How often is a diagnosis of ulcerative colitis changed to Crohnʼs disease and vice versa?

Abstract

U lcerative colitis (UC) and Crohn’s disease (CD) have historically been grouped under the heading of inflammatory bowel disease (IBD) due to an overlap in pathogenesis, symptoms, and treatments. The wide range of clinical phenotypes seen within this spectrum raises the possibility that we are dealing with more than 2 distinct diseases. Molecular profiling suggests that IBD comprises a heterogeneous family of polygenic inflammatory disorders.1 For example, the discovery of the link between mutations in the NOD2/ CARD15 gene and stricturing ileal CD has provided a genetic basis for this distinct clinical phenotype. Indirect markers of IBD, such as anti-Saccharomyces cerevisiae antibody (ASCA), perinuclear-staining antineutrophil antibody (pANCA), and outer membrane porin C (OmpC) have also been associated with specific disease subtypes and risk of IBD.2,3 It is likely that further advances in molecular classification will provide clinicians with a molecular taxonomy based on specific transcriptional profiles, similar to the data from breast cancer and leukemia. In most cases, CD and UC are readily differentiated based on clinical, radiological, pathological, and endoscopic features. UC only affects the colon and is associated with mucosal inflammation that starts in the rectum and extends proximally in a continuous manner. In contrast, CD can affect any part of the gastrointestinal tract from the mouth to the anus, typically features “skip lesions” of transmural inflammation, yet often spares the rectum. Perianal involvement (fissures, fistulae, and abscesses), ileal involvement, and granulomas are highly specific to CD. Certain extraintestinal manifestations such as erythema nodosum and peripheral arthropathy are usually associated with CD rather than UC, although recent population-based studies have reported weaker associations.4,5 Other clinical clues useful in differentiating between CD and UC are smoking history and the absence of rectal bleeding. A case-control study of 21 patients initially diagnosed with UC who were later classified has having CD identified only nonbloody diarrhea and weight loss as independent predictors of change in diagnosis 6 However, despite these obvious differences between UC and CD, there are certain instances where the diagnosis is not so clear-cut. In addition, an individual’s clinico-pathological features may evolve over time. Consequently, a subset of patients with a label of UC may eventually be reclassified as CD and vice versa. In patients with UC a number of endoscopic and histological findings can create uncertainty about the diagnosis. Backwash ileitis, which can resemble the ileitis of CD, occurs in 10%–20% of patients with extensive colitis.7 It is usually associated with severe pancolitis, but may occasionally occur in those with left-sided disease. Endoscopy of patients with distal UC may also reveal proximal areas of colitis, in the cecum or ascending colon, which are distinct from the distal disease. These “skip lesions,” or cecal patches, have been reported in up to 20% of patients. Prospectively, these lesions have not been associated with the development of CD.8 Granulomas, which are highly specific to CD, can occur in up to 5% of UC biopsies in the setting of ruptured crypts (cryptolytic granulomas).9 The diagnostic accuracy of pathologists in classifying IBD type is 60%–74%, with uncertainty occurring in patients presenting with fulminant colitis, those with treated disease, and those seen at the earliest stages of disease.10 Finally, rectal sparing and patchy colitis may develop over time in patients with UC, although this usually represents differential healing due to treatment rather than a natural evolution of the disease. Due to this crossover of features, and despite extensive work-up, 10% of patients with chronic IBD do not fit either diagnostic category cleanly and are labeled as having an “indeterminate colitis.” Over time, as their clinical features continue to develop, most of these patients will acquire a diagnosis of UC or CD.11 However, the importance of the distinction between CD and UC has become less important in recent years, as the medical therapies have become less disease-specific. With the recent FDA approval of infliximab for UC, only methotrexate, natalizumab, and antibiotics remain specific therapies for CD. The distinction between the 2 diagnoses is particularly important, however, for surgical therapy and prognosis. The vast majority of patients with UC who undergo ileal pouch–anal anastomosis (IPAA) do well, as opposed to those patients with CD who undergo IPAA, up to 50% of whom require revision or diversion 12,13 There are few data examining how frequently patients From the Center for Inflammatory Bowel Disease, Division of Gastroenterology, Beth Israel Deaconess Medical Center, Massachusetts, USA. Copyright © 2008 Crohn’s & Colitis Foundation of America, Inc. DOI 10.1002/ibd.20593 Published online in Wiley InterScience (www.interscience.wiley.com).