Colorectal cancer ranks as the third most common human malignancy and the second leading cause of cancer-related deaths in the U.S.¹ The overall risk of developing the disease is approximately 5% over a lifetime. Around 130,000 new cases of colorectal cancer were diagnosed and 57,000 deaths were attributed to it in both 1999 and 2000.^1,2

Most colon cancer arises from adenomatous polyps, which can take five to 15 years for malignant transformation.^3,4 The risk of developing carcinoma from a polyp is directly related to its size: essentially zero risk if the polyp is less than 5 mm, 1% risk if it is between 5 and 10 mm, 10% risk with size 10 to 20 mm, and at least 30% risk with polyps larger than 20 mm.⁵

Survival rates from colon cancer are related directly to the pathologic staging of the disease and exceed 90% when cancers are limited to the bowel wall.⁶ Whereas 75% of cancers found by screening in asymptomatic patients are confined to the bowel wall (Dukes’ A and B), more than half of those with symptoms have a more advanced stage (Dukes’ C and D).

Symptoms of colon cancer, such as anemia and change in bowel habits, are neither sensitive nor specific. Studies have shown that screening can reduce colon cancer mortality.⁶ The American Cancer Society recommends that screening begin at age 50 for asymptomatic, average-risk patients, with either yearly fecal occult blood testing with flexible sigmoidoscopy or double-contrast barium enema every five years, or a colonoscopy every 10 years.⁷

Fecal occult blood testing is the safest and least expensive tool, but it detects only 30% to 40% of colorectal cancers and 10% of adenomas.⁸ Because it examines only to the junction of the descending colon and sigmoid, sigmoidoscopy does not detect 32% of advanced cancers.⁹ Even one-time screening with fecal occult blood test and sigmoidoscopy would fail to detect 24% of patients with advanced colonic neoplasia.¹⁰ Double-contrast barium enema is more time-consuming, requires a good deal of patient positioning and cooperation, and has variable reported diagnostic accuracies.¹¹

Standard optical colonoscopy can biopsy and/or remove detected polyps, and is generally considered the gold standard, although miss rates of 18% of adenomas >6 mm have been reported on back-to-back colonoscopies.¹² Additionally, it is uncomfortable, requires colon cleansing and sedation, is time-consuming and invasive, has a small risk of perforation and death (colonic perforation in one in 500 to 1000 cases and death in one in 2000 to 5000 cases),¹³ fails to demonstrate the entire colon in up to 10% of patients,¹⁴ and is ineffective in examining areas of the colon blocked by masses or in areas of severe narrowing.

Colon imaging with helical CT and virtual reality was first reported by Vining et al in 1994.¹⁵ Abdominal axial images of the distended colon are taken in seconds during breath-holding and are used to create two-dimensional multiplanar reformation (MPR) or three-dimensional images of the colon. The performance data to date have been encouraging, with sensitivity and specificity for polyps greater than or equal to1 cm ranging from 75% to 91% and 90% to 93% respectively, as reported by per-polyp comparisons.^16,17 While most published reports include patients who were either symptomatic or highly suspected to have polyps, the use of virtual colonography for screening is supported in a recent study by Yee et al that showed similar performance characteristics in asymptomatic, average-risk patients as compared with high-risk, symptomatic patients.¹⁷

Patient Preparation

Adequate bowel preparation is necessary for virtual colonoscopy prior to obtaining CT images. Current approaches are similar to optical colonoscopy and involve mainly polyethylene glycol-electrolyte lavage solutions (PEG-ELS) that are rapid, safe, and effective. Many patients find it difficult, however, to drink the large volumes required over a two-day period, and they may experience side effects of nausea, vomiting, abdominal fullness, cramps, anal irritation, and sleep loss leading to poor compliance. A small-volume solution of oral sodium phosphate provides substantially less residual fluid in the colon, is better tolerated, is more likely to be completed, and could be given the day before the procedure, although there is concern about potential hemodynamic and electrolyte abnormalities.^18,19

Magnesium citrate oral solution may be an equally effective, potentially safer agent, and it can be combined with bisacodyl tablets and suppository for a less vigorous laxative program (LoSo Prep, E-Z-Em). The additional use of a modified low-residue diet to help satisfy hunger and cleanse the colon would further increase patient compliance.

