Prospective Study Comparing Multislice CT Colonography With Colonoscopy in the Detection of Colorectal Cancer and Polyps Venkatesh Munikrishnan, M.R.C.S.,* Alice R. Gillams, F.R.C.R.,† William R. Lees, F.R.C.R.,† Carolynne J. Vaizey, M.D., F.R.C.S.,* Paul B. Boulos, M.S., F.R.C.S.* From the Departments of *Surgery and †Academic Imaging, Royal Free and University College Medical School, London, United Kingdom PURPOSE: Multislice CT colonography is an alternative to colonoscopy. The purpose of this study was to compare multislice CT colonography with colonoscopy in the detection of colorectal polyps and cancers. METHODS: Between June 2000 and December 2001, 45 males and 35 females (median age, 68 (29–83) years) with symptoms of colorectal disease were studied prospectively. All patients underwent multislice CT colonography and colonoscopy, and the findings were compared. RESULTS: Colonoscopy was incomplete in 18 (22 percent) patients because of obstructing lesions or technical difficulty, and multislice CT colonography was unsuccessful in 4 (5 percent) because of fecal residue. Colonoscopy was normal in 26 patients and detected 29 colorectal cancers and 33 polyps in 35 patients, diverticulosis in 16 patients, and colitis in 3 patients. Multislice CT colonography identified 28 of 29 colorectal cancers with one false negative and one false positive (sensitivity, 97 percent; specificity, 98 percent; positive predictive value, 96 percent; negative predictive value, 98 percent). Multislice CT colonography identified all 12 polyps measuring ⱖ10 mm in diameter (sensitivity, 100 percent), 5 of 6 measuring 6 to 9 mm in diameter (sensitivity, 83 percent), 8 of 15 polyps ⱕ5 mm (sensitivity, 53 percent), and false-positive for 8 polyps. The overall sensitivity was 74 percent and specificity 96 percent. The positive predictive value for polyps was 88 percent, and the negative predictive value was 90 percent. Multislice CT colonography also detected 5 of 16 patients with diverticulosis (sensitivity, 31 percent; specificity, 98 percent) and colitis in 2 of 3 patients (sensitivity, 67 percent; specificity, 100 percent). In ten (13 percent) patients, extracolonic findings on multislice CT colonography altered management and included five patients with colorectal liver metastases. In 15 (19 percent) patients, there were incidental findings that did not demand further investigation. CONCLUSIONS: The results from this study indicate that the efficacy of multislice CT colonography in the detection of colorectal cancers and polyps ⱖ6 mm is similar to colonoscopy. Multislice CT colonography Presented at the meeting of The American Society of Colon and Rectal Surgeons, Chicago, Illinois, June 3 to 8, 2002. Address reprint requests to Professor Boulos: Department of Surgery, Royal Free and University College Medical School, Charles Bell House, 67-73 Riding House Street, London W1W 7EJ, United Kingdom.

allows clinical staging of colorectal cancers, outlines the whole length of the colon in obstructing carcinoma when colonoscopy fails, and can identify extracolonic causes of abdominal symptoms. [Key words: Colorectal cancer; Colonic neoplasms; Colonic polyps; Computed tomography] Munikrishnan V, Gillams AR, Lees WR, Vaizey CJ, Boulos PB. Prospective study comparing multislice CT colonography with colonoscopy in the detection of colorectal cancer and polyps. Dis Colon Rectum 2003;46:1384–1390.

