Two Colonoscopies Better at Predicting Future Polyp Risk

Results from two colonoscopies three years apart gave better information about whether a high-risk polyp would be found on a third exam than results from the second test alone.

Even if a second colonoscopy, done three years after the first, showed no adenomas at all, 8 in 100 study participants with high-risk polyps on their first exam had developed a high-risk polyp by six years when they had a third colonoscopy.

In the Aspirin/Folate Colorectal Polyp Prevention Study, 564 participants had three colonoscopies:  one at the beginning of the study, one three years later, and a third three years after that.  All of the people in the study had at least one polyp at the beginning of the study, but not all had polyps considered to be high-risk.

High risk patients had an adenoma 1 centimeter or larger, three or more adenomas, or a polyp with advanced pre-cancerous changes.

  • Patients whose first colonoscopy had high risk polyps but had no adenomas at all on a second study, had a 7.7 percent chance that their third exam would reveal a high-risk polyp.
  • Those with initial high-risk adenomas and a low-risk second exam had a 1 in 10 chance of a high-risk third exam.
  • Those who were low-risk on both a first and second colonoscopy had a less than 5 percent chance of having a high-risk polyp found during the third exam.
  • Patients were were high risk at their second exam had a 1 in 5 (20 percent) chance that an advanced polyp would be found during their third test.  For these patients, it made no difference whether they were high or low risk with their first colonoscopy.

Douglas J. Robertson, MD, MPH, and colleagues concluded,

Information from 2 previous examinations may help identify low-risk populations that benefit little from intense surveillance. Surveillance guidelines might be tailored in selected patients to use information from 2 previous examinations, not just the most recent one.

SOURCE: Robertson et al., Annals of Internal Medicine, Volume 151, Issue 2, July 21, 2009.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>