Applied Radiology

Dr. Shaffer is Professor of Clinical Radiology and Clinical Internal Medicine at the University of Virginia School of Medicine and Co-Director of the Division of Thoracoabdominal Imaging at the University of Virginia Health Sciences Center in Charlottesville, VA.

C olorectal cancer (CRC) is a lethal disease and is gaining attention as an important public heath issue. The American Cancer Society estimates that 129,400 new cases of CRC will occur in the United States in 1999. ¹ One out of every 20 Americans will develop this cancer within their lifetime. ^2,3 CRC is the second leading cause of cancer deaths, exceeded only by lung cancer. Approximately 56,600 Americans will die of CRC this year, and well over 6 million Americans alive today will succumb to this malignancy. ^1,4

Risk factors

A number of risk factors for CRC are well documented. ^3,5,6 A genetic predisposition to CRC is present in those persons with a family history of CRC or adenomas in first degree relatives, familial adenomatous polyposis (FAP) syndrome, or hereditary non-polyposis colorectal cancer (HNPCC) syndrome. Individuals who have a personal history of CRC or adenomatous polyps are at increased risk of having subsequent colon neoplasms. ^7,8 Chronic inflammatory bowel disease, especially ulcerative colitis, places a patient in jeopardy for acquiring CRC. Women who have had breast, ovarian, or endometrial cancer also may develop CRC as part of the Lynch II syndrome. ^2,5 However, to put things in perspective, it should be recognized that individuals at high and moderate risk make up only a small fraction of the general population, and 70 to 80% of CRC occurs sporadically in persons with no known risk factors. ^3,6

Adenoma-carcinoma sequence

Most CRC develops from adeno-
matous polyps--the well-recognized
"adenoma-carcinoma sequence." ⁶ An exception to this is found in patients with chronic inflammatory disease of the large bowel, where a malignancy can develop within dysplastic mucosa without passing through a polypoid stage. ⁵ Although most CRC arises from adenomas, the majority of adenomas do not become cancers. It is estimated that 95% of all 5-mm diameter adenomas will remain benign. ^9,10 Unfortunately, we have no way to identify the one adenoma in 20 that will become malignant. However, we do know that the larger the polyp, the more likely it is to harbor malignancy (table 1). ^11,12 An adenomatous polyp which is less than 10 mm in diameter has approximately a 1.0% chance of containing an invasive carcinoma. Cancer is found in 5 to 10% of polyps between 1 and 2 cm in size and in 11 to 46% of polyps over 2 cm in diameter. Based on autopsy data, the anticipated malignancy conversion rate for large adenomas is 3% per year. ¹³

CRC survival

The major determinant of CRC survival is the stage at which the cancer is discovered. ^14-17 A tumor confined to the mucosa (carcinoma-in-situ) has a 100% 5-year survival rate after surgery; the prognosis for survival becomes progressively worse for patients with a greater depth of colon wall invasion, local spread to adjacent structures, regional lymph node involvement, and distant metastases (table 2). Detection and endoscopic removal of precursor adenomas can prevent CRC, and early-stage cancer detection can lead to surgical cure. ⁶

Testing scenarios

There are three clinical scenarios in which testing for the presence of CRC or its precursors may be considered. The first is diagnostic testing of persons suspected of having CRC or polyps because of signs or symptoms of the disease. This would include individuals with documented intestinal bleeding, manifested either by visible blood per rectum or invisible bleeding detected by a fecal occult blood test (FOBT). Diagnostic testing also would apply to persons with unexplained anemia, change in bowel habits, or weight loss. Secondly, surveillance testing is appropriate for patients at increased risk of developing CRC because of a personal history of CRC, polyps, or inflammatory bowel disease, or a strong family history of CRC or adenomatous polyps. Lastly, the term CRC screening describes testing in average-risk individuals in the appropriate age group without signs or symptoms of the disease.

