Objective: To determine de incidence of colonic polyps in colonoscopies performed before scheduled and to identifY the clinical and endoscopic features that predicted this finding.
Methods: All patients who underwent at least two complete colonoscopies less than three years apart were retrospectively identified in our computerized database. We excluded patients with high risk of colonic neoplasm requiring a new colonoscopy in less than three years. We analyzed the incidence of polyps before the first and third year after the first study, and the clinical and endoscopic features related to this finding by means of multivariate logistic regression.
Results: 378 paired colonoscopies fulfilled criteria, 129 were performed less than one year apart (group 1), and 249 less than three years apart (group 2). Regarding surveillance colonoscopies, 19% of patients presented adenomas and 1.5% high grade dysplasia (HGD) in group 1, and 21% presented adenomas and 2% HGD in group 2. In group 1 fair preparation (P = 0.03), and prolonged colonoscopy (P = 0.02) on the first study were independent predictors to find any polyp on the second study before scheduled. In group 2 fair cleansing (P = 0.04), history of sessile polyps (P = 0.01) and 3 or more polyps in the first study (P = 0.01) were independent predictors to find any polyp, while more than 5 polyps in the first study predicate adenomas.
Conclusion: During the first year incident polyps related to difficult procedures (missed polyps?) while at 3 years the history of previous polyps was also important.