THE PATIENT WITH BLOODY DIARRHEA When a patient presents with bloody diarrhea, the diagnosis is colitis until proven otherwise. It is possible that the patient has some other diar- rheal disorders or hemorrhoidal bleeding, however, colitis needs to be considered. A primary care physician should be determining whether the symptoms have been acute (<4 weeks) or chronic (>4 weeks) or recur- rent, whether the patient has risk factors for different colitides (antibiotic use, past pelvic radiotherapy, a family history of inflammatory bowel dis- ease [IBD]), and if there are relevant physical findings (such as an abnor- mal abdominal examination or blood on the rectal examining glove). After coordinating routine blood work (most importantly, a complete blood count) and stool tests (bacterial culture, Clostridium difficile toxin, and ova and parasite testing, if appropriate), the next step is a referral for a colonoscopy. In a patient with a positive stool test for a Clostridium difficile toxin or other bacterial pathogens, treatment can be initiated and an endoscopy can be deferred. An endoscopy should be pursued if the patient has recurrent or persistent symptoms after treatment. Should the patient have a flexible sigmoidoscopy or a full colonoscopy? If he or she is not too ill, then it may be reasonable to properly prepare the patient for a full colonoscopy, which would facilitate assessment of the entire colon (hence, if the lower left side of the colon is normal, then the rest of the bowel can still be assessed) and the terminal ileum. In all cases of colitis, the ileum should be intubated to complete the assessment. If the patient is quite ill, then a flexible sigmoidoscopy may be sufficient, particularly if the left side of the colon is abnormal. At that point, regard- less of the disease extent, the endoscopist can take biopsy specimens to help sort through the diagnosis and also suction liquid stool out of the colon to be sent to the microbiology laboratory. The initial assessment at a colonoscopy is a retroflexed view of the ano- rectum to determine whether internal hemorrhoids exist and also whether there might be inflammatory changes in the lower 2 to 3 cm of the rectum. Ulcerative proctitis or radiation proctitis can affect only the very distal rectum, and this can be missed if a retroflexed view is not undertaken. The endoscopist should then describe the abnormalities: whether there is gran- ularity, loss of vascular pattern, friability, exudate, or ulceration, and whether the abnormalities are confluent or patchy or whether they dis- cretely skip in distribution. However, whatever findings of inflammation the endoscopist identifies, it is important to be aware that these findings are all nonspecific. Often, Crohn’s disease of the colon (in the absence of www.asge.org Clinical Update American Society for Gastrointestinal Endoscopy Editor: Ronnie Fass, MD Commentary: Indications for endoscopy in patients with inflammatory bowel disease (IBD) are diverse and are commonly prompted by acute events or the need for colorectal cancer screening. In this article, Dr Charles Bernstein provides a critical review of the value and yield of an endoscopy in various clinical scenarios where an endoscopy is commonly entertained in patients with IBD. The author emphasizes that an endoscopy should not routinely be used to evaluate disease activity in patients who are doing well and are in remission. A summary table (Table 1) provides 10 valuable tips about how to integrate an endoscopy into the management of patients with IBD. – Ronnie Fass, MD, Editor Vol. 15, No. 3 January 2008 THE ROLE OF AN ENDOSCOPY IN INFLAMMATORY BOWEL DISEASE Charles N. Bernstein, MD IBD Clinical and Research Centre, University of Manitoba, Winnipeg, Manitoba, Canada