Rationale for screening

The rationale for recommending CRC screening in the asymptomatic general population is that about 75% of all CRC occurs in people classified as "average" risk. ⁶ This malignancy is a major cause of morbidity and mortality, yet it is both curable and preventable. ¹⁸ Clinical studies have found that periodic screening and removal of adenomatous polyps decreases the incidence of CRC, and detection of early-stage cancers reduces CRC mortality. ⁶ Although most cancers develop from adenomas, the malignant potential of these precursor polyps is low and slow to develop. Only 5% of 5-mm adenomas become malignant, and it takes a minimum of 5 years, and an average of 10 years, for a small adenoma to develop into carcinoma. ^6,19 The demonstrated stability of most polyps and the generally long lead time for malignant transformation are reasons for optimism in determining strategies for detecting and removing precancerous polyps and early-stage cancers. ⁶

Adenomas can be found in approximately 25% of people by age 50 years, and the prevalence increases to 50 to 60% by the age of 80 years. ²⁰ The incidence of CRC has shown an increase after age 40 and accelerates sharply after age 50, with 90% of cancers occurring in persons 50 years of age and older (figure 1). ^21,22 Thus, a CRC screening strategy should logically begin no later than 50 years of age, and testing should be performed at intervals that allow timely detection of previously overlooked small polyps, as well as new lesions, that have enlarged to sizes of 10 mm or greater. ²³ It is projected that universal CRC screening has the potential to increase the current 5-year survival rate of 63% for whites and 53% for African Americans to 91% for all Americans. ^18,24 Sadly, fewer than 20% of American adults currently undergo CRC screening. ²⁵

Screening tests

There are at least six desirable attributes in any CRC screening test: 1) it should be widely available and easily accessible; 2) it should reliably detect premalignant polyps and early-stage cancer, not just advanced disease; 3) the test should reliably exclude the presence of the disease in normal individuals; 4) it should produce minimal discomfort to assure patient acceptance and compliance; 5) it must have a low risk of serious complications; and 6) the cost-benefit ratio must seem reasonable to the payer. Four generally-accepted tests are available for CRC screening: fecal occult blood test (FOBT), sigmoidoscopy, barium enema, and colonoscopy. Unfortunately, none of these tests is ideal for screening average-risk individuals. ²⁶

Fecal occult blood test

The FOBT is universally available, simple to perform, noninvasive, very safe, and of relatively low cost. It has been well documented that FOBT screening achieves a statistically and clinically significant reduction in death from CRC. ^3,6 Its biggest drawback, however, is that it is unreliable. ²⁷ In persons who have positive test results, CRC is found only in 2 to 17%; thus, the large majority of reactive tests are false-positive. The test also has a high false-negative rate. The average sensitivity reported for FOBT is in the range of 40 to 60% for cancers and 5 to 10% for adenomas. ^16,28,29 This is due to the intermittent nature, or even absence, of bleeding from these lesions. In addition, the FOBT has poor patient compliance, and it does not provide direct structural evaluation of the large intestine. The general recommendation that the FOBT be repeated annually is based on the theory that a test of low sensitivity may be more effective if repeated often. ²⁷ For those individuals who elect a periodic total colon examination with colonoscopy or double-contrast barium enema (DCBE), there is no need for an annual FOBT. ³

Sigmoidoscopy

This examination is widely available, minimally invasive, relatively safe, and only moderately expensive. Sigmoidoscopy is well tolerated by most individuals, so sedation is not commonly used. Reports of bowel perforation are rare. ²⁷ Examination with a flexible 60 cm scope is preferred over a 25 cm rigid sigmoidoscope because of the greater length of colon that can be evaluated. Still, the flexible instrument permits visualization of only about 40% of the large intestine and can be expected to detect the presence of no more than 40 to 60% of CRC and polyps. ^6,27,30 In patients under 50 years of age, this test may be considered adequate for screening because 75% of polyps will be located in the rectosigmoid colon. However, as patients grow older, an increasing number of tumors are located in the more proximal colon and beyond the reach of the sigmoidoscope (figure 2). ³¹

After age 60, a majority of polyps are beyond the reach of the sigmoidoscope, and this CRC screening strategy is no longer as effective or cost-effective.

