang- atic subjects (hence, screening) who adenomas are polypoid growths. But as ing from a low of 0.8 per 1000 person- were at average-risk and that informa- many as 20% to 30% of adenomas are flat years with biennial screening in the UK tion be provided on clinical outcomes. or depressed, which make them more study to a high of 4.6 per 1000 person- The clinical outcomes included mortal- difficult to detect and remove.10,11 years with annual screening in the Min- ity, cancer incidence, and identifica- The optimal means to prevent colo- nesota trial. Recent 18-year follow-up tion of adenomas. Studies to be of the rectal cancer remain uncertain. Evi- data from the Minnesota trial demon- highest quality were randomized con- dence for the efficacy of the com- strate that annual and biennial serial trolled trials (RCTs), which assessed monly practiced colorectal cancer FOBT screening reduces colorectal can- morbidity and/or mortality, followed by screening tests are reviewed. cer incidence by 17% to 20% as well.18 observational studies. Diagnostic accu- Fecal occult blood testing had 3 RCTs How the Test Is Performed. The racy studies were evaluated when RCTs and sigmoidoscopy had 1 randomized FOBT detects blood loss in the stool. and observational studies were not avail- trial and 3 case-control studies that as- It can detect blood loss caused by co- able or did not provide adequate evi- sessed mortality. No RCTs were found lorectal neoplasms, which tend to bleed dence. The highest quality evidence for colonoscopy, but 4 observational more than normal colonic mucosa. A available was included for each topic: for studies were identified in which asymp- variety of FOBTs are available,19,20 but FOBT, RCTs; for sigmoidoscopy, RCTs tomatic individuals underwent colonos- the Hemoccult II is most widely used and observational studies; for FOBT plus copy and in which adenoma and can- in the United States (Beckman- sigmoidoscopy, controlled trials, obser- cer incidences were assessed. Combined Coulter, Palo Alto, Calif). This test de- vational studies, and diagnostic accu- FOBT and sigmoidoscopy screening had tects the pseudoperoxidase activity racy studies; for double-contrast barium 5 trials that assessed clinical outcomes: found in hemoglobin when it inter- enema, diagnostic accuracy studies; and 1 trial assessed mortality, and the other acts with a guaiac-impregnated card in for colonoscopy, observational studies trials assessed incidence of polyps and/or the presence of a hydrogen peroxide de- and diagnostic accuracy studies. This re- cancer. Finally, for double-contrast veloper. A positive result is indicated view included additional literature to barium enema, no RCTs or observa- by the immediate appearance of a blue provide background material on the per- tional studies were identified that as- color on addition of the hydrogen performance, interpretation, and safety of sessed clinical effectiveness and 4 diag- oxide developer. The testing process re- colorectal cancer screening tests and to nostic accuracy studies were identified quires that the patient apply 2 distinct provide preliminary data on new, emerg- and evaluated. samples of 3 different stools to 6 test ing methods for colorectal cancer screen- card windows. Because the test de- ing, such as computed tomography (CT) Fecal Occult Blood Testing tects peroxidase or pseudoperoxidase colonography and stool-based molecu- Evidence of Clinical Efficacy. Results activity in stool, it is not specific for hu- lar testing. from 3 large RCTs of serial FOBTs con- man hemoglobin. Dietary substances ducted in Minnesota,12,13 United King- can result in false positive (eg, rare red EVIDENCE FOR COLORECTAL dom,14,15 and Denmark,16,17 involving meat, turnips, horseradish) or false CANCER SCREENING more that 250000 subjects followed for negative (eg, vitamin C) results. How- Death from colorectal cancer can be pre- up to 18 years, have consistently dem- ever, a recent systematic review found vented by the detection of early-stage dis- onstrated that serial FOBT reduces co- that a restricted diet does not reduce the ease that has not metastasized. The dis- lorectal cancer mortality (TABLE 1). positivity rate but it may reduce pa- ease itself can be prevented by the Screening with FOBT reduced colorec- tient compliance.21 detection and removal of colorectal ad- tal cancer mortality from 15% to A rehydration procedure that en- enomas, from which greater than 95% 33%,12-14 with the absolute risk reduc- hances the sensitivity of the Hemoc-

Table 1. Summary of Clinical Trials for Fecal Occult Blood Testing Mandel et al12,13 Hardcastle et al14 Kronborg et al16,17 Study Characteristics (United States) (United Kingdom) (Denmark) No. of study participants 46 551 150 251 61 933 Follow-up, y 18 7.8 13 Relative risk mortality with annual FOBT (95% CI) .67 (0.51-0.83) Not studied Not studied Relative risk mortality with biennial FOBT (95% CI) .79 (0.62-0.97) .85 (0.74-0.98) .82 (0.69-0.97) Absolute risk reduction for CRC death per 1000 subjects 4.6 (annual) 0.8 1.8 2.9 (biennial) No. of subjects needed to screen to prevent CRC death 217 (annual) 1250 555 344 (biennial) Abbreviations: CI, confidence interval; CRC, colorectal cancer; FOBT, fetal occult blood testing; RR, relative risk.