C

olorectal cancer is the second commonest cause of cancer1,2 and cancer-related deaths3 in the western world. In the United Kingdom, there were more than 32,000 new cases of colorectal cancer diagnosed in 1995, and in 1997 there were 17,000 deaths. In the United States, there were approximately 130,200 new cases and 56,300 deaths from colorectal cancer in 2000. The average lifetime risk of diagnosis and death from colorectal cancer is 5.6 percent and 2.5 percent, respectively.4 It is accepted that most colorectal cancers arise from preexisting adenomatous polyps. The detection and removal of these precursor adenomas will result in a decrease in the incidence and mortality rate of colorectal cancer.5–7 CT colonography (CTC) as a technique for examination of the colon was first described by Vining and Gelfand in 1994.8 They used volumetric CT data produced by a helical CT scanner to generate threedimensional, endoluminal-perspective images displayed in a cine loop, which simulate endoluminal views at colonoscopy. At present, CTC refers to a CT examination of the fully air-distended colon. The volumetric data of the entire colon is combined with advanced-imaging software, and the colon is examined at an offline workstation by using two-dimensional and three-dimensional images to generate virtual fly-throughs (virtual colonoscopy).

1384

Vol. 46, No. 10

MULTISLICE CT COLONOGRAPHY

Multislice CT has been introduced that differs from single-slice spiral CT in its ability to produce four simultaneous 1.25-mm thin slices in one tube rotation and is six times faster, allowing examination of the entire colon in a single breath hold. It also has a higher temporal and spatial resolution, allowing faster reconstruction time at a lower radiation dose than single-slice CT. The purpose of this study was to evaluate the ability of multislice CT colonography (MCTC) in detecting colorectal cancers and polyps in patients with symptoms of colorectal disease by comparing the results of MCTC with colonoscopy.

PATIENTS AND METHODS Study Group Between June 2000 and December 2001, 80 consecutive patients (45 males; median age, 68 (29–83) years) were recruited to undergo colonoscopy and MCTC. They were referred with symptoms that included change in bowel habit, rectal bleeding, abdominal pain, loss of weight, and a rectal mass (Table 1). The exclusion criteria included impending large bowel obstruction, pregnancy, and barium studies within the previous 14 days. Ethical approval was obtained from the Joint University College London/ University College London Hospitals Committee on the Ethics of Human Research. Written informed consent was obtained from all patients.

Technique MCTC examinations were performed according to standard protocol. A multislice CT scanner (Somatom Volume Zoom 4, Siemens, Germany) was used. The day before scanning, patients underwent standard bowel cleansing with two sachets of Picolax (Sodium Picosulphate, Ferring Pharmaceuticals, Berkshire, United Kingdom). All patients received 20 mg of intravenous Buscopan® (Hyoscine-N-butylbromide, Boehringer Ingelheim, Spain) to induce bowel paralTable 1. Symptoms on Presentation in 80 Patients Symptoms Change of bowel habit Rectal bleeding Abdominal pain Loss of weight Rectal mass Figures are number and (percentage).

Patients 46 (58) 38 (48) 24 (30) 9 (11) 19 (24)

1385

ysis to obtain better colonic distention. The patient was then placed in the left lateral position, a soft rubber tube was introduced into the rectum, and the colon was gently insufflated with room air to the maximum level tolerated by the patient. After insufflation, a standard CT scout film of the abdomen and pelvis was acquired to assess the degree of colonic distention, and additional air was insufflated as required. Omnipaque™ (Iohexol, Amersham, United Kingdom) intravenous contrast, 100 to 130 ml, based on body weight was administered before data acquisition. Images were acquired in a single breath hold usually lasting 40 seconds using 1-mm collimation, with variable table speed at 120 to 200 mA and 120 kV, and a matrix of 512 ⫻ 512. Datasets were acquired in both supine and prone positions. The 1-mm dataset was used for data analysis. All 80 MCTC examinations were reviewed on a Siemens workstation equipped with software capable of multiplanar and three-dimensional manipulations by two gastrointestinal radiologists. A time-efficient, two-dimensional, multiplanar image display technique with three-dimensional endoscopic reconstructions was used only to characterize lesions further if some doubt existed. The exact position of lesions on MCTC was identified by using a combination of abdominal topograms and multiplanar reformatted images. The average time to report a MCTC study was 15 minutes, which included image manipulation time. A final report for each examination was reached by consensus by the two radiologists. Colonoscopy was performed using standard endoscopy equipment (Olympus CF 230 liters, Olympus Optical Co., Tokyo, Japan; Pentax EC 3830 LK, Asahi Optical Co. Ltd., Tokyo, Japan). The location of each lesion was determined according to recognized morphologic features of the segments of the colon.

