Does acetylsalicylic acid or warfarin affect the accuracy of fecal occult blood tests?. Issue 6 (22nd May 2013)

• Record Type: Journal Article
• Authors: Gandhi, Sumeet; Narula, Neeraj; Gandhi, Samir; Marshall, John K; Farkouh, Michael E
• Is Part Of: Journal of gastroenterology and hepatology. Volume 28:Issue 6(2013:Jun.)
• Page Start: 931
• Page End: 936
• Publication Date: 2013-05-22
• Subjects: Gastroenterology -- Periodicals; Digestive organs -- Diseases -- Periodicals; Liver -- Diseases -- Periodicals; Gastroenterology -- Periodicals; Liver Diseases -- Periodicals; 616.33
• DOI: 10.1111/jgh.12201
• Languages: English
• ISSNs: 0815-9319

Abstract

<abstract abstract-type="main"> <title>Abstract</title> <sec id="jgh12201-sec-0001" sec-type="section"> <title>Background</title> <p>Current guidelines for screening of colorectal cancer do not offer specific recommendations for cessation of antithrombotic agents prior to fecal occult blood test (FOBT).</p> </sec> <sec id="jgh12201-sec-0002" sec-type="section"> <title>Aim</title> <p>To asess the accuracy of FOBT in patients taking acetylsalicylic acid (ASA) or warfarin.</p> </sec> <sec id="jgh12201-sec-0003" sec-type="section"> <title>Methods</title> <p>A literature search was conducted for studies that investigated the accuracy of FOBT in patients taking ASA and warfarin. The primary outcome was the pooled relative risk (RR) for true positive FOBT for detecting significant colonic neoplasia in patients taking ASA or warfarin compared with controls. The secondary outcome was a pooled RR for true positive in guaiac FOBT (g‐FOBT) compared with immunochemical FOBT (i‐FOBT).</p> </sec> <sec id="jgh12201-sec-0004" sec-type="section"> <title>Results</title> <p>Five observational studies included 759 patients taking ASA and 1652 control subjects. In patients taking ASA, pooled RR for true positive FOBT was 0.82 (95% confidence interval [CI] 0.73–0.93, <italic>P</italic> = 0.0009), pooled RR for true positive g‐FOBT was 0.69 (95% CI 0.60–0.79, <italic>P</italic> < 0.0001), whereas pooled RR for true positive i‐FOBT was 1.013 (95% CI 0.81–1.30, <italic>P</italic> = 0.8182). Five observational studies included 806 patients taking warfarin and 10 338 control subjects. In patients taking warfarin, pooled RR for true positive FOBT was 1.559 (95% CI 1.349–1.801, <italic>P</italic> < 0.0001).</p> </sec> <sec id="jgh12201-sec-0005" sec-type="section"> <title>Conclusion</title> <p>The results of our meta‐analysis demonstrate that in patients taking ASA, there is a decrease in the positive predictive value (PPV) of g‐FOBT but no significant difference in the PPV of i‐FOBT compared with control subjects for detecting significant neoplasia. In patients taking warfarin, the PPV of FOBT was increased for detection of colorectal cancer compared with control subjects.</p> </sec> </abstract>