©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, March 12, 2003—Vol 289, No. 10 1289 COLORECTAL CANCER SCREENING cult test, at the expense of specificity, moccult Sensa (Beckman Coulter, Palo is funding the Prostate, Lung, Colorec- can be performed by adding a few drops Alto, Calif) as the test of choice be- tal, and Ovarian screening trial, which of water to the stool samples before add- cause of its greater sensitivity than the is evaluating 60-cm flexible sigmoidos- ing the hydrogen peroxide to the test Hemoccult II and its greater specific- copy,32 and the UK FlexiScope Trial is windows.12 This procedure reduces the ity than the rehydrated Hemoccult.20 being conducted to study one-time positive predictive values of the test by Immunochemical FOBTs have the ad- screening sigmoidoscopy in subjects be- more than 50%. In the Minnesota Co- vantage of not requiring dietary or drug tween ages 55 years and 64 years.33-35 lon Cancer Control Society trial, of the restrictions, but these tests are more ex- These 2 large studies ultimately will en- guaiac-impregnated cards, 83% were pensive, and published data on effec- roll more than 250000 subjects, but no screened using this procedure, and as tiveness are limited.20 outcome data are yet available. a result, colonoscopy was performed on Based on evidence from large RCTs, The Telemark Polyp Study, a small 38% of the participants in the annu- FOBT should be repeated at least ev- randomized trial, demonstrated that ally screened group and 28% of the par- ery other year to be clinically benefi- one-time flexible sigmoidoscopy screen- ticipants in the biennially screened cial.12-14,16,17 An annual FOBT may of- ing could reduce colorectal cancer in- group during the first 13 years of the fer greater reductions in mortality than cidence. However, no reduction in co- study.12 By comparison, only 4% to 5% a biennial screening, but at an in- lorectal cancer mortality was observed of subjects in the UK and Denmark creased cost.3,13,25 in the screened group.36,37 studies underwent colonoscopy.14,16,17 Fecal occult blood testing itself is The major evidence supporting the No direct comparison has been made safe, but false positive results can lead effectiveness of sigmoidoscopy comes with respect to the efficacy of the rehy- to unnecessary further invasive tests, from well-designed, retrospective case- drated and nonrehydrated procedures such as colonoscopy, that have a mea- control studies.38-40 The landmark study for FOBT,12,14,16,17 and any incremental surable complication rate.26,27 by Selby et al38 found that rigid sigmoi- benefit of rehydration with respect to How to Interpret the Test Results. doscopy screening was associated with mortality reduction may be small in light The greater the number of test win- a 59% reduction in colorectal cancer of the increased costs due to a much dows that are positive, the higher the mortality (odds ratio [OR], 0.41; 95% higher false positive rate. In the Minne- positive predictive value of the test.18 confidence interval [CI], 0.25-0.69). sota Colon Cancer Control Society trial, However, even if only 1 of the 6 test win- The reduction in mortality from can- use of the rehydration procedure re- dows is positive, the overall test should cers within the reach of the sigmoido- duced the 13-year cumulative mortal- be considered positive, and these indi- scope was 70% (OR, 0.30; 95% CI, ity rate by 33% for colorectal cancer.12 viduals should be referred for complete 0.19-0.48), while there was no signifi- It has been argued that some of the mor- colonoscopy.28 This was the approach cant reduction in mortality seen from tality reduction in this trial was simply used by the Minnesota and Denmark cancers proximal to the reach of the sig- a consequence of the high rate of colo- FOBT studies.13,17 moidoscope (OR, 0.80; 95% CI, noscopy performed (ie, a similar ben- Performance Characteristics. Fe- 0.54-1.19).38 These results have been efit would have been demonstrated in cal occult blood testing performed on confirmed by 2 other well-designed the population if the number of colo- a single occasion for the detection of co- case-control studies by Newcomb et al39 noscopies performed in the study popu- lorectal cancer and adenomas shows and Muller and Sonnenberg,40,41 which lation had been randomly allocated).22 poor sensitivity. However, the key to the have expanded the findings to include The degree of the reduction in the rate success of FOBT lies in serial test- flexible sigmoidoscopy (TABLE 2). In of mortality has been clarified by addi- ing.29-31 In the UK and Denmark stud- addition, 2 case-control studies have tional data from the Minnesota trial and ies, FOBT screening detected 27% of the demonstrated a reduction in inci- additional mathematical modeling.23 patients in the intervention group who dence of colorectal cancer using flex- At the present time, professional or- developed cancer.14,16 In the Minne- ible sigmoidoscopy.40,42 ganizations, such as the World Health sota trial, after 13 years, 39% of pa- How the Test Is Performed. Flex- Organization,20 US Preventive Ser- tients in the biennial group and 49% of ible sigmoidoscopy generally is per- vices Task Force,3 and American Col- patients in the annual group who de- formed using a 60-cm flexible endo- lege of Physicians,24 do not recom- veloped colorectal cancer were identi- scope.42 Preparation for the procedure mend the rehydration FOBT procedure fied through FOBT screening.12 requires that at least 1 to 2 saline en- because of the uncertainties regarding emas are administered to the patient the effectiveness and cost, compared with Flexible Sigmoidoscopy morning of the examination. the nonrehydrated FOBT procedure. Evidence of Clinical Effectiveness.No The examination can be performed in Guaiac-based tests other than the He- completed large RCTs have demon- a physician’s office or in the hospital. Spe- moccult II also are available, but they strated the effectiveness of sigmoidos- cial training is required to perform the are not often used. Recently, the World copy in the prevention of colorectal can- procedure, but a variety of practitioners Health Organization endorsed the He- cer death. The National Cancer Institute (physician’s assistants, nurses, primary

1290 JAMA, March 12, 2003—Vol 289, No. 10 (Reprinted) ©2003 American Medical Association. All rights reserved. COLORECTAL CANCER SCREENING care physicians, gastroenterologists) rou- complete colonoscopy if any adenoma Prostate, Lung, Colorectal, and Ovar- tinely perform the procedure.44-48 The is found at sigmoidoscopy examina- ian sigmoidoscopy screening trial indi- procedure takes about 10 minutes to per- tion, while others feel that no addi- cated that the risk of advanced ad- form. The patient often experiences some tional procedure is required if only 1 enoma in the distal region of the colon tolerable abdominal pain. Sedation is not or 2 small adenomas are detected and 3 years after an initial negative sigmoi- administered, and the patient may drive removed at sigmoidoscopy examina- doscopy result was 0.8%, although the alone to the physician’s office or hospi- tion.9,25 These recommendations are incidence of cancer in this region of the tal and return to work immediately fol- based on a large number of studies that colon was less than 0.1%.58 Longer term lowing the procedure. correlate the findings in the distal re- follow-up of this study will further clarify The flexible sigmoidoscopy proce- gion of the colon and rectum, with the the importance of this finding. dure is safe when performed by ex- likelihood of finding a cancer or an ad- Performance Characteristics. Flex- perts. In a retrospective review of 49501 vanced polyp in the proximal region of ible sigmoidoscopy only examines a por- flexible sigmoidoscopy procedures per- the colon at complete colonoscopy ex- tion of the colon, and therefore impor- formed during a 10-year period, only amination (TABLE 3).31,51-56 tant colonic lesions will be missed even 2 perforations occurred (0.004%).26 After a negative sigmoidoscopy re- if the finding of any adenoma on sig- Similar low complication rates have sult in which no adenoma is found, the moidoscopy examination indicates a been reported from large population- standard recommendation is to repeat the complete colonoscopy. Only 20% to 30% based flexible sigmoidoscopy screen- screening examination in 5 years. This of colorectal cancers in the proximal re- ing programs in the United States and recommendation is based on case- gion are associated with an adenoma in the United Kingdom.34,35,49,50 control studies that have demonstrated the distal region that might be detected How to Interpret the Test Results. An that the protective effects of sigmoidos- at flexible sigmoidoscopy examina- important question is which lesions iden- copy appear to last at least 6 years38 and tion.59 Furthermore, recent observa- tified by sigmoidoscopy should prompt a prospective follow-up sigmoidoscopy tional studies on colonoscopy screen- evaluation of the entire colon. Guide- study in which the likelihood of finding suggest that one half of all advanced lines published by the 3 major US orga- ing an advanced adenoma or cancer was adenomas and cancers in the proximal nizations specializing in the gastrointes- 0 in the 3 to 4 years following a nega- region would be missed on sigmoidos- tinal tract state that the finding of an tive sigmoidoscopy result.57 However, copy examination.56,60 However, sigmoi- adenomatous polyp 1 cm or larger in di- preliminary data from the large-scale doscopy screening followed by com- ameter, or one with advanced histologic findings (eg, villous changes or high-grade dysplasia), or multiple pol- Table 2. Case-Control Studies of Mortality Reduction Associated With Sigmoidoscopy yps, at sigmoidoscopy examination re- Screening quires follow-up with complete colonos- Newcomb Muller and Study Characteristics Selby et al38 et al39 Sonnenberg40,41 copy, even if the lesions were removed No. of cases of colorectal cancer 261 66 4411 at the initial examination.9 Type of sigmoidoscope Rigid Rigid and flexible Rigid and flexible However, controversy remains re- Odds ratio (95% CI) for colorectal 0.41 (0.25-0.69) 0.21 (0.08-0.52) 0.41 (0.33-0.5) garding the appropriate follow-up for cancer death the finding of 1 or 2 small tubular ad- Interval of apparent protective effect, y 9-10 Not specified 6 enomas. Some experts recommend Abbreviation: CI, confidence interval.