Statistical Analysis The findings at colonoscopy were regarded as the criterion against which the results of MCTC were compared. Lesions identified on both MCTC and colonoscopy were considered to be true-positive, otherwise false-negative if not detected on MCTC. Polyps seen on MCTC but not on colonoscopy were considered to be false-positive. Polyp detection on MCTC was analyzed per polyp and per patient. Per-polyp assessment was based on polyp size and location. In perpatient assessment, a result was considered to be true-positive only if at least one polyp seen on MCTC

MUNIKRISHNAN ET AL

1386

was matched on colonoscopy. All other results were considered to be false-positive. Sensitivity, specificity, positive predictive values (PPV), and negative predictive values (NPV) were then determined (Table 2).

RESULTS Colonoscopy was incomplete in 18 (22 percent) patients because of poor bowel preparation in 7, technical difficulty (i.e., looping) in 4, occlusive colorectal cancer in 5, and sigmoid diverticular strictures in 2. MCTC was unsuccessful in four (5 percent) patients: three because of fecal residue and inability to retain air in one. There were no complications from either investigation. Minor side effects included abdominal pain in 16 (20 percent) patients and abdominal discomfort in 19 (24 percent) patients after colonoscopy, and abdominal discomfort in 8 (10 percent) patients after MCTC. Colonoscopy was normal in 26 patients and detected 29 colorectal carcinomas and 33 polyps in 35 patients, diverticulosis in 16 patients, and colitis in 3 patients. Of 29 histologically confirmed colorectal cancers seen on colonoscopy, 4 were located in the cecum, 1 in the ascending colon, 1 in the hepatic flexure, 1 in the transverse colon, 1 in the descending colon, 2 in the sigmoid, and 19 in the rectum. MCTC was normal in 34 patients. MCTC identified 28 of 29 colorectal carcinomas (Fig. 1) and missed a sigmoid carcinoma associated with a diverticular stricture. MCTC overcalled one lesion in a patient who had considerable fecal residue. The overall sensitivity and specificity for detection of colorectal cancer was 97 and 98 percent, respectively, with PPV of 96 percent and NPV of 98 percent (Table 3). Of 33 polyps seen on colonoscopy, 12 (36 percent) were ⱖ10 mm in diameter, 6 (18 percent) were 6 to 9

Dis Colon Rectum, October 2003

mm, and 15 (46 percent) were ⱕ5 mm. Thirty polyps were removed and examined histologically, two were destroyed by diathermy, and one was not retrieved. Of 30 polyps examined histologically, 27 were tubulovillous adenomas and 3 were hyperplastic. MCTC identified 25 of 33 colonic polyps seen on colonoscopy. MCTC was false-negative for eight polyps, of which one was 8 mm and the remaining seven polyps were ⱕ5 mm. MCTC was false-positive for an 8-mm polyp and seven polyps ⱕ5 mm. One 15-mm rectosigmoid polyp proximal to a rectal cancer was missed by colonoscopy but was diagnosed on MCTC and removed at operation. This lesion was not included in the polyp-detection analysis. By direct polyp matching, the sensitivity of CTC for polyp detection was 76 percent for all lesions. The sensitivity for detection of polyps ⱖ10 mm was 100 percent (12/12) and 83 percent for polyps sized 6 mm

Figure 1. Axial multislice CT colonography image showing a rectal cancer (white arrow) and a synchronous polyp (black arrow) in the sigmoid colon.

Table 2. Statistical Analysis Sensitivity ⫽ Specificity ⫽ PPV ⫽ NPV ⫽

true positive true positive ⫹ false negative true negative true negative ⫹ false positive

⫻ 100 ⫻ 100

sensitivity ⫻ prevalence 共sensitivity ⫻ prevalence兲 ⫹ 共100 ⫺ specificity兲 ⫻ 共100 ⫺ prevalence兲 specificity ⫻ 共100 ⫺ prevalence兲 共specificity ⫻ 关100 ⫺ prevalence兴兲 ⫹ 共100 ⫺ sensitivity兲 ⫻ 共prevalence兲

PPV ⫽ positive predictive value; NPV ⫽ negative predictive value.

