Yield of Fecal Immunochemical Test in Detection of Colorectal Cancer and Advanced Neoplasia in Veteran Population at Dayton VA Medical Center

Abstract

Purpose: Fecal occult blood test (FOBT) is a widely accepted, non-invasive technique for colorectal cancer (CRC) screening. According to national VA directive, a diagnostic colonoscopy must be performed within 60 days of the positive FOBT test. Since May 2009, the Dayton VAMC has replaced the guaiac-based fecal occult blood test (gFOBT) with fecal immunochemical test (FIT) using a cut-off of 100 mg/ml hemoglobin. Aim: To compare the yield of gFOBT versus FIT in the detection of CRC and advanced neoplasia (AN) in average risk veterans at Dayton VAMC and to determine compliance with a follow-up colonoscopy within 60 days of a positive test. Methods: We performed a retrospective chart review of all patients referred for follow-up colonoscopy after a positive FOBT screening test at Dayton VAMC between February 2008 and September 2010. Results: During the study period a total of 351 FOBT (171 gFOBT and 180 FIT) positive subjects were referred for colonoscopy. A total of 65% (229/351) patients completed a colonoscopy, 71% (122/171) with positive gFOBT and 59% (107/180) with positive FIT. Median interval between positive FOBT and colonoscopy was 74 days (Range 1-846 days). Colonoscopy was completed within 60 and 90 days of a positive test in 28% (65/229) and 66% (152/229) subjects, respectively. There were 96% males with median age of 62 years (Range 38-90 years) at the time of positive FOBT. Positive colonoscopy findings (hemorrhoids, diverticuli, polyps, AVMs, mass, inflammation, ischemia) noted in 80% (97/122) gFOBT vs. 88% (94/107) FIT group (p=0.11). The prevalence of AN was 19% (23/122) in gFOBT vs. 23% (25/107) in FIT group. Prevalence of AN in single positive gFOBT was 12% (8/69) vs. 13% (8/61) in FIT group. AN in gFOBT group was located proximal and distal to the splenic flexure in 22% (5/23) and 35% (8/23), respectively. In FIT group, AN was located proximal and distal to the splenic flexure in 12% (3/25) and 32% (8/25), respectively. Synchronous proximal and distal AN was noted in 44% (10/23) gFOBT vs. 56% (14/25) FIT group. CRC was diagnosed in 4% (8/229) patients with median age of 72 years (Range: 55-85 years) at diagnosis. Six cancers were distal and two proximal to the splenic flexure. Prevalence of invasive CRC was 5% (6/122) in gFOBT vs. 2% (2/107) in FIT group (p=0.28). Prevalence of adenomas 10 mm or more in size was 27% (17/63) in gFOBT vs. 35% (21/60) in FIT group (p=0.44). Conclusion: The prevalence of AN was higher in the FIT compared to the gFOBT group, possibly indicating that the FIT test might be more sensitive for detecting earlier stages of neoplasia. Less than one third of the subjects completed a colonoscopy within 60 days of a positive test, indicating need to improve compliance with follow-up colonoscopy.