Table 3. Rate of Advanced Proximal Neoplasm* According to Colorectal Findings in the Distal Colon Findings in Distal Colon, % (No./Total)

Hyperplastic Tubular Multiple Tubular Advanced Source Normal Polyp Adenoma Ͻ1cm Adenomas Ͻ1 cm† Neoplasm Lieberman et al31 2.7 (48/1765) 2.8 (13/464) 6.4 (35/543) 9.1 (4/44) 11.7 (32/274) Zarchy and Ershoff 51 Not reported Not reported 0.8 (1/124) Not reported 11.8 (12/102) Read et al52 Not reported Not reported 6.9 (13/189)‡ Not reported 28.6 (4/14) Schoen et al53 Not reported Not reported 2.9 (15/521) 2.4 (2/85) 5.9 (27/460) Wallace et al54 Not reported Not reported 1.6 (3/190) 10.4 (5/48) 7.4 (5/63) Levin et al55 5.3 (29/544) Not reported 5.0 (22/444) 6.3 (20/319) 8.8 (147/1665) Imperiale et al56 1.5 (23/1564) 4.0 (8/201) 7.1 (12/168) Not reported 11.5 (7/61)§ *Defined as invasive cancer or adenoma 1 cm or larger in diameter or with villous features or high-grade dysplasia. †Defined as 3 or more adenomas. ‡Includes adenomas with villous features. §Does not include adenomas 1 cm or larger.

be interpreted with some caution. In the 0.3% of subjects, including gastrointes- Figure. Cumulative Incidence of Colorectal 62 Cancer in the National Polyp Study Cohort US National Polyp Study, only 5 co- tinal tract bleeding, myocardial infarc- (NPS) lorectal cancers were detected in the tion, and stroke, and 3 subjects died study population for an incidence of 0.6 within 1 month of the screening exami- Expected Incidence cancers per 1000 years of subject follow- nation.60,65 Mayo Clinic St Mark’s up, and all the patients were asymptom- How to Interpret the Test Results. 4 SEER atic and the cancers were in early stages Colonoscopy is considered the crite- Observed Incidence NPS (4 in stage I and 1 in stage II). This low rion standard for detecting colorectal 3 incidence of metachronous cancer rep- cancers and adenomas. If an adenoma is resented a 76% to 90% reduction in can- detected, a repeat surveillance examina- 2 cer incidence compared with 3 refer- tion is generally recommended in 3 or ence populations, the Mayo Clinic 5 years, depending on the number, size, 1 cohort, the St Mark’s cohort, and the US and histologic findings of the adeno- Surveillance, Epidemiology, and End Re- mas removed.9 The recommendations for 62 0 1 2 3 4 5 6 7 sults Program cohort (FIGURE). Third, the appropriate surveillance interval af- Cumulative Incidence of Colorectal Cancer, % Cancer, Cumulative Incidence of Colorectal Years of Follow-up recent cross-sectional colonoscopy ter a positive finding at colonoscopy ex- screening studies indicate that colonos- amination are based on data from the US The observed incidence is compared with the expected copy is more sensitive than flexible sig- National Polyp Study.66,67 If no adeno- incidence based on data from 3 reference groups: the Mayo Clinic cohort, the St Mark’s cohort, and the US moidoscopy or sigmoidoscopy plus mas are detected, the test result is nega- Surveillance, Epidemiology, and End Results (SEER) Pro- FOBT for the detection of large adeno- tive. The recommendation for the 10- gram of the US National Cancer Institute. Reprinted 56,60 with permission from Winawer et al.62 mas and cancers. It has not been year repeat screening interval is based on shown but it can be assumed that in- indirect evidence. Case-control studies creased sensitivity would translate into of sigmoidoscopy screening suggest that plete colonoscopy for the finding of any increased effectiveness. the protective effect of endoscopy screen- adenoma most likely will have de- How the Test Is Performed. Colo- ing lasts about 10 years.38 In addition, in tected 70% or 80%, respectively, of all noscopy generally is performed by a gas- the US National Polyp Study, low rates the advanced neoplasms.60 troenterologist or a surgeon using a of metachronous adenomas or cancer 160-cm flexible endoscope. Extensive were seen after colonoscopic polypec- Colonoscopy Screening training is required to perform the pro- tomy during extended follow-up.62,67 Fi- Evidence of Clinical Efficacy. The ef- cedure safely and effectively.64 The day nally, 2 small prospective studies have ficacy of colonoscopy screening for the prior to the procedure the patient must found that the incidence of cancer is less prevention of colorectal cancer and co- only have consumed clear liquids and than 1% within 5 years after a negative lorectal cancer death has not been stud- then consume some form of purgative colonoscopy screening result.68,69 ied to date in RCTs. The National Can- (low-volume sodium phosphate purge Performance Characteristics. Colo- cer Institute is now sponsoring a pilot/ or high-volume polyethylene glycol noscopy screening can detect ad- feasibility study of colonoscopy purge). Sedation is administered and pa- vanced polyps and cancers that would screening,61 but results from a large tients cannot drive so they must be ac- otherwise be missed by sigmoidos- RCT, if undertaken, are not expected companied by another individual to es- copy and/or FOBT.31,56,60 Although con- to be available for many years. How- cort them home. The patient should not sidered highly sensitive and specific for ever, indirect evidence suggests that experience pain during the examina- the detection of colonic neoplasia, colo- colonoscopy is almost certainly a highly tion and often the patient cannot recall noscopy is not a perfect test, and le- effective screening test. the procedure. In a recent large study of sions can be missed.70-72 In one study First, the reduction in mortality dem- colonoscopy screening, a complete ex- in which tandem colonoscopies were onstrated in the FOBT screening stud- amination was possible in 98% of pa- performed by 2 expert examiners, the ies is attributable to the performance of tients, with a mean procedure time of 30 miss rates were 6% for adenomas 1 cm follow-up colonoscopy.12,14,16 Second, a minutes.60,65 Recovery time for the pa- and larger in diameter, 13% for adeno- reduction in colorectal cancer inci- tient is approximately 2 to 3 hours. mas 6 to 9 mm in diameter, and 27% dence has been demonstrated in the US Flexible sigmoidoscopy has been for adenomas 5 mm and smaller in di- National Polyp Study and the Italian shown to be a safer procedure than colo- ameter.71 Multicenter Study,62,63 2 large cohort noscopy, but colonoscopy is consid- studies of individuals who had adeno- ered safe.26,27 For instance, in a large study Combined FOBT and matous polyps removed at colonos- on colonoscopy screening, no deaths Sigmoidoscopy Screening copy. These studies were not about colo- were directly attributable to colonos- Evidence of Clinical Effectiveness. The noscopy screening, and they used copy and no colonic perforations oc- limitations of using FOBT and sigmoi- historical controls, so the results must curred.60,65 Major morbidity did occur in doscopy separately may be overcome