Vol. 46, No. 10

MULTISLICE CT COLONOGRAPHY

1387

Table 3. Analysis of all Pathology Detected Pathology

Colonoscopy

MCTC

MCTC Sensitivity (%)

MCTC Specificity (%)

PPV (%)

NPV (%)

Cancers 29 28 97 98 96 98 Polyps 33 25 74 96 88 90 Diverticulosis 16 5 31 98 79 85 Colitis 3 2 67 100 100 99 MCTC ⫽ multislice CT colonography; PPV ⫽ positive predictive values; NPV ⫽ negative predictive values. Table 4. Analysis of all Polyps Detected on MCTC Polyps

Colonoscopy

MCTC

Detection Sensitivity by Polyp (%)

Detection Sensitivity by Patient (%)

Detection Specificity by Patient (%)

76 100 83 53

74 100 80 36

96 100 99 97

All polyps 33 25 ⱖ10 mm 12 12 6 –9 mm 6 5 (1*) ⱕ5 mm 15 8 (7*) MCTC ⫽ multislice CT colonography. * False positives on MCTC.

to 9 mm (5/6). The detection rates fell to 53 percent (8/15) for polyps sized ⱕ5 mm. Analyzing the polyp detection further using per-patient detection, the sensitivity and specificity was 100 percent for polyps ⱖ10 mm. The sensitivity and specificity of 6-mm to 9-mm polyps was 80 and 99 percent, respectively, whereas for polyps ⱕ5 mm, the sensitivity and specificity was 36 and 97 percent, respectively. In per-patient detection analysis, the overall sensitivity and specificity for polyp detection was 74 and 96 percent, respectively, with PPV of 88 percent and NPV of 90 percent for all polyps (Table 4). MCTC identified 5 of 16 patients with diverticular disease of the colon with a sensitivity and specificity of 31 and 98 percent, respectively. Another six patients were noted to have diverticular thickening of the colon. MCTC also identified two of three patients with segmental colitis with a sensitivity and specificity of 67 and 100 percent, respectively (Table 2). In ten (13 percent) patients (6 males; median age, 77 (70–83) years), MCTC detected extracolonic pathology that altered management, including five patients with colorectal liver metastasis, one with primary hepatocellular carcinoma, two with ovarian tumors, and two with abdominal aortic aneurysms. In 15 (19 percent) patients, there were incidental findings, including 1 adrenal adenoma, 1 hemangioma of the liver, 7 patients with simple liver cysts, 3 with

simple renal cysts, and 3 patients with gallstones. None of these patients required further investigations. MCTC as a diagnostic investigation in this cohort of patients with colonoscopy as the criterion was accurate in 69 (86 percent) patients, false-positive in 3 (3 percent), and false-negative in 9 (11 percent) patients. The overall sensitivity and specificity of MCTC in the detection of colorectal disease were 82 and 93 percent, respectively, with PPV of 95 percent and NPV of 74 percent.

DISCUSSION CTC is a new diagnostic tool for the investigation of colorectal disease. Early results from several clinical studies using helical CT scanners have reported similar accuracy to colonoscopy in the detection of colorectal cancers and sinister polyps, i.e., ⬎10 mm in size. The recent introduction of multislice technology to CTC may have an impact on polyp detection as multislice scanners produce thinner CT slices and hence theoretically should define smaller lesions better. MCTC is currently being evaluated in many centers in North America and Europe, and the results are awaited. The speed of multislice CT is six times faster compared with single-slice helical scanners and produces 1-mm thin slices of the entire abdomen in a single