The barium enema and colonoscopy

Colonoscopy and the barium enema are the only currently available screening tests capable of providing structural evaluation of the entire colon from the rectum to the cecum. The American Cancer Society (ACS) terms both procedures "total colon examinations" (TCE). ³ Although published reports indicate that the air- or double-contrast barium enema (DCBE) and the single-contrast barium enema (SCBE) are approximately equivalent in sensitivity for the diagnosis of colonic cancer and large polyps, ^32-34 the DCBE is generally considered superior for detection of rectal lesions and small polyps, making it the radiologic screening examination of choice. ^27,35,36 The SCBE remains useful in selected screening situations, such as with very elderly and/or immobile patients. ³⁷

There is voluminous literature comparing DCBE and colonoscopy for the diagnosis of CRC and polyps, but most reports are retrospective studies in symptomatic patients or those with rectal bleeding. There are no published clinical studies specifically evaluating whether screening DCBE or screening colonoscopy alone reduces the incidence or mortality from colorectal cancer in people at average risk for the disease. ⁶ For comparative studies, colonoscopy usually is designated as the "gold standard." This may lead to an overestimation of the sensitivity of colonoscopy, because all false-negative colonoscopies are assumed to be true negatives. This comparison also leads to a bias against the specificity of the DCBE, as all false-negative colonoscopies are assumed to be false-positive barium enemas. ²⁶ The best controlled studies comparing colonoscopy and DCBE show that while colonoscopy detects more small polyps, there is little difference in the ability of these tests to detect large polyps (>=10 mm) and cancers. The combined false-positive rate (higher for DCBE) and rate of inadequate examinations (higher for colonoscopy) are about the same for the two procedures.

Schrock ³⁸ is likely correct in commenting that the "barium enema is neither as good as radiologists claim nor as bad as endoscopists believe. Colonoscopy is neither as good as endoscopists claim nor as bad as radiologists believe."

In the best hands, both colonoscopy and DCBE are reported to have a sensitivity of about 90% for detection of clinically significant polyps and CRC. ^30,39 However, both tests are operator dependent, and this level of lesion detection is unlikely to be achieved in clinical practice except by experts under ideal circumstances. Results from the studies with the fewest biases suggest that the sensitivity of the DCBE is between 50 and 80% for polyps of less than 10 mm, 70 and 90% for polyps greater than 10 mm, and about 90% for CRC. ^6,40 Studies using the barium enema as the gold standard consistently show that colonoscopy misses about 10 to 20% of confirmed lesions detected by barium enema. ²⁷ Studies using same day back-to-back colonoscopy as a gold standard show a miss rate for colonoscopy of 16 to 27% for polyps of 5 mm in diameter or less, 12 to 13% for polyps of 6 to 9 mm, and up to 6% for polyps measuring 10 mm or greater. ^41-43

While colonoscopy generally is considered superior to DCBE for detection of polyps of less than 6 mm in size, the rate of colonoscopic perforation is greater than the low incidence of malignancy in these small polyps; thus, the need to detect such small polyps has been seriously questioned. ⁴⁴

Why do colonoscopy and DCBE fail to diagnose some colorectal tumors? During colonoscopy, undetected lesions usually are missed either because the tumor site is not reached or the lesion is passed by the scope but not identified by the endoscopist. A prospective survey study done in 1984 by the American Society of Gastrointestinal Endoscopy showed that 26% of colonoscopies performed by the 672 responding society members were incomplete. ⁴⁵ Other studies report failure to reach the cecum in anywhere from 5 to 45%, with operator experience and newer scope designs being significant factors in attaining the higher rates of success. ²⁷ Other causes of incomplete colonoscopy are poor bowel cleansing, excessive bowel redundancy, benign or malignant stricture, and severe diverticulosis (figure 3). Because the right colon frequently harbors polyps and cancers, incomplete studies require that either a repeat colonoscopy or a barium enema be performed (figure 4). ⁶ When lesions are passed but not detected by colonoscopy, it usually is because the lesion was hidden behind a fold, particularly on the inside of a tight turn, such as that which occurs at colonic flexures (figure 5). ⁴⁶

Conversely, a barium enema rarely fails to reach the cecum and allow complete colon evaluation. When carcinomas are missed on a DCBE, the lesions are visible about 90% of the time on retrospective review of the films; thus, a technical problem or perceptive error by the radiologist usually is responsible for failure to report the lesion or properly interpret its nature. ^47-49 Because human error is a frequent cause of failure to make a correct diagnosis, it has been postulated that performing double-reading of barium enema examinations could raise the diagnostic sensitivity of the study, possibly to as high as 94%. ^26,49,50