1292 JAMA, March 12, 2003—Vol 289, No. 10 (Reprinted) ©2003 American Medical Association. All rights reserved. COLORECTAL CANCER SCREENING by performing the 2 tests in concert. population from a health mainte- mendation is to perform screening by These are widely practiced proce- nance organization, the addition of an double-contrast barium enema every 5 dures and little evidence in the pub- immunochemical FOBT did detect ad- to 10 years. lished literature supports combina- vanced adenomas and cancers in the Performance Characteristics. The tion testing. In the study most proximal region of the colon that oth- double-contrast barium enema exami- commonly cited to support combina- erwise would have been missed by nation is not as sensitive as the endos- tion testing, the investigators nonran- screening with flexible sigmoidoscopy examination in the detection of pol- domly allocated 12479 individuals ei- copy alone. However, more than 600 yps. It does appear that double-contrast ther to annual screening with FOBT individuals would need to be screened barium enema will detect a majority of combined with rigid sigmoidoscopy or by the immunochemical FOBT to de- advanced adenomas and cancers.81-84 In to rigid sigmoidoscopy alone.73 Pa- tect 1 additional advanced adenoma or the US National Polyp Study, 862 paired tient adherence to the protocol in both cancer that otherwise would have been double-contrast barium enema and colo- groups was poor. Colorectal cancer missed.77 Finally, in the Veterans Af- noscopy examinations were compared in mortality was lower in the combined fairs Cooperative Study of colonos- a surveillance population.85 In a large per- testing group after 5 to 11 years of fol- copy screening, the addition of FOBT centage of cases, the results of the double- low-up (0.36 deaths per 1000 per year to sigmoidoscopy would have in- contrast barium enema examination vs 0.63 deaths per 1000 per year), show- creased the percentage of patients iden- were false negatives, especially when the ing only a borderline statistical signifi- tified with advanced neoplasia from largest polyp found at colonoscopy was cance (P=.053).73 Given that the re- 70.3% to 75.8%.31 small. Even for polyps larger than 1 cm sults are marginal, the use of yearly rigid in diameter, the sensitivity of double- sigmoidoscopy in the protocol and the Double-Contrast Barium Enema contrast barium enema was approxi- poor compliance rates, generalizabil- Evidence of Effectiveness. Double- mately 50%.85 However, many of the ity of this study is tenuous. contrast (air-contrast) barium enema small polyps missed by double-contrast How the Test Is Performed. The stan- has been advocated as a screening barium enema examination may be not dard recommendation for patients who method for colorectal cancer, but to be clinically important, and therefore, a are undergoing colorectal cancer screen- date no published evidence from con- decreased sensitivity for detecting ad- ing is to have an FOBT performed ev- trolled studies is available examining enomas, especially those of small or me- ery year and to have sigmoidoscopy per- the effectiveness of this method. dium size, does not necessarily mean that formed every 5 years. In a year in which How the Test Is Performed. Pa- double-contrast barium enema is not an both tests are to be performed, the FOBT tient preparation for double-contrast effective screening test. should be completed first because a posi- barium enema is similar to that for colo- tive FOBT result would then require a noscopy. Sedation is not required, al- FUTURE DIRECTIONS OF complete colonoscopy to be performed though patients often complain of pain COLORECTAL CANCER and therefore eliminate the need for sig- and embarrassment.78 A trained radi- SCREENING moidoscopy. ologist must be present to perform the Two new promising screening technolo- How to Interpret the Test Results. procedure. Barium, followed by air, is gies are CT-assisted colonography, also If either the FOBT or the sigmoidos- instilled into the colon under gentle termed virtual colonoscopy, and stool- copy procedure has an abnormal re- pressure. The patient is then moved to based molecular testing. Before any new sult, then complete colonoscopy is in- different positions on an examination screening methods are routinely dicated. table while radiographs are obtained. adopted, they should be assessed in clini- Performance Characteristics. Two The procedure takes 30 to 60 minutes cal studies among average-risk patients large RCTs have demonstrated that to complete. This procedure has been that compare sensitivity and specificity combination testing will detect 4 to 5 shown to be safe; perforation of the co- for the detection of advanced polyps and times more large polyps and cancers lon is extremely rare, and serious com- cancers, cost, safety, and acceptability to than FOBT alone.74,75 However, in an- plications of any type occurring have patients with currently recommended other large randomized study, more been reported in approximately 1 in screening tests. Ideally, clinical trials that polyps and cancers were not detected 10000 examinations.79,80 assess the impact of these new screen- among patients undergoing FOBT and How to Interpret the Test Results. ing tests on colorectal cancer incidence sigmoidoscopy compared with pa- Polypoid lesions and masses detected and mortality should be undertaken, but tients who underwent sigmoidoscopy at double-contrast barium enema ex- this may be impractical. alone.76 These studies involved a single amination would indicate follow-up application of FOBT and not serial test- complete colonoscopy to verify the Computed Tomography ing, so the applicability of these find- presence of the lesions, to obtain a bi- Colonography Screening ings to clinical practice is unclear. In a opsy sample, and to remove the le- Virtual colonoscopy is a technique that study recently completed in a large sions if possible. The standard recom- uses data generated from CT or mag-

