Annual Scientific Meeting
Introduction: The American Cancer Society recommends that colorectal cancer (CRC) screening begin at 45 years. To further inform this issue, data on advanced adenoma (AA) and serrated polyp (SP) prevalence in this population are vital. We used the New Hampshire Colonoscopy Registry (NHCR) to examine age related prevalence of advanced findings in an average risk “screening” cohort < 50.
Methods: Indications for colonoscopies in adults < 50 are often diagnostic. We made the assumption that some indications (e.g. abdominal pain or constipation) might not be associated with increased risk of adenomas. To ensure an appropriate “average risk” population < 50, we included only those indications expected to yield AA prevalence rates similar to screening exams, and also tested this assumption. We excluded higher-risk indications, such as GI bleeding, occult blood, Fe deficiency anemia and abnormal imaging. We also excluded surveillance exams, family history of CRC, incomplete exams and poor bowel preparation. AAs were adenomas >1cm or with villous elements or high-grade dysplasia, or CRC. Clinically significant serrated polyps (CSSP) included large ( >1cm) SPs, any sessile SP or traditional serrated adenoma, and any proximal SP >5 mm. We present prevalence/adjusted risks for AA, adenomas and CSSP. Age was stratified into < 40 (reference), 40-44, 45-49, 50-54,55-59 and 60+ years. Covariates were sex, smoking and BMI. To test our hypothesis regarding indications, we examined the prevalence in screening versus the prevalence in included and excluded indications. We also added the presence of low-yield symptoms as a covariate.
Results: In our sample (n=42,600), AA prevalence increased significantly at 40 yrs. The prevalence of AA and CSSP was similar for ages 45-49 and 50-54. In the model, older age was associated with significantly increased risk for AA and adenomas (Table 2). Older age effect was less pronounced in CSSPs. The prevalence of AA was 4.4% for screening, 3.2% for the included diagnoses and 5.8% for the excluded diagnoses (p < 0.001), supporting our assumption regarding indication. Low-risk symptoms were not associated with an increased risk (OR=1.03: 0.84-1.13), further confirming our assumption.
Discussion: The prevalence of AA increased significantly in the 40-44 group as compared to those < 40. The increased prevalence for both AA and CSSP at age 45-49 was similar to that for those 50-54 years. These data support the suggestion that to optimize prevention, screening should begin at 45 years.