Both colonoscopy and DCBE are painful or, at best, uncomfortable procedures. It is not clear which is more acceptable to patients. Medications to provide analgesia, sedation, and amnesia are regularly given during colonoscopy but are rarely used during barium enema studies. Because of the delayed amnestic effect of drugs like medazolam (Versed, Roche Laboratories, Nutley NJ), it is not surprising that patients surveyed several weeks after undergoing the two procedures recall that they found the procedures to be equally well tolerated or that they preferred colonoscopy. ^51,52 The reverse is true when patients are questioned immediately following the two procedures, as was done in a randomized study by Stein of 190 patients agreeing to undergo CRC screening by both procedures. ⁵³ When her patients were questioned immediately afterwards, moderate to severe pain was reported by 85% of individuals during colonoscopy and by 46% during DCBE. The intensity of the pain on a scale of 1 to 10 (none­severe) was 6.0 for colonoscopy and 3.4 for DCBE. The length of time that the patients experienced pain was related to the length of the procedure, which averaged 31 minutes for colonoscopy and 14 minutes for DCBE.

Serious complications as a result of colonoscopy or DCBE are infrequent. However, perforation is 10 to 90 times more common during colonoscopy (1:500 to 1:1000) than during DCBE (1:10,000 to 1:46,000). ^6,30,39,54 Fatalities are 100 to 200 times more frequent with colonoscopy (1:5000 to 1:10,000) than with DCBE (1:1,000,000). ^27,54 In addition, with colonoscopy, reports indicate that 3:1000 have major hemorrhage and 5:1000 experience clinically significant respiratory depression. ^6,55

The radiation dose received by an individual during barium enema examination has been of concern to many patients and physicians. In reality, the dose is relatively small. During a barium enema, patients are exposed to 300 to 500 mrem, which compares to a radiation dose of 300 mrem during mammography. Considering the age and frequency at which screening is commonly recommended, a CRC screening strategy using barium enema every 5 to 10 years would deliver a lifetime dose that is considerably lower than that for annual screening mammography. There is no direct evidence that barium enema examinations cause clinically important increases in the risk of cancer or other tissue damage. ⁶

Little attention has been paid to the potential for disease transmission between patients from fecal contamination of instruments. The U.S. Food and Drug Administration and three state health departments found that 24% of fiber-optic endoscopes harbored a significant number of viable bacteria after they had undergone routine disinfection. As a result, they report that "outbreaks with substantial morbidity and mortality have occurred." ⁵⁶ Other scattered reports in the literature have described minor epidemics caused by contaminated endoscopes. ²⁷ To our knowledge, no studies have investigated the potential for viral transmission by contaminated endoscopes. Conversely, with barium enemas, disease transmission by instruments is not an issue, as only disposable administration sets and contrast media are used to perform the examination.

With the current focus on cost containment in medical care, it is important to appreciate that colonoscopy is much more expensive than DCBE: In most reported series, colonoscopy is 3 to 5 times more costly. In performing CRC screening of large populations, this cost differential is multiplied and becomes enormous. Also contributing to the economic impact of colonoscopy is the fact that the patient must have a recovery period of 1 to 2 hours following the administration of conscious sedation and can not immediately drive an automobile or return to work following the procedure.

Cost-effective screening

Because of a lack of large clinical trials comparing CRC screening tests in an average-risk population, computer models have been used to evaluate the cost-effectiveness of the four common screening tests: FOBT, flexible sigmoidoscopy, DCBE, and colonoscopy. Data largely derived from studies in symptomatic patients have been extrapolated to predict the most effective screening strategies for asymptomatic subjects.

After review of the applicable literature, Eddy discovered that the most cost-effective way to decrease the death rate from cancer is to perform a barium enema every 3 to 5 years (figure 6); ³⁹ slightly less cost-effective was colono-scopy performed every 5 years. Eddy found that more frequent testing was more costly and was ineffective in saving additional lives.