©2003 American Medical Association. All rights reserved. (Reprinted) JAMA, March 12, 2003—Vol 289, No. 10 1293 COLORECTAL CANCER SCREENING netic resonance imaging to generate rectal cancer screening if it costs ap- cancer. Several tests are available for co- 2-dimensional and 3-dimensional im- proximately 50% less than conven- lon cancer screening, including FOBT, ages of the colon.86-95 Usually, the co- tional colonoscopy, or was associated flexible sigmoidoscopy, double- lon is scanned while the patient is in with an initial compliance rate 15% to contrast barium enema, and colonos- the prone and supine position, and the 20% better than colonoscopy.99 copy. Direct and indirect evidence in- time it takes to acquire the images is less dicates that all the tests are effective, but than 5 minutes. Patients must un- Stool-Based Molecular Screening they differ in their sensitivity, specific- dergo a colonic preparation, as with Genomic alterations drive the ad- ity, cost, and safety. The available evi- double-contrast barium enema or colo- enoma to carcinoma DNA sequence, dence does not currently support noscopy. Research is ongoing that may and alterations in the neoplasm- choosing one test over another. In ad- eventually eliminate the need for co- specific DNA in colorectal adenomas dition, other colorectal cancer tests, lonic preparation.96 Following colonic and carcinomas have been well- such as virtual colonoscopy or stool- preparation, air or another gas, such as characterized.100 Colorectal epithelial based molecular testing, have the po- carbon dioxide, must be insufflated DNA can be extracted from stool tential to become important screening through a rectal tube to distend the co- samples and amplified, allowing for the tests in the future. lon to enhance imaging. The patient detection of mutations indicative of co- Funding/Support: Dr Walsh was supported by an 101 may experience some slight discom- lorectal neoplasia. Such stool-based American Cancer Society Cancer Control Career De- fort from the air insufflation, and no se- testing is appealing because it is non- velopment Award for Primary Care Physicians. Acknowledgment: We would like to thank Hai Emily dation is required. After the images are invasive, requires no special colonic Huang for her assistance with manuscript prepara- reformatted, they are reviewed by a ra- preparation, and has the capability of tion. diologist, a process that takes 20 to 40 detecting neoplasia throughout the en- minutes. Patients who have a sus- tire length of the colon. A recent study REFERENCES pected polyp or mass lesion are re- reported that mutations in the adeno- 1. Greenlee RT, Hill-Harmon MB, Murray T, Thun M. ferred for follow-up colonoscopy. matous polyposis coli (APC) gene could Cancer statistics, 2001. CA Cancer J Clin. 2001;51: 15-36. In the largest study published to date be detected in fecal DNA using a novel 2. Winawer SJ, Fletcher RH, Miller L, et al. Colorec- comparing virtual colonoscopy to con- method, digital protein truncation, tal cancer screening: clinical guidelines and rationale. Gastroenterology. 1997;112:594-642. ventional colonoscopy, polyp detec- among 26 (57%) of 46 patients with 3. United States Preventive Services Task Force. Clini- tion using virtual colonoscopy was posi- known colonic cancers or large pol- cal guidelines: screening for colorectal cancer: recommendation and rationale. Ann Intern Med. 2002;137: tive in 90% of cases in which yps (95% CI, 41%-71%), but in none 129-131. conventional colonoscopy found 1 polyp of 28 control patients without colonic 4. Rex DK, Johnson DA, Lieberman DA, Burt DW, Son- 102 nenberg A. Colorectal cancer prevention 2000: screen- and was negative in 72% of cases in neoplasms (95% CI, 0%-12%). How- ing recommendations of the American College of Gas- which conventional colonoscopy found ever, because the DNA alterations in co- troenterology. Am J Gastroenterol. 2000;95:868-877. no polyp.91 Good sensitivity for the de- lorectal cancer are heterogeneous, fu- 5. Smith RA, von Eschernbach AC, Wender R, et al. American Cancer Society guidelines for the early detection of large polyps and cancers us- ture assays will need to detect mutations tection of cancer: update of early detection guide- ing virtual colonoscopy has been re- in a number of genes typically mu- lines for prostate, colorectal, and endometrial can- 88-90 cers; also update 2000—testing for early lung cancer ported by several other investigators, tated in colorectal cancer in addition to detection. CA Cancer J Clin. 2001;51:38-75, quiz, 77- but not by all investigators.92-95 APC, such as Kirsten-ras (K-ras) and 80. 103-105 6. Morson B. President’s address: the polyp-cancer Virtual colonoscopy is a promising p53. A small pilot study of a pro- sequence in the large bowel. Proc R Soc Med. 1974; technology for colorectal cancer screen- prietary assay that looks for mutations 67:451-457. ing, but practical barriers and addi- in a number of genes was more than 7. Morson BC. The evolution of colorectal carcinoma. Clin Radiol. 1984;35:425-431. tional clinical studies are required be- 80% sensitive for the detection of large 8. Bond JH. Clinical evidence for the adenoma- fore it can be embraced as a screening adenomas or cancer and was 93% spe- carcinoma sequence, and the management of pa- 97 103 tients with colorectal adenomas. Semin Gastrointest option. The cost of virtual colonos- cific. Large-scale prospective stud- Dis. 2000;11:176-184. copy is high, and radiologists will need ies comparing the sensitivity and speci- 9. Bond JH. Polyp guideline: diagnosis, treatment, and surveillance for patients with colorectal polyps: Prac- to undergo extensive training to inter- ficity of this assay with FOBT and tice Parameters Committee of the American College pret the test results. Furthermore, it is colonoscopy in a colorectal cancer of Gastroenterology. Am J Gastroenterol. 2000;95: 3053-3063. not clear whether virtual colonoscopy screening population are under way 10. Rembacken BJ, Fujii T, Cairns A, et al. Flat and will be better tolerated and more ac- (David Ahlquist, MD, and Michael Ross, depressed colonic neoplasms: a prospective study of ceptable to the general population than MD, oral communication, September 1000 colonoscopies in the UK. Lancet. 2000;355: 98 1211-1214. conventional colonoscopy. A recent 2002). 11. Saitoh Y, Waxman I, West AB, et al. Prevalence cost-effectiveness analysis found that and distinctive biologic features of flat colorectal ad- COMMENT enomas in a North American population. Gastroen- virtual colonoscopy, even if 100% sen- terology. 2001;120:1657-1665. sitive and specific for colorectal neo- Colorectal cancer mortality can be re- 12. Mandel JS, Bond JH, Church TR, et al. Reducing mortality from colorectal cancer by screening for fe- plasia, would only be as cost-effective duced by screening all men and women cal occult blood: Minnesota Colon Cancer Control as conventional colonoscopy for colo- aged 50 years or older for colorectal Study. N Engl J Med. 1993;328:1365-1371.

