ABSTRACT
BACKGROUND AND AIM: Colorectal cancer (CRC) screening is based on colonoscopy or fecal occult blood tests, but is imperfect and costly. The Asia Pacific Colorectal Screening Score (APCS) is derived from age, sex, family history of CRC, and smoking history and has been validated in Asian populations. Validation in a Western population is, however, yet to be tested. METHODS: In a teaching hospital, patients underwent colonoscopy for standard indications and screening over 18 months. Data was collected on age, sex, family history of CRC, smoking, weight, ethnicity, and symptoms. Evaluation of the APCS to predict colonoscopy findings (polyps, adenoma, high risk adenoma, and CRC) was performed. RESULTS: A total of 645 patients were prospectively recruited (46.7% male, median age 57 years); 17.8% were average risk (AR), 50.9% were moderate risk (MR), and 31.3% high risk (HR) on APCS. High risk adenomas (AA) were seen in 14.9% of the HR, 5.2% MR, and 0.9% LR patients, P < 0.0001. Comparing HR and MR to AR patients demonstrated significantly elevated relative risk (RR) for AA: 17.1 (95% confidence interval [CI] 2.4-123; P = 0.0001), and adenoma 6.0 (0.80-44.3; P = 0.044). Comparing HR to MR groups for AA, the RR was 2.87 (1.62-5.06; P = 0.0001). Symptoms did not predict findings (odds ratio [OR]: 1.06 [0.75-1.48]; P = 0.75). Body mass index (BMI) <20 kg/m(2) was protective against colonic polyps (OR: 0.28, 95%CI: 0.11-0.74; P = 0.010), adenoma (0.08, 0.01-0.62; P = 0.015), and AA (perfect prediction, OR 2.35 × 10(-8)). CONCLUSIONS: APCS predicts colonic findings in a Western population, to a greater extent than in Asians, independent to symptoms. Low body weight carries a strong protective effect against colonic neoplasia.