In 1995, the Congressional Office of Technology Assessment published a comprehensive study which found that all four of the CRC screening strategies offer health benefits at a price considerably less than the benchmark value--roughly $40,000 per added year of life--commonly applied to preventive technologies. ⁵⁷ Despite using an assumption that the sensitivity of DCBE for polyps and cancer is only 70% in an average-risk screening population, they found that DCBE and colonoscopy are comparably cost-effective screening strategies at both 5-year and 10-year intervals (figure 7).

In 1997, an expert panel convened by the United States Agency for Health Care Policy and Research concluded from computer simulations that a screening barium enema every 5 years and screening colonoscopy every 10 years are approximately equivalent in the ability to save lives. In 1998, Glick et al reported the results of another computer model, also using a conservative assumption of 70% sensitivity for detection of polyps and cancer by DCBE compared to a liberal assumption of 90% sensitivity for colonoscopy. Despite this, their data also indicate that the DCBE is comparatively cost-effective as a screening procedure in average-risk individuals. ⁵⁸

All four of the above mentioned studies observed that screening sigmoidoscopy alone (at any interval) is less effective that the other screening strategies. It should be noted, however, that all of the commonly recommended strategies for CRC screening are more cost-effective than breast cancer screening with mammography and cervical cancer screening using the Papanicolaou technique. ^39,57

The American Cancer Society, American Gastroenterological Association, American College of Physicians, American College of Gastroenterology, American Society of Colon and Rectal Surgeons, American Society for Gastrointestinal Endoscopy, American College of Radiology (ACR), American Medical Association, and others are promoting CRC screening and the use of the DCBE as a primary screening tool in average-risk individuals. ^3,6,39,59,60 Most of their current recommendations are based on a landmark article in 1997 by Winawer et al, which established guidelines and rationale for CRC screening. ⁶ The highly respected and frequently quoted ACS guidelines are rather complex because they are tailored to the risk potential of each patient group; however, their recommendations are straightforward for the estimated 70 to 80% of the population who are at average risk (table 3). ³

Medicare reimbursement

Recently, the potential for universal CRC screening was greatly enhanced with the approval of Medicare coverage for such screening tests. Medicare payment was authorized by enactment of The Balanced Budget Act of 1997; ⁶¹ the details of Medicare coverage were left to the Secretary of Health and Human Services to define and were published in The Federal Register. ⁶² Medicare payment for CRC screening was implemented on January 1, 1998. Currently, the Healthcare Financing Administration (HCFA) is allowing Medicare coverage of CRC screening for persons aged 50 years and older (table 4). Under their rules, FOBT can be performed annually, and for average risk individuals, DCBE or sigmoidoscopy can be performed every 4 years. For high-risk individuals, DCBE or colonoscopy can be performed every 2 years.

Colonoscopy has not been approved for CRC screening in the general population due to its higher risk of morbidity and mortality and its much greater cost.

One would expect that Medicare payment for CRC screening would stimulate an increase in the number of barium enemas being performed in this country. However, in the year since the government authorized Medicare coverage for this service, many, if not most radiologists have seen no evidence of a jump in the volume of these studies. Instead, there has actually been a steady decline in the number of barium enemas performed in this country for the last 15 years, with a commensurate increase in the number of colonoscopies performed.

What is wrong? One obvious answer is that radiologists have been much less interested and effective in educating the public and referring physicians about the benefits of the DCBE than have the endoscopists regarding colonoscopy. In addition, many radiologists have focused their attention on mastering newer technologies. As a result, the quality of barium enema examinations has reportedly become inconsistent, and many referring physicians have lost confidence in the value of the examination. If the barium enema is to be utilized for CRC screening, individual radiologists must rededicate themselves to improving the quality of the study and convince their colleagues that it is an effective technique. ⁶³