1388

MUNIKRISHNAN ET AL

breath hold. Thus, a 1-mm collimation resulting in a dataset of approximately 800 slices poses a challenge to postprocessing hardware and software. Rogalla et al.,9 in a study of 23 patients with 56 endoscopically proven polyps evaluating multislice and spiral CT in polyp detection, used the raw data four times in the reconstruction of 50 percent overlapping images with 1-mm, 2-mm, 3-mm, and 5-mm slice thicknesses. The sensitivity of detection of polyps ⬍10 mm was highest with 1-mm and lowest with 5-mm slice thickness. With large polyps (ⱖ10 mm), the performance of spiral CT with 5-mm slice thickness was comparable to multislice CT. In a pilot study in this institution,10 there was no difference between 1-mm and 3-mm slice thickness for three-dimensional virtual colonoscopy, but the image resolution was better with 1-mm slice thickness for multiplanar reconstructions. A clean bowel is essential for both colonoscopy and MCTC examinations; whereas excessive residual fluid blurs the view and prevents the advancement of the colonoscope, this does not impede MCTC. The use of dual positioning for MCTC shifts any residual fluid, allowing a complete view of the air-distended colon. Intravenous contrast enhancement helps to differentiate residual fluid from colonic pathology and carcinoma from inflammatory change particularly in the presence of diverticular disease and better defines extracolonic disease. Morrin et al.11 and Morrin and Raptopoulos12 have shown that intravenous contrast improves the detection of colorectal polyps on CT colonography. Intravenous contrast is safe with a reported mortality of 1 in 100,000; the additional time and cost for its use is negligible. Amin and colleagues13–16 from this institution were among the first to report on the high sensitivity of CTC in the detection and staging of colorectal cancers. In this study MCTC detected all colorectal cancers with a sensitivity of 97 percent, specificity of 98 percent, PPV of 96 percent, and NPV of 98 percent, although the appearances were misinterpreted in a diverticular stricture and in the presence of fecal residue, a difficulty encountered in all imaging modalities. However, MCTC overcame the difficulty of outlining the colon proximal to an obstructing carcinoma. In six (20 percent) patients in whom colonoscopy failed because of an obstructing lesion, MCTC successfully defined the proximal colon in all and identified one synchronous adenomatous polyp. Fenlon et al.17 examining 29 patients with distal occlusive colorectal cancers by virtual colonoscopy demonstrated the proximal colon in 26 (90 percent) and was able to

Dis Colon Rectum, October 2003

identify all 29 cancers: 2 synchronous cancers and 24 polyps. The two cancers were confirmed intraoperatively, and postoperatively conventional colonoscopy in 12 patients confirmed 16 polyps and 2 other subcentimeter polyps missed by virtual colonoscopy. In another study CTC identified all colorectal cancers with a staging accuracy of 81 percent and was superior to barium enema in visualizing the colon proximal to obstructing cancers.18 Macari et al.19 similarly found virtual colonoscopy more effective than barium enema in completing examination of the colon in 20 patients when colonoscopy had failed. Several clinical trials evaluating helical CTC in the detection of colorectal polyps have reported a sensitivity and specificity approaching 90 percent for polyps ⬎10 mm, although the sensitivity fell to 50 percent or less for polyps ⱕ5 mm. In a blinded trial, Hara et al.14 prospectively examined 70 patients at high risk and reported on a patient basis a detection sensitivity and specificity of 75 and 90 percent in patients with adenomas ⱖ10 mm, 66 and 63 percent in patients with adenomas ⱖ5 mm, and 45 and 80 percent for patients with adenomas ⱕ5 mm in size, respectively. Yee et al.20 have recently reported in a series of 300 patients, an overall sensitivity and specificity for polyp detection of 90 and 72 percent, respectively. By using direct polyp matching, the overall sensitivity was 70 percent. The sensitivity was 90 percent (74 of 82) for polyps ⱖ10 mm, 80 percent (113 of 141) for polyps 5.0 to 9.9 mm, and 59 percent (178 of 301) for polyps ⱕ5 mm. Fenlon and colleagues16 in a prospective study of 100 patients at high risk for colorectal neoplasia using the more sensitive per-polyp analysis reported an overall sensitivity of 71 percent for detection of polyps, with a sensitivity of 91 percent for polyps ⱖ10 mm, 82 percent for polyps 6 to 9 mm, and 55 percent for polyps ⱕ5 mm. These remain the best results published using spiral CTC and are matched by the results in this study. Wessling et al.21 in a smaller study of 40 patients using MCTC reported a detection sensitivity of 100 percent for polyps ⱖ10 mm (3 of 3) and 86 percent for polyps sized 5.1 to 9.9 mm (6 of 7). Interestingly, this group reported a sensitivity of 77 percent (14 of 20) for polyps ⱕ5 mm that has not been observed in this or other studies. More recently, Macari and colleagues22 have reported their experience with MCTC in 105 patients that was particularly poor in the detection of polyps ⱕ5 mm. MCTC detected 13 of 14 polyps (93 percent) ⬎10 mm in diameter, 19 of 27 polyps (70 percent) measuring 6 to 9 mm, and 11 of 91 polyps (12 percent) ⬍5 mm.

