The parallel SIGGAR trials recently published in Lancet add to the growing body of literature regarding the utility of computed tomographic (CT) colonography in the detection of colorectal polyps and cancers. These papers reinforce the results seen in other large multicenter trials1-3 and echo the positive clinical experience with CT colonography over the past 10 years.
Comparing Modalities
These trials raise a number of important points. First, the performance of CT colonography is equivalent to optical colonoscopy and far superior to barium enema. At the 10-mm threshold, there was no significant difference between CT colonography and colonoscopy in rates of polyps detected within each study cohort (5.1% and 5.8% for CT colonography and colonoscopy, respectively; P = .53). In contrast, barium enema had a substantially poorer rate.
The CT colonography results become more impressive when the technical factors are taken into account. The trials recognized that CT colonography represented an evolving technology that would advance over the recruitment of the trial.4 Thus, while some institutions utilized state-of-the-art CT colonography technique, others used older or less rigorous parameters that would be considered far from optimal according to today’s standards. Besides these technical factors, the use of multiple readers with varying levels of expertise from over 20 teaching and general hospitals points to the robust nature of this imaging modality and the ability to transfer such performance to the community.
Second, extracolonic findings are an important added benefit from CT colonography. Additional diagnoses outside of the colorectum are possible due to the cross-sectional nature of this imaging technique (CT colonography is essentially a low-dose CT without intravenous contrast).
Although often cited as a possible negative consequence to the use of CT colonography, the feared high rate of additional imaging for clinically insignificant findings typically has not been seen.5,6 The SIGGAR trials also showed low rates, ranging between 7.5% and 10.1% for the two studies. For this minor additional cost, a large benefit was derived.
Combining both studies, 22 extracolonic cancers and 34 aortic aneurysms (3 cm or greater in diameter) were detected. This constituted 3.1% of the combined CT colonography populations (n = 1,810). In comparison, CT colonography detected colorectal cancers in 4.3% in this combined group. Thus, the number of important extracolonic diagnoses approaches the number of colorectal cancers despite being an incidental byproduct of a primarily colorectal examination.
Although it can be difficult to know how much improvement in survival is achieved with earlier detection of an extracolonic cancer as opposed to later, symptomatic detection, it is easy to see the large benefit of detecting an unsuspected aneurysm and treating when appropriate as opposed to discovery at symptomatic rupture. This extracolonic benefit would obviously not be possible at colonoscopy screening.
Other Colonic Investigations
On the other hand, the trials report a high rate of “additional colonic investigations” for CT colonography, ranging from 23.5% to 30%. Indeed, this result is highlighted, since this metric is a primary outcome measure of the CT colonography vs colonoscopy trial. This is a rather nonsensical comparison, since any polyp detected on CT colonography requires “additional investigation” or removal at colonoscopy, whereas any colonoscopy-detected polyp can be removed without an additional procedure. Logically, there would be a large difference, although I suspect that it is an easy way to ensure that the trial will achieve a statistically significant difference for one of its stated aims. If one more narrowly defines this measure (“additional colonic investigations”) as the rate of inadequate or nondiagnostic exams requiring additional evaluation, CT colonography actually outperforms colonoscopy, with 5.3% of CT colonography exams being inadequate and 11.3% of colonoscopies being incomplete.
Comparisons aside, the high referral rate from CT colonography to colonoscopy in this trial does pose concern if truly reflective of CT colonography use. It is discrepant with other large clinical series, in which the real-world rates have been much lower, at 8% to 10%.7,8 However, the disparate findings can be easily explained by the lack of a lower referral threshold at CT colonography in the SIGGAR trial. Most practices using CT colonography adhere to C-RADS (CT colonography reporting and data systems) guidelines and ignore polyps 5 mm or less in size.9 When a threshold is applied, referral rates are typically in the 10% range.
The National CT Colonography trial (ACRIN 6664; N = 2,531) reported that a 12% referral rate would result if a 6-mm threshold were used.1 In effect, the use of a 6-mm threshold places diminutive polyps (which are typically innocuous) in a 5-year follow up window. Long-term outcomes data of screened CT colonography patients with negative exams have shown that this is a safe and appropriate practice.10 The SIGGAR authors themselves point out that many of the referrals to colonoscopy “might be avoidable ... if referral is set at 8 mm.”
Closing Thoughts
As a final point, a previously published analysis from the SIGGAR trial reported that CT colonography is more acceptable to patients than colonoscopy.11 In practice, we have seen that the addition of CT colonography leads to improved screening participation.12 Hopefully, the SIGGAR results will provide the impetus to move this much needed technology into the community and into wider use. Given the current dismal state of colorectal cancer screening, in which nearly half of the eligible population is not adherent, it would appear that CT colonography should be added to the armamentarium of options to help eradicate this devastating but preventable cancer. ■
Disclosure: Dr. Kim is a consultant for Viatronix, cofounder of VirtuoCTC, and on the medical advisory board for Digitalartforms.
References
1. Johnson CD, et al: Accuracy of CT colonography for detection of large adenomas and cancers. N Engl J Med 359:1207-1217, 2008.
2. Regge D, et al: Diagnostic accuracy of computed tomographic colonography for the detection of advanced neoplasia in individuals at increased risk of colorectal cancer. JAMA 301:2453-2461, 2009.
3. Pickhardt PJ, et al: Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. N Engl J Med 349:2191-2200, 2003.
4. Halligan S, et al: Design of a multicentre randomized trial to evaluate CT colonography versus colonoscopy or barium enema for diagnosis of colonic cancer in older symptomatic patients. Trials 8:32, 2007.
5. Pickhardt PJ, et al: Unsuspected extracolonic findings at screening CT colonography. Radiology 249:151-159, 2008.
6. Zalis ME, et al: Diagnostic accuracy of laxative-free computed tomographic colonography for detection of adenomatous polyps in asymptomatic adults. Ann Intern Med 156:692-671, 2012.
7. Kim DH, et al: CT colonography versus colonoscopy for the detection of advanced neoplasia. N Engl J Med 357:1403-1412, 2007.
8. An S, et al: Screening CT colonography in an asymptomatic average-risk Asian population. Am J Roentgenol 191:W100-W106, 2008.
9. Zalis ME, et al: CT colonography reporting and data system. Radiology 236:3-9, 2005.
10. Kim DH, et al: Five year colorectal cancer outcomes in a large negative CT colonography screening cohort. Eur Radiol 22:1488-1494, 2012.
11. von Wagner C, Ghanouni A, Halligan S, et al: Patient acceptability and psychologic consequences of CT colonography compared with those of colonoscopy. Radiology 263:723-731, 2012.
12. Benson M, Pier J, Kraft S, et al: Optical colonoscopy and virtual colonoscopy numbers after initiation of a CT colonography program. J Gastrointest Liver Dis 21:391-395, 2012.
