THE FREQUENCY AND OUTCOMES OF FECAL OCCULT BLOOD TESTING AFTER COLONOSCOPY IN PATIENTS WITH ADENOMATOUS POLYPS

Abstract

Purpose: Colonoscopy is the primary modality used for colorectal cancer (CRC) screening in patients with prior adenomas. Though the use of annual fecal occult blood tests in concert with a screening sigmoidoscopy has been recommended for primary CRC screening, the utility of FOBTs after colonoscopy in a “high-risk” population has not been proven. Nonetheless, some providers are using FOBTs after colonoscopy. If FOBTs can improve diagnostic yield of colonoscopy, current guidelines might incorporate this strategy. If not, they increase costs without improving early detection of synchronous or metachronous lesions. This study examines the frequency and outcomes of FOBTs after colonoscopy in patients with prior adenomatous polyps. Methods: We retrospectively identified patients with adenomas found on colonoscopy between 1996 and 1999 from two large medical centers. Laboratory and endoscopy databases were then searched to identify significant endoscopic and pathologic findings, as well as FOBTs performed prior to subsequent colonoscopies. Significant findings were describes as tubulovillous/villous/high-grade histology/cancer, more than 2 adenomas or any greater than 1 cm. The analysis is descriptive with findings compared using Fisher's Exact Test and logistic regression. Results: 1,956 patients (68% male; mean age 67) with colonic adenomas were identified; 39% had FOBTs performed subsequently (49.4% single, 50.6% multiple—half sporadic, half annual). The mean number of FOBTs was 1.93 and 17.5% had at least one positive FOBT. Of these, over 80% had no significant pathology on any subsequent endoscopy. At the first subsequent endoscopy, individuals with positive FOBTs were 3-fold more likely to have a significant pathologic finding (P <0.001), but a minority of these FOBTs were performed within 6 months of the colonoscopy (and fewer than 2% had colonoscopies for FOBT positivity'). The sensitivity of FOBT prior to a colonoscopy in these high-risk patients ranged from 25–35% with a PPV of 10–25% (NPV consistently over 90%). Conclusions: The use of FOBTs after colonoscopy in patients with adenomas was higher than expected. One fifth of these patients had positive FOBTs, but these did not trigger endoscopy and had a poor sensitivity and PPV for significant pathology. Overall, FOBTs after colonoscopy, even in this high-risk population, do not improve diagnostic yield.