• Hermann Brenner, Jenny Chang-Claude, Lina Jansen, Christoph M Seiler, and Michael Hoffmeister.
• Division of Clinical Epidemiology and Aging Research, German Cancer Research Center, Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany. h.brenner@dkfz.de
• Ann. Intern. Med. 2012 Aug 21; 157 (4): 225232225-32.

BackgroundStudies have identified characteristics of adenomas detected on colonoscopy to be predictive of adenoma recurrence.ObjectiveTo assess the role of both colonoscopy-related factors and polyp characteristics on the risk for colorectal cancer after colonoscopic polyp detection.DesignPopulation-based case-control study (3148 case participants and 3274 control participants).SettingRhine-Neckar region of Germany.PatientsCase and control participants with physician-validated detection of polyps (other than hyperplastic polyps) at a previous colonoscopy in the past 10 years.MeasurementsDetailed history and results of previous colonoscopies were obtained through interviews and medical records. Case and control participants were compared according to colonoscopy-related factors (incompleteness, poor bowel preparation, incomplete removal of all polyps, and no surveillance colonoscopy within 5 years) and polyp characteristics (≥ 1 cm, villous components or high-grade dysplasia, ≥ 3 polyps, and ≥ 1 proximal polyp). Odds ratios (ORs) and attributable fractions were derived by using multiple logistic regression and the Levin formula.Results155 case participants and 260 control participants with physician-validated polyp detection in the past 10 years were identified. The following characteristics were significantly more common among case participants than among control participants: not all polyps completely removed (29.0% vs. 9.6%; OR, 3.73 [95% CI, 2.11 to 6.60]), no surveillance colonoscopy within 5 years (26.5% vs. 11.5%; OR, 2.96 [CI, 1.70 to 5.16]), and detection of 3 or more polyps (14.2% vs. 7.3%; OR, 2.21 [CI, 1.07 to 4.54]). Odds ratios ranged from 1.12 to 1.42 and CIs included 1.00 for all other variables. Overall, 41.1% and 21.7% of cancer cases were statistically attributable to colonoscopy-related factors and polyp characteristics, respectively.LimitationThis was an observational study with potential for residual confounding and selection bias.ConclusionColonoscopy-related factors are more important than polyp characteristics for stratification of colorectal cancer risk after colonoscopic polyp detection in the community setting.

22 keywords...

hide…