Purpose: To critically evaluate the current literature in an effort to establish the current role of radiologic imaging (computed tomography, magnetic resonance imaging, ultrasonography [US], fluoroscopy, conventional film radiography) in irritable bowel syndrome (IBS).

Materials and methods: The term "irritable bowel syndrome" was used to search Clinical Evidence, UpToDate, Cochrane Library, TRIP, and National Institute for Health and Clinical Excellence databases and the American College of Physicians Journal Club and Evidence-Based Medicine online. PubMed was searched by using medical subject headings ("irritable bowel syndrome;" "colonic diseases, functional;" "diagnosis;" "colonography;" "computed tomographic (CT)") and the dates January 1, 1985 to July 1, 2010. Appraisal was independently performed by two reviewers who followed the Oxford Centre for Evidence Based Medicine practice criteria.

Results: No systematic review (SR) specifically examined radiologic imaging in IBS; however, in the secondary literature, five relevant SRs or guidelines partially addressed this topic. A PubMed search identified 1451 articles, 111 of which at least partially addressed radiologic imaging. Of these, seven valid articles (two SRs and five primary research articles) were identified. The five primary research articles examined either colonic investigations (colonoscopy and barium enema examination) (n=5) or US (n=2) or both (n=2). Structural disease found infrequently in patients with IBS-type symptoms included diverticulosis, colorectal cancer, celiac disease, inflammatory bowel disease, and ovarian cancer. The incidence of structural disease in patients with concerning symptoms was low.

Conclusion: Although widely used, there is a surprising paucity of evidence guiding radiologic imaging in IBS. Radiologic imaging may not be required in patients with IBS without potentially concerning symptoms but should be considered where such symptoms exist, and choice of imaging study should be influenced by predominant symptoms. Definitive recommendations must await further research.