Vol. 46, No. 10

MULTISLICE CT COLONOGRAPHY

Other clinical trials corroborate the observation that the sensitivity and specificity of polyp detection fall as polyp size becomes smaller.15,23 Polyps ⬍5 mm are commonly hyperplastic and are not as easily detectable as adenomas of same size, probably because of their soft consistency, and are easily flattened out in an air-distended colon.16 The uncommon flat adenomas seen on colonoscopy may be difficult to define on CTC and constitute a risk, because they can show more aggressive behavior. One of the major advantages of CTC is its ability to evaluate extra colonic organs in the abdomen and pelvis, which is particularly relevant in staging of colorectal cancers. This avoids the need for CT or ultrasound scan to image the liver for metastasis if the colon is investigated by colonoscopy or barium enema. Other than the patient’s convenience, the cost benefit of CTC also should be considered. When CTC is performed as a screening or a diagnostic test, it usually targets the older population aged 50 years or older, which allows an opportunity for identifying extra colonic pathology. Hara et al.,24 in 264 consecutive patients undergoing CTC, reported extra colonic pathology in more than one-half of all patients, of which 11 percent required further investigation and treatment. They concluded that additional work-up of extracolonic CT colonography findings was relatively infrequent, but when indicated often was worthwhile. In this study, MCTC identified in 13 percent of patients (median age, 77 years) extracolonic pathology that influenced treatment and in 19 percent of patients incidental findings that were of no clinical relevance. None of these patients required routine abdominal CT or other investigations to confirm MCTC diagnosis, because the appearances were sufficiently convincing. MCTC ability to detect extracolonic disease is particularly useful in the older population, because multiplicity of abdominal or pelvic disease increases with age. As this technique becomes more acceptable, its cost effectiveness will have to be addressed. Although cost analysis was not included in this study, the estimated cost of a MCTC study in this institution is $180 compared with $570 for colonoscopy. The efficacy of MCTC in the detection of colorectal cancers and polyps (ⱖ6 mm in size) is comparable to colonoscopy and is safe with no complications. MCTC defines local and distant intra-abdominal disease spread and outlines the whole length of the colon when conventional diagnostic modalities fail because of obstructing carcinomas. Diagnosis of extracolonic

1389

pathology is an added advantage in the symptomatic patient.

