Is computed tomographic colonography ("virtual colonoscopy") as good as other methods for identifying polyps?

Computed tomographic (CT) colonography has a reasonable sensitivity as compared with colonoscopy (82%) for ruling out large (>10 mm) polyps, but its sensitivity is lower for smaller polyps. A 3-dimensional analysis of the CT was not better than a 2-dimensional CT. Colonoscopy is still the best available test for identifying polyps of any size. (LOE = 1a-)

Rosman AS, Korsten MA. Meta-analysis comparing CT colonography, air contrast barium enema, and colonoscopy. Am J Med 2007;120:203-210.  [PMID:17349438]

Meta-analysis (other)

Government

Various (meta-analysis)

The investigators conducting this analysis searched a single database (Medline) to identify studies comparing CT colonography with another test, usually colonoscopy, to determine its ability to identify polyps. Although they identified 30 studies, there may be additional studies available that they didn't identify. They did not evaluate the quality of the studies they identified, so we are left to assume they were performed to minimize bias. Four of the studies evaluated 3-dimensional reconstruction of the CT ("fly-through"), and 3 of these studies compared it with 2-dimensional CT. The sensitivity of CT colonography was highest for polyps larger than 10 mm, an average 82% (95% CI, 76%-88%). Sensitivity was less for identifying polyps 6 mm to 10 mm (63%; 52% - 75%) and 0 mm to 5 mm (56%; 42%-70%). There was no difference between the 2-dimensional and the 3-dimensional CT. The area under the curve (AUC) for a receiver operating curve (ROC) provides a single numerical estimate of a test's accuracy. The closer the AUC is to 1, the more accurate; a test with an AUC of 0.5 is not at all helpful. The AUC for colonoscopy of polyps larger than 10mm was 0.999; it was the same for polyps larger than 5 mm. Although colonoscopy was used as a blinded gold standard in some studies, this value is probably overinflated. CT colonoscopy had an AUC for polyps larger than 10 mm of approximately 0.9, and an AUC for polyps 5 mm to 10 mm of approximately 0.88. There were not enough studies to draw a ROC for barium enema; the results of the 2 studies, however, found showed that this method is inferior to both CT and optic colonoscopy.