13. Mandel JS, Church TR, Ederer F, Bond JH. Colo- (PLCO) cancer screening trial. Control Clin Trials. 2000; 52. Read TE, Read JD, Butterly LF. Importance of ad- rectal cancer mortality: effectiveness of biennial screen- 21(6 suppl):273S-309S. enomas 5 mm or less in diameter that are detected ing for fecal occult blood. J Natl Cancer Inst. 1999; 33. Atkin WS, Cuzick J, Northover JM, Whynes DK. by sigmoidoscopy. N Engl J Med. 1997;336:8-12. 91:434-437. Prevention of colorectal cancer by once-only sigmoi- 53. Schoen RE, Corle D, Cranston L, et al. Is colonos- 14. Hardcastle JD, Chamberlain JO, Robinson MH, doscopy. Lancet. 1993;341:736-740. copy needed for the nonadvanced adenoma found on et al. Randomised controlled trial of faecal-occult- 34. Atkin WS, Hart A, Edwards R, et al. Uptake, yield sigmoidoscopy? the Polyp Prevention Trial. Gastro- blood screening for colorectal cancer. Lancet. 1996; of neoplasia, and adverse effects of flexible sigmoi- enterology. 1998;115:533-541. 348:1472-1477. doscopy screening. Gut. 1998;42:560-565. 54. Wallace MB, Kemp JA, Trnka YM, Donovan JM, 15. Scholefield JH, Moss S, Sufi F, Mangham CM, 35. UK Flexible Sigmoidoscopy Screening Trials In- Farraye FA. Is colonoscopy indicated for small adeno- Hardcastle JD. Effect of faecal occult blood screening vestigators. Single flexible sigmoidoscopy screening to mas found by screening flexible sigmoidoscopy? Ann on mortality from colorectal cancer: results from a ran- prevent colorectal cancer: baseline findings of a UK Intern Med. 1998;129:273-278. domised controlled trial. Gut. 2002;50:840-844. multicentre randomised trial. Lancet. 2002;359:1291- 55. Levin TR, Palitz A, Grossman S, et al. Predicting 16. Kronborg O, Fenger C, Olsen J, Jorgensen OD, 1300. advanced proximal colonic neoplasia with screening Sondergaard O. Randomised study of screening for 36. Thiis-Evensen E, Hoff GS, Sauar J, Langmark F, sigmoidoscopy. JAMA. 1999;281:1611-1617. colorectal cancer with faecal-occult-blood test. Lan- Majak BM, Vatn MH. Population-based surveillance 56. Imperiale TF, Wagner DR, Lin CY, Larkin GN, Rogge cet. 1996;348:1467-1471. by colonoscopy: effect on the incidence of colorectal JD, Ransohoff DF. Risk of advanced proximal neo- 17. Jorgensen OD, Kronborg O, Fenger C. A ran- cancer: Telemark Polyp Study I. Scand J Gastroen- plasms in asymptomatic adults according to the distal domised study of screening for colorectal cancer us- terol. 1999;34:414-420. colorectal findings. N Engl J Med. 2000;343:169-174. ing faecal occult blood testing: results after 13 years 37. Hoff G, Thiis-Evensen E, Grotmol T, Sauar J, Vatn 57. Rex DK, Lehman GA, Ulbright TM, Smith JJ, Hawes and seven biennial screening rounds. Gut. 2002;50: MH, Moen IE. Do undesirable effects of screening RH. The yield of a second screening flexible sigmoi- 29-32. affect all-cause mortality in flexible sigmoidoscopy pro- doscopy in average-risk persons after one negative ex- 18. Mandel JS, Church TR, Bond JH, et al. The effect grammes? experience from the Telemark Polyp Study amination. Gastroenterology. 1994;106:593-595. of fecal occult-blood screening on the incidence of co- 1983-1996. Eur J Cancer Prev. 2001;10:131-137. 58. Schoen RE, Wissfeld J, Bresalier R, et al. The yield lorectal cancer. N Engl J Med. 2000;343:1603-1607. 38. Selby JV, Friedman GD, Quesenberry CP Jr, Weiss of repeat flexible sigmoidoscopy three years after a nega- 19. Allison JE, Tekawa IS, Ransom LJ, Adrain AL. A NS. A case-control study of screening sigmoidoscopy tive exam: results from the PLCO cancer screening trial comparison of fecal occult-blood tests for colorectal- and mortality from colorectal cancer. N Engl J Med. [abstract]. Gastroenterology. 2002;122:A355. cancer screening. N Engl J Med. 1996;334:155-159. 1992;326:653-657. 59. Lemmel GT, Haseman JH, Rex DK, Rahmani E. 20. Young GP, St John DJ, Winawer SJ, Rozen P. 39. Newcomb PA, Norfleet RG, Storer BE, Surawicz Neoplasia distal to the splenic flexure in patients with Choice of fecal occult blood tests for colorectal can- TS, Marcus PM. Screening sigmoidoscopy and colo- proximal colon cancer. Gastrointest Endosc. 1996;44: cer screening: recommendations based on perfor- rectal cancer mortality. J Natl Cancer Inst. 1992;84: 109-111. mance characteristics in population studies: a WHO 1572-1575. 60. Lieberman DA, Weiss DG, Bond JH, Ahnen DJ, (World Health Organization) and OMED (World Or- 40. Muller AD, Sonnenberg A. Prevention of colo- Garewal H, Chejfec G. Use of colonoscopy to screen ganization for Digestive Endoscopy) report. AmJGas- rectal cancer by flexible endoscopy and polypec- asymptomatic adults for colorectal cancer: Veterans troenterol. 2002;97:2499-2507. tomy: a case-control study of 32,702 veterans. Ann Affairs Cooperative Study 380. N Engl J Med. 2000; 21. Pignone M, Campbell MK, Carr C, Phillips C. Meta- Intern Med. 1995;123:904-910. 343:162-168. analysis of dietary restriction during fecal occult blood 41. Muller AD, Sonnenberg A. Protection by endos- 61. Winawer SJ, Zauber AG, Church T, et al. Na- testing. Eff Clin Pract. 2001;4:150-156. copy against death from colorectal cancer: a case- tional colonoscopy study preliminary results: a con- 22. Lang CA, Ransohoff DF. Fecal occult blood screen- control study among veterans. Arch Intern Med. 1995; trolled trial of general population screening colonos- ing for colorectal cancer: is mortality reduced by chance 155:1741-1748. copy. Gastroenterology. 