Standards and quality control

To improve the uniformity and quality of barium enema examinations, the American College of Radiology (ACR) has developed performance standards for these studies. ⁶⁴ The standards include: 1) appropriate training/certification of radiologists and radiologic technologists, 2) equipment standards and quality control procedures to maintain high-quality fluoroscopic and radiographic images, 3) effective bowel preparation to achieve a clean colon, and 4) strict quality control during the examination using a barium suspension specifically designed for the procedure, introducing an adequate volume of the contrast agent(s) to achieve complete colon examination, and using optimal patient positioning and radiographic technique for film documentation. The ACR quality controls specific to the DCBE are: 1) complete barium coating of the entire colon, 2) good distension of the colon with gas, and 3) demonstration of every segment of the colon in double contrast on at least two radiographs taken in different positions. The ACR also has developed a Barium Enema Quality Control Manual to help facilities establish and maintain a quality control program. ⁶⁵ This manual, published in 1998, follows the highly successful manual on mammography quality control, first published by the ACR in 1990.

Dr. Richard Mendelson, an Australian radiologist, expressed clearly the challenge that faces radiologists worldwide when he wrote, "The DCBE remains a potentially useful tool...but its continued existence and accuracy are highly dependent on adherence to strict quality control issues....There needs to be an acceptance by radiologists that good-quality barium radiology is a challenge and a goal to be strived for and that gastrointestinal (GI) radiology is a valid subspecialty." ²⁶

As new approaches to CRC screening--with blood and stool tumor markers or new imaging modalities such as CT and MR colonography--come to market, screening strategies will undoubtedly change. For the present, however, DCBE has been established as a safe, accurate, cost-effective, and widely-available total colon examination. Radiologists are obliged to perform the procedure well and educate patients and medical colleagues about the advantages of DCBE as a primary screening test for colorectal cancer and polyps. AR

References

1. Landis SH, Murray T, Bolden S, Wingo PA: Cancer statistics, 1999. CA Cancer J Clin 49:8-31, 1999.

2. DeCosse JJ, Tsioulias GJ, Jacobson JS: Colorectal cancer: Detection, treatment, and rehabilitation. CA Cancer J Clin 44:27-42, 1994.

3. Byers T, Levin B, Rothenberger D, et al: American Cancer Society guidelines for screening and surveillance for early detection of colorectal polyps and cancer: Update 1997. CA Cancer J Clin 47:154-160, 1997.

4. Seidman H, Mushinski MH, Gelb SK, et al: Probabilities of eventually developing and dying of cancer: United States 1995. CA Cancer J Clin 35:36-56, 1985.

5. Gore RM: Colorectal cancer. Clinical and pathologic features. Radiol Clin N Am 35:403-429, 1997.

6. Winawer SJ, Fletcher RH, Miller L, et al: Colorectal cancer screening: Clinical guidelines and rationale. Gastroenterology 112:594-642, 1997.

7. Otchy DP, Ransohoff DF, Wolff BG, et al: Metachronous colon cancer in persons who have had a large adenomatous polyp. Am J Gastroenterol 91:448-454, 1996.

8. Atkin WS, Morson BC, Cuzick J: Long-term risk of colorectal cancer after excision of rectosigmoid adenomas. N Engl J Med 326:658-662, 1992.

9. Ferrucci JT: Screening for colon cancer: Programs of the American College of Radiology. AJR 160:999-1003,1993.

10. Eide TJ: Natural history of adenomas. World J Surg 15:3-6, 1991.

11. Shinya H, Wolff WI: Morphology, anatomic distribution and cancer potential of colonic polyps. Ann Surg 190:679-683, 1979.

12. Muto T, Ishikawa K, Kino I, et al: Comparative histologic study of adenomas of the large intestine in Japan and England, with special reference to malignant potential. Dis Col Rectum 20:11-16, 1977.

13. Eide TJ: Risk of colorectal cancer in adenoma-bearing individuals within a defined population. Int J Cancer 38:173-176, 1986.

14. Gabriel WB, Dukes C, Bussey HJR: Lymphatic spread in cancer of the rectum. Br J Surg 23:395-413, 1936.

15. Astler VB, Coller FA: The prognostic significance of direct extension of carcinoma of the colon and rectum. Ann Surg 139:846-852, 1954.

16. Hutter RVP, Sobin LH: A universal staging system for cancer of the colon and rectum. Let there be light. Arch Pathol Lab Med 110:367-368, 1986.