A preferred approach to physical cleansing uses oral contrast agents to increase the density of the remaining material, which can then be automatically identified and removed by computer segmentation techniques. Barium sulfate, given with meals the day before the virtual colonoscopy, can be incorporated into residual stool, while iodinated contrast (Gastroview, Mallinckrodt) will label the remaining fluid. This approach may make it possible to completely eliminate all physical bowel cleansing.²⁰

The use of spasmolytic agents to prevent segmental collapse and spasm is debatable, although studies have reported no significant benefit for the routine use of glucagon.²¹

The examination begins by inflating the colon with approximately 1 to 2 liters of either air or carbon dioxide, introduced through a small rectal tube. Carbon dioxide, which is better tolerated by the patient, can be delivered through a handheld insufflator bulb or laparoscopic-type pressure-sensitive device.

CT scout views are used to assess colon distension. During a single 35 to 40-second breath-hold, continuous CT images with 100 mA or less are obtained, usually using the following parameters: single-slice CT, 5-mm collimation, 1.7 to 2 pitch, 1 to 3-mm reconstruction; or multislice CT, 2.5-mm collimation with 1.25-mm reconstruction, and an equivalent pitch of 6, yielding as many as 300 to 400 images for each sequence. Images are obtained in both supine and prone positions to improve distension and to move residual fluid, significantly improving performance in polyp detection.²²

Image Display

Two approaches to colon evaluation are possible, depending on whether 2-D or 3-D images are used for primary interpretation. CT colonography uses 2-D images, familiar to all radiologists, as the means of initial interpretation. On a computer workstation such as Navigator (GE Medical Systems), Virtuoso (Siemens Medical Systems), Vitrea 2 (Vital Images), Plug n View 3D (Voxar), and AccuView (Accuimage), scrolling up and down the axial images at lung window settings can sequentially evaluate the air-filled colon for polyps (Figure 1). Zoomed axial images or reconstructed views perpendicular to the colon surface can also be used. Improved detection and characterization of suspected lesions occurs with 2-D MPR images to simultaneously cross-reference one plane to the other two planes. Soft-tissue windows are reserved to evaluate areas of colonic collapse, wall thickening, pericolonic soft-tissue stranding, and extracolonic findings.²³ Narrow windows can be used for tissue characterization to detect fat within a lipoma, or air and retained barium within stool.

Because obtaining 3-D images is often cumbersome and time-consuming, they are reserved to confirm or further characterize 2-D findings, often to differentiate small polyps from folds. For technical reasons, these 3-D endoluminal views are frequently of only a small segment of the colon (Figure 2A), or even a small cubed section that can be rotated in different planes (Figure 2B). To reduce interpretation time with these particular systems, complete navigation throughout the colon is often not done. As a result, it is suggested that the radiologist primarily rely on the axial 2-D images for adequate interpretation of CT colonography, and use the supplemental 3-D images only for problem-solving.^24,25

A second approach to colon evaluation, known as virtual colonoscopy, uses a computer visualization system for image segmentation to construct a clean colon model and computer graphics to virtually navigate through the 3-D colon model. The V3D Colon Module (Viatronix) uses this approach, enabling both automatic and interactive endoluminal navigation as the primary method to evaluate the colon surface.^26-28

A colon model is created after electronic bowel cleansing in which oral contrast and opacified residual fluid/stool are removed. The fully automatic system segments the colon and generates its centerline to reduce the time needed for “setting up” the CT images for viewing and analysis (Figure 3).29 Guided navigation smoothly follows the centerline, like a flight path, toward the end of the colon. The direction of navigation can be selected and the current distance from the rectum displayed for all points along the centerline.