2002;122:T1560. selection for screening colonoscopy? JAMA. 1994; 42. Kavanagh AM, Giovannucci EL, Fuchs CS, Cold- 62. Winawer SJ, Zauber AG, Ho MN, et al. Preven- 271:1011-1013. itz GA. Screening endoscopy and risk of colorectal can- tion of colorectal cancer by colonoscopic polypec- 23. Church TR, Ederer F, Mandel JS. Fecal occult blood cer in United States men. Cancer Causes Control. 1998; tomy: the National Polyp Study Workgroup. N Engl J screening in the Minnesota study: sensitivity of the 9:455-462. Med. 1993;329:1977-1981. screening test. J Natl Cancer Inst. 1997;89:1440- 43. Rex DK, Lehman GA, Hawes RH, O’Connor KW, 63. Citarda F, Tomaselli G, Capocaccia R, Barcherini 1448. Smith JJ. Performing screening flexible sigmoidos- S, Crespi M. Efficacy in standard clinical practice of colo- 24. Ransohoff DF, Lang CA. Screening for colorectal copy using colonoscopes: experience in 500 sub- noscopic polypectomy in reducing colorectal cancer cancer with the fecal occult blood test: a background jects. Gastrointest Endosc. 1990;36:486-488. incidence. Gut. 2001;48:812-815. paper: American College of Physicians. Ann Intern Med. 44. Schertz RD, Baskin WN, Frakes JT. Flexible fiber- 64. Marshall JB. Technical proficiency of trainees per- 1997;126:811-822. optic sigmoidoscopy training for primary care physi- forming colonoscopy: a learning curve. Gastrointest 25. Winawer SJ, Fletcher RH, Miller L, et al. Colorec- cians: results of a 5-year experience. Gastrointest En- Endosc. 1995;42:287-291. tal cancer screening: clinical guidelines and rationale. dosc. 1989;35:316-320. 65. Nelson DB, McQuaid KR, Bond JH, Lieberman DA, Gastroenterology. 1997;112:594-642. 45. Cash BD, Schoenfeld PS, Ransohoff DF. Licen- Weiss DG, Johnston TK. Procedural success and com- 26. Anderson ML, Pasha TM, Leighton JA. Endo- sure, use, and training of paramedical personnel to per- plications of large-scale screening colonoscopy. Gas- scopic perforation of the colon: lessons for a 10-year form screening flexible sigmoidoscopy. Gastrointest trointest Endosc. 2002;55:307-314. study. Am J Gastroenterol. 2000;95:3418-3422. Endosc. 1999;49:163-169. 66. Winawer SJ, Zauber AG, O’Brien MJ, et al. Ran- 27. Zubarik R, Fleischer DE, Mastropietro C, et al. Pro- 46. American Society for Gastrointestinal Endos- domized comparison of surveillance intervals after colo- spective analysis of complications 30 days after out- copy. Guidelines for training non-specialists in screen- noscopic removal of newly diagnosed adenomatous patient colonoscopy. Gastrointest Endosc. 1999;50: ing flexible sigmoidoscopy. Gastrointest Endosc. 2000; polyps: the National Polyp Study Workgroup. N Engl 322-328. 51:783-785. J Med. 1993;328:901-906. 28. American College of Physicians. Suggested tech- 47. Wallace MB, Kemp JA, Meyer F, et al. Screening 67. Zauber AG, Winawer SJ, Bond JH, et al. Can surveil- nique for fecal occult blood testing and interpreta- for colorectal cancer with flexible sigmoidoscopy by non- lance intervals be lengthened following colonoscopic pol- tion in colorectal cancer screening. Ann Intern Med. physician endoscopists. Am J Med. 1999;107:214- ypectomy [abstract]. Gastroenterology. 1997;112:A50. 1997;126:808-810. 218. 68. Rex DK, Cummings OW, Helper DJ, et al. 5-year 29. Allison JE, Feldman R, Tekawa IS. Hemoccult 48. Schoenfeld P, Lipscomb S, Crook J, et al. Accuracy incidence of adenomas after negative colonoscopy in screening in detecting colorectal neoplasm: sensitiv- of polyp detection by gastroenterologists and nurse en- asymptomatic average-risk persons [see comment]. ity, specificity, and predictive value: long-term fol- doscopists during flexible sigmoidoscopy: a random- Gastroenterology. 1996;111:1178-1181. low-up in a large group practice setting. Ann Intern ized trial. Gastroenterology. 1999;117:312-318. 69. Avidan B, Sonnenberg A, Schnell TG, Leya J, Metz Med. 1990;112:328-333. 49. Levin TR, Conell C, Shapiro JA, Chazan SG, Nadel A, Sontag SJ. New occurrence and recurrence of neo- 30. Ahlquist DA, Wieand HS, Moertel CG, et al. Ac- M, Selby JV. Complications of screening sigmoidos- plasms within 5 years of a screening colonoscopy. Am curacy of fecal occult blood screening for colorectal copy. Gastroenterology. 2002;123:1786-1792. J Gastroenterol. 2002;97:1524-1529. neoplasia: a prospective study using Hemoccult and 50. Atkin WS, Edwards R, Wardle J, Northover JM, 70. Hixson LJ, Fennerty MB, Sampliner RE, Garewal HemoQuant tests. JAMA. 1993;269:1262-1267. Cuzick J. UK flexible sigmoidoscopy screening trial: HS. Prospective blinded trial of the colonoscopic miss- 31. Lieberman DA, Weiss DG. One-time screening for compliance, yield and adverse events [abstract]. Gas- rate of large colorectal polyps. Gastrointest Endosc. colorectal cancer with combined fecal occult-blood test- troenterology. 2000;118:A187. 1991;37:125-127. ing and examination of the distal colon. N Engl J Med. 51. Zarchy TM, Ershoff D. Do characteristics of ad- 71. Rex DK, Cutler CS, Lemmel GT, et al. Colono- 2001;345:555-560. enomas on flexible sigmoidoscopy predict advanced scopic miss rates of adenomas determined by back- 32. Prorok PC, Andriole GL, Bresalier RS, et al. De- lesions on baseline colonoscopy? Gastroenterology. to-back colonoscopies. Gastroenterology. 1997;112: sign of the Prostate, Lung, Colorectal and Ovarian 1994;106:1501-1504. 24-28.

