A t present, despite the claims of promoters of new diagnostic approaches, the diagnoses of Crohn's disease (CD) and ulcerative colitis (UC) depend on a combination of clinical, laboratory, radiographic, endoscopic, and histological criteria. Once these diagnoses are established by traditional means, it would be desirable to predict the clinical course and prognosis for an individual patient. There have been substantial advances in the development and commercialization of a number of serologic and genetic markers in patients with IBD. These markers have incrementally advanced our understanding of the underlying aberrant immune responses in patients with IBD, particularly in CD. Nonetheless, they have yet to be prospectively shown to be sufficiently sensitive and specific to inform a given patient of their future. We remain humbled by our insufficient skills when it comes to gazing into the crystal ball of prognosis. In this article we describe the case for the importance of accurate disease phenotyping by our traditional and time-tested clinical means. In addition, we will address the potential pitfalls in predicting clinical course of disease based on the currently available serologic and genetic markers.
Predictive Phenotypic Features of UCThe severity of UC is entirely based on its clinical phenotype. 1 Disease severity, which informs immediate treatment decisions, is categorized into mild, moderate, severe, and fulminant based on the assessment of simple, inexpensive, clinically relevant variables including stool number, rectal bleeding, fever, tachycardia, erythrocyte sedimentation rate, hematocrit, and endoscopic appearance. Predictive phenotypic information regarding UC has been available for many years and provides useful information to patients. A large case series (and therefore prone to referral bias) reported an analysis of 1116 patients with UC. Forty-six percent were found to have proctosigmoiditis, 17% left-sided colitis, and 37% pancolitis. 2 After a mean follow-up of more than 12 years, patients with pancolitis were at greatest risk for complications (colonic hemorrhage, toxic colitis, and surgery) and rates of taking medications and of being asymptomatic were 37% and 67%, respectively. A population-based study of 1161 Danish patients with UC reported that about half of patients were in remission at any time and 90% had an intermittent course. Disease activity in foregoing years indicated with 70%-80% probability that the disease would continue the following year. Most patients' lives were relatively little influenced by the disease and the probability of maintaining working capacity after 10 years was 92.8%. 3 If patients require treatment with corticosteroids, natural history studies can provide a guide as to the likely response. Analyzing population-based data from Olmsted County, Minnesota, researchers found that 34% of patients with UC were treated with corticosteroids and complete and partial remission rates were 58% and 26%, respectively. 4 Finally, it is known that smoking habit strongly ...