With real-time volume rendering at more than 10 frames/sec, interactive navigation of the endoscopic views can analyze the inside of the colon-even behind haustral folds-and obtain accurate 3-D measurements of suspicious structures. 2-D views corresponding to the navigation viewpoint are available for verification purposes. The specific image within any plane is easily correlated to the endoscopic view (Figure 4).

By turning, zooming, and rotating during the “fly-through” examination of the endoscopic views, the radiologist can view the entire colon surface instead of examining only a movie that can be played in forward and reverse. The visualized colon surface can be marked and measured during guided fly-through navigation (Figure 5). Areas of the surface not directly viewed by the observer are presented for direct evaluation.³⁰ Suspected abnormalities could be further evaluated using translucent rendering, a semitransparent view beneath the surface that represents structures of various densities in different colors.

Extracolonic Findings

Using both CT colonography and virtual colonoscopy, the colon wall itself, as well as extracolonic tissues, can be evaluated on 2-D images similar to standard CT exams. This is a unique benefit compared with other colorectal cancer screening techniques. In a group of patients with a high risk for polyps, one-half had some type of extracolonic abnormality, while 11% had highly clinically important findings, including lung nodules, abdominal aortic aneurysm, renal adenocarcinoma, and bowel containing inguinal hernia. Additional workup was often worthwhile and did not substantially increase the examination cost per patient.³¹

More Realistic Virtual Exam

Many of the early recommendations for examining virtual colonoscopy studies were based on workstations that could not easily perform 3-D endoluminal evaluation, making the reliance on axial images the only feasible approach. Researchers at Stanford University have shown, however, that if all surfaces of the colon lumen have been seen, 3-D endoscopic navigation has a higher sensitivity for polyp detection than 2-D axial images,³² allowing complete viewing of the entire colon surface, even behind haustral folds. Improved characterization of individual cases has been shown with 3-D display techniques.³³

Even if it is still preferable to initially evaluate the 2-D images, there is now a seamless, simultaneous correlation with the endoluminal views (Figure 6). Unlike the 2-D approach, it is also feasible to internally view within suspected abnormalities to help distinguish residual stool balls containing barium from true polyps, which appear similar during endoscopic evaluation, and possibly to differentiate adenomas from hyperplastic polyps (Figure 7). Rather than paging through hundreds of axial CT images to find small polyps that may be difficult to distinguish from colonic folds, a more realistic virtual endoscopic examination of the colon is possible. Only now can a legitimate comparison of the two different approaches to colon evaluation be performed.