REFERENCES 1. Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1997. CA Cancer J Clin 1997;47:5–27. 2. Jensen OM, Esteve J, Moller H, Renard H. Cancer in the European community and its member states. Eur J Cancer 1990;26:1167–256. 3. Silverberg E, Boring CC, Squires TS. Cancer statistics, 1990. CA Cancer J Clin 1990;40:9 –26. 4. Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics, 2000. CA Cancer J Clin 2000;50:7–33. 5. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Workgroup. N Engl J Med 1993; 329:1977– 81. 6. Muller AD, Sonnenberg A. Prevention of colorectal cancer by flexible endoscopy and polypectomy. A casecontrol study of 32,702 veterans. Ann Intern Med 1995; 123:904 –10. 7. Bond JH. Polyp guideline: diagnosis, treatment, and surveillance for patients with nonfamilial colorectal polyps. The Practice Parameters Committee of the American College of Gastroenterology. Ann Intern Med 1993; 119:836 – 43. 8. Vining OJ, Gelfand DW, Bechtold RE, Scharling ES, Grishaw EK, Shifrin RY. Technical feasibility of colon imaging with helical CT and virtual reality [abstract]. AJR Am J Roentgenol 1994;162(Suppl):104. 9. Rogalla P, Meiri N, Ruckert JC, Hamm B. Colonography using multislice CT. Eur J Radiol 2000;36:81–5. 10. Gillams AR, Munikrishnan V, Lees WR. Is 1 mm collimation (effective slice width 1.25 mm) essential for multislice CT colonography. Proceedings of the European Congress of Radiology, Vienna, Austria, March 1 to 5, 2002. Eur Radiol 2002;12 (Suppl):1001A. 11. Morrin MM, Farrell RJ, Kruskal JB, Reynolds K, McGee JB, Raptopoulos V. Utility of intravenously administered contrast material at CT colonography. Radiology 2000; 217:765–71. 12. Morrin MM, Raptopoulos V. Contrast-enhanced CT colonography. Semin Ultrasound CT MR 2001;22: 420 – 4. 13. Amin Z, Boulos PB, Lees WR. Technical report: spiral CT pneumocolon for suspected colonic neoplasms. Clin Radiol 1996;51:56 – 61. 14. Hara AK, Johnson CD, Reed JE, et al. Detection of colorectal polyps with CT colonography: initial assessment of sensitivity and specificity. Radiology 1997;205: 59 – 65. 15. Royster AP, Fenlon HM, Clarke PD, Nunes DP, Ferrucci JT. CT colonoscopy of colorectal neoplasms: twodimensional and three-dimensional virtual-reality tech-

1390

16.

17.

18.

19.

MUNIKRISHNAN ET AL

niques with colonoscopic correlation. AJR Am J Roentgenol 1997;169:1237– 42. Fenlon HM, Nunes DP, Schroy PC III, Barish MA, Clarke PD, Ferrucci JT. A comparison of virtual and conventional colonoscopy for the detection of colorectal polyps. N Engl J Med 1999;341:1496 –503. Fenlon HM, McAneny DB, Nunes DP, Clarke PD, Ferrucci JT. Occlusive colon carcinoma: virtual colonoscopy in the preoperative evaluation of the proximal colon. Radiology 1999;210:423– 8. Morrin MM, Farrell RJ, Raptopoulos V, McGee JB, Bleday R, Kruskal JB. Role of virtual computed tomographic colonography in patients with colorectal cancers and obstructing colorectal lesions. Dis Colon Rectum 2000;43:303–11. Macari M, Berman P, Dicker M, Milano A, Megibow AJ. Usefulness of CT colonography in patients with incomplete colonoscopy. AJR Am J Roentgenol 1999;173: 561– 4.

Dis Colon Rectum, October 2003

20. Yee J, Akerkar GA, Hung RK, Steinauer-Gebauer AM, Wall SD, McQuaid KR. Colorectal neoplasia: performance characteristics of CT colonography for detection in 300 patients. Radiology 2001;219:685–92. 21. Wessling J, Fischbach R, Domagk D, Lugering N, Neumann E, Heindel W. Colorectal polyps: detection with multi-slice CT colonography. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2001;173:1069 –71. 22. Macari M Bini EJ, Xue X, et al. Colorectal neoplasms: prospective comparison of thin-section low-dose multi-detector row CT colonography and conventional colonoscopy for detection. Radiology 2002; 224:383–92. 23. Fletcher JG, Johnson CD, Welch TJ, et al. Optimization of CT colonography technique: prospective trial in 180 patients. Radiology 2000;216:704 –11. 24. Hara AK, Johnson CD, MacCarty RL, Welch TJ. Incidental extracolonic findings at CT colonography. Radiology 2000;215:353–7.