17. Levini B, Raijman I: Malignant tumors of
the colon and rectum. In: Haubrich WS, Schaffner F, Berk JE (eds): Bockus Gastroenterology, ed 5, pp 1744-1772. Philadelphia, WB Saunders Co., 1995.

18. CDC Announcement 773, Federal Register 62(130):36528-36533, Washington, DC, United States Government Printing Office, 1997.

19. Muto T: The evolution of cancer of the colon and rectum. Cancer 36:2251-2270, 1975.

20. Williams AR, Balasooiya BAW, Day DW: Polyps and cancer of the large bowel: A necropsy study in Liverpool. Gut 123:835-842, 1982.

21. Ries LAG, Miller BA, Hankey BF, et al (eds): SEER Cancer Statistics Review, Bethesda, MD, National Cancer Center, National Institutes of Health, 1988-1992.

22. Fleshner B, Slater G, Aufses AH: Age and sex distribution of patients with colorectal cancer. Dis Colon Rectum 32:107-111, 1989.

23. Glick S, Wagner JL, Johnson CD: Cost-effectiveness of double-contrast barium enema in screening for colorectal cancer. AJR 170:629-636, 1998.

24. Landis SH, Murray T, Bolden S, Wingo PA: Cancer statistics, 1998. CA Cancer J Clin 48:6-29, 1998.

25. Trends in cancer screening--United States, 1987 and 1992. MMWR Morb Mortal Wkly Rep 45:57-61, 1996.

26. Mendelson RM: The role of the barium enema in the diagnosis of colorectal neoplasia. Australasian Radiology 42:191-196, 1998.

27. Gelfand DW: Colorectal cancer. Screening strategies. Radiol Clin N Am 35:431-438, 1997.

28. Ransohoff DF, Lang CA: Screening for colorectal cancer with the fecal occult blood test: A background paper. American College of Physicians. Ann Intern Med 126:811-822, 1997.

29. Wagner JL, Tunis S, Brown M, et al: Cost-effectiveness of colorectal cancer screening in average-risk adults. In: Young GP, Rosen P, Levin B (eds): Prevention and Early Detection of Colorectal Cancer, pp 321-356. Philadelphia, WB Saunders, 1996.

30. Brandeau ML, Eddy DM: The workup of the asymptomatic patient with a positive fecal occult blood test. Med Desc Making 7:32-46, 1987.

31. Bernstein MA, Feczko PJ, Halpert RD, et al: Distribution of colonic polyps: Increased incidence of proximal lesions in older patients. Radiology 155:35-38, 1985.

32. deRoos A, Hermans J, Shaw P, et al: Colon polyps and carcinomas: Prospective comparison of the single- and double-contrast examination in the same patients. Radiology 154:11-13, 1985.

33. Ott D, Scharling E, Chen Y, et al: Barium enema examination: Sensitivity in detecting colonic polyps and carcinomas. South Med J 82:197-200, 1989.

34. Johnson CD, Carlson HC, Taylor WF, Weiland LP: Barium enemas of carcinoma of the colon: Sensitivity of double- and single-contrast studies. AJR 140:1143-1149, 1983.

35. Smith C: Colorectal cancer: Radiologic diagnosis. Radiol Clin North Am 32:439-456, 1997.

36. Laufer I: The double-contrast enema: Myths and misconceptions. Gastrointest Radiol 1:19-31, 1976.

37. Gelfand DW, Chen YM, Ott DJ: Detection of colonic polyps on single-contrast barium enema study: Emphasis on the elderly. Radiology 164:333-337, 1987.

38. Schrock TR: Colonoscopy versus barium enema in the diagnosis of colorectal cancer and polyps. Gastrointest Endosc Clin N Am 3:585-610, 1993.

39. Eddy DM: Screening for colorectal cancer. Ann Intern Med 113:373-384, 1990.

40. Gelfand DW, Ott DJ: The economic implications of radiologic screening for colonic cancer. AJR 156:939-943, 1994.

41. Hixson LJ, Fennerty MB, Sampliner RE, et al: Prospective study of the frequency and size distribution of polyps missed by colonoscopy. J Natl Cancer Inst 82:1769-1772, 1990.

