My bias is that we can probably correct a significant portion of this problem by improving colonoscopy performance. First, everyone should use split-dose bowel preparations. There are now 10 randomized, controlled trials showing that splitting the prep—giving half of it on the day of the procedure—improves the preparation in the ascending colon. Second, we need all colonoscopists to photodocument the cecum. Finally, increased awareness and perhaps special training are needed to improve detection of flat and serrated polyps.
We have a lot of information that adenoma detection is operator-dependent and varies dramatically between endoscopists. We need information about whether interval cancers are clustering among individual endoscopists, as this would provide a strong hint about whether my bias that we can fix this problem is correct. We must reduce the operator dependency of colonoscopy. It's not good when a procedure that is so important for prevention of a common cancer is operator dependent. It's a flaw in the strategy.
DR. DOUGLAS K. REX is distinguished professor of medicine at Indiana University, Indianapolis.
Vitals