72. Bensen S, Mott LA, Dain B, Rothstein R, Baron J. cer in clinical practice. Gastroenterology. 1997;112: nography (virtual colonoscopy). Gut. 2000;47:126- The colonoscopic miss rate and true one-year recur- 17-23. 130. rence of colorectal neoplastic polyps: Polyp Preven- 83. Law RL, Longstaff AJ, Slack N. A retrospective 94. Miao YM, Amin Z, Healy J, et al. A prospective tion Study Group. Am J Gastroenterol. 1999;94:194- 5-year study on the accuracy of the barium enema ex- single centre study comparing computed tomogra- 199. amination performed by radiographers. Clin Radiol. phy pneumocolon against colonoscopy in the detec- 73. Winawer SJ, Flehinger BJ, Schottenfeld D, Miller 1999;54:80-83; discussion 83-84. tion of colorectal neoplasms. Gut. 2000;47:832-837. DG. Screening for colorectal cancer with fecal occult 84. Strom E, Larsen JL. Colon cancer at barium en- 95. Spinzi G, Belloni G, Martegani A, Sangiovanni A, blood testing and sigmoidoscopy. J Natl Cancer Inst. ema examination and colonoscopy: a study from the Del Favero C, Minoli G. Computed tomographic colo- 1993;85:1311-1318. county of Hordaland, Norway. Radiology. 1999;211: nography and conventional colonoscopy for colon dis- 74. Berry DP, Clarke P, Hardcastle JD, Vellacott KD. 211-214. eases: a prospective, blinded study. Am J Gastroen- Randomized trial of the addition of flexible sigmoi- 85. Winawer SJ, Stewart ET, Zauber AG, et al. A com- terol. 2001;96:394-400. doscopy to faecal occult blood testing for colorectal parison of colonoscopy and double-contrast barium 96. Zalis ME, Hahn PF. Digital subtraction bowel neoplasia population screening. Br J Surg. 1997;84: enema for surveillance after polypectomy: National cleansing in CT colonography. AJR Am J Roent- 1274-1276. Polyp Study Work Group. N Engl J Med. 2000;342: genol. 2001;176:646-648. 75. Rasmussen M, Kronborg O, Fenger C, Jor- 1766-1772. 97. Bond JH. Virtual colonoscopy—promising, but not gensen OD. Possible advantages and drawbacks of 86. Hara AK, Johnson CD, Reed JE, et al. Detection ready for widespread use. N Engl J Med. 1999;341: adding flexible sigmoidoscopy to hemoccult-II in of colorectal polyps with CT colonography: initial as- 1540-1542. screening for colorectal cancer: a randomized study. sessment of sensitivity and specificity. Radiology. 1997; 98. Akerkar GA, Yee J, Hung R, McQuaid K. Patient Scand J Gastroenterol. 1999;34:73-78. 205:59-65. experience and preferences toward colon cancer 76. Verne JE, Aubrey R, Love SB, Talbot IC, Northover 87. Rex DK, Vining D, Kopecky KK. An initial experi- screening: a comparison of virtual colonoscopy and JM. Population based randomized study of uptake and ence with screening for colon polyps using spiral CT with conventional colonoscopy. Gastrointest Endosc. 2001; yield of screening by flexible sigmoidoscopy com- and without CT colonography (virtual colonoscopy). 54:310-315. pared with screening by faecal occult blood testing. Gastrointest Endosc. 1999;50:309-313. 99. Sonnenberg A, Delco F, Bauerfiend P. Is virtual BMJ. 1998;317:182-185. 88. Fenlon HM, Nunes DP, Schroy PC 3rd, Barish MA, colonoscopy a cost-effective option to screen for co- 77. Levin TR, Allison JE, Sakoda L, et al. Adding new Clarke PD, Ferrucci JT. A comparison of virtual and lorectal cancer? Am J Gastroenterol. 1999;94:2268- fecal occult blood tests to sigmoidoscopy: what is the conventional colonoscopy for the detection of colo- 2274. added yield [abstract]. Gastroenterology. 2000;118: rectal polyps. N Engl J Med. 1999;341:1496-1503. 100. Fearon ER, Vogelstein B. A genetic model for co- A441. 89. Fletcher JG, Johnson CD, Welch TJ, et al. Opti- lorectal tumorigenesis. Cell. 1990;61:759-767. 78. Kim LS, Koch J, Yee J, Halvorsen R, Cello JP, Rockey mization of CT colonography technique: prospective 101. Sidransky D, Tokino T, Hamilton SR, et al. Iden- DC. Comparison of patients’ experience during im- trial in 180 patients. Radiology. 2000;216:704-711. tification of ras oncogene mutations in the stool of pa- aging tests of the colon. Gastrointest Endosc. 2001; 90. Hara AK, Johnson CD, Welch TJ, McCollough CH, tients with curable colorectal tumors. Science. 1992; 54:67-74. Harmsen WS. CT colonography: single- versus multi- 256:102-105. 79. Kewenter J, Brevinge H. Endoscopic and surgical detector row imaging. Radiology. 2001;219:461- 102. Traverso G, Shuber A, Levin B, et al. Detection complications of work-up in screening for colorectal 465. of APC mutations in fecal DNA from patients with co- cancer. Dis Colon Rectum. 1996;39:676-680. 91. Yee J, Akerkar GA, Hung RK, Steinauer- lorectal tumors. N Engl J Med. 2002;346:311-320. 80. Blakeborough A, Sheridan MB, Chapman AH. Gebauer AM, Wall SD, McQuaid KR. Colorectal neo- 103. Ahlquist DA, Skoletsky JE, Boynton KA, et al. Co- Complications of barium enema examinations: a sur- plasia: performance characteristics of CT colonogra- lorectal cancer screening by detection of altered hu- vey of UK Consultant Radiologists 1992 to 1994. Clin phy for detection in 300 patients. Radiology. 2001; man DNA in stool: feasibility of a multitarget assay Radiol. 1997;52:142-148. 219:685-692. panel. Gastroenterology. 2000;119:1219-1227. 81. Johnson CD, Ilstrup DM, Fish NM, et al. Barium 92. Cotton PB, Durkalski VL, Palesch YY, et al. Com- 104. Dong SM, Traverso G, Johnson C, et al. Detect- enema: detection of colonic lesions in a community parison of virtual colonoscopy and colonoscopy in the ing colorectal cancer in stool with the use of multiple population. AJR Am J Roentgenol. 1996;167:39-43. detection of polyps/masses [abstract]. Gastrointest En- genetic targets. J Natl Cancer Inst. 2001;93:858-865. 82. Rex DK, Rahmani EY, Haseman JH, Lemmel GT, dosc. 2002;55:A594. 105. Traverso G, Shuber A, Olsson L, et al. Detec- Kaster S, Buckley JS. Relative sensitivity of colonos- 93. Pescatore P, Glucker T, Delarive J, et al. Diagnos- tion of proximal colorectal cancers through analysis copy and barium enema for detection of colorectal can- tic accuracy and interobserver agreement of CT colo- of faecal DNA. Lancet. 2002;359:403-404.

A teacher affects eternity; he can never tell where his influence stops. —Henry Adams (1838-1918)