References

1. Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics, 2000. Ca Cancer J Clin 2000; 50:7-33.
2. Winawer SJ, Fletcher RH, Miller L, et al. Colorectal cancer screening: clinical guidelines and rational. Gastroenterology 1997;112:594-642.
3. Hill MJ, Morrison BC, Bussey HR. Etiology of adenoma carcinoma sequence in the large bowel. Lancet 1978;1:245-247.
4. Stryker S, Wolff B, Culp C, et al. Natural history of untreated colonic polyps. Gastroenterology 1987;93:1009-1013.
5. O’Brien M, Winawer SJ, Zauber A. The national polyp study: patient and polyp characteristics associated with high-grade dysplasia in colorectal adenomas. Gastroenterology 1990;98:371-379.
6. Mandel JS, Bond JH, Church TR, Snover D. Reducing mortality from colorectal cancer by screening with FOBT. NEJM 1993;328:1365-1371.
7. 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 1997;47:154-160.
8. Gazelle GS, McMahon PM, Scholz FJ. Screening for colorectal cancer. Radiology 2000;215:327-335.
9. Imperiale TF, Wagner DR, Lin CY, et al. Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colon findings. NEJM 2000;343:169-174.
10. Lieberman DA, Weiss DG. One-time screening for colorectal cancer with combined fecal occult-blood testing and examination of the distal colon. NEJM 2001;345:555-560.
11. Kewenter J, Breringe H, Engaras B, Haglind E. The value of flexible sigmoidoscopy and double-contrast barium enema in the diagnosis of neoplasms in the rectum and colon in subjects with a positive hemoccult: results of 1831 rectosigmoidoscopies and double-contrast enemas. Endoscopy 1995;27:159-163.
12. Rex, DK, Cutler CS, Lemmel GT, et al. Colonoscopic miss rates of adenomas determined by back-to back colonoscopies. Gastroenterology 1997;112:24-28.
13. Orsoni P, Berdah S, Verrier C, et al. Colonic perforation due to colonoscopy: a retrospective study of 48 cases. Endoscopy 1997;29:160-164.
14. Anderson ML, Heigh RI, McCoy GA, et al. Accuracy of assessment of the extent of examination by experienced colonoscopists. Gastrointest Endoscopy 119;38:560-563.
15. Vining DJ, Gelfand DW, Bechtold RE, et al. Technical feasibility of colon imaging with helical CT and virtual reality (abstr). Ann J Roentgenol 1994;162:104.
16. Hara A, Johnson CD, Reed JE, et al. Colorectal polyp detection using CT colography: initial assessment of sensitivity and specificity. Radiology 1997;205:59-65.
17. Yee J, Akerkar GA, Hung RK, et al. Colorectal neoplasia: performance characteristics of CT colonography for detection in 300 patients. Radiology 2001;219:685-692.
18. Macari M, Lavelle M, Pedrosa I, et al. Effect of different bowel preparations on residual fluid at CT colonography. Radiology 2001; 218:274-277.
19. Golub RW, Kerner BA, Wise, WE Jr, et al. Colonoscopic bowel preparations-which one? Dis Colon Rectum 1995;38(6):594-599.
20. Lakare S, Wan M, Sato M, Kaufman AE. 3D digital cleansing using segmentation rays. Proc IEEE Visualization 2000:37-44.
21. Yee J, Hung RK, Akerkar GA, Wall SD. The usefulness of glucagon hydrochloride for colonic distension in CT colonography. AJR 1999;173:169-172.
22. Fletcher JG, Johnson CD, Welch TJ, et al. Optimization of CT colography technique: prospective trial in 180 patients. Radiology 2000;216:704-711.
23. Johnson CD, Dachman AH. CT colonography: the next colon screening examination? Radiology 2000:216:331-341.
24. Macari M, Milano A, Lavelle, et al. Comparison of time-efficient CT colography with two- and three-dimensional colonic evaluation for detecting colorectal polyps. AJR 2000;174:1543-1549.
25. Dachman AH, Kuniyoshi JK, Boyle CM, et al. CT colonography with three-dimensional problem solving for detection of colonic polyps. AJR 1998;171:989-985.
26. Viswambharan A, Wax MR, Hong L, et al. Virtual colonoscopy: 3D visualization of the mucosal surface of the colon. Radiology 1996;201(P):565.
27. Hong L, Liang Z, Viswambharan A, et al. Reconstruction and visualization of 3D models of the colonic surface. IEEE Trans Nucl Science 1997;44:1297-1302.
28. Hong L, Muraki S, Kaufman AE, et al. Virtual voyage: interactive navigation in the human colon. Proc ACM SIGGRAPH 1997;27-34.
29. Bitter I, Sato M, Bender MA, et al. Automatic, accurate and robust colon centerline algorithm. Radiology 2000;217(P):370.
30. Wax, MR, Anderson, JC, Kreeger KA, et al. Do we see virtually all of the colon? Gastroenterology 2001;120:A-601.
31. Hara AK, Johnson CD, MacCarty RL, et al. Incidental extracolonic findings at CT colonography. Radiology 2000;215:353-357.
32. Beaulieu CF, Jeffrey RB, Karadi C, et al. Display modes for CT colonography part II. Blinded comparison of axial CT and virtual endoscopic and panoramic endoscopic volume-rendered studies. Radiology 1999;212:203-212.
33. McFarland EG, Brink JA, Pilgram TK, et al. Spiral CT colonography: reader agreement and diagnostic performance with two- and three-dimensional image-display techniques. Radiology 2001;218:375-383.