42. Cutler CS, Clark DW, Lehman GA, Rex DA: The colonoscopic miss rate for neoplastic polyps--a prospective study of back-to-back colonoscopies. Gastrointest Endosc 41:376, 1995.

43. Rex DK, Cutler CS, Lemmel GT, et al: Colonoscopic miss rates of adenomas determined by back-to-back colonoscopies. Gastroenterol 112:24-28, 1997.

44.MacCarty RL: Colorectal cancer: The case for barium enema. Mayo Clin Proc 67:253-257, 1992.

45. Gilbert DA, Shaneyfelt SL, Silverstein FE, et al: The national ASGE Colonoscopy Survey­
Analysis of colonoscopic practices and yield.
Gastrointest Endosc 30:143, 1984.

46. Miller RE, Lehman G: The barium enema. Is it obsolete? JAMA 235:2842-2844, 1976.

47. Ott DJ, Gelfand DW, Ramquist NA: Causes of error in gastrointestinal radiology: II. Barium enema examination. Gastrointest Radiol 5:99-105, 1980.

48. Kelvin F, Gardiner R, Vas W, Stevenson GW: Colorectal carcinoma missed on double contrast barium enema study: A problem in perception. AJR 137:307-313, 1981.

49. Thomas RD, Fairhurst JJ, Frost RA: Wessex regional radiology audit: Barium enema in colorectal carcinoma. Clin Radiol 50:647-650, 1995.

50. Markus J, Somers S, O'Malley B, et al: Double contrast barium enema studies: Effect of multiple reads on perception error. Radiology 175:155-156, 1990.

51. Williams CB, Macrae FA, Bartram CI: A prospective study of diagnostic methods in adenoma follow-up. Endoscopy 14:74-78, 1982.

52. Van Ness MM, Chobianian SJ, Winters C Jr., et al: A study of patient acceptance of double-contrast barium enema and colonoscopy: Which procedure is preferred by patients? Arch Intern Med 147:2175-2176, 1987.

53. Stein S: Which hurts the most? A comparison of pain rating during double-contrast barium enema examination and colonoscopy. Radiology 191:99-101, 1994.

54. Habr-Gama A, Waye JD: Complications and hazards of gastrointestinal endoscopy. World J Surg 13:193-201,1989.

55. Kaira L, Hamlyn A: Comparative evaluation of investigations for colorectal carcinoma in symptomatic patients. Post Grad Med J 64:666-668,1988.

56. Kaczmarek RG, Moore RM, McCrohan J, et al: Multi-state investigation of the actual disinfection/sterilization of endoscopes in health care facilities. Am J Med 92:257-261,1992.

57. United States Congress Office of Technology Assessment: Cost-Effectiveness of Colorectal Cancer Screening in Average-Risk Adults. Washington, D.C., United States Government Printing Office, 1995.

58. Glick S, Wagner JL, Johnson CD: Cost-effectiveness of double-contrast barium enema in screening for colorectal cancer. AJR 170:629-636, 1998.

59. Ferucci JT: Screening for colon cancer: Programs of the American College of Radiology. AJR 160:999-1003, 1993.

60. Report 7 of the Council on Scientific Affairs (I-98): Recommendations for colorectal cancer screening and surveillance in people at average and at increased risk. AMA Policy. Chicago, American Medical Association, http://www.ama-assn.org/med-sci/csa/1998/i98g.htm, 1998.

61. Conference report on H.R. 2015, Balanced Budget Act of 1997: Congressional Record, pp H6029-H6079, H6182-H6232. Washington, D.C., United States Government Printing Office, July 29, 1997.

62. Federal Register 62(211), October 31, 1997.

63. Kelvin FM: Screening for colon cancer with the barium enema: A new challenge for radiologists. ACR Bulletin 52:14-19, 1996.

64. American College of Radiology : ACR standard for the performance of adult barium enema. 1998 Standards, pp 71-73. Reston, VA, American College of Radiology, 1998.

65. American College of Radiology Committee on Barium Enema Quality Control: Barium Enema Quality Control Manual, pp 1-195. Reston, VA, American College of Radiology, 1998.

66. Miller RE, Lehman G: The barium enema. Is it obsolete? JAMA 235:2842-2844, 1976.