What is the incidence, prevalence, and natural history of indeterminate colitis?

The term “indeterminate colitis” (IC) was originally introduced in 1978 to describe surgical specimens from patients undergoing colectomy for inflammatory bowel disease (IBD), when the histological picture was not characteristic for either Crohn’s disease (CD) or ulcerative colitis (UC). Several studies have consistently shown that such a diagnosis could be made in 5%–15% of cases. However, up to twothirds of these patients could later be reclassified as affected by 1 of the 2 major forms of IBD when all clinical, radiological, and endoscopic evidence was taken into account.1,2 Subsequently, this term has been increasingly used to indicate nonoperated patients that were definitely affected by IBD confined to the colon but in which a clear diagnosis of either CD or UC could not be made. It has been recently recommended that the term “IBD unclassified” should be preferred to indicate these patients; nevertheless, the term “indeterminate colitis” appears to be constantly used in most epidemiological and clinical studies.3–5 Many population-based studies and hospital-based surveys have consistently found that, in adults, IC accounts for 3%–6% of all initial IBD diagnoses in Europe, as well as in Australia, Japan, Lebanon, Saudi Arabia, and South Africa; slightly lower figures have been occasionally reported from some south-European areas, and much higher figures (10%– 18%) in a single study from Canada. By contrast, the frequency of IC appears to be higher (up to 29%) among children, especially in northern Europe. To the best of our knowledge, no published data exist on the frequency of IC in American adults, whereas in children a frequency of 9% and of 13% has been found in Wisconsin and California, respectively.1 Very few data are available on the true incidence and prevalence of IC in the general population. In adults, an incidence of about 1/100,000/year and a prevalence of 3–7/ 100,000 were found in some European countries (Denmark, Hungary, Spain, Holland) but a higher incidence (about 2.5/ 100,000/year) has been reported in Scandinavia and Canada. The incidence of IC in children has been assessed in Sweden, France, and Wales, ranging from 0.12–0.7/100,000/year.1 It is generally believed that in most cases IC merely represents a provisional diagnosis; indeed, follow-up studies have shown that a vast proportion (up to 80%) of patients are eventually reclassified as being affected by either CD or UC, and up to 20% as not affected by IBD. However, a small subset of patients exists in which IC still remains the most accurate diagnosis a long time after disease onset (more than 10 years). It is likely that in such cases IC really represents a separate clinical entity. It has been suggested that serologic testing can help identifying these patients. In one study,6 a definite diagnosis of either CD or UC was eventually made in 40% of patients that were positive for either perinuclear antineutrophil cytoplasmic antibodies (ANCA) or anti-Saccharomyces cerevisiae antibodies (ASCA), but in only 10% or those that were initially negative for both. In more recent years, capsule endoscopy has been emerging as a promising tool to identify subtle small bowel abnormalities in patients with IC, thus allowing a change in diagnosis to CD.7 Very little is known about the clinical course of patients with “true” IC. It is generally assumed that such patients are clinically more similar to those with UC than those with CD. However, it has been occasionally reported that, when compared to patients with UC, patients with IC have a more aggressive disease course, with a higher risk or relapse, a more frequent need for immunosuppressants, and a higher rate of colectomy. Moreover, in patients with IC undergoing proctocolectomy with ileal pouch–anal anastomosis, the rate of postoperative complications is higher than in patients with UC and, in particular, the rate of pouch failure has been reported to be nearly 30%.8 No clinical trials have specifically evaluated the efficacy of any therapeutic agent in patients with IC, but the current recommendation is to treat these patients as those affected by UC are treated. In an open series the clinical response to infliximab was quite similar in patients with UC and IC. No consensus exists on which is the best surgical treatment for these patients, with some experts recommending permanent ileostomy and others believing that the complication rate mentioned above still warrants performing ileal pouch. anastomosis.